Search Results

This chapter provides an overview of the typical pathways for the development of negation, focusing on the languages of Europe and the Mediterranean. It considers the frequency and causes of Jespersen’s cycle, the emergence of new markers of negation from emphatic constructions that eventually come to replace the original marker of negation. It looks at the sources of new negative markers, from minimizers (‘not a bit’) and generalizers (‘not in any possible way’) and at the historical progress of the negative cycle that has occurred in almost every language in western Europe, showing considerable differences in detail despite overall commonality. The chapter also considers how expression of negative indefinites (e.g. ‘nothing, nowhere’) changes over time, including changes in negative concord systems. Common developments here include, for instance, the tendency for items to become more ‘negative’ (the quantifier cycle) and for free-choice items (‘anything you like’) to spread into negative contexts.

This chapter explores some determining factors for the development of indefinites participating in Negative Concord from Latin to Romance (with special attention to Old French and Old Italian). In particular, the discussion concentrates on the subclass of Romance n-words formed with the negative morpheme ne- / ni- < Latin nec. In its role as building block of the new indefinites, nec is a negative scalar focus particle meaning ‘even not’ (a use that first emerges in Post-Classical Latin); the cardinal numeral unus provides the scalar end point. With nec specific syntactic patterns are observed that quite systematically result in the redundant expression of negation, this occurring already in Classical Latin. These structures are easily prone to reanalysis once the prerequisites for Negative Concord are present. The chapter provides a reconstruction of how the facts emerging from Latin may help toward an improved understanding of optional Negative Concord in Early Romance.

Until-phrases are polarity sensitive when construed with telic event descriptions and polarity insensitive when construed with atelic event descriptions. They are interpreted like temporal frame adverbials in the former case and durative adverbials in the latter case. In addition, until is often referred to as “punctual until” in the former case and “durative until” in the latter case. The punctual and durative readings of until have been attributed to either lexical ambiguity or scopal ambiguity. This chapter examines lexical ambiguity and its implications for the interpretation of negation, and discusses the semantics of punctual until. It incorporates scalarity into the meaning of punctual until and considers negative polarity items as opposed to positive polarity items.

This chapter is a study of Latin indefinites in direct-negation contexts. These indefinites are interesting from a theoretical point of view because of their extreme dependence on the surrounding structural conditions, and because of the variety of their instantiations in different linguistic systems. Two phenomena of Latin grammar with wide-ranging implications for the development of Romance indefinites are discussed: the syntax of negation and the diachronic pathways followed by indefinites interacting with it. Latin is a Double Negation language, whereas Early Romance exhibits Negative Concord. The study proposes that this typological shift is linked to another major change from Latin to Romance, namely the change from OV to VO. Late Latin is analyzed as a ‘concealed’ nonstrict Negative Concord language, in which restrictions in the use of the ‘old’ negative indefinites emerge, as well as new patterns with (new) negative-polarity items.

This chapter argues that, while the creation of indefinites from generic nouns is grammaticalization in the form of upwards reanalysis from N to R, the quantifier and free-choice cycles do not in fact constitute instances of grammaticalization. Indefinites restricted to stronger negative-polarity contexts are not more functional than indefinites licensed in weaker negative-polarity contexts. Rather, it is argued that implicational semantic features requiring roofing by different types of operators situated in the Q head of indefinites, and in particular the way they are acquired in first language acquisition, are responsible for the diachronic developments. Negative concord items arise through an acquisitional mechanism maximizing the number of agreement relations in the acquired grammar consistent with the primary linguistic data.

Chapter 4 discusses presuppositionality, concord and (in a preliminary way) intervention phenomena as they relate to polarity. It has been noticed in the literature that some NPI licensing environments (overt only, or verbs like surprise) are DE only in a weak way (‘Strawson-DE’). However, adopting Strawson-DE as a licensing condition gives rise to several empirical and conceptual problems. It is argued that such problems can be overcome by splitting each clause in an assertive and a presuppositional component (á la Karttunen and Peters) under the assumption that exhaustification for weak NPIs only looks at the assertive component. This paves the way for an exhaustification based approach to strong NPIs: the latter are characterized by the fact that all components of their meaning (assertive, presuppositional and implicational) must be factored in (by the exhaustification process). This idea explains directly why strong NPIs can only be licensed by strongest members of negative scales (e.g. by no but not by few or less than n). Finally concord phenomena and N-words are analyzed as an overt syntactic realization of exhaustification. The observation that N-words tend to be indefinites is argued to be derived by the intervention of positive implicatures induced by indefinites under negation.

This chapter shows that, across languages, indefinites show a tendency to become increasingly restricted to ever-stronger negative polarity contexts, eventually becoming restricted to the scope of negation and beginning to enter negative-concord relations. It is shown that developments affecting indefinites can be grouped into two types of development: a quantifier cycle and a free-choice cycle. In the former, indefinites gradually change from (more) positive to (more) negative, in the latter, original free-choice indefinites become NPI-indefinites. It is also shown that negative quantifiers do not typically arise through either development, but through univerbation with a former negative particle. Apparently countercyclic developments can be observed as well. The chapter concludes with an overview of the typology of negative concord, and the diachronic connections between the types of negative concord.

Presupposition is a systematic discourse‐anchoring device in the semantics of natural language. Presuppositions are generated by the preconditions of predicates. Operational criteria are provided. Data show that two kinds of falsity must be distinguished, one that preserves and one that cancels presuppositions. This leads to a trivalent presuppositional logic. For predicate logic, this takes two forms: the basic‐natural logic of Chapter 3 and the Square of Opposition extended with a presuppositional component.

The main goal of this chapter is critically to address from a comparative perspective some of the most important issues in Romance negation, covering such topics as the source of different types of negators and negative polarity items; the cyclical interaction of reanalysis and analogy in the emergence and grammaticalization of Romance negators; issues surrounding the unidirectional development of Romance negative structures inherent in Jespersen’s Cycle and the discreteness of the three proposed stages; the development and loss of negative concord and the licensing of negative polarity items; the development and specialization of different negative markers in accordance with temporal, aspectual, modal and finiteness distinctions; multiple sentential positions for negation; negative raising; pleonastic negation. Specific topics dealt with include: form(s) and position(s) of the negative marker; Jespersen’s Cycle; postverbal negators (French, northern Italian dialects, Catalan); position of postverbal negators; form of postverbal negators; new preverbal negative markers; interaction between negation and verbal forms; negation and modality; negation and aspect; N-words and negative concord; negation and focus.

This chapter discusses the Composed Quantifier Argument, a widely accepted and apparently devastating argument against a syntactic view of Classical NEG Raising (NR), and argues that it is faulty and in no way conclusive in view of the general framework for understanding negation (NEG). The Composed Quantifier Argument depends on the existence of a variant of the Classical NR phenomenon involving not an overt main clause auxiliary instance of NEG (for example, in “Graham did not expect that she would arrive until Saturday”), but instead one or another negative quantifier phrase. The chapter considers examples, each of which has a Classical NR predicate in the main clause and a strict negative polarity item (NPI) in the embedded clause, but none of which manifests a matrix clause containing an overt NEG that could have raised from its embedded complement clause. It also shows that one can avoid wrong meanings and obtain the right ones without lexicalization of raised NEGs, and that Horn clauses licensed in composed quantifier structures strongly supports the syntactic view of Classical NR.

This chapter first situates the work against the background of the existing literature. It further introduces the key concepts and terminology used in the volume. After an explanation of the difference between standard negation and sentential negation, the cyclical changes affecting the expression of negation, as well as the typical types of interaction between indefinites and the expression of negation are introduced. Finally, the theoretical framework used in the volume is sketched. As the chapter considers both language-internal motivations for the changes affecting negative expressions and indefinites in the scope of negation, and language-external ones, the theoretical background for both are discussed in separate sections.

This chapter provides an overview of PSIs (polarity-sensitive items) in Arabic and briefly outlines the critical issues in the syntax of PSIs in Arabic and their contributions to Arabic syntax and linguistic theory. It also outlines the theoretical underpinnings of research on Arabic negation, relying on the author’s most recent publications. The author synthesizes the major and crucial findings from cross-linguistic studies on this topic and studies of PSIs in Arabic. This chapter also articulates the critical issues, methodology, and scope of content. A quick overview of polarity sensitivity reveals much of the microvariation in Arabic. Geographically, this variation comparatively grows at the phonetic, morphological, and lexical levels, moving from the west with the Moroccan dialect, into Egyptian and Jordanian along the Mediterranean, into the Arabian Peninsula with Qatari Arabic, and into the formal written Arabic.

This chapter focuses on certain contexts, dubbed Horn clauses, that demand the syntactic presence of a negative constituent. It first considers fronted negative polarity items (NPIs) by giving examples that represent standard cases of the Negative Inversion construction. The most obvious characteristic of the construction is that the extracted non-wh-constituent in the clause-initial position, termed Negative Inversion focus, co-occurs with subject-auxiliary inversion, which is obligatory. Sentences containing Horn clauses, such as Carl did (not) claim that penguins were mammals and neither did I and Carl claimed that penguins were not mammals (and neither did I), involve syntactic raising of a negation (NEG) from the embedded clause. The chapter proposes for Horn clause cases an analysis that treats examples as resulting from the raising via Classical NR of the NEG. It also shows that Classical NEG Raising (NR) out of Horn clauses is subject to the same set of island constraints holding for non-Horn clause island structures.

This chapter focuses on polyadic quantification, which exists when n-tuples (n > 1) of determiner phrases (DPs) yield a single quantifier interpreted as quantifying over n-tuples of individuals. It begins by considering the ordinary standard English sentence No one ever showed me anything, in which the any phrase is commonly assumed to be a negative polarity item (NPI), licensed by the c-commanding no one phrase. It then gives a nonstandard example, No one ever showed me nothing, which represents a distinct phenomenon known as negative concord and is deformed by various NEG deletions. In yet another example, No man loves any woman, in which, under polyadic interpretation, the two quantifiers are not interpreted independently. The chapter also discusses syntactic determiner sharing between the different DPs whose noun phrases denote the sets quantified over, the semantics of determiner sharing, and the realizations of polyadic quantifier structures.

This chapter considers three types of answers: weakly exhaustive, strongly exhaustive, and non-exhaustive/mention-some answers. Two lines of approach are discussed, one that locates the variation between strong and weak exhaustiveness in answerhood operators, another where variation is due to the optional presence of exhaustification operators in the question nucleus. The second issue addressed is the ambiguity between mention-some and mention-all readings. Under one view questions are ambiguous, with different contexts making different readings salient; under another view they only have mention-all readings but partial answers may count as complete answers in certain contexts. The possibility of three classes of predicates, those that select for strong exhaustiveness, those that select for both strong and weak exhaustiveness, and those that select only for weak/non-exhaustiveness is considered as evidence for these distinctions. Negative polarity items are also considered a diagnostic for the grammatical status of the divide between strong, weak, and non-exhaustiveness.

This chapter focuses on the assumption that the negative polarity item (NPI) phrase occupies a scope position as well as the object position, suggesting that NEG raising takes place only from the scope position. It also claims that [[NEG SOME] thing] must occupy a scope position, a common assumption in the syntax and semantics literature; that for a determiner phrase (DP) with a scope occurrence, the scope occurrence is the unique available launching point for NEG raising; and that if the NEGs of quantifier DPs raise out of nonscope positions, “overgeneration” will result in certain clear cases, whereas if such raising can only launch from scope positions, the overgeneration is avoided. In the sentence Rodney claimed that Evelyn did not own any cheetah, the scope of the quantifier DP is internal to the complement clause. The chapter also considers a structure with negation in the main clause in the sentence Rodney did not claim that Evelyn owned any cheetah.

This chapter focuses on the scope of strict negative polarity items (NPIs). It first considers scope issues concerning nonfinite complement clauses, with particular emphasis on similar pairs with analogous judgment markings, the relevance of stress contrasts to the ambiguities of NPI any forms as well as to those of other nominal NPIs, and the differential scope of the determiner phrase represented by anything (“Vaughn didn't accept to write anything about radiation”). It then turns to cases of infinitival complements containing strict NPIs, along with cases where the issue of high-scope confounds involve finite complement clauses. Finally, it suggests that any attempt to diagnose the presence or absence of Classical NEG Raising (NR) must always take into account the possibility of NEG raising out of main clause scope positions.

This chapter shows that Classical NEG Raising (NR) is sensitive to syntactic islands and considers a range of cases where it is blocked by island constraints, such as those involving clausal complements of nouns. At issue are examples invoking the Complex NP Constraint, clause-internal topics, truth predicates, wh-islands, clause-internal clefts, pseudoclefts, and Negative Inversion. The clear generalization is that Classical NR is never possible from an island. Such a generalization is especially striking for cases where all known semantic conditions on Classical NR are met (for example, for truth predicates), but Classical NR is still not possible. Because syntactic raising phenomena are subject to island constraints, it is possible to account naturally for the above generalization under the assumption that classical NR is a syntactic raising phenomenon. The chapter also examines island types that block strict negative polarity items (NPIs) but not nonstrict NPIs.

This chapter considers the role of nonfinite clauses in the distribution of Classical NEG Raising (NR). It first considers the literature bias toward finite hosts in standard Classical NR cases, along with proposals in nonsyntactic approaches that the Classical NR phenomenon is purely a function of the interaction of the meanings of the Classical NR predicates (CNRPs) with independent semantic or pragmatic principles. Since it is obscure how the semantic properties of CNRPs could be systematically different in their finite and nonfinite instances, on semantic/pragmatic treatments one expects that the same relations should systematically hold when the negation (NEG) sits in a nonfinite host. The chapter outlines the basics of the distinction between finite and nonfinite hosts and presents examples indicating the lack of a Classical NR reading for a NEG raised into a nonfinite clause and the impossibility of a strict negative polarity item (NPI) in the relevant complement clause. It also discusses the Classical NR Nonfiniteness Condition and composed quantifier cases.