The Is-Ought Problem

What is the Is-Ought problem? It is usually stated as the
problem of whether it is possible to derive normative
statements from descriptive statements; but to state the
problem at its most general level, it is the problem of
whether ANY moral statement can be literally true, and
hence potentially knowable. It is the problem of whether
there exist any moral FACTS in exactly the same sense as
there exist chemical facts, historical facts, or
mathematical facts. Since libertarianism is a normative
political theory, it is only natural to expect that great
libertarian thinkers would grapple with the Is-Ought
problem; after all, if there are no moral facts to be
known, then any normative theory would be senseless. It
would be akin to a theory about unicorns. Of course even if
the Is-Ought problem were solved, it would hardly establish
any particular moral doctrine; solving the Is-Ought problem
is a necessary condition for libertarian moral theory to be
established, not a sufficient condition. Interestingly,
libertarian political philosophers have spent even more
time on the Is-Ought problem than you would expect. Rand,
among others, popularized the problem. She certainly
stirred my initial interest in the question, but I found
her answer to be quite unsatisfactory. After several years
of thinking about the problem, I now think that I have a
very promising solution which I will presently expound. Now
it is very widely believed that there are only two sources
of knowledge: observation and deductive reasoning. This is
perhaps one of the few premises shared by philosophers as
diverse as Rand and Hume, though naturally they put
different spins on it. Now it is not too hard to show that
IF these are the only two sources of knowledge, then moral
knowledge is impossible. (Of course, just because we are
totally ignorant about something, we could not infer that
the thing did not exist; but, as with astrology, if a field
is shown to have no valid methods, then the validity of the
field itself falls into question.) So why can't observation
yield moral knowledge? Simply put, no matter how long you
look at something, listen to it, smell it, taste it, or
touch it, no moral conclusions arise. That seems fairly
obvious, but it has wide-reaching ramifications. For
suppose that we try to justify a moral conclusion with
deductive reasoning. The problem here is that deductive
reasoning merely shows that IF the premises are true, THEN
the conclusions are true, WITHOUT ESTABLISHING WHETHER OR
NOT THE PREMISES ARE TRUE. Therefore, for a deductive
argument to yield a true conclusion, we must know that the
premises are true, and must therefore have some
non-deductive means of knowing this if we are to avoid an
infinite regress. Normally, this is no problem, since we
can use observation to establish the truth of the premises.
But as we noted at the outset, moral conclusions can't be
reached by observation. But couldn't premises verified
through observation coupled with deductive reasoning yield
a moral conclusion? I answer that they could not. As a
general rule, a deductive argument can only reach a
conclusion within the basic subject matter of the premises.
You can't start with a premise about geometry and wide up
with a conclusion about history; nor can you take an
historical premise and yield a geometrical conclusion.
Deductive reasoning may yield new and interesting results,
but not about a totally distinct field of study than that
of your initial premises. So we seem to be in a quandary;
neither observation nor deductive reasoning can yield moral
knowledge. Fortunately, the quandary is self-created by the
initial premise. If we take the premise seriously, we will
notice that many NON-moral items are knowledge also fall
into question. Take, for three examples, the following
propositions: 1. Every effect has a cause; the same cause
always produces the same effect. 2. The argument ad hominem
is a fallacy. 3. 2+2=4 Notice: all three are non-moral; and
none of them could be known merely through observation or
deductive reasoning. We surely do not observe every effect
and every cause, then conclude that they always come in
pairs. But neither do we deduce the law of cause-and-effect
from another, more basic premise. So too with the logical
principle that the argument ad hominem is a fallacy; it is
not that we learn it by carefully staring at it; but
neither is it the product of a deductive argument. Or to
take the final case, we don't learn tha 2 and 2 must always
make four by observing groupings of 2's (though doing so
might surely help us grasp the principle), nor by deducing
it from anything else. But if we don't learn any of these
propositions by observation or deductive reasoning, how do
we learn them? I answer that the previous account of
knowledge makes a critical sin of omission: it assumes that
deductive, indirect use of reason is the entire faculty. I
say that there is also DIRECT reason, which we may also
call intellect or intuition. We use our direct reason when
we simply turn our intellects to a proposition and think
about it as honestly and critically as we can; and coupled
with sufficient intelligence, SOMETIMES we can immediately
see the proposition under consideration is true or false.
Thus, to validate the law of cause and effect, I turn my
intellect to the proposition and think about it to the best
of my ability; and eventually its truth becomes evident. So
too with the fallacy ad hominem: I think about the fallacy,
turning my intellect directly upon the issue, and see that
it is false. The same goes for 2+2=4. To sum up, the
problem with the theory that all knowledge comes from
either observation or deductive reasoning is that it
ignores the more basic faculty of direct reason; and the
best argument for this faculty of direct reason, besides
the introspective one, is that unless we allow for a
faculty of direct reason, almost everything that we call
knowledge turns out to be unjustified. I'd call that a
reductio ad absurdum if I ever saw one. Now how does this
help solve the problem of moral knowledge? I claim that
SOME moral propositions are learned by means of direct
reason. That is, we simply think about the propositions,
turning our intellects to them as honestly and critically
as we can, and then sometimes we immediately grasp their
truth. For example: Consider the proposition "Murder is
wrong." Turn your intellect to it as honestly and
critically as you are able. I claim that when I carry out
this thought experiment, the wrongness of murder becomes
evident to me. So too with other simple moral propositions.
When I wonder whether racism is wrong, or whether Hitler
was a bad man, when I apply my direct reason to the
problem, the answer is all too clear. Now of course, it
needn't be the case that ALL moral knowledge is direct. In
fact, I could only learn that Hitler was a bad man by the
cooperative use of all of my faculties: 1. Murder is wrong.
(Premise supplied by direct reason.) 2. Hitler was
responsible for many murders. (Premise supplied by
observation of incriminating evidence, testimony, etc.) 3.
Someone who deliberately commits many wrong acts is a bad
person. (Direct reason.) 4. Therefore, Hitler was a bad
man. (Deductive reason) The point is that for the argument
to even get of the ground, direct reason was necessary. It
might be that direct reason supplies only a tiny number of
valid moral principles, from which valid conclusions must
be deduced. My opinion is that the use of direct reason is
more frequent, but that is not the critical part of the
theory. The critical part is the admission that we
SOMETIMES use the faculty of direct reason to come to know
a moral proposition as literally true. -- Now as I said at
the outset, this theory is consistent with any substantive
moral views. Nevertheless, it is peculiarly consonant with
libertarian moral theory. Why? Well, it is a common
observation among libertarians that everyone follows
libertarian principles in his or her private life; it is
only where government is concerned that they grant a moral
sanction to the initiation of force. And if you asked your
average person why it was wrong to commit murders, or rob,
or defraud others, one popular answer would be: "That's
just common sense." Indeed it is; the principle of
non-initiation of force is just common sense; which is to
say, that even the simplest mind, if it honestly and
critically turns itself to the proposition that it is wrong
to use violence against peaceful persons, or rob them of
what they have produced, can immediately grasp its truth.
All that would then be required to establish libertarian
moral theory would be to couple this everyday insight of
direct reason with the premise, derived from observation,
that governments habitually violate the non-initiation of
force principle, and then use deductive reason to draw the
final inference that most, if not all, of what government
does is wrong and must be stopped at once. -- I would like
to thank Michael Huemer of the Rutgers philosophy
department for providing me with many of my key ideas on
this issue.