### Update 2020-09-04 ### (See below for short abstract) I have scheduled a Q&A for [14:00 (UK time) on Tuesday 8 September][1]. The link to join the Q&A (on Google Meet) is [here][2]. If you join then please use your full name, for security purposes. ---------- ### Abstract ### Krifka (1990) famously observed that a sentence like 1 has an ’event-related reading’ 1.1 in addition to the expected ’object-related reading’ 1.2. 1. Four thousand ships passed through the lock last year. 1. There are four thousand ships, each of which passed through the lock last year. 2. Four thousand times last year, a ship passed through the lock. The accounts of event-related readings by Krifka (1990) and subsequently Doetjes & Honcoop (1997) use a covert determiner introducing a cumulative measure relation on events. Barker 1999 criticizes accounts that put the onus on the determiner in this way, based on the observation that _different_ seems to block the event-related reading in 2, suggesting that the semantics of the noun phrase is playing a role in the derivation of the event-related reading. 2. Four thousand different ships passed through the lock last year. However, Barker (1999) does not provide a semantics for _different_ that would predict this effect, and neither has anyone else, to our knowledge, in the context of event-related readings. Building on ideas by Musan (1995) and Barker (1999), we propose to account for object/event-related readings, including the semantic contribution of _different_, as follows: - Stages, rather than individuals, are basic. We make use of a substage relation, and individuals can be recovered as maximal stages on this relation. - Predicates hold of stages (of individuals) at a narrow level of granularity. For example, suppose that John is (in the pre-theoretical sense) the agent of event _e_. Then agent(_x,e_) holds only of the stage _x_ of John which is temporally coterminous with _e_. - Both object- and event-related readings involve quantification over stages. The difference is in the degree of maximality that stages require in order to be counted. - The word _different_ introduces a maximality requirement of its own, thereby blocking the event-related reading. [1]: https://www.timeanddate.com/worldclock/fixedtime.html?msg=Event-releated%20readings%20and%20degrees%20of%20difference&iso=20200908T14&p1=1233 [2]: https://meet.google.com/tpc-zzea-qdi