Beata Stawarska - Saussure's Philosophy of Language As Phenomenology - Undoing The Doctrine of The Course in General Linguistics-Oxford University Press (2015) PDF

Saussure’s Philosophy of Language as Phenomenology
Saussure’s Philosophy
of Language as
Phenomenology
undoing the doctrine of the
course in general linguistics
beata stawarska
1
1
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide.
Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press

in the UK and certain other countries.
Published in the United States of America by

Oxford University Press
198 Madison Avenue, New York, NY 10016
© Oxford University Press 2015
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, by license, or under terms agreed with the
appropriate reproduction rights organization. Inquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data

Stawarska, Beata.
Saussure’s philosophy of language as phenomenology : undoing the doctrine
of the course in general linguistics / Beata Stawarska.
p. cm.
Includes bibliographical references and index.
ISBN 978–0–19–021302–2 — ISBN 978–0–19–021303–9 — ISBN 978–0–19–021304–6
1. Saussure, Ferdinand de, 1857–1913—Criticism and interpretation. 2. Structural linguistics.
3. Phenomenology. 4. Linguistics—History. I. Title.
P85.S18S73 2014
410.92—dc23
2014025383
9 8 7 6 5 4 3 2 1
Printed in the United States of America on acid-free paper

To Amelia
CONTENTS
Acknowledgments xi
Introduction 1
The speaking subject 2
Saussure’s general linguistics revisited 5
The book and the author function 9
The outline of contents 16
PA RT I | “Saussurean doctrine” and its discontents

1 The signifier and the signified 23
Arbitrariness of the sign 24
Graphic illustration: sign/signifier/signified 26
Critique of the nomenclature view of language 31
The nomenclature view of language and Bergson’s
philosophy of life 44

La langue: systemic, social, historical 47
2 Phonocentrism: Derrida 71
Natural symbolism of sound 72
Metaphysics of presence, and arche-writing 79
3
La langue and la parole, synchrony and
diachrony 85
Double essence of language 86
Dual way to language 97
PA RT I I | General linguistics: science and/or philosophy
of language
4 Involuntary assumption of substance, and points
of view in linguistics 109
5 Saussure’s general linguistics as linguistic

phenomenology 120
Predecessors: Kazan School of Linguistics,
Kruszewski 120
Saussure’s phenomenon 126
Effacement of consciousness in the Course 130
Consciousness structured like a language 134
Beyond the doctrine 147
6 Contributions to linguistic phenomenology: Hegel,

Husserl, Merleau-Ponty 154
Phenomenological science: Hegel and Saussure 155
Early structuralism: Jakobson and Husserl 169
Phenomenology of language: Merleau-Ponty and
Saussure 181
PA RT I I I | The inception and the reception of the

“Saussurean doctrine”: the Course
7 The editorial inception of the Course: Bally
and Sechehaye 193
Ghostwriting 194
Book reviewing 205
The famous formula 213
Elementary structures of kinship in academia 219
Reversal of order 225
A taste for great abstractions 230
8 Structuralist and poststructuralist reception

of the Course 239
Structuralism: east and west 242
Poststructuralism: Levi-Strauss, Lacan, Derrida 248
viii Contents
Appendix 1. English translations of the Course 257
Saussure séance 257
Saussure and Santa 260
Appendix 2. Saussure’s silence 265
Bibliography 273
Index 281
Contents ix
ACKNOWLEDGMENTS
R esearch for this book was made possible by the Humboldt Research
Fellowship for Experienced Researchers at the University of Heidel-
berg, Germany, 2009-11. I wish to acknowledge the generosity and hospi-
tality of my host Thomas Fuchs, and the reliable assistance of Rixta Fam-
bach throughout my visit. I benefitted from the exchanges, seminars, and
conferences with members of the Marie Curie ITN TESIS (“Towards an
Embodied Science of Intersubjectivity”) European research network, es-
pecially Hanne de Jaegher, Sonja Frohoff, Sanneke de Haan, Monika
Dullstein, Thiemo Breyer, Vasu Reddy, Alan Costall, Giacomo Rizzolatti,
Laila Craighero, and Michael Pauen. My home department at the Univer-
sity of Oregon welcomes innovative research in philosophy broadly con-
strued, and I am grateful to be a member of this cutting-edge program.
I wish to especially acknowledge the intellectual inspiration and enthusi-
astic support of my colleagues Bonnie Mann, Ted Toadvine, and Scott
Pratt in the Philosophy department. I also benefitted from the excellent
discussions with our Graduate students who were among the first intel-
lectual interlocutors for this project. I wish to recognize the participants
in my seminars on phenomenology, structuralism, and post-structuralism
at Oregon, especially Megan Burke, David Craig, Russell Duvernoy, and
Eric Rodriguez. My colleague and friend Joseph Fracchia offered un-
matched enthusiasm as well as helpful feedback on this book over a
number of years. I am grateful to Eva Simms (Duquesne University) for
her spirited encouragement. Christopher Norris (Cardiff University) and
Jonathan Culler (Cornell University) both offered valuable insights and
suggestions during the review stage, and I appreciate their input. Hallie
Stebbins, the acquisitions editor at Oxford University Press, shepherded
this book project with professionalism and patience. Russell Duvernoy
served as an editorial assistant in this venture, and was invaluable at the
proofreading and indexing stage. I gratefully acknowledge having re-
ceived an editorial assistantship subvention from the Philosophy Depart-
ment and a faculty publication subvention from the Oregon Humanities
Center at the University of Oregon. I received the publishers’ permissions
to reproduce figures from © Ferdinand de Saussure, translated by Roy
Harris, Course in General Linguistics (Bloomsbury Academic, an im-
print of Bloomsbury Publishing Plc), 2013; Ferdinand de Saussure, edited
and translated by Eisuke Komastu and Roy Harris, Saussure’s Third
Course of Lectures on General Linguistics (1910–11): from the Notebooks
of Emile Constantin (Elsevier Limited), 1993; and Ferdinand de Saussure,
edited by Simon Bouquet and Rudolf Engler, with the assistance of Antoi-
nette Weil, translated by Carol Sanders and Matthew Pires, with the as-
sistance of Peter Figueroa, Writings in General Linguistics (Oxford Uni-
versity Press) 2006. Some of the material published in article format in
Continental Philosophy Review, Philosophy Today, and CHIASMI Inter-
national is being reproduced here with the permission of the editors.
xii Acknowledgments
Saussure’s Philosophy of Language as Phenomenology
Introduction
T
ruisms are tricky. A truism goes without saying, but a question
or a problem can only be raised about something that can be said
without sounding too obvious for mention. Now, it is rarely ever
said, outside of introductory-level classes in philosophy, critical theory,
and modern cultures perhaps, that Ferdinand de Saussure is the official
founder of the structuralist movement in linguistics, the human sciences,
and philosophy. It hardly ever needs to be said that Saussure’s method
consists of a structural analysis of language viewed as a system of signs.
Yet another truism would be to say that structural analysis is opposed to
phenomenological description, since the former removes any reference to
the speaking subject while the latter adopts it as an irreducible datum. It
therefore goes without saying that Saussurean structuralism lies on the op-
posite end of the spectrum from the tradition of phenomenology. In the
upper philosophical circles at least, such things are rarely ever said.
I propose to revisit these interlinked truisms, to say them, and to show
them to be suspect. To accomplish this task I will venture outside of phil-
osophical parlors and into the lesser known pockets of intellectual activity
in parts of Europe where Saussure’s legacy is being reexamined in striking
and provocative ways. This research is conducted mainly in the special-
ized field of historical linguistics, and only partially represented in the
English-language publications. One could let this new research remain just
as specialized and pocketed as it is, but that would only harden the existing
blind spot in the philosophical reception of Saussure’s linguistics. Please
be reminded what a formidable influence Saussure’s linguistics exerted
upon the developments within contemporary European philosophy, as well
as the human sciences (psychoanalysis, anthropology, literary studies) in
postwar Europe and the United States. It laid the foundation for the struc-
turalist and the poststructuralist movements, made it possible to study all
human phenomena as objectively organized and structured systems, and
raised inquiries into human matters to an equal level with research in the
hard sciences. This structuralist euphoria eventually died down, and was
replaced by a poststructuralist critique of an assumed scientistic ideology,
positivist aspirations, and unexamined commitments to a closed and au-
tonomous system. The structuralist dream, if realized, would have made
its own perpetrators obsolete. And yet, the crisis of structuralism in the
humanities notwithstanding, the recent history of philosophy is dictated by
its rise and fall. The more philosophically refined permutations of the
structuralist science into a critique of the background beliefs and aspira-
tions held in such a venture remain in the grip of a semiotic program, and
retain the basic idea that signs signify with a degree of autonomy with
regard to subjects and are organized into systems which produce real ef-
fects. This poststructuralist view still sits uneasily in the company of a
phenomenological subject, if the two ever sit in the same discursive space.
My task will not be to rewrite the tangled history of structuralism, post-
structuralism, and phenomenology within contemporary continental phi-
losophy in toto, but to make a more modest and focused contribution to the
conversation by looking back to the official sources of the structuralist
movement in Saussure’s linguistics, to reexamine them in light of the
scholarly contributions of the last five decades which challenge the re-
ceived, structuralist view of Saussure, and, finally, to expose a phenome-
nological current in his own work. I will therefore be looking back to the
adopted foundations of structuralism, and if I trace some cracks in its bed-
rock, it is ultimately in the hope of suspending the institutionalized antago-
nism and enabling a renewed rapprochement between structure and sub-
ject based approaches to language and experience.
The speaking subject
The dominant scholarly view is that these approaches are at odds. Accept-
ing that linguistic expression is shaped by the received and sedimented
language systems would be contingent on sacrificing the subject on a semi-
ological altar. Saussure’s general linguistics is usually heralded as having
made this sacrifice indispensable, and for having thus initiated a structur-
alist turn away from speaking subjectivity. A representative sample of this
view can be found in Sturrock’s Structuralism, where he writes:
Many of us like to think that when we use language we control it more or

less totally and that it is we who determine the sequence of words or thoughts
2 Saussure’s Philosophy of Language as Phenomenology

each time we write or speak; we are not happy to allow that language itself
can prove more powerful than we are and that the association of the signs
we have already used may be determining the choice of the signs to come.
This loss of authority in the “speaking subject” or language-user is a most
important and contentious aspect of Structuralism, exploited to the full in
post-Structuralism . . . and it can be traced back to this insistence of Saus-
sure that the language-system impinges at every moment on language-
events. (Sturrock, 2003, p. 43)
The imagined relation between the speaking subject and the language
system suggests a battlefield where the speaker’s claims to authority and
control over linguistic expression engage in a tug of war with the lan-
guage’s own greater power to impose its set of historically weighed asso-
ciations onto the speaking present. Is this vision of a conflict for absolute
power waged between subject and system a datum of experience that a
speaking subject necessarily and universally has as soon as she begins to
speak or write? Or does it reflect an institutionalized antagonism between
traditions of inquiry which privilege subjective experience on the one hand
and semiological systems on the other, that is, phenomenology and struc-
turalism (and poststructuralism)? The latter conflict is well established and
long lived; it may even cease to appear as one out of many possible institu-
tional and doctrinal permutations and emerge rather as a direct conse-
quence of the way human reality itself segments into an area of subjective
expression on the one hand, and sedimented systems of signification on the
other. The more engrained the antagonism, the easier is it to assume that
each tradition presides over an autonomous, well-sealed, segment of real-
ity. The more entrenched the habit to preserve and police the borders be-
tween these two areas, the stronger is a sense that what lies on the other
side is foreign and strange. It is this apparent strangeness that can be used
to justify the envisioned conflict for power between the speaker and the
language itself; when internalized by the force of an inherited intellectual
tradition, this conflict can seem internal to the very experience of the
speaking subject. Linguistic expression seems then accompanied by alien-
ation and impotence in having to string the signs together “by the rule” and
not by intent. Yet if the proposed archaeology of this presumed surrender
to language’s greater powers holds (namely, that this perception is in-
formed by academic antagonisms and not a neutral datum of experience),
then it only proves that the views of what is involved in an experience of
speaking are cast within a net of institutionalized expectations about how
the subject and language fit together (not very well), what kind of relation
introduction 3
they enter into (conflict for total control), and how the subject feels when
deploying resources not of its own making (unhappy). Yet the speaking
subject never had total control over linguistic expression. There was no
happy Garden of Eden where we could speak freely, in an unburdened
present, and from which we would have one day fallen, and felt the unhap-
piness of an uprooted stranger ever since; the tower of Babel is a perma-
nent fixture of our landscape. The speaking subject never had god-like
authority to first let there be word; it does not experience a daily conflict
between coming to expression and borrowing from the tradition. If a sub-
ject were to undergo such a profound schism, it would not be a speaking
subject any longer.
Another vision of the speaking subject suggests an ambiguous zone
where the expressive intent is driven and guided by the received linguistic
resources, thinks according to them, but not necessarily by reiterating
worn-out clichés. According to this vision, the subject borrows from the
mighty powers of language and makes them her own—without ever being
able to claim total possession. The entry into language does not figure here
as an eviction from a happy home into unfathomable foreign lands; it feels
rather like drawing on an inexhaustible riches which can never be signed
with my proper name only, and yet enable me to speak and write in my own
words. Never a simple annexing of the other by the self, language can
avoid steering the subject toward impotence and alienation (even though it
may sometimes lead to both); it is a space of wonder that my innermost
powers of expression are on a perennial loan, without rights of ownership.
And yet expression is possible.
This ambiguous and porous space intersecting the received and the
proper is inhabited with some happiness by some speakers on some days,
and constitutes the contingent freedom of linguistic expression. This speak-
ing subjectivity is best approached by a combined descriptive and structure-
based mode of inquiry. As Kristeva argued already, the speaking subject
does not map onto the ego of transcendental phenomenology, this “peniten-
tiary” subject walled into the prison of knowledge, and better associated
with the ideal of a metalanguage than the language in actu (Culler, 2006).
The speaking subject may be better described as a subject on trial and in
process (en procès), hence worked through rather than competing for signi-
fying resources with the object language. Yet while Kristeva positions this
speaking subject in opposition to Saussure’s linguistics and its philosophical
offspring (ibid.), I argue that Saussure’s own, lesser known reflections help
to chart the very zone of linguistic experience undecided between renewing
subjectivity and sedimented signification that I believe Kristeva charts in

her own map of the semiotic and the symbolic relations (Kristeva, 1984).
These reflections contain an emphasis that “the language-system impinges
at every moment on language-events” (Sturrock, 2003, cited above), yet it is
not a vision of conflict between speaker and structure but rather a chiasm-
like interplay of entangled forces that mark the shifting place we occupy in
language. I argue that this lesser publicized variant of Saussure’s linguistics
helps to decipher linguistic experience, and safeguards the thinker from de-
ciding too quickly in favor of either phenomenological description of puri-
fied consciousness or structural analysis of semiological systems. It helps to
think anew about our being-in-language, and to move beyond the institu-
tionalized schism of these two methodologies.
Saussure’s general linguistics revisited
The reception of Saussure’s work has been largely based on the posthu-
mously published edition of the Cours de Linguistique Générale (1916).
This volume was ghostwritten and published by Charles Bally and
Albert Sechehaye in collaboration with Albert Riedlinger; while their
edition is typically assumed, in the philosophical circles at least, to offer,
in a book format, a simple recast of the lectures on general linguistics
that Saussure gave between 1907 and 1909 at the University of Geneva
by the students who attended them, it should be noted that Bally and
Sechehaye did not attend any of the lectures, while Riedlinger, who did,
may have been co-opted in the editorial process to increase its credibil-
ity and also expressed profound disappointment with the final product
(see ch. 8 for development). The publication of Godel’s Sources Manus-
crites (1957) and Engler’s Édition Critique (1989 [1967/1968]) docu-
ments in detail the discrepancies between Bally and Sechehaye’s 1916
redaction and the students’ lecture notes (and other source materials).
These critical works give evidence of a heavy editorial hand: Bally and
Sechehaye’s version changed the order of presentation, and altered the
contents and style found in the source materials, possibly according to
the editors’ own vision of general linguistics as objective science and in
response to the expectations relative to a classic academic book format
(see esp. chs. 1, 3, and 7 for development).
Scholarly research on Saussure’s linguistics of the last five decades has
radically shifted the ways in which Saussure’s intellectual heritage is inter-
preted. In response to the documented discrepancies between the edited
version of Saussure’s linguistics from the Course and the source materials
introduction 5
(autographed writings by Saussure, the student lecture notes from his
courses on general linguistics at the University of Geneva 1907–1911), as
well as the recently discovered and published autographed writings by
Saussure (Ecrits de Linguistique Générale, 2003/Writings in General Lin-
guistics, 2006), and in light of the revelations from the recently published
correspondence between the main stakeholders in Saussure’s estate (the
two editors of the posthumously published Cours de Linguistique Gé-
nérale, 1916/Course in General Linguistics, and Saussure’s students and
colleagues), contemporary scholars are shifting from the so-called first to
the second editorial paradigm of Saussure’s general linguistics.1 Saussure
scholars are gradually abandoning an earlier (the first) research paradigm
spanning over the last five decades, which compared the official version of
the Course on General Linguistics from the 1916 text with the source ma-
terials. While this approach helped to locate the manuscript sources of the
editorial rendering of the lectures (and also to establish a lack of manu-
script evidence for some of the claims made in the Course), and offered a
critical perspective on Bally and Sechehaye’s editorial legacy, it inadvert-
ently maintained the status of the Course as central reference in Saussure
scholarship, its documented shortcomings notwithstanding. In their pas-
sage to the second editorial paradigm, Saussure scholars are largely aban-
doning the Course and its structuralist legacy, and working solely with
historically authentic texts.
Anglophone scholarship has lagged behind these scholarly develop-
ments, in part due to an absence of an English-language critical edition of
the Course.2 Philosophical scholarship has similarly lagged behind, with
the Course still serving as the official version of Saussure’s linguistics in
scholarly publications and college-level pedagogy alike. I propose to follow
the lead of recent Saussure scholarship and offer a critical study of the ed-
itorial presentation of philosophically relevant problems in the Course in
1
As presented in Simon Bouquet, 1999.
2
Critical editions are available in the French (Engler, 1989 [1967/1968]) and Italian (de Mauro,
1967), translated into French (2005 [1967]). English-speaking readers have until recently had
access to Roy Harris’s 1972 translation (most recent printing from 2007), which includes the
translator’s introduction with some dismissive comments about the developments in critical
Saussure scholarship (discussed below in Appendix 1, Saussure séance). This translation was
reissued by Bloomsbury as part of the Bloomsbury Revelations Series in 2013; it includes the
original translator’s introduction (now titled “Introduction to the Bloomsbury Revelations
Edition”) as well as a new, ten-page-long “Introduction” (pp. xxix–xxxix), which acknowledges
the existence of source materials and the concurrent difficulty of associating Saussure’s general
linguistics with structuralism but endorses the Course based on the notion that “[Saussure’s] 1916
editors commendably made the best of a bad job” (p. xxxix), the “bad job” referring to Saussure’s
assumed failure to examine his ideas in requisite detail (ibid.).

light of the source materials (part I), a new and original interpretation of
Saussure’s Nachlass bearing on his conceptions of the science and/or phi-
losophy of language (part II), and a reconstituted history of the editorial
production of the Course as a site of the official Saussurean doctrine, and
its structuralist legacy (part III). I will follow the lead of contemporary
Saussure scholarship in the process and, considering the Anglophone
scholarly context in particular, employ both the first and the second edito-
rial paradigms in my engagement with Saussure’s general linguistics.
Throughout, I will take on the challenge raised by Bouquet when he
writes that “the thought of Saussure appears as a largely untouched territory
to be explored by the philosophers—philosophers of language, philoso-
phers of science, philosophers of knowledge, philosophers tout court—this
thought currently possesses the highest heuristic value for them” (Bouquet,
1997, p. viii, my translation). My own exploration of Saussure’s thought will
expose the shortcomings of the official doctrine associated with his name
and reclaim the philosophical dimension of his project by way of a linguis-
tic phenomenology. I do not, however, seek to dismiss the tremendous suc-
cess that a focus on the linguistic system attained in the relevant fields of
study, and I do not purport to offer the only possible or the “correct” reading
of Saussure’s general linguistics. My ultimate goal is to point out scholarly
possibilities opened up by a phenomenological interpretation of Saussure’s
Nachlass to the philosophical understanding of language as an ambiguous
sphere which intersects individual expression in the present with the histor-
ically sedimented societal conventions, and which is undecidable between
consciousness and the unconscious. I consider this ambiguity essential to
the situation of a speaking subject, and I regard Saussure’s Nachlass to be a
source of hitherto unacknowledged philosophical insights in this regard. I
simultaneously acknowledge the possibility, as well as the real existence, of
other interpretations and reclamations of the Nachlass. Consider in this
regard especially recent studies on the relation between Saussure’s general
linguistics and the romantic tradition (B. Gasparov, 2013), scientific linguis-
tics (Bouissac, 2010), agency within language (Starobinski, 1979), and in-
tertextuality (Thibault, 2005). I hope that my specific project will join their
Wade Baskin’s earlier translation of the Course was reissued by Columbia UP in 2011. Edited
by Perry Meisel and Haun Saussy, it contains a new introduction which seeks to both
acknowledge the recent developments in critical Saussure scholarship and rehabilitate the edited
version of the Course (this effort is discussed below in Appendix 1, ‘Saussure and Santa’). The
reissue also contains notes to the edited text, which run just four pages in length and consist
mainly of references to the critical editions by de Mauro and Engler. It does not therefore
constitute, nor does it claim to constitute, a critical edition of the Course in English.
introduction 7
efforts in fostering a scholarly space where multiple interpretations of Saus-
sure’s general linguistics can coexist as many alternatives to an exclusively
structuralist claim to his work.
Even though the project carried out here is critical insofar as it calls into
question the legitimacy of the Course and its structuralist legacy, it is also
the case that the value and interest of such a critique are themselves con-
tingent on the enormous success the Course attained in shaping the struc-
turalist (and poststructuralist) developments in the twentieth-century his-
tory of philosophy and the human sciences. The Course offered a powerful
conceptual paradigm and a methodological framework for the study of
cultural phenomena in a wide range of fields. It created a new and im-
mensely influential model for thinking about cultural signification in theo-
retical as well as applied contexts. The shortcomings of the Course docu-
mented here—its scientific dogmatism, methodological naïveté, logic of
opposing binaries—helped from a practical standpoint to establish the
measure of its success in supplying a practicable research tool for the
human scientist. As documented in more detail below (Appendix 2,Saus-
sure’s silence), it is unlikely that Saussure himself would have ever written
a comparable programmatic statement of general linguistics. An extremely
prolific writer who authored thousands upon thousands of manuscript
pages, he was in the grips of “epistolophobia” (a fear of writing) and suc-
cumbed to a “thirty-year silence” which deterred the publication of numer-
ous writing projects, including the project for a book on general linguis-
tics. It is therefore unlikely that Saussure would have attained the status of
a canonical figure in the twentieth-century history of philosophy and the
human sciences had it not been for the material presence of the Course on
the market of ideas.3
Any critical study of the Course is therefore itself enabled by the edi-
tors’ indelible success in making their version of Saussure’s teaching
widely accessible in a classic academic book format for decades. Herein
lays the performative paradox faced by any critical study of a Great Book:
the force of the critique depends to a degree upon the recognized impor-
tance of the object being critiqued. It is due to this performative paradox
3
Perhaps his status would have been that of a prodigious linguist whose career waned after he
had left the Parisian intellectual circles for Geneva in 1891, where he was surrounded by fewer
and less advanced students and “seemed to be settling into a decent provincial obscurity” (Culler,
1986, p. 23). See also Joseph’s monumental intellectual biography (2012, especially ch. 19 “The
end: 1911–1913”) for a detailed account of Saussure’s final years in Geneva, including the
vilifying personal attacks by the local journal Le Génévois, his deteriorating health, and his
withdrawal from public life.

that researchers have been moving away from the first to the second edito-
rial paradigm of Saussure’s general linguistics; as noted above, the first
paradigm, that is, a critical comparative study of the Course and the source
materials, effectively rehabilitates the Course as the central reference in
Saussure scholarship and reinstates the importance of the object under cri-
tique. The passage to the second editorial paradigm, that is a study focused
exclusively on Saussure’s Nachlass, offers an opportunity to assess the im-
portance of Saussure’s general linguistics according to the usual scholarly
standards of historical authenticity. It needs to be conceded, however, that
without the ambitious editorial venture that led to the production, structur-
alist appropriation, deconstruction, and, finally, critical revision of the
Course, the recent discovery and publication of Saussure’s Nachlass would
most likely present a lesser degree of interest to contemporary scholars in
philosophy and related fields. Consequently, the ideas expressed in the
Nachlass would most likely attract less attention than they deserve.
The book and the author function
The Course in General Linguistics is widely regarded as a locus of the

Saussurean doctrine, whose familiar oppositional pairings: signifier and
signified, speech (la parole) and language system (la langue), diachrony
and synchrony, became the hallmark of structuralism. Unsurprisingly, the
book’s material fate is closely bound up with that of the structuralist move-
ment itself, and the book’s rate of reprints and translations closely corre-
lates with the ups and downs of this movement. Thus, while the original
text saw five editions between 1916 and 1955, followed by another five
between 1955 and 1963, there were as many as twenty-three between 1964
and 1985 (Gadet, 1989, p. 113). While only five translations were issued
between 1916 and 1960, over twice as many (twelve) appeared between
1960 and 1980. This increased rate can be explained by the structuralist
claim of the Course in the 1960s as its historical point of origin, a founda-
tional text; this claim had a powerful enough grip on both the scholarly
and popular imagination to cement the view that structuralism was derived
directly from Saussureanism (albeit five decades after its original publica-
tion), or that Saussure’s ideas induced the development of the structuralist
doctrine as if by their own force. The Course became a little red book read
widely in the human sciences and philosophy, including structural linguis-
tics (Bloomfield, Greimas, Hjelmslvev), structural anthropology (Levi-
Strauss), semiology (Barthes), and psychoanalysis (Lacan); it was also read
introduction 9
deconstructively within poststructuralism (Derrida). The book’s publish-
ing success is closely tied to a programmatic reception of its contents.
It is also of note that, despite its specialized focus, the Course has been
a veritable best seller with a staggering rate of reprints and translations. At
the time of this writing, the popular English-language translation by Roy
Harris is in its seventeenth printing (copyright 1983) and was reissued as
part of the Bloomsbury Revelations Series, with a new introduction by
Harris, in 2013; the Wade Baskin earlier translation went through six edi-
tions, and was reissued with some pomp, accompanied by a new introduc-
tion and notes, by Columbia UP in 2011 (copyright 1959). This steady pres-
ence of the Course on the Anglophone market of ideas marks a considerable
editorial success considering the relative tardiness (by over four decades)
of the first English translation with regard to the original French edition.4
The Course has been widely translated into non-European and European
languages.5 Its seemingly endless reprint and translation process both re-
flects and shapes the perception of the Course as a member of the Great
Books series, and an indispensable element of our cultural heritage. This
iteration process can make questions about how and why this commodity
became produced seem obsolete; the product justifies itself through the
infinite chain of iteration, circulation, and reception. It does not appear as
a material artifact anymore, but as an external envelope to the living word
of Saussure.
Despite its massive presence on the market of ideas, the Course in Gen-
eral Linguistics is not a book—at least not according to the criteria for a
single-authored academic treatise that it is usually classified as. To be sure,
there is a (serially produced) volume with the above-mentioned title bear-
ing Ferdinand de Saussure’s name in the author slot on the title page. It is
a set of printed sheets bound together, which conforms to the standard
expectations of an academic treatise. Its contents are laid out in the usual
order, with a preface, an introduction, a series of chapters, appendices and
4
This tardiness does not imply that Saussure left no mark on the intellectual developments in the
United States prior to that date; however, it was filtered by Bloomfield’s appropriation, which
systematically avoided reference to Saussurean sources and emphasized the scienticity of
linguistics instead.
5
It was first translated into Japanese (1928; reedited in 1940, 1941, 1950). Other translations
followed suit: German (1931), Russian (1933), Spanish (1945, 1955, 1959, 1961, all in Buenos
Aires). The first English language translation from 1959 was followed by translations into Polish
(1961), Italian (1967, by T. de Mauro, including a critical apparatus and bibliography), Hungarian
(1967), Serbo-Croat (1969), Swedish (1970), Portuguese (1971), Vietnamese (1973), Korean
(1975), Turkish (1976), Albanian (1977), Greek (1979), Chinese (1980), and English again (UK,
Harris, 1986 [1983]).

an index. As citizens of the “civilization of the book” (Derrida, 1998, p. 8),
we will be drawn to its ponderable shape and architectonic structure; the
volume meets our expectations of a stable foundation for knowledge (ibid.,
p. xi). It holds a promise of a coherent doctrine whose internal organiza-
tion is reflected in the external layout of the component parts. It exerts a
fascination upon the reader who is offered a chance of mastery of a domain
mapped out from the beginning to the end.
Classified as a Great Book, the Course is usually considered a founda-
tional text which establishes a scientific program for the new discipline of
general linguistics whose main principles and terminology became gradu-
ally extended to other fields in the humanities. The Course is therefore
regarded as a privileged site of Saussure’s ideas, as Saussure’s work, ac-
cording to the usual association between the notion of a work and that of
an author, wherein the identity of the work is contingent on an (unacknowl-
edged) assumption of an identifiable author who deposited his principal
ideas therein. I will decipher this association drawing on Foucault’s analy-
sis of the author function, and relate it to the difficulties surrounding au-
thorship in the case of the Course.
Foucault argued that a philosopher or literary critic might declare the
disappearance—even the death—of an author and yet maintain “the priv-
ileged position of the author” nonetheless (2001, p. 10). The author privi-
lege is simultaneously maintained and disavowed within the idea of the
work. Even though a focus on the work shifts emphasis away from its rela-
tionship to the author, and turns to the structure, architecture, the intrinsic
form, the play of internal relations of the work itself, the very unity of the
work continues to be established in relation to the author. As Foucault
wonders: “Is [the work] not what the author has written?” (ibid., p. 11).
This empirical qualification is understandably weak from a theoretical
standpoint and it raises a host of issues (should a laundry list found among
Nietzsche’s notes be identified as his work and published, and if not, why
not?). In addition to its obvious theoretical weakness, the relation between
the work and the author also performs an (unacknowledged) ideological
function. The author serves largely to determine legitimate attribution of a
text, a book, or a work to a given individual—a function of special import
in the case of “great” literary authors, or the founders of science (ibid.), and
serves therefore as an instrument of exclusion as well. Foucault exposes
therefore the author function as an “ideological product” which serves to
impede free proliferation of meaning, and acts as “the principle of thrift”
that limits and excludes the free play and circulation of meaning (ibid.,
pp. 21–22).
introduction 11
If empirical factors typically regulate, or should regulate, to a degree at
least, the attribution of authorship such that the author functions not exclu-
sively in an ideological but also in a historically constrained manner (even
though in agreement with Foucault the author is always an ideological
product), then the inception and the dominant reception of the Course as
Saussure’s work become fraught with difficulties. There is the obvious dif-
ficulty of assigning an empirical author to a book ghostwritten by Bally
and Sechehaye after Saussure’s physical death (see ch. 7 for details of the
ghostwriting process). There is also the difficulty of reading the Course as
a programmatic statement for the human sciences and philosophy in an
effort to claim Saussure as the founder of the structuralist movement in the
post-Second World War France (see ch. 8 for details of the structuralist
reception of the Course). Finally, there is the difficulty of reading the
Course deconstructively in poststructuralism, for even though such a read-
ing focuses primarily on the structure, the play of internal relations and
tensions within the Course, it continues to assume an author privilege in
order to guarantee the book’s unity as Saussure’s work—however fragile
and fractured that unity may turn out to be (see chs. 2 and 9 for an engage-
ment with Derrida’s reception).
For what guarantees the unity of the Course if not its identification as
Saussure’s work, that is, the traditional assumption of an author—this
“genial creator” who deposits a wealth of signification into his products?
Only on that assumption can a deconstructive reading of the Course in Of
Grammatology, which exposes the manifest tensions between an emphasis
on sound or voice as a site of signification and a concurrent claim that sig-
nification emerges only within a relational host of differences without pos-
itive content, emerge. Only then does “Saussure” turn out to be a scholar at
odds with himself: he founds a positive semiological science with one hand,
while letting it break into a thousand pieces with the other. In sum, the
author function served the ideological role of legitimizing the Course as
Saussure’s work both as part of the editorial conception and of the domi-
nant structuralist and poststructuralist reception of the Course; what got
excluded in the process was the textual universe of the Nachlass. As a
result, the scholarly focus was largely directed to the familiar doctrine from
the Great Book—even if and when the de facto existence of the original
materials was duly noted (notably by Lacan, Levi-Strauss, and Derrida).
In the case of the Course in General Linguistics, the death of the author
emerges therefore as a problem of a distinctive kind. In this case, the
author was (empirically) dead before the book was written—but his name
exercised the ideological function of legitimizing the book as the work

nonetheless. Saussure’s name figures as the author on the title pages but he
did not even perform the functions usually assigned to an editor, such as
revision, verification, and correction of a completed work.6 The contents
and organization of the Course can perhaps be “attributed to” Saussure, ex
post mortem, but there is no established scholarly practice of listing an
attributee as either the author or the editor of a book—even though the
book had been issued and reissued, in multiple editions and many lan-
guages, under his name for decades. The author function which enabled
the identification of the Course as Saussure’s work is therefore determined
chiefly by ideological motives: its foundational role in establishing the sci-
ence of general linguistics, and the structuralist tradition in the human
sciences and philosophy, with Saussure’s name being assigned the role of
their official founder.
The ideological function of the Course as Saussure’s work can be tied
to the normative expectation that a recognized school of thought, such as
structuralism, be founded on a book, a Great Book, and that the book’s
author, a Past Master, be claimed as the school’s adoptive founder. This
scholastic element of the structuralist identity (the term is derived from
Latin schola or school, and is used here broadly to refer to normative ex-
pectations tied to establishing a school of thought, and not only in the
narrow sense related to European philosophy in the Middle Ages and its
elitism), and its concurrent claim to the Course as source of institutional
legitimacy, has largely overdetermined the book’s standing in academic
circles and beyond, and sidetracked the issues related to empirical author-
ship as well as the ideological exclusion of the Nachlass. The book’s dom-
inant reception largely eclipsed concerns related to the book’s inception.
The Course can therefore be likened to the Book writ large: in both
cases, reception overdetermines questions over empirically identifiable au-
thorship. In fact, the Course is usually termed a vulgate edition, an edition
6
Note that the French edition bears the comment “Published by Charles Bally and Albert
Sechehaye with Albert Riedlinger’s collaboration” on the title page. The Roy Harris ©1986
[1983] English translation renders this as “Edited by Charles Bally and Albert Sechehaye with
the Collaboration of Albert Riedlinger,” but the 2013 Bloomsbury reissue of this translation
drops the reference to Bally, Sechehaye, and Riedlinger from the title page, with Ferdinand de
Saussure figuring as sole author. Similarly, the 2011 reissue of the Wade Baskin translation by
Columbia UP drops the reference to Bally, Sechehaye, and Riedlinger from the title page; it
includes this mention instead: “Edited by Perry Meisel and Haun Saussy”—that is, it lists the
names of the two scholars who appended a new “Introduction: Saussure and His Contexts” to the
Course in 2011. Saussure figures as sole author (and Baskin as the translator, Meisel and Saussy
as the two editors). A contemporary Anglophone reader is therefore unlikely to glean Bally’s and
Sechehaye’s editorial involvement in the making of the Course from the most recent reissues of
the two English translations (see Appendix 1 for more discussion).
introduction 13
for the common people (borrowing the term from Jerome’s popular trans-
lation of the Roman and Hebrew Bible into Latin in the fourth and fifth
centuries [de Mauro, 2005 (1967)]), but this analogy is both illuminating
and potentially misleading. It is potentially misleading because a vulgate
edition suggests that a preexistent body of text has been translated in toto
from one language to another. That is not, however, the case with the
Course; it is not a direct translation of a preexistent collection from an elit-
ist to a vernacular language which would expand its audiences beyond the
learned institutions to wide masses. In the case of the Course, the “vulgar”
edition creates a volume from disparate sources for the first time, and picks
just one year from a three-year course in general linguistics delivered by
Saussure as the basis; it produces an impression of a completed system by
excising the elements of self-reflection and internal critique found in the
source materials; it reverses an inductive process of inquiry into a deduc-
tive order of demonstration, and frames tentative conclusions as a priori
axioms; it introduces a dogmatic preacher-like tone in place of a searching
one, and inserts its own formulations in the process, some lacking any
manuscript support. In sum, the editorial inception of the Course jars with
the notion of a popular translation, even if it were acknowledged that trans-
lations inevitably betray the original (see ch. 7 for detailed account of edi-
torial inception).
The analogy between the Course and the vulgate edition of the Bible is
illuminating, but in a somewhat different manner than the intended sense.
The Course functions as the Book writ large because it is a set of writings
whose validity is contingent largely on the retroactive collective recogni-
tion of it as the word issuing from the master himself, just as the validity
of the Bible is contingent on retroactive recognition by the members of the
Church as word of God. That the Course effectively became the scripture
of structuralism is contingent on the overwhelming force of such a post
factum appropriation of this book as a requisite foundation. The tremen-
dous success and impact of the Course can be partially deciphered by the
scholastic or school-founding potential harbored and resourced from a
Great Book. Within an institutionalized context where worshipful recep-
tion of foundational texts sustains the inception of legitimate schools of
thought, questions of empirically attributable authorship lose their gravi-
tas; the overriding commitment is to securing a bedrock and deriving a
programmatic statement for the school in question, not to the material and
institutional history related to the reclaimed text itself. In sum, the anal-
ogy between the Course and a popular edition of the Bible can be main-
tained within an institutional and historical context of European academia,

where the mechanism of founding a school of thought replicates that of
founding the Church (which is unsurprising considering the traditional
bond between the Church and the universities in the history of European
institutions).
Yet even though the Course cannot be characterized as a popular trans-
lation of a preexistent body of work, its relation to the source materials is
admittedly complex. Critical studies from the first editorial paradigm, no-
tably Engler’s critical edition and Godel’s study of manuscript sources,
offer ample evidence that numerous formulations found in the Course have
direct manuscript evidence, while others represent the editors’ bona fide
attempt to distill the source materials available to them into a set of intel-
ligible and coherent statements. Some of the most striking and memorable
formulations found in the Course—“It is . . . possible to conceive of a sci-
ence which studies the role of signs as part of social life . . . We shall call
it semiology” (1986, p. [33], 15); “Language (la langue) and speech (la
parole) are . . . interdependent; the former is both the instrument and the
product of the latter” (1986, p. [37], 19); “in language there are only differ-
ences without positive terms” (1986, p. [166], 118); “language is a social
institution” (1986, p. [129], 90)—have manuscript evidence.7 In sum, the
contents of the Course exhibit varying degrees of historical authenticity in
relation to the sources, and it goes without saying that the process of trans-
forming the “raw material” of multiple sets of student lecture notes as well
as Saussure’s autographed working notes into the traditional format of an
academic treatise must invariably alter its source materials in the process.
Bearing this stylistic discrepancy between the sources and the final prod-
uct in mind, it would be unreasonable to hope for a classic academic trea-
tise that fully captures the “real” and “authentic” Saussure.
A critical study of the Course is nonetheless warranted because this text
articulates the “official doctrine” associated with Saussure’s name, that is,
a set of oppositional and hierarchical pairings such as the signifier and the
signified, la langue and la parole, synchrony and diachrony, which became
appropriated as such within the structuralist tradition of inquiry. My pro-
posed critical study of the Course in part I bears therefore on the status of
this text as a site of the official Saussurean doctrine, and on the subsequent
7
References to the Course refer to the 1986 Roy Harris translation of Cours de linguistique
générale; unlike the recently republished Wade translation, the Harris translation preserves the
pagination of the French edition. The Harris translation is cited as follows: Course, 1986, page
numbers for the French in square brackets, followed by the page number in the English
translation.
introduction 15
coupling of Saussureanism and structuralism, which largely determined
the reception of Saussure’s general linguistics in twentieth- and twenty-
first-century philosophy and related fields—and led to the ideological ex-
clusion of the Nachlass. My study reexamines the official doctrine in light
of the materials from the Nachlass in order to expose a greater than here-
tofore acknowledged philosophical complexity within Saussure’s project.
It also prepares the ground for an alternative appropriation of Saussure’s
general linguistics from the source materials as a phenomenology of lan-
guage (part II). I will return to a critical study of the Course in part III,
where I enrich the conceptual critique with a detailed historical account of
the material and institutional processes that enabled, first, the editorial in-
ception, and then, the dominant structuralist reception of this book as the
official doctrine.
The outline of contents
Part I, ‘“Saussurean doctrine” and its discontents’, draws on materials

from the Nachlass to reexamine the host of dichotomies associated with
Saussure’s general linguistics as presented in the Course and later adopted
within structuralism: the signifier and the signified, la langue and la
parole, synchrony and diachrony. This part is composed of three chapters.
In ch. 1, ‘The signifier and the signified’, I note that the classic signifier and
the signified distinction helped to fuel the structuralist view that language
is a closed and autonomous system of signs. In the Course, the signifier is
defined as being arbitrary or unmotivated by reality; this definition sup-
ports the structuralist idea that the signifier gathers signification solely by
virtue of contrastive relations of difference to the other signs within the
structured system. I argue that this understanding of signification may
result in part from the editorial rendering of the source materials, that is, a
possible confusion between Saussure’s critique of the received view of lan-
guage as nomenclature (from nomen), with its focus on individual naming
relations between the signifier and the signified, and Saussure’s own con-
ception of language. The editorial organization of materials in the Course,
and the book’s dominant reception, further support this understanding.
In Saussure’s own conception of language, as developed in the student
lecture notes and the manuscript notes, the focus on an individual sign is pro-
visionally adopted and then surpassed in view of a more complex relational
understanding, wherein arbitrariness is not a property of individual signs but
extends onto the entire system of relations and differences between them.

Furthermore, a view of language as a sign system alone is deemed reductive
and abstract; a more concrete understanding defines language as a set of his-
torically sedimented conventions shared by a speaking community. Contrary
to the structuralist definition of language as a closed and autonomous sign
system, language is shaped by reality: the reality of social conventions as
they are transmitted and sedimented, as well as revised over time. Linguistic
arbitrariness is therefore defined positively in terms of historicized conven-
tional patterns of usage, and not purely negatively as lack of motivation. I il-
lustrate the bond between language and the evolving social reality by way of
contemporary research into the possibility of reclaiming and resignifying
harmful speech by historically marginalized groups.
In ch. 2, ‘Phonocentrism: Derrida’, I address Derrida’s influential read-
ings of the Course in Of Grammatology and Glas. I complicate Derrida’s
charge of phonocentrism, that is, the charge that Saussure privileges the
medium of sound and/or speech as a site of unmediated signifying pres-
ence, by reexamining the relevant sections from the Course in light of the
materials from the Nachlass. I document especially the extent of editorial
involvement in the sections discussed by Derrida that deal with the pur-
portedly “natural” expressions like onomatopoeias and interjections, and
with the relation between speech and writing. I make the case that ulti-
mately Derrida’s charge that Saussure’s linguistics is burdened by an alle-
giance to the metaphysics of presence carries little force. Their projects
turn out to be more closely related than previously thought: they share the
view that the sign is always already entrained or relativized by the lan-
guage system, and that the language system is contaminated by forces
deemed “external” to it.
In ch. 3, ‘La langue and la parole, synchrony and diachrony’, I examine
the remaining elements of the “Saussurean doctrine”: the pairings, la
langue and la parole, synchrony and diachrony, which are construed as
oppositional and violent hierarchies within the dominant structuralist re-
ception of the Course. I make the case that while the editorial presentation
of these pairings is in agreement with the source material on a number of
points, it tends to overstate the separation between the distinguished terms
and to underemphasize the importance of subjective viewpoint for estab-
lishing the distinction. I argue that Saussure’s methodological reflections
and the stated difficulty of finding the right orientation into the heteroge-
neous field of language, which echo Hegel’s stated difficulty of beginning
in philosophy, get occluded in the process. I develop the idea from Saus-
sure’s Nachlass that language has a double essence in that it exists both in
the present and in the past, and that the linguistic method is best figured as
introduction 17
a dual way to language wherein its two related facets: la langue and la
parole, synchrony and diachrony, can emerge.
Part II, ‘General linguistics: science and/or philosophy of language’,
looks to Saussure’s Nachlass for a philosophical reflection on language
study, with an emphasis on its phenomenological orientation. This part is
composed of three chapters (chs. 4–6). In ch. 4, ‘Involuntary assumption
of substance, and points of view in linguistics’, I raise the question of the
relation between scientific and philosophical approaches to language by
way of a distinction between scientific knowledge of language and reflec-
tion on linguistic experience. The former is committed to the standard of
scientific objectivity, while the latter preserves a subjective entanglement
within language as part of daily experience and usage. Even though it is
usual to interpret Saussure’s general linguistics as the foundation of a
modern linguistic science, with an emphasis on scientific objectivity, I pro-
pose that the interpretation is complicated by Saussure’s own reflections on
scientific methodology. I turn to materials from the Nachlass to develop
Saussure’s extensive critique of the dominant scientific methodology and
its metaphysical heritage. I represent Saussure’s critique of the notion that
language could map onto a set of self-standing things or substances, and
develop his emphasis on method and a comparative critique of points of
view in linguistics.
In ch. 5, ‘Saussure’s general linguistics as linguistic phenomenology’, I
reinterpret Saussure’s general linguistics as linguistic phenomenology. In
contrast to the institutionalized antagonism between phenomenological
and structure-based approaches within which the Course was received in
postwar France, Saussure’s own reflections on language grew on a terrain
saturated with phenomenological references and largely conciliatory. Re-
garded within its own formative context, Saussure’s linguistics was influ-
enced by the contributions to theoretical linguistics of Polish linguists Bau-
douin de Courtenay and Mikolaj Kruszewski, whose idea was to develop
“something like a phenomenology of language.” This phenomenological
impetus is manifest in the source materials of Saussure’s linguistics, popu-
lated as they are by references to the phenomenon, the speaking subject,
language acts, and consciousness. I show that this phenomenological im-
petus was partially lost within the edited version of the Course considering
the effacement of references to consciousness. I then analyze materials
from Saussure’s Nachlass to make the case for a consciousness structured
like a language, and use linguistic innovation by analogy with established
patterns as an exemplar of linguistic consciousness at work. The focus on
linguistic innovation makes it possible to go beyond the official doctrine

associated with Saussure’s general linguistics, considering that the distinc-
tions between la langue and la parole, synchrony and diachrony, are estab-
lished within the ongoing process of analogical innovation, without oppo-
sition or hierarchy.
In ch. 6, I offer additional resources for a linguistic phenomenology. I
first develop the idea that, with its admixture of philosophical reflection
and scientific rigor, Saussure’s general linguistics is best deciphered by
way of Hegel’s phenomenology, especially the notion of science founded
on the primacy of consciousness. I then go on to show that it was relatively
common before the Second World War, on an undivided European conti-
nent, to apply phenomenological resources to research in general linguis-
tics, as exemplified by the collaboration between the Russian linguist
Roman Jakobson and the German phenomenologist Edmund Husserl, and
their shared attention to structured organization of language and experi-
ence. Maurice Merleau-Ponty’s efforts to integrate the material from the
Course with Husserl’s writings in the 1940s and 1950s in order to develop
an emphasis on the speaking subject serve as another resource for linguis-
tic phenomenology.
Part III, ‘The inception and the reception of the “Saussurean doctrine”:
the Course’, enriches the philosophical critique of the official doctrine
associated with Saussure’s general linguistics with an account of the ma-
terial and institutional history that led to the production and reception of
the Course as a statement of the basic dichotomies: la langue and la
parole, synchrony and diachrony. This part is composed of two chapters
(chs. 7 and 8). In ch. 7, I make the case that the book’s editors, Charles
Bally and Albert Sechehaye, performed a double duty in the production
process. First, they projected some of their own methodological and con-
ceptual commitments onto the Nachlass in order to establish general lin-
guistics as a recognizable scientific discipline in their ghostwritten text.
Second, they received and replicated the basic principles of this science—
that is, the official doctrine—in a series of specialized articles in linguis-
tics and extensive book reviews of the Course. I illustrate this process with
concrete examples, such as the editorial insertion (and subsequent cita-
tion) of the concluding sentence into the Course, according to which the
proper object of linguistic study is la langue (language as a structured
system) studied in itself and for its own sake. This famous formula, which
effectively organizes the entire contents of the Course, became a motto
within structuralism, and was perceived as a challenge within phenome-
nology. I decipher its manifest success in being recognized as quintessen-
tial Saussureanism by situating it in the context of elementary structures
introduction 19
of kinship in academia, notably the institutionalized lines of succession
from master to disciple like from father to son. Another example of edito-
rial intervention which helped to cement the doctrine is a reversal of order
of presentation from the student lecture notes; Saussure’s introductory
survey of the diversity of languages (les langues) is pushed back to later
sections of the book, while the category of language as such (la langue) is
introduced early on in guise of an a priori axiom. This reversal, and the
concomitant separation between linguistic diversity and language as such,
can be partially deciphered by the editors’ commitment to abstractions,
especially to the law/fact dichotomy.
In ch. 8, I complicate the definition of structuralism as the predomi-
nantly French post-Second World War intellectual movement with a direct
lineage to Saussurean linguistics from the Course by reconstituting a more
complex story of structuralism in east and west, before and after the war. I
then discuss the structuralist and poststructuralist reception of the Course
by Levi-Strauss, Lacan, and Derrida, as a further development within the
process of establishing the Saussurean doctrine with its familiar set of op-
positional pairings and the ongoing ideological exclusion of materials from
the Nachlass.
In Appendix 1, ‘English translations of the Course’, I comment on how
the two English translations of the Course by Roy Harris and Wade Baskin
deploy various strategies to legitimize the Course in response to the ad-
vancements in Saussure scholarship. In Appendix 2, ‘Saussure’s silence’, I
offer a detailed account of Saussure’s overall reticence to publish in the
academic format (while writing thousands upon thousands of unpublished
manuscript pages) to make the case that behind the official façade of inac-
tivity one finds a secret becoming of a philosopher.

I “Saussurean doctrine”
and its discontents
1 The signifier and the signified
T
he received structuralist view of Saussure’s general linguis-
tics foregrounds the signifier/signified distinction as the single
most important contribution made by the Course. The termino-
logical complex sign, signifier, and signified has been described as “per-
haps Saussure’s most influential gift to Structuralism” (Sturrock, 2003, p.
35); this distinction—“once it has been successfully understood”—would
provide an unrivalled tool to the many disciplines practicing textual analy-
sis (ibid.). The proper understanding goes something like this: by tying the
graphic and/or acoustic sign to the signified idea in a manner that is arbi-
trary or unmotivated by reality, the signifier/signified distinction estab-
lishes the autonomy of language with respect to reality (ibid., p. 36). The
autonomy of language with regard to reality is established by the sign’s
gathering of signification via the mainly contrastive relations of difference
from other signs embedded within the structured system. Arbitrariness of
the sign is therefore typically received as evidence of the structuralist view
that language is a closed and autonomous system of internal relations. In
Harris’s words: “The essential feature of Saussure’s linguistic sign is that,
being intrinsically arbitrary, it can be identified only by contrast with co-
existing signs of the same nature, which together constitute a structured
system. By taking this position, Saussure placed modern linguistics in the
vanguard of twentieth-century structuralism” (1986, p. x).
This emphasis on signification emerging solely within the structured
system leads to a hierarchy of the signifier (signifiant) over the signified
(signifié), since signification seems no more than a result of contrastive rela-
tions between the (signifying) entities like words, gestures, or images, and
not due to their inherent expressive force. It is therefore not the category of
the sign, but the narrower category of the signifier that gains primacy in the
structuralist doctrine: the signifier’s authority is constrained exclusively by
relations to its likes within the system, while the system is not constrained
by anything of a different nature. Any notion of a signified transcendent to
the signifier can therefore be dismissed, with access to reality or the real
perpetually barred by the unsignifiable barrier between the signifier and the
signified. If the tripartite sign/signifier/signified distinction was Saussure’s
gift to structuralism, the recipient apparently preserved the middle part
only, and dispensed of the remainder like one would of giftwrapping.
Arbitrariness of the sign
This selective reception was enabled by the fact that arbitrariness is de-
fined in the Course with a focus on Saussure’s initial definition of the sign,
and that the developments, complications, and ultimate revision of this
initially held notion in the course of lectures on general linguistics were
relegated to subsequent chapters, some of which occur in the later sections
of the book. Specifically, the definition of the arbitrariness of the sign is
discussed in part I, ‘General principles, ch. I. Nature of the linguistic sign’;
the subsequent developments of this thesis in terms of mutability of the
sign in response to social and temporal factors occur in the subsequent
ch. II. ‘Immutability and mutability of the sign’. The developments related
to the sign in terms of linguistic value and the forces of the entire language
system which render arbitrariness relative occur only in part II, ‘Syn-
chronic linguistics’, chs. IV and VI. Even though a reader of the Course
can find evidence of the many complications and revisions that befall the
initial definitions of the sign: its acknowledged bond to the social world
and temporality (“[Language’s] social nature is one of its inner character-
istics”; “Time changes all things; there is no reason why language should
escape this universal law” [Saussure, 1986, p. [112], 77]), and its embed-
dedness within a network of relations to other signs (“in language there are
only differences without positive terms” [Saussure, 1986, p. [166], 118])
she may be unlikely to perceive them as organic elements of a discussion
begun in the better-known sections from part I, ch. I; the reader of the
Course may therefore miss the ways in which social and temporal factors
bear upon the initial thesis related to “the nature of the linguistic sign”—
considering especially that the dominant structuralist reception of the
Course paid little heed to the social and temporal dimensions of the sign
system, difficult as they would be to accommodate within a traditional sci-
entific program of study. I develop a more detailed comparison between
the editorial presentation and manuscript sources related to these issues
below (ch. III, ‘La langue: systemic, social, historical’).

By turning to the source materials, the reader can better appreciate how
the developments, complications, and revisions of the initial thesis of arbi-
trariness shift emphasis from the sign considered in relative isolation to the
constraining effects of the entire language system, as well as to the impact
of social conventions and the historical reality of language. As a result, lack
of motivation by external reality in the sign (a negative thesis) becomes
redefined positively as relative mediation by properties of the entire lan-
guage system, as well as by historically sedimented and socially transmit-
ted practices of language use. Contrary to the received structuralist view,
the sign is tied to reality, but reality ceases to be defined in a naturalist and
substantialist fashion as a “natural kingdom” or a rational order of things.
The arbitrary link between the signifier and the signified does not suspend
the link to reality tout court but to a metaphysical notion that reality is fixed
and preordained. The basic bond between word and world can then be pre-
served, but it gets figured more dynamically, as a sociohistorical process.
One can appreciate this alternative view by exposing the philosophical
complexity of Saussure’s argument from the source materials. The argument
combines linguistic and metaphysical considerations and develops a sustained
critique of the twin conceptions of reality in terms of substance metaphysics
and of the nomenclature view of language (from nomen—name). These twin
conceptions suggest that language forms a labeling system for an aggregate of
substances ordered in a fixed and rationally determined manner; language is
thus conceived as an order of words as immutable as the metaphysical order
of things it stands in for. Saussure’s writings offer a critique of these twin
orders of the word and the world, and suggest that language and worldview
continually evolve and thus escape a firm grasp of logical and natural deter-
minacy. This novel understanding ties language to the social reality of evolv-
ing patterns of usage and convention; as a result, the arbitrariness of the sign
(and the entire sign system) ceases to appear as a purely privative absence of
motivation by forces deemed “external” to language itself, and turns out to be
bound up with positively defined notions of social practice and its transgen-
erational transmission (as well as revision). In sum, language is not an autono-
mous and closed system where like only mixes with like; it is consistently tied
to, constrained, and sustained, by the world—the social world.
My exposition runs as follows. I examine the textual validity of the sections
dealing with the sign, signifier and signified distinction and arbitrariness of
the sign in the Course (part I, ch. I, sections 1 and 2) in light of the source
materials. This will help to showcase that the editors presented the sign along
the lines of the received view of language as a labeling system or nomencla-
ture, and thus occluded the novelty of Saussure’s own views. The editors also
the signifier and the signified 25

presented arbitrariness in a manner consistent with the notion of natural sym-
bolism of sound in language—the notion later critiqued, justifiably, by Der-
rida. Combined, the editorial presentations overshadow the philosophical
complexity of Saussure’s view; I will discuss how in the source materials the
claims about arbitrariness of the sign (and the sign system) serve the broader,
philosophical stakes of Saussure’s argument, and conclude by discussing the
intrinsic motivation of signs by social and historical forces.
Graphic illustration: sign/signifier/signified
Ch. I on “the nature of linguistic sign” (part I, General principles) consti-

tutes arguably the most influential section in the whole Course for philo-
sophical posterity. The chapter introduces the distinction between the sign
(signe), signifier (signifiant), and the signified (signifié),1 and then discusses
two related principles: the arbitrariness and linearity of the sign. The edi-
tors open the section with a mention of the received view of language as a
nomenclature, to which the view of language as a system of arbitrary signs
will be an alternative. Figure 1 illustrates the received view.
The figure is adopted from the student notes (Engler, 1989, [1087], D 186,
III C 278, p. 147; Saussure, 1993, p. 74) where it also figures as an illustra-
tion of the nomenclature view of language.2 The relevant passage reads:
It has often been mistakenly assumed that that in language (la langue) there
is only nomenclature (tree, fire horse, snake) . . . That’s a childish approach.
If we follow it for a moment, we shall easily see what the linguistic sign
consists in and what it does not consist in. We face a series of objects and a
series of names. (Saussure, 1993, pp. 74–75, emphasis added)
The figure illustrates the problematic thing-cum-name view which the

student is invited to provisionally adopt as part of a historical review and a
didactic exercise, before the alternative can be laid out.
The alternative is presented in the Course as the linguistic sign, a two-
sided psychological entity, composed of a “concept” and “acoustic image”
1
This terminology is not preserved in the Harris translation, which opts for “signal” and
“signification” respectively (Harris, 1986, p. 65); the Wade translation (2011) is preferable in this
regard. Note: signifiant is both a present participle and a substantive noun, i.e., it indicates both
the process of signifying and the entity in charge of it. It should not therefore be taken simply as a
thing but also as a verbal doing.
2
References to Engler’s Critical Edition of the Course have the following format: Engler, 1989,
[index number], top column designation of the source student lecture notes (where applicable),
text page number.

(image acoustique) (“sound pattern” in
• ARBOR Harris). While these terms figure in the man-
•
uscript sources, they are potentially ambigu-
ous and eventually abandoned. The notion
acoustic image could be taken to mean either
• EQUOS the material sound (studied in phonology) or
•
the mental representation of sound (belong-
ing to the province of psychology) (ibid., p.
etc. etc.
75)—the same methodological and concep-
figure 1 tual ambiguity that surrounds, in Saussure’s
(source: Saussure, 1986, p. [97], 65)
stated view, the notion of a “name” in no-
menclature. Furthermore, and more to the
point, a sound-based conception of the sign is suggestive of a natural origin
of language, where words could derive directly from for example natural
cries and where a unit of sound could be a self-standing carrier of significa-
tion independently and prior to the emergence of the language system. Lan-
guage would therefore be built up brick-by-brick from the already signify-
ing sound units, in agreement with the nomenclature view. Saussure
replaced the sound-based with the dedicated semiological vocabulary of
the signifying and signified in the lecture of May 19, 1911 (ibid., pp. 92–93).
This terminological change is not reflected in the Course where the earlier
and the later terminology is used interchangeably; this may obscure the
specificity of the language system and the science of semiology defined in
the chapter.
Having distinguished the two facets of the sign (concept/acoustic image),
the editors pursue:
These two elements are intimately linked and each triggers the other.
Whether we are seeking the meaning of the Latin word arbor or the word
by which Latin designates the concept “tree,” it is clear that only these con-
nections which are consecrated by language (la langue) appear to us to
correspond to reality. (Saussure, 1986, p. [99], 66).
Figure 2 is a set of two figures:
‘Tree’
Arbor Arbor
figure 2
(source: Saussure, 1986, p. [99], 67)

The right-hand figure is especially famed and oft reproduced. It consti-
tutes, however, another instance of editorial insertion without manuscript
support (Engler, 1989, [1110], p. 150), which has acquired the status of of-
ficial doctrine in philosophical circles. Specifically, the figure on the right
featuring a picture of the tree as well as the arrows in this (and in the pre-
ceding) figure are a free insert. The arrows seem to graphically illustrate
the point made in the above cited passage: “These two elements [the con-
cept and the acoustic image] are intimately linked and each triggers the
other,” but this line is as much of an editorial insert as its graphic illustra-
tion (ibid., [1108], p. 150). Furthermore, the arbitrary link seems to be set
up exclusively within the sign, as if by its own intimate linking and trigger-
ing force. The source materials emphasize however that the two facets,
which are of psychic nature, are both in the subject (dans le sujet) (ibid.,
[1094], D 186, p. 148); any other approach to the sign will be rejected as
misleading (fausse piste) (ibid., [1109], D186, J156, p. 150). The relation
between the signifying and the signified facets is therefore necessarily me-
diated by the speaking subject to whom the sign must signify something if
it is to signify at all; to render it as a purely objective process regulated by
language itself is to be on the false route. The intimate bond and necessary
relation between the signifying and the signified facets is an objective
aspect of this subjective association, and the two are both primary. The
sign does not signify in and of itself; the signification is contingent on the
speaker’s recognition and deployment of the sign in a shared context of
meaning making. In sum, the sign cannot be neatly isolated as an objective
datum for scientific study and the semiological relation cannot be con-
strued in purely objective terms.
Figure 2 is an editorial attempt to illustrate Saussure’s presumed intent,
but it confuses the reader at best, or makes it sound as though Saussure was
a nomenclaturist sans le savoir at worst. The figure establishes a presumed
parallel between, on the one hand, the Latin word arbor and the French
concept “arbre” (or “tree” in the English translation), and, on the other,
the Latin word arbor and the image of a tree. The figure to the left has a
manuscript source in Degailler’s lecture notes (see Godel, 1957, p. 82;
Engler, 1989, [1107], D 186, p. 150), but it is not coupled with the right-
hand one; by adding it the editors paste the graph previously used to illus-
trate the critiqued nomenclature view of language into the illustration of
Saussure’s presumed alternative view. As the manuscript sources demon-
strate, the figure featuring a series of pictorially represented objects, for
example a schematic symbol of a tree or horse on the one hand, and a
series of names in Latin like arbor and equos on the other, illustrates the

dominant view being undermined, and should not be mixed in with the
proposed view that language is a system consisting of two indissociable
signifying and signified facets. Such arbitrary (in the common sense of the
word) alterations to the manuscript material mar the entire section on the
sign (part I, ch. I, section 1) with inconsistencies. They suggest that the
concept “arbre” could be ultimately derived from a preexisting thing or
idea, and captured in a pictorial representation with some claim to univer-
sality (as long as one is versed in the relevant system of schematic symbol-
ization). One could then regard the little tree icon as an imprint of the
natural kind itself; the natural kind would receive a tripartite representa-
tion in a picture, an idea, and a word. The signifier would be the final ter-
minus in this classical representational chain, three times removed from
the thing itself. The notion that there is a fixed and immutable order of
things is therefore preserved, with the words from different languages like
“arbre,” “tree,” “Baum,” and so on, serving as the many “arbitrary” labels
for a stable core referent. We thus return to the classic view of natural
kinds mirrored in appropriate part of speech, the whole order being ration-
ally knowable and universally valid.
The section on the sign paves a path straight and narrow for the follow-
ing section on the arbitrariness of the sign (part I, ch. I, section 2). The
principle of arbitrariness is defined in a traditional or conventionalist
manner as a link between an idea and the sign with no internal or intrinsic
relation between the two (Saussure, 1986, p. [100], 67); the signifier is un-
motivated by the signified, with which it has no natural connection in real-
ity (ibid., p. [101], 68). For example, “there is no internal connection be-
tween the idea ‘soeur’ [“sister” in the English] and the French sequence of
sounds s – ö – r of which it is the signifier (signifiant); it could have been
represented by any other; differences between languages and the very ex-
istence of different languages serves as a proof: the signified ‘boeuf’ [“ox”
in the English] has b – ö – f as a signifier on one side of the border, and
o – k – s (Ochs) on the other” (my translation).
Arbitrariness emerges here as a purely privative concept, an absence of
natural determination of the signifier by the signified—as the above com-
parison between different languages demonstrates. The reader of these
definitions would be hard pressed to glean the novelty of Saussure’s ap-
proach, since the basic idea dates back to Plato and Aristotle as well as the
Stoics and the Sophists, even though the immediate source of the conven-
tionalist conception for Saussure would be the linguist Whitney (de Mauro,
2005, p. 442), who stated that “inner and essential connection between
ideas and word . . . there is none, in any language upon earth” (Whitney,

1887, p. 32; de Mauro, 2005, p. 442). This conventionalist notion of arbi-
trariness is consistent with and reinforces the nomenclature view, since
human convention would associate things/ideas and words originally dis-
tinct. It is still assumed on this model of arbitrariness that a preexistent,
fully fledged signified—a thing and/or an idea of the thing—subsequently
receives a linguistic label. However, as Jakobson rightly noted, the theory
of the arbitrariness of the sign presented in the Course:
is in blatant contradiction with the most valuable and the most fertile ideas
of Saussurean linguistics. This theory would have us believe that different
languages use a variety of signifiers to correspond to one common and un-
varying signified, but it was Saussure himself who . . . correctly defended
the view that the meanings of words themselves vary from one language to
another . . . Saussure himself cites “the difference in value” between the
French mouton and the English sheep . . . There is no meaning in and by
itself . . . In language there is neither signified without signifier nor signifier
without signified. (Jakobson, 1978, p. 111)
This alternative theory is borne out by the source materials, where

arbitrariness is extended to the idea or the signified; the idea or signified
is as arbitrary as the signifier, for its signification or meaning emerges
through the relation to other signifieds (and signifiers) in the system. For
example, the significations of the French mouton and the English sheep
carry different values, for the English system distinguishes between
sheep (animal in the field) and mutton (meat on the table), whereas the
French does not. The significations of sheep and mutton are therefore
constrained by the system-dependent differences between cognate terms
(sheep is alive, mutton is not), and cannot be framed as direct labeling
relations between a signifier and a signified. As a result, each sign gets
crossed by the relations to other signs in the language system—so much
so that one cannot isolate an individual sign and regard its signification
on its own. Arbitrariness cannot therefore be confined to the properties
of an isolated sign; this latter notion is a vestige of the traditional nomen-
clature view of language which Saussure submits to an extensive critique
in the manuscript writings and lectures on general linguistics. These
source materials put forward a different view, which ties linguistic arbi-
trariness, first, to the entire language system, and second, to the sociohis-
torical community of language users. Contrary to the received view that
the arbitrary linguistic sign establishes the autonomy of language with
respect to reality (Sturrock, 2003, p. 36), it is rather the case that the sign

is motivated by forces reigning within the language system (an intrasys-
temic motivation that a structuralist would accept), and is further moti-
vated by social and temporal forces usually deemed “external” to the
system of signs (an extrasystemic motivation that a structuralist would
not accept since it undermines the possibility of establishing a closed and
autonomous system in the service of scientific objectivity). As a result,
linguistic arbitrariness becomes revised as a species of motivation within
language by reality construed as a sociohistorical practice; this latter
notion is mutable and open to change, and thus distinct from the tradi-
tional view of reality as a preordained natural order, a divine decree, or a
metaphysics of substance.
Critique of the nomenclature view of language
The reader of the Course is unlikely to appreciate the philosophical com-

plexity of Saussure’s arguments, such as the critique of the dominant no-
menclature view of language. In this section, I will bring relevant source
materials and additional contributions from philosophy to bear on the re-
ceived material from the Course to reveal this complexity. The signifier/
signified distinction and the principle of arbitrariness become more intel-
ligible in light of the dominant philosophical view of language, which
forms the necessary backdrop to the more technical argument. This con-
text is necessary to decipher the specific claims made about the sign and
signification, and to bring out the larger stakes of the argument which
otherwise may appear too narrowly linguistic.
The first point of note is the title of the chapter in question: “the nature
of the linguistic sign.” The title is a faithful rendering of the one from Sau-
ssure’s lecture from May 2, 1911 (Saussure, 1993, p. 74). However, in the
appendix to this lecture, dating back to May 19 (Godel, 1957, p. 85), this
title was revised to “Language as a system of signs” (“La Langue comme
système de signes”). The editors ignored the title change, noting on the
margins of the Collation that “it does not seem at first sight that this title
be preferable,” citing as evidence that the first title seemed “rather natural”
according to Degailler’s notes (Engler, 1989, [1084], D 210, p. 147). How-
ever, one finds no such claim in the student’s notes (ibid.). Furthermore, the
published notes by E. Constantin confirm the title change (Saussure, 1993,
p. 92). The editors disregarded source materials in favor of perceived natu-
ralness and personal preference, even though the title change is significant,
as explained next.

The revised title was to indicate direct transition from ch. I on “Lan-
guage” (“La langue”) to ch. II on “Language as a system of signs” in
Saussure’s lectures; the latter would follow from the first in an identifiable
fashion, and maintain focus on the systemic organization of language as
a whole (and not on the sign regarded in isolation). By disregarding this
systemic focus the editors project a more traditional approach which
takes individual words or names like “tree” and “horse” to be the basic
constituents of language; in this way, they steer Saussure’s lectures in the
direction of the received (and critiqued) view of language as nomencla-
ture. According to the latter view, names serve as labels (étiquettes) for
preexisting and self-standing entities in the world, modeled on the meta-
physical notion of substance and grasped mentally by means of ideas.
The editorial emphasis on a one-to-one relation between the signifier and
the signified fits into this established understanding of language, and sug-
gests that Saussure implicitly adopted the view of language as composed
of signans and signatum—a traditional metaphysical notion held by the
Stoa and medieval logic, and later critiqued by Derrida due to a commit-
ment to an ideal possibility of attaining a full and unmediated signifying
presence. The nomenclature view of language, where names stand in for
things, corresponds to this basic model of signans—signatum. If Sau-
ssure’s own argument were to preserve the primacy of such notions, it
would be hard to appreciate the novelty of his argument—even if he
made a case for an arbitrary or unmotivated link between the signans and
the signatum.
In the Course, the section on the sign, signifier, signified opens: “For
some people a language (la langue), in its essential principle, is a nomen-
clature, that is, a list of terms corresponding to things” (Saussure, 1986,
p. [97], 65). The anonymous believers in the nomenclature view are made
more specific in the source materials: they are philologists as well as
philosophers (Engler, 1989, [1085], J 156, p. 147). The nomenclature view
is also referred to as “the traditional approach to language from a philo-
sophical angle” and as a default model of language which is either as-
sumed or explicitly put forward by the philosophers (Saussure, 2006,
p. [230], 162).3 Debating the nomenclature view therefore takes place on
philosophical grounds, with implied reference to the terms, theories and
3
Numbers in square brackets refer to the French pagination in the 2002 edition of Ecrits de
linguistique generale; the following number refers to the English pagination in Writings in
General Linguistics, 2006.

assumptions of broadly philosophical nature made throughout the history
of European thought from Aristotle via Augustine up to Wittgenstein.
Even though this view represents also the popular notion of language, it
extends beyond the casual reference made in the Course. Furthermore, as
a philosophical argument, it bears not only on the question of language
narrowly construed but also on the relation between language, thought,
and the world, and on the relation between language and time, as dis-
cussed below.
The editors of the Course render the critique of the nomenclature view
in a three-point list of objections:
1. it supposes that readymade ideas exist before words;

2. it does not specify whether the word is to be taken in a mental or
physical sense;
3. it makes believe that the relation between the word and the thing is
simple, which is far from being the case (elle laisse supposer que le
lien qui unit un nom à une chose est une opération toute simple, ce
qui est bien loin d’être vrai) (Saussure, 1986, pp. [97], 65–66).
This list is not devoid of manuscript support but is so brief that it

condenses the philosophical critique to a fault; the full manuscript
source found in Saussure’s notes for the semiology chapter in the planned
book on general linguistics is far more developed (see Saussure, 2006, p.
[230], 162; Engler, 1989, [1086–1091], pp.147–148; [1950–1956], p. 273).
Saussure opens by comparing the dominant philosophical model of lan-
guage to the scene from the Book of Genesis (2: 19–20), wherein Adam
gives each of the various assembled animals its name.4 The philosophi-
cal model of language is here cast in terms of a mythical narrative about
the origin of language—an approach which fails to describe language
on its own terms, and looks instead for underlying, natural or supernatu-
ral causes. The biblical scene features a single individual, Adam, who
baptizes each and every beast of the field, as if giving them proper
names. The animals themselves do not speak; they only receive the ap-
posite labels which enable man to organize God’s creations. Adam him-
self is alone; he knows neither intimate companionship nor human
4
“And out of the ground the Lord God formed every beast of the field, and every fowl of the air;
and brought them unto Adam to see what he would call them; and whatever Adam called every
living creature, that was the name thereof.”

society, and yet he can lay the foundations of language by dispensing the
first names.
This mythical narrative supports the nomenclature view of language.
This view privileges linguistic signs which function like names, that is,
which happen to correspond to an object given to the senses as a horse, the
fire, the sun. Yet language could be defined as a species of nomenclature
only if its sole function was to name physical objects and actions with ap-
posite nouns and verbs; this supposes also that the nameable reality was
already segmented into such basic categories, each one ready to be pinned
with a label.
As Saussure put it, “there is no clear reason, indeed quite the opposite,
for taking it as the paradigm (type) of language. Doubtless, for those who
wish consider it as such, this is, in a certain sense merely a case of choos-
ing a bad example. But implicit in this is a certain tendency which we can
neither misrecognize nor allow to pass relative to what language ultimately
is, that is, a nomenclature of objects. Of objects which are given before-
hand. First the object, then the sign, hence (and this we will always deny)
an external basis for the sign, and a depiction of language according to the
following relation:
*—— a
Objects *—— b Names
*—— c
figure 3
(source: Saussure, 2006,
p. [230], 162).
The nomenclature view is problematic not only because it makes biased

claims about language by reducing it to an inventory of names, but also
because it makes metaphysical claims about the world as a series of “ob-
jects given beforehand,” like a flock of animals which had been created by
God and passively received human-made names après coup. The identity
and the relations between objects would then be established prior to the
naming event, with signs serving as external envelopes for complete and
self-standing beings.
One finds probably the best example of the latter view in the rationalist
grammar of Port-Royal from 1660. The Port-Royal grammar maps the
rational structure of language onto the metaphysical ordering of the world
onto basic categories of object as available to rational thought.

As Arnauld and Lancelot put it:
The objects of our thoughts are either things, like the earth, the sun, water,
wood, what is ordinarily called substance, or else are the manner or modi-
fication of things, like being round, being red, being hard, being learned,
what is called accident . . . It is this which has engendered the principal
difference among the words which signify the objects of thought. For those
words which signify substances have been called substantives, and those
which signify accidents . . . have been called adjectives. (Arnauld and
Lancelot, 1975, p. 69, translation revised)
This passage captures the mirroring relation between the metaphysical

order of the world, with its “simple parts”: substance, accident, and proc-
ess, and the order of the word, with its simple “parts of speech”: nouns,
adjectives, and verbs—this time without implying a mythical origin but
resorting directly to the human reason’s ability to capture an immutable
order of things. This immutable order of things is reflected in a presumably
immutable order of ideas and words, whose role is simply to stand in for
the things themselves. The intelligibility of the view of language as no-
menclature is thus grounded in substance metaphysics, with its ordering of
the world in a series of basic and unchanging categories available to grasp
by the rational mind.
One finds more expressly philosophical expressions of such a mirroring
relation between metaphysics, thought, and language throughout the west-
ern philosophical tradition. Already Aristotle wrote:
Spoken sounds are symbols of affections in the soul, and written marks
symbols of spoken sounds. And just as written marks are not the same for
all men, neither are spoken sounds. But what these are in the first place
signs of—affections of the soul— are the same for all; and what these af-
fections are likenesses of— actual things— are also the same.” (1995, 16 a
1, p. 25)
Aristotle thus establishes a series of equivalences between an immuta-

ble order of universal ontology (actual things which are the same for every-
one) and a universal grammar of thought (affections which are the same
for everyone), which is then represented in the acoustic (spoken sounds)
and graphic symbols (written marks). The latter symbols differ widely
within the spectrum of the spoken and written languages, but that only

means that the symbols are unmotivated or “arbitrary” signs for a
shared stock of ideas and things whose order is unchanging. Impor-
tantly then, the notion of arbitrariness construed as a conventional and
varied way of labeling the universal code of ideas and things can be
located already in Aristotle, and does not constitute an original contri-
bution made much later by Saussure; Saussure’s own view will chal-
lenge the very notion of a preexistent order of ideas and things with his
conception of arbitrariness.
Augustine’s view of language belongs to the nomenclature tradition,
albeit he introduces a developmental perspective to explain how the
naming ability emerges:
I noticed that people would name some object and then turn towards what-
ever it was that they had named. I watched them and understood that the
sound they made when they wanted to indicate that particular thing was the
name which they gave to it, and their actions clearly showed what they
meant, for there is a kind of universal language, consisting of expressions of
the face and eyes, gestures and tones of voice . . . So by hearing words ar-
ranged in various phrases and constantly repeated, I gradually pieced to-
gether what they stood for, and when my tongue had mastered the pronun-
ciation, I began to express my wishes by means of them. (Augustine, 1961,
I. 8, p. 29)
Augustine’s ontogenetic account of language thus establishes a link be-

tween pointing or ostensive indication by bodily movement and hearing
the sound made to indicate an object, which serves as a practice ground for
learning to utter the sounds oneself. Augustine draws on the familiar sce-
nario of objects being named by words; Wittgenstein notes therefore:
“[Augustine’s] words . . . give us a particular essence of human language.
It is this: individual words in language name objects—sentences are com-
binations of such names. In this picture of language we find the roots of the
following idea: every word has a meaning (Bedeutung) . . . It is the object
(Gegenstand) for which the word stands” (Wittgenstein, 1953, p. 1). An
imaginary experiment featuring a language reduced to an inventory of
lexical terms for indicated physical referent is also found in the opening of
Wittgenstein’s Philosophical Investigations. It features a builder A and an
assistant B, with A using building stones: blocks, pillars, slabs, and beams,
and B passing them in the order needed by A. A and B would use a lan-
guage for this purpose, consisting of the words “block,” “pillar,” “slab,”
and “beam.” When A calls them out, B passes the stone he learned to bring

in response to the call. Wittgenstein concludes with a plea to the reader:
“Conceive this as complete primitive language” (ibid., p. 3). This primitive
scene is an ad absurdum attempt to enact the dominant philosophical
vision of language and expose the paucity of such an abstraction, in con-
trast to the richness and complexity of ordinary language in use.
Saussure’s approach appears similar to Wittgenstein’s critique of the
received philosophical models of language; both are enacted as part of a
rhetorical strategy of exposing their folly. Saussure’s own strategy assumes
apocalyptic proportions in a vision of the end of language and being.
Having dismissed the notion of a preexisting object, grasped by means of
an idea and labeled by the word (“It is surely unfortunate that one intro-
duces the objects designated by names (objets désignés) into it, as if they
were a primordial element, while they form no part of [language]” [Saus-
sure, 2006, p. [230], 162]). Saussure adds: “If an object could at any point
be a term which fixes the sign, linguistics would instantaneously cease
being what it is, from the top to the bottom; so too would the human mind”
(ibid., translation revised).
If ever realized, a direct encounter with the object designated by a name
would lead to a dramatic annihilation of the mind and terminate all lin-
guistic pursuits. Direct knowledge of an object is therefore neither a neces-
sary nor sufficient basis for language; instead, it sets up an unsurpassable
limit to what can be meaningfully said.
That is why, if entertained, the hypothesis of a world before the word
leads to a vision of total destruction of both: “The signified alone is noth-
ing, it dissolves into a shapeless mass. Likewise for the signifier” (Saus-
sure, 1993, p. 139). This is a per absurdum hypothesis of thought dissolv-
ing into amorphous mass:
The purely conceptual mass of ideas, the mass separated from the lan-
guage, is like a kind of shapeless nebula, in which it is impossible to distin-
guish anything initially. The same then goes for the language: the different
ideas represent nothing pre-existing. There are no a) ideas already estab-
lished and quite distinct from one another, and b) signs for these ideas.
(ibid., p. 138)
This passage does not narrate a myth of the origin of language: it does
not tell a story of how language came into being from out of a nebulous
mass, but dismisses any notion of a prelinguistic foundation as meaning-
less. Our starting point is a series of articulated ideas and signs; this series
exists because a whole chain of differences within the semantic as well as

the acoustic and graphic reality is always already set up. A fantasy of “a
mass separated from language” is entertained as an apocalyptic vision of
total destruction of the order of the word and the world, which illustrates
by force of a per absurdum hypothesis their necessary codependency.
Saussure’s argument follows here the guiding principles of Hegel’s
logic, according to which being without determination is an abstraction of
the understanding which ultimately reverses into nothingness (Hegel,
1991, p. 139). Out of this initial abstracted understanding, the speculative
moment emerges when unity is made contingent on opposition between
determinate forms (“The speculatively or positively rational apprehends
the unity of the determinations in their opposition, the affirmative that is
contained in their dissolution and in their transition” [ibid., p. 131]). Ac-
cording to Hegel’s logic, individual identity is ambiguously tied to differ-
ence, and determination to negation. These ideas transpire in Saussure’s
linguistics, and apply to the paired orders of the world and the word. At
both levels, individual identity is set up in in virtue of contrastive relations
with others; each idea has “contours” (Saussure, 1993, p. 140) which are
not secreted from within by any given unit of sound and/or sense, but op-
erate rather as relational borders between them. The system of identity and
difference, determination in negation, orients the relations within language
and being, the signifier and the signified, alike.
*
In his critique of the nomenclature view, Saussure notes also that many
linguistic signs do not fit into the model of name-for-identifiable referent:
“It is by accident that a linguistic sign happens to correspond to an object
captured by the senses (le sens) like horse, fire, sun, rather than an idea, as
in Greek ‘he put’” (Saussure, 2006, p. [230], 162). The nomenclature view
of language is thus contingent on an unacknowledged preference for signs
which can function like names bestowed upon animals and heavenly
bodies, and would be complicated by a choice of less obviously name-like
linguistic examples. And even in cases of seemingly simple words stand-
ing in for simple things like the sun, there is no universal underlying object
labeled by the corresponding expressions in different languages. As dis-
cussed in the lectures on general linguistics: “It is not possible [even] to
determine what the value of the word sun is in itself without considering
all the neighboring words which will restrict its sense. <There are lan-
guages in which I can say: Sit in the sun. In others, the word sun (= star)
does not have the same meaning (signification). The sense of a term de-
pends on presence or absence of a neighboring term.>” (Saussure, 1993,

p. 137). It is therefore impossible to export an interchangeable label out of
the many existing languages; the relevant term is embedded within its sig-
nifying milieu and shaped by relations to its neighbors. In the specific
terms used by Saussure, the signification of any individual term is bound
up with its value relations to other terms. Any translation effort must seek
to do justice to this relational web, and not only to the individual terms.
To take another example, “there is no such thing as the notion of ‘cher’
(dear) in itself before the language” (ibid., p. 139). If there was, the idea
would translate in toto from one language to another, but it does not. There
is no exact correspondence between the French cher and the correspond-
ing German term since the German distinguishes between lieb or “dear”
and teuer or “expensive,” whereas the French term does not. It follows that
the two sets of meanings only partially coincide with the other, and each
term is shaped in opposition to other terms within a language (ibid., p. 139).
Once we pay attention to linguistic diversity, we must shed the pretense of
a total and seamless translation from one language to another. The nomen-
clature view of language endorses an unacknowledged ideology of total
translation and suppresses linguistic diversity in favor of an assumed uni-
versal code of ideas. The signifier/signified schema is tributary of this ide-
ology in that it preserves the notion of a preexisting universal idea repre-
sented by an apposite unit of sound. According to the lecture notes,
however, the concept cum acoustic image is not “a primary schema” (ibid.,
p. 141), and signification is not a self-standing unit in linguistics; it is cou-
pled with the relations to the other cognate terms within a given language
(e.g., the contrastive relation between lieb and teuer in German, the single
term cher in French) and cannot be distilled on its own. In sum, Saussure’s
general linguistics is attentive to the unsurpassable differences between
languages and does not posit an underlying unitary structure beyond and
above the many known languages; the pictorial and ethnographic dimen-
sion of language is its irreducible trait.
Saussure extends the scope of the linguistic sign beyond the tradition-
ally privileged paradigm of a word or a name, and the concurrent focus on
the lexicon at the expense of the grammatical structure. The linguistic sign
is not interchangeable with the word or name, since grammatical catego-
ries also have a signifying function. For example, the notion of plurality is
signified by the category of number, without the latter being interchange-
able with words. While some languages distinguish only between the sin-
gular and the plural, others like Sanskrit and ancient Greek also include
the category of duality, applicable to twin-like or paired cases. The plural
number will have a different value in Sanskrit than in Latin: “Anyone who

assigns the same value to the Sanskrit plural as to the Latin plural is mis-
taken <because I cannot use the Sanskrit plural in all the cases where I use
the Latin plural>” (Saussure, 1993, p. 136). The category of plurality in
Sanskrit disambiguates between objects typically coming in pairs, and
larger aggregates of the many—whereas Latin distinguishes between the
categories of the one and the many only. The value of the Sanskrit plural
is therefore contrasted both with the singular and the dual number—while
the Latin plurality is not.
The difference in value thus serves as a limit to translation—a correc-
tive to the view that a preexistent idea of, for example, “plurality” is indif-
ferently labeled by arbitrary terms in Latin and Sanskrit. It turns out rather
that the idea itself emerges associatively, in relation to other terms indi-
cated in absentia. Plurality does not therefore translate in toto from a lan-
guage which marks the dual to a language that does not (in the former case
the plural contrasts both with the singular and the dual, while in the latter
the contrast applies to an individual only). Languages which do not mark
the dual number are therefore prone to regard the individual and the col-
lective as the two dominant numerical categories, while languages mark-
ing the dual contain also a category of twinhood or primary and indecom-
posable duality which is not a mere aggregate of individual parts (see
Stawarska, 2009; Humboldt, 1828). The nomenclature view of language
could not account for this interwoven or textual dimension of language, as
it posits only a series of seemingly disconnected and self-standing terms,
which enter into combinatorial relations only as a second step. It follows
that: “the real configuration is: a-b-c, outside of any knowledge of a func-
tional relationship such as * - a based on an object” (Saussure, 2006, p.
[230], 162), where * stands for an object knowable in advance. The “real
configuration” is therefore a language system where signs signify in virtue
of a relation to other signs within the language system, not an individual
sign linking up with a referent, whether in a simple or complex manner.
The revised chapter title in the student lecture notes is an expression of this
configuration; it will determine the whole development of Saussure’s
thinking about arbitrariness as a property of the language system as a
whole rather than of individual signs (or signifiers).
As Saussure notes:
The truly ultimate view of language, at least so far as we dare to speak of it,
is that there is never anything that can reside in a single term, and this be-
cause of the fact that linguistic symbols have no relation to what they ought
to designate, thus that a is incapable of designating something without the

help of b, and likewise b without the help of a, or in fact that both of them
are without value except through their reciprocal difference, or that neither
of the two has value, whether it be through any part of itself (for example,
“the root,” etc.), except by means of this same plexus of eternally negative
differences.
It is astonishing. But where in truth would the possibility of the contrary

lie? Where would be for a single instant the point of positive irradiation in
all of language once granted that there is no vocal image that responds
more than any other to what it must say? (Saussure, 1954, p. 63)
The thesis of arbitrariness (“that there is no vocal image that responds

more than any other to what it must say”) is astonishing because it leads to
purging the field of study of any entities existing in themselves (as a posi-
tive science of language like phonology has it), and frames the field of
study via an interplay of negative forces of difference with no solid core of
signification (the root in etymology, the unit of sound in phonology, the
articulatory movement in physiology of sound) left. The thesis of arbitrar-
iness is thus a profoundly disruptive hypothesis, which forecloses the pos-
sibility of establishing a positive science of language and undermines the
validity of substance metaphysics this science is couched in. Saussure’s
reflections on arbitrariness are therefore compatible with a philosophical
project to deconstruct the traditional notion of the sign as a positive entity
and a basic building block of language.
*
Saussure’s critique of the nomenclature view of language considers
also an assumed split or dualism between the idea and the word. The idea-
word dualism is not unlike the better-known ontological dualism separat-
ing the eternal and immutable mind (or spirit) from the perishable body.
In this dualist setup, the idea is cast in a linguistic form like the soul is
enveloped in the body, which alone is subject to change and decay; the
idea can be preserved in toto regardless of what happens to its external
linguistic envelope. This traditional metaphysical worldview is assumed
in the nomenclature view of language wherein “once an object has been
designated by a name, there is a whole (un tout) which will be subse-
quently transmitted, with no other phenomena to take into account! If an
alteration does occur, this is to be expected only on the side of the name,
in keeping with what is assumed, fraxinus becoming frêne” (Saussure,
2006, p. [231], 163).

The paired notion signifier/signified provides an alternative to this dual-
ist setup. This pairing replaces an earlier distinction between the acoustic
image and the concept, which may initially suggest a separation between
matter and mind but according to the lectures notes, both were conceived
as “psychic, and in the subject” (Saussure, 1993, p. 74). The acoustic image
is therefore not the material sound considered as brute physical datum, but
rather a mental imprint of the sound, tied to auditory perception. The new
terms signifier and signified are used to further a semiological conception
of language, emptied of any substantialist associations with physical sound
(and of the notion that a sign is a phonetic inscription replacing a unit of
sound). They foreground the uniquely semiological conception of language
as a system, and are tied to the consciousness of the speaking subject (and
are therefore both psychic in their indelible bond to subjectivity).
In opposition to the metaphysical separation between the idea and the
word, Saussure argues that the signifier and the signified are intercon-
nected and mutually dependent. Hence in some cases the alteration of the
sign changes the idea itself; for example, when the acoustic image of the
fourth and fifth century Latin necare changed to the French noyer, the
corresponding idea changed from “kill” to “drown” (ibid., pp. 98–99).
Similarly, two signs can become merged by phonetic alteration, such that
the ideas (to an extent determined not only by the two signs, but by the
other elements as a whole) will become merged. For example, Old German
has Dritteil for “third,” while Modern German has Drittel. In this case,
the signifying element has changed—but not without an impact on the
signification. While the Old German contains an identifiable Teil (part) in
its midst, the modern one does not. The former is therefore identified as a
composite of two, while the latter as a single word (ibid., p. 99). Finally,
if a sign becomes divided, “infallibly a meaning (un sens) attaches to this
difference which has just been born” (Saussure, 2006, p. [231], 163). In
sum, one must note “the complete irrelevance of a viewpoint taking as its
starting point the idea-sign relationship outside time, outside transmis-
sion, which along tells us, experimentally, what the sign is worth” (ibid.,
p. [231], 163).
The assumed dualism of an eternal idea and an external linguistic enve-
lope is ultimately derived from the philosophers’ inability to integrate tem-
porality into their thinking. This is Saussure’s third objection to the no-
menclature view: “That itself gives pause for thought concerning the effect
of TIME, that unforeseen factor wholly ignored in the philosophical ap-
proach, on the marrying of an idea and a name” (ibid., p. [231], 163).

The effect of time extends beyond the alterations affecting either the
name or the idea regarded on their own—in fact, “there would be nothing
striking, typical, proper to language, if there were only these two types of
alteration” (ibid., p. [231], 163). Temporality proper to language itself in-
tersects the signifier and the signified in a relation of mutual interdepen-
dency; this naturally makes the ideas themselves subject to mutation and
change. Importantly, the source materials do not support the structuralist
notion of a fixed and immutable system; such a notion emerges rather as
the philosophers’ abstraction removed from the concrete reality of lan-
guage in usage; it preserves the traditional notion of language as a static
order of words received from substance metaphysics, but drops the whole
ballast of things. But the basic notion of a rationally knowable and uni-
versally valid order of intelligibility remains unchanged. Structuralism
emerges therefore like a skeletal metaphysics from which the substance
has been purged.
In contrast, language (la langue) is a site of both continuity and change.
Saussure posits ceaseless transformation as an absolute principle in gen-
eral linguistics (Saussure, 2006, p. [158], 105), and derides the scholarly
view that, for example, the transition from Latin to French happened in a
single leap: “The people of France have never woken up and said bonjour
in French, where they went to bed the previous evening saying ‘good night’
in Latin” (ibid., p. [152], 100). The notion of a sudden delivery of a hitherto
unheard idiom supposes recourse to a magic wand, with the French emerg-
ing like Minerva from out of Jupiter’s brain “with the armor of Latin cov-
ering her flanks” (ibid.). The notion that French is a distinct offspring of
Latin is predicated on the notion that stasis is the basic condition in lan-
guage and change an accident; however, “[a] case of an idiom in a static
state of rest does not occur”—and any idiom left to itself is a site of ongo-
ing transformation (ibid.). The logos of language is therefore not bound by
the logical principle of nonidentity according to which becoming is an
accident affecting being from the outside. In the concrete reality of lan-
guage, change is the rule.
In sum, Saussure’s critique of the nomenclature view of language offers
a philosophical alternative to the traditional metaphysical idea of being as
self-standing and immutable substance, the implied referent of language as
nomenclature. His conception of language privileges becoming over im-
mutability, and a relational understanding of language and meaning over a
series of disconnected words and things. It thus offers an alternative to the
received structuralist idea of language as an autonomous and closed system,
not unlike the implied worldview within substance metaphysics. Insofar as

this philosophical background is hard to glean from the published version
of the Course, its reader is hard pressed to fully appreciate the intellectual
complexity of Saussure’s approach, and may be prone to read the tradi-
tional metaphysical and rationalist understanding of language (and reality)
into his project of general linguistics.
The nomenclature view of language and Bergson’s

philosophy of life
It is feasible that the editors of the Course projected their own unexamined
beliefs as they crafted a reconstructed and synthesized vision of the source
materials. Bally’s view of language around the time he worked on the
Course was, as he recounts in the 1913 essay “Le Langage et la vie” (Bally,
1965), heavily inflected by Bergson’s philosophy of life, and its implica-
tions for the understanding of language. Bally is partial to Bergson’s figur-
ing of life as incessant movement and fluidity, immediately available to
intuitive insight via introspection but in excess of intellectual understand-
ing, with its fixed logical categories, and hence to be studied objectively,
via observation. Bally is largely in agreement therefore with Bergson’s
claim that:
La pensée est un mouvement indivisible et les idées correspondant à chacun

de mots sont simplement les représentations qui surgiraient dans l’esprit à
chaque instant du mouvement de la pensée si la pensée s’arrêtait, mais elle
ne s’arrête pas. (cited in Medina, 1985, p. 97)
This opposition between the dynamis of living thought on the one hand
and the statis of ideas on the other leads to the conception of language as
imperfect symbolism of direct experience. We inescapably fail to translate
what the soul feels, “thought remains incommensurable with language . . .
language fixes the objective aspect of the feelings in the soul only” (ibid.,
p. 99). For Bergson, words are then but approximate labels (étiquettes) for
the reality of our thought, unable as they are to fix subjectivity, affectivity,
and movement:
Nous ne voyons pas les choses elles-mêmes; nous nous bornons, les plus sou-
vent, a lire des étiquettes collés sur elles. Cette tendance, issue du besoin,
s’est encore accentue sous l’effet du langage. Car les mots (a l’exceptions des

noms propres) désignent des genres. Le mot qui ne note de la chose que sa
fonction la plus commune et son aspect banal, s’insinue entre elle et nous . . .
Nous ne saisissons de nos sentiments que leur aspect impersonnel, celui que
le langage a pu note rune fois pour toutes parce qu’il est a peu près le même,
dans les mêmes conditions, pour tous les hommes . . . Nous nous mouvons
parmi des généralités et des symboles. (ibid., p. 100)
Bally adopts the polarized distinction between the immediate data of

singular thought on the one hand and the objective reality of linguistic
symbolism on the other, but maps it onto language itself. As such he rede-
fines Saussure’s distinction between la parole and la langue as subjective,
affectively laden, singular expression on the one hand, and objective ratio-
nal system, on the other. The former transpires in linguistic expression, the
latter in communication. The expressive dimension of language would be
at odds with the communicative one, since mutual understanding is contin-
gent on words expressing “simple, general, and abstract ideas,” and rela-
tions between them. Communication is therefore enacted at the expense of
expression, where ideas are “synthetic and singular and therefore incom-
municable” (Bally, 1965, p. 79). Communication is therefore figured as a
force sapping the vitality and affective charge of direct subjective experi-
ence, this pre- and extralinguistic datum which would find a de jure imper-
fect representation in the language destined for public use, with, for ex-
ample, words like death, suffering, injustice serving as “labels for cold and
abstract concepts” (ibid.) far removed from the living concrete reality of
the corresponding affects.
The Bergsonian heritage, with its view of language as an intellectual
label befitting exchangeable parcels of objective reality better than singu-
lar, subjective life provides a lens through which Bally reads Saussure in
his opening address as chair in general linguistics in 1913—and it can be
assumed, also in the editorial work on the Course. Saussure gets figured as
a “convinced intellectualist” due to his preoccupation with language rather
than affectivity (ibid., p. 157). We hear that Saussure’s “scientific tempera-
ment forced him to seek, and made him find, the regular, geometric and
architectural aspects of language” (ibid.). Bally’s own announced contri-
bution to general linguistics will be to expose the contrary force of speech
(parole), a site of unmediated expression of subjective life, in stated agree-
ment with Bergson. The Bergson read by Bally impacts the way Bally
reads Saussure. Hence the stated emphasis on an opposition between la
langue and la parole resonates directly with the Bergsonian dichotomy

between life and language but not with Saussure’s indissociable duality.5
The figuring of the arbitrary relation between the signifier and the signi-
fied as an unnatural and external relation of labeling a preexistent idea
with a conventional name is consistent with Bergson’s linguistic symbol-
ism but not with Saussure’s semiological system. Since Bally wholeheart-
edly accepted Bergson’s view, it is feasible that he reconstructed and syn-
thesized the principle of arbitrariness in accordance with this philosophical
and linguistic commitment. He notes: “When language (la langue) real-
izes its purpose, the linguistic sign becomes purely conventional or as
Saussure says arbitrary . . . The word is nothing other than a label (étiquette)
applied to a concept” (Medina, 1985, p. 101).
Bally thus collapses Saussure’s distinction between value relations be-
tween signs and signification of an individual sign into a single category of
pure concept; language emerges as a collection of word-labels (mots-
étiquettes) attached one by one to pure concepts (signifieds or values).
Both word and idea would be removed from the concrete sensorial reality
which the word and idea designate in parole—for example, the word
woman and its associated idea would be but an external garment for a sen-
sory image of a woman (ibid.).
The professed dichotomy between life and language resonates strongly
in Bally’s autobiographical note, a striking public confession of a struggle
to bring them together in his own experience:
Where does this obsession (hantise) with speech as function of life come
from? Here is the paradoxical cause I believe to have discovered.
One admits that to speak congruously about something, one needs to pos-
sess it fully. Now, in my case, it is less a possession than a privation. I want
to say that the one who wrote Le langage et la vie was rarely able to put
language in the service of life.
As far back as I look back in time, I was faced with inhibitions of speech. There
is first of all a timid and awkward boy, paralyzed whenever he has to speak to
grownups, seeking in vain the words and phrases to formulate innocent desires
or to convey simple messages (executer des faciles commissions) . . . Later, it is
5
Even Sechehaye remained unconvinced that Bally’s emphasis on pure affectivity was justified:
“Il y a autour de la langue . . . un halo de faits en devenir. La question est de savoir si ce
phénomène est d’origine purement affective. C’est dont je ne suis pas persuadé.” Similarly,
Sechehaye expressed doubts that the opposition between affective life and language could be
reconciled with Saussure’s distinction la parole-la langue, considering that one cannot draw a
precise distinction between them in the student lecture notes (Redard, 1982, p. 19).

the entry into life, and the disappointment, repeated without end, of being
unable to say the right words at the decisive moment, and of saying those that
should have been passed over in silence, and which one will always regret.
These sufferings have left an indelible mark upon me, and my inner rumi-
nations have made me see language as an entity both mysterious and attrac-
tive, similar to those paintings in a museum which one can contemplate
without ever appropriating them.
These painful experiences gave birth to reflections on the relation between

language and thought, and feelings. These obscure intuitions became more
precise and organized with time. I do not know if they have any value; they
are lived impressions, slices of life.
In sum, it was the faulty functioning of my speaking apparatus (le mauvais

fonctionnement de ma machine a parler) which incited me to look at it
more closely. When one handles a tool easily, one is not tempted to examine
how it is made. Had I been a master of my language and pen, I would per-
haps never have looked for what lies behind the words. (Redard, 1982, pp.
7–8, my translation)
La langue: systemic, social, historical
In this section I propose to make a case for the primacy of the language
system in its basic bond to sociality and temporality within Saussure’s
general linguistics. I will draw on the student lecture notes, especially
those by Constantin (the most complete source), and complement them
with autographed sources whenever possible. My commitment remains to
unveiling the philosophical complexity of Saussure’s understanding of
language that emerges therein—and risks getting obstructed otherwise. I
will focus on the planned chapters “absolute and relative arbitrariness in
language,” “immutability and mutability of the sign” (pp. 85–102) and un-
named (pp. 141–143), the concluding chapters of the third course on gen-
eral linguistics delivered by Saussure on May 5 and 19, 1911, which follow
the chapter with the (revised) titled “Language as a system of signs” (“The
nature of the linguistic sign” in the Course). Importantly, the discussion of
linguistic arbitrariness in the student’s lecture notes is not tied to the lin-
guistic sign considered in isolation, but to the conception of language as a
system of signs; arbitrariness is exposed as relative and mediated by the
semiological system, itself tied to the historically situated community of
language users. Linguistic arbitrariness is not an objective property of a

closed and autonomous system of signs, but effectively intersects systemic
“objectivity” with the “subjectivity” of historically sedimented and revis-
able social practices of language use.
Parts of my argument in this section were anticipated by Jonathan Cull-
er’s work.6 Culler acknowledges the importance of motivation for under-
standing the language system, that is, that signs are motivated by one an-
other even though they are also defined as unmotivated or arbitrary. He
concedes—in reference to relevant source materials—that relative arbi-
trariness is essential to understanding arbitrariness tout court (Culler,
2003, pp. 57–58), against his own earlier stated views (Culler, 1986, p. 30).
Culler concludes that the two principles—that the sign is arbitrary, and
that the linguistic system is a system of motivation—are interdependent.
He writes: “It is because the relation between signifier and signified is un-
motivated that la langue becomes a system of motivation” (Culler, 2007,
p. 127). Drawing on Derrida’s inspired reading of the Course in Glas, as
well as Saussure’s reflections on analogical creation in language and tex-
tual analysis of the anagrams, Culler proposes that “Language motivating
itself” may therefore be the most apt description of linguistic play (ibid.).
I am in agreement with Culler that the principle of arbitrariness devel-
oped in Saussure’s Nachlass is not to be confined to the properties of an
individual sign and extends to the interplay of motivating forces within
language as a whole. I will demonstrate below that such a systemic or in-
tralinguistic understanding of arbitrariness reflects, however, only a single
facet of a more complex process, and that arbitrariness would become ab-
stract if regarded solely as a system-bound phenomenon. While arbitrari-
ness precludes natural motivation (where nature refers to a preordained
order of natural kinds, as discussed above), it is positively defined as the
multitude of temporally sedimented social conventions, which exert a mo-
tivating force upon language. It follows that the “essence” of arbitrariness
extends beyond purely linguistic principles, to be interrogated within the
bounds of an internal language system alone, and that the forces of moti-
vation and play are not exclusively linguistic either.7 Finally, I will demon-
strate that social and temporal forces are at work even in the case of an
individual sign (“Why do we say man, dog? Because before us people said
6
Notably, L’Essentiel de l’arbitraire, 2003, and “The sign: Saussure and Derrida on
arbitrariness” published in The Literary in Theory, 2007.
7
I believe that Culler’s reflections on the role of performativity in language inspired by J. L.
Austin and in the process of enacting gender roles inspired by Judith Butler are largely in
agreement with this enlarged understanding of cultural signification.

man, dog”) (Saussure, 1993, p. 97), and so individual significations are not
absolutely unmotivated either but always already relativized by the prop-
erties of language as a whole as well as the apposite sociohistorical con-
ventions. Ultimately, it may therefore be misleading to attribute properties
like arbitrariness to an individual sign since the latter can hardly be main-
tained as a unit of analysis; such attributions may inadvertently return us
to the view of language as a string of nouns or nomenclature.
*
Some of the developments related to linguistic arbitrariness as a sys-
temic property of language as a whole do transpire in the Course. They are
taken up notably in part II, ch. IV on linguistic value, where the definition
of a linguistic term in terms of a simple union between a sound and a con-
cept is deemed “a great mistake” (une grande illusion), for one would need
to construct the system by adding up individual terms while the terms
emerge from an interdependent totality (Saussure, 1986, p. 112). It follows
that “the concepts are purely differential and defined not by their positive
content but negatively by their relations with the other terms of the system”
(ibid., p. [162], 115; see also entire section “The sign considered in its total-
ity” in part 2, ch. IV). This relational conception is fleshed out by way of
syntagms (linear sequences) and associations (intersecting webs within
which each individual term is embedded) in part II, chs. V (“Syntagmatic
and associative relations”) and VI (“Mechanisms of language”). Finally,
the question of arbitrariness is revisited in section 3, ch. VI, under the dis-
tinction between “absolute” and “relative” arbitrariness, where it is ac-
knowledged that language as a whole imposes a limitation upon arbitrari-
ness (ibid., pp. [180–184], 130–132).8 A relational conception of language
is therefore given its due within the Course.
However, the existing architecture of parts and chapters in the Course
makes it challenging for the reader to appreciate that this relational under-
standing retroactively revises the validity of the claims made in part I, ch. I
“Nature of the linguistic sign.” The discussion of a relational view of lan-
guage is pushed back to the midsection of part II (chs. IV, V, and VI), and
the reader may be hard pressed to receive this later discussion as an inte-
gral element of a sustained engagement with the problem of linguistic ar-
bitrariness, and to track the redefinition of this term from an initial (and
8
Note however that the distinction between absolute and relative arbitrariness is reflected in a
dedicated chapter in the lecture notes, but not so in the Course, where it is classified under the
third section of part II, ch. VI.

myopically focused) emphasis on the properties of an individual sign to a
broader view onto the play of interlinguistic solidarities and differences
within which the principle of arbitrariness must be couched.9 As a result,
linguistic arbitrariness becomes redefined by way of linguistic difference,
and effectively dethrones the solitary sign that reigns unchallenged in
part I, ch. I of the Course. The stylistic discrepancy between the traditional
format of an academic treatise and the lecture notes is significant in this
context: the architecture of chapters and parts suggests a series of self-
contained building blocks, while the lecture course allows for a back-and-
forth movement of ideas, where what comes later in the process may reflect
back on and alter what came before. Such internal redactions are inher-
ently difficult to render within a classic academic book format.
As for the discussion related to the social and historical forces operating
within the language system, they also transpire in the Course and are ad-
dressed specifically in ch. II on immutability and mutability of the sign,
part I. In this chapter, it is acknowledged that language (la langue) consid-
ered outside the social world would be made “artificial” (irréele), and that
the action of time combined with social forces helps to appreciate that
language cannot be simply changed at will despite its conventional and
arbitrary character (Saussure, 1986, p. [112], 77). Subsumed under a sepa-
rate chapter, this discussion does not emerge, however, as a necessary step
within an argument about linguistic arbitrariness; the reader may be hard
pressed to regard the turn to sociality and temporality as being integral to
understanding general linguistic principles. Its placement in the Course
between the sections on the arbitrariness of the sign (part I, ch. I), and the
distinction between synchrony and diachrony (part I, ch. III), which re-
ceived the most scholarly attention and “hypnotized” the readers of the
Course (who either adopted or rejected them wholesale) likely contributed
to a focus on the language system alone, at the expense of considerations
related to language change and the social context of use, within the domi-
nant structuralist reception (de Mauro, 2005, p. 448, n146). As a result,
ch. II received little if any attention in the structuralist readings of the
Course—it is one of the least referenced sections of the book. The discus-
sion of immutability and mutability of the sign puts pressure, however, on
the official view that general linguistics is an ahistorical discipline and that
language (la langue) constitutes an autonomous and closed system. If read
9
As Bouquet noted, it is ironic that this difficulty persists in the 1968 and 1974 critical editions of
the Course, which also leave out some the relevant material from the lesson of May 19, 1911
(Bouquet, 1997, p. 285).

deconstructively, against the stories told by its better-known neighbors in
the book, this chapter would effectively complicate the order of hierarchi-
cal dichotomies from the “Saussurean doctrine” associated with the
Course, and call into question the possibility of general linguistics being
realized by way of traditional scientific methods.
For reasons stated above, I propose to retrace Saussure’s engagement
with the theme of linguistic arbitrariness as it unfolds within the lecture
notes themselves, and point out the internal redactions this theme under-
goes in the process of delivery. In the lecture notes, the chapter on abso-
lute and relative arbitrariness opens with a reminder to the student about
the initially considered notion of absolute arbitrariness: “We posited as
evident truth that the link between the sign and the idea represented is
radically arbitrary” (Saussure, 1993, p. 85). This initial notion gets tested
and qualified as the lectures run their course. Arbitrariness, the students
are now told, admits of gradation: radical arbitrariness would consist in a
zero level of motivation, while relative arbitrariness would be motivated
to a degree. Saussure offers vingt (twenty) as an example of radical arbi-
trariness, where the term cannot be explained or justified via any other
terms; in the case of dix-neuf (nineteen), arbitrariness is relative since the
term is motivated by other coexisting terms, dix and neuf (nine and teen)
(ibid., p. 85); it therefore makes a reference to the idea neuf as it affects
numbers dix-sept to dix-neuf in the French (or teen as in thirteen, four-
teen, fifteen . . . nineteen in English). Similarly, the English plural ships
makes a reference to the idea of plurality as deployed in a whole series:
birds, flags, books, and so on, and is justified by this serial reference (ibid.,
p. 86). The relative degree of arbitrariness is determined therefore not
simply by a relation between one single term to another (e.g., as ship is
tied to ships) but by a general grammatical principle at work in many lexi-
cal instances. The latter are therefore internally wrought by grammar and
cannot be separated off into an inventory in a lexicon.
Note that the motivating forces described thus far are located within the
language system and not by an order of preexistent, extralinguistic reality.
Saying that linguistic signs are relatively arbitrary or partially motivated
does not then suggest any natural determination (causality, resemblance)
of the signifier by a transcendent signified. The degree of motivation em-
ployed in the radical/relative distinction is not taken in a naturalist but in a
semiological sense. Whatever rules the forces of relativity may follow in
motivating a sign, they are unique to the logos of language itself, and do
not necessarily map onto metaphysical and/or logical rules and principles.
Determining the scope of relativity in arbitrariness is then akin to

determining the hold of language on any of its distinguishable elements,
and tracing the many ways in which the system works on the sign. The
proper context for the arbitrarily motivated distinction thus shifts from the
usual relation between an individual sign (signans) to its signified (signa-
tum) to a relation between the sign and the language system as a whole; the
system of relations (or the entrainment of the sign by language, to use Der-
rida’s term from Glas) thus figures as the primary notion.
It may seem that the initially considered possibility of a radically ar-
bitrary sign would be immune to any determination by relations within
the language system. However, consider that this possibility, like the
nomenclature view of language discussed above, is employed in the lec-
tures as a didactic exercise, that is, it introduces a commonly held notion
which becomes displaced in favor of a more adequate view as the inter-
rogation proceeds. The style of presentation is as significant here as the
content: Saussure did not simple dictate a readymade doctrine to his
students, but explored ideas that took shape as the interrogation pro-
ceeded. Hence the aforementioned difficulty in projecting an axiomatic
style of presentation employed in the Course onto the evolving course of
lectures where what is entertained at first may be rejected at last; if the
academic presentation mimics the order of presentation from the living
lecture course, it fixes these initial assumptions into axiomatic dogmas,
and the remainder into trailing corollaries. The initial thesis of arbitrar-
iness thus figures in the Course as an axiom and all the developments of
the thesis as corollaries; in the lecture course, however, this initial thesis
morphs in the process of delivery. The order of discovery in the lectures
is in the case of arbitrariness inverse to the order of truth, and what
comes last retroactively revises what was posited at first. This clash be-
tween an exploratory and a dogmatic presentation style accounts for the
difficulty involved in transforming the lecture notes into the classic
format of academic treatise, which superimposes an order of truth onto
an order of discovery.
Consider the initial example again: vingt features a short, single-syllable
word which does not call attention to its component phonic or graphic
parts, or to its grammatical construction (e.g., the grammatical case, with
the assumed case being the nominative). The word appears as a non-
further-analyzable unit, which straightforwardly points to its intended
meaning, like a finger inviting the viewer to look at the indicated object
and not at itself. Absorbed in this task of ostensive indication, the language
user looks away from the word to the idea and the thing, oblivious of the
hold that the language system exercises on the word nonetheless. As the

lectures progress, the student is gradually made aware of how firm, even
though initially hidden, the hold of the system on each sign is. This realiza-
tion leads to eventually banning any notion of absolute arbitrariness as
irreconcilable with the primacy of language system. As put in Saussure’s
autographed notes for the Course in General Linguistics:
Reduction in any system of langue of absolute to relative arbitrariness; this

is what constitutes the “system.” If it was possible for a language (une
langue) to consist solely in naming objects (a dénommer les objets), all the
various terms in this language would have no relation among themselves;
the terms would remain as separate from one another as the objects them-
selves. (Saussure, p. [328] 233)
The traditional notion of language as nomenclature carries as its asso-

ciated thesis, absolute arbitrariness of the sign; both are committed to the
possibility of self-standing signs representing ideas and things, and both
need to be rejected after due consideration. In fact, the notion of absolutely
unmotivated arbitrariness is as problematic as the notion of natural moti-
vation (onomatopoeia, interjection) because both assume that signification
is a property of the sign standing in isolation from other signs, and have
not accomplished the necessary reduction of the sign as positive entity to
the more appropriate notion of language as a relational web without posi-
tive terms. So long as one privileges the one-to-one relation between the
signifier and the signified, it makes little difference if one construes this
relation as arbitrary convention or natural imitation; in both cases of zero
level and full force motivation, one remains wedded to the view of lan-
guage where words name things. In both cases, language appears as a
loosely strung together lexicon, untouched by grammar.
According to the development of the argument in the student lecture
notes, arbitrariness turns out to be, paradoxically perhaps, both radical and
relative. Radical because the terms of the language system are unmoti-
vated by any fixed order of things (natural law, divine decree); relative
because they are inextricably interconnected within a web of relations to
other signs and subject to the internal motivating forces within the system.
As a result, the unmotivated relation between the signifier and the signified
within an individual sign will get intimately cross coupled with the host of
relations between the signs, shifting the initial focus on an isolated signans
as a support of a signatum to “Language as a system of signs.”
The initial distinction between radical and relative arbitrariness gets
figured thus:

On the one hand we have this relation, already mentioned:
Concept And on the other hand this relation

Term A Term B
Acoustic image
figure 4
(source: Saussure, 1993, p. 89)
The figure on the left is the familiar illustration of the arbitrariness of

the sign, with the vertical bilateral vector representing the reciprocal rela-
tion between the signifying and the signified facets; the figure on the right
introduces another relation—to a coexistent term within language, with
the reciprocity deployed vertically between the terms (Saussure, 1993, p.
90). The new diagram thus shifts the initial focus on an isolated sign to the
one on the sign and the system. The following discussion and diagram add
further emphasis to this shift (Saussure, 1993, p. 96):
Every language forms a body and a system . . . it is this aspect which is not
entirely arbitrary, where one must recognize a relative rationality. The con-
tract between the sign and the idea is much more complicated.
It must not be considered like this: But like this:
figure 5
According to the diagram, the signifying/signified relation, initially con-

strued in one-to-one terms, does not suffice to capture the nature of the sign
and the principle of arbitrariness. Each sign is subject to the forces of rela-
tivity reigning within the system. However, the now acknowledged impact
of the system on the sign cannot be easily reconciled with the initial defini-
tion of the sign as a signifying/signified duality (Figure 6). Having reintro-
duced the diagram for the initial definition, Sau-
ssure notes: “In this view, the signification is the Concept
counterpart of the auditory image and nothing else. b) Auditory image
The word appears or is taken as an isolated and figure 6
absolute whole” (ibid., p. 135). A trap (piège) awaits (source: Saussure, 1993,
us in this view as in a cave (caverne): signification p. 135)

seems to be directly and completely engendered by the auditory image and
yet contingent on the mediating terms in the language system. It is unclear
how these two sets of relations: within the sign, between the signs, coexist
and how the vertical and horizontal vectors intersect. Or, from a different
angle, it is unclear how one distinguishes the relation within and between
the signs (ibid., p. 135).
The response to this paradox is not simply to graph the language system
onto the initial view featuring signs or words in isolation, but to adopt the
view of the system from the start (the preceding chapters would have been
organized in this manner [ibid., p. 137]). The sign will therefore no longer
be defined in terms of signification only, but also its value—a category
relative to the given term’s relations with other surrounding terms. To
return to the classic example, the French mouton does not have the same
value as the English sheep, since the former refers both to the animal “on
the hoof and on the table,” while the latter to the animal but not to the meat
(mutton) (ibid., p. 136). This simple lexical fact illustrates the far-reaching
philosophical idea that a word such as sheep cannot rise to the task of la-
beling a preexistent natural kind grasped by means of a universal idea,
since the ideas are shaped by what could be termed the semiological halo
or horizon of the individual word, that is, what surrounds and shapes it in
a given language. This semiological horizon is not simply external to the
individual word, circumventing a clearly delimited, bounded being; ac-
cording to the notes, signification and value overlap in part, even though
they cannot be collapsed.10
Saussure proposes therefore these additional diagrams to capture the
systemic view of language wherein all the terms are linked by relations:
a) a) a) a)
figure 7
This representation shifts focus from the individual terms and their sig-
nifications to the value relations between coexisting terms. This raises the
10
“It is perhaps one of the most subtle points in linguistics, to see how sense depends on but
nevertheless remains distinct from value. It is here that the difference between the linguistic view
and the narrow view considering language as a nomenclature breaks through” (ibid., p. 134, my
translation).

question as to how the two stand together? The concluding diagram in the
lecture illustrates that:
Signifying
figure 8
Signification as counterpart of the [auditory] image
and signification as counterpart of coexisting terms
merge (se confondent)
The systemic understanding of language is fleshed out by way of the two

kinds of relations in which individual words stand: syntagmatic and asso-
ciative (ibid., p. 128). The syntagmatic relation combines two or more units
in a consecutive relation which extends in a one-directional, linear manner.
The relation is often termed a syntagmatic solidarity to underscore the con-
nectedness between the succeeding terms, rather than a string-like succes-
sion. For example, contre tous (against all) relates successive terms on a
linear chain to produce the familiar set phrase. Such a syntagmatic relation
can also be noted within a word, for example, contremarche (counter-
march), a composite word in which two segments contre and marche can
be made out. The syntagm within a word thus breaks down the initial im-
pression of seamless unity and unveils the many-in-one complex harbored
in its midst. The syntagmatic relation plays out therefore both within and
without an individual word; it is unclear how the two could be set apart.
The associative relation ties any individual term to a whole series of
related terms. For example, enseignement belongs to an open series includ-
ing enseigner (to teach), enseignons (we teach), enseigne (teaches), and so
on, but also to the series including armament, rendement, and other simi-
larly formed signifiers, as well as the series based on the signified element,
such as instruction, apprentissage, education, and so on (ibid., pp. 129–
130). Both the signifying and the signified facets belong to such innumer-
able and criss-crossing series (ibid., p. 129).
The two relations are initially contrasted in the following way. The syn-
tagm underscores the extended dimension of linguistic terms, it is termed
an arrangement in praesentia since the terms must be laid out in an order
for the composite form to arise. In the case of association, the assemblage
is in absentia, since the associative series has properly speaking no begin-
ning and no end (ibid., p. 131), it is rather an open constellation with none
of the implicated terms being first or last. Even though the two orders of
relations are distinct and irreducible, they necessarily coexist. The notes

propose an analogy between these two orders and a perception of the col-
umns in an edifice:
If we compare them to the parts of a building: columns will stand in a cer-

tain relation to a frieze they support. These two components are related in
a way which is comparable to the syntagmatic relation. It is an arrangement
of two co-present units. If I see a Doric column, I might link it by associa-
tion with a series of objects that are not present <associative relations>
(Ionic column, Corinthian column). (ibid., p. 133)
While the associative relation is initially figured as a relation between

individual words, suggesting no more than a lexical inventory of self-
standing terms, the student is presented with a more complex, systemic,
understanding of this relation. The grammatical category of number dis-
cussed above is an example of how singularity and plurality are shaped by
presence or absence of the dual number in a given language. A similar
comment could be made relative to the category of tense:
<Idea of different tenses, which seems quite natural to us, is quite alien to
certain languages.> <As> in the Semitic system (Hebrew) there is no dis-
tinction, as between <present>, future <and past>; that is to say these ideas
of <tense> are not predetermined, but exist only as values in one language
or another. (ibid., p. 140)
Consider finally the classic case of a noun, used as paradigmatic under

the nomenclature view of language. While it is usual to think of a noun as
a self-standing category such as found in a lexicon, it is more appropriate
to regard it as a whole series of related terms. Rather than focus just on the
nominative form, for example, dominus or king, one should take the entire
flexional paradigm such as the Latin dominus, domini, domino—a group
based on associative relations (ibid., p. 133). Each individual term within
the group is dependent on the relations to the other ones. This relational
series is not to be thematized as a hierarchy beginning with a primary term
dominus and followed by its inflected derivatives, like the head of a family
followed by the dependents. The nomenclature view typically isolates the
nominative case out of the whole series, and casts it as a noninflected,
proper name for the designated referent. The nominative case becomes
therefore neutralized in its grammatical effects, as if it pertained only to
the simple fact of naming, assumed to be the privileged function of the
noun—rather than say, that of calling someone by their name (vocative), or

offering them a gift (dative)—an act that is reciprocal and implicates a
possibility of response.11
Isolating just the naming function of the noun out of the whole series is
unfounded from the point of view of language itself: the flexional constel-
lation does not have a beginning or an end (ibid., p. 133); any term from the
declensional series can be summoned up as well as any other (ibid., p. 134).
The series is not a hierarchy crowned with the nominative form, assumed
to be the neutral and default mode, and the remaining modes and functions
(vocative, dative . . . ) are not a secondary form, an Abweichung (deflection
or deviance) from the standard mode. The constellation encompasses and
impinges on all the flexional forms, making it impossible to isolate just one
as a paradigm for all. It is also ambiguous if the flexional series features a
number of individual words, or just one, divided, multiplied, spread across
the declensional spectrum, and if the inflected forms are related from
without or from within, and where the boundaries lie. No unitary, bounded,
indecomposable word or name remains.
Hence:
We must not begin with the word, the term, in order to construct the system.
This would be to suppose that the terms have an absolute value given in
advance, and that you have only to pile them up one on top of the other to
reach the system. On the contrary, one must start from <the system>, the
solidarity of the whole (tout solidaire); this totality may be decomposed
into particular terms from the rest, although these are not as easily distin-
guished as it seems. (ibid., p. 134)
Now if the unity of the word is unstable and relative, shot through with
the syntagmatic and associative vectors, extending horizontally and vertic-
ally in all directions, apparently missing a solid core, then the properness
of the naming relation traditionally assigned to it gets lost. Neither of the
serially related grammatical forms can aspire to the status of the proper
name, filled out by a proper idea and cashed out in an ostensive relation to
a thing in the world. Each one is shot through with the play of differences.
Ideas themselves now seem multiple, related, and distinct, clustering
around the differences within and between the forms.
11
The nominative case stems from Latin nominativus “pertaining to naming,” from nominatus,
past participle of nominare—1540s, “to call by name,” from L. nominat-, past participle stem of
nominare “to name,” from nomen “name.” The nominative case is traditionally assumed to offer
the basic and elementary form of the noun from which the remaining forms like vocative or
instrumental are derived.

These considerations lend further support to the revision of the initial
view of language based on an isolated sign. This view is untenable for it
tacitly assumes that an isolated sign—a self-standing, solid label com-
posed of a stretch of sound and linked up with an idea or unit of sense by
force of convention—could yield the complex web of solidarities and con-
trasts; however, no single sign can signify at all prior or independently of
these relations of solidarity and difference. The focus on the isolated sign
is effectively bound to the psychophilosophical view of ideas as self-
segmenting units of signification, subsequently labeled with names.
The familiar diagram representing the relation between the signified and
the signifier (Figure 9) is therefore demoted to a second-
ary role. According to the lecture notes, “we see that it
is doubtless justified but is only a secondary product of Signifying
value” (ibid., p. 139). The famed diagram illustrating the
sign/signifier/signified relation should therefore not figure 9
figure as an autonomous notion, but rather a product of (source: Saussure,
1993, p. 139)
a more primary and basic situation.
The principle of arbitrariness also gets redefined in the process: it is the
contract between the whole series of signifying and signifieds that is arbi-
trary (ibid., p. 138). Hence, it is the manner in which any given language
parcels up or articulates signification that is arbitrary, founded neither on
the universal order of nature nor ideas, and does not therefore fully map
onto other languages and their ways of meaning-making.12 Any head-on
access to meaning is therefore perpetually barred, with each sign implicat-
ing the whole series of signs and circulating signification across the series.
In Saussure’s terms, since the contract as a whole is arbitrary, the values
are entirely relative (ibid., p. 138). Arbitrariness thus gets reclaimed at the
end of the third course as the fundamental principle but refigured as pre-
condition of intralinguistic difference: “<If the sign were not arbitrary, one
would not be able to say that in the language there are only differences.>”
(ibid., p. 142)
The principle of arbitrariness becomes a condition sine qua non of there
being a system of differences or language at all. This relational understand-
ing exposes the forces of solidarity and motivation reigning within the lan-
guage system which a naturalist paradigm of the world and the nomenclature
view of language both cover over. The relational view gives way to a more
12
As de Mauro put it, “Saussure uses arbitrary because this adjective expresses the non-existence
of natural, logical, etc. reasons why the acoustic and semantic substance becomes articulated in a
certain determinate fashion” (de Mauro, 2005, p. 443, 137n).

positive understanding of language as a field of mutually motivating forces,
irreducible to causality and logical laws.
*
The systemic understanding of language developed up till now remains
abstract, mainly because the focus has been on the sign and the language
system (la langue) considered as an object dissociated from speech
(parole). According to the student lecture notes, such partial focus is prob-
lematic because “that would only be language (la langue) apart from its
social reality, and unreal (since comprising only one part of its reality). In
order for there to be a language, there must be a body of speakers using the
language. The language . . . is located in the collective soul right from the
start” (Saussure, 1993, p. 101). This social fact is part of the definition of la
langue (ibid., p. 101); in fact, this weight of the community would give
language a center of gravity (ibid., p. 97). Saussure reaffirms this point in
the autographed notes for the course: language (langue) divorced from
social reality (réalité sociale) becomes unreal (irréelle); importantly, a
speaking collectivity (masse parlante) is part of the “very definition” of
language itself (Saussure, 2006, p. [334], 238).
Language emerges therefore as a dual entity comprising the object-
side: a systemic organization of signs, and the subject-side: a community
of speakers with its stable and yet evolving conventions of usage. Even
though the socius is the implied necessary subject, it is not to be thought
of as an inventor or creator but rather a recipient of language from others,
especially from the past generations; still, the dominant patterns of usage
as well as any linguistic change is contingent on a consecration by the
society to take effect; language is therefore equally dependent on the
effect of time, which sediments and gives weight to past usage (Saussure,
1993, p. 97).
The corresponding diagrams will therefore bring in the social and then
the historical elements to complete the definition of language. The first
diagram looks like this:
While this graph does justice to the con-
crete reality of language by including its im-
plied subject, the socius, it is insufficient be-
cause it fails to consider the time factor, or it
considers language and society at a single point
Social body
in time. Only when frozen in an ahistorical
figure 10 present can language appear as a free system
(source: Saussure, 1993, p. 101) moving in a pure sphere of relations (ibid.,

p. 101). This ahistorical understanding is still congruent with the thesis of
an inherently arbitrary sign, since the latter similarly belongs to “an un-
constrained system built on logical principles alone, as a pure science built
on abstract relationships” (Saussure, 2006, p. [334], 239). The implied sub-
ject therefore could be a mastermind who laid the logical foundations for
language at the zero point of history, not unlike the primary geometer en-
visioned in Husserl’s The Origin of Geometry. Language is not, however, a
logical system independent of time: we speak a language borrowed from
others at a time immemorial, and language echoes without having a defi-
nite source. As put in the notes for the course: “Langue is not unconstrained
(libre), due to the principle of continuity or indefinite solidarity with previ-
ous ages” (ibid., p. [335], 238); temporality accommodates alteration of
language and a shift in the system of values which a purely logical concep-
tion does not (ibid). Temporality in language does not exclude change—
there is mutability to a certain extent (Saussure, 1993, p. 98).
Historical reality is therefore coupled with social reality, both being
central to the full reality of language. Language “considered at any time,
is always an inheritance from the past” (ibid., p. 94). It has always already
received from previous generations: “never has any society known its lan-
guage other than as a product more or less perfected by preceding genera-
tions and to be taken as is. In other words, we recognize a historical fact at
the origin of every state of the language” (ibid., p. 94). A story of the ori-
gins can therefore only be told by historicized subjects.
The more appropriate diagram is therefore:
The body of speakers is multiplied by

time, considered in time
figure 11
While it is usual to think of language (la langue) as a closed and auton-

omous system, Saussure compares it to a social institution, like a school or
church. Language is however unique in comparison to the other social in-
stitutions in that it has an unbreakable hold on our existence and pervades
all areas of our life. Saussure notes that language employs thousands of
signs, thousands of times per day, every day (Saussure, 1957, p. 10). We
may not always be at school or in church, but we are always “in” language,
just as much as language is “in” us. Language and society are therefore
intimately coupled, and one cannot be understood in all concreteness

without the other; this does not imply, however, that the community can be
posited as the implied creator before the advent of language; Saussure
notes that “since the aim of language, which is to make oneself under-
stood, is absolutely necessary in any human society as we know it, it fol-
lows that every society is characterized by the existence of language”
(Saussure, 2006, p. [179], 121). Language can therefore be compared to the
many customs and habits used within the society for communicative pur-
poses (ibid., p. [178], 120).
Language belongs together with other socially relevant signs, such as
maritime signs, but also the rules of politeness, all the rites and customs
whose character is eminently semiological and social (Saussure, 1957,
p. 19). It will have to do with the gestures and practices which are signify-
ing for social groups at certain times; Saussure notes that when a custom
loses its meaning, we are faced with a case analogous to the one when the
words of a language become unintelligible to the speaking subjects (ibid.,
p. 19). A study of signs is therefore a study of signification, which is so-
cially and historically inflected. The new study of language through the
lens of signs will depart from the dominant, folk and philosophical con-
ception of language as nomenclature, which suppresses the reciprocal de-
termination of coexistent values in language (ibid., p. 20). It will therefore
be a study of signification in and through the systemic organization of
signs, and not via the analytic approach which breaks the system down to
individual signs.
The collective dimension of language leads to a comparison with a
vessel in the sea whose course cannot be determined a priori, by its inter-
nal structure, just as the sign’s relation to a concept is not determined in-
ternally, by individual reason (ibid., p. 25). The community provides the
very environment (milieu) where language lives, and language “is made
for collectivity, not for an individual, like a vessel is made for the sea”
(ibid., p. 26). Saussure emphasizes that:
It is only when the system of signs becomes a thing of the community that
it merits the name, is a system of signs at all. It is because the totality of its
conditions of life is so distinct from this moment from everything which it
can constitute outside of [the community] that the rest appears unimportant.
(Saussure, 2006, pp. [289], 202–203)
Saussure emphasizes that the “semiological phenomenon” never leaves

this social collectivity behind. Even more, the social dimension of the sign
counts as one of its internal and not external elements, and language is

equally a semiological and social product (ibid., p. [290], 203). The image
of the vessel at sea should not then be construed as a container-like relation
where preexistent signs become subsequently inserted into the social world
like a pilot into a ship. Signs are internally worked through by the socius;
without the historically sedimented and daily renewed contract between
speaking subjects there would be no signs and no signification at all. Fi-
nally, putting the ship out to open sea means allowing for unpredictable
change: “if the community environment (milieu) changes everything for
the sign system, this environment is also the veritable site of development
towards which the sign system tends from the start” (Saussure, 2006, pp.
[289–289], 203). To anchor language in a social milieu is therefore to aban-
don the notion of a closed and autonomous sign system. “Langue, or indeed
any semiological system, is not a ship in dry dock, but a ship on the open
sea. Once it is on the water, it is pointless to look for an indication of the
course it will follow by assessing its frame, or its inner construction as laid
out in an engineer’s drawing” (ibid., p. [289], 202). Hence “only the ship at
sea may yield information about the nature of a ship, and, moreover, it
alone is a ship, an object available for study as a ship” (ibid.).
A rationalist and individualist approach to language would then be not
unlike the standpoint of an engineer who designed the internal structure of
the ship but who does not test it at sea, and so does not situate it within the
usual context of the maritime environment. Saussure thus moves deliber-
ately away from a Kantian understanding of language in terms of a priori
categories and relations housed within an individual mind; a Kantian ap-
proach to language would be comparable to one which derives the trajec-
tory of the vessel solely from its form, oblivious to the many unforeseeable
movements of the sea which effectively guide the vessel on its journey.
Forgetting the maritime environment in the case of a ship at sea is thus as
shortsighted as forgetting the social environment in which language runs
its course; attempting to thematize language via the categories of individ-
ual reason is then akin to some engineer’s attempts to narrate fantastic
stories featuring sea monsters and pirate attacks from the deck of a ship
which never left dry land.
If we accept the “Saussurean doctrine” with its hierarchical opposition
between la langue and la parole, we are likely to view the relation between
the social and the individual dimensions of language as being similarly op-
positional (la langue would be made up of social conventions, la parole of
individual acts of speaking). We would then need to assume that the speak-
ing individual is to be evacuated from a structured language system in
Saussure’s linguistics. The source materials offer however a more complex

narrative; la langue is defined as a social product (Saussure, 1993, p. 66),
tied to the many conventions and customs relative to usage; however, the
notion of la langue gets fleshed out in the context of speech (la parole),
which itself is undecided between an individual act and a social act since it
is interindividual, and composed of a minimum of two participants (“as far
as language is concerned, the individual act supposes two individuals”
[ibid., p. 67]).
Hence, even though Saussure defines la langue as a social fact, he does
not forsake the individual to attain the social, but rather approaches the
social from within the individual act: “The social act cannot reside else-
where than in the individuals added together but as for any [other] social
fact, it cannot be considered outside the individual” (ibid., p. 69).
The speaking subject is ambiguously situated between individual and
social facts; she exercises an irreducibly individual faculty, since each act
of speaking is tied to the individual and to the moment of utterance (ibid.,
p. 91); the individual remains therefore master (ibid., p. 69); however, the
speaking individual deploys a deposit (dépôt) of verbal images coordinated
among themselves and consistent among all individuals (ibid., p. 69); the
speaking individual is therefore a repository of socially determined pro-
cesses, and as such each individual speech act partakes in the social life of
la langue, even though the actual speech situation is limited to the two (or
more) speaking/hearing individuals. While distinct, the two facets of la
langue and la parole, social convention and individual speech act, are
bound in a circular and mutually determining relation. Saussure instructs
his students to take down verbatim the following remark (and insert it
before the chapter dealing with the linguistic sign):
There is nothing in la langue which has not entered <directly or indirectly>

through speech (parole), that is through the sum total of speech acts re-
ceived (des paroles percues), and, reciprocally, there is no speech possible
before the development of this product called la langue, which supplies the
individual with the elements for the composition of his speech. (ibid., p. 91)
The individual and the social, the spoken and the structured dimensions
of language intersect therefore in a chiasmatic relation, and are not radi-
cally opposed.
The socius is historicized; it is the necessary subject of language so
long as it is figured as recipient of signifying conventions from the past
generations, who similarly received and transmitted an inheritance via a
transgenerational chain without a determinable beginning. Language is

not purely linguistic—it is enabled and reinforced by social conventions of
use which are and have been transmitted from mouth to ear and from one
generation of speakers to another. Language is an ongoing contract within
a speech community which endures and is relatively resistant to change;
even if this relative stability cannot be explained naturalistically, it sup-
poses an inertial effect of shared investment in a tradition that an arbitrary
system of signs considered on its own cannot account for. This receptive
stance and acknowledged passivity of the language user before the sign
removes any remnants of voluntarism from language, considering that his-
torically sedimented social practice largely constrains individual free
choice in language use:
The power of time constantly holds in check the power we may call arbi-
trary [free choice]. Why do we say man, dog? Because before us people said
man, dog. The justification lies in time. It does not suppress arbitrariness,
and yet it does . . . [To sum up] the non-freedom of the signs which make up
a language depends on its historical side, or is a manifestation of the factor
time in the language, [on the continuity of the sign across generations].
(Saussure, 1993, p. 97)
Saussure acknowledges the sociohistorical weight of language on the

speaking individual:
In relation to the idea it represents, the signifier [sign], whatever it may be,
is arbitrary, appears to be freely chosen, is replaceable by another (table
might be called sable or vice versa). In relation to the human society called
upon to employ it, the sign is not free but imposed, and the social mass is
not consulted: it is as if the sign could not be replaced by another. This fact,
which to a certain extent seems to embrace contradiction of the non-
freedom of what is free, <this fact> could be called colloquially the Hob-
son’s choice/forced choice (carte forcé). <You say to the language:>
“Choose at random,” but at the same time “You have no right of choice, it
must be this or that. (ibid., p. 94)
The full sociohistorical view of language is thus coupled at the sub-

ject side with the nonfreedom of the speaking subject, subjected as she
is to the weight of traditional form which both enables and constrains
individual expression. And while the Saussurean doctrine stipulates that
the speaking individual is to be evacuated from the language system, the
source materials state that an individual alone is powerless to effect

lasting change; what is needed is the sanction and consecration of a
community. As the lectures notes affirm, both signification and value
depend on the sanction of the community: “<No value exists by itself,>
which means that the sign will only have its own value through the
<sanction of the> collectivity. It seems that in the sign there are two
values (a value in itself and one which comes from the collectivity!) but
at bottom they are the same” (Saussure, 1997, p. 15).
As a collectively sanctioned medium, language is therefore amenable to
a social analysis and social critique; it figures as a vehicle of traditional
values which can be both empowering and oppressive for the individual
(see also Thibault, 1997, p. 216). While Saussure does not develop an ap-
proach geared specifically at such an analysis and critique, his notion that
linguistic signification and value bear and depend on the stamp of collec-
tive approval exposes their contingency and opens up the possibility of
language practice creating a nontraditional set of values. Differently put,
as a socially constituted medium, language is open to the possibility of
transmission and revision through the evolving practices of usage (I con-
nect this potential to change with contemporary emphasis on linguistic
performativity in the final section of this chapter).
Arbitrariness could be defined as an unfounded tie between the signifier
and the signified only when considered from an individual viewpoint ob-
livious of the social and historical constraints. Considerations of the full
reality of language effectively derail the initially assumed primacy of the
sign/signifier/signified terminological complex. These semiological dis-
tinctions are useful but need to be tied to social practice and historical
process to have any concrete meaning; considered on their own, as purely
objective notions, they become as abstract and ahistorical as rules of sym-
bolic logic. The structuralist claim of the sign/signifier/signified as the
single most important gift made to general linguistics by Saussure falls
short of the much greater generosity and scope of his project.
The initially assumed notion of the arbitrariness of the sign becomes
relativized when considered from the point of view of the speaking sub-
ject: the sign could be deemed absolutely arbitrary only on the condition of
the speaking subject’s ability to exercise unconstrained free will of a libre
arbitre in making linguistic choices—as did the mythical primary lan-
guage user at the dawn of time in a story of origins. The socius exposes,
however, the folly of such a coupling between a dire and a vouloir as sole
basis for meaning or signification (or vouloir-dire); the socius forecloses
the possibility of absolutely unmotivated or voluntaristic expression in lan-
guage; to borrow Merleau-Ponty’s distinction, the speaking speech is

always heavy with the traces of spoken speech, hence laden with contin-
gency and ambiguity. The social and historical dimension of language thus
serves as a force of relativity constraining the arbitrariness in language at
its subject side, while the systemic organization of language relativizes it
at the object side. Combined, the socius and the system catch any free
floating signifier into their net.
The notion of the arbitrariness of the sign shares with the received no-
menclature view of language an implied individual subject arbitrating how
things in the world will be called. The narrative of the origin of language
typically supposes that once upon a time there was an individual like
Adam who instituted language by naming the animals. In the philosophi-
cal context, this origin narrative resurfaces in the social contract theory,
where a group of individuals laid down a foundation for language and so-
ciety by means of a “gentlemen’s agreement” in a mythical past. Saussure
openly critiqued the eighteenth-century philosophical notion of a social
contract, understood as a deliberate or voluntary consensus traceable back
to an assumed, historical or mythical, source point in time. First, it would
be impossible in principle to retrace an originary agreement or contract,
and so the latter is only “theoretical” (Saussure, 1957, p. 17). In fact, the
question of the origin of language does not have the importance one usu-
ally attributes to it, in fact, it does not pose itself since the moment of
genesis cannot be seized or seen (ibid., p. 22). The notion of social contract
is better thought of as a permanent condition of language rather than an
originary event; taken in this manner, the term highlights the collective
dimension of language, an ongoing contractual relation between language
users, which could be understood as a basic investment and trust in the
signifying potential of language for oneself and one’s community. The
philosophical idea of contract suggests a voluntary decision, but neither
individuals nor social groups can change language at will (ibid., p. 17). The
contractual dimension of language does not therefore issue from a volun-
tary agreement enacted at the zero point in time; it is traceable neither to a
temporal nor to an agentive source point.
A conception of language as a string of names lends itself to the individ-
ualist and voluntarist view, because one can plausibly entertain the idea of
an individual naming one or many things, but an idea of an individual
bringing about a whole system of complex grammatical rules and princi-
ples in toto and ex nihilo into being is much less likely to be entertained.
We baptize our children and give names to our pets; we therefore resonate
with the narrative of genesis because like Adam we may have given a
name to a creature which from that moment had been called by that name.

An invention of a full blown language system like Esperanto or radical
change affecting language as a whole would have “an assembly of logi-
cians and grammarians” rather than the social community as its necessary
subject (Saussure, 1993, p. 96); and even a successful artificial language
still depends on the social sanctioning by the language community across
generations to exist. Thus while the view of language as an agglomerate of
signs can be reconciled with the idea of individual free choice exercised at
an assumed instant of origin, the alternative view of language as a com-
plex, interwoven system of signs is more closely tied to a historical inheri-
tance of language from the past by a community of speakers who cannot
change it by force of individual will alone.
In sum, it matters little if the link between the signified and the signifier
is motivated (by natural resemblance, by symbolism of sound, by divine
decree) or arbitrary; Adam could have conceivably named the animals ar-
bitrarily, and the narrative of genesis can be reconciled with the notion that
the signifier has autonomy with regard to the signified. Any thesis of lin-
guistic arbitrariness which loses sight of how social practice and historical
sedimentation give language its weight is likely to recover the story of ori-
gins if it is to make sense of the apparent paradox that signifiers stick to our
lives with the force and tenacity that they do. Ultimately, it is the figure of
an absolute master guiding his disciples in a school of thought (with the
assumed backing of the founding father) that assumes, practically, this tra-
ditional role and privilege of a sovereign subject from the creationist myths.
Since the exclusive focus on the signifier/signified distinction leads to a
disregard of the social and historical reality of language, the structuralist
school is vulnerable to the critique usually reserved for the nomenclature
view of language. Both contain and conceal an ideological thesis that lan-
guage is as fixed and immutable as the rational order of things in nature
was previously thought to be. Such views put forth a “highly ideologically
constrained rationalization of the way language is used,” where language
“plays no role in the construction of the social reality of its users” (Thibault,
1997, pp. 23–24). Transcending the ahistorical rationalism implied in these
views of the natural and social world seems a necessary step in reconnect-
ing with the concrete reality of language, and reconsidering the possibility
of social change in history.
In conclusion, consider how the question about the basic bond between
language and the social reality of its users can be made concrete by way of
the phenomenon of harmful language, like racial slurs and sexist speech.
There is a debate in the scholarship about what constitutes the most apt
strategy of responding to words that hurt: regulation, including government

censorship of derogatory expressions, or reclamation wherein the expres-
sion gets resignified within alternative practices of usage, and loses some of
its potential to induce harm. The first approach could be deemed top-down,
since the state is called upon to intervene between the citizens and is
charged with the power to prohibit linguistic expression and punish their
users. The second is a bottom-up approach, since social groups themselves
are empowered to negotiate what words signify and what effects they pro-
duce; in a felicitous case of reclamation like that of queer, an expression that
previously carried the force of exclusion and contempt shifts to a celebra-
tory term of self-affirmation within the gay group.13
Reclamation illustrates that social processes go into the making and
remaking of linguistic signification and force. This process is unsurpris-
ing if we accept, in agreement with Bourdieu and Butler, that language is
a site where social forces and struggles for domination and recognition
routinely play out. This means that language is neither an inert and neu-
tral code that could be studied with scientific objectivity nor is it a sum of
subjective expressions by individual speakers. How many speakers does
it take to resignify queer? At what point does the new signification enter
the canon of acceptable usage? Just as in the distinction between la
langue and la parole, synchrony and diachrony (discussed in more detail
in ch. 3), the lines may be hard to draw and the sources of linguistic and/
or social change hard to point out. And yet at some point a term of con-
tempt becomes also a term of affirmation (at least in some contexts and
for some speakers), and enters the linguistic code as such; the expres-
sion’s potential to quasiautomatically cause social harm gets stifled, and
the phenomenological experience associated with it becomes compli-
cated and multivalent.
The process of reclamation can be deciphered with the help of Saus-
sure’s general linguistics. In agreement with Saussure, individual con-
sciousness is powerless in the face of established social conventions; this
does not imply, however, that all ties to consciousness and subjectivity are
cut considering that linguistic signification is, in agreement with Saussure,
13
Proponents of the regulation view include Charles R. Lawrence, Mari Matsuda, Richard
Delgado, and Kimberle Williams Crenshaw (Words That Wound: Critical Race Theory,
Assaultive Speech, and the First Amendment [Boulder: Westview, 1993]). Judith Butler, drawing
on Bourdieu, J. L. Austin, and others, proposes reclamation as a more desirable alternative
(Excitable Speech. A Politics of the Performative [New York: Routledge, 1997];
“Performativity’s Social Magic,” in Bourdieu: A Critical Reader, ed. R. Shusterman [Oxford:
Wiley-Blackwell, 1999], pp. 113–128); reclamation is tied to the basic view that language is
performative, a way of doing (and redoing, and undoing . . . ) things with words.

arbitrary, that is, constituted by the many vicissitudes of social convention
(and not a capricious play of signifiers among themselves). Saussure’s gen-
eral linguistics turns out therefore to be an unexpected ally within a practi-
cal as well as scholarly interest in the mutability of language, and sheds
light on efforts to resignify language and reclaim the terms of (historically
sedimented) domination. It helps to decipher the otherwise perplexing fact
that language is resistant to individual intent and can be worked through to
surrender some of its pernicious force, that it carries the weight of histori-
cally sedimented norms and enables novel expressions. There is a message
of liberation in Saussure’s linguistics, for it suggests that signification
emerges within the ambiguous terrain where the speakers borrow the ex-
isting signifying resources but the borrowing hands also leave their mark
on the product, and what they return differs somewhat from what had been
received. Perhaps the functioning of the cardinal principles from general
linguistics: arbitrariness, mutability, can be best studied within social
movements that make a renewed claim to linguistic meaning.

2 Phonocentrism: Derrida
D
errida’s reading of the Course in Of Grammatology has fa-
mously fashioned Saussure as being torn between an emphasis
on language as a system of differences and an attachment to the
primacy of the phone (sound); while the former emphasis purges the
sign system of any positive content and figures signification as a purely
relational field, the latter steers the whole project of general linguistics
in the direction of the metaphysics of presence which posits speech as a
site of direct and natural access to signification (and of which writing
would be but a secondary and derivative form). In this chapter, I will
revisit relevant sections from the Course (dealing, first, with “natural”
expressions like onomatopoeias and interjections in relation to linguistic
arbitrariness, and, second, with the relation between speech and writ-
ing) in light of Derrida’s influential reading in Of Grammatology and
Glas, situate them in light of the relevant materials from the Nachlass,
and ultimately complicate the presumed primacy of sound and/or speech
as a site of unmediated signifying presence within Saussure’s general
linguistics. I will make the case that Saussure’s general linguistics re-
jects the possibility of authentic or “pure” onomatopoeic expressions,
considering that the latter are subjected to linguistic rules in the same
measure as any other words. This resonates with Saussure’s understand-
ing of arbitrariness as a process of intralinguistic motivation whereby
any individual sign is informed by relations to other signs within the
system, as well as temporally sedimented social conventions (discussed
in ch. 1). In agreement with Derrida, the entrainment of the sign by the
language system has always already begun, and the contamination of
the language system by forces deemed “external” to it is a regular and
normal state.
Natural symbolism of sound
In the conclusion of part I, section 2 of the Course, the editors handle pos-
sible objections to the thesis of linguistic arbitrariness. The objections con-
cern the case of onomatopoeias and interjections, both arguably of a natu-
ral origin, imitating the sound associated with a given action or event, such
as the expression for an animal cry (e.g., cock-a-doodle-doo) or bodily
sensation (ouch). If both types of linguistic expression could be shown to
derive directly from natural sounds, they would put pressure on the gen-
eral validity of the principle of arbitrariness. Here is how the case is made
in the Course:
Onomatopoeic words might be held to show that the choice of signifier

(signifiant) is not always arbitrary. But such words are never organic ele-
ments of the linguistic system. Moreover they are far fewer than is generally
believed. French words like fouet (“whip”) or glas (“knell”) may strike the
ear as having a certain suggestive sonority. But to see that this is in no way
intrinsic to the words themselves, it suffices to look at their Latin origins.
Fouet comes from Latin fagus (“beech tree”) and glas from Latin classi-
cum (“trumpet call”). The suggestive quality of the modern pronunciation
of these words is a fortuitous result of phonetic evolution.
As for genuine onomatopoeia (e.g., French glou-glou (“gurgle”), tic-tac

“ticking (of a clock)”), not only is it rare but its use is already to a certain
extent arbitrary. For onomatopoeia is only the approximate imitation, al-
ready partly conventionalized, of certain sounds . . . once introduced into
the language, onomatopoeic words are more or less entrained (entrainée)
in the same phonetic and morphological evolution as other words. The
French word pigeon (“pigeon”) comes from the Vulgar Latin pipio, itself of
onomatopoeic origin, which clearly shows that onomatopoeic words them-
selves may lose their original character and take on that of a linguistic sign
in general, which is unmotivated. (Saussure, 1986, pp. [101–102], 69, trans-
lation revised; emphasis added throughout other than French and Latin
terms)
These two paragraphs offer a prodigious case study in intellectual am-

bivalence. While their overt purpose is to diffuse the worry that onomato-
poeic expressions endanger the thesis of linguistic arbitrariness, they seek
to accomplish this aim by either chasing onomatopoeias out of the realm
of language altogether via a mythical reference to their original character

(in which case they would not function as linguistic signs and be con-
strained by arbitrariness anyway), or by loosening the hold of arbitrariness
on onomatopoeias construed as linguistic signs—but only partly or to a
certain extent (in which case it would be perennially up for debate whether
or not they form an “organic” part of the language system). According to
the editorial presentation, genuine onomatopoeias derive directly from a
mimesis of natural sounds (gurgling, ticking), and are subsequently intro-
duced and entrained by the language system. Their original character is
then purely natural—belonging to language somewhat contrary to their
nature.
According to the source materials, onomatopoeias exist but are of lim-
ited importance. We may see them where there are none (e.g., il pleut (it
rains), from Latin pluit, previously plovit), and thus are ultimately hard
pressed to tell onomatopoeias apart from other linguistic expressions. Fur-
thermore, even onomatopoeias likes tic-tac or glou-glou are “so drowned
in the mass that they are subject to the same rules (passent sous le régime)
of any words” (Engler, 1989, [1152–1156], D 190, p. 156). They are there-
fore always already embedded within language, in agreement with the
notion that a linguistic sign signifies in virtue of being part of the language
system. The notion of a purely natural linguistic sign, derived directly
from a unit of sound, is therefore unthinkable; signs are not first intro-
duced into language as into a container and then compared and contrasted
with other signs; even the most seemingly primitive ones like grunts of
pleasure or pain signify thanks to their lateral differences within a system
of relations, however simple or complex. An analogous point can be made
about exclamations as about onomatopoeic expressions. Even though the
former may seem to be directly dictated by a natural bodily impulse, they
turn out to be conventional when one regards differences between lan-
guages (ibid.). The sources do not therefore allow exceptions to the general
thesis of linguistic arbitrariness; all signs without exception are shaped by
the workings of the system they are part of. The editorial exposition, how-
ever, muddies the claim in a number of ways.
First, Sechehaye chooses to qualify the (perhaps too radical) idea that
onomatopoeic expressions are subject to the same rules as all the other
words in a given language. While the sources unequivocally make this
point, Sechehaye introduces a hesitant tone by stating that onomatopoeias
are “more or less” subject to the phonetic, morphological, and so on, rules
governing language (Saussure, 1986, p. [104], 102; the English translation
dropped the qualifier); he adds on the margin of the Collation that they are
subject to a “special regimen” (Engler, 1989, [1156], p. 156). This editorial
phonocentrism 73
insertion casts a shadow of doubt on the general validity of the thesis that
language as a whole belongs to the order of signs and social conventions
(and does not have a natural origin and does not simply mirror the natural
world in any of its elements). This tone of doubt is amplified by another
editorial insertion relative to the onomatopoeias, namely that “such words
are never organic elements of a linguistic system” (Saussure, 1986, p. [103],
101; preserved in the English translation). There is no single manuscript
basis for this bizarre insertion (Godel, 1957, p. 126; Engler, 1989, [1148],
p. 156); yet its inclusion serves to put pressure on the primacy of the lan-
guage system and maintain the editorial attachment to an isolated linguis-
tic sign. The editorial presentation thus renders the overall argument con-
fused and inconsistent, covertly admitting the overtly rejected claim of
naturalism in language.
This does not prevent it from counting as Saussure’s official position, sub-
sequently cited and critiqued by the editor himself. In his 1932 oeuvre, Lin-
guistique générale et linguistique francaise, Bally notes that Saussure “de-
liberately ignored” (l’ignorait volontier) the symbolic or signifying dimension
of sound, “since his attention was solely focused on interjections (exclama-
tions and onomatopoeias) which he considered—possibly wrongly—as for-
eign to the language system” (Bally, 1965a, p. 129, emphasis added). The line
about onomatopoeias never being “organic elements” of a linguistic system
thus changes duty from editorial insertion to Saussure’s official position,
cited—and critiqued—by the editor himself. It is yet another example of the
to and fro movement between the Course and the literature tightly surround-
ing it, the editors first inserting what they subsequently inferred from the
book. Having become established as a successor to the post and linguist in
his own right, Bally can thus pursue, in 1932, in reference to Saussure’s tic-
tac example from the lecture notes that “without a doubt, an interjection is
situated on the margins of language when it is a pure and simple reproduc-
tion of a reflex or a noise,” but belongs to language as soon as it is used in a
complex phrase (ibid., p. 129). To stick to Saussure’s example, “I heard the
tic-tac of the clock” would be an example of such a complex linguistic phrase,
which crosses the line between nature and convention. Bally can thus make
the claim against Saussure that “interjections are phrases of a kind (à un
membre) with all the others” (ibid.), and therefore organic elements of lan-
guage after all.
The entire view of language becomes more organic, so to speak, or natu-
ral sounding in the process. Bally lists the many lexical forms in French of
sound-imitative origin (cliqueter, cliquetis, craquer, craquement, croasser,
miauler), and observes that speaking subjects may be unaware of the historic

origins for the presently felt expressivity of the sound, such that all that re-
mains is a strong intuition of a relation between the phonic form and the
impression yielded by its meaning (ibid., p. 130). This intuition would ex-
plain why “certain sounds and groups of sound produce sensory representa-
tions which can coincide with the effects produced by the meaning (signifi-
cation): the occlusive K underlies the impression of breakage or ‘cassure’
(casser, craquer, croquer, crever), P highlights that of explosion (péter, pé-
tiller)” (ibid.).
The Bally of Linguistique générale et linguistique francaise thus devel-
ops the idea, ventriloquized in the Course, now signed with his own name,
that signs are not arbitrary in any radical sense, since sound plays a natu-
rally motivating role within language as a whole, and not only in the case
of expressions of documented sound-imitative origin. The latter are there-
fore “organic elements” of language since language deploys “the symbolic
or signifying dimension of sound” (ignored by Saussure) across the board.
While Saussure would have admitted, as per the editorial exposition from
the Course, only a narrow class of onomatopoeias and interjections into
this realm of natural symbolism and suggestive sonority, Bally opens the
door wide to welcome all phonic forms. He would have thus developed and
expanded the predecessor’s overly narrow view. This original narrow view
is, however, the successor’s attributed fabrication. The case that expressions
like tic-tac can be made to fit into linguistic phrases need not be made,
since the stated view is that the former are subject to the linguistic rules like
any other. An expression like tic-tac does not therefore function as “authen-
tic onomatopoeia” (an expression found in the Course but not in the sources),
an instance of direct mimicry of natural sound to then be subjected to the
rules of the language system and embedded in linguistic phrases. This
work has already been done, the sign signifies in virtue of being constrained
by the language system. Rather than document authentic onomatopoeias
and derive sense from sound, the lecture notes set out to expose the illusion
that linguistic expressions derive signification from mimicry of natural
sounds (or that there are authentic onomatopoeias). Rather than try to distill
the sensory representations highlighting the sense “naturally” evoked by a
unit of sound: the Ks creating an impression and idea of “breakage” as long
as the signs sound something like cassure, casser, craquer, croquer, or
crever, the Ps creating an impression and an idea of “explosion” as long as
the signs sound something like péter or pétiller, the lecture notes empha-
size interlinguistic differences to conclusively untie language from any pre-
conceived notion of naturalness. Casser and péter may seem to have a nat-
ural motivating resource of signification-in-sound only as long as they are
phonocentrism 75
regarded from an unexamined, naïve, and narrow viewpoint of a monolin-
gual speaker with universalizing propensities, whose langue nationale
would feel as intuitively and irremediably expressive as God’s own lan-
guage must have felt before the tower of Babel rose to cast a shadow on such
notions. And yet these notions steer the concluding part I. 1., section 2 from
the Course in contrary directions, of radical arbitrariness of the sign on the
one hand, and of natural expressivity of sound (in some cases at least) on
the other. The latter direction, written into the Course and subsequently
developed under Bally’s own name, obscures the novelty of Saussure’s
thesis that language is not of natural origin, and that it is governed by its
own radical rules.
As it is, Saussure, the presumed author of the Course, defends a tradi-
tional notion of a sign spontaneously secreting signification, with no nec-
essary detour or delay which standing in a nexus of relations to other
signs within a language system would produce. Derrida’s inspired read-
ing of the two paragraphs on onomatopoeias can therefore, rightly,
ponder the possibility of knowledge of “authentic onomatopoeias,” con-
tingent as it would be on grasping the original instant predating their
having become drawn into language (Derrida, 1986, p. 92). This notion
of a pure origin of onomatopoeias would have transpired also in the ety-
mological claim that glas and fouet are not authentic onomatopoeias de-
spite their suggestive sonority since they result fortuitously from pho-
netic evolution. “Will one trust etymology and even a narrow concept of
etymology—historicist and unilinear, to analyze the functioning of a
linguistic signifier and recognize its ‘organic’ belonging to the ‘linguistic
system’? Does an element depart from language [langue] when it does
not conform to its presumed semantic origin?” (ibid.). Rather than de-
marcate the border between the inside of the system and its exterior, we
need to recognize that the “the process of being ‘drawn’ [entrainement]
has always already begun, which is neither an accident nor something
outside the system” (ibid., p. 93).
Derrida then wonders why Saussure would have chosen glas and
fouet as examples of expressions of (presumed) imitative origin—is
glas/knell supposed to imitate the sound of a bell, fouet/whip the sound
of lashing? Nobody can reasonably claim that we are dealing with gen-
uine onomatopoeias in this case, and so the examples are poorly chosen
from an empirical standpoint—unless we follow the lead of the text
itself, and consider that the knell may strike the ear with suggestive so-
nority like a whip. In this case, we need to acknowledge the existence of
motivating forces running throughout language as a whole—the forces

to which the ghostwriters of the Course fell prey as they made claims
about extralinguistic sounds. Before considering this possibility further,
the record needs to be set straight: the two examples of onomatopoeias
are editorial insertions with no basis in the source materials. The student
notes have a pluit (rain) example: “One sometimes says that pluit (‘rain’)
represents the sound of rain, but when one traces it back, one sees that
that’s not the case at all: previously plovit, etc.” (Engler, 1989, [1147–
1150], III C 242, p. 156).
In other words, the example found in the source materials features a
more common example of an expression commonly held to be mimetic in
the French, and exposes the illusory nature of the belief that pluit sounds
like the rain itself (to the French ear) by exposing its constitutive link to
another signifying term (the Latin plovit). Once exposed, the interlinguis-
tic relation between the signifying elements (pluit–plovit) forecloses the
possibility of the signified being rendered in the signifying by direct imi-
tation of its natural sound.
The editorial omission of the more common example in the French and
their insertion of the two unusual examples is puzzling. Were the editorial
ears struck more vividly by knells and whips then by the sound of rain?
Wouldn’t their idiosyncratic suggestibility to sound be at odds with the
methods and purposes of an empirical science program? Or was the sug-
gestibility associatively motivated by the intersections between the strike/
whip/knell signifiers, and thus already entrained by language, making it
impossible to derive any of the individual signifiers directly from a sound
source? Wouldn’t this intralinguistic motivation bring the sound-imitative
expressions into the fold of language, perpetually barring access to their
prelinguistic origin? It may be, however, that glas and fouet deploy a dif-
ferent kind of suggestiveness in the Course; they suggest that if even such
far-fetched cases could be reasonably construed as imitations of sound
within general linguistics, then language as a whole can be so construed.
This possibility only lends credence to Bally’s overt thesis of natural
symbolism of sound as later developed in his Linguistique générale et
linguistique francaise, and fits seamlessly into its linguistic program. The
glas example ghostwritten into the Course would thus better serve a nat-
uralist conception of language dear to Bally than Saussure’s nonnaturalist
understanding.
Derrida’s reading of the two paragraphs devoted to sound-imitative ex-
pressions proceeds to tie the notion of necessary entrainment of sound-
imitative expressions within language with its natural corollary: the notion
that the language system itself cannot maintain strict autonomy with
phonocentrism 77
regard to motivating forces (whether they are sounds found in nature, pho-
netic evolution deemed merely fortuitous by the Course, or intertextual
relations). The autonomy of the language system is as compromised as is
the authenticity of onomatopoeic expressions, the boundary between them
fluid, one impinging upon the other. The principle of arbitrariness itself
cannot therefore be safeguarded in its presumed radicality—arbitrariness
is motivated, or contaminated by forces deemed “external” to language.
Derrida writes:
what if mimesis so arranged it that language’s internal system did not exist,
or that it is never used, or at least that it is used only by contaminating it and
that this contamination is inevitable, hence regular and “normal,” makes up
a part of the system and its function, makes up a part of it, that is, also,
makes of it, which is the whole, a part of a whole that is greater than it. (Der-
rida, 1986, p. 94)
Importantly, these possibilities of the outside contaminating the inside

are also outlined within Saussure’s source materials. First, Saussure’s lec-
ture notes do not entertain the possibility of authentic or pure onomato-
poeic expressions, and emphasize that the latter are subjected to linguistic
rules in the same measure as any other words. This emphasis echoes Der-
rida’s notion that entrainment has always already begun. Second, as dis-
cussed in ch. 1, the thesis of arbitrariness is consistent, even inseparable,
from intralinguistic motivation—the latter is described in terms of rela-
tive arbitrariness and subsumed under the value-relations which shift the
emphasis away from an individual sign and its signification onto the pro-
cesses within the language system. Finally, these systemic constraints are
coupled with social and historical forces which shape and sustain but can
also unsettle the established habitus of linguistic usage. The language
system is therefore worked through both from “within” and from “with-
out.” In agreement with Derrida, contamination is inevitable—it is both
regular and “normal.”1
1
I am therefore in agreement with Culler when he argues, also in reference to the student lecture
notes, that there is no radical difference between Saussure’s and Derrida’s views of signification
(Culler, 2007, pp. 134–135). My own analysis in this chapter helps to appreciate why Derrida—
and other readers—would have thought it justified to charge the Course with privileging
presence, both in the case of the “sound-based” expressions, and of the relation between speaking
and writing. My emphasis on the importance of social and temporal factors in Saussure’s
conception of language lends additional support to Derrida’s claim that not only is the sign
entrained by the language system, but also the language system itself is contaminated by its
ambiguous “outside.”

Metaphysics of presence, and arche-writing
As a result of the editorial handling of source materials, the Saussure of

the Course emerges as the metaphysical traditionalist who maintained the
received notion of a sign considered as a positive unity of sound and sense.
This positivity would have been, pace Derrida, grounded in the traditional
metaphysical privilege attached to the phone, the spoken word or “vocal
image” understood as a site of signifying presence, deploying the “natural
bond” between sense and sound (Derrida, 1998, p. 35). Saussure’s reflec-
tions on speech and writing would then rejoin the traditional metaphysical
notion of a positive unity of signans and signatum, dating back to the Stoa
and medieval logic, where signification is tied to the metaphysical notion
of presence (Agamben, 1993, p. 155). This positivity would be best exem-
plified by a presumed sensory and intuitive plenitude of a speaking subject
who hears himself speak (where entendre connotes both the hearing and
the understanding of intended meaning); it would be endangered by a
transmission within the lifeless medium of a written text:
The epoch of the logos . . . debases writing considered as mediation of medi-

ation and as a fall into the exteriority of meaning. To this epoch belongs the
difference between signified and signifier . . . The difference between signi-
fied and signifier belongs in a profound and implicit way to the totality of the
great epoch covered by the history of metaphysics . . . This appurtenance is
essential and irreducible; one cannot retain the convenience or the “scientific
truth” of the Stoic and later medieval oppositions between signans and sig-
natum without also bringing with it all its metaphysico-theological roots.
(Derrida, 1998, pp. 12–13)
The adopted opposition between the signans and the signatum would
safeguard a problematic notion of a primary or transcendental signified,
that is “a concept simply present for thought, independent of a relationship
to language, that is of a relationship to a system of signifiers” (Derrida,
1981, p. 19). Despite his stated opposition to the nomenclature view of lan-
guage, Saussure would thus have reintroduced this traditional view through
the backdoor by retaining a positive notion of the sign which can express
an idea or concept directly via the medium of speech-sound, and only in-
directly and secondarily via its representation in writing. Even though
Saussure’s linguistics had been received within structuralism as a scien-
tific program that deliberately breaks away with a metaphysical heritage,
its continued alliance with a metaphysical epoch whose “concepts and
phonocentrism 79
fundamental presuppositions are quite precisely discernible from Plato to
Husserl, passing through Aristotle, Rousseau, Hegel, etc.” (ibid., p. 22)
would thus be exposed within a poststructuralist reception.
In fact, the metaphysical and scientific strands of Saussure’s linguistics
belong together according to Derrida. A metaphysical privilege attached to
the sign as a positive datum enables a positive science of language with a
determinate object of study: la langue. In “The Linguistic Circle of Geneva”
from Margins of Philosophy (originally published in December 1967), Der-
rida makes the case that Saussure’s linguistics provides an exemplary or
titular model for a science in a modern sense, that is, a human science with
aspirations of natural scienticity like Levi-Strauss’s structural anthropology
or Lacan’s psychoanalysis (Derrida, 2009, p. 139). Such scienticity can be
attained in the human science by satisfying two essential conditions. The
first condition is to define the object, the method, the proper field of study.
The second is to belong to “a determined and finite system of conceptual
possibilities, to a common language, to a reserve of oppositions of signs
(signifiers/concepts), which first of all is none other than the most ancient
fund of Western metaphysics” (Derrida, 2009, p. 140). General linguistics
would be counted as an exemplary science in the modern or structuralist
sense because its object and method fit into the received metaphysical
worldview where being is thematized as substance or presence, a positive
unity of sound and sense within any individual signifier.
This scientifico-metaphysical construal of Saussure’s linguistics un-
critically assumes the validity of the Course as official doctrine. It as-
sumes that the sign is only optionally inserted into the “plexus of eter-
nally negative differences” and can be celebrated as a positive phonological
plenum. Only on that condition can the science of signs be pushed back
into Western metaphysics (Agamben, 1993, p. 155). Similarly, the notion
that linguistics has an identifiable and autonomous object made up of
such positive signata unquestionably assumes the validity of editorial
presentation. However, source materials of Saussure’s linguistics tell a
more complex story. They enact a deconstruction of the metaphysical
worldview via a deliberate confrontation with the dominant philosophical
view of language as nomenclature and the concomitant view of being as
substance (discussed in ch. 1). Furthermore, the discussion of the sign in
the student lecture notes belongs to a much more elaborate argument
which steers away from an initially entertained signans/signatum distinc-
tion toward a systemic or web-like notion of language embedded in the
historically sedimented social practice of usage. Source materials under-
mine therefore a possibility of isolating a “determinate and finite system

of conceptual possibilities” out of the more complex reality of language.
Rather than validate metaphysics and ground modern science, source ma-
terials openly contest the validity of metaphysical and scientific methods
and terms in language study.
Derrida’s analysis of Saussure’s conception of writing in ch. VI “Repre-
sentation of a language by writing” from the Course targets this presumed
metaphysical privileging of pure presence directly. By denigrating writing,
Saussure would have sought to isolate a solid block of sound-cum-sense, and
keep the textual threats of loss (of the author and the referent), of spacing
(between and within the lines of text), of nonpresence (the silences between
sounds) at bay. Considering the influence this reception has played in craft-
ing the notion of Saussurean linguistics, it is useful to consider the validity
of the edited version in light of the source materials, and also to articulate
the role and function of Saussure’s comments about writing within his cri-
tique of the dominant disciplines and institutions dealing with language.
First, it should be noted that Saussure’s comments about writing are
part of the discussion relative to securing another approach to language
than the one adopted by the discipline of philology, with its exclusive focus
on the written document and its disregard of contemporary spoken lan-
guages. This leads to a purely historical conception of language without
raising the question of what language itself is. The rejection is then not of
writing per se, perceived as an external threat to the purity of speech, but
of exclusive focus on written documents and adoption of the “theoretically
fragile” reconstructive techniques combined with unguided fact gathering.
While Derrida’s focus in this discussion is object bound, or guided by on-
tological distinctions between the inner system of speech and the external
horizon of writing, Saussure’s focus is on the alternative methodological
approach in linguistics in comparison to the dominant historical study; his
negative claims about “writing” are part of a critique of the excesses com-
mitted by the discipline of writing, which values received grammatical
standards over and above the actual practices and conventions of usage.
There is also a concern with the excesses committed by the institutions of
writing, such as the Académie Française, whose imposition of a single lit-
erary standard and the policing of the French language from above go in
the way of dialectal diversity of the French language/s. In sum, writing
does not figure as a general category within linguistics (comparable, e.g.,
to the distinction between synchrony and diachrony), and its discussion is
tied to historically and geographically specific instances of language study
and institutional control. An exclusively metaphysical reception of Saus-
sure’s pronouncements loses sight of their purpose and context.
phonocentrism 81
Second, the citations accumulated by Derrida from ch. VI of the intro-
duction to the Course, which make a case for a suppression of writing in
favor of pure speech, sit uneasily with the source materials from the stu-
dent lectures notes. Notably, the citations from the Course deemed the
more “colorful” by Derrida, as well as those that are repeated like a mantra
in his text, are of editorial making. I’ll discuss them next.
Derrida writes: “[H]as it ever been doubted that writing was the cloth-
ing of speech? For Saussure it is even a garment of perversion and de-
bauchery, a dress of corruption and disguise, a festival mask that must be
exorcised” (Derrida, 1998, p. 35). As evidence, he cites from the Course:
“Writing veils the appearance of language; it is not a guise for language
but disguise (non pas un vêtement mais un tra/vestissement)” (Saussure,
1986, p. [52], 29; Derrida, 1998, p. 35). Yet there is no mention of such a
transvestite disguise in the sources (Engler, 1989, [572], p. 85). Derrida
goes on to expose the “historico-metaphysical presuppositions” within
Saussure’s assumed natural bond between the medium of the voice and
sense, implied in such a privileging of speech and ejection of writing as an
external clothing. Derrida repeatedly refers to this naturalness of the
sound-sense unit as presumed by Saussure, and cites from the Course:
“the superficial bond of writing is much easier to grasp than the natural
bond, the only true bond, the bond of sound” (Saussure, 1986, p. [46], 26;
Derrida, 1998, pp. 35–36). Yet this whole line is an editorial insertion
without support in the sources (Engler, 1989, [480], p. 73). Derrida repeat-
edly cites Saussure’s apparent invocation of the “tyranny of writing” (Der-
rida, 1998, pp. 38, 41; Saussure, 1986, p. [53], 31)—but this negatively
charged notion is of editorial making, the manuscript sources speaking
more neutrally of an “influence” of writing on language (Engler, 1989,
[607] D 50, III C 86, p. 88).
According to Derrida, Saussure’s commitment to a natural and sponta-
neous life of language leads him to view writing, together with artificial
languages like Esperanto, as a monstrosity. Derrida writes:
The perversion of artifice engenders monsters. Writing, like all artificial

languages one would wish to fix and remove from the living history of the
natural language, participates in the monstrosity. It is a deviation from
nature. The characteristic of the Leibnizian type and Esperanto would be
here in the same position. Saussure’s irritation with such possibilities drives
him to pedestrian comparisons: “A man proposing a fixed language that
posterity would have to accept for what it is would be like a hen hatching a
duck’s egg.” (Derrida, 1998, p. 38; Saussure, 1986, p. [111], 76)

According to Derrida, Saussure abhors artificial languages and writing
systems for ordinary languages alike because their fixity arrests the lan-
guage’s living flow and results in a deviation from language’s innermost
nature; this deviation would be dramatized by means of a family scene
involving one species caring for the offspring of another.
The citation from the Course is an editorial rendition of the line that
“language (la langue) is a bit like a duck hatched by a chicken” (Engler,
1989, [1275], p. 170). This line belongs to a broader discussion of the idea
that any language, including an artificial language like Esperanto, is sub-
ject to the “fatal law” (la loi fatale) of being at the mercy of social conven-
tion and transmission (ibid., [1273] III C, p. 170). Any language escapes by
necessity the control of any individual subject, such as the creator of Espe-
ranto himself, and gets integrated into “semiological life” (la vie sémi-
ologique) whose laws are far removed from the laws of creation or consti-
tution (ibid., [1274] II R, p. 170). The semiological life of language is
characterized by an ongoing circulation of signs within a speech commu-
nity, and their transmission and reception within successive generations
(ibid., [1275] II R, p. 170). Thus, the relative measure of success of an arti-
ficial language like Esperanto lies precisely its passage from the law of
creation/constitution to the law of circulation and transmission; this pas-
sage implies a loosening of ties to the individual will of its “creator,” and
suggests a condition wherein language floats together with the many suc-
cessive, crisscrossing, and receding waves of socio-semiological life.
Hence the analogy: “language (la langue) is a bit like a duck hatched by
a chicken” (ibid., [1275], p. 170). This cross-species hatching of an egg il-
lustrates that language is an object of forces other than the ones that led to
its creation; once brought into the world, the sign is open to adoption by
many guardians, and can become nested in various environments. Espe-
ranto, just like a freshly laid egg, is subject to the fatal law of adoption by
the many, with a number of potential purposes—after all, an egg can
become a symbol of life like a brightly colored Easter ornament; it can be
hatched and cared for by a different animal from the one that laid it; it can
be abandoned and remain unhatched.
The scenario envisaged within the lecture notes is therefore markedly
different from the one contained in the Course and later cited by Derrida.
According to the lecture notes, adoption and alterity belong to the very fate
of language, with the image of cross-species hatching illustrating such
necessary subjection to alien forces. According to Derrida citing from the
Course, this same image serves to illustrate the “perversion of artifice,” the
“deviation from nature” implied in the creation of artificial languages,
phonocentrism 83
removed as they are from “the living history of the natural language.” The
presumed monstrosity of artificial languages, and writing, would be tied to
“the wish to fix” this history once and for all (Derrida, 1998, p. 38). Yet
this analogy falls apart when one considers that according to the sources
any language (and this includes writing) resists whatever desire of fixed
and immutable order a presumed creator would have had when putting it
together; this necessary failure of individual control is a measure of lan-
guage’s success. What appears as an aberration from the standpoint of the
concrete reality of language is the presumed voluntary control by an indi-
vidual language-maker; what is monstrous is a desire for individual do-
minion over language—not the fact that individuals adopt a language they
have not themselves “laid.” The analogy between writing and unnatural-
ness falls apart because the image invoked illustrates an opposite kind of
scenario from the one implied in Derrida’s reading. The implied scenario
is of a presumably monstrous setup involving adoptive parenting arrange-
ments whose horror can be transferred onto other “deviations from
nature”—namely, the artifice of a living language being fixed by man. Yet
adoptive parenting can hardly accommodate the ideologically construed
monstrosity of a nurture breaking away from nature, of artificial fixity
projected onto a natural flow; instead, the kinship structure of adoptive
parenthood invites the thought of new possibilities opening up in an en-
counter with the other. That is why it is an apt image to illustrate the un-
predictable fate of signs in social life, of whatever making, in Saussure’s
own analogy. Used in the service of a very different from the intended
scenario, the image becomes a hollow appeal to the reader’s presumed
sense of horror when exposed to nontraditional structures of kinship. It is
reduced to a simple deterrent, with its rhetorical force shifting from an
imaged analogy between language and alterity to a cautionary tale about
monsters.

3 La langue and la parole, synchrony
and diachrony
I
n this chapter, I propose to examine the remaining elements of the “Sau-
ssurean doctrine”: the pairings, la langue and la parole, synchrony and
diachrony, which are construed as oppositional and violent hierarchies
within the dominant structuralist reception of the Course. I will discuss
how these distinctions are established in the Course in light of the source
materials. I will make the case that while the editorial presentation is in
agreement with the source material on a number of points, notably in that it
ties these distinctions to the duality of perspectives or viewpoints onto lan-
guage, it tends to overstate the separation between the distinguished terms.
Despite the acknowledgment of the importance of perspective for estab-
lishing the distinction between synchrony and diachrony, and la langue and
la parole, the latter are also presented as a preexistent, objective factum,
hence seemingly immune to perspectival configuration. I will argue that
Saussure’s stated difficulty of finding the right orientation into the hetero-
geneous field of language, which echoes Hegel’s stated difficulty of begin-
ning in philosophy, gets occluded in the process. The methodological com-
plexity involved in the study of language is therefore ironed out in favor of
a more manageable scientific program—at the risk of neglecting the philo-
sophical dimension of Saussure’s reflections in general linguistics.
The distinction between synchrony and diachrony is established in the
Course in part I, ch. III. This distinction is derived from a duality of per-
spectives or points of view that a scholar, as well as an untrained language
user, can adopt in relation to language, regarding it either as a set of rela-
tively stable and enduring states or temporal successions. The synchronic
viewpoint picks out language states [états de langue], that is, a time span
of relative stability in the life of language. This viewpoint focuses on the
relations that bind the coexisting terms into a system, as they are collec-
tively available to its speakers (Saussure, 1986, pp. [125], 87; [128], 89–90;
[138], 96–98). The language state is admittedly an approximate notion,
considering that the linguist needs to exclude from the timeslice the lin-
guistic changes deemed of little importance, in agreement with “the usual
simplification of data” (Saussure, 1986, p. [143], 100). It should therefore
not be construed as a preexistent factum but rather a methodologically ex-
pedient abstraction. The diachronic viewpoint, on the other hand, picks out
an evolutionary phase in the life of language, and studies relations between
successive terms that evade collective grasp and do not form a system
(ibid., p. [140], 98). Depending on the prevalence of drastic change or rela-
tive stagnation in a language, either a diachronic or a synchronic viewpoint
will be more appropriate (ibid., p. [142], 99). In agreement with the source
materials, the editors acknowledge that the distinction between viewpoints
is reflected within the inner duality of a science like linguistics, which
must consider now the axis of temporal successions, now the axis of the
simultaneities in its study (ibid., p. [115], 80). These two opposing and
crossing orders of phenomena relate to one and the same object (ibid.,
p. [116], 81), as it is considered from one angle or the other. This perspec-
tival distinction is illustrated by a comparison between a transversal and a
longitudinal cut of the stem of a plant, which reveals two possible perspec-
tives on the arrangement of the fibers within the plant (I revisit this exam-
ple in reference to the source materials below). The second illustration
offers a more famed analogy between language and the game of chess. The
editorial presentation of this analogy overstates the opposition between
contemporary states and evolving conditions, and attributes to Saussure a
view he overtly critiqued in the process, as discussed next.
Double essence of language
According to a classic analogy, language is like a game of chess. In any game,

one notes a distinction between, one the one hand, a contemporary arrange-
ment of the differentially powered pieces on the board, and, on the other hand,
the succession of individual chess moves as the game unfolds (Saussure, 1986,
pp. [125–127], 87–89). These two are distinct insofar as a spectator who has
followed the game from the start has no advantage over someone who hap-
pens to check on it at any particular moment (Saussure, 1986, p. [127], 88).
The contemporary arrangement on the board cannot be described in terms of
what happened previously, even in the most recent past (ibid.).

Applied to language study, the distinction between contemporary ar-
rangements and successive chess moves is used to drive a wedge between
synchrony and diachrony. One reads in the Course that the chess analogy
“confirms the radical distinction between the diachronic and the syn-
chronic” (ibid.—an editorial insertion; see: Engler, 1989, [1485], p. 197).
This last phrase resonates with other similarly dichotomous expressions;
one notes for example a reference to a “radical antinomy” (l’antinomie
radicale) (Saussure, 1986, p. [129], 90) between the synchronic and dia-
chronic facts in the Course (an editorial insertion; see Engler, 1989, [1512],
p. 201), while the lecture notes speak of difference and irreducibility of
static and evolutive notions; one finds a postulate of a “difference in nature”
(différence de nature) between successive terms and coexisting terms
(Saussure, 1986, p. [129], 90), while the lecture notes convey a difference
between the two approaches, but the latter is not construed as a natural fact
(Engler, 1989, [1451], p. 193). While one reads that “we must put each [lin-
guistic] fact in its own class and not confuse the two [synchronic and dia-
chronic] methods” in the Course (Saussure, 1986, p. [140], 98), according
to Degailler’s and Constantin’s lecture notes, these distinctions are made
on a theoretical plane and difficult to observe in practice (Engler, 1989,
[1658], p. 226). Similarly, in the manuscript writings, Saussure notes that
if we accept the analogy between any contemporary position in a chess
game and the language system (la langue), we must then raise the question
of how language can be historical—considering that, construed as a chess
position only, it “seems much better suited to abstract speculation” (Saus-
sure, 2006, p. [216], 151). In sum, the analogy between chess positions and
the language system illustrates one possible axis upon language (langage)
as a whole, it being understood that this synchronic perspective will en-
gender abstract and speculative notions if disentangled from the diachronic
one (see ch. 7, especially: “A taste for great abstractions,” for a discussion
of the editorial preference of such abstract notions).
In fact, in Saussure’s manuscript notes the game of chess analogy effec-
tively serves to underscore the inseparability of diachrony and synchrony, in
agreement with an emphasis on the dual nature of language. Commenting
on this analogy in the “Notes for an article on Whitney” (Saussure, 2006,
pp. [203–216], 140–151), Saussure observes that the game of chess analogy
has been used before him, but in a different and misleading manner:
Theorists of language before the foundation of the discipline and adepts of

linguistics since Bopp always considered the language system (la langue) to
be ONE chess POSITION of the chess game with nothing before or after it,
la langue and la parole, synchrony and diachrony 87

and have wondered about the respective value, power of the pieces, in this
position. (Saussure, 2006, p. [207], 143)
In other words, linguists like Bopp mistakenly thought of language as

nothing but a rule governed system, thus failing to give justice to its tem-
poral dimension (“with nothing before or after” a given position in a given
arrangement of pieces). The notion that language should be construed as,
ideally, a static and subjectless system is pace Saussure’s manuscripts an
abstraction attributable to Bopp; it is not endorsed but dismissed.
Importantly, this abstracted view ties to Bopp’s failure to see language as
a phenomenon, in relation to the speaking subject (Saussure, 2006, pp. [129–
130], 85–86). Language (langage) as a phenomenon is indissociable from
particular language acts (actes de parole), as performed by the speakers. La
langue is the ensemble of concordant forms this phenomenon assumes in a
collectivity of individuals at a given time. However, it must not be regarded
outside of the speaking subjects—isolating la langue as a system from the
phenomenal reality was the mistake made by Bopp and his school:
The misunderstanding that the school founded by F[ranz] Bopp fell victim
of was to assign to languages (les langues) an imaginary body and existence
outside of speaking individuals. An abstraction relative to language
(langue), even if made for good reasons, is in practice susceptible of limited
application only—it is a logical device—and being at the mercy of this ab-
straction, endowed with a body, was an even greater impediment . . . (Saus-
sure, 2006, p. [129], 85)
According to Bopp’s school, Saussure continues, la langue was the nec-

essary condition of langage, while langage was an application of la langue;
la langue was (falsely) believed to have been delimited and established
once and for all. It is therefore Bopp’s but not Saussure’s stated view that
la langue is a system separable from speaking subjects and opposed to
langage.1
1
Weber’s thoughtful objection to Saussure can therefore be transferred to Bopp: “For, if the game
of chess involves not merely the learning of its simplest rules, governing the movement of pieces,
or even the characteristics of individual positions, but the study of entire strategies, it is first of all
because what appears in Saussure’s discussion to be a self-identical ‘state’ or position is
structurally divided, split off from itself. For the significance of any particular position on the
chessboard is inseparable from the fact that there are two positions involved, and that there is
always the question of who has the next move? This small fact, involving nothing less than chess
as a game, disrupts the entire comparison of Saussure (and perhaps more as well). For it introduces

In fact, Saussure counters, the relation between la langue and le lan-
gage, or the system and the speech acts, is reciprocal: “We can now appre-
ciate that there is a constant reciprocity and that la langue draws both its
application and its unique source in the act of language (acte de langage),
and that langage is both the application and the generator of la langue . . .
reproduction and production” (ibid.) We find therefore a circular relation
between language organized as a system and an open series of possible
speech acts. Speech acts do not only simply instantiate but also generate
stable, identifiable patterns; they are productive as well as reproductive. In
sum, the analogy between language and a game of chess can be main-
tained as long as it is allowed that while the game is to be played “by the
rules,” the rules themselves are retroactively shaped by the practice of play,
and can possibly become revised as this practice evolves.
The editors partly acknowledge this interdependency between la langue
and la parole:
Doubtless these two objects are closely connected, each depending on the
other: language (la langue) is necessary if speech (la parole) is to be intelli-
gible and produce all its effects; but speech (la parole) is necessary for the
establishment of language (la langue), and historically speech facts always
come first . . . Language and speech are then interdependent; the former is
both the instrument and the product of the latter. (Saussure, 1986, pp. [37–38],
19, translation revised)
Having acknowledged this complex interplay of speech and system, the

editors assert: “But all this does not prevent them from being two abso-
lutely distinct things” (ibid.).
This editorial invocation of a presumed order of the things themselves,
which serves as an ontological prop for the methodological distinction be-
tween the linguistics of la langue and the linguistics of la parole, is clearly
at odds with the earlier emphasis on subjective viewpoints. It constitutes,
however, another instance of editorial insertion without manuscript sup-
port (Engler, 1989, [351], p. 57), and is designed to shift the emphasis to a
presumed objectivity in language, better suited to the implied normative
view of linguistics as a factual science.
the diachronic dimension of difference and alterity into what appears to be the closed system of
the synchronic state; as part of a game, the position in chess is inherently both a response and an
anticipation, involving the calculation of strategies which are neither entirely necessary nor
entirely arbitrary” (Weber, 1976, p. 932).

The implied normative view of science partially orients the editorial
presentation of the distinctions between synchrony and diachrony, and la
langue and la parole. While the editors duly acknowledge the consider-
ations of method and perspective in general linguistics, they tend to shift
the focus to the object itself, which would make the initial subjective con-
siderations redundant. For example, in the introduction, the science of gen-
eral linguistics is initially contrasted with other existing sciences by their
respective subject-object relations:
Other sciences work with objects that are given in advance and can then be
considered from different points of view; but not in linguistics. Someone
pronounces the French word nu (“naked”). A superficial observer would be
tempted to locate the concrete linguistic object there, but upon more careful
examination, one successively finds three or four quite different things, de-
pending on whether the word is considered as a sound, as the expression of
an idea, as the equivalent of the Latin nudum, etc. Far from being the object
that precedes the point of view, it would seem that it is the point of view that
creates the object; besides nothing tells us in advance that one way of con-
sidering the fact in question takes precedence over the others or is in any
way superior to them. (Saussure, 1986, p. [23], 8, translation revised)
This consideration of a viewpoint has ample textual support in the man-

uscript sources, where it figures as a mantra-like motif, and is deemed
crucial to any reform in sciences of language; it underscores the impor-
tance of methodological considerations in linguistics where the task is not
simply to locate the true object of study, but also to pick the best vantage
point for attaining it—in acknowledgment of the difficulties of making
such a choice, and the limitations that a choice of one perspective in favor
of the others necessarily carries with it.
The methodological emphasis on perspective is related to the inherently
dual or two-faced character of any object in linguistics (this point is also
borne out by the sources). No matter what observable aspect of language
we pick up, the one facet we focus on always relates to another one, which
is temporarily hidden from view. For example, a unit of sound in speech is
tied to its articulation; the same unit functions as a sign only if it points to
an idea or thought; the individual speech act is tied to a social convention;
the state of language in the present is an inheritance from the past (ibid.,
pp. [23–24], 8–9). This unsurpassable duality at the object end is bound up
with a subjective duality of viewpoints: language can be regarded in the
present and/or in the past from respectively a synchronic and diachronic

perspective; it can be regarded from the social point of view of generally
accepted conventions and/or from an individual viewpoint of the execution
of speech acts; it can be regarded from the point of the view of its audition
and/or articulation, the inscription and/or signification, and so on. In sum,
the principle of duality is bound up with the orientation or method, for it is
to the subject that language unveils one facet while keeping the other tem-
porarily hidden from view; in agreement with the Greek aletheia, truth
involves a moment of uncovering as well as of covering up. The object is
necessarily multifaceted and perspectivally configured in and by the rela-
tion to a subjective vantage point.
In the introductory chapters from the Course, these initial consider-
ations of duality and subject-object interdependency give way to a focus on
the object-side only, and to a simplifying reduction of the initially stated
complexity of language itself. This evacuation of the subjective viewpoint
and a shift to simple objects is deemed a result of the need to study lan-
guage scientifically. The editors raise this dilemma: “if we focus on only
one side of each problem, we risk failing to perceive the dualities pointed
out above; on the other hand, if we study language (langage) from several
points of view simultaneously, the object of linguistics appears to us as a
confused mass of heterogeneous and unrelated things” (ibid., p. [24], 9,
translation revised).
In other words, picking just one facet opens up language to the claim of
the other existing sciences (like acoustics, physiology), but maintaining
the complexity leads to a methodological chaos. What is a solution to this
predicament? Picking la langue as the object proper of linguistics.
There is only one solution, in our view, to all these difficulties: we must put
both feet on the ground of the language-system (la langue) and take it as the
norm of all other manifestations of language (langage). Actually, among so
many dualities, the language-system alone seems to lend itself to an auton-
omous definition and to provide a satisfying foothold.” (ibid., p. [25], 9)
This solution dictates a sole focus on a single facet of language as a struc-

tured semiological system (la langue), decreed the object proper of linguis-
tics. The initial considerations of duality and subjectivity are dropped as
incompatible with the implied normative ideal of science where knowledge
must be objective through and through, hence dissociated from subjective
perspective and released from an unmanageable complexity. The remaining
task for the scientist is to establish and secure the bounds of this object, by
separating it off from subjective practices like speech.

In agreement with a strategic shift from subjective to purely objective
considerations, the editors emphasize that the distinction or duality be-
tween system and speech inheres in the object itself, and that such an ob-
jective or ontological segregation into two fields dictates a hierarchy be-
tween two disciplines, linguistics of la langue and linguistics of la parole:
In establishing the proper place for the science of la langue within the over-
all study of langage, I have also outlined the whole of linguistics. All other
elements of langage—those that constitute la parole—freely subordinate
themselves to the first science, and it is by virtue of this subordination that
all the parts of linguistics find their natural place. (ibid., p. [36], 18)
This objective necessity is complicated by the source materials, which

cast la langue as a platform or a viewpoint onto other elements of lan-
guage (langage) (Saussure, 1993, p. 72); “The best way of examining this
speech part [of language] is by taking la langue as our point of departure”
(ibid., p. 73). La langue is then not simply a preexistent, classifiable object
but an orientation onto the complex and multiform phenomenon of lan-
guage (langage), which it helps to classify in turn. La langue remains
therefore rooted in the subjective, partial and perspectivally oriented
access to the totality of language (langage) and does not figure as a simple
object. Differently put, it is by adopting the point of view of synchrony
onto language that its system-like quality can be borne out; but that is why
language is not a closed and autonomous system that can be studied ac-
cording to the usual scientific expectations of objectivity. Furthermore,
there is no claim about la langue occupying the whole linguistic field; the
line “I have also outlined the whole of linguistics [in establishing the
proper place for the science of la langue within the overall study of lan-
gage]” is an editorial insertion without manuscript support (Engler, 1989,
[322], p. 52). This line resonates with the famous conclusion of the Course,
according to which “the only true object of study in linguistics is the lan-
guage (la langue) considered in itself and for itself (en elle-même et pour
elle-même)” (Saussure, 1986, p. [317], 230; my translation, italics in the
text). This conclusion is said to carry the fundamental thesis (l’idée fonda-
mentale) of the Course as a whole (ibid.), and has become an oft-cited
structuralist motto, but it constitutes an editorial insertion unwarranted by
the sources (see ch. 7, ‘Famous formula’, for development). Finally, the
subordination of speech to the language system is said to be happening
“almost by itself” in both Degailler’s and Constantin’s lecture notes (ibid.,
[325], D 183, III C 275, p. 52)—it being understood that it is from the

adopted general viewpoint of la langue that such a subordination takes
place, and not as a simple fact.
By advocating a linguistic study focused on la langue as the object
proper, the editors risk turning the corresponding synchronic point of view
into an absolute and totalizing perspective yielding an objectivity in the
received scientific sense. The initial claims that there are more than one
perspective onto language (synchronic and diachronic, individual and
social); that each perspective is partial and limited in scope; that the object
is indissociable from the perspective hence not objective in a straightfor-
ward sense—all risk receding into the background. The totalizing privi-
lege assigned to la langue as the proper object of linguistic study risks a
concomitant totalization of synchronic linguistics. Once the synchronic
point of view is picked as a proper perspective in linguistics tout court, it
seems no longer partial and limited by the other perspectives; after all, a
perspective is necessarily tainted by relativity, since it implies that one
could approach an object differently and otherwise. If the synchronic point
of view becomes absolute, it need not reflect back on itself, on what gets
revealed as well as concealed when the scientist approaches the object
from this particular angle. If synchrony becomes the only admissible per-
spective within the science of general linguistics properly so called, then
claims to unmediated linguistic objectivity can be made. Such claims sit
uneasily, however, in the company of Saussure’s autographed statements
that duality constitutes the “first and last” principle in linguistics (Saus-
sure, 2006, p. [17], 3), that there is no object in linguistics without a valid
point of view (ibid., p. [23], 8), and that linguistic generalizations constitute
a subjective property, they are “the product of covert operation of the
mind” (ibid., p. [23], 8). These statements suggest a philosophically com-
plex, self-reflective position that may be hard to reconcile with an ideology
of simple objects (see part II for further discussion of philosophy and sci-
ence in linguistics).
*
In the manuscript notes, Saussure is as critical of an exclusively syn-
chronic approach to language as he is of an exclusively historical. Histori-
cal grammar was, in his view, solely concerned with chess moves, and
regarded the positions unworthy of attention (Saussure, 2006, p. [207],
143). It therefore missed the very phenomenon whose evolution it pur-
ported to trace, unable to ground its claims about the identity of terms and
their relations. Saussure regards the existing systemic and historical ap-
proaches as equally grave errors (“it is hard to say which has deeper or

greater consequences” [ibid.]). He concludes that “a language (une langue)
can only be compared to the idea of the game of chess taken as a whole,
including both positions and moves, both changes and states in succes-
sion” (ibid., pp. [207], 143–144).
It follows that language does not divide up into a system of necessary
and immutable rules on the one hand and a historical sequence of contin-
gent facts on the other; rather, it is the case that language is inextricably
dual or has a “double essence” (ibid., p. [208], 144)—just like the actual
game of chess, regarded in its complexity, encompasses both a contempo-
rary layout of pieces on the chessboard and the series of steps leading up
to it. While Saussure contends that “no one would dream of describing a
position by mixing up what is and what has been, even ten seconds previ-
ously” (ibid., pp. [216–217], 151, translation revised), the impossibility of
mixing applies to the two ways of describing the phenomenon (either in the
present, or up to the present moment), and does not implicate that the
object itself is made up of two estranged parts which, like oil and water,
would not easily mix. The object itself is always already mixed or inextri-
cably dual; the problem is: how can you study it scientifically? One hears a
tone of exasperation over this “irritating duality” within language “which
always prevents it from being grasped” (ibid., p. [217], 151, translation re-
vised); elsewhere Saussure laments the “slippery substance of language”
(la substance glissante de la langue) (ibid., p. [281], 197, my translation),
which perpetually evades the scientist’s grip. Language’s Janus-like char-
acter is bound to be irritating to anyone who attempts to gain “a good in-
sight into the complex nature of the particular semiology that is language
(langage),” that is, who tries to not limit the insight to one aspect only
(ibid., p. [217], 151). The inherent complexity of language is never in ques-
tion: language is both historical and contemporary, both far and near,
absent and present, but we must struggle, so it seems, to say both in the
same breath, write both within a single line of text, encompass both under
the auspices of a single science.
Saussure inserts a “note to the reader” into the “Dual Essence of Lan-
guage” fragment with a “crucial point” that since “every fact of language
exists in both present and past realms,” they have not one but two rational
expressions, “both equally legitimate, and equally impossible to remove,
making two things out of one” (ibid., p. [45], 27).2 This insurmountable
2
Saussure adds that no pun is intended—he is not playing off the expression des deux choses
l’une (either of two things), which implies a necessary choice in favor of one or the other thing. A
“thing” here is a distinct object of thought (ibid.), and not a thing in-itself.

duality adds to the difficulty of studying language in accordance with the
guiding principles of science: “this very fact leads us to hesitate over the
nature of language, or to see how anyone can express its nature, since it is
fundamentally double: this is the central truth” (ibid., p. [208], 144). It fol-
lows that linguistics itself is a dual science (une science double), so pro-
foundly divided that it may not fit into the boundaries of a single discipline
(ibid., p. [210], 146). Yet this difficult disciplinary fit never serves to gloss
over the historical dimension of language—it is because of its historicity
that language belongs, academically, to a faculté des lettres and not to a
faculté des sciences naturelles, according to the first lecture delivered by
Saussure at the University of Geneva in November 1891 (ibid., pp. [145–
148], 95–98). Historical change cannot therefore be limited to the realm of
contingent phonetic developments; it belongs to linguistics as its proper
object.
If a distinction between synchronic and diachronic linguistics is to
ensue, it will reflect the two dominant viewpoints and not a natural border
within language itself. Even though language lends itself to a synchronic
study, it itself should not be figured as a static system inherently opposed
to historical change (and as a set of social conventions opposed to the
speaking individual). Saussure insists often enough on the temporal di-
mension of language for a dualist disjunction between the being and the
becoming in language to be justified. For example, even though he advo-
cates abandoning a historical approach to language (considering how use-
less historical studies have been), he acknowledges that “historical inquiry
may coincidentally throw considerable light on conditions governing the
expression of thought, mainly in affording proof that it is not thought that
creates the sign, but the sign that fundamentally guides the thought” (ibid.,
p. [46], 28). This reciprocal relation between thought and sign leads to a
paradox, in that language must be regarded in two irreducible ways: as ex-
pression of thought in sign in the present, and as transmission of sign from
individual to individual, and from generation to generation, through time
(ibid., p. [47], 28). While in expression, the sign is tightly associated with
thought, in transmission, this bond gets loosened, and thought is often lost
and altered in its passage (ibid.). Saussure concludes that the primordial
fact of expression is then lost in the transition of sign. We end up therefore
with an undecidable duality within langage between individual expres-
sion and transindividual (and transgenerational) transmission; such a dual-
ity does not imply an ontological dualism.
A historical perspective is then needed to dismantle the dominant view
according to which language serves as mere vehicle of complete thought,

and makes apparent how the vicissitudes of language itself shape the way
we think. Saussure emphasizes that the historical transmission of language
unties what he considers to be “the central phenomenon of language . . .
association of a thought with a sign” (ibid., p. [47], 28). The historical per-
spective is then ambiguous in that it both makes manifest the passivity of
the speaking subject with regard to the established conventions of use, and
runs the risk of untying the bond between the speaking subject and lan-
guage altogether, as in cases of phonetic evolution for which no thought
can account (or obsolete forms). This ambiguous admixture of symbolic
expression in the present with the transmission of symbols in time under-
scores the fundamental duality of the linguistic field.
As mentioned above, Saussure accuses philosophy of ignoring the tem-
poral dimension of language, that “unforeseen factor” (ibid., p. [231], 163);
this neglect leads to a problematic conception of logos as unchanging:
“What philosophers and logicians have missed . . . is that with the action of
time a system of symbols independent of the designated objects is itself
bound to undergo shifts which the logician cannot calculate” (ibid., p. [209],
145). Importantly, Saussure’s autographed notes advocate a renewed philo-
sophical analysis of the kind of logos appropriate to language—one which
pays heed to the actions of time. Yet, curiously, this study is not to be un-
dertaken under the auspices of the existing academic disciplines like his-
tory, historical linguistics, and philology. The latter are preoccupied with a
succession of external facts, without taking note of their internal organiza-
tion. The new discipline of linguistics as semiology advocated by Saussure
does not deny that language is a historical product; however, its interest lies
in examining how at any given moment language “represents nothing more
than the mind’s (l’esprit) latest compromise with certain symbols”; this
bond between the mind and the symbolic world captures the necessary con-
dition of there being language at all (ibid., p. [209], 145). These claims of
unmistakably philosophical character about the inalienably temporal and
mind-related dimension of language hang together with the emphasis that
language (la langue) forms a system; it is as if the systemic organization of
language could only be discovered from within, by following the internal
route between the symbol and the mind, this latest compromise stipulating
where meaning lies.
Saussure’s quarrel, then, is not with the many facets of language but
with the existing academic disciplines’ failure to do justice to the com-
plexity of language as a phenomenon. History does not properly illumi-
nate the intrinsic temporal dimension of language, philosophy and logic
study signs in disconnect from the mind, physiology of sound abstracts

meaningless sonic matter from out of the complex phenomenon of speech.
The divide between linguistics from a synchronic point of view and the
other disciplines laying a claim to language would then be an attempt to
break out of the existing typology of sciences and their corresponding
objects, in direction of an internal or phenomenological discipline. To be
sure, language must be relatively stable and rule governed to accommo-
date a synchronic approach; still, this relative stability does not imply a
radical antinomy with the evolving practices of usage. The difference be-
tween synchrony and diachrony, la langue and la parole, is perhaps best
illustrated by way of criss-crossings within a common linguistic terrain,
as discussed next.
Dual way to language
In the previous section I developed the idea that general linguistics is a

double science, comprising two related yet distinguishable facets, the his-
torical and the systemic. In this section I will spell out how this distinction
emerges in relation to an emphasis on viewpoints the linguist must alternate
between, as developed in the student lecture notes. This “subjective” em-
phasis will help to establish a philosophically complex, and self-reflective
understanding of language study that ultimately hearkens back to Hegel’s
methodological reflections on the difficulty of beginning.
In the introductory methodological remarks on general linguistics in
the First Course in General Linguistics (Saussure, 1996), Saussure notes
that there are two doors (or perspectives, or points of view) that open onto
language, itself having two facets:
1. The “static” side or facet of language “where everyone is at home,

which each person has an immediate sense, control of; that’s all that
makes up a state of the language (un état de langue): we speak and
therefore we are in a position to judge what we speak. <We can judge
a grammatical question.> Thus anyone can judge if in ‘having sung’
(avoir chanté) ‘having’ evokes an idea all by itself; or to what extent
we make the connections.”
2. The other side where “instinct of no use and which many do not even
suspect; the entire historical side of language, everything which is in
the past, <necessarily> escapes our immediate linguistic sense, and
has to be learned. We form a link <in the chain> of the history of
language (la langue); we see that link but not the chain” (ibid., p. 27).

The initial distinction between the two facets of language emerges in
relation to the two perspectives: the synchronic perspective of direct
access, or immediate linguistic sense akin to being at home in the lan-
guage we ordinarily speak, and the diachronic perspective of mediated
access dependent on historical knowledge of linguistic developments in the
past which does not need to inform the immediate sense of what is linguis-
tically appropriate in our daily usage and experience but which is domi-
nant in historical linguistics and philology. This distinction maps onto the
consciousness of language in actu and its unknown or unconscious side;
the link that we language users occupy belongs to this historical chain, but
our situated perspective of daily usage and experience does not extend onto
the totality of the chain. The perspectival distinction is then between the
daily commerce with language in its current state and the historical under-
standing of its developments, with the correlated points of view of the lan-
guage user—and by extension the general linguist: “It is important to note
that the static perspective concerns at the same time the speaking subjects
and the linguist” (Saussure, 1993, p. 125, translation revised)—and the
historian. The immediate perspective of language use informs the distinc-
tions made at the level of the new science of language, and it has primacy
over the historical approach because it opens a door in the home of lan-
guage where we live. However, the two perspectives remain ambiguously
interrelated.
In what follows, I’ll describe how the distinction between synchronic
and diachronic perspectives is established in the student lecture notes.3
I propose to characterize this process as a dual way to language, wherein
the linguist acknowledges an inherent duality of the object under investi-
gation, and brings one facet into focus without completely detaching it
from the other. This process is apparent in the way Saussure establishes a
series of distinctions within the linguistic field: between la langue and la
parole, and then between synchrony and diachrony.
In the lectures, Saussure invites the students to “consider among the
various spheres where le langage moves the special sphere which corre-
sponds to what I call la langue” (Saussure, 1993, p. 67). He proposes a
complex entity langage, within which one distinguishes the facet of la
langue (described as the social code of conventions) and la parole (active
and bound to the individual) (ibid., p. 70). Saussure insists that la langue
3
One finds a most targeted discussion of these questions toward the end of Constantin’s
notebooks, published in Saussure’s Third Course of Lectures on General Linguistics (1910–11).

and la parole must be treated separately within the emerging science of
general linguistics, each calling for its own theory:
Any chimerical attempt to bring these two parts of langage under the same
point of view will result only in a rather muddled discipline. The global
whole formed by langage is unclassifiable due to lack of homogenous unity.
There is then <in the study one part comprising the study of the individual
part of langage, of speech,> including phonation: this is the study of speech,
and a second part: part of langage lying beyond the will of the individual:
social convention, which is the study of la langue. (ibid., p. 92)
Even though la langue and la parole are deemed objects of two separate
studies, the metaphor of a more entangled, crisscrossed relation recurs in
the lectures, as if the two were mapped onto a chiasm out of which the lin-
guist must—as much as is possible—separate off the interconnected strands.
Saussure uses an analogy of the aspiring linguist who seeks to define its
object of study coming upon a junction, like in a branch line off of the main-
line (embranchement from embrancher, “to join”), a bifurcation where two
roads cross. Since the linguist cannot embark on the two routes simultane-
ously, he can either try to follow them separately or choose one of them. The
choice advocated by Saussure is to embark on the line of la langue (ibid., p.
92).4 Importantly, Saussure does not state that there is just one proper object
of study in general linguistics (la langue), but that the matter under investi-
gation is divided, bifurcated—and that a choice of path to follow must
somehow be made. He therefore raises the question whether one should
keep the name linguistics for both the study of la langue and la parole, and
opts for a duality of terms, linguistics of la langue and linguistics of la
parole (ibid., p. 92). Saussure’s advocated focus on la langue responds to the
need to pick a single road and have it guide one steps and one’s regard. And
yet following the main line does not annul its many intersections to the
branch lines; as Saussure puts it, “one should not conclude that in the lin-
guistics of la langue one should never glance (jeter de coup d’œil) on the
linguistics of speech” (ibid., p. 92).
This junction between linguistics of la langue and linguistics of la parole
is reinstated later on in the lectures when the difficulty of demarcating
4
“C’est là l’embranchement, la bifurcation que l’on rencontre immédiatement, savoir si c’est la
parole ou la langue qu’on prend comme objet d’étude. On ne peut s’engager simultanément sur les
deux routes, [il] faut les suivre toutes deux séparément ou en choisir une. Nous l’avons dit, c’est
l’étude da la lange que nous poursuivons pour notre part” (Saussure, 1993, p. 92).

between unstructured and sporadic instances of language usage on the one
hand and established general patterns on the other is acknowledged. Con-
sider a diachronic fact—for example, that the German ich was evolved into
ich war, in analogy with wir waren (unlike in English where I was survived
next to we were) (ibid., p. 119). This evolutive change is undecipherable with-
out a reference to facts of speech: “Because a few individuals began to say ich
war <by analogy>. It was only a fact of speech (fait de parole) and not a fact
of the language (fait de langue) as long as there were only a few individuals
who did it” (ibid., p. 119). How many individuals does it take to change the
status from a “fact of speech” to a “fact of language”? What marks the transi-
tion from a few speaking subjects to a collectivity whose stamp of approval
transcends the limitations of individual idiosyncrasy? These two facets
(parole/langue, individual/social) are so entangled that to try to separate
them is to effectively unravel both.
While acknowledging this pervasive ambiguity, Saussure advocates
again singling out one path to orient us to a dual phenomenon. “We cannot
mix the two paths” (ibid., p. 118, translation revised). As we approach the
second crossroads—the one between static/synchronic and diachronic
facts within la langue (the first crossroads involving la langue and la
parole) (ibid., p. 118), we the linguists are to embark on the path of syn-
chrony. “When the change becomes linguistic (quand le changement sera
fait langue), we study it” (ibid.).
The emphasis on la langue is then not on a closed and autonomous
structure, but rather on the historically sedimented social conventions of
usage which the linguist qua speaker recognizes as sufficiently stable and
enduring to be studied on their own. There is a judgment made here that
given linguistic patterns function as relevant and socially shared conven-
tions rather than fleeting instances of idiosyncratic usage—but the distinc-
tion is not watertight. The preference of a synchronic point of view onto
language is dictated by the need to orient ourselves to the complex phe-
nomenon—the object itself is temporal and dependent on the individual
speakers’ emergent and replicated instances of novel usage; still, the lin-
guist’s journey can hardly follow the two directions (contemporary and
historical, individual and conventionalized) at once. As before, this syn-
chronic orientation does not foreclose the glance or the foot wandering off
in the direction of diachrony, along the many trails of individual speech
acts. Saussure admits that it is useful to know the diachronic facts, which
show us “the passivity of the speaking subjects with respect to the sign”
(ibid., p. 109). The two crossroads effectively intersect with one another:
the distinction between la langue and la parole is undecipherable without

consideration of the temporal dimension of la langue, its historical excess
with regard to actual instances of speech; the distinction between syn-
chrony and diachrony is undecipherable without there having been reen-
acted speech acts in the past which leave a living (and mutating) residue in
the present. It therefore cannot be settled which crossroads is first within
language itself; even if the junctions can be distinguished, they seem to
sprout one in and out of the other like many intersecting grafts within a
common growth.
Hence, even though Saussure insists at times that synchronic and dia-
chronic facts are “perfectly distinct” (ibid., p. 109),5 he also highlights the ir-
reducibly temporal dimension of language, noting that la langue accommo-
dates the “infinite solidarity with previous ages” (ibid., p. 102) or that “La
langue, considered at any moment, however far back in time, is always an in-
heritance from the preceding moment” (ibid., p. 94). Henceforth, even though
language is to be regarded synchronically, it itself is always retroactive, laden
with the tradition, effectively undecided between the present and the past.
The insistence on a “perfect distinction” between synchrony and dia-
chrony is backed up in the lectures by recurring optic metaphors, consis-
tently advocating two incongruent points of view (these metaphors are
listed in the Course in part I, ch. III, section 4). The first metaphor and
graphic illustration figures the synchronic fact as a flat projection of the
historical reality; the diachronic fact is figured as a projected body, while
the synchronic fact its representation on a plane:
figure 12
This optic metaphor illustrates the dependency of the projected image

on the body and secures the possibility of studying projections indepen-
dently of any reference to the body itself (ibid.).
5
This is reference to an example of the French décrépit, which can now mean either dilapidated
(e.g., of a wall) or decrepit (of a person); this semantic range is historically explained by a
borrowing of the Latin decrepitus, which evolved after phonetic change into décrépit and
coexisted together with crêpir, décrêpir (the verbal forms related to décrépit meaning
“dilapidated”). A contemporary speaker of French perceives a connection between un homme
décrépit and un mur décrépit; from the point of view of the speaking subject in the present, the
two terms are related. This relation does not obtain, however, once the two terms are regarded
from the point of view of their evolution.

Horizontal section The second metaphor is also taken from
optics but is even more explicit on the inher-
ent unity of diachrony and synchrony (ibid., p.
124). According to this metaphor, the distinc-
tion between synchrony and diachrony plays
Vertical section out within one and the same body, such as a
stem, which can be sectioned lengthwise or
crosswise. The horizontal section of the stem
would reveal the cross-sectional arrangement
of its fibers, the vertical their growth over
figure 13 time. Diachrony would be the horizontal sec-
(source: Saussure, 1993, p. 124) tion following the length of the fiber along the
growth of the stem.
The patterns revealed in the horizontal and the vertical cuts thus offer a
certain perspective, a partial view onto a complex object. One can view
language either by means of the diachronic or the synchronic section,
without questioning its underlying unitary reality. Note also that the two
cross sections are undecided between a subjective point of view and the
objective slice revealed. Just as the path to be followed is undecided be-
tween the regard, the marching foot and the terrain, so is the cut along or
across a stem neither in the cutting hand and the seeing eye nor in the
sectioned plant. They form an ambiguous dual reality which coinvolves a
determinate orientation with the corresponding facet being exposed, as
well as traversed in the process.
Here may lie the profound philosophical dimension of general linguis-
tics: the famed distinction between synchrony and diachrony emerges
within this self-reflective inquiry into how the perspective adopted towards
language brings one facet into focus while leaving the other temporarily
out of sight. Yet this reflection is not solely specular in kind, since the meta-
phors offered to the student in general linguistics suggest an active explora-
tion of a terrain which writes a trail back into the very “object” being inves-
tigated. We continue to be in language as we speak of it; our speech returns
to language its various residual effects. A rigorous science of language
must accommodate this complex entanglement of subject and object which
the linguist maintains via his indissoluble speaker position in language. A
critic could charge Saussure with forcing the choice between synchrony
and diachrony, possibly out of frustration that language cannot be viewed
and traversed in both directions at once (which would suggest a desire for
instantaneous and unmediated access to the field, yet another offspring of
metaphysics of presence). Yet an equally plausible interpretation is this:

preserving a difference between synchrony and diachrony forecloses the
possibility of language being approached unreflectively, as a total object.
Once its dual, stable and evolving, essence is acknowledged, language
refers back to the duality of (synchronic and diachronic) perspectives
through which either facet is brought into focus. The distinction between
synchrony and diachrony thus retains a basic bond to the subject, and does
not sacrifice it in the name of neutral objectivity. The synchronic view is
privileged and yet this view is partial, and always ambiguously admixed
with its other.
*
Needless to say, acknowledging the orientational complexity of the lin-
guistic field poses a threat to the possibility of linguistic science. The ter-
rain is slippery at best. No term is fixed enough to offer a steady foothold.
What may seem like an objectifiable side of language turns out to be criss-
crossed with the subjective one, so much so that the basic terms (sign and
system, speaking subject and language) alternate and reverse depending on
one’s orientation. The point of view is irreducible in linguistics, and that is
why linguistics is not an objective science. What does it mean then to study
language scientifically?
I propose that this difficult if not infinite task involved in reflecting on
the complexity of language while seeking to study it within science is cen-
tral to Saussure’s general linguistics. It constitutes its continued appeal, but
may also shed some light on the manifest impasse of his project. In his
manuscript writings, Saussure posits, almost tragically, the “necessary ab-
sence of any point of departure” in linguistics; he suspects that the reader
who will have followed his future book on general linguistics from the
beginning to the end will recognize that it was impossible to follow any
rigorous order (un ordre très rigoureux) (Saussure, 2006, p. [198], 136):
We have arrived at each thing which we considered as a truth by so many

different ways (voies) that we confess not knowing which one is prefera-
ble. It would be necessary, in order to present all our propositions in an
appropriate manner, to adopt a fixed and defined starting point (point de
départ). But all that we have tended to establish in linguistics is that it is
wrong ( faux) to consider a single fact in linguistics as defined in-itself (en
soi). There is therefore necessarily no starting point, and we are sure that
the reader who follows our thinking attentively from one end to the other
of this volume will recognize that it was in fact impossible to follow a
rigorously defined order. We are obliged to submit to the reader the same

idea, up to three or four times, in a different form, because no one starting
point provides a more indicated foundation for the demonstration than
another. (ibid.)
The entire cited paragraph was deemed “unusable” by Sechehaye who

notes in pencil in the margins of his copy: “Here is a point of view which
our redaction effaces almost entirely” (cited in Matsuzawa, 2003, p. 320).
Yet Saussure’s plea to an imagined future reader who will have reached
the end of a volume that was never completed is useful in a number of
ways. It tells us that the book was not planned as a deductive thesis, and
its design was imagined quite differently from the deduction from basic
axioms format pursued in the Course. Such axiomatic presentation sup-
poses the validity of what was in question: a fixed starting point, from out
of which the whole doctrine could be spun out. Yet following Saussure’s
own reflections, language does not afford such fixed anchors. Further-
more, his planned book on general linguistics sought to reject the notion
of objectivity from natural science, and to develop an alternative under-
standing of scientific inquiry wherein the in-itself is bound up with the
for-itself—a complex notion which goes beyond the methodological and
metaphysical categories assumed in a purely fact-based science.
The editors apparently classified Saussure’s reflections cited above as
the disposable “variations” and “hesitations” (Saussure, 1986, p. [9], xvix)
added on to an otherwise determined doctrine. But the stated absence of a
natural entry into language is reiterated in Saussure’s writings, and forms
an identifiable thread within his reflections on general linguistics. Con-
sider this mantra-esque theme dispersed through the manuscript notes: “It
seems impossible in practice to give priority to any particular truth in lin-
guistics so as to make it the key starting point” (Saussure, 2006, p. [17], 3);
“in langue there is neither starting point nor indeed any fixed point what-
soever” (ibid., p. [40], 22); “It is never legitimate to consider an aspect of
language as previous or superior to any other and thus elevate it to the
position of starting point” (ibid., p. [197], 136); “If what I want to say about
[la langue] is true there is not a single point that could act as the evident
point of departure” (ibid., p. [281], 197). And even if the study of the lan-
guage system (la langue) is posited as being “precisely the starting point,”
then the “irritating duality” of language (langage), with its irreducible his-
torical dimension, comes to complicate the issue (ibid., p. [217], 151). In
sum, Saussure’s negative comments belong to the programmatic vision of
the science of general linguistics, and the explicit focus on la langue calls
for a different vision of science—one attuned to the complexity of language

and to the communal belonging and continual intersecting of subjects and
objects, expressions and conventions.
Interestingly, the “hesitations” noted in the “unusable” fragment cited
above, and elsewhere in the Writings, resonate profoundly with the diffi-
culty of beginning philosophical reflection, as stated by Hegel:
Philosophy forms a circle. It has, since it must somehow make a beginning,

a primary, directly given matter, which is not proved and is not a result. But
this starting point is simply relative, since from another point of view it ap-
pears as a result. (Hegel, 2008, p. xxiv)6
Saussure’s writings are making a similar point: within a system as

tightly woven as language, wherein the speaking subject is itself entangled,
all of the discernible elements (sounds, words, signs) seem at first to be
directly available and hence to offer a natural starting point; yet upon re-
flection they turn out to result from a greater unity which must be presup-
posed as their condition, and so effectively comes first. But how can all
that be said in the beginning? How can the beginning also reflect the end,
and everything that comes in between? How is the usual book format to
render this interweaving text? Is the book about language not just as much
about the pathways traced across its heterogeneous terrain, where the
routes of diachrony and synchrony continually crisscross and indicate the
trajectory of our own footsteps and our wandering regard? How to inaugu-
rate a science faithful to a reality so closely bound up with our own seem-
ingly endless peripatetic pursuit? How to lead into (intro-ducere) language
we already inhabit? Should the introduction contain all the metascientific
reflections attainable only at the end (as is the case in Hegel’s Phenomenol-
ogy and Logic)?
Saussure’s question: unde exoriar (Latin: from whence to begin or
emerge) becomes then all important in a reflection on language and in the
science of general linguistics. It is the question “not really pretentious and
in fact extremely modest and positive, that can be raised before attempting
to approach in any way the slippery substance of language (la substance
glissante de la langue)” (Saussure, 2006, p. [281], 197, my translation).
Hegel’s answer to the unde exoriar question could be to write a sequel
to a completed volume. Just as a book can only begin at the end, so does
the end become the beginning of the next. The Phenomenology is only
6
This question is also taken up within Hegel’s Logic: “With what must the beginning of science
be made?” (Hegel, 2010).

preparation for the Logic, and the latter designed to complete the system.
Saussure’s own answer, on the other hand, is to be found in the extant tex-
tual labyrinths compiled as the Writings in General Linguistics, which
were written with no introduction and no conclusion, and appear as end-
less, fragmented and unbounded as the projected system of signs itself.
Reading them is not unlike traveling across a slippery terrain, with lines
dropping off in midsentence like a rough cliff from which to hop onto the
next. The traveler has no means of settling if she has reached the destina-
tion, where she is and where she is going. There is no guidebook she can
consult, no book at all. She cannot but follow this labyrinth along the path-
ways created in part by her own footsteps and regard, and is guided by
them in turn. Maybe that is a faithful rendering of the subject’s implication
in a system of signs: an endless journey through a text which sheds all pre-
tenses of determinate beginning, middle, and end.7
7
In part I I developed a philosophical critique of the official doctrine associated with Saussure’s
general linguistics. In part III I enrich this critique with an account of the material and
institutional history that led to the production and reception of the Course as a statement of
doctrine. My account focuses on the editorial process of ghostwriting the Course in Saussure’s
name and then reviewing it in scholarly venues (ch. 7), and on the structuralist and
poststructuralist reception of the Course as Saussure’s work by Levi-Strauss, Lacan, and Derrida
(ch. 8). The reader interested in the full scope of my critique may therefore prefer to turn to part
III directly.

II General linguistics: science
and/or philosophy of language
4 Involuntary assumption of substance,
and points of view in linguistics
A
s a human science, general linguistics is faced with a dilemma. It
must meet the standards of scientific knowledge, which means that
it should abide by the standards of objectivity and employ the
method of external observation. The science of physics used to be regarded
as an exemplar of such an approach, thus setting the physicalist standard
for other disciplines aspiring to scienticity. However, since linguistics is a
science that deals with a human matter, it does not simply discover its
object for the first time but possesses an innermost familiarity with it in
daily experience and usage; experience and usage precede but also enable
scientific study in linguistics.
Before figuring as an object of study, language is encountered subjec-
tively, within the consciousness that a language user has of being involved
in language-bound practices of speaking, listening, and writing. To be
sure, there is a degree of automatism in language use, and language is
found in dreams as well as in carefully planned and delivered formal ad-
dresses. However language is not a foreign body forced into the stuff of
consciousness, even though it has a degree of agency of its own; if lan-
guage does not overlap with consciousness, it still belongs to it. This lin-
guistic consciousness holds primacy over scientific knowledge of lan-
guage; furthermore, linguistic consciousness is already a site of knowledge
in that the speaker discriminates between sense and nonsense, and makes
daily choices of le mot juste. One can therefore distinguish between an
originary knowledge of language in and through experience and usage,
and scientific knowledge which comes second (Pos, 1939, p. 356). Further-
more, even when adopted by a scientific discipline as its principal focus,
language cannot be reduced to the domain of known facts, and the linguist
does not cease being a language user. Linguistics does not therefore easily
fit into the mold of objective sciences; it may require an alternative notion
of scienticity—one contingent on employing the resources of philosophi-
cal reflection, and subjectivity.
The key question for a scholar of language is how to construe the rela-
tion between linguistic consciousness and linguistic knowledge. The schol-
ar’s conception of science bears directly on this question. A scholar who
adheres to an objectivist ideal may acknowledge the prior existence of lin-
guistic consciousness, but not endorse it as a scientific datum. Like a be-
haviorist in psychology, a linguistic physicalist may declare that conscious-
ness either lies outside the confines of objects amenable to scientific study
(even though it exists), or that it is reducible to physical processes (and so
has a lesser degree of reality in comparison, as it were). Whether it is of a
methodological or ontological bent, physicalism in linguistics advocates a
sole focus on language objectively construed and reduced to a purely ma-
terial process like articulation and reception of sound, which can be stud-
ied naturalistically within physiology.
Another approach is to acknowledge that the primary stance of linguis-
tic consciousness is unsurpassable, and that a science of language must
retain a footing within speaking subjectivity. This phenomenological ori-
entation calls for an alternative method within the human sciences. It must
shoulder the burden of proof that its approach remains scientific even
though it breaks away from an objectivist model, and that subjectivity can
be accommodated in science even though it does not fit into the mold of
objectifiable reality. This is an enormous if not an infinite task; yet argua-
bly this is the task taken up in Saussure’s general linguistics. Saussure’s
general linguistics seeks to respond both to the exigencies of science and
subjectivity, and it envisages a scientific reform that accommodates the
primary and unsurpassable stance of the speaking subject. The stance of
the linguist as scholar of language is therefore congruous with the stance
of a language user. Saussure’s response to the question of how linguistic
consciousness and linguistic knowledge relate preserves the primacy of
the former with regard to the latter, without giving up on the aspiration of
pursuing a rigorous scientific study. The method of this science will there-
fore not be external observation but participatory reflection on language as
experienced, lived, and employed in daily practice. This science will
remain bound to human experience, and ideally avert the crisis of meaning
that haunts any discipline aspiring to transform human facts into absolute
objectivity.

Saussure’s own writings develop a critique of the positive notion of sci-
ence. His basic claim is that there are no in-itself entities among linguistic
facts, and that their identity is bound up with the categories employed for
the sake of classification; before rushing out to accumulate the facts, the
scientist needs to reflect upon the basic categories employed in the process
of picking out and correlating entities for investigation. Put differently, the
scientist must interrogate the subjective perspective from which entities
can emerge at the object side of the relation, and ultimately opt for the most
appropriate perspective for language study—that of the speaking subject.
While the former task is largely critical in that it exposes the problematic
assumptions made in the existing sciences of language, the latter is con-
structive in that it devises an apt methodology for general linguistics. The
latter puts forward a new philosophical conception of linguistic science of
a phenomenological kind, which focuses on linguistic phenomena as they
are intrinsically coupled with and available to the perspective of linguistic
consciousness.
The notion of perspective is not only a property of general linguistics—
it figures as an ineliminable dimension of any scientific pursuit, including
the pursuit of a positive science, and therefore affects the branches of lin-
guistics that uncritically assume the validity of natural scientific method
and reduce language to a natural scientific object. This reduction is espe-
cially apparent in the attempt to fit language study into a physiology of
sound (Lautphysiologie), where linguistic claims are made based on a
study of sheer acoustic data devoid of signification and social value. Such
an approach inevitably leads to a crisis in the science of linguistics—the
scientific object is out of sync with the actual linguistic phenomenon, and
considerations relative to meaning get lost within a purely materialist un-
derstanding of reality.
The rejection of positively construed facts in the nineteenth and early
twentieth century was part of a larger methodological debate about the
foundations of science, which extended far beyond the scope of linguistics.
One finds examples of it in Wertheimer’s Gestaltpsychologie, in Weber’s
sociology and his notion of an ideal type—and most significantly for our
purposes, in Husserl’s phenomenological critique of the natural scientific
method. According to Husserl, the positivistic reduction of science to a
mere factual pursuit in the nineteenth century purged scientific study of
any considerations relative to meaning, value, and history (1970, pp. 5–7).
This reduction produced a crisis of scientific and existential order. The
positivist science has nothing to say to the vital need of meaning in human
life. Philosophical reflection on meaning and value is deemed unscientific,
involuntary assumption of substance, and points of view in linguistics 111

since removed from the study of positive facts. And yet especially the sci-
ences dealing with human matters, like psychology, are inevitably led back
to the enigma of subjectivity and resist being cast as objects pure and
simple.
Saussure’s reflections on general linguistics serve as an unexpected ally
in this phenomenological emphasis on the enigmatic entanglement of the
subject within science, and offer a concurrent diagnosis of a crisis in lin-
guistics conceived as a purely objective science. The source materials de-
velop a similar critique of the sciences dealing with human matters like
language as if it were a directly available natural fact, and develop an al-
ternative approach which reflects back on the subject’s involvement in the
matter under investigation, both in the everyday context of language use
and within language study. Rather than reduce this involvement to a merely
subjective obstacle to a preexisting standard of scientific objectivity, they
demonstrate that such standards misconstrue the concrete reality of lan-
guage, which is subjectively (and socially) mediated and shot through with
meaning. General linguistics must therefore admix philosophical reflec-
tion with empirical study to capture the reality of language, to turn back to
language itself. As such it goes against the grain of an assumed divide
between a philosophical and a practical science; language study can only
make headway if it moves on both tracks.
*
In the autographed manuscripts, Saussure advances a critique of the lin-
guistic methodology prevalent in his time, notably an unexamined meta-
physical commitment to the idea that reality is made up of substance-like
beings, and that language, like anything else, must fit into this mold. This
unexamined metaphysical commitment, which pervades the entire epis-
teme of the linguistic science, must be made explicit:
Since language does not, in any of its manifestations, present a substance . . .

and since all our distinctions, all our terminology, all the ways we express
ourselves are modeled on this involuntary supposition of a substance, it
must be accepted that the most essential task of a theory of language will be
to unravel the primary distinctions. (Saussure, 2006, p. [197], 136, transla-
tion modified)1
1
Saussure states also: “The concept of a basic substance, which is then enriched by attributes,
may not be entertained” (2006, p. [81], 55).

The initial task will then consist of unraveling these distinctions and
redefining the basic terms: “In our view, no theory should be drawn up
without this work of defining, despite the fact that this convenient approach
has up till now satisfied the linguistic public” (ibid.). This critical task
needs to be pursued despite the fact that such a methodological enterprise
may seem like an unnecessary deviation within a fact-based science of
language (Saussure, 1954, p. 55). This is the daunting and immense task of
showing to the linguist “what he is doing, by reducing each process to a
category laid in advance for it,” as Saussure put it to Meillet in his January
4, 1894 letter. Crucially, the linguist must refuse “any single fact as defined
in itself” (comme défini en soi) (Saussure, 2006, p. [198], 136, translation
revised).
Saussure interrogates the metaphysical prejudice of in-itself objectivity
at some length in the manuscript notes. It may be that any kind of theory
tends to assume its validity and work within its limits: “Our thought tends
continually to convert into substance the various actions required by langue.
It seems necessary as part of theory to adopt this conception” (ibid., p. [81],
55, my translation).
This tendency of any theory to convert the actions of language into met-
aphysical notions is manifest within the linguistics of his time, and Sau-
ssure documents the metaphysical worldview with numerous examples of
an “involuntary assumption of substance” in linguistics. Consider for ex-
ample the Neogrammarians (Jungrammatiker) who proposed a novel ap-
proach based on a scientific study of actual languages, both ancient and
modern, which were to be learnt and studied in requisite detail. The lan-
guages under consideration were almost exclusively Indo-European, and
the goal was to trace them back to an assumed ancestral source—a proto-
Indo-European, a common matrix from out of which the individual lan-
guages have sprung. The many differences between the members of the
Indo-European family were explained largely as effects of historical evolu-
tion from out of the unitary Indo-European prototype. They were phonetic
changes undergone by individual languages over time, and they could be
explained by means of phonetic laws which capture regular patterns within
phonetic change (e.g., skipping of weak vowels, adding of syllables to form
grammatical endings and resulting shift of stress, etc.). However, since the
majority of Neogrammatical research remained confined to dead languages
and focused on written documents, the phonetic laws were established by
means of a painstaking reconstitution of plausible sounds from the written
form and the modifications within the assumed vocal configurations as re-
flected in the changes of spelling over time. These laws were backed up by

a growing knowledge of the physiological apparatus by means of which
speakers produce articulate sound, as well as the properties of the sound
itself, which can be studied as a purely physical fact, independently of lin-
guistic meaning. The phonetic changes were then explained, scientifically,
as physiological and physical phenomena, bearing a set of acoustic and
articulatory properties. In sum, even though the new approach was both
historical and comparative, it remained wedded to a naturalist understand-
ing of language as a positive entity, comprising the properties and the pro-
duction of sound. This is in line with its aspirations to scienticity and solely
fact-based knowledge of language.
Saussure’s own linguistics is usually qualified as naively scientific in its
focus on the static system of la langue, but this qualification is at odds with
the linguist’s extensive critique of scientific methodology prevalent in his
time. According to the sources, a scientist typically privileges, without ac-
knowledgement or afterthought, entities that are simple, material, and in-
dependent of time—for example, a small unit of sound, such as the group
aka or the vowel a. This sound bite is purged of any meaning (significa-
tion) and idea of use (idée d’emploi) (Saussure, 1954, p. 55; 2006, pp. [197–
198], 136), and as an objectified quantum, becomes susceptible to natural
scientific analysis, that is, can be studied solely in terms of a concordant
series of articulatory and acoustic processes that went into its production.
The scientist arrives therefore at a causal account of articulated and audited
sound in terms of human physiology, independent of any reference to
human activities of meaning making. The question of how a unit of articu-
late sound—even as small as aka or a—could be at all isolated and picked
out within language study is not raised; it is assumed, in line with a positive
idea of knowledge, that sound is spontaneously segmented into units which
will subsequently function as signs; sounds would then be endowed with
signification après coup, the latter endowment being an external add-on to
a self-standing, substantial unit. Language can then loom as an independ-
ent material object, on a par with objects studied in natural science, or as a
substance with detachable attributes.
Yet even though a small unit of sound may meet all the requirements of
natural scientific analysis, it is so impoverished in comparison to language
facts properly so called that the scientist has effectively lost hold of the phe-
nomenon under investigation in the process; as a result, any findings rela-
tive to the physical object have no bearing on language itself. The natural
scientific approach to language has reduced the sign to a series of sound
waves and therefore deals only with “vocal figures” devoid of signification,
and not with signs (Saussure, 2006, p. [21], 6). This reductionist tendency

is coupled with a naturalist illusion—“the illusion of things being naturally
given” (Saussure, 2006, p. [199], 137), that is, of existing independently in
nature, outside of our awareness and practices of usage. This illusion is said
to run very deep, and pervade any objectivist approach to language. It re-
sults from a sole preoccupation with positive facts, while the accompany-
ing and enabling point of view gets glossed over. The naturalist illusion is
also tied to an unexamined metaphysical commitment to an unbridgeable
dualism of mind and matter, where the former circumscribes the domain of
signification within, and the latter the physical domain of sound without.
Saussure will replace this dualist stance with a principle of duality or the
“twofold essence of language,” with the sign and signification being indis-
solubly bound one to the other (see Saussure, 2006, p. [20], 6). Only then
can the complex character of language phenomena be preserved, and the
reductionist tendency of natural science foreclosed.
Saussure’s critique of the “involuntary supposition of substance” ad-
dresses the assumptions of a methodological nature made within positivist
linguistics, and the metaphysical scenario that accompanies it. Such a re-
flective inquiry may be classified as an exclusively “negative” task pursued
in philosophy but not in science; and yet, following the stated need to re-
consider the primary definitions within linguistics, and in accordance with
the guideline that “no theory should be drawn up without this work of de-
fining,” it is incumbent on the scholar of language to forsake a simple divi-
sion of labor between (negative) philosophy and (positive) science, and to
reflectively reorient the pursuit of gathering empirical knowledge about
language. The scientist must become a philosopher to not succumb to the
crisis of science.
This theoretical task consists in spelling out the interdependency be-
tween linguistic facts and the corresponding lens adopted by the scientist.
The idea here is that linguistic facts can only be grasped via such a lens,
and would not come into view without it. Just like an astronomer uses a
telescope to view distant planets, and a neuroscientist uses magnetic reso-
nance to image inner organs, the linguist adopts a determinate perspective
to orient to the phenomenon in question. This orientation may be limited
to employing a dedicated conceptual apparatus; still, the distinctive fea-
tures of each orientation have to be spelled out, and what it highlights as
well as what it obscures within the heterogeneous field of language needs
to be rendered explicit. For example, language can be viewed through the
lens adopted within the discipline of neural and motor anatomy; it will
then emerge as a physiological phenomenon with a set of neuromotor
properties. Language can also be viewed through the lens adopted within

philosophy and logic; it will then emerge as host of meaning-bearing con-
cepts and ideas which are employed in the making of logical arguments
and demonstrations. Language can be viewed through the lens of sociol-
ogy as one of the practices that maintain and/or disrupt the social fabric
via communicative use. It can be viewed through an archeological lens as
an abiding document of bygone eras, illustrating their customs and ways
of life. These and other disciplines lay a claim to the vast domain of lan-
guage by isolating and highlighting one of its many facets via their dis-
tinctive lens.
Consider some examples examined by Saussure of how a disciplinary
focus necessarily couples a linguistic fact with a subjective point of view
(regardless of whether the scientist reflects upon this). A linguist engaged
in a comparative historical study establishes comparisons by means of
identity relations between terms from contemporary languages, and also
from a given language’s own lineage. For example, the linguist can estab-
lish a relation of identity between the French chanter and the Latin can-
tare, and also between the Latin cantare and the Hottentot kantare. If left
without further qualification, the claim of identity between chanter, can-
tare, and kantare is viewed as an objective factum, and ultimately suggests
some common entity with a “solid form of existence,” such as an inde-
pendently existent vocal figure made up of a sound string kan-ta-re about
which it can be said that “it is” (Saussure, 2006, p. [199], 137). Such cate-
gorization of linguistic terms fits well into the usual criteria from sub-
stance metaphysics: a self-identical vocal figure like kan-ta-re is the sub-
stantial foundation, while anything added to it (meaning, historical
variations) in language is an attribute (ibid.).
The relations of identity just noted turn out however to be contingent on
different subjective criteria being deployed in each case, and any claim
about objective reality in language must be similarly tied to a subjective
viewpoint. The Latin cantare can be identified with the French chanter
according to the criterion of genetic identity, on etymological grounds, but
the Hottentot kantare can only be identified with the Latin cantare accord-
ing to the criteria of physiological and acoustic properties of the sound ut-
tered and received. Each identity is therefore bound up with its correspond-
ing criterion of what counts as common measure between compared terms;
it is therefore unfounded to make claims about exclusively objective reality
in language since each object is contingent on the perspective being ad-
opted. Even when we proclaim the existence of an identical vocal figure in
French and Hottentot and thus establish a material bridge between the two
terms, “we tacitly call on the ear to be the judge of identity,” and cannot

avoid “resorting to a highly positive mental operation” (une opération très
positive de l’esprit) (ibid., p. [199], 137). We would not, after all, say in the
same breath that the French chanter is identical to the Hottentot kantare,
and also to the Latin cantare. In the latter comparison, the claim of iden-
tity is of a different kind and arises “from another type of judgment” (ibid.).
Its judgment is relative to meaning, and situated on a temporal axis run-
ning from earlier to later language forms. No shared linguistic substrate
can therefore be isolated within the contemporary and the historical com-
parisons. In each comparison, the judgment is paired with a corresponding
set of objects (a unit of sound or vocal figure in physiology of sound, a unit
of meaning in historical linguistics), with no criteria indicating if the sets
are in fact the same thing. The thing category does not therefore fit the
study of language, and it is misguided to import the distinctions and termi-
nology modeled on the substance into it.
In sum:
There are different types of identity. That is what creates different types of
linguistic facts. Outside some relationship of sameness, a linguistic fact
does not exist. But the relationship of sameness depends on a variable view-
point, which one decides to adopt; thus not even rudimentary linguistic
phenomena can exist outside a defined viewpoint which governs the dis-
tinctions. (Saussure, 2006, p. [200], 138)
This acknowledged dependency of linguistic facts on the adopted view-

point implies that the scientific object is not immediately and directly
available as a positive datum but necessarily mediated by the categories
and criteria used to pick it out, and set it up for comparison with other enti-
ties. The logical law of identity between indiscernibles does not settle ob-
jective identity between linguistic facts, since a number of open questions
remain: From what standpoint do the facts first appear indiscernible? Does
the indiscernibility from standpoint A overlap with indiscernibility from
standpoint B? Which if any of the possible standpoints holds primacy over
the others? If there is no privileged perspective from which to approach
linguistic facts, can language make any claim to unity as an object, or is it
but a loose bundle of disjointed attributes?
A subjective orientation puts a heavy burden of relativity upon scientific
facts, and it may lead to outright skepticism if the void left by the seem-
ingly unconditional and objectively given facts cannot be easily filled by a
privileged perspective from which to approach them. Some of Saussure’s
statements are expressions of skepticism, as if all ground has been lost

from under the linguist’s feet. Yet one finds also a more positive notion that
any headway in general linguistics is contingent on accommodating this
acknowledged perspectival relativity within its own methodological prin-
ciples and practice. Saussure advocates a “comparative criticism of points
of view” (Saussure, 2006, p. [199], 137) as a necessary task in linguistics.
This comparative criticism should settle which point of view—if any—is
most adept to language study; the famed distinction between synchrony
and diachrony is related to this preoccupation with method and perspec-
tive in general linguistics.
Saussure’s comparative criticism makes a distinction between the true
and false points of view in linguistics. While in other disciplines there may
be preexistent objects available to different points of view, in linguistics
points of view are indissociable from the objects themselves (Saussure,
1954, p. 57). However, the point of view does not simply create the object ex
nihilo, and the object is not a result of unmotivated activity on the subject’s
part. Saussure notes that the point of view may or may not correspond to
linguistic reality. A false point of view onto language is at odds with, while
a true one in accordance with linguistic reality (ibid., p. 58). This does not
imply however that language is a preexistent object akin to the data of natu-
ral science. The concrete reality of language is nothing objective in the usual
sense; it is intrinsically bound to a subjective viewpoint insofar as facts of
language properly so called have a temporal dimension, signifying value
and are subject dependent (ibid., p. 58). A scientific approach must preserve
this primary entanglement within language which the linguist shares with
all other language users. It is in that sense that the point of view comes first
in language and linguistics alike. A naturalist approach purges the facts of
language of temporality and signification, and thus no longer deals with lan-
guage properly so called but with abstractions. Its point of view is therefore
necessarily falsifying. However, Saussure emphasizes that in the case of
both a true and a false point of view, there is no thing, no object, ever given
in-itself (en soi)—not even in the case of physiologically explained vocal
sounds, even though the latter seems to be most material and defined purely
in-itself (en soi) (ibid., pp. 57–58). Thus, while the point of view is inescap-
able in linguistics, naturalist or not, the task taken up by Saussure is to re-
cover the point of view most attuned to the pour soi dimension of language
where signification and speaking subjectivity mutually call upon each other.
The naturalist attitude seeks to efface itself in favor of the absolute objectiv-
ity of material facts. The reflective attitude promoted by Saussure preserves
the originary entanglement in language in order to safeguard its concrete
reality. Saussure’s preferred attitude is therefore phenomenological in that it

approaches language the way it is (already) given to us in experience and
usage, and models its method on this prereflective manifestation of meaning
to the speaker rather than on ideals of metaphysically construed objectivity
posited by the scientist. Saussure’s methodological attitude may therefore be
deciphered by way of Hegel’s conception of science (as argued below in
ch. 6, ‘Phenomenological science: Hegel and Saussure’).

5 Saussure’s general linguistics as
linguistic phenomenology
T
he retroactively constituted bond between Saussure the master
and his structuralist disciples creates an impression of uninterrupted
intellectual inheritance from like to like, and it is easy to lose sight of
the fact that the structuralist claim to Saussure’s linguistics was not seamless
but separated by some four decades from its inception, and marked by an aca-
demic climate quite different from the one Saussure himself was writing in.
While the structuralists such as Levi-Strauss, Lacan, and others reclaimed
Saussure’s linguistics within an intellectual climate of intense antagonism and
animosity between the phenomenological and structure-based approaches in
the 1950s and 1960s, Saussure’s own reflections on language grew on a terrain
saturated with phenomenological references (notably to Hegel’s phenomenol-
ogy), and largely conciliatory. It was relatively common before the Second
World War, on an undivided European continent, to apply phenomenological
resources to research in general linguistics, as Holenstein documented, by
means of the collaboration between the Russian linguist Jakobson and the
German phenomenologist Husserl. It is equally useful to recover the sphere of
influence the Course exercised upon the phenomenology of Merleau-Ponty,
especially his views of the speaking subject. Such an archeological task be-
longs to a larger effort to transcend the perceived divide between a phenome-
nological and a general linguistic type of inquiry; going before the divide may
point the way of going beyond it.
Predecessors: Kazan School of Linguistics, Kruszewski
It is notoriously difficult to trace Saussure’s own intellectual heritage outside

of linguistics narrowly construed, and some scholars have even suggested
that the brief list of acknowledged intellectual references for his own view
of language may suggest a poorly furnished reference library (Gambarara,
1972). Still, it is possible to shed some light on the phenomenological impe-
tus in Saussure’s work—even though one needs to venture into intellectual
territories less frequently traveled and not as well mapped out as is standard
in the European intellectual history. Following such an alternative compass
in the direction of Eastern Europe, before the Second World War, will help
to decipher the otherwise insoluble enigma: the presence of a phenomeno-
logical impetus within Saussure’s work.
Among the few directly acknowledged influences on Saussure’s ap-
proach to language we find the works of two Polish linguists, Baudouin
de Courtenay and Kruszewski, founding members of the Kazan School
of Linguistics.1 Their work has been almost totally eclipsed in the West,
in part for political reasons. As Percival notes, key representatives of the
early Eastern European structuralism belonging to the Prague Linguis-
tic Circle, notably Roman Jakobson and Vilem Mathesius, may have
found it “politic” at times to avoid mentioning Baudouin de Courtenay
when addressing an audience of linguists from Western Europe (2011,
p. 251). For example, when Mathesius presented a paper at the Second
International Congress of Linguists held in Geneva in August 1931, he
traced structural linguistics back to the ideas of “Saussure in the West”
and “Baudouin de Courtenay in the East”, but in the published confer-
ence proceedings he retained a reference to Saussure only. Jakobson is
also reported to have been bent on underplaying the importance of Bau-
douin de Courtenay. When asked directly about this, Jakobson is re-
ported to have said that “nobody would have listened to us, had we
talked about the Poles” (p. 252; see ch. 8 for further discussion of East
European structuralism).
These political considerations may explain in part why the impact of Bau-
douin de Courtenay and Kruszewski on Saussure’s general linguistics is
largely unrecognized, even though it may be substantial. Saussure himself
noted that the two scholars “got closer than anyone else to a theoretical view
of language, and this without leaving purely linguistic considerations” (2006,
p. [259], 185). And, in a different context, Jakobson goes as far as to state that
“[m]ost of the cardinal theoretical concepts and principles introduced by
Saussure go back to his older contemporaries, Baudouin de Courtenay and
1
See ch. 3 in Phonology in the Twentieth Century by Steven R. Anderson (Chicago UP, 1985), for
historical background on the Kazan school (the town of Kazan is situated in the European part of
Russia).
saussure’s general linguistics as linguistic phenomenology 121

Kruszewski” (1971, p. 717).2 For example, Saussure “obviously followed
Kruszewski” in setting up a distinction between two kinds of relations within
language: a paradigmatic one, based on selection of terms, and a syntag-
matic one, based on combination of terms into complex units (ibid., p. 719).
Kruszewski’s distinction itself could be further tied to Hegel’s dialectics,
considering a strong tradition of Hegelian influence in Russia (Jakobson,
1990, p. 3). Similarly, “The distinction between two linguistic attitudes—
synchronic and diachronic—was clearly outlined and exemplified by Baud-
ouin de Courtenay throughout the last third of the nineteenth century” (Ja-
kobson, 1971, p. 720). This distinction had a phenomenological equivalent in
Brentano’s lectures on descriptive psychology, whose approach offered a
much needed supplement to the dominant field of genetic psychology; an
emphasis on a synchronic approach to a phenomenon, studied on its own
terms rather than as part of a reconstructive effort to determine its historical
provenance, is therefore a shared feature of general linguistic and descriptive
phenomenological approaches.
Both Baudouin de Courtenay and Kruszewski were committed to de-
veloping a general theory of language and a theory of linguistics (de
Mauro, 2005, p. 340), but this commitment is especially apparent in
Kruszewski’s philosophically inflected work. At the outset of his master’s
thesis On Sound Alternation, Kruszewski states:
Everyone will agree that the subject matter of linguistics must be those
phenomena whose totality is called language, and that the ultimate goal of
this science must consist in the discovery of those laws which govern such
phenomena. (Kruszewski, 1995, p. 7)
Like Jakobson and Husserl in the early 1900s, Kruszewski seeks there-
fore to isolate a set of invariant a priori principles organizing the phenom-
enon under investigation. His general approach to language offers an al-
ternative to the genetic focus which dominated the linguistic sciences in
the 1800s, where a comparative study of Indo-European languages was
put in the service of the overriding goal to reconstruct a common protol-
anguage. Kruszewski charges this reconstructive approach with lack of
2
The full citation reads:
Most of the cardinal theoretical concepts and principles introduced by Saussure go back to
his older contemporaries, Baudouin de Courtenay and Kruszewski, but in the Course some of
these notions were presented in a more perspicuous and expanded manner, and an effective
emphasis was placed on the mutual solidarity of the system and its constituents, on their purely
relative and oppositive character, and on the basic antinomies which we face when we deal with
language. (Jakobson, 1971, p. 717)

scienticity: “Needless to say, none of this can be considered science”
(ibid., p. 7). This charge, leveled by an emergent scholar to his senior col-
leagues from well-established disciplines, is audacious if not reckless
(and may explain the difficulty of finding a publisher for his thesis in
Germany [ibid., p. xxi]). It calls for a radical reform of the linguistic sci-
ence along the lines of philosophical reflection on the basic and neces-
sary principles regulating the language phenomenon itself, rather than
inductive hypotheses based on facts accumulated without any guidelines
and criteria for setting up the comparisons and making derivations of
common traits. It raises the question of what constitutes a science of lan-
guage worthy of its name, and how the empirical and theoretical pursuits
should be related. Ultimately, Kruszewski concedes that even if the exist-
ing linguistic sciences were to retain their validity, the task of capturing
the laws which govern linguistic phenomena still lies ahead; these gen-
eral laws can then be put to the aid of specialized sciences, which study
the relationships of Indo-European languages and seek to reconstruct a
protolanguage.
In his view, the existing sciences of language are in a state of crisis due
to a lack of clarity about the guiding principles of study, and a commitment
to unexplicated and dubious axioms. Kruszewski diagnoses the following
false axiom embraced in the comparative study as if were an irrefutable
truth: “everything which was similar in languages a, b, c . . . was original
and inherited by them from their common ancestor language, language A;
everything which was dissimilar was secondary and arose later, within the
individual languages” (ibid., p. 44, italics in the original). For example, the
sound p of the Sankrit pitar-, the Greek pater-, and the Latin pater- would
be traceable back to an assumed common ancestor in the protolanguage.
However, such simple inductive derivation of prototypical features from a
series of common traits proves invalid in light of cases such as the Italian
pesce (fish) and the related German Fisch even though they share the same
sound. This commonality does not imply, however, that the shared sound is
derived from a common ancestor—the Italian pesce having developed
from Latin piscem, while the German Fisch from proto-Germanic fiska
(which makes the s sound derive in each case independently from the sk
sound combination).
Kruszewski concludes that “the simple empirical evidence of compari-
son is insufficient; at every step we need the assistance of deduction from
firmly established phonetic and morphological laws” (ibid., p. 45, italics
in the original). The inductive method based on accumulated facts must
then give way to a deductive science, grounded in a set of generally valid

and necessary laws—the linguistic a priori, which is used to guide empiri-
cal research and cannot simply be read off the data.
The practitioners of the special sciences dealing with language are not
themselves best suited for the task of scientific reform; their training in
philology and history leads them to adopt the views, aspirations, and meth-
ods of these disciplines within linguistic study, and to adopt the dominant
aim of historical and philological disciplines, that is, “the clarification of
our view of the past.” Such a focus naturally leads to a neglect of modern
languages, both present and past, and to a lack of consideration for the im-
mediate links between modern languages and the languages reconstructed
from the past. To this Kruszewski programmatically responds:
If . . . our aim is not to reconstruct the past but, in general terms, to discover
the laws which govern phenomena, then everything which is recent—
everything, in fact, which happens before our eyes—acquires a special sig-
nificance. It is either very difficult or completely impossible to discover
phenomenological laws [die Gesetze der Erscheinungen zu entdecken] in
what is dead or of the past. Under these circumstances we should not be
surprised that linguistics is impoverished with regard to generalizations.
(Kruszewski, 1995, p. 7)
This impoverishment is tied to methodological constraints: an exclusive

focus on past languages implies a preference for an inductive method.
Since the scholar of classical languages has no access to their phonetics,
that is, the sound elements of the signs employed by their speakers at the
time of usage, such facts can only be gathered in a piecemeal fashion, by
induction (ibid., p. 9). The study of living languages offers a more propi-
tious terrain for locating generally valid scientific laws than an inductive
reconstruction from written documents ever could (ibid., p. 48); the turn to
language as a living phenomenon is therefore an integral element of a sci-
ence of language in the true sense of the word.
This commitment to general phenomenology of language shapes
Kruszewski’s vision of his doctoral project, as outlined in his letter from
April 21 to May 2, 1882 to Baudouin de Courtenay. Kruszewski stresses
the need to develop a general science of language as an alternative to the
existing ones. This general linguistic science would be “a certain kind of
phenomenology of language” (coś w rodzaju fenomenologii je˛zyka), its
basis located in the language itself (moz·na znaleźć w je˛zyku trwale pod-
stawy takiej nauki) (Kruszewski, 1904, pp. 134–135; see also de Mauro,
2005, p. 340). The predecessors for this view would be found in the works

of Hermann Paul Prinzipien der Sprachgeschichte (Halle, 1880), which
focuses on language in general and its development rather than the recon-
struction of vanished ancestor languages. The specific emphasis on a phe-
nomenological orientation can be tied to Hartmann’s works. As Jakobson
notes, “The young linguist must have detected the concept of phenomenol-
ogy in Eduard von Hartmann’s Philosophie des Unbewussten (1875),
which H. Spiegelberg’s History of the Phenomenological Movement (1965)
views as ‘an isolated landmark on the way from Hegel to Husserl’” (Jako-
bson, 1971, p. 714). This philosophical inspiration sheds light on Krusze-
wski’s understanding of phenomenology as a science of a priori principles
whose field of study is situated within, even though it may extend beyond,
the immediate data of consciousness.
To the reader of Hartmann’s analysis (who draws on the works of
Schelling and Wilhelm von Humboldt), this is the paradox of language:
one the one hand, it is woven into the very fabric of our thought, and, on
the other hand, it resists conscious grasp. The phenomenology of language
will therefore partake in this paradox and occupy an ambiguous terrain: it
will be situated, partially at least, in the field of consciousness but it will
not be able to account for the phenomenon in terms of conscious produc-
tivity alone. Hartmann opens the chapter dedicated to the role of the un-
conscious in the origin of language (1893, pp. 293–300) with this poignant
reflection from Schelling:
As without language not only no philosophical, but no human conscious-

ness at all is conceivable, the foundations of language could not have been
consciously laid; and yet the deeper we penetrate into it, the more clearly
does it appear that its invention far surpasses in profundity those of the
highest conscious product. It is with language as with human beings: we
think we behold them come blindly into existence, and at the same time
cannot doubt their unfathomable significance even in the smallest particu-
lar. (cited in Hartmann, 1893, p. 293)
Language emerges here as an intimate stranger, which can never be

claimed fully as our own and yet emerges within the depths of our own
thinking. This ambiguously situates language both within and outside in-
dividual consciousness, and leads to the hypotheses of an unconscious
genius, a rational instinct, assumed as the underlying source of linguistic
ability in the works of Wilhelm von Humboldt as well as Heyse (Hart-
mann, 1893, p. 298). Language can be thematized as the hidden ground of
consciousness—not because all consciousness is linguistic, but because

consciousness of language finds language already speaking through it, is
enabled by it as it turns language into its object of study. This leads Hart-
mann to conclude, in agreement with Humboldt, that language points to a
common primitive foundation of universal spirit, of which an individual is
a “conditioned phenomenon” (ibid., p. 300). A phenomenological study
guided by this foundational aspect of language is therefore situated in an
ambiguous field which intersects consciousness with the unconscious, the
speaking subject with general conventions, individual expression in the
present with the intersecting trails of the past.
Saussure’s phenomenon
Once Saussure’s writings on general linguistics are resituated within their

own intellectual context, notably within the sphere of influence of the
Kazan School of Linguistics discussed above, it is less shocking and in fact
logical to discover a phenomenological orientation within his project.
Saussure’s repeated usage of the term phenomenon to describe language in
the Writings in General Linguistics (2006) may stem from a deliberate
choice to engage the phenomenological tradition in philosophy as part and
parcel of an envisaged reform of the linguistic science. In the Introduction
to the Writings, Carol Sanders acknowledges the broadly philosophical
approach to language adopted by Saussure in his manuscript writings,
where the ontological question of the nature or “essence” of language is
explicitly posed (Saussure, 2006, pp. xxv; [34], 17); this is in agreement
with the stated goal of Saussure’s general linguistics to serve as prepara-
tion for a philosophy of linguistics. Within these philosophical investiga-
tions, the essence of language will be defined as a phenomenon without
existence outside of the speaking subject. Saussure states directly that
“Language (langage) is a phenomenon” (Saussure, 2006, p. [129], 85);
henceforth, language is indissociable from particular language acts (actes
de parole), as performed by the speakers (ibid., pp. [129–130], 85–86).
Saussure draws therefore on resources from the phenomenological tra-
dition as part of his inquiry into the essence of language. Sanders com-
ments, “As a fluent speaker of German, Saussure was no doubt aware of
the contemporary resonance of the term ‘phenomenon’ in a philosophical
tradition with which he has not be usually been associated, that of Hegel
and Husserl . . . In particular his comments about the language act (cf. to-
day’s ‘speech act’) and the emphasis on the speaking subject, both show us
a different Saussure from the one most often associated with the Course”

(2006, p. xxv). Sanders perceptively adds that there are lines of research
worth pursuing in this regard (ibid.); the present monograph constitutes
one pursuit in this direction.
The term “phenomenon” is used in sufficiently precise a manner to
count as a terminus technicus in Saussure’s writings. Saussure distin-
guishes between internal phenomena, tied to consciousness, and external
phenomena, immediately graspable (Saussure, 2006, p. [17], 3). The sign
belongs to an internal perspective, as it is directly tied to consciousness,
without ever falling over to the side of brute physical facts (ibid., pp. [20–
21], 6). A subjective-objective duality pertains therefore to consciousness
itself when regarded through a semiological lens; here the primary struc-
ture is the relation between consciousness and linguistic facts (themselves
dual and complex). A sign like a familiar word is but “the habit acquired
by the speaking subject of making this string of sounds correspond to a
specific idea” (ibid., p. [281], 198); the sign exists only in relation to the
speaker (ibid., pp. [45], 26; [83], 56). The mention that a linguistic fact is
“determined for the consciousness of the speaking subjects” is ultimately
deemed superfluous, since the latter needs to be constantly presupposed
(ibid., p. [49], 29). The consciousness of the speaking subject is therefore
effectively assumed as a necessary prefix before any statement about ob-
jective linguistic facts—whether or not it is explicitly stated as such. It
forms a silent background to all claims made about language. However, it
would be a mistake to liken this inalienable stance of consciousness in lan-
guage to the usual transcendental prefix “I think” of the first person con-
sciousness found in Kantian and Husserlian phenomenology. Importantly,
the linguistic phenomenon is not (only) bound to the individual but also,
and primarily, to the social. Language—both langage and langue—must
be located “in their rightful place, that is in the speaking subject as a
human being and a social being” (ibid., p. [130], 86). The subject of the
language phenomenon is socially situated: “The individual, organized
with a view to speaking, may only use the vocal apparatus in the context
of his community—moreover, the individual only feels the need to use it
when interacting with that community . . . In this respect, then, the human
being is whole only through what he borrows from society” (ibid., p. [178],
120). The appropriate point of view within linguistic phenomenology is
then not confined to an individual thinker but situated at the intersection of
a speaking subject and the language community. Saussure states explicitly
that semiological phenomena are socially produced (ibid., p. [290], 203),
and undecipherable outside of an ever renewed social contract between the
speakers. It is therefore more useful to situate Saussure’s project in relation

to Hegel’s phenomenology, where the individual is similarly dethroned in
favor of the social situation, without abandoning the standpoint of con-
sciousness in the process.
One finds multiple instances of Saussure referring to the linguistic
phenomenon in the notes for a book on general linguistics (Saussure,
2006, pp. [224–228], 156–160). It figures as an entry in a planned index,
where Saussure distinguishes between phenomenon as an event or motus
and the resulting linguistic state (status); the former is tied to the dia-
chronic and the latter to a synchronic axis (ibid., p. [228], 160). In the
index, Saussure directly links the category of phenomenon with that of
motus or event rather than state, which could imply that synchronic lin-
guistics does not itself include the phenomenon category (ibid., p. [228],
160). Yet language events turn out to be as closely associated with the
resulting states just as synchronic and diachronic perspectives are corre-
lated in general linguistics.
For example, the linguist can observe the historical change of the Old
High German form faris to feris; this change can be interpreted by means
of a general law that “Old High German a changes to e in front of i: faru -
feris” (ibid., p. [223], 157). Such a law assumes however a linear notion of
language in time, wherein “linguistic phenomena constitute a single thread”
(ibid.); it is more in line with the complexity of language itself to posit an
earlier language state which includes faru as well as faris, and a later state
where it is faris (but not faru) that changes to feris. The resulting linguistic
arrangement will therefore feature a new relation: faru and feris, with only
one term having undergone a phonetic change, and the lateral relation being
reestablished between an unaltered and an altered term. Rather than posit
a direct link from event A at time X to event B at time Y, we need to pre-
serve the more complex, relational or textual arrangement at any point of
time, and record how this total arrangement gets refigured by local changes
in the system. It remains true that for example faris evolved to feris in the
Old High German, but account needs to be given of the fact that the related
term faru was preserved unchanged. The seemingly static linguistic form
is thus contingent on a perpetuation of the linguistic form in usage (or
status involves motus), it involves an active process and not exclusively the
final product, with the synchronic order being as much a “consequence of
our activity” (Engler, 1989, [2528], IIR 106, p. 379) as the diachronic evo-
lution is.
Historical development of language emerges then less as a series of iso-
lated strings of manifest mutations over time—the whole cloth is carried
along with them, and the remaining threads are implicated in the linguistic

history, even though they may remain exactly the same at time Y as they
were at time X (or may have been worn out and the cloth thread bare in
places). To be sure, the linguist can follow only one direction at a time—
the diachronic direction from event A to event B, or the synchronic one
from term a to term b ( faru-faris; faru-feris). But that does not force a
choice of following one sole direction in the course of linguistic study;
rather, the two directions of contiguous coexistence and evolutive change
must be preserved without being collapsed within the total language phe-
nomenon. In fact, when only one direction is opted for, the linguist either
exaggerates the historical aspect of language, imagining that it is simply
made up of phonetic accidents (simple events), or explains the (contingent)
event by means of principles applicable to a linguistic state, as in the initial
example of a linguistic claim that “Old High German a changes to e in
front of i: faru - feris” which projects strict regularity on a condition that
includes both an accident and an abiding, stable relation (Saussure, 2006,
p. [223], 157).
In the planned index, Saussure aligns the “phenomenon” with motus or
event, rather than state—to avoid this term being “transformed into an in-
admissible hybrid notion” (ibid., p. [228], 160). Yet similarly to the discus-
sion of the distinction between synchrony and diachrony in the lecture
notes on general linguistics, some ambiguity surrounds the distinction and
Saussure notes also, under the same index heading, that both event and
state “are in their respective spheres phenomena” (ibid.). The overriding
concern in the phenomenon as event and as state distinction is the one al-
ready traced in the distinction between synchrony and diachrony: to not
collapse the dual field of language into a unitary one (whether a linear suc-
cession of contingent events in time, or a fixed list of rules and principles);
if the term phenomenon was being used interchangeably in relation to
events in time and arrangements within the system, it would risk collaps-
ing the two perspectives into one.
His stated objections to “hybrid notions” notwithstanding, Saussure ties
the phenomenon directly to the language system, and thus critiques Seche-
haye for having misunderstood the grammatical problem of language and
for failing to capture the language phenomenon despite insisting on the
need to study language within psychology (Saussure, 2006, pp. [260–261],
186–187). General linguistics is therefore irreducible to a faculty psychol-
ogy, which would simply extend psychological description of mental pro-
cesses to language, and use thought as a yardstick for measuring the field of
expression—rather than attending to the proper laws governing this field.
Saussure makes a plea that the general linguist tease out “their specific,

whole unique properties within the phenomenon of language.”3 The lin-
guist is to remain faithful to the grammatical, that is, rule governed and
systemic dimension of language itself, but always in its bond to conscious-
ness (something that, in agreement with Sechehaye, Wundt failed to do
justice to in his introspectionist account of consciousness). In sum, the
notion of phenomenon is of general linguistic value, and applies both to the
synchronic and diachronic facets or threads within language—so long as
the two threads are not collapsed into a single line.
Effacement of consciousness in the Course
In this section I’ll develop the idea that the phenomenological dimension
of Saussure’s general linguistics is hard to glean from the Course due to an
effacement of references to consciousness and subjectivity. I will consider
how, having arguably established general linguistics as a science, the
Course puts forward specific claims about linguistic processes (notably
concrete entities, linguistic value, and analogy), which systematically gloss
over Saussure’s own insistence on subjective involvement therein. The edi-
tors do maintain Saussure’s point that the study of language system (la
langue) is contingent on subjective consciousness as part of the distinction
between synchronic and diachronic linguistics.4 As the reader progresses
3
Mr Sechehaye rightly berates Wundt for underestimating the problem of grammar, but himself
fails to grasp the idea sufficiently . . . The more the author tries to remove what he sees as an
unjustifiable barrier between the thought-form (la forme pensé) and thought, the more we see him
abandon his own aim, which would be to identify the field of expression, and to discover its laws,
not in terms of what they share with our general psychic make-up (notre psychisme en général),
but indeed in terms of their specific, unique properties within the phenomenon of language.
(Saussure, 2006, pp. [260–261], 186–187)
4
Part I, ch. III, section 5, Synchronic and diachronic linguistics: their methods and principles
contrasted:
Synchrony has only one perspective, that of the speaking subjects; and its whole method
consists in collecting evidence from them. In order to determine to what extent something is a
reality, it is necessary and also sufficient to determine to what extent it exists for the
consciousness of the subjects. (Saussure, 1986, p. [128], 89)
And subject matter:
Synchronic linguistics will be concerned with the logical and psycho-logical relations linking
coexistent terms and forming a system, such as they are perceived by an identical collective
consciousness.
Diachronic linguistics, on the contrary, will study the relations linking successive terms not
perceived by an identical collective consciousness and which replace each other without forming
a system among themselves. (ibid., p. [140], 98)
The bond between the perspective of the speaking subject and la langue is also maintained in
part I, “General principles,” section 2 (“Internal duality and the history of linguistics”), where it
is admitted that the linguist must suppress considerations relative to the past in order to “enter
into the consciousness of the speaking subjects” (ibid., p. [117], 81, my translation).

from this methodological distinction in part I to the field of synchronic
linguistics discussed in part II of the Course, however, the subjective in-
volvement tends to be suppressed—as if the choice of synchrony had per-
mitted access to an autonomous objectivity of language itself, with its
mechanisms and principles. This objective focus is in agreement with the
overall stated goal of studying language (la langue) “in itself and for its
own sake,” as formulated in the concluding lines of the Course.
The shift towards objectivity is apparent already in the introduction,
with references to “consciousness” being replaced by a “concept” in the
discussion of the circuit de la parole, against the material from the lecture
notes and manuscript notes (Saussure, 1986, p. [28], 11). One reads that:
The starting point of the circuit [of speech] is in the brain of one individual,
for instance A, where facts of consciousness which we shall call concepts
are associated with representations of linguistic signs or sound patterns by
means of which they may be expressed. (ibid.)
This passage (with no basis in either the student or the manuscript notes)
(Engler, 1989, [197], p. 37) directs emphasis to the sign objectively con-
strued, whose conceptual content could be studied independently of sub-
jective grasp. This is in marked contrast to the manuscript sources, accord-
ing to which the “constant view” (le point de view constant) in linguistics
is that both meaning (signification) and the sign are “fact[s] of pure con-
sciousness” (Saussure, 2006, p. [19], 4). Conceptual analysis of signs is not
advocated, since it would loosen the bond to consciousness.
Having replaced consciousness with concepts, the editors have free
hand to gloss over references to the former throughout the remainder of the
book, and to direct the emphasis to the sign (and the sign system) objec-
tively construed. For example, in part II, ch. V, section 1 “Syntagmatic and
associative relations,” the source materials emphasize that the associative
relation is a relation of consciousness (lien de la conscience), not of space
(Engler, 1989, [2000], D 265, S 2.38, III C 384, p. 282). The reference to
consciousness is missing in the relevant section of the Course; instead a
notion of “mnemonic virtual series” is introduced—a purely “conceptual”
notion, susceptible to objective study.
An objectivist tendency is found also in the discussion of concrete enti-
ties in language (Saussure, 1986, pp. [144–149], 101–105). The editors note
the difficulty of identifying concrete entities in language (e.g., cheval-
chevaux; ibid., p. [148], 104), and highlight the difference between natural
disciplines like zoology, astronomy, chemistry where a preexisting unitary

object can be identified (the animal, the stars and planets, molecular struc-
ture of particles), and linguistics, where no such preexistent entities exist.
Despite this stated difficulty, language is presented as an object of study in
linguistics:
The signs comprising a real language are not abstractions, but real objects.
Linguistics studies these objects and the relations between them. They may
be termed the concrete entities of that science. (Saussure, 1986, p. [144], 101)
The manuscript source for this passage emphasizes that the concrete
entities in science are also spiritual, that is, bound up with the spirit or
consciousness of the speaker (Engler, 1989, [263], D 180, S 2.6, J 152,
III C 272, p. 44). As such, they do not fit into the object category in a nat-
ural scientific sense but are contingent on the point of view of the speaking
subject. The Course glosses over this subjective dependency in favor of
objective processes; the only mention of the speaking subject in this chap-
ter effectively serves to trivialize it:
Anything which is significant in any way strikes [the speaking subjects] as

being a concrete unit . . . However, it is one thing to sense (sentir) this rapid
and subtle interplay of units, but quite another thing to give an account of it
by methodical analysis. (Saussure, 1986, p. [148], 104)
This passage figures the speaking subject as easily impressionable and

not a reliable guide in general linguistics. The speaker may well sense con-
crete units in language, but offer no clue how the serious business of me-
thodical analysis of signs is to proceed. The trivialization of the subject in
the second sentence is however an editorial insertion with no basis in the
student lecture notes or manuscript sources (Engler, 1989, [1738], pp. 239–
240). According to the sources, the speaking subject’s felt sense (sentiment)
consistently serves as arbiter of what is concrete in language (Saussure,
1957, p. 41). The student lecture notes all agree on this point that the con-
sciousness of the speaking subject serves as the basic methodological prin-
ciple in linguistics since it alone settles what is signifying in language and
what is not (Engler, 1989, [1737], II R 42, G 1.10b, B 26, II C 34, p. 239).
Signification is not an objective attribute of the sign, but contingent on the
semantic sentience of the speaking subject; the term sense should then be
heard simultaneously as the “objective” meaning/signification (as in “it
makes no sense”) and “subjective” sentiment (as in “having good sense”
and the connection to sense perception or “sensing”). Only by investigating

language insofar as it is felt by the speaking subject as being significant can
general linguistics remain anchored in concrete reality and avoid traffick-
ing in abstractions. Sense is undecided between the speaker and the signs,
and so the linguist is continually reminded to maintain the speaking sub-
ject’s position in language, including “dead” languages like Latin.
One finds another example of editorial effacement of references to con-
sciousness in a discussion of linguistic differences in the chapter dealing
with linguistic value in the Course (part 2, ch. IV, section 3). According to
the manuscript sources, consciousness perceives a differential relation be-
tween terms and not a single term in isolation:
While not immediately obvious, it becomes very clear on reflection that it

is precisely because the terms a and b are incapable as such of reaching the
regions of consciousness, which persistently perceives only the difference
a/b (laquelle n’apercoit perpétuellement que la difference a/b), that each of
the terms considered on its own remains open (or becomes free) to change
according to laws other than the ones resulting from constant penetration
by the spirit (d’autres lois que celles qui résultaraient d’une pénétration
constante de l’esprit). (Saussure, 2006, p. [219], 153)
In the Course the last line reads: “each sign remains free to change in
accordance with laws quite unconnected with their signifying function”
(Saussure, 1986, p. [163], 116). This passage (without a source in the stu-
dent notes) shifts emphasis from the point of view of consciousness and its
laws, to the laws tied to the sign and its signifying function, considered as
an objective factum.
Similarly, references to the speaking subject are glossed over in the dis-
cussion of analogical innovation in the Course (Saussure, 1986, p. [227],
165). According to Riedlinger’s lecture notes, the formation of a novel lin-
guistic form in analogy with the existing ones supposes the speaker’s felt
sense (sentiment) that the latter are signifying: On peut supposer que les
éléments existent pour la conscience de la langue (Engler, 1989, [2526],
I R 2.89, pp. 377–378). Thus a nonexistent but feasible word like in-
décorable, which is made in analogy with other in- prefixed words, is com-
posed of fragments readily available to the speaking subjects (a la portée
des sujets parlants) (ibid.). This subjective point of view is excised from
the presentation of analogical formation in the Course; it is simply stated
that an improvised word, like in-décorable exists potentially in language
(la langue) (Saussure, 1986, p. [227], 165). The linguistic potentiality ap-
pears therefore as an objective factum.

The objectivist tendency is found also in the presentation of classifica-
tion of language into basic parts like prefix, stem, and suffix. One reads in
the Course that:
The prefix precedes that part of the word recognized as its stem (radicale):
e.g. hupo- “under” in Greek hupo-zeugnumi (“bring under (the yoke)”). The
suffix is the element added to the root (racine) in order to form a stem (radi-
cale). (Saussure, 1986, p. [257], 186)
The corresponding passage in student lecture notes reads thus:
We observe . . . that <the prefix> is always at the head (en tête) of the word;
the prefix is recognized by its placement in front of the part recognized by
the consciousness of the speaking subject as stem. (Engler, 1989, [2817],
I R 2. 42, p. 427)
Further down in the same passage one reads that classification of basic
parts is not solely a grammarian’s task; speaking subjects already have a
sense of how language subdivides into parts. The corresponding section in
the Course drops the reference to the consciousness of the speaking sub-
ject, and shifts emphasis to the segmentation within the words themselves.
In sum, the editorial presentation partially projects the received concep-
tion of scientific objectivity onto the source materials and effaces the point
of view of the linguistic subject. This tendency is in line with a sole focus
on la langue as the proper object of study, but it introduces an incongruity
into the Course between philosophical reflection on the subject’s involve-
ment in synchronic linguistics on the one hand, and the received notion of
scientific objectivity on the other. Since the former idea is discussed as part
of the broad methodological considerations about the science of linguistics
but largely abandoned once the attention turns to specific linguistic facts,
like concrete entities, linguistic value, and analogy, the reader is likely to
conclude that consciousness and subjectivity can recede into the back-
ground once the foundation for the science of general linguistics has been
put into place.
Consciousness structured like a language
While references to consciousness tend to be effaced from the Course, the

student lecture notes and Saussure’s autographed writings advance the

claim that language is consistently tied to consciousness, in experience and
scholarly study alike. In fact, Saussure’s own writings are populated with
references to “consciousness” and “semi-consciousness”—terms which
belong primarily to the emerging discipline of psychology; his exposure to
the Wilhelm von Wundt’s experimental studies on consciousness in the
late 1870s at the University of Leipzig may explain in part the provenance
of this psychological interest.5 In a rare admission of intellectual debt,
Saussure notes that the writings by the psychologist Wundt, as well as
those by the linguists Wilhelm von Humboldt and Hermann Paul (on the
nature of words, prefixes, and suffixes) are “full of material which could go
towards the structure of the edifice [of linguistics], but that not one has
even sketched its foundations” (Saussure, 2006, p. [259], 185). It appears
also that Saussure envisioned general linguistics as a sister discipline of
psychology, and even forecast that the former will ultimately become
founded upon the latter: “Little by little psychology will take practical
charge of our science, because it will realize, not that langue is one of its
branches, but that it is the very basis of its own activity” (ibid., p. [109], 73).
Saussure’s multiple references to consciousness can be seen as direct ex-
pression of a psychological study of language.
The bond between consciousness and language is best developed by fo-
cusing on linguistic creativity which enables the production of innovative
forms of expression according to a model drawn from already established
forms and their relations; this phenomenon is technically termed “analog-
ical innovation (or creation).” While it might seem at first sight that linguis-
tic innovation would fall outside the domain of the language itself, Saus-
sure considers it integral to the functioning of language viewed as a system.
This approach is in agreement with a commitment to the inescapably dual
essence of language, intersecting linguistic stability with change over time.
A new element that is brought to light when focusing specifically on ana-
logical innovation is a direct link between the generative potential of the
language system on the one hand and the speaker’s expression and con-
scious motivation on the other. The discussion of linguistic change is
therefore directly related to a study of consciousness.
The reader of the Course would be hard pressed to glean the points just
made from the text. The placement of sections dealing with linguistic
5
Wilhelm Wundt pursued studies into consciousness in the first experimental laboratory of
psychology he established at the U. of Leipzig in 1879; Wundt was trained in medicine but was
appointed to the chair of philosophy in Leipzig in 1874; Saussure obtained his doctoral degree
from U. of Leipzig in 1880.

innovation in the book invites a limited interest in the phenomenon itself,
and creates an impression that the latter is of lesser import than the princi-
ples subsumed under general linguistics and synchronic linguistics. The
relevant discussion occurs in part III, “Diachronic linguistics,” following
the influential part I, “General principle,” and part II, “Synchronic linguis-
tics.” Part III in its entirety is an amalgam of disparate source materials:
the editors draw on a lecture from the first course (Godel, 1957, p. 61), two
lectures, relatively late, from the second (ibid., pp. 70; 74), and a lecture
from the third (ibid., pp. 78, 100). The editors refer to, in an order of their
own making, material from the lectures of the first course in the following
chapters, notably dealing with analogy (ibid., pp. 57–63). While Saussure
treats linguistic innovation in the first course, prior to establishing a dis-
tinction between synchronic and diachronic linguistics, the editors reverse
this order of presentation. This reversal creates an impression that dia-
chronic considerations of linguistic innovation and creativity are added on
après coup to a set of self-standing axioms setting the parameters of a
static language system (see de Mauro, 2005, p. 269n, for additional refer-
ences to the source materials and insightful comments). Furthermore, the
reader of the Course is unlikely to appreciate that analogy is construed as
“a general linguistic principle applicable to language (la langue)” (title of
the relevant section from the first course of Saussure’s lecture in general
linguistics); the popular Harris translation has “Analogy as the creative
principle of languages,” which misses the general aspect of the principle,
and suggests an observation from comparative philology; Baskin’s transla-
tion has language in agreement with the French edition, but neither has
retained the “general” aspect of the principle, which matters when the field
of study is general linguistics.
Analogy is a term borrowed from the Greek grammar, and distin-
guished from a simple anomaly. According to Saussure, linguistic innova-
tion constitutes a principled or reasoned process and not a one-time acci-
dent; it is contingent on active deployment of general linguistic rules and
not solely a result of fortuitous chance.6 In the first course of the lectures,
6
Cases of mere chance are found within phonetic change termed Lautverschiebung in German,
where one existent, valid form becomes substituted by another (e.g., Saussure, 1996, pp. 29, 31,
32). These cases involve an empirical process which is not rule governed; it is indifferent to the
linguistic meaning and the grammatical function of the material in question. This basic distinction
gets fleshed out in Saussure’s second lecture at the U. of Geneva from November 1891 thus:
These two great factors [of phonetic and analogical change] in linguistic renewal may be
opposed from many different viewpoints; we may say for instance that the first represents the
physiological and physical side of speech while the second reflects the psychological and mental
side of the same act, or that the first is unconscious, while the second is conscious, always

Saussure defines it as a psychological as well as a grammatical phenome-
non (Saussure, 1996, pp. 56, 63); analogy is psychological in that it can be
explained by the propensities of the human mind (even though it does not
depend on deliberate intention); it is also grammatical in that it “presup-
poses consciousness, the understanding of the relation between forms,
which implies that the forms are considered conjointly with the ideas that
they express” (ibid., p. 64).
For example, an author writing in French may coin a new term, such as
répressionnaire (in analogy with mission: missionaire = répression: X), or
an adjective firmamental/firmamentaux (from firmament; in analogy with
fin–final/finaux) (ibid., p. 62). These cases constitute veritable linguistic
creations since the novel form does not replace or modify a preexistent one
(there was no akin term prior to the innovation). Yet the novel form does
not emerge ex nihilo either but is guided by the established pathways con-
necting nouns for states and participants; nouns and adjectives; the singu-
lar and the plural forms of the nouns; and so on. Such analogical innova-
tiveness can be found within any literary, artistic creation (ibid.), and is
therefore directly relevant to the creative usage of language in literature.
It is usual to regard Saussure’s intellectual influence, notably his Neo-
grammarian teachers, as being antipsychologist due to the latter’s insist-
ence on strict phonetic laws (“Every sound change takes place according to
laws that admit no exception”). However, this view is inaccurate consider-
ing that they “assigned the central role in ad hoc changes to the process of
analogy, which is grounded in a psychological principle that associates
similarity in form with similarity in meaning or function” (Joseph, 2012,
p. 227). Saussure follows their lead when he terms analogical innovation—
and the kindred psychological process of association among similar or
contiguous ideas—a basic principle of linguistic activity. He ties it directly
to consciousness, or alternatively, the spirit:
An analogy provides the basis of the thinking behind the phenomenon. In

more general terms, the phenomenon represents an association of forms in
bearing in mind that the notion of consciousness is highly relative, such that there are two
degrees of consciousness, the higher of which remains that of pure unconsciousness (est encore
de l’inconscient pure) when compared to the degree of reflection which accompanies most of our
acts; these two orders of facts are also frequently opposed by saying that one concerns sounds
and the other grammatical forms, an unclear idea since the forms of a language can only be
sounds, though one can say that one [view] targets the form via the sound and the other via the
idea; one can also say that one represents purely mechanical operations, where one finds neither a
goal nor an intention, and the other intelligent operations, where one can discover a goal and an
intention. (Saussure, 2006, p. [159], 106, translation revised)

the spirit (dans l’esprit), which is dictated by the association of ideas rep-
resented. (Saussure, 2006, p. [161], 107)
The Neogrammarians adopted the conviction that associative combina-

tions of similar or contiguous ideas dominate the functioning of the mind
from the British empiricists, notably David Hume. For example, Herman
Paul noted:
Ideas are introduced in groups into consciousness, and hence as groups

remain in unconsciousness . . . These groups furnished at least originally by
the exterior world, now proceed to organize themselves in the mind of each
individual into far fuller and more complicated combinations, which come
to fulfillment for the most part unconsciously, and then proceed to operate
unconsciously. (cited in Amsterdamska, 1987, p. 184)
The Neogrammarians extended the psychological principle of associa-

tion to language organization and usage, and Saussure later adopted their
view that analogy is “basic to the psychological functioning of language
and that it reflect[s] the organization of language in consciousness” (ibid.,
p. 189).
The associative or combinatory activity allows individuals to speak by
bringing the existing linguistic forms into a manifold of combinations, as
well as to use language creatively, in an unprecedented fashion. Associa-
tion is therefore both a principle underpinning any individual speaking
activity, no matter how habitual and mundane, and a vehicle of linguistic
innovation in analogy with the existing models, where the already estab-
lished associative links pointed the way to novel forms of usage. The Neo-
grammarians believed that “the factors responsible for linguistic change
must also be active in everyday linguistic activity,” and an awareness and
understanding of the relations between linguistic forms is implied in the
two cases (ibid., p. 188). Analogical innovation is therefore not only a spe-
cialized case of creativity within literary language, but pertains just as
much to language in actu, where grasping linguistic meanings, ideas, and
their relations enables unprecedented yet understandable instances of
usage in ordinary speech. In sum, analogy may offer a window onto the
language of literature and the language of ordinary life, which harbors a
potential for inventive expression as well.
The formation of novel expressions in analogy with preexisting ones
may be most typical of the style of speakers least attached, accustomed, or
able to follow a fixed order of idiomatic phrases and familiar turns of

phrase: the poets, the foreigners, and the children. One often gleans the
deepest insights about one’s own language from those positioned on its
margins. They are less likely to suffer from—to revise Chomsky’s famed
phrase7—the linguistic blind spot of the (mature) native speaker, that is,
the habituation to the established linguistic patterns and norms from time
immemorial. This habituation may run so deep as to seemingly detach the
familiar patterns from fluctuating and contingent conventions of usage,
with their mixed heritage, and thus take on the guise of preordained natu-
ralness, which would sanctify them as the solely admissible into the canon
of correctness. As is well known, Chomsky equated grammaticality with
what is “acceptable to a native speaker” (Chomsky, 1975, p. 13). From the
standpoint of Saussure’s general linguistics, however, such claims of ac-
ceptability (to mature native speakers) remain at the mere surface level of
language, concerned as they are with the normative nativist expectation of
correctness; as a result, they miss the larger and more complex view of
language as an evolving and intermixed field of expression. Such claims
should therefore not be used as absolute measure of grammaticality; a
deeper insight into grammar is gleaned from the language’s generative po-
tential and openness to countless revisions and reappropriations—as in a
case of apparent violation of the standard norm. The scholar of language is
urged to loosen the blinders tied to the (mature) native point of view and its
naturalized intuitions; being on the margins may offer a better vantage
point onto language and grammaticality.
Consider children’s language as a case in point. Saussure notes that
“[n]o better idea [of the phenomenon of analogy] is given than by listening
to the speech of a three or four-year old child” (Saussure, 2006, p. [160],
107). For example, a child unfamiliar with the first person singular form of
the verb to come in the French, may draw on the existing knowledge of
other verbs and their first person singular forms and use them as model for
building the requisite expression. This child could generate a (nonexistent)
formation venirai (from venir, “to come”) on the model of similar sounding
verbs, like punir or choisir whose first person singular are respectively: je
te punirai and je choisirai. The link venir—je venirai is thus projected in
accordance with a recognizable and generalizable pattern. The novel form
is grammatically incorrect (the correct one being viendrais), and on hear-
ing it, the native expert should simply correct the child’s mistake. However,
there is a logic to the child’s superficially ungrammatical utterance—it is
7
“linguistic intuition of the native speaker” (Chomsky, 1965).

rule-governed and hence an “intelligent transformation” (ibid.), which per-
ceives and puts to work the manifest regularities within and between indi-
vidual words with a degree of inventiveness. It does not simply grasp any
of the individual words (like punir, choisir, venir) in isolation and it does
not regard the similarities between them as based solely on their physical
physiognomy. It detects and deploys the grammatical relations connecting
individual terms like a spiderweb, and does so in a consistent (albeit so-
cially unsanctioned and historically nonsedimented) fashion. The inven-
tiveness by analogy is true to the deep grammar of language while violat-
ing surface correctness: “There is nothing more consistent, nor more
logical and more accurate, than the reasoning that leads to venirai” (ibid.).
Analogical formations are found not only in children’s speech, they are
also a motor behind historical change in language. Henceforth, there are
historical cases of new and accepted grammatical forms emerging accord-
ing to the process employed in a surface mistake. For example, the existing
grammatical relation between
poussons and pousse
may exercise an analogical influence onto other existing relations, such as
trouvons and treuve
thus leading, through a process comparable to the fourth proportional, to

the later form trouve (Saussure, 1996, p. 64). The whole process rests on
understanding and deploying the grammatical relations between the “gen-
erating” or “inspiring” forms as a generalizable principle according to
which the new form can emerge as an X of the proportion.
Saussure compares this process to a drama with three characters: the le-
gitimate heir (e.g., treuve), the rival (e.g., trouve), and the collectivity which
engendered the rival form (e.g., poussons–pousse–trouvons) (Saussure,
1996, p. 61). In the first act of the drama, the rival form is installed next to
the earlier one, and so the two effectively coexist; it is only in the second act
that the primitive form falls into disuse and eventually disappears (ibid.,
p. 61). In contrast to phonetic evolution where a single linguistic form un-
dergoes change, and where the new form automatically annuls the old, an-
alogical innovation is contingent on a more complex linguistic arrangement
within a set of linguistic forms. It is irreducible to a transformation or meta-
plasm of an earlier form; Saussure deems it a creation or paraplasm where

an entire generative complex provides a matrix for the production of a new
linguistic expression, and where the form devised according to this bor-
rowed matrix can initially coexist with the earlier one as two out of the
many possible recipes for confectioning new linguistic forms (ibid.). Ana-
logical innovation is not a direct derivation of the new out of the old, which
would be a purely contingent and isolated phenomenon; it is a principled
change, predicated on an understanding of relations as a plan for projecting
similar yet different relational terms. The generative principle of language
is found in the replicable or iterable relation, and not in any one of the terms
standing on its own.
Saussure is therefore critical of the view that the generation of new lin-
guistic forms can be modeled on a birthing relation, like a procreation of a
child by a mother (e.g., the Latin cantare engendering the French chanter)
(Saussure, 2006, p. [153], 101). He uses the metaphor of fecundity in lan-
guage behind its fossilized form, a generative potential for new growth
which is likely to get covered over by too close an observance of serious
and correct usage. This fecundity extends beyond a one-to-one (body or
word) relationality to a more systemic one. It is the entire structure that can
be described as fecund or generative, that is, as bearing a seed of novel
form within it; a term unaffected by such a relational pattern would be
sterile, incapable of generating novel forms according to its own exemplar
(Engler, 1989, [2531], I R 2.80, p. 380). The mother does not give birth to
the mother tongue; she draws on its resources, and is also a child to the
tongue she speaks, given “the simple precondition that we all speak our
mother tongue” (Saussure, 2006, p. [153], 101).
To conclude, the two cases: child’s novel usage in superficially incor-
rect speech and the historical development of language both follow the
deep grammatical principle of novel formation in analogy with existing
patterns and principles. They differ only in that the former is not (or has
not been) socially sanctioned and historically habituated as acceptable
and correct usage; yet both the deviant and the accepted forms are legiti-
mate possibilities harbored within the deep grammar of language. Fur-
thermore, the deviant form is instructive in that it reminds the speaker that
things could be otherwise in language, without loss of grammaticality; the
surface grammar of acceptable usage is not a necessary and immutable
standard, but one synchronic slice within the diachronic life of language.
It matters little if the synchronic slice is all that a mature native speaker
finds acceptable at a given time; grammar is found in the depth of lan-
guage, within its acceptable forms as well as the innumerable generative

possibilities.8 From this perspective, a speaker should regard novel lan-
guage forms produced by an intermixing of a national language with that
of an immigrant population as an opportunity to gain insight into the
deep grammar of both languages engendered in their encounter, and not
as a deviation from a single standard of correctness. When one language
evolves in analogy with models drawn from another, the speakers can ap-
preciate the life of their language/s more than it is possible in cases of
faithful repetition of the standard form. Like literature and poetry, the
liminal language offered by a foreigner lets the native language speak of
itself, differently, and it redirects the native speakers to their field of ex-
pression by making it unhomely, no longer fully theirs, but not simply
alien either.
Saussure insists therefore that it is only on the surface that analogical
change emerges as a historical error and a mistake against language. In
enumerating the main errors committed by the linguists before him (from
Bopp to Schleicher, from 1816 to 1870–1875), Saussure lists insufficient
attention to “the incessant <daily> creation within the language [la
langue]; I mean analogy” (Saussure, 1997, p. 86). This lack of attention
accounts for the fact that linguists used to regard it as a violation of lan-
guage in its original state and only spoke of a “false analogy” (ibid.). For
the linguists before him, “everything that departs from the primitive order
seems not to be proper” (ibid.); there is an assumption that perfection was
reached within language’s primitive order, “without asking if the primi-
tive order was not preceded by another” (ibid.). In sum, “The language (la
langue) was . . . not allowed to use analogy; it seemed to be a kind of li-
cense, infraction, whereas it is the normal way for a language to renew
itself” (ibid., p. 86). Saussure issues a plea to recognize this “continual
work of renewal” and accept both the legitimacy and the universality of
analogical innovation (Saussure, 1997, p. 93); as put in a motto-like
phrase: “language always works (la langue travaille tout le temps); this
work is analogy” (ibid., p. 160). Innovation and renewal emerge therefore
as one of essential traits of the language itself and one of the key princi-
ples within the field of general linguistics (see conclusion of this section
for development).
8
Thibault notes, “The notion of synchrony shows . . . that variation and heterogeneity are
intrinsic characteristics of langue.” (Thibault, 1997, p. 90). He also links the idiosynchrony to the
temporal dimension, and to the evolutive potential within language. It would then be not only
individual instances of parole that enact instances of novel usage, but the fragmentation within
the language system is itself revealing of, as well as enabling, variation.

Thus far, analogical innovation has been explained by the principle of
the fourth proportional. Another possibility is offered by analysis wherein
individual words are broken down into smaller units that are sensed and
sanctioned by speakers as signifying to some degree, even though they
cannot be used as separate terms. The former principle focuses on the
whole word in relation to cognate forms (A is to B, as C is to D—punir is
to punirai, as choisir is to choisirai, and thus as venir is to venirai), the
latter principle focuses on the elements and the relations within each of the
words (A is composed of a and x, B is composed of b and x—pun-ir, chois-
ir, ven-ir; pun-irai, chois-irai, ven-irai). The fourth proportional does not
(need to) break up the perceived totality of the word, while the analytic
principle supposes a grasp of word parts. The component parts are grasped
as being both separable and recomposable units; hence one unit only, the
suffix can undergo a principled change (-ir–irai), while the stem remains
unchanged (pun-, chois-, ven-). In either of the two explanatory principles,
the final novel form is made in analogy with the existing elements and
their relations, as they play out between and/or within the words. They
differ mainly in their respective focus on the word as a bounded albeit
malleable unit, or as a more loosely conjoined pièce composé, whose com-
ponent parts may be grasped in relative isolation. The first principle is
more lexically, while the second more grammatically, oriented. Even
though Saussure at times views them as mutually exclusive (if one has ex-
planatory power, the other is redundant), they turn out to be as interde-
pendent as the lexicon and grammar are. In fact, they reflect the two dom-
inant ways of classifying language in grammar: either according to
complete words, or analytically, as in the modern Indo-European gram-
mar, and the Hindu and Sanskrit grammars, respectively (Engler, 1989,
[2546], I R 2.94, p. 381).
Consider why the principle of analysis is bound up with subjective
awareness in more detail. In the case of analysis of words into prefixes,
stems, suffixes, the latter are identified by the speakers’ sentiment. Saus-
sure terms them “the living units below the word” insofar as they do not
exist in themselves but in the consciousness of the speaker (Saussure, 1996,
pp. 71–72). A category like a prefix is not an objective notion established
by a grammarian, but constituted by subjective awareness of meaning and
tied to innovative usage:
Are there prefixes in language? This does not mean: have there been, or
again do grammarians distinguish them?, but: are there prefixes present in
the consciousness of those who use them? Certainly. I am justified in saying

that the French language knows the prefix re- (refaire, recommencer, re-
prendre). French knows also these units smaller than the word:
dé- (défaire, décomposer, décharger)
in- (inconnue, indelicate, indirect) (ibid., p. 72)
Saussure ties the different degrees of conscious awareness to the vary-

ing degrees of generative potential within existing language forms. This
point can be borne out by the subdivision of words into units like prefix,
root, and suffix. These traditional grammatical categories depend on con-
sciousness and usage, notably the potential to produce novel combinations
of familiar parts. Saussure notes:
The language (la langue) does not know the terms root (or radical), suffix,
etc. but we cannot deny it the consciousness and usage of these differences.
Analogical formation is evidence of this analysis made by the language (la
langue) itself, but it is necessary to realize the difference in function among
all these units, some of which are present in part, others completely present
to the consciousness of language (la conscience de la langue). (ibid., p. 70)
Relatively stable prefixes like the French re- are said to have a perfect
degree of signification and clear form since they can be deployed in the
service of improvised (yet understandable to a French language user)
forms such as redémissionner, recontempler, and so on (Saussure, 1996,
p. 72). One finds more ambiguous word parts, which belong to a lower lin-
guistic level because language (la langue) cannot endow them with the
meaning of a de- and recomposable terms. At the lowest level one finds
ambiguous cases, like séparer, séduire, sélection, where it is unclear
whether or not sé- functions as a prefix. It implies therefore only “<vague
consciousness that the element cannot be confused with a different cate-
gory>” (Saussure, 1996, p. 73, my translation). To settle its exact standing
one needs to determine whether or not it lends itself to inventive usage; as
the notes state: “The only absolute proof: usage made of <this prefix> by
creative analogy: to what extent could someone use a sé- in a new forma-
tion” (ibid., p. 73). Linguistic consciousness emerges therefore as a way of
doing things with words; its semantic sentiment is not passive recognition
but a putting of linguistic resources to work.
Analogical formation is bound up with consciousness both in the case
of socially sanctioned and historically sedimented words and nonexis-
tent but grammatically sound neologisms. Hence an improvised form

like in-décorable which “exists potentially (en puissance) in la langue”
(ibid.) draws its component parts from syntagms like décorer, décora-
tion: pardon-able, mani-able: in-connu, in-sense. Importantly, the latter
are said to lie within the speaking subject’s reach (a la portée des sujets
parlants), (Engler, 1989 [2526] I R 2.91, p. 377), and there is a degree of
consciousness involved already at the level of “comparison of materials
deposited in the store held by the language”; these materials are said to
“exist for the consciousness of the language” (pour la conscience de la
langue) (ibid., not included in the editorial version), and the will to con-
fection (confectioner) a new term from out of the existing ones. Further-
more, “la langue has consciousness not only of the elements but also of
the <influence> they exert one upon another when placed in a certain
order; la langue has the sense (sentiment) of their logical meaning (sens),
their order” (Engler, 1989, [2538], I R 2.92, p. 380).9 Instances of analogi-
cal innovation thus spring from the vast field of linguistic consciousness,
and are bound up with individual speaking subjects without being reduc-
ible to subjective intention (“It would be to misrepresent the whole of
psychology to offer <analogy as an> intention of speakers” [Saussure,
1996, p. 63]). The psychological nature of the analogical process is there-
fore not to be read in a narrowly subjectivist sense, and linguistic con-
sciousness functions to a degree as a prepersonal and anonymous field. If
Saussure ties analogy to consciousness, that does not imply that indi-
vidual consciousness is a master in the linguistic house.
Approached from the perspective of language acts, consciousness is
therefore not really opposed to the unconscious. As Saussure notes:
the notion of consciousness is highly relative, such that there are two degrees
of consciousness, the higher of which remains that of pure unconsciousness
(est encore de l’inconscient pure) when compared to the degree of reflection
which accompanies most of our acts. (Saussure, 2006, p. [159], 106)
Importantly, Saussure attributes a degree of unconscious automatism to

the language act; the latter is (typically) manifest to consciousness of the
9
Regrettably, the repeated mention of consciousness in the student lecture notes cited above is
excised from the Course—it is simply stated that an improvised word, like in-décorable exists
potentially in language (la langue) (part III, ch. IV, section 3 “Analogy”; Saussure, 1986, p. [227],
165; Engler, 1989, [2521], p. 376) and that the analogical creation can be explained by the fourth
proportional and/or analysis (part III, ch. IV, section 3 “Analogy,” Saussure, 1986, p. [228], 165;
Engler, 1989, [2538], I R 2.92, p. 380). The linguistic potentiality appears therefore as an
objective factum only.

speaking subject but it is not a direct product of conscious intention; the
speaking subject is animated by forces not of her own making, utterances
are sourced from grammatical depths that remain partially obscure and
independent from self-reflection.
Saussure casts therefore the distinction between consciousness and the
unconscious in language as one of degree rather than in kind:
There are many degrees of conscious or unconscious will; furthermore, of

all the acts which can be compared, the linguistic act (l’acte linguistique),
if I might call it that, is characterized by being the least reflected, the least
premeditated, as well as the most impersonal of all. That constitutes a dif-
ference of degree, which is so far-reaching as to have long appeared a fun-
damental difference, even though it is but a difference of degree. (Saussure,
2006, p. [150], 99)
The linguistic act is therefore ambiguously situated in the psychic space.

Insofar as it involves an understanding of meanings, ideas and their rela-
tions, it extends into the field of consciousness, but since it lacks the vigi-
lance and watchfulness of a self-reflective subject it belongs to the uncon-
scious. The linguistic act defies therefore a canonical phenomenological
equation of consciousness with self-consciousness; it comes closer to the
notions of pre- or nonreflective consciousness (Sartre) and anonymity
(Merleau-Ponty), as well as the psychoanalytic understanding of subjectiv-
ity as nonegological (Lacan).
It follows that consciousness in language is not an on or off phenome-
non but a gradational spectrum which extends into the subterranean levels
of unconsciousness at one end and the clarity of intuition at the other.
Hence the speaking subject has a feeling (sentiment) of the different ele-
ments composing a word, but this feeling emerges from the decomposition
of the word into component parts, which Saussure describes as being sub-
conscious; these elements are at the reach of the speaking subject who
exercises a degree of “industry” and “will” in the process (Engler, 1989,
[2526], I R 2.89, p. 377). Analogical innovation, and speech in general, is
therefore undecided between consciousness and the unconscious; the
speaker can pursue the process of decomposition and recomposition delib-
erately, but there is a degree of automatism to it as well; conscious involve-
ment is not (always) necessary but neither is it excluded in principle from
the generation of new forms; there is an “inner language” (langage in-
terieur) which precedes overt speech, but it is still a speaking to oneself (se
parler a soi-même) (ibid.). Thus an unconscious formation such as the

seemingly absurd neologism Norekdal which emerges in a dream indepen-
dently of deliberate intent can turn out to be meaningful thanks to an act
of interpretation, within a speech circuit shared by the analyst and the
analysand (see Freud, 1955).
Beyond the doctrine
Once we adopt a phenomenological approach within general linguistics,

and accept the principle of analogical innovation as a general linguistic
principle, we can transcend the confines of a “Saussurean doctrine” with
its emphasis on hierarchical dichotomies (signifier and signified, la langue
and la parole, synchrony and diachrony) and devise a more complex rela-
tional logic in its stead. According to the structuralist view, the language
system is closed and autonomous, but then analogical innovation in speech
would seem a sheer accident and an avoidable anomaly. Saussure regards
it however as a normal and permanent linguistic condition; in fact he goes
as far as to equate the totality of language with the process and the prod-
ucts of analogical formation:
Any language (une langue quelconque) at any moment is nothing other

than a vast web of analogical formations (un vaste enchevetrement de for-
mations analogiques), some very recent, others dating back so far that one
can only guess them. Asking a linguist to name some analogical formations
is therefore like asking a mineralogist to name some minerals, or an astron-
omer some stars; I say this at the outset so as to avoid any misconceptions
concerning the value given to these facts: they are neither exceptional nor
anecdotal, neither curiosities nor anomalies; rather, they are the most clear
substance (la substance la plus claire) of language (langage) everywhere
and at all time, its everyday history and the history of all times. (Saussure,
2006, pp. [161], 107–108, translation revised)10
If analogical formation acquires the status of a cardinal general linguis-

tic principle, interesting consequences follow for the entire field of general
linguistics: the stock of received dichotomies between diachrony and syn-
chrony, and language as speech and system, morphs into a web of com-
plex, crisscrossing, and nonhierarchical relations of interdependent terms.
10
Compare this forceful statement with the laconic line from the Course: “The role of analogy is
immense. It is always at work” (Saussure, 1986, p. 237).

The inherent logic of dichotomies and hierarchies associated with the
Course, wherein the first term is opposed and subsumed under the second,
thus gives way to another logic of chiasmatic interdependency between
nonhierarchical terms. First, analogy complicates, again, the distinction
between synchrony and diachrony. On the one hand, analogy is contingent
on the systemic relations of language which are deployed as a model; on
the other hand, the novel formation opens up a temporal relation which
cannot be contained within a synchronic arrangement: “Something is new,
therefore there has been a change. Here is an embarrassing question: if
there is change are we in the realm of the diachronic (nous nous mouvant
dans le diachronique)? We have indeed to say that this is a very delicate
point in the distinction between synchronic and diachronic” (Saussure,
1996, p. 58). Patois’ lecture notes describes this distinction as “a very dif-
ficult point to determine” (Saussure, 1996, p. 146), while Gautier’s notes
raise this double allegiance of analogy as “a self-contradiction” (Engler,
1989, [2591], G 2.28b, p. 390) (none of these complications transpire in the
published presentation of analogy in the Course). The analogical phenom-
enon itself is undecided between the synchronic and the temporal axes,
since too clearcut a distinction between them would effectively foreclose
the possibility of analogical innovation—and so would disentangle the
vast web of analogical formations that language is. To be faithful to the life
of language revealed in analogical innovation is to preserve both the (syn-
chronic) axis of the already instituted and relatively stable linguistic prod-
ucts and the (diachronic) axis of incessant praxis, which underpins the
daily process of language use, in the present and in the past.11 This criss-
crossing of synchrony and diachrony within analogical formation explains
why Saussure qualifies it as both an innovative and a conservative princi-
ple; analogy exhibits both a dynamic and a static force (Engler, 1989,
[2630], I R 2.98, p. 395).
The complication of synchrony with diachrony can only be introduced
by following the phenomenological lead of the speaker, from whose point
of view language is in actu or in process, whether or not it changes materi-
ally; it is for the speaker that language is undecided between the already
laid out net of grammatical pathways and the only partially mapped out
possibilities of individual expression; it is for the speaker that the stable
11
I therefore differ from Amsterdamska’s interpretation that Saussure established a firm
distinction between synchrony and diachrony in his discussion of analogy; this may be true of the
presentation in the Course, but not the source materials, where the distinction is more ambiguous,
and analogy emerges as a meeting point between the two axes.

and established language forms remain contingent on a speaking praxis—
whether the praxis involves repetition of the correct and the cliché, or a
playful, innovative, and partially unconscious speaking according to what
language can bear, and not only what is likely to carry the stamp of general
approval. The speaker is involved in any case, the motus intersects with the
status, the product with the process.
Second, analogy complicates the distinction between language as
speech (parole) and system (langue). It is rarely acknowledged in the
scholarship that this famed distinction emerges in the discussion of ana-
logical innovation in the first lecture course in general linguistics (Saus-
sure, 1996, p. 65). The distinction helps to map out analogical formation
according to the two facets:
1. the comprehension of relations between engendering forms (such as

nous poussons: je pousse = nous trouvons)
2. the engendered product, the X of the proportion (je trouve [previ-
ously je treuve]) (ibid., p. 64)
Only the form je trouve is executed within speech (parole); the enabling
forms rest at the level of the received fund, reservoir or treasury of language
(la langue) (ibid., pp. 64–65). Hence the distinction between, on the one
hand, “everything we say owing to the needs of discourse (tout ce qui est
amené sur les lèvres par les besoin du discours) and by an individual opera-
tion” and, on the other, “everything which is contained in the brain of the
individual, the deposit of forms heard and used and their meaning” (ibid.,
p. 65). Introduced in the first lecture course, the distinction between la
langue and la parole thus serves to disambiguate within the gradational
spectrum between the unconscious store of linguistic meanings, ideas and
their relations and the consciously available products. The former is consti-
tuted by the engendering forms which are located at the level of “half-
unconsciousness” (seulement pensées <ou plutôt senti dans une demi-
inconscience>) (ibid., p. 64); they do not translate into speech but remain
“subconscious, in the depths of thought” (ibid., p. 65). The engendered form
is alone produced in speech, and as such directly manifest to consciousness.
Still, the entire analogical process is spread out on a gradational spectrum
within the field of linguistic consciousness, between the hidden and the
manifest, the praxis and the product. Linguistic consciousness admits of
degree, and its manifest surface extends into a deeper grammatical dimen-
sion of speaking subjectivity. The linguistic unconscious is therefore not op-
posed to consciousness, just as a language act extends beyond an individual

signifying intention to a source within the deeper “half-unconscious” or
“subconscious” relational web. Consequently, la langue and la parole are
not really opposed; as previously noted, speech (parole) is effectively unde-
cided between consciousness and the unconscious.
The distinction between la langue and la parole is therefore not estab-
lished as an a priori claim but is arrived at by way of analogical innovation,
as a working principle. In contrast, in the official presentation in the Course,
“analogical formations are mentioned as providing historical evidence
(CLG 227) for a distinction [between langue and parole] already estab-
lished on a priori grounds” (Harris, 2003, p. 30). The distinction can there-
fore figure as yet another axiom within a deductive system—rather than a
double axis helping to map out a complex phenomenon whereby estab-
lished patterns give rise to novel expressions. Insofar as most scholars un-
critically accept the editorial presentation of the la langue/la parole dis-
tinction in the Course, and its structuralist uptake, they are unlikely to
trace its provenance in the lectures notes: “Today it is still probably true
that the majority of scholars who discuss ‘Saussure’s’ distinction between
langue and parole are completely unaware of its original emergence from
the doctrine of analogy” (ibid., p. 30). The majority of scholars would
therefore be hard pressed to appreciate how its original emergence from
the doctrine of analogy complicates the assumed logic of vertical dichoto-
mies associated with the distinctions between la langue and la parole, syn-
chrony and diachrony, the semiological system and speaking subjectivity.
According to the relevant sections from the Course, analogical innova-
tion is first noted in the sphere of speech, but this sphere is deemed to lie
“on the fringe of the language” (en marge de la langue) (editorial insertion
without manuscript source, Engler, 1989, [2518], p. 375); analogical inno-
vation is thus used as evidence of a presumed, already established hierar-
chy between the main linguistic field and the margins.
The editors thus insert this conclusion, of their own making: “Analogy
teaches us once again, then, to separate the language itself (la langue) from
speech (la parole). It shows us how speech depends on the language, and
allows us to put our finger on the operational linguistic mechanism” (Saus-
sure, 1986, p. [227], 164, translation revised; Engler, 1989, [2521], [2522],
p. 376).
The editors envision analogical innovation as a two step process: “Any
creation has to be preceded by an unconscious comparison of materials
deposited in the store held by the language, where the sponsoring forms
are arranged by syntagmatic and associative relations” (Saussure, 1986,
p. [227], 164).

The reader is likely to regard this “unconscious comparison of materials
deposited in the store of language” to be an objective process, best cast as
an “operational linguistic mechanism,” independent from speaking activ-
ity and removed from speaker consciousness. This is all the more the case
since the complex interdependency between consciousness and the uncon-
scious discussed in the source materials does not transpire in the editorial
vision of general linguistics. The reader is therefore likely to presume, with
no evidence to the contrary, that the two are distinct if not opposed.
In Saussure’s own account of analogical innovation, the speaker is not
simply expulsed from the domain of language proper onto linguistic mar-
gins; in fact, the analogical process would never get off the ground were it
not for the way a speaker’s consciousness extends into the depths of gram-
matical relations. The assumed hierarchical opposition between speech and
language is complicated by the source materials. According to the latter, all
language facts, especially the evolving ones, need to be regarded from the
side of speech and from the side of the reservoir of already established
forms (Engler, 1989, [2521], I R 2.23, p. 376). According to Riedlinger’s
lecture notes, language provides the elements of analogical formation, it
being a mistake to believe that the analogical formation emerges at the in-
stant it is made (ibid., [2525], II R 103, p. 377). The mistake consists, how-
ever, in a sole focus on the speech event at the expense of its preconditions;
that does not mean, however, that speech simply instantiates all the possi-
bilities laid out in advance.
According to Riedlinger’s notes, only an engendered form (such as je
trouve) is manifest and executed in speech. Here is the consequence: “We
have therefore to come face to face with the act of speech (se mettre en face
de l’acte de parole) in order to understand an analogical creation” (Saus-
sure, 1996, p. 65). The notes continue, with a tone of slight mockery, that
this new form “is not created in a meeting of scholars discussing the dic-
tionary”; rather, someone has to have improvised it in speech (a l’occasion
de la parole, du discours) (ibid., p. 65). The ordinary speaker’s novel usage
constitutes therefore a driving force and a sine qua non of analogical for-
mation; only as executed in a public medium can a novel form penetrate
the systemic dimension of language (la langue). There is therefore little
evidence that the speaker’s improvisation is a mere fringe added onto an
autonomous linguistic process. It is emphasized rather that the speech en-
genders new forms and rewrites the language code:
If it is true that we always need the fund of the language (la langue) in order
to speak, reciprocally, everything which enters the language (la langue)

was first essayed in speech a sufficient number of times for a durable im-
pression to have resulted: the language is but the sanctioning (consécration)
of what has been evoked <by> speech (la parole). (ibid., p. 65)
La parole is therefore a formative element of the entire analogical

process—and by extension, of langue construed as a web of analogical
relations. It turns out that a speech act is not narrowly tied to the context
of utterance, but inherently iterable. The utterance carries within it a po-
tential for infinite repetition and sedimentation into the socially shared
code of language (itself revisable, and contingent on social practice). The
usual opposition between speech and system is therefore undone by the
generative and conserving potential of the speech act itself; the speech act
is enabled by and yet it also rewrites the (enabling) code by an innovative
practice of usage. Just as the system of language is not a closed repertory
of products but a field of practical possibilities, so is the domain of speech
not limited to the here-and-now event of saying something to someone.
Any speech act carries an echo within itself, what has been said once can
be said again, by others, at other times, in other places. Speech contains
therefore the minimum requirements of rationality—repetition, and dif-
ference. To return the distinction between la langue and la parole to the
development of ideas in the course on general linguistics is therefore to
debunk its structuralist definition as a dichotomy between the deep, ena-
bling conditions and the epiphenomenal surface manifestations. Contrary
to the structuralist projection of a violent hierarchy onto the language
field, this distinction is chiasmatic and reversible, and bound up with
speaking subjectivity; the language field construed as a web of analogical
relations is never effectively closed.
In sum, the field of language forms a chiasm intersecting conscious de-
liberation with impersonal spontaneity—to speak is to inhabit a border-
land, where foreign influence breaks through familiar habits with such
regularity as to pass unnoticed. As a speaker, I continually draw on bor-
rowed resources of dubious origin to stake out the provisional claim to
being me. Speech never lies squarely in the sphere of ownness, but it is no
stranger to expression and confession. In this approach, consciousness
emerges as linguistically mediated, or, to reclaim a celebrated formula
hitherto reserved for its distant cousin: consciousness is structured like a
language. This does not imply that language is solely a phenomenon of
conscious awareness or that it is fully governed by conscious intention. But
neither is language a foreign body grafted onto conscious thought—it ori-
ents and guides thought like a series of pathways which afford being

followed in determinate ways and are themselves molded by the steps
being taken. Language both deploys an automatism predating conscious
intention and is so closely bound up with the twists and turns of expression
as to receive and bear its effects within its innermost structure. It cannot
therefore be decided between a phenomenology of consciousness and a
study of objective grammatical structures. Its field is that of linguistic
phenomenology.

6 Contributions to linguistic
phenomenology: Hegel, Husserl,
Merleau-Ponty
I
n this chapter, I propose to offer additional perspectives on linguis-
tic phenomenology by drawing on the works of Hegel, Husserl, and
Merleau-Ponty. I will first develop the idea that, with its admixture of
philosophical reflection and scientific rigor, Saussure’s general linguistics
may be best deciphered by way of Hegel’s phenomenology. This interpre-
tation responds to Riedlinger’s comment that the editors of the Course
sacrificed the philosophical depth of general linguistics; some of that depth
can potentially be restored by rethinking general linguistics as a phenom-
enological science. I will focus especially on Hegel’s view that phenome-
nological science is founded on the primacy of consciousness, and situate
Saussure’s view of language as a relational system within this context.1
In the remaining sections, I will make a case that linguistic phenome-
nology constitutes a relatively established trend within twentieth-century
phenomenology, especially in the works of Edmund Husserl and Maurice
Merleau-Ponty. The linguistic thread within their work is best followed by
way of Husserl’s collaboration and impact on Roman Jakobson, and
Merleau-Ponty’s extensive engagement with the work of Ferdinand de Sau-
ssure. Even though the relevant corpus is substantial and of undisputed
scholarly interest, it has received limited attention in the literature, perhaps
due to the still dominant view that phenomenology tends to privilege a
transcendental approach to experience, that is, a program limited to a
study of the data of purified, prediscursive consciousness, and contingent
1
I leave any substantive discussion of Saussure’s emphasis on the social and temporal
determination of the language system in relation to Hegel’s phenomenology for another study.
on recognizing the subject as a constituting source point of the origin of
meaning. Insofar as it is hard to imagine how a study of language would fit
into a program that privileges purified experience as its immediate field of
work, and posits subjective constitution to explain the origin of cultural
signification, the many developments within phenomenology which take
language as a starting point of inquiry have received less scholarly atten-
tion, and there is relatively little language-focused research being con-
ducted within contemporary phenomenology. Once this understanding of
phenomenology gets decentered in favor of a more open field, one notes a
number of elements for a linguistic phenomenology within the so-called
classical tradition itself.
Phenomenological science: Hegel and Saussure
Saussure’s project in general linguistics can be likened to Hegel’s phe-

nomenology because both embrace the notion of a total science, a science
not limited to the study of objects alone, but including the subject and the
pursuit of knowledge as well. Saussure’s distinction of two basic points of
view onto language: the synchronic and diachronic attain the two distin-
guishable albeit intersecting facets of the total phenomenon, langage;
combined, they would provide, in principle at least, an encompassing
view of language in toto, despite the fact that language itself is inherently
heterogeneous, and can never be simultaneously grasped along a syn-
chronic and diachronic axis (a pansynchronic view is impossible). The
duality of science emerges and is put in the service of a totalizing ambi-
tion; the projected linguistic science will be both historical and system-
atic. Yet it will also be tied to the underlying point of view—that of con-
sciousness or speaking subjectivity—which prefaces any statement about
objective facts, and must be included in a science if this science is to be
complete.
This programmatic vision of general linguistics echoes with Hegel’s
vision of total science from The Phenomenology of Spirit. The preface
outlines its program thus: “Only a science which grasps the whole of real-
ity can gain access to truth. It is therefore a mistake to exclude reflection
from the (pursuit of) truth; reflection is not just a dispensable means to an
end, or an unnecessary detour in consideration of truth: it forms its essen-
tial part. Science encompasses therefore not only the substance but also
the subject—namely what is reflected onto itself” (Hegel, 1977, p. 13).
Science therefore needs self-consciousness (ibid., p. 15); its scope must
contributions to linguistic phenomenology 155

not be limited to the in-itself but also to the for-itself. We typically assume
that consciousness and objects are separate; in immediate experience of
the senses and perception, but also in science, the substance appears ex-
ternal despite being immediately accessed; it does not appear as a prop-
erty of consciousness itself, mediated by consciousness. The phenomeno-
logical turn involves a break with this initial sense of immediacy and
alienation from the world; as discussed in more detail below, the sub-
stance returns to the subject or “shows itself to be essentially subject” as
consciousness gains in reflective understanding of its connectedness to
the world. “When it has shown this completely the spirit has itself for its
object just as it is . . . Being is then absolutely mediated” (ibid., p. 21). Fol-
lowing Hegel, the relation between consciousness and the world does not
map onto a fixed point occupied by the subject and the totality of objects
surveyed from afar. The true can only be understood (is actual) as a
system, which means that substance is essentially subject (ibid., p. 14), and
that both spread over the total reality that is spiritual through and through.
Being and thinking are thus tied into a systemic unity of the spirit; the
spirit includes not only an individual perspective on substance (“the single
individual is incomplete Spirit” [ibid., p. 16]) but also a trans- or suprain-
dividual standpoint in which individual perspectives inhere.
This general layout of the phenomenological project helps frame Saus-
sure’s program of general linguistics. This program, I propose, becomes
legible if regarded through the lens of Hegel’s directive to found a science
of the spirit which encompasses both the object and the subject of know-
ledge, and emphasizes their complex systemic interrelatedness. Saussure’s
repeated appeal to conduct a comparative criticism of points of view in
linguistics would emerge out of this basic conviction that statements about
objects are paired with subjective stances, and that reflection is needed to
both bring out this pairing and identify the points of view most apt to do
justice to the heterogeneous complexity of the phenomenon in question.
Reflection would then be deployed first in the service of the critical task of
uncovering a subjective point of view within a science claiming to be ob-
jective through and through (physiology of sound, comparative grammar),
and then, and most importantly, in the service of a positive task of devel-
oping an alternative scientific approach wherein the point of view is en-
dorsed and featured alongside any objective claims. This latter point of
view is that of consciousness or speaking subjectivity. Insofar as this (syn-
chronic) perspective is not uniquely tied to the speaking individual, but
implicates social conventions of usage in excess of individual will or even
understanding, it spreads over the socius in a way comparable to how

individual consciousness in Hegel extends onto the totality of the spirit.
The object language, or the semiological system, emerges in this perspec-
tive as spiritual through and through—as can be exemplified by Saussure’s
claim that both the signifying and the signified facets of the sign are to be
figured as psychic rather than physical, that is, that we are dealing with the
psychic imprint of the sound, as tied to an idea, and not with mere sound
vibrations that could be studied naturalistically: “The verbal <(acoustic)>
image is the sound transformed into mental sensations (sensations psy-
chiques) . . . The concept and the acoustic image are equally mental (psy-
chiques)” (Saussure, 1993, p. 68). It is from this mentalized conception of
speech sound that la langue can become instituted as a separable object of
study—it is defined as “a system of signs, in which the two parts of the sign
are mental” (ibid., p. 71).2
This claim becomes intelligible if paired with the subject-oriented claim
that signification depends crucially on the grasp of consciousness. In sum,
Saussure’s critique of the existing sciences of language and his envisaged
reform in general linguistics would be contingent on employing a philo-
sophical stance of reflection—or a philosophy of linguistics. Philosophy
would turn out then be a resource directly useful to science, and general
linguistics would be a science that thinks.
Consider how the project of Hegel’s Phenomenology sheds light on
Saussure’s program. Hegel’s entire phenomenological project is best
viewed via the meta-phenomenological idea (a global idea about con-
sciousness, rather than an idea directly reached from an unexplicated
standpoint adopted by consciousness) that metaphysical and scientific
claims are always paired with epistemological stances, or that any claim
made about being is always bound by a correlative mode of experiencing
(Russon, 2004, p. 222). As such, any claim about what is is undecipherable
outside of a phenomenological context, where the possible stances or posi-
tions of the experiencing subject are mapped out as necessary correlates of
the corresponding claims about what is. This phenomenological attitude
puts pressure on the validity of a naïve approach where such theoretical
concerns are deemed an impediment to fact gathering and advancement of
positive knowledge. Following phenomenology, knowledge seems positive
only at the lowest level stance of immediate sensuous certainty; it shows
itself to be mediated and thus shot with form from the more advanced
stance of the understanding.
2
Note that this claim about la langue is effectively established within the field of la parole—so
long as the latter is construed as a signifying and mind-related process.

The movement or stages in the progression of consciousness in Hegel’s
phenomenology can then be regarded as a resolution of the initial opposi-
tion between consciousness and its object, where the object is regarded as
external to consciousness itself. Passing through the stages of immediate
sensuous certainty, perception, and understanding (discussed in more
detail below), consciousness progresses via different possible relations to
its object. The first, most immediate relation is that of an opposition, where
the object is regarded as external to consciousness itself. The last involves
an internalization of object within consciousness and externalization of
consciousness itself. Initially consciousness divides into the content and
the form of cognition (or truth and certainty), with the material present
readymade apart from thought, and thought itself being empty; the mate-
rial subsequently provides the content for thought, which thought forms in
turn (Hegel, 2010, pp. 24–55). This initial opposition between conscious-
ness and object will become fully resolved when consciousness reaches
the stage of absolute knowledge; this culminating stage yields a realization
that consciousness is necessarily paired with its objects, and that science is
complete only if founded on this basic pairing. Pure science is then libera-
tion from the initial opposition of subject and object (ibid., pp. 28–29), and
of form and content; it is a science of objective thinking where facts belong
to pure thought and the concepts are necessarily bound up with conscious-
ness. The science founded on consciousness would not be purely formal or
lacking in materiality; rather, matter is always already bound up with form
(form is not external to it), and always already mediated by it (hence never
simple brute matter, but always relationally determined by concepts of
consciousness itself). As mediated by form, the matter in phenomenology
is internal rather than external to consciousness; both belong to pure
thought without collapsing one into the other.
Let me trace this gradual progression spanning the three stances ad-
opted by consciousness and leading up to absolute knowledge in more
detail. In this initial stance of immediate certainty, being is taken to be a
“this” or simple immediacy, the irreducible givens of experience in the
now. Importantly, such simple givens are regarded as primary in the em-
piricist tradition and in scientific positivism where reality is cashed out in
terms of simple data, like colors, sounds, and so on. Positive science would
then be located at the lowest level of Hegel’s dialectic, as a stance that may
come first but is most naïve in that it has not reflected on its enabling condi-
tions and the implication of the knowing subject in the object known. Ac-
cording to Hegel, this lowest stance enjoys an inexpressible immediacy of
being there which can be lived and yet cannot be described. Sensuous

immediacy cannot ultimately be a meaningful element of experience—it
turns opaque as soon as one tries to describe it, that is, compare it to other
nonpresent experiences, and thus introduce negation into the presumed
plenum.
The dialectic of sense certainty dispels the initial impression of imme-
diacy. Being there is not a guarantee of full possession of the event and
having an authoritative stance; “any philosophies that seek to found all
knowledge claims on the absolute guarantee of immediate sensory experi-
ence are . . . shown to be untenable” (Russon, 2004, p. 18)—the validity of
knowledge claims must exceed immediate presence in the here-and-now.
Furthermore, the presumed immediate givens (e.g., pure sound devoid of
signification) fail to yield anything durable or objective if grasped as iso-
lated atomistic data.
The pursuit of scientific knowledge thus moves to the stance of percep-
tion, where being is taken to be things, or determinate essential beings,
thematized in Western metaphysics under the heading of substance. From
the point of view of perception, things are the ultimate building blocks of
the world. Each thing is regarded as a central core carrying a host of de-
pendent properties or attributes. There is therefore an antinomy in percep-
tion: each thing seems to be unique and self-defined, and irreducible to the
aggregate of properties (e.g., colors, sounds) that can be ascribed to them
(a chair is not just a bundle of properties, it is). It seems therefore that a
thing needs to be posited as self-sufficient so as to salvage its uniqueness
and avoid reducing it to a bundle. This view is however untenable: “this
exclusivity and absolute singularity makes the thing indefinable, for there
is no way to express its determinateness without invoking terms that name
properties,” and properties are only determinate in relation (ibid., p. 26).
A thing or a substance cannot therefore be an ultimate building block of
reality; the identity of each thing is contingent on its determination in rela-
tion to others—uniqueness and dependency are tied together, rather than
opposed. Contrary to Aristotle’s metaphysics, things or substances are not
ultimate ontological unities; rather, they are appearances reflective of the
total system of forces at work in nature. The progression of consciousness
entails therefore a movement away from the metaphysical commitment to
the primacy and irreducibility of substance (what Saussure terms “the invol-
untary assumption of substance”) in favor of a single unified system in
which things are embedded and contingent on. This is the level of under-
standing where being is taken to be a comprehensive system of forces, and
where being-for-self is tied to being-for-others. The metaphysical foundation
for a science cannot then be located in a substance; the latter presupposes a

more originary relational unity. The task of science will accordingly change
from the Aristotelian observation of multiple beings to a study “of the oth-
erwise invisible forces of which these things are the visibility, the appear-
ance” (ibid., p. 31). Science must therefore be comprehensive and systematic
to reach a ground that does not upon reflection turn out to be itself supported
by additional metaphysical structures.
According to the progression outlined above, being can appear as im-
mediacy, multiple determinacy, and systemic comprehensiveness depend-
ing on the stance adopted by the knowing consciousness. The pursuit of
phenomenology consists then in unveiling how the different metaphysical
claims are “structural features within experiential attitudes” (ibid., p. 224).
It follows that “[t]he object for us—for the phenomenological observer—is
the set of experiential shapes where each shape is experienced by us as a
paired structure of subjective comportment and metaphysical commit-
ment” (ibid., p. 224). Within phenomenological science, the notion of a
self-standing object (whether a sense datum, or a substance) is only an ab-
straction out of this more basic and original paired structure rather than a
source of “objective” or true knowledge about what is. True knowledge
supposes an originary coupling between subject and object, embedded
within a systemic unity encompassing reality in toto.
The paired structures of metaphysical claims and experiential attitudes
are themselves explored from the stance of absolute knowledge. That
means that we do not simply project our own metaphysical commitment
(according to any of the possible stances of consciousness) but that these
metaphysical commitments themselves become our object. In the process,
absolute knowledge will sort out the inherent contradictions of the implicit
metaphysical stance implied within each stage of consciousness. The dis-
tinctive feature of absolute knowing is that it explicitly takes metaphysics
as its object—unlike in the three stages where such metaphysical commit-
ment remained unthematized. Absolute knowing is then a stance onto the
three stances distinguished hitherto, it is a metastance which is contingent
on the former (there is no consciousness without the stage of immediate
sensuous certainty), but which illuminates its structurally paired character
for the first time, bringing its own resources to bear on the existing posi-
tions (like Saussure’s critical analysis of the different points of view as-
sumed without being explicated within the sciences of language). While
each of the three stages can be termed a point of view onto being, it is the
point of view adopted in absolute knowing that sheds light on the points of
view implied in the different modes of experience available to conscious-
ness. This point of view therefore makes apparent that a point of view is

(always already) implied in any claim about what is, or that what is is in-
separable from what appears. As such, absolute knowledge is necessary
for each stage to come to know itself.3 From the point of view of absolute
knowing, the phenomenology of conscious experience and the science of
being are conjoined, and belong in equal measure to the global project of
a science, the science of spirit.
In what follows I’ll make a case that Hegel’s phenomenology, in its
progression toward the stance of absolute knowing where systemic organ-
ization and consciousness become (discovered as) inextricably paired,
provides a useful interpretative framework for Saussure’s project in gen-
eral linguistics. This Hegelian framework helps to decipher Saussure’s
critique of the positive orientation in the sciences of language, and pro-
vides resources for his own preliminary program for a science where the
object side of (systemically organized) language is inextricably paired
with the subject side of consciousness. This reinterpretation lends intelli-
gibility to Saussure’s repeated employment of terms such as conscious-
ness, phenomenon/a, and speaking subject in the manuscript writings on
general linguistics, a practice effectively undecipherable in light of the
dominant structuralist interpretation of his work.
Recall first Saussure’s critique of approaches to language modeled on
natural science in sound physiology and comparative grammar. Sound
physiology is committed to the primacy of sound data stripped of any sig-
nification—sheer content that would only subsequently be shaped by form
(ideas, concepts, meanings). Sound would be traced back to the articula-
tory process involved in its emission, yet without it being asked whether
articulation itself may be a process bound up with meaning since there is
nothing within the process of sound emission alone which enables the sci-
entist to identify given sound blocks as basic units rather than others. This
view is therefore founded on external phenomena only, observable and
calculable phonetic facts (such as the aka or a sound bites) effectively ab-
stracted from language phenomena properly so called. These data may
seem immediately given and positive, and as such a firm basis for making
objective claims. Yet upon reflection, sound data alone seem necessarily
anchored in a more basic reality—of a signifying unit, a sign or a word.
Saussure considers at length the linguistic standing of the word within
the second course of lectures on general linguistics, as part of an exten-
sive reflection on “concrete entities” in language. He opens with a word of
3
Since absolute knowing is itself a stance of conscious experience, it must also project a
metaphysical commitment. It is an absolute idea, with being as dialectic.

warning: “Linguistics is not altogether simple in its principle, <in its
method>, in the research it comprises, because language (la langue) is not
simple” (Saussure, 1997, p. 1).
This simultaneous complexity of the subject matter and method, lan-
guage as the what to be investigated and the manner of investigating it, is
at odds with the initial impression that language is “close at hand.” Saus-
sure cautions that language may be too close, like the glass of the specta-
cles through which we grasp other objects but which itself is unavailable
to view (ibid.). What is immediately graspable is to be put aside; the whole
metaphor of simple seeing is put into question, a proclaimed distrust with
regard to any immediate intuitive insight. Hence, “language is not what
strikes us anthropologically” (ibid., p. 16)—that is, the myopic perspec-
tive on the immediately available sign, and what goes into producing it
(sound, idea). The student is advised not to lay all the trust in these first
intuitions.
The assumed transparency of language becomes exposed as an illusion
(ibid., p. 1). Language could provide direct access to meaning as if via
transparent glass only if the points of access were fixed once and for all,
with no equivocation in the present or mutation across time; the possibility
of direct access to meaning is thus contingent on an illusion of stability
and stasis in language. This illusion is tied to the view that language could
be defined in terms of its “immediately graspable elements” (Saussure,
1997, p. 16)—that is, the meanings of individual signs. A time honored
candidate for immediately graspable elements is found in the classification
of language into parts of speech, which were to help distinguish between
different types of words (ibid., p. 23). This classification suggests that there
are classes of words, like nouns, adjectives, verbs, which have a referential
function in conformance with the organization of the world into basic
kinds of beings. This view of language maps onto the view of external re-
ality from substance metaphysics, and supposes that basic parts of speech
map directly onto the basic metaphysical entities; as in the grammar of
Port Royale, a substantive noun maps onto a substance, an adjective onto a
property, a verb onto an action. These parts of speech would therefore
seem as fixed as the metaphysical kinds they name; they would maintain
the basic unity of the substance; they would not be susceptible to further
analysis or breaking down into component parts.
This classical grammatico-metaphysical classification is undone, how-
ever, when one considers language from the perspective of usage. For
example, a common expression like bon marché does not map onto a
speech part but can be counted as a concrete entity in ordinary French

(ibid., p. 23). Traditional classification of language into parts of speech
does not reliably yield irreducible unities. Could the more general cate-
gory of a word provide a basic, irreducible element within language? It
seems not. For example, the same word is pronounced differently depend-
ing on its placement in a phrase or sentence (le mois—as in le mois de
decembre/un mois et demi),4 or the word changes depending on number
(cheval-chevaux) (ibid., p. 19). If one were to disregard this plurality of
forms in the name of some primary, basic form, like the infinitive form for
the verb or the nominative for the noun, one would disregard the concrete
level of language and traffic in grammatical abstractions. As long as one
stays at the level of the concrete plurality, no primary unity can be easily
identified. Furthermore, we find signification not only in the lexicon. The
categories usually classified as syntactical or grammatical in a narrow
sense are also meaning making (Saussure, 1957, p. 65). For example, the
grammatical categories of number and tense are signifying, since they tell
us something about, for example, the amount of items being considered,
or the time frame associated to a given action. As with the lexicon, we
find here meaning making oppositions between, for example, the singular,
dual and plural number, or between the present, perfect, past, future tense.
Just as in the lexicon, so in grammar it makes no sense to assume a single,
self-standing category, in isolation from the system of relations.
Any given category in language is established via systemic contrasts to
at least one (and typically more) other term in the associative group. Any
linguistic category is inflected by language—there are no absolute terms
but only relative ones. For example, the distinction between the e and a in
the German gebe-gab (or a and o English came-come) signifies a temporal
distinction between an accomplished and a present or habitual action; the
grammatical categories are signifying and differential, they form an inte-
gral part of the semiological system; traditional grammatical distinctions
between lexicon and syntax fall apart. Wherever we turn, whenever some-
thing in language appears to provide a sound bedrock, we are redirected
somewhere else in the system, where everything seems entwined with eve-
rything else.
Saussure concludes therefore:
Aucun système ne serré comme la langue: serré implique précision des

valeurs (la moindre nuance change les mots); multiplicité de genres de val-
eurs; multiplicité immense des termes, des unités en jeu dans les système;
4
The mois in the first phrase is pronounced mwa, the second mwaz.

réciproque et stricte dépendance des unités entre elles: tous est syntactique
dans la langue, tous est système. (ibid., p. 69)
A substance-based view of language turns out upon reflection to be un-

derwritten by a hitherto unthematized commitment to a relational under-
standing as a system. Just as the things in Hegel’s Phenomenology turn out
to be embedded in a system of “the otherwise invisible forces” (Russon,
2004, p. 31), signs in Saussure’s general linguistics emerge as the visibility,
the appearance of a larger semiological system and the manifold relations
between individual signs. The forces are not apparent at first because they
circulate in the interval between (and within) signs rather than being re-
vealed by each sign considered in isolation; while signs appear compact
and unitary in the initial fascination with what is simple and immediately
given, they point to a whole host of necessary lateral relations to other signs
(and within each sign) once the attitude of consciousness has progressed
from sense certainty to a more reflective understanding that the meaning of
each sign is conditioned by contrastive, associative and syntagmatic rela-
tions to other signs. In the process, consciousness begins to appreciate that
a complex system of forces of difference, association, and syntagmatic soli-
darity, which run like an endless stream of vectors across the whole lin-
guistic field, enables the grasp of meaning in any individual sign. The basic
unity in language is therefore the whole system of relations.
A manifest failure to locate substantive unities within language is thus
a measure of success, since it exposes the folly of the entire quest for stable
foundations, and points us toward a novel, nonsubstantial understanding of
language. There are no substantive unities since all the meaning bearing
phenomena identifiable in language represent a relation. “Everything is
relation” (tout est rapport) in language (Saussure, 1957, p. 68). No return to
things themselves is possible in the study of language—according to Saus-
sure’s notes, the relation between things both preexists and determines the
things themselves (le lien qu’on établit entre les choses préexiste . . . aux
choses elles-mêmes, et sert à les déterminer) (Saussure, 1954, p. 57); the
basic terms in linguistics are therefore relations. There are no preexistent
unities, whether at the level of sound or sense; language does not conform
therefore to the rational categories derived from substance metaphysics,
since nothing in it is substantial, there are no bounded beings to be found
at the bottom of the sea. However, each related term is itself relationally
constituted, through the implication of sense in an element of sound. The
relata are themselves sites of (internal) relations. This leads Saussure to
state that “all the phenomena are relations between relations” (Tous les

phènomenes sont des rapport entre des rapports [Saussure, 1957, p. 68])—
or it is the differences, the oppositions between (and within) the relata
which can be termed phenomena in linguistics, and that means also: which
are the ultimate unities (ibid.).
Importantly, the understanding of language as systemic and relational
belongs to the realm of phenomenology even though it departs from the
initial myopic preoccupation with individual words. This is in fact the par-
adox, in Saussure’s view, of general linguistics: there is no difference be-
tween the levels of phenomena and of the unities to be found there (ibid.,
p. 67). This paradox is intelligible in light of the fact that access to concrete
linguistic reality is secured by the criterion of consciousness:
Ce critère [de ce qui est abstraction pure et de ce qui est concret] est dans
la conscience de chacun: ce qui est dans la conscience des sujets parlants,
ce qui est ressenti à un degré quelconque, c’est la signification. Et on
pourra dire alors que le concret réel, pas du tout si facile à saisir dans la
langue, c’est ce qui est ressenti, ce qui à son tour équivaut à: ce qui est
significatif à un degré quelconque. (ibid., p. 41)
It is therefore the experience of signification that speaking subjects have

that orients us to the question of the concrete reality of language, and
serves as the basic criterion in the search for its basic categories and prin-
ciples. The quest for concrete reality in language should not however be
equated with a species of phenomenological naivety in linguistics, and the
study of linguistic phenomena is not to be equated with an unreflected im-
pression of familiarity. Importantly, Saussure ties consciousness with re-
flective progression from a naïve commitment to the priority of the sign as
signifying unit to a reflectively refined understanding of the signification
as mediated by the system. As in Hegel’s phenomenology, the metaphysi-
cal claim about the foundational role played by the invisible system of re-
lations for any visible element is paired with the epistemological notion of
a subject who can transcend the initial fascination with the data of imme-
diate sense in favor of a more complex understanding, all without losing a
foothold in consciousness.
Consider this progression of reflection to a systemic understanding of
language in more detail. In the final note from the Orangery manuscripts,
Saussure puts forward the notion of “integration or retrospective reflection
[postméditation-réflexion]” as a method to be employed in linguistics. In-
tegration would capture “the active life of language [langage] in toto”
(Saussure, 2006, p. [87], 60, my translation). Integration or reflection would

offer a complete view of the sign system as available and contingent on the
subject. Like Hegel’s absolute knowledge, it would permit a retrospective
survey of the system of relations as tied to consciousness, but with a direct
focus on the being of the sign.
Saussure opens his methodological proposal with a thought experiment
featuring a primitive language composed of just two terms: ba and la. It
would be a fundamental error to think that this language à deux reduces to
two individual signs plus their correlated significations. For the two can
function as signs—for example for solids and nonsolids respectively—
only on condition that they are apprehended by the mind not only as posi-
tive concepts, but also as contrasted or differentiated values. This means
that understanding the sign ba implies understanding also that ba is not la,
and understanding la implies also understanding that la is not ba—with
each sign carrying in it, as it were, a trace of the absent other.
As Saussure puts it,
The mind (esprit) will locate a principle of classification in the simple fact
that there exists a ba/la difference, and no other, enabling it to regularly
classify everything under the first or the second of these heads (for example
the distinction between solid and nonsolid); at this point, the sum of posi-
tive knowledge will be represented by the common features attributed to ba
things and by the common features attributed to la things; even though the
features are positive, it [the mind] only sought the negative feature which
allows ba and la to be told apart; it never tried to unite and coordinate, only
to differentiate. (Saussure, 2006, p. [88], 60, translation revised)
The contrastive relation between the two signs can then only be estab-
lished by the forces of negativity deployed by the mind. Understanding the
meaning of either ba or la signs is contingent upon both the identity of
either one and its mediation by difference to the other. Each individual sign
is then shot through with negativity, and spread over the entire semiologi-
cal system—which in this case is just a system à deux, but could in princi-
ple include an innumerable number of terms and relations. The small scale
system envisaged here offers a miniature version, as it were, of the existing,
complex, languages, and helps bring insights relative to the latter into an
otherwise hard to attain focus. The important point is this: the mind’s me-
diation is manifest even in a language system as “primitive” as the one
envisaged here; signs signify in virtue of a contrastive relation between
them (the value), which is established by the mind’s power of negation (the
mind “never tried to unite and coordinate, only to differentiate”). A system

of contrastive relations in language is therefore not an objective factum dis-
covered in nature and ready to be catalogued in a positive science; it is
contingent on the mind’s differentiating potential and needs to be studied
in this necessary (albeit nonobjectifiable) relatedness and negativity. The
being of the sign is contingent on the mind—to encounter a sign as signify-
ing something is also to indirectly grasp the mind’s work already done, the
host of contrastive relations already set. The study of signs—even if based
on a small and (presumably) static system featuring two monosyllabic
terms (there have of course to be at least two for either one to signify)—
turns out therefore to be more complex than it initially seems, and in excess
of the immediately available givens, the signs ba and la. It includes the
contrastive value relation between the signs as established and recognized
by the subject for whom there is a significant difference between solids and
nonsolids. The study of the sign system can then only be complete if it ac-
commodates both the “objective” relation of difference between the signs
and the “subjective” differentiation by the mind; the total view must en-
compass both the system and the spirit in a reconstructive reconstitution.
What would happen if the system resembled anything like the complex
web we actually find in natural languages? There the terms and the pos-
sible interrelations are endless, but also the whole system is constantly
evolving:
In every existing sign, a determined value becomes integrated, retroactively

elaborated (postélaborer) . . . which is only ever determined by the sum of
signs present or absent at a given time; and as the number and reciprocal
and relative aspects of these signs change from one moment to the next in
an infinite number of ways, the result of this activity, for each sign, and for
signs as a whole, changes also from one moment to the next in an incalcu-
lable fashion. (Saussure, 2006, p. [88], 60)
In a system as complex and constantly evolving, the mind’s work is

harder to pinpoint—it seems that the mind would now spread over a much
greater sum of signs (present or absent at a given time), where the recipro-
cal and relative interrelations is not only innumerable but changing in an
infinite number of ways. Having conceded that the mind’s work is part and
parcel of the miniature model, the signifying system à deux, we now have
to concede the mind’s implication in a signifying system we effectively put
to use every day . . . Or better, following Saussure and Hegel, we need to
accommodate the collectivity of minds, past, present, and future, for whom
language signifies and through whose usage it is alive.

But note that the mind here (as in the simple system initially envisaged)
is not simply added on to an objective setup; to encounter a system of signs,
of whatever complexity, is to witness the mind’s work already done, since
the reciprocal and relative relations exist through the mind’s power of dif-
ferentiation (whether the relations span a static duo or an evolving infinity
of terms and relations). The relation between the mind and language or
spirit and system cannot therefore be mapped onto two fixed terms; the
active life of language is contingent on the activity of the mind, but the latter
is discovered within language, while language itself is a site of its many
acts. Language is therefore not a sheer product of the spirit’s work, but a
medium of its praxis. Hence a profound ambiguity in their relation, which
makes it effectively impossible to array either the mind/spirit or language/
system at the subject or the object end of a relationship. As active and ac-
tively signifying, language does think and thus belongs to the side of the
subject, at least according to the usual subject-predicate classification. Yet it
makes sense to say that language thinks or works through me, or that the
sign means something (to me, to us, to the community); in both cases, the
subject’s role falls on the sign’s (and language’s) own shoulders—it is active,
alive, guides me (us) in thinking, speaking, and writing. Anyone having
made herself at home in a second (or third, or fourth) language will appreci-
ate this experience of being guided by its patterns and principles, having
meaning light up in a joke or reference that suddenly makes sense, but
always according to the way this language thinks, and not as direct or even
indirect translation of meanings inherited from the first language. Here con-
sciousness shows itself to be semiologically mediated in its own meaning
bearing structures; the linguistic system ceases being an external object for
it, and becomes integral to its own meaning making acts (including the acts
of thinking about language as sheer object). Even though language assumes
the object pole in science, this pole is not a fixed term. Language shifts to
the object pole when being studied, back to the subject pole when it signi-
fies. The subject and object pole are therefore shifters for relative and recip-
rocal roles assumed according to the dominant situation (language use, lan-
guage study) rather than labels designating fixed terms; language itself is
undecided between them, even though it accommodates both.
This undecidability between the subject and object must be preserved in
a science of general linguistics founded on the concrete reality of lan-
guage. We therefore find it playing out within the relation between the
scientist and the subject matter under study. The project of establishing the
science of general linguistics is itself predicated on exposure to the many
workings of the sign, its inheritance from others and from the past. The

scientist can then always retroactively reflect on this “passivity before the
sign,” and her own preexisting entanglement in the language about to be
studied. This retrospective reflection belongs to the business of linguistic
science if the latter purports to be concrete and complete. The role of the
scientist would then be to adopt the stance of absolute knowledge, and re-
trace the progression of consciousness in its various structural inflections,
spell out the basic unity of consciousness and the complex evolving system
of signifying relations which could not have arisen without its mediating
and negating potential but which are still separated, minimally, from the
retroactive reflective glance. The integrative and retrospectively reflective
science of general linguistics seeks to encompass this entire field.
Early structuralism: Jakobson and Husserl
There are multiple resources for a linguistic phenomenology within Hus-

serl’s writings; the latter were explicitly adopted within the early East Eu-
ropean structuralist program which precedes the post-Second World War
era when a forced choice between phenomenology and structuralism
became de rigueur. The reception of Husserl’s phenomenology by the early
East European structuralists like Jakobson was unique in that it highlighted
the possibility of a phenomenological analysis of structural types, distinct
from the dominant conjunction between phenomenology and intuition,
which privileges prediscursive experience available to conscious grasp in
the mode of phenomenological seeing. This possibility (of a phenomeno-
logical analysis of structural types) hinges on the phenomenological goal
of securing a priori foundations of phenomena—including language, that
is, the necessary forms and patterns of relationships organizing the field, a
grammar comprising a set of laws which stipulate how meaning is com-
bined out of simple into complex elements, and how it can be transformed.
In the terms used by Husserl, “the laws of complex meanings set forth the
requirement of . . . significant unity, i.e. the a priori patterns in which mean-
ings belonging to different semantic categories can be united to form one
meaning, instead of producing chaotic nonsense” (Husserl, 2001, p. 49).
This a priori grammar would constitute an alternative approach to
meaning and language to the dominant approach from the empirical sci-
ences. As Husserl put it:
Modern grammar thinks it should build exclusively on psychology and other

empirical sciences. As against this, we see that the old idea of a universal, or

even of an a priori grammar, has unquestionably acquired a foundation and
a definite sphere of validity, from our pointing out that there are a priori
laws which determine the possible fields of meaning. (ibid., p. 49)
This emphasis on the a priori would lead against the grain of our age,
with its focus on empirical data and neglect of theory (ibid., p. 73). In Hus-
serl’s words: “even though all basic insights lead back to the a priori, our
age’s sense for it almost threatens to wither away. I therefore fairly take up
the cudgels for the old doctrine of a grammaire générale et raisonnée, a
philosophical grammar, for its obscure, undeveloped intention aiming at
the ‘rational’ in speech, in the true sense of the word, and in particular at
the ‘logic’ of speech or its semantic a priori” (ibid., p. 73). This ambition to
secure a set of general rational laws responds to Kruszewski’s stated aim
to base the science of language on a deductive system of generalizations,
distinct from the inductive methods of the empirical sciences. It responds
to the goal of studying language on its own terms, rather than those bor-
rowed from cognate disciplines and their scientific programs.
It is usual to associate Husserl’s phenomenological method with indi-
vidual consciousness, but consider how his analysis of meaning in the Log-
ical Investigations leads him to affirm both the involvement of conscious-
ness in its meaning giving acts and the structuring effects of grammatical
kind which operate independently of individual intention. The discussion
of the grammatical distinction between the so-called categorematic and
syncategorematic expressions from the Investigation IV is a case in point.
The categorematic expressions are those that can stand on their own, while
the syncategorematic ones require completion, and are therefore parts of a
more comprehensive whole. Importantly, this grammatical distinction can
only be made at the level of meaning—as a type of expression: “To pin
down . . . the inner ground that enables some expressions, and not others,
to stand as finished locutions, one must . . . go back to the semantic realm,
and point out there the need of completion that attaches to certain non-
independent meanings” (ibid., p. 58).
Following Husserl’s analysis, the independence and nonindependence
of meaning is contingent on the character of a meaning giving act. A con-
crete act can involve several acts which live in a whole as either independ-
ent or nonindependent parts. An act of meaning can as such be complex,
made up of component acts: “A meaning, accordingly, may be called ‘in-
dependent’ when it can constitute the full, entire meaning of a concrete act
of meaning, ‘non-independent,’ when this is not the case” (ibid., p. 59). The
presentation or the intentional act is therefore the basis for settling

meaning, and syncategorematica like together, with, and, or, cannot
achieve a fulfillment of meaning or intuitive understanding—except in the
context of a wider meaning whole (ibid., p. 60). The question becomes then
how such incomplete expressions combine with others to form complete
ones—without leading to a mere jumble or heap. Husserl notes that mean-
ings can be combined to form new meanings only using connective forms,
and that there are a priori laws governing all meaning combinations. These
laws transcend subjective intention and depend on inherent laws of lan-
guage itself: “As regards the field of meaning, the briefest consideration
will show up our unfreedom in binding meanings to meanings, so that we
cannot juggle at will with the elements of a significantly given, connected
unity” (ibid., p. 62).
This inability is a law of essence, rather than a sign of some deficiency
on our part; it is an objective impossibility, rooted in the pure essence of
the meaning realm, and involves a range of essential kinds, semantic cat-
egories. For example in a proposition of a type “S is P” such as “Tree is
green,” S is filled out by nominal material, P by adjectival material—and
these cannot be substituted for by just any grammatical category. We may
produce foolish, false ridiculous meanings—for example, “this blue raven
is green” (ibid., p. 63), but when we transcend the bounds of semantic cat-
egories we do not achieve a unity of sense (as in, e.g., “this careless is
green,” “more intense is round,” etc. [ibid., p. 63]). In sum, there is an a
priori law regulating possible combinations according to the very nature of
the constituent parts, and even though we can construct verbal strings such
as “if the or is green,” “A tree is and”—such strings have no graspable
meanings. Since meanings are combined according to the principles of
form (or syntax, or category), meaning and (grammatical) form are inter-
connected, there is a grammar or structure which organizes meaning
above and beyond subjective grasp. There is no fulfillment without gram-
mar, the meaning giving intentions are entwined with grammatical catego-
ries in the formation of complex wholes out of component parts. So, the
question of meaning (and meaninglessness) is settled grammatically:
The grammatical expression of the a priori incompatibilities and compati-

bilities here in question, as of the pertinent laws governing meaning-
combinations, must in part be found in the grammatical rules governing
the parts of speech. If we ask why our language allows certain verbal com-
binations and disallows others, we are to a large extent referred to contin-
gent linguistic habits, to matters of mere fact concerning language, which
develop in one way in one speech-community and another way in another.

In part, however, we encounter the essential difference of independent and
non-independent meanings and, closely involved therewith, the a priori
laws of the combination and transformation of meanings, laws which must
be more or less revealed in every developed language, both in its grammar
of forms and in the related class of grammatical incompatibilities. (ibid.,
p. 68, emphasis added)
Considerations relative to language are therefore not confined to a con-

tingent linguistic layer which could be in principle stripped away to reveal
a purified field of meaning. Meaning is not a field of simple phenomenal
seeing—it includes also a theory of meaning structures and their formal
constitution—the elementary structures (ibid., p. 68); it deals with se-
mantic categories, for example, nominal, adjectival, propositional (ibid.,
p. 69)—hence it effectively adopts the existent categorization of language
into parts of speech. The task facing the phenomenologist is then the fol-
lowing: “We must fix the primitive forms of independent meanings, of
complete propositions with their internal articulations, and the structures
contained in such articulations. We must fix, too, the primitive forms of
compounding and modification permitted by the essence of different cat-
egories of possible elements” (ibid., p. 69). These forms help to establish
the basic laws of meaning: the laws which discourage nonsense (Unsinn)—
and laws which discourage absurdity (Widersinn). In sum, Husserl inter-
sects an analysis of meaning in terms of meaning giving intentions/acts
with a grammatical analysis of forms and combinatory principles where
combination of simple parts into complex wholes is contingent on both
intuitive fulfillment and structural laws. This demonstrates that a phe-
nomenology of language, even in Husserl’s own project, does not fit into
a model of subjective constitution of meaning, and that there is room for
a structural analysis of language itself, according to its guiding principles
and rules within a phenomenological conception of language. And while
it is questionable whether such an analysis should simply adopt the re-
ceived view of the parts of speech as basic categories (a point emphasized
by Saussure in his general linguistic work), and whether the bounds be-
tween sense and senselessness can be fixed as firmly as the analysis above
suggests (a point later developed by Derrida), Husserl’s analysis provides
an exemplar of working at the intersection of subjective and systemic un-
derstanding of meaning and language, and as such is instructive to the
project of linguistic phenomenology.
It is therefore unsurprising that Husserl’s works were directly integrated
into structural linguistics. In addition to the Logical Investigation IV

discussed above, Jakobson drew on Husserl’s Cartesian Meditations in his
own structural analysis. This influence is less surprising when one consid-
ers the congruity between Husserl’s phenomenology and the analysis of
general principles. Like Kruszewski, the Husserl of the Cartesian Medita-
tions proposes to “put out of action all the convictions we have been ac-
cepting up to now, including our sciences” (Husserl, 1999, p. 7); that does
not mean rejecting science tout court but interrogating the idea of science
and grounding scientific judgments in experiential evidence, which Hus-
serl defines as the direct givenness of a state of affairs to consciousness.
Only judgments derived from experience of the affairs present as they
themselves can count as scientific (ibid., p. 14). This manifestation of expe-
riential evidence is tied to intuition, in agreement with the principle of all
principles that “every originally giving intuition is a legitimating source of
cognition [knowledge].” What is disclosed to intuition would have the
mark of apodictic evidence, that is “single peculiarity of being at the same
time the absolute unimaginableness of [the states of affairs’] non-being,
and thus excluding in advance every doubt as ‘objectless,’ or empty” (ibid.,
pp. 15–16). While apodictic evidence seems to have a limited range with
only the living present being adequately given to consciousness (ibid.,
p. 22), Husserl extends the range of apodicticity beyond adequate given-
ness, or an originally giving intuition of the living present. Surrounding
the immediate data of intuition there is an open horizon of “what is strictly
non-experienced but necessarily also present” (ibid., p. 23). Our field of
work is not confined to the momentary cogito, but includes the set of uni-
versal structures organizing the experience with the necessity of an a
priori principle, like intentionality, and temporality, which intersect the
subject with the otherness of objects and the future and the past in a set of
systemic relations. The method of phenomenology cannot therefore be that
of immediate intuitive seeing, since the structural types involve relational
categories and an interplay of presence and absence, rather than the self-
standing data of intuition. Unsurprisingly, the early structuralists like Ja-
kobson will be drawn and inspired by such an analysis of structural types
and invariant features, and apply it to a search for the a priori principles of
language.
Intentionality is one of such universal and apodictic structures of sub-
jectivity (ibid., p. 28). Conscious acts bear a relation to what is worldly
within themselves (ibid., p. 32), and not a simple effect of objects confront-
ing us from the outside (ibid., p. 42). The tripartite ego—cogitationes-
cogitatum structure is therefore a universal element of the transcendental
field itself. Contrary to the empiricist view, consciousness is not a blank

slate to be filled with content—it is structured in determinate ways, and
there is a fit between the structural type of experience and the object being
experienced. “Thus each passing cogito intends its cogitatum not with an
undifferentiated blankness, but as a cogito with a describable structure of
multiplicities, a structure having a quite definite noetic-noematic composi-
tion” (ibid., p. 40). We therefore do not perceive disconnected sensations
but meaningful unities; thanks to the work of objective synthesis, the ob-
jects of perception, for example, a die, are given continuously as a unity in
multiplicity of appearances—such as variation in spatial distance (near-
far), mode of presentation (from one side or another), and variations within
the individual aspects of the objects such as color and shape (which change
according to the degree of illumination, or the angle from which they are
viewed).
The condition of objective synthesis is found therefore in the temporal
unity of consciousness and the tripartite retention-now-moment-protention
structure (Husserl, 1999, p. 18). For example, the perception of a cube ex-
tends beyond what is immediately given to view to include the anticipated
rear side, and thus it includes an open horizon of the future (protention);
rather than constituting a saccadic succession of discrete temporal mo-
ments, perception has the quality of an extended flowing movement where
the just-seen (retention) is preserved within the current moment like the
comet’s tail. This temporal organization of perception in the extended
present is distinct from both recollection and expectation, which are both
contingent on the flowing temporality of consciousness itself. This con-
sciousness of internal time is deemed the most primary cognition, the
foundation of all experiences (ibid., p. 43).
While visual perception of a cube offers a paradigmatic example of a
universal structure of subjectivity, Husserl distinguishes between different
“structural types” of intentionality, in addition to perception, such as recol-
lection, fantasy, expectation (ibid., p. 51). Each one has its unique mode of
grasping the object (as experienced in the flesh and blood, as having been
experienced in person, as to be experienced, as experienced in an as-if
mode). Husserl concludes that “any ‘Objective’ object, any object whatever
(even an immanent one), points to a structure, within the transcendental
ego, that is governed by a rule” (ibid., p. 53). All this illustrates that con-
sciousness is not a chaos but a system, and the phenomenological task con-
sists in filling in the system with the many and all possible types. Even
though Husserl does not consider this possibility, language can be made to
fit into this taxonomy of structural types of consciousness and be studied
as part and parcel of the phenomenological project thus described.

Unsurprisingly, Husserl’s Cartesian Meditations together with the
Logical Investigations came to provide central methodological guide-
lines for the early East European structuralists. Jakobson especially inte-
grated elements from both—but focusing less on the canonical themes of
the phenomenological movement like the analyses of consciousness dis-
cussed in the Fifth and Sixth Investigations, and more on the Third and
Fourth Investigations “On the theory of wholes and parts” and its appli-
cation to the specific problem of meaning in language in “The distinction
between independent and non-independent meanings and the idea of pure
grammar” (discussed above). As Holenstein observed, Jakobson’s focus
was not merely incidental and agreed with Husserl’s own views of the
foundational role of these sections for the better known, later ones. Hus-
serl noted: “I have the impression that this Investigation is all too little
read. I myself derived great help from it: it is also an essential presuppo-
sition for the full understanding of the Investigations which follow” (in
Holenstein, 2005, p. 13). Jakobson was one of the few readers to appreci-
ate that the structural analysis of meanings provides the basis for the
phenomenological analysis of consciousness. He could therefore regard
Logical Investigations, up until the 1960s, as “one of the most inspiring
contributions to the phenomenology of language” (ibid., p. 13). Jakobson
also carefully read the Cartesian Meditations, especially the emphasis on
the set of structural types of consciousness and the open horizon struc-
ture of experience just discussed (see ibid., pp. 14–15; for further discus-
sion of thematic connection between Jakobson and Husserl, see Holen-
stein, 2005).
Husserl’s contribution to structural linguistics, especially to Jakobson’s
work, is largely unacknowledged in contemporary scholarship, both in lin-
guistics and philosophy (ibid., p. 11). Considering that the official history
of the phenomenological movement by Spiegelberg makes no reference to
Husserl’s impact on the early East and Central European structuralism, it
is unsurprising that the phenomenological strand within structural linguis-
tics falls squarely on the blind spot of a contemporary reader who is un-
likely to recognize this historia obscura of phenomenology (ibid., p. 11).
However, the intellectual context within which the early Eastern and Cen-
tral European structuralism developed was marked by none of the antago-
nism between the structuralist and phenomenological approaches which
characterized the French intellectual climate in the 1960s. Phenomenology
fell on a receptive ground and became integrated into the structuralist pro-
gram in the 1910s and 1920s; the early structuralist reception of phenom-
enology was of a both-and type, both Husserl and Saussure, in contrast to

the later either-or stance, where the return to Saussure was typically com-
bined with turning away from Husserl.
It is notable in this regard that the East Europeans were far ahead of
their French counterparts in turning to Husserl’s work for inspiration and
guidance. Consider the translation and reception rate of Husserl’s works
into the Russian and into French respectively as indication thereof. Lo-
gische Untersuchungen (1900/1901) was first translated into Russian in
1909, while the four volumes were first published in French between 1959
and 1963 (Spiegelberg, 1965, p. 731). Philosophie als strenge Wissenschaft
(1911) came out at the same time as the German original, while it was
translated into French by Q. Lauer in 1955 (ibid., p. 731). Ideen I (1913) was
translated by Ricoeur into French in 1951, but the book was widely dis-
cussed in Russia already in 1914. As Husserl’s student Gustav Spet wrote
to his teacher from Moscow on February 26, 1914, “everyone is talking
about phenomenology; there are even special societies for studying phe-
nomenological problems . . . The opinion of phenomenology is everywhere
high and positive; phenomenology is considered as the first and new step
of philosophy” (cited in Holenstein, 2005, p. 12). The French reception of
Husserl’s phenomenology was begun some one and a half decades later:
Husserl presented the two lectures of Einleitung in die transzendentale
Phaenomenologie at the Sorbonne in 1929 (translated into French by Ga-
brielle Peiffer and Emmanuel Levinas in 1931 as Méditations Cartési-
ennes); Levinas’s Théorie de l’intuition dans l’oeuvre de Husserl was pub-
lished in 1930. Yet the distinctiveness of the Eastern European reception
of Husserl lies not only in its relative advance over the Western one, but
also in its programmatic vision. The early reception of Husserl belongs to
a period of intense intellectual interest in language both in philosophy and
science; structural analysis is regarded compatible with the phenomeno-
logical method because of a shared commitment to locating general prin-
ciples proper to the phenomenon itself. Its approach is therefore both theo-
retical and empirical, less vested in safeguarding the purity of philosophy
as an autonomous discipline and more in illuminating the phenomenon
from as many angles as needed.
It may be that Western hegemony dictates the usual association of struc-
turalism with intellectual developments in the France of 1960s; the split
between structuralism and phenomenology is a direct consequence of this
assumed single point of origin of the structuralist tradition on the western
side of Europe, and its dissociation from the phenomenological tradition.
As soon as one accepts that structuralism was not only born once, and that
its Parisian birthplace belongs to a larger map of fecund intellectual centers

that includes the schools of Moscow, Prague, and Kazan, then the relation
between the phenomenological and the structuralist traditions becomes de-
cidedly more complex. Instead of outright antagonism and animosity, one
finds continuity and collaboration, without any strict dividing lines being
drawn between the methods and subject matter.
Looking back on the beginnings of the structural trend in general lin-
guistics in the late twenties and early thirties (Jakobson, 1971, pp. 713–714),
Jakobson notes that structural linguistics is reproved for a supposed es-
trangement from philosophy whereas “in reality the international protago-
nists of this movement had close and effective connections with phenome-
nology in its Husserlian and Hegelian versions” (ibid., p. 713). Specifically,
Jakobson notes that the Moscow Linguistic Circle was a site of “continuous
and ardent debates led by Gustav Spet—in Husserl’s opinion, one of his
most remarkable students” dealing with the linguistic use of the Logische
Untersuchungen (1900/1901). The Moscow Circle also debated Edmund
Husserl’s and Anton Marty’s project of returning to a universal grammar as
conceived by the rationalism of the seventeenth and eighteenth centuries
(Port Royale), and anticipated by the medieval philosophers of language
(Aquinas). The later founder of the Prague Linguistic Circle, Vilem Mathe-
sius, was also heavily influenced by phenomenology, especially the Bren-
tano school, via two of Husserl’s friends, T. G. Masaryk and Marty. As for
the Prague Circle, “Husserl’s ideas and his memorable personal address on
November 11, 1935—‘Phaenomenologie der Sprache’—met with a respon-
sive welcome” (ibid., pp. 713–714).5
In light of this complex intellectual heritage, it is unsurprising that an
early structuralist may describe, without deflating, the speaking subject’s
involvement in linguistic meaning. Jakobson, for example, notes that there
are degrees of freedom in language: at the lowest level of the phoneme,
understood as a bundle of distinctive features like nasalized/nonnasalized,
vocalized/nonvocalized, the freedom of the individual speaker is null; yet
the distinctive features are nothing more than a “negative charge,” and
while they are at the service of meaning, they are themselves devoid of
meaning (Jakobson, 1978, p. 109). The freedom to combine phonemes into
words is still largely limited, although it allows for neological formations;
5
It should also be noted that members of the Copenhagen Linguistic Circle like Viggo Broendal
referred to “Husserl’s penetrating meditations on phenomenology” when making a claim for the
need to study the structure of language as a nonderivable and autonomous object (Broendal,
1939). Hendrik Pos (1898–1955), a Dutch student of Husserl, also advanced a phenomenology of
language together with the theory of structural linguistics (Pos, 1939); his reflections exerted
some influence on Merleau-Ponty’s phenomenology of language.

it is at this level that meaning emerges. As we move to the level of forming
sentences, the individual’s freedom increases, as it does further still with
the combination of sentences into utterances and with the creation of novel
contexts—such as the creation of a literary genre or a poetic style unique
to an individual or a group of individuals (Jakobson, 1956, p. 74). In sum,
the speaking subject is not simply a plaything of larger linguistic forces
over which she exerts no control; it is more appropriate to say that lan-
guage is a layered phenomenon, with levels of distinctive interphonemic
features lying beneath conscious grasp and influence (and also at the
threshold of meaning), and levels of creative usage and expression contin-
gent on the subject’s meaning making—but always as claimed by the re-
ceived history, idiomatic turns of phrase, and shared conventions, that are
not exclusively of her own making. Phenomenologico-linguistic analysis
cannot therefore decide between a focus on the subject and/or the struc-
tured system; the phenomenon itself resists such a split, and invites the
scholar to preserve, without collapsing, the two angles onto a complex and
heterogeneous field. Such a dual subjective and systemic focus is typical of
the early structuralism; it is only some four decades later that structural
analysis of language will become construed as a critical response and al-
ternative to phenomenological description of subjectivity.
However the historical genealogy of structuralism and phenomenology
gets written, a contemporary reader may still resist too intimate a rapport
between the two traditions. Does not phenomenology presuppose the notion
of a constituting subject which is in principle at odds with the structuralist
emphasis on the system based generation of meaning? Foucault’s com-
ments are exemplary of such a view when he reflects on how phenomenol-
ogy became gradually eclipsed by structuralism in the postwar period:
So the problem of language appeared and it was clear that phenomenology

was no match for structural analysis in accounting for the effects of mean-
ing that could be produced by a structure of the linguistic type, in which the
subject (in the phenomenological sense) did not intervene to confer mean-
ing. And quite naturally, with the phenomenological spouse finding herself
disqualified by her inability to address language, structuralism became the
new bride. (Raulet, 1983, p. 198)
This quasinatural eclipse of phenomenology by structuralism rests on

the assumption that phenomenology is contingent on the notion of the sub-
ject in the phenomenological sense—that is, in the sense of a source point
of meaning, an engine of the constitution of the world, a remnant of an

idealist understanding of the human being’s place in nature. Only on such
an assumed basis is the disjunction between phenomenology and structur-
alist analysis necessary. However, this disjunction is not found in Husserl’s
own corpus, notably the lesser read sections of the Logische Untersuchun-
gen or the sections of the Cartesian Meditations read with an eye to struc-
tural analysis; nor is such a disjunction assumed by the pre-Husserlian
phenomenologically oriented scholars like Hartmann or Kruszewski.
There may therefore be at least two ways of conceiving and practicing
phenomenology: as a transcendental philosophy of consciousness, situated
in the idealist lineage, and as a descriptive method aimed at teasing out
invariant features of the phenomena themselves. The former risks an
eclipse when confronted with complex phenomena like language because
of its exclusive attachment to the subjective constitution of meaning. The
latter one has the advantage of inclusivity: it does not chase the subject out
of its field of study, but situates it at the junction with systemic processes
found within language. In other words, the latter opts for an arrangement
of cohabitation with the two brides—whatever the risks and tensions in-
volved in such an arrangement may be.
The distinction between the (at least) two conceptions of phenomenology
can be borne out by their methods. The first suspends natural scientific va-
lidity in an effort to retrieve the things themselves (even if the phenomenon
like language were no thing at all, but a relational system). This suspension
or epoche facilitates an investigation of the object in its own lawfulness
(and not according to laws borrowed from other types of being, like the de-
velopmental trajectory of a living organism, or physical attributes of brute
matter). If language is intrinsically shot through with meaning, then its laws
must preserve meaning if they are to be phenomenologically valid. Study-
ing language phenomenologically amounts then to capturing what belongs
to language as it is manifest to us in experience—this is the dual require-
ment of phenomenological reduction (of preconceived notions borrowed
from other fields with the aim of attaining language itself) and phenomeno-
logical attitude (a relation coinvolving the object and the subject of study).
Phenomenological reduction is then not a Cartesian reduction of language
for the sake of attaining a purified field of transcendental consciousness; it
is rather a reduction of preconceived notions about language in the hope of
attaining an understanding grounded in the language itself. Just as one way
to phenomenology is by the science of psychology, there is also a possibility
of a via linguistica, a way by the science of linguistics, so long as the latter
eschews received notions borrowed from other fields and interrogates the
phenomenon itself. When linguists like Kruszewski advocate developing a

phenomenology of language, their commitment is to such a focus on lan-
guage itself; this focus does not imply recourse to a prelinguistic conscious-
ness, the canonical notion of classical phenomenology. If the phenomenon
itself is structurally organized and lends itself to structural analysis, then an
appropriate phenomenological line of approach includes structural notions.
If the phenomenon itself is independent (at least in part) of subjective inten-
tion, then an appropriate phenomenological approach must bracket the ide-
alist preconception of subjective intentions as the basis of ideas. This ap-
proach could be called an ordinary phenomenological one.
The second type of phenomenology is the one presupposed by Foucault,
and implied in the antagonist rendering of phenomenological and structural-
ist traditions. It can be called transcendental-phenomenological, with mean-
ing regarded exclusively in terms of subjective constitution by conscious-
ness. This type is received from the parts of Husserl’s corpus which received
the most attention within the phenomenological movement “proper.” Its ad-
vantage lies in demarcating an autonomous field of philosophical study, that
is, transcendentally purified consciousness, and extending the tradition of
idealism whose primary goal is the analysis of consciousness. This advan-
tage is however also its limitation; historically, language has largely dropped
out of phenomenological focus, and the bridge to empirical approaches to
linguistic phenomena has been all but burnt.
While an antagonistic construal of the relation between phenomenology
and structuralism assumes the validity of the transcendental understand-
ing of phenomenology as a study of objects exclusively in terms of their
constitution by consciousness, the ordinary phenomenological attitude
where objects are investigated according to their own structural lawful-
ness is assumed within the earlier rapprochement between the two tradi-
tions. The ordinary phenomenological attitude was adopted by Jakobson,
and it serves as the best ally of structural analysis. It is also the attitude
adopted by Saussure in his reflections on general linguistics, following
Kruszewski’s lead. In Saussure’s linguistics, language is studied independ-
ently of naturalistic considerations, and in terms of its own intrinsic laws.
Unsurprisingly, Saussure adopts a phenomenological approach and termi-
nology in his autographed writings, with an emphasis on the speaking sub-
jectivity, and the concrete reality of language. Such a phenomenological
stance adopted by the “founding father of structuralism” strikes one as an
aberration and a scandal only if one conceives of phenomenology exclu-
sively in the narrow, transcendental sense. However, this conception is un-
warranted from a historical point of view (it was not assumed by the schol-
ars in question), and it is sterile as a strategy (it robs phenomenology of

rights to study language in its complexity). I propose therefore that linguis-
tic phenomenology be bound to the ordinary phenomenological method—
and that Saussure, together with Husserl and Jakobson, be coopted as pow-
erful allies in that venture.
Phenomenology of language: Merleau-Ponty and Saussure
Merleau-Ponty was one of the few, if not the sole philosopher who identi-
fied a phenomenological dimension within Saussure’s linguistics—a feat
rendered even more remarkable by the fact that his reception was histori-
cally limited to the official doctrine laid out in the Course. Merleau-
Ponty’s reception of the Course is unique in its high tolerance for the com-
plexity if not the paradox of general linguistics, where the distinguished
levels of language as system and speech turn out to be reciprocally inter-
woven and mutually conditioning rather than hierarchically layered and
mutually opposed. Unlike the later structuralist readers of the Course
whose hermeneutic strategies are put in the service of deriving a scientific
program for the human sciences from general linguistics and biased in
favor of an unexamined notion of scientific objectivity, Merleau-Ponty
maintains the ambiguous conjuncture of the objective and the subjective in
language, in accordance with the precepts of phenomenology. He may, in
accordance with the later structuralist readers of the Course, regard gen-
eral linguistics as foundational for the human sciences and philosophy—
but without sacrificing philosophical reflection for the sake of scientific
success in the process. His approach thus demonstrates that a philosophi-
cally complex reading of the Course is indeed possible—albeit exception-
ally rare, and largely eclipsed by the dominant structuralist reception.
Merleau-Ponty was in fact concerned with language even before his ex-
posure to Saussure in the late 1940s, as evidenced by “The Body as Ex-
pression, and Speech” chapter from the Phenomenology of Perception
(1945), where a gestural theory of meaning and expression is laid out as a
basis for understanding language. It is the encounter with Saussure, how-
ever, that inaugurated an over a decade long engagement with linguistics.
One can identify therefore a veritable “linguistic” phase within his overall
philosophical trajectory, albeit, with a decidedly non-structuralist empha-
sis on language as living speech.
Merleau-Ponty’s engagement with linguistics can be dated back to the
1947–1948 course at the University of Lyon on “Language and Commu-
nication” (unpublished; summarized in Silverman’s Inscriptions: After

Phenomenology and Structuralism, 1997), followed by the 1949–1950
course at the University of Paris on “Consciousness and the Acquisition
of Language” (published under the same title, 1973). Between 1950 and
1952 Merleau-Ponty worked on a booklong project dealing with linguis-
tic and literary experience, tentatively titled The Prose of the World (un-
finished and published posthumously, 1973a). He authored a series of
essays dealing to some extent with the problem of language, notably the
1951 “Phenomenology and the Sciences of Man” (in The Primacy of
Perception and Other Essays, 1964) and “The Philosopher and Sociol-
ogy” (in Signs, 1964a), the 1952 “Phenomenology of Language” and “In-
direct Language and the Voices of Silence” (also in Signs), the 1953 “An
Unpublished Text by M. Merleau-Ponty: A Prospectus of His Work” (a
prospectus presented as part of his candidacy to the College de France,
published in Primacy of Perception) and In Praise of Philosophy (an in-
augural address to the College de France, published under the same title,
1988). From 1953 to 1954 he gave the lecture series on “The Sensible
World and the World of Expression,” “Studies in the Literary Use of
Language,” and “The Problem of Speech” at the College de France
(summarized in the Themes from the Lectures at the College de France,
1970); the 1959 essay “From Mauss to Claude Levi-Strauss” (in Signs)
also belongs to the “linguistic” phase.
Like Saussure before him, Merleau-Ponty did not complete a booklong
treatise dealing with language. His major work dealing with philosophy of
language, The Prose of the World, remains unfinished. It was half-completed
when Merleau-Ponty applied to the College de France, yet apparently he lost
interest in the project around 1952–1953, and abandoned it in 1959 (see Le-
fort’s introduction to The Prose of the World). His other essays and lectures
dealing with language were never unified into a coherent body of work. As
in Saussure’s case, this may be a testimony to the difficulty of the task at
hand, rather than a mere failure, and one finds plentiful resources for a phe-
nomenology of language within the extant texts. It is also notable that
Merleau-Ponty’s phenomenological reception of Saussure only gains addi-
tional support in light of the manuscript writings by Saussure himself; his
unfinished work emerges therefore as a more authentic appropriation of Sau-
ssure’s contributions to general linguistic theory than the official doctrine,
and its structuralist reception.
Merleau-Ponty regards a phenomenological approach to language as a
much needed remedy for the crisis engendered by the existing scientific or
observational approach. The scientific approach is directed toward an al-
ready established or instituted language, for example, a body of written

texts studied in philology, or a system of phonological, morphological, and
syntactic relations subjected to structural analysis. It therefore regards lan-
guage solely “in the past” (Merleau-Ponty, 1964a, p. 104), and as an aggre-
gate of externally related elements without intrinsic unity. To adopt an ex-
clusively empirical study of language means then to “pulverize [it] into a
sum of fortuitously united facts” (ibid., p. 39), and therefore to miss the
preexistent unity of language as a communicative medium shared by a
community of users. The crisis engendered within an unreflected empiri-
cal study by such inevitable fragmentation of language into atomic facts
can be rectified by integrating a phenomenological approach into linguis-
tics. Here language gets recovered as a circumscribed field of subjective
expression and intersubjective communication. In Merleau-Ponty’s words:
Taking language as a fait accompli—as the residue of past acts of significa-

tion and the record of already acquired meanings—the scientist inevitably
misses the peculiar clarity of speaking, the fecundity of expression. From
the phenomenological point of view (that is, for the speaking subject who
makes use of his language as a means of communicating with a living com-
munity), a language regains its unity. It is no longer the result of a chaotic
past of independent linguistic facts but a system all of whose elements co-
operate in a single attempt to express which is turned toward the present or
the future and thus governed by a present logic. (ibid., p. 85)
The phenomenologist adopts the speech situation as an inalienable

ground of any empirically sound inquiry into language—including a study
of languages from the past:
To know what language is, it is necessary first of all to speak. It no longer

suffices to reflect on the languages lying before . . . in historical documents
of the past. It is necessary to take them over, to live with them, to speak
them. It is only by making contact with the speaking subject that I can get a
sense of what other languages are and can move around in them. (Merleau-
Ponty, 1964, p. 83)
The recourse to speech or lived language would then amount to me-

thodical subjectivism, an internal take on language even within its histori-
cal dimension, and history would emerge less as a sequence of external
events derived from dusty documents, but rather as a coexistence of the
contemporary speaker with the extinct subjectivities, an intimate copres-
ence to a system of expression, which includes all other presents too

(Merleau-Ponty, 1973a, p. 25). Diachrony is but a sequence of synchronic
arrangements:
When I discover that the social is not simply an object but to begin with my
situation, and when I awaken within myself the consciousness of this
social-which-is-mine, then my whole synchrony becomes present to me,
through that synchrony I become capable of really thinking about the whole
past as the synchrony it has been in its time, and all the convergent and dis-
cordant action of the historical community is effectively given to me in my
living present. (Merleau-Ponty, 1964a, p. 112)
The phenomenological turn to the speaking community, situated in

time, provides a remedy to the crisis within a solely objective linguistic
science, and needs to be injected into the empirical study of language; like
Kruszewski and Jakobson, Merleau-Ponty regards the set of phenomeno-
logical methods as a useful and workable scientific program rather than a
self-standing and self-sufficient reflective inquiry. Phenomenology is not
an alternative to science, but a guide of how to reform science from a
purely objective model to a subjective and objective one. Importantly,
Merleau-Ponty locates such a phenomenological reform within “certain
linguistic investigations” which “anticipate Husserl’s own” (ibid., p. 105),
and notes that “certain linguists . . . without knowing it tread upon the
ground of phenomenology” (ibid.). The linguists treading upon phenome-
nological grounds feature especially Ferdinand de Saussure, whose gen-
eral linguistics is taken up in Merleau-Ponty’s 1953 Course at the College
de France on the problem of speech (Merleau-Ponty, 1970). Receiving gen-
eral linguistics in a decidedly non-structuralist manner, Merleau-Ponty
states that Saussure adopted “speech as his central theme” in the Course
(Merleau-Ponty, 1970, p. 19); Merleau-Ponty’s own course would have
sought “to illustrate and extend the Saussurean conception of speech as a
positive and dominating function” (ibid., p. 20).
On Merleau-Ponty’s reading, Saussure’s semiology is a science founded
on speaking subjectivity: the subject must transcend the signs toward their
signification in speaking; they only hold it in abeyance (Merleau-Ponty,
1964a, p. 88). Similarly, the very definition of the sign as composed of two
inseparable signifying and signified facets can only be offered from the
perspective of living speech, rather than as an objective property of the
sign itself; in Merleau-Ponty’s terms, while “the rigid distinction between
sign and signification . . . seemed evident when one considered instituted
language alone,” it “breaks down in speech where sound and meaning are

not simply associated” (ibid., p. 18). The structuralist view that the signi-
fied is transcendent to the semiological system would then be guilty of bias
in favor of the instituted language alone, and disregarding the possibility
of linguistic expression where speech, like gesture, is saturated with mean-
ing. In agreement with the student lecture notes on general linguistics, the
thesis of absolute arbitrariness of the sign gets relativized as soon as the
sign rejoins its immediate milieu of signifying practice, and ceases to be
viewed formally, within a logical set of operations. As a result, language
ceases to appear as a closed and autonomous system—language must be
spoken, and there is contingency in its logic (ibid., p. 88). Instead of a dis-
tinction between the deep structural level and the surface phenomenon of
speech, there is a crisscrossing or a chiasm between expression and lan-
guage systems: “Already in Saussure [CLG] . . . speech is far from being a
simple effect [of language], it modifies and sustains language just as much
as it is conveyed through it” (ibid., p. 19). Speech can therefore rewrite the
code of language, and its actions feed back into the source.
The turn to language as a signifying practice recovers the essential
unity of language which a purely historical (and external) approach must
miss:
Saussure shows admirably that if words and language in general, consid-

ered over time—or, as he says, diachronically—offer an example of vir-
tually every semantic slippage, it cannot be the history of the word or
language which determines its present meaning . . . Whatever the haz-
ards and confusions in the path of the French language . . . it is still a fact
that we speak and carry on dialogue, that the historical chaos of lan-
guage is caught up in our determination to express ourselves and to un-
derstand those who are members of our linguistic community. (Merleau-
Ponty, 1973a, pp. 22–23)
Merleau-Ponty thus locates a core of reason or logos at work within lan-

guage insofar as it facilitates communication and mutual understanding
within a speech community. It is this reason that binds language together
from within—but its logic is tainted with contingency, accidents and ex-
ceptions are always possible and become amassed in its net without yet
leading to pure chaos (Merleau-Ponty, 1964a, p. 88). “Saussure has the
great merit of having taken the step which liberates history from histori-
cism and makes a new conception of reason possible . . . The mutations in
every signifying apparatus, however unexpected they may seem when
taken singly, are integral with those of all the others, and that is what

makes the whole remain a means of communication” (Merleau-Ponty,
1973a, p. 23). Saussure would then have identified a pact of reason within
language—a basic trust in this medium of world- and self-disclosure, an
ongoing contract with one another that no amount of deceit has undone; as
speakers, we are the custodians of this commonwealth and our Mitsein
maintains its, however elusive and fragile, unity.
Merleau-Ponty both locates a phenomenological orientation within
Saussure’s linguistic works, and finds an echo of Saussure’s emphasis on
living speech within Husserl’s writings, notably the Ideen and Krisis. Hus-
serl’s work would be a subsequent iteration of Saussure’s earlier turn to
living speech and subjectivity: “Husserl does not say it . . . but it is hard
not to think of Saussure when Husserl insists that we return from language
as object to the spoken word” (Merleau-Ponty, 1964a, pp. 105–106). Hus-
serl would be “approximating the task which Saussure had set before him-
self: to return to the speaking subject in its linguistic context—which for
Husserl is a fullness and for Saussure a system of differences” (Silverman,
1997, p. 160). Merleau-Ponty thus complicates the received view of Saus-
sure and Husserl as foundational figures for the two opposing schools of
thought: structuralism and phenomenology; the reader can instead envis-
age a phenomenology predating its presumed foundation, since outlined
in Saussure’s work and then echoed in Husserl’s; the reader is also made
to realize that the foundation of structuralism is fractured at best, and
strangely contaminated by its presumed opposite.
The phenomenological orientation shared by Husserl and Saussure can
be fleshed out under the heading of “evidence and difference” (to adopt the
distinction from Roggenbuck, 1997). Both Husserl and Saussure seek after
the essences, and suspend reference to a natural scientific approach to the
phenomena under investigation. While Husserl adopts the possibility of
direct access to the essences via the evidence of intuition, Saussure opts for
an indirect access, mediated by the differential character of the semiolog-
ical system. Saussure thus lets language itself guide the phenomenological
method, and the possibility of attaining the essential meaning of language
is subjected to the diacritical principle governing any linguistic meaning.
It is the logos of language that dictates the methods of linguistic phenom-
enology, as the task of phenomenology would consist in heeding to what
language itself is saying about itself, following its lead. Phenomenology of
language reverses therefore into language as (and of) phenomenology, lan-
guage as a site of signifying donation, already instituted and thus predating
subjective intention, and yet necessarily enacted and revealed to a speaking/
listening subject, and therefore not subjectless. This phenomenological

notion of language undercuts the transcendental/structuralist divide be-
tween the signifying subject and the semiological system; the system is
already there, and we are continually borrowing from its resources when
we listen and speak. And yet the system is neither closed nor complete, it
does not expel the subject like a foreign body, but lies open to new speakers
and novel instances of usage and inhabitation. It lives in and through
speech, but is not exclusively configured by consciousness, intention, and
intuition. Nor does it evacuate these subjective terms altogether. Each term
gets relativized through the relation to what was thought to be excluded by
it, and thus gets shot with indirectness, difference, incompleteness.
This is in outline the approach of linguistic phenomenology: it occupies
the ambiguous juncture of the borrowed and the self-made, the contempo-
rary and the transgenerational, the novel and the sedimented, from which
the subject speaks. This juncture is the meeting point of the two tracks
running across the field of language, itself dual or even divided between
“two languages”: le langage parlé and le langage parlant, the sedimented
language and speech language as an institution and language “which cre-
ates itself in its expressive acts, which sweeps me on from the signs toward
meaning” (Merleau-Ponty, 1973a, p. 10). The phenomenological orienta-
tion to speech is then already entangled within language, and cannot re-
trieve the hypothetical standpoint of the universal and timeless constitut-
ing consciousness that Merleau-Ponty attributes to the early Husserl
(Merleau-Ponty, 1964a, p. 85). It would not consist in the Kantian task of
determining the conditions of possibility for any and all possible lan-
guages. The philosopher, rather is “the one who realizes that he is situated
in language, that he is speaking” (ibid.). Linguistic phenomenology does
not seek to retrieve the “conditions without which” there is no language,
and which a purely thinking subject could enumerate, before an entry into
language properly so called. From the position of a speaking subject, such
a return instantiates the myth of origin, and a vain attempt to step outside
the boundaries of experience for the sake of objectivity. But Husserl’s ob-
jection to Kant retains full force, for any attempt to capture the preexisting
conditions of experience transcends the boundaries of situated experience
and falls outside the province of phenomenology, whose methods are de-
scriptive and steeped in the concrete reality of what appears. Husserl’s
objection applies to a study of linguistic experience (as to any other),
which is a sine qua non situation for any practitioner of phenomenology; it
therefore makes it possible to pursue the phenomenological study in a rig-
orous and methodical manner from within the linguistic field, without the
need to recover its preconditions from the mythical past of pure thought.

A rigorous phenomenological method retains the ambiguous position of a
subject of philosophical thinking, who is both a beneficiary of language,
“enveloped and situated [with]in it” (Merleau-Ponty, 1964a, p. 85), and a
source of new growth and interminable fermentation which coherently
deforms the existing fabric of language, without dissolving it.
Merleau-Ponty acknowledges the transformative effects that a refocus-
ing of study from consciousness to language has on the discipline of phe-
nomenology as such. Whereas a phenomenology of consciousness is tied
to the perspective of constitution, it is perennially haunted by the problem
of others, who appear theoretically excluded from the subjectively figured
world (who would not be truly other, but serial subjects). Recentering phe-
nomenology in speech dissolves these problems, for the self finds itself
situated in a preconstituted world, always already with others (ibid., p. 95).
Hence Merleau-Ponty concludes:
When I speak or understand, I experience that presence of others in myself

or of myself in others which is the stumbling-block of the theory of inter-
subjectivity, I experience that presence of what is represented which is the
stumbling-block of the theory of time, and I finally understand what is
meant by Husserl’s enigmatic statement, “Transcendental subjectivity is in-
tersubjectivity.” To the extent that what I say has meaning, I am a different
“other” for myself when I am speaking; and to the extent that I understand,
I no longer know who is speaking and who is listening. (ibid., p. 97)
Speech introduces a shared medium which is the “solution” to the prob-

lem of others, without dissolving them; it is an ambiguous third that medi-
ates the relation between the two; it is neither wholly my own nor the
others, and that is why we inhabit it both. A phenomenology of sociality
cannot therefore bypass a phenomenology of speech: speech is an evolved
form of social being, and sociality comes into sharper focus when regarded
through its lens.
Merleau-Ponty’s claim that Saussure was treading upon the ground of
phenomenology is unusual in light of the instituted association between
Saussure and structuralism. Pace the Course and its canonical interpreta-
tion, Saussure’s linguistic science would have the language system (la
langue) as its direct and sole object; it would be distinguished from a sci-
ence of speech (la parole), which was left undeveloped by Saussure, and
focused on the phonetic evolutions of languages over time; it would view
speech primarily as a psychophysiological process and would thus share its
object with the disciplines of phonology and phonetics. Only the linguistics

of la langue could then be demarcated as a new and autonomous scientific
discipline, distinct from the existing empirical sciences of phonetics, pho-
nology, and philology. Its method would be synchronic study of a static and
rule governed system of signs. In the linguistics of la parole, the focus
would be diachronic, covering the contingent and external (to the language
system) developments of speech productions in history. The separation of
the two scientific approaches would be motivated (or possibly motivate in
turn) a separation into two distinct areas of study—structure and speech.
Saussure would thus have laid the foundations of a structuralist approach
to language, as an alternative to the phenomenological focus on the signi-
fying subject. Structuralism would be a science, that is, an objective ap-
proach, unaffected by the subjective experience of language use in speech.
Saussure would then be a traditionalist in terms of method, but an innova-
tor in terms of the object, the latter being neither a loosely hanging collec-
tion of empirical facts about particular languages, nor a series of written
records, nor the functioning of the vocal apparatus and its products, but the
systematic and rule governed organization at the deep level which provides
the conditions of possibility for any surface phenomenon. La langue would
serve as a condition of possibility and set of inherent principles governing
any of its manifestations. Such a study seems at the opposite end from the
phenomenological commitment to the primacy of speech and the inalien-
able standpoint of the experiencing subject. Did Merleau-Ponty simply
misunderstand the basic claims of Saussure’s linguistic, and read into it
what obviously wasn’t there?
Merleau-Ponty recognizes the primacy of the synchronic perspective in
Saussure’s study; on his reading, however, synchrony is bound to the sub-
jective (albeit socially modulated) experience of speech. The two catego-
ries in Saussure’s linguistics would then be “a synchronic linguistics of
speech (parole),” and “a diachronic linguistics of language (langue)”
(Merleau-Ponty, 1964a, p. 86); with the synchronic linguistics of speech
revealing “at each moment an order, a system, a totality without which
communication and linguistic community would be impossible” (Merleau-
Ponty, 1973a, pp. 22–23).
Merleau-Ponty appears, on first sight, guilty of a double oversight in his
reading of Saussure. He gives primacy to la parole over la langue, whereas
the primary object of linguistic or semiological study would have been la
langue. Furthermore, he raises the possibility of a “synchronic linguistics
of speech (parole),” distinguished from a diachronic linguistics of lan-
guage (langue), in disregard of the oft lamented “fact” that Saussure failed
to deliver a linguistics of speech, and of the usual alignment of synchrony

with la langue and diachrony with la parole. Unsurprisingly then, Ricoeur
charges Merleau-Ponty’s unorthodox distinction between the synchronic
linguistics of speech and the diachronic linguistics of language with being
no more than an error (Ricoeur, 1967, p. 12). Merleau-Ponty is charged
with a misguided attempt to incorporate objective structures into the sub-
jective point of view, and to force the historical weight of language into the
present of the spoken word. A synchronic linguistics of speech would then
itself be a double misnomer, since speech—on a structuralist reading of
the Course, is neither a properly linguistic, nor a synchronic entity, and de
facto falls completely outside a scientific study of language. It is not sys-
tematic and rule governed in the way la langue is, and for this very reason,
it does not lend itself to a snapshot-like view in the present. Speech would
then be a purely empirical and historical datum, relegated to a natural sci-
entific study within phonetics and physiology, as well as to historical lin-
guistics; it would not be amenable to the new semiological program de-
vised by Saussure.
Upon closer view, Merleau-Ponty’s “error” contains, albeit in an embry-
onic form, a more faithful response to Saussure’s project than the received
structuralist one. It maintains a commitment to subjectivity within a philo-
sophically inflected study of language. Merleau-Ponty gleaned this com-
mitment from the published edition of the Course, but the source materials
of Saussure’s thought lend direct support to such a phenomenological inter-
pretation. Saussure may not have expressly posited a “synchronic linguistics
of speech,” but he does accord priority to the speaking subject in linguistics,
and insists on the need to study language as an act and a phenomenon tied
to experiencing consciousness in the manuscript sources. His emphasis on
the “language acts” (actes de langage) approximates Merleau-Ponty’s com-
mitment to language as speech. They share an emphasis on the social di-
mension of language, without yet reducing the speaking individual to an
impersonal set of societal conventions. They emphasize a reciprocal inter-
dependency between the daily praxis and the historically sedimented insti-
tution of language, against the more widespread view of a hierarchy of
levels. If it made sense to evoke a spirit of Saussureanism, Merleau-Ponty’s
phenomenology of language would have come closest to distilling it.

III The inception and the reception
of the “Saussurean doctrine”:
the Course
7 The editorial inception of the Course:
Bally and Sechehaye
W
ithin twentieth-century philosophy and human sciences,
the Course functions primarily as a site of the official doctrine
closely associated with structuralism, that is, as a statement of
the familiar oppositional pairings between the signifier and the signified, la
langue and la parole, synchrony and diachrony. Considering the hold that
structuralism had on Saussure’s linguistics within twentieth-century Euro-
pean philosophy and related fields, it is common to consider these opposi-
tional pairings as a distinctive feature and a shared trait of Saussureanism
and structuralism. They figure as the general principles that were applied to
linguistic study in the Course in the 1910s, and then extended to the broader
field of philosophy and the human sciences within post-Second World War
French structuralism (and then challenged within poststructuralism, nota-
bly by Derrida). This assumed continuity between a “protostructuralist”
doctrine and French structuralism can also be rendered by the fact that it is
common in the scholarship to define structuralism proper as an intellectual
movement with a distinctive Saussurean lineage, and to exclude the strands
of structuralist scholarship that do not share in the Saussurean legacy (see
ch. 8 for development).
The shared Saussurean/structuralist commitment to the familiar oppo-
sitional pairings between the signifier and the signified, la langue and la
parole, synchrony and diachrony, was made possible by the production,
replication, and reception of the Course as a site of the official doctrine.
The emergence and impact of these intellectual ideas is therefore undeci-
pherable without understanding the material and institutional history that
led to their production and reception—that is, the editorial production of
the Course in the early 1900s and the dominant structuralist reception of
the Course in the 1950s and 60s. In this chapter, I propose therefore to crit-
ically examine the official doctrine not exclusively by way of a philosoph-
ical critique (which bears especially on the logic of violent hierarchies
deployed within these familiar dichotomies) but by way of reconstruction
of the material and institutional history that went into the making and re-
ceiving of the Course as a privileged textual locus of this dichotomous
logic. I contend that the emergence and impact of the “Saussurean doc-
trine” on the developments in twentieth-century philosophy and related
fields is effectively undecipherable without understanding its relation to
the material and institutional history that led to the production and domi-
nant reception of the Course. The editors, Charles Bally and Albert Seche-
haye, performed a double duty in the production process: they projected
their own methodological and conceptual commitments onto the source
materials in order to establish general linguistics as a recognizable scien-
tific discipline, and subsequently received and replicated the basic princi-
ples of this science in a series of dedicated book reviews of their ghostwrit-
ten text. The editors’ role extends beyond the initial inception to a scholarly
reception of the Course as official doctrine. Considering that it would be
especially difficult for an Anglophone reader to access and/or assimilate
the materials related to the editorial production (and reception) of the
Course otherwise, I will reconstitute this process in requisite detail. I will
then discuss the structuralist and poststructuralist reception of the Course
by Levi-Strauss, Lacan, and Derrida, as a further development within the
process of establishing the Saussurean doctrine (ch. 8).
Ghostwriting
The editorial history of the Course is revealing of the functioning of aca-

demic institutions in twentieth-century Europe, and beyond. It offers an
interesting case study of how personal influence, prestige, and at times,
intimidation and silencing of dissidents act as a real force driving the course
of ideas. That Bally and Sechehaye were able to publish a book in Saus-
sure’s name is contingent on an elaborate institutional feat: gaining sole
control of the student lecture notes from the course on general linguistics
that neither of them attended; undermining the efforts by other scholars
and the students themselves to publish the lecture notes in their entirety by
portraying them as unreliable witnesses of Saussure’s teaching; usurping
the role of direct disciple to Saussure and ghostwriting a book in the mas-
ter’s name; claiming to read Saussure’s own ideas and intentions after his

death, as if communicating directly with his ghost. This editorial history
merits studying in some detail, especially in light of the increasing aware-
ness of how a material and institutional context enables and constrains any
intellectual inquiry, including an inquiry into philosophical ideas.
The self-appointed editors of the Course were Charles Bally (1865–
1947) and Albert Sechehaye (1870–1946). Both were engaged as Privat-
dozent at the University of Geneva, each author of a book on linguistics
(Traité de stylistique française, 1909, and Programmes et méthodes de la
linguistique théorique, 1908, respectively) and a handful of other articles
at the time the edition was undertaken. They counted among the small
circle of Saussure’s students at the University of Geneva, Switzerland,
Bally having followed Saussure’s lectures from 1893–1906 (Godel, 1957,
p. 16; De Mauro, 2005, p. 344), and Sechehaye from 1891–1893. The edi-
tors took part in Saussure’s lectures on comparative grammar, Gothic,
Lithuanian, and Sanskrit. According to a widespread scholarly view, the
editors also attended the lectures in general linguistics taught by Saussure
from 1907 to 1911, and were therefore in a privileged position to compile a
volume in general linguistics based on a redaction of their own lecture
notes from the course after Saussure’s untimely death. This view is ex-
pressed, for example, by Antoine Grégoire, who notes the following about
the Course in a review from 1923:
Sirs Bally and Sechehaye realized the pious undertaking (la pieuse entre-
prise) of redacting the notes they took during the course in linguistics given
by the illustrious F. de Saussure. (Mounin, 1968, p. 176)
This popular view is false. Neither Bally nor Sechehaye attended a

single lecture on general linguistics given by Saussure from 1907 to 1911.
The editors of the Cours de linguistique generale were not themselves the
auditors of the courses they subsequently assumed sole charge of. This
absence is even more striking considering that they became the firsthand
authors of some of the influential formulations found in the Course. Their
ghostwriting of a volume in Saussure’s name may therefore not be simply
an exercise of filial piety, but also of (usurped) filial privilege which Bally
and Sechehaye successfully assumed after Saussure’s death.
Bally and Sechehaye publicly situated themselves as disciplines of notre
grand maître Ferdinand de Saussure (Bally, 1908), and univocally as-
sumed this privileged bond in their Preface to the Cours (Bally and Seche-
haye, 1916, p. 9). The professed master, however, is of comparative gram-
mar and Sanskrit, not general linguistics (Amacker, 2000, p. 208). The
the editorial inception of the course 195

master-disciple projection may well have been a sign of genuine admira-
tion and respect, assuming at times the form of religious piety for a godlike
figure (Redard, 1982, p. 21). It also secured the filial right of property over
the remaining intellectual estate, enabling Bally and Sechehaye to serve as
self-designated apostles to Saussure, out on a mission to convert his oral
lessons on general linguistics into a legible and lasting testament. The
Course bears therefore a closer tie to the format of a testament than to a
Vulgata editio, that is, a popular translation of the Roman and Hebrew
Bible, to which it is usually compared. Like the Book of the New Testa-
ment, the Course offers a written record of oral teaching of a past master
by his (presumed) disciples who capitalize on the (presumed) direct cove-
nant with the master and usurp the privileges associated with being the
(presumed) direct witnesses to his earthly teachings as members of the
small circle of Saussure’s students in Geneva.1 This combined process of
bearing witness to the master’s will and partaking in an alliance with the
master is enabled by the broad semantic spectrum of the Latin testamen-
tum which means both “last will disposing of property” and a “covenant,
pact, alliance” (the term could then have been used in the sense of a “cov-
enant” in the account of the Last Supper, but also interpreted as Christ’s
“last will”). If the Course combines the rhetorical forces associated with a
document construed as a testament, it is empowered to mark the special
pact between the teacher and the apostles and to pass down the teacher’s
last will disposing of his (intellectual) property, like from father to son.2
These combined strategies gain efficacy within the broad semantic spec-
trum associated with testamentary practices.
Public expression and recognition of direct covenant with a master-
disciple relation was and continues to be of service in a traditional Euro-
pean academic institution. Since intellectual heritage tends to pass directly
from master to disciple like from father to son, the filial right to handling
the paternal property is simply assumed once the filial bond is publicly
1
The group of students who attended Saussure’s course on general linguistics included P. F. Regard,
Albert Riedlinger, Louis Caille, Leopold Gautier, Francois Bouchardy, Emile Constantin, George
Dégallier, Marguerite Sechehaye-Bourdet, and Francois Joseph (Engler, 1989, pp. x–xii).
2
Testamentum was used by Latin ecclesiastical writers to translate the Greek diatheke. “With the
profane authors this latter term means always, one passage of Aristophanes perhaps excepted, the
legal disposition a man makes of his goods for after his death. However, at an early date, the
Alexandrian translators of the Scripture, known as the Septuagint, employed the word as the
equivalent of the Hebrew berith, which means a pact, an alliance, more especially the alliance of
Yahweh with Israel. In St. Paul (1 Corinthians 11:25) Jesus Christ uses the words ‘new testament’
as meaning the alliance established by Himself between God and the world, and this is called
‘new’ as opposed to that of which Moses was the mediator” (Catholic Encyclopedia [CD-ROM],
New Advent, 2003).

recognized. After all, the relations of influence and prestige in academia
tend to flow along the same channels as those found in the greater social
and political world; even though they are not usually written into the public
record, or acknowledged as the enabling and constraining conditions of
the history of ideas, academic alliances effectively shape the production,
dissemination, and reception of intellectual content just as they determine
the passage of property and power in the other areas of life. They are usu-
ally kept behind the scenes of the actual performance but even though
obscure, they provide the necessary backstage. The editors of the Course
were expert in directing this backstage process, as discussed next.
Bally and Sechehaye initiated a popular demand for a book on general
linguistics seven days after Saussure’s death (who died on February 22,
1913 at the age of fifty-five). In a laudatio professed on March 1, 1913,
Bally enters stage to issue an unequivocal call to action, using combined
rhetorical strategies of funeral lament and question begging:
Oh, why did he [Saussure] confine the treasures of his genius (génie) to a
small circle of disciples? So many of his lessons could have been published
right away; so many ideas buried in the student notes, so many riches laid
in personal manuscripts, too secretly guarded! Will none of it ever see the
light? Will we resign ourselves to let the many flashes of his unique spirit
(esprit) die down? (Bally, 1988, p. 74, my translation)
These lessons have been piously written down by his students; they would
give rise to a beautiful book (un beau livre); will it never see the light?
(ibid., p. 73, my translation)3
Bally actively creates a sense of urgency around the fact that Saussure
failed to bestow a book upon posterity. A beau livre is declared a filial
right, the only vessel worthy of safeguarding the treasures of Saussure’s
genius which otherwise risks fleeing into the maze of student notes or
Saussure’s own personal documents. Un beau livre will be a bottle in which
to catch the génie, the ghost of his genius. It will also be an academic tomb-
stone, a publicly recognized memorial which mediates the relation between
3
Ah! Pourquoi a-t-il [Saussure] réservé au cercle restreint de ses disciples le trésors de son génie!
Tant de ses leçons auraient pu être aussitôt imprimées; que d’idée enfouies dans des notes
d’étudiants, que de richesses enterrés dans des manuscrits personnels, trop jalousement gardés!
Tout cela ne montrera-t-il jamais le jour? Nous résignerons-nous à voir s’éteindre tant d’éclairs
jaillis de cet esprit unique? (p. 74)
Ces leçons ont été pieusement recueillies par ses élèves; le livre qui en sortirait serait un beau
livre; ne verra-t-il jamais le jours? (p. 73)

the dead and the living, between the ghost of the father and the self-elected
son and intellectual successor. The book will serve to contain the specter
of Saussure in a portable dogma, it will prevent the specter from making
visitations of his own accord, as in Hamlet’s ghostly father coming back
from the dead to haunt the living. Using the assumed link of filiation, the
privileged speech position from which to address oneself to the parental
ghost, Bally and Sechehaye, will communicate with the spectral Saussure
and write the book as if it were dictated by the past master himself.4
They will also effectively undermine alternative publication attempts,
notably the attempt to publish Paul Regard’s lecture notes by Antoine Mei-
llet and Regard himself. Meillet was a renowned Parisian linguist, Saus-
sure’s student and friend of long date; Paul Regard was one of the students
who attended Saussure’s lectures on general linguistics in Geneva. Meillet
highly estimated Regard the linguist, and endorsed the latter’s monograph:
Contributions à l’étude des prepositions dans la langue du Nouveau Tes-
tament (1918).
Bally’s express opinion of the student, and of the publication project,
clashed with Meillet’s. Having learnt of the publication project from
Regard himself, Bally consults with two of Saussure’s students, Leopold
Gautier and Marguerite Sechehaye-Burdet.5 Their stated views, communi-
cated by Bally to Meillet in a letter from May 29, 1913 are as follows:
They agree exactly on the following points: even though the principles in
the teaching have not changed much, each course . . . has its own character
and originality, and the details of one can enrich the other two without lead-
ing to a contradiction . . . Everyone insists on this point: whatever the pub-
lication format, the work should not be based on the lecture notes of one
student of a single year . . . (Amacker, 1989, pp. 102–103, my translation)
This expression of the students’ own views on the matter suggests that
publishing the majority if not the totality of the lecture notes on general
linguistics would be the best testimony to Saussure’s teaching. For his part,
Bally puts it to use against the Meillet/Regard planned publication project,
4
See Derrida, 1994, p. 12, for a discussion of the filial privilege and spectral speech.
5
Marguerite Sechehaye-Burdet was a famed psychoanalyst, and author of Autobiography of a
schizophrenic girl (New York: Grune and Stratton, 1956 [1951]); A new psychotherapy in
schizophrenia: relief of frustrations by symbolic realization (New York: Grune and Stratton,
1970 [c.1951]); and Symbolic realization (New York: International Universities Press, 1952). She
married Albert Sechehaye in 1908 and, as was the custom at the time, was often referred to in the
correspondence as Mme Albert Sechehaye, i.e., by her husband’s first and last name. This creates
confusion as to her identity relative to that of her husband.

based as it would be on a single student’s lecture notes only (albeit from
the first two years of the course). Bally opposes the Meillet/Regard initia-
tive even though it is obviously compatible with a more representative pub-
lication of the student lecture notes suggested by the students themselves,
and would have formed an important part thereof. To tip the scales in favor
of his own venture, Bally slips this piece of information to Meillet:
I have to add an important item, which is confidential and will, I hope, remain
between us: I have learnt from a reliable source, from a person who read Re-
gard’s notes, that these notes, however conscientious they may be, do not rep-
resent the spirit of Saussure’s teaching (l’esprit de l’enseignement de S.), and
that sometimes they completely distort it. I cannot ascertain the validity of this
piece of information, but it agrees with my impression of Regard’s working
methods, which are apt to seize the details but not the totality of a question.
All this gives me pause. I hope that you will see in this way of proceeding
only a desire to safeguard a memory respected by all; it may be better to
defer action so as to avoid disagreeable surprises. If you permit, I will keep
you informed about what I undertake in all this . . . (Amacker, 1989, 103)
In the May 29 letter, Bally has thus assumed the role as sole editor of
Saussure’s estate, with Meillet relegated to the role of an observer from
afar. Only Bally would be able to capture the very spirit of Saussure’s
teaching, without being slowed down by the detours of the students’ lec-
ture notes from the course in general linguistics. This textual document is
judged an external and ill-fitting body from which the spirit should be
liberated. Only Bally would be able to synthesize the question in its total-
ity, if not in the requisite detail. The professed objective of seizing Saus-
sure’s spirit foreshadows the method employed by Bally and Sechehaye in
the process of the ghostwriting of the Course where the editors will draw
on Saussure’s imputed authorial intentions in order to reconstruct the sci-
ence of general linguistics in a great synthesis (discussed below). Regard’s
own methods are dismissed as falling short of such divinatory and total-
izing ambitions. He gets discredited by a piece of anonymous slander, to
be secretly guarded between the gentlemen.
On May 31, Meillet writes back to issue consent:
As I wrote to Mme de Saussure, the project I envisaged with the young

Regard is abandoned; this project was always contingent on your agree-
ment, and is out of the question if you have a different view on the matter.
(Amacker, 1989, p. 104)

Meillet adds:
It is hard for me to appreciate the project you describe. I have great reserva-
tions about posthumous publications, and that is why I suggested to Regard
the project you are aware of. I would probably be even wearier of combin-
ing the different courses. But you are better placed than I am to appreciate
things, and you have access to sources I don’t. In any case, I will be grateful
if you kept me informed about what you have decided, once you have for-
mulated a definitive project. (ibid.)
Meillet will expand on his misgivings about this editorial venture in

two book reviews of The Course (discussed below).
Bally gets elected chair in general linguistics at the University of
Geneva, previously occupied by Saussure, on June 20, 1913 (Sechehaye
would replace him as chair in 1939). On June 22, he begins “an examina-
tion of my master’s notes” (un dépouillement des notes de mon maitre)
(Redard, 1982, p. 14), where dépouillement can mean both close study and
skinning, stripping, or shedding, as in skinning an animal, stripping the
bark of a tree or a person of their clothes—or stripping the notes to reveal
the master’s spirit? In a letter to Niederman, Bally concedes Meillet’s dis-
agreement with the editorial venture, but makes little of it, since Meillet
“only saw incomplete notes by one student who did not follow the last
course, the most important [one].” And noting the risks, adds, “It would be
criminal in my view to give nothing to the public” (ibid., p. 15). What the
public demands is a book?
So it seems. Having assumed the chair in general linguistics, Bally reit-
erates the call for a beau livre in his inaugural address delivered on Octo-
ber 27, 1913 (Bally, 1965, pp. 147–160). In a tone of lingering funeral
lament, but having now assumed the filial and scholarly right of a direct
successor to the post, Bally speaks with unquestioned authority about
Saussure’s contributions to general linguistics and provides their first offi-
cial compte rendu. He once again notes Saussure’s reticence to share his
thoughts about this emergent discipline, despite the fact that, if written
down, even approximately (d’une façon grossière), they would have made
a beau livre (ibid., p. 148).
Bally adds that Saussure would possibly have left without telling his
secret, had it not been for the circumstances that so to speak forced him to
deliver it (ibid., pp. 148–149).
The circumstances of this delivery, an imperfect substitute for the
unborn beau livre, would have been created by Saussure’s appointment to

the chair of general linguistics in December, 1906 and the three year
course on general linguistics. Bally regrets not having been able to attend
the delivery itself, and raises the hope that, thanks to the existence of de-
tailed student notes, the lectures will not remain “eternally unpublished.”
Or rather that the guiding principles and methods will be inferred from the
extant material. For:
The three courses differ in detail, each is original in its own way, but a to-
tality of great principles unites them all, and rises like a solid trunk of an
oak in the midst of the whimsical foliage ( frondaisons capricieuses) of its
branches. That’s what distinguishes F. de Saussure from other linguists of
his generation: his detailed researches became embedded in the natural set-
ting of his general views; that is why what can be made out (qui se devine)
is even more interesting than what appears (qui apparait). (ibid., p. 149)
The method for writing the beau livre in general linguistics must there-
fore be of a divinatory kind, a conjuring up of Saussure’s definitive doc-
trine, an imaginary clearing of the obscuring and meandering growth to
reveal the monumental column hidden in its midst. Since Saussure did not
write such a clear doctrine, a quasisupernatural insight will be required to
make out his hidden intentions behind the labyrinthine textual universe
left behind, and to erect a solid block of the new science above the web of
writing from the students’ notebooks. Bally does not hesitate to name the
guiding principles of the Saussurean doctrine already in October 1913:
there is a rigorous distinction between la langue and la parole, with only
la langue serving as “the true object of linguistics,” and la parole being
deemed “a secondary and contingent fact” (ibid., p. 151). This hierarchy
would be mirrored in a dualism of static and historical linguistics, with
only the former attaining language as a system (ibid., p. 152). Since the
public has no access to the manuscript sources themselves, this compte
rendu of Saussure’s general linguistics attains unrivalled monopoly and
becomes the source of the official Saussurean doctrine.
The 1916 editorial preface to the Course provides an additional but
partial perspective on the progress of writing the book. Bally and Seche-
haye note:
All these fortunate enough to attend these seminal lectures [i.e., the lectures
on general linguistics given by F. de Saussure at the U. of Geneva in 1907,
1908–1909, 1910–1911] were disappointed that no book resulted from them.
After the master’s death, we hoped to find something in his papers, which

Mme de Saussure kindly made available to us, a faithful or at least adequate
reflection of his brilliant courses (ces géniales leçons). We envisaged the
possibility of a publication based upon simple copyediting of Saussure’s
own notes, together with the lecture notes taken by students. Our disap-
pointment was enormous: we found nothing, or almost nothing which cor-
responded to his students’ notebooks. Saussure destroyed bit by bit the
rough copies in which he sketched out his lectures day by day. In his desk
drawers we found only old jottings which, although not without value, could
not be put to use and combined with the contents of the three lecture courses
. . . There was thus no alternative but to rely on the notes taken by the stu-
dents who had attended the three courses of lectures. (1986, p. xviii, trans-
lation corrected)
As noted in their editorial preface, Bally and Sechehaye initially con-

sidered collating the material from the three courses, drawing on the notes
of Saussure students Louis Caille, Lèopold Gautier, Paul Regard, and
Albert Riedlinger for the first two, and Marguerite Sechehaye-Burdet,
George Dégallier, and Francis Joseph for the third course. They enlisted
Albert Riedlinger in the collaboration on the first two courses, who at-
tended the lectures during the first two years and made detailed notes
(Sechehaye made a complete handwritten copy of these notes; they are
kept at the Bibliothèque Publique et Universitaire in Geneva). Riedlinger
did not attend the third course, and so could obviously not be consulted on
their content—a significant fact considering that Riedlinger is officially
listed as collaborator of the 1916 edition, and that the third course was
picked by the editors as direct basis for the Course. Sechehaye thus com-
piled the notes of Marguerite Sechehaye-Burdet, George Dégallier, and
Francis Joseph. Even though Sechehaye is said to have worked alone on
the third, one can reasonably expect that Marguerite Sechehaye-Burdet,
his wife and auditor of the course, played some role in the process; unlike
Riedlinger, she did attend the third year of the course.6
The editors discovered that the compared sources did not always fully
overlap; at times the editors overtly disagreed with the formulations they
found. The unpublished Collation, based on the material from the third
course and housed at the Bibliothèque Publique et Universitaire de Genève
6
Thanks to Godel’s Sources Manuscrites, we know that in addition to the student lectures notes,
Bally and Sechehaye had the following materials at their disposal: Saussure’s autographed notes
from around 1894 (published in CFS 12) dealing with general linguistics, and a sketch for a
planned tribute to the American linguist Whitney from the same year; somewhat brief
preparatory notes for a few out of the three courses on general linguistics.

(Sechehaye, unpublished MS, pp. 432–433) is instructive in this regard.
Bally and Sechehaye’s marginal notes on the collated material express
doubt and frustrations with the sources at hand (Engler, 1989, [321], p. 52;
[332], p. 54). The editors sometimes alter the material according to their
own interpretation of Saussure’s intent, even when the sources agreed
among themselves (ibid., [261], p. 43; [1122], p. 152).7 The editors’ working
methods are also different, with Bally but not Sechehaye advocating a col-
lation of the materials from the three courses; Sechehaye seemed initially
more hesitant to adopt a global perspective and would have preferred re-
constituting Saussure’s course page by page to a great synthesis; he also
admitted the difficulty of reconstructing what Saussure said considering
the limited source they were working with (Redard, 1982, pp. 16, 18–19).
It is also Sechehaye who concedes publicly that the systematization as well
as the reorganization of the contents in the Course is of the editors’ own
making:
Perhaps it is necessary to recall that the order of contents in the Course in

General Linguistics is not Saussure’s own. The three courses he offered
were situated on three different planes. The book’s editors were obliged to
adopt a more or less systematic order which seemed appropriate to them.
(Sechehaye, 1927, p. 234n, my translation)
In the published preface to the Course, the editors do not discuss the
details of the collating process; there is no mention of the alterations made
to the manuscript sources, of the ironing out of differences for the sake of
a systematicity and organization they deemed appropriate. The editors do
note the difficulties of reconciling the constantly renewed and evolving
flow of oral teaching with the format of a scientific book. They declare that
the heteroclite character of the extant material made publishing it in its
original form “impossible.” They ultimately decide to reconstruct Saus-
sure’s own, presumed, intentions, in a classical narrative form, and to base
the reconstruction upon but a third of the existing textual material:
We eventually hit upon a more daring, but also, in our view, more rational
solution. We would attempt a reconstruction, a synthesis. It would be based
upon the third course of the lectures, but make use of all the material we
had, including Saussure’s own notes. The task then was of re-creation, all
7
For some additional examples of editorial distortions of manuscript sources, see Harris, 2003,
p. 35.

the more difficult (malaisé) because it needed to be completely objective. It
was necessary, in every case, to go to the bottom of each thought, and try to
see it in its definitive form, freed from the variations and hesitations (flotte-
ments) typical of an oral lecture (leçon parlée); then, to put it in its proper
place (milieu naturel), with all the various parts presented in an order con-
forming to the author’s intention, even when this intention could only be
inferred (se devinait) rather than perceived (qu’elle n’apparaissait). (Saus-
sure, 1986, pp. [9], xviii–xix, translation revised)
The professed editorial solution aims at complete objectivity and ration-

ality by removing discrepant and deviant data from an expected, coherent
and invariant model. The diachronic stratification of the lecture course
gets masked by a unitary or synchronic architecture of parts and chapters
whose order does not reflect the development of the lectures themselves;
the resulting reorganization covers over the exploratory and evolving
nature of Saussure’s thinking/speaking in actu, which has been likened to
another exemplar of “impossible” writing, that by later Wittgenstein (de
Mauro, 2005, p. 408). It creates an impression of a completed system,
which lays out firm foundations for a new science of linguistics but effec-
tively silences the critical and philosophical voices heard in the sources.
On Bally and Sechehaye’s own admission, the “reconstruction” and
“synthesis” of the lecture notes is fragmentary and incomplete; and yet it
would, curiously, conform to the author’s own intention (“even when this
intention could only be inferred (se devinait) rather than perceived (qu’elle
n’apparaissait)”), with each thought planted in its “proper place.” In a
single paragraph, Bally and Sechehaye thus express a humble acknowl-
edgement of the limitations of their editorial venture and transubstantiate
the final product into official doctrine whose contents and order of exposi-
tion are divined directly from the authorial intention to which they hold the
successors’ rights as self-appointed heirs apparent to Saussure’s estate.
The editors disguise the disquietingly spectral character of such a process
with a guise of scholarly seriousness and complete objectivity. After all, as
the readers of Shakespeare’s Hamlet know, the scholar type Marcellus is
ill fitted to speak to the ghost directly; it is the link of filiation that binds
Hamlet to his father’s ghost and opens up a direct channel between the pa-
ternal specter and the son. The editors of the Course occupy both the filial
and the scholarly positions; they are visited by a specter like Hamlet and
they follow the precepts of rationality and objectivity like Marcellus. As
such they correct the presumed fault with Saussure’s teaching that Bally
diagnosed already in his “Maitres et Disciples” speech as too quick witted;

literally, Saussure’s spirit would have been led astray (les écarts trop vifs
de l’esprit) in the course of oral delivery (Bally, 1908, pp. 3–4)—and so,
presumably, needed to be returned to the path straight and narrow.
The editorial process was completed in 1915. Bally writes to Nieder-
mann on May 21 that final touches are added to the Course and on June 24
that the editorial process is complete (Redard, 1982, p. 17). In October, the
book is projected for publication with Payot in the beginning of 1916, and
in March the publication process is in its last phase. The book comes out
in May (“later than we would have hoped,” Bally writes [ibid.]). The whole
process of collating and redacting the material and publishing the book
took about three years—which testifies to a certain rush on the editors’
part, with Bally having announced the guiding principles of Saussure’s
general linguistics already in his inaugural lesson from October 1913
(Godel, 1957, p. 96).
Book reviewing
Bally and Sechehaye were both actively involved in the making of the
Saussurean doctrine, with its host of basic principles: the hierarchical di-
chotomy between la langue and la parole, synchrony and diachrony. Their
involvement extends, however, beyond the public presentations and the
writing of the basic principles of this doctrine into the Course to include a
subsequent reception of the same principles from the Course in a series of
dedicated book reviews. This scholarly reception helps to cement the con-
tents of the Course as official doctrine, and to seemingly distance the edi-
tors, notably Sechehaye, from the writing process by suggesting a stance of
an independent jury responding to a completed work.8
8
The first of such “internal reviews” by scholars associated with the making of the Course is by
Jules Ronjat (1864–1925), a French linguist consulted by the two editors. Ronjat wrote a
hagiographic review for Journal de Genève (June 26, 1916), a reputable journal for the general
public, in which he praised the book for being accessible to “any person capable of [paying]
attention and equipped with general culture”; the Course, he wrote, does not unnecessarily daunt
the reader with technical terms (and defines the terms it does employ), and is to be recommended
as summer reading for the upcoming holidays, which many inhabitants of Geneva like to spend
in the country. The editors are showcased as direct inheritors of the master’s unfinished work,
who, guided by his genius, execute their due succession rights in completing his project:
“At the end of a life too short to realize all the projects of his brilliant intelligence [Ferdinand
de Saussure] left behind a sketch of an oeuvre which definitively founds the general science of
human language as well as disciples capable of bring this oeuvre to completion—a scientific
testament of rare brilliance, which we salute as a dawn of a new era of progress initiated by a
genius (génie) which will guide (illuminera) us for a long time, after his mortal envelope has
disappeared” (Ronjat, 1916, p. 1).

Sechehaye writes a series of internal reviews which help establish the
Course as the official Saussurean Credo in scholarship. The earliest review
(“Les Problèmes de la langue a la lumière d’une théorie nouvelle,” Revue
Philosophique, 1917) amounts in fact to an extensive thirty page article
which maps out the field of general linguistics in historical and systematic
detail. Within it, Sechehaye casts the Course as an expression of the “Sau-
ssurean doctrine” (p. 11), a system derived from a finite set of axiomatic
principles:
The doctrine of Ferdinand de Saussure can be reduced to a certain number

of principles or theses, since it contains a series of affirmations which can
be deduced successively from a common principle. (Sechehaye, 1917, p. 1,
my translations)
The first principle of this doctrine is “the distinction between la langue

and la parole within the totality of loosely defined phenomena called lan-
gage” (ibid., p. 11). Sechehaye emphasizes the autonomy of la langue with
regard to la parole: “La langue, separated from la parole, the action of
speaking, and considered in itself, is the object of a very particular science
and demands to be exactly known” (ibid., p. 12).
In addition to this usual statement of priority of la langue over la parole,
we find one related to synchrony and diachrony:
There are two types of linguistic problems: the synchronic problem, which
concerns the language states, organized systems, and the diachronic prob-
lem, which concerns the transformations affecting the many parts of which
linguistic drama is made up of. The scholar is to never confuse these two
orders of questioning, which are by nature separate and irreducible one to
the other. (ibid., p. 23)
Sechehaye’s 1917 review of the Course thus reiterates what Bally pro-
posed already in 1913 as the official compte rendue of the Saussurean doc-
trine, (la langue serves as “the true object of linguistics,” while la parole
is “a secondary and contingent fact”) there is a split between static and
historical linguistics, with only the former attaining language as a system
(Bally, 1965, pp. 151–152). Repeated like a mantra that the disciples attrib-
ute to/acquire from the master, the doctrine is elevated to the realm of in-
controvertible eternal ideas, the quintessential truths of general linguistics.
(It should be added that the 1917 review adds a level of complexity to the
“doctrine” by emphasizing, unlike other reviews, the importance of value

relations between signs in linguistics and the differential character of lin-
guistic values; both principles defy the notion that language is a simple
nomenclature made up of isolated labels that mechanically associate with
corresponding ideas.)
In the 1927 article L’école genevoise de linguistique générale, Seche-
haye draws on the “Saussurean doctrine” (the “essential doctrines which
maintain the entire system,” that is the distinction between la langue and
la parole, and the need to isolate la langue as the object proper in linguis-
tics (1927, ibid., pp. 221–222]) to make a case for the so-called Geneva
School of Linguistics. He also defends the Course as a site of undiluted
Saussureanism. Sechehaye writes:
by a long process of selection and critique from the few documents at their
disposal, they [Bally and Sechehaye, aided by Riedlinger] succeeded in de-
riving a systematically organized posthumous oeuvre, within which French
linguists found the authentic thought, manner and sometimes the style of
Saussure. (ibid., p. 219, my translation)
However, some “French linguists,” notably Meillet, disagreed with this

assessment (see discussion below).
In 1940 Sechehaye undertakes yet another extensive review of the con-
tents of the Course, totaling almost fifty pages (Sechehaye, 1940, pp. 1–48).
The overt purpose of the review is to respond to the critiques of the Course.9
Sechehaye develops a threefold distinction between synchronic linguistics
(la linguistique synchronique ou des états de langue), diachronic linguis-
tics (la linguistique diachronique ou des évolutions de la langue), and the
linguistics of organized speech (la linguistique de la parole organisée ou
du fonctionnement de la langue). Having framed the relation between syn-
chrony and diachrony as one applicable solely to the domain of la langue,
Sechehaye proposes that the linguistics of la parole serves as an interme-
diary between the first two. In doing so, Sechehaye claims to have cor-
rected Saussure’s mistake of confusing diachrony with la parole in the
Course (“[Saussure] usurped things which logically (en bonne logique)
belong to the science of organized parole within diachrony”), and of thus
9
The book was initially received and reviewed primarily in linguistics—significant linguists like
Jespersen, Meillet, Bloomfield, and Schuhardt all wrote reviews of the Course (see Koerner,
1972, pp. 69–76, for a detailed list). The reviews tended to be muted and some, like Jespersen’s,
were overtly critical, especially of the dichotomies. Few reviewers noted the difficulties in
interpreting the contents in the Course in light of its editorial history, e.g., Benveniste (1966)
noted that the Course is a posthumous book, redacted based on student lecture notes.

failing to disambiguate between accidental occurrences of forms of speech
on the one hand and the possibility that such accidents find their way into
institutionalized linguistic usage on the other (ibid., p. 23). Yet Sechehaye’s
bonne logique seems to follow from his own professed taste for great ab-
stractions, according to which la langue is opposed to la parole as neces-
sary is to the contingent, and abstract is to the concrete (see “A taste for
great abstractions” below for development).
A propensity for abstraction is also apparent in Sechehaye’s account of
the Course as an exemplar of the “Saussurean method,” that is:
a method of analysis which consists in putting at the center of linguistics the

problem of la langue . . . in all its logical abstraction, and in subordinating lin-
guistic thinking in its entirety to the exigencies of this abstraction. (ibid., p. 2)
This method leads directly to the “famous distinction” between la

langue and la parole (ibid., p. 1), as well as between synchrony and diach-
rony (ibid., p. 7).
Sechehaye concedes that some passages from the Course suggest a mu-
tually determining relation between la langue and la parole which would
complicate the Saussurean doctrine as just laid out (ibid., p. 8). Still, he
deems this notion an in principle unsatisfactory “simple reciprocity” (ibid.,
p. 8), and notes that “the master was led into error by his usual mental
tendencies (tendances familières a son esprit)” (ibid., p. 8). The tendency
in question is the one that Bally-the-editor wrote into the Course, but which
Sechehaye-the-reviewer now critically receives: “The dominant and cen-
tral place he attributed in principle to la langue in his doctrine prevented
him from attributing a subordinate position to it” (ibid., p. 8).
Committed to the la langue/la parole hierarchy as a matter of principle,
Saussure could only resort to formulas such as “simple coordinations” and
“reciprocal services” to frame the relations between la langue and la
parole. These two negatively colored terms (the coordinations are simple,
reciprocity a mere service) are of Sechehaye’s own making, yet he derives
them from Saussure’s own presumed “taste for paradoxical formulas,”
such as the thesis that all linguistic phenomena are essentially double
(ibid., p. 8). Saussure’s general linguistics is thus presented as a species of
dogmatism wedded to a set of basic dichotomies; any exception to this as-
sumed “Saussurean method” is framed as a futile attempt to join what has
already been put asunder.
This last review projects a deliberately dogmatic reading of the Course
as “Saussurean orthodoxy” (Sechehaye, 1939, p. 19; 1940, p. 30) to be

approached with religious respect by any follower/believer in Saussurean
linguistics ( fidèle de la linguistique saussurienne) (Sechehaye, 1940,
p. 28), faithful to the spirit of our master (ibid., p. 48). This theolinguistic
narrative issues around the time when Sechehaye is at the height of reli-
giosity, consumed by what has been referred to as optimist spiritualism
(Fryba-Reber, 1994, p. 192; Amacker, 2000, 13n). The specter of Saus-
sure whom Sechehaye continues to channel is but a natural complement
to this scenario.
This spectral Saussure will be kept alive for decades, even after the
critical work of Godel had begun the long process of reassessing the offi-
cial doctrine. Fifty years after Saussure’s death, Benveniste left a detailed
description of the source materials to others, and opts to simply reconsti-
tute “its basic principles in the drive which animated and even formed
[that work]” (Benveniste, 1971 [1966], p. 29), retaining “the doctrine
which the disciples of Saussure put into shape and published” (ibid., p. 37).
It is only on the assumption of such a doctrine that Saussure can be said to
have accomplished his destiny: “Beyond his earthly life, his ideas spread
further than he could have imagined, and his posthumous destiny has
become, so to speak, a second life which henceforth mingles with ours”
(ibid., p. 40).
*
In Les trois linguistiques, Sechehaye partially acknowledges the complex-
ity of the editorial situation surrounding the Course. He notes that Saus-
sure would never have consented to publish the student lecture notes from
the course in general linguistics. Yet any “posthumous violence” that may
have been involved in failing to heed “the venerated master’s reservations”
would have been assuaged by the evident success of the product itself
(Sechehaye, 1940, pp. 2–3). That there is another kind of posthumous vio-
lence performed by the publishers of the Course—the violence inflicted on
the texts recording Saussure’s lectures by the students and onto Saussure’s
own texts, the violence inflicted onto the students’ own reputation and
ability to bear witness to Saussure’s teaching—remains unsaid.
Saussure’s own students, Regard and Riedlinger (and Meillet), on the
other hand, voice a critical perspective to the editorial undertaking. In the
preface to his dissertation (1918), Regard notes that Saussure had scruples
to “confine his infinite genius to the finite book form” (le génie infinie
redoutait la forme fini d’un livre) (Regard, 1918, p. 3). Even though he is
largely appreciative of the efforts to publish the Course, Regard bemoans
some of the distortions of the sources in the book. He notes especially an

artificial separation between linguistic change and its larger context (“ex-
ternal conditions of language”), which deprives the former of reality and
turns it into an inexplicable abstraction (ibid., p. 11). This implies that for a
student who attended Saussure’s lectures, a linguistic process can only be
studied “in itself and for its own sake” as a result of problematic dissocia-
tion between the phenomenon of language and the concrete and evolving
practice of usage. Furthermore, change goes hand in hand with conserva-
tion of linguistic traits—this connection was explicitly made in the lec-
tures (ibid., p. 11). Regard notes also:
A student who attended a large part of Ferdinand de Saussure’s lectures,

and was familiar with a number of the documents the publication [of the
Course] draws on, is going to be inevitably disappointed to discover that the
exquisite and captivating charm of the lectures is gone. Would publication
of the [student] lecture notes not have more faithfully preserved Ferdinand
de Saussure’s thinking, with its force and originality, despite being repeti-
tive on some points? And would not the very variations which the editors
apparently fear to make shown be of particular interest? (ibid., p. 6, my
translation)10
These excellent questions remained largely without a response until the

1950s.
Riedlinger, the official collaborator of Bally and Sechehaye in the edi-
torial process, gives no immediate public expression of his own views. It is
only in 1957, a little over a decade after Bally’s and Sechehaye’s deaths,
that he writes to a fellow student to explain his refusal to write an article
requested by the latter for the Tribune de Genève:
It would be impossible for me to represent the true greatness of F. de S.

without comparing him to Bally . . . Bally butchered general linguistics, as
Godel’s work will demonstrate beyond any discussion . . . What is even
worse is the scrapping of the wonderful 100 pages long introduction to the
second course . . . Godel sees the quintessential Saussurean thinking in this
10
Un élève qui a entendu lui-même une part importante des leçons de Ferdinand de Saussure sur
la linguistique générale, et connu plusieurs des documents sur lesquels repose la publication,
éprouve nécessairement une désillusion à ne plus retrouver le charme exquis et prenant des leçons
du maître. Au prix de quelques redites, la publication des notes de cours n’aurait-elle pas
conservé plus fidèlement la pensée de Ferdinand de Saussure, avec sa puissance et son
originalité? Et les variations elles-mêmes que les éditeurs paraissent avoir craint de mettre au
jour n’auraient-elles pas offert un intérêt singulier?

introduction. But Bally, though very gifted as far as observing linguistic
facts is concerned, did not have his master’s philosophical sense. (cited in
Bouquet, 1999)11
One can therefore wonder about the extent of the loss of philosophical
complexity of Saussure’s work in its official published version—a question
that merits analysis in its own right, but becomes especially pertinent in
light of the tremendous impact of the Course on the development of philo-
sophical movements in the twentieth century, in Continental Europe, the
United States and beyond. If the dominant structuralist reception of the
ideas from the Course rests on a presentation lacking in Saussure’s “philo-
sophical sense,” then it is of interest to the philosophers themselves to re-
cover some of the philosophical complexity from the source materials, and
resist the already instituted habit of receiving the vulgate edition as the
Saussurean doctrine.12 The reader can also wonder whether the professed
collaboration with Riedlinger in the editorial process could have been used
to add a veneer of credibility to Bally and Sechehaye’s attempts of recon-
struction and synthesis, as compared to Regard and Meillet’s alternative
attempt to publish the notes in the original form. Riedlinger’s participation
could have been used as the official stamp of the insider’s approval of the
nonparticipants’ reconstructive attempts; the collaborator and signatory
apparently feared to cross the party line and to express overt disagreement
with the Course up to the late 1950s, around the time when the second
generation of linguists at Geneva began to question the validity of Bally
and Sechehaye’s construct. This protracted silence may have been neces-
sary for the official Saussurean doctrine to resound with unrivalled mo-
nopoly in the field.
11
Il me serait impossible de donner une idée de la vraie grandeur de F. de S. sans le comparer à
Bally . . . Bally a sabré la linguistique générale, ce que le travail en cours de Godel démontrera
sans discussion possible . . . Plus grave encore est la suppression complète de la magnifique
introduction de 100 pages du deuxième cours . . . Godel voit dans cette introduction la
quintessence de la pensée saussurienne. Mais Bally, très doué par ailleurs pour l’observation des
faits linguistiques, n’avait pas le sens philosophique de son maître.
12
Such a task is less satisfying or even less manageable than the reading of a portable paperback,
but it is instructive; for example, the reader may begin to ponder how her practical preference for
one packaging of ideas over another shapes their circulation and reception (e.g., Engler’s critical
edition of the Course is not a portable volume, because of its size and weight, and it is rare; the
published student lecture notes come in a three-volume set, which it was impossible to find for
purchase at the time of this writing); the reader may therefore appreciate the material historicity
of the ideas themselves, and realize, if need be, that the “inner” intellectual content is tied to its
“external” envelope like the soul to the body.

As for the French linguist Meillet, his editorial and interpretative con-
cerns relative to a posthumous mixing and matching of source materials
are forcefully articulated in his own book reviews of the Course. Meillet
highlights the impudent gesture (le parti hardi) involved in constructing a
book “which the master did not make, which he would have without a
doubt never made” (Meillet, 1916, p. 32). He reiterates the concern the fol-
lowing year, noting that:
there is a lot of impudence (hardiesse) in a project consisting of publishing,

under F. de Saussure’s name, a course which not only he did not supervise
the redaction of, but whose very arrangement does not correspond to any of
the three lecture series on general linguistics he taught. Bally and Seche-
haye and the family of F. de Saussure judged it essential to offer to the
public a doctrine developed by the master and passed down to his disciples.
(Meillet, 1917, p. 50)
The familial scene implied in the posthumous publishing of the Course,

which features the master conceiving the official doctrine, and the trail-
ing disciples who inherit it, is here acknowledged as a conceit of Bally
and Sechehaye’s (and the bereaved family of F. de Saussure’s) own
making—a conceit that effectively enabled the publication of the book in
Saussure’s name.
Meillet further observes that the many sets of student lecture notes
from the three-year-long course in general linguistics differed among
themselves—and that it is regrettable to let these differences perish within
a single, systematic volume (Meillet, 1916, p. 32). As a former student of
Saussure at l’Ecoles des Hautes-Etudes in Paris, Meillet regrets also the
loss of poetic force and imagery in the Course, as compared to Saussure’s
oral discourse (ibid, p. 33). In Meillet’s own account of Saussure’s teach-
ing, “he made [his students] love and feel the science he taught; his think-
ing was that of a poet and often gave an imaged form to his exposé, a form
one could never forget” (Meillet, 1913, p. clxx). Editorial expectations of a
beau livre apparently excluded such poetic excesses and indelible imagery
as overly subjective and out of place in serious scientific scholarship. Meil-
let notes that “one finds a bit of all that in the book, but, the brilliance
(l’eclat) of the images is weakened, and their number is limited” (Meillet,
1916, p. 33).
Throughout the more substantial 1916 review, Meillet acknowledges the
difficulty of reviewing a book “where one does not know if the elements
subject to a critique come from the author or the editors” (ibid., p. 36).

And yet such a critique is advanced in the review nonetheless: an emphasis
on la langue at the expense of la parole raises the difficult problem of how
the observations of la parole can lead to claims being made about la langue
(ibid., p. 35); this problem can only become aggravated by the advances in
the field of phonetics (ibid.). Furthermore, the separation between linguis-
tic change and its “external” conditions reduces the former to an abstrac-
tion devoid of reality and ultimately inexplicable (ibid.). The very notion of
a closed and autonomous language system is therefore rendered suspect—
Meillet deems the problems associated with it to be “fanciful” (chiméri-
que) (ibid.). Finally, Meillet notes that the notion of diachrony implies an
emphasis on concrete historical facts, as well as the process of transmis-
sion of language actually being spoken—which cannot be accomplished
without also taking into account the diversity of social conditions, and the
entire social structure (ibid., p. 36). In sum, the official Saussurean doc-
trine with its usual host of dichotomies: la langue and la parole, synchrony
and diachrony, is deemed inadmissible in the book reviews penned by Mei-
llet. Meillet objects to the production process that led to the posthumous
publishing of the Course in Saussure’s name as much as the contents of the
final product itself. Contrary to Sechehaye’s assessment that the “French
linguists found the authentic thought, manner and sometimes the style of
Saussure” in the Course (Sechehaye, 1927, p. 219; cited above), Meillet ap-
parently found a number of discrepancies and inadmissible abstractions.
The famous formula
An important element of the Saussurean doctrine consists in the primacy

of la langue over the domain of la parole (as well as over the plurality of
particular languages [les langues], discussed in the next section). In this
section, I will discuss one strategy that establishes the primacy of la
langue: the editorial insertion of the concluding lines according to which
language (la langue) is an in-itself objectivity akin to a closed and autono-
mous system of signs. This insertion was subsequently cited as Saussure’s
own word by the editors in their writings in linguistics and book reviews
of the Course. Having gained the status of a “famous formula,” the con-
cluding lines of the Course functioned subsequently as the structuralist
motto—and were also cited as evidence of incompatibility between struc-
turalism and phenomenology.
The contents of the Course as a whole are framed by an opening ques-
tion and a concluding reply. The question is hinted at in the opening lines

of the introduction where the idea of “the one and only true object of
study” in linguistics is introduced (Saussure, 1986, p. [13], 1); it is stated
explicitly at the beginning of ch. III of the introduction: “What is the inte-
gral and concrete object of linguistics?” (ibid., p. [23], 8, my translation).
The question receives a definitive response in the last lines of the
Course:
the only true object of study in linguistics is the language (la langue) con-
sidered in itself and for itself (en elle-même et pour elle-même). (ibid.,
p. [317], 230, my translation, italics in the text)
This response is said to support the fundamental thesis (l’idée fonda-

mentale) of the course as a whole (ibid.). Everything sandwiched in be-
tween the question and the response is thus made to fit into a program-
matic search for an objective datum for the emergent science of general
linguistics to identify as being truly its own.
The closing line of the Course (“the only true object of study in lin-
guistics is the language (la langue) considered in itself and for itself”)
has acquired the status of quintessential Saussuranism in the structuralist
and phenomenological circles alike, and it helped to establish the impres-
sion of a seamless transition from Saussureanism to structuralism. This
formulation was, however, freely inserted by the editors; there is not a
single textual locus in the manuscripts to justify its addition (Godel, 1957,
pp. 119, 181). A source for this apocryphal statement may be found in the
editors’ own reconstruction of Bopp’s views (Bouquet, 2003, p. 12)—but
the latter were overtly critiqued by Saussure for their bias in favor of static
language states (see ch. 3). The insertion of the last sentence thus illus-
trates the editors’ role as the ghostwriters of the Course, dictating their
own commitment to the holistic priority of the language system over
speech into what later became the Magna Carta of modern linguistics.
Unsurprisingly, this sentence became programmatic in the development
of the structuralist doctrine. As de Mauro notes, “according to structural-
ist linguistics, to respect Saussure means to ignore the disequilibrium of
the system, the synchronic dynamics, the social conditions, mutability,
the link between the latter and the various historical contingencies—the
whole flotsam of linguistic phenomena thanks to which language is form”
(de Mauro, 2005, p. 476n). He concludes: “The addition of the last phrase
is a seal of the editorial manipulation of Saussure’s notes, and bears in
part the responsibility for the exclusivist attitude of structuralism” (ibid.).
The scholars are now in agreement that if Saussure became lauded by

posterity as a pioneer of structuralism, it is in part due to such editorial
manipulation.13
The phrase may also bear in part the responsibility for the exclusivist
attitude of phenomenology—notably Paul Ricoeur cites the phrase as a
principal evidence that classical phenomenology needs to distance itself
from the perceived semiological challenge raised by structuralism. As he
put it:
We know how far Saussure went along the path thus opened up towards a lin-
guistics of language, which is synchronic before being diachronic: language is
henceforth a system of signs defined solely by their differences . . . As the last
sentence of the Course in General Linguistics states: “The only real object of
linguistics is language considered by and for itself.” (Ricoeur, 1974, p. 16)
For Ricoeur, the last sentence of the Course expresses in a nutshell the
possibility of signification being deployed by a closed and autonomous
system, solely in virtue of the contrastive relations between signs. This
stated primacy of structure over process would make phenomenological
descriptions of speech, and their references to a signifying subject, null
and void. This destructive possibility “presents such a radical challenge to
phenomenology that it may justly be said that phenomenology will not sur-
vive unless it can properly reply to this challenge” (ibid., p. 14). Ricoeur’s
vision of the ensuing relation between phenomenological and structural
approaches to language is therefore not unlike that of a wrestling match:
“Only through a hand-to-hand combat with the presuppositions of semiol-
ogy will phenomenology reconquer the sign’s relation of transcendence or
its reference” (ibid., p. 251).
Both phenomenology (as understood by Ricoeur) and structuralism will
inevitably adopt exclusivist attitudes:
The privilege . . . conferred on consciousness in an idealist conception of

reduction is radically incompatible with the primacy that structural linguis-
tics accords to language over speech, system over process, structure over
function. In the eyes of structuralism, this absolute privilege is the absolute
prejudice of phenomenology. With this antinomy the crisis of the philoso-
phy of the subject reaches its extreme point. (ibid., p. 257)
13
Curiously, even Meillet cites this editorial insertion as the fundamental idea of the course
(Meillet, 1916, p. 33). Hence even some of the scholars who voiced misgivings about the value of
the Course received this conclusion as official Saussurean doctrine.

Ricoeur’s response does not locate any middle ground between the
system and subject based approaches to language and signification, but
only deepens a perceived antinomy by refiguring speech, diachrony, and
subjectivity in terms of pure transcendental consciousness. This only leads
to a protective enclosure within the subjective field, with no bridge to the
(purportedly) closed and autonomous system of signs, la langue. The free
editorial insertion written into the Course by Bally and Sechehaye thus led
also to an antagonistic attitude in phenomenology, which, in Ricoeur’s
hands, reclaimed its transcendental heritage and refocused signification on
the signifying intentions deployed by the speaking subject.
*
Bally and Sechehaye played a double role in the process of establishing
the familiar set of dichotomies: la langue and la parole, synchrony and
diachrony, as official doctrine: they publicly presented and ghostwrote
these basic principles into the text of the Course, and subsequently re-
ceived the same principles from the Course as Saussure’s own word. This
process can be documented in relation to the famous formula: the editors
performed a double duty in that they inserted textual material unwar-
ranted by the sources into the Course, and subsequently cited their inser-
tion as if it came from Saussure’s own hand. This familiar process of writ-
ing into- and -from the Course helps to cement the famous formula as
official doctrine and to seemingly distance the editors from the making
process by suggesting a stance of an independent jury responding to a
completed work.
In Linguistique générale et linguistique francaise, Bally claims to be
effectively following Saussure’s own method:
which F. de Saussure summarized in the last phrase of his Course in Gen-

eral Linguistics: “the only true object of study in linguistics is the language
(la langue) considered in itself and for itself.” (Bally, 1965a, p. 17)
Following the Saussurean method will consist, according to Bally, in

teasing out the general traits and internal tendencies of the French lan-
guage, and considering it strictly from a linguistic point of view (ibid.).
As for Sechehaye, he notes that:
the fate of la langue is completely removed from the speaking subjects’ psy-
chology. This thesis is, as one knows, developed in the final pages of the
Course and this doctrine is summarized in the famous formula found in the

conclusion of this work: the only true object of study in linguistics is the lan-
guage (la langue) considered in itself and for itself. (Sechehaye, 1940, p. 26)
The editors thus combined forces to write their creations into the Course,
and to obscure their role by adopting a stance of a respectable pupil receiv-
ing the master’s teachings. Their self-assigned task of projecting the entire
system of Saussure’s thought according to the author’s inferred intention
onto the Course, “all the more difficult (malaisé) because it needed to be
completely objective” (Saussure, 1986, pp. [9], xviii–xix, cited above), is
pushed to the extreme of false attribution of fabricated sources in their later
citations from the Course.14 One step ahead of their apologetic statements
from the preface, and in the interest of complete objectivity perhaps, the
citations are no longer framed as a hypothetical result of editorial inference
from purported authorial intentions but as an effective result of the autho-
rial intention now simply assumed to be the generative source of the “famous
formula found in the conclusion of the [Course].” The editorial insertions
now perform double duty as inference and evidence; or rather they change
duty from the former, too subjective perhaps, to the latter, objective, one.
Saussure’s séance thus reaches a full circle: the ghost which has been
called has come and spoken, as if by his own accord and by force of his
own intention. The séance described in the editorial preface can thus
recede into oblivion, since the tables have been turned, the summoning
process no longer needed, the sorcerer simply echoing the master’s own
voice. In a twist on the story of the Sorcerer’s Apprentice, in which the
master returns in time to break the spell that has gotten out of the appren-
tice’s hand, and thus to save the day, the ghost summoned by Bally and
Sechehaye is the old master himself, his apparitions now fully in the ap-
prentices’ hand. The readers will therefore not find a cautionary reminder
in the concluding lines of their text that powerful spirits should only be
called by the master himself:
As a spirit
When he wills, your master only
Calls you, then ’tis time to hear it.15
14
The fallacy of a false attribution occurs when an advocate appeals to an irrelevant, unqualified,
unidentified, biased, or fabricated source in support of an argument. A more deceptive and
difficult to detect version of a false attribution is where a fraudulent advocate goes so far as to
fabricate a source in order to support a claim.
15
Denn als Geister
Ruft euch nur, zu seinem Zwecke,
Erst hervor der alte Meister.

Instead, they will find a structuralist motto dictated by a master too
ghostly to break the spell.
The famous formula meets the three criteria typically attributed to de-
lusional belief: certainty, incorrigibility, and falsity of content (Jaspers,
1997). As indicated by Jaspers (ibid., p. 95), the delusion refers to the ex-
ternal process only (ibid.), and no claim is being made about inner mental
states—such as an experience of delusion, or a change to personality. The
delusion is therefore not understood here as an individual belief on the edi-
tors’ part but as a process involving institutional mechanisms and relations
of power that enabled certain texts to define Saussurean linguistics. Bally’s
and Sechehaye’s above discussed texts were rendered effectively incorri-
gible by lack of access to the source materials, locked in the Geneva li-
brary, and by lack of any critical appraisal of their editorial work in light
of the unpublished sources, which was motivated in part by the editors’
unwillingness to publish these sources and to thus make their own publica-
tions open to scrutiny during their lifetime (the second wave of Saussure
scholarship began to emerge after the editors’ death). The editors’ unwill-
ingness to let the manuscripts see the light of day effectively contributed to
the mystification of Saussure’s linguistics, closing, for a time at least, any
space for a critical consciousness about the dominant constructions of ac-
ademic and popular knowledge about it.
Bally’s and Sechehaye’s statements were rendered apodictically certain
by virtue of an established view that the two men possessed the succes-
sors’ rights to Saussure’s estate and could offer direct testimony of Saus-
sure’s own ideas. This view was institutionalized when Bally became ap-
pointed to the chair of general linguistics at the University of Geneva
(previously occupied by Saussure) in 1913. It was also made possible by
the references to the so-called Geneva School of Linguistics—a term
which Charles Bally used at a gathering honoring Ferdinand de Saussure
with a presentation of Mélanges (a volume of contributions by his friends
and students [Sechehaye, 1927, p. 217]), with an emphasis on a presumed
bond between Saussure and his disciples. Saussure himself, however, con-
tested the exclusivist connotations of a “school,” noting that “to reform is
to not be a school” (quand on reforme, on n’est pas une école) (Engler,
1987, pp. 55–56).
Sechehaye makes references to the Geneva School to strategically situ-
ate Bally and himself as Saussure’s direct successors. He writes:
Ferdinand de Saussure, the initiator and the master laid down the general
principles. He impressed upon his students an intellectual discipline of the

first order. He bequeathed to them a solid doctrine upon which they could
build. Bally devoted himself to the reality of living language, and devel-
oped a method destined to sustain both the science and the teaching of
language, while Sechehaye, a theoretician by principle, devoted himself to
revising grammatical notions and methods. (Sechehaye, 1927, p. 240)
This familioscholarly scene stages the relation between Saussure on the

one hand and Bally and Sechehaye on the other as being a master-initiator
who first lays a foundational stone upon which his disciples erect an edi-
fice; the master establishes basic principles of a doctrine which the disci-
ples develop. The scene thus fits seamlessly into the scholasticism of the
European academic institutions, wherein an intellectual movement gains in
legitimacy if it is recognized as an officially established school of thought.
The existence of the Course, a Great Book written by a past master, is an
integral element of this scholastic arrangement: it functions as the meta-
phorical and material bedrock for the successors to build on. The envisaged
kinship relation within the school building process is that of a special bond
linking the patriarch to his legal heirs. I will examine this kinship relation
in more detail next.
Elementary structures of kinship in academia
The apocryphal last line of the Course enjoyed the status of an oft-cited
motto for decades. It is venerably cited in the work of the Danish structur-
alist linguist’s Louis Hjelmslev, a cofounder of the Copenhagen Circle of
linguistics (founded in 1939, together with Viggo Brondal). Hjelmslev’s
approach to language was extreme in its reductionism: he developed a dis-
cipline of glossematics (from Greek “language”), a purely structural kind
of linguistic research focused on language as a form without substance. Its
ultimate objective is to reduce the phonetic as well as the semantic and
referential dimension of language in favor of an idealized algebraic nota-
tion. Such an approach effectively abandons the territory of language study
in any concrete sense for the sake of pure logical and mathematical models,
and was critically received by Roman Jakobson, the cofounder of the Lin-
guistic Circle of Prague in 1926, one of the most prominent and prolific
representatives of the structuralist movement in linguistics broadly con-
strued (Dosse, 1997, p. 56).
Following Hjelmslev, the final sentence from the Course both captured
the fundamental idea of Saussure’s lectures and provided direct impetus to

his own formalist approach to language (Hjelmsev, 1972, p. 101). As he
noted, “my endeavor is on the side of langue studied and conceived as a
mere form, as a pattern independently of the usage” (ibid., 103); a formalist
reductionism would therefore follow directly from the concluding state-
ment which picks out la langue as its object of study. Glossematics would
then provide a direct implementation of the fundamental idea of general
linguistics. When Hjelmslev marginalizes extralinguistic reality in favor
of a presumed underlying structure of the internal linguistic order, he
claims to be developing a method of immanent linguistics already found
there. The linguist’s task is primarily this: “to simply draw all the conclu-
sions possible from the final sentence of the Course: ‘linguistics has as its
unique and true object language considered in itself and for its own sake’”
(Hjelmslev, 1972, p. 106, my translation).
Hjelmslev duly notes that his approach was sanctified by one of Saus-
sure’s disciples:
The late Professor Charles Bally, who was the successor of Saussure in the
chair of linguistics in the University of Geneva, wrote a letter to me some
few months before his death in which he said: “Vous poursuivez avec . . .
constance l’idéal formulé par F. de Saussure dans la phrase finale de son
Cours de linguistique générale” [You pursue . . . in a sustained manner the
ideal formulated by F. de Saussure in the last phrase of his Course in Gen-
eral Linguistics]. (Hjelmslev, 1972, p. 101)
Bally’s personal note to Hjelmslev included by the latter in an otherwise

dry scientific article strewn with algebraic notations thus serves to institute
a logistic program of language stripped of any reference to sensibility, sig-
nification and the social conventions of usage, as loyal Saussureanism. A
purist structure read off the final sentence of the Course would on Bally’s
informed judgment have already been written into it by Saussure himself.
The direct lineage of discipleship is thus extended by Bally to Hjelmslev,
it being understood that Bally is empowered to thus profess a novice into a
Saussurean community in virtue of his own privileged standing with the
master himself. The bonds of filiation add in Hjelmslev’s case as previ-
ously in Bally’s own case, to the scholarly legitimacy of their projects;
Hjelmslev’s immanentism ceases to appear as just one out of many possi-
ble appropriations of an elusive text—it becomes an instance of direct suc-
cession, of writing a postscript to the master’s book as if under his name,
picking up the project where he left it off. It would be as foolish to chal-
lenge such scholarly filiation as it would be to undermine the legality of the

son’s inheritance rights to the father’s property; patrilineal succession de-
termines the passage of ideal and material goods. Bally’s endorsement
helps Hjelmslev become officially recognized as the legal heir to Saus-
sure’s teaching in wider academic circles; another prominent structuralist
linguist, Algirdas Julien Greimas, presents Hjelmslev in a preface to his
Prolegomena to a Theory of Language as “The true and perhaps only suc-
cessor of Saussure who has been able to make his intentions explicit and
formulate them definitely” (Dosse, 1997, p. 68). Hjelmslev would have di-
vined and explicated Saussure’s own intentions as Bally had done before
him; according to Hjelmslev’s divination, Saussure’s implicit intention
would have been to commit a rare excess of formalist reductionism in lan-
guage study.
The structuralist appropriation of “Saussure’s own” ideas by Bally and
Sechehaye ran along the retroactively constituted lines of direct succession
from master to disciple like from father to son. It is therefore enabled (and
constrained) by what could be termed, to employ a term of another struc-
turalist scholar, Claude Levi-Strauss, the elementary structure of kinship
(parenté) within European academia. Let me interrogate this family scene
in what follows. The master-disciple relation sublimates the biological
bond of paternity into a cultural bond which thus preserves the prima facie
naturalness of such a relation, and produces the expectation of sameness
(or at least marked similarity and continuity) between the linked genera-
tions and their works. The master-disciple bond in the academic circles is
similar to the more broadly established continuation of patrilineal descent
through a passing down of the father’s proper name and property; this
process is legally enshrined in the rights of inheritance of the so-called
heir apparent—historically, the eldest son, whose right to inherit is inde-
feasible as long as he outlives the property holder (in distinction from the
heir presumptive whose right may be defeated by the birth of a nearer
heir)—except by exclusion under a valid will. Such publicly recognized
rights of succession need not in principle follow a biological bond between
the ancestor and the successor nor do they need to be confirmed by the an-
cestor’s will; as long as the relation seems to mimic patrilineal descent (as
in the case of a master-disciple relation in academia), it fits the logic of he-
redity, and may serve as a channel for the transmission of goods (both ma-
terial and intellectual property), and the establishment of a legal heir (such
as succession to an academic post, e.g., a chair in general linguistics). This
privilege of an heir apparent can in some cases be assumed retroactively,
after the master’s death (in distinction from the usual legal setup where the
heir is already assumed as such during the lifetime of the property holder,

and/or is named in the holder’s will). So long as the relation is publicly
recognized as being like that of a legitimate heir, it may assume the force
of direct and indefeasible succession comparable to a birthright. This proc-
ess helps to explain why the master-disciple relation retroactively con-
tracted by Bally and Sechehaye, and then extended to Hjelmslev, estab-
lished their role as Saussure’s direct heirs and legitimate disposers of his
ideas proper.
The master-disciple relation is a stable social structure in European ac-
ademia, both in the early twentieth century, and in contemporary Euro-
pean academic culture, where it is commonplace to situate scholars in re-
lation to their Doctorvater (and increasingly Doctormutter) regardless of
how far along they are in their academic career. This does not mean that
the master—disciple bond routinely predetermines the disciple to continue
the master’s work; a breakaway disciple is possible and not uncommon.
However, the expectation of discipleship is unshaken by such concrete in-
stances of pursuing a via negativa, and a negative response does not in the
public mind call into question the preexistent bond to the teacher. A rene-
gade disciple is still presented and perceived as such: someone who chose
to pursue a different path from the one indicated by his apprenticeship,
like a prodigal son who may one day be accepted back into the fold. The
preexistent bond is as much exposed by a negative as it is by a positive ap-
propriation; not all children choose to further the family tradition, but that
does not deny the existence of a family. Not all disciples follow in their
master’s footsteps, but that does not deny that a path for the disciple to
follow had been laid in advance. An estranged child, a wandering disciple,
are locked into the bond as if with an unbreakable force of nature.
The examples of adopted discipleship in Bally’s case, or the one of dis-
cipleship bestowed by another disciple in Hjelmslev’s case, both fit into
this logic of heredity; the perceived naturalness of these bonds and the
privileges of direct access and transmission of Saussure’s own ideas fur-
ther down the line can be deciphered within an already constituted and
accepted elementary structure of social relations in this context. Thanks to
its perceived quasinatural strength, the bond is untouched by the disap-
pearance of the mortal envelope; the disciples can freely communicate and
transmit Saussure’s ideas after his death.
This mimicry of the order of nature in the academic social structures
illuminates at least in part the great value attached to being publicly recog-
nized as a loyal heir; such recognition dispenses with the need to justify
one’s interpretation and appropriation of the master’s work; in fact, the
relation ceases to appear as a case of interpretation and seems to no longer

be mediated by texts. This effectively clears the field of any other interpre-
tations and appropriations as being unorthodox. Yet this privilege comes
with a price: if the elementary structure of academic kinship follows
mainly the line of (adopted) heredity, there is a risk of immanentism, self-
enclosure, and isolation. I draw on Simone de Beauvoir’s review of the
Elementary Structures of Kinship (1949) to bear this point out.
First, recall Levi-Strauss’s distinction between the facts of nature and
the facts of culture from his study of elementary structures of kinship:
while the former are universal, the latter are subject to norms, and as such
may vary (Levi-Strauss, 1969). The distinction between facts of nature and
facts of culture serves to situate the prohibition of incest at their borderline.
“Here . . . is a phenomenon which has the distinctive characteristics both
of nature and of its theoretical contradiction, culture. [It] has the universal-
ity of bent and instinct, and the coercive character of law and institution”
(ibid., p. 10). Placed against this backdrop, the prohibition of building
social alliances on the basis of being of the same blood appears as an am-
biguous or hybrid phenomenon: it is as universal as a natural fact, and yet
it incarnates a cultural norm, and belongs to the set of laws organizing
human society. The prohibition’s function is then not, as previously as-
sumed, merely negative and constraining; it opens up channels of reciproc-
ity between groups, similar to the ones involved in offering, receiving, and
reciprocating a gift. The prohibition subserves the positive interest of
building alliances beyond one’s immediate kin; as such it is a medium de-
fining the larger structure of the human society which cannot be straight-
forwardly derived from the (perceived) order of nature. Nature may facili-
tate these societal relations through the pregiven fact of paternity but it
does not predetermine any determinate social order, in Levi-Strauss’s own
case, a patriarchic order where reciprocal relations implicate men as sub-
jects while women serve as an object of exchange. Put differently, even
though paternity is a biological given like mating, the patrilineal inherit-
ance of the rights to the paternal name and property by the male offspring
is not—even if it were universal in scope, like a natural fact; it is contin-
gent on there being a set of societal values and norms already in place. A
patrilineal inheritance (of material goods and/or ideas), like Levi-Strauss’s
prohibition of incest, is ambiguously situated between nature and culture
in that it institutes a stable pattern of societal relation which seems to
mimic a preexistent biological bond but derives from a set of norms; its
perceived naturalness is an effect of cultural practice.
The ultimate advantage of Levi-Strauss’s analysis of the elementary
structures of kinship is that it illuminates the social values enacted by

exogamy in particular, and of an openness to the other in general. As
Beauvoir noted, under Levi-Strauss’s analysis, the prohibition serves to
“prevent a group from becoming self-enclosed (se figer sur lui-meme) and
to maintain the possibility of exchange through contact with other groups”
(Beauvoir, 1949, p. 946). However, if the dominant lines of force pass
along those of quasinatural descent, and undervalue the line of exchange
and alliance between less “naturally” akin groups, there is a risk of social
and intellectual isolation, an inbreeding of customs and ideas, a lack of
influx of new blood. As Beauvoir emphasized in her review of Elementary
Structures of Kinship, the danger of consanguineous, endogamous and any
other in-group relations is not their presumed unnaturalness, but rather the
social isolation and congealment they produce. Incest—taken in the broad
sense of a preference for seemingly natural, in-group alliance—is prohib-
ited out of social and not natural/physical considerations; inbreeding
harms the society by putting a dam on the circulation and exchange of
precious goods—these goods are women within Levi-Strauss’s analysis of
patriarchic society, and “Saussure’s ideas” within an extended study of el-
ementary socioacademic structures proposed here. Insofar as the elemen-
tary social structure in an academic institution mimics primarily the rela-
tion of heredity along the same sex, its transcendence is greatly limited,
and the related risks of traditionalism in the professed doctrine, elitism and
lack of diversity among the members of the group arise. Following Levi-
Strauss’s and Beauvoir’s analysis, a social order attached primarily to he-
reditary ties along the male line comes with the danger of increasing im-
manence and isolation.
On Beauvoir’s reading, Levi-Strauss’s study is not a narrow anthropo-
logical analysis of matrimonial preference across a selection of societies
but a philosophical case for the vital importance of the self to encounter
and form alliances with an other; such openness to alterity is a mark of
transcending the order of nature and instituting the order of culture in the
human realm. This alterity is in Beauvoir’s own work best represented by
sexual difference, and the task is then to cultivate relations of reciprocity
between women and men, and not, as in Levi-Strauss’s account, between
men only (Beauvoir, 2010). While Levi-Strauss describes the universal
order of patriarchy as if it were unchanging because of its presumed univer-
sality, Beauvoir exploits its cultural hence changeable status; this opens up
the possibility of resisting the dominant order and instituting a different
one. Ultimately, Beauvoir puts pressure on Levi-Strauss’s own distinction
between nature and culture; it may be more appropriate to speak of two
cultures, one which values immanence and sameness, another one which

promotes transcendence and difference. Both are normative and value
laden, whatever their scope. To put the point differently, if we accept that
norms such as the prohibition of incest are (always) already in place in the
human realm, it is impossible to isolate a precultural layer untouched by the
institutions; similarly, a seemingly natural and generally sanctioned bond
between the master and the disciple in the academic context is itself an
effect of a set of values privileging relations of direct succession and same-
ness over the relations of reciprocal alliance and difference. Whether we
are dealing with issues of matrimonial preference or succession by an heir,
there is a choice of alliance involved which goes beyond whatever could be
attributed to a preexistent order of things. And while the former, following
Beauvoir’s reading of Levi-Strauss, may open the door to an alternative
culture if based on reciprocity within and across difference, the latter privi-
leges a culture of self-same continuity along (assumed) heritage lines.
If it is granted that the elementary structure in European academia is
the one of succession and sameness (e.g., it is still common in some Euro-
pean institutions to hire their own graduates in the name of maintaining
continuity in an intellectual tradition), then we can shed light on the other-
wise perplexing fact that scholars like Bally, Sechehaye, and Hjelmslev
were successful in becoming recognized as exemplars of faithful Sau-
ssureanism in academic contexts. The fact that their interpretation was
instituted and received as unfiltered Saussureanism gains some intelligi-
bility within such an institutional and societal context, but it would also
lose its guise of presumed inevitability, and become exposed as a mark of
an academic culture which values sameness over difference, and filial con-
tinuity over exogamous contestation. Such a denaturalization of the ele-
mentary societal structure ultimately helps to loosens the bond between
Saussureanism and protostructuralism; both appear as culturally and so-
cially motivated processes, based on an order of norms rather than nature,
and as such subject to revision—regardless of how widespread and seem-
ingly universal they may be. This opens the door for alternative readings
and appropriations in particular, and for an alternative academic culture in
general.
Reversal of order
In this section I continue to examine the process of ghostwriting the Sau-

ssurean doctrine into the Course but now with an emphasis on how the
selection and organization of the book’s contents served to establish the

primacy of language (la langue) over a plurality of particular languages
(les langues). The process consist in reversing the original layout of the
lecture notes, such that language (la langue) gets placed at the head of the
book while the survey of the diversity of languages (les langues) gets
pushed to the far end; this reversal produces the impression that la langue
is a preexistent and self-standing object while the many different lan-
guages are its contingent realizations. The ample historical material deal-
ing with the many languages (les langues) from the student lecture notes is
thus marginalized in favor of an abstract category of la langue, which is
construed as an a priori axiom, presented to the reader right from the start.
This reverses the order of presentation from the student lecture notes,
where a targeted discussion of la langue is deferred to the second part of
the third course, and only sketched out. The marginalization of the histor-
ical material and the reversal of the organization from the order of an on-
going (and perhaps never ending) discovery in the student lectures notes to
an order of assumed dogmatic knowledge produces a similar effect to the
one noted above in relation to the “famous formula.” They serve to une-
quivocally establish the primacy of language system (la langue) over its
contingent realizations in the Course by way of, respectively, the book’s
written contents and its form. I will make the case that the editorial proc-
ess of establishing the primacy of la langue in the Course was enabled in
part by Sechehaye’s professed penchant for great abstractions.
Pace the preface to the Course, the volume is based largely on the third
course on general linguistics; the order of presentation runs, however, in
reverse. The main parts distinguished in the book are: Introduction, Pho-
nology, General principles of signs (part one), Synchronic linguistics (part
two), Diachronic linguistics (part three), Geographical linguistics (part
four), and Retrospective linguistics/conclusion (part five). This ordering
reverses the one found in the student lecture notes for the third course,
which begins with issues related to geographical diversity of les langues
and the causes of this diversity. A general theory of the linguistic sign was
begun only after the historical survey of known language families (les
langues) had been completed; as a result, the material dealing with les
langues is pushed to the margins (in the Course, the discussion of lan-
guages and their diversity is found only in parts three, four, and five; they
are the focus of part four on Geographical linguistics), while the material
dealing with la langue turned into the centerpiece (introduction, parts one
and two).
The reversed ordering reflects a presumed authorial intention to discuss
general principles of linguistic theory before turning to specific instances,

and to thus oppose the so-called internal to external linguistics. A general
understanding of the language system (la langue) would need to precede
any considerations relative to the historically known languages (les
langues). The order of exposition serves to justify the isolation of la langue
as a proper and privileged object of study, behind which a historical survey
of languages could trail as a series of empirical variations on a basic theme.
This lends an aura of a deductive thesis to the Course, where the reader
encounters axiom-like statements about la langue early on, with little jus-
tification or background. The reader of the student lecture notes, however,
is unlikely to receive an impression of a dogmatic doctrine, since la langue
is tackled last, as an a posteriori claim and an ideal object culminating a
concrete linguistic study; it is an entity in search of its own definition, and
presented in guise of an enigma.
The lecture notes effectively undermine the primacy and independence
of a language system. Saussure begins the third course with “Geographical
linguistics,” and offers a wide overview of language families (including
the Semitic languages in addition to Indo-European). It was only in the
second semester (April, June, and July 1911) that he turns to general con-
siderations relative to la langue. Although deemed an “essential part of the
subject” by the editors (Saussure, 1986, p. [7], xvii), this discussion was
never completed in the lecture course itself. According to the concluding
lines of the third course, “only the external part [of linguistics] is close to
being complete” (Saussure, 1993, p. 143a); the internal study of la langue
itself is never realized; it figures therefore like a sought after ideal, and not
an already secured foundation. Furthermore, the contributions made in the
lectures to an internal linguistics of la langue are deemed incomplete,
since they would have neglected the evolutionary or diachronic dimension
in favor of synchrony (ibid.). In sum, the third lecture notes present la
langue in the context of an in-process exploration, whose object emerges
as an elusive category, and an intersection of synchrony and diachrony.
The survey of les langues serves as an introduction to general consider-
ations relative to la langue in the third course. According to the schedule
announced on November 4, 1910, this lecture course divided into: 1. les
langues, 2. la langue, and 3. faculté et exercise du langage chez les indivi-
dus. The editors scrapped the part on les langues from their exposition,
regarding its inclusion in Saussure’s lectures to be externally motivated
and inessential to general linguistics: “The requirements of the curriculum
. . . obliged him [Saussure] to devote half of each course to a historical and
descriptive survey of the Indo-European languages, and the essential part
of this subject was thus considerably reduced” (Saussure, 1986, p. [7], xvii,

translation revised).16 However, it may be that the survey was not as redun-
dant as the editors make it sound; it may in fact have formed an organic
part of the lecture series (Harris, 2003, p. 19). It is significant to note in this
regard that the survey dominates the second course (1908–1909), taking
up some the three fifths of the total (based on Riedlinger’s notes, Godel,
1957, pp. 53–76); it would have therefore been more substantive than the
“obligatory half” suggests. This extensive survey in the second year con-
trasts to be sure with a much briefer one offered in the first, which takes up
around one twentieth of the whole (still not fitting into the presumed
“obligatory half” requirement); its role changes also depending on the
year: it serves as an introduction in the third, and as a lengthy concluding
part in the second lecture series. It remains therefore unclear what role the
survey plays within Saussure’s lectures, and while the relation between les
langues and la langue may be getting worked out without becoming fully
resolved in the course of the lecture series, the two are considered in
tandem throughout. In sum, “suppressing an intrinsic part of a course can
hardly pass muster as a conventional editorial practice. Here we see [a]
reason why the students’ perspective on Saussure’s teaching can claim to
be unique. They, unlike the subsequent readers of the CLG, were in a po-
sition to relate the ideas on general linguistics to the complementary pic-
ture of an established body of linguistic facts that the lecturer took care to
present in tandem. Somehow, the two had to be understood as related”
(Harris, 2003, p. 21).
The interconnection between les langues and la langue is apparent
throughout the whole series of lectures on general linguistics. Below I
offer a general outline with comments as illustration.
The first course (1907) includes the following parts: Phonology or laut-
physiologie; Historical linguistics, phonetic and analogical changes; Rela-
tion between units (unités) perceived by the speaker in synchrony (subjec-
tive analysis) and by the units isolated by historical grammar, such as
roots, suffixes and others (objective analysis); Popular etymology; Issues
of reconstruction. The first course has an “external” focus—its main inter-
est is in issues relative to the historical aspects of language. These subjects
were relegated by the editors to the marginal parts of the Course, such as
appendices and the last chapters (part III: Diachronic linguistics, chs. II–V;
16
The introduction and the first, second, and fourth parts are sourced from the student lecture
notes from the last (third) course; the third part and the appendices draw on the notes from the
first, while the supplement, as well as select chapters, include material from the notes from the
second course.

Appendix A: Subjective and objective analysis; Appendix B: Subjective
analysis and determination of units smaller than the word). However, the
perspective of synchrony and subjective analysis is already present, and
distinguished from an objective or external historical approach. Further-
more, the concept of analogy and analogical change, distinguished from
purely phonetic evolution of the sonic dimension of language, is discussed
at length. Analogy is deemed a “general creative principle of language”
(principe général des créations de la langue), a factor responsible, simul-
taneously, for innovation and conservation in language as such (la langue)
(discussed in more detail in ‘Beyond the doctrine’, ch. 6). Principles of
presumed general linguistic validity are therefore introduced alongside
and as part of a situated analysis of concrete examples of languages from
the present and the past, without language as such being posited as an in-
dependent object.
The second course (1908–1909) is devoted to the following: Relation
between a theory of signs and theory of language (la langue), with defini-
tions of linguistic system, units, identity and value; From this corpus of
basic definitions two methodological perspectives or points of view in a
study of linguistic facts are derived: a synchronic and a diachronic
perspective—and the problems associated with them, namely how to coor-
dinate the two perspectives in a system—not as a sequence but rather as an
interwoven web: (la langue est un système serré . . . [Saussure’s conversa-
tion with Riedlinger from January 19, 1909, cited in Godel, 1957, pp. 29–
30]). Then follows the survey of Indo-European languages, which forms a
substantial part of this course, as already noted. Since the survey serves as
an “introduction to general linguistics” in the lectures, and even as prepa-
ration for a course on the philosophy of linguistics (un cours philosophique
de linguistique) according to Saussure’s conversation with Riedlinger
mentioned above (Godel, 1957, p. 30), it is obviously tied to general lin-
guistic considerations. Even though this course is more general in orienta-
tion from the first one, it is also undecided between an internal and exter-
nal focus.
The proposed plan for the third course (1910–1911) is: 1. les langues, 2.
la langue, 3. faculté et l’exercise due langage chez l’individu (as announced
on November 4, 1910). This course combines the general order of the
second course with the historical analysis of the first. The problem of les
langues, or “external linguistics” is connected with the question of la
langue, as already noted. One finds therefore a dialectical interplay be-
tween general and historical considerations in the third course as well as
throughout the lecture series, with general considerations relative to la

langue being tied to the survey of particular languages. The study of lan-
gage as individual faculty is barely sketched in the lectures themselves
(but developed further in the autographed sources).
A taste for great abstractions
The primacy assigned to la langue in the Course can be partially traced

back to one of the editors’ own methodological and conceptual com-
mitments. In his Programme et méthodes de la linguistique théorique:
Psychologie du langage, Sechehaye develops a project of theoretical
linguistics, that is, a linguistics which construes language as an ab-
stract idea:
Just as much as we can conceive an idea of language in general above the

idea of particular languages, we can imagine a science which studies this
phenomenon in the sense of an abstract idea . . . this is what we propose to
call theoretical linguistics. (Sechehaye, 1908, pp. 9–10, my translations)
Theoretical linguistics is thus elevated above historical linguistics, and

its study of contingent facts affecting a plurality of languages. Its object of
study is language in general.
The project of general linguistics is guided by Adrien Naville’s topol-
ogy of the sciences, and the distinction between the empirical sciences
which aspire simply to observe and describe facts (les sciences des faits)
and the rational sciences which seek to identify invariable laws (les sci-
ences des lois) (ibid., p. 4). These two basic sciences may well study one
and the same object such as the sun; the former would describe it as being
hot and luminous, while the latter would draw on general principles from
mechanics, physics, and geometry. Their methodological approaches are
therefore different; factual sciences remain at the level of the contingent
content of perception, the phenomenon (ibid., p. 6), while the sciences of
laws grasp an intelligible form, and seek to identify “the general and the
necessary behind the contingent” (ibid., p. 4). Their truths are independent
of time and place, and are universally valid. Even though Sechehaye seems
to oppose the methodological naivety of factual sciences to the sophistica-
tion of the rational ones, he envisages, ideally, an alignment of the two
with the method of induction being used to distill general laws, and the
same laws being used in a process of deduction to derive possible applica-
tions (ibid., pp. 5–7).

In the specific context of theoretical linguistics, Sechehaye advocates
the following process:
Theoretical linguistics begins . . . with the fact but one does not return to
it; one moves up to the general principles and remains at that level; one
defines them, lists them, justifies them rationally as much as that is possi-
ble, and, equipped with these principles, instead of trying to reconstruct
the reality of history in an imperfect way, one edifies by deduction a gen-
eral system of possibilities, of which reality is but a contingent application.
(ibid., p. 8)
Assuming the validity of absolute determinism, as well as the human

ability to know all the applicable conditions and laws, the realm of possi-
bility will one day coincide with that of reality, and everything will appear
to us as an effect of “one sole and eternal cause” (ibid., p. 8). As a result,
“the whole world, with all its multiplicity, will be for us but a vast theorem
demonstrated by the science of laws, and which history confirms experi-
mentally” (ibid.).
This is, in a gist, the methodological framework that Sechehaye later
applies in the process of collating the source materials related to Saus-
sure’s general linguistics. The distinction between the sciences of empiri-
cal facts and general laws serves to drive a wedge between a survey of
known languages on the one hand, and a science of universal linguistic
laws on the other. The two are ordered chronologically in a sequence, with
the study of known languages serving as a stepping stone onto the higher
plane of general linguistic laws.
As Sechehaye put in his editorial notes:
He will derive by abstraction that which is universal from the languages. He

will therefore be faced with a totality of abstractions by means of which he
will have access to, as much as it is possible, to la langue; in other words,
we will study that which can be gleaned of la langue within different lan-
guages (les langues) within these abstractions.
It is therefore necessary to pass through a study of languages (les langues)

to arrive at a study of language (la langue). (cited in Harris, 2003, p. 52)17
17
See Harris, 2003, pp. 51–54 for a slightly different take on this material; Harris emphasizes the
difficulty faced by the editors in reconstituting Saussure’s usage of la langue as a general term,
and their overall goal to offer foundations of a modern science of language in the Course.

Having adopted this methodology of abstraction of la langue from les
langues, the editors have free hand to reorganize and alter the contents
from the third course as they see fit. Initially, their plan was to preserve the
order of exposition followed by Saussure in the lectures, that is 1. Les
langues, 2. La langue, 3. La faculte et l’exercise du langage chez les indi-
vidus. As Harris notes, “At this stage in the editorial process, clearly, the
program Saussure had devised for the third course was being taken very
seriously and followed as closely as possible. That approach was later
abandoned by the editors in favor of the quite different arrangement ad-
opted in the 1916 text” (2003, p. 51).
The initial fidelity to the source materials was apparently undone by
Sechehaye’s penchant for abstractions; la langue changes status from a
generalization to an abstraction adopted like an a priori axiom.
Even though he deployed the abstracting method to alter the source ma-
terials from the third course, Sechehaye ends up attributing this method to
Saussure himself, and argues that the master and the disciple shared a pen-
chant for abstractions:
If we dare to make a comparison with F. de Saussure, we would say that his

disciple [Sechehaye refers to himself] shares with him the taste—we say
simply the taste (le gout)—for great abstractions and these mental visions
(ces vues de l’esprit) which surpass and dominate the facts. (Sechehaye,
1927, p. 234, my translation).
To the reader of Programme et méthodes de la linguistique théorique

and of the editorial notes from the Collation, this taste for abstract notions
which surpass and dominate the facts seems to be uniquely the editor’s
own, but Sechehaye projects them onto the source materials of Saussure’s
general linguistics, and then infers them as a common trait passed down
from master to disciple like from father to son. The contents of the student
lecture notes are organized and systematized according to this presumed
(or divined) common taste, and the abstract category of la langue is posited
above and beyond the factual multiplicity of languages. In sum, “Seche-
haye’s Collation of the students’ notes as the basis for the 1916 edition
clearly reflects the editors’ intention to reorganize Saussure’s third course
along lines that clearly conformed to the requirements of the editors [to
present a general theory of la langue] rather than to accurate transmission
of what Saussure’s students had annotated” (Thibault, 2005, p. 670).
There is no evidence in the student lectures notes or the autographed
materials that a clear fact–law distinction is assumed and that a two step

passage from one to the other is advocated or even considered possible.
The general linguistic question is being raised within the broad horizon of
many languages—as many as possible in fact (Saussure, 1993, p. 10).18 The
notion of language in general emerges in an encounter between languages,
and not at a higher level of abstraction. If there is an identifiable logos
within the many different languages, it would be found in the interstices
between them, as a possibility of communicating across difference and as
an unbridgeable limit to the translatability from one idiom to another. The
quest for a logos in language does not lead to a universal and generic su-
prastructure to be projected upon the varied diversity.
The source materials emphasize that any consideration of language as
such (la langue) is tied up both with a survey of the many languages (les
langues)—but also with a consideration of the human language faculty
(langage). One finds therefore an undecided ambiguity between these dis-
tinct yet related terms, which prevents any easy isolation of one as sup-
posed foundation of the others. Firstly, la langue is not framed as a closed
system, but rather as a general principle applicable to—and indissociable
from—concrete languages. Hence Saussure dictates to his students on
May 5, 1911:
This being our understanding of language (la langue), it is clear that we

have access to it only through a series of different languages. We cannot
grasp it except through some specific language, any one. The language (la
langue), this word in the singular, what is its justification? I intend it as a
generalization, that which will turn out to be true for any given language,
without having to specify which one. (Saussure, 1993, p. 78)
This theoretical consideration of the language system is to be inserted

at the end of ch. I titled Geographical diversity of the language (la langue),
and thus it is raised already in the context of a survey which the editors had
scrapped. The relation between the two parts is therefore not external, and
considerations of general linguistic nature are raised already in the over-
view of specific languages. The general category of la langue has a firm
foothold in the study of les langues—the emphasis on the former helps to
approach the different languages by way of their shared traits and princi-
ples, as a logos located within any and all specific languages, and made
manifest especially via a comparison—and contrast—between them. The
focus on la langue does not simply dispense with comparative philology
18
See also Harris, 2003, p. 52.

for the sake of general linguistics, but offers a unitary perspective on the
languages being compared—something that is lacking within the tradi-
tional comparative approaches. Saussure can therefore blame the compar-
ative grammarians for dealing with the languages only, and for failing to
trace their shared logos or la langue—without yet having to abandon the
territory of languages (les langues) in the process. The emphasis on la
langue does not therefore imply a turn away from linguistic diversity and
contingency to an autonomous and closed system of signs.
Saussure’s autographed writings also emphasize that la langue cannot
be dissociated from les langues. The distinction les langues–la langue is
presented as one between a diversity of languages the linguist encounters
during the course of study and the subsequent generalizations:
Languages, found all over the globe are the linguist’s concrete object
(l’objet concret) of study. Langue is the name which may be given to what
linguists will be able to derive from these observations of all languages,
through time and across space. (Saussure, 2006, p. [307], 215).
One may wander about the validity of Saussure’s proposed “generaliza-

tion, that which will turn out to be true for any given language”—what
criteria would be used to distinguish between essential and universal fea-
tures on the one hand and particular and accidental on the other? Can such
a generalizing pursuit be sustained considering the irreducible diversity of
languages? The least that can be said is that this question is raised at the
level of linguistic multiplicity, and not assumed a priori as an abstract and
universal law. The road to generalizations leads through the diversified
linguistic field, and it is not assumed in advance that the ideal end of this
journey will ever be reached.
In sum, the structure of the course laid out in Saussure’s autographed
notes dictates the order of inquiry; the survey of languages comes first in
the course and in linguistics alike, since the linguist can only make lin-
guistic generalizations on the basis of linguistic diversity, which needs to
have been established first. The book on general linguistics would then
open with a survey of languages, and the survey would constitute an inte-
gral part of the project of making general claims. As Saussure’s notes
make apparent, general linguistics is a linguistics of generalizations based
on the many particular languages; it is not a self-standing discipline whose
one and true object of study is la langue; such an exclusive and unitary
focus emerges rather as an inadmissible abstraction at odds with Saus-
sure’s commitment to linguistic diversity.

According to Saussure’s notes for the third course, linguistic diversity
comes first in language experience and study:
The plurality of linguistic forms in the world, the diversity within language
(la langue) as we move from one country to another, or one region to an-
other, that’s the primary observation (la constatation primordiale), imme-
diately available to everyone . . . I say that this geographical diversity is the
primary fact for a linguist and indeed any person in general. Language
variation over time necessarily escapes the observer, but variation in space
cannot possibly do so. (Saussure, 2006, p. [307], 215)
Saussure concludes that the two variations (in space and in time) are
ultimately inseparable, but only the former is “immediately given”
(ibid.). The observer situated in the present state of language is therefore
not exposed to a homogenous block, and even a speaker within a mono-
lingual culture can and should cultivate awareness of linguistic diversity
through exposure to a variety of languages and dialects. It is therefore
incumbent on a linguist to become immersed in the languages and ap-
preciate their differences as part and parcel of the process of making
general claims. Only through realization of linguistic diversity can a lin-
guist, and any language user, gain “consciousness of language” (con-
science de la langue) (ibid.). Saussure’s preference for a synchronic per-
spective (the immediately given) is therefore consistent with an emphasis
on plurality and difference; the synchronic point of view implies an id-
iosynchronic appreciation of the plurality of idioms coexisting at any
point in time.
Just as questions relative to les langues and la langue belong together,
so do the ones relative to langage and la langue. Saussure addresses this
terminological ambiguity within an inaugural lecture at the University
of Geneva, where he returned from Paris in 1891 to teach Sanskrit and
Indo-European languages. The specific problem raised in this (and the
remaining) lecture is the justification and place of general linguistics
(although historical linguistics is also represented) within the academic
institution. The specific question is whether or not the object of study
for the science of language (la science du langage) is le langage or la
langue considering that langage can be viewed as a faculty of producing
articulate sound, hence a natural endowment of the human organism
when compared to other animals (Saussure, 2006, pp. [145–146], 95). It
is therefore unclear if the study of language belongs to the natural or the
human sciences.

Even though Saussure does not settle this question in the first lecture, he
clearly states considerations relative to la langue and langage are indisso-
ciable, even though they can be distinguished in principle:
Langue and langage are but one thing; one is the generalization of the
other. To claim to study the phenomenon of language (le langage) without
bothering to study its manifestations in various languages (les langues) is
an entirely vain and empty undertaking; on the other hand, studying lan-
guages (les langues) while forgetting that they are regulated primordially
by certain principles which make up the very idea of language (langage) is
a task even more devoid of any serious meaning, of any veritable scientific
basis. (Saussure, 2006, p. [146], 96)
One could deflate this whole terminological ambiguity by stating that

Saussure has not yet clarified the distinction between la langue and le lan-
gage around 1891. However, a similar ambiguity arises in the course of
lectures on general linguistics (1907–1911), which begin by listing the mul-
tiple and interrelated focal areas of interest to the linguist:
Adopting an internal principle, we could define linguistics as: the science of

language (le langage) or languages (les langues). But the question then im-
mediately arises: what is language (le langage)? Now even for a linguist
who has a view of his science as a whole it is very difficult to determine the
nature of the linguistic phenomenon of the language (la langue). It would be
illusory to attempt it at the outset, within the short time at our disposal.
(Saussure, 1996, p. 1)
The opening lines thus set up an open, relational field between langage,
les langues, and la langue, without prioritizing or separating any term
from the others. Saussure uses langage and langue interchangeably, and la
langue does not have a distinct identity within this constellation. It is
therefore not clear if there is a single object (la langue, le langage) or
many (les langues), whether the object belongs to the human faculty to
produce and process articulate sound—or to the more general set of rules
and conventions shared in a society, and in what relation they stand (in-
ductive or deductive, condition of possibility or generalization). And while
la langue is introduced as a possible object of study, it is under the guise
of an enigma whose nature evades the linguist rather than a readily classi-
fiable entity.

This open ended formulation is taken up again in the opening lines of
the chapter on linguistics:
One can hesitate about the best plan. It is more advantageous to place cer-
tain general ideas at the end of the course rather than at the beginning. This
is why we do not wish to define the nature of language (langage). This
would make itself the object of a course: one would have to remark that lan-
guage (langage) is not an immediately classifiable object. <One would find
no genus>, no <larger> category to which a specific difference is added to
form the definition of the object. (Saussure, 1996, p. 27)
The question about the object of study raised in the first course is taken
up again in the third (in the second part devoted to general linguistics,
April 25, 1911). There the focus becomes sharpened on la langue as an
essential part of langage, with the proviso that it is only a part (Saussure,
1993, p. 66). This linguistic generalization could in principle be studied on
its own; it is a “separable totality” (un tout séparable), a unitary category
which the mind can grasp (une unité satisfaisante pour l’esprit) (ibid.,
p. 66). La langue is then described as a social product enabling an individ-
ual to employ the language faculty; it seems then that la langue could be
opposed to langage as a collective product and an ensemble of social con-
ventions can be opposed to an individual faculty and exercise of langage.
Such a split is, however, untenable: as soon as Saussure posits la langue as
an independent object, he subsumes it together with la parole (speech)
under the larger unity of langage (ibid., p. 70).
According to Saussure’s notes, the object of general linguistics is the
heterogeneous entity langage, whose reality is so multiple that it would be
unclassifiable were it not for the possibility of arranging it on a series of
dual terms. Saussure distinguishes a number of pairings in langage: the
two sides of the sign (signifier and signified), the speaking individual and
the society (or mass), langue and society, langue and parole, with parole
being tied to the acoustic production (unlike langue, which belongs to the
order of established conventions in a community), and to the individual
will (as contrasted with social passivity) (Saussure, 2006, pp. [298–299],
208–209).19 Neither of the terms within these pairings has primacy; while
19
These distinctions bear on langage, and not, as the English translation in Writings (ibid.) has it,
on la langue—an unhappy mistranslation that would make langue both the general category and
a member of the category.

they lend themselves to being distinguished, it is also the case that one is
undecipherable without the other. Thus one must, for example, distinguish
between la langue and the society, but only on the condition of acknowl-
edging that “langue is social, or else does not exist” (Saussure, 2006,
p. [298], 208). Rather than posit a unitary object, la langue, Saussure
adopts therefore a more complex notion of an indissociable duality which
serves to organize the heterogeneous field of langage into a series of basic
distinctions. This is in agreement with the conception of general linguis-
tics as a dual science and the notion that the law of duality cannot be tran-
scended (infranchissable) (ibid.).

8 Structuralist and poststructuralist
reception of the Course
S
tructuralism emerged as a new theory of meaning or significa-
tion which displaces the signifying subject in favor of an autonomous
semiological system, and sets up a host of rule governed, oppositive,
and combinatorial relations between signs that alone constitute and explain
signification. Structuralism embodies a manifest paradigm change that
swept the European academic world in the twentieth century; it enabled the
study of any human phenomenon (matrimonial preference, prescribed at-
titude to one’s kin, folk tales, and even sport) on the basis of the relations
governing between its signifying elements.
One of the signature marks of structuralism is reliance on the opposi-
tional pairings of signifier and signified, la langue and la parole, and syn-
chrony and diachrony. Drawing on these hierarchical and dichotomous
terms as part of an effort to map out the many aspects of human reality
(social arrangements, neurotic symptoms, literary genres) may in fact be a
hallmark of structuralist activity, one which distinguishes it from the other
traditions of inquiry (Barthes, 1972, p. 213). Structuralism (at least in its
French postwar version) can therefore be defined as a direct application of
the “Saussurean doctrine” to the many fields within the human sciences as
well as philosophy.
The term “structuralism” covers a large and varied territory of know-
ledge, and arguably one finds structuralist elements throughout the written
history of Western philosophy in the many attempts to characterize objects
in terms of a combination of structural elements within a system (Culler,
2006, p. 5). However, it is standard to define structuralism in the proper
sense of the term as the intellectual movement that displays a direct line-
age to Saussurean linguistics from the 1916 Course; a number of scholarly
works devoted to structuralism testify to this trend. Culler notes that “the
term structuralism is generally used to designate work that marks its debts
to structural linguistics and deploys a vocabulary drawn from the legacy of
Ferdinand de Saussure . . . There are many writings, from Aristotle to
Noam Chomsky, that share the structuralist propensity to analyze objects
as the products of a combination of structural elements within a system,
but if they do not display a Saussurean ancestry, they are usually not
deemed structuralist” (ibid., p. 5). Sturrock states, “The founding father of
structural linguistics in Europe, and the man frequently looked on as the
patron of the whole Structuralist movement, was the Swiss linguist Ferdi-
nand de Saussure” (Sturrock, 2003, p. 26). And Dosse observes that struc-
turalism’s (in the proper sense) “central core, its unifying center, is the
model of modern linguistics and the figure of Ferdinand de Saussure, pre-
sented as its founder” (Dosse, 1997, p. 43). The structuralist “return to
Saussure” would belong to the period’s prevailing theme of “returning” to
foundational figures, like Marx and Freud (ibid.). In sum, the identity of
structuralism as a distinctive intellectual movement is closely bound up
with its historical foundation in the Great Book authored by/attributed to
Ferdinand de Saussure.
Therein lies the strange paradox of structuralism in the proper sense: it
privileges an autonomous semiological system as part of its doctrine but
claims a contingent historical subject and body of work (oeuvre) as a
source of its own legitimacy. One way of deciphering this paradox is by
exposing an unacknowledged entanglement of structuralism within the
very history of Western metaphysics it purportedly breaks all ties with.
Derrida offers such a critique in the essay “Structure, Sign, and Play,”
where he writes:
The concept of structure and even the word “structure” itself are as old as
the episteme—that is to say, as old as Western science and Western philos-
ophy . . . structure—or rather the structurality of the structure—although it
has always been at work, has always been neutralized or reduced, and this
by a process of giving it a center or of referring it to a point of presence, a
fixed origin. (Derrida, 1978, p. 278)
The “structure” implied in structuralism is thus organized around a

stable center, such as substance, subject, consciousness, God, or man,
which is believed to tap into a site of invariable presence, a plenum of nec-
essary being around which secondary and derivative beings (like attri-
butes, objects, the unconscious, the world, the animal) arrange themselves
at a respectful distance, like satellites. The center is therefore, paradoxi-
cally, a sine qua non of a structured system, but it is rendered immune to

the “structurality of the structure”; that is, the center is excluded from the
field of play which should apply to all the elements within the system, and
make them permutate, transform, and substitute for one another (in agree-
ment with the basic law of linguistic arbitrariness, according to which no
single sign has a proper signification and a proper place within the system).
The center is fixed and frozen as a transcendental signifier that names a
transcendental signified in a direct and unmediated fashion. According to
Derrida, a structurality of the structure can emerge therefore only through
a decentering movement, where the center is exposed as a no-thing: not an
ontological category but rather a function within the system, which can be
filled out in a number of ways (Macksey, 1970, p. 271).
While Derrida applies the metaphysical notion of a centered structure
to the work pursued by the practitioners of structuralism (notably Levi-
Strauss), it applies just as well to the above cited definitions of structural-
ism by its proponents (Culler, Sturrock, Dosse). According to their defini-
tions structuralism in the proper sense revolves around a fixed and
inalienable center: Saussure-the-author-of-the-Course. This is a process
“of giving [structuralism in the proper sense] a center or of referring it to a
point of presence, a fixed origin” (Derrida, 1978; cited above). Since “it has
always been thought that the center which is by definition unique, consti-
tuted that very thing within a structure which while governing the struc-
ture escapes structurality” (ibid., p. 279), the central place occupied by
Saussure-the-author-of-the-Course within structuralism escapes a possi-
bility of substitution of this sign by another, and excludes a possibility of
multiple origins of structuralist activity being traced. The structurality of
the structure is thus “neutralized and reduced” in the way structuralism is
usually defined; it follows that metaphysical mechanisms are replicated
not only within structuralism itself (Derrida’s claim), but also within dom-
inant scholarly narratives about structuralism as a school of thought. An
alternative narrative about structuralism, one that preserves the structural-
ity of the structure, would open up a space for a free play between the
many signifiers involved in the production, reception, and dissemination of
structuralist activity. After all, “the absence of the transcendental signified
extends the domain and the play of signification infinitely” (ibid., p. 280),
and the structuralist play can extend onto less well traveled territories
(Eastern Europe), and less typical timeframes (pre-Second World War).
Such a redefinition of structuralism would let go of the usual tropes of a
founding father, a lineage of descendants/disciples, and a dedicated school.
On this redefinition, structuralist activity becomes decentered, that is, de-
coupled from the canon of past masters, the cult of Great Books, and a
struc tur alist and poststruc tur alist recep tion of the course 241
single intellectual capital in the city of light. Like the structured systems
themselves, structuralist activity becomes more diffuse and eager to trans-
gress the borders between the center and the periphery. I believe that such
a decentering may help to heal the antagonism between phenomenology
and structuralism; on a broader definition, structuralism was contaminated
by phenomenology already at its inception, as evidenced by the work of
Saussure from the Nachlass (and the work of his predecessors), and in the
early Eastern European structuralist developments. The perceived antago-
nism can therefore hardly be dictated by an insurmountable incompatibil-
ity of methods. In fact, these methods are bound to intersect when we
follow Saussure’s lead and regard language as a phenomenon. Language
(but other phenomena would fit the mold as well) is effectively undecidable
between experience bound description and structural analysis; these two
modes of inquiry can be productively combined once we leave open the
possibility that both depart from their “classical” mode and “proper” defi-
nition in the process.
Structuralism: east and west
Regardless of the usual definition of its proper sense, structuralism in the

twentieth-century does not have a fixed and unitary origin. Historians typ-
ically trace at least two distinct loci for claiming Saussure’s general lin-
guistics as a foundational reference. The first was the meeting of Russian
and Swiss linguists at the First International Congress of Linguistics at the
Hague in 1928 (Dosse, 1997, pp. 44–45). This event offered the first op-
portunity for the Prague structuralists to present their group (the Prague
Linguistic Circle) and its central ideas at an international forum; another
congress followed suit in 1929, and the group presented a manifesto-like
statement of their core beliefs in the “Theses of the Prague Circle” (see
“Theses Presented to the First Congress of Slavic Philologists” in Steiner,
1982). At the Hague Congress, Russian linguists like Jakobson, Karcevski,
and Trubetzkoy, and the Genevans Bally and Sechehaye “made common
references to Saussure in their description of language as a system” (Gadet,
1989a, cited in Dosse, 1997, p. 44). This was also the first time that the
term “structuralism” was employed (by Jakobson)—a term that does not
figure in Saussure’s own writings, with “structure” being used infrequently
even in the Course, and not in a technical sense. It is regrettable that this
early locus of structuralist thought goes largely unrecognized in the schol-
arship; Steiner explains this oversight by two misconceptions: first, the

view that Prague structuralism was confined to specialized linguistic
study, and even more narrowly to phonology, whereas its topics extended
into various areas of esthetics, and ranged from literature to visual arts and
folklore; second, the notion that Prague structuralism was a restatement of
basic tenets of Russian formalism, whereas it offered a new paradigm of
knowledge (Steiner, 1982, p. x).1
The second locus is found in the 1950s return to Saussurean linguistics
in France by Merleau-Ponty in phenomenological philosophy, Levi-Strauss
in anthropology, Barthes in literary studies, and Lacan in psychoanalysis.
Curiously, Saussure was not received well within the field of linguistics
itself, but became a central reference across the human sciences—possibly
in a concerted attempt to provide an alternative to the stodgy academic
culture associated with the Sorbonne, which was positioned as the “bearer
of scholarly legitimacy and dispenser of classical humanities” in postwar
France (Dosse, 1997, p. xix). What united the French readers of Saussure
was the possibility of locating an adaptable scientific model for sociology,
anthropology, psychoanalysis, and literature in general linguistics. They
adhered to the Saussurean doctrine, with its assumed priority of synchrony
over diachrony and of a system of rules and relations over subjectivity; this
assumed hierarchy seemed to secure an objective basis for the many
human sciences, since any semiological study or any study of signs in
social life (phonemes, kinship patterns, neurotic symptoms, literary codes)
now seemed equipped with an independently verifiable object of study.
Levi-Strauss famously claimed that structural linguistics played a compa-
rable role for the social sciences to the one played by nuclear physics for
the physical sciences (Levi-Strauss, 1963, p. 33). And Lacan noted that
language defined as a structure within the unconscious attained the status
of a scientific object in the modern sense, this revolution in knowledge en-
abling linguistics to serve as a pilot science for the human sciences at large
(Lacan, 2006, p. 414).
If there may have been two returns to Saussure, one in the late 1920s
and then one in the 1950s, each received Saussure in a distinctive way.
The early readers like Jakobson were critical of the presumed doctrine
and the many dichotomies organizing the Course. In fact, the members of
the Prague Linguistic Circle differed from their Geneva Circle counter-
parts in a marked emphasis on a necessary dialectical integration between
synchrony and diachrony (as well as the system and the subject). As put in
1
See Steiner’s chapter “The roots of structuralist esthetics” in the anthology The Prague School
(Steiner, 1982) for more details on the second point.
the programmatic theses (which the former group presented to the First
Congress of Slavic Philologists in Prague in 1929), “There is no insur-
mountable barrier between the synchronic and diachronic methods, as
claimed by the Geneva School” (Steiner, 1982, p. 6, italics in the original).
The authors noted the illogicality of the notion that linguistic change is
nothing but a destructive and purposeless intervention, heterogeneous
from the point of view of the system itself. Change reflects rather the
needs of the system itself (its stabilization, and realignment), making a
diachronic study indispensable to an understanding of the system itself
(ibid.). Such a nonpolarizing reception of Saussure would have been made
possible in part by Jakobson’s background in phenomenology, especially
Husserl but also Hegel (see Holenstein, 1976, 2005). In sum, the Eastern
European structuralism depended on its alliance with phenomenology to
effectively deconstruct the “Saussurean doctrine”.
On the other hand, the later French structuralist readers typically cele-
brated the doctrine as a resource enabling the human sciences to rise to the
dignity of the natural sciences. This scientism of the French structuralists
accounts for the emphasis on the separable existence of structural systems
from subjective experience, including that of language from the speaking
subject. Interestingly, this separation between language regarded as objec-
tive code and its practice of usage was not made as strongly by the early
linguistic readers of Saussure like Jakobson, but was made with all force
by predominantly nonlinguists like Levi-Strauss and Lacan. Levi-Strauss
and Lacan both saw in linguistics a series of adaptable algorithms that
apply broadly to the human life, are hardwired into the unconscious both
at the individual and the social levels, and thus belong to the realm of uni-
versally applicable hard facts, as studied in the natural sciences. This sci-
entist agenda helped to cement the one party line reading of the Course, a
reading that removed any trace of ambiguity relative to the distinctions
between language and speech, synchrony and diachrony, system and sub-
ject, and projected a logic of violent hierarchies of dominant and subordi-
nate terms. The famous formula culminating the Course according to
which the language system was separable and could be studied on its own
constituted an ideal reference for such a pursuit; the chapters dealing with
the social and historical reality of language were altogether glossed over
within a structuralist reading. While the Course with its unambiguous
statement of a doctrine lends itself to a structuralist expropriation, the
French structuralist readers were partial in picking out just the elements fit
for a scientific program from the Course and passing over the less easily
assimilable ones. Derrida’s deconstructive reading of the Course in Of

Grammatology is unique in its emphasis on the manifest tensions between
the metaphysico-scientific elements and the rest, notably between the pos-
itive notion of the sign on the one hand and the view of language as a
system of differences without positive terms on the other.2
Prague structuralism has become largely eclipsed by its French succes-
sor in scholarship—so much so that according to the relevant entry in The
Princeton Encyclopedia of Poetry and Poetics, structuralism “has been
assimilated and developed in various ways in other countries but it re-
mains in its most distinctive and characterizable form a French movement”
(Preminger, 1974, p. 378). This erasure of Prague structuralism from struc-
turalism’s history in Continental Europe by the scholars in the West may
be an expression of Francocentrism; still, the process was started in the
East shortly after the end of the Second World War, when the Communist
takeover established Marxism as official doctrine (Steiner, 1982, p. 66). As
Jakobson reports in July, 1949:
The terror of the Moscow authorities in the field of linguistics is for the
present much more fraught with consequences than it was before the war.
For now the whole scholarly life of the satellite nation is affected. During
the past academic year the brilliant linguistic activities of Czechoslovakia
began to be regimented on the Moscow level. The Prague Linguistic Circle
had to declare and publish a series of repentances and to disavow its ties
with Western scholarship and with its own past, to repudiate structuralism
and to rally around the banner of dialectical materialism and Marx’s doc-
trine. (cited in Steiner, 1982, p. 65)
This postwar repudiation of Prague structuralism has been surprisingly

effective, and demonstrates, if need be, the extent to which a life of ideas
is enabled—and in times of crisis, tragically undermined—by a political
and institutional context. The Saussurean doctrine became cemented as
the one party line reading of Saussure’s general linguistics as a result of
this silencing process.
*
Considering the dominant status of the French structuralism and the near
eclipse of the Eastern European one, it is unsurprising that the former
2
In line with his predecessors, the Derrida of Of Grammatology ignored, however, the social and
historical reality of language discussed in the Course. He largely maintained the structuralist
notion of a closed and autonomous system of signs (or signifiers), seemingly immune to social
influence.
maintained a firm hold on Saussure’s general linguistics in the twentieth-
century. Yet the identification of structuralism with a Saussurean ancestry
remains a curious fact considering the movement’s own professed goal to
transcend a historical explanation in favor of a structure-based one; for
example, Sturrock’s Structuralism simultaneously claims Saussure as the
founding father of structuralism (Sturrock, 2003, p. 26) and dismisses “the
fact that [the Course] is not a work of Saussure” as “unimportant” since it
is the soundness of the ideas themselves that overrides “our knowing
whether they are given in precisely the form Saussure himself would have
wanted” (ibid., p. 27). Yet if the ideas themselves are sound, why the need
for a backing from a founding father wielding a Great Book? If the oppo-
sitional pairings like signifier and signified, language system and speech,
synchrony and diachrony are inherently valid, why is the structuralist ac-
tivity so consumed with establishing a direct connection to a contingent
diachronic event (the delivery of lectures on general linguistics from 1907–
1911, the publication of the book in 1916), and to an individual speaking
subject (Ferdinand de Saussure, 1857–1913)? Nothing in the structuralist
program suggests the practical importance taken on by these historical
facts. One can decipher this allegiance to a founding father as a metaphysi-
cal desire for a stable center, but this explanation does not address the
practical side of the matter. I submit that this allegiance is necessary be-
cause structuralism in the accepted or “proper” sense is more than a study
of ideas with universal validity; it is an institutionalized movement, a
school (schola) within European academia, which finds legitimacy and
cohesiveness in a retroactive claiming of a past master and his body of
work as its foundation. One notes therefore the coexistence of two loosely
tied strands in in the way (the French) structuralism defined itself: one the
one hand, as a doctrine of inherently true ideas, notably the oppositional
pairings (signifying/signified, synchrony/diachrony, system/subject), and,
on the other hand, as a member of institutionalized European academia
whose legitimacy is not conferred by the sheer force of these ideas them-
selves but by conforming to the established scholastic practices of forming
an intellectual movement (not unlike a religious one). Considering this co-
existence of doctrinal and scholastic elements within the structuralist iden-
tity, it is unsurprising that its representatives, Lacan, Levi-Strauss, and
Derrida, will steer at one time in the direction of pledging allegiance to the
Great Book (and its author, a past master), and, at another, in the direction
of affirming the soundness of the ideas themselves.
As the Course became the foundational stone for the structuralist school,
Saussure himself became retroactively crowned as a structuralist—and the

label “Saussurean structuralism” is commonly used to describe Saussure’s
conception of language as a system of signs (Sturrock, 2003, p. 34). One
even finds comments that “One of the oddities of the Course in General
Linguistics, given the influence it has had, is that it nowhere contains the
word ‘structure’” (ibid., p. 28)—as if one naturally expected to find the iden-
tifying label (or some version thereof) for the chosen people within the book
later claimed as the source point of origin of their movement; yet the per-
ceived oddity of an apparent omission is enabled by the already accepted
equivocation between structuralism and Saussureanism.
It is interesting to consider how seamless the transition from Saussure-
anism to structuralism may have seemed in the 1950s and 1960s France.
Mounin’s Ferdinand de Saussure ou le structuraliste sans le savoir (Segher,
1968) is a representative specimen thereof. A contemporary reader of this
book expects to find some reasons why Saussure was a structuralist even
though he didn’t (or couldn’t) know it. Instead one simply finds an orthodox
appraisal of the usual hierarchical dichotomies: langue-parole, synchrony-
diachrony, presented as official Saussurean doctrine, and backed by a se-
lective reading of the Course. The case that Saussure was a structuralist
does not therefore even need to be made.
Mounin deploys this unexamined assumption as leverage against the
likes of Meillet, whose 1916 critical review of the Course is praised for its
finesse and detail, but chastised as “misleading in regard to the essential”
(fourvoyant sur l’essentiel). What makes a critical review of the Course
essentially misleading is that it voices reservations about the authenticity
of the text, insists on the lacunae in Saussure’s thinking about general lin-
guistics against the notion of a complete and closed system, laments the
paucity of space devoted to the question of historical change at the expense
on the dominant focus on stasis, and lends undue importance to phonetic
processes (Mounin, 1968, p. 162n). All that adds up to “incomprehension
of the substance of a work (ouvrage)” whose organic completeness Meillet
would have failed to appreciate (ibid.).3
This is a curious charge indeed: if reservations about the authenticity of
the text ghostwritten by Bally and Sechehaye (together with the other crit-
ical points made by Meillet) can be deemed misleading on the essential, it
is hard to fathom how any reservations about this opus could be made
without falling into the trap of incomprehension and deliberate deceit on
3
Similarly, Merleau-Ponty gets accused of committing a contresens by refusing the distinction
between synchrony and diachrony (Mounin, 1968, p. 80); the sole admissible reception of
Saussure could then only have been along the structuralist lines of thought.
the dissident’s part. This all out defensiveness can only be deciphered if
one considers that the inviolability of the doctrine is contingent on the un-
assailable authenticity of the Book. To touch the Book is to potentially
unsettle the assumed foundation of the structuralist program. The Book
can only function as the master’s word for the disciples to follow if its in-
tegrity is beyond and above scrutiny; if an intellectual movement like
structuralism finds legitimacy in its foundational sources, to question the
empirical authenticity of the sources is to potentially endanger the move-
ment’s standing in academia. The usual metaphors comparing the Course
to a vulgate edition of the Bible illustrate the case in point: the Course was
received and retroactively established as structuralist scripture, and thus
empowered to secure the foundation of the structuralist movement. Schol-
ars typically note that despite the difficult situation of the manuscripts left
after Saussure’s death, the Course served as “the sole basis of structural-
ism, just as the Vulgate served as the sole basis of biblical interpretation for
the Catholic Church” (Frank, 1989, p. 28). Such acknowledgements of tex-
tual complexity surrounding Saussure’s work typically leave the founda-
tion intact: one can express hope for an eventual reformation of Saussure
scholarship similar to the one of the Church (ibid.), and, like a good Catho-
lic, continue to read the Vulgate, and nothing but the Vulgate (and only the
usual, oft-cited, excerpts from the Vulgate).
Poststructuralism: Levi-Strauss, Lacan, Derrida
The two previously distinguished strands within the structuralist identity—

an emphasis on the inherent validity of structuralist ideas on the one hand
and a legitimation by recourse to a Great Book on the other—both inform
the way in which the structuralist and poststructuralist scholars like Lacan,
Levi-Strauss, and Derrida received the Course. Their reception is deter-
mined by the author function discussed in part I: even though they acknowl-
edge the existence of Saussure’s Nachlass, they continue to adhere to the
Course as exclusive site of Saussure’s work in general linguistics. Their
reception actively maintains received ideas from the Course as official
doctrine.
Lacan was exposed to Saussure’s work through Levi-Strauss. This ex-
change led to the famed Rome report (Ecrits), which makes a case that
modern linguistics should become a methodological guide in psychoanaly-
sis in line with its piloting role in anthropology. Lacan delved into the text
of the Course around 1953 (Dosse, 1997, p. 105); this reading is reflected in

the seminal essay “The Instance of the Letter in the Unconscious” (Ecrits).
This essay makes an explicit claim for locating the structure of language
within the unconscious. Specifically, Saussure’s distinction between the
signifier and the signified gets turned into an algorithm S/s, which helps
establish linguistic scienticity within psychoanalysis. It is generally agreed
that the algorithm deviates from the account of the relation between the
signifying and the signified facets of the sign from the Course. Lacan ele-
vates the signifying facet to a dominant position and relegates the signified
to a secondary one; this hierarchy is reflected in the “S” for the Signifier
being written in the upper case and the “s” for the signified in the lower.
The signified content thus gets separated or barred from the signifying
chain by a barrier that itself resists signification. This leads to the notion
that signs are embedded in an autonomous structure wherein they perpetu-
ally refer to other signs without ever centering on a core meaning; any re-
maining notion of such a solid core resistant to the process of substitutions
would be a mirage.
While Lacan “attributes” this hierarchy of the Signifier over the signi-
fied to Saussure, it deviates from an emphasis found even in the Course on
the coprimacy of the signifying and signified facets of the sign; this copri-
macy is figured in an image of the two inseparable sides of a sheet of paper
(Saussure, 1986, p. [157], 111), according to which to sever one side of the
sign from the other would be like trying to peel one side off of a sheet.
Consider the logic of Lacan’s attribution of the S/s formalism to Saussure.
He writes:
The sign written in this way [S/s] should be attributed to Ferdinand de Saus-
sure, although it is not reduced to this exact form in any of the numerous
schemas in which it appears in the printed version of the various lectures
from the three courses he gave in 1906–7, 1908–9, 1910–11, which a group
of his devoted disciples (que la piété d’un groupe de ses disciples) collected
under the title, Cours de linguistique gén érale—a publication of prime im-
portance for the transmission of a teaching worthy of the name, that is, that
one can stop only on its movement (on ne peut arrêter que sur son propre
mouvement). (Lacan, 2006, p. 415)
One notes first an attribution of the S/s formalism to Saussure despite

the acknowledged fact that it does not figure in the book; like the editors of
the Course before him, Lacan positions himself therefore in a disciple role
from which he can attribute ideas to Saussure directly, independently of
source materials (even though their existence is dutifully acknowledged).
His strategy mimics the “more daring solution” of reconstruction and syn-
thesis, the “completely objective” recreation undertaken by Bally and
Sechehaye in their prior handling of the student lecture notes. The struc-
turalist reception of the Course by Lacan thus mirrors the book’s inception
by Bally and Sechehaye; the latter initiated a subsequently established tra-
dition of trading in Saussure’s own ideas alongside a noncommittal ac-
knowledgement of their textual complexity. It is as if in the special case of
the founding father’s own ideas the devoted disciple can bypass the signi-
fying chain constituted by the texts to access the signified itself.
While Lacan claims to traffic in Saussure’s ideas independently of the
text, he invokes quasi-religious metaphors to bestow credibility upon the
Great Book, the Course in General Linguistics. In disregard of the histor-
ical facts of the matter, the volume is presented as a simple collection of
the student lecture notes, compiled by the very students themselves (!). The
making of the Course gets construed as an inspirational process pursued
by Saussure’s direct disciples, animated as they were by the force of devo-
tion (piété) to transmit the master’s teachings. This “transmission” of
Saussure’s teaching into the Book gets figured as being as unstoppable as
the teaching itself; the force of the teaching seems directly at work within
its (unstoppable) transmission, all bathed in an aura of convert devotion
and filial piety. The Course can thus be venerated as direct revelation of
the master’s word.
The historical facts of the matter, namely that the volume is not a simple
collection of the lectures, and that it does not result from the efforts by
Saussure’s disciples (in general linguistics), are covered over; should they
become revealed, they could still be dismissed as contingent and extrane-
ous to the underlying notion that teaching perpetuates itself along the lines
of filial piety and convert devotion. The volume would then be but a vehicle
of such an unstoppable process—a process that Lacan’s own devoted dis-
ciples were unlikely to stop, despite the master’s reasoning being couched
in a piece of seductive sophistry.
*
Levi-Strauss is more attuned to the editorial complexity surrounding Sau-
ssure’s work. In his discussion of the distinction between the synchronic and
diachronic, he acknowledges that “the recent documents show that the mas-
ter’s thought has at times been forced and schematized by the editors of the
Course” (Levi-Strauss, 1983 [1976], p. 16). The assumed opposition between
synchrony and diachrony may be a result of the editors’ work, and if it seems,
as it did to Trubetzkoy and Jakobson, forced and schematized, it is largely as

a result of the editorial presentation of the material. Levi-Strauss thus cau-
tiously replaces a direct reference to Saussure with one to “the editors of the
Course in General Linguistics.” It is for the editors that “an absolute opposi-
tion exists between two categories of facts: on the one hand, that of gram-
mar, the synchronic, the conscious; on the other hand, that of phonetics, the
diachronic, the unconscious” (ibid.). As he proceeds to make the case for
synchrony extending beyond the conscious grasp, Levi-Strauss reverts to the
direct reference to a “de Saussure” who “seems to deny the reality of a struc-
ture which is not immediately given” (ibid., p. 17), and thus in principle
forecloses the possibility of a grammatically organized system operating at
the level of the unconscious. A direct critique thus reassumes Saussure as the
single author bearing responsibility for the ideas contained in his book. It is
Saussure (and not the editors) that emerges as a thinker excessively commit-
ted to the primacy of consciousness in his emphasis on the availability of
synchronic systems to conscious awareness. That Saussure acknowledges
the role of the unconscious in language, and employs a complex understand-
ing of consciousness as a gradational spectrum extending into the uncon-
scious in the source materials becomes a moot point; it is the Course that
dictates the official Saussurean doctrine and silences any expression of dis-
senting ideas, even when they may come from Saussure himself.
Even though he concedes that the Course is largely a product of edito-
rial making, Levi-Strauss slips between references to Saussure, Bally, and
Sechehaye, and the Course itself (1983 [1976], p. 17). This floating refer-
ence helps to maintain the chain of substitutions between the (presumed or
implied) author of the Course, Ferdinand de Saussure, the editors (whose
authorial role is acknowledged in passing but without upsetting the pri-
mary reference to Saussure as the source point of ideas in the Course), and
the Course, a scripture of structuralism. This practice of respectfully nod-
ding to the primary texts from the Nachlass while remaining solely fo-
cused on the doctrine from the Book has become standard in scholarship;
it reflects the ideological functioning of the author in relation to the offi-
cially recognized work, as well as the institutional motive to secure a foun-
dation for intellectual movements.
*
Derrida’s reception of Saussure’s Course is interesting in its manifest ten-
sion between, on the one hand, a general critique of the “civilization of the
book,” and, on the other, a firm attachment to the book in the special case
of Saussure. The critique of the civilization of the book deconstructs our
common expectation that the book “with its ponderable shape and its
beginning, middle and end” can be mastered and possessed through
knowledge; it would unsettle the human expectation of a stable center and
foundation which the book seemingly provides (Derrida, 1998, p. xi).
These attachments and expectations both get displaced in favor of an end-
less, unbound text: “The text has no stable identity, no stable origin, no
stable end. Each act of reading the ‘text’ is a preface to the next” (ibid.).
Such stated openness in face of an infinite textual nexus is curiously at
odds with Derrida’s practice which assumes the book as sole receptacle of
Saussure’s ideas. Derrida’s extensive deconstruction of Saussure’s alleged
privileging of speech over writing, and his attachment to the sign (signa-
tum) as privileged locus of signification—this unexamined metaphysical
heritage maintained within a scientific project on general linguistics—is
captive to the canonical volume of the Course, and never ventures into the
open labyrinth of Saussure’s Nachlass. And yet the latter would respond
better to a deconstructive call for unbound textuality. The method of de-
construction is still wed to an identifiable deconstrandum, a condition sine
qua non of its possibility and usefulness. Derrida states:
I have chosen to demonstrate the necessity of this “deconstruction” by priv-

ileging the Saussurean references, not only because Saussure still domi-
nates contemporary linguistics and semiology; it is also because he seems
to me to hold himself at the limit: at the same time within the metaphysics
that must be deconstructed and beyond the concept of the sign (signifier/
signified) which he still uses. But Saussure’s scruples, his interminable
hesitations, particularly in the matter of the difference between the two
“aspects” of the sign and in the matter of “arbitrariness,” are better realized
through reading Robert Godel’s Les sources manuscrites du cours de lin-
guistique générale. (1957, p. 190ff)
Suffice it to say here that it is not impossible that the literality of the Course,
to which we have indeed had to refer, should one day appear very suspect
in the light of unpublished material now being prepared for publication.
(Derrida, 1998, p. 329, 38n, italics added)
Derrida acknowledges that the literality of the Course assumed in his

deconstruction of Saussure’s doctrine may one day become suspect. A loss
of literality could ultimately reveal that the doctrine does not really exist
and therefore does not need to be deconstructed, which would make Der-
rida’s reading obsolete. We find a second—this time strikingly different—
take on the question of literality further down in the note:

Up to what point is Saussure responsible for the Course as it was edited and
published after his death? It is not a new question. Need we specify that,
here at least, we cannot consider it to be pertinent? Unless my project has
been fundamentally misunderstood, it should be clear by now that, caring
very little about Ferdinand de Saussure’s very thought itself, I have inter-
ested myself in a text whose literality has played a well-known role since
1915, operating within a system of readings, influences, misunderstandings,
borrowings, refutations, etc. What I could read—and equally what I could
not read—under the title A Course in General Linguistics seemed impor-
tant to the point of excluding all hidden and “true” intentions of Ferdinand
de Saussure. If one were to discover that this text hid another text—and
there will never be anything but texts—and hid it in a determined sense, the
reading that I have just proposed would not be invalidated, at least for that
particular reason. Quite to the contrary. Besides, at the very end of their
first “Preface,” the editors of the Course foresee this situation. (Derrida,
1998, p. 329, 38n)
This response is forcing the reader’s hand—either Derrida’s adherence

to the literality of the Course goes unchallenged, or the questioning reader
simply reveals her prior commitment to metaphysics of presence, and fun-
damental misunderstanding of the project of deconstruction. Any potential
critic thus gets, provisionally at least, silenced, since any reservation gets
cast as a search for Saussure’s very thought itself (including Saussure’s re-
sponsibility for a book edited and published after his death). Framed in
this manner, any dissent is derided as soon as it is voiced, since it is as-
sumed to bear exclusively upon Saussure’s innermost and hidden authorial
intentions and his responsibility for personal remnants from after the
grave. Yet while such mentalistic attitudes can be found in the way the edi-
tors of the Course proceeded in their rendering of Saussure’s own, pre-
sumed, ideas—they should not be attributed by default to its readers. It is
the editors who claim in the preface to access Saussure’s intentions as if
they were their own. To claim to be able to anticipate how the reader takes
them up (as books subject to authorial control and ownership) is to assume
an ability to read the readers’ very thoughts themselves, including whether
or not they misunderstood deconstruction. The intended perlocutionary
effect of Derrida’s statements is this: close the debate, let the book Of
Grammatology have its final word on Saussure (and that in an endnote),
and so elevate this book above the grip of a deconstructive reading.
Yet the defensive tone of Derrida’s comments barely hides uneasiness
about privileging Saussurean references with a sole focus on the Course.
The uneasiness has to do with this: Derrida emerges as a dupe of the civi-
lization of the book. He is drawn to the ponderable unity of the volume on
general linguistics, but strangely not to the existing web of manuscripts,
lecture notes, and letters, that constitute the textual universe related to
Saussure’s general linguistics. Derrida prefers to deconstructively disman-
tle the book’s assumed unity, but stays clear of the text where deconstruc-
tion of the book is already underway. He chooses as sole reference a volume
that deconstruction (as far as I understand it) should find suspect, because
it imputes all hidden and “true” intentions to Ferdinand de Saussure re-
gardless of textual evidence. It is puzzling that Derrida spares the editors
of the Course this charge but directs it proactively against all potential and
future readers of his book.
The editors, he notes, foresee “this situation” at the end of their first
“Preface” anyway. But what the editors foresee at the end of the preface is
the charge of responsibility for publishing a text which makes it hard for
the critics to distinguish “between the master and [his] interpreters”; they
express readiness to “receive the blows rather than tarnish the one whose
memory is dear [to them]” (Saussure, 1986, p. xx, translation revised). This
readiness to receive the blows in lieu of the beloved master would have
been a grandiose gesture indeed if only Bally and Sechehaye had been in
the right place to assume the martyr’s responsibility for the master’s deeds.
To receive the blows, Bally and Sechehaye should have been the rightful
executors of Saussure’s estate, and the classic academic book format should
have best fit Saussure’s project—yet both claims are up for debate, and
cannot be simply assumed.
Neither the editors—nor Derrida—foresee an alternative development,
wherein the authority of the Course becomes displaced by the texts from
the Nachlass, such as the student lecture notes whose publication efforts
were effectively stifled by the editors themselves, or Saussure’s auto-
graphed sources that were held under lock and key in the library at the
University of Geneva. Neither the editors nor Derrida seem able to foresee
that the foreclosure of textual resources which was needed to foreground
the Course as the sole Saussurean doctrine could one day be reversed.
Like Bally and Sechehaye, Derrida is silent on this foreclosure; he magni-
fies the potential critics’ effort to locate Saussure’s true intentions, and
minimizes the whole textual universe on whose ruins the book was built.
But yet another foreseeable situation is this: the authority of the book gets
displaced by the text, and Saussure scholarship shifts focus from the
former to the latter. This situation, and an endorsement of the second edi-
torial paradigm, constitutes in fact the dominant trend in current Saussure

scholarship, even though it was foreseen by neither Bally and Sechehaye
nor Derrida. It goes beyond the initial efforts to offer bibliographical and
critical resources for reading the Course to an increasing focus on the
source materials themselves; this second wave in Saussure scholarship
may one day make the book obsolete, and marks therefore a more radical
and practical gesture of protesting the civilization of the book than its
perpetual mourning.
If one turns the edge of deconstruction back onto the text of Of Gram-
matology, one notes a tension between what it overtly says and what it ef-
fectively does. The text declares the end of the book and the beginning of
writing, but performs a philosophy of the book which has not yet begun the
textual reading and writing it advocates. It states the necessity of castrating
the book’s phallic authority but all from the safety of the shadow projected
by it, and without venturing into the criss-crossing maze of the text that
would, it seems, directly respond to a call for infinite textuality. Of Gram-
matology declares the death of the book but does not venture away from its
grave. There is then a similar tension in Of Grammatology between what
a text says and what it does as the one it attributes to other texts (by Saus-
sure, by Rousseau). Despite the defensiveness of Derrida’s comments, Of
Grammatology cannot simply deflect the possibility of a deconstructive
process turning back onto itself. Even though its author addresses the
reader from the grande hauteur of a master laying down the principles to
a disciple aspiring to “understand deconstruction,” the tension between the
constative and the illocutionary dimensions are present in the deconstruct-
ing as much as in the deconstructed text: Of Grammatology performs a
book-bound reading while declaring its end. In a delicious tension ripe for
deconstruction, Derrida practices a philosophy of the book while advocat-
ing the primacy of writing.
APPENDIX 1 | English translations of the Course
I
n this section, i will comment on how the two English-language editions and transla-
tions of the Course deploy various strategies to legitimize the book despite its compli-
cated material and institutional history. In the absence of a corrective usually provided
by a critical edition, the English-language editions of the Course bolster the common
association between Saussure’s general linguistics and the official doctrine in the Anglo-
phone scholarly world.
Saussure séance
Harris’s translation of the Course (which has until 2011 been the sole English-language
version of the Course easily available on the market) acknowledges the constructed
character of this redaction but seeks to minimize the controversy, and ultimately to dis-
miss critical scholarship as a set of crude complaints with little research value.
Harris writes in the introduction:
When Saussure died in 1913, he left no manuscript setting out his theories in
detail. What was published three years later as the Cours de linguistique gé-
nérale was put together by his colleagues, mainly from lecture notes taken by
his pupils. The notes in question have now—belatedly—been published in full
by R. Engler in his critical edition of the text (1967–74). On the evidence of this
material, it has sometimes been suggested that by no means all the ideas in the
Cours de linguistique générale are a faithful reflection of Saussure’s. (Saussure,
1986, p. xii)1
What should the reader make of the comment that “by no means all the ideas in
the Cours de linguistique générale are a faithful reflection of Saussure’s”? The
1
This introduction has been reprinted as: “Introduction to the Bloomsbury Revelations Edition”
in the 2013 reissue of Harris’s original translation by Bloomsbury (pp. xiv–xxii). A new
introduction (pp. xxix–xxxix), which briefly discusses the source materials of the Course, has
been appended to the Bloomsbury edition.
suggestion has a ring of anonymity to it, which is surprising considering that some of
the leading European scholars in historical linguistics, like Robert Godel, Rudolf
Engler, Tullio de Mauro, Simon Bouquet, and others are being referenced here. Their
groundbreaking research is turned into a “sometimes made suggestion,” as if it were
an offhand comment made over dinner. This research is therefore effectively silenced
as soon as it is mentioned, considering especially that the Anglophone reader is un-
likely if at all able to trace the reference to the relevant works, that most of contem-
porary Saussure scholarship is largely unknown outside of general and philosophical
circles, rarely mentioned in the main English-language scholarly publications on
Saussure, and not usually translated into the English language. The book fallacy
continues therefore to run its course in the Anglophone Saussure scholarship, and the
English language translator can continue to regard his translation of the Course as
primary reference (but see Harris’s Reading Saussure for an informed and critical
perspective).
Having dismissed the non-English-language Saussure scholarship, the above citation
presents its charge that “by no means all the ideas in the Cours de linguistique générale
are a faithful reflection of Saussure’s.” By no means all the ideas? A curious phrase. It
could mean either that “not all” the ideas in the book are a faithful reflection of the ideas
Saussure himself had, or that such a faithful reflection is precluded, and hence it is “by
no means” possible to faithfully reflect the ideas Saussure himself had in a book bearing
his name. Such de jure impossibility of one-by-one correspondence between the ideas in
the book and in Saussure’s mind is easily granted, considering that the reader would be
hard pressed to access the latter and lay them side by side with the former for the pur-
pose of comparison and contrast. Rather than voicing the de jure impossibility of such a
naïve correspondence between authorial intention and textual evidence (which would be
laudable in principle, but hardly innovatory), the passage steers in the direction of a de
facto impossibility, suggesting that an exact reflection of Saussure’s ideas in the book is
“by no means” possible anyway. The professed worry that some ideas in the book are
unlike some ideas the author himself had then carries little weight. The principal heuris-
tic strategy advocated throughout remains that of getting as close as possible to the au-
thor’s own signifying intentions, understanding clearly what he meant to say—a telepa-
thy of sorts, where reading is not unlike a séance and the book a medium summoning
the ghosts.
The translator continues:
It is . . . a somewhat crude critical procedure constantly to compare the pub-

lished text of the Course with the available notes, and complain that the editors
have misrepresented Saussure every time a discrepancy is found. There may be
discrepancies both of detail and of arrangement. But what they prove is another
matter. If we take the published text as a whole, there is no convincing reason
for supposing that it seriously misrepresents the kind of synthesis towards
which Saussure himself was working when he died. That synthesis is necessar-
ily hypothetical, a projection of what might have happened had Saussure lived.
But if its validity is questioned on quite basic points, then we are driven to one
or the other of two equally unlikely conclusions. Either Saussure’s closest col-
leagues and sympathizers were not able fully to understand his thinking on
258 Appendix 1
linguistic topics; or else Saussure himself had inadvertently misled them, while
at the same time managing not to mislead his pupils. (Saussure, 1986, pp.
xii–xiii)
Which one shall it be: a charge of a feeble mind on the part of the editors, or a charge
of selective (albeit inadvertent) manipulation on Saussure’s own part? Both charges are
figured as unlikely, and couched in the form of a rhetorical question which implicitly
assumes the opposite of what it overtly entertains. Clearly, Bally and Sechehaye were
both versed in linguistics, and so able to understand Saussure (even though their “being
able to understand his thinking fully” cannot be settled in any usual manner); the edi-
tors would not have likely been “inadvertently misled” by Saussure, whatever that may
mean. The specific charges are therefore presented in a manner that removes any trace
of their plausibility; they convey also the familiar heuristic strategy of a ghost-sum-
moning kind. In an uncritical endorsement of the editors’ professed methods, the trans-
lator assumes their task consisted in restoring Saussure’s thinking on linguistic topics
as faithfully as a full understanding allows, barring any advertent or inadvertent con-
ceit on the linguist’s part. The scholars who “constantly compare” the extant texts, that
is, the book of the Course on the one hand, and the Nachlass on the other, are said to
be “complaining” about the discrepancies between them, as if their extensive critical
work were not unlike a dinner conversation where the suggestion is made that the food
on the table does not compare to the one served last night. The scholars are then called
to order in the name of a curious notion, “the kind of synthesis towards which Saussure
himself was working when he died.” Is the claim being made that Saussure was work-
ing (on a hypothesis, no less) at the time of his death? Is the implied referent of “Sau-
ssure” a living-dead specter, still working in the afterlife? Speaking through the lips of
the living?
Let’s read on. Saussure’s synthesis is deemed “a projection of what might have hap-
pened had Saussure lived.” A counterfactual projection of syntheses in the past is shaky
evidence, especially when working with texts. Yet this evidence gets projected onto the
reader by a force of a per absurdum argument: if you do not accept this projection, you
will either have to accuse the editors of the Course of knowing little about linguistics, or
accuse Saussure of misleading the editors ex post mortem. Since neither of the forced
choices is appealing, the reader is led into the corner where a Saussure séance is being
held. The reader cannot but accept that the writers and translators of the Course com-
municated with Saussure’s ghost directly.
Yet what if the séance goes awry? Familiar with Goethe’s Der Zauberlehrling, trans-
lated and popularized in English as the Sorcerer’s Apprentice, the reader may waver
against calling powerful spirits. As the poem has it, an apprentice may not easily rid
himself of the magic powers he carelessly summoned by usurping the master’s own
words in his absence. Consider the apprentice’s cries as the magic he deployed without
master’s permission gets out of hand:
Stop now, hear me!

Ample measure
Of your treasure
We have gotten!
Appendix 1 259
Ah, I see it, dear me, dear me.
Master’s word I have forgotten!
cries the apprentice as rising waters threaten to drown the house.
Upon the master’s eventual return, the apprentice acknowledges:
Sir, my need is sore.

Spirits that I’ve cited
My commands ignore.
According to this cautionary tale, the apprentice is well advised to refrain from citing
the spirits and wielding the master’s word in his absence. Yet unlike the presumed
master dictating the narrative of the Course in General Linguistics, the master invoked
in Sorcerer’s Apprentice manages to return before it is too late, and remedies the damage
being done while he was gone.
Saussure and Santa

The other recent edition of the English-language translation of the Course on General
Linguistics (Columbia UP, 2011) seeks to acknowledge the developments in critical
scholarship dealing with Saussure’s linguistics while rehabilitating the validity of the
book. For such a synthesis between critical and conservative stances to be possible one
would need to integrate the first and the second editorial paradigms, whereas most con-
temporary Saussure scholars are moving away from the former toward the latter. Fur-
thermore, the possibility of such a synthesis uncritically assumes the ideological privi-
lege associated with Saussure, the implied author of the Course, and rehabilitates
legendary tales about Saussure in disregard of empirical evidence.
Recall that according to Foucault, the author function includes two interrelated but
distinguishable elements: empirical verification and ideological force. In the case of
scholarly texts, one typically assumes that the existence of an individual or a group of
individuals that wrote the text in question is historically demonstrable. This rather naïve
empirical postulate is closely bound up with the ideological force of authorship, in that
references to the author are routinely deployed to stamp some texts as canonical works,
and to exclude or marginalize other texts. There are instances, however, when the author
exercises an ideological function independently of empirical constraints. In the case of
legendary tales about a figure of distinction who has acquired a heroic and quasimythi-
cal status the tale is largely immune to historical constraints; it is an element of the lore
shared by a given people, and its abiding power is bound up with ongoing ritual prac-
tices, such as the retelling of the story, regardless of its empirically verifiable status.
In the case of legendary tales, it is possible but by no means necessary that the histor-
ical details of the figure’s life be known, and even if they are, they have no direct bearing
on the legendary figure itself. For example, a firm believer in Santa Claus need not know
that the legend is tied to the historical figure of Saint Nicholas of Myra (270–343) who
had a reputation for secret gift giving,2 and it would matter just as little if Santa turned
2
Charles W. Jones, Saint Nikolaos of Myra, Bari, and Manhattan: Biography of a Legend
(Chicago: University of Chicago Press, 1978).
260 Appendix 1
out to not be so tied, or only very tenuously. Thus, even though it makes sense to say that
there is a “one and the same” being who can be described as both a historical figure and
a popular personage who annually performs his assigned duties at Christmas, we main-
tain a distinction between the historical and fictional characters, as made apparent by the
fact that those who knew Saint Nicholas of Myra would not be said to have met Santa
Claus in person. It belongs to the essence of Santa that he is a legendary creature who
travels by sled across the arc of the sky in the company of elves. We tell stories about
Santa which are largely unconstrained by historical evidence relative to Saint Nicholas
of Myra. If a document written by Saint Nicholas of Myra had emerged—not an unlikely
supposition considering that Nicholas is believed to have served as a reader at liturgy,
that is, read aloud excerpts from the scripture, and was thus endowed with the rare gift
of literacy—in which he explicitly condemned excessive attachment to material goods
and other practices which can best be described in contemporary terms as commodity
fetishism and capitalist consumerism, it is very unlikely that the current set of cultural
practices, including the expectation that Santa will lay the latest iPad model under the
tree, would be terminated as a result. These cultural practices and expectations may one
day reach an eclipse, but it will not be as a result of factual revelations and scholarship
in the material history related to the saint.
As the Santa example illustrates, legendary narratives about a figure of distinction
can exist independently from historical ones. Such narratives are not confined to the
common lore but permeate also scholarly constructions and perceptions of certain his-
torical figures as quasimythical heroes, described as, for example, “the greatest living
philosopher,” “the best known representatives of a movement,” “the founding father of a
given intellectual tradition,” and other heroic epithets. The canonical history of Western
philosophy teems with similar references, and it is not surprising that the heroic tale of
Ferdinand de Saussure, the founding father of structuralism in philosophy and beyond,
continues to be told, heeded, and disseminated in the classroom and at conferences. In
Saussure’s case, the heroic role has been successfully exercised in relative disconnect
from empirical evidence, largely as an ideological function tied to the status of the
Course as Saussure’s work. The editors of the most recent English-language presentation
of the Course (a reissue of the 1959 translation by Wade Baskin, Columbia UP, 2011, ed.
Perry Meisel and Haun Saussy) continue to adhere to this legendary and ideological
construct.
In the introduction, the editors acknowledge the complexity of the editorial situation
and discuss the many divergences between the source materials and the received version
of Saussure’s linguistics in some detail. They opt to synthesize all the textual material,
regardless of its historical authenticity. They declare, “Among the many possible Sau-
ssures, the Saussure some readers have long desired to meet, the conjectural ‘real Saus-
sure,’ the Saussure who understands Saussure emerges hesitantly from the various notes
and the manuscript materials. A reprinting of the 1959 translation of the 1916 Course
cannot fail to evoke him” (2011, p. xxxiii). But this last claim—that a reprinting of an
English-language translation of a posthumous redaction “evokes” real Saussure only
begs what is, at best, in question, namely that the construct retains validity in light of
recent advancements in Saussure’s linguistics. In line with its allegiance to a legendary
Saussure, the above-cited passage also invokes a semidirect relation between the scholar
and the reader (“the Saussure some readers have long desired to meet”), not unlike that
Appendix 1 261
of a child around Christmas time, hopeful to soon be able to sit on Santa’s lap. The edi-
tors have a similarly cozy relation with the linguist: they claim to be reconstructing
“what Saussure has to say” by disentangling the correct understanding from the misun-
derstanding of the source materials.3 The spectral channel to Saussure himself initially
claimed by the first two editors of the Course in the 1910s is now extended to the two
new editors almost a century removed in time and an ocean in space.
Little if any room for scholarly conversation about the new materials is therefore left
open; the authentic sources need to be subsumed, so it seems, under a new kind of syn-
thesis, with the unchanged official doctrine hitherto associated exclusively with the
Course now needing to accommodate the source materials from the Nachlass as well.
The source materials are thus acknowledged only so long as they can fit with what we
already know about Saussure’s linguistics, but what to do if and when they don’t? They
risk being excluded and marginalized yet again, in another bout of selective reading ac-
cording to an assumed paradigm of linguistic science.4 Furthermore, the preferred edi-
torial strategy of referring indiscriminately to the material from the Course and the
source materials (e.g., pp. xviii, xxvi) can only create its own set of paradoxes, and would
call for a new body of critical literature to establish the validity of this editorial vision of
“what Saussure has to say.”
The editors of the 2011 reissue of the Course opt to preserve the Saussure legend and
patch it with pieces culled from Saussure’s own corpus, instead of following the lead of
the recent Saussure scholarship which advocates the second editorial paradigm, that is
abandoning the Course in favor of a sole focus on the Nachlass. By doing so, they priv-
ilege the legendary narrative about Saussure over the recently discovered writings by
Saussure himself, and disregard empirical criteria for establishing historical authorship.
Meisel and Saussy concede:
Though the original editors were convinced that their work was a reverent recon-
struction, later manuscript discoveries have shown that they are responsible for a
number of Saussurean myths. Thus Saussure the Semiologist, like a medieval
saint or hero, survives in a legend compiled from diverse written sources record-
ing a vanished oral tradition. (Saussure, 2011, p. xxii)
The question, as they rightly note, is then why put yet another edition of the Course,
“the home of this legendary Saussure” on the market:
Why republish it? Why not publish a corrected edition, one taking into account as
far as possible the differences between what Bally, Sechehaye, and Riedlinger
wrote in Saussure’s name and what he said or wrote himself? (ibid., p. xxii)
3
In their preface, the editors describe their task as being to “describe in detail the many versions
of the Course discovered in other sources since 1916, including the consequences each one has
for understanding, and misunderstanding, what Saussure has to say” (Saussure, 2011, p. ix,
emphasis added).
4
This is the case even though the editors emphasize that “Saussure is a philosopher” (p. xvi), who
seeks to find a solution to “a series of familiar, and presumably intractable, problems in the
history of ideas” (p. xx)—and distinguish their edition from the one by Harris the linguist, “little
interested in Saussure’s philosophical breakthroughs” (p. xviii).
262 Appendix 1
Meisel and Saussy do not consider the second editorial paradigm: it would make their
own editorial enterprise null and void. Instead, they respond to the charge stated above:
While we are now aware that Saussure is not identical with the presumed author
of the posthumous 1916 publication, it is the 1916 Saussure who has exerted the
immense influence on twentieth-century linguistics, literary study, and social sci-
ence . . . The legendary Saussure is the effective Saussure, the Saussure of record.
(ibid., p. xxii)
A reader familiar with the 1916 publication did not need to wait until 2011 to become
aware that Saussure is not its presumed author; that much was disclosed by Bally and
Sechehaye in their editorial preface to the book. What we are now aware of, thanks to
recent developments in Saussure scholarship, is that the legendary narratives associated
with Saussure are being revised and possibly relinquished in light of empirical evidence
from the historically authentic corpus.
The editors prefer to maintain the legend alongside the authentic texts due to the con-
sideration that “the Saussure known to Saussure specialists substitutes for the figure of
legend as little as Saint Nicholas of Myra can replace Santa Claus” (ibid., p. xxii).
Is it a good scholarly practice to treat Saussure on a par with Santa? Viewed as a leg-
endary figure, Saussure, like Santa, is largely unconstrained by historical evidence—
one can continue believing in one as one would in the other. The legendary Saussure,
like Santa, works mainly through his tangible effects, he offers a gift that keeps on
giving: a best-selling book. This legendary Saussure, fashioned by a posthumous con-
struction narrated according to the rules of a heroic tale, tells the story of how Saussure
laid the foundations for the science of general linguistics and structuralism. If Saussure
is like Santa, then the structuralist narrative will remain unrivaled in Saussure studies,
just like consumer capitalism maintains its hold on the contemporary version of Saint
Nicholas of Myra. But the Santa comparison makes light of the empirical standards of
authorship, wherein a historically verifiable corpus takes precedence over second- and
thirdhand reports, regardless of whether the latter have been ritually received and retold
as official doctrine for decades. It effectively forecloses the possibility of advancement
within the field of Saussure’s linguistics in relation to philosophy. Yet once the paradigm
shifts to a more scholarly one, then the historical corpus will begin to constrain and, in
case of conflict, override the legend.
Appendix 1 263
APPENDIX 2 | Saussure’s silence
I
t is unlikely that Saussure would have ever written a programmatic statement of
general linguistics comparable to the Course. His overall reticence to publish in the
academic format (while writing thousands upon thousands of unpublished manu-
script pages) strongly suggests the contrary. Yet the lamented absence of a beau livre
in general linguistics is not a simple failure, or an easily avoidable and rectifiable calam-
ity, and Saussure’s prolonged silence begins to speak in its own right if considered within
the context of his intellectual biography. There one finds evidence of a turmoil of intel-
lectual activity and uninterrupted writing, as well as an increasing sense of crisis in
linguistics construed as a positive science of language. Behind the official façade of in-
activity one finds a secret and difficult becoming of a philosopher who is increasingly
dissatisfied with the existing approaches and methods in scientific linguistics, and who
declares the need for—but also dreads—a radical reform.
Saussure’s publication record issued during his lifetime is surprisingly brief. It in-
cludes a list of relatively short and technical articles, many of which were the required
communications to the Société de linguistique de Paris regularly published in its Mé-
moires. Other pieces include invited contributions to diverse volumes of Mélanges Lin-
guistique, motivated by friendship and/or deference to the editors. Saussure published
just one book during his lifetime: the Mémoire sur les système primitif des voyelles dans
les langues indo-européennes, deemed by Meillet le plus beau livre of comparative
grammar ever written. The Mémoire was Saussure’s thèse universitaire, satisfying the
requirement for graduation from the University of Leipzig (1877). It was published in
December 1878, when Saussure was just twenty-one.
The book was audacious, and, according to Saussure’s autobiographical notes, more a
result of his own intellectual labors than his studies in Germany (Saussure, 1958, p. 15).
In it, Saussure famously proposes a hypothesis far ahead of its time: the existence of the
so-called sonant coefficients (les nasales voyelles), subsequently known as laryngeals, to
explain the development of long vowels in Proto-Indo-European. What is striking to a
linguist and a philosopher alike about the Mémoire is its method: Saussure bases his hy-
pothesis on an internal reconstruction of underlying forms from the observable relations
between the existent ones; this analysis is only possible in view of language as a rule-
governed system (Bouquet, 1997, p. 63, 1n). Such a systemic approach is an intellectual
construct, a philosophical thesis about essential traits of language (or what language nec-
essarily must be qua language), which is put in the service of interpreting and inferring
empirical data relative to specific languages. The Mémoire thus sets Saussure’s method
apart from that of his contemporaries:
Saussure began the task of defining languages as systems based on differential

and relational terms rather than on the basis of the material properties of their
phonetic substance. The Mémoire is significant for the break that it represents
with the atomistic and substance-based approaches of nineteenth-century com-
parative linguistics. (Thibault, 2005, p. 667)
Even though Saussure is not expressly reflecting on the reform in linguistic method-
ology that this early study effectively calls for, the latter can be viewed as a practical
expression of what later became articulated as a systemic theory of language with its
concurrent critique of the atomism and “the involuntary assumption of substance” made
in traditional linguistics. Saussure would then have applied a systemic approach to lan-
guage first, and brought it under a reflective focus later on.1
One finds an even more precocious expression of a systemic approach in the posthu-
mously published Essai pour réduire les mots du grec, du latin et de l’allemand a un
petit nombre de racines (found in the papers acquired by Harvard University, Cahiers
Ferdinand de Saussure, 32, 1978, pp. 73–101), autographed by Saussure before turning
fifteen. Dismissed by him later as an enfantillage (Saussure, 1958, p. 17), this piece in-
spired by Adolph Pictet’s Origines indo-européennes (studied by Saussure in his early
teens) proposes a theory that derives all existing languages from a structure constituted
by just three consonants (ibid., p. 17). Regardless of its credibility, the Essai testifies to
an early commitment to a systemic vision of language, pursued by the young Saussure in
relative solitude, like a private vision of the ultimate key to the manifest enigma of
language.
The Mémoire was followed by Saussure’s doctoral dissertation, De l’emploie de
génitif en Sanskrit, a short technical article, issued in Leipzig in February 1880. This
article was to be the last piece published by the author out of his own initiative. The
following shorter pieces, with the exception of two articles on declension and intonation
in Lithuanian (Indogermanische Forschungen) and the Phrygian inscriptions (Recher-
ches archéologiques de l’Asie centrale), were written out of an obligation (Bouquet,
1997, p. 65). As Meillet notes,
After the Mémoire, whose publication can be explained without a doubt by the
audacity of first youth, F. de Saussure did not believe to have sufficiently devel-
oped the theory of any linguistic fact that could be exposed to the public. He
wasn’t one of those who rush to publish their ideas before they have matured,
before having realized all their difficulties and before having devised a complete
1
For a detailed account of Saussure’s linguistic contributions in the Mémoire and the book’s
complicated reception see ch. 7 in Joseph, 2012, pp. 221–249.
266 Appendix 2
and coherent system. Overly concerned with having to produce definitive work,
he only broke the silence to publish rather short notes, often simple footnotes (bas
de pages) of the Mémoires of the Société de Linguistique. His last articles came
out without a doubt as a result of feeling obliged to contribute some pages to cer-
tain collections; the scruple with which he kept his word, the zealousness with
which he remained associated to these collective editions were touching for those
who knew with what repugnance he embarked on any publication. (1913, clxxi,
my translation)
In 1910, the scholar himself acknowledged, not without horror, having succumbed to
“thirty years of silence” since the publication of the Mémoire (1879). “Mes trente ans de
silence sont vraiment effrayants a sentir derrière moi en face de la haute distinction qui
semble vouloir récompenser mon labeur scientifique,” he writes in a letter to Louis Havet
after having been elected a corresponding member of l’Institut de France (Redard, 1976,
p. 348). Saussure is then in the grips of epistolophobia—a fear of writing compounded
with a disgust at having this fear (Jakobson, 1985, p. 222). The three-decade-long silence
reflects, however, only on the public life of Saussure, as perceived by the academic es-
tablishment. Behind this façade of inactivity, we find a private Saussure writing thou-
sands of manuscript pages on subjects ranging from methodological and conceptual
issues within general linguistics to problems in ancient Indian theosophy; Vedic litera-
ture and Hindu mythology; textual studies of Germanic legends; a quest after anagramic
inscriptions within Greek and Latin poetry and Latin prose; notes on speaking in tongues
by the famous medium Helen Smith. Saussure’s writings testify to a sustained reflection
on all things having to do with language, but either broadly construed and hence lying
outside of the province of scientific linguistics, or pursued as a critique and an envisaged
reform of the scientific linguistic discipline itself. The writings unpublished during
Saussure’s life are therefore contestatory of scientific linguistics in a double sense, by
either focusing on marginal or “external” phenomena of language, or by rethinking the
basic methods and categories of linguistic science.
One significant body of work is constituted by some 638 sheets and 995 pages of
manuscript pages stored in the Houghton Library of Harvard University (known as the
Harvard Manuscripts, they are catalogued as bMS Fr 266 (1)–(9); selected excerpts with
interpretative commentary from the collection have been published by Herman Parret as
Les manuscrits saussuriens de Harvard, in Cahiers Ferdinand de Saussure, 47, 1993).2
2
Thibault (2005) identifies the following major themes in the Harvard Manuscripts: (1) the study
of the functional roles of the Indo-European phonemes in relation to the opposition between
consonants and sonants, (2) the syllabic basis of articulation, (3) the concept of the phoneme as
“unité phonétique” (phonetic unit), (4) the form and substance of speech sounds, (5) the
combining of phonemes in parole, (6) the role of the voice and of the ear, (7) the physiological
and physical dimensions of speech sounds, (8) intention and will as agencies that modulate
speech sounds in parole, and (9) the diverse temporal spheres of parole. In the other manuscripts
in this collection, Saussure’s notes cover many diverse topics. These include (1) the Armenian kh
final, (2) the Sanskrit genitive, (3) the absolute genitive, (4) Vedic literature, (5) a discussion of a
book by Paul Oltramare (1907) on ancient Indian theosophy, (6) the Indo-European a, (7) Vedic
and Hindu mythology, (8) ancient Greek linguistics, and (9) a draft of Saussure’s doctoral thesis
(see also Parret, 1995).
Appendix 2 267
The Harvard Manuscripts are dated by Maria Pia Marchese to the period between 1881
and 1885 based on the authors cited by Saussure (Saussure, 1995, p. xiv). The 177 pages
of unpublished material catalogued as bMS Fr 266 (8), which bear the handwritten title
Phonétique (Phonetics), are generally believed to be notes and reflections for a treatise
on phonetics that Saussure never completed (Maria Pia Marchese published a critical
edition of the complete manuscripts catalogued as bMS Fr 266 (8) as: Phonétique: il
manoscritto di Harvard, 1995; see also Jakobson “Saussure’s Unpublished Reflexions
on Phonemes,” 1969). The materials from the planned treatise on phonetics form an im-
portant part of Saussure’s overall project to develop a theoretical view of language as a
system; they put forward a semiological phonetics, whereby sounds and successions of
sounds are identified in terms of their value for an idea (Saussure, 1995, p. 120), and not
as atomic parcels of physical sound. Saussure’s phonetics is concerned therefore with
how linguistic ideas are articulated in conjunction with the way in which the ear distin-
guishes between material sounds; this articulation, made at the level of perceived
speech, supposes therefore an access to an understanding of language as a system of
relational terms.
In addition to the Harvard collection, a substantial body of Saussure’s manuscripts is
preserved and catalogued in the Bibliothèque Publique et Universitaire in Geneva (see
Godel, 1957; “Inventaire des Manuscripts de F. de Saussure Remis à la Bibliothèque
Publique et Universitaire de Genève” [1960]). This collection includes Saussure’s exten-
sive notes on the Germanic legends of Niebelungslied, which are dated to a period
roughly from late 1903 to 1913, and possibly even until Saussure’s death in 1913. In the
notes, Saussure approaches the Germanic legends as systems of signs whose constitutive
elements take on a symbolic status as they became gradually removed from their histor-
ical source. The legends emerge therefore as textual sites of signification, irreducible to
the historical events they narrate. Saussure put some of this material in the service of his
teaching in 1904. Yet a planned book on the subject was never published in Saussure’s
own hand, even though he filled some eighteen notebooks with notes. (A representative
selection of the material has been made available to Saussure scholars in a critical edi-
tion of the notes, with editorial commentary, by Anna Marinetti and Marcello Meli
[Saussure, Ferdinand de, Le Leggende Germaniche: Scritti Scelti e Annotati (Padua,
Italy: Libreria Editrice Zielo–Este, 1986)].)
Yet another body of written work is constituted by Saussure’s quest to reconstitute the
anagrams or “the words upon words” in Saturnine Latin poetry. This project dating back
to 1906–1909 seeks to tease a second text out of the fragments isolated in the interior of
a poem, typically from the proper nouns. This research was extended to Greek and Latin
epic, lyric, and dramatic poetry, and Latin prose. A search for anagrams can be dis-
missed on empirical grounds as a self-fulfilling prophecy since it is prone to accumulate
evidence in favor of its hypothetical assumption as it progresses. And yet it raises a prop-
erly philosophical question of whether a hidden text, if it were to exist “upon the words,”
would be traceable to a deliberate choice on the composer’s part or a product of chance
(Starobinski, 1971, p. 119). Whatever one makes of empirical evidence in favor of the
anagrams, the theoretical distinction between intentional activity deployed by an indi-
vidual subject and processes inherent in language as such is valid and nontrivial; this
distinction centers on the role of individual consciousness and the unconscious in lan-
guage, which is also raised within Saussure’s project in general linguistics. Saussure
268 Appendix 2
filled close to 140 notebooks with reflections on the anagrams, without publishing any of
it. Jean Starobinski made this material available in an insightful study of the scientific
status of Saussure’s research on the anagrams, his doubts surrounding the entire process
and its eventual termination, in: Let mots sous les mots: les anagrammes de Ferdinand
de Saussure (Gallimard, 1971; translated as Words upon Words: The Anagrams of Fer-
dinand de Saussure, Yale UP, 1979).
Finally, of interest related to the “marginal” linguistic phenomena are Saussure’s
notes on glossolalia or speaking in tongues. Saussure is invoked repeatedly in Theodore
Flournoy’s Des Indes a la planète Mars (1983; From India to the Planet Mars: A Case
of Multiple Personality with Imaginary Languages, Princeton UP, 1994) as an expert of
Sanskrit, the language purportedly spoken by the clairvoyant under the adopted name
Helen Smith, who, when in trance, adopted the personality of a Princess Simandini.
Smith had a spiritual double, Leopold, who wrote down Princess Simandini’s messages
in this language. Saussure took part in the spiritualist séances featuring Mlle. Smith. He
sought to analyze the written messages, which he identified as “Sanskritoid”; his reports
are cited in Flournoy’s book.
Saussure was therefore anything but silent when regarded from within the perspec-
tive of his largely solitary pursuits; his solitude and separation from the establishment
could have been as deliberately contestatory as Nietzsche’s descent into the abyss before
conceiving a radical reform, a new philosophy of morals (see Bouquet, 1997, p. 67). In a
similar vein, Saussure may have been conceiving a radical reform, a new philosophy of
linguistics, in response to the stated crisis in the European sciences of dealing with lan-
guage as a positive fact. This is an immense task which calls for a renewed reflection
about the nature of language and appropriate methods of study; I’d like to consider that
it is the immensity of this task that accounts in part at least for Saussure’s reticence to
make his own work public during his lifetime. His silence should therefore be heard in
its proper resonance and not too hastily lamented as a loss and a lack to be filled.
This sense of crisis and need of a radical reform is expressed with force in a letter to
Antoine Meillet from January 4, 1894. The letter was written at the time (1893–1894)
when Saussure was working on a book on general linguistics, and it reveals the many
misgivings he felt about this project. Saussure concedes the magnitude of the task lying
ahead of him, as well as nostalgia for the historical studies only seemingly out of place
in linguistics. Having updated Meillet about the slow progress of his writings on Lithu-
anian intonation and accentuation, he professes both his disgust and the great difficulty
involved in writing “even ten lines of good sense (sens commun) on linguistic matters”
(Benveniste, “Lettres De Ferdinand De Saussure À Antoine Meillet,” 1964, CFS: 21, 95).
It seems like everything needs to be started afresh, a new basis laid before a single line
on a specific linguistic problem can be written.
As he pursues: “Preoccupied as I have been for a long time above all with the logical
classification of linguistic facts and with the classification of the points of view from
which we treat them, I realize more and more the immense amount of work that would
be required to both show the linguist what he is doing, by reducing each process to a
category laid in advance for it, as well as the conceit (l’assez grande vanité) of what one
can effectively do in linguistics” (ibid., p. 95, my translation).
The task then is double: to expose the unexamined assumptions about the subject
matter and the method which effectively guide the linguist in his work—the assumptions
Appendix 2 269
which Saussure will trace to the enduring influence of categories from substance meta-
physics and naïve realism in science, and the resulting conceit reigning in linguistics.
Saussure decries “the absolute ineptitude of the current terminology” and points to “the
need of a reform, and, for this purpose, of showing what kind of an object language in
general is.” This professed need does not figure, however, as an inspiration for future
research but rather as a dread: it would have spoiled Saussure’s pleasure in working with
languages, and forced him to work with language in general against his dearest wish.
The crisis in linguistics would thus lead him to write a book “despite himself” (malgré
moi), a book where he would have to explain, with neither passion nor enthusiasm, why
there isn’t a single term in linguistics to which he could attribute any meaning at all
(ibid., p. 95). Note that this reflective effort to glean the nature or essence of the object
being investigated was a guiding question for Saussure already in the Mémoire. It is a
question of “the very foundations of the subject (données élémentaires), without which
everything remains unanchored, arbitrary and uncertain” (Mémoire, 2009 [1879], p. 1).
The question of the foundations does not, he emphasizes, belong to the transcendent
order of pure speculations (ibid.); it is a question that scientists and philosophers must
ask in unison if any headway is to be made relative to questions about the being of lan-
guage, and how best to approach it.
Saussure mentioned a need for a book reflecting on the basic categories used in lin-
guistics in November 1891, in the third lecture at the University of Geneva (2006, pp.
[163–173], 109–116). He wrote: “One day, a very special and interesting book will be
written on the role of the word as the main element of distortion in the science of words”
(ibid., p. [166], 111). The problem diagnosed in his observation that the word is an ele-
ment of distortion will be later developed as the dominant philosophical idea that lan-
guage is a species of nomenclature made up of names (nomen); this linguistic idea is
bound up with a metaphysical commitment to substance as a basic unit of reality, and
implied direct referent of a name (what Saussure describes as “an involuntary assump-
tion of substance” in linguistics). Saussure put his energies into this planned book around
1893–1894, as documented by the “Notes for a book on general linguistics” (Writings,
2006, pp. [197–203], 136–140), over a decade prior to the course on general linguistics
at the University of Geneva (1907–1911). As he told a student, M. L. Gautier, on May 6,
1911, it was before the 1900s that he was especially preoccupied with questions relating
to general linguistics and philosophy of language; he did not expand on these reflections
later, and expressed misgivings about teaching such complex material as part of a stand-
ard university course, considering especially that he was ridden with many doubts.3 This
admission lends substantial weight to the unpublished manuscript notes as a principal
site of Saussure’s reflections on general and philosophical issues related to language. The
planned book project was definitively abandoned by the time Saussure taught the course.
As he shared with another student, A. Riedlinger, on January 19, 1909, “As for a book on
this subject, one cannot consider it: a book has to offer the definitive thought of its
3
Je me trouve placé devant un dilemme: ou bien exposer le sujet dans toute sa complexité et
avouer tous mes doutes, ce qui ne peut convenir pour un cours qui doit être matière à examen. Ou
bien faire quelque chose de simplifié, mieux adapté à un auditoire d’étudiants qui ne sont pas
linguistes. Mais à chaque pas, je me trouve arrêté par des scrupules. Pour aboutir, il me faudrait
des mois de méditation exclusive. (Godel, 1957, p. 30).
270 Appendix 2
author” (Godel, 1957, p. 30). Saussure apparently did not consider the traditional book
format appropriate for the unfinished, if not infinite, process he was engaged in.
Saussure’s sense of a crisis in linguistics may explain the painstakingly slow progress
of his publications at the time, that is, the two essays on Lithuanian intonation and ac-
centuation. Saussure read the first one to the Société Linguistique de Paris in 1889; the
paper was well received, and Saussure hoped that his study of Lithuanian accentuation
would eventually provide independent empirical evidence that the vowel system hypoth-
esized in the Mèmoire was correct. He therefore declined to publish the paper immedi-
ately in the journal of the Société, and planned to develop it further in a book-length
manuscript. As the notebooks from around that time show, the projected book would
have dealt with issues relative to general linguistics broadly construed, and included dis-
cussion of terms like opposition, value, and difference, which are key to the theoretical
account of language from the Course (Joseph, 2009, pp. 188–189). Saussure did not con-
sider this work ready for publication, however, and eventually set the hundreds of pages
of notes and drafts aside to embark on a project related to general theoretical issues in
linguistics: synchrony and diachrony, their relation to phonetics and morphology, and
the nature of linguistic signs (“Double Essence Manuscripts,” Writings, 2006, pp. [17–
88], 11–88). Apparently Saussure deemed the latter project more promising as far as
meeting the publication requirements relative to securing an “ordinary” or permanent
post at the U. of Geneva, which included a book or a series of articles with high impact
(Joseph, 2009, p. 189). However, the resulting pages of draft could not be brought to sat-
isfactory completion either.
This protracted history of repeated editorial miscarriages—a failure to produce pol-
ished pieces which would be palatable to the linguistic establishment—is productive
insofar as it demonstrates or rather enacts the crisis of the linguistic science itself, the
impossibility of developing a novel approach to a linguistic problem (such as Lithuanian
accentuation and intonation) within the confines of traditional theory. The impossibility
of developing this novel approach would result from the meaninglessness and logical
incoherence pervading the basic terms in linguistics (Benveniste, 1964, pp. 95–96).
Saussure’s silence is therefore a direct expression of the impossibility of breaking new
ground so long as the dominant categories dictate what counts as real in language and
how it should be investigated. Hence the practical and embarrassing consequences: de-
spite the fact that the essay on Lithuanian read to the Société in 1889 presented no “ma-
terial difficulties,” Saussure “dragged” (fait traîner), on his own admission, its publica-
tion for over a year (and then sent off to the editor, Louis Duvau, the original draft from
1889 plus one paragraph on a minor point and an announcement that a second part would
follow (cited in Joseph, 2009, p. 189); this piece would have seen the light of day “with-
out succeeding to avoid logically odious expressions, because a decidedly radical reform
would be needed for that purpose” (Benveniste, 1964, p. 96). The pieces that were even-
tually published were thus written decidedly “malgré moi”—not only despite but even
against oneself. The announced second piece was never published, even though, as Saus-
sure informed Meillet in a letter from September 23, 1907, it was not only written but
also largely typeset (ibid., p. 108). The promised link between Lithuanian accentuation
and the controversial vowel system from the Mémoire, which could have gathered em-
pirical evidence from a living European language for the validity of the historical recon-
struction of the Indo-European language system, was therefore never publicly made
Appendix 2 271
(Joseph, 2009, p. 198)—an absence that only deepened the thirty-year silence following
the publication of the Mémoire.4
Saussure emerges therefore as a scientist who enacts a crisis of his discipline (and the
end of the book?) with his life of seclusion. According to Agamben,
Saussure represents . . . the precious instance of a philologist who, caught in the

net of language, felt, as Nietzsche did, the insufficiency of philology, and who had
to become a philosopher or succumb. Saussure did not abandon linguistic study as
Nietzsche had done, but, closing himself for thirty years in a silence that appeared
inexplicable to many, interrupted only by the publication of mélanges of brief
technical notes . . . pursued to the limit an exemplary instance of the impossibility
of a science of language within the western metaphysical tradition. (1993, pp.
152–153, emphasis added)
Saussure’s silence would then carry within it an echo of the secret becoming of a
philosopher of language behind the scenes of scientific research in linguistics. Saussure
did not succumb in the face of the crisis of the science of linguistics, even though he may
have enacted “the impossibility of a science of language within the western metaphysi-
cal tradition” by his manifest impasse. Through a disengagement with the established
forums and formats of scientific linguistics, Saussure would have attempted a reduction
of Western metaphysics and the sciences based upon it, in view of testing their validity
and ultimately developing—in outline at least—an alternative metaphysics and method-
ology. The impasse is then productive insofar as it opens up a space for critical distanc-
ing from the traditional model of science and for a renewed reflection on both the subject
matter and the requisite methods in a regrounded study of language. Such reflection
bears on the very foundations of science, and as such it must summon philosophical
forces. Its task will be to respond to the metaphysician’s question: what is language, if
not an object modeled on the traditional notion of substance? And to the methodologist’s
question: what approaches, other than the natural scientific approach, can best accom-
modate the kind of being that language is?
4
The essay in question (“The Accentuation of the Lithuanian Language”) was read to the Tenth
International Congress of the Orientalists, held at the University of Geneva, on September 8,
1894. It was in fact the only international congress in which Saussure read a paper, since all the
other papers were delivered to the Société in Paris (Joseph, 2009, p. 182). The paper laid out a
new linguistic law, later termed Saussure’s law (ibid., p. 191). An essay spelling out this law was
published, together with another piece on Lithuanian, in a prestigious journal: Indogermanische
Forschungen. The first came out in 1894, the second, stating Saussure’s Law, in 1896. These are
the only two articles Saussure published in a scholarly journal, other than the Société’s. The two
pieces were published in time to secure a permanent position at the U. of Geneva for Saussure on
October 23, 1896.
272 Appendix 2
BIBLIOGRAPHY
Agamben, Giorgio. “The Barrier and the Fold” in Stanzas: Word and Phantasm in West-
ern Culture. Minneapolis: U of Minnesota P, 1993.
Amacker, René. “Correspondance Bally-Meillet (1906–1932).” Cahiers Ferdinand de
Saussure 43 (1989): 95–127.
Amacker, René. “Le Developpement des idées saussuriennes chez Charles Bally et
Albert Sechehaye.” Historiographia linguistica 27.3 (2000): 205–264.
Amsterdamska, Olga. Schools of Thought: The Development of Linguistics from Bopp
to Saussure. Dordrecht: D. Reidel, 1987.
Anderson, Stephen R. Phonology in the Twentieth Century: Theories of Rules and Theo-
ries of Representations. Chicago: U of Chicago P, 1985.
Aristotle. De Interpretatione. The Complete Works of Aristotle. Rev. Oxford Transla-
tion. Ed. J. Barnes. Vol. 1. Princeton: Princeton UP, 1995.
Arnauld, Antoine, and Claude Lancelot. General and Rational Grammar: The Port-
Royal Grammar. Ed. and trans. Jacques Rieux and Bernard E. Rollin. The Hague:
Mouton, 1975.
Augustine. Confessions. Trans. R. S. Pine-Coffin. London: Penguin Classics, 1961.
Bally, Charles. “Maître et disciples.” Journal de Genève 18 (1908): 32–33.
Bally, Charles. Traité de stylistique française. Vol. 1. Paris: Klincksieck, 1921 [1909].
Bally, Charles. Mélanges de linguistique offerts à Charles Bally. Geneva: Georg, 1939.
Bally, Charles. Le Langage et la vie. Vol. 34. Geneva: Droz, 1965 [1925]. [A modified
version of the 1913 essay under the same title appears in this anthology].
Bally, Charles. Linguistique générale et linguistique française. Bern: A. Francke, 1965a
[1932].
Bally, Charles. Unveroffentlichte Schriften/Comptes rendus et essais inedits. Bonn: Ro-
manist Verlag, 1988.
Barthes, Roland. Critical Essays. Evanston: Northwestern UP, 1972.
Beauvoir, Simone de. “Review of The Elementary Structures of Kinship.” Paris: Les
Temps Modernes, 1949.
Beauvoir, Simone de. Le Deuxième Sexe. Paris: Gallimard, 1986.
Beauvoir, Simone de. The Second Sex. Trans. Constance Borde and Sheila Malovany-
Chevallier. New York: Alfred A. Knopf, 2010.
Benveniste, Emile. “Lettres de Ferdinand de Saussure à Antoine Meillet.” Cahiers Fer-
dinand De Saussure 21 (1964): 93–130.
Benveniste, Emile. Problèmes de linguistique générale. Vol. 2. Paris: Gallimard, 1966.
Benveniste, Emile. Saussure after Half a Century. Trans. Mary E. Meek. Miami: U of
Miami P, 1971.
Bouissac, Paul. Saussure: A Guide for the Perplexed. London: Continuum, 2010.
Bouquet, Simon. Introduction à la lecture de Saussure. Paris: Editions Payot & Rivages,
1997.
Bouquet, Simon. “La Linguistique générale de Ferdinand de Saussure: textes et retour
aux textes.” Paper presented at Congrès ICHOLS, September 1999. Available online
at http://www.revue-texto.net/Saussure/Sur_Saussure/Bouquet_Linguist-gen.html.
Bouquet, Simon. “Les Deux Paradigmes éditoriaux de la linguistique générale de Ferdi-
nand de Saussure.” Cahiers Ferdinand de Saussure 51 (1998): 187–202.
Bouquet, Simon. “Saussure après un siècle.” Cahier de l’Herne: Ferdinand de Saussure.
Ed. Simon Bouquet. Paris: Editions de l’Herne (2003): 11–15.
Bouquet, Simon, ed. Cahier de l’Herne: Ferdinand de Saussure. Paris: Editions de
l’Herne, 2003.
Broendal, Viggo, and Louis Hjelmslev. Acta Linguistica. Copenhagen: Munksgaard,
1939.
Caton, Steven C. “Contributions of Roman Jakobson.” Annual Review of Anthropology
16 (1987): 223–260.
Chomsky, Noam. Aspects of the Theory of Syntax. Cambridge: MIT Press, 1965.
Chomsky, Noam. The Logical Structure of Linguistic Theory. New York: Plenum, 1975.
Culler, Jonathan. Ferdinand de Saussure. Rev ed. Ithaca: Cornell UP, 1986.
Culler, Jonathan. “L’Essentiel de l’arbitraire.” Cahier de l’Herne: Ferdinand de Saus-
sure. Ed. Simon Bouquet. Paris: Editions de l’Herne (2003): 52–61.
Culler, Jonathan, ed. Structuralism: Critical Concepts in Literary and Cultural Studies.
Vol. 4. London: Routledge, 2006.
Culler, Jonathan. “The Sign: Saussure and Derrida on Arbitrariness” in The Literary in
Theory. Stanford: Stanford UP (2007): 117–136.
de Mauro, Tullio. Ludwig Wittgenstein: His Place in the Development of Semantics. Vol.
3. Dordrecht: D. Reidel, 1967.
Derrida, Jacques. Writing and Difference. Trans. Alan Bass. Chicago: U of Chicago P,
1978.
Derrida, Jacques. Positions. Chicago: U of Chicago P, 1981.
Derrida, Jacques. Glas. Trans. John P. Leavey, Jr., and Richard Rand. Lincoln: U of Ne-
braska P, 1986.
Derrida, Jacques. Specters of Marx: The State of the Debt, the Work of Mourning, and
the New International. Trans. Peggy Kamuf. New York: Routledge, 1994.
Derrida, Jacques. Of Grammatology. Trans. Gayatri Chakravorty Spivak. Baltimore:
Johns Hopkins UP, 1998.
Derrida, Jacques. “The Linguistic Circle of Geneva” in Margins of Philosophy. Trans.
Alan Bass. Chicago: U of Chicago P (2009): 137–154.
Dosse, François. History of Structuralism. Minneapolis: U of Minnesota P, 1997.
Engler, Rudolf. “Charles Bally, Kritiker Saussure’s?” Cahiers Ferdinand de Saussure 41
(1987): 55–63.
274 Bibliography
Flournoy, Théodore, Marina Yaguello, and Mireille Cifali. Des Indes à la planète Mars.
Paris: Editions du Seuil, 1983.
Flournoy, Théodore, Mireille Cifali, and Sonu Shamdasani. From India to the Planet
Mars: A Case of Multiple Personality with Imaginary Languages. Princeton: Princ-
eton UP, 1994.
Foucault, Michel. “What Is an Author?” Contributions in Philosophy 83 (2001): 9–22.
Frank, Manfred. What Is Neostructuralism? Minneapolis: U of Minnesota P, 1989.
Freud, Sigmund. The Interpretation of Dreams. Trans. James Strachey. New York: Basic
Books, 1955.
Frýba-Reber, Anne-Marguerite. Albert Sechehaye et la syntaxe imaginative: contribu-
tion à l’histoire de la linguistique saussurienne. Geneva: Droz, 1994.
Gadet, Françoise. Saussure and Contemporary Culture. London: Hutchinson Radius,
1989.
Gadet, Francoise. “Le Signe et le sens.” DRLAV, Revue de linguistique 40 (1989a).
Gambarara, Daniele. “La bibliothèque de Ferdinand de Saussure.” Genava 20 (1972):
319–368.
Gasparov, Boris. Beyond Pure Reason: Ferdinand de Saussure’s Philosophy of Lan-
guage and Its Early Romantic Antecedents. New York: Columbia UP, 2013.
Godel, Robert. Les Sources manuscrites du Cours de linguistique générale. Geneva:
Droz, 1957.
Goethe, Johann W. The Sorcerer’s Apprentice. Livingston: Officina, 1993.
Grégoire, A. “Compte rendu du ‘Recueil des publications scientifiques de Ferdinand de
Saussure.’” Originally published in Revue belge de Philosophie et d’Histoire 2.3
(1923). Reprinted in George Mounin, Saussure ou le structuraliste sans le savoir
(Paris: Seghers, 1968): 176–178).
Harris, Roy. Saussure and His Interpreters. Edinburgh: Edinburgh UP, 2003.
Hegel, Georg W. F. Phenomenology of Spirit. Trans. A. V. Miller and J. N. Findlay.
Oxford: Clarendon, 1977.
Hegel, Georg W. F. The Encyclopedia of Logic. Trans. T. F. Geraets, W. A. Suchting, and
H. S. Harris. Indianapolis: Hackett, 1991.
Hegel, Georg W. F. Philosophy of Right. Trans. S. W. Dyde. New York: Cosimo, 2008.
Hegel, Georg W. F. The Science of Logic. Trans. G. di Giovanni. Cambridge: Cambridge
UP, 2010.
Hermann, Paul. Prinzipien der Sprachgeschichte. Halle: Max Niemayer, 1880.
Hjelmsev, Louis. “Structural Analysis of Language” in Readings in Modern Linguistics.
Ed. B. Malberg. Stockholm: Läromedelsförlagen, 1972.
Holenstein, Elmar. Roman Jakobson’s Approach to Language: Phenomenological
Structuralism. Bloomington: Indiana UP, 1976.
Holenstein, Elmar. Jakobson und Hegel: die Wurzeln des osteuropäischen Strukturalis-
mus in der “russischen ideologischen Tradition” und in der deutschen Romantik.
Tokyo: U of Tokyo P, 1985.
Holenstein, Elmar. “Jakobson and Husserl: A Contribution to the Genealogy of Structur-
alism” in Edmund Husserl: Critical Assessments of Leading Philosophers. Ed. R.
Bernet, D. Welton, and G. Zavota. London: Routledge, 2005: 11–48. [Originally
published in The Human Context 7 (1975): 61–83. Trans. E. Eng].
Humboldt, Wilhelm. Ueber den Dualis. Berlin: s.n., 1828.
Bibliography 275
Husserl, Edmund. The Crisis of European Sciences and Transcendental Phenomenol-
ogy: An Introduction to Phenomenological Philosophy. Trans. David Carr. Evan-
ston: Northwestern UP, 1970.
Husserl, Edmund. Cartesian Meditations: An Introduction to Phenomenology. Trans.
Dorian Cairns. Dordrecht: Kluwer Academic, 1999.
Husserl, Edmund. Logical Investigations. Vol. 2. Trans. J.N. Findlay. London: Rout-
ledge, 2001.
Jakobson, Roman and Morris Halle. Fundamentals of Language. The Hague: Mouton,
1956.
Jakobson, Roman. “Saussure’s Unpublished Reflections on Phonemes.” Cahiers Ferdi-
nand de Saussure 26 (1969): 5–14.
Jakobson, Roman. Selected Writings. Vol. 2. Berlin: de Gruyter, 1971.
Jakobson, Roman. Six Lectures on Sound and Meaning. Cambridge: MIT Press, 1978.
Jakobson, Roman. Selected Writings. Vol. 7. Berlin: Mouton, 1985.
Jakobson, Roman. On Language. Ed. L. R. Waugh, and M. Monville-Burston. Cam-
bridge: Harvard UP, 1990.
Jaspers, Karl. General Psychopathology. Vol. 2. Baltimore: Johns Hopkins UP, 1997.
Jones, Charles W. Saint Nikolaos of Myra, Bari, and Manhattan: Biography of a Legend.
Chicago: U of Chicago P, 1978.
Joseph, John E. “Why Lithuanian Accentuation Mattered to Saussure.” Language &
History 52.2 (2009): 182–198.
Joseph, John E. Saussure. Oxford: Oxford UP, 2012.
Koerner, E. F. K. Bibliographia Saussureana, 1870–1970. An Annotated, Classified
Bibliography on the Background, Development, and Actual Relevance of Ferdi-
nand de Saussure’s General Theory of Language. Metuchen: Scarecrow, 1972.
Kristeva, Julia. Revolution in Poetic Language. New York: Columbia UP, 1984.
Kristeva, Julia. “The Speaking Subject” in Structuralism: Critical Concepts in Literary
and Cultural Studies. Ed. Jonathan Culler. Vol. 4. London: Routledge, 2006.
Kruszewski, Mikołaj. Szkice je˛zykoznawcze [Linguistic sketches]. Vol. 1. Warsaw: Piotr
Laskauer, 1904.
Kruszewski, Mikołaj. Writings in General Linguistics. Ed. E. F. K. Koerner. Amster-
dam: J. Benjamins, 1995.
Lacan, Jacques. Ecrits: The First Complete Edition in English. New York: W. W. Norton,
2006.
Lévi-Strauss, Claude. Structural Anthropology. Trans. C. Jacobson and B. Grundfest
Schoepf. New York: Basic Books, 1963.
Lévi-Strauss, Claude. The Elementary Structure of Kinship. Boston: Beacon, 1969.
Lévi-Strauss, Claude. Structural Anthropology. Vol. 2. Chicago: U of Chicago P, 1983.
Macksey, Richard, and Eugenio Donato. The Languages of Criticism and the Sciences
of Man: The Structuralist Controversy. Baltimore: Johns Hopkins UP, 1970.
Matsuzawa, Kazuhiro. “Notes pour un livre sur la linguistique générale.” Cahier de
l’Herne: Ferdinand de Saussure. Ed. Simon Bouquet. Paris: Editions de l’Herne
(2003): 319–322.
Médina, José. “Charles Bally: de Bergson à Saussure.” Langages 19.77 (1985): 95–104.
Meillet, Antoine. “Nécrologie de Ferdinand de Saussure (1857–1913).” Paris: Imprim-
erie nationale, 1913. [Reprinted as an appendix in de Saussure, Ferdinand. Mémoire
276 Bibliography
sur le système primitif des voyelles dans les langues indo-européennes. Cambridge:
Cambridge UP, 2009].
Meillet, Antoine. “Compte rendu du Cours de linguistique générale.” Bulletin de la So-
ciete de Linguistique de Paris 64 (1916): 20, 32–36. [Reprinted in George Mounin,
Saussure ou le structuraliste sans le savoir (Paris: Seghers, 1968): 166–168].
Meillet, Antoine. Review of the Course in General Linguistics. Revue Critique d’Histoire
et de Litterature 84.1 (1917): 49–51.
Merleau-Ponty, Maurice. The Primacy of Perception and Other Essays on Phenomeno-
logical Psychology, the Philosophy of Art, History, and Politics. Ed. and introd.
James M. Edie. Evanston: Northwestern UP, 1964.
Merleau-Ponty, Maurice. Signs. Trans. Richard C. McCleary. Evanston: Northwestern
UP, 1964a.
Merleau-Ponty, Maurice. Themes from the Lectures at the Collège De France, 1952–
1960. Trans. John O’Neill. Evanston: Northwestern UP, 1970.
Merleau-Ponty, Maurice. Consciousness and the Acquisition of Language. Trans. Hugh
Silverman. Evanston: Northwestern UP, 1973.
Merleau-Ponty, Maurice. The Prose of the World. Ed. Claude Lefort. Trans. John
O’Neill. Evanston: Northwestern UP, 1973a.
Merleau-Ponty, Maurice. In Praise of Philosophy and Other Essays. Trans. John Wild,
James Edie, and John O’Neill. Evanston: Northwestern UP, 1988.
Merleau-Ponty, Maurice. Phenomenology of Perception. Trans. Colin Smith. London:
Routledge, 2002.
Mounin, Georges. Ferdinand de Saussure, ou le structuraliste sans le savoir. Paris:
Seghers, 1968.
Parret, Herman. “Les Manuscrits saussuriens de Harvard.” Cahiers Ferdinand De Saus-
sure 47 (1993): 179–234.
Parret, Herman. “Réflexions saussuriennes sur le temps et le moi.” Cahiers Ferdinand
De Saussure 49 (1995): 85–119.
Percival, Keith. “Roman Jakobson and the Birth of Linguistic Structuralism.” Sign
System Studies 39.1 (2011): 236–262.
Pos, Hendrik J. “Phénoménologie et linguistique.” Revue internationale de philosophie
1 (1939): 354–365.
Preminger, Alex, Frank J. Warnke, and O. B. Hardison. Princeton Encyclopedia of
Poetry and Poetics. Princeton: Princeton UP, 1974.
Raulet, Gerard. “Structuralism and Post-Structuralism: An Interview with Michel Fou-
cault.” Telos 55 (1983): 195–211.
Redard, Georges. “Ferdinand de Saussure et Louis Havet.” Bulletin de la Société de
Linguistique de Paris 71.1 (1976): 313–349.
Redard, George. “Charles Bally Disciple de Ferdinand de Saussure.” Cahiers Ferdinand
de Saussure 36 (1982): 3–23.
Regard, Paul. Contribution a l’étude des prépositions dans la langue du Nouveau Testa-
ment: thèse pour le doctorat ès-lettres présentée à la faculté des lettres de
l’Université de Paris. Dissertation. Paris: Paul F. Editions Ernest Leroux, 1918.
Ricoeur, Paul. “New Developments in Phenomenology in France: The Phenomenology
of Language.” Social Research 34.1 (1967): 1–30.
Ricoeur, Paul. The Conflict of Interpretations. Evanston: Northwestern UP, 1974.
Bibliography 277
Ritzer, George. Encyclopedia of Social Theory. Thousand Oaks: Sage, 2005.
Roggenbuck, Simone. “Differenz und Evidenz. Zu Saussure und Husserl.” Vox roman-
ica 56 (1997): 1–9.
Ronjat, Jules. “Le Cours de linguistique de F. de Saussure.” Journal de Geneve (June 26,
1916): 1.
Russon, John E. Reading Hegel’s Phenomenology. Bloomington: Indiana UP, 2004.
Saussure, Ferdinand de. De L’Emploi du génitif absolu en sanscrit. Leipzig: Impr. JG
Fick, 1881.
Saussure, Ferdinand de. Cours de linguistique générale: publié par Charles Bally et
Albert Sechehaye avec la collaboration de Albert Riedlinger. Lausanne: Libraire
Payot, 1916.
Saussure, Ferdinand de. “Notes inedites de F. de Saussure.” Cahiers Ferdinand de Saus-
sure 12 (1954): 49–71.
Saussure, Ferdinand de. “Cours de linguistique generale (1908–9). IIe Cours: Introduc-
tion.” Cahiers Ferdinand de Saussure 15 (1957): 3–103.
Saussure, Ferdinand de. “Souvenirs d’enfance et de jeunesse.” Cahiers Ferdinand de
Saussure 16 (1958).
Saussure, Ferdinand de. Corso di linguistica generale. Ed. Tullio de Mauro. Roma: Lat-
erza, 1967.
Saussure, Ferdinand de. “Essai pour réduire les mots du grec, du latin & de l’allemand a
un petit nombre de racines.” Cahiers Ferdinand de Saussure 32 (1978): 73–101.
Saussure, Ferdinand de. Le Leggende germaniche. Ed. A. Marinetti and M. Meli. Este:
Zielo, 1985.
Saussure, Ferdinand de. Course in General Linguistics. Ed. C. Bally and A. Sechehaye,
with A. Riedlinger. Trans. Roy Harris. LaSalle: Open Court, 1986. [Reissued by
Bloomsbury, 2013].
Saussure, Ferdinand de. Cours de linguistique générale. Edition critique par Rudolf
Engler. Vol. 1. Wiesbaden: Harrassowitz, 1989 [1967/1968].
Saussure, Ferdinand de and Eisuke Komatsu. Troisième cours de linguistique générale
(1910–1911): d’après les cahiers d’Emile Constantin/Saussure’s Third Course of
Lectures on General Linguistics (1910–1911): From the Notebooks of Emile Con-
stantin. Oxford: Pergamon, 1993.
Saussure, Ferdinand de. Phonétique. Il manoscritto di Harvard. Vol. 3. Padua: Unipress,
1995.
Saussure, Ferdinand de, Eisuke Komatsu, and George Wolf. Premier cours de linguis-
tique generale (1907): d’après les cahiers d’Albert Riedlinger/Saussure’s First
Course of Lectures on General Linguistics (1907): From the Notebooks of Albert
Riedlinger. Oxford: Pergamon, 1996.
Saussure, Ferdinand de, Eisuke Komatsu, and George Wolf. Deuxième cours de linguis-
tique generale (1908–1909): d’après les cahiers d’Albert Riedlinger et Charles
Patois/Saussure’s Second Course of Lectures on General Linguistics (1908–1909).
Oxford: Pergamon, 1997.
Saussure, Ferdinand de. Ecrits de linguistique générale. Ed. S. Bouquet, R. Engler, and
A. Weil. Paris: Gallimard, 2002.
Saussure, Fedinand de. Cours de linguistique générale. Critical ed. Tullio de Mauro.
Payot, 2005 [1967].
278 Bibliography
Saussure, Ferdinand de. Writings in General Linguistics. Trans. C. Sanders, M. Pires,
with A. Weil. Oxford: Oxford UP, 2006.
Saussure, Ferdinand de. Mémoire sur le système primitif des voyelles dans les langues
indo-européennes. Cambridge: Cambridge UP, 2009 [1879].
Saussure, Ferdinand de. Course in General Linguistics. Ed. Perry Meisel and Haun
Sussy. Trans. Wade Baskin. New York: Columbia UP, 2011.
Saussure, Ferdinand de. Course in General Linguistics. Trans. Roy Harris. Introd. Roy
Harris. London: Bloomsbury, 2013.
Sechehaye, Albert. Programme et méthodes de la linguistique théorique: psychologie
du langage. Paris: H. Champion, 1908.
Sechehaye, Albert. “Les Problèmes de la langue à la lumière d’une théorie nouvelle.”
Revue philosophique de la france et de l’étranger 84 (1917): 1–30.
Sechehaye, Albert. “L’Ecole genevoise de linguistique générale.” Indogermanische
Forschungen 44 (1927): 217–241.
Sechehaye, Albert. Évolution organique et évolution contingentielle. Geneva: Georg,
1939.
Sechehaye, Albert. Les Trois Linguistiques saussuriennes. Zurich-Erlenbach: Eugen
Rentsch Verlag, 1940.
Silverman, Hugh J. Inscriptions: After Phenomenology and Structuralism. Evanston:
Northwestern UP, 1997.
Spiegelberg, Herbert. The Phenomenological Movement: A Historical Introduction. The
Hague: Martinus Nijhoff, 1965.
Starobinski, Jean. Les Mots sous les mots: les anagrammes de Ferdinand de Saussure.
Paris: Gallimard, 1971.
Starobinski, Jean. Words upon Words: The Anagrams of Ferdinand de Saussure. New
Haven: Yale UP, 1979.
Stawarska, Beata. Between You and I: Dialogical Phenomenology. Athens: Ohio UP,
2009.
Steiner, Peter. The Prague School: Selected Writings, 1929–1946. Austin: U of Texas P,
1982.
Sturrock, John. Structuralism. Introd. Jean-Michel Rabate. Oxford: Wiley-Blackwell,
2003.
Thibault, Paul J. Re-reading Saussure: The Dynamics of Signs in Social Life. London:
Routledge, 1997.
Thibault, Paul J. “de Saussure, Ferdinand” in Encyclopedia of Social Theory. Ed. George
Ritzer. Thousand Oaks: Sage, 2005.
von Hartmann, Eduard. Philosophy of the Unconscious: Speculative Results According
to the Inductive Method of Physical Science. Trans. Eduard Coupland, William
Chatterton. 2nd ed. Vol. 1. London: Kegan Paul, Trench, Trubner, 1893.
Weber, Samuel. “Saussure and the Apparition of Language: The Critical Perspective.”
MLN, 91.5 (1976): 913–938.
Whitney, William D. Language and the Study of Language: Twelve Lectures on the
Principles of Linguistic Science. New York: Scribners, 1887.
Wittgenstein, Ludwig. Philosophische Untersuchungen/Philosophical Investigations.
Trans. G. E. M. Anscombe. Oxford: Blackwell, 1953.
Bibliography 279
INDEX
Note: The letter ‘n’ following locators refers to notes
academic book format, 5, 8, 50, 254 as ghostwriter, 5, 12, 194–204, 217,

Agamben, Giorgio, 272 247, 259; thesis of sound’s natural
aletheia, 91 symbolism, 74–77
analogy, 130, 133, 134, 136–150, 229; Barthes, Roland, 9, 243
analogical creation, 48; analogical Baskin, Wade, 7n, 10, 13n, 20, 136, 261
innovation, 133, 136, 138, 141, 142, Beauvoir, de Simone, 223–235
146; as cardinal general linguistic Benveniste, Emile, 207n, 209
principle, 147; language and chess, Bergson, Henri, 44–47
86–9. see also language Bopp, Franz, 87, 88, 142, 214
arbitrariness, linguistic, 52, 70, 71, 72, 73, Bourdieu, Pierre, 69
241, 252; ‘arbitrariness of the sign’, 16, Brentano, Franz, 122, 177
23, 24–26, 50, 54; conventionalist Butler, Judith, 48n, 69
notion of, 29, 30, 36, 46; defined
positively, 17, 40; radical arbitrariness, categorematic expressions, 170
41, 51, 76, 78, 185; relative Chomsky, Noam, 138, 139, 240
arbitrariness, 47, 48, 49, 53, 59; social ‘Civilization of the book’, 11, 251,
dimension of, 31, 65, 66, 67, 68. see 254, 255
also Saussurean doctrine consciousness, 5, 19, 69, 128, 154, 173,
associative relations, 49, 56, 57, 131, 150; 182, 184; editorial effacement of,
associative principle, 138 130–134; Hegelian progression of,
author function, 9–16, 248, 260. see also 155–161, 164, 165, 166, 169; linguistic
death of the author consciousness, 109–111, 134–147, 149,
152, 153; and phenomenology, 18,
Bally, Charles, 6, 13n, 242, 251, 255, 262, 125–127, 170, 175, 179, 180, 188; as
263; Bergson’s influence on, 44–46; as semiologically mediated, 168; and the
book reviewer, 205–208, 210, 211; speaking subject, 42, 109, 127, 151,
discipleship, 212, 218–222, 225, 250, 155, 190; and structuralism, 215, 216,
254; double role of, 19, 194, 216–222; 240; temporality of, 174, 187, 235;
and the unconscious, 7, 98, 150, 151, 193, 194, 248–55; “Structure, Sign,
251, 268. see also the unconscious; and Play”, 240. see also
speaking subject; phenomenology Deconstruction
Constantin, Emile, 31, 47, 87, 92, 98n, diachrony, 16, 81; and ambiguity, 129,
196n 244; and analogy, 148, 150; and chess
Cours de Linguistique Générale, see analogy, 87; editorial presentation of,
Course in General Linguistics 90, 92, 93, 250; and method in general
Course in General Linguistics, 7, 91, 92, linguistics, 118, 271; pairing with
97, 101, 120, 122n, 126, 220, 265; synchrony, 9, 15, 17, 19, 97, 206–208,
arbitrariness of the sign, 49, 50, 52, 213, 216, 246; as Saussurean doctrine,
53, 72; axiom format of, 14, 52, 104; 15, 17, 18, 85, 98, 147, 193, 205, 213,
Derrida’s reading of, 48, 71, 76, 78n, 216, 239; and the social, 50, 69; and
79, 81–83, 244, 245n, 251–255; speech acts, 100, 101; and
doctrine of the sign, 24–27, 30, 32, structuralism, 9, 15, 16, 50, 247; and
47, 249; editorial manipulations, 44, subjectivity, 190, 243; unity with
45, 74, 75, 87, 154, 202, 204, 231n; synchrony, 102, 103, 105, 184, 227. see
effacement of consciousness in, also general linguistics; Saussurean
130–134, 145n; ghostwriting of, 5, 12, doctrine; synchrony
77, 197, 199, 203, 214, 225–230, 257,
258, 259; as Great Book, 8, 10, 11, Ecrits de Linguistique Générale, see
219, 248, 250; Merleau-Ponty’s Writings in General Linguistics
reading, 181, 184, 188, 190; Édition Critique, see Engler, Rudolf
philosophical complexity of, 31, 33, editorial manipulations, 9, 31, 92,
44, 135, 154; presentation of analogy, 130–134, 194–205, 213, 257–263; and
133, 136, 147n, 148, 150; reviews of, the book format, 10–16; insertions, 77,
195, 200, 205–210, 212; and 217; famous formula, 19, 92, 213–219,
Saussurean doctrine, 9, 15, 19, 51, 80, 226, 244; ghostwriting, Sechehaye,
85, 193, 213, 215n, 216, 217, 243, 271; Albert; reversal of contents, 20, 136,
second editorial paradigm of, 6, 9, 225–230, 232. see also Bally, Charles
254, 255, 258, 261, 262; Structuralist Engler, Rudolf, 5, 26, 258; Critical
reception of, 16–18, 20, 23, 50, 106n, Edition (Édition Critique) of, 7n, 15,
194, 239, 241, 242, 246, 247; as a 211n, 257
Vulgate edition, 13, 14, 196, 248. see “epistolophobia”, 8, 267n
also Saussurean doctrine; editorial Esperanto, 68, 82, 83
manipulation
Culler, Jonathan, 48, 78n, 239, 241 Flournoy, Theodore, 269
Foucault, Michel, 11, 12, 178, 180, 260
de Courtenay, Baudouin, 18, 121, 122, 124
Dégailler, George, 28, 31, 87, 92, 196n, 202 general linguistics, 9, 14, 30, 38, 39, 45,
de Mauro, Tulio, 6n, 7n, 59n, 214, 258n 77, 106n, 129, 139, 161, 200, 206, 218;
death of the author, 12 analogy in, 135, 136, 142, 147, 149,
Deconstruction, 9, 80, 252–55 151; and change, 43, 50, 70; and the
Derrida, Jacques, 12, 26, 32, 172, 241, Course, 5–9, 11, 12, 24, 53, 197, 203,
244, 246; Glas, 17, 48, 52, 71–84; Of 215, 216, 220, 226, 248, 250, 257; di-
Grammatology, 17, 71–84, 245, 245n; chotomies of, 208; as a double science,
and poststructuralism, 10, 20, 106n, 96–97, 99, 238; editorial creation of,
282 Index
205, 227–232, 234, 253; and meta- Hjelmslev, Louis, 219–222, 225
physics of presence, 71, 252; method Holenstein, Elmar, 120, 175
of, 51, 90, 111, 118, 263; paradox of, Humboldt, Wilhelm von, 125, 126, 135
109, 165, 168, 181; and phenomenol- Hume, David, 138
ogy, 18, 19, 110, 112, 120, 121, 126, Husserl, Edmund, 61, 80, 125, 126,
130, 154, 156, 180; philosophical com- 184–188, 244; Cartesian Meditations,
plexity of, 44, 85, 102, 112, 154, 157; 175; collaboration with Jakobson, 19,
Saussure’s lectures on, 195, 196, 198, 120, 122, 154, 169–181, 244; critique
199, 201, 209, 212, 234–237; Sau- of natural sciences, 111; Logical
ssure’s planned book on, 33, 103, 104, Investigations, 170; phenomenology
109, 128, 202n, 265, 267, 269–271; of language, 127, 169–18
science of, 11, 13, 80, 90, 93, 99, 103,
105, 134, 194, 214; and semiology, intentionality, 173–174
164, and sociality, 47, 69; and speech,
184; and structuralism, 2, 16, 23, 66, Jakobson, Roman, 30, 215, 219, 242, 243,
152, 177, 242–247; and subjectivity, 2, 245, 250, 268; collaboration with
110, 132, 133, 135; and temporality, Husserl, 19, 120, 122, 125, 154, 169–181;
47, 169; as total science, 155. see also and phenomenology, 184, 244
Course in General Linguistics;
language; structuralism Kant, Immanuel, 127, 187; Kantian
ghostwriting, 5, 11, 12, 19, 77, 106n, approach to language, 63
194–205, 214, 225, 247. see also Kazan School of linguistics, 120–126, 177
editorial manipulations kinship, 20, 84, 219–225, 243. see also
Godel, Robert, 5, 15, 202n, 209, 210, master-disciple relation; patrilineal
211n, 252, 258 descent
grammar, 51, 53, 130, 136, 139, 140, 141, Kristeva, Julia, 4–5
142, 163; a priori or universal, 35, 38, Kruszewski, Mikolaj, 120–126, 170, 180:
169–172, 175, 177; comparative and phenomenology of language, 18,
grammar, 143, 156, 161, 195; historical 173, 184
grammar, 93, 228, 265; Port-Royal
grammar, 34, 162. see also Lacan, 80, 106n, 120, 146, 194, 243, 244,
Neogrammarians 248; Lacan’s reading of Saussure,
Greimas, Algirdas Julien, 9, 221 249–50
language: analogy with chess, 86–88, 94;
Harris, Roy, 20, 26n, 27, 231n, 232, 258, autonomy of, 23, 78, 213; double
262; on the arbitrariness of the sign, essence of, 86, 94, 97–109, 115;
23; English language translation, 10, ethnographic dimension of, 39;
13n, 15n, 136, 257; 1972 translation, general theory of, 226–237, 266, 270;
6n. see also Course in General grammatical laws of, 170–177; and
Linguistics metaphysics of presence, 79–84, 113;
Hegel, G.W.F., 38, 80, 119, 122, 125, 244; nomenclature view of, 16, 25, 26,
difficulty of beginning, 17, 85, 97, 27–29, 31–47; onomatopoeia in,
105; phenomenology of, 19, 120, 126, 74–78; and phenomenology, 16,
128, 155–169, 177; Phenomenology 120–147, 155–169, 181–193;
of Spirit, 155, 164. see also philosophical understandings of, 5, 7,
phenomenology 18, 270–272; pictorial dimension of,
Index 283
39; relational understanding of, 43, 49, Meillet, Antoine, 113, 198, 199, 211, 213,
129, 141, 147, 154, 164, 236, 266, 268; 215n, 265, 266, 269, 271; criticism of
scientific study of, 90, 91, 93, 95, 110, editorial process, 200, 209, 213, 247;
112, 205n, 214–217, 265; as review of the Course, 207, 212
semiological, 2, 9, 15, 27, 91, 268; as Merleau-Ponty, Maurice, 66, 146, 154,
sign system, 15, 16, 17, 19, 92, 243, 247n; phenomenology of
151–153, 201, 206, 226–227, 242; language, 154, 177n, 181–193;
social convention in, 24, 25, 30, 47–71, speaking subject, 19, 120
74, 83, 100, 127; and structuralism, metaphysics of presence, 17, 71, 79–85,
1–3, 91, 150, 178, 247; and 102, 253
subjectivity, 15, 88–89, 91, 109, 111; Mounin, George, 247
temporal dimension of, 78n, 88, 90,
94, 96, 101, 245n; and the Nachlass, 7, 17, 18, 19, 71, 242, 259;
unconscious, 145–146, 149, 150, exclusion of, 12, 13, 16, 20, 248, 252,
243–244, 249, 251, 268. see also 252; linguistic arbitrariness in, 48; and
general linguistics; speech acts; second editorial paradigm, 9, 254,
unconscious 262. see also editorial manipulation
la langue, 9, 15–20, 43, 83, 198, 213–217, Naville, Adrien, 230
220, 226–239; and analogy, 133, 136, Neogrammarians, 113, 137, 138. see also
142, 145; and arbitrariness, 46; grammar
complexity of, 103–105; and the Nietzsche, Friedrich, 11, 269, 272
Course, 24; distinction with la parole, nomenclature view of language, 16, 29,
45, 85, 87–94, 96–101, 149, 150–152, 31–47, 49, 52, 55n, 62, 68, 79, 207,
162–165, 201; and the nomenclature 270; and sound based conceptions of
view, 26, 27, 31, 32; as object of study, signs, 27, 53. see also metaphysics of
80, 134, 157, 188–190; and Saussurean presence; and substance metaphysics,
doctrine, 9, 85, 87–94, 147, 193, 25–7, 30, 57, 59, 67, 80
205–208; and semiology, 15; and the
social, 47–71; as a static system, 114; onomatopoeia, 17, 53, 71–77
and subjectivity, 130, 131, 144. see objectivity, 48, 103, 110, 187, 213;
also Course in General Linguistics; objectivist tendency of editors, 89, 93,
general linguistics; language; la 131, 134, 204, 217; Saussure’s
langue, la parole interrogation of, 104, 113, 118; and
Levinas, Emmanuel, 176 science, 18, 31, 69, 92, 109, 112, 119,
Levi-Strauss, Claude, 12, 120, 182, 221, 134, 181
241, 244, 250, 254; culture/nature
distinction, 223–225; reception of the Paul, Hermann, 125, 135, 138
Course, 20, 106n, 194, 243, 246, 248; la parole, 157, 206–208, 213; and the
and structural anthropology, 9, 80. see Course, 9, 15–19, 193; distinction with
also structuralism la langue, 45, 46n, 69, 85, 87–94,
logos, 43, 51, 79, 96, 185, 186, 233, 96–101, 149, 150–152, 201, 208;
234 editorial presentation of, 89, 90, 131,
205, 216; and Saussurean doctrine, 9,
master-disciple relation, 196, 221–2. see 85, 147, 205, 237, 239; and the social,
also patrilineal descent; kinship 63, 64; and temporality, 188–190. see
Mathesius, Vilem, 121, 177 also Course in General Linguistics;
284 Index
general linguistics, la langue; of, 16, 20, 147, 213; and structuralism,
language 2, 85, 194, 211, 239, 243, 244;
patrilineal descent, 19–20, 221, 223. see synchrony/diachrony, 17, 19, 85, 247.
also kinship; master-disciple relation see also diachrony; general
phenomenon, 18, 48, 88, 111; of analogy, linguistics; la langue, la parole;
139, 141, 148; duality of, 100; synchrony
language as, 88, 92, 96, 114, 123, 124, Scholasticism, 13, 14, 219
178, 210, 242; Saussure’s use of, 62, séance, 6n, 217, 257–259, 269
96, 100, 126–130, 161, 236; speech as, Sechehaye, Albert, 6, 13n, 46n, 210, 225,
97, 185. see also phenomenology 259, 262; as disciple, 195–198, 200,
phenomenology, 4, 120, 127, 128, 160, 212, 218, 219, 221, 222, 254; double
173; and ambiguity, 181; and role of, 19, 194; as ghostwriter, 5, 12,
consciousness, 153, 188; Hegel’s 104, 199, 201–205, 211, 217, 247, 250;
phenomenology, 155, 157, 158, 161, and onomatopoeia, 73; penchant for
164, 165; linguistic phenomenology, 7, abstractions, 226, 230–232; as
16, 18, 19, 110, 127, 153, 154, 169, 187; reviewer, 207–209, 213; Saussure’s
phenomenology and structuralism, critique of, 129, 130; and Saussurean
1–3, 18, 176–181, 186, 213, 215–216, doctrine, 216, 225, 242. see also
242; phenomenology of language, 124, ghostwriting; séance
172, 175, 182, 190; and Saussurean semiology, 9, 215, 252; Saussure’s
doctrine, 244; as a science, 125, 184; conception of, 15, 33, 96; as a science,
and the semiological challenge, 27, 184, semiological conception of
215–216. see also consciousness; language, 42, 83, 94; semiological
speaking subject halo, 55; ‘semiological phenomenon’,
phenomenological subject, see speaking 62. see also general linguistics;
subject language
Prague Linguistic Circle, 121, 177, 219, signifier, 9, 15, 16, 23–24, 37, 38, 39, 56;
242, 243, 245 and arbitrariness, 26, 29, 30, 40, 46,
Prague structuralism, see Prague 48, 51, 66; and nomenclature view, 32;
Linguistic Circle and onomatopoeia, 72; 76, 77;
oppositional pairing, 193, 237, 239,
Regard, Paul, 196n, 198, 202 246, 249; and presence, 79, 80, 241,
Riedlinger, Albert, 5, 13n, 154, 196n, 252; signifier-signified relation, 23–24,
202, 207, 209–11, 262; 25, 31, 42, 53, 59, 68, 147;
correspondence with Saussure, 229, sociohistorical dimension of, 65, 67,
270; lecture notes of, 133, 151, 228 70, 245n. see also general linguistics;
Ricoeur, Paul, 176, 215, 216 language; Saussurean doctrine; and
temporality, 43
Sartre, Jean-Paul, 146 Smith, Helen, 267, 269
Saussurean Doctrine, 65, 215n; the social convention, 69–71, 74, 83, 220,
Course as the official site of, 7, 9, 15, 237; and individual speech acts, 69,
194, 206, 247, 251, 254; editorial 90, 91, 95, 99, 156; and la langue/la
creation of, 201, 205, 207, 208, 225, parole distinction, 63–65, 237; and
245; hierarchical dichotomies, 51, 63, temporality, 17, 25, 48, 71, 100. see
85, 147, 213, 247; la langue/la parole, also la langue; language, la langue,
17, 19, 85, 247; oppositional pairings la parole
Index 285
social critique, 66–70 substance metaphysics, 25, 31, 35, 41, 43,
speaking subject, 2–5, 62–66, 132–134, 113, 164, 270
178; ambiguity of, 7, 64, 103, 105; and syncategorematic expressions, 170
the community, 100, 127; and synchrony, 19, 69, 81, 97, 251; and
consciousness, 146, 149, 155, 156, 161; analogy, 148, 150; and chess analogy,
and history, 74, 96, 98, 101n; and 87; in the Course, 16, 50, 205,
metaphysics of presence, 79; and 213, 216, 247; editorial presentation
methodology of linguistics, 111, 118, of, 90, 92, 93, 250; and method of
130n, 165, 177, 190; and general linguistics, 118, 129, 130n,
phenomenology, 18, 19, 110, 120, 126, 131, 271; pairing with diachrony, 9,
180, 183, 186, 187; and semiology, 42, 15, 17, 19, 206–208, 213, 216, 246;
150, 184; and the signifier/signified as Saussurean doctrine, 15, 17, 18, 85,
relation, 28; and speech acts, 88, 145; 98, 147, 193, 205, 213, 216, 239; and
and structuralism, 1, 152, 216, 244, the social, 69, 100, 184; and
246. see also consciousness; structuralism, 9, 193, 239; and
phenomenology; subjectivity subjectivity, 189, 228, 229, 243,
speech act, 100, 101, 126; iterability of, 244; and temporality, 142n, 227;
152; reciprocal relation with system, unity with diachrony, 101, 102, 103,
89; and social convention, 64, 90, 91. 105. see also diachrony; Saussurean
see also diachrony; language; doctrine
speaking subject; synchrony syntagmatic relations, 56
Spet, Gustav, 176, 177
Starobinski, Jean, 269 temporality, 47, 174; and language, 43, 47,
structuralism, 2, 43, 182, 189, 225, 239, 60, 61, 88, 118; and nomenclature
251; Eastern European structuralism, view, 42, 43; universal law of, 24, 50,
20, 121, 169, 175, 242–245; relation to 173. see also language
phenomenology, 1, 3, 19, 176, 178, the unconscious, 7, 136n, 137, 138, 240;
180, 186, 213, 215; and Saussure’s Lacan’s analysis of, 243, 244, 249;
general linguistics, 6n, 79, 262, 263; and language, 98, 125, 126, 145, 146,
Saussurean doctrine as foundation of, 149–51, 251, 268n. see also
9, 13, 14, 16, 188, 193, 214; signifier/ consciousness; subjectivity;
signified distinction, 23, 24; standard language
definition of, 239, 240–45, 246, 247,
248. see also Saussurean doctrine Weber, Max, 111
Sturrock, John, 2, 240, 241, 246 Weber, Samuel, 88n
subjectivity, 4, 44, 130, 146; and Werthheimer, Max, 111
intentionality, 173, 174, 216; and Wittgenstein, Ludwig, 33, 36, 37, 204
linguistic consciousness, 149, 155, writing: “arche-writing”, 79–84;
156; and linguistics as a science, 110, Derrida’s analysis of, 79–84, 255;
112, 118, 134, 190; phenomenological “impossible writing”, 204; relation
description of, 178, 180, 184; and with speech, 17, 71, 78n, 252. see also
semiology, 42, 91, 150, 184; and the Derrida; deconstruction
social, 48, 69, 186, 188; and Writings in General Linguistics, 6, 32n,
structuralism, 2, 152, 243. see also 106, 126, 237n
consciousness; speaking subject Wundt, Wilhelm von, 130, 135
286 Index

Beata Stawarska - Saussure's Philosophy of Language As Phenomenology - Undoing The Doctrine of The Course in General Linguistics-Oxford University Press (2015) PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Beata Stawarska - Saussure's Philosophy of Language As Phenomenology - Undoing The Doctrine of The Course in General Linguistics-Oxford University Press (2015) PDF

Încărcat de

Drepturi de autor:

Formate disponibile

Saussure’s Philosophy of Language as Phenomenology

Oxford New York

Oxford is a registered trade mark of Oxford University Press

Published in the United States of America by

© Oxford University Press 2015

All rights reserved. No part of this publication may be reproduced,

You must not circulate this work in any other form

Library of Congress Cataloging-in-Publication Data

Printed in the United States of America on acid-free paper

PA RT I | “Saussurean doctrine” and its discontents

5 Saussure’s general linguistics as linguistic

6 Contributions to linguistic phenomenology: Hegel,

PA RT I I I | The inception and the reception of the

8 Structuralist and poststructuralist reception

The speaking subject

Many of us like to think that when we use language we control it more or

2 Saussure’s Philosophy of Language as Phenomenology

4 Saussure’s Philosophy of Language as Phenomenology

Saussure’s general linguistics revisited

6 Saussure’s Philosophy of Language as Phenomenology

8 Saussure’s Philosophy of Language as Phenomenology

The book and the author function

The Course in General Linguistics is widely regarded as a locus of the

10 Saussure’s Philosophy of Language as Phenomenology

12 Saussure’s Philosophy of Language as Phenomenology

14 Saussure’s Philosophy of Language as Phenomenology

The outline of contents

Part I, ‘“Saussurean doctrine” and its discontents’, draws on materials

16 Saussure’s Philosophy of Language as Phenomenology

18 Saussure’s Philosophy of Language as Phenomenology

20 Saussure’s Philosophy of Language as Phenomenology

Arbitrariness of the sign

24 Saussure’s Philosophy of Language as Phenomenology

the signifier and the signified 25

Graphic illustration: sign/signifier/signified

Ch. I on “the nature of linguistic sign” (part I, General principles) consti-

The figure illustrates the problematic thing-cum-name view which the

26 Saussure’s Philosophy of Language as Phenomenology

Figure 2 is a set of two figures:

the signifier and the signified 27

28 Saussure’s Philosophy of Language as Phenomenology

the signifier and the signified 29

This alternative theory is borne out by the source materials, where

30 Saussure’s Philosophy of Language as Phenomenology

Critique of the nomenclature view of language

The reader of the Course is unlikely to appreciate the philosophical com-

the signifier and the signified 31

32 Saussure’s Philosophy of Language as Phenomenology

1. it supposes that readymade ideas exist before words;

This list is not devoid of manuscript support but is so brief that it

the signifier and the signified 33

The nomenclature view is problematic not only because it makes biased

34 Saussure’s Philosophy of Language as Phenomenology

This passage captures the mirroring relation between the metaphysical

Aristotle thus establishes a series of equivalences between an immuta-

the signifier and the signified 35

Augustine’s ontogenetic account of language thus establishes a link be-

36 Saussure’s Philosophy of Language as Phenomenology

the signifier and the signified 37

38 Saussure’s Philosophy of Language as Phenomenology

the signifier and the signified 39

40 Saussure’s Philosophy of Language as Phenomenology

It is astonishing. But where in truth would the possibility of the contrary