Tuesday, December 13, 2011

Hume and reasons

Simon Blackburn was the first person to teach me philosophy when I was an undergraduate, and I'm sure he has influenced me more than I realize. He was a brilliant teacher, managing to make Hume seem all at once very clear, very controversial, and indisputably correct. He also helped me enormously when I was applying to graduate schools, going far beyond what I might reasonably have asked of him. It's good to remember what I owe him. But I can't resist quibbling with some of what he writes in his defense of  Hume in The Stone. According to Blackburn, Hume is misunderstood by his critics:
The most visible example of this is the rumpus surrounding the famous passage in which Hume declares that reason by itself is inert, and has no other office than to serve and obey the passions. The mountains of commentary this has excited include accusations that Hume is a skeptic about practical reasoning (whatever that might mean); that he is a nihilist who cannot have any values; that in his eyes nothing matters; that he is too stupid to realize that learning that a glass contains benzene instead of gin might extinguish your desire to drink from it; that he constantly forgets his own theory; and indeed, in the words of one contemporary writer — the frothing and foaming and insolence here reach a crescendo — that philosophers like Hume only avoid being “radically defective specimens of humanity” by constantly forgetting and then contradicting their own views. 
The contemporary writer in question is Allen Wood. Here Blackburn quotes Wood at greater length, writing that, as Wood sees it:
Humeans must say that there are no reasons for anything—nothing matters. They are rank nihilists! Nicely illustrating how to combine poverty of imagination with vulgarity of tone, one of the commentators included here, Allen Wood, describes them as ‘either radically defective specimens of humanity who are incapable of feeling respect for anyone or anything, or else every time they do feel it they commit themselves to contradicting their own metaethical theories’. Golly.
It seems to me that Wood is saying that the Humean theory would only be true if people were much worse than they actually are. Whether that's true or false, I don't think that saying it is so vulgar, insolent, frothing, or foaming.

Reasons are tricky things because they can be thought of as motives or as justifications. Hume's claim that reason is the slave of the passions covers both these ideas: reason does not move us to action without some accompanying emotion, and reason ought not to move us unless in the service of some (presumably good) emotion, according to him. The 'ought' that follows the 'is' (in "Reason is, and ought only to be, the slave of the passions") is a little odd, but not disastrously so. If reason cannot be anything else then why say that it ought not to be anything else? If it could be something else then how can Hume be so sure that it isn't? I assume he means that it isn't and we shouldn't pretend, or try to act, otherwise.

It depends what you mean by 'reason' though, surely. Hume seems to mean something like 'the calculative faculty,' and I suppose that is a slave of the passions. But if by 'reason' we mean something closer to sanity, then reason is not such a slave. It isn't rational in that sense of 'reason' to prefer the destruction of the world to the scratching of one's finger.

Schopenhauer says that all actions are the result of the agent's character combined with a motive, where I think the motive is taken to be something like a carrot and a relevant feature of one's character might be loving carrots. The carrot alone does not cause my movement towards it, but then neither does reference to my love of carrots unless we know (that I believe) there is a carrot there. This picture seems all right as far as it goes, but it seems to rely on the fact/value distinction perhaps a little too much. If I avoid pain do we need to add that my character happens to have aversion to pain as one of its features? Or if I do the right thing do we need to add that I happen to be the kind of person who values rightness in order to make sense of my behavior?

It's tempting to say that we only don't need to because it goes without saying. But I wonder. Can we really understand something's being painful without understanding it as to-be-avoided? Or something's being right without understanding it as desirable? If something is desirable other things being equal, we don't need to add "and other things are equal in this case." It is analytic that the painful is to be avoided (other things being equal) and the right (if there is such a thing) is to be done, isn't it? Which I suppose is just a way of saying that some concepts are normatively loaded. "Because there were carrots in it" is not an intelligible reason for running into a burning building, unless I know that the runner really loves carrots. But "Because my daughter was in it" is a perfectly intelligible reason. And not because we add an unspoken "And people love their daughters."

If, as Blackburn says, "reasons have become the Holy Grail of contemporary philosophy," then it seems that philosophers would do well to pay more attention to grammar, to the definition and use of words like 'reason'. But I suppose I would say that.

[UPDATE: when I first conceived this post I thought I had something to say. Now I'm less sure that that's the case, especially given how vast the literature on this subject is and how little I even try to say about it here. But my sense is that a lot of time has been spent arguing about what reasons really are and whether reason is really the slave of the passions, and so on. If the answer to these questions is "It depends what you mean by 'reason'," or "In this sense/these cases Yes, but in that sense/those cases No," then much of this energy might have been wasted.]

9 comments:

  1. But I suppose I would say that.

    Yes, you would...and you would be right!

    ReplyDelete
  2. Thanks! I hope so. (But then I would, ... etc.)

    ReplyDelete
  3. "'Because there were carrots in it' is not an intelligible reason for running into a burning building, unless I know that the runner really loves carrots. But 'Because my daughter was in it' is a perfectly intelligible reason. And not because we add an unspoken 'And people love their daughters.'"

    I'm a little trepidatious of arguing against the idea that one's daughter being in a burning building is an intelligible reason to run into it - it's a rhetorically weak position to take - but without some unspoken reason added like "people love their daughters," I genuinely do not know how the physical facts alone provide the reason. My love for my daughter is likely so ingrained that the facts and my love for my daughter don't seem to need separating in the moment, but when I reflect on it, I notice that the physical facts are different from my love for my daughter.

    And, when you add the parenthetical "other things equal" this seems to open the door for the gap-filling work the passions (or values) do, according to Blackburn (and Hume).

    If a child in the 18th Century needed surgery to survive, but the surgery would be extremely painful, this seems to show that pain itself does not come packed with the meaning "to be avoided." I know this example is not new to you, so I'm not using it as a knock down argument. But, I'm not sure I understand something being analytic "other things equal." (I'm open to other examples)

    And the fact that the painful is to be avoided seems analytic, to me, is because we add our passions to the situation. At least, the facts don't seem to mean, "run into the building and save your daughter," not upon reflection, even if that's the it seems in the moment I run into the building.

    ... I'm not sure if I've done anything other than present the opposite opinion from you, rather than provide a reasoned counter argument... It's tough to get one's bearings on this terrain.

    ReplyDelete
  4. When I wrote

    "And the fact that the painful is to be avoided seems analytic, to me, is because we add our passions to the situation. At least, the facts don't seem to mean, 'run into the building and save your daughter,' not upon reflection, even if that's the it seems in the moment I run into the building."

    I should have clarified that the impression of obviousness of something like "pain is to be avoided" is that our passions are so ingrained into the way we frequently use the terms that it seems analytic. But the example of necessary surgeries before anesthetic seems to show that the meanings are not tied as tightly as we normally assume. "Pain" is a sensation. Whether it is to be avoided depends, on a lot of things, in my humble opinion.

    ReplyDelete
  5. without some unspoken reason added like "people love their daughters," I genuinely do not know how the physical facts alone provide the reason.

    I'm a little trepidatious of arguing against Hume and Blackburn, and I'm not sure that I have an argument as such, but here's what I want to say: the physical facts are not the point. A sentence such as "The man ran into the burning building to save his daughter" belongs to a language, or language game, that is not that of physics or disinterested science, at least as I am thinking of it. "Man" here does not mean "male member of the species homo sapiens," and "daughter" does not mean "female offspring of the male." If we extract the implications of words like "man" and "daughter" from the sentence, then , yes, we have to add back in something like reference to passions in order to understand the situation it describes. But, if I can use Wittgensteinian cliche for a while, in the language game that is its home no such extraction or addition is either necessary or useful.

    My use of the word "analytic" might have been ill-advised, because I don't mean that "This will hurt, therefore I should not do it" or "My daughter is in danger, therefore I must save her" are necessarily good arguments. It's not like "If Socrates is a bachelor then he is unmarried." But I do think that there is more to concepts like man and daughter than might be mentioned in a minimal or technical definition of these terms, and part of this something more is normative. Not in the sense that if x is a man then y, but in something like the sense: if x is a man then, other things being equal (or: normally), y.

    I don't think it's that we merge our passions into our thoughts about our daughters and pain in a way that causes confusion about the meaning of "daughter" and "pain" unless we carefully separate the passions from the facts, in other words. Pain is a good example, because the definition of it involves reference to its being an unpleasant sensation. And I don't know how to make sense of unpleasantness without some reference to avoidance. So I do think that analytically pain is to be avoided. The kind of example you bring up involving necessary painful surgery shows why I add the "other things being equal" caveat.

    I'm not sure, as I say, that any of this amounts to an argument, but I hope it makes my position a little clearer.

    ReplyDelete
  6. I really appreciate you taking the time to explain your view. It's very refreshing to discuss (what seems like) a disagreement with someone on a blog and feel improved by it, rather than frustrated, which is my typical experience.

    We've both stated our views without claiming to posses a knock-down argument, so perhaps I can try a different tack:

    *If language picks out preexisting things in the world ready for the picking, like labels, then it seems like we have to add our passions in order to make "he ran into the burning building because his daughter was in it" intelligible. Surely in everyday discussion it is intelligible without making love for daughter explicit, but that doesn't seem to settle anything between us, since it could be that this habit is just very ingrained, or on the other hand that it's normatively built into language from the beginning. If language, on the other hand, doesn't merely pick out things in the world, but plays a more (or perhaps even exclusively) social function, then it seems likely that Hume's observations here are feeble or... peevish or something like it. Thoughts?

    *It's becoming fashionable for hard-nosed rationalists (for lack of a better term) like Sam Harris to deny the fact/value distinction. But doesn't one need a somewhat Wittgensteinian view of language in order to make that claim? That is to say, while exalting science for picking out true things and condemning all forms of post-modernism and relativism, isn't it a bit odd to turn around and meld facts and values the way Harris does? It seems like you can't go half scientism and half Wittengenstein, in other words; you have to choose, right?

    Maybe I'm being vague, but I'm interested in circling the argument to find my bearings. Maybe after I'll just reflect. Thanks again for your time.

    ReplyDelete
  7. Internet philosophers are supposed to be humorously cranky jerks or plain-speakers who don't suffer fools gladly (i.e. not very nice), but I'm not very good at playing that role.

    If language, on the other hand, doesn't merely pick out things in the world, but plays a more (or perhaps even exclusively) social function, then it seems likely that Hume's observations here are feeble or... peevish or something like it.

    Yes, I think that language does play a social function and does not merely pick out things in the world. Whether this makes Hume's observations feeble or peevish, I'm not sure, but feeble might be close. We learn language in the course of living, interacting with other people, trying to get what we want, and so on. The biological definition of 'mother' is not wrong, and there are engineering meanings of 'male' and 'female' too, but not everything that we might reasonably call the meaning of words like 'mother' is included in this kind of passion-free sense of 'meaning.' And the scientific sense of words is not the primary one but an abstraction from the primary, 'human' sense. Or so it seems to me. (I hope this isn't too obscure.)

    It seems like you can't go half scientism and half Wittengenstein, in other words; you have to choose, right?

    This certainly sounds right to me. I'm not sure, though, what you have in mind about Harris, because I don't know his work well enough. Do you mean his idea that science can answer ethical questions? I think you do. If we want empirical ethics we should, I think, turn to J.S. Mill rather than Sam Harris. And then we should read Frege on Mill's empiricist philosophy of mathematics. And then read Wittgenstein. But you don't need Wittgenstein or Frege to see how crazy Harris is.

    But that's not very nice of me to say.

    ReplyDelete
  8. Isn't suggesting that Harris is "crazy" overstating things a bit? He's wrong, as are most people who, being of a naturalist/empiricist bent, look at utilitarianism, see numbers, and think, "well, the theory involves quantification, so it must be right..." I guess this is crazy (maybe Hard Times should be more standardized reading?), but not uncommon...

    ReplyDelete
  9. You're right, Harris's kind of view is not uncommon. But I'm still inclined to think of it as crazy in a Brave New World or Hard Times kind of way. Perhaps wrongly, I tend to think that Mill found out the hard way that Bentham's kind of ethics is literally incompatible with sanity. And I think (without having read much of his work) that Harris is not as good a philosopher as Bentham.

    ReplyDelete