Jonathon D. Jones
PHL 320
February 21, 2002
Max Black and Humean
Skepticism
In
this essay I will argue that the Humean problem of induction is only truly
problematic when a strange, impossible definition is given to the term
“reasonable”. I will begin by
explaining what it is I understand Hume’s induction problem to be, and to try
to flesh out the issues relevant to my case.
I will then examine Max Black’s proposed solution to the problem, and
show in what ways this solution is useful and why it is ultimately
unconvincing. In this latter context I
will invoke the work of Wesley Salmon, and then try to solve the problem that
Salmon poses.
Hume’s
problem of induction is that inductive reasoning is not, in fact,
reasonable. That is, we are not justified
in reasoning inductively. This is
because he believes that, in order to justify induction, we must use some form
of the Uniformity Principle. This
Uniformity Principle (henceforth noted as UP) states “[t]hat instances, of
which we have had no experience, must resemble those, of which we have had
experience, and that the course of nature continues always uniformly the same”
(Hume 89). He also believes that “we
must provide one of two types of justification for UP: (a) Show that UP is the conclusion of a
deductive argument, or (b) show that UP is based on experience” (Crumley
15). He shows that it is not possible
to prove this principle deductively because of problems of circularity, and
that to show that it is based on experience is to be similarly circular. That is, providing evidence for something
and using this as a justification for a believe is precisely what induction is
all about, and so one ends up justifying induction through induction. (Crumley 14-16)
The
first method of justification, deduction, isn’t particularly useful to look
at. While it is true that, if induction
were deductively sound, we would have good reason for believing it (at least,
as good a reason as we do for anything else deductively sound), this would mean
that induction is, at base, merely a part of deductive reasoning, rather than a
separate entity. As such, it is neither
surprising nor particularly harmful to be unable to validate induction
deductively.
The
second method, however, is precisely the issue that Max Black addresses. By eliminating the first we conclude that
the only justification for induction is induction, and so the issue at hand
moves from the justification of induction to whether this justification is
acceptable. Max Black believes that it
is, or that at least it is unreasonable to expect further justification.
Black
spends a great deal of time proving induction inductively, and trying to do so
in a convincing matter. While this
isn’t particularly interesting, perhaps, it does at least show that induction
is not internally inconsistent (which would provide a very nice reason to
discard it all together). As a method
of showing induction to be reasonable, however, Black’s own words should
suffice: “you cannot persuade somebody who distrusts and rejects all inductive
argument by means of an inductive argument” (Black 40).
One
way he shows this is to compare induction to deduction. That is, if attention is turned from
induction to deduction, the same problem is found – that deduction can only be
shown deductively. This seems to be no
more or less circular than justifying induction inductively. He says that “in this respect inductive
argument is in no case worse than deductive argument. There are limits to what can be shown by means of deductive argument
concerning the validity of deductive argument; and it would be unreasonable to
demand more of inductive argument” (Black 40).
This
argument doesn’t quite prove the skeptic wrong, however. By itself, it really doesn’t give the
skeptic any reason to believe that humans are, in fact, reasonable in using
induction. However, it does bind the
two types of reasoning together in such a way that a new argument can be made,
but which Black doesn’t actually make.
What he has shown is that it is no more reasonable to trust in deduction
than it is to trust in induction. That
is, if the skeptic truly believes it unreasonable to use induction, he must
also discard deduction. But this seems
odd, for this very reason: that deduction and induction are the only forms of
reason that we have. For something to
be “reasonable” or “rational” means precisely that there is some reason for
believing it, and the only tools available for justifying these reasons are
deduction and induction. And so to say
that humans are unreasonable because we use induction and deduction is either
to say that we are unreasonable because we are reasonable, or to give some
meaning to “reasonable” that is beyond the realm of induction or
deduction. The first option simply
doesn’t make sense, and the second depends on changing the requirements of
“reasonable” to something that is not what anyone thinks of as
“reasonable”. Instead, a new term,
“quesanable”, could be attributed to this higher realm of justification. At base, however, no one is concerned with
“quesanbleness”; the question at hand is whether we are reasonable.
Wesley
Salmon, however, does seem to deal with the question of “quesanableness”,
although he doesn’t do so directly. He
says, “I think it is to be presumed
that most readers can recognize some instances of inductive correctness, and
when we ask for a justification of induction we do not presume otherwise. We merely ask for the grounds of such
recognition. Black seems to feel this
request is improper. He has failed to
show, however, why it is improper, beyond showing that it is difficult to
fulfill” (Salmon 47). That is, he
agrees with Black that induction is correct a good deal of the time, and that
most people recognize that. What he
seems to ask, though, is what standards we are using for recognizing that
induction is a correct method to use.
And he correctly points out that Black doesn’t deal directly with why we
ought not to pursue this line of questioning.
This
is precisely where the queasonable issue becomes pertinent. The question Salmon seems to be asking is
whether we should be seeking quesonableness, and if not, then why not. My answer given so far, that quesonableness
is simply not something that is normally sought, would perhaps be less than
satisfying. If, for instance, no one
aspired to be good according to some moral principle would not mean that there
is no reason for them to do so. Also,
the fact that no one thinks about quesonableness is no more reason to fail to
seek it; mere ignorance of something does not lessen the quality of it. Again, the mere fact that I do not know all
the principles of morality in the Hindu religion does not mean those principles
are not worthy ones to seek.
However,
there is, I think, a good reason not to worry about quesonableness, and this is
simply that the problem would not be solved, but simply expanded. That we have a reason for deductive and
inductive thinking would validate those, perhaps, but then we would have to
begin anew in determining a reason for quesonable thinking. And assuming we could find non-circular
reasons for believing queason, those reasons would also be subject to the same
type of inquiry, to infinite regress.
It
seems, then, that if one were to say that trusting in induction is
unreasonable, they would have to essentially say that it is impossible to be
reasonable. The requirement that a
system of reason must be non-circular in order to be justified is an impossible
requirement to fulfill, as I have shown.
But if it is impossible to be reasonable, there seems to be no point in
attempting to be so. In fact, again,
the person seems to be using a new definition of “reasonable”, specifically
“using a system of reason that is non-circular”. And I could assign this another silly term, like leasonable, but
I think the point is clear enough.
Essentially, I believe this is simply too harsh a requirement for
reasonable, as such should be discarded.
And if it is, our original, normal definition of “reasonable” is used,
simply to use induction or deduction.
It is fairly evident, I think, that we are at least this kind of
reasonable, and since this is the definition used, I believe we are, in fact,
reasonable creatures.
What
I have shown, then, is that Hume’s problem of induction is only really problematic
when an unpalatable and unattainable definition of “reasonable” is used. I have shown that Black provides a good
start to the problem, but that his solution is ultimately unconvincing to
skeptics of induction. And I’ve
attempted to address the problem that Salmon brings up; that is, I’ve attempted
to show that it is improper and non-valuable to try to provide reasons for
induction. My conclusion, then, is that
as long as being reasonable is something that is possible to be, humans are, in
fact, reasonable.
1.
Black,
Max. Caveats and Critiques. Ithaca and London: Cornell University Press,
1975.
2.
Crumley,
Jack S II. An Introduction to
Epistemology. Mountain View,
California: Mayfield, 1999.
3.
Salmon,
Wesley. “Should We Attempt to Justify
Induction?” Philosophical Studies
8 (April 1957): 33-48.