In Praise of the Cognitive Emotions

Originally presented as a special lecture in May 1976 at the 129th annual meeting of the American Psychiatric Association, this paper was also given as a special lecture in 1976 at Brooklyn College. A variant constituted the John Dewey lecture for 1977, sponsored by the John Dewey Society. Versions of the paper were also presented at Harvard University, The University of Notre Dame, Brandeis University, and Long Island University.

The mention of cognitive emotions may well evoke emotions of perplexity or incredulity. For cognition and emotion, as everyone knows, are hostile worlds apart. Cognition is sober inspection; it is the scientist's calm apprehension of fact after fact in his relentless pursuit of Truth. Emotion, on the other hand, is commotion—an unruly inner turbulence fatal to such pursuit but finding its own constructive outlets in aesthetic experience and moral or religious commitment.

Strongly entrenched, this opposition of cognition and emotion must nevertheless be challenged for it distorts everything it touches: Mechanizing science, it sentimentalizes art, while portraying ethics and religion as twin swamps of feeling and unreasoned commitment. Education, meanwhile—that is to say, the development of mind and attitudes in the young—is split into two grotesque parts—unfeeling knowledge and mindless arousal. My purpose here is to help overcome the breach by outlining basic aspects of emotion in the cognitive process.

Some misgivings about this purpose will, I hope, be allayed by a preliminary word. My aim, to begin with, is not reductive; I am concerned neither to reduce emotion to cognition nor cognition to emotion—only to show how cognitive functioning employs and incorporates diverse emotional elements—these elements themselves acquiring cognitive significance thereby. I am emphatically not suggesting that cognitions are essentially emotions, or that emotions are, in reality, only cognitions. Nevertheless, I hold that cognition cannot be cleanly sundered from emotion and assigned to science, while emotion is ceded to the arts, ethics, and religion. All these spheres of life involve both fact and feeling; they relate to sense as well as sensibility.

Secondly, though applauding the cognitive import of emotions, I do not propose to surrender intellectual controls to wishful thinking, nor shall I portray the heart as giving special access to a higher truth.1 Control of wishful thinking is utterly essential in cognition; it operates, however, not through an unfeeling faculty of Reason but through the organization of countervailing critical interests in the process of inquiry. These interests of a critical intellect are, in principle, no less emotive in their bearing than those of wayward wish. The heart, in sum, provides no substitute for critical inquiry; it beats in the service of science as well as of private desire.

Finally, I concede it to be undeniable that certain emotional states may be at odds with sound processes of judgment and decision making. Overpowering agitations may derail the course of reasoning; greed, jealousy, or lust may misdirect it; depression or terror may bring it to a total halt. Conversely, the effect of rational judgment may well be to moderate, even wholly to dissipate, certain emotions by falsifying their factual presuppositions: Anger fades, for example, when it turns out the injury was accidental or caused by someone other than first supposed; fear evaporates when the menacing figure becomes the tree's dancing shadow. It does not follow from these cases, however, that emotion as such is uniformly hostile to cognitive endeavors, nor may we properly conclude that cognition is, in general, free of emotional engagement. Indeed, emotion without cognition is blind and, as I shall hope particularly to show in the sequel, cognition without emotion is vacuous.

EMOTIONS IN THE SERVICE OF COGNITION

Considering now the various roles of emotion in cognition, I divide the field, for convenience, into two main parts, the first having to do with the organization of emotions generally in the service of critical inquiry, and the second having to do with specifically cognitive emotions. Under the first rubric I shall treat: (a) rational passions,2 (b) perceptive feelings, and (c) theoretical imagination, and I turn first to the rational passions, that is to say, the emotions undergirding the life of reason.

Rational Passions The life of reason is one in which cognitive processes are organized in accord with controlling rational ideals and norms. Such organization involves characteristic patterns of thought, action, and evaluation comprising what maybe called rational character. It also thus requires suitable emotional dispositions. It demands, for example, a love of truth and a contempt of lying, a concern for accuracy in observation and inference, and a corresponding repugnance of error in logic or fact. It demands revulsion at distortion, disgust at evasion, admiration of theoretical achievement, respect for the considered arguments of others. Failing such demands, we incur rational shame; fulfilling them makes for rational self-respect.

Like moral character, rational character requires that the right acts and judgments be habitual; it also requires that the right emotions be attached to the right acts and judgments.3 "A rational man," says R. S. Peters, "cannot, without some special explanation, slap his sides and roar with laughter or shrug his shoulders with indifference if he is told that what he says is irrelevant, that his thinking is confused and inconsistent or that it flies in the face of the evidence."4 The suitable deployment in conduct of emotional dispositions such as love and hate, contempt and disgust, shame and self-esteem, respect and admiration indeed defines what is meant, quite generally, by the internalization of ideals and principles in character. The wonder is not that rational character is thus related to the emotions but that anyone should ever have supposed it to be an exception to the general rule.

Rational character constitutes an intellectual conscience; it monitors and curbs evasions and distortions; it combats inconsistency, unfairness to the facts, and wishful thinking. In thus exercising control over undesirable impulses, it works for a balance in thought, an epistemic justice, which requires its own special renunciations and develops a characteristic cognitive discipline. There is, however, no question here of the control of impulses through a "bloodless reason,"5 as control is exercised through the structuring of emotions themselves. Rationality, as John Dewey put it,

is not a force to evoke against impulse and habit. It is the attainment of a working harmony among diverse desires. . . . The elaborate systems of science are born not of reason but of impulses at first slight and flickering; impulses to handle, move about, to hunt, to uncover, to mix things separated and divide things combined, to talk and to listen. Method is their effectual organization into continuous dispositions of inquiry, development and testing. It occurs after these acts and because of their consequences. Reason, the rational attitude, is the resulting disposition. . . . The man who would intelligently cultivate intelligence will widen, not narrow, his life of strong impulses while aiming at their happy coincidence in operation.6

This coincidence, I emphasize, requires appropriate organization of feelings and sentiments in the interests of intelligent control.

PERCEPTIVE FEELINGS

Having seen the role of emotions in the internalization of rational norms, let us consider now their employment in perception. For they are not only interwoven with our cognitive ideals and evaluative principles; they are also intimately tied to our vision of the external world. Indeed they help to construct that vision and to define the critical features of that world.

These critical features—however specified—are the objects of our evaluative attitudes, the foci of our appraisals of the environment. Our habits and judgments are keyed in to these appraisals; we define ourselves and orient our action in the light of our situation as appraised. Characteristic orientations are associated with distinctive emotional dispositions, and both involve seeing the environment in a certain light: Is it, for example, beneficial or harmful, promising or threatening, fulfilling or thwarting?7 The subtle and intricate web relating adult feeling and orientation to adult perception of the environment is a product of evolutionary development, to be sure, but also of the special circumstances of individual biography. Acquiring human significance through biographical linkage with critical features of the environment, our feelings come indeed to signify—to serve as available cues for interpreting the situation.

Fear of a particular person, for example, presupposes that that person is regarded as dangerous—danger being a critical feature of the environment calling for a special orientation in response. There need, however, be no independent evidence, in every case, of the threat we sense: The characteristic feeling that has become associated for us with past dangers itself serves us as a cue. Interpreting that feeling as fear, we at once characterize our own state and ascribe danger to the environment. Indeed, we may thence proceed to an explicit attribution of danger, prompted by cues of feeling. Pursuing a more abstract direction in forming our cognitive concepts, we may, further, come to describe a certain situation as terrifying, ascribing to it, independently of our own state, the capacity to arouse fear. Thus employing the emotions as parameters, we gain enormous new powers of fundamental description, while abstracting from actual conditions of feeling.

The notion that aesthetic experience, for example, is peculiarly and purely a matter of emotion ignores such manifold connections of feeling and fact—both fact as embodied in the art work and fact as represented therein. Relative to the latter, H. D. Aiken writes,

Just as in ordinary circumstances an emotional response is the product of a perceived situation which is apprehended by the individual as promising or threatening, so the expressiveness of an imaginative work arises, at least in part, from the fact that it provides a dramatic representation of an action of which the evoked emotion is the expressive counterpart. And such a representation must be understood as such if the expressive values of the work are to become actual; without it such emotion as the observer might experience would have no ground, and if, by a miracle, it could be sustained, it would still remain the private, dumb, inexpressive importation of the observer himself. As such it would be nothing more than an accidental, adventitious subjective coloring which, having no artistic basis in the thing perceived, would be devoid of aesthetic relevance to it. Aesthetically relevant emotion in art is something which is expressed to us by the action or gesture of the work itself; it is something aroused and sustained by the work as an object for contemplation, and it is found there as a projected quality of the action.8

That the emotion is thus tied to a representational understanding of the work of art does not imply, however, that this understanding must be antecedently fashioned, in complete isolation from the feelings. This point must be especially emphasized since the familiar notion of the work of art as "an object for contemplation" may carry contrary, and therefore misleading, connotations. In fact, I believe, the very feelings through which we respond to the content of a work serve us also in interpreting this content. Reading our feelings and reading the work are, in general, virtually inseparable processes.

The cognitive role of the emotions in aesthetic contexts has been emphasized by Nelson Goodman in a recent discussion. He writes,

The work of art is apprehended through the feelings as well as through the senses. Emotional numbness disables here as definitely if not as completely as blindness or deafness. Nor are the feelings used exclusively for exploring the emotional content of a work. To some extent, we may feel how a painting looks as we may see how it feels. The actor or dancer—or the spectator—sometimes notes and remembers the feeling of a movement rather than its pattern, insofar as the two can be distinguished at all. Emotion in aesthetic experience is a means of discerning what properties a work has and expresses.9

The general point is, of course, not limited to the aesthetic realm since, as I have earlier emphasized, the emotions intimately mesh with all critical appraisals of the environment: The flow of feeling thus provides us with a continuous stream of cues significant for orientation to our changing contexts. Indeed, as Goodman remarks,

In daily life, classification of things by feeling is often more vital than classification by other properties: we are likely to be better off if we are skilled in fearing, wanting, braving, or distrusting the right things, animate or inanimate, than if we perceive only their shapes, sizes, weights, etc. And the importance of discernment by feeling does not vanish when the motivation becomes theoretic rather than practical. . . . Indeed, in any science, while the requisite objectivity forbids wishful thinking, prejudicial reading of evidence, rejection of unwanted results, avoidance of ominous lines of inquiry, it does not forbid use of feeling in exploration and discovery, the impetus of inspiration and curiosity, or the cues given by excitement over intriguing problems and promising hypotheses.10

THEORETICAL IMAGINATION

Mention of the context of theory brings us to the third role of emotions in the service of cognition, that of stimulus to the scientific imagination. This role is virtually annihilated by the stereotyped emotion-cognition dichotomy. For this dichotomy assigns all feeling and flair, all fantasy and fun, to the arts and humanities, conceiving the sciences as grim and humorless grind. The method of science is miserly caution—to gather the facts and guard the hoard. Imagination is a seductive distraction—a hindrance to serious scientific business.

This doctrine is, in fact, the death of theory. Theory is not reducible to mere fact-gathering, and theoretical creation is beyond the reach of any mechanical routine. Science controls theory by credibility, logic, and simplicity; it does not provide rules for the creation of theoretical ideas. Scientific objectivity demands allegiance to fair controls over theory, but fair controls cannot substitute for ideas. "All our thinking," said Albert Einstein, "is of this nature of a free play with concepts; the justification for this play lying in the measure of survey over the experience of the senses which we are able to achieve with its aid."11

The ideal theorist, loyal to the demands of rational character and the institutions of scientific objectivity, is not therefore passionless and prim. Theoretical inventiveness requires not caution but boldness, verve, speculative daring. Imagination is no hindrance but the very life of theory, without which there is no science.

Now the emotions relate to imaginative theorizing in a variety of ways. The emotional life, to begin with, is a rich source of substantive ideas. Drawing from the obscure wellsprings of this life, the mind's free play casts up novel patterns and images, exotic figures and analogies that, in an investigative context, may serve to place old facts in a new light. The dream of the nineteenth-century chemist F. A. Kekule will provide a striking illustration. He had been trying for a long time to find a structural formula for the benzene molecule. Dozing in front of his fireplace one evening in 1865, he seemed, as he looked into the flames, "to see atoms dancing in snakelike arrays. Suddenly, one of the snakes formed a ring by seizing hold of its own tail and then whirled mockingly before him. Kekule awoke in a flash: he had hit upon the now famous and familiar idea of representing the molecular structure of benzene by a hexagonal ring. He spent the rest of the night working out the consequences of this hypothesis."12

The emotions serve not merely as a source of imaginative patterns; they fulfill also a selective function, facilitating choice among these patterns, defining their salient features, focusing attention accordingly. The patterns developed in imagination, that is, carry their own emotive values; these values guide selection and emphasis. They help imagined patterns to structure the phenomena, highlighting factual features of interest to further inquiry. "Passions," as Michael Polanyi has said, "charge objects with emotions, making them repulsive or attractive; . . . Only a tiny fraction of all knowable facts are of interest to scientists, and scientific passion serves ... as a guide in the assessment of what is of higher and what of lesser interest."13

Finally, the emotions play a directive role in the process of applying the fruits of imagination to the solution of problems. The course of problem-solving, as has already been intimated, is continually monitored by the theorist's cues of feeling, his sense of excitement or anticipation, his elation or suspicion or gloom. Moreover, imagined objects encountered in thought by the problem-solver affect his deliberation emotively, as real objects do, and influence his decisions in analogous ways. "In thought as well as in overt action," says Dewey, "the objects experienced in following out a course of action attract, repel, satisfy, annoy, promote and retard. Thus deliberation proceeds."14 There is, no doubt, much yet to be learned about the interaction of emotions and imagination in all the ways I have sketched, and in others as well. It should, however, even now, be evident that creation is fed by the emotional life in the sphere of science no less than in the spheres of poetry and the arts.

COGNITIVE EMOTIONS

We have, until now, concerned ourselves with the organization of emotions generally in the service of cognition. I want now to deal with two emotions that are, in a sense to be explained, specifically cognitive in their bearing—the joy of verification and the feeling of surprise.

In what sense do I speak of an emotion as specifically cognitive? Consider first the notion of moral emotions, conceived as those resting upon suppositions of a moral sort: Thus, indignation, for example, rests upon the supposition of a moral grievance—a piece of injustice, and remorse presumes that one has in fact done something wrong. If the relevant moral suppositions are false or lack evidential foundation, the respective emotions may be thought unreasonable, but if these suppositions are not made at all, that is to say, if the suppositions do not exist, the emotions in question can hardly, in normal circumstances, be said to have occurred. Now I propose, analogously, to consider an emotion specifically cognitive if it rests upon a supposition of a cognitive sort—that is to say, a supposition relating to the content of the subject's cognitions (beliefs, predictions, expectations) and, in cases of special interest to us, bearing upon their epistemological status.

It is important to avoid misunderstanding of the terminology I have chosen here. It is indeed true that all suppositions may be considered cognitive in a broad sense, inasmuch as they make factual claims expressible in prepositional form; moreover, emotions generally, as I have maintained, presuppose the existence of such claims concerning critical features of the environment. However, when I characterize an emotion as specifically cognitive, I mean more than this. In particular, I mean not simply that it presupposes the existence of a factual claim but that the claim in question specifically concerns the nature of the subject's cognitions (and, in cases of interest, is epistemologically relevant to them). A cognitive emotion, I should further emphasize, is thus decidedly an emotion, but an emotion of a certain kind, specifiable by its cognitive reference as just explained.15

THE JOY OF VERIFICATION

In his well-known paper of 1934 on "The Foundation of Knowledge,"16 Moritz Schlick provides an example of such a cognitive emotion in outlining his theory of science, giving primary place in his theory to the joy that accompanies the fulfillment of an expectation. Cognition, in Schlick's view, has, from the earliest times, always been predictive but the value of reliable prediction lay originally in its practical sendee to life. "Now in science," h*e writes, " . . . cognition ... is not sought because of its utility. With the confirmation of prediction the scientific goal is achieved: the joy in cognition is the joy of verification, the triumphant feeling of having guessed correctly."17 Such moments of joy are, in Schlick's opinion, of central importance in understanding scientific purpose. "They do not in any way," he says, "lie at the base of science; but like a flame, cognition, as it were, licks out to them, reaching each but for a moment and then at once consuming it. And newly fed and strengthened, it flames onward to the next. These moments of fulfillment and combustion are what is essential. All the light of knowledge comes from them."18

Now one need not agree with Schlick's general view of science in order to acknowledge that the satisfaction of a theoretical forecast may indeed occasion joy. Nor is it required that we concur with the extravagant suggestion that all predictive success brings elation. It may, for example, be countered that routine successes based on theory frequently, perhaps typically, go unnoticed, while soberly predicted events may be so dreadful as to occasion not joy but sorrow or despair. Nevertheless, we can agree that the fulfillment of a prediction may indeed crown an investigative achievement in science, producing in its wake what Schlick calls a "triumphant feeling of having guessed correctly."

This joyful feeling I consider a cognitive emotion, because it rests on a supposition (with epistemological relevance) as to the content of the guess in question: It presumes that what has happened is what had, in fact, been predicted. Without such presumption, this joy of verification cannot be said to occur. Whether the presumption is true, or is based on adequate grounds, is another story. Certainly one assumes that the emotion in question may be criticized as unreasonable if it can be shown that what has in fact happened had not, in fact, been predicted.

Can such a criticism, however, actually be entertained as a matter of psychological fact? Is it not, rather, true that our expectations are so powerful as consistently to warp our observations to fit? A whole library of psychological writings testifies to the powerful tendency of expectation to create its own confirmations in experience. The general theme of this testimony may be indicated by the role of normal cues in perception: The perceptual identification of objects "proceeds on the basis of cues normally sufficient to select the objects in question; when these cues fail in fact, we tend anyway to see them as having succeeded."19 Bruner, Goodnow, and Austin comment on this point as follows:

If [a bird] has wings and feathers, the bill and legs are highly predictable. In coding or categorizing the environment, one builds up an expectancy of all of these features being present together. It is this unitary conception that has the configurational or Gestalt property of "birdness". Indeed, once a configuration has been established and the object is being identified in terms of configurational attributes, the perceiver will tend to "rectify" or "normalize" any of the original defining attributes that deviate from expectancy. Missing attributes are "filled in" ... , reversals righted ..., colors assimilated to expectancy.20

Some philosophers have further maintained that scientific observation itself is systematically theory-laden—presupposing the very theories it is naively thought to test.21 If indeed we are, as suggested, so blinded by our own theoretical beliefs as to be incapable of acknowledging anything that might contradict them, we can hardly take the joy of verification to represent a cognitive triumph of science. Rather, we must count it an unearned and deluded joy, resulting not from a happy match between theory and experience but solely from our desperate rigging of experience to make it fit.

This conclusion, as I have elsewhere argued, seems to me too extreme for the facts.22 It is undeniable that our beliefs greatly influence our perceptions, but neither psychology nor philosophy offers any proof of a preestablished harmony between what we believe and what we see. Expectations have the function of orienting us selectively toward the future, but this function does not require that they blind us to the unforeseen. Indeed, the presumption of mismatch between experience and expectation underlies another cognitive emotion: surprise. The existence of this emotion testifies that we are not, in principle, beyond acknowledging the predictive failures of our own theories, that we are not debarred by nature from capitalizing upon such failures in order to learn from experience. The genius of science is, in fact, to institutionalize such learning by wedding the free theoretical imagination to the rigorous probing for predictive failures.

THE SIGNIFICANCE OF SURPRISE

Surprise is a cognitive emotion, resting on the (epistemologically relevant) supposition that what has happened conflicts with prior expectation. Without such presumption, surprise cannot be supposed to occur, although the truth of the presumption may, of course, be questioned in particular cases. Surprise must, in any event, not be confused with mere novelty. A novel—that is to say, a hitherto unencountered—contingency may well be anticipated in thought, while a familiar phenomenon, juxtaposed with available theory, may profoundly surprise. Thwarting expectation, the surprising element may indeed provoke the revision of theory, even the reorganization of categories, thus producing novelty as a result. It is itself, however, never a mere matter of novelty, but always of conflict with prior belief. The concept of unexpectedness, it should be noted, is too weak to make this critical distinction, for it covers both the case of a feature that has simply not been anticipated, and that of a feature that has been positively ruled out by anticipation.23

To the extent that we are capable of surprise, the possibility that our expectations are wrong is alive for us and thus our joy in verification, if it occurs, is not utterly deluded. Receptive to surprise, we are capable of learning from experience—capable, that is, of acknowledging the inadequacies of our initial beliefs, and recognizing the need for their improvement. It is thus that the testing of theories, no less than their generation, calls upon appropriate emotional dispositions.

Receptivity to surprise involves, however, a certain vulnerability; it means accepting the risk of a possibly painful unsettlement of one's beliefs, with the attendant need to rework one's expectations and redirect one's conduct. To be sure, where the relevant beliefs are weakly held, or relatively segregated, or of peripheral significance for one's basic orientation—or where the required alterations are likely to be readily effected—the risk may be easily borne, even by the cautious. Surprise may, in such circumstances, not in fact distress but amuse-even enchant, as will be evident from even brief reflection on the role of surprise in humor, in music, in literature, and generally in the arts. Moreover, there are, in all realms of life, pleasant surprises, where the value of the unexpected event, or even of the unsettlement itself, outweighs the stress of disorientation and the concomitant costs of revision in belief.

One cannot, however, reasonably count on all—even most—surprises in life to be thus amusing or pleasant; it must be conceded that a general openness to surprise involves a real risk of epistemic distress. This risk may to varying degrees become palatable, even exciting; certainly accepting it is one of the normal requirements of rational character. Yet it is a risk of possibly painful disorientation, and it requires emotional strength to face and to master. To commit oneself to learning from experience is, in short, a significant attitude—supported by mature reflection, to be sure—but exacting a price in return for the prospect of improvement in one's system of beliefs.

Three alternative attitudes promise an avoidance of the price by erecting wholesale defenses against surprise. Since surprise presumes prior expectation, a defense may be sought, to begin with, in the rejection of all expectation—in effect, the denial of all belief. This is the attitude of the radical skeptic, who hopes to make himself immune to surprise by any contingency through renouncing all anticipations to the contrary, that is to say, all anticipations without exception. Of whatever happens, he says, in effect, "It doesn't surprise me since I never expected it not to happen!" A second—apparently opposite—attitude is that of utter credulity or gullibility—the acceptance of all beliefs or expectations without distinction. Here the formula in response to every contingency is "I'm not surprised! I expected that too!" Both radical skepticism and radical credulity are, however, alike forms of epistemic apathy: To reject all expectations is to be indifferent to each, while to accept all as equally good is in actuality to choose none, having no reason to expect anything at all rather than something else. It is no wonder that these seemingly contrary attitudes have so often been remarked to be psychologically akin, and together opposed to the selective hypothesis-formation characteristic of scientific thought. "Complete doubt," as Peirce noted, is "a mere self-deception" and no one who follows the method of radical skepticism "will ever be satisfied until he has formally recovered all those beliefs which in form he has given up."24

Moreover, each of these two alternative attitudes exacts its own heavy price. Neither can in fact be realized as a genuine option over a significant area of conduct. The skeptic, despite himself, forms positive expectations in executing his actions, while the radically credulous person, generally hospitable to inconsistencies, perforce rules out certain contingencies in carrying through the activities of daily life. Only in a local and intermittent way can these attitudes be attempted. They are perhaps more accurately described as poses or pretenses, the effect of which is, however, perfectly real—to aid the denial of responsibility for one's beliefs and so to block the possibility of their improvement through the educative medium of surprise.

The third attitude promising a defense against surprise is that of dogmatism. Unlike the radical skeptic and the total believer, the dogmatist is perfectly firm about the beliefs he espouses and the beliefs he rejects. He blocks surprise not by disclaiming responsibility for his doctrines, but rather by denying all experience that purports to contradict them. He not only avoids the systematic testing of his beliefs; he closes off the very possibility of recognizing negative evidence, by early and stout denial of its existence. Theory-laden to the point of blindness, his observations are predictably positive, the joy he takes in verification thus unearned and hollow. Dogmatism is also a difficult attitude to maintain if only because (as Peirce saw)25 it is impossible to filter all negative indications in advance by a systematic method. Yet it can be carried a long way, preventing the acknowledgment of surprise, and, hence, the application of new and surprising experience to the improvement of initial beliefs and orientations. Dogmatism, no less than skepticism and gullibility, conflicts with the effort at such improvement. To accept this effort, with its associated vulnerability to the unsettlement of surprise, is to choose a distinctive emotional as well as cognitive path.

But how, it may be asked, is receptivity to surprise possible? Surprise is, after all, unsettling; it risks the distress of disorientation and the potential pain of relearning. In similar vein, Schlick contrasts the joy of verification with the disappointment of falsification26—the disappointment following upon the violation of beliefs in which we had put our trust. How can one counsel receptivity to surprise: Is this not an impossibly mixed emotion, like elation at despair, or happiness at depression?

We must, first of all, reject the suggestion that surprise is always painful. If Schlick's notion of uniformly joyful verification is to be rejected as extravagant, in line with our earlier remarks, his parallel notion of uniformly disappointing falsification must equally be criticized. Some falsifications are, as we have previously intimated, delightful, some disruptions of expectation pleasantly exciting, some occasions of relearning fraught with engaging challenge. Schlick, I suggest, confuses expectation with hope, but the two are clearly separable, and only the former is necessary for surprise.

Yet if surprise is not always painful, surely it sometimes is: It must therefore be conceded, at the very least, to be uncertain in its quality. The question then recurs in a new version: How can one counsel receptivity to uncertainty? Here, however, an immediate reply is forthcoming. The original version of the question raised the issue of impossibly mixed emotions, whereas the present version no longer does so. For uncertainty is not an emotion; it is rather a prospect or condition, while the feeling of uncertainty mixes readily with receptive and aversive attitudes. Uncertainty is indeed consistently faced in varying ways: Some persons tend to shrink from, while others tend to welcome the prospect. The receptivity to surprise that is implicated in the capacity to learn from experience is, in any case, perfectly coherent in its emotional composition.

Moreover, receptivity to surprise is not to be confused with elation or happiness. It is rather the capability of acknowledging surprise than joy in its occurrence that is here in point. Analogously, to acknowledge one's grief does not entail being elated by it. Acknowledgment itself is a possible and a significant attitude, opening the way beyond the acknowledged circumstance.

Yet receptivity is, of course, not enough to characterize the testing phase of inquiry. How we cope with surprise, once it is acknowledged, is of critical importance. Surprise may be dissipated and evaporate into lethargy. It may culminate in confusion or panic. It may be swiftly overcome by a redoubled dogmatism. Or it may be transformed into wonder or curiosity, and so become an educative occasion. Curiosity replaces the impact of surprise with the demand for explanation;27 it turns confusion into question. To answer the question is to reconstruct initial beliefs so that they may consistently incorporate what had earlier been unassimilable. It is to provide an improved framework of premises by which the surprising event might have been anticipated and for which parallel events will no longer surprise.

Critical inquiry in pursuit of explanation is a constructive outcome of surprise, transforming initial disorientation into motivated search. There is, as we have seen, no mechanical routine that guarantees success in the search for explanatory theory. Yet, an emotional value of such search is to offer mature consolation for the stress of surprise and the renunciation of inadequate beliefs.28 Achieving superordinate status in the economy of science, the value of inquiry becomes, indeed, autonomous, pressing new explanations deliberately into situations of risk, testing their vulnerability in novel ways, exposing their implicit predictions systematically to the chance of new surprise.

The constructive conquest of surprise is registered in the achievement of new explanatory structures, while cognitive application of these structures provokes surprise once more. Surprise is vanquished by theory, and theory is, in turn, overcome by surprise. Cognition is thus two-sided and has its own rhythm; it stabilizes and coordinates; it also unsettles and divides. It is responsible for shaping our patterned orientations to the future, but it must also be responsive to the insistent need to learn from the future. Establishing habits, it must stand ready to break them. Unlearning old ways of thought, it must also power the quest for new, and greater, expectations.29 These stringent demands upon our cognitive processes also constitute stringent demands upon our emotional capacities. The growth of cognition is thus, in fact, inseparable from the education of the emotions.