Navigation – Plan du site

Implicit stage topics

A case study in French
Karen Lahousse

Résumés

Il a souvent été proposé que les éléments spatio-temporels en position initiale de phrase spécifient le cadre de l’événement dénoté par la proposition et ont une interprétation thématique ou topicale. Alors que les topiques spatio-temporels explicites ont souvent été étudiés, Erteschik-Schir (1997, 1999) propose l’idée que les topiques spatio-temporels, ou topiques scéniques (stage topics) peuvent aussi être implicites.
Dans cet article, nous offrons des arguments en faveur de la notion de topique scénique implicite. Nous montrons qu’un certain nombre de cas d’inversion nominale en français, une configuration syntaxique qui est favorisée par la présence d’un topique scénique explicite, s’expliquent par la présence d’un topique scénique implicite. Le fait que les topiques scéniques implicites interagissent avec la structure syntaxique de la même façon que les topiques scéniques explicites constitue un argument empirique en faveur de leur existence.

Haut de page

Texte intégral

1. Introduction

1It has often been proposed that sentence-initial spatio-temporal elements specify the frame in which the whole proposition takes place and are topical/thematic. Whereas considerable attention has been paid to explicit spatio-temporal topics, Erteschik-Shir (1997, 1999) argues that spatio-temporal topics, or stage topics, can also be implicit. In this article, I concentrate on this notion of implicit stage topic and show that the existence of implicit stage topics and the specific constraints they are subject to are confirmed by the distribution of nominal inversion in French.

2The structure of the article is as follows. I first give a brief overview of the literature on spatio-temporal topics (section 2.1.), and present and refine Erteschik-Shir’s hypothesis according to which the presence of an implicit stage topic is only justified if its content is deictically or ‘discoursally’ specified, i.e. recoverable from the speech context or the narrative context (section 2.2.). In the second part of this article, I provide empirical evidence with respect to the distribution of nominal inversion in French which shows that, in the same way as overt stage topics (section 3), covert stage topics also (section 4) interact with linguistic structure.

2. The notion of ‘stage topic’

2.1. Frame-setting and overt stage topics

3Many researchers observe that certain types of adverbials are interpreted differently in sentence-initial position and in sentence-internal or sentence-final position (cf., among others, Jackendoff, 1972; Guimier, 1996; Laenzlinger, 1996: 50; Berthonneau, 1987; Le Querler, 1993; Grosu, 1975; Cinque, 1990; Chetrit, 1976; Charolles, 2003; Melis, 1983).

4For example, Melis (1983: 186) states that example [1a] has the interpretation [2a], while [1b] has the interpretation in [2b]:

[1a]Au XVIIe siècle, la condition paysanne était dure.
‘In the 17th century, the peasants’ fate was hard.’

[1b]La condition paysanne était dure au XVIIe siècle.
‘The peasants’ lot was hard in the 17th century.’
(Melis, 1983 : 186)

[2a]‘soit le XVIIe siècle; pour ce siècle, il est vrai que la condition paysanne était dure’
‘given the 17th century, for this century, it is true that the peasants lot was hard’
(Melis 1983:186)

[2b]‘soit la condition paysanne; à propos de celle-ci, il est vrai qu’elle était dure au XVIIe siècle
‘given the peasants’ lot; about this, it is true that it was hard in the 17th century’

5Although formulations differ, the main idea in many of the above-mentioned works is that these elements in sentence-internal or sentence-final position modify the state of affairs expressed by the verb, whereas in sentence-initial position, they specify the frame in which the whole proposition takes place.

6In addition to the definition of topics as ‘what the sentence is about’ (cf. among others, Kuno, 1972; Dik, 1989; Reinhart, 1981; Gundel, 1989; Lambrecht, 1994), topics are sometimes defined as the frame in which the event denoted by the proposition takes place. This is explicit in Chafe’s (1976 : 50) definition of what he calls ‘Chinese style topics’: “The topic sets a spatial, temporal, or individual framework within which the main predication holds (the frame within which the sentence holds)”. According to Chafe (1976 : 50), these topics are not what the sentence is about, but they “limit the applicability of the main predication to a certain restricted domain”. Chafe’s ‘Chinese style topics’ thus have a ‘frame-setting’ function, as defined by Jacobs (2001 : 656): “In (X,Y), X is the frame for Y if X specifies a domain of (possible) reality to which the proposition expressed by Y is restricted”. The same idea is explicit in Charolles’ (2003) characterization of ‘framing adverbials’ (adverbiaux cadratifs).

  • 1  Besides the spatio-temporal frame-setting, Chafe (1976) also distinguishes an ‘individual framewor (...)

7The most typical examples of frame-setting elements are spatio-temporal frame-setting elements1. According to Chafe (1976: 50-51), the temporal framework is expressed in English by the use of sentence-initial temporal adverbials in sentences like Tuesday I went to the dentist, and by the sentence-initial locative PP in sentences like In Dwinelle Hall people are always getting lost. Spatio-temporal frame-setting elements have often been argued to function as the ‘topic’ of the clause, (cf. Chafe, 1976; Nikolaeva, 2001; Le Querler, 1993: 177; Jacobs, 2001; Combettes, 1996: 94-95; Charolles, 2003; Reinhart, 1981; Gundel, 1989; Sgall et al., 1986: 202; Erteschik-Shir, 1997, 1999). In Erteschik-Shir (1997, 1999), spatio-temporal frame-setting elements are called ‘stage topics’, which are defined in the following way:

“A stage topic (sTOPt) defines the spatio-temporal parameters of the utterance. Stage topics may be overt (‘this afternoon’, ‘on Park Avenue’), or discoursally implied (the here-and-now)” (Erteschik-Shir, 1999: 124). “The term “stage” here (…) refer[s] to the Time/Place at which the event expressed by the sentence takes place.” (Erteschik-Shir, 1997: 26-27)

  • 2  In Erteschik-Shir (1997/1999) and Vallduví (1992/1994), following proposals made by Reinhart (1981 (...)
  • 3  In Lambrecht’s (1994) framework, this amounts to saying that, just as regular topics, stage topics (...)

8In the file-card model for information structure2, stage topics, just as regular topics, are represented by a card. Cards represent existing discourse referents, i.e. referents that belong to the ‘common ground’, which is defined as “that part of the information state which the hearer has in common with the speaker” (Erteschik-Shir, 1997: 6). Consequently, the referent of a stage topic must be known to both the speaker and the hearer3; in order to count as a stage topic, a location must be identifiable by both the speaker and the hearer.

9As Erteschik-Shir (1997: 27) argues, a stage topic is typically a sentence-initial spatio-temporal phrase. However, a sentence-internal or sentence-final spatio-temporal phrase can also function as the stage topic of the clause, provided that it has been previously introduced into the discourse. Hence, although this is not the default interpretation, a spatio-temporal phrase that is not in initial position can be topical if it is mentioned or implied in the previous context: “anteposition is a possible but not a necessary trace of the thematisation of a constituent” (Berthonneau, 1987: 67). By contrast, a sentence-internal or sentence-final spatio-temporal phrase that is not mentioned or implied in the previous context will be interpreted as (part of) the new information, i.e. the focal or rhematic part of the sentence.

10For instance, [3a] can be the answer to both [3b] and [3c]:

[3a] La Première Guerre mondiale a éclaté en 1914.
‘World War 1 broke out in 1914.’

[3b] Quand est-ce que la Première Guerre mondiale a éclaté?
‘When did World War 1 break out?’

[3c] Que s’est-il passé en 1914?
‘What happened in 1914?’

11Hence, [3a] has two different interpretations, which also correspond to two different intonational patterns. When [3a] functions as the answer to a question as [3b], the temporal phrase en 1914 ‘in 1914’ is considered as the ‘focus’ of the clause, since it specifies the value of a missing element in the preceding context. In such contexts, the main stress of the clause naturally falls on the temporal phrase. In contrast, when [3a] functions as the answer to [3c], the temporal phrase will be interpreted as the (stage) topic of the clause, since it really indicates the spatio-temporal stage at which the event expressed by the sentence takes place. With this interpretation, the main stress of the clause falls on the past participle, and the temporal phrase is pronounced with an ‘afterthought’-stress. Thus, a sentence like [3a], where the temporal phrase is not in sentence-initial position, has two potential interpretations, but the temporal phrase can only be interpreted as a stage topic if it is mentioned (or implied) in the discourse context.

12Clauses with a spatio-temporal phrase in sentence-initial position, like [4a], are not ambiguous in this sense (cf. also Charolles, 2003 for similar contrasts with other types of adverbials): whereas [4a] is appropriate as an answer to [4c], it is not a legitimate answer to [3b], repeated here as [4b]:

[4a] En 1914, la Première Guerre mondiale a éclaté.
‘In 1914, World War 1 broke out.’

[4b] # Quand est-ce que la Première Guerre mondiale a éclaté?
‘When did World War 1 break out?’

[4c] Que s’est-il passé en 1914?
‘What happened in 1914?’

2.2. Covert stage topics

13In contrast with the other authors mentioned above, Erteschik-Shir (1997, 1999) explicitly states that stage topics may not only be ‘overt’, i.e. explicitly realized in the sentence, as in all the above-mentioned examples, but also ‘covert’ or ‘implicit’, as in the following examples:

[5a] STOPtit is raining
‘It’s raining’
(Erteschik-Shir 1997: 27)

[5b]Nous sommes arrivés en Espagne. sTOPtIl pleuvait.
‘We arrived in Spain. It was raining.’

  • 4  In Erteschik-Shir’s model, the here-and-now of the discourse context is represented by a card that (...)

14According to Erteschik-Shir, stage topics are only covert when their content is deictically or ‘discoursally’ specified. In a conversation, an utterance like [5a] has an implicit deictic stage topic, i.e. a stage topic that refers to the here-and-now of the speech context4. In a narrative, the covert stage topic is licensed when its content can be recovered from the previous discourse context. The spatio-temporal parameters of the second clause in [5b] are determined by the previous context: it was raining in Spain and at the moment of the speaker’s arrival. Covert stage topics are thus licensed if their interpretation is recoverable either from the deictic discourse context either from the preceding narrative context.

  • 5  This implicit stage topic corresponds, as Erteschik-Shir (1997) argues, with Kratzer’s (1989) inte (...)
  • 6  From a theoretical point of view, the notion of covert stage topic as a formalization of the spati (...)

15In Erteschik-Shir (1997)’s model of information structure, the existence of covert stage topics is motivated in the following way. In line with Strawson (1964) and Reinhart (1981), the author considers the topic of a clause as the pivot for the assessment of this clause. The assessment of a clause consists in examining the truth of the predicate with respect to the topic. An immediate consequence is that “every sentence must have a topic since if topics are the pivots for assessment it is crucial that every sentence have one” (Erteschik-Shir, 1997: 10). As for clauses that do not have an explicit topic, the author argues they have an implicit topic provided by the spatio-temporal parameters of the utterance, i.e. an implicit stage topic5. In this respect, a sentence like [5a] above is assessed by examining the truth of ‘if it is raining’ with respect to the implicit stage topic, i.e. the here-and-now of the utterance. In the same way, Erteschik-Shir (1997: 27) argues that all out-of-the-blue sentences are assessed with respect to an implicit deictic stage topic. It follows from this that Erteschik-Shir’s (1997) hypothesis of the existence of covert stage topics is particularly motivated on a theory-internal basis6.

16I would like to refine Erteschik-Shir’s (1997, 1999) conception of covert stage topics in three respects.

17First, although the author does not explicitly tackle the issue, the existence of covert temporal stage topics seems to be easier to justify than that of covert locative stage topics. Saying that a clause contains a covert stage topic amounts to saying that this clause is linked to the previous context, and it is easier to assume temporal linkage between clauses in a narrative that do not contain any explicit temporal or spatial phrase, than spatial linkage. It is indeed commonly accepted that clauses are temporally linked to the preceding context. For instance, de Swart (1993: 248) argues that “an independent sentence is anaphoric in the sense that it picks up its reference point from the preceding context”. In section 4.2.3., I will provide further evidence against the existence of covert locative stage topics.

  • 7  Some authors even claim that verbal tenses anaphorically refer to the temporal context in a way si (...)

18Second, to the extent that the notion of topic is a linguistic concept, topics must be formally, i.e. lexicogrammatically, evoked in the sentence (cf. Lambrecht, 1994). In other words, I argue that only those covert topics whose presence is, in some way or another, formally indicated in the clause are linguistically relevant. Covert temporal stage topics have a formal expression in verbal morphology (provided the verb carries a tense-indicating morpheme) and it is generally accepted that verbal tenses refer to the temporal context of the clause7. In section 4.1., I will provide further evidence showing that adverbs and connectives, besides verb morphology, indicate the presence of a covert stage topic.

19Thirdly, it is clear that the hypothesis of the existence of covert stage topics would be stronger if it were supported by empirical evidence. More particularly, the existence of covert stage topics should be confirmed by empirical data showing that they interact with linguistic structure in the same way overt stage topics do. This is the purpose of the remainder of this paper, where I will show that nominal inversion in French is triggered by the presence of overt (section 3) and covert stage topics (section 4).

3. Overt stage topics and nominal inversion in French8

3.1. Data and generalization

20It is well known that nominal inversion in French frequently occurs after sentence-initial locative [6a] or temporal [6b] PPs, an instance of inversion that is also called ‘locative inversion’ (cf. Blinkenberg, 1928; Le Bidois, 1952; Jonare, 1976; Togeby, 1982-1985; Marandin, 2001; Bonami et al., 1999; Korzen, 1983/1985; Tasmowski, Willems, 1987; Fournier, 1997 and many others).

[6a]Dans la cour régnait l’animation habituelle.
lit. In the courtyard reigned the usual hustle and bustle.
‘In the courtyard, the usual hustle and bustle reigned.’
(Brincourt)

[6b]En septembre apparaissent les grosses araignées. Elles tissent leurs toiles scintillantes et polygonales d'une branche à une autre.
‘In September come the fat spiders. They spin their glittering polygonal webs from branch to branch.’
(Simon)

21It has often been claimed that the possibility for locative inversion to occur is due to the topical nature of the sentence-initial elements (cf. Tasmowski, Willems, 1987; Fournier, 1997). However, this hypothesis makes false predictions insofar as this assumption does not explain why inversion is not possible after domain adverbs and clitic left-dislocated elements, which are also considered as topical elements:

22Sentence-initial domain adverb

  • 9  Henceforth, I use VS and SV in the examples to indicate verb-subject and subject-verb word order, (...)

[7a] VS9: * Logiquement devraient s’arranger les choses.
lit. Logically would have to be arranged the things.
SV: Logiquement, les choses devraient s’arranger.
lit. Logically, the things would have to be arranged.
‘Logically, things should get better.’

[7b] VS: * Légalement peuvent être organisées des élections.
lit. Legally can be organized elections.
SV: Légalement, des élections peuvent être organisées.
lit. Legally, elections can be organized.
‘Legally, elections can be organized.’

Sentence-initial clitic left-dislocated element
[8a]VS: * Ce pays, l’envahit l’armée.
lit. This country, it invades the army.
SV: Ce pays, l’armée l’envahit.
lit. This country, the army it invades.
‘This country, the army is invading it.’

  • 10  Based on the attested example (i):
    (i) (…) la morne campagne du nord (…), dont les quais semblent p (...)

[8b]VS: * Les quais, les déserte la foule.
lit. The plaforms, them leaves the crowd.
SV: Les quais, la foule les déserte.10lit. The platforms, the crowd them leaves.
‘The platforms, the crowd is leaving them.’

23Clitic left-dislocated elements, domain adverbs and stage topics all have a thematic or topical interpretation, but stage topics differ from the other two classes of elements in that they refer to the spatio-temporal frame of the clause. Hence, rather than stating that nominal inversion is accepted in clauses introduced by a topical element, the interim descriptive generalization of the distribution of nominal inversion may be formulated as follows:

[9]Hypothesis I
Nominal inversion is allowed after an overt stage topic.

3.2. Further confirmation

24According to Erteschik-Shir (1997, 1999), there are no constraints on the specific syntactic expression of a stage topic; it is not restricted to PPs and it may be a quantified phrase. Hence, for generalization [9] to hold, nominal inversion should be allowed after quantified PPs and after any phrase that indicates the scene at which the event takes place, irrespective of its syntactic form. This prediction is borne out, as shown by the following attested examples, where nominal inversion occurs after a quantified locative phrase [10a], after a participle phrase [10b] and after an embedded temporal clause [10c]:

[10a]Chaque mercredi paraissait un hebdomadaire radiophonique très en vue, Radio-Hebdo, qui publiait, outre les programmes de la semaine à venir, tout un supplément photographique (…)
lit. Every Wednesday came out a famous radio magazine…
‘Every Wednesday a famous radio magazine came out, Radio-Hebdo.Besides the next week’s programs, it would have a big photo supplement (…)’ (Tournier)

[10b]Cela exécuté, venait l'heure de liberté totale, sans contrôle, vouée à la fantaisie sentimentale, que dans mon for intérieur je disais amoureuse.
‘This done, came a time of unchecked and utter freedom given over to the sentimental fantasizing that in my heart of hearts I called love.’ (Simonin)

[10c]Quand on bassine vient l'allergie!
lit. When one is stressed comes the allergy.
‘When you’re stressed, the allergy strikes!’
(Blier)

25Interestingly, the hypothesis [9] also explains the following observation made by (Tasmowski, Willems, 1987):

[11]Sur ce tableau se promènent trois mouches.
lit. On this painting are walking three flies.
(Tasmowski, Willems, 1987: 181)

[11a]  # ‘in this painting three flies are walking’
# the painting = representation

[11b] ‘on this painting three flies are walking’
the painting = object

26According to the authors, the clause with nominal inversion in [11] cannot have the interpretation glossed in [11a], with the painting understood as a representation of three flies walking somewhere. In this interpretation, the painting itself is not the spatial scene in which the walking event takes place, and, hence, is not the stage topic of the walking event. In contrast, ce tableau ‘this painting’ must be interpreted as a concrete object having a surface, such as the canvas itself, a frame or a protective glass pane where three flies are walking on [11a]. In this interpretation, the painting’s surface is the spatial scene the walking event takes place on, and is thus interpreted as the stage topic of the walking event. The fact that in the clause with nominal inversion in [11] the sentence-initial locative phrase can only be understood as a stage topic, confirms our initial hypothesis in [9], according to which nominal inversion is explained by the presence of an overt stage topic.

27Our hypothesis [9] also directly accounts for the fact that nominal inversion is not allowed after a directional PP, cf. (Marandin, 1997) and (Kerleroux, Marandin, 2001: 298), as the following examples show:

[12a] * Vers la ville s’étend un marais nauséabond.
lit. Towards the town stretches a putrid swamp.

  • 11  This example is acceptable when the man is not in motion, and, hence, when the verb se balancer is (...)

[12b] * Contre le mur se balançait un homme11.
lit. Against the wall swayed a man.

[12c] * Hors du fleuve sautèrent des poissons de lune.
lit. Out of the river jumped sunfish.

[12d] * Dans le ravin se jetaient les soldats acculés par l’ennemi.
lit. Into the ravine jumped the soldiers cornered by the enemy.

[12e] * Dans la pièce entrait Jean.
lit. Into the room entered John.
(Marandin, 1997)

28Directional PPs denote the endpoint of the event, but not the scene at which the event takes place before it reaches the endpoint; they do not indicate the stage topic of the ongoing event. As predicted, they do not allow nominal inversion to their right.

29In conclusion, all the data presented in this section confirm the generalization according to which nominal inversion is licensed by an overt stage topic. Let us now examine whether this hypothesis can be extended to other cases of inversion.

4. Covert stage topics and nominal inversion

4.1. Sentence-initial adverbs or connectives

30Besides locative and temporal PPs, nominal inversion is also allowed after several types of adverbs and connectives. For instance, inversion is allowed after locative adverbs, whether they are deictic or anaphoric [13], e.g. ici ‘here’, ‘there’, de là ‘from there’, derrière ‘behind’, un peu plus loin ‘a bit further’, dehors ‘outside’, partout ‘everywhere’, ailleurs ‘elsewhere’ and after temporal adverbs or connectives, whether deictic or anaphoric [14], such as alors ‘then’, après ‘after’, enfin ‘finally’, puis ‘ then’, ensuite ‘then’, aussitôt ‘immediately’, tout d’abord ‘first of all’, parfois ‘sometimes’, bientôt ‘shortly’, soudain ‘suddenly’, de nouveau ‘again’, tout à coup ‘all of a sudden’, déjà ‘already’, maintenant ‘now’, aujourd’hui ‘today’:

[13a] [Je m’inscrivis au club de tennis de V.] se nouaient des idylles avec…lit. I joined V.’s tennis club. There were tied idylls with…
‘I joined V.’s tennis club. There idylls were tied with…’
(Hébrard and Velle)

[13b]Odilon se leva soi-disant pour allumer la terrasse.
Dehors tombait une pluie venteuse (…)
lit. Outside fell a windy rain.
‘Odilon got up, supposedly to turn on the terrace lights.
Outside blew the wind and the rain.’
(Queffelec)

[14a] L’heure était venue de se souvenir. Alors commença pour les miens la traversée de la douleur comme celle d’une épaisse forêt (…).
‘The time had come to remember. Then began for my loved ones the arduous passage through the pain like through a dense forest (…).’
(Hébrard and Velle)

[14b]Je trottinais par-derrière, (…). Et soudain surgirent six hommes noirs.
‘I was trotting along behind, (…). And suddenly out came six black men.’
(de Grèce)

  • 12  My translation. The original text reads: “une localisation temporelle ayant déjà fait l’objet d’un (...)
  • 13  I assume that deictic adverbs, just like anaphoric adverbs, are dependent on the presence of a cov (...)

31The sentence-initial adverbs in [13] and [14] above are not stage topics, to the extent that they do not refer directly to a temporal or spatial location at which the event expressed by the sentence takes place. However, these adverbs refer to a temporal or a spatial location in an indirect way: when used in a narrative, deictic adverbs such as ici ‘here’, ‘there’, maintenant ‘now’, and aujourd’hui ‘today’ refer to the time and place of the previous narrative context, and the other adverbs and connectives mentioned above are anaphoric in the sense that they refer to the spatio-temporal scene of the previous discourse. As Borillo (1998) argues, these adverbs and connectives can only be interpreted with respect to a “temporal location that has already been the object of a calculus in the interpretation of the preceding discourse”12; they connect the clause in which they occur to the temporal and/or spatial location of the preceding context. Thus, rather than being a stage topic in themselves, these adverbs, precisely because of their anaphoric nature, refer to a (covert or overt) stage topic or co-occur with a (covert or overt) stage topic.13 It is then predicted, on the one hand, that these adverbs are only licensed when the clause contains an overt or covert stage topic, and, on the other hand, that it is this stage topic that favors nominal inversion.

32As for the first prediction, the examples [13] and [14] illustrate the fact that anaphoric adverbs and other connectives occur in clauses in which the content of a stage topic is recoverable from the preceding context. By contrast, as [15] shows, anaphoric adverbs do not occur in clauses that are uttered out of the blue and, hence, do not contain a covert stage topic:

[15]Pourquoi as-tu l’air si troublée?
‘Why are you looking so confused?’

[15a] * Soudain, le téléphone a sonné.
‘Suddenly, the telephone rang.’

[15b] * Alors, le téléphone a sonné.
‘Then, the telephone rang.’

[15c]Le téléphone a sonné.
‘The telephone rang.’

[15a’]J’étais en train de dormir et soudain, le téléphone a sonné.
‘I was sleeping and suddenly, the telephone rang.’

[15b’]J’étais en train de dormir et alors, le téléphone a sonné.
‘I was sleeping and then, the telephone rang.’

  • 14  For some speakers, nominal inversion is awkward in the contexts (16a’) and (16b’):
    (i) Pourquoi as- (...)

33The clauses in [15a] and [15b] are not appropriate answers to the question in [15]. This is due to the presence of the adverbs soudain ‘suddenly’ and alors ‘then’, since the corresponding example without these adverbs is acceptable [15c]. By contrast, the adverbs soudain ‘suddenly’ and alors ‘then’ are allowed in the answers [15a’] and [15b’], where the first clause acts as the temporal location, and, hence, the stage topic, of the second clause14. In short, contrasts such as those in [15] confirm that the presence of a spatio-temporal adverb or connective is correlated with the presence of either an overt stage topic or a covert stage topic whose content is recoverable from the previous discourse or from the speech context.

34The second prediction, according to which inversion occurring after an initial adverb [13 and 14] is not licensed by the deictic or anaphoric adverb itself, but rather by an implicit stage topic, is confirmed by the contrasts in [16] and [17]:

[16a] V Adv S
[Un silence se fit.] Jaillit alors une clameur… .
lit. A silence fell. Arose then an outcry…
‘A silence fell. Then an outcry arose … .’
(Kerleroux, Marandin, 2001: 299)

[16b]* V PP S
* Jaillit à 4 heures une clameur…
lit. Arose at 4 o’clock an outcry…

[17a] V Adv S
Coexistent ici les données anciennes.
lit. Coexist here the old data.
‘Here, the old data coexist.’
(Le Querler 1993:178)

[17b] * V PP S
* Coexistent dans la mémoire les données anciennes.
lit. Coexist in the memory the old data.

  • 15  Cf. Le Bidois (1952:30-31) and Jonare (1976:38) for more examples of this type.

35As these examples illustrate, nominal inversion is allowed when an anaphoric or deictic adverb is in sentence-internal position [16a and 17a]15, but it is not allowed when a PP denoting a location is in sentence-internal position [16b and 17b]. These contrasts can be explained if, as I showed above, adverbs such as alors ‘then’ and ici ‘here’ are only allowed in a context that contains a stage topic, whether covert or overt. It is this (covert) stage topic which licenses inversion in [16a] and [17a], just as it does in [13] and [14]. On the contrary, a PP only denotes a stage topic when it is in sentence-initial position or when it is mentioned in the previous context (cf. supra). Hence, the PP in [16b] and [17b] is not a stage topic itself, neither does it refer to a covert stage topic, and, as predicted, inversion is not allowed. This contrast thus confirms that, in clauses that contain a spatio-temporal adverb, inversion is licensed by the presence of the covert stage topic whose presence is lexicogrammatically indicated by the spatio-temporal adverb.

  • 16  The hypothesis in (19) also allows a unified account of the two classes of inversion distinguished (...)

36The two preceding sections have thus shown that nominal inversion in French is licensed by the presence of a stage topic, be it an overt stage topic (typically represented by a sentence-initial spatio-temporal phrase, irrespective of its syntactic realization) or a covert stage topic, whose content is recoverable from previous discourse and whose presence may be indicated in the clause by an adverb. Hence, these data allow for the following hypothesis, which refines that given in [9]16:

[18]   Hypothesis II:

[18a] Nominal inversion is allowed after an overt stage topic

[18b] Nominal inversion is allowed in clauses introduced by a covert stage topic.

37In the following section, I will show that absolute inversion, i.e. nominal inversion in contexts where the verb is the first element of the clause, constitutes another piece of empirical data in favor of this hypothesis and, hence, in favor of the existence of covert stage topics.

4.2. Absolute inversion

  • 17  In this article I do not take into consideration instances of absolute inversion due to the restri (...)

38Let us now consider absolute inversion, i.e. inversion where the verb is the first element of the clause17:

[19]Arrive le général Vint le printemps Naît un conflit.
arrives the generalcame the spring arises a conflict
(Tasmowski, Willems, 1987: 182)

39It is clear that such instances of inversion cannot be accounted for by the presence of an overt stage topic, since no overt spatio-temporal element precedes the verb in these examples. However, could such cases be accounted for by the presence of a covert stage topic? One possible explanation may be found in (Tasmowski, Willems, 1987: 183) who claim that absolute inversion cannot appear at the beginning of a text, and that it corresponds to locative inversion (cf. supra) introduced by an implicit adverbial ‘at this point in the narrative’. Thus, given that absolute inversion cannot appear at the beginning of a text, it needs to be linked up with a previous discourse context, where the content of a covert stage topic ‘at this point in the narrative’ can be recovered. In other words, the distribution of absolute inversion is explained by the presence of a covert stage topic.

  • 18  The distribution of absolute inversion has also been described to some extent by Blinkenberg (1928 (...)

40Although several authors mention examples of absolute inversion, only Le Bidois (1952) proposes a detailed classification of the contexts it appears in18. The purpose of the following sections is then to test whether the presence of a covert stage topic can be argued for in the contexts mentioned by Le Bidois (section 4.2.1.) and to present the results of my own corpus-based research, which show that this is indeed the case (section 4.2.2.).

4.2.1. Le Bidois’ (1952) examples

41Le Bidois (1952) observes that absolute inversion is allowed in scenic indications, which are used in plays and scripts to indicate stage directions, as in the following examples:

[20]Entre Stolberg. Entre l’Archiduc. Entrent le roi et Piéchèvre.
lit. Enter Stolberg. Enter the Archduke. Enter the King and Piéchèvre.
(Romains, cited in (Le Bidois, 1952: 23))

42According to Le Bidois, examples of this type are almost fossilized, and the verb is always in the present indicative. These cases of absolute inversion can be easily accounted for by the notion of stage topic: as Le Bidois (1952: 23) mentions, such examples are equivalent to Attention! (Il) entre (quelqu’un)!‘Attention! There comes somebody!’. Hence, these expressions signify that somebody arrives at the present scene and at the present moment, i.e. there is the perception of an entrance. This also explains why the verb morphology is restricted to the present  indicative. In these examples, thus, the content of a covert stage topic is deictically recoverable, and absolute inversion is licensed by this covert stage topic.

  • 19  According to Jonare (1976:35), rester ‘to stay’ is the verb which occurs most frequently in absolu (...)

43As the following examples illustrate, absolute inversion is also possible with the verbs suivre ‘to follow’ and rester ‘to stay’19:

[21a]Les hommes, le verbe haut, discutaient de l’affaire.
Suivaient les coiffes, les robes noires, des jeunes filles par bandes.lit. The men, in high and mighty tones, were discussing the case.
Followed the headdresses, the black clothes, of the young girls in groups.
‘The men were discussing the case in high and mighty tones.
After them came women in headdresses, in black, young girls in groups.’
(de Chateaubriand, cited in (Le Bidois, 1952: 21))

[21b]Toutes les difficultés ne s’étaient pas trouvées aplanies.
Restait la grave question des places…lit. All the difficulties were not yet smoothed over.
Remained the serious matter of places…
‘All the difficulties were not yet smoothed over.
There remained the serious matter of places…’
(Benoit, cited in (Le Bidois, 1952: 24))

44Le Bidois (1952: 21) holds that absolute inversion is allowed with suivre ‘to follow’ because this verb marks a succession in time or space, and is equivalent to adverbs like puis ‘then’ and ensuite ‘then’. Moreover, sentences with nominal inversion examples like [21a] are not acceptable at the beginning of a text; for such constructions to be licensed, the preceding context must specify the location with respect to which the postverbal subject is situated. In other words, the verb suivre ‘to follow’ indicates that the referent of the subject follows the event denoted by the first clause; the first clause specifies the content of the covert stage topic of the second clause. For example, in [21a], the young girls follow the men, and, hence, are situated with respect to the men.

45Concerning absolute inversion with rester ‘to remain’ as in [21b], Le Bidois (1952: 24-25) explains such cases by the ‘weak semantic value’ of this verb, which gives it the possibility to function as the psychological subject, i.e. the “group of ideas which is first present in the mind of the speaker”, or “the thing one speaks about” (Le Bidois, 1952: 346). In more recent terms, Le Bidois thus considers the verb rester in [21b] as the topic of the sentence. However, because a verb alone, and especially a semantically weak one like rester ‘to remain’ does not denote an entity or location, it is difficult to consider it a topic. In contrast, it can be argued that sentences like [21b] contain a covert stage topic: such clauses cannot occur discourse-initially, and, moreover, it is always with respect to a given situation that something remains. Thus, in examples of absolute inversion with rester ‘to remain’, the first clause determines the (abstract) location of the event denoted in the second clause, and, hence, determines the content of the covert stage topic of the second clause.

46As Le Bidois (1952: 20-23) observes, absolute inversion is not restricted to written stage constructions nor to the verbs rester and suivre, but also occurs in other contexts. Such contexts are: clauses with postverbal temporal subjects [22a], with postverbal subjects denoting an “important event” [22b] or a person [22c]. The verbs in these clauses are verbs indicating the occurrence of an event or a sequence of events, such as venir ‘to come’ in [22a] and arriver ‘to arrive’ in [22b], or verbs of movement, such as venir ‘to come’ in [22c].

[22a]Vint l’ère des grenades.
lit. Came the era of grenades.
‘Then came the time of the hand-grenade.’
(Romains, cited in (Le Bidois, 1952: 20))

[22b]Arrive la lettre d’un ami.
lit. Arrives the letter from a friend.
‘The letter from a friend arrives.’
(Pourrat, cited in (Le Bidois, 1952: 20))

[22c]Vint une servante qui prononça quelques mots.
lit. Came a maid who pronounced some words.
‘A maid came by and said something.’
(Bedel, cited in (Le Bidois, 1952: 20))

47In order to test the hypothesis that inversion in the heterogeneous set of contexts mentioned under [22] is also licensed by a covert stage topic, it is necessary to examine the discourse context they appear in. This is the purpose of the following section.

4.2.2. The discourse context of absolute inversion: corpus research

  • 20  For technical reasons inherent to the search engine of Frantext, it is impossible to search for ex (...)

48Most authors do not supply the context of the examples they quote, in spite of its extreme importance for the distribution of inversion. In order to examine absolute inversion in context, I ran a corpus test in Frantext, an on-line categorized corpus of French texts assembled by the Institut National de la Langue Française, on all the novels and essays of the database published between 1970 and 2000. Results showed 46 examples of absolute inversion occurring with the verbs venir ‘to come’, arriver ‘to arrive/happen’, surgir ‘ to appear suddenly’and (ap)paraître ‘to appear’20. These examples can be classified into several groups.

49First of all, in 26 examples – more than 50 % of all occurrences –, there is a temporal postverbal subject, as in [23]:

[23a]L'obscurité était si épaisse, le vacarme des flots si puissant que l'on n'y voyait pas à un pied et que toute voix, même la plus déchirante, était aussitôt étouffée. Vint le moment où la taraque, couchée sur le côté, hésita à sombrer.
lit. Came the instant when the vessel, lying on its side,hung on the brink of sinking.
‘The darkness was so thick, the thunder of the waves so deafening that a foot ahead nothing was visible and all human cries, even the most piercing, were lost. Then came the instant when the vessel, lying on its side, hung on the brink of sinking.’
(Lanzmann)

[23b]J'adorais livrer mes textes en ces petits matins grisâtres et pluvieux. Achard corrigeait, annotait, expédiait à la compo. Et arrivait l'instant magique: il m'ouvrait le placard d'Ali Baba.
lit. And arrived the magic moment:…
‘I loved delivering my texts on these grey and rainy dawns. Achard would correct, annotate and send them to the typesetters. Then came the magical moment: he opened Ali Baba’s cupboard for me.‘
(Manoeuvre)

[23c](…) parfois les larmes aux yeux quand il faisait beau, et quand, impitoyablement enchaîné à son instrument, il entendait les cris de ses camarades qui s'amusaient en plein air.
Vint sa seizième année.
Son talent s'épanouissait avec une plénitude incomparable.
lit. Came his sixteenth year.
‘(…) sometimes tearful when the weather was fine, and he, mercilessly bound to his instrument, would hear his friends shouting and having fun outdoors. Then came the year he became sixteen. His talent was blooming in all its unique richness.’
(Tournier)

  • 21  Why is it that examples of absolute inversion with a temporal subject are so frequent? Interesting (...)

50In these examples, the clause marks a succession with respect to the previous context. In other words, the clause with absolute inversion contains a covert stage topic whose content is recoverable from the preceding context, and the postverbal subject identifies a new reference time, which is situated with respect to this covert stage topic21.

51In seven examples, the event denoted by the inversion structure is the immediate consequence of the event denoted by the preceding clause/context:

[24a]Elle sonne. Arrive une infirmière: “Ah ! Mais madame, ce n'est pas l'heure.”
lit. She rings. Arrives a nurse:…
‘She rings. A nurse arrives: “Oh! But madam, it’s not time yet.”’
(Dolto)

[24b]Soudain vacarme. Télémaque a tiré la tenture
Apparaît une volière extravagante bourrée d'oiseaux somptueux, coucoudetus jumanovillensis, pygalectron vulgaris, nyctomelos caliphi, tetropteryx velatus, (…)lit. A sudden loud racket. Telemachus pulled the drape.
Appears an extraordinary aviary, crammed with magnificent birds:…
‘A sudden loud racket. Telemachus pulled the drape.
An extraordinary aviary appeared, crammed with magnificent birds: coucoudetus jumanovillensis, pygalectron vulgaris, nyctomelos caliphi, tetropteryx velatus, (…)’
(Bory)

52In these examples, the event in the first clause denotes the immediate cause of the event in the second clause, and, hence, determines the content of the stage topic of the second clause. Similarly, inversion also appears in clauses which are connected to a previous clause by the coordinating conjunction et ‘and’:

[25a]Parfois s'ouvre une autre porte et apparaît la silhouette mince et sombre de Monsieur Péréverzev...lit. Sometimes opens another door and appears the thin and dark silhouette of Mr. Pereverzev.
‘Sometimes another door opens and Mr. Pereverzev’s thin dark silhouette appears…’
(Sarraute)

[25b]Elle vieillissait mais c'était comme Cécile Sorel, elle resterait éternellement jeune. On disait “la Miss” et surgissaient un sourire gouailleur et des jambes parfaites.lit. We had to say “the Miss” and appeared a slightly cheeky smile and perfect legs.
‘She was getting old but it was like Cecile Sorel, she would stay young forever. You only had to say “the Miss” and there appeared that slightly cheeky smile and those perfect legs.’
(Sabatier)

53This use of the coordinating conjunction et ‘and’ reminds that of the et ‘and’ that frequently appears in contexts with the infinitif de narration (‘narrative infinitive’):

[26a]Le lendemain, pas de Salavin. Et, cette fois, Edouard de s’inquiéter.‘Next day, no Salavin. And this time, Edward was worried.’
(Duhamel, cited in (Grevisse, 1986: 1314))
[26b] Je m’écriai: “Voilà notre homme” et mes collègues d’applaudir, et le roi d’agréer.I cried out, “There’s our man” to my colleagues’ applause and the king’s approval.’
(Châteaubriand, cited in (Grevisse, 1986: 1314))

  • 22  Cf. Zribi-Hertz and Diagne (2003) on inflectionally deficient clauses in Wolof which create a succ (...)

54In these examples, the second infinitive clause is linked to the previous spatio-temporal context through the use of the conjunction et ‘and’22. Hence, the content of a covert stage topic can be recovered in examples with the coordinating conjunction et ‘and’, and absolute inversion is allowed.

55In the 13 remaining examples, the clause with inversion clearly presents the occurrence of a new event, or of a new person with respect to the scene described in the preceding context:

[27a]Ayant appris leur détention à la Maison de France, je les avais fait libérer et ils avaient travaillé jour et nuit à préparer cette féerie d'un soir. Vint le clou de la fête :
lit. Came the climax of the party:…
‘Having learned they were being held in the Maison de France,
I had had them freed and they had worked night and day to prepare this one-night extravaganza. The climax of the celebrations came when…’ (de Grèce)

[27b]Parmi les écoliers nommés, peu poursuivraient leurs études car, promis au monde du travail, la cérémonie marquait pour eux un point final à l'univers de l'école. Vint la distribution des prix, classe après classe.
lit. Came the distribution of the prizes, class by class.
‘Among the schoolchildren named above, only few would pursue their studies, for the working world was their destiny. The ceremony marked for them the end of their schooling. The prizes were given out, class by class.’ (Sabatier)

[27c]Cecilia avec son violon, Marco avec sa clarinette, ils sourient, nous font signe avec leurs instruments, de loin... Flottements... Accords... Tout le monde s'assoit... Arrive le chef d'orchestre, Eliahu Inbal, un Israélien...
lit. Arrives the conductor, Eliahu Inbal, an Israeli…
‘Cecilia with her violin, Marco with his clarinet, smiling, bob their instruments at us, far away … Stirrings… Tuning… Everyone sits down… The conductor, Eliahu Inbal, an Israeli, arrives.’ (Sollers)

[27d]Napoléon se fait servir ce qu'il appelait un déjeuner-dîner. Arrive son frère Jérôme suivi de tous les généraux; ne prêtant guère de crédit aux enseignements fournis par l'entourage, Napoléon fait le tour de ses armées.
lit. Arrives his brother Jérôme followed by all the generals;…
‘Napoleon has himself served what he called a lunch-cum-dinner. His brother Jerome arrives with all the generals in train; having little confidence in information provided by the entourage, Napoleon reviews his troops.’ (Rheims)

56In conclusion, in all the examples of absolute inversion collected from my corpus search as well as those given by Le Bidois (1952), the event denoted by the absolute inversion construction immediately follows the event in the previous context; it denotes the occurrence of a new event or moment, or the appearance of a new person with respect to the immediately preceding spatio-temporal context. In other words, in all these cases, absolute inversion occurs in a context where the content of a covert stage topic can be recovered from the discourse context.

4.2.3. Covert temporal stage topics: further confirmation

57The hypothesis that absolute inversion occurs in clauses which contain a covert stage topic correctly predicts the unacceptability of absolute inversion in the answer to a so-called out-of-the-blue question:

[28a] Que s’est-il passé?
‘What happened?’

[28b] SV:   Jean est arrivé.
‘John arrived.’
VS:  * Est arrivé Jean.  
lit. Arrived John.
‘John arrived.’

58Although a temporal location is implicit in question [28a], it does not function as the stage topic of the event in the answer. This is confirmed by two independent arguments. Firstly, the temporal location implicit in the question is not lexically or grammatically indicated in the answers in [28b]. As [29] illustrates, anaphoric expressions too are unacceptable in the answer to an out-of-the-blue question (irrespective of word order):

[29a]Qu’est-ce qui s’est passé?
‘What happened?’

[29b] SV: - # Jean est alors arrivé.
‘John arrived then.’

59This shows that the event referred to in the answer to an out-of-the-blue question is not temporally linked to the event denoted in the question. Hence, the answers in [28b] do not contain a stage topic whose content would be recoverable from the preceding context.

60The claim that the temporal location implicit in a question is not a stage topic in the answer to that question is further confirmed by the fact that the temporal location implicit in the question does not necessarily correspond with the tense morphology of the verb in the answer. In other words, the location implicit in the question is not referred to by the tense morphology of the verb in the answer, and, hence, is not a linguistically relevant covert stage topic. This is illustrated by [30], where the tense morphology of the verb in the answer does not correspond with the tense morphology of the verb in the question:

[30] [Context: Peter arrives at home. He sees his wife is crying. He asks:]

[30a]Que s’est-il passé?‘What happened?’
[The woman answers that Jean, their son, who is a journalist, has decided to go and work in a faraway country:]

[30b]Jean partira à l’étranger.‘John is going abroad’

61Hence, a temporal location implicit in an out-of-the-blue question does not function as a stage topic in the answer to such a question. The fact that the answer to an out-of-the-blue question does not contain an implicit stage topic, correctly predicts the unacceptability of absolute inversion in this context.

62In section 2.2. I argued that only those stage topics whose presence is lexically or grammatically indicated in the clause are linguistically relevant. The presence of a covert stage topic may be indicated by a temporal or locative anaphoric expression, such as a temporal or locative pronoun or adverb (cf. section 4.1.). Moreover, when there is no pronoun or adverb in the clause, the tense morphology of the verb may indicate the presence of a covert temporal stage topic, but it is clear that the tense morphology does not indicate the presence of a covert locative stage topic. Thus, when the clause does not contain an explicit anaphoric expression, it is predicted that absolute inversion can only be licensed by a covert temporal stage topic. This explains the ungrammaticality of absolute inversion in examples such as [31a]:

[31a]J’entrai dans la chambre. * Dormaient les enfants.
lit. I entered in the room. Were sleeping the children.

[31b]Dans la chambre dormaient les enfants.
lit. In the room were sleeping the children.

[31c]J’entrai dans la chambre. dormaient les enfants.
lit. I entered in the room. There were sleeping the children.

63In [31a], the location of the sleeping children is the spatial location dans la chambre ‘in the room’ introduced in the first clause. Nevertheless, the example is ungrammatical. This is due to the fact that the presence of the spatial location dans la chambre ‘in the room’ is not lexicogrammatically indicated in the clause with inversion (by the presence of a pronoun, a connective or an adverb). Hence, this implicit location is not a linguistically relevant stage topic, and, as predicted, it does not license inversion. When the location is referred to in the clause, by a PP, as in [31b], or an adverb, as in [31c], inversion is allowed.

5. Conclusion

64This article concentrated on the notion of implicit stage topic, a temporal topic that’s not explicitly realised in the clause. It was argued that implicit stage topics are subject to two constraints: on the one hand, their content must be recoverable either from the preceding context (in a narrative) or from the speech context, and, on the other hand, their presence must be lexicogrammatically indicated in the clause.

65Whereas in Erteschik-Shir (1997) the existence of implicit stage topics is argued for on the basis of theory-internal arguments, this article presents empirical evidence. I have shown that a variety of instances of nominal inversion in French, a syntactic configuration that is influenced by the presence of an overt stage topic (i.e. topical or thematic spatio-temporal phrases), are explained by the presence of a covert stage topic. The fact that implicit stage topics interact with syntactic structure in the same way as overt stage topics do is a strong empirical argument in favor of their existence.

Haut de page

Bibliographie

Berthonneau, A.-M. 1987. La thématisation et les compléments antéposés. Travaux de Linguistique 14/15: 63-82.

Blinkenberg, A. 1928. L’ordre des mots en français moderne. Copenhague: Bianco Lunos Bogtrykkeri.

Bonami, O., Godard, D., Marandin, J.-M. 1999. Constituency and word order in French subject inversion out-of-the-blue question. Constraints and Resources in Natural Language Syntax and Semantics, G. Bouma, E.W. Hinrichs, G.-J.M. Kruiff, T. Oehrle (eds). CSLI Publications. Stanford: Stanford University. 21-40.

Borillo, A. 1998. Les adverbes de référence temporelle comme connecteurs temporels de discours. Temps et discours. S. Vogeleer, A. Borillo, C. Vetters, M. Vuillaume (sous la dir. de). Louvain-la-Neuve: Peeters. 131-145.

Chafe, W.L. 1976. Givenness, contrastiveness, definiteness, subjects, topics, and point of view. Subject and topic. C.N. Li (ed.). New York: New York Academic Press. 27-55.

Charolles, M. 2003. De la topicalité des adverbiaux détachés en tête de phrase. Travaux de Linguistique: adverbiaux et topiques 47. M. Charolles, S. Prévost (sous la dir. de). Louvain-la-Neuve: Belgique. 11-51.

Chetrit, J. 1976. Syntaxe de la phrase complexe à subordonnée temporelle. Etude descriptive. Paris: Editions Klincksieck.

Cinque, G. 1990. Types of A’ dependencies. Cambridge: MIT Press.

Combettes, B. 1996. Facteurs textuels et facteurs sémantiques dans la problématique de l'ordre des mots: le cas des constructions détachées. Langue française 111: 83-96.

Davidson, D. 1967. The Logical Form of Action Sentences. The Logic of Decision and Action. N. Resher (ed.). Pittsburgh: University of Pittsburgh Press. 81-95.

de Swart, H. 1993. Adverbs of quantification: a generalized quantifier approach. Ph.D.Diss. University of Groningen.

Dik, S.C. 1989. The theory of functional grammar.Dordrecht et Providence: Foris.Part 1: The structure of the clause.

Erteschik-Shir, N. 1997. The dynamics of focus structure. Cambridge: CUP.

Erteschik-Shir, N. 1999. Focus structure and scope. Grammar of focus. G. Rebuschi, L. Tuller (eds). Amsterdam/Philadelphia: John Benjamins. 119-150.

Fournier, N. 1997. La place du sujet nominal dans les phrases à complément prépositionnel initial. La place du sujet en français contemporain. C. Fuchs (sous la dir. de). Louvain-la-Neuve: Duculot. 97-132.

Fuchs, C. 1997. La place du sujet nominal dans les relatives. La place du sujet en français contemporain. C. Fuchs (sous la dir. de). Louvain-la-Neuve: Duculot. 135-178.

Grevisse, M., Goosse, A. 1986 (12th ed). Le bon usage. Grammaire française. Louvain-la-Neuve: Duculot.

Grosu, A. 1975. The position of fronted wh-phrases. Linguistic Inquiry 6: 588-599.

Guimier, C. 1996. Les adverbes du français. Le cas des adverbes en –ment. Paris-Gap: Ophrys.

Gundel, J.K. 1989. The role of topic and comment in linguistic theory. New York-Londres: Garland.

Heim, I. 1982. The semantics of definite and indefinite noun phrases. Ph.D.Diss. University of Massachusetts.

Hinrichs, E. 1986. Temporal anaphora in discourses of English. Linguistics and Philosophy 9: 63-82.

Jackendoff, R. 1972. Semantic Interpretation in Generative Grammar. Cambridge: MIT Press.

Jacobs, J. 2001. The dimensions of topic-comment. Linguistics 39: 641-681.

Jonare, B. 1976. L'inversion dans la principale non-interrogative en français contemporain. Stockholm: Almqvist och Wiksell.

Kamp, H. 1981. A theory of truth and semantic representation. Formal methods in the study of language. J. Groenendijk, T. Janssen, M. Stokhof (eds). Amsterdam: Mathematisch Centrum. (Reprinted in J. Groenendijk, T. Janssen, M. Stokhof (eds). 1984. Truth, interpretation and information. Dordrecht: Foris. 1-41.)

Kampers-Manhe, B., Marandin, J.-M., Drijkoningen, F., Doetjes, J., Hulk, A. 2004. Subject NP inversion. Handbook of French Semantics. F. Corblin, H. de Swart (eds). CSLI Publications, Stanford.

Kerleroux, F., Marandin, J.-M. 2001. L’ordre des mots. Cahier Jean-Claude Milner. J.-M. Marandin (sous la dir. de).Lagrasse: Verdier. 277-302.

Korzen, H. 1983. Réflexions sur l'inversion dans les propositions interrogatives en français. Revue Romane 24 (numéro spécial): 50-85.

Korzen, H. 1985. Pourquoi et l'inversion finale en français: étude sur le statut de l'adverbial de cause et l'anatomie de la construction tripartite. Copenhague: Université de Copenhague (Institut d'études romanes Copenhague).

Kratzer, A. 1989. Stage-level and individual-level predicates. Papers on Quantification. J.-M. Marandin (ed.). Department of linguistics, University of Massachusetts, Amherst.

Kuno, S. 1972. Functional sentence perspective. Linguistic Inquiry 3: 269-320.

Laenzlinger, C. 1996. Adverbs, pronouns, and clause structure in Romance and Germanic. Amsterdam: John Benjamins.

Lahousse, K. 2003a. La distribution de l’inversion nominale en français dans les principales non interrogatives et les subordonnées circonstancielles. Linguisticae Investigationis 26 (1): 101-136.

Lahousse, K. 2003b. La complexité de la notion de topique et l’inversion du sujet nominal. Travaux de Linguistique: adverbiaux et topiques 47. M. Charolles, S. Prévost (sous la dir. de). Louvain-la-Neuve. 111-136.

Lahousse, K. 2003c. The distribution of postverbal nominal subjects in French. A syntactic, semantic and pragmatic Analysis. Unpublished Ph.D.Diss. K.U. Leuven (Belgium) and Université de Paris 8 (France).

Lahousse, K. 2005. ‘Focus VS’: a special type of French NP subject inversion. Romance Languages and Linguistic Theory 2003 (Current Issues in Linguistic Theory 270). T. Geerts, I. van Ginneken, H. Jacobs (eds). Amsterdam/Philadelphia, John Benjamins Publishing Company. 161-176.

Lahousse, K. 2006a. NP subject inversion in French: two types, two configurations. Lingua 116 (4): 424-461.

Lahousse, K. 2006b. L’assertion et l’inversion du sujet nominal dans les subordonnées adverbiales. Linguisticae Investigationes 29 (1): 113-124.

Lambrecht, K. 1994. Information structure and sentence form. Cambridge: Cambridge University Press.

Le Bidois, R. 1952. L'inversion du sujet dans la prose contemporaine (1900-1950).Paris: Artrey.

Le Querler, N. 1993. Les circonstants en position initiale. 1001 circonstants. C. Guimier (sous la dir. de). Caen: Presses Universitaires. 159-184.

Le Querler, N. 1997. La place du sujet nominal dans les subordonnées percontatives. La place du sujet en français contemporain. C. Fuchs (sous la dir. de). Louvain-la-Neuve: Duculot. 179-203.

Lerch, E. 1939. Die Inversion im modernen Französisch. Mélanges de linguistique offerts à Charles Bally. Genève: Georg. 347-366.

Marandin, J.-M. 2001a. Unaccusative inversion in French. Romance languages and linguistic theory 1999. Y. D'Hulst, J. Rooryck, J. Schroten (eds). Amsterdam: Benjamins.

Marandin, J.-M. 1997. Dans le titre se trouve le sujet. Ou: l’inversion locative en français. Mémoire d’habilitation. Paris: Université de Paris 7.

Melis, L. 1983. Les circonstants et la phrase. Leuven: Presses Universitaires de Louvain.

Nikolaeva, I. 2001. Secondary topic as a relation in information structure. Linguistics 39: 1-49.

Partee, B. 1984. Nominal and temporal anaphora. Linguistics and Philosophy 7: 243-286.

Reinhart, T. 1981. Pragmatics and linguistics: an analysis of sentence topics. Philosophica 27: 53-94.

Sgall, P., Hajičová E., Panavová, J. 1986. The meaning of the sentence in its semantic and pragmatic aspects. Dordrecht: Reidel.

Strawson, P.F. 1964. Identifying reference and truth values. Theoria 30: 86-99.

Tasmowski L., Willems D. 1987. Les phrases à première position actancielle vide, Par la porte ouverte (il) entrait une odeur de nuit et de fleurs.Travaux de linguistique 14-15: 177-191.

Togeby, K. 1982-1985. Grammaire française. Copenhague: Akademisk Vorlag.

Vallduví, E. 1992. The informational component. New York: Garland.

Vallduví, E. 1994. Information packaging: a survey. Ms. University of Edinburgh, Report prepared for WOPIS.

Zribi-Hertz, A. 2003. Réflexivité et disjonction référentielle en français et en anglais. Essais sur la grammaire comparée du français et de l’anglais. P. Miller, A. Zribi-Hertz (eds). Saint-Denis: Presses Universitaires de Vincennes. 135-175.

Zribi-Hertz, A., Diagne, L. 2003. Déficience flexionnelle et temps topical en wolof. Typologie des langues d’Afrique et universaux de la grammaire. P. Sauzet, A. Zribi-Hertz (eds). Paris: L’Harmattan.

Haut de page

Notes

1  Besides the spatio-temporal frame-setting, Chafe (1976) also distinguishes an ‘individual framework’, which is illustrated by the sentence-initial constituent in Chinese sentences equivalent to Those trees, the trunks are big. This corresponds to the ‘notional framework’ (cadrage notionnel) in Le Querler (1993:178) and the ‘frame’ (le cadre) in Danon-Boileau et al. (1991:117).

2  In Erteschik-Shir (1997/1999) and Vallduví (1992/1994), following proposals made by Reinhart (1981) and Heim (1982), the speaker’s assumptions about the hearer’s knowledge are not considered to be a set of pragmatic presuppositions (as in Lambrecht 1994), but as a well-organized knowledge-store, a file that consists of a certain number of ‘cards’. Each card represents a referential entity (a ‘discourse referent’), and contains a record listing properties attributed to the entity, or to situations, relations or states of affairs in which the entity is involved. According to Reinhart’s (1981) well-known metaphor, this organization essentially resembles the organization of a library’s subject catalogue.

3  In Lambrecht’s (1994) framework, this amounts to saying that, just as regular topics, stage topics are part of the pragmatic presuppositions, which are defined as “the set of propositions lexicogrammatically evoked in a sentence which the speaker assumes the hearer already knows or is ready to take for granted at the time the sentence is uttered” (Lambrecht 1994:52).

4  In Erteschik-Shir’s model, the here-and-now of the discourse context is represented by a card that is always accessible.

5  This implicit stage topic corresponds, as Erteschik-Shir (1997) argues, with Kratzer’s (1989) interpretation of Davidson’s (1967bn) spatio-temporal argument.

6  From a theoretical point of view, the notion of covert stage topic as a formalization of the spatio-temporal linkage between sentences is also appealing to the extent that it enables us to draw a parallel with other types of topics, which can also be implicit. Zribi-Hertz (2003) indeed argues that, whenever a pronoun appears in a sentence, it must be licensed by the presence of an implicit or explicit topic in the clause, which binds the pronoun.

7  Some authors even claim that verbal tenses anaphorically refer to the temporal context in a way similar to the relation between a pronoun and a previously mentioned referent (cf. Kamp 1981, Hinrichs 1986, Partee 1984, and many others following them).

8 The examples with nominal inversion will be presented as follows: first the original French example in italics, then a translation in English (preceded by ‘lit.’) which keeps the VS word order as in French, followed by a ‘good translation’. The ‘lit.’ translation has been chosen instead of glosses to make non-French-speaking persons ‘feel’ the VS word order, and will only be given when the word order in French and English are different.  

9  Henceforth, I use VS and SV in the examples to indicate verb-subject and subject-verb word order, respectively.

10  Based on the attested example (i):
(i) (…) la morne campagne du nord (…), dont les quais semblent plus larges et plus vides qu'ailleurs, quand les déserte la foule des champs de courses. (Gracq)

11  This example is acceptable when the man is not in motion, and, hence, when the verb se balancer is interpreted as ‘to hang’ rather than as ‘to sway’. In this case, the sentence-initial PP is a stage topic.   

12  My translation. The original text reads: “une localisation temporelle ayant déjà fait l’objet d’un calcul dans l’interprétation du discours qui précède” (Borillo 1998:131).

13  I assume that deictic adverbs, just like anaphoric adverbs, are dependent on the presence of a covert stage topic, which refers to the here-and-now of the utterance. As Vallduví (1994) and Erteschik-Shir (1997) argue, the here-and-now is always present in the (pragmatic) representation of a clause.  

14  For some speakers, nominal inversion is awkward in the contexts (16a’) and (16b’):
(i) Pourquoi as-tu l’air si troublée?
‘Why are you looking so confused?’
a. J’étais en train de dormir et soudain a sonné le téléphone.
lit. I was sleeping and suddenly rang the telephone.
b. J’étais en train de dormir et alors a sonné le téléphone.
lit. I was sleeping and then rang the telephone.’
This is due to the register of the interaction in (i): since this is a piece of informal discourse, it is not surprising that inversion is awkward, as it appears above all in formal, written language. Nonetheless, independently from inversion, the contrast in (16) still shows that temporal adverbs are only allowed in contexts where the content of a stage topic can be recovered from the preceding discourse, or when the clause itself contains an overt stage topic.

15  Cf. Le Bidois (1952:30-31) and Jonare (1976:38) for more examples of this type.

16  The hypothesis in (19) also allows a unified account of the two classes of inversion distinguished by Bonami, Godard and Marandin (1999) and Marandin (2001a). These authors postulate the existence of, on the one hand, ‘locative inversion’, which includes inversion after temporal PPs and locative PPs and adverbs, and, on the other hand, ‘unaccusative inversion’, which occurs in clauses introduced by temporal adverbs (among other contexts). All the examples belonging to these two classes are uniformly treated by my descriptive generalization (19), since they all instantiate cases where inversion is favoured by a (covert or overt) stage topic.

17  In this article I do not take into consideration instances of absolute inversion due to the restrictive (i.e. exhaustive) focalisation of the subject (cf. Lahousse 2003a), such as (i). This type of inversion, which I call ‘focus inversion’ (cf. Lahousse 2005 and Lahousse 2006a), includes but is not restricted to ‘heavy subject NP inversion’ (Bonami, Godard and Marandin 1999:21) or ‘elaborative inversion’ (Kampers-Manhe et al. 2004:75):
(i) Se sont qualifiés pour les demi-finales des championnats de France amateurs samedi soir à Dijon: mouche, Rabak Khalouf; coq, A. Consentino, Acquaviva; plume, Lasala; Lainé (…).
lit. Were qualified for the semi-finals in the amateur championship of France Saturday evening in Dijon: Rabak Khalouf, flyweight; A. Consentino, Acquaviva, bantamweight; Lasala, featherweight; Lainé (…)
(L’Aurore, cited by Jonare, 1976: 40)

18  The distribution of absolute inversion has also been described to some extent by Blinkenberg (1928:88-94), Lerch (1939) and Jonare (1976). Since the contexts mentioned by these authors are all included in Le Bidois’ classification, I only mention the latter.

19  According to Jonare (1976:35), rester ‘to stay’ is the verb which occurs most frequently in absolute inversion.

20  For technical reasons inherent to the search engine of Frantext, it is impossible to search for examples of absolute inversion with all verbs. The verbs used for this test are the verbs which appear most frequently in inversion.

21  Why is it that examples of absolute inversion with a temporal subject are so frequent? Interestingly, this is not restricted to absolute inversion: in 22.8% of the quand-clauses with nominal inversion I found in the Frantext corpus (1950-2000), the subject is also temporal (cf. Lahousse 2003c). In my view, temporal subjects are very frequent in absolute inversion and in quand-clauses because the new reference time encoded in the postverbal subject contrasts with the preceding reference time, i.e. the covert stage topic in absolute inversion and the moment referred to by the temporal conjunction. Indeed, the contrastive focalization of the subject is one of the factors that favor nominal inversion in contexts where it is otherwise not allowed (cf. Lahousse 2006b).

22  Cf. Zribi-Hertz and Diagne (2003) on inflectionally deficient clauses in Wolof which create a successive semantic effect and require the anchoring of the event to a temporal point provided by the immediately preceding context. According to the authors, this is best translated by a narrative infinitive in French, preceded by the coordinating conjunction et ‘and’. In other words, the sequence [et ‘and’ + narrative infinitive] in French indicates that the clause is temporally linked to the immediately preceding context, and, hence contains an implicit stage topic.

Haut de page

Pour citer cet article

Référence électronique

Karen Lahousse, « Implicit stage topics », Discours [En ligne], 1 | 2007, mis en ligne le 02 avril 2008, consulté le 16 décembre 2017. URL : http://journals.openedition.org/discours/117 ; DOI : 10.4000/discours.117

Haut de page

Auteur

Karen Lahousse

Fonds national de la Recherche Scientifique - Flandres
Katholieke Universiteit Leuven
Chargée de recherches
Karen.Lahousse@arts.kuleuven.be

Haut de page
  • Logo PUC
  • OpenEdition Journals