From: Robin Hanson (rhanson@gmu.edu)
Date: Thu Mar 02 2000 - 12:03:48 MST
At 11:18 AM 3/2/2000 -0500, you wrote:
> > It seems you have granted that ... Everything we can know about ethics
> > *is* in principle learnable via science. You might claim that progress
> > in this science will be terribly slow, or that it will stop at some
> > limit to scientific progress. ...
>
>No, I haven't contradicted myself. The hypothetical you offered me was
>that, through some scientific process, we might predict what I would think
>about ethics at the end of inquiry. Surely, then, I can't know more about
>ethics than I would know at the end of inquiry, so, you argue, everything
>I can know about ethics is an empirical matter.
>
>However, if no scientific process can DO that, then everything I know
>about ethics is NOT an empirical matter, since we can't actually run the
>experiment. So if you agree with me that we'll never have a process like
>this, then you'll be forced to agree with me that ethics cannot be done
>purely empirically.
To be more precise, such a science might predict your beliefs at any stage
of a cognitive process; I see no reason to focus on some mythical "end."
And let's not get distracted by the straw-person of perfect prediction.
It could be plenty useful if predictions were just much better than random.
And let's keep our eye on the key claim at issue: can one study ethics
via scientific methods? You complained that evolutionary psychology folks
were pretending to study ethics, while instead ethics is just a different
world; no factual statements are relevant to judging ethical statements.
I am saying that ethics is part of science, just as other cognitive and
social sciences are. You *can* learn about what is right by empirical
study.
>I countered with TWO objections, the main one being the first, that, a
>priori, a prediction like that is impossible. (I'd primarily hinted at
>this argument, since you seemed to agree with me that it couldn't be
>done.) To predict what I would think about ethics at the end of inquiry,
>the process would have to predict what I would come to believe on the
>basis of the answer that it would give me; in other words, it would have
>to predict the result of its own computation. A priori, this can't
>happen.
The standard of a science of ethics isn't the ability of it to predict
some mythical "end" of ethical inquiry, any more than the standard of
solar physics is to predict the final end of the sun, billions of years
after humans may have vastly changed the galaxy.
You can't say that science is irrelevant if it can't predict
perfectly. And it is far from obvious that ethical opinion is some
maximally opaque computation which can only be predicted by running it.
Certainly many other things that brains do can be predicted, even though
in principle brains could make it very hard to predict them, if they
tried hard at it.
>And, unfortunately, getting close just doesn't cut it in this case, as it
>might in sociology, economics, or general psychology. Knowing a little
>about human interaction is useful, but I already know A LITTLE about what
>I think about ethics (and what I'm going to think). Knowing somewhat
>more, while potentially helpful, still won't render ethics a purely
>empirical matter unless it can tell me EVERYTHING I'll think about ethics,
>which it can't, thanks to the problems I describe above.
You're trying to make this a choice of extremes: either science must
predict everything exactly, and do it soon damn it, or the subject (ethics)
is just not a scientific subject; it is fundamentally different.
This same attitude would say "scientists shouldn't pretend to be
understanding love; there's no way science could exactly predict who I
would love at the end of the universe, since I could work hard to
make that unpredictable. Thus love and science are just two different
things."
Robin Hanson rhanson@gmu.edu http://hanson.gmu.edu
Asst. Prof. Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030
703-993-2326 FAX: 703-993-2323
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:27:08 MST