Frederik, Howard, lists,
I've gone back and read Howard's post *"*RE: [biosemiotics:6635] Re:
[PEIRCE-L] Natural Propositions" Sat, 06 Sep 2014 20:59:06 -0400
https://list.iupui.edu/sympa/arc/peirce-l/2014-09/msg00124.html .
I just don't see much psychologism in Howard's views, even though he
claims that psychologism has its place and seems to regard himself as at
least partly psychologistic. He doesn't claim that psychology
establishes the validity of a deductive inference or the ampliative
character of a non-deductive inference (its conclusion's claiming
something that its premisses do not claim). That would be psychologism.
Instead mostly Howard focuses on heuristics, creativity in inference,
etc., as the place for psychological understanding of inference. I just
wouldn't call that psychologism. It's not the traditional meaning of the
word 'psychologism.'
Inferential statistics is a field that focuses on induction. I don't
think that there will be such a field focusing on abductive inference
except for the special sciences themselves; they are where abductive
hypothetical explanations are explicitly formalized (key heuristical
guesses in the processes of more general classes of research are often
not made explicit at all). So, for example, chemistry is the place to
learn the particulars of plausible chemical hypothesis generation.
Deductions _/from/_ hypotheses - those are mathematical. The
methodeutical level, the selection and strategizing of hypotheses for
testability and actual testing, is complex and not strictly abductive.
The special science concerned with mind or intelligent behavior is
psychology, and it does seem to be the most general place for the
riddles of _/how/_ one generates cogent guesses, plausible hypotheses,
etc., which generation is heavily context-dependent, belief-laden, etc.,
thus lending itself more to idioscopic study than to cenoscopic or
mathematical study.
In "Logical Machines" Peirce wrote: "Every reasoning machine, that is to
say, every machine, has two inherent impotencies. In the first place, it
is destitute of all originality, of all initiative. It cannot find its
own problems; it cannot feed itself. It cannot direct itself between
different possible procedures." (I should note that Peirce here does not
use the word "reasoning" in the sense that he gives it elsewhere, that
of conscious deliberate inference.) Of course now there are computer
programs that learn. These of course are not mere implementations of
propositional logic. So far I've hard of no computer programs that do
creative mathematics. What general way would there be for a computer
program to recognize that a deductive result is surprising or nontivial?
There could be specific algorithms of course, that would work up to a
point. The traditional categorical syllogisms are ways to ensure a
modicum of seeming novelty in a valid deduction.
I'm not aware of any general useful mathematization of the new aspect or
the nontrivial aspect which a worthwhile deduction's conclusion brings
to its premisses, aspects in whose universal absence no mind would
bother with deduction. The 'novel' aspect has sometimes been called
psychological. Peirce expected there to be no way to quantify
verisimilitude in the sense of the likeness of the inductive conclusion
to the sample data, the sense in which the conclusion seems a more
veteran version of the sample, the opposite of seeming novel. Now, I'm
not saying that nontriviality, novelty, and verisimilitude are, like
Peirce's version of natural simplicity, forms of appeal to inborn
instinct. They seem forms of appeal to something that has come to be
called 'intuition' - not in Peirce's sense of 'intuition', but rather in
a loosened sense of 'instinct' - developable but not necessarily inborn
in a particular form attuned to idiosocopic nature. I don't know what
Peirce would have thought of this, but I would not rule out that
psychology in Peirce's broad sense, also called by him psychognosy, may
have something essential to contribute to the understanding, for
example, of the "aspectual" or perspectival nontriviality or 'depth' (in
the sense of the diametrical opposite of triviality) that guides pure
mathematical research. One might further note that each of these
seemingly psychological aspects seems to _/compensate/_ for the formal
character of its respective mode of inference. I don't think there's any
psychologism in my views. In other words, I'm granting as much of
Howard's claims as I can understand of them (he obviously knows a great
deal more about things like AI than I do), and arguing that they're not
psychologistic, and that Howard may be mistaken in regarding himself as
having tendency properly called 'psychologistic'. We may be arguing over
words, as Jon Awbrey worries.
In his above-cited post, Howard says, "The evidence so far leads me to
think embodied cognition is most important." I'd say, important for
what? There are different objectives of research, different guiding
research interests, especially across classes of research that differ
vastly in scope, and various people value objectives variously.
Frederik is pursuing the very idea of application of cenoscopic
semiotics to biology. This requires him to discuss semiotics at a very
cenoscopic level, more general in scope than reflects the usual
interests of biologists. The development and application of some of
mathematical analysis to physics requires some pure-mathematical work
and understanding of mathematical analysis more general in scope than
reflects the usual interests of physicists. A better example might be
work in theoretical statistics for the purpose of application in special
sciences. Where these examples that I've offered seem to fall shortest
is that they don't reflect complications arising from the reflexivity
involved in philosophy and psychology, complications that vaguely seem
to me to be playing a role in the present discussion.
Best, Ben
On 9/10/2014 6:40 AM, Frederik Stjernfelt wrote:
Dear Howard, lists -
Reality certainly is too complex for any one model. That is one of the
reasons we have disciplines.
I have never said anything against the discipline of psychology which
has many merits and many important tasks. My claim is just that logic
and psychology have different objects and different tasks. So I remain
as anti-psychologistic as ever. For anti-psychologism is not at all
against psychological research - it is against the erroneous belief
that issues of truth, reasoning and validity of thought should be
reducible to psychology (Peirce, of course, was also interested in
psychology and even counts as an early pioneer in experimental
psychology for his and Jastrow's investigation of subliminal perception).
A reason why I think it is important now again, after 100 years, to
address the issue of anti-psychologism is discussed in the conclusion
of ch. 2 of NP - namely the fact that neuropsychological research in
many cases proves to reintroduce psychologism via the back door. To
preempt misconceived counterarguments - this does not imply any
scepticism on my part vis-a-vis the neurosciences as such. Quite on
the contrary, I follow recent neuroscience with interest, even
enthusiasm. But scientific progress sometimes blind researchers to
believe that their claims have vaster applicability than warranted by
data. For instance, when we hear investigators of the reward system of
the brain claim to be able to explain economics from such study, or
when investigators of Broca's and Wernicke's areas claim that language
as such can be explained from the functions of those brain structures,
or when scholars of neuroaesthetics may claim that beauty and art are
explainable from brain findings, I think elementary ontological
misconceptions are involved - and that the error, again, lies in
psychologism in a new garb.
The progresses of neuroscience made possible by the development of
brain imaging techniques during recent decades - PET, fMRI, MEG etc. -
are important, among other things, for their charting of how the human
brain processes economics, language, art etc. But that is not at all
the same as claiming that these phenomena unilaterally depend on
particularities of the human brain.
For my own part, I am currently involved in a project experimentally
investigating joint diagrammatical reasoning - how people collectively
solve diagrammatical tasks by communicating
(http://jointdiagrams.org). This gives many interesting findings, but
I do not think those findings should be interpreted as reducing
diagrams to psychological processes.
Best
F
Den 10/09/2014 kl. 03.37 skrev Howard Pattee <hpat...@roadrunner.com
<mailto:hpat...@roadrunner.com>>
:
At 04:20 PM 9/9/2014, Frederik wrote:
Dear Howard, list
This [complementarity of models] is an important issue. I do not
think it is a choice between a psychological and a non-psychological
model which must compete. It is a distinction between two levels of
description . . .
HP: That helps. As you all know, Stan and I believe that reality is
too complex for one model. Adequate models of reality require
complementary hierarchical levels. So I infer that you can also
accept psychological models and are not as anti-psychologistic as you
sound in/NP/or as Peirce obviously was/./
Enough of that, Onward to Dicisigns.
Howard
-----------------------------
PEIRCE-L subscribers: Click on "Reply List" or "Reply All" to REPLY ON PEIRCE-L
to this message. PEIRCE-L posts should go to peirce-L@list.iupui.edu . To
UNSUBSCRIBE, send a message not to PEIRCE-L but to l...@list.iupui.edu with the
line "UNSubscribe PEIRCE-L" in the BODY of the message. More at
http://www.cspeirce.com/peirce-l/peirce-l.htm .
