Richard Montague: When Logic Met Language

A logician and the study of meaning

Oct 10, 2025

Richard Montague was not a linguist. He was a logician trained in mathematics and philosophy, more at home with symbols and proofs than with everyday speech. Yet in the late 1960s and early 1970s, he made a claim that impacted both philosophy and linguistics:

“There is no important theoretical difference between natural languages and formal languages.”

To put it more plainly, Montague argued that human languages like English, French, or Chinese could be studied with the same tools used to analyse the precision of logic and mathematics. Many found this outrageous. Natural language was messy, ambiguous, full of idioms and irregularities. Logic was neat, rule-bound, and supposedly immune to confusion. Montague’s point was that the gap between them was not as wide as people thought.

What did this mean in practice?

Every student read a book.

At first glance, this looks straightforward. But it hides two possible meanings:

All the students read the same book.
Each student read a different book.

This kind of ambiguity is everywhere in natural language. Montague showed that you can capture these differences using logical notation, in the same way mathematicians work with formulas. His approach treated natural language as if it were a formal system, where the meaning of a whole sentence comes from the meaning of its parts and the way they are combined. This principle, known as compositionality, became one of the central pillars of modern semantics.

By applying the full machinery of logic to human language, Montague opened the door to a different way of studying meaning. But he also left behind work so dense and technical that linguists needed a bridge to make sense of it, and that bridge was built by Barbara Partee.

Montague’s Big Idea

Montague was trying to solve a basic problem: how do we explain meaning in natural language with the same clarity we use in mathematics? Traditional philosophy had long been interested in meaning, but often in a vague, descriptive way. Montague insisted on precision.

The cornerstone of his project was formal semantics: the idea that natural language meaning could be captured in a system of rules and symbols. Instead of treating sentences as unstructured wholes, Montague showed how you could break them down into parts and assign each part a logical role.

If Alice left, Bob stayed.

The meaning here depends on how we understand the logical connection between “Alice left” and “Bob stayed.” Montague’s system mapped this relationship directly onto logical operators like “if…then,” making explicit what is often only implicit in everyday speech. This was not about stripping language of its richness, but about showing that the logical skeleton of meaning is always there, even in the most ordinary sentences.

This approach allowed linguists and philosophers to tackle long-standing puzzles:

How do we interpret quantifiers like every, some, or most?

Quantifiers: every, some, most

- Every student passed the exam means there were no exceptions.

- Some student passed the exam means at least one did, but possibly more.

- Most students passed the exam means more than half.

Montague’s system allowed these words to be treated with mathematical precision. For example, “every” can be modelled as: “for all x in the group of students, x passed.” This shows how logic captures subtle but important differences in meaning.

Why do sentences with the same words sometimes mean different things?

Same words, different meanings

Visiting relatives can be annoying.

- Meaning 1: Relatives who visit you can be annoying.

- Meaning 2: The activity of visiting relatives can be annoying.

Montague Grammar explains how one string of words can map onto two different logical structures, making the ambiguity explicit

How can we formally explain the difference between “the” and “a”?

“The” vs. “A”

- The cat is on the mat. (implies a specific, identifiable cat)

- A cat is on the mat. (introduces some cat, not yet identified)

Montague formalised this by treating “the” as referring to a unique entity, while “a” signals the existence of at least one such entity. These small differences in articles have large consequences for meaning, and his system captured them in logical form.

Montague’s answer was that natural language could, and should, be treated with the same seriousness as logic.

The Clash of Traditions

Montague’s work landed in the middle of another “(r)evolution” in linguistics: Chomsky’s rise to dominance. Chomsky had shown that syntax, the structure of sentences, could be studied as a formal system. But his project often treated semantics as secondary. For Chomsky, meaning was messy compared to the elegance of syntax.

Montague reversed the emphasis. He argued that meaning was just as amenable to formal study as structure. This put him at odds with the Chomskyan mainstream, which often considered semantics as peripheral.

The tension was clear. Where Chomsky focused on the internal rules of grammar, Montague wanted to show how those rules linked directly to meaning. To some, Montague seemed to be trespassing into territory that linguists thought belonged to them alone. But in hindsight, both projects were complementary. Montague’s formal semantics eventually became a natural partner to Chomsky’s generative syntax, even though the two men never collaborated.

Barbara Partee and the Bridge

Montague’s writing was not intended for linguists. His papers are dense with symbols, set theory, and logical proofs. Even professional linguists struggled to make sense of them.

Enter Barbara Partee, a linguist who had studied with Chomsky but became fascinated by Montague’s work. She realised that Montague Grammar had the potential to revolutionise linguistics but only if someone could make it accessible.

How Linguist Barbara Partee Pioneered a Field by Studying What She Loved | alum.mit.edu — **Barbara Partee**

Partee translated Montague’s ideas into a form linguists could use. She explained his technical innovations in plain language, applied them to actual linguistic data, and trained a generation of linguists in formal semantics. Without her, Montague’s work might have remained a footnote in the history of philosophy. With her, it became one of the foundations of modern semantics.

Partee’s Example: Every farmer who owns a donkey beats it
At first, the sentence seems straightforward. But in formal semantics, it poses a challenge: the pronoun it must refer to a donkey that appears inside another part of the sentence. This kind of dependency was notoriously difficult to represent before Montague’s framework.

Partee showed how Montague’s system could model it precisely. The sentence says:

“For every person who is a farmer, if that farmer owns at least one donkey, then that farmer beats that donkey.”

In logical form, this becomes:

∀x (farmer(x) ∧ ∃y (donkey(y) ∧ owns(x,y)) → beats(x,y))

Breaking it down:

∀x means for every x (for every individual).
farmer(x) means x is a farmer.

∃y means there exists at least one y.
donkey(y) means y is a donkey.

owns(x,y) means x owns y.
beats(x,y) means x beats y.

The arrow (→) means if … then ….
So it reads:
For every individual x, if x is a farmer and there exists some individual y such that y is a donkey and x owns y, then x beats y.

Partee used this to show that natural language sentences, however complex, can be translated into precise logical forms. This was Montague’s core insight made readable.

A Simpler Explanation

You are trying to explain what’s tricky about this sentence.

When you say, “Every farmer who owns a donkey beats it,” you clearly mean that each farmer beats his own donkey.

But the semantic puzzle is: the word “it” comes after “a donkey” but doesn’t refer to one single donkey. It changes each time, depending on which farmer we’re talking about.

So, in your head, you’re doing something like this:

Farmer 1 owns Donkey 1 → he beats Donkey 1.

Farmer 2 owns Donkey 2 → he beats Donkey 2.

Farmer 3 owns Donkey 3 → and so on.

Grammar doesn’t easily show how that pattern works, but Montague’s logic could.
His system used symbols (like x for farmer and y for donkey) to keep track of who owns what and who beats what.

Legacy and Relevance Today

Montague died tragically young in 1971, at the age of 40, the victim of a violent murder. His career was cut short, but his influence continued to grow.

Today, Montague’s ideas live on in several fields:

Linguistics: truth-conditional semantics, compositionality, and the study of quantifiers all trace back to Montague.
Philosophy: his insistence on logical rigour shaped contemporary debates about meaning, reference, and truth.
Computational linguistics: Montague’s project is echoed in natural language processing, where computers try to model meaning in ways he anticipated.
Artificial intelligence: attempts to formalise human language for machines build directly on his legacy.

When you ask a voice assistant, “Did every email get answered?”, the system must distinguish between one specific email and all emails. That kind of fine-grained distinction is exactly the kind of problem Montague’s approach was designed to handle.

His legacy shows that his claim was not just provocation. Treating natural language like a formal system turned out to be one of the most fruitful moves in modern linguistics.

Conclusion

Richard Montague’s life was brief, but his impact was lasting. He forced linguists and philosophers to take semantics seriously, and to treat meaning with the same analytical depth as syntax. His bold claim — that there is no theoretical gulf between natural and formal languages — continues to guide research in linguistics, philosophy, and artificial intelligence.

Yet Montague was never an easy writer, and his work would likely have remained locked in the world of logic had Barbara Partee not carried it into linguistics. Together, they changed the course of semantics.

Montague reminds us that even the messiest parts of human language can be studied with clarity and rigour. What once seemed chaotic turns out, under the right lens, to be governed by patterns every bit as precise as mathematics.

LinguisticallyYours' Substack

Discussion about this post

Ready for more?