Critique of "The Intuitionist Ethics"
Gordon G. Sollars
> But unless you think that we have all the theories
> we're ever gonna have at conception, then there must be a way of getting
> *new*
> theories. You mention "variation and selection" below, but I don't suppo
> se you
> think that we, as cognitive beings, are doing those at random; we engage in a
> rational process of varying particulars of a theory, and selecting among the
> options by comparing predictions to observations. Right?
I'm not sure that we have any deep understanding of how we get new
theories - that would seem to require a detailed understanding of
how human imagination and creativity works. To say that we engage in a
"rational process" strikes me as an after-the-fact dressing up what
really goes on, since we "know" that we are rational and that only
rational processes have value, etc. But, just in case any Objectivists
are reading this, I do not mean to imply that the process is "irrational"
either.
...
> The predictions come after the theory, and the (dis)confirming observations
> later
> still.
In a logical, but not necessarily a temporal, sense.
> That there are any number of theories that could have predicted som
> ething
> makes no never mind when it comes to saying whether the prediction *this* t
> heory
> made was confirmed or not. Please don't confuse confirming a prediction with
> confirming a theory.
OK, but confirming a prediction can depend on "knowledge" from other
theories, which in turn might be disconfirmed.
...
> But we typically assume we *have* been careful about error
> s, and
> proceed on that assumption till reality comes along and smacks our faces fo
> r being
> so impertinent.
I am a big fan of the "smack of reality", but I hope you realize that
that "smack" does not speak for itself. It tells us that /something/ is
wrong in the total theoretical superstructure, but it does not come with
instructions about what to fix.
...
> So do you think evolution has provided us with the ethical theories we li
> ve our
> daily lives by? Do you think that all the new ethical theories you will ev
> er need
> will spring from your subconsciousness fully formed? If you answered no to
> both of
> those questions,
Yes to the first; no to the second.
> then (it seems to me) you must believe that we need (somet
> imes) to
> do conscious theory formation in the field of ethics,
You seem to be opposing evolutionary processes with conscious processes.
But consciousness is the result of an evolutionary process. (Or so I
believe - and you?)
...
> > Now, Rawls's theory of justice tells us (predicts) that we will have a
> > just society if certain conditions are met. But, ultimately, these
> > conditions require centralized control of the economy (of course, I could
> > be wrong about that - perhaps Rawls's Difference Principle is satisfied
> > by laissez faire - but that is a different problem). But the Soviet
> > observation disconfirms (it doesn't work) centralized control of the
> > economy. So his theory is false.
>
> But what observations were his theory based on?
I am giving Rawls's theory a realist construal for my own purposes. He
would say that his theory was based upon ethical "considered judgments"
not "observations", such as that persons have a right to freedom of
belief. His theory was specifically designed to account for such
judgments. As a moral realist, I take such judgments to be literally
true (or false), and so to report facts (when true). I take facts to be
the sort of thing that discomfirm theories. If you think that some facts
are too complex (in some sense) to serve as "observations", this might be
the source of some confusion between us.
> > Of course, Rawls can say that there were other factors involved, so that
> > the observation does not really show his theory of justice to be false.
>
> The observation doesn't show even that centralized control doesn't work -
> - it's
> just evidence in that direction. Besides, does Rawls's theory say that a
> just
> society wouldn't ever (or even soon) collapse?
We should be so lucky that we always have observations that provide
conclusive evidence. Perhaps we have such in physics (I doubt it, but
I'll leave that to the experts), but not in ethics. Does that (assumed)
difference defeat moral realism? I don't think so.
Rawls does have a discussion of justice and stability, but it has been
many years since I have read it. Rather than review that, let me ask if
you think that social institutions that survive for more than a single
human generation are ones that "work"? For me, even three to four
generations is not enough to show what works.
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
> [...]I may say that I should grade papers without meaning that I
> am currently motivated to grade papers. I may be averse to the idea
> (I'm feeling tired, or lazy, or only interested in something else).
> It's possible that I have not been considering the possibility, and it's
> only in retrospect that I say I should have been grading papers. What
> *I* am reporting when I say I should be (have been) grading papers is
> that grading papers is something I (would have) approved of in that
> situation.
Sorry, but I don't follow this. What situation? In the situation we
have been given, you (or I) did not grade the papers. Is your
parenthetical "would have" a counterfactual? That wouldn't make any
sense (would it?). Perhaps you are saying that you /in fact/ approve of
grading the papers but did not do it?
I would say you thought you ought to grade the papers, but you did not
feel motivated to do it. Why interpret "thinking that you ought" as a
feeling of approval? Do you think that it must be a feeling so that it
has the power to overcome other, contrary feelings? Or some other
reason?
> Approval is a mildly pleasant feeling (and disapproval a
> mildly unpleasant one -- both stronger when the object of my
> (dis)approval is myself),
Any reason for that greater strength?
> and so it often motivates me to act in those
> ways. But lack of (or conflict in) present motivation does not deny the
> should.
It would seem to make sense to ask if you should have the [feeling of
approval to grade papers]. It that "should" a sort of feeling of
approval as well?
...
> I suspect that the most common intent is a naive realist one
> -- that one is reporting a should-fact. I don't agree that there are
> such facts, and so I would take their intended interpretation to be
> false -- but that does not mean that I take my own should-statements to
> be false, nor that I reject *their* should-statements as worthless.
Indeed, it would seem the simplest course to assume that their "should"
statements are analogous to yours, reporting the same sort of feeling of
approval on their part.
...
> and so what I take to be my basic moral attitudes can be
> interpreted as moral premises by others and only meta-ethical claims
> will be misrepresented thereby.
Right. We could hold different meta-ethical views and yet be in ethical
agreement. That's one reason why I'm not too afraid of discovering that
moral realism is wrong. ;-)
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
...
> Gordon G. Sollars:
> > I'm not sure that we have any deep understanding of how we get new
> > theories - that would seem to require a detailed understanding of
> > how human imagination and creativity works.
>
> theories is a rational process?
Note that I was talking about theories in general, not just ethical
theories; and I did not say it was an irrational process, either.
> I'm quite surprised. I sort of had the
> impression that you felt that reasoning was the sine qua non of morality --
> "good reasons" and all that we went thru last year.
Right. But what does that have to do with the /development/ of a good
theory? Of course a good theory, once developed, will /be/ rational and
people will be able to reason with it.
No other comments??
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
> Mark Young:
> >> So I'm taking it that you *don't* think that the development of new
> >> ethical theories is a rational process?
>
> > Note that I was talking about theories in general, not just ethical
> > theories;
>
> is not rational, but ethical theory development is? If not, then what
> is the import of this response?
It was you who wrote "development of new ethical theories" above, so no,
in fact, my response was to make it clear that I was not making a
distinction between ethical theories and theories in general. If that
was already clear to you, I was being unnecessarily cautious.
> > and I did not say it was an irrational process, either.
>
> Right. So, not rational and not irrational. But you don't care to
> say any more about it.
I'm not sure what else to say. I think that theory development proceeds
by flashes of insight, leaps of intuition, imaginative hunches, etc.,
although the end product is tidied up and put in a logical form. But
perhaps I am wrong; I admit I have not made a serious study of the matter
myself.
...
> > No other comments??
>
> I was hoping for debate, not just comments tossed back and forth. If
> you're still too busy for that, I'll wait.
Well, I am a little busy right now. However, I thought the real
substance (such as it is ;-)) of our debate was over what served as
observations in moral theories, how such theories could be disconfirmed,
etc. You didn't even bother to label that part "comments tossed back and
forth".
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
> Gordon G. Sollars:
> > [...]As a moral realist, I take such judgments to be literally
> > true (or false), and so to report facts (when true). I take facts to be
> > the sort of thing that discomfirm theories. If you think that some facts
> > are too complex (in some sense) to serve as "observations", this might be
> > the source of some confusion between us.
>
> require an observer, but an observation does.
I don't see how this affects my claim that facts are the sort of things
that disconfirm theories. If a fact is not noted by anyone, then of
course no one will be aware that a theory has been disconfirmed. But,
surely, a realist would not want to claim that such theory was correct
simply because the fact had not been noticed.
...
> I distinguish between judgements and observations. One can observe (as
> Rawls does) that people make certain ethical claims, and infer that they
> have made ethical judgements supportive of those claims. But one does
> not observe an ethical judgement, any more than one observes a thought.
My position is that ethical judgments can be true or false. A moral
theory can be proposed that attempts to account for the true ones. That
theory can be falsified, as I claimed Rawls's is falsified. (We are also
arguing, below, over whether my example is a meaningful case of
falsification.)
> I will not rule out something as an observation simply because it relies
> on pre-existing theory -- what I object to are "observations" that need
> the theory they're supposed to be in support of in order to be made --
> circular justification.
People make various ethical judgments all the time without relying on any
particular moral theory. Rawls did not make the judgment that we ought
to have freedom of conscience because he had his theory; rather he took
it as good for his theory that it agreed that we ought to have freedom of
conscience.
> And I didn't get from this response an example of an ethical-content
> observation.
And I guess you won't. Any such thing I give you, you will call an
ethical judgment, not an observation. From the start, I have, I
hope, talked about moral statements and the facts (I think) they report.
These can be likened to the role of observations in a physical theory,
but I don't expect the analogy to hold up perfectly.
> > We should be so lucky that we always have observations that provide
> > conclusive evidence.
>
> That's not my beef. The observations were centrally_controlled(su) and
> fails(su). You said that Rawls's theory entailed
> centrally_controlled(s)
> if just_society(s). You said that the observations disproved the
> theory.
> I say that they're not the right kind of observations to disprove the
> theory, even if we assume that just_society(s) entails ~fails(s). The
> implication is the wrong way round.
Mark, it's too late in the day for me to decode this "(s)" and "(su)"
symbolism, sorry. Please lay this point out in some other way for me.
...
> > Rather than review that, let me ask if
> > you think that social institutions that survive for more than a single
> > human generation are ones that "work"? For me, even three to four
> > generations is not enough to show what works.
>
> I don't know what you mean. If it survives for one generation, then it
> works for one generation. If it survives for three generations, then it
> works for three generations. Do you insist that something has to
> survive
> forever in order for it to "work" at all?
No, or I would have used larger numbers than "three or four". If
implemented, Rawls's political system might "work" for a long time in a
wealthy society. The point is that a centrally-planned economy is not
self sustaining. A snapshot in time is not sufficient to judge a
political system.
> Does a human body not "work"
> because it doesn't last more than four generations?
A human body is not a political system.
> I take it from this paragraph, tho, that you think that a just society
> would be a lasting society. What would your basis be for such a claim
> (if, indeed, you do claim it)?
A just society might not be a lasting society for any number of exogenous
factors. The point is that it should not be self defeating. I think it
would be pointless to consider self-defeating political systems as just.
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
> That situation is an example of the kind of situation where I
> would have said we should have been grading papers, tho. It is a
> situation where I would have approved of paper-grading behaviour, if
> we had been doing any.
Why would you have approved of it?
...
> Let me try again. When I say "You should do X", I mean that I would
> approve of you doing X. There is no occurrent feeling of approval
> implied by the statement -- quite the opposite, since if you were
> doing X it would be quite odd for me to say that you should be doing
> it.
I'm still puzzled. Above I thought you might be saying that you feel
approval toward grading papers (when there are papers to grade). Here
you seem to say that if you were to grade papers you would feel approval.
Or is it both?
...
> > Do you think that it must be a feeling so that it has the power to
> > overcome other, contrary feelings? Or some other reason?
>
> I think it's a feeling because (a) I feel it (that is, the feelings of
> approval and disapproval), and (b) no theory I know of satisfies the
> requirements I have for me to consider it objective. I have spoken of
> those requirements before, so I won't get into them right now.
OK. But, of course, I can also feel approval when I do what I should, so
I don't see (a) as a reason to identify the feeling of approval with the
"should". As to (b), I have thought for some time that we have different
requirements for considering something to be objective. I guess I should
bite the bullet and review your posts on this point in detail.
...
> > Right. We could hold different meta-ethical views and yet be in ethical
> > agreement. That's one reason why I'm not too afraid of discovering that
> > moral realism is wrong. ;-)
>
> I'm not sure what would count as evidence against moral realism, on your
> view.
I grant that that is a reasonable question at this point. Realism in
general seems to be a stance that we adopt to make the gathering of
evidence possible. I'll have to give it some more thought.
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
...
> > A moral
> > theory can be proposed that attempts to account for the true ones.
>
> operationally, I mean. After all, people do make contrary judgements.
When it comes to moral judgments, you push the disagreement, I push the
overlap. People make contrary claims about facts as well. Your
"operationally" suggests that you think that there is some privileged
method of getting at what is true, whereas it seems to me that we must
investigate the truth of statements in a variety of ways. I don't know
of any infallible method of separating true statements from false ones.
I think, e.g., the judgment that it is morally proper for all human
beings to die in agonizing pain at 12:00 A.M. is false. Further, I think
that there is wide-spread agreement on this. I think that you have four
replies to this. One is the skeptical, "If I were to deny it, could you
prove I was wrong?". As I have said before, I don't find that position
to be interesting, if only because you always have that reply at hand for
any number of iterations regardless of what I reply.
Second, (and perhaps this is your position) you could say that it is just
a mistake to say that such a judgment could be true or false. If I had
any direct way to refute this, then moral realism (at least as I
understand it, and here I follow Sayre-McCord - see his introductory
chapter in /Essays on Moral Realism/) would be established - but I don't.
However, I don't think that you have any way to directly support the
claim that it is "just a mistake", either. So what we are left with,
short of some breakthrough, is looking at the general adequacy and
implications of meta-ethical theories that say it is a mistake versus
those that say it is not. That is way I have been asking you about your
"approval" view of shoulds.
Third, it could be that "wide-spread agreement" is all that we have, and
the judgment is true although everyone thinks it is false. Perhaps, for
example, every person did some terrible thing such that all dying in
agony is the proper punishment. Or, perhaps, God wants to start over,
and His will is always the morally proper thing. How could we come to
know that everyone was wrong? Well, by critically analyzing the claims
that everyone did such things or that God wants to start over. Here, I
suppose, you would say that, e.g., there is no "fact of the matter" about
whether there are any such terrible things, so such a claim could not
actually be analyzed for truth. That brings us back to "second".
Fourth, you could point out that the judgment I've claimed is false is
not a very pressing concern, and that there are many other moral
judgments about which there is considerably more disagreement. But that
just tells us that the moral world is more resistant to analysis than
some (most?) aspects of the physical world. Or its evidence for the
second case, if you have a good reason to think that realism implies the
same degree of "analyzability" for all real things.
Did I miss something?
...
> Yes, but Rawls is, according to what you said before, taking the
> ethical judgements as the basis of his ethical theory. What he wants
> (I take it) is a consistent set of axioms that produces the best match
> between its theorems and those judgements. As such, this is a
> modified cultural relativism -- not an objectivism.
Only if the ethical judgments are not true. As I said, I don't think
Rawls is a moral realist, but that doesn't preclude giving his theory a
realist spin. Of course it is possible that all we get is consistency.
...
> It is no contradiction to say that A judges that X while B judges that
> not X.
It should be clear that I don't fully agree with that. A moral judgment
(or an aesthetic one - remember Louis Armstrong!) can be a true or false
statement on my view, so if A and B make such judgments there is a
contradiction. Note I am not claiming that all moral (or aesthetic)
judgments are true or false - some statements (meaningless or self-
referential ones, say) simply can't be judged true or false.
> It is a contradiction to say that A observes that X and B
> observes that not X. You can make a misjudgement,
Which I would say means that you make a false statement (if you express
the misjudgment).
> but you cannot make
> a misobservation (you can have a misperception, but that's not the
> same thing).
So by "observation" you mean "a correct perception"? But the idea of a
/correct/ perception presupposes realism, doesn't it? And that is what I
am doing about moral matters. Observers have perceptions, some of which
we have good reason to take as correct and hence as observations (if I
understand you). People make ethical judgments, some of which we have
good reason to take as correct.
> Hence ethical judgements are not observations.
Having noted my misgivings over this argument (as opposed to this
conclusion), I don't mean to press too hard, as I think I suggested last
time, on these judgments being observations, as opposed to being "like"
observations. They are statements which are true or false, and any true
or false statement can figure in (dis)confirming a theory.
...
> According to you, his theory requires that any just society be
> centrally controlled. Also according to you (I assumed it before, and
> you confirm it below), a just society will be a lasting society (and
> we will assume that Rawls agrees with that, and that it is part of his
> theory as well). To show Rawls' theory false we would need one of the
> following:
>
> (A) a just society that is not centrally controlled;
>
> (B) a just society that does not last; or
>
> (C) that centrally controlled states cannot last.
>
> What we have is
>
> (C') a centrally controlled society that does not last.
>
> This observation counts as evidence for (C), but is far short of what
> one would require to consider (C) justified, and so is far short of
> what is required for a claim of falsification to be justified.
OK, I'll try to get there in two steps. I take (C') as evidence
consistent with Hayek's (and Mises) theoretical critique of centrally
planned economies. If centrally planned economics did not fail for
endogenous reasons, Hayek's view would be falsified. Since I take
Hayek's economic theory as the best available in this context, I claim
Rawls's theory of justice is wrong.
> The observation (C') can be taken as falsifying Rawls's theory only if
> we assume (or can show) that the Soviet Union was a just society under
> Rawls's theory -- in which case it falls into category (B) above.
I don't think so. If Rawls's theory depends on something that can not be
done, it can not be correct. He holds that "justice is the first virtue
of social institutions". If social institutions that follow his model
must collapse, then his model of justice fail to deliver the key virtue.
Now, it is true that I am depending on Hayek's theory. So I admit that
the statement or "observation" that is my counterexample to Rawls is a
rather complex conjunction of other statements. Perhaps I can improve on
that later.
...
> > A just society might not be a lasting society for any number of
> > exogenous factors. The point is that it should not be self
> > defeating. I think it would be pointless to consider self-
> > defeating political systems as just.
>
> ??? Don't you hold that it a matter of fact as to what is just or
> not? And isn't also it a matter of fact as to whether some political
> system is self-defeating or not? Then how can "consideration" come
> into it? What does "pointlessness" have to do with it?
I meant that I could see no reason to call such a system "just". It's
possible to stipulate that "just" includes such systems, but I don't
think that that is how the term is actually used.
--
Gordon Sollars
gsol...@pobox.com
Gordon G. Sollars
...
> Gordon G. Sollars:
> > Why would you have approved of it?
>
> en on
> ourselves. I approve of that.
May I assume that if I were to keep asking this question, you would
always respond with a reason /because/ you approved of that reason?
...
> I identify the should not with the feeling, but with the fact that I woul
> d have
> the feeling. I make this identification because it is the simplest assum
> ption
> that allows consistent treatment of should-claims. Of course you would feel
> approval when you do what you (think you) should -- if you wouldn't appro
> ve of
> doing X, then you wouldn't think you should do X. Of course you would feel
> disapproval when I do what you think I should not -- if you didn't, you would
> not think I should not do it.
But I don't think there any "of course" here. For example, I think that
a person can fail to feel approval for what he thinks he should do. A
father might think that he should not be bigoted, yet find that he does
not feel approval that his daughter should marry a man of another
race or religion. Of course, you could say that he /must/ feel /some/
sort of approval, because that's what it means to say he thinks he should
not be bigoted. But then you are merely asserting your theory.
> We
> have here A if and only if B. Identifying them seems like a reasonable t
> hing to
> do.
>
> (A = M (dis)approves of N doing X. B = M thinks N should (not) do X.)
>
> Now you want to introduce C (= N *should* (not) do X) as a fact that explains
> B. I say fine -- but your elaboration of C also has to explain A -- not
> because
> I have identified A and B, but because A and B are coextensive.
My elaboration of C by itself does not have to explain A; A could follow
from M's training (by, say, his parents or himself) to feel approval
towards what others (including himself) *should* do. Indeed, if we are
lucky, that is exactly how we feel.
...
> To have evidence against moral realism, you must take *something*
> as real, but you must not insist that morality is real.
Right. I have been toying with the idea of coordination games here.
Consider a "theory" for what side of the road to drive on. There are
two, apparently equally effective, yet contradictory ways to do this. So
there appears there can be no "fact" about what is the correct "side of
the road" theory. I am tentatively thinking that if we had evidence that
choosing a moral theory was like choosing a "side of the road" theory,
then morality would not be real.
> This is why I'm
> interested in what you take to be the observations that your moral realism is
> based on. To call something an observation is to make a commitment to the
> reality of what has been observed.
But it seems that you do not take that to be a sufficient condition for
something to be an observation. On the correspondence theory of truth
(which I hold), to call a statement true is to make a commitment to the
reality of some fact described by the statement. So I think that we can
get to realism relying on true statements generally, rather than on what
you mean by observations.
...
> I take colour vision as the paradigm for ethics to emulate.
I'll try to think more about this later; it might indeed be promising.
--
Gordon Sollars
gsol...@pobox.com