I've been thinking about the duty to keep promises a little recently, and it strikes me that there's something odd about such duties.

Imagine a group of people who know everything: they always know how they should act, how the others should act, and so on. Obviously, they all know the same things, and all agree. Would such a group have need for promises? Presumably not. Knowing that someone will do something is a perfectly good substitute for their promising to do so. So it seems that promising only makes sense under conditions of epistemic ignorance.

(Holists can jump off the bandwagon here: epistemic ignorance might enable promise-keeping to be a duty without being the source of that duty.)

But then it looks like a very odd sort of moral duty to keep promises. For we aren't obliged to keep them for distinctly moral reasons, but rather for epistemic ones. Promise keeping is mandatory because it makes the information you've given people reliable. You make it true that their expectations are fulfilled.

Presumably the reason the duty to keep promises is still moral is because the duty is to do an action, even if that action is itself required for epistemic benefit. So the reason we have to keep promises is a theoretical reason for action. That seems to make promise-keeping parallel to practical reasons for belief: when the justification for some belief is not the epistemic status of that belief, but some more practical benefit such believing brings.

But now things are really confusing: Many want to deny that there can be such things as practical reasons for beliefs. Will they want to say the same thing about theoretical reasons for action? That is, will they want to deny the existence of duties to keep promises merely because of the epistemic foundations of such duties?

The classic Rawlsian worry with utilitarianism is that it ignores the seperateness of persons. In particular, it is thought to do so because it aggregates across people. By concering itself only with the aggregate of your utility and mine, it seems to ignore the fact that these units of utility belong to different people. And, the thought goes, that's a morally relevant fact.

There are various ways of responding to this. One is to point out that in at least some senses, utilitarians do acknowledge the seperateness of persons. The view that we should aggregate utility across people is just a view on the relevance of the seperateness of persons, and not a way of ignoring it.

Another would be to suggest that any plausible view will have to aggregate at least sometimes. Even those with the most individualists theories want to have disaster-prevention clauses, which allow their theories to allow aggregation of benefits and losses in extreme situations.

But I wonder if there's another response: Why accept that utilitarianism needs to be cast as an aggregative theory at all? If an agent behind the Rawslian veil of ignorance would aim to maximise their own expected utility, this is generally thought to result in utilitarianism. But if the view is justified in this way, the resulting form of utilitarianism would be non-aggregative. The theory says, for any individual, ignoring their particular position in society, maximise their expected utility. If that is our aim, then we should make all of the choices that utilitarianism classically implies that we ought to. But not, on this view, because that maximises aggregate utility. Rather, we are doing it because it maximises expected utility for each agent behind a veil of ignorance.

That seems to allow utilitarianism to avoid the seperateness of persons objection. (I considered related issues here.)

[1] If one thinks that this framework commits you to some publicity principle, this should be ammended to be the theory that utilitarianism implies that we should endorse.

Slow posting right now, I find myself a little busy this term.

So it looks like humans can feel happy, feel pleasure, be satisfied, feel enjoyment, and so on. These, I take it, are synonyms for what we tend to call hedonic states. They are the states, whichever they are, that hedonists think we should promote.

But how should we define hedonic states? One view is this: they are to be defined in some purely natural way, as the presence of some chemical, or similar. But that just seems to miss out the fact that being in a state like this feels good. What distinguishes hedonic states from others is that we like them.

So a natural second account is that hedonic states are mental states which we desire to have. But I don't think this can be made to work either, since it's the character of those states that makes us want them, and not vice-versa.

So here's a tentative suggestion: Hedonic states are those which it is good to have. This is a shamelessly normative account, but seems to me to fit the bill pretty well. Hedonic states are just any states which are intrinsically good. The restriction to only intrisic goods is necessary to exclude true beliefs, good dispositions of character, and the like, which I take it may be good, but are not pleasures. On this account, they are ruled out since their goodness is generated by the relationship between the state and something else (namely, the truth and good acts).

Normative accounts of mental states are sometimes appealing (direction of fit accounts of belief and desire sometimes takes this form), and it seems to me to perfectly capture what it is for some state to be a pleasure. Some state is pleasurable if it is a mental state that is intrinsically good.

Objections welcome.

"[T]he slogan "There's probably no God. Now stop worrying and enjoy your life" could now become an ad campaign on London buses".

In fact, since they reached their fundraising target, it will now be running on London buses. They've raised over £36,000 towards the costs of the adverts, which seems to me to be a very large amount of money. Seeing as their target was only £5,500, they've done rather well.

I don't really have much to add on this, though I'd be interested to know how effective such adverts will actually be.

Libertarians (in the sense I am talking about) believe that (a) humans have free will, and (b) that free will is incompatible with physical determinism. So they think that when people act freely, they are acting in some way not determined by the facts before the action.

The classic objection to this is that it looks like what we left with is not freedom, but chaos. Freedom isn't merely the ability to act in ways totally unrelated to your concerns, interests, and what you see reason to do. We want to say that we are at our most free when we act in accordance with those things that we value. The basic thought then, is that libertarianism amounts to the view that there is indeterminacy, indeterminacy is really a matter of chance, and acting randomly isn't freedom at all.

Now consider weakness of will. The problem here, at its most general, is this. Imagine a set of conditions sufficient to explain an action: the action recognises overwhelming reason to do P. Under normal conditions, this can fully explains why an agent P's. But in cases of weakness of will, this overwhelming reason fails to make it the case that the agent P's. The agent, despite herself, does something other than P. So the conditions that were meant to be sufficient to explain the action sometimes do not generate an action. But how can they then be sufficient to generate it?

The obvious thought, once we put these two issues together, is that there's some connection between the two. Libertarians say that no set of conditions is sufficient to explain an action, since agents can always choose otherwise. Weakness of will relies on the thought that no set of conditions is sufficient to explain an action, since agents do sometimes, stupidly, choose otherwise. In fact, it looks a little bit like libertarians are defending the freedom as the possibility of weakness of will, and that weakness of will only makes sense on the libertarian view of freedom. In turn, perhaps we are best off rejecting both.

But I can't imagine I'm the first to draw this connection between these issues. Anyone?

At the end of Smith's collection of papers (Ethics and the A Priori), Smith has a nice little dialogue between "cog" and "noncog", debating the nature of normativity and normative judgements.

During the course of the dialogue, various objections surface against each person's view, and they gradually formulate responses to the objections. Both cog and noncog take this process to be illustrative of something.

Cog sees his ability to respond to noncogs objections to his view as evidence that he must be right. There must be something fundamentally right with it, or else it would have died more easily. Noncog doesn't see it like this. He thinks the reverse. There must be something fundamentally wrong with cogs view, or else these objections wouldn't keep resurfacing.

Of course, the same meta-issue arises over noncog's view. Various objections surface and responses are made, and cog sees this as evidence of noncog's fundamental error, and noncog sees it as evidence of his fundamental success.

There's something to this description of what goes on at a more abstract level in philosophy. There are many objections which I think rely on irrelevant nuances of the presentation of the target. There are other cases in which I think the exact reverse: that the view in question only survives the objection by relying on irrelevant nuances of the presentation of the objection.

Are there any general ways of telling such cases apart?

It's tempting to think that it should be done partly in terms of whether the objection or response is ad hoc. That is to say, whether the resulting view/objection retains some overall sense of unity.

Another important factor will be whether the view or objection is modified, or whether it is replaced, in face of problems. A view that survives objections (or an objection that survives responses) only by giving up its central components to it seems to be one that has died a death of a thousand qualifications. Views and objections are not successful if they have to be abandoned to be defended.

Thoughts?

Can anyone help me with a fairly simple question: What's the relationship between cognitivism and realism?

I understand cognitivism to be the view that the attitudes under discussion (moral, aesthetic, scientific, whatever) are beliefs. We might think of them as representations of the world.

Realism is obviously a messy term, but the key sense is easily enough understood: the truths/facts under discussion (again, moral, aesthetic, scientific, whatever) are real facts, and so independently of what anyone thinks about them.

I can't help but feel it should be easy to state the relationship between these two, but I'm struggling.

At first glance I want to say that agents hold cognitive attitudes about something only if they are realists. But this is obviously wrong, since agents might hold cognitive attitudes without having given any thought to issues of realism at all.

Another suggestion is that agents should hold cognitive attitudes when they are realists. But that claim is really slippery, since it's not clear what sense of "should" is in play here. An epistemic one? And even if we can sort that out, it seems that the relationship between these two is more intimate than that. Mere one-way normative implication doesn't seem to capture just how it is that it seems that these two views should be put together.

(Matters seem worse once we introduce views like fictionalism.)

I want to say that these views are counterparts of one-another, both parts of some larger picture of how some domain works. But I don't see how to fill out that thought in any manner that makes sense. Any suggestions? As I say, I can't help but feel that I'm missing some obvious answer to this.

Sorry for the lack of posting, but I'm a little busy as the term starts. I've finally gotten around to reading Aristotle's Nicomachean Ethics, and found this gem:

"the present enquiry does not aim at theoretical knowledge like the others (for we are inquiring not in order to know what virtue is, but in order to become good, since otherwise our inquiry would have been of no use)..."
(1103b)

I suspect few moral philosophers these days would think of their studies as including psychological facts about how best to make people good. It seems that Aristotle thought of inquiry that excluded such facts as without use.

Just a small note: The problem of evil is the worry that an all-knowing, all-powerful, wholly benevolent God cannot exist if there is evil in the world.

But it's not clear how damaging this worry is. Why can't believers acknowledge that God is not benevolent? This doesn't make God evil, or unjust, or wrong. Unless, that is, we assume some kind of utilitarian view, but I suspect that most believers are opposed to that kind of view anyway.

I take it that the justness of God is the central tenet believers want to retain, not the benevolence. Perhaps that allows them to acknowledge the existence of evil without tension in their view.

I haven't had the chance to work my way through it yet, but the British Academy has just published a report, "Punching our weight: The Humanities and Social Sciences in Public Policy Making".

As I say, I have yet to take a look in any detail, but it may be of interest to some readers. The summary of recommendations is here.

Edit: Forgot to mention the British academy press release:
"Humanities and Social Science disciplines are not 'punching their weight' in contributing to public policy making as they could and should be. [...] 'Our findings show that... there remains considerable scope to improve the effectiveness of public policy making through increased use of humanities and social science research.'"