On the Primacy of Cognition Richard S. Lazarus University of California, Berkeley Abstract: Zajonc and I differ greatly in our concep- out of ongoing transactions with the environment tualization of emotion and its relations with cognition, that are evaluated. as well as in our evaluation of the evidence. My reply Cognitive activity is a necessary precondition of is in two parts. First, I discuss the boundaries of emo- emotion because to experience an emotion, people tion as a phenomenon and whether sensory preferences must comprehend—whether in the form of a prim- can be regarded as emotions; second, I make an anal- itive evaluative perception or a highly differentiated ysis of the evidence Zajonc regards as supporting his symbolic process—that their well-being is implicated claims for the independence of cognition and emotion in a transaction, for better or worse. A creature that and the primacy of emotion. My aims are to sharpen is oblivious to the significance of what is happening the philosophical and empirical issues that underlie for its well-being does not react with an emotion. our disagreement and to emphasize the indetermi- This same point has been stated cogently in various nancy of the issue of cognitive versus emotional pri- ways by numerous multidisciplinary scholars who re- macy. This latter issue is less important than the task sponded recently to an article on a general psycho- of exploring the cognitive contents or meanings that biological theory of emotions by Panksepp (1982) shape each kind of emotional reaction. Finally, I offer appearing in The Behavioral and Brain Sciences. a brief programmatic statement about what cognitiv- The conception that the meaning or s i g n i f i c a n ce ists can do to advance our understanding of emotion of a transaction is crucial to emotion forces us to over the life course. restrict its definition to some psychophysiological phenomena and to reject others as outside its p u r v i e w. The searching question is what an emotion is or is not. Zajonc evades this question. Thus, he takes me The latest riposte by Zajonc (1984) has, in my view, to task for doing what any good theorist should do not helped to clarify our understanding of the rela- with definitions, b u t he does not do himself—namely, tionship between cognition and emotion. Zajonc takes specify the phenomena of interest. Emotion, for ex- my reasoning (Lazarus, 1982) to task in two major ample, is not just physiological arousal, though such ways. First he complains that my position cannot be arousal is one of the traditional defining attributes. falsified because I defined emotion as requiring cog- Arousal can be produced by exercising vigorously or nitive appraisal, then that I have ignored the evidence entering a hot or cold room. Doing this will produce that emotion can occur without cognitive activity, an emotion only if we appraise the encounter (e.g., which he cites. I believe he is wrong about my episte- the physical and social conditions and the bodily state mological position and wrong that the evidence sup- it produces) as having a bearing on our well-being, ports the primacy of emotion or its independence as when, for example, it presents some physical danger from cognition. The body of this reply consists of a or brings blissful relief from discomfort. discussion of the definitional issue and why I think Startle is a reaction that has long but erroneously the empirical case he makes is specious. My objective been included under the rubric of emotion. Ekman is to sharpen the issue and sustain my position and (in press) presents new experimental evidence that that of like-minded cognitivists. startle might better be regarded as a reflex, like the knee jerk, because it does not behave as do other The Definitional Issue reactions we call emotions. He examined facial and Definitions do not arise out of the blue; they are an bodily responses under four conditions: when subjects integral part of a theory that helps delimit the phe- did not know at what moment a blank pistol would nomena of interest and organize observations. In my be fired; when they did know the moment; when they view, emotion reflects a constantly changing person- tried to inhibit the startle reaction; and when they environment relationship. When central life agendas attempted to simulate a genuine startle. Ekman con- (e.g., biological survival, personal and social values trasted emotions with startle. He found that the startle and goals) are engaged, this relationship becomes a was easy to elicit and was consistently the initial re- source of emotion. Therefore, an emotional experi- sponse to a gunshot; in contrast, there is no single ence cannot be understood solely in terms of what elicitor that will always call forth the same emotion happens inside the person or in the brain, but grows in adults. Moreover, the startle response could not 124 February 1984 • American Psychologist Copyright 1984 by the American Psychological Association, Inc. Vol. 39, No. 2, 124-129 be totally inhibited; in contrast, emotions can. Nor definable solely by behavior, subjective reports, or could the startle be simulated correctly by any subject. physiological changes; its identification requires all On the basis of these findings, the startle response three components, since each one can be generated seems fixed and rigid in comparison to emotions, and by conditions that do not necessarily elicit emotion, once elicited it seems to run its course. Thus, from as in the example of arousal by exercise given above. the perspective of a relational and cognitive concep- An emotion researcher must worry about which re- tualization of emotion, I think we should exclude sponse states or processes can be called emotions and startle from the rubric of emotion, as I advocated in which cannot; meeting one or even two of the three my earlier discussion in this j o u r n a l (Lazarus, 1982). response criteria is not enough. Historically, there has been much concern about With respect to the debate between Zajonc and diverse phenomena for which e m o t i on terms are em- me, a critical question is, On what basis should pref- ployed, such as moods, sentiments, emotion traits, erences (e.g., for taste, smell, or photographs of faces) and actual emotional reactions (cf. Ortony & Clore, be regarded as emotions? We must ask a similar q u e s - 1981). Moods usually refer to sustained general states, tion about aesthetic reactions to a pretty picture, a such as sadness and contentment, that may or may pleasing sunset, a stirring piece of music, or a fine not be considered emotions depending on theoretical piece of literature. Some years ago, while examining and definitional conventions. Sentiments refer to emotion and feeling in psychology and art, Arnheim characteristic ways a person evaluates an object (per- (1958) pointed out that emotion in artistic experience son, idea, thing); they operate as dispositions to react is not merely a passive receipt or apprehension of emotionally to that object but are not in themselves information, but requires active, involved partici- emotions. Another questionable category of emotion pation. As with sensory preferences, what may pass consists of personality traits, such as cheerful, which for an emotional (aesthetic) response may be nothing could in some instances describe an actual emotion, more than a pro forma statement that implies emotion as in "I feel cheerful," but could in other instances but does not necessarily reflect it, as when one casually describe a trait, not an actual emotional response, as says, "That's a pretty picture" in a manner more in "I am a cheerful person." Some emotion terms indicative of labeling than emotion. are heavily detached, lukewarm, or cold, such as in- Although preferences can involve emotions, even terested, whereas others, such as enraged, are hot. As strong ones, they often seem to fall at the ambiguous Ortony and Clore pointed out, it makes a considerable borderline between emotion and nonemotion. On the difference how such terms are used by subjects in one hand, statements of preference can be "cold cog- emotion research. nitions" expressing merely a social requirement to In the 1940s and 1950s, a period characterized make a choice, or on the other hand, they can be by the scientific outlook sometimes called logical pos- expressions of genuine emotional involvement. In the itivism, the dominant view in psychology was that research Zajonc cites, we do not know whether in emotions could not be defined and studied as such expressing a preference (e.g., "I like him more") sub- but represented intervening variables (cf. Brown & jects are experiencing an emotion, as indicated by Farber, 1951; Lazarus, 1968; Lazarus, Kanner, & multileveled response criteria, or merely expressing Folkman, 1980). Although similar attitudes still exist, an intellectual choice. If the latter, then preferences the more restrictive treatment of emotions in the past must be excluded from the category of emotions; if has given way to a view that allows much greater the former, they fall under the rubric of emotion. latitude in their study; this view depends heavily on Zajonc fails to come to grips with this problem, and what subjects report, but can be supplemented by he makes no mention of the alternative states that simultaneous assessments of behavioral and physio- could be indicated when subjects give reports about logical response variables. Emotions are commonly their sensory preferences. conceptualized and studied as an organic mix of action Zajonc defines cognition as requiring some kind impulses and bodily expressions, diverse positive or of transformation of a present or past sensory input, dysphoric (subjective) cognitive-affective states, and and in his current article he notes that untransformed physiological disturbances.