Richard S. Lazarus’s Appraisal Theory I
Richard S. Lazarus’s Appraisal Theory I
Emotions as Intentional States
Abstract and Keywords
The American psychologist Richard Lazarus played an important role in the post-World War II history of research on the emotions. This chapter offers an analysis of the challenges he faced in his attempts to account for the meaning of the emotions. Lazarus's ideas about the role of "appraisal" or cognition in emotion were often tentative and confused, in part because of the difficulty he had in deciding what kind of claim it is that emotions are intentional states or actions. Is the claim fundamentally a constitutive-conceptual one, according to which it belongs to the very "grammar" of the emotions that they are intentional states? Or is the claim a causal argument about how emotions are aroused? Are those two kinds of claims incompatible, or can one adopt both a conceptual-grammatical and a causal explanation of the affects? Lazarus did not find it easy to answer these questions, even as he pursued a major research program designed to do so. The aim of this chapter is to examine Lazarus's experiments on the emotions and appraisal in the light of these difficulties.
More recently, especially among social and humanistic psychologists, the contextualist perspective has been adopted. The root metaphor is the historical act in all its complexity … The phenomena of interest will be intentional acts, acts designed to transform or transfigure the world; and the acts are performed by persons as agents, not as mechanical automata … Unlike many theorists in the recent academic tradition, I do not treat “emotion” as detachable from social contexts.1
These remarks by Theodore Sarbin in 1986 can serve as the starting point for an analysis of the contribution made by the American psychologist Richard S. Lazarus (1922–2002) to the study of emotions between the 1960s and the early years of the twenty-first century. Sarbin made his comments as an intervention in the well-known debate between Lazarus and Robert Zajonc about the role of cognition in the emotions. That debate had started six years earlier, in 1980, when, without naming Lazarus, who nevertheless understood that he was one of the main targets of the attack, Zajonc published the first in a series of papers contending that the affects did not depend on cognition, as several psychologists had argued, but could occur without extensive cognitive coding. Lazarus responded to Zajonc by insisting that emotions flow from cognitive processes, and the battle was joined. It lasted for many years. Indeed, in an important sense it is still not over.
What was the controversy between Zajonc and Lazarus really about? It has more than once been suggested that it was not about very much, that the dispute between the two men was not an empirical or even a theoretical one but was merely verbal or definitional: the two men were in fact describing the same thing, namely, the speedy, apparently automatic, involuntary response to emotionally salient stimuli, but one of them, Zajonc, called that response “noncognitive” whereas the other, Lazarus, called it “cognitive.”2 But this interpretation risks glossing over fundamental (p.130) issues. Setting aside the question of the validity of Sarbin’s adherence to a “dramaturgical” or “role-playing” theory of emotional action, his remarks have the merit of zeroing in on the most central question at stake in the dispute—the question of intentionality.3
In Lazarus’s theory of the emotions, intentionality was a fundamental concern. It was his commitment to the idea that emotions involve the cognition of objects—which is to say that they involve questions of meaning—that makes the idea that they are intentional states a central issue in his work. The claim that emotions are intentional states has had a distinguished history: in the philosopher Franz Brentano’s phenomenology; in Sigmund Freud’s psychoanalysis, in which emotional states can be unconscious because repressed (or rather, the original object of the emotion may be repressed and hence unavailable for consciousness, since Freud tended to think that emotions themselves were in some sense always conscious); in Wittgenstein’s remarks on the philosophy of psychology, in which he repeatedly questioned William James’s non-intentionalist reduction of emotions to bodily sensations; in Magda Arnold’s post-Brentano and post-Freudian approach to the emotions based in part on Sartre’s intentionalist arguments; in Anthony Kenny’s Wittgenstein-inspired discussion of emotion; in Robert Solomon’s existential-phenomenological discussion of affect; in Sarbin’s social-constructionist approach to emotion; and in the work of more recent emotion theorists, such as that of Alan Fridlund, who has offered a “Behavioral Ecology” or “Communicative” theory of the emotions in which the intentionality of animal action is a given; in Peter Goldie’s treatment of the emotions; and so on.
Yet Lazarus’s approach to the topic was often tentative and confused. In particular, as time went on he had to confront the question of what kind of claim it is that emotions are intentional states or actions. Is the claim fundamentally a constitutive-conceptual one? Could one say in a Wittgen-steinian spirit that it belongs to the very “grammar” of the emotions that they are intentional states—that emotions are necessarily about some object or event, so that the meaning of the various cognized objects or situations or events to which emotions are directed is therefore one of its constitutive or essential features? Or is the claim that emotions are intentional (p.131) states an empirical assertion about the causal link between emotions and their objects, so that what is required is scientific evidence about that link and about the causal mechanisms involved? Are those different kinds of claims incompatible or can one adopt both a conceptual or “grammatical” and an empirical position with respect to the nature of emotion? And if the claim is an empirical one, what methods—introspective, behavioral, physiological, ethological—are best suited to determining the causal link between cognition and affect? We will see that Lazarus did not find it easy to answer these questions. Indeed, his life’s work can be regarded as a kind of litmus test of the difficulties involved in formulating a cogent position on the cognitive-intentional nature of the emotions. His arguments proved hard to evaluate precisely because it was not clear to others and even to himself at what level he was operating.
Exacerbating Lazarus’s situation were several other factors. In the first place, the concept of intentionality involves some of the most intractable problems in the philosophy of mind, so it is hardly surprising that as a busy research scientist without philosophical training, Lazarus could not master the problems involved. Wittgenstein’s famous remark that “in psychology there are experimental methods and conceptual confusions” is especially pertinent in this regard, for the controversy over the nature of the emotions raised deep conceptual questions that not only confounded psychologists but troubled philosophers as well. A further complicating factor was that the post–World War II “cognitive revolution,” which challenged behaviorism by making it respectable for American psychologists to study mental processes once again, coincided with the cybernetic revolution, with the result that cognitive psychology was quickly captured by computational assumptions. Although the assimilation between cognitive psychology and information theory was not inevitable, their historical convergence, as Shanker has superbly demonstrated, can be traced to shared epistemological commitments of a Cartesian, dualistic-mechanist kind. The result was that cognitive psychologists were tempted to explain the operations of the mind in subpersonal computational-mechanical terms, which tended inevitably to marginalize the whole person and deflect attention from the meaningful mental contents that were the ostensible focus of psychological interest.4
One of Lazarus’s more interesting objections to Zajonc’s noncognitive theory of emotion, an objection inspired in part by Hubert Dreyfus’s well-known critique of artificial intelligence, was that it was indebted to a false (p.132) picture of how the mind works because it mistakenly assumed that cognition and meaning could be built up out of meaningless “bits” of information by hypothetical information-processing systems in the mind-brain. But we shall see that Lazarus himself was unable to resist some of the same computational assumptions, with the result that he risked losing control of his most important insights. In particular, his theory about the role of cognitive appraisal in emotions rested on the presupposition that there exists a gap or separation between the person and the world, which it is the function of cognition or appraisal to close, with the result that our emotional evaluations of objects are not direct and immediate but indirect and mediated—as if the meaning of objects for the individual had to be tacked on to neutral percepts of some kind.
In fact, as I will argue, Lazarus’s entire picture of appraisal as involving inner cognitions intervening causally between the person and the world—a picture he shared with information-processing theorists—was a mistake, one that led to several dead ends. As I noted in the introduction, an “embodied world-taking cognitivism” of the kind recently advocated by Phil Hutchinson, and inspired by Wittgenstein and the philosopher John McDowell, might have saved Lazarus from some of these errors. Hutchinson accepts the Kantian thought that “intuitions [or perceptions] without concepts are blind” in order to insist, in the terms brilliantly developed by McDowell, that our perceptions are conceptual or cognitive “all the way out,” which is to say that they are inherently conceptual. The resulting account of emotions is a cognitivism that emphasizes the ways in which humans and other animals are alive to aspects of the world—not to the disenchanted world of the modern natural sciences that stands external to minds, but to the cognized, conceptualized world. In Hutchinson’s case, it is a cognitivism that, because it does not make the attribution of propositional attitudes or speech central to a theory of the intentionality of the emotions, is not vulnerable to the charge that it is unable to account for emotions in nonhuman animals.5 But Lazarus himself was far from being able to articulate such a position. Instead, he adopted views that tended to undercut some of his best insights and, in an otherwise laudatory impulse to relate his ideas to those of his contemporaries and critics, offered (p.133) several concessions and accommodations that also had the overall effect of undermining the coherence of his approach.
In order to elucidate these and related points, in this chapter I shall discuss Lazarus’s early work on subliminal perception or “subception,” his subsequent experiments on stress and emotion, and his early theorizations of appraisal. In the following chapter I will turn to the debate generated in the 1980s by Zajonc’s noncognitive claims about affect in order to assess Lazarus’s response. In that chapter I will also consider Lazarus’s later writings, focusing on certain criticisms made by his peers that throw light on the continued difficulties Lazarus faced in his effort to establish a science of emotions defined as intentional states.
Subception, Perceptual Defense, and the “New Look” in Psychology
“Subception” is the name Lazarus coined to describe an individual’s automatic discrimination of threat or danger without conscious awareness of the stimulus. An early experiment on that topic, published with Robert McCleary in 1951 when Lazarus was starting his career at Johns Hopkins University, involved conditioning subjects to a set of nonsense syllables associated with painful electric shocks and, in a later session, comparing the subjects’ psychogalvanic skin responses to the shock syllables and to syllables that had not been associated with a shock. Lazarus and McCleary reported that even when subjects incorrectly identified which syllables had been tachistoscopically flashed on a screen, because the speed used was below the recognition threshold, they nevertheless produced larger psychogalvanic skin responses to the shock syllables than to the non-shock ones. The authors concluded that subjects could discriminate between the threat and non-threat syllables even in the absence of conscious awareness of the presented stimulus. In other words, some sort of judgment of the stimulus had occurred at an unconscious level, as indicated by the differential skin responses.6 As Lazarus subsequently observed of the implications of his findings: “In effect, people can instantly ‘recognize’ (p.134) that a situation is dangerous or benign and act accordingly without having much information about it and/or without consciously having examined the cognitive premises on which the appraisal is based.”7
Lazarus and McCleary’s experiment belonged to the movement launched by Jerome Bruner, George Klein, and others in postwar American psychology known as the “New Look” in perception. This was a movement that challenged the classical view of perception as involving the subject’s passive registration of external stimuli by emphasizing instead the active, dynamic role of individual motivation and personality in the perceptual process.8 The “New Look” owed a debt to Freudian ideas about unconscious motivation although, owing to hostility to psychoanalytic concepts among most American psychologists, Lazarus deliberately avoided the term “unconscious” when reporting his results and introduced instead the word “subception” (from subconscious perception).9
Although the phenomenon of perceptual defense was hardly new, the experiments performed on the topic in the postwar period generated a huge controversy that lasted for many years.10 As Norman Dixon showed in an interesting book published in 1971 at a time when perceptual defense was still the topic of much dispute, the problems associated with this work were indeed in part technical and methodological. Other factors also influenced the course of the debate—for example, many psychologists rejected the idea of unconscious or subconscious discrimination because (p.135) they feared its commercial exploitation in the form of subliminal advertising. But the most basic objection stemmed from the very idea of intentional defense—the idea that the mind could intentionally block out threatening stimuli by making unconscious evaluations of their affective meaning. Dixon commented in this regard that
Few findings in psychology have generated more heat than the discovery that recognition thresholds depend upon the emotional connotation of that which is recognized. To psychologists years hence, this academic pother may well seem rather odd since emotional connotation is but a parameter of meaning, and few would deny, even nowadays, that the ease or difficulty with which one recognized somebody or something depends upon their meaning or personal significance … Why, then, all this fuss over perceptual defence?11
One reason for the fuss, as Dixon explained, was that the notion of perceptual defense, or subception, seemed paradoxical because it implied that the mind simultaneously knows and does not know the same thing. Numerous critics voiced that objection (Subliminal Perception, 221).12 As (p.136) the eminent perception researcher James Gibson objected: “This seems to imply unconscious defence mechanisms governing perception as well as motivated behavior—wishful perceiving. But to say that one can perceive in order to not to perceive is a logical contradiction” (cited in Dixon, Subliminal Perception, 228–29). Of course, the psychoanalytic concepts of the unconscious and repression were meant to solve that paradox or contradiction. But many theorists rejected Freud’s solution as unsatisfactory because it seemed to merely reproduce the problem, not solve it, by implying the existence of an inner ego or homunculus capable of deciding which stimuli are acceptable and which must be defended against and repressed because they are threatening to the individual.13
In an attempt to solve the paradox of perceptual defense, investigators proposed many alternative solutions. These alternatives often took the form of neurophysiological theories in which the unconscious discrimination of stimuli was viewed as a consequence of a peripheral or central nervous system “gating” or “filtering” of sensory inputs and flows.14 Those physiological approaches were themselves sometimes informed by the new cybernetic ideas, with the result that subception effects were treated as the consequence of hypothetical information-processing mechanisms capable of discriminating between different system inputs. The dualism (p.137) inherent in such approaches, according to which what is mental must be conscious and what is not conscious must therefore be neurophysiological or corporeal, is revealed in Dixon’s remark that
If only conscious minds can “discriminate” and “recognize,” then obviously it is absurd to talk of discrimination without awareness or perceptual pre-recognition, but if we take these words to signify not the consciousness of the act, but merely the operation it performs the difficulty evaporates … [I]t is not difficult to build a machine which is programmed to “discriminate” between different inputs, and to classify the resultant discriminanda according to some prearranged principle. Nor is it difficult to conceive of this classification having an effect on subsequent read out.
In this formulation, the concept of subception as an unconscious mode of meaning-making is transformed into a concept of subception as a mechanical response—as if, unknown to the subject himself or herself, sub-personal processes or “programs” do the perceiving and judging that have been denied the mind and indeed the person.
Of course, as Dreyfus argued in his 1972 critique of cybernetics and artificial intelligence which appeared a year after Dixon’s book on subliminal perception, the latter’s widely shared belief that a machine could easily be programmed to discriminate stimulus inputs turned out to be naïvely optimistic. Dreyfus suggested that it was impossible to derive meaning from bits of meaningless information, so that attempts to frame issues of meaning in terms of information theory were bound to fail because they begged all the crucial epistemological, psychological, and ontological questions—a prediction that was soon borne out when efforts to build intelligent machines faltered or failed.15 Nevertheless, for many researchers in cognitive science the attraction of cybernetics and information-processing theory was very great.
A defining moment came in 1974 when the psychologist Matthew Erdelyi in a “New Look at The New Look” reformulated subception in strictly information-processing terms. Erdelyi argued that the paradox of perceptual defense, according to which the subject has to know something in order to defend against knowing it, could be solved by treating cognition not as a single, unified phenomenon but as a multicomponent filtering and encoding process involving several parallel subsystems operating selectively (p.138) at different levels and with different outcomes.16 On this hypothesis, there was no paradox about subception because affective and other signals could be managed by hypothetical information-processing systems running independently of conscious processing and before conscious cognition could kick in. In the next chapter, we shall see that Erdelyi’s cybernetic model of unconscious defense contained the seeds of Zajonc’s theory of emotion according to which affective signals are handled entirely separately from cognition because separate information-processing or neural pathways are involved.17
It could be argued that Erdelyi’s cybernetic hypothesis was no more of a solution to the paradox of subception than Freud’s psychoanalytic approach, since it merely restated the problem of stimulus recognition and defense in computational terms while begging every conceivable question about how the information-processing system could, by performing the necessary filtering processes, “recognize” contextual cues and create semantic meaning from “information” held in raw storage. But for many psychologists Erdelyi’s proposal had the irresistible advantage that, unlike psychoanalytic or quasi-psychoanalytic theories of ego defense of the kind Lazarus and other New Look theorists had put forward to explain the phenomenon of subception, it avoided any appeal to unconscious intention by reformulating mentation in strictly computational-corporeal terms.
In those early years Lazarus was gripped by the idea that affective and motivational factors played a dynamic role in perception, and he tended to stick to ego-psychology explanations of the phenomenon of psychic defense. He was not tempted by approaches that reduced perceptual and emotional processes to the workings of neurophysiological or information-processing systems (even if his own formulations sometimes (p.139) strayed in that direction). Instead, after leaving Johns Hopkins for Clark University in Massachusetts, in 1957 he moved to the Department of Psychology at the University of California at Berkeley, where he would spend the remainder of his career. During the 1960s at Berkeley he carried out an important series of experiments on stress, in the course of which he extended and refined his views about the part played by unconscious defense in psychic life. But it was during his efforts to theorize the cognitive appraisal processes involved in stress that problems in his theorizing began to surface.
Lazarus’s Experiments on Stress
In the 1950s Lazarus took up the problem of stress in an impressive laboratory program of research that lasted until 1972. The concept of stress had come into prominence in World War II when pilot error and other forms of impairment due to fatigue had become a topic of concern in the military. Physicians in the postwar period concerned with caring for survivors of the concentration camps were also dealing with the phenomenon of stress in the form of post-traumatic symptoms and the “survivor syndrome.”18 So were doctors and others investigating the effects of natural disasters and other life-threatening conditions, although during the early postwar years there was not much contact between the clinicians and the scientists who were carrying out experiments in the laboratory.19 By the time Lazarus published his first book on stress in 1966, stress had garnered so much attention that he complained that it was virtually impossible for him to do justice to or even fully cite the huge literature that had already been published on the topic.20
(p.140) The general thrust of Lazarus’s work on stress, which he offered as a continuation of the “New Look” in psychology, was to reject the dominant tendency to regard the stress threat as objectively “out there as an attribute of the stimulus” and to argue instead that the meaning of a stimulus depended on the contributing role of personality and individual differences. According to him, in a critique aimed at the majority of his fellow researchers who regarded stress as a purely external, objective event, the stressful nature of a stimulus could not be treated as objective or predetermined by the experimenter because its significance was also a function of a “person’s appraisal of the meaning of the stimulus for the thwarting of motives of importance to him”—which is to say, on “the interpretation the person makes concerning its future significance.”21
Since apparently similar stressful conditions meant different things to different individuals—since a situation that appeared threatening to one subject might not be threatening to another—Lazarus suggested that the role of the person in the stress response had to be taken into account. Instead of assuming, as most stress researchers did, that subject populations were homogeneous, Lazarus therefore emphasized the importance of individual differences. Coping styles and strategies—which is to say, cognitive styles—might vary depending on a person’s specific motivational history, with the result that the stressor or stimulus condition could not be isolated from the question of individual meaning. In short, the accent of Lazarus’s work on stress fell on the connotation of the situation or event for the individual and on the defensive appraisals he or she deployed in order to cope with danger. Lazarus suggested that the entire field of stress research had to start over so that the role of personality factors and individual differences could be considered.22 Over time, he (p.141) shifted attention from the concept of stress to that of emotion, because he regarded emotion as the more embracing term.23
Lazarus’s aim was to devise experiments in which the relationship between specific stressor conditions and individual responses could be operationalized, manipulated, and controlled. In a typical stress experiment, the scientist induced reactions by subjecting individuals to stressful situations under certain conditions, such as having them perform tasks under time pressure, or exposing them to painful-stressful stimuli such as small electric shocks or taboo words. The subjects’ responses were assessed by measuring changes in psychogalvanic skin reflexes, heart rate, and other gauges of autonomic arousal, as well as by evaluating subjective reactions, moods, and personality characteristics through the use of questionnaires and interviews. One of the challenges Lazarus hoped to meet in designing his experiments was to deploy stress stimuli that were as realistic as was possible within the confines of the laboratory in order to fully engage his experimental subjects—college students, for the most part. That is why starting in 1959 he began exploiting the stress-inducing qualities of Sub-incision, a short, silent, black-and-white anthropological film, showing naked Aborigines in the Australian bush undergoing puberty initiation rites. The film included scenes of adolescents, who appeared to volunteer for the procedure, repeatedly wincing and writhing in pain as older men in the tribe ritualistically cut their penises with sharp stones.24 We have already (p.142) encountered this film in the work of Ekman, who borrowed it from Lazarus (see chapter 2, n. 47). Lazarus’s rationale for using Subincision was that it provided a more vivid and immersive, if vicarious, experience of stress or threat to experimental subjects than did the tamer and more deceitful stimuli hitherto used in experiments of this kind.
As his research program advanced, in an elegant series of experiments he and his co-researchers sliced Subincision into short clips so as to be able to assess as precisely as possible the influence of the film content. Autonomic changes in the viewing subjects were monitored continuously during film screenings; fluctuating experiential reactions were assessed by questionnaires, post-film interviews, and other methods; and a “neutral” film was used as a control. In some of the experiments, other stress films, such as workshop accident films, were also employed to manipulate cognition and evaluate stress responses. Lazarus and his research group were able to demonstrate “huge main effects” of the contents of Subincision on their viewing subjects, as evidenced by large increases in autonomic arousal and related measures. Thus, the stressor film produced much higher levels of response than did the control film. But the stressor film also produced more individual variability in mood changes, coping mechanisms, and patterns of autonomic arousal.
So on the one hand, Lazarus and his team found large, consistent film effects on autonomic indices and subjective reactions. But on the other hand, they found many individual differences. Indeed, while some subjects expressed extreme distress, including disgust and horror, at viewing the scenes of genital cutting, others denied being disturbed. In an attempt to discover which scenes were the most upsetting, Lazarus and his team tried to pinpoint the parts of the film where the greatest autonomic arousal tended to occur. Here, too, they detected a peak reaction at the initial scenes of cutting. Nevertheless, some subjects showed little response to those scenes but reacted strongly to other moments in the film. Lazarus thought one of his most interesting findings was that those who denied being disturbed by the stressor film nevertheless registered autonomic changes of great magnitude, so there was a “baffling” discrepancy between subjective report and the psycho-physiological reaction, a discrepancy whose significance was not clear.
The discovery that some viewers denied that they were upset by the movie suggested to Lazarus that they were unconsciously using defensive (p.143) coping strategies. This led him to recognize that how the film content was presented could directly influence outcomes. Thus, Lazarus and his team found that if they added a soundtrack to the film clips that took a distant, intellectual, and detached anthropological view of the proceedings, or one using an official-sounding travelogue voice that took a denying attitude toward the scenes (as if the cutting were a happy experience for the boys), the threat of the film, as measured by autonomic changes and subjective responses, was reduced. But if a “trauma” soundtrack was added, emphasizing the horror of the situation, the dread of the Aborigine boys, and the harmful consequences that would befall some of them, the stressful response was increased. At the same time, individual differences in viewing responses continued to manifest themselves, so that the experiments revealed the existence of both typical changes and individual variations due, as Lazarus proposed, to personality differences and other motivating factors. Thus, while he did not deny that there were classes of conditions that served as stressors for most people, he also proposed that the more complex the content of the stressor or the situation, the more likely it was that individual differences relating to motives, needs, and personality would produce different areas of sensitivity. In a series of experiments conducted first in Japan and then back in the United States, Lazarus extended his studies to examine the role of cultural differences in emotional responses.25
(p.144) Lazarus’s central finding in these experiments was that what mattered in stress was the consciously or unconsciously cognized meaning of the situation for the individual. At first he deployed the familiar psychoanalytic concept of ego defense to explain how subjects coped with stress. But as his work progressed and as he began to read more widely in the literature on stress and the coping process, “appraisal” emerged as a key concept. He owed the term to the work of Magda Arnold (1903–2002), a pioneer of emotion research who in 1960 had published an important book on emotions that decisively influenced the development of Lazarus’s theoretical ideas.26 Thus, in his first reference to the appraisal concept in 1964, Lazarus and his coauthors cited Arnold for arguing persuasively that “an emotion implies an evaluation of a stimulus as either harmful or beneficial.” They stated that “In our own view … a stimulus must be regarded by the person as a threat to his welfare in order for stress responses to be produced. Thus, the same stimulus may be either a stressor or not, depending upon the nature of the cognitive appraisal the person makes regarding the significance for him.”27 A year later, Lazarus and his team again acknowledged a debt to Arnold by observing that the short-circuiting of threat by defensive processes was accomplished by “modification of appraisal—that is, the interpretation, the recognized meaning—of the events in the stimulus film. The theoretical sequence posited by this view is similar to the conceptual statement put forward by Arnold (1960) and is consistent with Schachter and Singer’s (1962) emphasis on cognitive processes as critical in determining emotional reaction.”28 From then on, Lazarus repeatedly credited Arnold with being the first to provide a “truly systematic use of the concept [of appraisal] in a serious theoretical treatment of the field of emotion.”29
What is important to grasp about Arnold’s influence on the development of Lazarus’s views is what she got right about appraisal—and what she got wrong. The first thing to note in this regard is that Arnold presented her general position on emotion as a phenomenological one. As Reisenzein has helpfully observed, her debt to the phenomenology of Brentano and Husserl was largely indirect, through the influence of their intellectual heirs and successors, such as Scheler and especially Sartre.30 In his little book The Emotions: Outline of a Theory (1948), Sartre had argued with reference to the writings of Husserl that if psychologists wanted to understand the emotions they would have to grasp the essence of human reality, not reduce the human being to the sum of isolated, externally observed psycho-neurological and behavioral “facts.” The essence of human reality included, crucially, the meanings and values that constituted consciousness. According to Sartre, then, the essence of the emotions as organized kinds of consciousness is that they are directed toward the world of objects such that signification or meaning inheres in them. As Sartre wrote, emotion “is strictly to the extent that it signifies.”
For Sartre, therefore, the relation of emotions to their objects is not an accidental or contingent one but a necessary or essential feature, and any properly phenomenological psychology of emotions has to start from that premise. It has to recognize that, since emotion does not exist as a corporeal phenomenon, because “a body cannot be affected, for want of power to confer meaning on its own manifestations,” a phenomenological approach to the affects must accept the purposive-intentional character of emotional reactions. From this point of view, the essence of emotion is not the mere consciousness of an affective state but a consciousness of something—of a real (or imagined) object. “[T]he man who is afraid is afraid of something,” Sartre observed, and the affected subject and the affective object are thus “bound in an indissoluble synthesis.” According to Sartre, even in the case of apparently objectless feelings, such as “one of those indefinite anxieties which one experiences in the dark, in a sinister and deserted passageway, etc., one is afraid of certain aspects of the night, of the world.”31 (This is an important claim in the context of the later debates over the cognitive nature of emotions in which, as we saw in (p.146) the introduction, theorists such as Paul Griffiths have routinely appealed to the objectless character of some emotions, such as some depressive states, to argue for the non-intentional nature of the basic affects.)
Arnold endorsed Sartre’s views. She stated in this regard that “Sartre sees the ‘signification’ of emotion in its unique relation to reality. Emotion always has an object reference. A man is always afraid of something, angry at something. As long as the emotion lasts, it maintains that focus on the object. Emotional consciousness is an unreflective awareness of the object, not a person’s awareness that he has an emotion” (her emphasis).32 As she also affirmed: “In his insistence that emotion can be understood only as a special relationship of the human being to the external world, Sartre’s theory is a much needed corrective. There is no doubt that any emotion refers to an object and clings to it as long as it remains an emotion. It is also true that emotional awareness is unreflective, immediate, object-bound. Nevertheless, it remains a reaction to the object, and therefore is active, not passive” (EP, 160). This last statement also brings out the importance Arnold attached to the unreflective nature of emotional judgments or appraisals. In her view many of the evaluations involved in appraisals are not deliberate or intellectual in nature but more like the rapid, intuitive judgments a person makes when performing a skill such as throwing a ball, driving a car, or playing the piano. A reflective judgment might follow the intuitive or non-reflective one.33
So far, so good. But there were other aspects of Arnold’s approach to emotion that from the start compromised her analysis. In particular, in spite of her assertion, following Sartre’s lead, that phenomenologically the appraisal of the object is immediate and unreflective, she nevertheless proceeded as if there was a gap between the person and the intended object, a gap that had to be filled by a cognitive appraisal process intervening between them. The result was that in Arnold’s analysis perception and appraisal were split apart into two separate processes. “Both perception and emotion have an object,” she wrote, “but in emotion the object is known in a particular way. To perceive or apprehend something means that I know what it is like as a thing, apart from any effect on me. To like or dislike it means that I know it not only objectively, as it is apart from me, but also that I estimate its relation to me, that I appraise it as desirable or undesirable, valuable or harmful for me, so that I am drawn toward it or (p.147) repelled by it” (EP, 171). To which she added: “Before anything can have ‘meaning’ for us, it must be seen as a thing (must be perceived) and must also be seen in some relationship to us (it must be appraised). Meaning comes with appraisal” (EP, 171). On this model, perception gives me “objective” knowledge or a “simple apprehension” of an object in the world but not its meaning for me, because to evaluate its meaning for me I have to make a separate assessment of its personal significance. Or, as Arnold also stated: “To know or perceive something and to estimate its effect on us are two distinct processes, and appraisal necessarily presupposes perception … To estimate how [an object] affects us personally … seems to require a further step beyond perception … Following upon perception and completing it, appraisal makes possible an active approach, acceptance or withdrawal, and thus establishes our relationship to the outside world” (EP, 176).
One way Arnold had of describing the appraisal process was to suggest, in terms borrowed from William James, that what appraisal does is to turn “cold” perceptions, which lack emotional meaning, into “hot” ones, in which the object perceived becomes affectively significant for the individual. As she put it: “When we examine the features to be explained in emotion, we find that at the base of every emotion there is some kind of perception or awareness of an object, a person, or a situation, which in some cases becomes emotional, in other cases remains (in the words of William James) a ‘cold perception.’ Therefore, it stands to reason that the perception that arouses an emotion must be somehow different from the mere perception of an object as such, which does not arouse an emotion” (EP, 93).34 But here, objections immediately arise. From what ostensibly neutral perspective can I perceive an item in the world as an objective “thing” in itself without regard to its meaning for me? How can I see an object “as such” before it acquires any personal meaning? It’s not at all clear that these formulations, which imply a splitting apart of the perception of the object from its significance for the person, are compatible with Arnold’s self-proclaimed phenomenological standpoint. Moreover, once such a gap between perception and personal meaning is posited, then it is natural to assume that the scientist’s goal must be to discover the causal processes or mechanisms by which the gap is closed. But this suggests that for Arnold the relationship between emotions and their objects is not an inherent and necessary one, as Sartre had argued, but is merely (p.148) contingent—as if, for her, the relationship is the consequence of a merely empirical correlation or connection.
Lazarus made the same troubling moves. “The cognitive interpretation and the emotional effect of a stimulus is [sic] determined not only by the objective properties of the stimulus but also by factors within the psychological structure of the individual such as motivation and knowledge,” he wrote in an early statement.35 Or, as he also observed:
The process of appraisal depends on two classes of antecedents. One consists of factors in the stimulus configuration, such as the comparative power of the harm-producing condition and the individual’s counter-harm resources, the imminence of the harmful confrontation, and degree of ambiguity in the significance of the stimulus cue. The second class of factors that determine the appraisal are within the psychological structure of the individual, including motive strength and pattern, general beliefs about transactions with the environment, and intellectual resources, education, and knowledge.36
Lazarus posited the same distinction between perception and appraisal in the following statement:
For threat to occur, an evaluation must be made of the situation, to the effect that a harm is signified. The individual’s knowledge and beliefs contribute to this. The appraisal of threat is not a simple perception of the elements of the situation, but a judgment, an inference in which the data are assimilated to a constellation of ideas and expectations. If you change the background of cognition, the same situation will now have a very different significance, perhaps no longer signaling harm to the individual … While the objective nature of the stimulus configuration contributes importantly to the appraisal process, it is always in interplay with the psychological structure of the individual.
(Lazarus, PS, 44, 55; his emphasis)
One can see why Lazarus wanted to say things like this. He wished to acknowledge the fact that features of a situation that do not bother one person do cause distress or emotional upset in another—as, for example, in the case of the viewers of the film Subincision, some of whom reported that they were not at all dismayed by the scenes of genital mutilation while (p.149) others reported feeling disgust. It was as if the “same” film aroused quite different emotions in different people because of the different evaluations they placed on it. But in what sense was the film “stimulus” the “same” for both types of viewers? The problem in Lazarus’s analysis begins when, in the passages just quoted, he differentiates between the “elements of the situation” or the “data” in the “stimulus configuration” on the one hand, and the meaning or emotional evaluation that is conferred on these same elements or data on the other—as if appraisal is a matter of imposing meaning on a percept that is not in itself meaningful to the individual subject until the appraisal process intervenes. In other words, Lazarus proceeded as if he believed that meaning has to be added, tacked on, to a supposedly impartial or unbiased and therefore not yet fully meaningful perception of some kind.
He used the same language as Arnold when he described the process of appraisal as one in which a person goes from a “cold” perception to a “hot” appraisal or cognition (PS, 52). He gave as an example the way an ordinary knife can suddenly appear threatening to an individual. The threat value of a knife, he wrote, “depends on inferences about the intentions of the holder. We deal with knives all the time, but it is only a particular pattern of cues (for example, the way it is held, the expression on the face, or the content of the words spoken) that suggest [sic] that there is any danger. And the significance of this pattern is established through experience” (PS, 395). What seems problematic here is Lazarus’s idea that in this situation the knife first appears as a neutral and unthreatening object but then is subsequently inferred to be threatening on the basis of cues of a more personally meaningful kind.
I can put the difficulty in this way: however much Lazarus tried to emphasize the role of context and the framing process in emotion, he couldn’t give up, or work his way around, the idea of a “given” state of affairs in perception, one that could in principle be objectively or neutrally described independently of the person actually encountering the scene. But who has this objective or neutral perception, and to whom does it belong? To be “objective” in the way Lazarus conceived it, the perception would have to be experienced from the perspective of no one in particular.37 Moreover, by positing the need for an appraisal process ancillary or supplementary to perception and designed to mediate between the impersonally perceived object and the individually motivated (p.150) subject, Lazarus was proposing that emotional meaning is not immediate and direct, as Sartre had claimed, but is indirect because it is an “inference” made from an initially “neutral” percept when the latter’s significance is subsequently or belatedly recognized. But this was to assert that the relationship between emotion and the intentional object was not a constitutive or essential one, as Sartre had argued, but was accidental or external—as if the significance of the object was not inherent in emotion but depended contingently on what the object happened to mean to a particular individual owing to his or her particular history and the mediation of appraisal. It is the concept of “mediation” itself that is problematic here, because it suggests that meaning, including affective meaning, is not directly apprehended but has to be added on to objects in the form of supplementary, personal interpretations.
It is interesting that Lazarus did not derive more help in thinking about these matters from another important text that he read at this time, the philosopher Anthony Kenny’s short book Action, Emotion and Will (1963), which concerned the intentionality of emotion and action. Kenny’s views reflected the influence of Wittgenstein rather than of Sartre, which is to say that his discussion was framed in the terms of ordinary language philosophy, or in what might be called “grammatical” terms, rather than in phenomenological-existential terms. Kenny was influenced not only by Wittgenstein but also especially by the philosopher Elizabeth Anscombe (a student of Wittgenstein), whose short book Intention (1959) has by now achieved the status of a classic but whose difficult and sophisticated arguments Kenny was attempting to work out in a more readable form.38
In his book, Kenny maintained that emotions could not be understood as intentional actions by virtue of their being related to some features of an agent’s supposed inner mental states functioning as the causes of his or her affects. He made use of Wittgenstein’s arguments against the idea of a private language and of private mental states to contend against the common picture of the emotions as inner psychic events that were only contingently connected to their objects. “Wittgenstein has shown that a (p.151) purely mental event, such as Descartes conceived emotion to be, is an Unding [absurdity],” he wrote:
Any word purporting to be the name of something observable only by introspection, and merely causally connected with publicly observable phenomena, would have to acquire its meaning by a purely private and uncheckable performance. But no word could acquire a meaning by such a performance; for a word only has meaning as part of a language; and a language is something essentially public and shareable. If the names of emotion acquire their meaning for each of us by a ceremony from which everyone else is excluded, then none of us can have any idea what anyone else means by the word. Nor can anyone know what he means by the word himself; for to know the meaning of a word is to know how to use it rightly; and where there can be no check on how man uses a word there is no room for “right” or “wrong” use.
Kenny traced that mistaken causal view of the emotions back to the influence of Descartes, commenting that
In arguing against the notion of the “private object” Wittgenstein chose as his example a sensation, pain; no doubt because it is in such cases that it seems most plausible to suggest that a word might acquire a meaning by a private ostensive definition. With the emotions, the Cartesian idea of a purely mental event runs into extra difficulty. Emotions, unlike pain, have objects: we are afraid of things, angry with people, ashamed that we have done such-and-such. This feature of the emotions, which is sometimes called their “intensionality,” is misrepresented by Descartes, who treats the relation between a passion and its object as a contingent one of effect to cause … [I]t is well known how difficult it is to give any account on Cartesian principles of a causal relation between physical and spiritual events. Here I will remark merely that the intensionality of the emotions adds a further reason for denying that emotion-words acquire their meanings by some private ostensive definition. Many people are attracted by the idea that the meaning of the word “pain” is learnt by picking out a recurring feature of experience and associating it with the sound of the word. It is much less plausible to suggest that the meaning of “fear” is learnt in the same way, when we reflect how very different from each other fears of different objects may be. What is the feature of experience, recognizable by introspection (p.152) without reference to context, which is common to fear of mice and fear of waking the baby, fear of overpopulation and fear of being overdressed, fear of muddling one’s sequence of tenses and fear of hell?
Kenny went on to suggest that the same erroneous Cartesian assumptions informed recent attempts to explain emotions by identifying their psychological or physiological causes, and tried to show in some detail why such approaches were conceptually and empirically inadequate. It is not that he rejected experimentation out of hand. He regarded as useful the methods being used to measure the intensity of emotions as precisely as possible based not only on what people said but also on the measurement of bodily changes, including facial expressions and psychogalvanic skin reflexes when subjects were exposed to shocks or films or other stimuli (these were just the kind of methods Lazarus himself was employing). But he cautioned that too many psychologists mistakenly believed that by measuring these phenomena they were measuring emotion itself. In his judgment, they were not. His argument was not just that scientists had so far been unable to detect distinct physiological or behavioral patterns associated with specific emotional states, although that was one of his points. Thus, he noted with reference to recent experiments that the same physiological and behavioral patterns had been detected in quite different affective states (a finding that undermined the kind of claims Tomkins had just made to the effect that different emotions could be defined by distinct patterns of physiological-behavioral response, though Kenny did not mention Tomkins). But Kenny’s objection to such approaches was also the conceptual one that in such experiments psychologists were obliged to treat certain affects, such as hope, love, vanity, or fear of a world war, which were not easily connected to precise bodily changes, as degenerate or aberrant cases of emotion—as if a chronic fear of war was different from sudden fright only in being less intense and therefore less objectively measurable.39
As Kenny saw it, emotions were not mere behaviors but motivated actions whose status, as such, experiments were in principle incapable of capturing. We might put it that for Kenny emotions relate to their objects (p.153) not as a result of causal mechanisms but by virtue of conceptual necessity. He observed:
The precisely measurable bodily phenomena which psychologists study are not identical with the feelings of which they are characteristic; for unlike the feelings they have a merely contingent connection with motivated behavior. There is a conceptual connection also between a feeling and its object, whereas the physiological processes studied by psychologists lack intensionality. Bodily changes may be the vehicle of an emotion, but they are not themselves emotion. As Aristotle said, a man who defines anger as a bubbling of the blood about the heart gives only the matter without the form.
According to Kenny’s Wittgensteinian approach to the emotions, understanding another person’s emotional state is not a matter of inference, as Lazarus appeared to suggest, but of the direct apprehension or perception of facial and gestural expressions and the mastery of the grammar or meaning of emotion terms. He therefore suggested that all efforts to attribute emotions to putative inner mental states linked causally and contingently to external objects were confused or mistaken.
It is impossible to do justice here to the further details of Kenny’s difficult arguments. He has been described as one of the first philosophers in the analytic tradition to defend the role of cognition in emotion, which is presumably why his book attracted Lazarus’s attention.40 Yet, if we ask what Lazarus got out of Kenny’s text, the answer is—not much. Lazarus cited Kenny’s book in one of his earliest attempts to outline a cognitive theory of emotion in a paper he and his colleagues gave at a symposium held at Loyola University in 1968.41 The symposium, organized by Arnold herself, gathered together an array of psychologists offering disparate, often competing views. Plutchik, Tomkins, and Schachter were there, among other well-known conference participants.
In their presentation to the meeting, published in the proceedings two years later, Lazarus et al. commented on the difficulty of defining emotions by observing that theorists had often assumed there must be some one “characteristic” unique to emotions. But that research had failed to reveal the one “thing” to which the noun “emotion” referred. This was not because of experimental or introspective failures, they suggested, but because (p.154) of the kind of category to which emotion words belonged. Emotion words do not refer to things, they proposed, but rather to “syndromes” in the sense that a disease is a syndrome. As syndromes, emotions are not characterized by a single symptom or set of symptoms, nor do they have one center or locus.
By comparing emotions to disease syndromes, Lazarus et al. appeared to be offering a natural-scientific approach to the affects. But then they seemed to change direction by mentioning Kenny’s book. “In addition to referring to syndromes,” they declared in citing Kenny’s text, “emotion concepts are also relational; that is, they typically imply an object, just as the concept ‘answer’ implies a question” (“TCTE,” 212). This statement is self-divided because in it the authors appear simultaneously to be making a “grammatical” claim in the spirit of Kenny’s arguments—that emotions imply an object in the same way that the concept “answer” implies a question—and an empirical claim at odds with Kenny’s position. For if, according to Lazarus et al., emotion concepts “typically” imply the existence of objects, then they may also atypically lack them, suggesting that the relationship between emotions and their objects is not a necessary and conceptual one, as Kenny had argued, but is only a contingent matter. Perhaps, though, it would be more accurate to say that the question-answer analogy proposed by Lazarus et al. was not “grammatical” in any Wittgensteinian sense at all, but was merely definitional. For if you want to know what an “answer” is, you can look the word up in a dictionary, which will tell you that an answer is a response to a question. But the meaning of “emotion” cannot be established by consulting a dictionary in that way, for there is no comparable definitional sense of the word, which is why Kenny thought it was necessary to undertake a conceptual analysis. In short, the authors’ question-answer analogy was not adequate to the issue.
Indeed, what is notable about Lazarus et al.’s reference to Kenny’s book is their failure to grasp the import of the latter’s arguments. It is true that in the same paper Lazarus, Averill, and Opton made a few comments about the relevance of ordinary language usage and terms for psychology. They also cited a recently published paper on the value of ordinary language philosophy itself by his student and coauthor Averill, who had also read Kenny’s work.42 But not only did Lazarus et al. quickly pass on to other (p.155) topics, I think it is fair to say that the essential issues with which Kenny was dealing eluded them.43 In particular, the authors did not appreciate the conceptual-Wittgensteinian considerations that drove Kenny’s investigation and instead aligned his work with the kind of causal investigation of the emotions to which the latter was in fact opposed. Lazarus et al. proceeded as if what was required for an understanding of emotions defined as “syndromes” was to identify the inner cognitive appraisal processes causing emotional states and to link these states to their contingently attached external objects. We might put it that Lazarus et al. “psychologized” and operationalized the issue of intentionality in ways that pulled against Kenny’s Wittgensteinian-inspired diagnosis. (I am reminded here (p.156) of Stanley Cavell’s brilliant thought that the achievement of Wittgenstein’s Philosophical Investigations was to de-psychologize psychology.)44 In sum, the authors adopted the very paradigms and modes of analysis that Kenny rejected and in doing so showed that they had not understood the stakes of the latter’s book. They implicitly assumed a Cartesian view of the mind as a private psychic domain separated from a world of objects onto which the individual mind imposed its personal cognitions and evaluations. They thereby individualized and subjectivized affective meaning.
Lazarus, Averill, and Opton’s inability to comprehend what Kenny was trying to do is not surprising. Philosophical work does not circulate in other disciplines the way empirical research does, and it usually takes several decades before it is actually understood and digested in ways that allow it to have an impact outside philosophy itself. It is only recently that philosophers have begun to focus on the question of intentionality in terms that throw light on what Kenny was trying to accomplish when he contended that, since mental states such as intentions are connected with their objects not by empirical laws but conceptually, mental states cannot be considered the cause of the emotions. As Stoutland has pointed out, at the time Kenny published his book the work of the “school” of the philosophy of intention and action to which he belonged, stemming from Wittgenstein’s and Anscombe’s thought, was neglected or even disdained by most mainstream philosophers, who argued that the philosophy of intention required lawlike, empirical generalizations that causally connected (p.157) the phenomenon to be explained with the conditions that gave rise to it.45 If the majority of philosophers rejected Wittgenstein’s and Anscombe’s undertaking, what chance did a busy experimental psychologist like Lazarus have of appreciating Kenny’s arguments?
Besides, even if Lazarus had been able to master and accept them, their implications would have been unclear. Did it follow from Kenny’s discussion that all experimental approaches to the affects would have to be given up as misguided? Or was the import of his conclusion rather that certain kinds of experimental strategies were ill advised—for example, those that imagined that the goal was to identify the putative internal psychic mechanisms thought to cause emotional reactions—but that more ecological or ethological approaches, involving the study of intact animals in their natural settings, might be warranted and valuable? Those questions do not seem to have even occurred to Lazarus at this time.46 All in all, it is understandable that when Lazarus first read Kenny’s book he was unable to grasp its mode of argumentation or its diagnosis of the problem of the intentionality of the emotions. It is interesting that he found his way to Kenny’s work in those years, but a comprehension of its implications for psychology was beyond his reach.
It might seem that I have attributed more importance to Kenny’s book than is warranted by Lazarus’s reference to it in one paper. My rationale for paying attention to it is both historical and theoretical. In fact, Lazarus cited Kenny’s text more than once during the 1970s precisely because it appeared to him relevant to certain fundamental questions about the nature of the emotions.47 And the same issues at stake in Kenny’s book would (p.158) continue to concern not just Lazarus but other emotion theorists as well, some of whom also read Kenny’s text.48 Indeed, as recently as 2013 the well-known philosopher of biology Paul Griffiths, whose views I mentioned in the introduction and will be discussing in chapter 6, cited Kenny’s book in the course of reviewing current emotion research in philosophy.49 That Griffiths favors a non-intentionalist account of the so-called basic emotions at odds with Kenny’s intentionalism does not detract from the interest of the fact that, to this day, the issue of intentionality and indeed Kenny’s views continue to haunt the field of the affective neurosciences.
It is worth noting that if Lazarus believed that affective meaning was to be considered the outcome of causal, psychic mechanisms or processes, then his position raised empirical expectations that he would soon find challenging to meet. For example, he needed to show experimentally that the appraisal process was separate from and prior to the resulting emotional response, a difficult requirement given that, as we shall see, according to Lazarus himself emotional reactions to appraisals are instantaneous, so there is virtually no time gap between the two “events.” The (p.159) causal role of appraisal in emotion was also going to be hard to demonstrate empirically if, as he also believed, appraisal is frequently unconscious and automatic and hence is unavailable for conscious self-report.
Some of these issues did not surface as a challenge until the 1980s, when Zajonc launched the controversy over the role of cognition in emotion. In the meantime, Lazarus spent the 1970s completing his laboratory studies of stress and then turning to the clinic as offering a more natural setting for the study of affect. Unhappy with what he perceived as the limitations of laboratory research approaches, he decided to pursue his by-now-influential appraisal approach by pursuing in-depth studies of how individuals react to different life threats in more ordinary, everyday settings. The shift to the clinic yielded a series of reports in which Lazarus attempted to enlarge and make more precise his views about appraisal and coping processes and to address some of the methodological challenges involved in conducting appraisal research across time and in different clinical groups and situations.50 But in the course of pursuing his agenda, Lazarus committed himself to certain views that further compromised his position on the intentionality of the emotions. It is to this aspect of his work that I now turn.
Differentiating the Emotions
Among the views to which Lazarus was committed was that emotions are differentiated into natural classes or types. It is generally accepted by today’s appraisal theorists that both Arnold and Lazarus endorsed the notion of discrete emotions as central to their respective enterprises. Not only do they associate the rise of Arnold’s and Lazarus’s appraisal concept with the claim that emotions are segregated into distinct, universal categories, they also link Arnold’s and Lazarus’s work to the idea that each basic emotion is tied to a specific type of appraisal. Thus, Roseman and (p.160) Smith, two prominent appraisal theorists, have recently suggested that among a number of specific problems the early appraisal theories were trying to solve, the first was how to account for the differentiated nature of emotional responses.51 This is how Lazarus himself viewed the matter.
In her 1960 book, Arnold had already expressed her opinion that emotions formed distinct categories and that each distinct emotion was accompanied by a specific pattern of bodily activity.52 Early on, Lazarus expressed a similar belief, even though he conceded that convincing evidence was still lacking. Thus, in 1966 he stated his assumption that the affects “are different from each other even though they appear to overlap and coexist” and claimed that there was “no good reason to assume the negative position that the names for different affects do not refer to distinguishable states and that the person is never capable of accurately and consistently reporting about them” (PS, 321). He repeated this belief on several occasions, stating in 1968, for instance, that the analysis of emotions conceived as distinct response syndromes made “mandatory, at some point, the classification of emotions.” He remarked that “I have taken the position that anger, fear, grief, euphoria, love, etc., are distinctively different in their three major response components, the physiological, motor, and cognitive. A consequence … is that research must be (p.161) aimed at distinguishing descriptively the properties of different emotional reactions.”53 He often declared that one of the greatest hindrances to the study of emotions was the lack of an adequate scheme for their definition and classification and called for a nosology of emotional phenomena. He held on to this position for the rest of his career. In one of his last papers, he joined his appraisal theory to the discrete emotions position by stating that “The theoretical construct that eventually came to be emphasized by those adopting a discrete emotions outlook is appraisal, its central issue being what an individual must think and want in order to react with each of the emotions.”54
As this last remark suggests, Lazarus linked the idea of discrete emotions to the existence of corresponding patterns of cognitive appraisal or meaning. Thus, as early as 1966 he stated that “different cognitive processes are associated with different affects. And these cognitive processes determine the form of coping with threat” (PS, 323). He reiterated this point, stating his view that it was likely that “one day specific physiological patterns, specific cognitive patterns, and specific behavioral patterns, in association with given eliciting conditions, will be identified, and shown to be organized in such a way as to clearly distinguish one emotional state from another” (his emphasis).55 The notion that discrete emotions are attached to characteristic cognitive patterns or coping styles became a central motif of his work.56 Lazarus thus called for a research program designed to identify objectively the cognitive, behavioral-expressive, and autonomic components of different emotional reactions.57
(p.162) It is not difficult to motivate historically Lazarus’s belief that emotions form discrete types or categories. For one thing, it seemed to him on phenomenological grounds that we experience different emotions. In this regard, he was reacting against the majority consensus of the time, which held that emotions could not be differentiated into discrete categories because the evidence for characteristic behavioral or physiological patterns of response associated with specific emotions was lacking. As Lazarus explained, for theoretical reasons the “emotion-as-motivation” theorists treated emotion as merely an intervening variable; the drive theorists likewise downplayed the idea of emotional differentiation; and the behaviorists tended to make a single emotion, such as anxiety, carry all the theoretical weight. The net result was that emotion was treated as an undifferentiated state of drive or arousal rather than as a differentiated array of experiences and somatic and action patterns. “The great importance attached to the drive or motivational properties of emotion led to the virtual disappearance of emotion as a substantive topic in psychology journals,” Lazarus et al. noted in 1982. “Aspects of emotion that were not readily accommodated in these terms were neglected … Psychology suffered a severe form of tunnel vision.”58
So strong were the doubts about the existence of discrete emotions that, as Lazarus et al. observed in the same paper, in 1941 and 1962 Duffy had argued that the emotion concept was worse than useless, and in 1951 Brown and Farber had likewise rejected emotion as a concept suitable for scientific study. Lazarus et al. even criticized Schachter’s well-known cognitivist model on the grounds that although the latter emphasized the role of cognition in emotion, he treated cognition as involving a labeling of diffuse states of arousal in terms too close to that of the drive activation theorists to satisfy Lazarus’s convictions about affective differentiation.
Against this background, it is comprehensible that in 1966, at the very moment he began to pay attention to the emotions, Lazarus welcomed Silvan Tomkins’s “somewhat unsystematic” attempt to define the affects as discrete states (PS, 321). It is also intelligible that early on he was receptive to Ekman’s views about the existence of discrete emotions and the leakage of nonverbal behaviors as sources of information about specific emotions.59 Repeating the opinion that one of the greatest barriers (p.163) to the investigation of emotion had been the lack of an adequate classification, Averill, Opton, and Lazarus in a paper of 1969 offered what they called a “first approximation” of this kind. First, they identified three major components of any emotional response system: “input variables (stimulus properties),” the “appraiser subsystem,” and “output variables (response categories).” Input variables or stimulus properties included “information” specific to objects with emotional meaning, nonspecific information such as novelty, intensity, ambiguity, as well as input due to the response feedback from the individual. The appraiser subsystem included the individual’s primary and secondary appraisals of the stimulus situation, a subsystem that Averill, Opton, and Lazarus compared to the “decider system” in J. Miller’s general systems theory, thereby implying a problematic equivalence between their appraisal theories and the latter’s information-processing approach—a topic to which I will return in the next chapter. The authors then defined the output variables as involving three different emotional response categories: the cognitive, the expressive, and the instrumental. The cognitive response reactions included reappraisals or other modes of defense, and the authors here made a pitch for the importance of such cognitive modes of response.60
Especially interesting is what the authors had to say about the “expressive” and “instrumental” categories of response. They described expressive reactions as “stereotyped and rather discrete, e.g., weeping, laughing, and certain facial displays,” and said of them that such expressive reactions “typically are not intentionally directed toward any goal.” They included in this group “biologic expressors” such as animal displays and the fixed action patterns described by ethologists, as well as “acquired expressors” of the kind learned in childhood, as described by Ekman and Friesen and others. They contrasted those biological expressive responses with instrumental responses, consisting of “coordinated behavior sequences directed toward some goal, e.g., avoidance of harm or injury to another.” The distinction between expressive and instrumental reactions was not considered hard and fast, and the authors’ further subdivision of (p.164) instrumental responses into “symbols,” “operators,” and “conventions” was not especially clarifying.61
What should be remarked on is that in this paper we see the extent to which Lazarus’s position was beginning to resemble Ekman’s developing neurocultural theory of the emotions. From one point of view this is understandable because, as the authors report, Ekman was collaborating with Lazarus at just this time on experiments designed to determine the influence of cultural differences on emotional responses. In fact, one of the studies described by Averill, Opton, and Lazarus appears to have been the prototype for Ekman’s soon-to-be-famous study of the (so-called) “spontaneous” reactions of Japanese and American students to stressful films. In 1963–1964, Lazarus had spent a year in Japan carrying out a series of experiments designed to elucidate the cultural differences, if any, between Japanese and American coping responses, by comparing their respective reactions to the Subincision film. In their first report of this cross-cultural study, Lazarus and his colleagues reported that although the Japanese subjects responded to the film in a manner quite similar to that of their American counterparts, there was one major cross-cultural difference. Specifically, unlike their American counterparts, whose levels of physiological arousal as measured by skin conductance fluctuated throughout the film viewing process in response to the sequence of threatening and less threatening or “benign” scenes, Japanese subjects not only demonstrated consistently high levels of skin conductance during both the control and stress films, but exhibited very little variation corresponding to the threatening or benign scenes of the film. In other words, the Japanese subjects appeared to be in a continuous state of physiological arousal during the film viewing, an arousal not reflected in their self-reports. After ruling out various possible explanations for these results, Averill, Opton, and Lazarus proposed that the psychological experiment itself was threatening to the Japanese because, unlike Americans, they were not used to being observed or evaluated under such experimental conditions and therefore found the experience highly stressful. The Japanese were therefore reacting to the entire situation and not to variations in the stimulus content.62
(p.165) That finding prompted further experiments back in the United States to determine whether the results changed if the Japanese were allowed to adapt to the experimental situation. Preliminary findings, also reported by Averill, Opton, and Lazarus, indicated that the Japanese students continued to show higher levels of skin conductance throughout the experiment, even as their self-reports failed to register their apprehensions. At this juncture, Ekman entered the scene as a co-partner with Lazarus in carrying out yet another experiment, this time focused on the facial expressions of the experimental subjects during the film viewings.63 In other words, Lazarus and his group now incorporated Ekman’s interest in facial expression into their experimental protocols. In a pilot study described in the same paper by Averill, Opton, and Lazarus, hidden observers rated the expressiveness of the Japanese and American subjects as they watched control and stress films. Motion pictures were also taken of the faces and, in a further variant, the authors reported videotaping the faces of both Japanese and American subjects as they viewed a specially edited film containing scenes from four stress films, while physiological changes were continuously recorded.
This last experiment appears to have been the prototype of the famous experiment on the difference between “spontaneous” reactions of Japanese and American students to stress films carried out by Ekman and Friesen. As discussed in chapter 2, in that experiment, first reported in (p.166) 1972, Ekman claimed to have demonstrated the way in which cultural display rules controlling for polite behavior when in the presence of an authority figure caused the Japanese students, unlike the Americans, to cover over with polite smiles their natural-universal disgust response at seeing such stress films.64
On the basis of that experiment, which went on to acquire iconic status, and with reference to his cross-cultural judgment studies, Ekman argued that Tomkins was right to suggest the existence of a limited number of discrete, basic emotions or “affect programs,” programs or patterns of emotional response that could be disguised by learned, culturally determined “display rules” but that under the right conditions would “leak out.” In the same way, Lazarus began to suggest that there exist phylogenetically evolved, wired-in mechanisms of response, such as facial expressions, which could be camouflaged or inhibited, especially in humans, by the superimposition of culturally determined norms and behaviors, but which were a biological given or predisposition. It is as if Lazarus saw Ekman as providing a particular piece of the emotion puzzle (the facial expression piece), without believing that Ekman’s views compromised his own commitment to the “cognitive” piece (the cognitive appraisal piece) of the same puzzle.
Yet for Lazarus to proceed in this way was to risk undermining the coherence of his position. By declaring, as Averill, Opton, and Lazarus did in their 1969 paper, that facial expressions are not directed toward any goal, Lazarus risked severing emotion from the intentional object in ways that conformed to Tomkins’s and Ekman’s views about the non-intentional status of the basic emotions but were not coherent with his own professed intentionalism. As I have shown in previous chapters, for Tomkins and Ekman the way to understand the emotions is that they are elicited or “triggered” by what we might call the emotional object, but the object is nothing more than a stimulus or trip wire for an inbuilt behavioral response. We might put that in Tomkins’s and Ekman’s account the intentional object of the emotion is turned into the causal trigger or “releaser” of the emotional “program” or response. The issue for them was to determine the causal conditions activating the automatic discharge of the affective response. On this model, emotional responses could and did interact with the purposive-cognitive systems in the brain, but they were in principle independent of them.
The trouble with such formulations for the coherence of Lazarus’s project (p.167) is that the intentional object as such tended to disappear from view. Indeed, in their 1969 paper Averill, Opton, and Lazarus mentioned the intentional object almost as an afterthought and with no clear connection to the theoretical framework that preceded it. Thus, having offered a provisional classification of the emotions, the authors added with yet another reference to Kenny’s 1963 book that “the mere listing of reactions, no matter how complete, is insufficient to specify a response as emotional. Emotional concepts are relational, that is, they refer to responses in relation to objects, which may include complex situational and social factors (Kenny, 1963). In this respect, an emotion, which implies an object, is like an answer, which implies a question” (his emphasis).65 But this by-now-familiar statement hangs, and when the authors again compared emotions to disease syndromes, the reader is left wondering whether, in referring to Kenny’s work once again, Lazarus and his coauthors believed they were offering a conceptual analysis of emotions as intentional states or making an empirical claim about the intentional object—or both.
Moreover, as his ideas about the various components of emotions and their differentiation and classification evolved, Lazarus began to further develop the notion that appraisals fall into characteristic patterns according to whether there is a particular environmental harm or benefit for the person involved. He called these appraisal patterns “core relational themes” and tied those themes to specific emotions: thus, the appraisal of loss was linked to the emotion of sadness, and so on. His formulations came so close to Tomkins’s and Ekman’s “affect program” views as to be almost indistinguishable from them. In his much later book Emotion and Adaptation (1991), summarizing his life’s work (Lazarus retired from teaching that year), he commented:
If a person appraises his or her relationship to the environment in a particular way, then a specific emotion, which is tied to the appraisal pattern, always follows. A corollary is that if two individuals make the same appraisal, then they will experience the same emotion, regardless of the actual circumstances. I think of this as a psychobiological principle, which provides for universals in the emotion process of the human species and probably applies to other animals, too. In other words, we are constructed in such a way that certain appraisal patterns and their core relational themes will lead to certain emotional reactions. This biological principle is similar in function to (p.168) the concept of affect programs … though my version of it is highly flexible, especially with respect to the input; therefore, there is considerable variation in the exact details of the emotional response.
The psychobiological principle is essential for a cognitive-motivational-relational theory of emotion, because it implies a degree of universality—or commonality—in the emotion process, which seems evident observationally. (his emphasis)66
Here, Lazarus explicitly recognized the resemblance between his views and the Tomkins–Ekman “affect program” approach to the emotions, even as he claimed he attached greater importance than either Tomkins and Ekman did to the influence of individual experiences on the input side of the affective process.67
But, as the anthropologist Richard Shweder observed in a sympathetic yet shrewdly critical review of Lazarus’s “great book,” in which he focused on this passage, Lazarus’s core relational themes were analyzed in such abstract terms, such as “irrevocable loss” rather than “irrevocable job loss” or “irrevocable loss of a child,” that, despite his cognitivism, he ended up “short of fully defining the emotions by reference to their ‘objects’ (such as loss of a job in contrast to loss of a child).” The result was that Lazarus’s appraisal theory made it possible to treat two mental appraisals as equivalent, regardless of differences in the events or objects they were about.68 As Shweder put it: “It is assumed [by Lazarus] that mental processes do not function concretely and that, in the production of an emotional experience, there is no fundamental qualitative difference between, for example, ‘I feel the way I feel because my child has died’ and ‘I feel sad because I was fired from my job’” (“EY,” 323; his emphasis). I would supplement Shweder’s remarks by pointing out that the differences (p.169) between those two experiences of loss involve not just a difference in the two kinds of objects but also a difference in the very meaning of the concept of loss itself. On the one hand, the object in each case of loss is distinct, and that is important. On the other hand, the very grammar of loss is also different in each case. If we had to paraphrase the different kinds of losses involved in losing a child versus losing a job, we would soon find that our descriptions would diverge, because losing a child usually involves suffering and despair of a kind and depth that losing a book, or a game, or even a job does not.
Shweder went on to note in this regard the “dual ontological status” of Lazarus’s abstract core relational themes, in that the themes had both a constitutive and a causal-empirical status. By 1991 this idea was something Lazarus himself had come to embrace by proposing that the appraisal-emotion relationship was both a synthetic (or causal) and a “logical” (or “analytic”) one. But Shweder demurred. It was not just that he questioned whether appraisals were the kind of things that could be assessed empirically, especially if, as Lazarus himself believed, such appraisals were frequently automatic and unconscious. Shweder’s fundamental objection was to the very idea that there were invariant causal connections between two types of events in the natural world, the making of an appraisal and a specific emotional response. He did not think the relationship between appraisals and emotions was a causal one at all, but rather was a matter of “conceptual necessity.” Lazarus was right to think it was a “foregone conclusion” that, for example, sadness and loss are tied to each other, but “not as events in the empirical world.” They are bound to one another, Shweder argued, because the appraisal condition (loss) is intrinsic to our concept of what it is like to be sad. As he put it:
We are not biologically constructed in such a way that mental event A (the appraisal or loss) and mental event B (the experience of sadness) must go together. It is the idea of sadness that is so constructed. Internal to the idea of sadness is a connection to loss that we have no choice but to employ if we are to interpret others as sad. The link between the appraisal condition and the emotion is part of an a priori conceptual architecture, which is available for us to put to use in our attempts to comprehend others as persons and to arrive at a reading of their emotional life.
On this basis, Shweder suggested that many readers would query whether the postulated mental appraisals that Lazarus claimed were a causal condition of the emotion were anything other than a “reified redescription of the meaning of the emotion.” “It left me wondering,” Shweder wrote, (p.170) “whether it is really necessary to reify meanings (describing them as if they were antecedent causal events) in order to acknowledge their central role in our mental life” (“EY,” 323). Identifying himself as “in some sense or other” a cognitive appraisal theorist, Shweder nevertheless doubted that Lazarus’s psychobiological principle was evident observationally. He concluded:
I do not even believe that it is the kind of formulation (a “hypothesis”) that might one day be supported or disproved by new evidence from cross-cultural research. In my view, cognitive appraisal theory is not so much a theory as a framework of concepts for generating interpretations about the mental life of others … I am skeptical … because I believe that the … way appraisals and emotions are “tied to” or “bound to” one another is a matter of conceptual necessity, not causal necessity … The connections and links built into cognitive appraisal theory (loss and sadness, transgression and guilt, etc.) are not there as a result of being observed. They arise out of the meanings inherent in our emotional state concepts (the “folk psychology,” to use the contemporary philosophical parlance) that we use to “mind-read” the subjective states of others.
It is not clear to me that Lazarus was ever able to reply adequately to these fundamental criticisms and concerns.70 How he would try to do (p.171) so is the story of the attempts he made, starting in the 1980s, to respond to the challenges posed by Zajonc and other critics on the basis of a position already marred by conceptual confusions and contradictions. These developments are the topic of the next chapter.
(1.) Theodore R. Sarbin, “Emotion and Act: Roles and Rhetoric,” in R. Harré, ed., The Social Construction of Emotions (Oxford, 1986), 83.
(2.) H. Leventhal and K. R. Scherer, “The Relationship of Emotion and Cognition: A Functional Approach to a Semantic Controversy,” Cognition and Emotion 1 (1987): 3.
(3.) In their interesting discussion of Carl Stumpf’s Brentano-influenced intentionalist account of the emotions, Reisenzein and Schönpflug have rightly observed that the concept of intentionality has tended to be a crucial missing term in the post–World War II controversy over the cognition-emotion relationship. Rainer Reisenzein and Wolfgang Schönpflug, “Stumpf’s Cognitive-Emotional Theory of Emotion,” American Psychologist 47 (1) (1992): 34–45.
(4.) Stuart Shanker, Wittgenstein’s Remarks on the Foundations of AI (London and New York, 1998).
(5.) Phil Hutchinson, Shame and Philosophy: An Investigation in the Philosophy of Emotion and Ethics (Basingstoke and New York, 2008); Hutchinson, “Emotion-Philosophy-Science,” in Emotions and Understanding: Wittgensteinian Perspectives, eds. Ylva Gustafsson, Camilla Kronqvist, and Michael McEachrane (Basingstoke and New York, 2009), 60–80; John McDowell, Mind and World (Cambridge, 1994).
(6.) Richard S. Lazarus and Robert A. McCleary, “Autonomic Discrimination Without Awareness: A Study of Subception,” Psychological Review 58 (1951): 113–22. The authors coined the term “subception” in an earlier paper: Robert A. McCleary and Richard S. Lazarus, “Autonomic Discrimination Without Awareness: An Interim Report,” in Perception and Personality: A Symposium, eds. Jerome S. Bruner and David Krech (Durham, NC, 1949 and 1950), 171–79.
(7.) Richard S. Lazarus, “A Cognitivist’s Reply to Zajonc on Emotion and Cognition,” American Psychologist (February 1981): 223. Or, as he also later noted in a comment that brings out the intentionalism of his views about perceptual defense, when people appraise a situation unconsciously, the emotions experienced or displayed seem to make little or no sense “because the intentional premises are hidden.” Richard S. Lazarus, “Cognition and Motivation in Emotion,” American Psychologist 46 (April 1991): 363.
(8.) On the “New Look” in American psychology, see Jerome S. Bruner and Cecile C. Goodman, “Value and Need as Organizing Factors in Perception,” Journal of Abnormal and Social Psychology 42 (1947): 33–44; Jerome S. Bruner and George S. Klein, “The Function of Perceiving: New Look Retrospect,” in Perspectives in Psychological Theory, eds. S. Wapner and B. Kaplan (New York, 1960), 61–77.
(9.) As Lazarus observed, “psychologists in those days were simply not ready to assimilate comfortably the idea of unconscious processing, especially if it contained the dynamic implication of Freudian thought that sometimes ideas are kept out of awareness as an ego-defense.” Lazarus, “Cognition and Motivation in Emotion,” 360.
(10.) For a discussion of the bitter and sustained controversy over perceptual defense, see for example M. H. Erdelyi and B. Goldberg, “Let’s Not Sweep Repression Under the Rug: Toward a Cognitive Psychology of Repression,” in Functional Disorders of Memory, eds. J. F. New Kihlstrom and F. J. Evans (New York, 1979), 378–79.
(11.) Norman F. Dixon, Subliminal Perception: The Nature of a Controversy (London, 1971), 179. Cf. W. P. Brown, “Conceptions of Perceptual Defence,” British Journal of Psychology, Monograph Supplements (Cambridge, 1961). As Dixon noted, “far from being an unlikely phenomenon, perceptual defence was merely a laboratory demonstration of processes akin to those underlying hysterical blindness and psychogenic deafness” (Subliminal Perception, 180). Yet the notion of perceptual defense, or subliminal perception, proved immensely controversial. Of the great number of contradictory results produced during the debate over subliminal perception, Dixon wrote: “Truly, we are in deep waters” (Subliminal Perception, 188). He went on to observe that each new methodological departure only served to introduce new problems, and added: “One begins to suspect that there is something inherent in the logic of the situation which always precludes a satisfactory solution” (Subliminal Perception, 186). Commenting on the possibility that demand characteristics might have influenced the results of some experiments, Dixon observed that “some experimenters, some academic departments, and even some countries, invariably provide positive evidence for subliminal effects, while others with almost equal monotony do not. This suggests that other things being equal, differences in the belief system of the experimenter concerned is a crucial and deciding variable, and that this communicates itself in some way to the experimental subject, and makes the subject receptive or resistant to subliminal effects” (Subliminal Perception, 242).
(12.) Leo Postman, Jerome S. Bruner, and Elliott McGinnies had already voiced that concern in “Personal Values as Selective Factors in Perception,” Journal of Abnormal and Social Psychology 43 (1948): 142–54. “One may inquire at this point,” they wrote, “How does the subject ‘know’ that a word should be avoided? In order to ‘repress,’ he must first recognize it for what it is. We have no answer to propose … Of only one thing we can be fairly sure: reactions occur without conscious awareness of what one is reacting to. Psychological defense in perception is but one instance of such ‘unconscious’ reaction” (153).
(13.) On the dilemmas of Freud’s solution to the problem of self-deception, see Simon Boag, “Realism, Self-Deception, and the Logical Paradox of Repression,” Theory and Psychology 17 (2007): 421–47. As Walter Benn Michaels has remarked, the paradox is at once psychological and epistemological: it is psychological in that according to Freud’s theory of repression it involves the refusal to know something based on the theory of repression; and it is epistemological because it involves the paradox of self-deception according to which the subject simultaneously does and doesn’t believe something, an impossibility formulated by Wittgenstein in his discussion of Moore’s paradox. See Walter Benn Michaels, “The Death of a Beautiful Woman: Christopher Nolan’s Idea of Form,” www.electronicbookreview.com, October 1, 2007.
(14.) Dixon proposed just such a solution: “Subliminal perception can be envisaged as very largely a physiological phenomenon, and needs to be explained in physiological terms. It is concerned with neural processes initiated by a physical stimulus and terminating in a physiological response, whether this involves activity of the speech muscles or activity of the autonomic nervous system. By definition, there is no psychological (mental) content to this process, and, therefore, no solely psychological explanation of the phenomenon can possibly be adequate. To talk of unconscious feelings, memories, wishes, and fantasies is a contradiction in terms, and adds little by way of explanation” (Subliminal Perception, 244).
(15.) Hubert Dreyfus, What Computers Can’t Do: The Limits of Artificial Reason (New York, 1972).
(16.) Matthew Hugh Erdelyi, “A New Look at the New Look: Perceptual Defense and Vigilance,” Psychological Review 81 (1974): 1–25.
(17.) As Erdelyi himself noted of his 1974 contribution: “There is probably no defense process discussed by Freud that does not naturally lend itself to a computer articulation. … We have already seen that certain logical and philosophical objections to defense processes collapsed with the advent of the computer analogy. If the computer could be programmed to simulate perceptual defense effects, then all the paradoxes that supposedly rendered the process impossible must be in error. Of if a persuasive philosopher such as Sartre proves that a system cannot lie to itself (and that, therefore, there cannot be self-deception and, therefore, no repression), the idiot savant computer can set matters straight immediately by showing that it can deceive itself and thereby undeceive us brighter human savants.” Erdelyi and Goldberg, “Let’s Not Sweep Repression Under the Rug: Toward a Cognitive Psychology of Repression,” 390.
(18.) Ruth Leys, From Guilt to Shame: Auschwitz and After (Princeton, 2007).
(19.) For the history of the stress concept, see especially Stress, Shock, and Adaptation in the Twentieth Century, eds. David Cantor and Edmund Ramsden (Rochester, NY, 2014); and Cary L. Cooper and Philip Dewe, Stress: A Brief History (Maiden, MA, 2004).
(20.) Richard S. Lazarus, Psychological Stress and the Coping Process (1966), viii–ix; hereafter abbreviated PS. Lazarus suggested that, although the issues encompassed by the word “stress” had often been considered under the rubric of “emotion,” the word “stress” was adopted because it connoted the negative aspects of emotion, such as fear, anger, and depression, and emphasized disturbances in or failures of psychological and biological adaptation, an emphasis not found in the concept of emotion. He also suggested that in the era of behaviorism, the engineering concept of stress might have been more acceptable to psychologists than the word “emotion,” with its implied reference to states of mind (PS, 10). See also Richard S. Lazarus, “From Psychological Stress to the Emotions: A History of Changing Outlooks,” Annual Reviews 44 (1993): 1–22.
(21.) Richard S. Lazarus, “A Laboratory Approach to the Dynamics of Psychological Stress,” American Psychologist 19 (1964): 404.
(22.) Richard S. Lazarus, The Life and Work of an Eminent Psychologist: Autobiography of Richard S. Lazarus (New York, 1998), chapters 6–8; Richard S. Lazarus, James Deese, and Sonia Foster, “The Effects of Psychological Stress Upon Performance,” Psychological Bulletin 49 (4) (1952): 293–371; Richard S. Lazarus and C. W. Eriksen, “Effects of Failure Stress Upon Skilled Performance,” Journal of Experimental Psychology 43 (1952): 100–105; Richard S. Lazarus and J. Keenan, “Anxiety, Anxiety-Reduction and Stress in Learning,” Journal of Experimental Psychology 46 (1953): 55–61; Richard S. Lazarus and Robert W. Baker, “Psychology,” Progress in Neurology and Psychiatry 11 (1956): 253–71; Richard S. Lazarus, “Motivation and Personality in Psychological Stress,” Psychological Newsletter 8 (1957): 159–93; Richard S. Lazarus, “A Program of Research in Psychological Research,” in Festschrift for Gardner Murphy, eds. J. G. Peatman and E. L. Hartley (New York, 1960), 313–29; William Vogel, Robert W. Baker, and Richard S. Lazarus, “The Role of Motivation in Psychological Stress,” Journal of Abnormal and Social Psychology 56 (1) (1958): 105–12; William Vogel, Susan Raymond, and Richard S. Lazarus, “Intrinsic Motivation and Psychological Stress,” Journal of Abnormal and Social Psychology 58 (2) (1959): 225–33.
(23.) Richard S. Lazarus, “Emotions and Adaptation: Conceptual and Empirical Relations,” in The Nebraska Symposium on Motivation, ed. W. J. Arnold (Lincoln, NE, 1968), 175–76.
(24.) Subincision, which was made by the anthropologist Géza Róheim in the 1940s, had already been used to study castration and mutilation fantasies experimentally. See J. B. Schwarz, “An Empirical Test of Two Freudian Hypotheses Concerning Castration Anxiety,” Journal of Personality 24 (1956): 318–37; and A. Aas, Mutilation Fantasies and Autonomic Response (Oslo, 1958). For a contemporaneous discussion of the meaning of the subincision ritual, see G. Róheim, “The Symbolism of Subincision,” American Imago 6 (1949): 321–28; J. E. Cawe, N. Djagamarra, and M. G. Barrett, “The Meaning of Subincision of the Urethra to Aboriginal Australians,” British Journal of Medical Psychology 39 (1966): 246–53; and Ashley Montagu, Coming Into Being Among the Australian Aborigines: The Procreative Beliefs of the Australian Aborigines (London, 1974). Lazarus came to recognize that the content of the film was very complex, so it was not clear what aspects of it were most important to each individual viewer. Themes included fantasies of castration, sexual exposure, mutilation, and pain, all of which were confounded. He eventually concluded that the film was too complex to serve as a useful instrument of further research, and his turn away from laboratory to clinical research seems to have been motivated in part by this insight.
(25.) Joseph C. Speisman, Janet Osborn, and Richard S. Lazarus, “Cluster Analyses of Skin Resistance and Heart Rate at Rest Under Stress,” Psychosomatic Medicine 23 (4) (1961): 323–43; Richard S. Lazarus, Joseph C. Speisman, Arnold M. Mordkoff, and Leslie A. Davison, “A Laboratory Study of Psychological Stress Produced by a Motion Picture Film,” Psychological Monographs 34 (1962): 144–56; Richard S. Lazarus, Joseph C. Speisman, and Arnold M. Mordkoff, “The Relationship Between Autonomic Indicators of Psychological Stress: Heart Rate and Skin Conductance,” Psychosomatic Medicine 25 (1) (1963): 19–30; Richard S. Lazarus, “A Laboratory Approach to the Dynamics of Psychological Stress,” American Psychologist 19 (1964): 400–411; Richard S. Lazarus and Elizabeth Alfert, “Short-Circuiting of Threat by Experimentally Altering Cognitive Appraisal,” Journal of Abnormal and Social Psychology 69 (1964): 195–205; Joseph C. Speisman, Richard S. Lazarus, Arnold M. Mordkoff, and L. A. Davison, “The Experimental Reduction of Stress Based on Ego-Defense Theory,” Journal of Abnormal and Social Psychology 68 (4) (1964): 367–80; Richard S. Lazarus, Edward M. Opton Jr., Markellos S. Nomikos, and Neil O. Rankin, “The Principle of Short-Circuiting of Threat: Further Evidence,” Journal of Personality 33 (1965): 622–35; Edward J. Malmstrom, Edward M. Opton Jr., and Richard S. Lazarus, “Heart Rate Measurement and the Correlation of Indices of Arousal,” Psychosomatic Medicine 27 (1965): 546–56; Richard S. Lazarus and Edward M. Opton Jr., “The Study of Psychological Stress: A Summary of Theoretical Formulations and Experimental Findings,” in Anxiety and Behavior, ed. Charles D. Spielberger (New York, 1966), 225–62.
(26.) For biographical information about Arnold and discussions of her contributions, see Cognition and Emotion 20 (7) (2006), a special issue devoted to Arnold, edited by Stephanie Shields and Arvid Kappas.
(29.) Lazarus, The Life and Work of an Eminent Psychologist, 169. Lazarus added that Arnold’s book influenced him to abandon the term “perception,” which he had originally used “in its broadest sense to imply personal meaning,” and instead to employ the term “appraisal” systematically “because it clearly denoted evaluation” (169). This suggests that it was the substitution of Arnold’s term “appraisal” for that of “perception” that played a role in Lazarus’s mistaken separation of these two functions or processes, discussed below.
(30.) Rainer Reisenzein, “Arnold’s Theory of Emotion in Historical Perspective,” Cognition and Emotion 20 (7) (2006): 920–51.
(31.) Jean-Paul Sartre, The Emotions: Outline of a Theory (New York, 1948), 51–52.
(32.) Magda Arnold, Emotion and Personality (New York, 1960), vol. 1, Psychological Aspects, 158–59; hereafter abbreviated EP.
(33.) See in this connection Randolph R. Cornelius, The Science of Emotion: Research and Tradition in the Psychology of Emotions (New York, 1996), 117.
(34.) For useful recent discussions of James’s views, about the import of which there is still no consensus, see “William James and His Legacy,” a special section devoted to his work in Emotion Review, January 2014.
(37.) In his 1966 book on stress, in a section entitled “Examples Where Appraisal Deviated from the Objective Facts,” Lazarus defines “the point of view of the objective facts” as facts judged “consensually” (PS, 109).
(38.) Anthony Kenny, Action, Emotion and Will (London, 1963; hereafter abbreviated as AEW. A few years later the philosopher Norman Malcolm offered a related Witt-gensteinian critique of the attempt to explain intention and other mental concepts by appealing to internal “cognitive processes,” focusing especially on Chomsky’s hypothesis of the existence of an innate grammatical structure. See his “The Myth of Cognitive Processes and Structures,” in Cognitive Development and Epistemology, ed. Theodore Mischel (New York, 1971), 385–92.
(39.) As we saw in chapter 1, Tomkins in 1962 had indeed proposed that there exists a number of discrete emotions or “affect programs” with signature physiological and behavioral characteristics. Kenny in 1963 did not cite Tomkins but relied on reports of experiments in Robert S. Woodworth’s influential textbook Experimental Psychology ( 1954, 2nd ed., with Harold S. Schlosberg), as well on more recent findings (AEW, 34ff).
(41.) Richard S. Lazarus, James R. Averill, and Edward M. Opton Jr., “Towards a Cognitive Theory of Emotion,” in Feelings and Emotions: The Loyola Symposium, ed. Magda N. Arnold (New York and London, 1970), 207–32; hereafter abbreviated as “TCTE.”
(42.) In another paper, Averill in stated in words very close to Lazarus’s that “The concept ‘emotion’ implies an object, just as the concept ‘answer’ implies a question. That is, emotional terms refer to relational as well as to intrinsic properties of the response.” Averill added: “This should not, of course, be taken to mean that an emotion is more than its manifestations, any more than an answer is more than the statement which expresses it. The concept of object as it applies to emotional response is quite complex (Gosling, 1965; Kenny, 1963) and is not discussed here. At the risk of oversimplification, it may simply be said that each emotion is logically appropriate to a certain class of objects” ( James R. Averill, “Grief: Its Nature and Significance,” Psychological Bulletin 70 : 742 ). Gosling’s 1965 paper, cited by Averill, was a critical review of Kenny’s 1963 book in which the author discussed the complexity of the distinction between the concepts of “object” and “cause.” Gosling focused especially on the familiar question of (so-called) “objectless emotions,” such as states of “pointless depression,” states that, he argued, posed an obstacle to Kenny’s intentionalist position. See J. C. Gosling, “Emotion and Object,” Philosophical Review 74 (4) (1965): 486–503.
(43.) In yet another paper, Averill referred to works by Ryle, Wittgenstein, Austin, and other authors, including papers by philosophers Bedford and George Pitcher that critiqued Cartesian-causal theories and emphasized instead the intentionality and object-directed nature of emotions in terms not unlike Kenny’s. Averill’s article was a brave attempt to offer criticisms of efforts to achieve scientific precision about language by operationalizing meaning, in the course of which he made some pertinent comments about Wittgenstein’s and Ryle’s critiques of the notion of private psychological states and their reification as mental events. See James R. Averill, “Operationism, Metaphysics, and the Philosophy of Ordinary Language,” Psychological Reports 22 (1968): 861–87. But it is a sign of the difficulty of understanding Wittgenstein’s contribution to ordinary language philosophy that in the course of his article Averill cited as having “much merit” (884) Jerry Fodor and J. J. Katz’s famous (or infamous) attack on the philosopher Stanley Cavell’s remarkable essays “Must We Mean What We Say?” and “The Availability of Wittgenstein’s Later Philosophy,” whose aim was precisely to defend Wittgenstein’s procedures against Fodor and Katz’s empirical-positivist misunderstandings and to say what he thought those procedures are, properly understood. In this connection, see Stanley Cavell, “Must We Mean What We Say?,” Inquiry 1 (1958): 172–212; Stanley Cavell, “The Availability of Wittgenstein’s Later Philosophy,” Philosophical Review 71 (1962): 67–93; J. A. Fodor and J. J. Katz, “The Availability of What We Say,” Philosophical Review 72 (1963): 57–71; Stanley Cavell, “Aesthetic Problems of Modern Philosophy,” in Must We Mean What We Say? (New York, 1969), 73–96; Richard Henson, “What We Say,” American Philosophical Quarterly 2 (1) (1965): 52–62; and Stanley Bates and Ted Cohen, “More on What We Say,” Metaphilosophy 3 (1) (1972): 1–24.
(44.) Cavell writes: “We know of the efforts of such philosophers as Frege and Husserl to undo the ‘psychologizing’ of logic (like Kant’s undoing Hume’s psychologizing of knowledge): now, the shortest way I might describe such a book as [Wittgenstein’s] Philosophical Investigations is to say that it attempts to undo the psychologizing of psychology, to show the necessity controlling our application of psychological and behavioral categories; even, one could say, show the necessities in human action and passion themselves” ( Cavell, “Aesthetic Problems of Modern Philosophy,” in Must We Mean What We Say, 91 ). In a footnote Cavell quotes from and then comments on a passage from Wittgenstein’s text: “Consider, for example, the question: ‘Could someone have a feeling or ardent love or hope for the space of one second—no matter what preceded or followed this second?’ (Investigations, ¶583). We shall not wish to say that this is logically impossible, or that it can in no way be imagined. But we might say: given our world this cannot happen; it is not, in our language, what ‘love’ or ‘hope’ mean; necessary in our world that this is not what love and hope are. I take it that our most common philosophical understanding of such notions as necessity, contingency, synthetic and analytic statements, will not know what to make of our saying such things” (91). Everything challenging Lazarus’s attempt to establish a science of emotions is contained in these words. The question posed by them is: what kind of scientific enterprise in psychology would make sense in the light of the considerations Cavell raises?
(45.) Frederick Stoutland, “Introduction: Anscombe’s Intention in Context,” in Essays on Anscombe’s Intention, eds. Anton Ford, Jennifer Hornsby, and Frederick Stoutland (Cambridge, MA, 2011), 6–8.
(46.) Moreover, I do not want to suggest that Kenny’s analysis of the nature of emotions as intentional actions was so brilliantly and coherently carried out that its message ought to have been obvious. On the contrary, not only were the issues he was dealing with inherently difficult, it’s not even clear how sound was his own grasp of Wittgenstein’s ordinary language procedures and their implications for psychology. As the recent collection Essays on Anscombe’s Intention makes evident, the interpretation of Anscombe’s discussion of intention and action on which Kenny drew remains to this day a topic of debate and contestation.
(47.) See Richard Lazarus, Masatoshi Tomita, Edward Opton Jr., and Masahisa Kodama, “A Cross-Cultural Study of Stress-Reaction Patterns in Japan,” Journal of Personality and Social Psychology 4 (1966): 622–33; James R. Averill, Edward H. Opton Jr., and Richard S. Lazarus, “Cross-Cultural Studies of Psychophysiological Responses During Stress and Emotion,” International Journal of Psychology 4 (2) (1969): 91; and Richard S. Lazarus and James R. Averill, “Emotion and Cognition: With Special Reference to Anxiety,” in Anxiety: Current Trends in Theory and Research, ed. C. D. Spielberger (New York, 1972), 253–54. In these experiments, the general patterns of reaction were assessed on three dependent variables: skin conductance, distress rating, and a modified Nowlis Adjective Check List of Mood. The latter did not measure basic emotion categories but the following items: Concentration, Aggression, Pleasantness, Activation, Deactivation, Egotism, Social Affection, Depression, and Anxiety. For details concerning the Nowlis checklist, which originally contained 145 adjectives, see Vincent Nowlis, “The Development and Modification of Motivational Systems in Personality,” in Current Theory and Research in Motivation: A Symposium (Lincoln: University of Nebraska Press, 1953), 114–38; and V. Nowlis and H. Nowlis, “The Description and Analysis of Mood,” Annals of the New York Academy of Science 65 (1965): 345–55.
(48.) See for example W. G. Parrott, “The Role of Cognition in Emotional Experience,” in Recent Trends in Theoretical Psychology: Proceedings of the Second Biannual Conference of the International Society of Theoretical Psychology, April 20–25, 1987, eds. W. J. Baker, L. P. Mos, H. V. Rappard, and H. J. Sam (New York, 1988), in which, in the process of critiquing Jamesian-feeling approaches to the emotions and defending a cognitivist position, the author states: “It is well known that this ‘Cartesian’ view of emotion succumbs to a variety of criticisms. Wittgensteinians have pointed out that this view falls prey to the problems of any ‘private language’ (Kenny, 1963)” (331).
(49.) Paul E. Griffiths, “Current Emotion Research in Philosophy,” Emotion Review 5 (2) (2013): 215–22. See also Andrea Scarantino, Explicating the Emotions, unpublished doctoral dissertation, University of Pittsburgh, 2005, which cites and discusses aspects of Kenny’s 1963 book in connection with the problem of intentionality and the latter’s emphasis on the role of the formal object in emotion. Of course, Kenny would not have endorsed Scarantino’s attempt to naturalize the intentionality of the emotions by appealing to Ruth Millikan’s teleosemantic theory.
(50.) For Lazarus’s reasons for transitioning to clinical research, see Richard S. Lazarus and Raymond Launier, “Stress-Related Transactions Between Person and Environment,” in Perspectives in Interactional Psychology, eds. L. A. Pervin and M. Lewis (New York, 1978), 287–327; Richard S. Lazarus and Susan Folkman, “Transactional Theory and Research on Emotions and Coping,” European Journal of Personality 1 (3) (1987): 141–69; Richard S. Lazarus, James C. Coyne, and Susan Folkman, “Cognition, Emotion and Motivation: The Doctoring of Humpty-Dumpty,” in Psychological Stress and Psychopathology, ed. R. W. J. Neufeld (New York, 1982), 218–39; Richard S. Lazarus, “Vexing Research Problems Inherent in Cognitive-Mediational Theories of Emotion and Some Solutions,” Psychological Inquiry 6 (1995): 183–96; and Richard S. Lazarus, The Life and Work of An Eminent Psychologist, chapter 9.
(51.) See Ira Roseman and Craig A. Smith, “Appraisal Theory: Overview, Assumptions, Varieties, Controversies,” in Appraisal Processes in Emotion: Theory, Methods, Research, eds. Klaus R. Scherer, Angela Schorr, and Tom Johnstone (Oxford, 2001), 3–19. See also Angela Schorr, “Appraisal: The Evolution of an Idea,” in ibid., 20–34. Appraisal theorist Phoebe Ellsworth has recently observed that basic emotion theories were the dominant psychological theories at the time she and other authors in the 1980s tried to show that the emotions proposed by the basic emotion theorists could be associated with a distinctive combination of appraisals. However, the author points to two “regrettable consequences” of that step: first, it suggested that appraisal theorists were just like the basic emotion theorists in positing the emotions as natural kinds, a view Ellsworth now rejects in the light of criticisms by Barrett (2006) and others; and second, the commitment to the basic emotions distracted attention from certain fundamental assumptions of appraisal theory, namely (1) that appraisals are continuous, and not categorical; (2) that a person can feel emotional even if the combination of appraisals does not correspond to any of the basic emotions; and (3) that emotional experience is not a state, but a “process” in which changes in appraisals, bodily responses, and action tendencies feeding back into each other transforming the emotional experience. See Phoebe C. Ellsworth, “Appraisal Theory: Old and New Questions,” Emotion Review 5 (2) (2013): 126–28.
(52.) Magda B. Arnold, EP, especially chapter 10. See also Reisenzein, “Arnold’s Theory of Emotion in Historical Perspective,” 920–51, on Arnold’s debt to William McDougall’s instinct theory for her ideas about the existence of “discrete” or “basic” emotions.
(54.) Richard Lazarus, “Relational Meaning and Discrete Emotions,” in Appraisal Processes in Emotion: Theory, Methods, Research, eds. Scherer, Schorr, and Johnstone, 37.
(56.) “The point is that each emotion involves its own particular kind of appraisal,” Lazarus wrote in 1970, “its own particular kind of action tendencies, and hence its own particular constellation of physiological changes which are part of the mobilization to action … [T]he motor-expressive features (facial expressions or bodily postures) are in many cases universal in the species, though modifiable by social custom.” Lazarus, Averill, and Opton, “Towards a Cognitive Theory of Emotion,” 218.
(57.) “[T]he cognitions determining different emotional responses (syndromes) must be identified, and in turn, this leads naturally to a search for the antecedent conditions, both in the stimulus configuration and within the psychological structure of the individual, which govern the expression of the component reactions … [T]his plan … must be filled in by substantive theoretical propositions about the intervening cognitions of anger, fear, grief, joy, etc., and the conditions under which they will emerge.” Ibid., 229.
(59.) See for example Lazarus, “Emotions and Adaptation: Conceptual and Empirical Relations,” 208, citing Paul Ekman, “Communication Through Non-Verbal Behavior: A Source of Information About an Interpersonal Relationship,” in Affect, Cognition and Personality: Empirical Studies, eds. S. S. Tomkins and C. E. Izard (New York, 1965), 390–442. See also James R. Averill, Edward M. Opton Jr., and Richard S. Lazarus, “Cross-Cultural Studies of Psychophysiological Responses During Stress and Emotion,” 88, for references to Paul Ekman and Wallace V. Friesen, “The Repertoire of Nonverbal Behavior-Categories: Origins, Usage, and Coding,” Semiotica 1 (1969): 49–98.
(62.) R. S. Lazarus, E. Opton, M. Tomita, and M. Kodoma, “A Cross-Cultural Study of Stress-Reaction Patterns in Japan,” Journal of Personality and Social Psychology 4 (1966): 633–23. Lazarus made the same point in Averill, Opton, and Lazarus, “Cross-Cultural Studies of Physiological Responses During Stress and Emotion,” 97.
(63.) Ibid., 98, n. 4. For Ekman and Friesen’s report on the origin of these experiments, see their “The Repertoire of Nonverbal Behavior: Categories, Origins, Usage, and Coding,” 81: “This research is being conducted in collaboration with Lazarus, Averill, and Opton, utilizing their stress-inducing procedure … In our study we take motion pictures of the subject’s facial expressions and hand movements without his knowledge, while he watches a stress film and a neutral film. We have collected pilot data, utilizing very brief samples of nonverbal behavior both in Japan and in the U.S. Our analysis of these records involves both the application of FAST and the collection of Japanese and U.S. observers’ interpretation of the affect shown in the Japanese and U.S. films. The FAST analysis is designed to determine whether the same muscles move in response to stress in both cultures, and whether the same display rules are exhibited in the two cultural groups. The observers’ interpretations of the stimuli will reveal whether both cultures interpret similarly the stress reactions of members of their own and of another culture. It is too early to report results from this study, other than the impressions that the procedures have worked and information is obtainable from the experiments.” In Paul Ekman, Wallace V. Friesen, and Silvan S. Tomkins, “Facial Affect Scoring Technique: A First Validity Study,” Semiotica 3 (1971): 57, the authors cite an unpublished manuscript, P. Ekman, R. S. Lazarus, W. V. Friesen, E. T. Opton, J. R. Averill, and E. J. Malmstrom, “Facial Behavior and Stress in Two Cultures.” As far as I have been able to determine, that paper was never published in this form.
(64.) Paul Ekman, “Universals and Cultural Differences in Facial Expressions of Emotion,” in Nebraska Symposium of Motivation 1971, ed. J. Cole (Lincoln, NE, 1972), 207–83.
(66.) Richard S. Lazarus, Emotion and Adaptation (New York, 1991), 191.
(67.) For the convergence between Lazarus’s views and Ekman’s position on discrete emotions, see Paul Ekman, The Face of Man: Expressions of Universal Emotions in a New Guinea Village, foreword by Richard S. Lazarus (New York, 1988); Paul Ekman, “An Argument for Basic Emotions,” Cognition and Emotion 6 (3–4) (1992): 187; Richard Lazarus, “Universal Antecedents of the Emotions,” and Paul Ekman, “Antecedent Events and Emotion Metaphors,” in The Nature of Emotion: Fundamental Questions, eds. P. Ekman and R. J. Davidson (New York and Oxford, 1994), 146–49, 163–17; and Paul Ekman, “What We Become Emotional About,” in Feelings and Emotion: The Amsterdam Symposium, eds. Antony S. R. Manstead, Nico Frijda, and Agneta Fischer (Cambridge, 2004), 119–35.
(68.) Richard A. Shweder, “Everything You Ever Wanted to Know About Cognitive Appraisal Theory Without Being Conscious of It,” Psychological Inquiry 4 (4) (1993): 322–42; hereafter abbreviated as “EY.”
(69.) For Shweder, the “ontological” status of appraisal is still a central topic. In 2014 he identified the same conflict between “causal” versus “constitutive” accounts of appraisal at work in four different theoretical approaches to the emotions, associated with the work of Agnes Moors, Jessica Tracy, Batja Mesquita and Michael Boiger, and Lisa Feldman Barrett, respectively. He argued with reference to Lazarus’s earlier views that “the connection between an emotion concept (such as shame or pride) and its ‘appraisal’ condition is not falsifiable, but is rather a constitutive feature of its very meaning.” See Richard A. Shweder, “Comment: The Tower of Appraisals: Trying to Make Sense of the One Big Thing,” Emotion Review 6 (4) (2014): 324.
(70.) For a very different critique of the same 1991 book by Lazarus, undertaken from the perspective of a researcher committed to a neurobiological approach to the emotions, see Jan Panksepp, “Where, When, and How Does an Appraisal Become an Emotion? ‘The Times They Are A’Changing,’” Psychological Inquiry 4 (4) (1993): 334–42. Here, Panksepp suggests that the “affect programs” mentioned by Lazarus as comparable to his psychobiological principle are isomorphic with the highly conserved, self-sufficient “emotional command systems” that Panksepp identifies as the fundamental affect systems of the mammalian brain. Panksepp thus suggests that the basic emotional systems function independently of the later-developing cognitive processes that may eventually be involved in the emotions. For Panksepp, appraisal thus comes late to the operation of the organism’s fundamental emotional systems. For Pank-sepp’s recent defense of his version of the basic emotions approach and Lisa Feldman Barrett’s post-Fridlund and post-Russell critique of the idea of emotions as discrete natural kinds, a critique in which Barrett implicates Lazarus, see Lisa Feldman Barrett, “Are Emotions Natural Kinds?,” Perspectives on Psychological Science 1 (1) (2006): 28–58; Jan Panksepp, “Neurologizing the Psychology of Affects: How Appraisal-Based Constructions and Basic Emotion Theory Can Coexist,” ibid., 2 (3) (2007): 281–96; and Lisa Feldman Barrett, Kristen A. Lindquist, Eliza Bliss-Moreau, Seth Duncan, Maria Gendron, Jennifer Mize, and Lauren Brennan, “Of Mice and Men: Natural Kinds of Emotions in the Mammalian Brain? A Reply to Panksepp and Izard,” ibid., 2 (3) (2007): 297–312.