EPA02403 Finno Ugric Lang 2012 01-02 067-081
EPA02403 Finno Ugric Lang 2012 01-02 067-081
The study presents experimental findings on new information subjects in Finnish. The
main answering strategies that emerge in the collected data are discussed in light of
recent studies within the cartographic framework (Belletti 2001, 2004, 2005). In null
subject languages subject inversion is typically adopted in contexts in which the subject
is new information. Conversely, in non null subject languages other strategies emerge,
such as in situ focalization (e.g. English) and cleft sentences (e.g. French). Finnish is
particularly interesting for its nature of partial null subject language (Holmberg et al.
2009). The unavailability of VS structures of the type observed in null subject languages
and the presence of XPVS structures can be accounted for assuming that Finnish does
not have a referential pro and the EPP can be satisfied by other lexical elements.
1 Introduction
Null Subject Languages (NSL) such as Italian typically adopt subject-verb inversion,
resulting in the VS order, when answering with a full clause to questions concerning the
identification of the subject of the clause. In contrast, in Non Null Subject Languages
(NNSL) such as e.g. French and English VS structures are excluded and other types of
answers are typically adopted: (reduced) clefts and SV structures (in situ focalization
henceforth), in which the subject is associated with a particular intonation (Belletti 2001,
2004, 2008, Belletti, Bennati & Sorace 2007). (1)-(3) illustrate the equivalent question-
answer exchanges in Italian, English and French, respectively.
∗
I am especially grateful to the anonymous reviewers of FULL whose detailed and insightful
comments have considerably improved this contribution. I also thank all the participants at the 11th
International Congress for Finno-Ugric Studies in Pilicsaba, where this paper was first presented.
Finno-Ugric Languages and Linguistics Vol. 1. No. 1–2. (2012), 67–81. http://full.btk.ppke.hu
ISSN: 2063-8825
68 New Information Subjects in Finnish: an Experimental Study
Burzio 1986 and subsequent literature). The phenomenon is often referred to as Free
Inversion2 (henceforth FI, cf. Belletti 2001, 2004). The recent studies quoted above have
proposed that discourse factors are also highly relevant in the distribution of FI/VS in
NSLs. This kind of inversion is thus not ‘free’ in the sense that it can occur freely since it
is discourse-related; it is typically adopted in contexts as (1) where the subject is
interpreted as the subject of new information (Belletti 2001, Belletti 2004b, Belletti,
Bennati & Sorace 2007). Belletti (2001, 2004) has proposed that a low vP-peripheral
focus position is present in the clause structure, which hosts new information subjects in
NSLs (e.g. Italian). As mentioned, recent work on answering strategies (cf. Belletti 2009)
has shown that this position is not made use of in the same way in NNSLs, which thus
adopt different structures to focalize the new information subject (as in (2)-(3) above).
Consequently, a relation exists between the possibility to instantiate FI/VS, with the
subject interpreted as new information, and the null subject nature of the language (see
also Belletti & Shlonsky 1995, Hulk & Pollock 2001, Kayne 2005, Sheehan 2010 a.o. for
discussion on inversion in Romance).
Updating the terminology of the traditional account (e.g. Rizzi 1982) it is assumed
that a small pro satisfies the relevant EPP property of the relevant high subject position of
the clause, thus yielding VS.
In a crosslinguistic perspective, it has been observed that the strategies adopted in
different and unrelated languages3 fall within the alternatives illustrated in (1)-(3) (Belletti
2009). In recent studies (Holmberg et al. 2009, Holmberg & Sheehan 2010) it is proposed
that Finnish has a special status with regard to the Null Subject Parameter: it is a Partial
Null Subject Language (PNSL), cf. also Huang (2000) for a classification of PNSLs or
semi NSLs. It allows 1st and 2nd person null subjects but it does not allow 3rd person null
subjects, except in some embedded clauses with the null subject ‘controlled’ by a higher
argument; Finnish also does not allow verb-initial impersonal clauses.4
The present work presents an experimental study which reports novel findings
from an oral elicitation task which tests the use of new information subjects in L1
Finnish. The research questions which are addressed are: (i) What answering strategies
are available in Finnish in contexts in which the subject is a new information subject? (ii)
Does Finnish make use of the dedicated vP-internal focus position, which in the
approach referred to above and adopted here, hosts new information subjects? The
main results tell us that in a PNSL like Finnish new information subjects can be generally
interpreted as such in preverbal position and no FI/VS is instantiated. However, the
possible activation in Finnish of the vP peripheral new information focus position will be
discussed in the light of a different word order, XPVS, which has also emerged in the
collected data. It will be proposed that the Finnish XPVS order instantiates a different
way to satisfy the EPP property of the high subject position of the clause, different from
the one characteristically exploited in a NSL (e.g. Italian through small pro).
2 Free Inversion has different properties from other inversion structures in Romance languages,
such as French Stylistic Inversion (Kayne & Pollock 1978, 2001) discussed in Belletti (2001, 2004b) in
comparison to FI.
3 Belletti (2009) discusses data coming from several typologically different and diachronically
unrelated languages such as Italian, English, French, European Portuguese, Romanian, Paduan,
Japanese, Norwegian, Malayalam, German, Hungarian, Basque, Gungbe.
4 There are some exceptions to this general pattern, as discussed in Holmberg (2005, 2009) and
Holmberg & Nikanne (2002). These exceptions are not crucial for the present discussion and will not
be discussed here.
Lena Dal Pozzo 69
2 Theoretical background
Following Belletti (2001, 2004b), we assume, along the lines of the cartographic approach
(Rizzi 1997, 2004, Cinque 2002), that a new information postverbal subject is located in
the Specifier of a low focus position, a dedicated position for new information elements
(Belletti 2001, 2004b). A clause-internal vP periphery with a FocusP surrounded by Topic
projections is identified, as in (5a), parallel to the fine-grained clausal left periphery5
assumed by Rizzi (1997, 2004).
The analysis proposed by Belletti (2001, 2004b) in (5b) assumes that the new
information subject moves to the low dedicated position, where it is interpreted, while a
silent pro fills the preverbal subject position in order to satisfy the clausal EPP feature. As
discussed in the Introduction, the FI/VS order is constrained by discourse factors, so
that typically in sentences such as (5b) the subject is interpreted as a new information
subject.
Under this approach the traditional idea that a relation between the preverbal pro
and the postverbal subject holds, is revisited as follows: Belletti (2005) assumes a
doubling derivation inspired by the analysis in Sportiche (1988) for floated quantifiers
(see also Torrego 1995, Kayne 1994, Rouveret 1989 for different accounts in the same
vein) common to various structures such as clitic doubling, clitic left/right dislocation,
and so-called strong pronoun doubling. All these structures exhibit two nominals with
the same thematic role and case. It is proposed that in sentences like (6) illustrating the
strong pronoun doubling construction a Big DP exists in which both the functional
word, the pronoun, and the lexical noun phrase are merged.
Belletti (2005) proposes that FI/VS structures can be derived in a similar way, as
illustrated in (7):
5The cartographic approach assumes a detailed architecture of the clause composed of distinct
functional heads and their corresponding projections which are directly visible to the interpretative
systems. According to Rizzi (1997, 2004) the left periphery of the clause has as a structure along the
following lines: [ForceP [TopP [IntP [TopP [FocusP [TopP [FinP [IP … ]]]]]]]]
70 New Information Subjects in Finnish: an Experimental Study
(7) IP
ty
proi I'
ty
verràk …
comes FocP
ty
[ti [DP2 Gianni ]]j Foc'
John ty
…
vP
ty
tj v'
ty
tk
In this configuration pro and Gianni are generated in the Big DP, pro moves to a
nominative case position and Gianni is stranded in the vP-peripheral new information
focus position. Nominative case-marking of the postverbal subject is a consequence of
the doubling computation. According to this proposal, the EPP feature triggers the
movement of the part of the Big DP corresponding to pro.6
The assumption is, following Belletti (2005) that in the doubling derivation of VS
structures in FI/VS pro is a silent personal (referential) pronoun sharing features with the
postverbal lexical subject by virtue of their relation in the Big DP, rather than an
expletive pronoun.
Languages such as Brazilian Portuguese (BP) have progressively lost the null-
subject property with referential subjects. BP has also lost (free) subject inversion
structures. The analysis summarized assumes that these two properties are correlated and
they can be explained under the doubling derivation according to which a referential pro
is present in the preverbal position, as illustrated above.7
6 See Holmberg (2005, 2010) for a different refinement of the traditional analysis (Rizzi 1982) on
NS couched within the MP in terms of definiteness. The feature [+ referential] of Rizzi is replaced by
a [uD]-feature (a definiteness feature that is present in I with which a null subject enters in an Agree
relation); languages divide into those that have such feature in I, hence allowing null subjects (a
deficient pronoun phi-P in consistent NSLs) and those that do not have this feature (non NSLs or
PNSLs).
7 The present account does not exclude the possibility of having an expletive rather than a
Doubling configuration, as pointed out by an anonymous reviewer. Biberauer (2010) a.o. extensively
investigates, within the MP, expletives in NSL and a subset of PNSLs is discussed with regard to the
nature of Spec,T. A categorical difference between expletives in NSLs and NNSLs is proposed as
well. In the same spirit, Roberts (2010) suggests an alternative analysis to the Rizzian one for null
subjects in consistent NSLs based on Holmberg (2005), cf. also Holmberg (2010).
Lena Dal Pozzo 71
From the above we conclude that having pro is a necessary but not sufficient
condition for instantiating Verb-Subject structures.8 A straightforward consequence is
that only NSLs, which can have a silent pro in the canonical subject position, allow for
the kind of movement involved in free inversion structures and NNSLs typically use
other kind of structures in context in which the subject is new information, as
exemplified in (2)-(3). As for PNSL the question arises as to whether new information
subjects are implemented in the NNSL or in the NSL fashion. In Guesser (2007) the
same experimental design first used by Belletti & Leonini (2004) and adapted to Finnish
in the present study has been adapted to Brazilian Portuguese, another language classified
as a PNSL in Holmberg et al. (2009). The results show that in BP different strategies are
adopted to focalize the new information subject: SV, clefts (which include reduced clefts,
pseudo-clefts and truncated clefts, see Guesser 2007), while VS structures of the type
observed in a NSL like Italian seen above were not observed. This supports the idea that
referential pro is a necessary condition for instantiating VS structures as proposed by
Belletti (see section 2) and adopted in Guesser (2007). The present experimental study
aims at testing whether this is true for another PNSL, namely Finnish.
Crucially, if an expletive pro were sufficient to allow for FI/VS, we would expect
FI/VS also in PNSLs such as Finnish and Brazilian Portuguese (BP) as these languages
have an expletive pro (cf. Duarte 2000, Figueiredo-Silva (1996); Rodrigues (2002), Ferreira
(2004), Guesser (2007), Modesto (2008) for discussion and analysis on null subjects in BP
and Holmberg & Nikanne 2002, Holmberg 2005 on overt and null expletives in Finnish).
Given the similarities between BP and Finnish (they do not have referential pro and both
have a null expletive), we expect that similarly to what observed for BP (Guesser 2007)
no FI/VS structures are available in Finnish. The results (cf. section 5) will show us that
things are rather more complicated.
The experimental task that was used in the present study was first created by Belletti &
Leonini (2004) and then also used by Belletti, Bennati & Sorace (2007) to test the
acquisition and use of postverbal subjects in Italian L2. Adaptations of the experimental
task in different languages have been done to collect data in L1 Brazilian Portuguese
(Guesser 2007), L1 Croatian (Kras 2010) and L1 Finnish (in the present study).
Noticeably, the same design and materials were used in all studies.
The experimental task consists of 22 short videos in indoor settings with female
and male subjects (see Appendix I for tokens of the task). It aims at creating the ideal
discourse-pragmatic conditions for question-answer pairs in which the subject is new
information focus. The task was presented individually to the participants through a
Power Point presentation, the videos were the same for all the participants and they were
not randomized. Using the same experimental design crosslinguistically maintains
8 This is also attested by data on languages that have a positive setting for the null subject
parameter but do not allow verb-subject structures, e.g. Bantu languages (Nicolis 2005). Also studies
in second language acquisition strengthens the dissociation between the availability of pro and that of
postverbal subjects: Belletti and Leonini (2004) and Belletti, Bennati and Sorace (2007) investigated
the use of null subjects and postverbal subjects in contexts in which the subject is new information in
Italian L2. In both studies it is shown that a target use of null subjects is not correlated to a target use
of postverbal subjects.
72 New Information Subjects in Finnish: an Experimental Study
4 Results
2010, Guesser 2011) it seems that in Finnish a vP-peripheral focus position is extensively activated in
these structures. The issue is left for future research.
12 The following abbreviations are used for grammatical cases: ACC=accusative; NOM=
nominative; INE= inessive. Following common practice PAST is used to indicate past tense and SG for
singular.
Lena Dal Pozzo 73
Table 1 provides the total amount of answers classified for verb type and type of
answers:
a)
Verb class SV VS O(DP)VS O(pr)VS SO(pr)V Cleft R. Cleft Exist. cleft Tot.
Trans. 82% 0% 2% 8% 0,3% 3,8% 4,2% 0,0% 100%
Unacc. 88% 7% 0% 0% 0% 2,3% 2,3% 0,0% 100%
Unerg. 84,9% 4,6 0% 0% 0% 3,9% 3,9% 2,6% 100%
b)
Verb class SV VS O(DP)VS O(pr)VS SO(pr)V Cleft R. Cleft Exist. cleft Tot.
Trans. 234 0 5 24 1 11 12 0 287
Unacc. 38 3 0 0 0 1 1 0 43
Unerg. 129 7 0 0 0 6 6 0 152
Table 1: Total amount of answers for verb type
100%
SV
80% AdvVS
O(DP)VS
60%
O(pr)VS
40% SO(pr)V
20% cleft
reduced cleft
0%
exist.cleft
Trans. Unacc. Unerg.
As expected, FI/VS of the type found in NSL languages are not observed under
the same discourse-pragmatic circumstances. The result is consistent with elicited data
collected through the same experimental design in BP (Guesser 2007, Dal Pozzo &
Guesser 2011) and it supports the assumption that a referential (3rd person) null subject,
which neither BP nor Finnish have, is a required condition to instantiate FI/VS
structures in addition to adequate (new information) discourse conditions, which were
controlled for by the contexts of the elicitation task.
I suggest that the SV(O) order in Finnish is an instance of in situ focalization, a
subject focalization strategy to which typically NNSLs such as English resort (see
references quoted), with the derivation in (12). The subject is in its canonical preverbal
subject position13 (Spec, FinP according to Holmberg & Nikanne 2002 and as assumed in
Kaiser 2006), in which it is focalized as new information.
Assuming the representation in (11) for SV(O) with S new information, apparently
supports the idea that Finnish can be assimilated to NNSLs: the new information subject
is focalized in its canonical preverbal position and no FI/VS (nor activation of the
dedicated vP-peripheral focus position) emerges. However, SV(O) is not the only
strategy that emerges in our data. The second quantitatively relevant strategy consists of
the XPVS order, in which we postulate that the low vP-peripheral position dedicated to
new information elements is activated, as discussed in the following section.
4.1 XPVS
The XPVS order is attested in 10% (28/287) of the total amount of answers with
transitive verbs, resulting in OVS, and in 5,1% with unergative and unaccusative verbs,
resulting in AdvVS. At the discourse level, XPVS is possible when XP is a topic in the
sense of known/given information and S is new information (cf. Vilkuna 1995,
Holmberg & Nikanne 2002).
Turning the discussion to the OVS order, syntactically at least two alternative
analyses come to mind: (i) OV is first obtained by topicalization of the object in the low
part of the clause and then the OV chunk is fronted into the left periphery, as in (11),
and S is in the same preverbal position as in (10); (ii) as consistent NSLs like Italian, the
new information subject is in the low vP-peripheral focus position, as in (5) and repeated
in (12).14
13 Cardinaletti (1997, 2004) identifies a number of subject positions in the preverbal field which
are assumed to be quite uniform across languages. In mapping the IP at least two different positions
are identified (Spec,AgrSP for the syntactic subject and Spec,SubjP for the the subject of predication).
In the present work we abstract away from the discussion.
14 At first sight another alternative consists of assuming a structure parallel to V2 languages, as in
(i):
(i) [CP O1 V2 [TP S3 [t1 t2 t3 ]]]
This is, however, immediately falsified by examples such as (13), where the subject is preceded by
auxiliary verb and main verb, and other sentential material.
Lena Dal Pozzo 75
The analysis in (11) is soon ruled out by word order facts. As a matter of fact, sentences
such as (13a-b) show us the impossibility of such a representation for XPVS orders.
Postulating movement of the OV chunk to a topic position in the clausal domain with
the subject in the preverbal position would exclude having Aux or Neg Aux between O
and V. These are nevertheless grammatical sentences15, (14) illustrates the basic (neutral)
word order.
A better way to account for these structures comes from the alternative analysis outlined
in (ii) above, which assumes that the vP-peripheral focus position is activated. I suggest
that this position is where the new information subject is located in XPVS structures. As
said earlier, Finnish does not have a referential pro which could satisfy the EPP. If we
assume that the EPP can also be satisfied by other lexical elements (see Holmberg 2010),
in the Finnish XPVS structure it is the XP element that satisfies the EPP16. This is
reminiscent of Holmberg & Nikanne’s (2002) proposal of Finnish as a topic prominent
language. Consequently, the orders in (13) can be derived by assuming movement of the
object to the preverbal EPP position. Fronting of the object in the left-periphery is also
correlated to discourse factors: in (13a-b) the object can be interpreted as known/given
or contrastive/corrective (depending on the intonation).
Hence, we can formulate the following:
15 I thank an anonymous reviewer for raising the issue and suggesting a possible way to account
1997). Locative Inversion typically occurs with intransitive verbs which take a locative argument, as
represented in (i) for English and in (ii) for the equivalent in Finnish:
(i) In the corner sat a man
(ii) Nurkassa istui mies
corner.in sat man
76 New Information Subjects in Finnish: an Experimental Study
Notice that (15) is intended under the discourse contexts in which the subject is new
information, as discussed earlier. Notice also that this is a tentative generalization and a
more extensive discussion based on data from different PNSLs is left for future
research17.
Thus, the assumption put forth by Belletti (see references quoted above) can be
further developed in the following way:
(16) Only full Null Subject Languages allow for FI/VS in contexts of (new information)
subject focalization. Non Null Subject Languages typically adopt different
strategies such as in situ focalization (English) and cleft strategies (French, Brazilian
Portuguese). Partial Null Subject Languages such as Finnish can have a “mixed
pattern” consisting of in situ focalization and focalization of the new information
subject in the vP-peripheral postverbal position through a different way to satisfy
the EPP.
In conclusion, this section examined two possible ways to account for new
information subjects in Finnish: in situ focalization and activation of the dedicated focus
position in the vP periphery. Most importantly, postulating in situ focalization (similar to
NNSLs such as English) for SV(O) structures is not in contraposition with an activation
of the vP-peripheral focus position in XPVS structures. Moreover, this position
dedicated to new information elements seems to be active also in the cleft structures that
emerged in the data (see fn. (12)). Hence, PNSLs such as Finnish (and BP) seem to have
a wider set of possible strategies to adopt, than NSLs and NNSLs, under the discussed
discourse contexts.
5 Final remarks
The present study aimed at investigating within the cartographic framework whether
Finnish, which is assumed to be a partial null subject language, allows for the subject-
17 An open question arises from sentences such as (i) and (ii): FI seems to be excluded in Finnish
also with first and second person, which can always have a silent subject pronoun. The equivalent
sentences are pragmatically appropriate in the same contexts in Italian.
(i) a. Kuka puhui?
who spoke?
b. #Puhuin minä
spoke I
(ii) a. Kuka siellä (on)?
who there (is)
b. #Olen minä
is I
’It’s me’
Lena Dal Pozzo 77
verb inversion analyzed as free inversion, which typically involves a low vP-peripheral focus
position, in contexts in which the subject is a new information subject. Moreover, it
investigated which other word orders might be exploited in the contexts discussed.
The research is based on the observations coming from previous theoretical and
empirical research that propose that a positive setting of the null subject parameter is
necessary to instantiate free inversion in NSLs such as Italian in contexts in which the
subject is a new information subject. Even though Finnish is almost fully a null subject
language it does not allow for VS structures of the kind observed in NSLs (examples (1)
and (5)). It was observed how the data from Finnish compares with the results from the
Brazilian Portuguese adaptation of the experiment. In BP, which is also considered a
PNSL (Holmberg et al. 2009), the typical strategies in the contexts of the experimental
task consist of Subject-Verb structures and clefts of various type (cleft, reduced clefts and
truncated clefts, Guesser 2007). The unavailability of FI/VS in Finnish and BP is a direct
consequence of the absence of referential null subjects in these languages. If an expletive
pro were sufficient then instances of FI of the type observed in NSLs such as Italian
could have been observed in the collected data.
The empirical data are novel and they have been collected using the same
methodology as previous studies on the topic. The data show that the preferred
answering strategy in Finnish is SV(O), which also represents the canonical word order.
Hence, Finnish does not show instances of subject focalization through a FI strategy of
the kind observed in NSL such as Italian in contexts in which the subject is new
information focus (Belletti 2001, 2004, 2009). However, even though FI/VS of the NSLs
type is excluded, it is proposed that the observed XPVS strategy involves the activation
of the vP-peripheral focus position dedicated to new information subjects. Since Finnish
does not have a referential pro that could satisfy the EPP, it is assumed that also other
lexical elements can satisfy the EPP in a language like Finnish, yielding the XPVS
structure found in the data. Hence, in Finnish a postverbal subject is possible only when
there is an overt element in the preverbal sentence-initial position, namely an object for
transitive verbs, an adverb/adverbial for intransitive verbs.
78 New Information Subjects in Finnish: an Experimental Study
Scene 1: The phone rings. The young woman answers and talks with a friend of hers. A
young man enters the room and asks who has called.
Question: Who has called? (target item)
Further questions:
1) Who has answered? (target item)
2) What she was doing when the phone rang? (filler item)
Lena Dal Pozzo 79
References
Guesser, Simone. 2011. La sintassi delle frasi cleft in Portoghese Brasiliano. PhD Dissertation.
University of Siena.
Guesser, Simone. 2007. Il soggetto nullo incassato in Portoghese Brasiliano: problemi
legati al movimento a posizioni tematiche. Rivista di Grammatica Generativa: 39-64.
Holmberg Anders & Nikanne Urpo. 2008. Subject doubling in Finnish: the role of
deficient pronouns. In S. Barbiers, O. Koeneman, M. Lekakou (eds.). Microvariations
in Syntactic Doubling. Bingley, Emerald, 325-349.
Holmberg, Anders & Nikanne Urpo., 2002. Expletives, Subjects and Topics in Finnish.
In P. Svenonius (ed.). Subjects, expletives and the EPP. Oxford: Oxford University
Press.
Holmberg, Anders & Sheehan, Michelle., 2010. Control into finite clauses in partial null-
subject languages. In T. Biberauer, A. Holmberg, I. Roberts, M. Sheehan (eds.).
Parametric Variation: Null Subjects in Minimalist Theory. Cambridge: Cambridge
University Press.
Holmberg, Anders, Nayudu, Aarti, & Sheehan, Michelle. 2009. Three partial null-subject
languages: a comparison of Brazilian Portuguese, Finnish and Marathi. Studia
Linguistica 63:1. Wiley-Blackwell Publishing.
Holmberg, Anders. 2002. Deriving OV order in Finnish. In P. Svenonius, (ed.). The
Derivation of VO and OV. Amsterdam: Benjamins.
Holmberg, Anders. 2005. Is there a little pro? Evidence from Finnish. Linguistic Inquiry
36:4, 533-564.
Holmberg, Anders. 2010. Null Subject Parameters. In T. Biberauer, A. Holmberg, I.
Roberts, M. Sheehan (eds.). Parametric Variation. Cambridge: Cambridge University
Press.
Huang, Yan. 2000. Anaphora. A cross-linguistic study. New York: Oxford University Press.
Hulk, Aafke C. & Jean-Yves Pollock. 2001. Subject positions in Romance and the theory
of Universal Grammar. In A. Hulk and J-Y. Pollock (eds.). Subject inversion in
Romance and the theory of Universal Grammar, 107-162. Oxford: Oxford University
Press.
Jaeggli, Osvaldo. 1982. Topics in Romance Syntax. Dordrecht: Foris.
Johns, Christopher. 2007. Interpreting agreement. Durham University dissertation.
Kaiser, Elsi & Trueswell, John. C. 2004. The role of discourse context in the processing
of a flexible word order language. Cognition 94:113-147.
Kaiser, Elsi & Trueswell, John. C. 2006. Negation and the left periphery in Finnish.
Lingua 116:314-350.
Kayne, Richard. 1994. The Antisymmetry of Syntax. Cambridge: MIT Press.
Kayne, Richard. 2005. Movement and silence. New York: Oxford University Press.
Kayne, Richard., Jean-Yves. Pollock 2001. New thoughts on Stylistic Inversion. In (eds.)
A. Hulk and J-Y. Pollock Subject inversion in Romance and the theory of Universal
Grammar. Oxford: Oxford University Press.
Kayne, Richard & J-Y. Pollock. 1978. Stylistic inversion, Successive Cyclicity, and Move
NP in French. Linguistic Inquiry 9:595-621.
Kras, Tihana. Testing new information subjects: Croatian version of Belletti & Leonini
2004. manuscript. University of Edinburgh.
Modesto, Marcello. 2008. “Topic prominence and null subjects. In T. Biberauer (ed.) The
limits of syntactic variation, 375–409. Amsterdam: John Benjamins.
Rizzi, Luigi. 1982. Issues in Italian Syntax. Dordrecht: Foris.
Rizzi, Luigi. 1997. The Fine Structure of the Left Periphery. In L. Haegeman (ed.)
Elements of Grammar. Dordrecht: Kluwer Publications.
Lena Dal Pozzo 81
Rizzi, Luigi. 2004. Locality and Left-Periphery. In Belletti, A. (ed.) Structures and Beyond.
The cartography of Syntactic Structures, vol. 3, 223-251. New York: Oxford
University Press.
Rodrigues, Cilene. 2002. Morphology and null subjects in Brazilian Portuguese. In D.
Lightfoot (ed.), Syntactic Effects of Morphological Change, 160-178. Oxford: Oxford
University Press.
Rouveret, Alain. 1989. Cliticisation et temps en portugais européen. Revue des Langues
Romanes 2:337-371.
Sheehan, Michelle. 2010. Free’ inversion in Romance and the Null Subject Parameter. In
T. Biberauer, A. Holmberg, I. Roberts & M. Sheehan (eds.). Parametric Variation:
Null Subjects in Minimalist Theory, 231-262. Cambridge: Cambridge University Press.
Sportiche, D., 1988. A Theory of Floating Quantifiers and its Corollaries for Constituent
Structure. Linguistic Inquiry 19:425-449.
Sportiche, Dominique. 1996. Clitic Constructions. In J. Rooryck and L. Zaring (eds.).
Phrase Structure and the Lexicon, 213– 276. Dordrecht: Kluwer.
Sportiche, Dominique. 1998. Partitions and Atoms of Clause Structure: Subjects, Agreement, Case
and Clitics. London/New York: Routledge.
Torrego, Esther. 1995. On the Nature of Clitic Doubling. In H. Campos and P.
Kempchinsk (eds.) Evolution and Revolution in Linguistic Theory, 399-418. Washington
DC: Georgetown University Press.
Vilkuna, Maria. 1995. Discourse Configurationality in Finnish. In (ed.) É.K. Kiss,
Discourse Configurational Languages. Oxford & New York: Oxford University Press.
Vilkuna, Maria. 1989. Free Word Order in Finnish. Its Syntax and Discourse Functions.
Helsinki: Suomalaisen Kirjallisuuden Seura.