what would
be wrong or inadequate with interpreting in ZF types as

sets fulfilling certain conditions to avoid dealing with a theory of
types or with order-sorted logic?


> Could you say more about what conditions do you have in mind?

> As JM pointed out, I can't see why someone would want to avoid to "dealing 
> with a theory of types or with order-sorted logic" either.


Hi  Joao and Bruno


I'm sorry I took so long to answer your questions, but I wanted first to read  
all the literature

you pointed to, and carefully reflect about my answer.


Here it goes...



It is natural  to think of formal logic as a language to express and reason 
about mathematical arguments. In fact, two of the main areas of foundational 
studies, more properly of mathematical logic, have to do with syntax (including 
proof calculation) and semantics.


However, pragmatics is also an important aspect of a working language.

According to the article " Why Is Discrete Math Hard To Teach?"  in 
https://rjlipton.wordpress.com/  there, the author proposes 'viewing learning 
discrete math as learning a new language'.


He mentions Viète and Descartes referring to their development of symbolic 
notations to express and solve numerical equations, which meant a great advance 
for their time with respect to the  unwieldy rhetorical way of reasoning of the 
Arabs for solving algebraic problems. This was really a breakthrough that 
boosted further notational and calculational development that allowed to reason 
mathematically in a more effective way.

However, four centuries after that, and in spite of a century  of developments 
in the foundations of mathematics, and in mathematical logic in particular, it 
seems that there have not been anymore advances in this respect. The author of 
the above-mentioned article thinks that reasoning must be done through 
rhetorical arguments.

He does not conceive that an analogous achievement to the one done by Viète and 
Descartes for algebraic reasoning can be (or has been) done for logical 
reasoning. Well, he must have a point there. So called natural deduction, 
sequent, tableaux and resolution systems are actually, artificial ways of 
proving, perhaps appropriate for machine theorem-proving, but not compatible at 
all with the "rhetoric" with which most mathematicians express  their proofs.

However,  Avigad in his article “Clarifying the nature of the infinite” states:
"Logic can be construed as the science of symbolic representations of logical 
notions, and the means of calculating with them. This viewpoint underlies 
Leibniz’ notion of a calculus ratiocinator."
And also: "Logic [may be viewed] as a branch of algebra or as a mathematical 
tool in the tradition of Boole and Peirce."

Only from a theoretical or a potential point of view can Logic be considered a 
mathematical tool useful for (human) reasoning and calculation . But, it does 
not seem to offer anything of practical value to the working mathematician in 
terms of  complementing or clarifying  the rhetoric with which he expresses his 
proofs.

I am not against explaining our intuitions using natural (human) languages but, 
in the same way that the algebraic notations and calculations of Viète and 
Descartes help to clarify and avoid obscure rhetoric in the daily practice of 
mathematics, we should expect something analogous from a predicate logic 
deduction system.

This is why I find the theory of types, in particular the mimicking of logic by 
way of the Curry-Howard isomorphism, very awkward and contrived as a practical 
tool for effective  mathematical reasoning.
Please, do not misunderstand me, I considered the Curry-Howard isomorphism a 
nice and interesting result; and for those interested in Logic as an object of 
study (and not as a practical tool to express and reason about mathematical 
arguments ) it is most surely, an area for deep and attractive results.

True, there are important advances in mechanical theorem proving based on the 
calculus of construction, Martin-Lo¨f's dependable theory of types, hott, and 
so on. But in absence of explanations in terms of the rhetoric used by 
mathematicians or at least in terms of a formalization compatible with it, they 
will only work as oracles whose cryptic messages will only be understood by the 
initiated. Needless to say, this situation is unpalatable for the average 
mathematician.

I wonder which way is easier, adapting machines to the way people reason and 
express their proofs, or the other way around, adapting people to the way 
machines better do and express their proofs. Both ways seem to be considerable 
challenges.

(Maybe I would not have a problem working with order-sorted logic,  Having to 
put up with the clutter
necessary to express and prove (by hand) a simple argument using those 
alternative foundations of mathematics is what I do not feel attracted to.)

Abrazos calurosos.

Jaime

-- 
Você está recebendo esta mensagem porque se inscreveu no grupo "LOGICA-L" dos 
Grupos do Google.
Para cancelar inscrição nesse grupo e parar de receber e-mails dele, envie um 
e-mail para [email protected].
Para postar neste grupo, envie um e-mail para [email protected].
Visite este grupo em https://groups.google.com/a/dimap.ufrn.br/group/logica-l/.
Para ver esta discussão na web, acesse 
https://groups.google.com/a/dimap.ufrn.br/d/msgid/logica-l/BY2PR0101MB0886D1AB225F51495B99B025AD490%40BY2PR0101MB0886.prod.exchangelabs.com.

Responder a