As Steve and Manyul announced last month, with each new issue of Dao the blog will host a discussion of one of the issue’s articles, and the journal will make that article freely available online. Here I’m kicking off the series with a discussion of Loy Hui-chieh’s “On the Argument for Jian’ai” (Dao 12.4, available here).
Loy’s article treats the Mohists’ main argument for inclusive care (jiān ài 兼愛), focusing on the role played in it by appeals to virtues such as filial piety that are inevitably partial. Fundamental to his treatment is the view (which I share) that inclusive care did not require absolute impartiality—it did not imply that we have equal obligations to all people, or that we should treat them the same, or feel the same about them. Loy thus undermines one common sort or argument against the Mohists, that inclusive care is incompatible with the partial virtues and is therefore morally dubious. However, this does not mean that the Mohists’ own appeals to the partial virtues succeed, and Loy goes on to argue that they do not. I’ll sketch Loy’s argument, and then make critical comments on two points.
A sketch of Loy’s argument
As Loy reads it, the Mohists’ main argument for inclusive care proceeds by identifying a series of evils, offering a unifying diagnosis of those evils, and then proposing inclusive care as a cure.
The evils fall into two groups. Some involve people’s failures to possess various role-specific and partial virtues, including filial piety and loyalty. The others involve people harming either other people or various outgroups (other lineages or states for example) in order to benefit themselves or their ingroups.
The Mohists claim these evils arise because people do not care sufficiently for one another. (The “sufficiently” begs an interpretive question that Loy treats at pp. 492–493.) Loy adds to this, writing that (in two versions of the main argument) “[the Mohists'] diagnosis of the problem is that it is mutual lack of care that led to reciprocal harming among people” (p. 492). For the purposes of his larger argument the key claim here is that all of the evils that the Mohists had identified consisted in people harming one another.
And similarly with the cure that the Mohists propose. This is of course that people care inclusively for one another, but Loy again goes further, this time saying that inclusive care is supposed to work as a cure because it will prevent people from harming one another and (perhaps) get them benefiting one another more (more on that issue in a moment).
You can see how this is supposed to work with the second group of evils the Mohists identify. A lineage head cares for his own lineage more than for other lineages, so he is willing to mobilize his own lineage to harm the other ones; that’s an evil. But if he cared sufficiently for the other lineages, then he would not be willing to bring harm upon them, and the evil would be cured.
But the first group of ills do not yield the same treatment, and that is Loy’s objection to the Mohists’ argument. Here is how it would have to go. A son cares for himself more than for his father, so he is willing to harm his father to benefit himself; this is the son lacking filial piety. If he cared appropriately for his father, though, he would be filial, because he wouldn’t harm his father and (perhaps) would also benefit him appropriately.
The problem is that filial piety does not work that way—it is not simply a matter of whether or how one benefits or harms one’s parents. (And similarly for the other partial virtues.) Loy thus concludes that the Mohists’ diagnosis of the ills they identify is incompatible with the inclusion among those ills of people’s lack of filial piety (and other partial virtues).
Care and benefiting
My sketch of Loy’s argument evaded the question of what inclusive care amounts to. Here we face two pressures. The first is the widely-shared sense that inclusive implies impartial. The second is the Mohists’ obvious endorsement of a range of partial virtues. How can they advocate impartial care alongside filial piety (for example)? Is it not obvious that filial piety requires us to care more for some than for others?
Aside: the reasons for thinking that inclusive care is impartial are a lot weaker than the reasons for thinking they endorsed partial virtues; they are weak enough that if there’s no way to reconcile impartial care with partial virtue then the appropriate conclusion is probable that inclusive care would not be impartial after all.
Loy entertains three accounts of what inclusive care might demand: equal concern to benefit all people, equal concern to help all people, and refusal to harm anyone (pp. 497–498). The third of these does not go far enough: the Mohists regularly associate inclusive care with benefiting people. The first sounds like it goes too far, for (as Loy in effect recognizes at p. 497) it seems incompatible with partial virtues. So what about the second?
The idea, I take it (drawing still on p. 497), is that some people are in need, and we should be just as concerned to help them to the point where they are no longer in need as we would be concerned to help ourselves if we were similarly in need. It follows that we need not be concerned to benefit all people equally, because inclusive care is irrelevant to the things we do to benefit people who are not in need.
I do not see how this can be right. Consider an elderly couple who would be in need if their children did not provide for them. Does the children’s filial benefiting (as I will call it) count as helping? If it does not, then (on the proposed interpretation) it seems that it must be irrelevant to inclusive care and thus that the children could still count as caring appropriately for their parents even if they did not provide for them. This is manifestly not the Mohists’ view. Except: this would leave the parents in need; which is to say that inclusive care would require the children to engage in filial benefiting if and only if they did not engage in it. Surely that is mad. So filial benefiting must count as helping. But then (on the proposed interpretation) the children are supposed to be equally concerned to provide for all people, and again that is manifestly at odds with the Mohists’ views (and the helping interpretation inherits all the problems of the benefiting interpretation).
So I don’t think this is a promising way to reconcile inclusive care with partial virtue. I’ve got an account I prefer, but won’t get into it here; it’s in my “Mohist Care” (PEW 62.1).
The Mohists’ diagnosis
Recall Loy’s account of the diagnosis the Mohists’ offer of the world’s ills: these ills all involve people being willing to harm one another. The problem is that this seems a poor way to characterize a person’s lack of a partial virtue. If a son is unfilial, this does not seem to mean just that he is willing to harm his father. But if Loy’s account of the Mohists’ diagnosis is correct, the Mohists seem to be assuming that unfiliality just is a willingness to harm; and they seem to be reducing filial piety to a refusal to harm and a willingness to benefit.
Loy’s account gains its main support from these statements (and the subsequent parallels about the converse relationships):
Subjects and sons not being filial to rulers and fathers is what is called disorder. Sons care for themselves but not for their fathers, so they injure their fathers to benefit themselves. Younger brothers care for themselves but not for their older brothers, so they injure their older brothers to benefit themselves. Subjects care for themselves but not for their rulers, so they injure their rulers to benefit themselves. This is what is called disorder. (Book 14)
Here the Mohists do seem to equate a lack of filial piety with a tendency to a harm, and thus invite Loy’s objection. But these statements (along with the subsequent parallels) are the only ones in all three versions of the main argument that do this. Nowhere else do the Mohists suggest any reduction of filial piety to benefiting.
Consider the account of the cure the Mohists propose from the same version of the main argument:
If the world cared for one another inclusively, if they cared for others like they cared for themselves, would any still be unfilial? If they saw their fathers, older brothers, and rulers like their own persons, how could they practice unfiliality? (Book 14)
(There again follow parallel claims about the converse relationships.)
Here there is nothing about benefiting and harming; the Mohists treat unfiliatily as itself an ill.
Similarly, in the second version of the Mohists’ diagnosis, we find them describing people harming one another when dealing with the second group of ills that Loy identifies, ills such as states going to war with other states. But with the first group of ills, the ones relevant to the partial virtues, they say only this:
If ruler and subject do not care for one another they are not gracious and loyal, if father and son do not care for one another then they are not loving and filial, if elder and younger brother do not care for one another they are not harmonious. (Book 15)
That is, the Mohists relate the ills in question—the lack of the partial virtues—to a lack of care but not to a willingness to harm. Loy’s direct statement to the contrary (p. 492, citing this passage) is just wrong. And, again, there is nothing in the subsequent account of the Mohists’ cure that implies that the partial virtues require only that we benefit others.
Loy’s discussion of the third version of the main argument (p. 495) is much briefer because this version of the argument is itself very brief. As far as I can tell, Loy does not claim that this version of the argument reduces the partial virtues to benefiting, and certainly he would be wrong to do so.
In sum, Loy’s objection to the Mohists depends on an account of the diagnosis they offer of the world’s ills, but that account is motivated by just one passage in just one version of the Mohists’ argument. This might be reason to take issue with that one version of the argument. But whether or not the Mohists recognized this problem, they avoided it in later versions.