This is something I came across while thinking about expressing a 'nested' choice, i.e. you make a choice where one of the options entails making another choice, for example, at college you could choose to study a single major A, or enter another stream where there is two majors, B and C which you must decide on at a later date.

If I were to treat the word 'or' as an operator using the mathematical 'order of operations' I would have something like:

A or (B or C) actually, however the order of operations here are slightly wrong, as in Mathematics we would have to decide on B or C first, it does however represent the idea that the other choice is nested.

Is there any sense of using words such as 'or' as operators that take two binary arguments? As with 'or' we must use it twice, but its just between words, there's no sense of whether it is just bridging words or taking whats on either side as an 'input'.

  • I can't answer for anything but US English, but the sense of 'and' and 'or' is very muddled, because 'or' sometimes means an excluding choice and 'and' can mean that you take both branches. So getting computer program requirements correct can seem to ordinary people like an Inquisition! Order and nesting is rather a Bridge Too Far perhaps.
    – Scott Rowe
    Apr 28, 2022 at 10:07
  • 1
    Order has been implicitly stipulated in the linear production rules of most r.e. languages including regular and CFLs… Apr 29, 2022 at 22:37
  • Linguistics has something similar, scope, but it's more commonly considered for quantifiers and negation. See this Q&A from Linguistics.SE. Apr 29, 2022 at 23:56
  • Of course. Informal languages have a lot of previous assumptions. When you say "I'm going to get a beer with Tim", it does not mean that Tim is part of the beer, or that you will first get the beer, and then will meet with Tim. It will not be in five years. It is assumed that you are a human, you can drink, the beer is liquid, it has alcohol, etc. To use human language you need first long years of living experience.
    – RodolfoAP
    Feb 17, 2023 at 13:35

1 Answer 1



If you are speaking conceptually, then all mathematical statements are true whether in natural or artificial language. 'Three plus six divided by two' simplifies to 'six'. However, note that the natural language expression evaluates to the answer by way of conceptual convention, not a property of the language per se. But I don't think that's what you're asking.

Rules that Govern Language Sounds, Syntax, and Semantics

Now, it seems to me you're asking are there rules that govern syntax of morphemes, and to that the answer is a resounding yes. The rules that construct a language in a sense generate it, and while rules have been observed by grammarians for thousands of years, Noam Chomsky rocked the world with his theory of generative grammar.

Ray Jackendoff expresses a more modern theory of the rules of operations on pg. 125 of his Foundations of Language with his model of tripartite architecture in essence dividing up the rules of natural language into phonological, syntactic, and conceptual formation.

Therefore, one has to be clear to articulate which domain of rules one is talking about. Chess and math have rules and orders of operations, for instance, black and white alternate moves in chess, and these are conceptual constraints. English and German have syntactical rules, such as modifiers generally go before the nouns, and in English the verb occurs in the middle of the sentence, but for German, complex predicates have verbs listed at the end. And of course both have phonological rules such as how dipthongs and blends occur or how phones correspond to graphemes.

Operations that Govern Transformation

Relevant to philosophy of language is how multiple syntactical expressions might represent the same proposition.

S1. Bob dropped the pizza. (Active voice)
S2. The pizza was dropped by Bob. (Passive voice)

S1 is transformed into S2 by a rule which roughly consists of swap the order of the subject (Bob) and the direct object (the pizza), and transform the verb using an auxiliary copula (was). This rule is regular in English, so any active sentence can be transformed from active to passive.

SVO -> O(was/were)VS

Typlogogically, this is significant because English like Spanish is SVO.

Rules of Syntax and Semantics with Logical Connectives

Lastly, do rules govern the use of logical connectives in English? Sure. A list is generally expressed as {e1, e2, e3[,] and/or e4} (with the square brackets indicating the use of the Oxford comma as optional (regrettably ;). Thus we have standard and non-standard (indicated by *) expressions with logical operators:

S3. The boy has a mitt, bat, and ball.
S4. The boy has a mitt, bat and ball.
S5. The boy has the mitt, bat, or ball.
*S6. The boy has a mitt and batt, ball.
*S7. The boy has or a mitt, ball, ball.

And note, that nesting of linguistic structures generally employs punctuation for clarification when written to economize on word use:

Bob has the bat, mitt, and ball; or he has the skis and the poles; or he has both the bat, mitt, and ball, and the skis and the poles.

And generally, where concepts are ambiguous, for instance, is 'or' inclusive or exclusive, there are rules of syntax for clarification.

S8. Bob has the bat or ball. (inclusive or exclusive?)
S9. Bob has either the bat or ball but not both. (exclusive)
S10. Bob has the bat or ball or both. (inclusive)


What's important is that when you talk about rules and operations in natural language, you differentiate whether you're talking about the phonological domain, the syntactical one, or conceptual one. In fact, philosophers of language expand even further into pragmatics which explores ideas like performativity, implicature, and direction of fit.

  • Using 'convention' to express mathematical operations in natural language is interesting, although not the point of my question, I'd agree that really however you define it, 'mathematical language' can be expressed through the mean of natural language through 'conversion words' but really the 'mathematical/logical language' only exists through the symbols.
    – Confused
    Apr 28, 2022 at 17:31
  • 1
    We can 'describe' the symbols or 'describe' the operation and that allows us to interpret a symbolic statement, but its not a property of the language as you say.
    – Confused
    Apr 28, 2022 at 17:35
  • Interestingly, words circumscribe concepts, like in languages where the only expressions of numerosity are none, one, and many. This aspect of language is called expressivity, and the more expressive the language, the more distinctions it draws in concept-space if you will. Learning rules and using language is therefore central to modeling conceptual reality. Your question is an excellent question because you are attempting to broaden your own set of rules regarding the predication of the phrase 'Rules govern X' where X is aspects of natural language...
    – J D
    Apr 28, 2022 at 17:44
  • My own take on mathematics is essentially it is a language for converting the experience of numerosity, shape, relations, operations, truth, and direction. Good luck!
    – J D
    Apr 28, 2022 at 17:45
  • Of course, I'll need it for sure
    – Confused
    Apr 28, 2022 at 18:03

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .