I don't even know where to begin here, Michael. Your bafflement probably comes 
from psychology textbooks typically being vague and uninformed about the 
distinction. Let me put the question the other way for you: Why would you 
regard them as being the same, putting them both under the very vague heading 
of "reasoning," rather than making then entirely distinct categories. 

There is a formal process that explains most kinds of deductive logic. We know 
exactly how to get validly from premises to conclusions (where "valid" means 
that truth will be "preserved" from premises to conclusions). That formal 
process began to be uncovered in Ancient Greece (and probably before), but was 
still divided between "propositional" and "syllogistic" (quantified) branches 
until Bertrand Russell and Alfred North Whitehead combined the two in the early 
20th century (in their Principia Mathematica). Since then, deductive logic has 
focused mostly on extensions of that (2nd order logic) and various "deviant" 
logics (modal, multi-valued, relevance, etc.). The problem is that deduction is 
always the unfolding of specific knowledge that was "folded" into knowledge we 
already (believed we) had. If all ravens are black, then of course the raven in 
my backyard right now is going to be black. Sometimes deductions are not that 
obvious -- they lead to realizations of implications we did not previously 
recognize -- but it can never result in truly new knowledge, which is what the 
"new" scientists of the 17th century aimed to produce.

Induction is an entirely different ballgame. Although Aristotle did some 
primitive work on it, how it actually operated was pretty much a mystery until 
the mid-19th century. The general idea was that, somehow, the collection of 
enough individual observation of some phenomenon would eventually add up to 
evidence sufficient for one to make a general claim (Francis Bacon is often 
credited with a view like this, though his position was actually more 
sophisticated). David Hume famously "exposed" the whole thing by noting that 
nothing of that sort was ever going to work because, no matter how many 
positive instances one had collected, it would only ever take one additional 
negative instance to bring down the whole thing (classically illustrated by the 
discovery of black swans in Australia -- all the white swans Europeans had 
observed to that point didn't mean a thing, and the whole project of 
determining that "All swans are white" collapsed in an instant). By the bye, 
Popper's famous "falsificationism" is an effort to take Hume seriously by 
attempting to convert science into a form of deduction by only attending to 
refuting instances (which, unlike confirming ones, are deductively valid). 
Although it evaded the old Humean "problem of induction,"  it left science with 
no way to make positive statements about the world (which, clearly, it does, 
more or less successfully, all the time).  Popper was not the only one to 
attempt to turn science into a form of deduction. Carl Hempel did as well, with 
his Deductive-Nomological theory. It had its problems as well (in what way does 
observing the prediction in the conclusion contribute to the truth of the 
covering law in the premises? Problem of induction again), but it did give 
science a way to make positive claims again. 

Induction finally began to get formalized when it was wedded to probability 
theory. There are bits of this in the 18th and 19th centuries, but it didn't 
really get going until the 20th.  Probability theory solves (or, at least, 
addresses) the problem by providing a framework for rigorously deducing 
conclusions that have probabilities in them (and so the uncertainty is 
quantified). That is, one may not be able to conclude that "All swans are 
white," but I can conclude that there is, say, "a probability of .8 that the 
next swan I see with be white." There are some names psychologists will 
recognize in this story: Ronald Fisher and Karl Pearson, among others. 
Unfortunately, psychology seems to be stuck in the frequentist formulations of 
these folks and, despite repeated efforts since the 1960s, can't seem to move 
on to more adequate Bayesian formulations of the late 20th century (though 
several other sciences have).

There is also a third form of reasoning that is often conflated with induction. 
It was identified by the late 19th-century American 
scientist-philosopher-semiotician Charles Sanders Peirce. He noted that the 
process by which we attempt to inductively confirm a hypothesis with evidence 
(induction) is entirely distinct from the process by which we choose a 
particular hypothesis to be confirmed in the first place. Peirce labeled the 
latter process "abduction." Psychologists, to the extent that they pay 
attention to this process at all, tend to conflate it with induction.  

There is no reason, a priori, to expect that these processes will be 
underpinned by the same psychological processes. Indeed, the natural assumption 
would seem to be that they are pretty distinct, psychologically. That is why 
cognitive textbooks should separate them (probably into different chapters). 

I hope this helps.

Chris
---
Christopher D. Green
Department of Psychology
York University
Toronto, ON M3J 1P3
Canada

[email protected]
http://www.yorku.ca/christo/
==========================



On 2012-03-16, at 8:14 AM, Michael Britt wrote:

> A typical cognitive psyc chapter explains these two terms and the difference 
> between them.  I'm always left with a feeling of "Okay. Interesting, but so 
> what?".  My apologies to those who have studied logic.  But aside from being 
> fodder for a multiple choice question, ("Which of the following is an example 
> of inductive reasoning?"), and excuse the emphasis on pragmatism, but why do 
> we teach these two concepts?  
> 
> 
> Michael A. Britt, Ph.D.
> [email protected]
> http://www.ThePsychFiles.com
> Twitter: mbritt
> 
> 
> 
> 
> 
> 
> ---
> You are currently subscribed to tips as: [email protected].
> To unsubscribe click here: 
> http://fsulist.frostburg.edu/u?id=430248.781165b5ef80a3cd2b14721caf62bd92&n=T&l=tips&o=16755
> or send a blank email to 
> leave-16755-430248.781165b5ef80a3cd2b14721caf62b...@fsulist.frostburg.edu


---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=16761
or send a blank email to 
leave-16761-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to