1
Against Cognitivism
I
Writing about the multifaceted legacy of Jerome Bruner, Clifford Geertz has us contemplate the centrifugal history of psychology. Beyond the predictable disagreements of a formerly young science, one can hardly miss a certain tendency to walk on incongruent theoretical paths. This sometimes happens at the same historical moment, or in the same person, Bruner himself being such an example. Geertz nods to Pirandello: From the outside at least, psychology
looks like an assortment of disparate and disconnected enquiries classed together because they all make reference in some way or other to something or other called âmental functioningâ. Dozens of characters in search of a play.
(Geertz 2001, 19)1
That this picture is only available to someone who is not captive to the problem, or the apparent problem, of claiming a unitary domain for psychologyâin other words, of circumscribing the mentalâwill not be surprising. But, in only being fit for so serene a person, the image probably downplays the inside perspective of those who staged one or another volte-face in this field, and the substantiality of the stakes. The pressure to uncover a common dialect in the Babel of psychology has been real, and a part of the quest for scientific citizenship. Consonantly, psychology has suffered not only ruptures but also revolutions. It is not only that people decided to go their own way in their research and manner of describing and explaining findings, as one would, perhaps, in the humanities; parties have repeatedly argued that their way was the way of getting things right for everybody doing psychology.
As if echoing Ryle, who compares psychology with medicineââthe name of a somewhat arbitrary consortium of more or less loosely connected inquiries and techniquesâ (1949, 323)âGeertz suggests that, with Bruner turning psychology cultural, one arrives in the vicinity of anthropologyâthe âother hopelessly miscellaneous and inconstant scienceâ (2001, 23). Perhaps, but one need not remind oneself of Brunerâs own lament at the beginning of Acts of Meaning about the fragmentation of psychology (1990, ixâx), or of his earlier career as a founder of the cognitive revolution, to understand that psychology does not seem to be comfortable with such a self-image. As chromosomes during cell division, this uneasiness becomes visible at times of change. For example, the two major turning points in twentieth-century academic psychology2âthe behaviorist movement, which dominated the first half of the century and which at the end of the 1940s was still considered revolutionary by Ryle (1949, 328), and the cognitive revolution of the 1950sâhave both been immodest in scope.
In the case of behaviorism, we tend to recognize this as a mistake. We also recognize that one characteristic way in which immodesty manifested itself was philosophical speculation. Here, I am not only referring to the pointless attempts to translate all suspect vocabulary in behaviorist newspeak. Hawkish dispositions are also exemplified by the formulaic rendering, âsolving,â or rejection of classical topics (thinking, learning, deviance, human nature), by pop-science campaigning, or by the âdeductionâ of overly ambitious sociopolitical consequences from oneâs psychology.3 This philosophical halo has since evaporated, as few people continue to think that the rigor of behaviorism simply is scientific rigor in psychological clothes. There are standards of scientific respectability that a research program has to meet, and this often is an issue worth discussing, but one of the clearest signs of conceptual hubris is the parading of candidate scientific commandments as religious relics. The significance of this should not be obscured by the inevitable messiness of empirical research. Perhaps no scientific program can live up to its own rhetoric, but in the case of behaviorism the rhetoric itself should beâand has beenâthe subject of legitimate worries. The philosophical halo inherent in this rhetoric was symptomatic for the mutation of the behaviorist research program into an ideology of the mind.
If this summary and unexceptional diagnosis of what went wrong with behaviorism is correct, what can one say about the kind of psychology which emerged after the cognitive revolution? To continue with the language I used above, has cognitive psychology been not only a critique of behaviorism, but also a successful critique of ideology? In asking this question, I propose to contribute to an ongoing argument to the effect that, in the wake of the cognitive revolution, the now-dominant conception of the mind has itself become an ideology. The core of this ideology (which we might call âcognitivismâ) is the computer metaphor as a model for mental states and processes, and the idea that the language of psychology is to be made explanatorily good by being shown to be, at bottom, a language of computation. More recently, this language is thoroughly colored by the attention given in the field to neural loci where said computation presumably takes place, but this does not change the nature of the problem. Like behaviorism, cognitivism unifies the mental by providing a criterion for it (computation), but unlike at least some strands of behaviorism, cognitivism also reifies the mental by requiring computational substance from its explanatory terms. We recognize mental phenomena by the fact that they are computational phenomena involved in the control of behavior; what people call, for example, âbeliefâ qualifies as mental because it can be given a computational description (unification). Moreover, belief is such and such a computational pattern (reification).
In the cognitive sciences, this view is, more or less, the foundation of the administrative jargon, and it is taken to express a scientific breakthrough, indeed, the definitive mark of doing away with the dogmatism of behaviorists. While acknowledging at all times the progress made in psychology in the last decades, the areas most involved in the cognitive sciences here included, I think it is a mistake to associate this progress with cognitivism. Cognitivism is in fact a source of confusion in thinking both about explanation in psychology, and about the folklore of the mind that has been organizing our lives since forever. As with the behaviorists, or with Freud, a Copernican revolution has not taken place in psychology, advertisements to the contrary notwithstanding.
Now, even if this is true, we need to start by recognizing that we are far from such an accepted diagnosis for what followed the cognitive revolution, although the idea of ideological continuity is not particularly original.4 Even the worries expressed by some of the very people who founded the cognitive sciences in the 1950s, such as Bruner (1990) or Chomsky (2000), remain peripheral. Philosophical criticism seems even less influent. There have by now been numerous such warnings, involving very different threads of thought, from Searleâs rejection of the syntactic engine metaphor (e.g., 2003a), to Putnamâs change of mind about functionalism (e.g., 2001, especially chapter 5), or to the criticisms on Wittgensteinian lines put forward by authors such as P.M.S. Hacker (e.g. 2007) and Stuart Shanker (1998). These criticisms do not all carry significant bite, perhaps, and may be plagued by problems of their own. But added to inside worries they should give some pause to the idea that the computer metaphor is a particularly illuminating stance on what minds are. Philosophical worries, especially when expressed by practicing scientists, should not share the fate of philosophy as a disciplineâoften quietly escorted to the corner of the formerly reputable, but now silly, seniors.
This barrage of skepticism, one should note, cannot be uniformly assimilated to (and dismissed as) that which naturally fits those who conceive of psychology as closer to the social sciences and even to the humanities than to physics or biology. Predictably, cognitivism could not be considered an authentic renaissance of the mind from such a perspective, as we will see later in the book. But clearly one need not agree with this softer view of psychology and the unsurprising opposition to the mechanization of the mind which it has generated to be critical of cognitivism. Let psychology be as natural a science as it pleases; one can still resist the idea that psychology is scientific to the extent that it is cognitive in the sense of mentalistic and computational. This criticism which originates, as it were, closer to home is more damaging. Nevertheless, for the regions of the academic world that are the inheritors of the cognitive revolution this does not seem to be a time of insecure self-reflection. Maybe the thought is that, since the mind returned to the scientific picture, resulting psychologies are immune to the ills of behaviorism. Part of this may very well be true, but attention is needed to the part which is not.
One way to give credibility to the view that there should be parallel worries about the behaviorist and the cognitivist attempts to explain everything mental with a single set of tools is to notice that some of the more excessive claims of behaviorism were not results of the specific posits of this school, but of a deformed, and historically endemic, image of what counts as scienceâof what psychology should strive for. If physics is the science of everything physical, whatever that may be, then psychology should be the science of everything psychical, whatever that may be. If physics aims to explain everything physical in terms of particles, elementary forces, and so on, then psychology should explain everything psychical in terms of whatever the psychical equivalents of particles and forces may be.5 The point then is not that the cognitive revolution failed in its attempt to reverse the specific misconceptions of behaviorism, but that it inherited, like behaviorism before it, a hollow ideal. The revolution was not, in this sense, revolutionary enough.
It is telling that, if radical behaviorists felt they had to ostracize the âmentalâ for not being scientific, Jacobin cognitivists arranged for its return, but only with the naturalization appropriate for an alien. This embarrassment deserves, perhaps, more therapy than it has received, and it is unnecessary. Nothing tragic should follow for psychology if the mind unravels, if it turns out to be different kinds of things which we understand in various ways, not all of which, to say the least, are compatible with the explanatory language of cognitivism. The unity of the mental, like that of psychology, has always been, after all, illusory, and a sign of mauvaise foi. If I am not mistaken then, we continue to be in the position of warning that the appetite for theorizing about everything mental in one move constitutes a symptom of philosophical addiction rather than scientific virtue.
The existence of many traditions within psychology is itself the proximal antidote of grand generalizations and a proof that even the best ideas and the most convincing explanations and analogies do not naturally extend to all mental phenomena. Important differences, for example, can easily be found even among the various approaches to the mental life of typical, mature humansâpsychologyâs paradigmatic6 subject matter. Consider, for example, the important differences between social psychology and the schools of cognitive psychology focused on the isolated individual.7 Cognitive dissonance (Festinger, Riecken, and Schachter 2008) and the magical number seven (Miller 1956) were born in the same year (1956), but they do not belong in the same conceptual habitat. Nor is it obvious that there should be a hierarchical relationâreduction, analysisâbetween such concepts. Manifestos aside, it does not add much to the discussion to ask which kind of concepts (or which branch of psychology) is more fundamental. This only becomes a problem if one starts by stipulating that the mind must be, for all intents and purposes, a computer, or a neural network, or an electrochemical engine. But this is to get things upside down, since, as a matter of historical and conceptual fact, we do not start, even in psychology, with a regimented concept of the mental. The mental is already in front of us, as a matter of our nature and socialization. We need psychology to study it, not to identify it.
It is enough to accept that the fragmentation of psychology is neither accidental, nor a pubertal detail, to be en route to a criticism of cognitivism which may even prove constructive. To confront cognitivism head-on would be a mistake, and in more than one sense. One should not get carried away by a certain natural weariness about a discourse that happens to be dominant at a certain moment. There is a legitimate task of exposing the limits of received wisdom, but this goal should be accomplished with discernment and charity. It would be a grave error to think that one can sort decades of research into valuable and worthless with a few conceptual remarks. Here, the aim will be to erode those philosophical commitments which I consider ideologicalâa way of talking about findings, not their authenticity, will be the eventual target.
A partial and contextualized engagement is, in practice, unavoidable since objections need to be specific, even if their systemic significance is what interests us most, and since different regions of cognitive psychology have been made more or less vulnerable by cognitivism. It leads nowhere to measure a colossal theoretical edifice with an overstated disapproval. Moreover, even if there is some sort of urgency in being lucid about the dominant psychology of our day, it should not be forgotten that it is a general tendency to reify and unify the mental which forms the horizon of this criticism. The commitments in question embody philosophical dispositions that have affected psychology throughout its historical development. This is why, in the following, time will be spent describing at least in part the intellectual past of the puzzles we face today. They are not new.
II
History, then, will be our ally in the effort to emphasize the plurality of psychology. As in most situations, a historical investigation might be the best one can do to illuminate the present. Here, however, I will only be able to offer a fragmentary historical overview. Most of the weight of the criticism must rest elsewhere, in how these fragments are put to use. One can capitalize on there being, and there having been, many kinds of psychology in more than one way. Geertzâs ruminations, to which I gestured above, constitute one such example. In the following, I propose to rely on what may be called a âtopologicalâ argument.
The fragmentation of psychology preserves some of the pre-theoretic fault lines of its subject matter. In the Introduction, I prosed the following perspective on this fragmentation: there is an equator of psychological preoccupationsâthe typical, mature, socially competent human beingâand there are also marginal regions. Our self-understanding is grounded in our life at the equator, in our familiarity with, or being at home in, our social universe. Making sense of what people do and why, of what they say and think, begins here, in the ordinary. The concepts that allow for understanding others and ourselvesâbelief, intention, motive, and so onâbelong in this social environment; they are given meaning by being regularly used as part of familiar and shareable practices: interpretation, answering questions, prediction, giving reasons, finding excuses, and so on.
As we move further away from the ordinary, none of the above can be taken for granted; we arrive at the margins of psychology. The margins are to a certain extent a matter of context. The notion, as already explained, is meant to capture a gradual erosion of confidence or transparency in making sense of others; it does not carry an ethical judgment. The figures of the infant (not yet fully developed human) and of the madman (beyond shared human life) can be safely situated in these outer regions, but they are not, by far, the only such examples. This is a geography psychology inherits from the nonscientific world essentially due to the fact that it takes over the conceptual repertoire of the âfolk.â The common psychological parlance or âfolk psychology,â even if notoriously laissez-faire in use, is mainly about what typical adults feel, think, or doâand why. Of course, we talk in mentalistic terms even when we are quite far from the paradigm, and this includes our reflections on the mental lives of small children and of bizarre fellow humans, as well as those about more distant characters, such as other animals, machines, and the inhabitants of philosophical thought experiments. But even if we do ordinarily talk about all of them in âfolk psychologicalâ terms, we cannot claim the same degree of confidence as in the case of typical, mature individuals. Puzzling situations are to be expected, since they come with the territory. We know full well that the soil can suddenly turn slippery, that we can exceed at any moment the explanatory grip of everyday discourse. We thought the child was scared of the dog when she ran, but now she hugs it; well, who knows what she thought. We cannot ask the child and, even if we could, if she is very young, we would not consider her report decisive. The question is if anything else can help settle the case. The notion of margin should also capture the claim that this question must be answered in the negative. The idea of the gradual erosion of folk psychological explanations (or simply put of understanding), at least in the case of bizarre behavior, has been on the philosophical agenda, notably on that of the radical interpretation/translation literature, and also on that of the eliminativists (Stich 1983, Churchland 1996). It has been observed, for example, that our interpretive capacities eventually break down when facing behavior that cannot be assimilated to any rational paradigm we recognize. Or that the impotence (and perhaps the indifference) of folk psychological âtheorizingâ in the case of mentally atypical individuals is evidence as to its being a terrible theory of the mind. I will try to put the idea to a different use. To the extent that we are successful in the remote regions, according to these threads of literature, we manage to illuminate them as âmore-of-the-same.â So, for example, we impose our logic upon the native, as Quine8 has it, or we hope for a mature science of the brain9 that will explain behavior both typical and atypical, both verbal and infraverbal, as the eliminativists repeatedly advised. This âmore-of-the-sameâ picture does a poor ...