Critical Precis of “Is Sympathy Naïve? Dai Zhen on the Use of Shu to Track Well-Being” by Justin Tiwald
By Tobias A. Fuchs
I am very grateful for the opportunity to comment on Justin Tiwald’s incredibly rich paper. It’s been a pleasure to engage with the many interesting ideas in it, some of which resonate strongly with my own views. I am myself a defender of both the idea that we can learn about well-being by making use of something like sympathy, as well as a view very similar to the particular account of well-being his paper sketches. But while I am very sympathetic (no pun intended) to these views, I am skeptical of some of the connections Justin draws, and I doubt that sympathy is quite as powerful as he might have it.
Before raising some questions for discussion, I will summarize Justin’s paper. Since I am not expert on Dai Zhen, I will not address exegetical issues regarding Dai’s work. I should also note that this paper is from 2009, and Justin tells me that his views on some of the issues may have evolved. But I leave it to him to set the record straight in discussion, should it become relevant. One more point: the central concept here is shu, which Justin/Dai Zhen understand as a form of sympathetic understanding. Following Justin, I will use that term (or simply “sympathy”).
Dai Zhen (“Dai”), an 18th century Chinese philosopher, is known for his criticism of then dominant Neo-Confucians. He argues that by discouraging the use of our desires in moral deliberation, his adversaries undermine sympathetic understanding of others. Healthy desires and shu are needed to know what harms and benefits people. Since this knowledge is crucial to moral understanding, we need shu for proper moral deliberation.
What is often missed is just how central and rich a notion shu is for Dai. Justin sets out to rectify this, by arguing for a particular understanding of how sympathy tracks well-being. In the course of this, he presents Dai’s account of the human good and argues for a certain view of self-interest.
Dai’s account of shu starts from a classical formulation in the Analects: “What you would not desire yourself, do not inflict upon others” (15.24). Shu, or sympathetic understanding, tells us how to treat others by considering how we ourselves desire to be treated. It’s a fairly familiar idea that sympathy helps us to understand what benefits and harms people, but it’s less clear how it does this.
We usually conceive of sympathizing as imagining ourselves in the place of another or others (the person or persons we aim to sympathetically understand), imagining what it would be like to be in their shoes. This can then inform our deliberation of how to act, insofar as it affects those others, but we might be wrong about what they want. Their desires might be different from ours, and some desires can be bad for the desiring agent to satisfy.
One move would be to imagine what the other(s) would desire if they were fully informed and rational. But this is not how Dai sees it. Rather than imagining ourselves in the other’s place, and then thinking about how to ‘correct’ that picture, Dai holds that sympathy alone can tell us what’s really good for a person.
In order to show this, Dai must establish that sympathy can correctly tell what is good for a person from what is irrelevant or even detrimental to their well-being. In terms of desires, this means distinguishing which desires really are in our interest to satisfy. So what are the ‘good’ desires?
In moral deliberation, according to Dai, we need to attend specifically to those desires that “belong to us by nature and have a universal or near-universal status in ordinary human beings.” (p. 148). So desire satisfaction is for Dai only a derivative good – it is only instrumentally valuable to achieve a more fundamental good, namely what he calls “life fulfillment” (sui sheng).
Well-being, on Dai’s account, is life fulfillment, which requires certain goods and humans accordingly have basic needs. So if we want to promote others’ well-being, we ought to promote the fulfillment of those of their desires that are conductive to life fulfillment. And that is what morality requires of us. “It is to the particular desires that arise from the structural requirements of living … that the truly virtuous or “humane” (ren) person is supposed to attend” (p. 149).
Assuming that shu/sympathetic understanding helps us to track the well-being of others by projecting certain desires onto them, how can we ensure that it is reliable, i.e. sorts out those desires that really contribute to the others’ well-being?
We might have some independent way to make that restriction. Sometimes we might disregard somebody’s apparent desires when they’re obviously irrational, for example if our friend Wang wants to give up his career in order to watch the home team win. However, a more systematic approach would be to regulate one’s own desires, since those “go into our deliberations in the first place”, through a process of self-cultivation. Zhu Xi, one of Dai’s philosophical opponents, advocated this latter approach. For Dai, this is too demanding. Shu is supposed to help ordinary people in their moral deliberation. He therefore believes that Shu itself can determine which desires count.
This brings us to the titular question: is sympathy (shu) naïve? By naïve shu, Justin means shu is mere perspective taking (imagining oneself in the other’s place, “feeling as he would feel and wanting as he would want”). We normally think of sympathy as naïve in this sense, because we – raised in the modern (western) tradition – are used to give authority to the individual’s subjective assessment of what’s good for them. To know what’s good for Mary, it suffices to know what she would want (at least want under ideal circumstances). So we think taking her perspective is all we need.
However, this view commits us to regard rationally held altruistic desires as impossible. Because if whatever I would rationally and fully informed want is by definition good for me, then self-sacrificial desires are not possible (lest one absurdly took self-sacrifice to enhance the sacrificing person’s well-being). This is the problem of self-sacrifice, generally seen as a serious problem for desire views of welfare (see Overvold 1980; Heathwood 2011). But self-sacrificial desires are quite common: parents do want to sacrifice for their children – and vice-versa (recall the important Confucian virtue of filial piety).
Self-sacrificial desires, Justin argues, “point to a more nuanced account of self-interest (si).” (p. 154). He quotes Dai, who sees self-interested care for oneself and care for those near and dear both as forms of humane love (ren). What do these forms of benevolence share? “All creatures of blood and qi know the love of life and fear of death, and therefore pursue benefit and avoid harm.” Seeking someone’s benefit and avoiding harm just is having a benevolent attitude towards them. So benevolence towards myself and benevolence towards others have this in common: love of the person’s life. Self-interest must therefore be more than rationally desiring. It requires humane love (ren) towards myself. We need this love (ren) towards ourselves to understand what’s in our self-interest. Ren does some of the work of distinguishing desires that are good for us from those that aren’t.
Furthermore, this humane love or benevolence toward someone which, according to Dai, aims at fulfilling that person’s life, contains shu, or sympathetic understanding within it. This might be why self-interest requires humane love – the sympathetic understanding of ourselves contained therein tells us which of our desires are important for our life fulfillment. By reflectively “returning to ourselves”, that we recognize the desires we share with other human beings qua human beings and see that they truly count.
This is the knowledge essential for proper moral deliberation and shu is essential in acquiring this knowledge. Shu, on this view, is not naïve. It itself provides us with insight into what is good for people and is therefore fundamental to moral understanding, as well.
1. Should we mix accounts of sympathy with accounts of well-being?
For Dai, knowing what’s good for a person is knowing which of their desires are worth satisfying. Accordingly, if shu can tell us what’s good for people, it must be able to tell us which of their desires are good for them to satisfy. The paper therefore proceeds to discuss problems with the basic idea that desire-satisfaction is what constitutes our well-being. The development of an account of sympathy is interwoven with the discussion of well-being.
But these are two separate things. I can be a hedonist and entertain a variety of views about how sympathy works, and what it can accomplish. I can be a perfectionist and entertain those same views. Conversely, I might believe that sympathy is naïve, and still hold the very view of well-being that is presented in this paper. Even if sympathy was able to tell us what really is good for us, this should in principle be compatible with any number of theories of welfare.
Why even frame the whole discussion in terms of desires? Of course this is because for Dai – and most of his interlocutors – shu had to do with figuring out what the other may desire or not desire to have done to them. But this is not how we now think of sympathy/empathy. Sure, desires are part of the picture when we imagine what it’s like for someone. But that’s not necessarily what sympathy is about. Suppose a relative tells me that she’s suffering from debilitating back problems. It’s no secret that she desires to be rid of those problems. But the one thing that would help me most to properly sympathize with her, would be for me to have had a similar affliction myself. Knowing what it is like is what would make for good sympathy.
It wouldn’t make a difference if it turned out, surprisingly, that such painful back problems were actually good for people. That’s partly because accounts of sympathy are a different matter than accounts of well-being.
2. Do we even need shu?
Speaking of Dai’s view of well-being, we might wonder whether it doesn’t undermine his case for the importance of shu. On Dai’s account, our well-being should be understood as “life fulfillment”. This means that we fulfill those needs we have simply in virtue of being human beings (and which we therefore share with other human beings). Needs that “arise from the structural requirements of living.” (p. 149). We have desires for those goods we need to fulfill our lives. And it is the satisfaction of those desires that benefits us. Other desires may contribute nothing, or even detract from, our welfare. “Those [desires] that really count are the core of basic desires that we share in common with other human beings.”
a. The question thus arises: if those desires are universal (or near universal), then why do we need shu, especially such a complex and powerful shu as Dai advocates? It seems that all we need is a good grasp of what those universal desires are (understanding of which might be facilitated by the fact that we have them, too) and an accurate description of the subject’s situation, in order to know what they lack and need. Provided we have good judgment, this could be a completely ‘cognitive’ task, sympathy not required.
Of course, Dai argues that it is shu which tells us which the relevant desires are in the first place: “It is…sympathy that informs our evaluations of well-being, and not the other way around.” (p. 157). Even so, once we have learned – through shu – what the right desires are, we should be good to go.
b. Another reason to question whether shu is necessary comes from an entirely different direction. There are several other concepts Dai brings to bear to flesh out his full account of shu. A central notion here is love – the humane love, or humanity (ren) essential to (or perhaps equivalent to) benevolence and care. “To care for oneself is to be humane toward oneself,” says Dai (p. 154). Shu is contained in humanity.
Justin tells us that our ability to take the attitude of humanity, or loving care, toward ourselves, enables us to tell apart what we want from what we want for our own sake, i.e. distinguish the desires that are in our interest to satisfy from the others. “Needless to say it is not sufficient for this insight- and Dai Zhen is eager to supplement our deliberations with the refinement in judgment and knowledge of the world that comes with successful self-cultivation- but it is necessary all the same.” (p. 156). In other words, humanity (ren) is necessary but not sufficient to distinguish ‘good’ desires from bad or irrelevant ones. If ren contains shu, and ren is merely necessary but not sufficient for this knowledge, then shu is also at best necessary but not sufficient. So shu is not sufficient for the knowledge we’re after.
If we need to supplement ren (and therefore shu) with the refined judgement and knowledge of the world that comes with successful self-cultivation, then on what grounds can Dai maintain his rich account of shu which by itself can do that work?
3. Does Dai pack too much into sympathy?
At this point, it is time to finally say something about Dai’s account of shu directly. On Dai’s account, shu, or sympathetic understanding, tells us what truly harms and benefits the person sympathized with (whether ourselves or others). Surveying our own desires with shu, we learn which of those we should satisfy in order to enhance our well-being, and which one’s would be irrelevant or harmful to satisfy.
This is surprising, if we think of sympathy as the emotion by which we can feel for others. As an emotion, sympathy must follow certain patterns. Regardless of which particular theory of emotions one embraces, emotions are first and foremost a way of reacting to something that happens in the world or in one’s mental life. Emotions are always triggered by some stimulus. Fear, for example, is triggered by something that appears threatening to us (see e.g. Deonna & Teroni 2012). Sympathy, is a reaction to perceived harm or misery of some sort. My friend tells me of receiving her 20th rejection letter in a row, and I react by feeling sympathy for her. But if this is the case, it cannot be that sympathy tells me that my friend is badly off, because without it already seeming to me that she is badly off, I wouldn’t feel sympathy in the first place.
But perhaps this is not what is meant here by “sympathy”. Indeed, much of what Dai says appears to be better captured by our term “empathy”. Empathy is not an emotion, but rather a mechanism, or a capacity, for, getting at what others feel, to put it broadly. As such, it does not need to be triggered in the way emotions are, but can be activated at will. It can also sometimes work ‘automatically’. There are many conflicting accounts of empathy (see e.g. Batson 2009; Coplan & Goldie 2011; Maibom 2014), but most agree that it is what allows us to take another’s perspective. Empathy therefore comes closest to what Justin calls “naïve shu” – a “mere” perspective taking. Just as he says, we can empathize with our friend Wang and thus recognize how much it means to him to see the home team win, before we make a – separate – judgement that Wang is irrational in wanting to even give up his career to watch the game. (An interesting question to raise here would be how we make that judgement).
Such mere perspective taking, however, “cannot possibly be a dependable indicator of well-being on its own” (p. 147). One other way mentioned to help empathy out is to self-cultivation. This seems plausible. Generally, the better we understand our own feelings and desires, the better able are we to make sense of others (on a related view, see Kauppinen 2014). And we saw in the discussion of my previous question how Dai himself “is eager to supplement our deliberations with the refinement in judgment and knowledge of the world that comes with successful self-cultivation” (p. 156).
Nevertheless, as an account of how shu works, Dai rejects this as too demanding. Everyday people should be able to use shu as is in their moral deliberation. Justin sees it therefore as more natural to understand sympathy as itself telling us what’s really good for the people we sympathize with. “Much as grief for the passing of a loved one, for example, lays bare the ways in which the deceased contributed to our personal fulfillment…so, too, does sympathy help to focus and make more vivid the facts that are salient for determinations of well-being.” (p. 151).
Since for Dai, shu can tell which kind of desires contribute to well-being and which one’s don’t, grief, by this analogy, would tell us which kind of actions towards us contribute to our fulfillment, and therefore we learn from our grief which kinds of contributions the deceased made to our life benefited us. This does not strike me as an accurate description of grief. As an emotion, a kind of sadness, grief is a reaction to perceived loss. Unreflective grievers may not be able to say anything else than that they feel very sad about the death. But reflective people, especially with some life experience, may understand that their grief is a reaction to perceived loss, and therefore know that they must be experiencing this death as a loss (see Nussbaum). From this, one might deduce that one must see the deceased as something of value, and upon further reflection, one might come to remember all the ways in which the deceased contributed to one’s life in valuable ways.
In the same way, when we empathize with someone, we might simply feel, or apprehend, their sadness (or whatever). But if we want to use this in deliberating about what to do about it, we may be at a loss. Sure, there are cases where it may be obvious what the problem is. For example, if my sister is wailing as I accidentally put down a heavy box on her foot, I as much as anyone can tell that what is needed is to lift the box off her foot. But in more nuanced cases, it takes much experience and information to know what to do.
To me at least, it seems more natural to see empathy as naïve, and deliberation or knowledge of what the other needs as a separate matter. When I understand that Wang is irrational in his desire to watch the game, it is not my sympathy that lets me know. Rather, my sympathy tells me just how much it means to him – that’s why I sympathize with him when I tell him that he needs to go to work instead.
To be sure, sympathy can help us understand the plight of others, but it is far from a reliable source of knowledge about well-being. Not only because we can be mistaken about the facts. Other emotions can get in the way, as when we don’t feel bad for someone because we are upset at them (see Hooker 2015). Or we might overestimate some harm because our sympathy is heightened due to our love for the harmed. At the same time, we may underestimate the harm a stranger is suffering, because our sympathy for her is less than it would be for a loved one (see Hoffman 2000). This effect may be amplified if the harmed are many in numbers, in far-away places (see Prinz 2015).
If we could be sure that sympathy was accurate, then we could directly infer facts about harm from it. As I’ve argued elsewhere, that’s because sympathy is only fitting if the target person’s well-being is negatively affected. But this does not mean that actual sympathy therefore must be accurate. The most we can hope for is to ‘cultivate’ it. But that we could ever be so wise, cultivated, and benevolent that our sympathy was guaranteed to accurately track well-being is surely unrealistic. It may be an ideal to strive for, like that of the sage, but even Confucius himself did not claim this status for himself. Sympathy is helpful in thinking about well-being, but it is not as powerful as Dai would have it.
I will leave it at that. There is much more in this paper than I am able to comment on or even mention here (for example similarities and differences between Dai’s account and Darwall’s view of welfare as rational care [Darwall 2002], which Justin cites as inspiration, especially in relation to Justin’s points about self-interest). I hope to have been able to do justice to Justin’s multifaceted work, and to raise some interesting points for discussion.
– Batson, C. D. (2009). “These Things Called Empathy: Eight Related but Distinct Phenomena,” in Decety. & Ickes, eds. (2009). The Social Neuroscience of Empathy. Cambridge, MA: MIT Press.
– Darwall, S. (2002). Welfare and Rational Care. Princeton: Princeton University Press.
– Deonna, J. A. and Teroni, F. (2012). The Emotions: A Philosophical Introduction. NY: Routledge.
– Coplan, A. & Goldie, P., eds. (2011) Empathy: Philosophical and Psychological Perspectives. Oxford: Oxford University Press.
– Heathwood, C. (2011). “Preferentism and Self‐Sacrifice”. Pacific Philosophical Quarterly 92 (1), 18-38.
– Hoffman, M. L. (2000). Empathy and moral development: Implications for caring and justice. New York: Cambridge University Press.
– Hooker, B. (2015). “The Elements of Well-being”. Journal of Practical Ethics, 3 (1), 15-35.
– Kauppinen, A. (2014). “Empathy, Emotion Regulation, and Moral Judgment”. in Maibom, 97-121.
– Maibom, H. L. (2014). Empathy and Morality. New York: Oxford University Press
– Overvold, M. C. (1980). “Self-Interest and the Concept of Self-Sacrifice,” Canadian Journal of Philosophy 10, 105–118.
– Prinz, J. J. (2011). “Against Empathy”. Southern Journal of Philosophy, 49 (s1), 214-233.