LECTURE #5: FORMS OF INFERENCE
- so far we have focused our attention on
- what is the definition of a deductive argument?
- an argument where the conclusion follows from the premises with certainty.
- if the premises are true, the conclusion must also be true.
- we have done so because, in some respects, deductive arguments are the simplest to reconstruct
and to criticize.
- but, as we have seen, there are other forms of argument besides deductive.
- a complex argument may contain inferences that are nondeductive as well as deductive.
- [gun control argument overhead]
- the argument we looked at last week, for example, contains deductive inferences, an argument
from analogy, and overall has the structure of a convergent argument.
- sometimes the premises of a deductive argument will be supported by nondeductive arguments.
- purely deductive arguments are relatively rare.
- tonight, we’re going to consider four forms of nondeductive arguments:
analogical, and convergent.
- we’re going to find out how each one works and then consider some strategies for criticizing
- as usual, we’ll practice these techniques in class.
- any questions?
- the term
is sometimes used to refer to any argument form that is not deductive.
However, it is commonly used to refer to arguments with statistical premises.
- consider the following argument: [1
- this argmt is similar in form to one of the deductive argmt forms we’ve studied.
- which form? [predicate instantiation]
- do the premises support the conclusion?
- I would say that they do support it, but not in the sense that they make the conclusion certain.
- instead, if the premises are true, then the conclusion is
to be true.
- how likely? [66%]
- would this argmt (assuming the premises were true) convince you to take Modes of Reasoning?
- it might, but then again, it might not.
It depends upon other factors.
66% is not that high.
We’d like to see higher.
- now consider this argmt: [2
- now it just says “most” Modes students do better on essays.
But how much is most?
- again, the premises support the conclusion, but they don’t lend a
- we’d much rather see a quantified figure than “most.”
- “most” could be anywhere from 51% to 99%.
When I see an argmt of this form, it makes me
wonder why they didn’t give exact figures.
- why do you think they didn’t give exact figures?
- probably because the figure is not that high (say, 66%) and they wanted to hide that fact.
- so although an argmt like this one is technically successful, it is
successful than one which
gives more specific information.