Our talk is concerned with Minimal Sufficiency (MS) as in (1) and (2). (Grosz 2011 coined the term, but the data had been around for decades.)

1. Tylenol alone will help here already.
2. If the weather is just so-so, I will go hiking already.

MS sentences share two characteristics: (i) conditions that are surprisingly easy to fulfill suffice to for a a state-of-affairs to obtain; (ii) a scalarly interpreted 'only' word takes local scope (alone in (1), just in (2)). The existing literature on MS has largely focused on these two properties, and on the correct modeling of "upward scalarity" (Berends & Kaufmann 2009, Grosz 2011, Coppock & Lindahl 2014, Liu 2017, Panizza & Sudo 2020, Wimmer 2022). We add the following major empirical points to the discussion: (i) a split between event-internal and event-external MS, with different, though similar notions of sufficiency underlying these two types (event-internal sufficiency providers bear theta-roles, event-external ones do not); (ii) some languages display optional, some obligatory particle doubling (concord); (iv) upward scalarity is mereology-based with event-internal MS, and informativeness-based with event-external MS; (v) Chinese offers particularly interesting and recalcitrant data defying a standard conditional treatment for at least some biclausal MS constructions.
Our analysis will argue that, particularly for the recalcitrant cases from Chinese, an implementation in terms of correlative constructions is needed (Dayal 1995, 1996). More conservative analyses remain available and, we suggest, are more intuitively justified for the simpler cases. On the syntactic side, we offer a morphosyntactic account of the phenomenon of scalar particle doubling, which is crucial for the Chinese case. Lastly, we will demonstrate how the syntactic side of our proposal makes the right predictions concerning the discovery of new MS data types.

Abstract

Abstract

We are going to present our ongoing work on Chinese constructions as in (1).

1. Xiwang

   hope

   mei-you

   not-exist

   shi

   matter

   cai

   CAI

   hao.

   good

   
   lit.: 'Only if [hopefully] nothing is the matter (will it be) good.'
   'If only there will be no trouble!' (cf. Hole 2004:263)

We follow Kaufmann (2017) in calling a sentence like (1) a conditional evaluative construction (CEC): (some kind of) conditional that serves to make a modal statement, and whose consequent boils down to an evaluative predicate -- hao 'good' in (i). CECs are attested in Chinese (Hole 2004, 2006), Japanese (Kaufmann 2017, Yang 2020), Korean (Chung 2019, 2020) and Vietnamese. As slightly different functional elements in the four unrelated, or only loosely related, languages converge in bringing about an overall analogous function, we like to think of CECs as a potential sprachbund phenomenon.
The goal is to discuss prospects for a syntax-and-semantics for Chinese CECs, building on already available analyses like Kaufmann's (2017). A compositional semantics for Chinese CECs minimally needs to make well-motivated assumptions about the interactions of (a) the antecedent (b) the evaluative predicate and (c) additional functional material, most crucially for us scalar particles like cai 'only.then' as in (1) and its frequent alternate jiu 'already.then'. At least for cai, it is fairly clear that it induces universal modal force just like 'only if' does in Korean (Chung 2019).
A special challenge is posed by cases of modal concord (MC) like (1): the seemingly redundant co-occurrence of bouletic xiwang 'hope' in the antecedent and bouletic hao 'good' in the consequent. For more basic CECs like [if p, good], we will follow Kaufmann (2017) and Yang (2020) in treating [if p] as a definite description of worlds, and 'good' as a predicate thereof (Sode 2018). For MC-CECs like [hope p, good], we would like to argue that they are conjunctions of two roughly equivalent modal claims with a correlate anaphoric uptake of p in the consequent. Klinedinst & Rothschild's (2012) discussion of two types offers a potential way of unifying the MC- and the more basic cases as involving two types of conjunction.

Abstract

Abstract