Philosophical Reflection on Emotions and Shame

Explanations of emotion fall in to one of two philosophical camps. Those whose work falls within the paradigm initiated by the 19th-century philosopher, psychologist and physician William James, are referred to as Jamesians.ⁱⁱ In short, you are a Jamesian if you argue that emotions are essentially physiological responses to causal stimuli, and the awareness of these responses. Emotion as affect.

In opposition, one finds the cognitivists, a position in the philosophy of emotions traceable back to the Stoic movement in Ancient Greece, but with contemporary adherents in contemporary cognitive psychology, ecological cognition and postanalytic philosophy. You are a cognitivist if you argue that emotions are essentially thoughts about, or directed towards, something in the world. Emotion as (embodied) thought.ⁱⁱⁱ

One could advance one of many psychological or sociological theories of emotion without referring to either of these two underlying philosophical positions, but in doing so, in advancing your theory, you will commit to one of them, and this commitment would then constrain you. One is so constrained because whether acknowledged or not, these philosophical commitments will be operative in one’s theorising.

This philosophical dispute has consequences for addressing, or treating emotions, where they present difficulties. If we conceive of emotions as essentially physiological changes triggered by causal stimuli in the environment, our focus will be on understanding the causal mechanisms, and perhaps preventing the cause. On this view, the person experiencing the emotion is subject to that emotion, the emotions can be depicted as ‘passions’ and a person who is in an emotional state is passive.

Depicting the emotions as passions leads us to depict the person having the emotion as akin to a stimulus-response processor, responding to certain causal inputs with certain outputs (emotional responses). While this has the virtue of being a naturalistic depiction of emotion, it is somewhat at odds with the complexity of human psychology and mind–world relations; it also faces the somewhat difficult problem of accounting for the meaning emotions have for those experiencing them. On this view, we understand human emotional responses on the model of an artefact such as a smoke alarm responding to smoke in the environment.

If one is a cognitivist, the picture is different, a little more complex, and, we propose, more appropriate to our nature. Here, our emotions are meaningful responses to a world which we experience as having meaning for us, what in some philosophical contexts is referred to as the lifeworld. On this conception, emotions are no longer appropriately depicted as passions and the bearer of emotions is no longer necessarily passive but, rather, has scope for some emotional agency, for it is at least logically (or grammatically) possible to achieve control over one’s emotions, if not necessarily possible in practice, on specific occasions.

Within the cognitivist approach to explaining emotion there are a number of different sets of commitments one might have, which relate to the nature of the cognitive constituents one employs in the explanation. So, one might be a judgementalist (eg, Solomon2), where one depicts emotions as constituted and type individuated by judgements. Here one is also, therefore, committed to a representationalist picture of cognition, where the constituents of cognition are mental representations of the world, with propositional form. On this view, the emotion is a propositional attitude. So, where judgementalists identify the constituents of emotion as judgements, other philosophers identify the cognitive constituents as ‘thoughts’ (Nussbaum3), ‘evaluative beliefs’ (Taylor4) or ‘construals’ (Roberts5); we can, therefore, refer to this group of emotion theorists as ‘propositional attitude cognitivists’ (Pro-Cogs), for they are all committed to the claim that the constituents of emotion are propositional attitudes. Pro-Cogs conceive the person as a belief (or judgement) machine, akin to software programmes running on hardware components, which output contentful information in response to certain patterned inputs.

In contrast, and at the other end of the cognitivist spectrum, we find those who forgo or reject representationalism, and depict emotions as a specific class of affordances or evaluative perceptions (eg, Hutchinson6). We call this version of cognitivism ecological cognition (Eco-Cog), and the explanation becomes less theoretical and abstract and instead focused on persons who have gone through a process of enculturation, who inhabit a world which has meaning and contains loci of significance for them, given their enculturation (Bildung/Second Nature) and their particular interests.^iv While this framework forgoes brevity and neatness, perhaps, we gain in understanding and explanatory power, for it is sensitive to the way in which the world has meaning, and the specific meaning it has for this person might be in response to that person’s enculturation and interests.

We here want to pick out two gains we believe arise from taking up the Eco-Cog perspective on emotion.^v

1. An emotion is not conceived of as an affect (Jamesianism), nor is conceived of as constituted by beliefs (Pro-Cog). Rather, we conceive of emotion as a dynamic relation between, on the one hand, a person, who in addition to their biological constitution is endowed with a specific culture, and on the other hand, the lifeworld. This means that theoretical solutions will not cut it, because the generality of the theory will mean it is unable to capture the dynamism and occasion, context and cultural specificity of the relationship. A theory of shame will fail to adequately capture the nature of the emotion. Rather, we must engage with the individuals experiencing the emotion and work to understand the way in which they relate to the world and how that brings them to the emotional state.

2. The approach we recommend overcomes the ‘problem of emotional recalcitrance’ (see D’Arms and Jacobson7 and Griffiths8), oft-cited as a reason to forgo Pro-Cog in favour of Jamesianism: an example of emotional recalcitrance being fear of flying, experienced while concurrently believing (correctly) that flying is the safest mode of transport. On the Eco-Cog approach, belief is not playing the explanatory role which it plays in Pro-Cog, so, we can retain the cognitivist insight without facing the problem of recalcitrance. This pays dividends when we turn our attention to shame, for it is a characteristic feature of shame that it is experienced in the absence of a set of beliefs that would serve to explain the shame. Indeed, often it is the converse: the beliefs held tell the person experiencing shame that they have no grounds for their shame, yet the shame persists.

Emotions and Their Objects: The World-Directed Nature of Emotion

Cognitivist explanations of emotion depict emotions as object oriented: emotions have objects, and the conceptual analysis of emotion, by which we establish their identity criteria, includes these objects. On this account, shame’s object is status diminishment, of one’s self (one’s being), as an individual or as a member of a group. To be stigmatised, as an individual or member of a stigmatised group, is to be subject to such a diminishment in status. Shame is the emotional manifestation of such stigma. In philosophical terms, shame and stigma are internally related, for they are not related as two separate entities but are internal to the meaning and identity of each other.

Let us take some examples from the writings of Primo Levi, and his discussions of shame in If this is a Man 9 and later in The Drowned and the Saved.10 We do so because, as we noted above, our analysis should be focused on people expressing and experiencing the emotions we are studying.

Primo Levi’s writings testify to a number of sources of status diminishment that give rise to shame:

• Humanity was diminished because the crimes of Auschwitz had taken place: so humanity bore the stigmata engendered by the crimes of the Holocaust (the bureaucratised and industrialised dehumanisation, mass murder and genocide). Levi’s shame testifies to this stigma.

• In addition, Levi felt diminished owing to the accident of survival. He experiences shame because survival was not connected to his worth as a person, because death and survival in Auschwitz were radically decoupled from one’s goodness or badness, or any discernible rationale, becoming a matter of chance, decided by the whim of a psychopath or perverse bureaucratic system. This diminishes the status of the individual, post-Auschwitz. It seems that for Levi his survival felt unearned, and while others died, this amounts to a diminishment.

• There also seems to be a diminishment which comes from a mismatch between how one acted in Auschwitz—even allowing for judging these as irreproachable in the context of the camp—and the standards one sets for oneself, or that are reasonably expected of one by one’s honour group, in the pre and post-Auschwitz context. Here, what is stigmatised is the person acting as they did in the camp, viewed from the perspective of the postcamp culture, and the recognition that the viewer and viewed are the same person (or group).

There are a number of things that we can learn from these examples that are central to philosophical discussions of shame and that, we will argue, are relevant to HIV-related shame and stigma. We here want to focus on stigma-as-object-of-shame.

Emotions and Objects

Our first observation is that while emotions have objects, these need not be concrete, but can be diffuse and abstract. The object of fear might well be the bear standing before us on the mountain path, though even here, it is our taking, our evaluative perception, of the bear-as-a-threat-to-our-continued-wellbeing and the conceptual resources that draws upon, that constitute it as an object of fear. However, this said, the object of our fear in this example is there before us in very real and literal sense. In contrast, we might also experience fear of failure, and here the object of fear is the idea of our failing to achieve that which is either expected of us by our peers or by ourselves. While the bear has a kind of concreteness and is directly perceived in the act of fearing, the future failure exists only as a possible future outcome, and as such is neither concrete nor directly perceptible.

Returning to shame; as we saw in the first example taken from Levi’s writings, what had become stigmatised was ‘humanity,’ and Levi experiences shame as he acknowledges this, for his identity is in part constituted by his own humanity—not only his membership of the species Homo sapiens, but how ‘humanity’ as a moral concept, what affords us the evaluative terms humane and inhumane, for example, features in our identity. Humanity is stigmatised owing to the crimes of the Holocaust and individual members of humanity experience shame in acknowledging this stigma and their membership of the stigmatised group.

So here is gain 3:

3. Objects can be concrete, simple and directly perceived or diffuse, complex and indirectly conceived. If the unfolding of history can stigmatise the very concept of humanity bringing shame for Holocaust survivors like Levi, then this tells us something about how some acts become stigmatised over time, through political change, for example, and lead to shame where there was none before.

Shame, Honour Groups, Autonomy and Heteronomy

Discussions of the role of audience or honour group have been prominent in philosophical work on shame5 7 17 and the related question as to whether shame is an essentially heteronomous or autonomous emotion (eg, refs. 6 11–13). The question is this: does the negative evaluation of others, taking the role of audience or honour group, play an essential role in the emergence of an individual’s shame? If one answers ‘yes,’ then one is implying in that answer that shame is heteronomous: shame is dependent on the judgement of others. If one answers ‘no,’ then one allows that shame can be autonomous: shame is not dependent on the judgement of others. This question cuts to the heart of how we might address shame.

Shame can appear to be characteristically heteronomous, as we see in the kind of shame that an individual experiences when they have fallen short of some societal standard. Consider the third in the list of shame experiences we find in Levi’s writings. Here Levi experienced shame because his actions in Auschwitz did not meet the standards expected of individuals in post-Auschwitz, post-War, European society. This is an example of heteronomous shame, because the shame emerges from the mismatch between a standard which is external to the individual, which resides in an actual audience evaluating the individual’s action or in a social rule or norm. Both actual audience and social norm provide a standard against which the person’s status is diminished.

Alternatively, one might argue that shame is essentially autonomous, where shame comes from a mismatch between one’s prevailing assumptions about one’s own character and the more considered, honest, reflective judgements made about that character. We might seek to project an image of our character as morally upstanding, caring and sharing. But, when we take time to reflect, when we are honest with ourselves about the totality of our behaviour over time, we see a mismatch and we see we are not the person we assumed ourselves to be and who we sought to project to others. This is an example of autonomous shame.

This discussion has persisted in the philosophical literature on shame because it cuts to the heart of our moral psychology. From the perspective of (modern) morality, if shame were to be essentially heteronomous then it can play no significant role in an individual’s moral psychology, because to be a moral agent demands that one act on one’s own moral judgements, even when these are at odds with the judgements of the honour group or established social norms. If shame is to be an emotion worthy of a place in analyses of our moral psychology then it cannot merely track the dominant social attitudes.

We draw on Hutchinson’s6 arguments to propose that shame can appear to operate, from case to case, either autonomously or heteronomously, but that ultimately the distinction collapses. For a person to take an event as shameful is for them to have read it as such, for it to have had that meaning for them. That person, this individual, has acknowledged the event as shameful. Looked at this way, shame meets the criteria for being autonomous.

However, there are certain enabling conditions that are required for this autonomy. The world has meaning for individuals because individuals go through processes of enculturation, which include the learning of a (public) language; in turn, learning a language is to learn to register, acknowledge and communicate the loci of significance in our world. In this sense, we can see the heteronomous sources of shame. The event is shameful because it has that meaning for us. It has that meaning in virtue of our reading it this way and thereby seeing it as shameful. What enables our reading of the world in this way, seeing it under this aspect, is our second nature—our character—which is formed in and through our life with others.

While shame is intimately related to our social, encultured nature, it is therefore a mistake to assume this makes it necessarily heteronomous in a way which disqualifies it from having moral significance. Understanding that shame is not necessarily heteronomous, but can be autonomously arrived at should not be to deny that the social sources of our enculturation, the development of our moral character, serve as enabling conditions for our life with shame.

Shame can be heteronomous, such that shame is an acknowledgement, a taking-on-board, of the judgements (or morally loaded perceptions) of others about one’s self, and in so doing considering oneself, one’s being, to be in some way evaluatively diminished. While shame can also be autonomous, such that shame serves as testament to a mismatch between the sense of self one assumes and seeks to project to others and the self that one considers oneself to be on reflection.

Insisting it must be one or the other has consequences: if one holds that shame is always autonomous, then that will lead one to focus any attempt to alleviate shame solely on the psychology of the individuals who bear shame. Conversely, if one were to assume that shame is always heteronomous, then that will lead one to identify that which is in need of change as being the social norms of which the honour group (the shame-instantiating audience) are an embodiment. In this latter, heteronomous sense, addressing shame might be a political, cultural and social task, in addition to being a psychological task. This is our fourth gain.

4. We propose that addressing and treating shame can be, and often is, a political, cultural and psychological task.