Normally we classify all arguments
into one of two types: deductive
and inductive.
Deductive arguments are those meant to work because of their pattern
alone, so that if the premises are true the conclusion could not be
false. All other arguments are considered to be inductive (or
just non-deductive), and these are meant to work because of the actual
information in the premises so that if the premises are true the
conclusion is not likely to be false. The difference is between
certainty (we can be sure the conclusion is correct) and probability
(we can bet on the conclusion being correct).
We now go one step further. A
deductive argument with the right form is considered to be valid, regardless of the truth
of the premises. When the premises are in fact true and the
argument is valid, then we call it sound.
Inductive arguments can be seen as strong (the conclusion is more
likely to be true because of support provided by the premises) or as weak. When an
inductively strong argument does have true premises, we call it cogent.
How
strong does an argument have to be to be acceptable? A good rule
to start with is that the more is at risk, the more likely you want the
conclusion to be correct. For instance, in a civil case (the kind
that occurs when one person sues another) a jury is asked to decide
between two sides based simply on the preponderance of the evidence,
and typically there can be a split decision among the jurors.
However, in a criminal case there is obviously more at stake (it could
be a person's freedom or possibly his life), and so the jury is asked
to decide unanimously on the basis of there not being a reasonable
doubt about their verdict. In everyday life, you would expect a
stronger argument about where to transfer for the last two years of
college than you would about what movie to see next weekend.
All
arguments then can be classified as valid or invalid. If valid,
they are sound or unsound. If invalid, they are strong or weak
and then, depending on the premises, cogent or not cogent.Note that a strong argument by
definition cannot be valid, and a valid argument by definition cannot
be strong.
Some
additional notes: an argument that misuses a form (what we will
call a formal fallacy)
may not be valid but then we need to look at it
in terms of inductive strength. Also, an argument may be
technically
sound (valid with acceptable premises) but still not a "good" argument
because of some informal fallacy
(another kind of mistake in the
reasoning but one not related to the pattern). Most typically
this could be a
problem of what we call begging the question, when the premises would
be acceptable only if someone already accepted the conclusion as
true.
(We'll see more about this later on.)
In the first part of the course we
are going to look more closely at the form taken by deductive arguments
that involve complete statements with a premise expressed as a
conditional relationship (one that can be restated with the phrases
"if" or "only if").
Inductive arguments can be seen as
involving reasoning based on the similarities of things or events
(reasoning by analogy), reasoning based on inferences from a limited
group to a much larger one (inductive generalizations and statistical
arguments), reasoning about what is likely to take place in the future
or have taken place in the past (think of explanations such as those a
jury is called up to make in a trial), and especially reasoning that
sets out to decide cause and effect relationships. We will be
looking at all this in more detail in the second half of the
course.
A
final point to be considered is how strong is a claim (the type of
statement that might become a conclusion in an argument). Saying
that Jack will get a perfect score on his exam is a stronger claim than
saying he will do well on it. A good working rule for evaluating
arguments intended to prove such claims is that the stronger the claim,
the better the evidence should be. For instance, knowing that
Jack is a good student and is studying hard might be enough to justify
saying he will do well on his exam, but we would need more evidence
before we can say he will get a perfect score. We would have a
much stronger case for this if we also knew the test was comparatively
easy.
