The goal of this ISF-funded workshop is to deal with the interaction between (a) domain widening (as originally discussed in Kadmon & Landman’s 1993 work on any, and as developed / debated in Chierchia 2013, and (b) the fact that some scalar alternative sensitive expressions give rise to mirativity (‘beyond expectations’ / low likelihood) or evaluativity (‘beyond norms / standards’) inferences,  as well as (c) the way these facts are related to negative (or positive) polarity-sensitivity. 

Examples of expressions like that are still and already (as in Lydia is still / already 3 years old), giving rise to inferences of earlier / later than expected or than the norm, respectively (cf. Krifka 2000), only and even (as in John only / even solved 6 problems in the exam), giving rise to inferences of small / large than expected or than the norm, respectively (cf. Klinedinst 2005, Zeevat 2008, Alxatib 2013, Greenberg 2018, 2019, see also copular all, as in All that John ate was a banana discussed, giving rise to a “smallness” effect, discussed in Homer 2019), and in years and until with punctual predicates (as in John hasn’t had a seizure in years / John hasn’t arrived until 5), giving rise to the inference that the event happened farther back in the past / farther forward into the future than expected or than the norm, respectively (cf. Iatridou & Zeijlstra 2019), an effect that in the case of until only arises when it applies to negated perfective predicates.

While some theories take such mirativity / evaluativity effects to be hardwired in the lexical entry of the relevant expressions, other suggested to derive (some of) them from mechanisms similar to ‘domain widening’. Such mechanisms can be roughly characterized as involving (a) universal quantification over alternatives, (b) taking the alternatives to be ordered on some scale relative to the ‘prejacent’ or the focused element, and/or (c) taking these alternatives to be contextually salient / relevant / entertainable, so they end up representing contextual expectations / contextual norms (and the prejacent or focused element ends up deviating from such contextual expectations or norms).

In this workshop we would like to take a general perspective at such observations and ideas, and examine the extent to which  they can be sharpened, generalized and contextualized within wider theoretical frameworks dealing with alternative sensitivity, contextual sensitivity and polarity sensitivity across constructions. 

Research questions include (but are not limited to) the following:

·       What are reliable diagnostics to distinguish between expressions for which mirativity / evaluativity is hardwired and those in which it is  derived in the way sketched above? (and what are other ways that mirativity / evaluativity effects can be derived (cf. ‘comparative noch’ in Umbach 2012, often translated as even: John is taller than Harry and Bill is noch taller))

·       What is the source of contextual saliency of alternatives giving rise to such derived mirativity / evaluativity effects? And are these sources uniform for all relevant expressions? Options include  prosodic and information structure factors (as with Chierchia’s ideas about the effects of contrastive stress with any), lexical ones (e.g. a lexicalized constraint on the substitution source in the algorithm for constructing alternatives in Fox & Katzir 2011, Katzir 2014), and perhaps some combination of these.

·       What is the principled connection between such derived mirativity / evaluativity effects and polarity sensitivity? Why is it that many mirative / evaluative particles are often polarity-sensitive?

·       Can ‘domain widening’ in the wide sense cover also operators which strictly speaking do not induce domain alternatives? (e.g. by widening sets of degrees on a scale, which correlate with the prejacents of only and even)?

·       Are there principled differences between the way mirativity / evaluativity effects is derived with expressions which obligatorily trigger alternatives (as with any under Chierchia 2013 for any, Iatridou & Zeijlsra 2019 on in years / punctual until) and those which operate over alternatives (e.g. only)?

·       Are there principled differences between overt and covert alternative sensitive expressions regarding mirativity / evaluativity effects derived using such mechanisms, e.g. between only, which trigger mirativity/ evaluativity effects and exh, which was reported not to trigger such effects (cf. Crnič 2012, Greenberg 2019)?

·       Are there mirativity / evaluativity effects derived from domain widening with expressions which are NOT analyzed as alternative sensitive (cf. Zanuttini & Portner 2003 on exclamatives)

·       How can mirativity (‘beyond expectations’) and evaluativity (‘beyond standards’) effects be precisely distinguished? Are the former subsumed by the latter? Does this distinction correlate with the distinction between ‘distributional’ and ‘functional’ standards with gradable adjectives (discussed in e.g. Kagan & Alexjaenko 2010, Bylilina 2011, Solt 2012). Are such standards really operative with alternative sensitive constructions which are not themselves gradable (cf. Axatib 2013 Greenberg 2015, 2019)?