Accueil > Manifestations > Colloques, journées d’étude > Colloques 2016-2017 > Existential Constructions in Typological Perspective

Dernière modification : 19 janvier 2017

Existential Constructions in Typological Perspective

International Workshop




Locative and implicit arguments in Italo-Romance copular constructions
Delia Bentley1, Francesco Maria Ciconte2 & Silvio Cruschina3
1 University of Manchester, UK
2 University of Puerto Rico
3 University of Vienna, Austria

In this talk we analyse a cluster of Italo-Romance copular constructions which have strikingly similar morpho-syntactic properties. These constructions are illustrated in (1a-c) and (2a-c) (Capitals in the idiomatic translations indicate focus).



  • Ce stanno le ciavatte, sotto lu lettu.
  • CL stay.3PL the slippers under the bed
  • ‘THE SLIPPERS are under the bed.’ 

(Terni, Umbria)



  • Guarda : ce sta Maria.
  • Look.IMP CL stay.3SG Maria
  • ‘Look ! THERE’s Maria.’



  • Ce sta lu terremotu : ce stanno li fiji in pericolo.

  • CL stay.3SG the earthquake CL stay.3PL the kids in danger


(2)  a.
  • A-i é le pantofle, sota’l let.
  • EXPL CL be.3SG the slippers under the bed
  • ‘THE SLIPPERS are under the bed.’
(Turin, Piedmont)



  • Guarda : a-i é Maria.

  • Look.IMP EXPL CL be.3SG Maria
  • ‘Look ! THERE’s Maria.’



  • A-i é ’l taramòt : a-i é ëd le masnà an perìcol.

  • EXPL CL be.3SG the earthquake EXPL CL be.3SG of the kids in danger

The word order is VS in all cases. In (1a-c) the copula STARE hosts the etymologically locative clitic ce (<ECCE HIC, Rohlfs 1969, or HINCE, Maiden 1995 : 167), while in (2a-c) the copula ESSE hosts etymologically locative i (<HIC/ILLIC/IBI/ILLI, Benincà 2007). In (2a,c), ESSE patently fails to agree in number with the post-copular noun phrase. The morpho- syntactic comparability between the members of the two clusters in (1a-c) and (2a-c) suggests that the constructions under discussion require a unified analysis.

We discuss historical and synchronic evidence that challenges this hypothesis. In particular, in earlier stages of Italo-Romance, the etymologically locative clitic occurred in complementary distribution with a clause-internal locative phrase, thus proving to be synchronically locative (Ciconte 2008). In Modern Italo-Romance, this complementary distribution is solely observed in a subgroup of the above constructions, illustrated in (1a-b) and (2a-b). Note that, in (1a), (2a) the locative phrase ‘under the bed’ is not part of the same prosodic and syntactic unit as the rest of the clause. Unlike the clitic ce/i in (1a-b) and (2a-b), the cognate and homophonous clitic in (1c), (2c) is not referential, but rather context dependent, in that its semantic contribution can only be understood in the discourse context. In a group of dialects, for example the Venetan dialect of Belluno, this clitic is missing altogether in the counterpart of (1c), (2c), while it is required in the counterpart of (1a-b), (2a- b). The two sets of constructions in (1a-b), (2a-b) and, on the other hand, (1c), (2c) also differ in terms of their presuppositions.

Following Francez (2007, 2010), Cruschina (2012) and Bentley (2015), we claim that a distinction ought to be drawn between two principal types of copular structure, the one being characterised by a topical locative argument (cf. 1a-b, 2a-b) and the other by an implicit spatio-temporal argument (1c, 2c) (see also Koch 2012). Although a unified analysis is unwarranted, we attempt to capture the morpho-syntactic comparability of the constructions in question.

Haut page

Locative and possessive existentials : from Late Latin to Modern Italo-Romance
Delia Bentley1, Francesco Maria Ciconte2 & Silvio Cruschina3
University of Manchester, UK
2 University of Puerto Rico
3University of Vienna, Austria

In this talk we provide a comparative analysis of the two principal types of existential construction that are found in Italo-Romance (Bentley et al. 2015). We illustrate these types in (1a-b) and (2a-b), respectively (Capitals in the idiomatic translations of the examples indicate focus).

 (1)  a.
  • Stattə tranquillə : nge stongə iə.
  • stay.IMP.REFL calm CL stay.1SG I.NOM

  • ‘Do not worry : THERE IS ME.’
(Eboli, Campania)



  • Ng’è u tərramotə : nge stannə i criature ca sə ponnə fa malə.

  • CL be.3SG the earthquake CL stay.3PL the kids who REFL can.3PL do hurt


(2)  a.
  • No te preoccupare : ave a mìe.
  • NEG REFL worry have.3SG to I.ACC
  • ‘Do not worry : THERE IS ME.’
(Martano, Puglia)



  • Ave lu terremotu : ave (li) piccinni in periculu

  • have.3SG the earthquake have.3SG the kids in danger

Examples (1a-b) exhibit a copula derived from Latin STARE, which originally meant ‘stand upright’ or ‘be situated’ (Pountain 1982 : 144 ; see also Peral Ribeiro1958 : 149f.) and later came to be associated with locative as well as a subset of ascriptive copular constructions in some subfamilies of Romance (Leonetti 2015, Bentley & Ciconte 2016, Bentley to appear). This copula hosts the proclitic nge (<ECCE HIC, Rohlfs 1969, or HINCE, Maiden 1995 : 167) and bears finite agreement with the post-copular nominal form. Nge is etymologically locative. In fact, it appears not to have lost its locative function in the copular constructions of early Italo-Romance (Ciconte 2008, 2015).

Examples (2a-b) exhibit a copula derived from HABERE, which figured in a possessive construction of Latin. The post-verbal nominal form takes accusative case, as is clearly testified by differential object marking : the first-person form mie ‘me’ in (2a) is marked with a, unlike the noun phrase (li) piccinni ‘the children’ in (2b). HABERE fails to spell out finite agreement.

The analysis of Late Latin and early Italo-Romance evidence fully supports the hypothesis that the existential constructions under discussion originate from locative and, respectively, possessive sources, thus following a path of development that is very well attested cross- linguistically (Creissel 2013, 1014, Gaeta 2013). Contrastingly, the analysis of Modern Italo- Romance indicates that these constructions have not preserved their original meanings and argument structures (Bentley 2015), although the type in (2a-b) clearly maintains syntactic properties of possessives (Manzini & Savoia 2005, Bentley & Cruschina to appear). Importantly, a wide range of variation is attested in the Modern Italo-Romance existential constructions originating from the one or the other type. Reporting the findings of fine- grained cross-dialectal comparison, we offer a synchronic analysis of the semantics-syntax linking in the two types of existential construction, claiming that - qua existentials - they only differ in syntactic terms. However, Modern Italo-Romance has locative copular constructions that are morpho-syntactically comparable to both types of existential. Whether the same can be said of modern possessive constructions is a moot point.

Haut page

Negating existence : a cross-linguistic, corpus-based account
Bert Cappelle1, Anne Carlier1, Benjamin Fagard2 & Machteld Meulleman3
University of Lille 3, CNRS-UMR 8163 « Savoirs, Textes, Langage » (STL), France
2 CNRS (Lattice) ENS – PSL – University Paris 3, France
3 Université de Reims Champagne-Ardenne, EA 4299 CIRLEP, France

Negative existentials (e.g. Unicorns do not exist), which paradoxically deny the existence of what they presuppose to exist, have long puzzled philosophers, culminating in the early-twentieth-century debate between Russell and Meinong (Cartwright 1960, Clapp 2009, Reicher 2015, Smith 1985). Linguists, however, have only recently turned their attention to negative existentials, especially from a typological point of view (Croft 1991, Kahrel and Van den Berg 1994, Veselinova 2013, 2014). We here investigate the claim that in most languages of Western Europe, there are no ‘special’ negative existentials as these make use of standard negation applied to positive existentials (Veselinova 2013). Our overall assumption is that, though negative existentials do display standard negation, they have distinct syntactic, semantic and pragmatic properties.

First, we hypothesize that if something doesn’t exist at all, it is hardly relevant to specify where it doesn’t exist and how it doesn’t manifest itself. In a sample of 300 positive and negative existentials in Dutch, English and French (extracted from the ParaSol corpus), positive existentials indeed appear to contain an explicit Ground (e.g. There’s a strange cat in the garden) and/or a presentational verb such as appear, erupt, occur etc. significantly more often (p<0.01) than negative existentials, which rely more strongly on grammaticalized structures (e.g. there+be ; il y+avoir in French).

Second, we assume that negative (there-)existentials mainly express backgrounded states and therefore rarely make use of an aorist (e.g. the passé simple in French), which is typically used in foregrounded events. This prediction is confirmed by results obtained for a sample of ca. 500 existentials in French, Italian, Rumanian, Portuguese and Spanish (again from the ParaSol corpus, with a 60/40 positive/negative existential ratio) : aorists are found overwhelmingly in positive existentials (52 positive vs. 7 negative instances). Moreover, we hypothesize that the lexical verb exist or its equivalent in other languages is used for ‘categorical’, predicative judgements about entities whose ontological status can be debated and therefore mainly occurs in negative existentials (e.g. Black holes don’t exist), in contrast to grammaticalized structures with there or equivalent, which are more typically used to introduce a new entity and are therefore relatively more common in positive environments. This meaning-based asymmetry prediction is borne out for English and confirmed in a set of 442 existentials of either type (‘exist’ vs. grammaticalized structure) in Dutch, French, German, Italian, Portuguese and Spanish (p<0.001).

Third, we observe that some negative existential idioms with there (or equivalent in other Western languages) have no positive counterpart. Many of these have a modal meaning and special illocutionary force (e.g. There’s no (use) stopping them ‘one can’t stop them’ ⇒« ‘let’s not (try to) stop them’) or involve scalar reasoning : negative polarity items and comparatives trigger a reinforced reading (e.g. There isn’t a shadow of a doubt ⇒ »‘There isn’t any doubt whatsoever’ ; There’s no better X than Y ⇒ ‘Y is the best X’).

We conclude that negative existentials, even in Western languages, are syntactically different from positive existentials and that these differences reflect their distinct semantico-pragmatic properties.


  • Cartwright, Richard L. 1960. Negative existentials. Journal of Philosophy 57, 629–639.

  • Clapp, Lenny. 2009. The problem of negative existentials does not exist : A case for dynamic semantics. Journal of Pragmatics 41(7), 1422–1434
  • Croft, William. 1991. The Evolution of Negation. Journal of Linguistics 27, 1–39.
Kahrel, Peter and René van den Berg (eds.). 1994. Typological studies in negation. Amsterdam : John Benjamins.
  • Reicher, Maria. 2015. ‘Nonexistent objects’. The Stanford Encyclopedia of Philosophy., accessed 12 January 2016.
  • Smith, Janet Farrell. 1985. The Russell-Meinong Debate. Philosophy and Phenomenological Research 45, 305–350.
  • Veselinova, Ljuba. 2013. Negative existentials : A cross-linguistic study. Rivista di Linguistica 25(1), 107– 146.
  • Veselinova, Ljuba. 2014. The negative existential cycle revisited. Linguistics 52, 1327–1369.

Haut page

There and have existentials in English and French : The case of abstract nouns
Bert Cappelle1, Vassil Mostrov2 & Fayssal Tayalati1
1 University of Lille 3, CNRS-UMR 8163 « Savoirs, Textes, Langage » (STL), France
2 University of Valenciennes, CALHISTE, France

There is a close connection between location and possession in the linguistic expression of existence (e.g. There are three bedrooms in the house / The house has three bedrooms). Both constructional strategies are cross-linguistically well documented (inter alia, Clark 1978, Freeze 1992, Koch 2012, Lyons 1967, Wang & Xu 2013). It has been suggested in the literature (Van de Velde 2003, adopted in Cappelle & Tayalati 2015, Paykin & Van de Velde 2015) that the choice between these two existential constructions is governed by the (non-)autonomy of the existential entity : only the there existential is possible when that entity is autonomous with respect to its location/possessor (e.g. There’s a man in the garden / *The garden has a man) and only the have existential is possible, alongside a structure with an adjectival phrase, when it is not (e.g. *There’s a round shape in the table / The table has a round shape / The table(’s shape) is round). This paper focuses on human-property-denoting abstract nouns (beauty, courage, love, etc.), which, to varying degrees of acceptability, allow both constructions (e.g. There was in that man a deep love for his children / That man had a deep love for his children). Building on studies which have shown that abstract nouns can be ranked from being highly autonomous to being highly dependent (Van de Velde 1995, Mostrov 2010), we here hypothesize that the acceptability of there and have existentials with abstract nouns is largely dependent on their degree of autonomy. We submitted questionnaires involving 72 French and 72 equivalent English sentences containing an abstract noun occurring in 6 different constructions (among which the two existential constructions under study) to native speakers of French and English, respectively, asking them to grade the sentences’ acceptability on a five-point scale. Using the statistical technique of Principal Component Analysis, we find that for both languages, acceptability ratings of abstract nouns for the six constructions demonstrate that speakers roughly classify these nouns along a primary axis of ‘temporal stability’. Abstract nouns which in the literature have been classified as being highly autonomous (esp. French colère and tristesse, as well as the latter’s English equivalent sadness) are ‘attracted’ to the there rather than the have existential and to constructions expressing a situation of limited duration (e.g. Emily’s sadness didn’t last long ; Emily went into a state of sadness), while they are ‘repulsed’ by the genitive of quality (e.g. ??Emily is of a great sadness). Likewise, we find that those abstract nouns previously classified as conceptually highly dependent on their location/possessor (French beauté, maigreur, poids, tempérament, as well as their English equivalents beauty, skinniness, weight, temperament) tend to be ‘attracted’ by the genitive of quality (e.g. Louise is of a fiery temperament) and, especially for poids and tempérament and their English equivalents, to the have rather than the there existential. These findings confirm the psychological reality of previously proposed classes of abstract nouns and reveal interesting positive and negative correlations among there and have existentials and other sentence patterns.


  • Cappelle, Bert & Fayssal Tayalati. 2015. Autonomous entities and the choice between there and have existentials. Spring Colloquium of the Graduate Student Association, 28 March 2015, University of North Carolina at Chapel Hill.
  • Clark, Eve V. 1978. Existential, locative, and possessive constructions. In : Joseph H. Greenberg (ed.), Universals of Human Language, volume 4, pp. 4.85-126. Stanford : Stanford University Press.
  • Freeze, Ray. 1992. Existentials and other locatives. Language 68(3), 553-595.
  • Koch, Peter. 2012. Location, existence, and possession : A constructional-typological exploration. Linguistics 50(3), 533-603.
  • Lyons, John. 1967. A note on possessive, existential, and locative sentences. Foundations of Language 3 : 390–396.
  • Mostrov, Vassil. 2010. Etude sémantique et syntaxique des compléments adnominaux en à et en de dénotant des parties. Ph.D. dissertation. Villeneuve d’Ascq, University of Lille 3.
  • Paykin, Katia & Danièle Van de Velde. 2015. Possession, localisation et existence en russe et en français, International colloquium “Existential Predication in Natural Languages : Values, Operations, Structures, Modalities”, 10-11 April 2015, Paris, Inalco.
  • Van de Velde, Danièle. 1995. Le spectre nominal. Des noms de matières aux noms d’abstractions. Leuven/Paris : Peeters.
  • Van de Velde, Danièle. 2003. Existence et localisation. Ms., séminaire de DEA. University of Lille 3.
  • Wang, Yong & Xu, Jie. 2013. A systemic typology of existential and possessive constructions. Functions of Language 20(1), 1-30.

Haut page

Existence and Information Structure : Comparative Perspective in Romance Languages
Anne Carlier1, Benjamin Fagard2, Lena Karssenberg3, Karen Lahousse3 & Machteld Meulleman4
1 University of Lille 3, CNRS UMR 8163 « Savoirs, Textes, Langage » (STL), France
2 CNRS (Lattice), ENS – PSL – Université Paris 3, France
3 Katholieke Universiteit Leuven, Belgium

4 University of Reims Champagne-Ardenne, EA 4299 CIRLEP, France

Since John Lyons’ (1967) seminal paper, it is widely accepted that there is a cross-linguistic correlation between predicates expressing existence, possession and location. On the basis of earlier proposals of Clark (1978) & Bickerton (1981), Koch (2012) redefines the relationship between existential predication and possessive and locative predicates as represented in the figure below :

PNG - 83.4 ko

Existential predication can thus be lexicalized in terms of rhematic possession or in terms of rhematic location. Both patterns are attested in Romance (e.g. Fr. il y a ; Sp. Hay ; It. c’è, …).

This analysis raises the following research questions :

  • How can we define the information structure conveyed by sentences containing a grammaticalized existential predicate ? Are they systematically entirely rhematic or all-focus ?
  • To what extent can existential markers function as information structure markers ?


Our paper will be devoted to the second research question and will examine the discourse function of existential predicates in Romance. Taking into account the fact that the Romance languages lost at different degrees the flexibility of constituent order which was characteristic of Latin (De Mulder & Lamiroy, Lamiroy & De Mulder 2011, Carlier, De Mulder & Lamiroy 2012 ; Lahousse & Lamiroy 2012), the central hypothesis we will argue for is that there is a correlation between the variety of discourse functions of the existential predicate and the degree of syntactization of constituent order in each Romance language.


  • Bentley, Delia. 2015. Existentials and Locatives in Romance Dialects of Italy. Oxford : Oxford University Press.
  • Bickerton, Derek. 1981. Roots of Language. Ann Arbor : Karoma.
  • Carlier A., W. De Mulder & B. Lamiroy (2012). The Pace of Grammaticalization in a Typological Perspective. Folia Linguistica 46/2, 287-302.
  • Clark, Eve V. 1978. Locationals : existential, locative, and possession constructions. In Joseph H. Greenberg (ed.), Universals of Human Language. Vol. IV : Syntax, 85–126. Stanford : Stanford University Press.
  • Creissels, Denis . Forthcoming. Existential predication in typological perspective.<>
  • Cruschina, Silvio. 2015. Patterns of variation in existential constructions. sogloss 2015, 1/1 : 33-65.
  • De Mulder, W. & B. Lamiroy. 2012. Gradualness of grammaticalization in Romance. The position of French, Spanish and Italian. In : Davidse, K. et al. Eds. Grammaticalization and language change : new reflections, 199-227. Amsterdam : J. Benjamins.
  • Koch Peter. 2012. Location, existence, and possession : A constructional-typological exploration. Linguistics 50–3, 533 – 603.
  • Lahousse, K. & B. Lamiroy. 2012. Word order in French, Spanish and Italian. A grammaticalization account. Folia Linguística, 46, 2, 387-417.
  • Lamiroy, B. & W. De Mulder 2011. Degrees of grammaticalization across languages. In : Narrog H. & B. Heine (eds). The Oxford Handbook of Grammaticalization, 302-317. Oxford : OUP.

Haut page

Existentials in Relative Clauses – a comparison of French il y a and Spanish hay
Charlotte Coy
University of Tübingen, Germany

Carlson (1977) observed that English there is can only be used in so-called amount relatives (1a) and is impossible in relative clauses with singular count nouns as antecedents (1b). Dobrovie- Sorin/Beyssade (2004, 2012) adapt this idea to French and claim that similar restricions hold for French il y a (2a,b). Both Carlson and Dobrovie-Sorin/Beyssade explain these restrictions with the well-known definiteness effect on existentials (cf. Milsark 1977).

(1) a. Every man there was on the life-raft died. (= 6b in Carlson 1977 : 521)
  b. * The man that there was in Austria likes Bob. (= ex. 14c in Carlson 1977 : 526)
(2) a. Rome est le plus splendide musée qu’il y ait au monde. La quantité de chefs-d’oeuvre qu’il
 y a dans cette ville, c’est étourdissant ! (Frantext catégorisé)
  b. *Le livre qu’il y avait dans la bibliothèque de ma grand-mère a brûlé. (= ex. 92 in Dobrovie-Sorin/Beyssade 2012 : 124)
(3) a. Gómez rebuscó entre los papeles que había encima de su mesa. (CORPES XXI)
  b. Cogí la grapadora que había sobre su mesa y me puse a inspeccionarla. (CORPES XXI)

In Spanish, however, relative clauses containing the existential hay do not display similar restrictions (3a,b), although the definiteness effect with hay is even stronger than with il y a (Meulleman 2012, cf. also Escandell Vidal/Leonetti 1998). The definiteness effect may thus not be the only reason for the restrained use of il y a in relative clauses. Furthermore, whereas the chief function of existentials is normally to introduce new referents into the discourse (cf. Lambrecht 1994), this cannot be the case here. Since the pivot of the existential is encoded as the antecedent of the relative clause, it is already introduced before the existential occurs.

In my presentation, I will thus address two questions :

  • which reasons beyond the definiteness effect explain the differing use of il y a and hay in relative clauses ?

  • which functions do il y a and hay serve in these contexts ?

My presentation is based on an extensive corpus study in Frantext catégorisé, CORPES XXI and C Oral Rom.


  • Carlson, G. (1977) : « Amount relatives ». In : Language 53, 520–542.
  • CORPES XXI = Real Academia Española : Banco de datos (CORPES XXI) [online]. Corpus del Español del Siglo XXI. <;;.

  • C Oral Rom = Cresti, E./Moneglia, M. (2005) : C-Oral-Rom : Integrated Reference Corpora for Spoken Romance Languages. Amsterdam/Philadelphia.

  • Dobrovie-Sorin, C./ Beyssade, C. (2004) : Définir les indéfinis. Paris.
— (2012) : Redefining Indefinites. Dordrecht.

  • Escandell Vidal, M. V./Leonetti, M. (1998) : « Construcciones existenciales y oraciones de relativo ». In : G. Rufino (ed.) : Atti del XXI Congresso Internazionale di Linguistica e Filologia Romanza. Tübingen, 261-272.

  • Frantext = Base textuelle FRANTEXT, ATILF - CNRS & Université de Lorraine. <;;.
  • Lambrecht, K. (1994) : Information structure and sentence form. Cambridge.
  • Meulleman, M. (2012) : « Degrees of Grammaticalization in three Romance Languages : a comparative analysis of existential constructions ». In : Folia Linguistica 46, 417-451.
  • Milsark, G. (1977) : « Toward an Explanation of Certain Peculiarities in the Existential Construction in English ». In : Linguistic Analysis, 3, 1-30.

Haut page

Existential predication in the languages of the Sudanic belt
Denis Creissels
University of Lyon, France

The Sudanic belt (Clements & Rialland 2008), aka Macro-Sudan belt (Güldemann 2008), is a large belt of northern sub-Saharan Africa from the Atlantic Ocean to the Ethiopian plateau. Some important structural characteristics are particularly frequent among languages spoken in this area irrespective of their genetic affiliation (Westermann 1911, Greenberg 1959), and are not found with a comparable frequency in the genetically related languages outside of this region, which suggests an important role of language contact. Recent areality hypotheses dealing with the Sudanic belt have focused on features such as labial-velar stops, labial flaps, implosives and other “nonobstruent” stops, nasal vowels and lack of contrastive nasal consonants, ATR vowel harmony, tone, “lax” polar question markers, logophoricity markers, S-(Aux)-O-V-X and V-O-Neg order patterns. My talk deals with an areal feature of the Sudanic belt not mentioned so far in the literature : the particularly high frequency of a type of existential predication which is relatively rare at world level.

Existential predications (There is a book (on the table)) provide an alternative way of encoding the prototypical figure-ground relationships also denoted by plain locational clauses (The book is on the table), from which they differ in the perspectivization of figure-ground relationships (Borschev and Partee 2002, Partee and Borschev 2004 & 2007).

Probably more than half of the world’s languages are devoid of a special predicative construction encoding the existential perspectivization of figure-ground relationships (Creissels, Forthcoming). In many cases (Avar, Basque, Czech, Finnish, Kabyle, etc.), constituent order in locational predication is flexible, and variation in constituent order provides a rough equivalent of existential perspectivization :

Finnish (Uralic ; Huumo 2003 : 464)
Poika on pih-lla. vs. Piha-lla on poika.
boy is in_yard   in_yard is boy
‘The boy is in the yard.’   ‘There is a boy in the yard.’

However, some of the languages devoid of a dedicated existential construction also have a rigid constituent order in locational predication. In such languages, in the absence of indications provided by definite/indefinite markers or focus markers, the same locational clauses can be used indiscriminately in contexts that would trigger a choice between locational and existential predication in other languages :

Mandinka (Mande ; pers.doc.)
Wùlôo yíròo kótò.
dog is tree under
’The dog is under the tree.’ or ’There is a dog under the tree.’

Languages with rigid order in locational clauses and no contrast between locational predication and a dedicated existential predicative construction are particularly common in the Sudanic belt. In my talk, after presenting the types of locational and existential predication found in the languages of the Sudanic belt according to their geographical and genetic distribution, I will discuss possible correlations with other areal features of the Sudanic belt.


  • Creissels, Denis . Forthcoming. Existential predication in typological perspective.< ;
  • Borschev, Vladimir & Barbara Partee. 2002. ‘The Russian genitive of negation : theme-rheme structure of perspective structure ?’. Journal of Slavic Linguistics 10. 105-144.
  • Clements, G. N. & Annie Rialland, 2008. ‘Africa as a phonological area’. In Bernd Heine & Derek Nurse (eds.), A Linguistic Geography of Africa. Cambridge : Cambridge University Press, pp. 36-85.
  • Güldemann, Tom. 2008. ‘The Macro-Sudan belt : towards identifying a linguistic area in northern sub-Saharan Africa’. Bernd Heine & Derek Nurse (eds.), A linguistic geography of Africa. Cambridge : Cambridge University Press, 151-185.
  • Westermann, Diedrich H. 1911. Die Sudansprachen [The Sudanic languages]. Hamburg : L. Friederichsen & Co.
  • Partee, Barbara H. & Vladimir Borschev. 2004. ‘The semantics of Russian genitive of negation : the nature and role of perspectival structure’. In Kazuha Watanabe & Robert B. Young (eds.), Proceedings of Semantics and Linguistic Theory (SALT) 14. Ithaca NY : CLC Publications. 212-234.
  • Partee, Barbara H. & Vladimir Borschev. 2007. ‘Existential sentences, BE, and the genitive of negation in Russian’. In I. Comorovski and K. von Heusinger (eds.), Existence : Semantics and Syntax. Dordrecht : Springer. 147–190.

Haut page

Pure existentials vs. pure presentationals : finding an existence out(side) of place
Livio Gaeta
University of Turin, Italy

Existential constructions are usually defined as sentences in which an entity is associated with some location. In spite of the apparent simplicity of the so-called locative paradigm it leaves unaccounted cases in which existential constructions cannot be immediately connected with any locative basic format. This opens door for identifying pure existentials which only predicate the existence of an entity not necessarily connected with a physical location. In the paper, pure existentials will be contrasted with pure presentationals which can be investigated in a similar way by abstracting away from the physical dimension typical of presentational constructions, with the general aim of sketching a brief typology of the possible source constructions underlying these constructions.

Haut page

Indefinite subject and locative in French and Hungarian
Zsuzsanna Gécseg1 & Laure Sarda2
1 University of Szeged, Hungary
2 CNRS (Lattice) ENS – PSL – University Paris 3, France

Existential constructions have been widely characterized as having a marked syntax, where the pivot (entity whose existence is asserted) appears in post-copula position as illustrated in (1a) (cf. among others Mc Nally 2011, Bentley et al 2013). The presence of a locative (usually called the coda) has also been considered constitutive of existential constructions, although not always obligatory (cf. among others Leonetti 2008). The usual terminology to refer to constituents of the existential construction is laid out in (1b). Alternatively, Creissels (2014) suggests that an existential construction can be conceived of as an inverse locational predication between a Figure and a Ground (1c). Partee and Borschev (2006) considered that the difference between a locational and existential construction lies on the Perspectival Structure : existential sentences are location-centered whereas locational sentences are Thing-centered.

(1) a. There is a cat under the chair
  b. Proform copula pivot coda
  c. GramLoc predicate Figure ground
  d.     Thing Location

In this paper, we study the relationship between the position of an indefinite subject and the presence vs. absence of a locative. We adopt a contrastive perspective comparing French sentences as in (2) with their Hungarian translations as in (3).

(2) Un tout jeune chat jouait sous les chaises (Hugo, Les misérables p. 455)
  a very young cat play PAST under DEF.PL chair.PL
  A very young kitten was playing about among the chairs’ [English Translation in Parasol]


 (3) Egy fiatal macska játszadozott a székek alatt. [Hungarian Translation in Parasol]
  a young cat play.past DEF chair.PL

In the French sentences (i.e. (2)), the indefinite subject appears in preverbal position. We explore parameters leading to an existential reading of those sentences despite their unmarked [SVX] word order. They are considered functionally equivalent with the prototypical existential construction, which in French is characterized by a marked word order, with a post-copular subject as in (4) :

(4) il y avait un chat (qui jouait) sous les chaises
  Expl proform have.past a cat (who was playing) under chairs

In the Hungarian translations, the picture is more complex. Hungarian is usually characterized as a discourse-configurational language, that is, the syntactic position of the constituents is not determined by their grammatical function but by their discourse status. There are two positions available for indefinite subjects in the preverbal field : the Topic or the Focus/Verbal Modifier position, and they also may appear postverbally (Kálmán 1985, É. Kiss 2002). Existentials in Hungarian can be realized in several different word orders, all of which differ from the word order used in canonical locational clauses. As a general rule, the subject of an existential sentence appears in the comment part of the sentence : either in a preverbal comment position or after the verb. The sentence may also contain a locative expression that fills the Topic position or appears postverbally (É. Kiss 2002, Viszket 2004, Kálmán 1995, Maleczki 2008, 2010, Hegedűs 2013).

This paper addresses the following questions : (i) what are the parameters influencing the obligatory vs. facultative presence of a locative in utterances with indefinite subject in French and in Hungarian ? (ii) Which relationships hold in Hungarian between the position of the indefinite subject and the obligatory vs. facultative presence of a locative ?


  • Bentley, D., Ciconte, F.M., Cruschina, S., 2013, “Existential constructions in crosslinguistic perspective”, Rivista di Linguistica, 25, 1, 1-13.
  • Creissel, D., 2014, “Existential predication in typological perspective”, 46th Annual Meeting of the Societas Linguistica Europaea, Split, 18 - 21 September 2013.
  • É. Kiss, K. (2002) : The  Syntax of Hungarian. Cambridge : Cambridge University Press.
  • Hegedűs, V. (2013) : Non-verbal predicates and predicate movement in Hungarian, PhD-thesis, Tilburg University
  • Kálmán, L. (1985) : “Word order in neutral sentences” In : Kenesei, I. Approaches to Hungarian (vol. 1). Szeged : JATE Press, 13-23.
  • Kálmán, L. (1995) : “Definiteness effect verbs in Hungarian”. In : Kenesei, I. (ed.), Approaches to Hungarian Vol 5. Szeged : JATE Press, 221-242.
  • Leonetti, M. 2008. “Definiteness Effects and the Role of the Coda in Existential Constructions”. in A. Klinge and H. Hoeg-Müller (eds), Essays on Determination. Amsterdam : John Benjamins, 131-162.
  • Maleczki, M. (2008) : “Határozatlan argumentumok” [Indefinite arguments], In : Kiefer, F. (ed.), Strukturális magyar nyelvtan 4. A szótár szerkezete [A structural grammar of Hungarian, volume 4 : The structure of the lexicon], Budapest : Akadémiai Kiadó, 129-184.
  • Maleczki, M. (2010) : “On the definiteness effect in existential sentences : Data and theories”, In : Bibok, K. & Nemeth. T., E. (eds.), The role of data at the semantics-pragmatics interface, Berlin and New York : Mouton de Gruyter, 25-56.
  • McNally, L.E., 2011, “Existential sentences”, in Maienborn, C., Von Heusinger, K., Portner, P. (éd.), Semantics : an international handbook of natural language meaning, vol. II, Berlin : Mouton de Gruyter, 1829-1848.
  • Partee, B. H., and Borschev, V. (2006) : Information structure, Perspectival Structure, diathesis alternation, and the Russian Genitive of Negation. In Proceedings of Ninth Symposium on Logic and Language (LoLa 9), Besenyőtelek, Hungary, August 24–26, 2006, eds. Beáta Gyuris, László Kálmán, Chris Piñón and Károly Varasdi, 120-129. Budapest : Research Institute for Linguistics, Hungarian Academy of Sciences, Theoretical Linguistics Programme, Eötvös Loránd University.
  • Viszket, A. (2004) : Argumentumstruktúra és lexicon [Argument structure and the lexicon]. PhD dissertation, Budapest : Eötvös Loránd University.

Haut page

The Finnish existential clause : structure, meaning, discourse function, and quantification
Tuomas Huumo
University of Turku, Finland

At least since the 1950’s, the structurally and semantically idiosyncratic clause type known as the existential clause has been one of the most intensively studied topics in Finnish syntax. Distinctive morphosyntactic features of the construction include its inverse XVS word order, the lack of subject-verb agreement (in case the construction is analyzed as having a subject at all), and the partitive case marking of the existential S argument (abbreviated as Se). The Se was traditionally analyzed as a grammatical subject, but its subjecthood has been questioned by many scholars since the 1970’s.

From an early focus on verb semantics and the conditions of using the partitive-marked Se, subsequent work has proceeded towards construction-level analyses of the existential construction, as well as analyses of its discourse function. It has been demonstrated, for instance, that the Se is not a clause-level topic or a semantic starting point for the predication (in the sense of Langacker 1991:293 ), and that its referent is typically not tracked in subsequent discourse. Such construction-level points of view were already touched upon by early scholars including Ikola (1954) and Schlachter (1958), and this line of research has been continued e.g. by Huumo (2003) and, from an interactional-linguistic point of view, by Helasvuo (2001). In a recent joint paper, Huumo and Helasvuo (2015) have argued against the analysis of the Se as a grammatical subject, basing their argument on its grammatical, semantic and discourse properties, which all deviate from those of canonical subjects.

In my presentation, I will start with an overview of the syntactic and semantic nature of the Finnish existential clause, to be followed by a concise research history of the construction : why is it that this clause type has evoked so much interest and how have the problems it poses been dealt with in approaches of different times ? I will then discuss in more detail the grammatical, semantic and discourse functions of the Se, with an emphasis on the aspectual and quantificational meanings expressed by the construction. Traditionally, a partitive-marked Se has been analyzed as a designator of an unbounded quantity of an indefinite referent (or referents, if the Se is in the plural), and this often affects the clause-level nominal aspect of the construction, making it unbounded accordingly. I will introduce new insights to the analysis of the Se from the point of view of mass and number quantifiers, which are quite common in the Se but have been largely ignored by previous approaches. In general, a quantified Se is understood as designator of a bounded quantity even if marked with the partitive case (as first pointed out by Yli-Vakkuri 1979). This has important consequences to the clause-level aspectual meaning of the construction, especially in cases where its aspectual nature is based on quantificational features of the Se.


  • Helasvuo, Marja-Liisa 2001. Syntax in the making : The emergence of syntactic units in Finnish conversation. Studies in discourse and grammar 9. Amsterdam : John Benjamins.
  • Huumo, Tuomas. 2003. Incremental Existence : The World According to the Finnish Existential Sentence. Linguistics 41(3) : 461–493.
  • Huumo, Tuomas and Helasvuo, Marja-Liisa 2015. On the subject of subject in Finnish. In : Helasvuo, Marja-Liisa and Tuomas Huumo (eds.), Subjects in constructions : canonical and non-canonical. Constructional approaches to language 16. Amsterdam : John Benjamins.
  • Ikola, Osmo 1954. Suomen lauseopin ongelmia I – III. Virittäjä.
  • Langacker, Ronald W. 1991. Foundations of Cognitive Grammar. Vol. II : Descriptive application. Stanford : Stanford University Press.
  • Schlachter, Wolfgang 1958. Partitiv und Inkongruenz beim Subjekt des Finnischen. – Finnisch-Ugrische Forschungen 33 s. 3–95.
  • Yli-Vakkuri, Valma 1979. Partitiivisubjektin toiset juuret : eräs kvantiteetin ilmaisujärjestelmän ilmentymä. Kallio, Jussi, Häkkinen, Kaisa and Kytömäki, Leena (eds.), Sanomia. Juhlakirja Eeva Kangasmaa-Minnin 60-vuotispäiväksi 14.4.1979. Publications of the Departmnt of Finnish and general linguistics 9 : 155–192. Turku : University of Turku.

Haut page

A multifactorial quantitative account of existential constructions across languages and language families
Karolina Krawczak1, Frédérique Becquet2 & Laure Sarda2
Faculty of English, Adam Mickiewicz University, Poznan, Poland
2 CNRS (Lattice) ENS – PSL – University Paris 3, France

This study is part of the project Space Time and Existence (labex TransferS, 2012-2016), which is a cross-linguistic investigation of existential constructions. The project has resulted in a richly annotated database of existential constructions in English and their functional equivalents in 12 other languages, i.e., French, Italian, Spanish, Portuguese, Romanian, German, Dutch, Russian, Polish, Greek, Hungarian and Arabic. This multilingual database amounts to approximately 1200 examples extracted from Master and Margarita by Bulgakov via the Parasol platform (v. Waldenfels 2011). The data extraction procedure used English as the primary language. Accordingly, a hundred occurrences of [there + be] were first extracted from the English translation and then aligned with the other languages.

The present study adopts the profile-based approach (Geeraerts et al. 1994 ; Gries 2003 ; Gries & Stefanowitsch 2006 ; Glynn & Fischer 2010 ; Glynn & Robinson 2014) for the analysis of the metadata. This approach applies multivariate statistical modeling to manually annotated data in order to identify frequency-based usage patterns associated with the linguistic phenomenon under investigation. Our principal goal is to reveal such language-specific patterns for existential constructions. More specifically, in doing so, we intend to test a set of hypotheses focused on differentiating between “essential” and “contingent existentials”. Some of the usage properties that are particularly revealing in this regard include :

  • Copula type
  • Construction type (personal vs. impersonal)
  • Transitivity
  • Telicity
  • Presence of locative adverbials
  • Referential properties of the Figure/ Pivot NP (generic vs non generic)
  • Part-whole relationship between referents of Figure/ Pivot and Ground/ Coda.


Our tentative results show structural preferences of languages or language families to express different types of existence. Weighting parameters has enabled us to provide a map of the expression of existence, illustrating prototypical existential structures, but also giving an inventory of a large range of alternative strategies found cross-linguistically in the translational equivalents. Importantly, in addition to presenting the frequency-based patterns typical of existential constructions, our study also zooms in on a number of less frequent structures, which, despite their sparseness, represent interesting cases.

Selected References

  • Geeraerts, Dirk, Stefan Grondelaers, & Peter Bakema. 1994. The Structure of Lexical Variation : Meaning, naming, and context. Berlin : Mouton de Gruyter.
  • Glynn, D. & J. Robinson (eds.). 2014. Polysemy and Synonymy. Corpus methods and applications in Cognitive Linguistics. Amsterdam : John Benjamins.
  • Glynn, D. & K. Fischer (eds.). 2010. Quantitative Methods in Cognitive Semantics : Corpus-driven approaches. Berlin : Mouton de Gruyter.
  • Gries, St. Th. 2003. Multifactorial Analysis in Corpus Linguistics. London : Continuum.
  • Gries, Stefan Th. & Anatol Stefanowitsch. 2006. Corpora in Cognitive Linguistics : Corpus-based approaches to syntax and lexicon. Berlin : Mouton de Gruyter.
  • v. Waldenfels, R. 2011. Recent developments in ParaSol : Breadth for depth and XSLT based web concordancing with CWB. In : Daniela M., and Garabík, R. (eds.), Natural Language Processing, Multilinguality. Proceedings of Slovko 2011, Modra, Slovakia, 20–21 October 2011. Bratislava, 156-162.

Haut page

Existential Sentences Cross-Linguistically : Variations in Form and Meaning
Louise McNally
Universitat Pompeu Fabra, Spain

In this talk, which reports on McNally (2016), I present an overview of the range of structural variation found cross-linguistically in sentences that have been identified as “existential”, arguing that this variation in form is accompanied by variation in meaning. I first offer examples of existential sentences based on distinguished existential predicates, as well as on copular, possessive, and expletive or impersonal constructions, examples of which are shown in (1)-(4), respectively.

(1) Güaha buteya gi hälum kahun ´ais.
  sg.exist bottle inside box ice
  ‘There’s a bottle in the icebox.’
  (Chamorro, Chung 1987, example 5a)


(2) Piha-lla on poika.
  yard.ade be.pres.3sg boy.nom
  ‘There is a boy in the yard.’
  (Finnish, Huumo 2003, example 4)


(3) Il y a trois chambres et deux salles de bain…
  il LOC has three bedrooms and two bathrooms
  ‘There are three bedrooms and two bathrooms…’


(4) Es gibt viele Gänseblümchen in meinem Garten.
  it gives many daisies in my garden
  ‘There are many daisies in my garden.’
  (German, Adapted from Czinglar 2002, example 2a)

I then turn to variation in the compositional semantics and discourse function of these structures. I submit that the attested semantic and pragmatic variation argues against universalist approaches to the syntax of existentials such as that in Freeze (2001), and is aligned instead with the variationist approach to existentials defended in Gaeta (2013).

Selected references

  • Freeze R. 2001. Existential constructions. In Language Typology and Language Universals, ed. M Haspelmath, E König, W Oesterreicher,W Raible, pp. 941–53. Berlin/New York : de Gruyter.
  • Gaeta L. 2013. Existential constructions : a semasiological perspective. In Argument Structure in Flux, ed. E van Gelderen, M Cennamo, J Barðdal, pp. 477–509. Amsterdam : Benjamins.
  • McNally, Louise. 2016. Existential Sentences Cross-Linguistically : Variations in Form and Meaning. Annual Review of Linguistics 2, 211-231.

Haut page

Natural phenomena and spatio-temporal locatives
Katia Paykin1 & Machteld Meulleman2

University of Lille 3, CNRS UMR 8163 « Savoirs, Textes, Langage » (STL), France
2 University of Reims Champagne-Ardenne, EA 4299 CIRLEP, France

Within various theoretical frameworks, it has often been emphasized that natural phenomena constructions contain a locative PP, either situated on the (sub)syntactic structure level or as a part of the information structure (cf. among others, Bolinger 1973, 1977, Erteshik-Shir 1997, Taylor 2001). However, the analysis of such expressions has been mostly limited to the domain of weather in general and precipitations in particular, while other natural phenomena, if mentioned at all, have been assumed to behave in the same way. The aim of the present contribution is (a) to investigate how spatio-temporal locatives function in the expression of weather, and (b) to verify whether the general class of natural phenomena is really homogeneous in this respect.

Drawing on examples from three different languages, English, French and Russian, we argue that weather utterances, independently from their syntactic encoding, can convey both thetic and categorical judgments. In the case of a thetic reading, they either contain an explicit locative or establish a link with the speaker’s location, as in (1). This also applies to cyclic phenomena such as sunsets and tides, as in (2), and natural disasters such as earthquakes and volcano eruptions, as in (3).

(1) RU Tuman / Vetrenno (v Tule).
  fog / windy (in Tula)
  ‘There is fog / It is windy (in Tula).’
(2) FR Le soleil se couche / C’est la pleine lune / C’est la marée haute (à Arcachon).
(3) EN There is an earthquake / The volcano is erupting (in Italy).

In the categorical reading, independently of the syntactic construction used, weather expressions have a true spatio-temporal topic which is characterized by a true predicate denoting a kind of climatological property of a given place or a seasonal time-span, as in (4). However, with cyclic phenomena, the habitual reading is pragmatically impossible, regardless of the spatio-temporal locative, as in (5). In the case of natural disasters, habitual sentences with an explicit locative function like weather expressions, but since the occurrence of natural disasters is by definition unpredictable, temporal locatives cannot denote a particular season, as in (6).

(4) RU V Pariže // Vesnoj často tuman / vetrenno.
  in Paris // in-spring often fog / windy
  ‘In Paris // In spring there is often fog / it is often windy.’
(5) FR *C’est souvent la pleine lune (à Paris / en hiver).
(6) EN There are often earthquakes in Japan / lately /  ??in winter.

The fundamental difference between thetic and categorical weather judgments can be made explicit through the use of adverbs meaning ‘outside’. Thetic weather judgments, as in (7), authorize these adverbs, which refer to the observer of the phenomenon. However, in habitual weather judgments, such adverbs can no longer appear, as illustrated in (8). In the case of other natural phenomena, the presence of an observer can be expressed in thetic judgments with cyclic daylight phenomena, but not with tides nor with natural disasters, as shown in (9).

(7) RU Na dvore / na ulice/ zo oknom dožd’ / vetrenno.
  on yard / on street / behind window rain / windy
  ‘It is raining / windy outside.’
(8) FR  ??A Paris il pleut souvent dehors.
(9) a. EN Wake up ! The sun is already coming up outside.
  b. EN  ??It is high tide outside.
  c. EN ??There is an earthquake / a volcano eruption outside.

Thus, we argue that the linguistic expression of weather phenomena is characterized by the particular behavior of spatio-temporal locatives in both thetic and categorical weather judgments, at least in the three languages under study. Although other natural phenomena share some characteristics of weather events, we show that this semantic class should nevertheless be subdivided conceptually into those phenomena that are cyclic and non-cyclic, on the one hand, and atmospheric and non-atmospheric, on the other hand.


  • Bolinger, D. (1973) “Ambient it is Meaningful Too”, Journal of Linguistics 9, 261-270.
  • Bolinger, D. (1977) Meaning and Form. London : Longman.
  • Erteshik-Shir, N. (1997) The Dynamics of Focus structure. Cambridge University Press, Cambridge.
  • Taylor, K. (2001) “Sex, Breakfast, and Descriptus Interruptus”, Synthese 128, 45-61.

Haut page

To Be and its Negation in Latvian
Daniel Petit
École normale supérieure - PSL, France

A striking peculiarity of the Baltic languages (Old Prussian, Lithuanian, Latvian) is the way they express the existential verb and the copula ‘to be’. Whereas Old Prussian, the only documented West Baltic language, extinct at the end of the 17th century, still had an Indo-European-looking paradigm (e.g. asmai ‘I am’, ast ‘he is’, ni asmai ‘I am not’, ni ast ‘he is not’), Lithuanian and Latvian, the two remaining languages of the East Baltic subgroup, have developed new forms of the verb ‘to be’ and of its negation. In Lithuanian the third person is yrà ‘he is’, probably an emphatic particle, and its negation nėrà ‘he is not’, originally a contamination of *nėsti (< *ne-esti) and yrà. More striking is the situation in Latvian, where the third person is ir ‘he is’ (= Lith. yrà), but its negative nav ‘he is not’. Suppletion between positive and negative is a typological rarity in the Indo-European languages, the only parallel being Polish jeść ‘there is’, vs. nie ma ‘there is not’, limited to the existential verb. The question raised by the Latvian paradigm is first whether the suppletive relationship between ir and nav first arose in existential contexts and was secondarily extended to the other uses of ‘to be’, including those as a copula, and second where the negative form of ‘to be’ comes from. It can be shown that nav is based on a longer form *nevaida, originally the negative of a verb *vaidât ‘to appear, to show itself’. This analysis points to an existential origin of Latvian nav(‘there does not appear, there is not’) and this shows that a negative existential is not only the negation of a positive existential. In Baltic, it can be argued that the unbalance between ir and nav reflects a chronological difference in the elimination of nominal (a-verbal) sentences in the positive and in the negative.

Haut page

Existence and Perception
Laure Sarda1, Machteld Meulleman2, Charlotte Danino3, Benjamin Fagard1 & Eva Soroli4
1 CNRS (Lattice) ENS – PSL – University Paris 3, France
University of Reims Champagne-Ardenne, EA 4299 CIRLEP, France
3 Université Sorbonne Nouvelle, France
4 University of Lille 3, CNRS UMR 8163 « Savoirs, Textes, Langage » (STL), France

That existence manifests itself through perception, or that “to be is to be perceived” has long been at the center of philosophical debates. « Is sound only sound if a person hears it ? » Dating back to George Berkeley’s riddle of the sound of a tree falling down in an uninhabited island, the question is whether the tree (and the sound it produces when falling down) exists outside of human perception. In other terms, the question is : “can we assume that there is a non-observed reality ? And how can we know it ?”

The study of the linguistic expression of existence applies to some extent to this philosophical question. Working on possible translations of existentials in parallel corpora (, we noticed that prototypical English constructions as in (1a, 2a and 3a) could sometimes be translated with perception verbs as in (1b, 2b and 3b). 

(1) a. There were chairs of extraordinarily complex construction
  b. On y voyait des fauteuils extraordinairement compliqués
    One there see.PAST DET.IND.PL chairs extremely complex
         (Parasol, Bulgakov)


(2) a. Above all, there was not even any scar left on the neck.
  b. Und das schönste, nicht mal eine Narbe war am Hals zu sehen.
    and the prettiest, not once a scar was at the neck to see
         (Parasol, Bulgakov)


(3) a. He could predict a thunderstorm when there was not the least cloud
  b. Katalavene oti erhotan bora protou fani esto ki ena synefaki
    Understand.PAST that was.coming thunderstorm before is.visible even and one cloud
     (Parasol, Perfume)

This first observation led us to assume a functional parallelism between ‘pure’ There+BE existential constructions (1a, 2a and 3a) and perception verb expressions with existential meanings (1b, 2b and 3b), thus noting some common properties between the domains of existence and perception. The question we address here is, how far does this equivalence go ? In order to investigate this, we compare properties and constraints associated with existential or perceptual constructions from a cross-linguistic perspective in French, Spanish, English, German and Greek. We illustrate the extent to which visual perception verbs can be used to express existence, and show that the overlap between the two domains is only partial.

We present clusters of morphological, syntactic, semantic and pragmatic properties that tend to correlate with either the existential construction or the perceptual construction. Then, we suggest a conceptualization of those two clusters of properties as characterizing the two poles of a continuum running from “pure existential” to “pure perception”. 

The existential construction tends to be impersonal, thetic and atelic, with a post-copula indefinite Pivot (Figure). On the other hand, the perception construction is prototypically personal, categorical and telic, with definite as well as indefinite NPs in object position. Those opposite clusters of properties are somewhat fused in some expressions as (1b, 2b and 3b) above or (4) and (5) below, where perception verbs enter into a construction borrowing some properties from existential constructions such as impersonality, atelicity and indefinite NP.

(4) No se veían leyes raciales desde el 1938
  neg reflx from Det.def 1938


(5) Vlepoume olo ke ligotero pnevma enosis ke omonias stous thesmous aftous
  See.pres.1pl less and less spirit union.GEN and concord.GEN in.the institutions these
  ‘There is less and less spirit of union and concord in these institutions’

The study describes how the domain of existence spread out from the prototypical existence of conceptual reality (types) to the existence of perceptible reality (token). Similarly, perception exceeds the concrete domain of physical apprehension of an entity or event to gain metaphorical senses. We can SEE, for example, representations or mental images without physically perceiving them. They are apprehended though the same way as perceptible objects in the world, at least verbally. We suggest a multi-dimensional characterization of possible linguistic constructions that express existence along a cline taking into account not only their morphological, syntactic and semantic features but also the pragmatic factors having to do with evidentiality and direct or mediated access to referents.

Selected references

  • Bentley, D., Ciconte, F.M. & Cruschina, S. (2013). Existential constructions in crosslinguistic perspective. Rivista di Linguistica, 25 (1), 1-13.
  • Creissel, D. (2014). Existential predication in typological perspective, 46th Annual Meeting of the Societas Linguistica Europaea, Split, 18 - 21 September 2013.
  • Kleiber, G. (2003). Indéfini, partitif et adjectif : du nouveau. La lecture individualisante, Langages, 151, 9-28.
  • Viberg, Å. (2008). Swedish verbs of perception from a typological and contrastive perspective. In de los Ángeles Gómez González, M., Lachlan Mackenzie, J. & González Álvarez, E. M. (eds.). Languages and Cultures in Contrast and Comparison. Amsterdam : John Benjamins. 123–172.
  • Vogeleer, S. (1994). L’accès perceptuel à l’information : à propos des expressions un homme arrive / on voit arriver un homme. Langue française, 102, 69-83.

Haut page

The French verb manquer : a negative existential predicate ?
Danièle Van de Velde
University of Lille 3, CNRS UMR 8163 « Savoirs, Textes, Langage » (STL), France

In this paper I try to describe the two existential uses of the french verb manquer, a verb that seems to combine non existence with modality, meaning that something does not exists that ought to exist.
The first use exhibits a form of locative alternation where the theme may be either subject (l’eau manque aux réfugiés) or prepositional complement (les réfugiés manquent d’eau). The other use has a typical existential, impersonal, construction where the theme can occupy either the subject or the post verbal position (il manque deux boutons à ma veste / deux boutons manquent à ma veste).
The similarities between manquer and the so-called « abundance verbs » suggest that both types incorporate a verbal quantifier : in the case of manquer1 this incorporated quantifier has to be interpreted as meaning an « insufficient » quantity, while for manquer2 its meaning would be a quantity « smaller than ».
At this point, our analysis makes clear that manquer in both constructions always presupposes the background of existence and that the scope of the negation it contains is only the (relative) quantity of non existing entities : in fact, manquer works in a very similar way to y avoir. Finally, our conclusion is that inexistence, like existence cannot be posited but on a background of already existing entities.

Manquer : un prédicat existentiel négatif ?

La communication porte uniquement sur les emplois existentiels de manquer, et elle en distingue deux.
Le premier (manquer1) est un emploi syntaxiquement « personnel » à alternance, dans lequel le verbe a deux arguments qui peuvent échanger leurs positions (l’eau manque aux réfugiés / les réfugiés manquent d’eau).
Le second (manquer2) est un emploi fondamentalement impersonnel, où l’argument interne peut monter en position de sujet (Il manque deux boutons à ma veste / Deux boutons manquent à ma veste).
La parenté de manquer1 avec les verbes « d’abondance » nous conduit à faire l’hypothèse que ce verbe incorpore un quantificateur, exprimable par l’adverbial « en quantité insuffisante ». Cette hypothèse peut être étendue à manquer2 mais avec la variante « en quantité inférieure à x, de y ».
Dans tous les cas, manquer est plus proche de y avoir (avec lequel il partage la propriété de monotonie croissante), que de ne pas y avoir. Pour ce qui est du négatif incorporé dans la signification de manquer, il apparaît bien dans les adjectifs insuffisant et inférieur utilisés pour la décrire, mais ce négatif présuppose toujours un fond positif.
Nous concluons donc que comme l’existence, l’inexistence ne se pose pas dans l’absolu, mais présuppose toujours déjà l’existence.


haut de page