>
> I think I was talking about two styles of logical -inference-,
> deduction and induction. Kant might say that it is inductive or -
> empirical- and hence probable, not necessarily a part of the topic
> definition.
>
> http://groups.google.com/group/talk.philosophy.misc/msg/d55108c0ba7d042e
Immortalist wrote
SUMMARY OF SECTION 1.6
In this section, we discuss the essential nature of deductive and of
inductive arguments. The core of the difference between deductive and
inductive arguments lies in the strength of the claim that is made
about the
relation between the premisses of the argument and its conclusion.
In deductive arguments, the conclusion is claimed to follow from its
premisses with absolute necessity; in inductive arguments, the
conclusion is
claimed to follow from its premisses only with some degree of
probability.
A deductive argument is valid if its premisses do provide conclusive
proof
of its conclusion; otherwise it is invalid. But the terms "validity"
and
"invalidity" do not apply to inductive arguments, which are appraised
with
other terms.
The addition of new premisses may alter the strength of an inductive
argument, but a deductive argument, if valid, cannot be made more
valid or
invalid by the addition of any premisses.
Introduction to Logic Irving M. Copi
I'm afraid I do not think this sort of thing 'explains the essential
nature of deductive argument'. Specifically, it does not get to the
bottom of why and how a valid deductive argument is supposed to make
its conclusion necessary. Instead through keeping on repeating that a
valid deductive argument makes its conclusion necessary and then
giving instances of arguments that are supposed to be known as
absolutely necessary it creates a myth that we know why a supposedly
valid deductive argument is true and necessary.
The case is a bit like Socrates asking 'What is good?' and then
someone giving him lots of examples of good things. Similarly we want
to understand WHY the conclusion of a supposedly valid deductive
argument is necessarily true given the premises. Giving examples of
arguments we supposedly agree are 'logically valid' does not do this,
anymore than giving instances of things you assert, or that people
generally agree, are good tells Socrates WHY any of these things are
good. Also giving distingui****ng characteristics of valid deductive
arguments does not explain why they are valid e.g. that they proceed
from universal premises to a particular conclusion. It once again
ASSERTS that they are valid and then notices a feature which seems to
belong to the class of things asserted to be valid deductive
arguments. Just as if someone were to reply to Socrates 'Anything the
gods think is good, is good'. The philosophical question is WHY does
THAT make it good? Similarly we want to know why a valid deductive
argument (assuming there are such things), is valid? What MAKES it
valid?
If you reply 'It is necessary' my rejoinder will be 'What makes it
necessary?' If you reply 'It is deductively valid.' We are going round
in circles. If you reply (for instance) 'Its premises are general and
its conclusion is particular.' I will, once again, want to know why
that should make it necessarily true and deductively valid?
One possible response is. 'When we get to valid deductive arguments we
have reached rock bottom. We can see that they are necessarily true,
but if we are to try and give a satisfactory answer as to why they are
necessarily true we are liable to have to give valid deductive
arguments in our explanation. But in that case we would be assuming,
and so taking for granted, just what we are supposed to be
explaining.'
A difficulty with this response is that in logic and with valid
deductive arguments, we are supposed to be able to KNOW that we are
right and that the conclusion follows from the premises. But if you
can't give an account of how you know something, without begging the
question, you seem only to be left with an assertion that you do know
it, however certain you are of this assertion. But lots of people
assert lots of things, and often they are wrong. This surely wont do
for logic, which in valid deductive arguments is supposed to provide
the acme of what it is to JUSTIFIABLY assert some conclusion given
some premises.
Another possible response (which surely aught to come up on a group
discussing Kant!!) is that in a valid deductive argument the
conclusion is ANALYTIC on the premises. That is, if you look at the
premises in a valid deductive argument you will find that you have all
ready got the conclusion of the argument asserted in the premises.
What is asserted in the conclusion does not go beyond what is all
ready asserted, or given, in the premises. Since you've all ready got
the conclusion included in the premises, if you assert the premises
you must be justified in supposing the conclusion. Another way of
trying to put this point might be to say that in a valid deductive
argument the premises are SUFFICIENT to establish the conclusion
because if the premises are true that fact includes the truth of the
conclusion within it.
What we do in a valid deductive argument, according to something like
this response, is inspect the premise of the argument, or the meaning
of them, and see that they assert the conclusion. But this raises
another problem----WHERE IS THE MEANING?!!
If, for instance, we can't directly look at the premises and equally
directly see the meaning of 'a man' (for instance) then this
particular idea of logical deduction as something that we can perform
privately and come to correct conclusions that are immediately known
to be justified as necessarily true wont work. If we have to know how
the words in the premises are used, and if meaning is not an object
that can be inspected, as Wittgenstein argues at length it is not,
then this whole idea of logical deduction turns out to be a myth! Any
person thinking they are making an immediately valid logical
deduction, actually has to do something much more complicated and
outside of their control than is required by this myth of logical
deduction.---They have to know how certain marks or sounds are used by
their community. But this is just as complicated and un-logical as
knowing the relation****ps between any contingent matters of fact in
the objective world.
This conclusion is covered up by someone who thinks they 'know the
meaning of the words before them', asserts that some conclusions are
necessarily true given some premises, and then explains how they can
be known to be necessarily true by saying they are deductively, or
logically valid, given those premises. [This is still not quite the
same as saying that if we have certain propositions in the premises
(propositions being conceived as states of affairs, mind independent
truths, or mind independent assertions) these can include such
propositions as are in the conclusion, and so make it logically valid
given those premises].
In my previous but one post I tried to describe how na=EFve realism and
some science can view factual situations as self sufficient in terms
of the continuation of the objects, properties and processes which
apparently constitute those situations. In this way, I claim, we can
get an apparently philosophically satisfactory form of explanation,
because the premises can appear sufficient to produce what occurs in
the situation. And that this form of philosophically satisfactory
explanation can be pursued without having to try and make sense of how
the mind connects its phenomena, as Kant supposes is philosophically
necessary for understanding 'our understanding' of situations, and the
myth of logical deduction supposes must be fundamental to provide the
acme of philosophically justified reasoning.
http://members.lycos.co.uk/causalrealism/
|
