Anna Carlier - Selected Publications#

My research is in the field of morphosyntax and focuses on two major categories, viz. nominal determiners (articles, demonstratives, possessives, quantifiers) and auxiliary verbs combined to non-finite verb forms as markers of aspect and argument structure (e. g. ‘be’ + past participle as a marker of passive voice, and perfective, resultative, stative aspect, as well as more recent work on ‘be’/’go’/’come’ + gerund in several Romance languages). The following publications cover a selection of topics that particularly fascinated me.

1. FROM PREPOSITION TO ARTICLE: THE DEVELOPMENT OF THE PARTITIVE ARTICLE IN FRENCH AND IN ROMANCE

(1) Carlier A. (2007). From preposition to article: the grammaticalization of the French partitive. Studies in Language 31: 1, 1-49. (115 citations) https://doi.org/10.1075/sl.31.1.02car

This study addresses two major questions:

(i) Why did French develop an article for indefinite non-singular reference, whereas most other languages endowed with an article system leave this type of reference unmarked?

(ii) How can we, in the pre-article stage, account for the use of the preposition de in syntactical functions that are in principle incompatible with the use of a preposition?

On the basis of a detailed and rigorous study of a diachronic parallel corpus, a new hypothesis is proposed concerning the morphosyntactic status of DE in the medieval partitive construction, called the one-sided preposition hypothesis, and in the so-called ‘partitive’ article. Moreover, this diachronic study highlights the syntactic and morpho-syntactic evolutionary trends specific to French in order to offer an insightful explanation as to why the partitive construction has become a fully grammaticalized article in French, whereas this evolution has not occurred in other Romance languages.

This paper has been very influential and has become a reference for later publications on partitive constructions in several Indo-European and other languages. For my part, I have extended my research into a comparative perspective between Romance languages, which currently also includes Occitan and Catalan. The following publication offers a first version of this comparative study:

(2) Carlier, A. & B. Lamiroy (2014). The grammaticalization of the prepositional partitive in Romance. In S. Luraghi & Th. Huumo Eds. Partitive Cases and Related Categories, 477–519, Berlin: Mouton-De Gruyter. (73 citations) https://doi.org/10.1515/9783110346060.477

I also contributed a chapter on this topic to the Oxford Research Encyclopedia of Romance Linguistics, eds F. Gardani & M. Loporcaro, 2021). In a recent study, I argued against the widespread view that reduces the partitive article to a morphological marker of gender and number, showing that it has referential properties that distinguish it from bare mass and plural nouns expressing indefiniteness.

(3) Carlier A. (2021). Du/des-NPs in French: a comparison with bare nouns in English and Spanish. In Giusti G. & Sleeman P. Eds. Partitives determiners, partitive pronouns and partitive case, 79-113. Berlin: De Gruyter (Linguistische Arbeiten) (https://www.degruyter.com/document/doi/10.1515/9783110732221-003/html)

2. FROM DEMONSTRATIVE TO DEFINITE ARTICLE

(4) De Mulder W. & A. Carlier (2011). Definite Articles. In Heine Bernd & Narrog Heiko Eds The Oxford Handbook of Grammaticalization, 522-534. Oxford: Oxford University Press. (100 citations) https://doi.org/10.1093/oxfordhb/9780199586783.013.0042

Research into the emergence of definiteness marking in Late Latin (IPSE/ILLE) has led us to reject a common hypothesis analyzing the development of the definite article as the result of a semantic weakening of the demonstrative, consisting in the loss of its deictic force (i. a. Lyons 1999). Instead, we develop the hypothesis that distal deixis plays a crucial role in the transition from demonstrative to definite article and accounts for the core meaning of the definite article, which consists in expressing unique reference by activating a frame of accessible knowledge which the referent is associated with.

Contrary to the proximal demonstrative(s), the distal demonstrative denotes a referent that is not fully identified or even not accessible within the immediate context of its occurrence. It mobilizes specific knowledge shared by speaker and hearer in order to achieve the identification of the referent (Himmelmann 1997). The distal demonstrative becomes a definite article when the presumably shared knowledge necessary to identify the referent is no longer necessarily anchored in the speech situation and presented as specific to the speaker and the hearer, but rather as stereotypical or shared by all members of the speech community. Drawing on Hawkins' (1978) distinction between pragmatic and semantic definiteness, this hypothesis leads us to postulate that the semantic definiteness stage is preceded by a pragmatic definiteness stage, where the distal deixis feature is still effective.

3. FROM UNITY NUMERAL TO INDEFINITE ARTICLE

As regards the grammaticalization of the indefinite singular article, my major contribution is to have redefined the primitive value of the unity numeral as a source expression, thus providing a better account for early as well as modern uses of the indefinite article derived from it. The notion of unity has to be understood not only in a quantitative sense ‘only one’ but has also a qualitative dimension ‘one and the same (in nature)’. Hence, when becoming an article, "un" introduces either a new discourse entity, either a new type of entity or category. The second, qualitative dimension explains why the article "un" is used from the earliest texts on with non-count nouns, mass or abstract (e.g. “une oscurité si grant”, Graal) and in predicative positions as early as Classical Latin. It also enables us to understand exclamatory uses such as “Il est d’un prétentieux” (lit. ‘he is of a pretentious’) and why its use is mandatory in generic NPs evoking a non-established category (“Un/*Le singe à queue longue est un habile grimpeur”).

I also conducted a study on the curious phenomenon of the plural indefinite article “uns”:

(5) Carlier, A. (2016). The capricious evolution of the indefinite plural article uns and its relationship with lexical plurality in medieval French. Lingvisticae Investigationes 39:2 : 309-334. https://doi.org/10.1075/li.39.2.06car

G. Guillaume argues that the disappearance of this ephemeral article is due to an internal contradiction (unity numeral vs grammatical feature of plural). Theoretical arguments concerning semantic change and empirical arguments (the survival and grammaticalisation of this formof the plural form of the article derived from UNUS in Ibero-Romance languages) go against this hypothesis. It is shown that the evolution towards the stage of the indefinite plural article, which was on-going in Middle French, was interrupted because of an independent evolution, i. e. the rise of a new mode of marking grammatical number, not based on suffixes, but on semi-free morphemes preceding the noun. This disappearance therefore coincides with the emergence of the partitive article.

4. OTHER NOMINAL DETERMINERS: DEMONSTRATIVES, QUANTIFIERS, POSSESSIVES

The semantics of demonstratives has occupied me on several occasions. The following publication reflects research in progress on the shift from the ternary, person-oriented system in Classical Latin (HIC/ISTE/ILLE) to the binary system as it exists in Old French (CIST/CIL), through the study of the transitory stage observed in Late Latin.

(6) Carlier A. (2017). Le système des démonstratifs en cours de restructuration en latin tardif : une séparation des rôles référentiel et pragmatique de la deixis. Langages 208, 29-51. https://doi.org/10.3917/lang.208.0029

This study, based on a detailed quantitative and qualitative analysis of demonstratives in the PaLaFra corpus, shows how this restructuring affects the semantics of all the demonstratives. It deals with the empirical problem of the high frequency of HIC in Late Latin, enigmatic since this demonstrative does not survive in any Romance language. Following R. Kirsner (1979), the distinction between distal and proximal demonstratives is defined in terms of 'strong' vs 'weak': proximal demonstratives have a stronger deictic force than distal demonstratives, and as such have a dual role: they serve to identify referents that require an important identification effort (referential role), on the one hand, and to highlight referents that are highly salient by virtue of their intrinsic properties or in context (pragmatic role), on the other. It is argued that once ISTE encroaches upon the sphere of the 1st person, traditionally the domain of HIC, a new division of roles emerges, assigning a referential role to HIC and a pragmatic role to ISTE.

I have also explored the diachronic evolution of quantifiers like MULTUM (Latin) and BEAUCOUP (French), oscillating between verbal and nominal quantification, which raise issues as to morphosyntactic categorization. I have notably highlighted that the categorial trajectory of MULTUM and BEAUCOUP is identical: stemming from a (pro)nominal expression, they evolve into adverbs quantifying the verbal process and subsequently reintegrate into the class of nominal determiners.

Finally, I have also conducted, in collaboration with S. Simonenko, a study on the evolutionary trajectory of long and short possessives in three Romance languages—French, Spanish, and Portuguese—using statistical modeling on large corpora.

This research has earned me expertise on the topic of nominal determination, reflected not only in multiple invited conferences but also in contributions to major publication projects such as the Grande Grammaire de l’Histoire du Français (eds C. Marchello-Nizia et al., De Gruyter : 2022, 5 chapters), l’Encyclopédie grammaticale du français (chapter on quantifiers, forthcoming), la Oxford Research Encyclopedia of Romance Linguistics, eds F. Gardani & M. Loporcaro, OUP : 2021, chapter on partitives), The Oxford Handbook of Grammaticalization, eds H. Narrog & B.Heine, OUP: 2011, 2nd ed: 2021, chapter on Definite articles), The Wiley Blackwell Companion to Diachronic Linguistics (chapter 'Changes in the determiner inventory across time', forthcoming).

5. AUXILIARY VERB CONSTRUCTIONS: ARGUMENT STRUCTURE AND ASPECT

This research axis, while less prolific in publications, has occupied me throughout my career, especially through the doctoral theses I have supervised (A. Buchard, G. Viale, J. Vangaever). I choose here to present a less recent study, bringing original hypotheses to a yet classic topic.

(7) Carlier A. (2002). Les propriétés aspectuelles du passif. Cahiers Chronos 10, 41-63. (63 cit.) DOI/ISBN : 9789042011335

While the French passive construction ‘be’ + PST.PTCP is primarily a marker of diathesis, it is associated to non-systematic interpretative effects of resultative state and iteration (Blanche-Benveniste 1984), which are still poorly understood.

This study critically examines and refutes two previous hypotheses accounting of the aspectual features of the French passive voice (i.e. the existence of an adjectival passive besides the verbal one, and the perfective aspect inherent to the past participle as a source of the resultative aspect of the passive voice ‘be’ + PST.PTCP). It is argued instead that the very construction of the passive of which ‘be’ is a component yields a stative interpretation. This hypothesis accounts for the two semantic changes triggered by the passive construction when the verbal predicate corresponds to a telic process – resultative state and iteration (in the presence of the agent complement) (resp. Le vin est servi/ Le vin est servi par le sommelier) –, whereas no aspectual shift occurs when the verbal predicate corresponds to a state (Pierre est aimée de Marie).

The study next focuses on the interaction between different aspectual levels (lexical aspect, grammatical aspect related to voice, grammatical aspect related to tense) by elucidating their hierarchical relationships. On the basis of an analysis of the relative scope hierarchy between the morphological marker of the passé composé and that of the passive (cf. Benveniste, "Structures des relations d'auxiliarité"), it is demonstrated that there is iconicity between the morphological structure of the verb form and the aspectual hierarchy, i.e. (((lexical aspect) grammatical aspect related to voice) grammatical aspect related to tense).

DIACHRONY & COMPARISON

With respect of the languages studied, my work deals with Romance languages, with a focus on French. However, a typological perspective is always present. The originality of my approach lies in combining a long-term diachronic perspective, from late Latin to the contemporary stages of Romance languages, on the one hand, and a comparative perspective between Romance languages, on the other. These two perspectives are mutually enriching, in light of the hypothesis of the Romance Grammaticalization Cline put forward in the following paper:

(8) Carlier A., W. De Mulder, B. Lamiroy (2012), The pace of grammaticalization in a typological perspective. Folia Linguistica 46/2 “The pace of grammaticalization in Romance”, 287-302. (37 cit.) https://doi.org/10.1515/flin.2012.010

It is argued in this paper that, within the Romance language family, languages and language varieties differ in a consistent way with respect to a range of grammaticalization phenomena, and can be ordered on a scale, according to their degree of grammatical innovation in relation to their common ancestor, Latin. With respect to a variety of phenomena, French, for instance, proves to be a highly innovative language, whereas Spanish has moved less far from its Latin source, with Italian occupying an intermediate position, between French and Spanish. This is why the study of a particular phenomenon in contemporary Spanish can, for example, shed light on a comparable phenomenon in medieval French.

The Romance Grammaticalization Cline hypothesis has proved to have a strong explanatory power for Romance languages and generated vivid scholarly debate precipitating numerous subsequent studies. It has been corroborated for various linguistic features (e.g. word order, causative constructions, prepositions, connectives, indefinite articles, partitive articles, possessives, existential constructions, motion verbs, negation, subjunctive, tense, aspect and mood, color terms). Other studies have located a specific Romance language or language variety (e.g. Catalan, Piedmontese) on the Romance cline. This list of publications also includes our study on nominal determiners in Romance languages.

(9) Carlier A. & B. Lamiroy (2018). The emergence of the grammatical paradigm of nominal determiners in French and in Romance: Comparative and diachronic perspectives. Canadian Journal of Linguistics 63:2, 141-166. https://doi.org/10.1017/cnj.2017.43

With hindsight, it seems to me today that this Romance Grammaticalization Cline is rather an epiphenomenon. What is fundamentally at play is a global restructuring of the grammatical architecture of the language, changing the balance between syntax, morphology and parts-of-speech system (Hengeveld, Rijkhoff & Siewierska 2004; Ledgeway 2011), but at different rates and to different extents. This shift in balance between syntax, morphology and parts-of-speech system, at different rates to different extents is therefore a crucial parameter of language diversification within the Romance language family.

METHODOLOGY: CORPUS LINGUISTICS

From a methodological point of view, I am not only a user of digital corpora, but I have also contributed to build them. The latest achievement is the PaLaFra corpus, a corpus of Merovingian Latin and Old French texts from ca. 500 until 1250 AD, conceived with a special attention to the sociolinguistic properties of each text, as well as to differences between text types and genres. This corpus, elaborated in the framework of a funded ANR-DFG project, in according with philological standards, PoS-annotated and lemmatized, has disclosed a new field of investigation for the linguistic community. A first volume has been produced based on a pilot version of the PaLaFra corpus.

(10) Carlier, A., & Guillot-Barbance, C. Eds. (2018). Latin tardif, français ancien. Berlin, Boston: De Gruyter. doi: https://doi.org/10.1515/9783110551716

It serves as a springboard for a more ambitious publication project, which will be published in early 2025 under the title “Bridging the gap between Late Latin and Old French: A handbook of morpho-syntactic change” (eds A. Carlier , L. Danckaert , C. Guillot-Barbance, M. Selig and J. Vangaever). https://www.degruyter.com/document/isbn/9783110617764/html

Based on the observation that, of all the Romance languages, French differs most strongly from its Latin ancestor, this handbook aims to understand how this situation came about, by offering a comprehensive account of the morpho-syntactic changes that took place during the transition from Latin to French. Each chapter is co-authored by specialists of Late Latin and of Old French (40 contributors) and is empirically based on the PaLaFra corpus.

For my own research, which combines diachronic and comparative perspectives between Romance languages, I encountered the problem that it is very difficult to empirically study the process of diversification of Romance languages in the absence of an electronic corpus enabling us to compare the diachronic evolution of Romance languages. Within the framework of the CoRaLhis project (Idex-funded), I have thus started to conceive a historico-comparative corpus, observing the constraints of the philological quality of the texts, and a balance between text genres, and which will no less than ten languages belonging to the three major Romance sub-branches, viz. Gallo-, Italo-, and Ibero-Romance, over the period from the 13th to the 18th C. Once completed, this corpus will be made available in free access to the scientific community for the benefit of Romance studies.

REFERENCES

Blanche-Benveniste, C. (1984). Commentaires sur le passif en français, Travaux du Cercle linguistique d'Aix-en-Provence 2: 1-23.

Benveniste, E. (1974). Structures des relations d'auxiliarité, in: Id., Problèmes de linguistique générale 2, Paris: Gallimard.

Hawkins, John (1978). Definiteness and indefiniteness. A Study in Reference and Grammaticality Prediction. London: Croom Helm.

Hengeveld, K., J. Rijkhoff & A. Siewierska (2004). Parts-of-Speech Systems and Word Order. Journal of linguistics 40, no. 3 (2004): 527–570.

Himmelmann, Nikolaus (1997). Deiktikon, Artikel, Nominalphrase. Zur Emergenz syntaktischer Struktur. Tübingen: Niemeyer.

Ledgeway, A. (2012). From Latin to Romance: Configurationality, Functional Categories and Head-Marking. Transactions of the Philological Society 110(3): 422–442.

Kirsner R. (1979). Deixis in discourse: An exploratory quantitative study of the modern Dutch demonstrative adjectives, in T. Givón (ed.), Syntax and semantics XII, New York, Academic Press, 355-375.