Research

The Present Perfect Puzzle refers to the fact that in English (and some other languages), the present perfect cannot be used with a past temporal adverbial (e.g. #Mary has watched the football game yesterday), despite the fact that the event time does fall into the past time denoted by the adverbial. In addition, languages with the present perfect differ in whether they have this constraint.

My analysis: the present perfect has the weaker version of Extended Now semantics (Gronn and von Stechow 2017), which allows the reference time of a present perfect sentence to be either an Extended Now interval that includes the speech time, or a past interval. In principle, the present perfect should be compatible with the reference time being a past interval provided by the adverbial. However, in languages that have the Present Perfect Puzzle effect (English, some variaties of Spanish, and several Romance languages in Italy), the anaphoric past tense is a presuppositionally stronger alternative to the present perfect. By Presupposed Ignorance Principle (Spector and Sudo 2017), the past tense must be preferred over the present perfect.

On the other hand, in languages that do not have the Present Perfect Puzzle (the spoken version of the standard variety of French, German and Italian), the anaphoric simple past (anaphoric past tense + perfective aspect, which competes with the perfective present perfect in these languages) is absent, allowing the present perfect to take over.

This analysis has the advantage of not relying on semantic contradiction (as some previous analyses do), and makes a much better prediction of the crosslinguistic variation given the distribution of the alternatives for the past reading.

The literature on tense has two major groups of analyses for the (unembedded) past tense: pronominal (i.e. anaphoric) and existential. However, there is a group of data where the English past tense presents conflicting evidence for both analyses, namely, the obligatory use of the English past tense in contexts without an obvious antecedent.

Pointing at a church: Who built this church?

There are several proposals in the previous literature that address this issue, and they mostly fall into 2 groups: (i) lexical ambiguity between an anaphoric and present perfect like reading, where the English past tense may spell out a present reference time and a perfect (i.e. τ(e)≺t) aspect (Kratzer 1998); (ii) lexical ambiguity between an anaphoric and a non-anaphoric reading (Michaelis 1994: English past tense is unmarked for anaphoricity; Gronn and von Stechow 2016: freely inserted covert definite/indefinite articles; Matthewson et al. 2019: the English past tense is lexically ambiguous between an anaphoric reading and an existential reading).

As pointed out by Matthewson et al. (2019), if the non-anaphoric past tense has some sort of present perfect reading, it is not clear why the English present perfect is actually prohibited in those examples. Matthewson et al. also note that the distribution of these non-anaphoric past tense is restricted, which they attribute to salient domain restriction.

I argue instead that these examples reflect the uniqueness presupposition of the English past tense. The non-anaphoric past tense is felicitous in contexts where the existence and uniqueness (with respect to a certain temporal property) of the reference time can be guaranteed. The presupposition projection pattern also matches that of unique definites, instead of existentials or indefinites. In other words, the English past tense is more similar to the definite article (plus an NP) than to pronouns. In addition, in languages such as (the standard variety of) German, the past tense does not have this ambiguity and behaves as expected from an anaphoric past (Kratzer 1998). The German past tense also suggests that the non-anaphoric past in English cannot be the result of some sort of pragmatic accommodation of the antecedent, since such mechanisms would likely to be universal. In the recent literature on definite descriptions, it has been accepted that uniqueness and anaphoricity are separate sub-concepts of definiteness, and that languages may use distinct morphology for each. The English and German past tenses together show that we have similar divisions in the temporal domain, and may provide insights to future research on the crosslinguistic variation of tense/aspectual constructions.

The tenseless analyses of Mandarin Chinese (Klein et al. 2000; Lin 2006, 2007, 2010) argue that the temporal interpretation of Mandarin Chinese comes from the context and the help of past-shifting aspect particles (e.g. the perfective -le). The tensed analyses of Mandarin Chinese include Sybesma (2007), which argues that tense is an agreement particle and languages differ in whether it is morphologically explicit. Recent analyses such as Sun (2014) argue for a covert non-future tense along the lines of Matthewson (2006).

There is no consensus in the literature on the status of the rich collection of aspect markers in Mandarin Chinese, especially the perfective markers. There is also a lot of disagreement in the judgements, possibly due to misclassification of the often homophonous particles. I argue that Mandarin Chinese has three particles with the standard perfective semantics (∃ e [τ(e)⊆ t]): (i) the perfective sentence-final -le; (ii) the verbal -le; (iii) -guo. The perfective sentence-final -le is presupposition-less; the verbal -le is an anaphoric particle, presupposing an event antecedent; -guo presupposes non-resultativeness (that there does not exist a result state of the -guo-marked event that answers the topic question, cf. Portner (2003)).

The previous literature established the set of ‘perfect’ readings, including experiential/existential, resultative, recent past, hot news, the Present Perfect Puzzle, the lifetime effect, and the lack of narrative progression (McCawley 1971; McCoard 1978; Portner 2011, a.o.). On the other hand, it has been noted that the present perfect in some languages other than English, as well as similar tense/aspect constructions in other languages, falls into the category of a ‘general purpose past perfective’ (Bertrand et al. 2022), sharing some properties with the English present perfect while not being subject to constraints such as the lifetime effect and the Present Perfect Puzzle.

I propose that the general-purpose past perfectives are presuppositionally neutral tense/aspect constructions that allow the standard past perfective reading. If a language has presuppositionally stronger alternatives for the past perfective (anaphoricity, uniqueness, etc.), by the Presupposed Ignorance Principle (PIP), the standard, pre-suppositionally neutral, past perfective will be felicitous only if the pre-suppositionally stronger alternatives cannot be used. I argue that this competition is the source of many of the perfect readings observed. I further argue that the cross-linguistic variation in this respect follows from the available alternatives languages have. I illustrate this idea with three groups of languages: (i) English; (ii) French, German, Italian; and (iii) Mandarin Chinese, each illustrating a different set of alternatives available. This analysis allows me to decompose various perfect readings that come from different sources and make better predictions regarding which of these readings a tense/aspect construction in a given language has. The alternatives examined are summarized in the table below.

This project focuses on three aspects of before- and after clauses in Mandarin Chinese: (i) temporal semantics of the NONFUTURE tense (Sun (2014), cf. Matthewson (2006) and Aonuki (2021)); (ii) overt aspectual marking and potential covert aspect semantics in before- and after-clauses, in comparison to that in matrix and relative clauses; (iii) the interaction of before and after with different kinds of eventualities and change-of-state/inchoative particles. More details to follow as I progress.

Parts of this project include: (i) to establish the bias profile of Mandarin Chinese polar question forms; (ii) provide a formal analysis of the strength of evidential and epistemic biases which affect the distribution of certain polar question forms; (iii) provide diagnostics for high negation in Mandarin Chinese to cover the full range of data. Details will be updated as we progress.