topic and focus in old french v1 and v2 structures · 2014. 3. 29. · hypothesis that the old...
TRANSCRIPT
1
Topic and Focus in Old French V1 and V2
structures
1. Introduction
A growing body of work explores the role of Information Structure
(henceforth IS) on word order variation and change. The
cartographic approach to syntactic structures (e.g. Cinque 2002;
2006; Cinque and Rizzi 2008; Rizzi 1997; 2004), despite its
shortcomings (Craenenbroeck 2009), has shown that specific
positions in the clause appear to be dedicated to topic and focus
constituents (Belletti 2004; Benincà 1999; 2006; Benincà and
Poletto 2004; Benincà and Munaro 2010), and a growing number of
authors argue that changes in word order reflect changes in how a
language expresses these notions (e.g. Benincà 2006; Roland
Hinterhölzl 2009; Hinterhölzl and Petrova 2009; 2010; Laenzlinger
2006; Kroch and Santorini 2009; and the various papers in Batllori
and Hernanz 2011). The present paper centers on Old French (OF)
V1 and V2 structures with the aim of bringing to light the possible
correlations between the informational role of phrases and their
position.
2
Since the end of the 19th c., it has been observed that the Old French
finite verb tends to be found in second position in main clauses
(Thurneysen 1892; Foulet 1928; Skårup 1975).
(1) messe e matines ad li reis escultet
mass and matins has the king heard
‘The king has attended mass and matins’
(1100, Roland 11;139)1
This verb-second, or V2, property is well documented in German
and other Germanic languages. Within the Principles and
Parameters’ approach of the 1990’s, V2 was analyzed as having the
verb under C and some constituent in Spec,CP, this position
corresponding to Skårup (1975)’s place du fondement, or German’s
Vorfeld (Adams 1987; 1988; Platzack 1995; Roberts 1993; Vance
1995). This is represented schematically in (2) (using the
copy+delete representation of move):
(2) [CP messe e matines ad [IP li reis ad escultet messe e
matines]
A number of authors have recently challenged the V2 analysis of
Old French, in part based on the argument that V1 and V3
constructions are more frequent in Old French than in either modern
German (Ferraresi and Goldbach 2002; Kaiser 2000; 2002; Kaiser
and Zimmermann 2011; Rinke and Meisel 2009) or Middle High
3
German (1050-1350) (Elsig 2009). Among these, Rinke and Meisel
(2009) claim, on the basis of a study of two early 13th c. texts
(Villehardouin and Les 7 sages de Rome), that in Old French the
preverbal position is a Topic position. In declaratives, the verb
would surface under Tense; the preverbal position (Spec,TP) would
be a Topic position, and the VP would contain the Information
Focus portion of the clause.
(3) [TP XPTop V [VP …...]Foc ]
For the authors, this means that
1) subject-inversion would be “contingent on the topicalisation of
a non-subject constituent” (p. 112)2 , and
2) subjects would move to the preverbal position to avoid their
being interpreted as Information Focus or as part of a thetic
sentence (“An incompatibility of the post-verbal subject with
an interpretation as information focus or as part of a thetic
sentence would cause the subject to move to the pre-verbal
position.”, p.109).
In this perspective, OF contrasts with contemporary German, where
the informational role of the preverbal position is not restricted to
Topics, but may also be occupied by a Focus or an adverb that is
neither Topic nor Focus. On the other hand, OF turns out to be
similar to Old High German if, as argued by Hinterhölzl and
4
Petrova (2010; 2011), the OHG verb served to separate the
aboutness Topic from the comment (for these authors, the evolution
to modern German would result from a neutralization of the
information structural specialization of the sentence-initial
position). According to Rinke and Meisel, in being a Topic-initial
language, OF would have a grammar similar to that of
contemporary Romance null subject languages like Italian, Spanish
or Portuguese (but see Remberger 2010 for evidence that Focus
elements may be initial in these languages).
In the research presented here, we study Old French V1 and V2
constructions from the perspective of these information structural
accounts. On the basis of previous research, we expect early 12th c.
texts to allow preverbal Focus constituents (Marchello-Nizia 1995;
2009; Rickard 1962) (this research is discussed below under 5.2).
However it may be that the language passed through a transitional
Topic-initial stage (TVX) before becoming SVO (see Fleischman
1992: 463). The purpose of the present paper is therefore to test the
hypothesis that the Old French grammar passed through a stage
corresponding to Rinke and Meisel’s characterization, in which:
1) The preverbal position is reserved for Topics.
2) Subjects move to the preverbal position in order to avoid being
interpreted as part of the Focus.
5
We will provide arguments against both claims. It is important to
realize that the second claim does not necessarily follow from the
first one, since there may be more than one topical constituent in a
clause. Thus, Speyer (2010) showed that, in German, when two or
three of the following types of topical constituents are present in a
clause, (1) Information Topic, (2) Scene-setting Topic, (3) element
standing in a poset relation to some discourse set, the competition
for the vorfeld position is resolved according to the following
ranking: scene-setting >> poset >> I-topic; the other topic(s) are
realized postverbally. A similar competition may occur in a
language where the verb is in second position and the preverbal
position is reserved for a Topic. If the ranking observed in German
for the vorfeld applies to a TVX language, we expect to observe
sentences with a Scene-setting Topic in preverbal position and a
postverbal subject fulfilling the role of Information Topic. In such a
language, 1) would be verified, but not 2). If, however, the Old
French verb marks the boundary between Topic and Focus
constituents, we would expect the verb to appear in third position
whenever there are two topical constituents in the clause, and in
fourth position if three topical constituents are present.
6
2. Information structure – Definitions
Notions like Topic and Focus are notoriously complex. In the
present paper, the following operational definitions are adopted :
Information Topic : “The Topic constituent identifies the entity or
set of entities under which the information expressed in the
comment constituent should be stored in the [Common Ground]
content.” (Krifka 2008) As formulated by Krifka, this definition
applies specifically to the entity which is the subject of discourse,
i.e. the Aboutness topic (Reinhart 1982) or Information Topic (I-
Topic). An I-Topic may co-occur in a clause with other types of
topics, like frame-specifying topics, which are generally clause-
external and have different functions (cf. Andréasson 2007;
Nikolaeva 2001). An I-Topic is typically a definite constituent,
referring to an entity given in the discourse or accessible in the
Common Ground (this corresponds to Vallduví 1993’s link).
Information Focus : The Information Focus (I-Focus) is linked to
the pragmatic principle of Progression (e.g. Charolles 1978) which
states that if a sentence is to be informative, it must contain material
that is new in relation to previously known information (the
background). The I-Focus of the sentence is the new information
that should be stored in the Common Ground.
7
According to Gundel & Fretheim (2004), these notions of
Information Topic and Information Focus correspond to relationally
given/new information respectively, and are equivalent to the
notions of Theme/Rheme, or Topic/Comment. However, it has been
pointed out that the Comment is not always equivalent to the I-
Focus, since it may contain an informative part (the I-Focus per se)
as well as background (uninformative, relationally given)
information. For example, in (4), from Krifka (2008: 42, ex. 41),
married her in sentence B is background information within the
comment (Büring 2007: 5; Vallduví 1993’s tail).
(4) A: When did [Aristotle Onassis]Topic marry Jacqueline
Kennedy?
B: [He]Topic [married her [in 1968]Focus]]Comment
Thetic sentences are all-focus. This is the case with the second
sentence in a dialog like: What happened? The telephone rang.
Contrastive Focus : A Contrastive Focus (C-Focus) is defined by
Gundel & Fretheim (2004) as being “material which the speaker
calls to the addressee’s attention, thereby often evoking a contrast
with other entities that might fill the same position.” The notion of
contrast is also present in Krifka’s (2008) definition of Focus as
indicating the presence of alternatives. Contrastive Focus may be
marked by prosody or by expressions like even, only, also.
8
Contrary to the I-Focus, a C-Focus may be part of a Topic, as in the
following example:
(5) A: What do your siblings do?
B: [My [SIster]Focus]Topic [studies MEDicine]Focus, and
[my [BROther]Focus]Topic is [working on a FREIGHT
ship]Focus.
Krifka (2008: 44) talks of Contrastive Topics in that case (a
Contrastive Topic being defined as an aboutness Topic containing a
Focus). (On the difference between I-Topics and C-Topics, see
Lipták 2010).
With these notions in mind, let us turn to our study.
3. The Corpus
The data are extracted from the MCVF corpus of Old French parsed
texts (Martineau et al. 2010, and Penn supplement). We selected the
19 texts dated between 980 and 1309. From these texts, we
extracted all positive declarative matrix clauses with a full DP
subject. We report here our analysis of V1 and V2 clauses (V3+
clauses were kept for later investigation).
One characteristic of the corpus in its present state is that the 12th c.
texts are all in verse except for the Quatre Livres des Rois (1170).
This is due to the lack of 12th c. prose texts, all surviving texts of
9
some length being in verse. Conversely, all of the 13th c. texts in the
corpus are in prose, with the unfortunate consequence that there is
(near) conflation of the text genre and time variables. In the figures
to be presented, both the date and the text genre will be indicated.
4. V1 clauses
If the OF verb marks the boundary between Topic and Focus
constituents, we would expect the post-verbal subjects in V1
clauses to correspond to the I-Focus. In order to verify this
prediction, we selected strict V1 clauses (i.e. not introduced by a
coordinator, since coordinated V1 declaratives may have their own
characteristics, Labelle and Hirschbuhler 2005).
Let us first consider sentences other than those where the verb
introduces direct discourse. Taking the context into account, we
coded the subjects in all these sentences as being I-Topics, I-Focus,
or Unclear (see Cook and Bildhauer 2011 for a discussion of the
difficulties facing the coding of information structure). The results
are displayed in Table 1. While there is a dominance of I-Focus
(rhematic) subjects, the fact that over a third of the postverbal
subjects were coded as I-Topics confirms Rouveret (2004: 196)’s
comment that “It does not seem that the postverbal position in [OF]
V1 sentences is pragmatically specialized.”
10
I-Focus I-Topic Unclear Total
51 35 17 103
49.5% 34% 16.5%
Table 1 - Informational role of postverbal subjects in V1 declaratives
The following examples illustrate subjects that were coded as being
Topics:
(6) Curecerent s’ en les princes des Philistiens
got-angry REFL GEN the princes of-the Philistians
‘The princes of the Philistians got angry at this.’
(1170; QLR1-2;1332)
(7) Cunuit Brandans a l’ air pluius
knew Brendan from the air rainy
Que li tens ert mult annüus.
that the weather was very worrysome
‘Brendan knew from the wet wind that the weather was
worrysome’
(1120; Brendan, 56;675)
Figure 1 illustrates the evolution of V1 declaratives over time (the
numbers indicated at the bottom of the graph correspond to the total
number of relevant examples in the texts). In Roland (1100), the
11
large majority of postverbal subjects were coded as being Focus,
but in Brendan (1120) and the Quatre livres des Rois (1170), they
were more often Topics. The construction dropped out of the
language at the beginning of the 13th c. when the language became
more strictly V2. This is why there are few examples after 1170;
three texts of the corpus dated between 1205 and 1267 do not
appear in the figure because they contain no V1 declarative. After
1267, V1 declaratives re-appear. They contain no clear Topic
subject, but there are only 7 examples.
Figure 1
Let us now turn to verbs introducing direct discourse. Here V1
declaratives alternate with V2 declaratives, some of type SV(X),
and some of type XVS. The three constructions are illustrated
schematically in (8-10). We did not consider parentheticals and
cases where the verb of saying follows the direct discourse, since
12
there is no word order variation in these cases (the subject is always
postverbal).
(8) Dit Roland: « ... » = VS
‘Says Roland: ….’
(9) Roland dit: « ... » = SV
‘Roland says: ….’
(10) a. Ço dit Roland: « ... » = cataphoric object VS
‘This says Roland : ….’
b. Donc dit Roland: « ... » = adverb VS
‘Thus says Roland : ….’
As far as we could see, the subject has the same informational
function in the three constructions: it signals a new speaker or a
change of speaker. In the three constructions, the subject is almost
always definite (there are 7 indefinite subjects out of 580 V1
declaratives, distributed across the 3 constructions). With the
disappearance of VS declaratives at the beginning of the 13th c., we
may ask the following question: by which construction are VS
declaratives replaced? If the postverbal position of the subject in VS
declaratives was due to its informational role, we would expect VS
sentences to be replaced by XVS sentences. However, Figure 2
shows that between 1177 and 1225, VS sentences, which drop to
13
zero percent, are replaced by SV(X) sentences more often than by
XVS sentences.
Figure 2
This suggests that the VS order in declaratives resulted from a
desire to place the verb first, and not from a desire to place the
subject in postverbal position. When speakers started to avoid
placing emphasis on the verb in declaratives, a V2 construction was
used instead, and subjects were largely preferred over either objects
or adverbs to fill the preverbal position (cf. Labelle and
Hirschbuhler 2005 for an analysis).
5. V2 clauses
Let us now turn to V2 clauses. Two questions are adressed in this
section: Firstly, do postverbal subjects belong to the comment or are
the I-Focus of the clause? Secondly, what is the informational role
of preverbal constituents?
14
5.1 Subjects
Since definite subjects tend to be topics while indefinite subjects
make bad topics & are often found in thetic sentences, we may ask
ourselves whether there is a marked tendency to find definite
subjects preverbally and indefinite subjects postverbally. We
therefore coded the nature of subjects in non-coordinated V2
sentences. Figures 3a and 3b show that both preverbally and
postverbally, subjects are overwhelmingly definite.
Figure 3
15
If we concentrate our attention on the 192 indefinite subjects, we
find a small tendency to place indefinite subjects postverbally, but
clearly no exclusion of indefinite subjects from the preverbal
position: 44.7% of indefinite subjects are found preverbally in verse
texts (12th c.) and 43% in prose texts (1170 and 13th c.). The
following examples show that these indefinite preverbal subjects
may be I-Focus.
(11) (Une forest aveit entur, Trente liwes ot de lungur.)
(There was a forest about, It was thirty leagues long)
Un seinz hermites i maneit
A saint hermit LOC lived
‘A saint hermit lived there’
(1180; Marie de France, Eliduc 182;3715)
(12) Une musteile vint curant,
A weasel came running
‘A weasel came running’
(1180; Marie de France, Eliduc 187;3815)
(13) Doi gentil homme du paÿs,
two gentlemen of-the country,
qui pas ne l’ amoient,
who not NEG him liked
saillirent hors a un trespas,
16
jumped out at a passage
‘Two gentlemen of the country, who didn’t like him, jumped
out at a passage’
(1267; Cassidorus, 643;4056)
Considering only texts containing more than ten indefinite subjects,
we observe a small decrease in the tendency to place indefinite
subjects in preverbal position (Figure 4), but the percentage of
preverbal indefinite subjects at the end of the period is still 35%.
Therefore, it does not appear that there is a constraint against
rhematic (I-Focus) subjects in preverbal position.
Figure 4
To summarize, the distribution of definite vs indefinite subjects in
the corpus does not support the hypothesis that preverbal subjects
are always Topics or that postverbal subjects always part of the
comment or Information Focus of the clause. The following
17
sections will provide examples of overt postverbal subjects which
are I-Topics.
5.2 Preverbalconstituentsdistinctfromasubject
Recall that Rinke & Meisel claim that, while in German the
constituent in the prefield may be a Topic, an Informational Focus,
a Constrastive Focus or an adverb that is neither topic nor focus, in
Old French the preverbal constituent would always be a Topic or an
adverb linking with the preceding discourse, which Rinke & Meisel
consider as a type of Topic. In the present section, we consider
XVS sentences with the aim of determining whether the preverbal
position is a Topic position, and whether the postverbal subject is
always interpreted as being part of the information Focus portion of
the clause.
Contrastive Focus. Let us first observe that a contrastive focus
may occupy the preverbal position.
(14) Meïsmes a l' empereour sont les lermes venues
even to the emperor are the tears come-PRT
aus yex,
to-the eyes
‘even to the emperor, the tears came to the eyes’
(1267; Cassidorus, 664;4447)
18
Rickard (1962, 19) provides many other examples of clause-initial
C-Focus constituents in Old French texts, like the following, where
vaincu in the second clause contrasts with mort in the previous
clause:3
(15) Vous m’ avez mort sans recouvrier,
You me have dead without recourse,
mais vaincu ne m’ avez vous mie.
but defeated NEG me have you not
‘You killed me without recourse,
but you have not defeated me.’
(Merlin in prose, ii. p. 52)
In addition, if we are correct in considering that, in (16)-(18), the
information conveyed by the preverbal subject is the new
information conveyed by the sentences (here of type SV(X)) , these
examples provide further confirmation that a preverbal constituent
may be a contrastive focus.
(16) Meïsmes la pucele y fu
even the girl LOC was
‘even the girl was there’
(1267; Cassidorus, 149;1211)
(17) Sul David é Jonathas le sourent.
only David and Jonathas it knew
19
‘only David and Jonathan knew it’
(1170; QLR1-2;757)
(18) Nes li oisel s’ an istront fors;
even the birds REFL GEN go-FUT away
‘even the birds will leave’
(1177; Yvain, 13;394)
Preverbal objects. Marchello-Nizia (1995) studied the role of
preverbal direct objects in two texts: Roland (1100) and La Queste
del Saint Graal (1225). In Roland, Marchello-Nizia did not notice
any restriction on the informational role of preverbal objects.
However, she observed that they tend to be more often rhematic (I-
Focus) than thematic (I-Topic) (pp. 99-100). By contrast, the OV(S)
order is more restricted in the Queste, where it serves mainly to
thematize the object; when the object is rhematic, the OVS order is
marked and is found mainly in idioms of type mander saluz (send
greetings) (19), or when O is either strongly focused or
cataphorically linked to the discourse.4
(19) Saluz vos mande li bons chevaliers
Greetings you send the good knight
‘The good knight sends you his greetings’
(Queste, p. 30, in Marchello-Nizia, 1999: 44)
20
The change in the informational role of the preverbal object is
confirmed in our corpus. Taking into account the context, we coded
all preverbal objects in V2 sentences as being Topic, Focus or
Unclear. As shown in Figure 5, a preverbal object tends to be the I-
Focus of the clause before 1205 but the I-Topic after 1225. It is
important to realize, though, that OVS with a preverbal Focus
object is grammatical until the end of the period studied; at the end
of the period this construction still makes up twenty to forty percent
of preverbal objects.
Figure 5
In the present corpus, when the preverbal object is an I-Focus, the
postverbal subject is generally interpreted as the Topic. This
situation is illustrated in (20)-(23).
(20) .XX. escheles ad li reis anumbrees.
twenty columns has the king counted
‘the king has counted twenty columns’
(1100; Roland, 112;1459)
21
(21) et divers chanz chantoit chascuns;
and various songs sang each-one’
‘and they all sang various songs’
(1177; Yvain, 15;453)
(22) La main destre leva adonques la dame,
The hand right raised then the lady
‘Then the lady raised the right hand’
(1177; Yvain,202;7065)
(23) La maniere comment il pristrent la cité de Baudas
The manner how they took the city of Baudas
et le calife nous conterent les marcheans;
and the caliph us told the merchants
‘The merchants told us how they took the city of Baudas and
the caliph’
(1309; Joinville,289;3370)
(this example is also discussed in Rickard 1962: 29)
This means that the particular analysis of Rinke and Meisel
according to which the subject moves to the preverbal position to
escape being interpreted as part of the information Focus is
incorrect. In Old French, a postverbal subject may be the I-Topic
while the preverbal position is occupied by an I-Focus. In other
22
words, the verb does not mark the boundary between Topics and
Focus constituents.
Preverbal predicates and quantifiers. Let us now turn to other
types of preverbal constituents. The preverbal field may contain
constituents that are not candidates to topichood and that must be
interpreted as being part of the comment. These include quantifiers,
non-finite verbs, adjectives and other predicates. Examples are
given in (24-27); the postverbal subject was coded as being the I-
Topic of the clause.
(24) Mut est Lanval en grant esfreie! =QP
much is Lanval in great fright
‘Lanval is greatly afraid’
(1180; Marie de France, Lanval, 196; 78.1592)
(25) Trenchet li ad li quens le destre poign, = NF-V
cut-PRT him has the count the right hand
‘The count has cut his right hand’
(1100; Roland, 142;1926)
(26) Malade ot geü longuemant la pucele, = Adj
sick has laid a-long-time the girl
‘The girl had been sick for a long time’
(1177; Yvain, 177;6235)
(27) Male chose est murmure, =Pred. NP
23
bad thing is whisper
‘Whispering is a bad thing’
(1279; Somme Royal, 1,64;1758)
Such constituents were always rare in preverbal position compared
to adverbials and PP’s. Table 2 shows that their percentage is higher
in verse texts (12th c.) than in prose texts (mainly 13th c.). This
might suggest that the preverbal position is indeed more frequently
rhematic in 12th c. texts than in 13th c. texts.
Adjective Non Fin. Verb Pred.NP
Verse 2.7% 2.0% 0.6%
Prose 0.6% 0.3% 0.2%
Table 2 - Distribution of adjectives predicative NP’s and non finite verbs in
preverbal position
Adverbials and PPs. We coded the informational status of
subjects in V2 sentences introduced by an adverbial (in which we
included temporal DP’s like le lendemain (the next day), le jeudi
(on Thursday)). Figure 6 presents the results of this analysis. It can
be shown that the postverbal subject is generally a Topic, and that
no observable change is apparent within the period considered.
24
Preverbal adverbials may be locative or temporal. In that case, they
are generally frame setting (Hanging Topics or Scene Setting
Topics in Benincà and Poletto (2004)’s terms). In that function,
they are topical, but do not constitute the I-Topic of the clause, i.e.
the “entity or set of entities under which the information expressed
in the comment constituent should be stored in the [Common
Ground] content.” Indeed, when the preverbal field is filled by a
Scene Setting Topic and the clause contains a postverbal full DP
subject, this subject is generally the I-Topic of the clause, as in the
following examples, which illustrate the type of situation described
by Speyer (2007) for German.
(28) La fist Cassidorus plus d’ armes
There made Cassidorus more (feats) of arms
que onques Achilles ne fist devant Troies.
than ever Achilles NEG made in-front-of Troy
‘There, Cassidorus made more feats of arms than Achilles
ever made in front of Troy’
Figure 6
25
(1267;Cassidorus,26;627)
(29) Le jour de la saint Marc me dit le roy
The day of the Saint Marc me told the king
que a celi jour il avoit esté né;
that at that day he had been born
‘On Saint Marc’s day, the king told me that he was born that
day’
(1309; Joinville,305;3574)
In the majority of cases where the postverbal subject is coded as I-
Topic, the preverbal adverb is lors, puis, après, donc, adont, a
temporal adverbial which links with the previous discourse without
being clearly scene-setting or topical. The particle si may also
occupy the preverbal position:
(30) Dont sot Ydoines vraiement que Diex l’ amoit,
Then knew Ydoine truly that God her loved
‘Then Ydoine truly knew that God loved her.’
(1267; Cassidorus,351;3031)
(31) Apres assegia Karles Cordis.
After assieged Karl Cordis
‘After that, Karl assieged Cordis’
(1220; Pseudoturpin-2,285;422)
(32) Si demora laienz Perceval avec s’ antain.
26
thus stayed therein Perceval with his aunt
‘Thus, Perceval stayed there with his aunt’
(1225; Queste, 107;2806)
As for PP’s, when they are preverbal, they generally link with the
previous discourse without constituting the I-Topic of the clause. In
the following examples, the postverbal subjects constitute the
I-Topics.
(33) Por ce panse mes sire Yvains qu’ il l’ ocirra
For this thinks my lord Yvain that he him kill-FUT
premieremant;
first
‘For that reason, my lord Yvain thinks that he will kill him
first’
(1177; Yvain,102;3563)
(34) Del colp chancelad li gluz
At-the blow faltered the giant
‘The giant faltered at the blow’
(1170; QLR1-2,.470)
In rare cases, the preverbal PP clearly provides new information (is
I-Focus). This is illustrated below. In these examples, the postverbal
subject again constitutes the I-Topic.
(35) Au geu de la verté l’ a prise Lunete,
27
At-the game of the truth her has taken Lunete
molt cortoisemant.
very curteously
‘Lunete has led her into the game of truth, very curteously’
(1170; Yvain,201;7046)
(36) En grant effrei erent amdui.
In great fright were both
‘Both of them were greatly afraid’
(1180; Marie de France, Guigemar, 476;359)
(37) E en sa maín destra tient l’ imagre una clef
and in its hand right holds the statue a key
que nus ne li puet tolír.
that no-one NEG it-DAT can take
‘and the statue holds a key in its right hand, that nobody can
steal from it’
(1220; Pseudoturpin-1,269;151)
To summarize, in sentences of type XVS, the postverbal subject
may be the I-Topic of the clause; the preverbal constituent may be
topical, or it may be (part of) the I-Focus of the clause. We
conclude that throughout the 12th and 13th c.,
1) the verb does not delimit the boundary between Topic
constituents to its left and Focus constituents to its right;
28
2) the preverbal constituent may be an I-Focus, contrary to the
hypothesis that the language became TVX during the period;
3) there is no rule that forces subjects to raise to preverbal position
in order to avoid being interpreted as part of the Focus.
6. Comparison with other Germanic languages
Before concluding, let us compare the distribution of preverbal
constituents in Old French and in Germanic languages.
Bohnacker and Rosén (2007, 34 & 36) compare the prefields in
German and Swedish, and they arrive at the figures in Table 3.
Commenting on these figures, the authors observe that both
Swedish and German tend to start declaratives with a subject, to let
the subject coincide with the theme and topic, and to place the
theme before the rheme. But Swedish has a stronger tendency to
place the rheme after the verb, to start with an element of low
informational value and with a phonologically light element (e.g.
expletive, det, så), and to use few fronted objects. Although the
constructions with a heavy rhematic object found in German are
considered grammatical in Swedish, they are dispreferred. Swedish
typically fronts objects that are themes. This explains the difference
in the percentages of fronted objects in German (about 7%) and in
Swedish (about 3%).
29
Subjects Objects Adverbials Others
German
newspapers
54% 6.6% 36.8% 2.5%
Germaninformal 50% 7% 42% 1%
Swedishnewsp. 64% 2.3% 30.8% 3%
Swedishinformal 73% 3% 23% 2%
Table 3 - Constituents in the prefield in German and Swedish (from
Bohnacker & Rosen 2003)
The corresponding figures in our data are given in Table 4. It can be
seen that the percentages of the various types of preverbal
constituents in the two Tables are of the same order of magnitude.
The comparison reinforces the conclusion that during the 12th and
13th c., OF was a V2 language of the Germanic type.
Subjects Objects Adv.&
PP’s
Others
OFVerse 58% 7% 29% 10%
OFProse 64% 3% 33% 2%
Table 4 – Constituents in the prefield in Old French
30
Considering the percentage of preverbal objects, it is similar in
verse texts (12th c.) to what is found in German, while the
corresponding percentage in prose texts (mainly 13th c.) is similar to
what is found in Swedish. This observation, compounded with the
fact that, as we saw, (1) fronted preverbal objects tend to be
rhematic in verse texts and are more often thematic in prose texts,
(2) the percentage of “other” constituents like predicates and non
finite verbs is lower in prose texts, and (3) there is a small decrease
in the tendency to find indefinite subjects preverbally, suggests that
the use of the prefield in OF V2 sentences may have evolved from a
use more typical of that of contemporary German to one more
typical of that of contemporary Swedish. These changes might
indicate a change towards a preference for keeping focal
constituents in postverbal position. It is clear, however, that Topics
are not restricted to the preverbal position during the period
considered.
7. Conclusion
The present research leads to the conclusion that, from an
information structure viewpoint, OF is very similar to a V2
language of the Germanic type until the end of the 13th c. As in
German, the preverbal position of an Old French sentence could be
filled by an I-Focus, by a C-Focus, by an I-Topic, by a distinct type
31
of topic, like a scene-setting topic, or by an adverb that was neither
topic nor focus. Clearly, Old French was not strictly Topic initial at
any point during the period considered, even though, just as in
Germanic languages, the preverbal position tended to be filled by a
topical constituent. In addition, nothing prevented postverbal
subjects from being interpreted as the Topic of discourse.
However, a change is observed in the use of the left periphery. The
12th c. grammar is statistically closer to German and the 13th c.
grammar, closer to Swedish. This change might indicate a change in
progress towards a grammar where focal constituents remain in
postverbal position.
8. References
Adams, Marianne P. 1987. “From Old French to the Theory of Pro-
drop.” Natural Language & Linguistic Theory 5 (1): 1–32.
———. 1988. “V2 Effects in Old and Middle French.” Revue
québécoise de linguistique théorique et appliquée 7 (3): 13–
39.
Andréasson, Maia. 2007. “The Architecture of I-Structure.” In
Proceedings of LFG07, ed. Miriam Butt and Tracy
Holloway King, 26–43. CSLI Publications. Stanford.
http://csli-publications. stanford. edu.
32
Batllori, Montserrat, and M. Lluïsa Hernanz (eds.). 2011.
Generative Diachronic Syntax: Word ordre and information
structure. Catalan Journal of Linguistics, Volume 10.
Barcelona: Universitat Autonoma de Barcelona.
Belletti, Adriana. 2004. “Aspects of the Low IP Area.” In The
Structure of CP and IP. The Cartography of Syntactic
Structures, Vol. 2, ed. Luigi Rizzi, 16–51. Oxford: Oxford
University Press.
Benincà, Paola, and Nicola Munaro. 2010. Mapping the Left
Periphery. The Cartography of Syntactic Structures, Vol. 5.
Oxford: Oxford University Press.
Benincà, Paola. 1999. “The Position of Topic and Focus in the Left
Periphery.” In Current Studies in Italian Syntax. Essays
Offered to Lorenzo Renzi, ed. Guglielmo Cinque and
Giampaolo Salvi, 39–64. Amsterdam: Elsevier-North
Holland.
———. 2006. “A Detailed Map of the Left Periphery of Medieval
Romance.” In Cross-linguistic Research in Syntax and
Semantics: Negation, Tense and Clausal Architecture, ed.
Raffaella Zanuttini, Héctor Campos, Elena Herberger, and
Paul Portner, 53–86. Washington, D.C.: Georgetown
University Press.
33
Benincà, Paola, and Cecilia Poletto. 2004. “Topic, Focus, and V2.”
In The Structure of CP and IP. The Cartography of
Syntactic Structures, Vol.2, ed. Luigi Rizzi, 52–75. Oxford:
Oxford University Press.
Bohnacker, Ute, and Christina Rosén. 2007. “How to Start a V2
Declarative Clause: Transfer of Syntax vs. Information
Structure in L2 German.” Ed. Merete Anderssen and Marit
Westergaard. Nordlyd 34 (3). Papers from the Language
Acquisition Workshop XXnd SCL 2006, Tromsø: CASTL:
29–56.
Büring, Daniel. 2007. “Semantics, Intonation, and Information
Structure.” In The Oxford Handbook of Linguistic
Interfaces, ed. Gillian Ramchand and Charles Reiss, 445–
474. Oxford, UK: Oxford University Press.
Charolles, Michel. 1978. “Introduction aux problèmes de la
cohérence des textes.” Langue Française 38 (1): 7–41.
Cinque, Guglielmo. 2006. Restructuring and Functional Heads. The
Cartography of Syntactic Structures, Vol. 4. Oxford: Oxford
University Press.
Cinque, Guglielmo, and Luigi Rizzi. 2008. “The Cartography of
Syntactic Structures.” STiL (Studies in Linguistics) 2: 42–58.
34
Cinque, Guglielmo, ed. 2002. Functional Structure in DP and IP.
The Cartography of Syntactic Structures, Vol.1. Oxford:
Oxford University Press.
Cook, Philippa, and Felix Bildhauer. 2011. “Annotating
Information Structure: The Case of Topic.” In Beyond
Semantics. Corpus-based Investigations of Pragmatic and
Discourse Phenomena., ed. Stefanie Dipper and Heike
Zinsmeister, 3:45–56. Bochumer Linguistische
Arbeitsberichte.
Craenenbroeck, Jeroen van. 2009. “Alternatives to Cartography: An
Introduction.” In Alternatives to Cartography, ed. Jeroen
van Craenenbroeck, 1–14. Berlin: Mouton de Gruyter.
Elsig, Martin. 2009. “Verb Second Effects in Old French and
Middle High German: Considerations from the Language
Border.” Research Center 538: Multilingualism; H1
Multilingualism as Cause and Effect of Language Change:
Historical Syntax of Romance Languages.
Ferraresi, Gisella, and Maria Goldbach. 2002. “V2 Syntax and
Topicalisation in Old French.” Linguistische Berichte 189
(Feb): 3–25.
Fleischman, Suzanne. 1992. “Discourse and Diachrony: The Rise
and Fall of Old French SI.” In Internal and External Factors
35
in Syntactic Change, ed. Marinel Gerritsen and Dieter Stein,
433–473. Berlin: Mouton de Gruyter.
Foulet, Lucien. 1928. Petite syntaxe de l’ancien français. 3e édition
revue. Paris: Honoré Champion.
Gundel, Jeanette K., and Thorstein Fretheim. 2004. “Topic and
Focus.” In The Handbook of Pragmatics, ed. Laurence R.
Horn and Gregory Ward, 175–196. Malden, MA: Blackwell
Publishing.
Hinterhölzl, Roland, and Svetlana Petrova, ed. 2009. Information
Structure and Language Change: New Approaches to Word
Order Variation in Germanic. Vol. 203. Mouton de Gruyter.
———. 2010. “From V1 to V2 in West Germanic.” Lingua 120 (2):
315–328.
———. 2011. “Rhetorical Relations and Verb Placement in Old
High German.” In Salience: Multidisciplinary Perspectives
on Its Function in Discourse, ed. Christian Chiarcos, Berry
Claus, and Michael Grabski, 173–202. Munich: De Gruyter
Mouton.
Hinterhölzl, Roland. 2009. “The Role of Information Structure in
Word Order Variation and Word Order Change.” In
Information Structure and Language Change: New
Approaches to Word Order Variation in Germanic, ed.
36
Roland Hinterhölzl and Svetlana Petrova, 45–66. Berlin:
Walter de Gruyter.
Kaiser, Georg A. 2002. Verbstellung und Verbstellungswandel in
den Romanischen Sprachen. Tübingen: Niemeyer.
Kaiser, Georg A. 2000. Dialect Contact and Language Change: A
Case Study on Word Order Change in French. Hamburg.
Working Papers in Multilingualism,
Sonderforschungsbereich 538. Universität Hamburg.
Kaiser, Georg A., and Michael Zimmermann. 2011. “On the
Decrease in Subject-verb Inversion in French Declaratives.”
In The Development of Grammar: Language Acquisition
and Diachronic Change: In Honour of Jürgen M. Meisel,
ed. Esther Rinke and Tanya Kupisch, 355–381. Amsterdam-
Philadelphia: John Benjamins Publishing Company.
Krifka, Manfred. 2008. “Basic Notions of Information Structure.”
Acta Linguistica Hungarica 55 (3-4): 243–276.
doi:10.1556/ALing.55.2008.3-4.2.
Kroch, Anthony S., and Beatrice Santorini. 2009. “The
Comparative Evolution of Word Order in French and
English.” Paper presented at the XIth Diachronic Generative
Syntax Conference (DIGS XI), University of Campinas,
Brazil. www.ling.upenn.edu/~kroch/handouts/digs11.pdf
37
Labelle, Marie, and Paul Hirschbuhler. 2005. “Changes in Clausal
Organization and the Position of Clitics in Old French.” In
Grammaticalization and Parametric Variation, ed. Montse
Batllori, Maria-Lluïsa Hernanz, Carme Picallo, and Francesc
Roca, 60–71. New York: Oxford University Press.
Laenzlinger, Christopher 2006. “Le rôle de l’interface syntaxe-
structure informationnelle dans la variation de l’ordre des
constituants dans la phrase.” Nouveaux cahiers de
linguistique française 27: 53–81.
Lipták, Anikó. 2010. “The Structure of the Topic Field in
Hungarian.” In Mapping the Left Periphery: The
Cartography of Syntactic Structures, Vol. 5, ed. Paola
Benincà and Nicola Munaro, 163–198. Oxford: Oxford
University Press.
Marchello-Nizia, Christiane. 2009. “VO Vs V(…)O en français.” In
Historical Linguistics 2007: Selected Papers from the 18th
International Conference on Historical Linguistics,
Montreal, 6-11 August 2007, ed. Monique Dufresne,
Fernande Dupuis, and Etleva Vocaj, 109–121. Amsterdam-
Philadelphia: John Benjamins Publishing Company.
Marchello-Nizia, Christiane. 1995. L’évolution du français: Ordre
des mots, démonstratifs, accent tonique. Paris: Armand
Colin.
38
Martineau, France, Paul Hirschbühler, Anthony S. Kroch, and Yves
Charles Morin. 2010. “Corpus MCVF (parsed Corpus),
Modéliser le changement : Les voies du français”.
Département de français, University of Ottawa. CD-ROM,
first edition.
Nikolaeva, Irina. 2001. “Secondary Topic as a Relation in
Information Structure.” Linguistics 39-1: 1–49.
Platzack, Christer. 1995. “The Loss of Verb Second in English and
French.” In Clause Structure and Language Change, ed.
Adrian Battye and Ian Roberts, 200–226. New York: Oxford
University Press.
Reinhart, Tanya. 1982. “Pragmatics and Linguistics: An Analysis of
Sentence Topics in Pragmatics and Philosophy.”
Philosophica 27: 53–94.
Remberger, Eva-Maria. 2010. “Left Peripheral Interactions in
Romance.” Workshop on“ Focus, Contrast and Givenness
in Interaction with Extraction and Deletion”. Handout .
University of Tübingen, SFB 833-A7, 26./27. March 2010
Rickard, Peter. 1962. “The Word-order Object—verb—subject in
Medieval French.” Transactions of the Philological Society
61(1): 1–39.
Rinke, Esther, and Jürgen M. Meisel. 2009. “Subject-inversion in
Old French: Syntax and Information Structure.” In
39
Proceedings of the Workshop “Null -Subjects, Expletives,
and Locatives in Romance”, ed. Georg A. Kaiser and E.M.
Remberger, 93–130. Arbeitspapier Nr. 123. Fachbereich
Sprachwissenschaft, Universität Konstanz.
Rizzi, Luigi, ed. 2004. The Structure of CP and IP. The
Cartography of Syntactic Structures, Vol. 2. Oxford: Oxford
University Press.
Rizzi, Luigi. 1997. “The Fine Structure of the Left Periphery.” In
Elements of Grammar: A Handbook of Generative Syntax,
ed. Liliane Haegeman, 281–337. Dordrecht: Kluwer
Academic Publishers.
Roberts, Ian G. 1993. Verbs and Diachronic Syntax: A Comparative
History of English and French. Vol. 28. Dordrecht: Kluwer
Academic Publishers.
Rouveret, Alain. 2004. “Pronominal Clitics and the Left Periphery
in Old French.” Bulletin de la Société de linguistique de
Paris 99(1): 181–237.
Skårup, Povl. 1975. Les premières zones de la proposition en
ancien français: essai de syntaxe de position. Études
romanes de l’université de Copenhague 6. Copenhague:
Akademisk Forlag.
Speyer, Augustin. 2010. “Filling the Vorfeld in Written and Spoken
Discourse.” In Discourses in Interaction, ed. Sanna-Kaisa
40
Tanskanen, Marja-Liisa Helasvuo, Marjut Johansson, and
Mia Raitaniemi, 263–290. Amsterdam-Philadelphia: John
Benjamins Publishing Company.
Thurneysen, Rudolf. 1892. “Zur Stellung Des Verbums Im
Altfranzösischen.” Zeitschrift Für Romanische Philologie
16: 289–307.
Vallduví, Enric. 1993. The Informational Component. Philadelphia:
University of Pennsylvania.
Vance, Barbara. 1995. “On the Decline of Verb Movement to Comp
in Old and Middle French.” In Clause Structure and
Language Change, ed. Adrian Battye and Ian G. Roberts,
173–199. Oxford: Oxford University Press.
Zaring, Laurie. 2010. “Changing from OV to VO: More Evidence
from Old French.” Ianua. Revista Philologica Romanica 10:
1–18.
List of Old French texts
Code Approx.
Date
Reference
LEGER 980 Saint Léger. Étude de la langue du manuscrit de Clermont-Ferrand suivie
d’une édition critique du texte avec commentaire et glossaire, éd. par
Joseph LINSKILL, Paris : Droz, 1937, viii + 193 p. Penn Supplement to
MCVF.
ALEXIS 1090 La vie de saint Alexis, éd. par Christopher STOREY, Genève : Droz (TLF,
no 148), 1968, 157 p. Penn Supplement to MCVF.
41
ROLAND 1100 La Chanson de Roland, éd. par Gérard MOIGNET, texte établi d’après le
manuscrit d’Oxford ; traduction, notes et commentaires, 3e édition revue et
corrigée, Paris : Bordas (Bibliothèque Bordas), 1969, 320 p. MCVF.
BRENDAN 1120 The Anglo-Norman Voyage of St. Brendan, éd. par Ian SHORT et Brian
MERRILEES, Manchester : Manchester University Press (Anglo-Norman
Text Society), 1979, vii + 136 p. [BENEDEIT, Le voyage de saint
Brendan]. MCVF.
WILLELME 1150
Lois de Guillaume le conquérant en français et en latin, éd. par John E.
MATZKE, textes et étude critique, préface historique par Gh. Bémont,
Paris : Picard et fils, 1899, 33 p. [Les Leis Willelme] (dates betw. 1150 &
1170). MCVF.
QLR 1170 Li Quatre Livre des Reis, éd. par Ernst Robert CURTIUS, Dresde/, Halle :
Max Niemeyer (Gesellschaft für Romanische Litteratur, no 9), 1911, xcv +
244 p. Supplement to MCVF.
YVAIN 1177 Chrétien de Troyes, Le Chevalier au lion (Yvain), éd. par Mario ROQUES,
Paris : Champion (CFMA, no 89), 1960, xxx + 266 p. MCVF.
LAIS 1180 MARIE DE FRANCE, Les Lais, éd. par Jean RYCHNER, Paris : Honoré
Champion (CFMA, no 93), 1981 [1971] , xlv + 317 p. MCVF.
CHIEVRES 1194 Maurice-Aurélien ARNOULD, « Le plus ancien acte en langue d’oïl : la
charte-loi de Chièvres (1194) [Rasse VIII de Gavre et Nicolas IV de
Rumigny] », dans Hommage au professeur Paul Bonenfant (1899-1965).
Études d’histoire médiévale dédiées à sa mémoire... Bruxelles. 1965,
p. 85-118. MCVF.
AUCASSIN 1200 Aucassin et Nicolette : chantefable du XIIIe siècle, éd. par Mario
ROQUES, Paris : É. Champion, 1936, 105 p. MCVF.
CLARI 1205 Robert DE CLARI, La Conquête de Constantinople, éd. par Philippe
LAUER, Paris : Champion (CFMA), 1924, 130 p. MCVF.
PSEUDO-
TURPIN
1220 Theodor M. AURACHER « Der Sogenannte poitevinische Übersetzung
des Pseudo-Turpin », dans Zeitschrift für romanische Philologie
(Tübingen; éditeur : Gustav Gröber), t. I, 1877, p. 259-336. MCVF.
QUESTE 1225 La Queste del Saint Graal, roman en prose du XIIIe siècle, éd. par Albert
42
PAUPHILET, Paris : Champion (CFMA, no 33), 1923, 301 p. MCVF.
AGNES 1226 Anonyme. Sermon anonyme sur Sainte Agnès, texte du 13 siècle. ed.
Robert Taylor. Strasbourg : Centre de philologie et de littérature romanes,
1969. MCVF.
CASSIDORUS 1267 Livre de Cassidorus empereur de Constantinoble. Le roman de Cassidorus,
publié par Joseph Palermo, Paris, Picard (Société des anciens textes
français). Supplement to MCVF
SOMME 1279 FRERE LAURENT, La Somme le roi, éd. par Édith BRAYER et Anne-
Françoise LABIE-LEURQUIN, Paris/Abbeville : Société des anciens
textes français / Paillart, 2008, 596 p. et un cahier d’illustrations en
couleurs. Penn Supplement to MCVF.
ROISIN 1283 Le Livre Roisin : coutumier lillois de la fin du 13e siècle, publié avec une
introduction et un glossaire par Raymond MONIER, préface d’Alexandre
de Saint-Léger, Paris : Domat-Montchrestien (Documents et Travaux
publiés par la Société d’Histoire du Droit des pays flamands, picard et
wallons, no 2), 1932, xxxv + 175 p. MCVF.
JOINVILLE 1309 JOINVILLE, Jean sire de, Vie de saint Louis, éd. bilingue par Jacques
MONFRIN, Paris : Dunod (Classiques Garnier), 1995, cxxxixi + 485 p.,
MCVF.
1 References to OF examples contain first the approximate date of
the text, then the abbreviated name of the text, followed by a page,
verse, or section number (according to the text), followed by the
phrase structure number in the corpus.
43
2 In the present paper, as in Rinke and Meisel (2009), postverbal
subjects include subjects described as of the Germanic as well as of
the Romance inversion type.
3 The focal nature of the preverbal contrastive constituent comes out
clearly in (1), a sentence with a null subject:
(1) Del tranchant, non mie del plat, le fiert ...
with-the cuting not a-bit with-the flat, him hits
‘with the cutting edge, not with the flat side, (he) hits him’
(1177 ; Yvain, 128.4433)
4 Zaring (2010) studied the OV order with (non-initial) non-finite
verbs, in two different texts (Le Roman de Perceval ou le Conte du
Graal by Chrétien de Troyes (c. 1185) and La Conqueste de
Constantinople by Geoffroy de Villehardouin (c. 1205)), and found
the same evolution. It thus seems that what changes is not the nature
of the clause-initial position but more generally what allows an
object to precede the verb.