Please don't give me the Wikipedia or dictionary explanations, we're looking at mutual influence.

In computer science the primitive expression of syntax would be as the format for instructions to the processor: "move x to w" or "w,x mov"; "add y,w" or "w+y"; etc. The syntax is determined ultimately by processor architecture, we can think of it as the rules that determine whether the computer can 'understand' the semantic (data) content. Given correct syntax, the computer knows what to do with the data but it is the programmer that must process the semantic content into a good syntax. From syntax, the computer can learn nothing, but a programmer can learn about the architecture of the computer.

For natural language, syntax is less important. The semantic content can be successfully transmitted even if the message syntax is imperfect. In addition (extra) semantic content can be encoded in non standard ways by variations in syntactic form; for example: "John is kicking the ball!" also mean he is kicking high, and "The ball is being kicked be John." means he is kicking deep... In addition to pragmatics, semiotics, and other linguistic features, humans can manipulate syntax to convey meaning.

It seems there are more than one level available for the processing of language. There is the literal (computer) level, then there is an adaptive (self-programming), communal (interactive) level of language usage. All of this brings up three questions, any one of which I would appreciate an answer to:

1) Is there similar issues in formal languages, and how are they treated?

2) Are there any literature that explores this in the context of "brain-mind interaction", "consciousness ", "self-awareness" etc?

3) What implications are there for artificial intelligence research, and has there been any publications?

I realize this is rather 'fringe' for Philosophy, overlapping with linguistics, computer and cognitive science, but there is a common thread (I think) that could have strong epistemic implications.

  • I think it is unclear what you would accept as "similar" issues. "It seems there are more than one level available for the processing of language." is this not to some extent the idea of "interpreted languages" but maybe you would not accept that as a "simular" issue? Jan 1 '19 at 15:19
  • 1
    See Conceptual semantics for an example of theory stressing the interplay of syntax and semantics. See also Lexical semantics. Jan 1 '19 at 17:22
  • @JesperHybel As I understand it one of the reasons for formal languages is to mitigate against ambiguities that may be introduced by syntactical influence. From a computer science perspective, interpreters would certainly qualify as another level from the "literal" language. I would think syntactic-semantic interplay would definitely impact natural language interpretation. However I don't know enough about formal and meta languages to gauge how much of a factor this issue is to them, or how much work has been done in the area.
    – christo183
    Jan 1 '19 at 19:30
  • One question at a time. Jan 3 '19 at 2:32
  • @curiousdannii From the Question: "... three questions, any one of which I would appreciate an answer to:" - in other words pick one. And of course three different questions, with largely similar body text would attract vacuous 'duplicate' flags...
    – christo183
    Jan 3 '19 at 6:06

The formal computer treatment of syntax and semantics is denotational semantics, e.g., https://en.wikipedia.org/wiki/Denotational_semantics and domain theory, e.g., https://en.wikipedia.org/wiki/Domain_theory That is, the denotational "semantic function" maps syntax in the computer language (which is typically specified by a BNF grammar https://en.wikipedia.org/wiki/Backus%E2%80%93Naur_form) to its meaning in an appropriate domain (often the space of (computable) N-->N functions, which is a formal domain as per the wikipedia link).

There's also operational semantics https://en.wikipedia.org/wiki/Operational_semantics as well as algebraic semantics https://cseweb.ucsd.edu/~goguen/pubs/asbk.html (I'm not immediately seeing a good link for this). What you really want is a good textbook for this stuff, e.g.,

The Formal Semantics of Programming Languages, http://www.cin.ufpe.br/~if721/intranet/TheFormalSemanticsofProgrammingLanguages.pdf

Semantics of Programming Languages, https://mitpress.mit.edu/books/semantics-programming-languages (I'm not immediately seeing a free and legal pdf)

You might also want to look at Plotkin's "Pisa Notes" lectures, http://homepages.inf.ed.ac.uk/gdp/publications/Domains_a4.ps

as well as tons of other stuff googling denotational semantics and/or domain theory.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.