The Odd Couple The Compatibility of Social Construction

The Odd Couple: The Compatibility of Social Construction & Evolutionary Psychology Ron Mallon University of Utah & University of Hong Kong & Stephen Stich Rutgers University

The BIG Picture: What I hope to show in this talk • Evolutionary Psychologists (EPs) & Social Constructionists (SCs) seem to have deep empirical disagreements about – the extent to which normal humans share innate mental mechanisms which • were shaped by natural selection • strongly constrain our psychology, our social interactions & our institutions

• While there may indeed be some important disagreements between SCs & EPs on these points, there is another, less obvious issue dividing them – A philosophical disagreement about the meaning and reference of ordinary words for mental states and social phenomena – words like anger, disgust, gender & homosexuality

• When this philosophical dispute is made explicit, it becomes clear that nothing much turns on it, and it can be easily set aside • When the philosophical dispute has been set aside, the empirical disputes between SCs & EPs look much less serious – Rather than being adversaries, they look more like natural partners

Overview of the Talk I. Getting clearer on terminology (or why ‘social constructionism’ isn’t always a dirty word) II. A quick sketch of some SC work on the emotions III. A quick sketch of an emerging EP consensus about the emotions IV. What’s left to fight about? Are emotions universal or culturally local?

V. The philosophy that underlies this dispute: The “thick description” theory of the meaning & reference of emotion terms i. ii. What it claims How it underlies the debate over the universality vs. cultural locality debate VI. How to sidestep the problem posed by the thick description theory and make it clear that SCs and EPs need not be adversaries, since their theories are compatible & complement each other

I. Getting clearer on terminology (or why ‘social constructionism’ isn’t always a dirty word) The terms Social Constructionism & Evolutionary Psychology are both used for a variety of different views Social Constructionism is sometimes used as a label for a (quite nutty) metaphysical view that denies the mind-independence of all reality These radical social constructionists think that everything is “socially constructed” including atoms, galaxies and dinosaurs

• I’ll be focusing on a more modest and sane (and prima facie plausible) version of SC which holds that – important features of human psychology & social life are • culturally caused and • local in character • SCs who fit this description need not endorse every aspect of what Tooby & Cosmides call the Standard Social Science Model

• Evolutionary Psychology is sometimes used for a specific cluster of views associated with Cosmides, Tooby & Pinker.

• Evolutionary Psychology is sometimes used for a specific cluster of views associated with Cosmides, Tooby & Pinker. – We think of this as High Church Evolutionary Psychology

• Evolutionary Psychology is sometimes used for a specific cluster of views associated with Cosmides, Tooby & Pinker. – We think of this as High Church Evolutionary Psychology • Though we don’t propose to offer a definition, our view of EP is decidedly Low Church – and more inclusive.

II. The Social Constructionist Approach to the Emotions A primary concern of SCs concerned with the emotions is to describe the rich multifaceted, culturally local network in which the emotions are embedded. Since understanding what informants say is a matter of great importance, SCs pay careful attention to a number of aspects of emotion discourse & behavior in the target culture, including:

– the (often complex) circumstances under which people claim they or others experience the emotions picked out by various emotion words – the pattern of inferences drawn when someone is believed to be experiencing the emotion – the pattern of interactions that exist (or that people believe to exist) among emotions and other mental states – the ways in which both emotions & discourse about emotions interact with the moral, political & economic lives of people in that culture

When done well, the resulting “ethnopsychological” accounts result in fascinating “thick descriptions” (Geertz) of patterns of interaction that differ in important ways from the patterns in which our own emotions and emotion discourse is embedded.

An example: Catherine Lutz’s study of the emotions of the Ifaluk

• song is an Ifaluk emotion akin to (what we call) anger – it has a strong moral component • to count as feeling song an Ifaluk must be justifiably angry at another person who has engaged in morally inappropriate behavior – thus two people can’t be song at each other – there are many other Ifaluk words for emotions akin to anger that do not involve this moral dimension – there is no generic Ifaluk term that picks out all these sorts of anger – one sort of behavior that can provoke song is the violation of a taboo (e. g. working in the taro garden while menstruating) – another is ker, sort of excited happiness which can produce inappropriately loud talk or “showing off”

– song does not lead to physical violence – it does lead to • • refusal to eat or speak with the offender gossiping about the offender threats of fasting or even suicide threats to burn down the offender’s house – when a person recognizes that someone’s is song is directed at them, they typically experience an emotion the Ifaluk call metagu, a sort of fear or anxiety that leads to • calmer, more appropriate behavior • corrective action – paying a fine – apologizing – sending something of value to the aggrieved parties

Person 1 Person 2 Ker Misbehavior [happiness/ excitement] Misbehavior [fear/anxiety] Metagu Good, calm behavior Song [justifiable anger] Song

III. Evolutionary Psychology: An Emerging Consensus on the Emotions Ekman: To explain his well known finding that facial expressions of some emotions are crosscultural universals, Ekman posited that these emotions are subserved by affect programs – – – largely automated or involuntary suites of coordinated emotional responses subserved by innate mental mechanisms that are the product of natural selection present in all normal members of the species

– downstream of the affect program, the behavior that an emotion produces is strongly influenced by culture • display rules • posture, tone of voice • self reports • more complex patterns of cognitive, behavioral & social activity

– upstream • Ekman posited an “appraisal mechanism” which monitors stimuli & triggers the appropriate affect program • by the mid-1990 s Ekman had come to think that just about all the activity of the appraisal mechanism is affected by culturally local factors • Lazarus proposes that the emotion triggering mechanism includes an innate set of abstractly characterized conditions (“core relational themes”) – Anger: A demeaning offense against me & mine – Fear: An immediate, concrete & overwhelming danger – Sadness: Having experienced an irrevocable loss • determining when these conditions have been satisfied requires culturally local beliefs & information about culturally local norms, goals & values

Robert Levenson’s Ekman-Inspired Bio-Cultural Model Recruited Response Tendencies Antecedent Conditions Measurable responses Subjective Experience Facial Program Interpersonal Emotion Prototype Self-Report Facial Expression Voice Tone Vocalization Program Intrapersonal Motor Behavior Motor Program Physiologic Support Appraisal System Physiologic response Cultural learning: Display and feeling rules

Robert Levenson’s Ekman-Inspired Bio-Cultural Model Recruited Response Tendencies Antecedent Conditions Measurable responses Subjective Experience Facial Program Interpersonal Emotion Prototype Self-Report Facial Expression Voice Tone Vocalization Program Intrapersonal Motor Behavior Motor Program Physiologic Support Appraisal System Affect Program Physiologic response Cultural learning: Display and feeling rules

Norms Values Goals Beliefs Subjective Experience Facial Program Interpersonal Emotion Prototype Self-Report Facial Expression Vocalization Program Voice Tone Intrapersonal Motor Program Physiologi c Support Appraisal System Motor Behavior Physiologic response Cultural learning: Display and feeling rules

Norms Values Goals Beliefs Downstream Cognitive & Social Activity Subjective Experience Facial Program Interpersonal Emotion Prototype Self-Report Facial Expression Vocalization Program Voice Tone Intrapersonal Motor Program Physiologi c Support Appraisal System Motor Behavior Physiologic response Cultural learning: Display and feeling rules

IV. What’s Left to Fight About? Are emotions universal or culturally local? Far from being incompatible with each other, it looks like the evolutionary psychology and social constructionist approaches to the emotions are complementary – ethnopsychologies like the one that Lutz provides provide details about culturally local aspects of belief, values, etc. – Cognitive models like Levinson’s give an account of the psychological mechanisms underlying emotions & explain how innate, evolved mechanisms interact with culturally local beliefs & values.

Norms Values Goals Beliefs Downstream Cognitive & Social Activity Subjective Experience Facial Program Interpersonal Emotion Prototype Self-Report Facial Expression Vocalization Program Voice Tone Intrapersonal Motor Program Physiologi c Support Appraisal System Motor Behavior Physiologic response Cultural learning: Display and feeling rules

So what’s left to fight about? • Evolutionary psychologists maintain that central parts of the emotion system are innate and present in all normal humans, and when they are triggered they produce emotions (like fear, anger & sadness) which are cultural universals. – Since these mechanisms are homologous to mechanisms in other species, many non-human animals also experience emotions.

• Social constructionists insist that emotions are a culturally local phenomenon and thus that people very different cultures have very different emotions. – – – song & metagu are Ifaluk emotions which outsiders do not experience amae (a pleasant sense of helplessness and desire to be loved) is Japanese emotion which Westerners do not experience (Harré) accidie (boredom or disgust with fulfilling one’s religious duty) is an emotion that was once common in the West but has now disappeared.

– – – anger is unknown among the Inuit (Briggs); “anger, as a specific emotion, is not universal across cultures (Averill) sadness does not exist among Tahitians (Levy) there are no universal emotions and there may well be some cultures in which there are no emotions at all (Shweder)

V. What’s Going On Here? The philosophical issue that underlies the dispute Far from being incompatible with each other, it looks like the evolutionary psychology and social constructionist approaches to the emotions are complementary – ethnopsychologies like the one that Lutz provides provide details about culturally local aspects of belief, values, etc. – Cognitive models like Levinson’s give an account of the psychological mechanisms underlying emotions & explain how innate, evolved mechanisms interact with culturally local beliefs & values.

So why do they disagree about the universality or cultural locality of emotions?

So why do they disagree about the universality or cultural locality of emotions? Blame Canada Philosophy

• The most widely discussed philosophical account of the meaning & reference of terms for the emotions & other mental or psychological states is due to David Lewis David K. Lewis 1941 -2001

– jargon: • • • meaning is (roughly) what a person must know to understand the term reference is what the term picks out meaning ≠ reference: ‘the morning star’ & ‘the evening star’ have the same reference, but different meaning

– some central ideas of Lewis’ theory • • ordinary language “mental state” terms can be treated as theoretical terms are implicitly defined by theory in which they are embedded – the definition is just a long description that includes everything theory claims about the thing the term refers to • so I will call this sort of theory about the meaning of a theoretical term a description theory – these implicit definitions are holistic because (i) theory implicitly defines all of its theoretical terms, and (ii) the entire theory contributes to the meaning of every theoretical term

• theory which implicitly defines mental state terms is commonsense psychology = “our extensive, shared understanding of how we work mentally” which is “common knowledge among us” (Lewis) • mental state terms express “package deal concepts” since commonsense psychology provides an implicit definition for all of them • if Lewis’ theory is correct then a culture’s folk psychological theory implicitly defines the emotion words they use

– Ethnopsychologies like the one provided by Lutz are intended inter alia to provide an account of the culture’s folk psychology (their shared understanding of how they work mentally) • Question: How much of the detail in a rich ethnopsychology contributes to the meaning of ordinary mental state terms? austere description theories opulent description theories

– What about reference? • since the meaning of a “theoretical” term is given by a long description, the natural idea (& the one urged by Lewis) is that the term refers to whatever fits the description • but we can’t require the fit to be perfect, since if we did, then any small error in theory would make all theoretical terms refer to nothing • Lewis suggests that a term refers to whatever more or less fits the description – but that is intentionally vague

high accuracy low accuracy at one end of the spectrum are high accuracy accounts of reference that require that most descriptions fit; at the other end are low accuracy accounts that allow a referent to fit many fewer descriptions

high accuracy low accuracy austere opulent

high accuracy low accuracy thick description austere opulent a thick description account of the meaning & reference of ordinary mental state terms is opulent & high accuracy

• But what does all this have to do with the dispute about whether emotions are universal or culturally local? – Assume that thick description account is correct

• Among the Ifaluk it is common knowledge (i. e. part of their folk psychological theory) that – – – if a woman goes into the taro garden when she is menstruating, it will provoke song if a man goes into the birthing house, it will provoke song if you are really song at me, I can’t be song at you when a person realizes that someone is song at him, he typically experiences metagu etc. , …. • There is no emotion, among us, that fits all of these descriptions • So there is no emotion in our culture that counts as an instance of song; song does not exist here.

• • Much the same argument can be run in the opposite direction Among us, it is common knowledge (i. e. part of our folk psychological theory) that – – – if someone shouts racial epithets it is likely to provoke anger if some gives someone else “the finger” it is likely to provoke anger if you are really angry at me, I can be angry at you too when one person provokes anger in another person, it will often lead to a heated exchange of words and will occasionally lead to physical confrontation & violence etc. , …. • There is no emotion among the Ifaluk that fits all of these descriptions • So there is no emotion among the Ifaluk that counts as an instance of anger; anger does not exist among the Ifaluk.

• N. B. These arguments make no assumptions at all about the psychological mechanisms underlying the emotions. Thus they are compatible with any account of those mechanisms, including the account favored by evolutionary psychologists. All that the arguments require are • – – the thick description account of meaning & reference the fact that there are substantial differences between our folk psychology (= what we commonly believed about mental states) and Ifaluk folk psychology (= what they commonly believe)

• Is this the argument that leads Social Constructionists to the conclusion that emotions are culturally local? • Though they don’t use the philosophical jargon (or take account of important philosophical distinctions like meaning vs. reference) a good case can be made that they do indeed have something like this argument in mind.

“Across languages, the range of implications, suggestions, and connotations of psychological state terms do not easily map, at least not lexically; and to adequately understand the meaning of the terms in either language is to understand a good deal about different local systems of values and particular ways of life. (Shweder, 1994) “Emotion words are treated here as coalescences of complex ethnotheoretical ideas about the nature of self and social interaction. … To understand the meaning of an emotion word is to be able to envisage (and perhaps to find oneself able to participate in) a complicated scene with actors, actions, interpersonal relationships in a particular state of repair, moral points of view, facial expressions, personal and social goals, and sequences of events. ” (Lutz, 1988)

• Obviously, Evolutionary Psychologists and others who reject the view that emotions are culturally local must reject the thick description theory of meaning & reference. • How much of a problem is this?

high accuracy thick description low accuracy Producing an alternative to the thick description theory is no problem at all. There are lots of alternatives on the market.

In addition to all these, philosophers like Putnam and Kripke have offered alternative “causal/historical” theories of reference according to which the reference of a term fixed by the circumstances in which it was introduced and its transmission history in the language. If any of these alternative accounts of meaning & reference is correct, then the argument for cultural locality collapses.

• So the situation is this: – Those who think that emotions are culturally local are (tacitly or explicitly) adopting a thick description account of the reference of emotion terms. – Those who think that emotions are universal are (tacitly or explicitly) adopting some other account of the reference of emotion terms.

VI. Who’s Right? And Why It Doesn’t Matter To settle the universal vs. culturally local dispute, it looks like we have to determine which account of reference is correct. And that might sound like bad news because philosophers and linguists aren’t even close to figuring out which account of reference is correct. Even worse,

In Deconstructing the Mind, I argued that the issue may well be moot, since neither philosophers nor linguists have given us a coherent account of what a theory of reference is supposed to do, thus we don’t really have a clear idea of what it would be to get a theory of reference right.

• But actually this is not such bad news since, for two rather different reasons, it doesn’t much matter which side is right. – Reason 1: The dispute is almost entirely isolated from the empirical & theoretical claims that SCs & EPs want to make.

• A SC who thinks that emotions are culturally local can accept everything in the Levinsonstyle EP model and everything an EPist wants to say about the evolution of the components of this model.

Norms Values Goals Beliefs Downstream Cognitive & Social Activity Subjective Experience Facial Program Interpersonal Emotion Prototype Self-Report Facial Expression Vocalization Program Voice Tone Intrapersonal Motor Program Physiologi c Support Appraisal System Motor Behavior Physiologic response Cultural learning: Display and feeling rules

• A SC who thinks that emotions are culturally local can accept everything in the Levinsonstyle EP model and everything an EPist wants to say about the evolution of the components of this model. • All the SC need insist on is that many beliefs, norms, goals & values are culturally local. And this is not a claim that (sensible) EPists are in the least inclined to dispute.

• Reason 2: – No matter who is right about the meaning & reference of ordinary language emotion terms, each side can easily say what it wants to say with the help of a bit of technical language. – To see the point, let’s once again assume thick description account is correct.

• The EPs must concede that anger (the emotion picked out by the ordinary English word) is not universal. • But they can still maintain that there is a family of emotions in different cultures, all subserved by the same innate affect program & “emotion prototype” or “core relational theme” that subserve anger in our culture.

Norms Values Goals Beliefs Downstream Cognitive & Social Activity Subjective Experience Facial Program Interpersonal Anger Emotion Prototype Self-Report Facial Expression Vocalization Program Voice Tone Intrapersonal Motor Program Physiologi c Support Appraisal System Motor Behavior Physiologic response Cultural learning: Display and feeling rules

• We can then simply introduce a technical term, core anger, to refer to all of these emotions.

Norms Values Goals Beliefs Downstream Cognitive & Social Activity Subjective Experience Facial Program Interpersonal Anger Emotion Prototype Self-Report Facial Expression Vocalization Program Voice Tone Intrapersonal Motor Program Physiologi c Support core anger Appraisal System Motor Behavior Physiologic response Cultural learning: Display and feeling rules

• EPs who have conceded that anger is not universal can claim, instead, that core anger is universal. • And this, surely, captures what they really want to claim.

Conclusions The most conspicuous disagreement between SCs & EPs who study the emotions is not an empirical disagreement about minds, cultures or evolution. Instead, it disagreement about the universality or cultural locality of the emotions.

Conclusions That dispute is rooted in a philosophical dispute about the meaning & reference of ordinary language emotion terms. And nothing much depends on that philosophical disagreement. It does not matter who is right.

Conclusions Once the philosophical dispute is set aside, it should be easier for Social Constructionists and Evolutionary Psychologists to stop seeing each other as enemies and start seeing each other as natural allies in the attempt to understand the emotions.

Blessed are the peacemakers; for they shall be called the children of God Matthew 5: 9