Ljuba Veselinova

Lexicalization of negation and its role for the development of negation markers


The term standard negation (SN) refers to the negation markers in simple declarative sentences with an overt verb predicate like (i) Maria does not smoke. The negation markers in sentences such as (ii) Britt is not a teacher (a non-verbal sentence) and (iii) There are no ghosts (an existential sentence) are excluded from large-scale comparative studies on SN. In languages from Western Europe, the negation markers in sentences like (ii) and (iii) are identical with SN. However, this is not the case in many other languages elsewhere in the world. For instance, in Serbian, SN is expressed by a particle ne as in (iv) Maria ne peva ‘Maria does not sing’; two different negators are used for the negation of non-verbal sentences like (v) Tom nije nastavchik ‘Tom is not a teacher’ and existential sentences like (vi) Nema divlijh machaka ‘Wild cats do not exist’. I use the term special negators to refer to the negation markers restricted to sentences like (v) and (vi).


The current project has two parts, one is synchronic and the other diachronic. For the synchronic part of the project, I will produce a description of both SN and special negators in a sample of several language families. In the diachronic part of the project, my focus is on the evolution of special negators, ways in which they interact with SN, and finally, how they themselves may become sources for SN. The ultimate goal of this study is to test and revise existing models of the evolution of negation in human languages.

Stockholm University, Linguistics

2010-2017
The evolution of negation is often discussed in terms of a grammaticalization process dubbed by Dahl (1979) Jespersen Cycle. Within this process, negation markers are seen to originate from emphatic elements in the negative phrase which gradually lose their sense of emphasis and are eventually interpreted as general verbal negators. Croft (1991) has suggested negative existentials as another source for negation markers. This latter author presented his hypothesis under the name Negative Existential Cycle (hereafter NEC). The NEC is a variationist model that builds on cross-linguistic data. Despite recent renewed interest in cyclical processes in language change and negative cycles in particular, the cycle suggested by Croft (1991) has received little attention. Typically, it is quoted at face value, cf. Mosegaard Hansen (2011) for a relatively recent reference. The main goal of this work has been to test it by more traditional methods of historical linguistics, e.g. applying it on historical-comparative data from a number of different families. Since negative existentials can also be seen as lexicalizations of negation, e.g. concepts that can be semantically decomposed into a negative component and a positive sense, I also strived to gain a better understanding of which positive senses tend to be most frequently fused with a negator.
Abbreviations: DET-determiner; EX-existential; FUT-future; NEG-negative; NMZN-nominalization; PRS-present; PST-past; SG-singular; SN-standard negation; TA-tense-aspect; 1-first person; 3-third person
All language classifications cited here follow Glottolog 2.3, http://glottolog.org.
 

The main results
As stated above, the NEC was proposed by Croft (1991) as a way of modeling the evolution of standard negation markers from negative existentials. As shown in (1a), in Turkish, SN, or verbal negation, is expressed by a suffix -mA (the quality of the vowel is determined by vowel harmony); the negation of existential sentences is expressed by the semi-verb yok which replaces its positive counterpart var, cf. (1b-c).
(1) Turkish (Turkic, Common Turkic, [...], Oghuz) (Eyüp Bacanl?, p.c.)
a. Gel-me-yecek
come-NEG-FUT
'(S)he will not come.'
b. Su yok-tu
water NEG.EX-PST
'There was no water.'
c. Su var-d?
water EX-PST
'There was water.'
The NEC predicts three stable stages, i.e. having the same negative for all verbs, having different negatives for existential and non-existential verbs, and having the negative existential used as the negative marker for all verbs (again). There are also three transitional stages in between the stable ones. In the NEC, negative existentials are seen to gradually break into the domain of SN by their uses for the negation of verbal predications in specific tense-aspect-mood categories or other well delimited contexts. For instance, in Bulgarian, the negative existential njama, shown in (2a), is also used to negate verbal predications with future time reference as in (2b). The particle ne, cited in grammars as the SN-marker, is used for the negation of verbal predications with non-future time reference (2c).
(2) Bulgarian (Indo-European, South Slavonic) (own data)
a. Dnes njama voda
Today NEG.EX water
'There is no water today.'
b. Maya njama da pee
Maya NEG.EX to sing.3SG.PRS
'Maya is not going to sing.'
c. Maya ne pja
Maya NEG sing.3SG.PST
'Maya did not sing.'
Thus Bulgarian is taken to represent a transitional type, e.g. languages where the negative existential is expanding into the verbal domain.
In my work I covered eight language families from North Africa, Eurasia, Oceania and South America: Berber, Slavic, Uralic, Dravidian, Turkic, Polynesian, Arawak and Tucano. The collected data suggest the following generalizations.
First, negative existentials enter the domain of SN via several different pathways. These include (i) subordination processes and predicate concatenation; (ii) the reanalysis of an external negator into a negator external to the proposition; (iii) a direct inheritance of a construction; (iv) the use of negative existentials with nominalized verb forms cf. (3). Among these pathways, only (iv) shows a great extent of cross-linguistic recurrence.
(3) Selkup, Taz dialect (Uralic, Samoyedic) (Béata Wagner-Nagy, p.c.)
man il?-ptä-m? cää?ka
1SG live-NMZN-1SG be.absent.3SG.SUBJECT
'I didn't live' literally 'my living [is] absent'
Second, via the pathways listed above, negative existentials typically take over very specific parts of verbal negation. That is, they come be used as verbal negators for a particular tense-aspect-mood category. These partial take-overs of verbal negation tend to last for very long periods of time. So from a diachronic point of view, they look like stable states.
Third, negative existentials are most likely to fully take over the domain of SN in languages where SN is expressed by a complex clause. In my family based sample, this is observed in Polynesian languages, illustrated in (4) below by M?ori.
(4) M?ori, (Austronesian, [...] Nuclear Polynesian, Eastern, Tahitic), (Harlow 2007: 161)
K?ore te tamaiti e tangi ana
NEG DET child TA weep TA
'The child is/was not crying' literally 'There is no child [who] is crying.'
In (Veselinova 2014), I show that the productivity of subordination processes in Polynesian languages has led to several renewals of their SN markers and to at least one round of completion of the NEC. Using the comparative method, I have been able to estimate the length for this evolution to about 2000 years. In the literature on linguistic cycles, it is typically noted that the rate of change varies considerably from one language to another. Thus the duration of a cycle is hard to predict. However, the evolution of negative existentials into markers of standard negation appears to be consistently of a lengthier kind, see also above on the longevity of partial take-overs.
Finally, the cross-linguistic and likewise the historical-comparative data show that negative existentials are extremely wide-spread in the languages of the world and that they are constantly renewed. This has implications for the cyclical processes they enter. Negative existential cycles tend to stretch over long periods of time and if completed, they immediately start anew (Veselinova 2016). In (Veselinova 2013a) I provide evidence that negative existentials are better seen as a functional domain of their own. The distinction between negation of actions and negation of existence is cognitively basic and consistently maintained. Consequently, I suggest that the extended duration of the transitional stages in the NEC as well as its perpetual renewals are cognitively motivated.
I have organized the collected data in a database and visualized them on a dynamic map at http://arcg.is/1NYh54X. This way my results become accessible and verifiable.
I also did a more general study on lexicalizations of negation; for it I used both the family based sample detailed above and also a stratified sample with world coverage; the latter consists of 97 languages. My database contains lexical expressions for 65 negative senses which can be grouped into broader semantic domains. The most common is by far the domain of non-existence/non-possession; followed by cognition ('not-know'), modality ('cannot') and finally tense-aspect ('not-yet'). I am preparing a manuscript to be submitted to Studies in language.
 

New research questions generated during the project
The data collected here brought to light a number of cross-linguistically common lexicalizations of negation, some of which give rise to TA categories restricted to the domain of negation. Thus this project led to my current large scale study of asymmetries between TA systems in the affirmative and in the negative domain.
 

International visibility
I participated in international conferences on a yearly basis. My work on the negative existential cycle has been received very well. In January of 2015 I was contacted by Lexington Books, https://rowman.com/LexingtonBooks who are interested in publishing a book on this topic. To this end, I initiated collaborative work with Arja Hamari from the University of Helsinki. We are organizing a conference on this topic in May 2017 where we have solicited the participation of scholars who have expertise on specific language families and are interested in the issues discussed here. A peer reviewed volume edited by myself and Hamari will be complied after the conference.

Outreach to the non-scientific community
I was invited to several non-academic fora to speak about the language situation in the modern world and also on the kind of work a typologist does. During these presentations, I also took the opportunity to highlight the historical work presented here.
 

The two most important publications
Veselinova (2014) and Veselinova (2016) are the most significant outcomes of this project. The first one appears in one of the most prestigious journals in the field of linguistics; thus it made the work known to the entire linguistic community. The second one reflects a more mature reasoning about cyclical processes and the evolution of negation. Both of these works are widely cited and have sparked a lively academic debate.

Publication strategies
I have seen to publish in peer reviewed venues with high impact factor. At the same time, I have also made copies of my work freely available at sharing sites such as www.academia.edu.

References
Croft, W. 1991. The Evolution of Negation. Journal of Linguistics 27.1-39.
Dahl, Ö. 1979. Typology of sentence negation. LINGUISTICS 17.79-106.
Harlow, R. 2007. M?ori: A Linguistic Introduction Cambridge: Cambridge University Press.
Miestamo, M. 2005. Standard Negation: The Negation of Declarative Verbal Main Clauses in a Typological Perspective Berlin: Mouton de Gruyter.
Mosegaard Hansen, M-B. 2011. Negative Cycles and Grammaticalization. The Oxford Handbook of Grammaticalization, ed. by H. Narrog & B. Heine, pp. 570-79. Oxford: Oxford University Press

PUBLICATIONS


Veselinova L. 2013a. Negative existentials: a cross linguistic study. Italian Journal of Linguistics: Special Issue on Existential Constructions 25.107-46.
Veselinova L. 2013b. Suppletion. Oxford Bibliographies in Linguistics, ed. by M. Aronoff. New York: Oxford University Press.
Veselinova L. 2014. The Negative Existential Cycle Revisited. LINGUISTICS 52.1327-69.
Veselinova L. 2016. The Negative Existential Cycle through the lens of comparative data. The Linguistic Cycle Continued, ed. by E. van Gelderen, 139-87. Amsterdam/New York: John Benjamins Publishing Co.
Veselinova L.. 2016. Random Samples. In Wörterbücher zur Sprach- und Kommunikationswissenschaft (WSK) Online (eds) S.J. Schierholz & H.E. Wiegand. Berlin: De Gruyter.
Veselinova L.. 2016. Sampling Procedures. In Wörterbücher zur Sprach- und Kommunikationswissenschaft (WSK) Online (eds) S.J. Schierholz & H.E. Wiegand. Berlin: De Gruyter.
Veselinova, Ljuba (with Hedvig Skirgård). 2015. Special Negators in the Uralic Languages: Synchrony, Diachrony and Interaction with Standard Negation. Negation in Uralic Languages, ed. by M. Miestamo, A. Tamm & B. Wagner-Nagy, 547-99. Amsterdam/New York: John Benjamins Publishing Company.

WEBSITES
http://arcg.is/1NYh54X
www.ling.su.se/negative.existentials.cycle

CONFERENCE PARTICIPATION
May 24-26, 2016. Standard and special negators in Arawak and Tucano languages. Amazonicas VII, Leticia, Colombia.
August 1-3, 2015. Not-yet expressions in the languages of the world: a special negator or a separate cross- linguistic category. 11th Biennal Conference of the Association for Linguistic Typology (ALT 11), Albuquerque, New Mexico, USA.
May 1-3, 2015. Not-yet expressions in the languages of the world: a special negator or a separate cross- linguistic category. Diversity Linguistics: Retrospect and Prospect. Closing Conference at the Department of Linguistics, Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany.
April 25-26, 2014. The Negative Existential Cycle Revisited: Part II. Workshop on Linguistic Cycles II Arizona State University, Phoenix AZ, USA.
August 15-18, 2013. Lexicalized negative senses: a cross-linguistic study. 10th Biennal Conference of the Association for Linguistic Typology (ALT 10), University of Leipzig/MPI Leipzig, Leipzig, Germany.
April 11, 2013. Negative existentials and the constructions they require. Workshop on constructions. Stockholm University.
August 29 - September 1, 2012. Applying the Negative Existential Cycle on the Uralic Languages. Poster oresentation at the 45th Annual Meeting of the Societas Linguistica Europea (SLE), University of Stockholm, Stockholm, Sweden.
June 29-30, 2012. Negative existentials – a cross linguistic study. Manchester Symposium on Existentials, University of Manchester, UK.
November 24-25, 2011. Special Negators in the Uralic Languages. Workshop on Negation in the Uralic Languages, University of Stockholm, Stockholm, Sweden.
July 23-26, 2011. The Negative Existential Cycle Revisited. Presentation at the 9th Biennal Meeting of the Association for Linguistic Typology, University of Hong Kong, Hong Kong, China.
    Sampling Isolates (with Oliver Bond). Presentation at the 9th Biennal Meeting of the Association for Linguistic Typology, University of Hong Kong, Hong Kong, China.

INVITED TALKS
March 14, 2014. Stora och små språk. Vad gör en språktypolog. Invited talk to the Linguistic Olymplics (Swedish LingOlympiaden), Stockholm.
December 3, 2013. Stora och små språk. Vad gör en språktypolog. Invited talk daycare and primary school teachers, BoService 24, Stockholm.
November 9, 2013. Stora och små språk. Vad gör en språktypolog. Invited talk during SU på Plattan (A cooperative enterprise between Stockholm University and the City Council of Stockholm where lecturers from the university deliver lectures to the general public).