Site Loader
Rock Street, San Francisco

As humans, we are
hardwired to use our knowledge of the past to construct our expectations for
the future. When making predictions, we oftentimes—both consciously and
consciously—take into account previous experiences. We use our past knowledge
to predict a wide array of topics, from weather patterns to sports. This
practice is especially prevalent in various fields of science—medicine, physics,
chemistry, etc. What arises from this idea of basing future predictions upon
previous experiences is a complication referred to as the traditional problem of induction.

problem of induction brings forth the question of whether we are rationally
justified in utilizing a system of scientific inductive logic rather than
another system of inductive logic, such as astrology, to make predictions about
the future. Skyrms (2000) defines inductive logic as that which “is used to
shape our expectations of that which is unknown on the basis of those facts
that are already known.” Inductive logic contains epistemic probabilities,
which are based on the knowledge that we currently hold, along with the degree
of strength of the argument brought forth. We say that an argument with a high
epistemic probability is likely to occur in most cases and that which holds a
low epistemic probability is not likely to occur in most cases.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

first to raise the problem of induction was Scottish philosopher David Hume,
who states that there is no rational justification of inductive logic. His
argument is built on the premises that for an argument to yield a high
inductive probability, either it must be a valid, deductive argument, or a
strong inductive argument, yet neither is able to do the job.

the case of deduction, Hume is correct in that a deductive argument cannot justify
reasoning by induction because it is simply impossible to create one that can
do so. A deductive argument must be valid—that is to contain true premises
(which may only deal with knowledge of the past) that lead to a guaranteed
conclusion, and must be non-ampliative, meaning that the extent of the information
specified in the conclusion cannot reach outside of the premises. Simply put,
the conclusion cannot make any claim not contained in the premises. Because we
do not know if the future will be like the past, we are unable to use a deductive
argument to justify induction. If the conclusion were to make a prediction
about the future, then this would contradict the necessity of validity and
non-ampliativity. If we managed to create a deductive argument that had a high
epistemic probability, this would imply that a true conclusion would stem from
the premises most of the time, therefore a true conclusion is not guaranteed
one hundred percent of the time. Also, because the premises may only contain
information about the past or present, a conclusion that makes a prediction
about the future would give an ampliative argument.

also argues that an inductive argument is unable to justify induction because
this leads to a question begging, circular argument. As Skyrms states, “…if we
attempt to rationally justify scientific induction by use of an inductively
strong argument, we are in the position of having to assume that scientific induction is reliable in order to prove that
scientific induction is reliable…” In this case, we are assuming the truth of
that which we are attempting to prove. When we state that, “because scientific induction
has worked in the past, scientific induction will stand to work in the future,”
we are using our conclusion to support our premises and vice-versa.

underlying basis of scientific induction is the principle of uniformity of
nature, which states that the future will be like the past. If something
happens time and time again in the past, the principle of uniformity of nature
maintains that it will continue to occur in the future. What the traditional
problem of induction then boils down to is whether nature is in fact uniform. Hume
would say no, reiterating his stance that inductive arguments—which are based
on the principle—are circular and beg the question. However, many would believe
that nature is uniform. After all, the more an event has occurred in the past,
the more likely it is to occur in the future, right?

brings us to the point of counterinduction, a system of inductive logic that is
directly opposed to the principle of uniformity of nature. Whereas the
uniformity of nature states that the future will mimic the past, counterinduction
holds that the future will not resemble the past. A prediction that one might
maintain as highly probable in terms of scientific induction would be
considered highly improbable to a counterinductivist, and vice-versa. An
example of each argument would be as follows:


All F’s have been observed to be G’s

The next F will observed to be G’s


All F’s have been observed to be G’s

The next F will not be observed to be G

An everyday example of
counterinduction is known as the “gambler’s fallacy;” the more an event happens
in the past, the more one believes that it is less likely to occur in the
future. The system of counterinduction supports Hume’s claim that we are not
justified in believing the future to be like the past.

are significant rebuttals to Hume’s argument that inductive logic can be
rationally justified. In the case of the inductive justification of induction,
one may state that the argument, in fact, not circular. Whereas Hume states that
scientific induction is circular because we are assuming the truth of what we
are trying to prove, a supporter of scientific induction may say that we are
indeed justified in using inductive logic by using a system of levels. At each
level, there is a rule that gives an inductive probability. As the number of
levels increases, and the rules of inductive probability come to be true at
each level, we are gradually more justified in our predictions. For instance,
if we observe 99 apples and each has been observed to be green, we can predict
the 100th to be green as well, and we are justified in doing so
because the rules at level 99, 98, 97, etc. have all been correct thus far.

solution is referred to as the pragmatic
justification of induction, as offered by Hans Reichenbach. He attempts to
solve the traditional problem of induction with a deductive argument and holds
that although we may not be able to prove beforehand that an inductive argument
will be true, scientific induction is just as useful as any other method of
induction. We cannot make accurate predictions if we make none at all. His
argument is as follows:

                        Nature is either uniform or not

                        If nature is uniform, scientific
induction will succeed.

                        If nature is not uniform, then no
method will succeed.

                        If any method of induction succeeds,
then scientific induction will succeed.

Although Reichenbach does create a
valid, deductive argument, he fails to come up with any solution to the
discussion of the principle of uniformity of nature.

In summary, there
exists a conflict on the topic of the traditional problem of induction. There
are some, such David Hume, who believe that we cannot rationally justify induction
either deductively or inductively. Then there are others, like Reichenbach, who
believe that we induction is justified. There advantages to each argument. Hume
is correct in that we are unable to predict whether the future will be like the
past. However, I agree with Reichenbach in that scientific induction has been
successful in the past and we should continue to utilize the system until it
fails—which it may or may not. Regardless of the case, science cannot afford to
cease making predictions simply because we do not hold the knowledge regarding
nature’s uniformity.

Post Author: admin


I'm Dora!

Would you like to get a custom essay? How about receiving a customized one?

Check it out