John S, List,

The question of what kinds of activities counts as mathematical inquiry and 
what kinds of systems count as mathematical systems is a question that is 
answered (on Peirce's account), first, by looking at the purposes that properly 
guide the inquiries, and second by looking at the methods that should be used 
to accomplish those purposes. The character of the diagrams that are 
constructed on the basis of the precepts is a tertiary concern.

When it comes to mathematics "I mean by its general hypothesis the substance of 
its postulates and axioms, and even of its definitions, should they be 
contaminated with any substance, instead of being the pure verbiage they ought 
to be. We have to make choice, then, between a division of mathematics 
according to the matter of its hypotheses, or according to the forms of the 
schemata of which it avails itself." (CP 4.246)

We could ask the further question of where the mathematical study of the 
systems drawn from games such as chess, go, checkers and tic-tac-toe fit within 
its major branches. In order to answer this question, Peirce indicates that the 
division should proceed on the following grounds:

Let us, then, divide mathematics according to the nature of its general 
hypotheses, taking for the ground of primary division the multitude of units, 
or elements, that are supposed; and for the ground of subdivision that mode of 
relationship between the elements upon which the hypotheses focus the 
attention. CP 4.248

As such, the question of whether or not one or another kind of game counts as 
mathematics is determined, on Peirce's account, mainly by the purposes that 
guide the formation and use of the general hypotheses that serve as the 
starting points for the more deductive parts of inquiries. What is more, the 
primary division between the mathematical study of these games is made first on 
the grounds of the kinds of multitudes that are involved and second on the 
grounds of the kinds of relations that hold between the elements. So, we should 
ask:  where does the mathematical study of games such as chess, go, checkers 
and tic-tac-toe fit within the main branches of mathematics. Which are the 
study of finite and discrete systems, which are the study of denumerable or 
first-abnumerable infinite and discrete systems, and which are the study of 
higher order abnumerable infinite and continuous systems?

Let us apply Peirce's distinction between the practical arts and sciences to 
this classification of mathematical systems.  In a sense, the counting, 
measuring and calculating that was done for many centuries by ancient Chinese, 
Egyptian, and Indian mathematicians was certainly mathematical in character, 
but some of these activities (like much of our own) is really mathematics as a 
practical art. Doing mathematics in a more scientific spirit requires, it 
seems, an understanding of the purposes that govern the activities and the 
methods that should be employed. The main purpose of scientific inquiry in 
mathematics is the same as the guiding purpose of all scientific inquiry which 
is to conduct experiments that are designed to help us learn the truth 
concerning general questions about which we are currently experiencing real 
doubt. (CP 8.115)

In "The Logic of Mathematics, an attempt to develop my categories from within," 
Peirce asks three questions:


1.     1. what are the different systems of hypotheses from which mathematical 
deduction can set out,

2.     2. what are their general characters,

3.     3. why are not other hypotheses possible, and the like?

The answers to these questions--including especially the third--will help us 
sort out what kinds of purposes and methods are really mathematical as sciences 
and which are not. When it comes to the character of systems of hypotheses that 
lie at the basis of the scientific study of games such as chess, go, checkers 
and tic-tac-toe, what is necessary in order for the mathematical deduction of 
theorems to proceed?

In his discussions of how mathematical inquiries often get started, Peirce 
points out that someone such as an ancient Egyptian property tax collector, a 
Chinese astronomer, or an Indian engineer asks a practical question that 
couldn't be answered by the normal means, so the mathematicians were called in 
for assistance. At this point, the mathematician went to work constructing 
diagrams that could serve as little "models" of the problem. In order for the 
models to be of any service, they had to be simpler than the real problem. Once 
the "models" and the rules for manipulating the diagrams were established, the 
mathematicians tries to discover what follows--necessarily--from those diagrams 
and rules. At this point, the mathematician is guided by a very different 
purpose than the tax collector, astronomer or engineer because the 
mathematician cares, first and foremost, about the purpose of finding the truth 
about what really does or doesn't follow consistently from the idealized 
hypotheses. It is a different question altogether--having a different 
purpose--when we ask what can be learned about the actual world from the 
results of the mathematical inquiries concerning this idealized systems.

So, I am drawn to the conclusion that most people who play games such as chess 
and go are not engaged in mathematical inquiry--at least not in a scientific 
spirit. They just want to win the game. Perhaps I am being uncharitable to most 
people who play these games, or perhaps I'm being too much of a stickler about 
what is requisite to be guided by the proper aims and methods, but this is the 
way I interpret the texts cited above.

--Jeff


Jeffrey Downard
Associate Professor
Department of Philosophy
Northern Arizona University
(o) 928 523-8354
________________________________________
From: John F Sowa <[email protected]>
Sent: Thursday, March 9, 2017 7:09 AM
To: Jerry LR Chandler; Peirce-L; Clark Goble
Cc: Benjamin Udell; Frederik Stjernfelt; Jeffrey Brian Downard; Jeffrey 
Goldstein; Jon Alan Schmidt; Ahti-Veikko Pietarinen
Subject: Re: [PEIRCE-L] Truth as Regulative or Real; Continuity and Boscovich 
points.

Jerry, Clark, list,

In my response to Jeff B.D., I was defending the claim that board
games are versions of mathematics.  But I definitely do *not* restrict
math to board games or to set-theoretic models.

Jerry
> Many mathematicians reject set theory as a foundation for mathematics

Yes. Peirce did and so do I. The four board games I cited illustrate
diagrammatic reasoning.  But those diagrams use only discrete set
theory.  Peirce also considered continuous diagrams, and so do I.
I would also allow diagrams for any mathematical structures anyone
might propose or discover -- including quantum-mechanical diagrams.

>>> JFS
>>> Thanks for the reference.  On page 134, Béziau makes the
>>> following point, and Peirce would agree:
>> JYB
>> Universal logic is not a logic but a general theory of different
>> logics.
> Jerry
> Analyze this quote. Is [JYB] saying anything more beyond
> a contradiction of terms?

Peirce's semiotic is a general theory of all kinds of sign systems.
Those systems include, as special cases, all natural languages and
all versions of formal logic.  I agree with Montague that the
underlying semantics of NLs and formal logics are essentially the
same, but I would add that formal logics are weaker than NLs.

I interpreted JYB as saying that universal logic is a theory about
logics in the same sense that CSP's semiotic is a theory about logics.
But JYB's notion of universal logic is weaker than CSP's semiotic.

>> JYB
>> This general theory is no more a logic itself than is
>> meteorology a cloud.
> Jerry
> What the hell is this supposed to mean? Merely an ill-chosen metaphor?

My interpretation of JYB:  Universal logic is to any particular logic
as meteorology is to clouds.

Jerry
> Chemical isomers are not mathematical homomorphisms because of the
> intrinsic nature of chemical identities. Thus, this reasoning is
> not relevant to the composition of Boscovichian points.

I would not impose any restrictions on the kinds of diagrams or the
mappings that define similarity.  If you can define a Boscovichian
diagram for chemistry, I believe that Peirce's notion of diagrammatic
reasoning could accommodate that diagram.

Implication:  Instead of defining a special kind of logic for every
kind of subject matter, I would just change the kinds of diagrams
-- quantum mechanical diagrams, Boscovichian diagrams, or whatever
mathematical structures anyone might discover or imagine.

JLRC
> Semiotics is not, in my view, a foundation for logic which is
> grounded on antecedent and consequences.

That is a Fregean view of logic, not a Peircean view.  For his
Begriffsschrift, Frege chose implication, negation, and the
universal quantifier as his primitives.

For his algebraic logic, Peirce started with Boolean algebra and
added quantifiers.  But he later switched to existential graphs.
The early version distinguished Alpha (Boolean) from Beta (which
added the line of identity).  But he later started with relational
graphs (existence and conjunction) and added ovals for negation.

For beginning students, Boolean algebra is too abstract.  It just
represents an NL sentence with a single letter like 'p'.  Peirce's
relational graphs are a better starting point because they can be
translated to and from actual NL sentences.  As a pedagogically
sounder approach, I follow Peirce's later tutorials (circa 1909).
See the first 25 slides of http://www.jfsowa.com/talks/egintro.pdf

Note slides 3 and 4 which come from Peirce's own intro in MS 145.
In slide 8, I discuss one of CSP's examples that has a direct
mapping to and from RDF -- the basic notation for the Semantic Web.

Many people believe RDF is a good starting point for logic.  I hate
the RDF notation, but I use the comparison to show semantic webbers
how a real logic can be defined on top of something like RDF.

Also note CSP's rules of inference (slide 25).  They are grounded
in the need to preserve truth (as determined by endoporeutic).  And
they apply equally well to Kamp's Discourse Representation Structures,
which Kamp designed for NL semantics.

Note slide 31, which presents two *derived rules of inference*
that are implied by the rules in slide 25.  These derived rules
emphasize generalization and specialization.  I believe that it is
more appropriate to say that logic is a theory of generalization
and specialization.  That includes implication as a special case
(p implies q iff p is more specialized than q).

There is much more to say, some of which I say in the slides
http://www.jfsowa.com/talks/ppe.pdf .  See slides 39 to 60.
In particular, note slide 59 about Turing oracles.

Clark
> The problem with the game theoretical view of mathematics is
> the question of realism.

I'm not sure what you mean by "game theoretical view".
There are three options, with some similarities among them:

  1. The idea that games like chess are mathematical systems.

  2. The point that Peirce's endoporeutic may be characterized
     as an example of Hintikka's game theoretical semantics.

  3. Wittgenstein's debate with Turing. (I prefer LW's side.)

Clark
> there’s a difference between how we use the language of
> mathematics and what the objects of mathematics are. That
> is what are the relationship between the game and reality.

The issues of nominalism vs. realism are orthogonal to all three
of these kinds of games.

Clark
> I think we have to think through carefully what sort of game we
> are playing if we’re going to use that as our metaphor.

Yes. But I believe that both nominalists and realists could adapt
any of the three "game theoretical views" to metaphors that are
compatible with their ways of thinking and talking.

Summary:  What I'm trying to emphasize is the fundamental
importance of diagrammatic reasoning for logic, mathematics,
language, science, and everyday life.  The model-theoretic
semantics used to define truth in formal logics is a special
case of diagrammatic reasoning.

John
-----------------------------
PEIRCE-L subscribers: Click on "Reply List" or "Reply All" to REPLY ON PEIRCE-L 
to this message. PEIRCE-L posts should go to [email protected] . To 
UNSUBSCRIBE, send a message not to PEIRCE-L but to [email protected] with the 
line "UNSubscribe PEIRCE-L" in the BODY of the message. More at 
http://www.cspeirce.com/peirce-l/peirce-l.htm .




Reply via email to