Gut Reactions: A Perceptual Theory of Emotion

by Jesse J. Prinz

Gut Reactions is an interdisciplinary defense of the claim that emotions are perceptions in a double sense. First of all, they are perceptions of changes in the body, but, through the body, they also allow us to literally perceive danger, loss, and other matters of concern. This proposal, which Prinz calls the embodied appraisal theory, reconciles the long standing debate between those who say emotions are cognitive and those who say they are noncognitive. The basic idea behind embodied appraisals is captured in the familiar notion of a "gut reaction," which has been overlooked by much emotion research. Prinz also addresses emotional valence, emotional consciousness, and the debate between evolutionary psychologists and social constructionists.

Genres: Philosophy, Psychology, Nonfiction, Emotion, Science

288 pages, Paperback

First published January 1, 2004

About the author

Jesse J. Prinz is a Distinguished Professor of Philosophy and director of the Committee for Interdisciplinary Science Studies at the City University of New York, Graduate Center. He lives in New York.

Reviews

[Alina · Rating 3 out of 5]

Prinz is a strange thinker. On the one hand, he is well-informed of the experimental psychological literature on emotion, and is good at presenting its findings that are most relevant to philosophy. But then, on the other hand, he gives absurd answers to his starting question, and his arguments are often bad. This made for an interesting reading experience; it was great to see this broad survey on psychology and philosophy on emotion, and how the two approaches may relate; but it was confusing to see such poor arguments and misguided conclusions. I'll summarize chapters 1-3, where Prinz gives the heart of his account. After that, I'll give my objections.

Prinz's main question, just like that of most philosophers of emotion, is: What is emotion? What does emotion essentially consist in, as opposed to what are the merely necessary causal preconditions for emotion to hold (e.g., feelings, beliefs, actions, intentions)? (I will return to below why I think this way of framing the question is misguided). Prinz's answer is that William James and Lange were right: Emotion consists in perceptions or conscious feelings of bodily changes that come with an emotional response to something. For example, fear consists of the feeling of one's heart racing, hands sweating, etc. Prinz adds his own twist to this view. The way by which emotion represents a bodily change can be called an "embodied appraisal," which he defines in chapter 3.

In chapter 1, Prinz surveys main approaches thinkers have taken towards understanding emotion. Here's a sampling. Darwin pointed out that bodily changes that come with emotion prepare an animal for behavioral responses that are conducive for the animal's satisfying its needs and interests, in light of an environmental event. This paved the path towards certain thinkers who prioritize action and action-disposition as the mental type with which emotion should be identified (e.g., Gilbert Ryle)—call this a somatic theory. In psychology and the cognitive sciences, the mind is understood as composed of various functions (e.g., memory, categorization systems, attention, reasoning)—call this 'encompassing theories'. Emotion can be understood as consisting in systematic changes across these functions. On another approach, cognitive theories of emotion prioritize the role of thinking in emotion. Emotion might be understood as identical to thought, or thought is essential to emotion but requires more ingredients.

In chapter 2, Prinz focuses on the appraisal literature. Today, the most popular approach to emotion in psychology may be called "dimensional appraisal theories," as a variety of cognitivist theories. On these, emotion is understood as tightly related to, if not constituted by, appraisal, which is a sort of cognitive event that involves valuing an object in light of its properties that are relevant to one's interests and needs. He opposes theories like the cognitive labeling theory on which one's labeling one's emotion state is constitutive of the emotion itself. He opposes views on which emotion should be seen as cognitive because we assess emotions for rationality or justifiability. But most of his arguments are pretty bad. For example, he attempts to refute conclusions from empirical studies that go against his thesis by attacking features of methodology; but these very same features hold of the studies that he cites in support for his thesis. Also, he says that emotions can't be cognitive because we don't have control over them as we do with our judgments. That's a silly point; we do have some control over our emotions, even if it differs from that of judgment, and why think that the exact sort of control we find in the case of judgment is necessary to a state's being cognitive? Prinz later in the chapter raises the question of what it means for something to be cognitive, and his answer is a strange one: a state must be under our "direct control" in order to be cognitive. But he does not satisfactorily spell out what it means for us to be in control over a state, and what it means for this control to be direct, rather than mediated. I also really didn't like how Prinz dismisses many proposals for defining cognition on the basis of these conflict with what he deems is folk intuition. But shouldn’t cognition, at this preliminary stage methodologically, be regarded as a family resemblances concept for the time being, and be open to revision? Prinz no where defends that cognition is a natural kind. This is another example of his sloppy reasoning.

In chapter 3, Prinz explores what it means for emotion to consist in a perception of bodily change, which is also identical to what he calls an "embodied appraisal." An appraisal is minimally a representational state. Prinz takes up Dretske's framework for making sense of mental representation. Such states must carry information and can be erroneous. For example, a frog may have a perceptual representation of a fly if this mental state delivers it the information that there is a fly, and this information is correct (there is no misfiring of the perceptual detector that's responsible for representing flies). Prinz argues that emotions represent bodily changes (on the basis of what seems like mere stipulation that we've evolved to have perceptual detectors whose natural function is to represent such changes.) We should distinguish between the causal factors that activate a perceptual detector, on the one hand, and the things that are represented as a result, on the other hand. Many philosophers think that the formal or intentional object of emotion is a relational property, or something that an object has in relation to the subject's interests (e.g., the object of sadness is loss.) Prinz holds that emotion, as an embodied appraisal, involves both perception of bodily changes and non-cognitive apprehension of the relevant relational property; he thinks that the two are as tightly related as the perception of furriness and tail wagging is to the non-cognitive apprehension of the creature standing before oneself is a dog. But then Prinz goes on to say that emotions are fundamentally about appraising bodily changes that you’ve perceived.

Here are my objections (in no particular order). His conclusion left me confused: what does he mean by what emotion is "about"? Rarely do our emotions seem to be about our bodily changes. I am sad not because I feel tears running down my cheek, but rather because my hamster has died; and it is the death of my hamster that my emotion is about. Maybe I was a sloppy reader, but it could also be that Prinz is not clear about his position. Moreover, if Prinz does want to commit to that the non-cognitive apprehension of the relevant relational property (e.g., loss of my hamster) is constitutive of emotion, he does not address why this is non-cognitive rather than cognitive, and does not advance the literature with respect to making sense of what this cognitive-esque component consists in. How is it different from the propositional content of judgments? (I've been wondering about the possibility that the cognitive content of emotion is non-propositional, but nevertheless "meaningful," and have been trying to make sense of the nature or structure of that meaning, and how it psychologically relates to propositional meaning).

Second, unfortunately, it seems that Prinz overhastily assumes that various appraisal theories construe "appraisal" in more or less the same way; it ultimately consists in a judgment, as a basic sort of cognitive act. While certain psychologists do understand appraisal in this way, some do not. Frijda and Scherer, for example, define appraisal in relation to the various functions that are operative in the emotion processes that constitute an emotion episode, so any sort of mental state can amount to an appraisal during an emotion episode, as long as it performs this role. This misunderstanding is significant. If appraisal is understood as a particular sort of mental state, in order to explain its various forms found across different emotion experiences, one would have to say that it consists in many different sorts of mental states. Prinz believes that this is the case, and he chides appraisal approaches for being inelegant, for thinking that emotion involves many different components, without explaining how they hand together. But if appraisal is understood in in this functionalist way, this difficulty is eliminated; there is a unified sort of mental state that constitutes emotion at any particular emotion episode, but this unified sort of state may vary between episodes, and they count as emotion insofar as they all fit an overall functionally-defined template.

A sad feature of Prinz's analysis is that he chides these psychologists for being misguided and failing to define terms they use, like cognition and appraisal. But there is beauty in this approach; one can just have the term implicitly defined in light of the functions one has given to emotion, and this can lead to new pictures of the various mental functions and how they work together, which can then improve how one thinks about these folk psychological concepts. Prinz’s approach seems to force us to bring forth folk intuitions as the gold standard, for making sense of and assessing proposals; the psychologist’s way allows us to potentially modify and at least add informative detail to these folk concepts.

Third, I found Prinz's bringing up Dretske's analysis of mental representation really interesting. What does it mean for there to be error if a state is indeterminate regarding its information, and the sort of information contained rather serves to sets constraints for later more determinate informational states? In this case, it would seem to be only assessable with respect to appropriateness or helpfulness in light of the subject's goals (either epistemic or practical). The relevant question would be: will this state help me get to avoiding error and achieving my goals? I've been playing with the idea that emotion is a process, and always at its early stages, and often over its entire duration, information is kept in an indeterminate state as such. In this case, emotion would not be representational, on Dretske's definition. Emotion is not assessable for correctness, but only only for conduciveness to avoiding errors later down the road (as long as the emotion does not violate only basic physical or low-level facts, like whether a certain object exists or not).

Fourth, one of Prinz's main arguments for his thesis is based in the experimental literature on fear circuits (cf. LeDeoux). There are two fear circuits found across many mammalian brains; one does not require processing in cortical areas, and the other does not. This indicates that it's possible for there to be fear caused by visual stimuli alone, without conscious apprehension of what's going on. Prinz thinks that this means that emotion is fundamentally noncognitive. But that does not follow. Such studies establish only that certain autonomic nervous processes can be triggered directly by visual stimuli. This leaves it open that once an animal starts sweating or having a pounding heart, this will force it to look at its environment, to search for a threat. If there is a threat, the animal will enter fear, and if not, it will calm down.

All of this leads to my main beef with this book.

Fifth, Prinz, among most philosophers of emotion, assume that emotion must be identical with a certain type of mental state, which is dissociable with other types which are merely the necessary causal preconditions or consequences of emotion. But why think that? Why not think that it is somewhat arbitrary where to draw the line, demarcating the sorts of activity that are part of the emotion, from those which are not part but are surrounding the emotion? Prinz uses the analogy to sunburn. Sunburn consists in skin damaged in a certain way, and its necessary causal precondition is UV rays. The rays themselves don't constitute the sunburn. But the way this analogy should apply to emotion is not straightforward. Sunburn is a purely physical object, persisting over time. We can ask about its material makeup. In contrast, emotion is a psychological thing, which morphs rapidly over time (e.g., in being anxious, one may start off vaguely fretful, and then with escalating thoughts, it spins out into intense fear). The issue of how to carve up "parts" of a psychological thing is significantly different from that regarding the "parts" of a material thing. This seems to indicate that there is no line between the parts v. surrounding conditions of emotion (on the condition that all candidates are equally necessary for emotion); perhaps there are practical and theoretical considerations that make it more useful to prioritize certain conditions over others, but in reality, all are equally "constitutive."

So once we reject the very question that Prinz starts off with, we can see that perception of bodily changes is a component of emotion, but is no more necessary to emotion than conscious apprehension and evaluation of an object in the world is necessary to emotion. Many different types of mental activity is necessary and constitutive of emotion, and these activities are interdependent even. I do appreciate, however, that Prinz raises the question of what "binds" together the diverse elements of emotion, if such diverse elements do exist. This is perhaps a worthwhile question. My thought at this time is that our conscious awareness of these elements lets them interact (e.g., perception of bodily change, thought of what's happening in the world), and the bare fact of our being psycho-material unities, persisting over time, also seems to satisfy the explanatory demand.

[Juan · Rating 4 out of 5]

Me da risa que uno de los filósofos afectivos más importantes sea un twink peliazul.

[Adam R · Rating 3 out of 5]

The book offers a nice interdisciplinary overview of theories of emotions. Prinz argues for an account he dubs an 'embodied appraisal account of emotions' (EA). On EA, emotions are perceptions of bodily states that represent core relational themes and are essentially valient. Core relational themes are, roughly, appraisals of your relation to the surrounding environment. For instance, the emotion of fear registers things like increased heart rate, quickened breathing, and muscle tenseness and represents the core relational theme DANGER. On Prinz's view, emotional valence is basically a reinforcement mechanism and is either positive (b/c it reinforces that you do some behavior) or negative (b/c it reinforces that you not do some behavior). Also, on his view, emotions can be either conscious or unconscious and form a natural kind. Thus, he rejects what he thinks is A. Rorty's disunity thesis. The primary motivation for his view, I take it, is that it allows us to take all of the good stuff from theories of emotions on which emotions are purely bodily sensations that do not represent and all the good stuff from theories on which emotions do represent without all the extra baggage that many take to be problematic. Generally, I like the book. Prinz provides a very lucid commentary of very difficult issues in philosophy of mind, psychology, cognitive science, and neuroscience. Still, my colleagues and I found many of the central arguments unconvincing, but interesting nonetheless. There are also an unbelievable number of typos. I recommend reading the book for anyone interested in theories of emotion. While we found the arguments to be lacking, the book generated excellent discussion, which I attribute primarily to Prinz's clarity of presentation and ability to put forward profound hypotheses in a clear and concise manner.

[Matthew A LaPine · Rating 5 out of 5]

Great book. This is exactly the kind of interdisciplinary work needed for understanding. Will be challenged and perhaps corrected in places (esp. as neuroscience progresses), no doubt, but a very needed contribution.

[LemontreeLime · Rating 3 out of 5]

Read part of this for a paper I had to write, and really interested in reading the rest of it eventually!