Formal linguistics uses mathematical and logical tools to represent and analyze language structure. It seeks to express linguistic knowledge as precise, testable systems rather than intuitive descriptions.
Begin by examining a simple grammatical phenomenon (e.g., word order) and asking how you would state it precisely enough that a computer could verify it. Compare informal descriptions with formal rules.
Imagine you tried to describe the rules of chess to someone using only ordinary English prose. You could say "the rook moves in straight lines" and "the knight moves in an L-shape" — but as the rules get more complex (castling, en passant, check conditions), natural language becomes imprecise and ambiguous. Eventually you'd want a notation system precise enough that a computer could verify any board state automatically. Formal linguistics is the project of building that kind of system for language — replacing intuitive descriptions with precise, computable rules.
The core ambition is to separate competence (the abstract knowledge of a language — the grammar in a speaker's head) from performance (how language is actually used in real-time, with hesitations, errors, and shortcuts). Formal linguistics targets competence: it tries to write down, in mathematical notation, what every fluent speaker implicitly knows. This is a bold claim: that the fuzzy, creative, culturally-embedded thing called "knowing a language" can be at least partly captured in a formal system. The history of the field is a series of attempts to make that claim increasingly precise — and to discover where formal systems fall short.
The tools come largely from logic and mathematics: set theory, formal grammars, automata theory, lambda calculus, and model theory all appear in different subfields. Phonology uses rule systems and later constraint rankings to capture which sound sequences are allowed in a language. Syntax uses phrase structure grammars and transformations to generate grammatical sentences and block ungrammatical ones. Semantics uses predicate logic and possible-worlds models to represent meaning precisely enough to reason about entailment, negation, and quantification. Pragmatics uses game theory and probability to model how context shapes interpretation.
What ties these together is a shared methodology: state a phenomenon, formalize a rule that predicts it, test the rule against data (native speaker judgments, corpora, cross-linguistic patterns), and revise when the rule over- or under-generates. The goal is not to write down every sentence of English — that's impossible — but to write down a finite system of rules that can, in principle, generate all and only the grammatical sentences of the language. This generative ambition distinguishes formal linguistics from traditional descriptive grammar and from corpus linguistics. It is linguistics as mathematics — and the payoff is a set of tools precise enough to inform computational systems, theories of language acquisition, and the study of linguistic universals across all human languages.
This is a foundational topic with no prerequisites.
No prerequisites — this is a starting point.