This volume represents the fruit of a workshop organized by the editors at the 55th Annual Meeting of the Societas Linguistica Europaea in Bucharest in August 2022, the hundredth anniversary of Otto Jespersen’s classic (Reference Jespersen1922) Language: Its Nature, Development and Origin. That volume contains one of the earliest uses of the term ‘negative concord’, the use of two seemingly negative elements in a language to convey a single semantic negation as with the negator non ‘not’ and the negative concord item nessuno ‘no one’ in Italian Non è venuto nessuno ‘No one has come’. It turns out that the term was actually coined, in Danish, by Jespersen in his (Reference Jespersen1891) dissertation, so we are in fact dealing with a little over a hundred years of negative concord.

The editors open with a very interesting and useful overview of the history of research on negative concord (‘Negative concord: The first 133 years’, pp. 9–52). The basic idea is there from the earliest discussions in the nineteenth and early twentieth centuries: negative elements tend to ‘pile up’, nevertheless yielding only a single negative interpretation, so that they may be thought of as agreeing with one another in the same way that concord exists elsewhere in language, as with verbs showing morphological agreement with their subjects, etc. They trace the relatively recent growth in use of the term ‘negative concord’ from Labov’s (Reference Labov1972) work on African American Vernacular English in the late 1960s and 1970s through generative work of the 1980s and 1990s, typological work of the 1990s and work on non-standard or historical English grammar from the 1970s. As they note, negative concord is not especially common in the world’s languages, but is prominent in Europe and in creoles, hence the frequency of its occurrence tended to be overstated in earlier work, less so as the diversity of the world’s languages has come to be more fully acknowledged.

Another shift in perspective has been the change from viewing negative concord as the property of a particular construction to viewing it as the property of an entire language. Van der Auwera and Gianollo point out that this is not unproblematic, with a single language capable of using negative concord with some items or clause types and not with others, and some languages using negative concord some, but not all, of the time. They also highlight the contribution of dialect data and the study of language contact to the field, with diachronic, corpus and acquisition-based work also having become increasingly important. A positive feature of work on language contact that they note is that it has suffered from less of a divide between generative and typological work than in many other areas of syntactic research.

The terminological minefield that bedevils research on negative concord is a significant problem for the area (and the variation in terminology and definitions even among the contributors to this volume still confirms this as an ongoing difficulty to some extent). Martin Haspelmath, in his chapter ‘Negindefinites and negative concord: Concepts, terms and analyses’ (pp. 53–75), attempts to address this by providing precise definitions for a number of core concepts. Perhaps inevitably, this involves the creation of even more terminology, via the new term ‘negindefinite’ (from ‘negative indefinite’), defined (p. 57) as ‘a pronoun (or determiner) which either (i) can express negation in isolation, or (ii) always occurs in the scope of a clausal negation meaning’. Negative concord is then a construction involving a negindefinite and another negative form (either a clausal negator or another negindefinite) yielding a single negation meaning. Negindefinites that can occur in negative concord are consequently called concord negindefinites. He proposes replacing ‘double negation’ for cases where two negative elements cancel each other out with ‘duplex negation’, because the former is sometimes also used to mean bipartite negation (use of two elements to express clausal negation, as with French ne…pas). Finally, negative amalgamation is where a single element expresses both clausal negation and a negindefinite, as with English nothing (an ‘amalgam negindefinite’). The attempt at terminological clarity here is to be welcomed, although it comes at the cost of yet more terminological proliferation, and more rigorous definition of existing terms and propagation of those definitions might be a better way ahead.

Karen De Clercq (‘Negative concord and TAM: A new perspective’, pp. 77–105) looks at the morphological interaction of standard negators and negative concord items, comparing them to the interactions that are observed in many languages between the form of the standard negator and the tense/form of the verb. She looks at two case studies, Berber and Egyptian Arabic. In Quebliyeen Tamazight Berber, negative concord items on their own are insufficient to trigger the changes in verb forms found with the standard negator. In Egyptian Arabic, the form of the standard negator changes according to tense/aspect, but the presence or absence of a negative concord item has no effect on this alternation or vice versa. De Clercq interprets this as meaning that interactions between the form of the standard negator and verb are not carried over to negative concord items, a fact which favours approaches that treat markers of standard negation and negative concord items as very distinct. She develops such a nanosyntactic approach along these lines.

Pierre Larrivée (‘Non-strict negative concord: A correlate of focus movement?’, pp. 107–22) discusses obligatory non-strict negative concord, where a postverbal negative concord item requires the sentential negator, while a preverbal one cannot co-occur with the sentential negator. While an optional form of this is found in many languages, its obligatory form is observed only in Romance and Kartvelian. He links this to the fact that these languages have overt focus movement: focus movement of the negative concord item triggers agreement with the clausal negation, causing it to delete. As Larrivée observes, there are (a diverse set of) languages with focus movement, and not very many with obligatory non-strict negative concord, so the relation is not a particularly strong one. The status of all the optional non-strict negative concord languages is also left rather unclear on this proposal.

Petr Rossyaykin (‘The universal force of (strict) negative concord’, pp. 123–47) also considers non-strict negative concord, arguing that it is not a unified phenomenon and that there may be no universal account applicable to all languages that exhibit it. His main focus, however, is on languages with strict negative concord and negative concord items of the form wh-stem + scalar-additive particle (‘and even’), such as Russian or Hungarian, arguing that these do present a unified system, where negative concord items are universal quantifiers that move to a focus position agreeing with the negative marker. Although both the languages involved and the details of the analysis are different, there is a similarity of spirit shared by both Larrivée’s and Rossyaykin’s approaches. This general approach is continued in Emil Ionescu’s chapter (‘N-words: In defence of wide scope. A view from Romanian’, pp. 149–72), which looks at strict negative concord in Romanian. It follows Rossyaykin in arguing that negative concord items are universal quantifiers scoping over negation, discussing various data that support this view and offering ways to explain apparent exceptions.

Nikos Liosis (‘Negative concord in Medieval Greek: Lexical asymmetry and the role of correlative negation’, pp. 175–206) presents data from the history of Greek tracing the complexities of the shifting negative concord systems of the language. As is well known, Ancient Greek was a non-strict negative concord language, while Modern Greek is a strict negative concord language. Liosis highlights the strong variability in use of non-strict negative concord in earlier stages of Greek and in Greek varieties today. Negative concord occurs occasionally in Classical Greek with the element uðé ‘not even’. This is used to renew negative concord items, spreading furtively from a few postverbal instances at first, but forming the basis for the spread of negative concord, initially as part of a non-strict negative concord system, later in all syntactic positions, yielding the strict negative concord of Modern Greek.

Alexander Letuchiy (‘What does the linear position of negative concord items tell us about their properties?’, pp. 207–38) deals with word-order asymmetries that have been observed between negative concord items and equivalent plain noun phrases, including pronouns, in Russian. He provides corpus evidence confirming these differences, showing, above all, that, as direct objects, negative concord items appear more frequently in preverbal position than ordinary noun phrases or pronouns. In general, negative concord items have a tendency to appear immediately before the negative marker that licenses them, while positioning of negative concord items near the right-edge of the clause is associated with intonational emphasis. These properties are attributed to the association with negation rather than any phonological weakness of the kind found with object pronouns.

Egbert Fortuin and Thom van Hugte (‘Grammar or error? Non-compositional (expletive/paratactic) negation in Dutch sentences with zonder “without”’, pp. 239–67) investigate expletive negation in Dutch zonder (dat) ‘without’ clauses. They report the results of a judgement-task experiment which shows that, although speakers generally reject expletive negation in this context, they are more tolerant of it when there is a negatively connotating element in the main clause and when an element is focused in the subordinate clause, confirming the result of previous corpus-based work. They consider whether use of niet ‘not’ in these clauses is a speech error or part of speakers’ grammars, preferring the latter interpretation on the basis that the role of the negative marker is to negate the proposition of non-occurrence of the event in the subordinate clause (i.e. to mark the fact that ‘you can’t do x without doing y’ does not entail ‘you don’t do y’).

The final two chapters deal with variation in negation in English. Laura R. Bailey and Claire Childs (‘Two negatives in Tyneside English questions: Negative concord or double negation?’, pp. 269–302) discuss aspects of negation in Tyneside English. Standard English is reported, on the basis of linguists’ grammaticality judgements (Kramer & Rawlins Reference Kramer, Rawlins, Lima, Mullin and Smith2011), to make an interpretative distinction between outer negation (Didn’t she go on holiday? Yes means ‘She did go on holiday’) and inner negation (Did she not go on holiday? Yes means ‘She didn’t go on holiday’). Tyneside speakers favour auxiliary contraction (we’ve not) over negative contraction (we haven’t) with auxiliaries. This might suggest that the distinction between inner and outer negation is less clear-cut for them. The authors’ experiment indeed fails to find this distinction to be particular clear among a group of Tyneside participants; however, the experiment lacks a control group of non-Tyneside participants for comparison, and it may be that the distinction would not emerge particularly robustly from experimental data with any group of English native speakers. More research is needed to address this possibility.

Tyneside English also has a construction where negative and negative tag questions seem to contain a concording negative element (Didn’t she not go on holiday? She went on holiday, didn’t she not?). In Bailey and Childs’ experiment, Tyneside speakers mostly judged a Yes response to Didn’t she not go on holiday? to mean ‘She did go on holiday’ and a No response to mean ‘She didn’t go on holiday’. This is the same interpretation as Standard English outer negation Didn’t she go on holiday?. The authors analyse this as a form of negative concord, with both -n’t and not bearing uninterpretable negative features that concord with a null interpretable negative feature. For tags, the experiment shows that speakers attribute somewhat less certainty to the speaker when didn’t he not? or did he not? are used after a positive anchor statement, than when didn’t he? is used. This is consistent with the existing view that the former are seeking information, while the latter is seeking confirmation.

The final chapter, by Stephanie Rotter and Mingya Liu (‘Less formal and more rebellious – An experiment on the social meaning of negative concord in American English’, pp. 303–32), focuses on the social meaning attached to the non-standard use of negative concord in English. Rotter and Liu asked speakers to rate sentences containing negative concord (I didn’t get no results) and negative polarity items (I didn’t get any results) for various grammatical and social features. Negative concord sentences were well understood by participants as conveying a single negation, but were rated less ‘grammatical’. It is hard to know how participants interpreted the word ‘grammatical’, a notoriously difficult term for laypeople, but the results seem to suggest that ‘prescriptively correct’ contributed to the responses. As might be expected from the non-standard status of negative concord in English today, a speaker using negative concord was judged to be of lower socioeconomic status, less educated, less formal, but more rebellious.

The chapters demonstrate the interplay of theory and insights from corpus and experimental data that has increasingly become a feature of work in this area. A number of contributions indeed provide empirical evidence to test or bolster existing claims in the literature. While the focus is mainly on European languages, this is to be expected given the areal distribution of negative concord in the world’s languages. The idea that negative concord items, at least in some languages, might be universal quantifiers outside the scope of negation is explored by a number of authors. Several adopt or extend the idea that negative concord is agreement with an abstract interpretable negative feature that licenses uninterpretable negative features on all the concording elements within its scope (Zeijlstra Reference Zeijlstra2004), obviating the need to decide which element provides the ‘true’ negative interpretation. All in all, the volume showcases the vitality of current research on negative concord set in a rich historical tradition. It attests to the range of typological and formal work currently being undertaken, built on a foundation of rich corpus material and experimental results.