Instrumental Inference

Author(s): George Dickie

Source: The Journal of Aesthetics and Art Criticism, Vol. 42, No. 2 (Winter, 1983), pp. 151-
154
Published by: Wiley on behalf of The American Society for Aesthetics
Stable URL: http://www.jstor.org/stable/430659 .
Accessed: 31/05/2013 08:50
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Wiley and The American Society for Aesthetics are collaborating with JSTOR to digitize, preserve and extend
access to The Journal of Aesthetics and Art Criticism.
http://www.jstor.org
This content downloaded from 193.225.200.89 on Fri, 31 May 2013 08:50:45 AM
All use subject to JSTOR Terms and Conditions
GEORGE DICKIE
Instrumental
Inference
THERE HAS BEEN a considerable discus-
sion inthe literature of aesthetics of
how,
if at
all,
reasons and
principles play
a
role in the evaluation of art. Monroe
Beardsley
has been aresolute defender of
principles
as
necessary
for criticismto be
a rational
activity,
and his views do not
appear
to have
changed
since he first dis-
cussed the
topic
in 1958. In this
paper
I shall discuss the nature of the
princi-
ples
which are involved,
according
to
Beardsley,
in
justifying
critical evaluations
and also the nature of the inference which
yields, according
to
Beardsley,
critical
evaluations.
My
remarks amount to a
tidying up
of
Beardsley's theory.
InAesthetics: Problems in the Philos-
ophy of
Criticism
Beardsley
begins by
taking
the
general
form of critical
evaluation to be of the form:
This work of art is
good (the
critical evaluation).
because it has such and such a
property (the
rea-
son).
He then
goes
on to claimthat some rea-
sons involve
generality.
For
example,
if
the reason
given
is "This work of art is
unified,"
this
implies
that there is a
gener-
al
principle
such as
"Disunity
is
always
a
defect in works of art" or
"Unity
is a
good-making
feature of works of art."
Some reasons do not involve as
great
a
degree
of
generality,
because
they
referto
properties
which are
good-making only
when
accompanied by
other
properties.
Finally,
at this initial
stage, Beardsley
GEORGE DICKIE is
projfssor of philosophyi
at the Uni-
versity of Illinois at
Chicago
Circle.
raises the
question
of whether critical
evaluation
arguments
are deductive or in-
ductive. His first conclusion is that the in-
ference to "This work of art is
good"
cannot be deductive because a false
gen-
eral
principle
would be
required
to make
the
argument
valid. The
following
kind of
argument
would,
Beardsley
thinks, be re-
quired:
1. All
highly
unified works of art are
good.
2. This work of art is
highly
unified.
3.
Therefore,
this work of art is
good.
Beardsley
first
claims, quite rightly,
that
the
general principle
in the
argument
is
false; he further
claims,
wrongly,
that
consequently,
critical evaluation
argu-
ments cannot be deductive. It would seem
then that the inference to "This work of
art is
good"
must be inductive, and
Beardsley strongly suggests
that it is. He
does
not, however, explicitly
draw this
conclusion and
says
that he will not be
able to settle this issue until certainother
questions
are answered. I do not
believe,
however,
that
Beardsley
ever returns to
the
question
of the natureof the inference
inan
explicit
and theoretical
way
inAes-
thetics: Problems in the
Philosoplhy
of
Criticism or in
any subsequent
book or
essay.
Nor does he ever illustrate his
apparent
view that critical evalution
argu-
ments are inductive in character
by
con-
structing
a
sample argument.
Perhaps
the first
thing
to notice is that
although Beardsley
asserts that a
principle
such as "All
highly
unified works of art
Copyright
1983. The Journal
of Aesthetics
and Art Criticism
This content downloaded from 193.225.200.89 on Fri, 31 May 2013 08:50:45 AM
All use subject to JSTOR Terms and Conditions
DICKIE
are
good"
is
false,
he asserts that weaker
principles
such as
"Unity
is a
good-mak-
ing
feature of works of art" are ac-
ceptable.
This weaker
principle may
be
restated
equivalently
as "A unified work
always
has some
good
init." "Has some
good
in it" is
very
weak and is true of
works from
magnificent
ones to those of
the least
possible
value.
By
contrast "is
good"
is
narrowly
focused and is true
only
of works which exceed a certain
threshold of value. With this weaker
prin-
ciple
as a
premise,
the
following
deduc-
tive
argument
can be constructed:
1. A unified work of art
always
has
some
good
in it.
2. This work of art is unified.
3.
Therefore,
this work of art has some
good
in it.
So, using Beardsley's principle,
a deduc-
tive critical evaluation
argument
can be
constructed, although
its conclusion is
much weaker than the conclusion
Beardsley
used in
making
his
argument
against
critical
arguments
as deductive.
Beardsley
has inmind as a conclusion of
anevaluative
argument
"This work of art
is
good,"
But the conclusions of such ar-
guments
have evaluative
predicates
which
range
from'bad" and "some
good
init"
through "good"
to
"magnificant."
At this initial
stage
of his
argument,
Beardsley
concludes that there are three
general
evaluation
principles:
1.
Unity
is a
good-making
feature of
works of art.
2.
Intensity
is a
good-making
feature of
works of art.
3.
Complexity
is a
good-making
feature
of works of art.
What is clear at this
point
is that even the
statement that awork of art is
unified,
in-
tense,
and
complex, together
with these
three
principles,
will not
deductively yield
the conclusion that the work of art is
good
but
only
that it has some
good
init.
The real moral of the
story
thus far is
that whatever the
logical
nature of the in-
ference to an evaluative
conclusion,
a
strong
conclusion such as "This work of
art is
good" requires
a
strong premise
or
premises. And,
I think that
Beardsley's
instrumental
theory
of value involves
strong principles
which can serve as
strong premises, although Beardsley
does
not seemto have been aware of themor
at least does not make
any explicit
use of
them.
The first
step
in the
development
of
Beardsley's theory
of value is to maintain
that works of art (he
says,
"aesthetic ob-
jects")
have a
function, namely,
to induce
aesthetic
experiences.
Given this
step,
he
gives
the
following
definition of
"good
aesthetic
object":
"X is a
good
aesthetic
object"
means "X is ca-
pable
of
producing good
aesthetic
experiences
(that is, aesthetic
experiences
of a
fairly great
magnitude)' (p. 530).
He then
says,
"I
propose
to
say, simply,
that
'being
a
good
aesthetic
object'
and
'having
aesthetic value'
mean the same
thing" (p. 351).
Given this
equation,
he
gives
the fol-
lowing
definition:
"X has aesthetic value" means "X has the ca-
pacity
to
produce
an aesthetic
experience
of
fairly great magnitude
(such an
experience
hav-
ing value)" (p. 531).
The
equation
of
"good
aesthetic
object"
and
"having
aesthetic value"
is,
I
think,
a nonlethal
misstep.
To
say
that some-
thing
has aesthetic value is to
say
some-
thing quite
broad-it covers cases from
the smallest
possible
value to the
greatest
possible
value. To
say
of
something
that
it is a
good
aesthetic
object
is to
say
something
more
narrowly
focused-it cov-
ers cases which
surpass
a
threshold,
cases
which have
enough goodness
or value to
be
good.
In the remainder of the discus-
sion in his
book, Beardsley
uses the ex-
pression
"aesthetic
value,"
but I think he
uses it to mean what is meant
by
the ex-
pression "good
aesthetic
object."
Recall now
Beardsley's
definition of
"good
aesthetic
object."
152
This content downloaded from 193.225.200.89 on Fri, 31 May 2013 08:50:45 AM
All use subject to JSTOR Terms and Conditions
Instrumental
Inference
"X is a
good
aesthetic
object (means)
X is capa-
bleof
producing good
aesthetic experiences (that
is,
aesthetic
experiences
of a
fairly great magni-
tude)."
The definition
speaks explicitly
of
good
aesthetic
experiences
and asserts that
good
aesthetic
experiences
are to be
equated
with aesthetic
experiences
of a
fairly great magnitude.
This is
equivalent
to
claiming
that the
following generaliza-
tion is true:
1. Aesthetic
experiences
of a
fairly
great magnitude
are
always good.
The definition of
"good
aesthetic
object"
yields
also the
following
conditional in
which I substitute "work of art" for
"aesthetic
object"
in order to limit the
discussion to works of art.
2. If awork of art is
capable
of
produc-
ing
a
good
aesthetic
experience,
then the
work of art is
(instrumentally) good.
Now if ina
given
case we have the
prem-
ise:
3. This work of art can
produce
anaes-
thetic
experience
of
fairly great magni-
tude.
From1 and 3 we can
get
4. This work of art can
produce
a
good
aesthetic
experience.
From2 and 4 we can
get
5. This work of art is
(instrumentally)
good.
All the
steps
inthis
argument
are deduc-
tive,
so the
argument
is a deductive
argu-
ment. What about the
premises
of the
argument?
Premise
1,
"Aesthetic
experi-
ences of
fairly great magnitude
are
always
good,"
is
clearly
an inductive
generaliza-
tion, presumably
derived from
multiple
observations that aesthetic
experiences
which
equal
or exceed acertain
degree
of
magnitude
are
invariably good.
There is
no
logical problemhere;
the
argument
is
a
straight-forward
inductive
generaliza-
tion that
property
A is
invariably
accom-
panied by property
B. Premise 2 is
simply
a
particularapplication
of the definitionof
"instrumentally good."
Premise
3,
"This
work of art can
produce
an aesthetic ex-
perience
of
fairly great magnitude,"
is not
a
generalization
but a
singularreport
of an
experience
of the kind used inthe
support
of
premise 1,
which is a
generalization.
A
parenthetical
remark is inorderhere.
There are also other
principles
available,
given Beardsley's theory,
which canfunc-
tion in
arguments
similar to the
sample
argument being developed. Examples
of
such
principles
are: "Aesthetic
experi-
ences of less than
fairly great magnitude
are
always
not
good,"
"Aesthetic
experi-
ences of a
fairly
low
magnitude
are al-
ways bad,"
and the like. The
expressions
"fairly great magnitude"
and
"fairly
low
magnitude"
refer to threshold
points
which mark the
dividing
lines
respectively
between
good
and not
good
and between
bad and not bad aesthetic
experiences.
The
sample argument being developed
is
complete,
but
premises
of the kind
Beardsley
mentions as the
premises
of
critical
arguments, namely, reason-prem-
ises such as "This work of art is unified"
and "This work of art is
intense,"
have
not been mentioned. Neither have
princi-
ples
such as "A unified work of art al-
ways
has some
good
in it" been men-
tioned.
It
is, however, certainly
true that state-
ments such as "This work of art is uni-
fied,"
"This work of art is
intense,"'.and
"This work contains a touch of humor"
are used
by
critics in connection with
their critical evaluations. The
question is,
"Given
Beardsley's theory,
does he
give
the
right
account of the relation between
critical evaluations and reason-statements
which refer to
properties
of works of
art?"
Aesthetic
experience
is basic inBeards-
ley's
evaluational
theory.
Whatever value
a work of art
has,
derives fromthe value
of anaesthetic
experience
it can
produce:
a
good
work of art is
good
because it can
produce
a
good
aesthetic
experience,
a
poor
work of art is
poor
because it can
153
This content downloaded from 193.225.200.89 on Fri, 31 May 2013 08:50:45 AM
All use subject to JSTOR Terms and Conditions
154
produce only
an aesthetic
experience
which is less than
good,
and so on. Inor-
der for a
qualified
critic to know that a
work of art is
good,
he must first have an
aesthetic
experience produced by
the
work and find it to be
good.
Once the
critic knows the aesthetic
experience
is
good,
he knows the work of art which
causes it is
(instrumentally) good,
and he
is ina
position
to construct the deductive
argument given
above which
begins
with
the
premise
"Aesthetic
experiences
of
fairly great magnitude
are
always good"
and ends with the conclusion"This work
of art is
(instrumentally) good."
Thus,
on
Beardsley's view,
one does
not have to have
premises
such as "This
work of art is unified" in order to con-
clude of a
given
work that it is
good.
One
does not evenhave to be
overtly
aware of
the value
properties
of the work for them
to
produce
the aesthetic
experience.
So
the
question
of how
inductively
to draw
such conclusions as "This work is art is
good"
from
premises
which refer to
prop-
erties of the work
plus
some
generaliza-
tions about
properties
does not arise for
Beardsley's theory, despite
the fact that
he thinks it does.
As noted
above, however,
reason-state-
ments such as 'This work is unified" and
"This work is intense" are used
by
critics
in connection with critical
evaluations,
and it is clear that
they
have an
important
role to
play
in criticism. What is that
role? No set of such reason-statements to-
gether
with the weak
principles
with
which
they
connect can
imply any specific
evaluationother than that a work has
some
good
in it. On
Beardlsey's theory
and instrumentalistic theories like
it,
rea-
son-statements and weak
principles
cannot
imply
that a work of art is
poor,
DICKIE
mediocre,
or
good. Only strong principles
about aesthetic
experiences
and state-
ments about awork's
capacity
to
produce
anaesthetic
experience
of acertain
mag-
nitude can
imply
such
specific
evaluations.
The
only
role that remains for reason-
statements and weak
principles
is that
they
canbe used to construct
explanations
of
why
awork is
poor, mediocre,
or
good
after
it has beendetermined that the work is
poor,
mediocre,
or
good by
means of
strong prin-
ciples.
Consider this
argument:
1. A unified work
always
has some
good
init.
2. This work of art is unified.
3.
Therefore,
this work of art has some
good
init.
Such an
argument
serves as a
partial
ex-
planation
of
why
awork has the value it
has as well as anindicationof the source
(unity)
of the value. The total set of rel-
evant reason-statements
plus
the weak
principles
will constitute the total
explana-
tionof a work's
specific value,
but
they
will not
imply
that
specific
value.
There
are, then,
in
Beardsley's theory
three
principles
which he
explicitly
dis-
cusses and
develops
and which he takes
to be the
generalizations
involved in
argu-
ments which
yield specific
critical
evaluations. These are the
principles
in-
volving unity, intensity,
and
complexity.
These three
principles
are weak
princi-
ples.
These are also
strong principles imp-
licit in
Beardsley's theory, principles
which refer to aesthetic
experiences
rather thanto art. It is the
strong principles
which
yield specific
critical evaluations. The
weak
principles,
which refer to works of
art,
functionto
yield explanations
of the sources
of value inart.
This content downloaded from 193.225.200.89 on Fri, 31 May 2013 08:50:45 AM
All use subject to JSTOR Terms and Conditions