Maribel Romero

Besides their ordinary cardinal and proportional meanings, many and few have been argued to allow for a ‘reverse proportional’ reading. This reading has later been characterised in two opposite directions: Cohen’s reading where the proportion \ matters and Herburger’s where it does not. We develop a compositional analysis that derives the correct truth conditions for both characterisations of Westerståhl-style sentences while maintaining conservativity, assuming a standard syntax/semantics mapping and reducing their context-dependence to mechanisms independently needed for degree constructions in general. In a nutshell, mirroring the decomposition of other degree expressions like tall, many is decomposed into the parametrized determiner many and the operator POS, where POS combines with a contextually salient comparison class C matching the alternatives triggered by some element X\ in the sentence. Non-reverse readings obtain when X\ is external to the original host NP and reverse readings when X\ is internal to the host NP. Cohen’s truth conditions for Westerståhl-style sentences are derived as a reverse proportional reading and Herburger’s interpretation as a sub-case of the non-reverse cardinal reading.

Besides their cardinal and proportional readings, many and few have been argued to allow for a ‘reverse’ proportional reading that defies the conservativity universal. Recently, an analysis has been developed that derives the correct truth conditions for this reading while preserving conservativity. The present paper investigates two predictions of this analysis, based on two key ingredients. First, many is decomposed into a determiner stem many and the degree operator POS. This predicts that other elements may scopally intervene between the two parts. Second, non-reverse and reverse readings arise from the versatile association possibilities of POS, as independently witnessed in the grammar. In languages in which the same versatile association profile is granted to the superlative degree operator -est, reverse readings are expected to arise for most as well. Both predictions are borne out.

In (1b), for the most part induces a so-called Quantificational Variability Effect (QVE) on the NP the linguists from the East Coast, yielding roughly the interpretation ‘most of the linguists from the East Coast came to NELS’. We claim that the two constructions above differ in the domain where they apply, producing similar but not identical quantificational interpretations over the NP. In particular, we argue that most of the NPs applies to the nominal domain, while for the most part applies to the verbal domain. Our claim is based on two sets of novel semantic data. First, we show that the distribution of most of the NPs is parallel to that of all the NPs in terms of its selective compatibility with collective predicates. To account for this data, we extend Brisson’s (1998, 2003) analysis of all the NPs to most of the NPs, concluding that most is an ∃-quantifier introducing a group of a certain proportion. Second, we show that, when for the most part gives rise to a QVE on a definite NP, the collective interpretation is not available. We develop a semantic analysis of for the most part as a verbal modifier that explains the lack of collective readings and that extends to interpretations other than QVE. The structure of the paper is as follows: in section 2, we introduce some general background on events and distributivity that are relevant to the current paper. In section 3, we propose the analysis of most of the NPs, followed by the analysis of for the most part in section 4. Section 5 concludes the paper and discusses further issues.

Superlative adjectives accompanied by certain modal adjectives like possible (e.g. John bought the largest possible present) are ambiguous between a reading where possible is a regular noun modifier and a reading paraphrasable as ‘as Adj as possible’, called ‘modal superlative reading’. Three interesting restrictions have been observed in the literature. First, possible and some other adjectives ending in -able, but not potential and probable, support the latter reading. Second, when the modal adjective appears postnominally, only the modal superlative reading is available. Third, prenominal possible needs to be in a local configuration to -est in order for the modal superlative reading to arise. Using LF structures independently motivated for degree constructions, the present paper develops a compositional semantic analysis of the modal superlative reading, makes correct new predictions concerning this reading, and—by reconciling previous, opposed syntactic analyses—allows us to derive the three empirical restrictions above. The key innovations are: (i) the previously proposed constituent [possible ▲ellipsis] is interpreted as an amount relative clause, and (ii) this constituent is treated as overtly expressing the comparison class argument of -est

Montague’s analysis of the well-known temperature paradox poses a problem for Gupta’s syllogism, whose surface syntax differs from the temperature syllogism in the addition of the intensional adverb necessarily. Lasersohn (2005) argues that the puzzle arising from these syllogisms can be solved if one adopts the Fregean presuppositional treatment of definite descriptions, and concludes that the temperature-Gupta puzzle provides an argument in favor of such treatment. This paper shows that the analysis of definite descriptions is in fact orthogonal to the puzzle. Instead, it will be shown that the differences between the two syllogisms stem from the temporal interpretation of their premises.

Superlative sentences with modal modifiers like possible give rise to the so-called 'modal superlative reading' . The present paper uses this reading to investigate an open issue in degree constructions: whereas two different lexical entries have been argued to exist for the comparative morpheme -er , it is not clear whether two entries are needed for the superlative morpheme -est. The paper argues that, with 3-place –est, otherwise unmotivated syntactic material would to have to be postulated and that, even with this material, not all modal superlative examples would be assigned correct truth conditions. In contrast, 2-place -est can generate the modal superlative reading in all the cases, as shown in Romero . Modal superlative sentences, thus, provide evidence that 2-place –est is needed in the grammar

Specificational sentences show Connectivity Effects (Akmajian 1970, Higgins 1979, Halvorsen 1978, Jacobson 1994, among others). For example, an NP like no man embedded in a relative clause in general cannot bind a pronoun outside the relative clause, as illustrated in (3a); but in specificational copular sentences this binding is possible, as in (3b). This effect is called Variable Binding Connectivity. Similarly, the NP a unicorn cannot be interpreted de dicto with respect to the embedded verb look for in (4a); but a de dicto reading is possible in the specificational sentence (4b) (Opacity Connectivity). Connectivity will be used in this paper as a diagnosis for specificational sentences

This paper describes a novel pedagogical software program that can be seen as an online companion to one of the standard textbooks of formal natural language semantics, Heim and Kratzer (1998). The Penn Lambda Calculator is a multifunctional application designed for use in standard graduate and undergraduate introductions to formal semantics: Teachers can use the application to demonstrate complex semantic derivations in the classroom and modify them interactively, and students can use it to work on problem sets provided by the teacher. The program supports demonstrations and exercises in two main areas: (1) performing beta reduction in the simply typed lambda calculus; (2) application of the bottom-up algorithm for computing the compositional semantics of natural language syntax trees. The program is able to represent the full range of phenomena covered in the Heim and Kratzer textbook by function application, predicate modification, and lambda abstraction. This includes phenomena such as intersective adjectives, relative clauses and quantifier raising. In the student use case, emphasis has been placed on providing “live” feedback for incorrect answers. Heuristics are used to detect the most frequent student errors and to return specific, interactive suggestions.

Preposed negation yes/no (yn)-questions like Doesn''t Johndrink? necessarily carry the implicature that the speaker thinks Johndrinks, whereas non-preposed negation yn-questions like DoesJohn not drink? do not necessarily trigger this implicature. Furthermore,preposed negation yn-questions have a reading ``double-checking'''' pand a reading ``double-checking'''' p, as in Isn''t Jane comingtoo? and in Isn''t Jane coming either? respectively. We present otheryn-questions that raise parallel implicatures and argue that, in allthe cases, the presence of an epistemic conversational operator VERUMderives the existence and content of the implicature as well as thep/ p-ambiguity.

In English, a non-wh-question may have a disjunctive phrase explicitly providing the choices that the question ranges over. For example, in (1), the disjunction or not indicates that the the choice is between the positive and the negative polarity for the relevant proposition, as spelled out in the yes/no (yn)-question reading (2) and in the answers (2a,b). Another example is (3). The disjunction in (3) can be understood as providing the choices that the question ranges over, hence giving rise to the alternative (alt-)reading in (4) and eliciting the answers in (4a,b). (Cf. Karttunen (1977) and Higginbotham (1993) for the semantics of yn/alt-questions).

There are two main approaches to the scopal properties of the N’-restrictors of which-phrases. One line attributes widest scope within the interrogative clause to the entire which-phrase, outside the question formation operator, often assumed to reside in C0. The result is Karttunen ’s question denotation --exemplified in --,1 whose distinctive feature is that the semantic contribution of the N’- restrictor of the which-phrase is represented outside the so-called question nucleus, i.e., outside the subformula “p=…”. The second main avenue interprets the N’-restrictor of which-phrases somewhere inside the question nucleus, that is, under the scope of the question formation operator and possibly under other, further embedded operators. The outcome is Hamblin’s --or unselective binding—question meaning, as in. This paper will be concerned with this second line, which I will refer to as the base position line

Connectivity, found in a number of constructions involving typically a trace of movement or gap, is the effect by which a constituent behaves grammatically as if it occupied not its surface position but the position of the gap. The phenomenon is central to the debate between defendants of Direct Compositionality –where the semantics is read off the ‘visible’, surface syntax– and the defendants of the so-called Logical Form (LF) –according to which semantics is computed on an abstract syntactic representation, LF, obtained after applying some transformations to the surface syntax. The present paper is concerned with connectivity in specificational copular sentences. A simple specificational copular sentence is given in (1), where the post-copular constituent Smith identifies the actual value of the subject N(oun) P(hrase) the murderer. More complex examples reveal connectivity effects, as shown in (2)-(4) (Akmajian 1970, Higgins 1973, Halvorsen 1978, among others).