Dear Ken,
I agree with all sides of this which is probably illogical but there you go, I'm a human.
But, of more use than humour, in this case, for me, is what we might take from Aristotle's Rhetoric. That is, while we might want to argue using syllogisms, we seek to persuade using enthymemes. That is, when not building bridges, we are seeking to persuade others, for various reasons, using fragments of language logic to do so. Again I urge that we can all benefit from a study of rhetoric not simply to avoid the pitfalls of being seduced by smart talkers using twisted logics (Sophists for Socrates) but rather to take advantage of our ability to see the possible errors in our ways. I enjoy my unintended errors when I can capture the fault.
Again, I would draw attention to some of the the other members of the "duct" family: induct, conduct, subduct (seduct), reduct.
Cheers
keith
>>> Ken Friedman <[log in to unmask]> 06/21/11 8:40 AM >>>
Dear Terry,
Having dipped into this conversation earlier, I'm going to risk dipping in again
to narrow the frame. You're arguing here that logic has uses in checking the validity
of sentences, truth claims, and reasoning. For me, at least, this was never an issue.
I use Aristotelian logic all the time to clarify issues.
The assertion that logic is not a model of human thinking or reasoning has to do
with a claim that machines or automated systems can design.
My argument against that claim is that machines or automated systems cannot engage
in discovery as human do, and therefore, they cannot design. Logic does not model the
process of human thought and logic-based systems cannot think as people do.
I'd agree that logic is among the many useful tools we have for checking or testing
claims. This may includes using logical steps or logic machines to check preliminary
results at different steps in the creation process.
Einstein always argued that scientific propositions were free creations of the imagination.
He also asserted that one must subject these free creations to stringent testing. One of
his great guides was deductive Euclidean geometry.
I don't think that anyone is saying that logic has no uses. Designers would benefit from
the ability to test propositions and truth claims more often than they do. What is being
said here is simply that "predicate first-order logic is not a model of human thought and
reasoning." People don't think or reason through predicate first-order logic, though they
may well use predicate first-order logic to check what they think.
Yours,
Ken
|