Frank Veltman (1996), Journal of Philosophical Logic 25(3):221–261 — DOI-10.1007/BF00248150
Key takeaways
- The slogan and the test. “You know the meaning of a sentence if you know the change it brings about in the information state of anyone who accepts the news conveyed by it.” An update system ⟨L, Σ, [ ]⟩ is additive (each sentence has a static content 0[φ], and σ[φ] = σ + 0[φ]) iff updates are total and satisfy Idempotence, Persistence, Monotony and Strengthening (Proposition 1.2). Only where one of these fails does the dynamic view say anything the static view cannot.
- Tests are not information. σ[might φ] is σ if φ is consistent with σ, and the absurd state otherwise. “Might φ” adds nothing about the world; it checks the state, so it can be accepted and later rejected (“Maybe it’s John… It’s Mary… Maybe it’s John” is incoherent). There are no “might-φ worlds”.
- Three validities. Updating the minimal state with the premises (valid₁), any state (valid₂), or classical acceptance-preservation (valid₃) coincide in additive systems and come apart otherwise. Valid₁ is non-monotonic and is characterised by Sequential Monotony, Sequential Cut and Reflexivity.
- Defaults. “Normally ψ” refines an expectation pattern (a preorder of worlds by normality) and is persistent; “presumably ψ” tests whether ψ holds in the optimal worlds. Restricted rules “if φ, normally ψ” need a frame of patterns, one per domain. Coherence (every domain has a normal world) and applicability (a default applies within s if no domain above s has all its normal worlds violating it) make more specific rules win, leave the Nixon diamond undecided, validate a defeasible modus tollens and keep Independence (an exception in one respect is normal in others), where Reiter, Delgrande, Asher and Morreau, and the inheritance algorithm of Horty, Thomason and Touretzky each differ somewhere.
Standing in the record
Filed on 2026-10-09 at the owner’s request from the manuscript
bibliography of 2026-10-09 (work what-survives-translation), which
considered it and dropped it from the final reference list, as one of the
dynamic-semantics works (with Stalnaker and Heim) the exchange named. See
the curation entry of that day.
Read on 2026-10-09 (NOTE-617). logic is primary because the paper
is a logic of default reasoning; philosophy-of-language and
linguistics hold the update-semantic account of meaning and of epistemic
modals. Its footnote 1 traces the dynamic notion of meaning to Stalnaker
(LIT-811 is his “Assertion”), Kamp, Heim’s dissertation
(LIT-838) and Gärdenfors, and its direct inspiration to Groenendijk
and Stokhof’s dynamic predicate logic, which the record does not hold.
With the Heim and Krifka readings of the same day it is a source of THEORY-162.