by Daniel A. Kaufman
 Philosophers sacralize moral obligation and maintain that moral considerations are always overriding of all others, and yet ordinary people (as well as philosophers in their ordinary lives) hold actions done from earnest desire in much higher esteem than those done from duty. “Don’t just do it out of a sense of duty” is a commonly expressed sentiment, and who, upon hearing that someone is doing something for you solely out of duty, hasn’t at least thought that you’d rather they not do it at all? Indeed, such common wisdom is this that it is standard in the raising of children.
 That people should sometimes do what they aren’t obligated to do also seems a common and comprehensible human attitude that modern moral philosophy is ill-equipped to make sense of. “You should be generous” expresses such an attitude, and it’s only one of many such expressions one hears in ordinary conversation.
Of course, philosophy can offer an analysis of this in terms of hypothetical imperatives: for example, People should sometimes do what they aren’t obligated to do, as it is a good way of currying favor or some such thing. But I would maintain that this doesn’t provide a good reconstruction of this common attitude. Nor is it well-construed via Kant’s conception of “imperfect duties,” which are simply obligations that apply under practical constraints.
No, the idea here is that one doesn’t want to be the sort of person who only does what’s required or what will invite punishment or blame if not done, and as observed in , this it is a wisdom so common that we have a stock of expressions articulating it, many of which are employed in the raising of children. The reasons why it’s (usually) a bad idea to live this way are many and cut across a number of categories, in a manner that is resistant to the abstractions, simplifications, and [small-‘r’] rationalism required of most modern moral philosophy.
In discussing this with my colleague and boss, Elizabeth Foreman, she said that it’s a good part of the reason why she dislikes act-centered moral theories – a category to which virtually every major modern moral philosophy belongs – and prefers agent-centered ones. Certainly, a virtue ethic will have an easier time making sense of this common attitude than will Utilitarianism or Kantianism.
 Yet another concept for which modern moral philosophy seem incapable of giving an accurate reconstruction is that of the supererogatory.† Philosophers define it as that for which one is rightly praised if one does it, but not rightly blamed if one fails to do it, but this is a poor analysis of the ordinary concept, the sense of which is “going above and beyond the call of duty.”
The philosophical treatment of supererogation will include all of the most mundane of Kant’s imperfect duties, as well as acts so noble they go beyond what duty requires, but this combination of the prosaic and the holy is not what is not what people mean to invoke when they employ the expression. Indeed, given that modern moral philosophy sacralizes the obligatory, as mentioned in , it is almost analytically ill-suited to making sense of the idea of something being above and beyond it.
 Last week, while I was visiting the superlative Robert Talisse in Nashville, I asked him why the following wasn’t the best argument for liberal democracy one could give: Because you won’t win every election and you can’t kill all your opponents.
It seems to me that where a prudential case is available, it’s always preferable to a theoretical one: fewer assumptions to take on board, no appeal to spooky or otherwise dubious entities, and persuasive in a straightforward, concrete way that requires little by way of elaborate or abstract thinking and is most likely to “travel well” cross-culturally. And given that politics, like ethics, is a practical discipline, whose purpose is to address how we should actually live, this last point is especially important.
Robert wasn’t especially dazzled by my formulation of the fundamental principle of liberal democracy, but he also couldn’t really deny it, which led to an interesting conversation on the usefulness of theory in practical areas of philosophy.
 Over dinner, for which Robert’s friend and colleague Scott Aiken joined us, I maintained that philosophers overestimate both the efficacy and the aptness of arguments. I hadn’t been very moved by certain first-order moral claims they were making or the arguments for them, and at one point Robert said of my intransigence that tenacity isn’t an argument. He’s right, of course, and that’s when I said the thing I said about arguments.
In one sense, there’s nothing strange about philosophers’ overestimation of arguments. It is our primary modality, after all, and as a general matter, people overestimate the value of the thing they’re best at or do the most. But in another sense it’s quite weird, as philosophers (like Soylent Green) are people. And surely, as people, we must have noticed that other people – and not just the stupid ones – don’t value or employ arguments anywhere close to as much as we do.
It’s worth thinking about what arguments actually get you. If you’ve managed to put together a valid one, it means that if your premises are all true, then your conclusion must follow. Of course much of the time – and in the case of the sorts of arguments I was talking about with Robert and Scott, pretty much all of the time – some or even most of the premises one is relying on are controversial or even actively contested, and just as often there is no way to confirm their truth. (In any ethical or political dispute, many of the premises on which a judgment or imperative is based, are going to be value judgments.) What this means is that many if not most arguments one encounters are just a bunch of complicated conditionals, for which the people offering them cannot confirm the antecedent. Arguments can thus function as a kind of bluff (in ordinary discourse especially, but in academic discourse as well), giving the aura of having demonstrated something, when all you’ve really done is offer a bunch of “if-thens” that you know you can’t cash out.
If you want to get people to do something or change their minds on matters of value, the most important thing you can do is get them to care, and arguments are just about the worst way to do that (though facts may be essential). If you want them to change their non-axiological beliefs or practices, it may be worth asking yourself: (a) why do you want this? (b) how important is it that the other person agree with you? (c) how sure are you that you are correct? and (d) is it the sort of thing that really can be demonstrated by an argument, or is it merely bluffed by means of one.
†Throughout I have been speaking of offering a “reconstruction” of a common practice, attitude, way of speaking, etc. In the Analytic philosophical tradition from which I hail, the purpose of a philosophical theory is to provide an analysis and rational reconstruction of something first-order: a practice; attitude; belief; theory; etc.
With regard to ethics then, as W.D. Ross put it in The Right and the Good, “the main moral convictions of the plain man [are] not opinions which it is for philosophy to prove or disprove, but knowledge from the start,” which means the job of moral philosophy is to provide an analysis and reconstruction of actual ethical practice, not invent a practice and then prescribe it. After all, as Bernard Williams has pointed out in “The Human Prejudice”:
It is a muddle between thinking that our activities fail some test of cosmic signiﬁcance, and (as contrasted with that) recognizing that there is no such test. If there is no such thing as the cosmic point of view, if the idea of absolute importance in the scheme of things is an illusion, a relic of a world not yet thoroughly disenchanted, then there is no other point of view except ours in which our activities can have or lack a signiﬁcance.
And with regard to philosophy more generally, as Stanley Rosen observed in Metaphysics in Ordinary Language (and which I’ve quoted before):
Ordinary language is retained as a basis… of all technical dialects. It is this basis that serves as the fundamental paradigm for the plausibility or implausibility of theoretical discourse and particularly of philosophical doctrines.”