Documente Academic
Documente Profesional
Documente Cultură
by
Jon Elster
and evidence. They have nothing in common with what has traditionally been
thought of as science. In many countries, these paradigms dominate large parts of
anthropology, psychology, sociology and political science.
Many scholars will accept this rough description of a sad state of affairs.
They will add, however, that the discipline of economics provides an outstanding
instance of a genuinely scientific approach to social phenomena. I shall contest
this claim. In my opinion, economics as well as its offshoots in political science
offer many examples of hard obscurantism. While the arguments are coherent and
compelling, because of their mathematical form, their relevance for understanding
social phenomena is often close to nil. Forty years ago, my compatriot Ragnar
Frisch, one of the first winners of the Nobel prize in economics, wrote that
“econometrics must have relevance to concrete realities—otherwise it degenerates
into something which is not worthy of the name econometrics, but ought rather to
be called playometrics”. The importance of playometrics, and of science-fiction
economics more generally, has steadily increased since that time. Up to a certain
point, I am not against playfulness – but not if it also claims to be science!
Soft obscurantists tend to criticize hard obscurantists, and vice versa. When I
criticize both sides, each of them may easily assume that I belong to the other side.
Alternatively, each side may take me for an ally in their fight against the other
side. Neither situation is particularly pleasant. Fortunately, I am not alone in this
enterprise. A number of public-spirited scholars have taken time off from their
own work to criticize and denounce obscurantism in detail, sentence by sentence
or equation by equation.
In this talk I shall try to do three things.
First, I want to document and illustrate the pervasiveness of obscurantism in
the social sciences. This will take up the greatest part of the talk. It goes without
saying that in the span of one hour, I cannot go deeply into the theories I shall
discuss and denounce. Yet I believe that with one exception, to which I shall
return, I could back up my statements with detailed and specific criticism.
3
I begin with the bottom right-hand corner. Lévi-Strauss’s early work on kinship
systems, although deeply flawed empirically, was not obscurantist. His later work on
analysis of myth seems to me, however, impenetrable and lacking in intersubjective
validity. It is closer to numerology or astrology than to science. In France and
elsewhere, it was nevertheless deeply influential, by holding out the prospect of
creating the sociological analogue to Mendeleff’s periodic table of the elements. Since
the heyday of structuralism is over, and the theory seems to have left few traces, I shall
not discuss it any further.
Functionalism, however, is well and alive, as I discovered upon my recent return
to France. While functionalism as a mode of analysis was born in sociology and
anthropology, it is I believe, a very general attitude, rooted, I believe, in our desire to
find meaning in the world. To give a flavor of the approach, I can cite the claim that
the function of feuds and vendettas is to keep population size at a sustainable level.
While this may perhaps be an effect of feuding, the term “function” implies that feuds
can be explained by that effect. But this is getting thing backwards: science explains
phenomena by their causes, not by their effects.
The functionalist ideas of Michel Foucault and Pierre Bourdieu retain an immense
and immensely pernicious influence on French social science. According to these
writers, social practices of the most varied kind can be explained by their tendency to
maintain the hegemony of dominant groups. If some practices appear to have a
different effect, that fact is itself explicable by the “need of the system” to maintain a
facade of impartiality. This functionalist mind-set is readily identified by the pervasive
use of verbs (or verbal nouns) without subjects, as in the statements I have italicized in
the following excerpt from Foucault’s enormously influential Discipline and Punish:
But perhaps one should reverse the problem and ask oneself what is served by the
failure of the prison: what is the use of these different phenomena that are
continually being criticized; the maintenance of delinquency, the encouragement
5
Anyone reading Sigmund Freud’s original work might well be seduced by the
beauty of his prose, the elegance of his arguments and the acuity of his intuition.
But those with a grounding in science will also be shocked by the abandon with
which he elaborates his theories on the basis of essentially no empirical
evidence. This is one of the reasons why Freudian-style psychoanalysis has long
since fallen out of fashion: its huge expense – treatments can stretch over years –
is not balanced by evidence of efficacy (Nature editorial October 15 2009)
The fact that the memories of victims and witnesses can be false or inaccurate
8
even though they believe them to be true has important implications for the legal
system and for those who counsel and treat victims of crimes. Some
psychotherapists use techniques that are suggestive (along the lines of, «you don’t
remember sexual abuse, but you have the symptoms, so let’s just imagine who
might have done it»). These can lead patients to false beliefs and memories,
causing great damage to the patients themselves and to those who are accused. In
one Illinois case, psychiatrist Bennett Braun was accused by his patient, Patricia
Burgus, of using drugs and hypnosis to convince her that she possessed 300
personalities, ate meat loaf made of human flesh and was a high priestess in a
satanic cult. By some estimates, thousands of people have been harmed in similar
ways by well-meaning providers who apply a ‘cure’ that ends up being worse
than the disease. (Elizabeth Loftus)
In a Norwegian case, a father was accused of sexual abuse of his daughter. On the
basis of her statements, an expert psychologist testified that the sharp fence posts in the
child’s drawing of a house surrounded by a fence very likely had a sexual significance.
She affirmed, moreover, that the number of posts in the fence very probably indicated
the number of occasion on which the child had been abused. The child’s father spent
two weeks in jail, in a security cell, was barely acquitted of incest, but his life was
ruined. Later, the child confessed that it was all an invention. The alleged perpetrator
was the victim.
It goes without saying that in one sense, Marxism has caused untold harm.
Whether Marxism as a theory, rather than a practice or a rationalization for actions
undertaken on other grounds, has done actual harm is not obvious. The Stalinist idea of
“objective complicity”, while consistent with some functionalist aspects of Marxism,
probably reflects a very general tendency of the human mind. The more specific idea
of “crisis maximization” or la politique du pire – things have to get worse (be made
worse) before they can get better – does perhaps have a more direct Marxist ancestry.
9
Here again, however, there are many non-Marxist examples of this strategy. After
1790, Louis XVI used it, for instance. On a more general level, however, Marx’s
successors took over his intellectual hubris and notably his belief in an historical
progression with a desirable, inevitable and foreseeable outcome. The most serious
charge that can be laid against Marx is that his contempt for what he called “bourgeois
rights” probably had a direct influence on the massive human rights violations that his
successors carried out.
Let me conclude this discussion of soft obscurantism by a provocative question:
might the science of man, as well as society itself, have been better off if Marx and
Freud had never lived? I am not prepared to argue for an affirmative answer to this
question, but I do not think it can be dismissed out of hand.
I now move on to discuss what for many will be more controversial and surprising
claims about hard obscurantism.
There have been many attempts to explain human behavior using quantitative
models. The most prominent ones are perhaps rational-choice theory, including game
theory, agent-based modeling, and network theory. What little I know about the last
two tends to make me skeptical, but I cannot make any strong claims about them. A
long familiarity with rational-choice models has, however, fostered a skepticism that
is, I hope, better grounded.
Before I state the grounds for my skepticism of rational-choice theory, let me
emphasize that for some purposes it is highly valuable. In my opinion, the
development of rational choice theory, as a conceptual tool for understanding human
action and interaction, has been the greatest achievement of the social sciences to date.
I shall give some examples later. It is, moreover, an indispensable policy tool for
changing behavior, for the simple reason that people usually respond to incentives in
predictable ways. What I deny is that it has a privileged status as an explanatory tool.
To achieve explanatory success, a theory should, minimally, satisfy two criteria: it
10
should have determinate implications for behavior, and the implied behavior should be
what we actually observe. These are necessary conditions, not sufficient ones.
Rational-choice theory often fails on both counts. The theory may be indeterminate,
and people may be irrational. In what was perhaps the first sustained criticism of the
theory, Keynes emphasized indeterminacy, notably because of the pervasive presence
of uncertainty. His criticism applied especially to cases where agents have to form
expectations about the behavior of other agents or about the development of the
economy in the long run. In the wake of the current economic crisis, this objection has
returned to the forefront. Before the crisis, going back to the 1970s, the main
objections to the theory were based on pervasive irrational behavior. Experimental
psychology and behavioral economics have uncovered many mechanisms that cause
people to deviate from the behavior that rational-choice theory prescribes.
Disregarding some more technical sources of indeterminacy, the most basic one is
embarrassingly simple: how can one impute to the social agents the capacity to make
the calculations that occupy many pages of mathematical appendixes in the leading
journals of economics and political science and that can be acquired only through
years of professional training? Here are some frequently made answers to this
question, with my rejoinders:
1. We should accept a theory if it produces correct predictions even if we do
not understand how it does so (quantum mechanics). Rejoinder: the social
sciences produce very few correct predictions, except for the short-term
impact of small changes in economic variables.
2. Even though agents are incapable of intentional maximizing in complex
situations, natural or social selection will eliminate non-maximizers.
Rejoinder: this is mere handwaving. No explicit model exists.
3. Individual errors will cancel each other in the aggregate. Rejoinder: there
is no reason to think that errors are symmetrically distributed around the
correct answer.
4. An expert billiard player whose experience helps him figure out the angles
would be utterly incapable of solving the relevant equations; yet he acts as if
he could. Rejoinder: nobody can be an expert in all the situations economists
11
With regard to irrationality, I shall only produce a list of robust mechanisms that
have been shown to generate behavior that violates rational-choice theory:
Many scholars – not just rational-choice theorists – find the rapidly expanding
array of mechanisms uncomfortably large. Some are also worried by the fact that
12
with the data. You have to make assumptions in order to make progress. You
have to give the models the benefit of the doubt. Where’s the harm? (David
Friedman.)
I shall return to the question of harm. First, however, there is the issue of
waste. I believe that much work in economics and political science that is inspired
by rational-choice theory is devoid of any explanatory, aesthetic or mathematical
interest, which means that it has no value at all. I cannot make a quantitative
assessment of the proportion of work in leading journals that fall in this category,
but I am confident that it represents waste on a staggering scale. I believe, but less
confidently, that the same is true is true of many instances of data analysis.
The amount of harm is more difficult to assess. I do not know of any detailed
study discussing the importance of modeler hubris in the 1998 collapse of Long
Term Capital Management, costing investors 4.5 billion dollars, or in the current
financial crisis. Greed, short-termism and deregulation may have been more
important than unwarranted confidence in the Nobel-prize-winning models.
Although one can cite many examples of fund managers telling their clients that
according to their models a crisis of the magnitude of what has happened since
2007 would only occur once in n years, n being some very large number, it
remains to be shown that these managers actually believed in the models and used
them as decision premises. After all, as we have learned, they had very little to
lose if the models got it wrong.
This being said, I find it hard to believe that excessive belief in the efficiency
of markets and the rationality of market participants did not play some role in
generating the crisis. Since the information that is reflected in prices is a public
good, nobody has an incentive to produce it. This free-rider problem generates
mechanical diversification of assets as a substitute for due diligence. The “Warren
Buffett remedy”, similar to the “shoe-leather sociology” that David Freedman
15
In 1975, Isaac Ehrlich's analysis of national time-series data led him to claim
that each execution saved eight lives. Solicitor General Robert Bork cited
Ehrlich’s work to the Supreme Court a year later, and the Court, while
claiming not to have relied on the empirical evidence, ended the death
penalty moratorium when it upheld various capital punishment statutes in
Gregg v. Georgia and related cases.
Sunstein and Adrian Vermeule defend the death penalty on these grounds. To be
sure, they qualify their argument as a conditional one, valid only if the empirical
claims stand up. Yet they display considerable hubris when, as legal scholars with
no econometric training, they assert that “One leading study finds that as a
national average, each execution deters some eighteen murders” and proceed to
discuss the implications. If they had read David Freedman, they would have
understood that the idea of executing people on the basis of any kind of statistical
analysis is morally unacceptable. There is just too much uncertainty about the
facts to warrant this irreversible action.
If you are willing to grant me the claim that soft and hard obscurantism are
sufficiently widespread to justify worry, you may have asked themselves about
their causes. As I said at the outset, the causes may be psychological,
sociological, or institutional.
At the psychological level, hard obscurantism may be due to the widespread
belief that “social science” can or must become a science, on the model of the
natural sciences. While talk about “physics-envy” may be too strong, there may
be an unconscious desire to emulate the most prestigious scientific disciplines.
The highly sophisticated theories hold out the promise of satisfying this desire. At
a more mundane level, the high prestige of mathematical social science goes
together with very high salaries. A young scholar with a talent for mathematics
may easily be lured by the prospect of rising to the top of the profession. At the
same time, the normal workings of self-deception may prevent him from
understanding that he is weaving a web of froth. Finally, self-selection may cause
recruitment into these professions to be biased in favor of scholars who are
subject to the relevant forms of hypertrophy and atrophy to begin with.
To understand the psychological roots of soft obscurantism, I suggest that we
turn to the phenomenology of scholarship. When we believe we have found the
17
could be, groups of parents came together to pledge in public that they would not
bind the feet of their daughters nor marry their sons to women whose feet were
bound.
Mind-binding perpetuates hard obscurantism because departments have to
train their students in abstruse mathematical techniques. If they don’t, the students
will not get a job in elite departments, just as Chinese women without their feet
bound could not get married. Mind-binding differs from foot-binding, however, in
an important respect. Even before foot-binding was abolished, it was widely
perceived as absurd and perverse, a perception that eventually led to its demise.
By contrast, most chairs of departments of economics and political science
probably do not perceive themselves as being in a bad equilibrium. To the extent
that individuals scholars are seized with occasional doubts, as, being human, they
can hardly fail to be, a look at what their colleagues are doing may assuage their
worries or at least prevent them from speaking up. This remark brings me to the
second interaction-based mechanism, pluralistic ignorance.
This idea dates from 1835, when Hans Christian Anderson published his tale
about the “Emperor’s New Clothes”. It was given a more theoretical formulation
five years later, in the second volume of Tocqueville’s Democracy in America,
and then rediscovered by psychologists in the early nineteenth century. In an
extreme case, pluralistic ignorance obtains when no member of a community
believes a certain proposition or espouses a certain value, but each believes that
everybody else holds the belief or the value. In the more common case it obtains
when only a few members hold the belief or the value in question, but most of
them believe that most others do.
In the case of economic and statistical models, pluralistic ignorance would
obtain if each scholar, although secretly worried about the fragility or irrelevance
of the procedures, kept quiet because of the perception that his colleagues are
19
firmly convinced of their validity. There are several mechanisms that might be at
work here. From my own experience I know very well how a scholar’s
confidence in his own judgment can be undermined by the fact that the majority
thinks differently. How could all these people, who are certainly smarter than I
am, be so wrong? Also, even with unshakeable self-confidence a scholar might
worry that speaking up might cause ostracism and career obstacles.
Finally, hard and soft obscurantism will be hard to eradicate because of recent
institutional developments. Today, the scientific community increasingly seeks to
base tenure decisions and funding decisions on quantitative criteria, notably on
citation counts. To the extent that these criteria are used together with qualitative
criteria, which are based on some competent evaluators actually reading the
literature, this development may not be a bad thing. There is a real danger,
however, that high citation numbers are now turning into a sufficient criterion for
excellence. Since hard obscurantistst tend to cite other hard obscurantists, and soft
obscurantists tend to cite other soft obscurantists, and since there are plenty of
scholars in each camp, they can perpetuate themselves as a group by citing one
another.
The high citation counts of soft obscurantists are especially striking. If we
look at the number of citations on Google Scholar, we find the following
numbers:
These are rough numbers. Different search engines give different result. Yet
the order of magnitude is the same.
The first five are soft obscurantists. The last five are not obscurantist at all,
but represent what I believe to be the best or among the best scholars in,
respectively, psychology, economics, political science, sociology and
anthropology over the last half century. As you can see, not even the best serious
scholars can compete with even a marginal soft obscurantist.
By citing the last five names, I have proposed an implicit definition of non-
obscurantist social science. An explicit definition is hard to give. I could simply
say that good social science respects facts, logic, and common sense. But I shall
try to go beyond that trite statement, by offering some examples of simple, robust
causal mechanisms.
I have already cited two examples: the idea of a bad equilibrium, and that of
pluralistic ignorance. A third example can be taken from Tocqueville’s
Democracy in America.
Our fathers had a peculiar opinion of marriage. Having observed that what
few marriages of inclination were made in their time almost always badly,
they resolutely concluded that it was very dangerous to consult one’s own
heart in such matters. Chance struck them as more clairvoyant than choice
21
He then goes on to show why this conclusion was unjustified. Suppose that in
an aristocratic society two young people marry for love. By going against the
current they will tend to encounter the hostility of their friends and relatives, a
situation that, as Tocqueville says, “soon breaks their courage and embitters their
hearts”. In a society in which this practice was general, this effect would not arise.
He also points to another mechanism that explains why we cannot generalize from
exceptional cases to the general case. For a man to marry for love in societies in
which this practice is uncommon he must have, in Tocqueville’ words, “a certain
violent and adventurous cast of mind, and people of this character, no matter what
direction they take, rarely arrive at happiness or virtue”. And one might add that a
marriage of two people of this disposition is even less likely to be happy. For me,
this argument is a model of how to think about society.
A further example is provided by an analysis of the puzzling fact that people
sometimes burn their bridges or their boats. It would seem that no rational
individual would ever throw away a resource that might prove useful. Any such
behavior must be irrational. Yet Thomas Schelling showed that it can make
perfect sense. Suppose that the general of Country I is debating whether to invade
Country II to take possession of a contested piece of territory:
If both generals are rational, Country I will attack and Country II will retreat.
Yet if the general of Country II burns his boats or his bridges, making retreat
22
impossible, he will have to fight. Because the general of Country I can anticipate
this reaction, he will choose to negotiate. In other words, by discarding one of his
options, the general of Country II can achieve a better outcome. Sometimes, less is
more.
The analyses of bad equilibria and of the rationality of burning one’s bridges
rely on game theory, one of the most powerful tools of the social sciences. It is the
perfect tool for understanding the conceptual structure of human interaction. It can
help us understand why people sometimes do not cooperate even when it is in
their objective interest to do so, and why in other situations they do manage to
cooperate. Game theory is only useful, however, to the extent that it does not
impute excessive cognitive abilities to social agents. When it does so, it ceases to
be social science and turns into science fiction, or hard obscurantism.
Thomas Schelling once told me that before he wrote his path-breaking work
The Strategy of Conflict, in which he showed how it could be rational to burn
one’s bridges, he spent a year or two reading randomly in military history. I
believe in fact that history should be one of the two unifying disciplines of the
social sciences. By virtue of their knowledge, good historians can pick out the
“telling detail” as well as the “robust anomaly”, thus providing both stimulus and
reality check for theorists. The other unifying discipline is what one might call the
science of choice. Today the study of choice is divided among psychology, micro-
economics, and behavioral economics. In the future I believe these will fuse into a
single discipline, relying on laboratory experiments as well as on field studies.
If I am right, my conclusions also have implications for graduate training in
the social sciences. Today, students of sociology have to read Durkheim, Marx
and Weber; students of economics read Adam Smith and Ricardo; students of
political science read Aristotle, Locke, Hobbes and Rousseau. Without denying
that these readings can be useful, I find it scandalous that students are not also
required to read works in social history, economic history, and political history. At
the same time, they should learn how to do laboratory experiments and field
23