By Duncan Lucas, PhD. and Maria McManus
Carroll Izard is a widely cited theorist and researcher on emotion, possibly more well-known today than his mentor, Silvan Tomkins. People familiar with Izard’s theories might be interested in understanding the similarities with and differences from Tomkins’s theories, which this article attempts to provide.
Carroll Izard worked closely with Silvan Tomkins on three projects in the 1960s, though their professional careers diverged in 1972, with the publication of Izard’s Patterns of Emotions. Their falling-out over questions of attribution left Tomkins with a complicated legacy. Tomkins’s theory, already complex, dense and far-reaching, needed talented champions. As we shall see, Izard advanced Tomkins’s theories only up to a point, while confounding the theoretical legacy with conceptual conflicts and neglecting to fully credit Tomkins as the source of much of what made Izard’s work salient. While making important contributions to the science of emotions by overseeing numerous studies, Izard was not always an advocate of Tomkins’s theories, so it is important to disentangle how Izard’s work supported and abandoned key aspects of Tomkins’s theories.
Izard is most known for his Discrete Emotions Theory (DET), which he detailed in his well-known Human Emotions (1977). Compared with Tomkins, Izard lays out five slightly different key assumptions:
- There are ten fundamental emotions;
- Each emotion is uniquely motivational and phenomenological;
- Fundamental emotions “lead to different inner experiences and different behavioral consequences” (43);
- Emotions interact, and may activate, amplify, attenuate one another;
- Emotions “interact with and exert influence on homeostatic, drive, perceptual, cognitive, and motor processes” (43).
While these are clearly adopted from Tomkins’s Affect Imagery Consciousness (1962, 1963), there are several important theoretical differences, the most significant of which are Izard’s use of the word ‘emotion’ instead of ‘affect’, which establishes foundational distinctions between the two theories, and Izard’s ten emotions rather than nine affects.
Izard uses ‘emotion’ as the unit of analysis; Tomkins uses ‘affect’
While Izard uses the word ‘emotion’ as the basis of his theory, Tomkins uses the term ‘emotion’ sparingly, preferring instead to invest in a precise and distinct definition of the term ‘affect.’ As concepts, ‘affect’ and ‘emotion’ are anything but interchangeable. For Tomkins, ‘affect’ was a component part of ‘emotion’. If one is to understand the inner workings of a complex system, it is important not to confuse the ingredients with the final product. Eggs, milk, flour and sugar are not the same as cake, scones or cookies. Nuts, bolts and gears are not the same as an engine. For Tomkins, an affect is an innate biological response that combines with cognition, memory, and other affects to create emotions. As such, affects are a unit of analysis at a wholly different level than emotions and so they should not be conflated.
Fundamentally, for Tomkins, ‘affects’ are neuro-bio-chemical adaptions inherited through evolution, and they are the primary motivators of humans. Based on reading Tomkins, Basch (1976) and Nathanson (1992) clarified the tripartite affect-feeling-emotion distinction as biology-psychology-biography. As affect is biology, so feeling is awareness that an affect has been triggered thus making it psychological. Emotion then is a more complex and compounded individual, biographical aggregation of affect-feeling with memory.
For Tomkins, it was critical to investigate affect as a specifically defined component of emotion because the language of emotion has too many ambiguous labels and phenomenological discrepancies. Tomkins describes his efforts to make sense of commonly-used language of emotions.
“Indeed, it is the ambiguity of language with respect to affect per se which defeated me in my attempt to create an affect dictionary . . . I studied several thousand English words for two years and had eventually to give up the attempt because of the great variety of admixtures of affect with cognitive, behavioral, and event references which made it impossible to code for an unambiguous affect of reference . . . Important as subtle distinctions between affect complexes are, it is nonetheless critical that such complexes not be confused with the very restricted number of biological primary affects. Because there are an infinite number of the former, these are already forming the basis of competing classifications and lend themselves to adversary debate and magnification and theoretical confusion. The confusion arises because there is no theoretical basis for deciding between classifications of combinations—any more than one could define a limited classification of types of sentences, in contrast to a question of how many letters there were in any specific alphabet.” (Affect Imagery Consciousness III page 51-52).
While Tomkins is looking to the biological primary affects as a means of systematically slicing through the ambiguity of emotional language, Izard, arguing for a differentiated set of basic emotions, places less value on the biological basis of emotion. Izard writes: “Emotions, like all functions of the organism, have neural and biochemical substrates. I do not consider neural and biochemical substrates as emotion or as ‘cause’ or ‘determinants’ of emotion. The natural or spontaneous functioning may instigate emotion, but in general they are viewed as the neuroanatomical and biochemical bases of emotion” (1972, 24). While this last sentence sounds distinctly like Tomkins’s ‘affect’, Izard doesn’t say so, nor does he consider these biological responses as important to understanding the emotional system as Tomkins does. To extend the recipe metaphor: where emotions are the finished dishes and affects are the ingredients, Izard might be seen as a meal planner who wants to understand the interrelation of appetizers, entrees and desserts. Tomkins could be imagined as the chef who has gone out to shop for the dinner, poring over the quality of each tomato, garlic bulb and leaf of basil that will make up the dish, then studying their combination and transformation in detail.
Izard and Tomkins hold different views of the biology of emotion
In fact, at the biological level, Izard is at odds with Tomkins, particularly as it relates to whether or not the biological emotional response is accompanied by expression. Izard writes, “In DET, an emotion is a particular set of neural processes instigating efferent processes that may or may not lead to an observable expression but that always lead to a unique conscious experience” (Izard, 2000). Izard suggests the neural processes underlying emotion will result in a conscious experience of emotion but not necessarily an emotional expression. Tomkins’s theory posits the reverse: the body, especially the skin, musculature, and autonomic system, always expresses the affective underpinning of emotion, which may or may not rise to the level of consciousness. Ekman’s well-supported work on micro-expressions is based on the assumption that affect is first and foremost a bodily response. See Ekman’s Emotions Revealed (2007) and Telling Lies (2009).
Izard and Tomkins name different emotions, and name them differently
Izard differs with Tomkins with regard to how emotions/affects are labeled and conceptualized, with the difference more pronounced in certain emotions than in others. For example, in Izard’s Discrete Emotions Theory, shame and guilt are identified as two distinct emotions, while Tomkins describes them as variations of experiences associated with the affect shame-humiliation. Tomkins uses a hyphenated format to refer to affects (enjoyment-joy, interest-excitement, distress-anguish, etc.) The hyphenation emphasizes that the qualitative aspect of an affect can occur at different intensities. While we may describe the extremes of these intensities as different emotions in everyday language, they are based on the same affective biological response occurring at different volumes (rage is anger with the volume turned up to 11). The hyphenated approach to describing affects captures the dynamic process of an affective experience and rejects the implications of the traditional noun-based nomenclature: that emotions are things that you can put on a stainless steel table and measure with a ruler. Emotions occur in time, as affects happen in the body and combine with memory and cognition. For Tomkins, guilt is a variant of shame-humiliation, though with a moral-cognitive valence around the bio-process (Exploring Affect, page 85), and so guilt is a socialized emotion rather than an affect.
Another example of differential labeling is Izard’s use of ‘sadness’ rather than ‘distress’—terms which are for Izard (1991) the same basic experience. Yet Tomkins did not equate sadness with distress. For Tomkins, sadness is closer to what Izard himself would call a ‘dimensional’ perception rather than a discrete affective experience. Dimensional measurements such as pleasant-unpleasant, positive-negative, and sadness-happiness are more easily quantified in a laboratory than are discrete affects. Sadness is arguably less discrete than distress-anguish. Sadness suggests a psychological malaise, or the ‘blues’, which might be a low-level, steady state distress combined with a cognitive appraisal that there is no effective remedy to the distressing situation. Sadness suggests a mood, or an emotion sustained across time. A baby who is a little cold, unpleasantly hungry, or out of proximity from a protective caregiver is distressed, maybe even anguished; sadness is not a correct description for such an early raw affective experience.
Here it is important to underscore that, unlike Izard, Tomkins was definitively not a proponent of discrete emotions. He proposed that there are discrete affects, occurring in the body on a continuum of intensity, that combine to make up an infinity of overlapping, nuanced and idiosyncratic emotions. In light of the excerpt on language quoted above, we might even imagine Tomkins cringing at any attempt to describe emotions as discrete. Significantly, Tomkins’s hyphenation of affect names, as in shame-humiliation or distress-anguish, allows for both discrete observations (qualitative) and dimensional observations (quantitative).
Izard identified ten ‘basic emotions’ (1977, 1979), for which he typically used a single word label, though he sometimes adopted the hyphen (in parentheses in the chart below) in the more detailed explications of specific chapters of Human Emotions (1977). Tomkins named nine basic elements of the affect system, among which he distinguished whether they are innately activated, stand-alone affects; drive auxiliaries (experiences associated with the hunger drive and co-opted to function in the affect system) such as disgust; or affect auxiliaries such as shame-humiliation, which is an affect that occurs only in the context of positive affect. Here is a list comparing the two.
|Izard’s Discrete Emotions (1977)||Tomkins’s Affect System|
|Innately activated affects|
|Distress (-anguish), Sadness (1991)||Distress-anguish|
|Contempt||Dissmell (replaced contempt in 1990)|
The reader who is not familiar with Tomkins’s concept of ‘dissmell’ may be puzzled by the word. Tomkins replaced his original concept of contempt with this highly specific term which refers to the body’s reaction to a foul odor and the desire to keep the source of the offending smell at a distance. Tomkins noted in 1990 that the same facial expression and bodily reactions come into play when we experience contempt, and therefore ‘dissmell’ is a more apt description of the embodied affect than the cognitively complex ‘contempt’. Dissmell is a drive auxiliary, a function of the hunger drive which has been recruited to function as an affect, imbuing an experience with a particular valance and meaning: “Keep that stinky thing away from me.”
Tomkins’s concepts of affects, drive auxiliaries, and affect auxiliary are unique. No theorist, including Izard, has embraced the idea of the ‘auxiliaries’ as a way to conceive of aspects of human affective experiences. Further, many citations of Tomkins that list his affects, or ‘core emotions’ as they are too often and incorrectly called, only cite eight of them because so few people find their way to the transcript of “Inverse Archeology” as contained in Exploring Affect, and so do not realize that dissmell replaced contempt. Initially, Tomkins conceived the singular drive auxiliary contempt-disgust (1961), before declaring that construct a “huge mistake” in 1990 (Exploring Affect page 394), thus separating and distinguishing disgust (gustatory drive auxiliary) from dissmell (olfactory drive auxiliary) from contempt, the “learned composite of anger and dissmell” (Exploring Affect page 394, emphasis added). For Tomkins, disgust and dissmell in their affect-mode guises are changed in status “from drive-reducing acts to acts that also have a more general motivating and signal function, both to the individual who emits it and to the one who sees it” (Exploring Affect page 84). In their affect-modes, disgust judges scenes as if they are bad food, and so are rejected, and dissmell signals object (people, things, ideas) rejection as if they smell foul and must therefore be kept at a distance. The concept of the drive auxiliary in an affect-mode is an exclusive Tomkins masterstroke.
The prodigious scope and import of Tomkins’s script theory is another important distinction between Izard and Tomkins. Late in his career, around the year 2000, Izard (with Ackerman) lays out seven basic assumptions in his approach to emotion research:
- Emotions are primary motivators;
- Discrete emotions function to organize perception, cognition, action, and thus shape personality and behaviour;
- “Personally significant situations typically activate a coherent pattern of interacting emotions” (2000);
- Emotion and behavioural relations “develop early and remain stable over time” (2000);
- Sustained patterns of emotion based motivation, organization, and behaviour shape personality;
- Individual differences in activation, frequency, and duration of emotions are major determinants of personality;
- Though emotions are adaptive by nature, maladaptive behaviours may nonetheless result.
None of these assumptions differs in any meaningful way from Tomkins’s original postulates. It should be noted that assumptions 3-7 echo Tomkins’s 1990 descriptions of scripts, the “sets of ordering rules for the interpretation, evaluation, prediction, production, or control of scenes” (Exploring Affect page 334). Yet Tomkins’s script theory moves well beyond Izard’s work in its focus on complex structures of personality, detailing the processes and products of cognition and information management in affect socialization and the ensuing scripts that control motivation and behavior. While Izard points to cognition and recognizes the thorough interdependence of emotion and cognition within personality formation, Tomkins, to a much greater degree, draws from disparate sources, synthesizes new insights, and pushes his contributions to understanding human beings well beyond Izard’s Differential Emotions Theory or DET.
What might have led to Izard’s departure from these critical elements of Tomkins’s theory of an affect system? One can only speculate, but it may have something to do with prioritizing the provable descriptive power of the theory over the explanatory power. Izard, like Ekman, started in the 1960s by researching the universality of emotional expression as evidenced by facial expression. In both cases, they focused their research on what they could test for and measure in the laboratory. If they couldn’t quantify an expression, then they de facto began to ignore it. That is why, for example, Ekman omitted shame-humiliation from his research. He couldn’t—or didn’t—know how to test for it. (In fact, Ekman says as much. See Emotions Revealed, page 235.) Izard tried to be a little more inclusive in his testing processes or objectives, but he was always constrained to a greater or lesser degree by experimentation. They had to be pragmatic. Tomkins had no such constraint; he was a ‘pure’ theorist, free to postulate a comprehensive system based on the science at the time and extraordinarily finely-tuned observational powers. Tomkins was well ahead of his time, and emotion researchers are still far from catching up.
Basch, M.F. “The Concept of Affect: A Re-examination,” Journal of the American Psychoanalytic Association. 24:759-777. 1976.
Ekman, Paul. Emotions Revealed. New York: St. Martin’s Griffin, 2007.
Ekman, Paul. Telling Lies: Clues to Deceit in the Marketplace, Politics, and Marriage (Revised Edition). WW Norton & Company, 2009.
Izard and Ackerman. “Motivational, Organizational, and Regulatory Functions of Discrete Emotions,” Handbook of Emotions, 2nd edition. Lewis and Haviland-Jones, Eds. New York: Guilford Press, 2000. 253-264.
Izard, C.E. “Emotions as Motivations: An Evolutionary-Developmental Perspective,” Nebraska Symposium on Motivation 1978: Human Emotion. Lincoln, NE: University of Nebraska Press, 1979. 163-198.
Izard, C.E. Patterns of Emotions. New York: Academic Press, 1972.
Izard, C.E. Human Emotions. New York: Plenum Press, 1977.
Nathanson, D.L. Shame and Pride: Affect, Sex, and the Birth of the Self. New York: W.W. Norton, 1992.
Tomkins, S.S. Exploring Affect. E.V. Demos, Ed. Cambridge, UK: Cambridge University Press, 1995.
Tomkins, S.S. Affect Imagery Consciousness. Vol. 1: The Positive Affects. New York: Springer, 1962.
Tomkins, S.S. Affect Imagery Consciousness. Vol. 2: The Negative Affects. New York: Springer, 1963.
Tomkins, S.S. Affect Imagery Consciousness. Vol. 3: The Negative Affects. New York: Springer, 1991.
Tomkins, S.S. Affect Imagery Consciousness. Vol. 4: Cognition. New York: Springer, 1992.