I present a biased look at data relating to the form, meaning, understanding, and acquisition of comparative sentences with more. I highlight two major points: (i) comparatives provide a potentially unique case study for examining the interplay of current ideas in formal semantics, generative syntax, and cognitive psychology; (ii) we can give unified explanations for the otherwise disparate phenomena observed here only by interpreting our semantic theory along explicitly cognitive lines. The upshot is not so much a rejection of more traditional views as providing a window on their scientific limitations.

Wellwood, A. (2019). What more is. Philosophical Perspectives.

official link (open access)

Comparatives can invoke various dimensions for comparison, but not anything goes: more coffee invokes volume or weight, but not temperature, while more coffees invokes number, but not volume or weight. In general, the extant literature assumes that the difference between more coffee/coffees reflects a morphosyntactic ambiguity of more, such that it spells out MUCH-ER with bare nouns, and MANY-ER with plural nouns. Semantically, MUCH introduces a variable over measure functions, with constraints, whereas MANY introduces a cardinality function. I argue for an alternative, univocal theory based on the decomposition MUCH-ER, and account for the observed patterns of constrained variability by means of a stronger condition on the selection of measure functions than has previously been proposed.

Wellwood, A. (2018). Structure preservation in comparatives. Semantics and Linguistic Theory, 28.

official link (open access)

Comparative constructions like More people have been to Russia than I have are reported to be acceptable and meaningful by native speakers of English; yet, upon closer reflection, they are judged to be incoherent. This mismatch between initial perception and more considered judgment challenges the idea that we perceive sentences veridically, and interpret them fully. It is thus potentially revealing about the relationship between grammar and language processing. This paper presents the results of the first detailed investigation of these so-called 'comparative illusions'. We test four hypotheses about their source: a shallow syntactic parser, some type of repair by ellipsis, an incorrectly-resolved lexical ambiguity, or a persistent event comparison interpretation. Two formal acceptability studies show that speakers are most prone to the illusion when the matrix clause supports an event comparison reading. A verbatim recall task tests and finds evidence for such construals in speakers' recollections of the sentences. We suggest that this reflects speakers' entertaining an interpretation that is initially consistent with the sentence, but failing to notice when this interpretation becomes unavailable at the than-clause. In particular, semantic knowledge blinds people to an illicit operator-variable configuration in the syntax. Rather than illustrating processing in the absence of grammatical analysis, comparative illusions thus underscore the importance of syntactic and semantic rules in sentence processing.

Additional materials related to the published experiments and our preliminary experiments can be found on Github.

Wellwood, A., R. Pancheva, V. Hacquard, and C. Phillips. (2018). The anatomy of a comparative illusion. Journal of Semantics, 35(3).

read pdf | official link

Recent proposals in the semantics literature hold that the negative comparative less and negative adjectives like short in English are morphosyntactically complex, unlike their positive counterparts more and tall. For instance, the negative adjective short might decompose into little tall (Rullmann 1995; Heim 2006; Büring 2007; Heim 2008). Positing a silent little as part of adjectives like short correctly predicts that they are semantically opposite to tall; we seek evidence for this decomposition in language understanding in English and Polish. Our visual verification tasks compare processing of positive and negative comparatives with taller and shorter against that of arguably less symbolically-rich mathematical statements, A > B, B < A. We find that both language and math statements generally lead to monotonic increases in processing load along with the number of negative symbols (as predicted for language by e.g. Clark and Chase 1972). Our study is the first to examine the processing of the gradable predicates tall and short cross-linguistically, as well as in contrast to extensionally-equivalent, and putatively non-linguistic stimuli (cf. Deschamps et al. 2015 with quantificational determiners).

Tucker, D., B. Tomaszewicz and A. Wellwood. (2018). Decomposition and processing of negative adjectival comparatives. In The semantics of gradability, vagueness, and scale structure: Experimental perspectives, Cognition and Mind series.

read pdf | official link

The event analysis is only rarely incorporated into degree-theoretic treatments of adjectival comparatives. I propose a neodavidsonian account of predications like Ann was happy that involves quantification over both states and events. This 'double-eventuality' analysis is motivated primarily by how stage-level gradable adjectives interact with temporal for-phrases in two classes of comparatives, which I differentiate as 'low' versus 'high' attachment of the comparative morpheme. Low attachment comparatives express canonical degree readings (more available), while high attachment involve comparing numbers of occasions (available more). I resolve these patterns by positing a stative core for adjectives, and the possibility of mapping properties of states to properties of events. Low attachment interpretations, then, involve comparison of states, and high involve comparison of (pluralities of) events. I show that the analysis extends to other cases where states and events can do work for adjectives outside of comparatives.

Wellwood, A. (2016). States and events for S-level gradable adjectives. Semantics and Linguistic Theory 26.

official link (open access)

How much meaning can a morpheme have? Syntactic and morphological analyses generally underdetermine when distinctions in meaning between two forms are due to (i) the presence of an additional syntactic head or to (ii) different information coded on the same head. Surveying patterns across hundreds of languages, Bobaljik (2012) hypothesizes that superlative forms universally consist of a comparative morpheme plus an additional superlative morpheme, e.g., tallest is underlyingly [ SUP [ CMPR [ TALL ] ] ]. Bobaljik's hypothesis includes, in part, a speculative proposal for a universal limit on the semantic complexity of morphemes. We offer a concrete basis for this proposal, the 'No Containment Condition' (NCC). The NCC is a constraint on grammars such that they cannot contain a certain semantic representation for a unitary head, if that representation can be constructed out of the semantic representations of two heads. Illustrating the proposal, we take Bobaljik's analysis of forms like tallest further, into [ [ [ CMPR SUP ] MUCH ] TALL ]. Based in semantic analysis, our suggestion introduces Bresnan's (1973) classical analysis of comparatives into the decomposition of superlatives.

Dunbar, E. and A. Wellwood. (2016). Addressing the two interface problem: comparatives and superlatives. Glossa.

official link (open access)

This paper explores the hypothesis that all comparative sentences—nominal, verbal, and adjectival—contain instances of a single morpheme that compositionally introduces degrees. This morpheme, sometimes pronounced much, semantically contributes a structure-preserving map from entities, events, or states, to their measures along various dimensions. A major goal of the paper is to argue that the differences in dimensionality observed across domains are a consequence of what is measured, as opposed to which expression introduces the measurement. The resulting theory has a number of interesting properties. It characterizes the notion of ‘measurement’ uniformly across comparative constructions, in terms of non-trivial structure preservation. It unifies the distinctions between mass/count nouns and atelic/telic verb phrases with that between gradable and non-gradable adjectives. Finally, it affords a uniform characterization of semantically anomalous comparisons across categories.

Wellwood, A. (2015). On the semantics of comparison across categories. Linguistics & Philosophy, 38(1).

read pdf | official link

Determining the semantic content of sentences, and uncovering regularities between linguistic form and meaning, requires attending to both morphological and syntactic properties of a language with an eye to the notional categories that the various pieces of form express. In this dissertation, I investigate the morphosyntactic devices that English speakers (and speakers of other languages) can use to talk about comparisons between things: comparative sentences with, in English,more… than, as… as, too, enough, and others. I argue that a core component of all of these constructions is a unitary element expressing the concept of measurement.

The theory that I develop departs from the standard degree-theoretic analysis of the semantics of comparatives in three crucial respects: First, gradable adjectives do not (partially or wholly) denote measure functions; second, degrees are introduced compositionally; and three, the introduction of degrees arises uniformly from the semantics of the expression much. These ideas mark a return to the classic morphosyntactic analysis of comparatives found in Bresnan (1973), while incorporating and extending semantic insights of Schwarzschild (2002, 2006). Of major interest is how the dimensions for comparison observed across the panoply of comparative constructions vary, and these are analyzed as a consequence of what is measured (individuals, events, states, etc.), rather than which expressions invoke the measurement.

This shift in perspective leads to the observation of a number of regularities in the mapping between form and meaning that could not otherwise have been seen. First, the notion of measurement expressed across comparative constructions is familiar from some explications of that concept in measurement theory (e.g. Berka 1983). Second, the distinction between gradable and non-gradable adjectives is formally on a par with that between mass and count nouns, and between atelic and telic verb phrases. Third, comparatives are perceived to be acceptable if the domain for measurement is structured, and to be anamolous otherwise. Finally, elaborations of grammatical form reflexively affect which dimensions for comparison are available to interpretation.

Wellwood, A. (2014). Measuring Predicates. University of Maryland linguistics dissertation.

read pdf

Bresnan (1973) posited that more is uniformly analyzed as much-er, whether it appears with adjectives (more intelligent, redder) or nouns (more soup). On the earliest degree-semantic analysis of such constructions, much appears but is semantically inert: it serves to morphologically mark the presence of the degree argument which is introduced by adjectives and nouns (Cresswell 1976). I present an alternative analysis, one suggested by Cresswell himself: on this account, the degree argument is introduced by much. I first show how the interpretation of this morpheme as a structure-preserving mapping to the domain of degrees is motivated by data from nominal and verbal comparatives, and then how it extends to adjectival comparatives. To accomplish this, I argue that adjectives express predicates of states, and interact with degrees only in composition with much. The upshot is a theory in which much universally provides the mapping to degrees for comparison by more, regardless of the syntactic category it combines with.

Wellwood, A. (2012). Back to basics: more is always much-er. Proceedings of Sinn und Bedeutung 17.

read pdf | official link

This squib investigates parallels between nominal and verbal comparatives. Building on key insights of Hackl (2000) and Bale & Barner (2009), we show that more behaves uniformly when it combines with nominal and verbal predicates: (i) it cannot combine with singular count NPs or perfective telic VPs; (ii) grammatical properties of the predicates determine the scale of comparison—plural marked NPs and habitual VPs are compared on a scale of cardinality, whereas mass NPs and perfective (atelic) VPs are (often) compared along non-cardinal, though monotonic, scales. Taken together, our findings confirm and strengthen parallels that have independently been drawn between the nominal and verbal domains. In addition, our discussion and data, drawn from English, Spanish, and Bulgarian, suggest that the semantic contribution of more can be given a uniform analysis.

Wellwood, A., V. Hacquard, and R. Pancheva. (2012). Measuring and comparing individuals and events. Journal of Semantics, 29(2).

read pdf | official link