Hans Kamp

Hans Kamp: The Logic Hidden in Ordinary Talk

Classical logic handles a sentence beautifully. It has always struggled with the second sentence — the one that says "he" or "it" and assumes you have been paying attention. In 1981 a Dutch logician teaching in London published a paper arguing that the unit of meaning is not the sentence at all but the discourse, and that to understand language is to build a representation and keep updating it. Formal semantics has been working in his shadow ever since. Kamp's own description of himself: "not really a linguist," "not really a philosopher," "not really a logician."

Texel, Leiden, Amsterdam

Johan Anthony Willem Kamp was born on 5 September 1940 in Den Burg on Texel, the Dutch island at the top of the North Holland coast. He went to the Murmellius Gymnasium in Alkmaar from 1952 to 1958 and then to the University of Leiden, where from 1958 to 1961 he read physics and mathematics. What happened next set the pattern of his career: he got hooked on logic, and moved to the University of Amsterdam from 1961 to 1965 for a *doctorandus* in logic, foundations of mathematics and the philosophy of the empirical sciences. He has described E. W. Beth, of the Amsterdam school, as one of the three strongest influences on his thinking.

Montague's Student

In 1965 he went to UCLA, and the choice was decisive. His supervisor was Richard Montague, who was then constructing the programme that treats natural language as a formal language amenable to exactly the same model-theoretic treatment as the languages of logic. His doctoral committee also included Chen Chung Chang, David Kaplan, Yiannis Moschovakis and Jordan Howard Sobel — as concentrated a group of logicians as any student could sit before. He took the PhD in philosophy in 1968 with a dissertation on *Tense Logic and the Theory of Linear Order*, proving a functional completeness result: the temporal operators *since* and *until* suffice to define all the others over continuous linear orderings. Alongside Montague and Beth, he credits Arthur Prior, the founder of modern tense logic, as a formative influence — in each case for methodological rigour and intellectual honesty rather than for doctrine.

Now

His 1971 paper "Formal Properties of 'Now'" is a small classic. The word *now* is a nuisance for standard model-theoretic semantics: its reference depends on the moment of utterance, but it can also be embedded inside operators that shift the time of evaluation, and the two indices come apart. Kamp's solution — double indexing, evaluating expressions against two temporal coordinates at once — became standard equipment in the semantics of indexicals and modality.

He was by then moving constantly: instructor at UCLA in 1968, assistant professor at Cornell from 1968 to 1970, Amsterdam in 1970–71, then thirteen years lecturing at University College London and Bedford College from 1971 to 1984, with a professorship at the University of Massachusetts, Amherst in 1983 and one at the University of Texas at Austin from 1984 to 1988.

The Turn to Discourse

"A Theory of Truth and Semantic Representation" appeared in 1981 and introduced Discourse Representation Theory. Its central move is a reversal. Montague semantics assigns truth conditions sentence by sentence; DRT interposes a level of representation — a discourse representation structure — which the hearer constructs incrementally as the text proceeds, introducing markers for the things being talked about and conditions on them, and which is then interpreted against a model. Anaphora across sentence boundaries, which had embarrassed compositional semantics for decades, becomes natural: the pronoun in the second sentence picks up a marker the first sentence put on the board.

The consequence is that meaning becomes dynamic. A sentence is no longer a static package of truth conditions but an instruction to update a representation, and context is not a fixed backdrop but something the discourse itself continuously builds. Kamp published "The Paradox of the Heap" the same year, on vagueness and the sorites — the two 1981 papers together mark him out as a logician drawn precisely to the places where formal treatment is hardest.

From Discourse to Logic

He moved to the Institute for Natural Language Processing at the University of Stuttgart in 1988 as professor of formal logics and philosophy of language, and stayed. In 1993 he and Uwe Reyle published *From Discourse to Logic*, the full-dress treatment that turned DRT from a striking paper into a working system with enough coverage that computational linguists could implement it. Around it he ran a series of large collaborative projects — the European Dyana initiative from 1989 to 1995, the German special research centre SFB 340 on computational linguistic methods from 1989 to 2000, FraCaS from 1995 to 1997, and a DFG logic group from 2000 to 2002 — which is how a framework acquires a generation of users.

The research kept widening: temporal reference, presupposition, context dependence, prototypes and compositionality (with Barbara Partee, 1995), and non-literal language — metaphor and metonymy — which is unusual territory for a formal semanticist. His Jean Nicod Lectures were published by MIT Press as *Thinking and Talking about Things*, and a volume of his selected writings appeared as *Meaning and the Dynamics of Interpretation*.

Why Hans Is Called a Genius

The claim rests on one structural insight and the discipline to make it work. Kamp saw that the sentence was the wrong unit — that if you want to explain how people actually understand connected text, you need an intermediate representation that grows as the discourse grows, and that this representation can be given a rigorous model-theoretic interpretation rather than being waved at as psychology. Getting both halves is the hard part. Plenty of people had intuitions about discourse and context; the achievement was building something formal enough to be proved things about and implemented on a computer, while remaining recognisable as a model of what a listener does.

The supporting quality is range. Very few people carry a functional completeness proof in tense logic, an analysis of the sorites paradox, a theory of presupposition and a working account of metaphor in one career, and Kamp's self-description — not really a linguist, philosopher or logician — is the accurate report of someone who has been genuinely useful in three fields by refusing to be domesticated by any of them.

The counter-case is worth stating plainly. The 2024 Rolf Schock Prize was awarded to Kamp jointly with Irene Heim, and prizes get shared when more than one person arrives — the dynamic turn in semantics was in the air around 1980, and DRT is one solution among several rather than the only possible one. It is also notationally heavy, and rival dynamic frameworks have taken market share. And Kamp's reputation is entirely internal to a technical discipline: he has no public profile, no popular book, no idea that has escaped into general culture. Within formal semantics his standing is close to unquestioned. Outside it, almost nobody has heard of him.

Legacy

The honours have accumulated steadily: the Jean Nicod Prize in 1996, corresponding membership of the Royal Netherlands Academy of Arts and Sciences in 1997, an honorary doctorate, fellowship of the American Academy of Arts and Sciences in 2015, and the Rolf Schock Prize in Logic and Philosophy in 2024. The deeper legacy is that the questions changed shape. Before DRT, semantics asked what a sentence means; after it, the standard question is what a sentence does to a context — how it updates the state of an ongoing interpretation. Every system that tracks discourse referents across sentences, in theoretical linguistics or in computational language processing, is working in a tradition that begins with a paper Kamp wrote in 1981 about how we know who "he" is.

Achievements

• Jean Nicod Prize — 1996
• Held posts at University of Stuttgart and Massachusetts Institute of Technology
• Fields: computational linguistics, analytic philosophy and computing