Hume's Guillotine

by Julia Galef


I hope Massimo won't start regretting his generous invitation for me to co-blog with him (hi readers! great to be here!) if I kick things off by immediately and publicly disagreeing with him. He and I have been having a debate on moral philosophy for the last few weeks, and after the twentieth iteration of the same arguments we decided it makes sense to invite you all to weigh in, at the very least because we're tired of the sound of our own voices by now. Massimo asked me to lay out the debate, and then he'll follow up with his own post next week.

So, I agree with Massimo that moral reasoning is possible, given a set of initial axioms. (Axioms are the starting assumptions on which all of your moral judgments are based, like the concept of certain fundamental rights, or tit-for-tat justice, or protecting individual liberty, or maximizing total happiness). Where I disagree with him is over his belief that it is possible to use scientific facts to justify selecting one particular set of initial axioms over another.

Roughly speaking, Massimo starts with biological and neuroscientific facts such as "Human welfare requires things like health, freedom, etc." and "Humans are wired to care about each other's welfare," and from these he derives the conclusion, "Therefore, it is moral to act in a way that increases those things which are necessary for human welfare." In my opinion, this is an example of what is sometimes called the naturalistic fallacy: telling me scientific facts doesn't tell me how to act on those facts, and the alleged point of moral principles is to tell me how to act. Science can tell me that if I want to make other people happier, then treating them in certain ways -- giving them health, freedom, and so on -- will accomplish that goal. But science can't tell me whether making other people happier should be my goal.

Alternately, you could use evolutionary biology and neuroscience to argue that being kind to others is the best way to maximize one's own happiness, thanks to the way our brains have become wired over the course of our evolution as social animals. I agree that there's some truth to this claim, but I deny that we can derive any moral principles from it -- it implies only an appeal to self-interest that happens, through lucky circumstances, to have positive consequences for others. (Furthermore, if your moral imperative takes this form, the implication is that if for some reason I were wired differently, then being unkind would not be immoral.)

The difficulty of deriving facts about how people ought to behave from facts about how the world is was most famously articulated by David Hume in his A Treatise of Human Nature (1739):

"In every system of morality, which I have hitherto met with, I have always remark'd, that the author proceeds for some time in the ordinary ways of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when all of a sudden I am surpriz'd to find, that instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible; but is however, of the last consequence. For as this ought, or ought not, expresses some new relation or affirmation, 'tis necessary that it shou'd be observ'd and explain'd; and at the same time that a reason should be given; for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it."

This is called the "is-ought problem", or sometimes "Hume's Guillotine" (because it severs any connection between "is"- and "ought"-statements). My understanding is that Hume is generally believed to have meant not just that people jump from "is to "ought" without sufficient justification, but that such a jump is in fact logically impossible. There have been a number of attempts to make that jump (here's a famous one by John Searle), though I've found them pretty weak, as have other people with much more philosophical expertise than me.


With that in mind, I can't see any way in which a claim of the kind Massimo is making -- "doing X increases human welfare, therefore X is the moral thing to do" -- could logically hold, unless you're simply defining the word "moral" to mean "that which increases human welfare," in which case the statement is tautologically true. But I'm not sure what we gain by simply inventing a new word for a concept that already exists.

Fortunately, even though I think the blade of Hume's guillotine is inescapably sharp in the philosophical world, I don't think it has the power to sever much in the real world. Because, thanks to some combination of evolutionary biology and social conditioning, I do enjoy being kind, and I do want to reduce other people's suffering -- and I would want to do those things even without a rational justification for why that's "moral." And I believe most people would feel the same way.

But if someone didn't care about other people's welfare, I couldn't accuse him of irrationality. He would be committing no fallacy in his reasoning, nor would he be acting against any of his own preferences. (If he wanted to increase human welfare and yet he knowingly acted in a way that reduced human welfare, then I could legitimately call him irrational.)

Massimo, I believe I've represented our disagreement accurately, but please correct me if I haven't! *thwack* Ball's in your court!