Here is my latest reply to Mark Sulkowski's latest criticisms. 
1. Am I myself unclear about precisely what "direct reason" is?  No, I
am unclear about what "garden-variety intuition" is.  The latter term is
used in many ways, usually with negative connotations.  Since it has no
univocal meaning, I do not speak of garden-variety intuition if I can
help it.
2. Is direct reason necessarily/by definition "self-critical and
thoughtful"?  Now that is an interesting question.  The answer is yes.
Thus, if someone just told me whatever they felt in their gut, I
wouldn't call it direct reason; nor would I consider it a valid means of
justifying a belief.  But if someone thought carefully and
self-critically about a proposition and came to a conclusion, I would say
that direct reason is at work.  (This is different from saying that they
are CORRECT, of course.  But they have at least used an
in-principle-valid faculty.  Analogously, if someone solved equations by
coin flipping, I would not call it deduction/indirect reason; but if
someone, correctly or incorrectly, thought about whether one proposition
implied another, I would say that their faculty of indirect reason was
at work.)
3. Proof/explanation/knowledge.  There's a lot going on here, so I'll
try to take each position separately.
  a. Conclusive proof vs. proof within a reasonable doubt.  Well, I'll
be quick to agree that oftentimes, proof within a reasonable doubt is
the best we have.  So far we agree.  But I go further, and say that
sometimes we can know something beyond a reasonable doubt without any
"proof" at all.  That is, without any sort of deductive derivation.
Rather, sometimes we think about something, then see that it must be or
probably is true.  As a paradigm example, take the proposition that the
argument ad hominem is a fallacy.  I don't prove it; rather I see that
it is true on its own merits, directly.
But I do commend you for recognizing the merits of probable proofs.
Indeed, the parallel between deductive/indirect reason and direct reason
extends here; for just as I sometimes know that a deductive argument is
PROBABLY right, so too do I sometimes know that an insight of direct
reason is PROBABLY right.  It's not an all-or-nothing kind of thing.
Maybe the misapprehension that direct reason is supposed to be
infallible is what creates resistance to the notion?
  b. "I place a high degree of confidence on beliefs based on
observations under reasonable conditions of observation."  So do I.  I
also place a high degree of confidence on insights of direct reason with
a high degree of evidence.  What makes observation so superior to the
intellect that we should play down the intellect as much as possible? In
fact, if you re-read your statement, you will notice that even you turn
to the intellect to justify your observations; for what determines what
conditions of observation are "reasonable"?  Surely not observation
itself, for then we would have a circular argument.
Don't exagerrate the importance of this, but I would argue that ONE of
the multitude of functions of direct reason is to guide and inform even
our momentary observations.  Thus, I see a face; I think that it is my
friend Andy.  But then the face is lost in the crowd.  Now what can I do
to determine if I "really saw" Andy?  Normally, I think about the
situation and figure out if it was likely that we would have been there.
The intellect plays a role even in what appears to be a case where only
observation is needed. 
  c. Do we need an explanation other than "it's obvious"?  Well, that
would seem to depend on whether or not the fact is really obvious.  As a
general principle, no explanation will be useful unless the explanation
is more obvious than what is explained; otherwise we increase rather
than reduce the likelihood that we are right.  How can you argue with a
principle like that?  I really don't see how.  Now it might be the case
that there are a small number of super-obvious explanations that explain
a lot of difficult-to-understand facts.  But I doubt it.  It seems much
more likely that there are a lot of easy-to-understand facts, but very
few completely satisfying explanations. 
  Is it obvious that a speeding car will hurt you upon impact?  Well, I
think that we are confusing psychological with epistemological
obviousness, here.  Once you've had a lot of experiences with moving
objects, their dangerousness is "obvious."  But when you think about the
logical justification of the belief, it is actually a product of:
      i. Observations of speeding objects and human bodies
      ii. Knowledge of the law of cause-and-effect.
      iii.  A determination of the appropriate inductive breadth based
on i.  (That is, if you observe a blue Pontiac hit a squirrel, should
you induce that all large heavy objects will injure all living beings;
that all blue objects will hurt all rodents; that all Pontiacs will
injure all mammals? etc.  Observations alone don't determine a
hypothesis, since an infinite number of hypotheses are consistent with
every set of observations.  Direct reason picks among the hypotheses to
try to figure out which is right).
       iv. Deductive reason which puts these all together.

Now it would be nice to have a general rule that "explains" your
conviction that a speeding car will hurt you.  But this only makes sense
if the four-point argument for the general case is easier to make than
the four-point argument for the particular case.

4. Axioms & proof-end-points.  So you admit that we need some axioms.  I
agree.  So how do you think you come to know these axioms, if not by
direct reason?  Remember that you need a non-circular explanation, which
I don't think your view can provide.  Now I used to follow Rand,
thinking that there were a small number of axioms (three, to be precise)
and that everything else had to be justified by these.  And I thought
that we knew an axiom because its opposite entailed a
self-contradiction.  Now I think that that is ONE way to know an axiom,
ONE way that makes a proposition evident to my direct reason.  But there
are many more axioms in the broader sense of judgments which I see to be
true.  Not via proof, but because of the evidence of the propositions
themselves.
Now you state that merely because there is a problem of finding
end-points to proofs, that doesn't show that there is a solution.  In
general you're right; a problem's existence does not imply a solution.
However, I take skepticism to be a self-contradictory view; and I think
I've shown that demanding proof for everything leads to skepticism.  And
since the demand for endless proofs isn't either internally consistent
(since it is unproven), I reject the demand.  I claim that the problem
of finding end-points to proofs is a self-created problem, created by a
misunderstanding of the utility of proofs.  Which is to derive the less
evident from the more evident, not to justify everything.

5. Getting from perceptions to concepts.  I think this whole issue is
more and more of a red herring every day.  Concepts may help you express
what you want to express, but if you described the same facts with
different concepts, the truth of your claims would not change.  I focus
not on "conceptual" knowledge, whatever that is, but on knowledge
acquired with the intellect vs. knowledge acquired with observation.

6. What does this have to do with morality? (My question, not yours.)
Well, I think that direct reason can bridge the is-ought gap, and
nothing else can.  How can you start with descriptive premises and get a
prescriptive conclusion?  You can't unless you smuggle in a moral
premise.  If you deny the existence of direct reason, you're stuck.  I
say that we come to know the moral premise with direct reason.  That
provides a non-circular explanation of moral knowledge that I have never
seen anyone else provide.  And an extremely clear, clean, and simple
explanation, I might add.  Few philosophical exercises are more painful
than reading moral philosophers trying to prove obvious moral truisms
with premises that are much less obvious than the moral truisms
themselves.
                  --Bryan
                        
 
From lsanger@magnus.acs.ohio-state.edu Thu Feb 16 19:51:27 1995
Received: from beauty by ponyexpress.princeton.edu (8.6.9/1.7/newPE)
	id TAA12080; Thu, 16 Feb 1995 19:51:25 -0500
Received: from [128.146.24.116] by beauty (8.6.4/4.940426)
	id TAA29408; Thu, 16 Feb 1995 19:49:58 -0500
Date: Thu, 16 Feb 1995 19:49:58 -0500
Message-Id: <199502170049.TAA29408@beauty>
To: (Recipient list suppressed)
From: lsanger@magnus.acs.ohio-state.edu (ASP-Disc)
Subject: Pure Thought
Status: RO

From: Bryan Douglas Caplan <bdcaplan@phoenix.Princeton.EDU>

What is "pure thought"?  Well, as I use the term it is just a 
synonym for intellectual as opposed to empirical modes of acquiring
knowledge.  But I get ahead of myself.  Roughly, I think that there
are four ways of acquiring knowledge.
1. Extrospective observation (the five senses etc.)
2. Introspective observation (examining one's own mind by introspection)
3. The deductive use of the intellect
4.  The direct use of the intellect.

I believe that #4 is the most controversial.  What I have in mind is
what Mike calls "intuition", and what other thinkers have called
understanding, judgment, and so on.  How does this work?  Well, it
is pretty simple.  Think about about deductive use of the intellect
on e.g. a mathematical proof works.  Typically, you think about the
premises, puzzle over them, and then see that another mathematical
claim follows from them.  Of course, you might prove that step as
well by other intermediate steps; but eventually you just see a step
short enough to be obvious, and make it. Period.

Now the direct use of the intellect is the same; EXCEPT, that rather
than judging whether A follows from B, it judges DIRECTLY whether A
is true.

But aren't all such judgments empirical?  I don't think so.  What is
a "non-empirical judgment" like?  Well, it is a lot like a deductive
judgment, only it considers a claim immediately rather than mediately.
Do you see how a deductive judgment is intellectual rather than
observational?  I claim that a direct intellectual judgment has the
same quality of being intellectual rather than empirical.

I think that if you want examples, I'll just send you my short essay
on the subject.
    --Bryan