The Authority of Empathy and the Ground of Sentimentalism

 

Draft: First paragraphs (not viewable on mobile devices)

 

 

The man whose sympathy keeps time to my grief, cannot but admit the reasonableness of my sorrow…He who laughs at the same joke, and laughs along with me, cannot well deny the propriety of my laughter. On the contrary, the person who, upon these different occasions, either feels no such emotion as that which I feel, or feels none that bears any proportion to mine, cannot avoid disapproving my sentiments on account of their dissonance with his own.

    

- Adam Smith, The Theory of Moral Sentiments[1]

 

 

 

However we do it – and it seems we do it in a few different ways – we empathize. When observing or considering the emotions of another person, we sometimes feel an emotion to a sufficient degree and quality to satisfy us that our emotion and the emotion of this other person are “in accord,” as I will put it. Almost as obvious, as Adam Smith keenly observed, we frequently make judgments about the propriety of emotions on the basis of empathy, defined is this very loose way. This is clearest when the “way” to empathy is one person imagining the situation of another. That is, when one person reflects on that conglomerate of objects, both social and environmental, that the other person’s emotion is a response to – a process many would classify today as “simulation.” If, by dint of such reflection, we have the same affective response to that situation as the agent herself, then the consequent empathic accord prompts us to approve her emotion, in some sense. And if our feelings diverge, or to the extent that they do, this lack of empathic accord typically prompts us to disapprove the emotion – again, in some sense. For ease, let us hereon call this assessment based on whether or to what degree literal empathic accord is observed, empathic judgment.

 

Obviously, this is a rough and ready starting point. It is the task of Part 1 to substantiate it. This will involve two principal points of clarification. The first is the notion of “approval,” which will require a few rather fine distinctions. In part, this is because the relevant normative judgment turns out to be novel.[2] Second, because many different phenomena pass for ‘empathy’ these days, the nature of empathy will also need clarification. But in this case I’ll offer coarser distinctions. For, my goal in this essay isn’t to offer a theory of empathy per se. Nor am I here interested in theory of mind debates and their the nitty-gritty debates about empathy’s epistemic role, e.g. in producing knowledge of what others feel by justifying judgments of the form “S has emotion e.”[3] Instead, to reiterate, my aim is to clarify and defend our practice of judging emotions empathically, and for this end some rough classification of empathy should suffice.

Importantly, however, this is not the only goal. Or better, because more precise, my goal isn’t to substantiate the starting claim only through the argument, in Part 1, about the nature of empathic approval. For ultimately the descriptive metaethics of Part 1 is instrumental to staking out, in Part 2, a broader view about the normative grounds of empathic judgment and in turn the prospects of ethical sentimentalism more generally. Let me explain.

 

Sentimentalists must eventually connect their metaethics to a general theory of morality. Given that (as is true in this paper) the “object” of the judgment in question usually just is a sentiment or emotion,[4] this means explaining the significance of emotion to morality. Classically, this connection was made straightforwardly in virtue of taking emotions as the motives for action and the central factor in determining reward and punishment, gratitude and resentment. David Hume and Adam Smith, for example, both advanced the agenda of ethical sentimentalism this way. However, I will not. In fact, at the end of Part 1, when we turn to possible objections to my account of empathic judgment – such as, does empathy with nasty emotions really entail approving them? –  my reply will at one point recall the following contention: the justification for an emotion is distinct from whether that (justified) emotion justifies any given action.[5] This essay takes no position on the latter question, and in the end complicates it, for two reasons. First, I will argue that the empathic standard is always in principle in conflict with other normative standards of emotion like prudence and the much discussed idea of “fittingness,” which asks whether an emotion’s evaluative outlook matches the evaluative “facts,” i.e. whether the object of any emotion F is really F-able (e.g. crudely, my anger is fit when the object of my anger is genuinely frustrating or outrageous; my fear is fit when its object is really fearful or dangerous, etc).[6] Correspondingly, empathic judgment does not settle the normative question of right feeling: it isn’t a de facto all-things-considered judgment of “how to feel.” Second, compared to these other standards, we will see that empathy actually issues a very weak kind of approval – and in then end, probably not even aptly called “approval.” Correspondingly, empathic judgment issues reasons-to-feel that are easily swamped by the reasons derived from other normative standards for emotion. Taken together, then, empathizing with another person’s emotion turns out a far cry from licensing her to act on that emotion.

 

Of course, the foregoing raises an obvious question, which brings us back to the point at hand: If empathy has such weak normative significance, why investigate it? Or, what amounts to the same thing, in absence of the kind of straightforward classical connection those like Smith offered, how does my inquiry advance sentimentalist agenda? At this point I can answer this in only the most preliminary way, but here it is: although the content of empathic judgments is very weak, the form of this judgment reveals itself as grounded in an awesome normative authority, namely, the fundamental worth or status we ascribe to ourselves as moral agents – albeit, as we shall see, not agents qua rational but qua affective. And thus through the currency of empathic judgment sentimentalism gets a purchase on one of the more urgent human aspirations of the modern age: the notion of human dignity. Defending this claims is the goal of Part 2.

 

 

 

[1] Smith, TMS I.i.3.1. (The Theory of Moral Sentiments (TMS), (ed.) D. D. Raphael and A.L. Macfie (Indianapolis: Liberty Fund Press, 2001). References are to Part N°, Section N°, Chapter N°, and paragraph N°.

 

[2] Smith argued that approval expressed a distinctive pleasure in “mutual sympathy,” or what I’ve called empathic accord. What isn’t perfectly clear on Smith’s view is the normative content of this approval. Smith called it a judgment of propriety, but by contemporary lights “propriety” is ambiguous between a few possible meanings. I say a bit more about this and the connection to Smith later on.

 

[3] Hume also posited a normative role for empathy (sympathy). But he developed this account in Book 3 of the Treatise, well after dissecting sympathy’s epistemic role T. 2.1.11 (though, of course, not in terms of producing knowledge).

 

[4] The term ‘emotion’ had only just emerged when Hume and Smith were writing. And while I acknowledge possible differences between emotions and sentiments (from both a historical and contemporary perspective), let alone ‘passions’ or ‘feelings,’ it would take us too far afield to examine these distinctions. Moreover, partly for the reasons I express in this sentence, it is not clear that any such differences would be directly relevant to my own inquiry. For a historical primer on the evolution of emotion terminology and concepts, see Thomas Dixon, From Passions to Emotions: The Creation of a Secular Psychological Category (New York: Cambridge University Press, 2003).

 

[5] See also Charles Griswold -- who makes a more sustained argument for this point.

 

[6] The terminology of fittingness has been popularized principally by the work of Daniel Jacobson and Justin D’Arms, who, through a series of articles over the last two decades have been advancing the argument that fittingness is the central normative notion in “neo” sentimentalist metaethics. See e.g. Justin D’Arms and Daniel Jacobsen, “The Moralistic Fallacy: On the ‘Appropriateness’ of Emotions,” Philosophy and Phenomenological Research 61:1 (2000) p. 68.