PreprintPDF Available

Clitic Climbing without Restructuring in Czech and Polish

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract

Clitic climbing, i.e. the realization of one or more clitics in a syntactic constituent hierarchically higher than the clitics’ licencing predicate, has been accounted for in terms of a restructuring approach. The embedded infinitive the clitics are extracted from has been assumed to be structurally deficient – that is, a bare VP. Due to the lack of projections above the lexical V-head, clitics escape the infinitival domain to get their morphosyntactic features licenced in the matrix clause. However, the predictions of the restructuring approach do not withstand a corpus linguistic examination and are falsified by empirical data. Clitic climbing cannot be adequately accounted for by syntax proper and alternative accounts have to be taken into consideration seriously. It will be proposed to refer to the notion of information structure as a feasible explanatory account of clitic climbing.
version 30th Sep 2021
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Clitic Climbing without Restructuring in Czech and Polish*
Irenäus Kulik
Abstract
Clitic climbing, i.e. the realization of one or more clitics in a syntactic constituent hierarchically higher than the
clitics’ licencing predicate, has been accounted for in terms of a restructuring approach. The embedded infinitive
the clitics are extracted from has been assumed to be structurally deficient that is, a bare VP. Due to the lack of
projections above the lexical V-head, clitics escape the infinitival domain to get their morphosyntactic features
licenced in the matrix clause. However, the predictions of the restructuring approach do not withstand a corpus
linguistic examination and are falsified by empirical data. Clitic climbing cannot be adequately accounted for by
syntax proper and alternative accounts have to be taken into consideration seriously. It will be proposed to refer to
the notion of information structure as a feasible explanatory account of clitic climbing.
1. Introduction
Clitic Climbing (CC) is the realization of a pronominal or reflexive clitic in a syntactic
constituent hierarchically higher than the licensing predicate. Junghanns (2002a: 66)
schematizes CC as in (1a), whereby a constituent α embeds a constituent β. CC is analysed as
movement of the clitic (CL) from β to α. (1b) paraphrases the scheme in a theory-neutral way.
1
The gap e co-indexed with CL captures the fact that CL is linearized in α, but subcategorized for
by the verbal predicate in β. In principle, an arbitrary number n of phrase boundaries may
intervene between α and β (with {0} denoting the set of natural numbers including zero).
(1) a. [α CL … [β … tCL … ]]
b. [α CLj [n [β ej ]]] , n {0}
* Different versions of this paper were presented at FDSL-14 (Leipzig, 24 Jun, 2021) and SLS-16 (Urbana-
Champaign, 35 Sep, 2021). I am grateful to Uwe Junghanns, Hagen Pitsch, Luka Szucsich, Peter Kosta, Steven
Franks, and Jacek Witkoś for discussion and suggestions. I am much indebted to Martina Berrocal, who kindly
helped me with the Czech data. I kindly thank Susanne Wurmbrand, who provided me with information on
recent developments in the study of restructuring phenomena. Finally, I thank an anonymous reviewer for
his*her comments. All improvements are thanks to the aforementioned individuals, while all shortcomings
remain in my sole responsibility.
1
I will attempt to phrase my arguments in a theory-neutral way, in order not to impose a specific approach on
the reader. For instance, I will not opt for a particular analysis of long-distance dependencies in terms of e.g.
movement or copy-and-delete.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
There is an extensive body of research literature on CC in Romance, which significantly
inspired research on Slavic. In the Italian (Ital) example (2a), the direct object clitic lo ‘him-
ACC’ follows the embedded infinitive legger(e) ‘read’ it is argument of. This is the local or in
situ realization of the clitic.
2
CC is found in (2b) with the object clitic being realized before the
finite verb of the matrix phrase, which will be also referred to as non-local placement.
(2) a. Martina vuole1 legger2 =lo2.
3
(in situ/local)
Martina want-PRS.3SG read-INF =him-ACC
b. Martina lo2 vuole1 leggere2. (CC/non-local)
‘Martina wants to read it.’ (Ital; Spencer & Luís 2012: 163164)
The remainder of the paper is organized as follows: Section 2 briefly comments on the syntactic
status of clitic pronouns in West Slavic. Section 3 provides a concise overview of basic clitic
climbing properties in West Slavic. Section 4 is the core of the paper and tests the correlates of
restructuring empirically focusing on accusative case licencing (4.1), the absence of an
underlying subject in the infinitive phrase (4.2), the dependence of the infinitive’s temporal
reference upon the matrix verb’s tense information (4.3), and the all-or-nothing quality of clitic
climbing (4.4). Section 5 addresses the role of information structure for clitic climbing.
Concluding remarks are given in section 6.
2. The Status of West Slavic Clitics
It is common to distinguish Polish (Pol) and Czech (Cz) clitics along the lines of Zwicky’s
(1977) simple/special-clitic dichotomy. The second position clitics in Czech are considered
special clitics, whereas Polish clitics being distributed rather freely are not. However, the often-
made statement that Polish clitics are typologically peculiar in comparison to clitics in other
West and South Slavic languages, turns out to be controversial on closer inspection.
4
Therefore,
2
Glossing follows the Leipzig Glossing Rules in a simplified form. Only those linguistic expressions are
glossed, which are immediately relevant for the issues under discussion.
3
Clitics will be highlighted in boldface. I adopt the integer-index-convention from Hana (2007), Rosen (2014),
and Hansen et al. (2018) to indicate the structural hierarchy between the verbal heads as well as the
subcategorization relations between a verbal predicate and its dependents.
4
On the one hand, Rappaport (1988), Dziwirek (1998: 89), Kupść (2000), Borsley & Rivero (1994), Franks
(2009, 2010), and Franks & King (2000) treat Polish as a language without second position clitics hence, not
possessing special clitics. On the other hand, Rothstein (1993), Birnbaum & Molas (2012), Urbańczyk (1976),
Damerau (1992: 102), Veselovská (1995: §4.8 fn. 23), and Dimitrova-Vulchanova (1999: 85) consider Polish to
be a second position clitic language essentially. Spencer (1991: 390) regards Polish as a special clitic language.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Czech and Polish clitics will be treated alike throughout the paper. I restrict myself to the set of
unambiguous short pronominals that has been recognized for Czech and Polish independently
and which I summarize in tab. 1 below (cf. Fried 1994, Avgustinova & Oliva 1997, Rosen 2001,
Junghanns 2002b, Petkevič 2009 for Czech, Kupść 2000 for Polish).
5
I follow the spirit of
Dotlačil (2007) and – most recently Adam (2019) in refraining from hypothesizing about the
exact syntactic status of the short pronouns, e.g. whether they are syntactic phrases or heads, or
whether they are weak rather than clitic pronouns.
6
Table 1: Unambiguous short pronominals in Czech and Polish
Czech
ho
mi
ti
mu
se
si
Polish
cię
go
mi
ci
mu
się
se7
‘youACC
‘himACC
‘meDAT
‘youDAT
‘himDAT
REFLACC/GEN
REFLDAT
3. Clitic Climbing in West Slavic
As in the Italian example (2), CC occurs from embedded infinitives in Czech and Polish.
8
Several scholars point out that infinitive-hood is a necessary, but not a sufficient condition for
CC (cf. Junghanns 2002a: 69, Golden 2003: 221–222, Kupść 2000: 58). The infinitival domain
must not be introduced by a subordinator.
9
Note that the clitic, which is subject to climbing, is
not necessarily an argument, e.g. REFL of a reflexive tantum or in impersonal constructions.
5
I adopt a traditional terminology from Slavic studies here, which rests upon the formal distinction of “short”
(e.g. Cz. mu, ho) vs. “long” pronouns (e.g. Cz. jemu, jeho). Note that terminology differs between authors. While
the short pronouns are referred to as konstantní přiklonky ‘constant clitics’ in Czech linguistics (cf. Trávníček
1959, Rosen 2001, Hana 2007), Avgustinova & Oliva 1997 propose the term pure clitics, Junghanns (2002b)
coins Lexikalische Klitika ‘lexical clitics’.
6
The anonymous reviewer pointed out that Czech short pronouns are true second position clitics, whereas Polish
short pronouns are weak pronouns. This point of view is reminiscent of Cardinaletti & Starke’s (1999) proposal
of a tripartite typology of pronouns. Since there is no general consensus on this matter, it appears that the
typology of clitics in Slavic still needs further investigation. For an alternative view, cf. Jung & Migdalski (this
volume), who propose an extension of Cardinaletti & Starke’s approach to a four-way classification.
7
Short dative reflexive se occurs in colloquial Polish, but remains unconsidered in most analyses (e.g. Spencer
1991, Kupść 2000). Rubadeau (1996: 137) claims that “Polish […] does not have a clitic form of the dative
reflexive”. On the other hand, Urbańczyk (1976: 58) discusses se in his survey of Polish dialects. Franks & King
(2000: 150) list se among the Polish clitics, but note that it “is used only in the spoken language”. Aguado &
Dogil (1989) explicitly take se into consideration.
8
The anonymous reviewer pointed out that CC is possible from a subset of morphologically finite da-clauses in
Serbian, i.e. from subjunctive-like da-clauses (cf. Progovac 1993, 1996 for indicative/subjunctive distinction
relevant for the availability of CC among other things). However, the indicative-subjunctive distinction is
irrelevant for CC in West Slavic, which is sensitive to the infiniteness of the embedded verb.
9
The term subordinator is meant to broadly cover elements introducing subordinate clauses of different kind, i.e.
(i.) complementizers introducing argument clauses, (ii.) subordinate conjunctions introducing adjunct clauses,
and (iii.) relative pronouns and adverbs introducing relative clauses.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Therefore, I adopt the term dependent from dependency grammar as a general notion for clitics
licenced by a verbal head. CC occurs in a variety of syntactic constructions, i.e. raising, subject
and object control, and the accusative with infinitive (abbr.: AcI, from Latin accusativus cum
infinitivo)
10
(cf. Junghanns 2002a, Golden 2008, Kupść 2000).
11
Note that clitic climbing is
ungrammatical in object control constructions in Romance, but not in Slavic (cf. Golden 2008:
315). Note also that Standard Polish and its vernacular do not possess the AcI construction (cf.
Przepiórkowski & Rosen 2005: 33, Kupść 2000: 96).
12
It has been observed that CC is not
obligatory and clitics may be realized in situ as in Italian (2a). In the same way, both a- and b-
examples are grammatical in Czech (3) and Polish (4). The question arises then, why CC does
come into being and what are the conditions for the local vs. non-local realization of the clitics.
(3) a. Asi ho2 chtěla1 usušit2 pomalu. (Cz; Junghanns 2002a: 82)
perhaps him-ACC want-PST.F dry-INF slowly
b. Asi chtěla1 usušit2 ho2 pomalu.
‘Perhaps she wanted to dry it slowly.’
(4) a. Jan go2 chciał1 obudzić2 o szóstej. (Pol; Kupść 2000: 60)
Jan him-ACC want-PST.SG.M wake-up at six
b. Jan chciał1 obudzić2 go2 o szóstej.
‘Jan wanted to wake him up at six o’clock.’
It has been proposed to account for CC in Slavic in terms of a restructuring approach by Rezac
(2005) for Czech and Aljović (2004) for Bosnian-Croatian-Montenegrin-Serbian (BCMS).
13
According to such an approach the optionality of CC is only an alleged one. While clitics must
remain in situ in true bi-clausal structures, they are forced to climb under restructuring, which
is underlyingly mono-clausal due to the structural deficiency of the embedded so-called
restructuring infinitive (RI). Being bare VPs, RIs lack the vP- and TP-shell.
14
Several correlates
10
The accusative with infinitive is known as Exceptional Case Marking (ECM) in Generative Grammar.
11
AcI verbs are verbs of perception semantically. Veselovská (1995: §9.5) states that CC is ungrammatical from
AcI constructions in Czech and Rezac (2005: 108) tells that Czech does not have true AcI verbs. The data
discussed in the present paper contradict these positions.
12
The lack of AcI is a general property of Polish syntax, it is not a particular feature of the Polish clitic system.
The AcI is attested in several diatopic substandard varieties of Polish (cf. Urbańczyk 1976: 56).
13
Cf. Rizzi (1982) on restructuring in Italian and Wurmbrand (2001) for a general analysis of restructuring
properties on the basis of German and Japanese.
14
Note that Wurmbrand (2001 and subsequent work) proposes a multi-way distinction of restructuring, which is
not limited to binary parametrization. We cannot take these proposals into consideration here due to space
limitations and must leave them for future discussion.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
have been put forward to support the restructuring analysis: (a) RIs are unable to license
accusative case, (b) RIs do not have an underlying subject (= PRO), (c) RIs do not constitute a
binding domain for Principle B, (d) either all clitics climb as a consequence of the infinitive’s
structural deficiency or none, (e) RIs are temporally dependent upon the matrix verb’s tense.
Criteria (a)(d) are taken from Rezac (2005), criterion (e) is taken from Todorović (2012).
4. Clitic Climbing and Correlates of Restructuring
In what follows I will test the hypothesis that CC is dependent upon restructuring by assessing
the above-mentioned correlates empirically towards corpus data from the Český Národní
Korpus ‘Czech National Corpus’ (ČNK) and the Narodowy Korpus Języka Polskiego ‘National
Corpus of Polish’ (NKJP) respectively.
15
If restructuring enforces CC, then clitics are expected
to remain in situ in configurations, where restructuring is not effective.
4.1. Case Licensing
Due to the lack of vP/TP, RIs are unable to license accusative case. Clitics climb in order to
receive case in the matrix phrase then.
16
Lenertová (2004) and Dotlačil (2004) recognize
independently for Czech that CC into passivized matrix domains contradicts this argument. It
is generally known from Burzio’s generalization (cf. Burzio 1986) that passivized verbs are
unable to licence accusative case. Data like (5) and (6) contradict the case-based argument, as
the case of the clitic ho ‘him-ACCcannot be licenced in the passivized matrix verb, but only by
the embedded infinitive. Lenertová’s (2004) and Dotlačil’s (2004) arguments are corroborated
by examples (7) and (8) for Czech and Polish respectively.
(5) (Přivezl puk za švýcarskou branku,) (Cz; Lenertová 2004: 159)
bring-PST.SG.M puck behind Swiss goal
ale tam ho3 byl1 donucen2 předat3 Lubinovi.
but there him-ACC be-PST.SG.M forced-PASS.SG.M give-INF Lubin-DAT
15
In particular, the Czech data are drawn from the subcorpus SYN version 8 (syn v8) (cf. Křen et al. 2019,
Hnátková et al. 2014). For Polish, I searched the full NKJP corpus through the Poliqarp search engine (cf.
Przepiórkowski et al. 2012).
16
It is irrelevant for the purpose of the present study, how case licensing is technically implemented.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
‘(He brought the puck behind the Swiss goal,) but there he was forced to give it to Lubin.’
(6) Pavel ho3 byl1 nucen2 zničit3. (Cz; Dotlačil 2004: 88)
Pavel-NOM him-ACC be-PST.SG. force-PASS.SG.M destroy-INF
‘Pavel was forced to destroy it.’
(7) a. […], kdo by ho3 byl1 oprávněn2 zbavit3 z
who COND him-ACC be-PST.SG.M entitle-PASS.SG.M relieve-INF of
odpovědnosti za osud Ruska.
responsibility for fate Russia-GEN.SG
‘…, who would have been entitled to relieve him of his responsibility for the fate of
Russia.’
b. […], kteří ho3 byli1 připraveni2 zatknout3.
who him-ACC be-PST.PL.M.ANI prepare-PASS.PL.M.ANI arrest-INF
‘…, who were prepared to arrest him. (Cz; ČNK syn v8)
(8) a. […] że już nigdy nie będę1 cię3 zmuszona2 oglądać3.
that already never NEG be-FUT.1SG you-ACC force-PASS.SG.F look-INF
‘… that I will never be forced to look at you, again.’
b. bo z powodu drżenia twoich rąk będę1 cię3
because from reason tremor your hands be-FUT.1SG you-ACC
zmuszony2 wrzucić3 do KF
force-PASS.SG.M throw-INF to KF
‘because of your hands’ tremor I will be forced to throw you to the KF [= kill file]’
(Pol; NKJP full)
4.2. Missing Subjects
The lack of vP yields RIs without having an underlying subject (PRO). Rezac (2005: 114) states
that RIs “will not constitute a binding domain of their own, and coreference between a
pronominal argument of the infinitive and any argument of the upstairs verb should be
blocked.” He provides the minimal pair in (9ab). In (9a) the embedded clitic ji ‘her-ACCis
co-referential with both the matrix subject Anna (index a) and a distinct discourse referent
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
beyond the sentence-level (index b). Co-reference between Anna and ji is established by
mediation through PRO. In (9b) the clitic has climbed due to restructuring. As a consequence,
there is no clause boundary between matrix and subordinate domain, thus, co-reference between
Anna and ji is excluded. Rezac (2005) accounts for (9b) by a violation of Binding Principle B,
according to which [a] pronominal is free [i.e. unbound] in its governing category [i.e. clause]”
(Chomsky 1981: 188). Principle B determines semantic co-reference by syntactic non-co-
membership, which does not reveal anything about clause boundaries here. Co-reference
between Anna and ji is still excluded by Principle B in presence of a clause boundary, for both
subject and clitic are co-members of the matrix domain, cf. (9c).
(9) a. [ Annaa mu dovolila [ PROa políbit jia/b nashledanou ]].
Anna-NOM him-DAT allow-PST.SG.F kiss-INF her-ACC good-bye
b. [ Annaa mu ji*a/b dovolila políbit nashledanou ].
c. [ Annaa mu ji*a/b dovolila [ (PRO) políbit nashledanou ]].
‘Ana permitted him to kiss her good-bye.’ (Cz; Rezac 2005: 114)
Rezac (2005: 114115) further states that neither matrix argument binds subject-oriented
anaphora svým ‘one’s-POSS.PL.DAT’ in (10b) in contrast to (10a). As the clitic je ‘them-ACC’ has
climbed, restructuring must have occurred and PRO is missing thence. However, co-reference
between the matrix subject and the anaphorical possessive pronoun should be still expected in
a restructuring context. In fact, Dotlačil (2007) and Skoumalová (2005) judge (10b)
grammatical with both interpretations, such that embedded svým is bound by either matrix
argument (Pavel, Janovi) despite CC. These judgements are corroborated by the corpus data in
(11) and (12). First, the matrix subject stavitel ‘constructor’ binds the possessive anaphor své
‘one’s’-POSS after CC in (11) as expected. Second, and even more intriguing, example (12)
shows that the matrix object clitic mu ‘him-DATbinds the embedded possessive anaphor své
in spite of the climbed embedded clitic ho ‘him-ACC’. While the binding relations in (11) are
expected under standard assumptions in any mono-clausal domain, the binding facts in (12) are
best analysed by assuming an underlying subject in the embedded infinitive (i.e. PRO under
standard Generative assumptions).
17
(10) a. Pavela přikázal1 Janovib dát2 je2 svýma/b přátelům.
17
I do not adopt Hornstein’s (1999) proposal in abandoning the raising-control distinction, which has originally
been the main motivation for the assumption of PRO (cf. Przepiórkowski & Rosen 2005 for a similar account in
HPSG). Cf. Culicover & Jackendoff (2001) and Landau (2003) for a critique.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Pavel-NOM order-PST.SG.M Jan-DAT give-INF them POSS friends-DAT
b. * Pavela je2 Janovib přikázal1 dát2 svýma/b přátelům.
‘Pavela ordered Janb to give them to hisa/b friends.’ (Cz; Rezac 2005: 114115)
(11) Stavitela mu2 nechtěl1 vnucovat2 svéa mínění […].
constructor-NOM him-DAT NEG-want-PST.SG.M impose-INF POSS opinion
‘The constructor didn’t want to impose his opinion on him …’ (Cz; ČNK syn v8)
(12) ( A právě [NP. “ten myš” ]a se sourozencib zalíbil
and exactly this-SG.M mouse REFL sibling-SG.DAT please-PST.SG.M
natolik, že požádal,)
so-much that me-ACC ask-PST.SG.M
abych mu1/b ho2/a dovolil1 použít2 v jedné svéb písničce.
so-that him-DAT him-ACC allow-PST.SG.M use-INF in one POSS song
‘(And [my] sibling liked exactly “this he-mouse” so much that he asked me,)
if I would allow him to use it in one of his songs.’ (Cz; ČNK syn v8)
Another argument that challenges the predicted binding correlations of restructuring has been
put forth by Golden (2008: 316) for Slovene (Sln). She observed that certain object control
constructions are semantically ambiguous, although the embedded clitic has climbed (13). She
argues that the ambiguities might be only captured with a PRO-analysis. The ambiguity of
Czech (14) and Polish (15) likewise calls for an analogue of a PRO-analysis for the infinitive.
(13) a. Janez ji1/2 jih2 je dovolil1 kupiti2.
Janez-NOM her-DAT them-ACC AUX.3SG allow-PST.SG.M buy-INF
i. ‘Janez allowed her to buy them.’
ii. ‘Janez allowed (someone) to buy them/it for her/them.’ (Sln; Golden 2008: 316)
b. Jaz sem ji1/2 ga2 dovolil1 poslati2 po pošti.
I-NOM AUX.1SG her-DAT him-ACC allow-PST.SG.M send-INF by mail
i. ‘I allowed her to send it by mail.’
ii. ‘I allowed (somebody) to send it to her by mail.’ (Sln; Golden 2008: 312)
(14) [...], strýc mu1/2 ho2 nedovolí1 přečíst2. (Cz; ČNK syn v8)
uncle him-DAT him-ACC NEG-allow-PRS.3SG read-INF
i. ‘The uncle doesn’t allow him to read it.’
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
ii. ‘The uncle doesn’t allow (someone) to read it to him.’
(15) […] każą1 mu1/2 go2 rozebrać2. (Pol; NKJP full)
order-PRS.3PL him-DAT him-ACC deconstruct-INF
i. ‘… they order him to deconstruct it.’
ii. ‘… they order (someone) to deconstruct it for him.’
4.3. Temporal Reference
It has been argued that the RIs’ temporal reference is dependent upon the one presupposed by
the matrix verb. RIs are ungrammatical with a temporal adverb, which refers to a time frame
deviating from the matrix verb’s one. Wurmbrand’s (2001) German (Ger) example (16a)
provides a grammatical utterance without restructuring. The main verb encodes the past tense
(morphosyntactically encoded by the analytical perfect form), but the embedded infinitive
refers to the future by the time adverb morgen ‘tomorrow’. On the other hand, the presence of
the time adverb is ungrammatical in a restructuring context like (16b). Example (17) from
Aljović (2004) suggests that the same holds for Slavic. The presence of the time adverb sutra
‘tomorrow’ in the embedded clause is grammatical in the BCMS example (17a), as long as the
pronominal clitic ga ‘him-ACC’ is in situ. When restructuring occurs and the clitic climbs, then
the realization of the time adverb yields the utterance ungrammatical (17b).
(16) a. Hans hat beschlossen (morgen) zu verreisen.
Hans have-PRS.3SG decide-PTCP (tomorrow) to travel-INF
‘John decided to go on a trip (tomorrow).’
b. Hans hat versucht (*morgen) zu verreisen.
Hans have-PRS.3SG try-PTCP (tomorrow) to travel-INF
‘John tried to go on a trip (*tomorrow). (Ger; Wurmbrand 2001: 73)
(17) a. On želi1 da ga2 (sutra) Jovanu predstavi2.
he want-PRS.3SG that him-ACC (tomorrow) Jovan-DAT introduce-PRS.3SG
b. * On ga2 želi1 da (sutra) Jovanu predstavi2.
‘He wants to introduce him to John tomorrow.’ (BCMS; Aljović 2004: 193)
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Lenertová (2004) notes that the aforementioned argument does not hold for Czech, where CC
co-occurs with the embedded infinitive’s independent temporal reference. The clitic ho ‘him-
ACC’ in (18) has climbed to the matrix domain headed by the past tense verb rozhodl ‘decide-
PST.SG.M’. However, the realization of the temporal adverb příště ‘next time’ or adverbial PP
na moment ‘for a moment’ within the infinitive’s domain is grammatical.
18
Lenertová’s
observation is corroborated for Czech (19) and Polish (20) by corpus data. Note that climbing
is grammatical irrespective of whether the time adverb(ial)s intervene between matrix verb and
embedded infinitive or not.
19
(18) Místo toho se1 ho2 rozhodl1 na moment / příště ignorovat2.
instead-of this REFL him-ACC decide-PST.SG.M on moment next-time ignore-INF
‘Instead, he decided to ignore him for a moment / next time.’ (Cz; Lenertová 2004: 157)
(19) a. Přitom ho2 chtěla1 odstartovat2 příští sobotu při příležitosti
but-in-fact him-ACC want-PST.SG.F launch-INF next Saturday at occasion
oslav 700 let od udělení městských práv Sokolovu.
celebration 700 years from awarding city rights Sokolov
‘But in fact, [the town’s administration] wanted to launch it on the occasion of the
700th anniversary of Sokolov receiving its town charter.’
b. Lidé, kteří se2 chtěli1 zítra večer bavit2 při filmu
people who REFL want-PST.PL.M tomorrow evening entertain-INF at film
Borat mají smůlu.
Borat have-PRS.3PL bad-luck
‘Those people, who wanted to enjoy the Borat movie tomorrow evening, have bad
luck.’ (Cz; ČNK syn 8)
(20) a. Ja się2 postanowił-em nie podrapać2 jutro o 12.15 […].
I-NOM REFL decide-PST.SG.M-M.1SG NEG scratch-INF tomorrow at 12.15
‘I decided not to scratch myself tomorrow at 12:15 … .’
b. ja mu2 zdecydowałem1 się1 odpowiadać2 [o ile na
I-NOM him-DAT decide-PST.SG.M-M.1SG REFL reply-INF [at how-much on
jakieś posty będzie warto] po 24 godzinach
18
The time adverb(ial)s are underlined for ease of exposition.
19
This fits Junghanns‘s (2002a: 66) observation that the cascade of verbs, which constitutes an environment for
CC, does not form a verb cluster (Germ. Verb[al]komplex) in Czech, i.e. they do not need to be contiguous.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
some posts be-FUT.3SG worth] after 24 hours
‘I decided to respond to him [as far as some posts will be worth it] after 24 hours.’
(Pol; NKJP full)
4.4. All or Nothing
CC has been deemed an all-or-nothing phenomenon” (Rezac 2005: 111), whereby either all
embedded clitics climb or none (cf. Aljović 2004: 194 for a similar position regarding BCMS).
Due to the RI’s structural deficiency, the clitics escape the infinitival domain to satisfy their
formal requirements in the matrix phrase, where they are placed in the respective clitic cluster.
Diaclisis of co-dependents poses a problem for such an approach then.
20
In the empirically
attested Serbian example (21), both the pronominal clitic mi ‘me-DATand the reflexive clitic
se ‘self’ are subcategorized for by the embedded verb vrti ‘spin-PRS.3SGof the da-clause.
However, it is only the pronominal clitic that climbs, the reflexive remains in situ. As
restructuring is supposed to affect all embedded clitics equally, the approach is unable to predict
differences in the distribution of co-dependent clitics. This is also true for diaclisis in Czech
(22) and Polish (23). Again, both reflexive and pronominal clitics are co-dependents of the
embedded infinitives. Only the reflexive clitics occupy the clausal second position, whereas the
pronominal clitics do not build a cluster with the reflexives and appear further to the right.
21
While this behaviour would appear unsurprising for Polish given the rather peculiar status its
clitic system is assigned, the occurrence of the same pattern in Czech is unexpected.
22
Note that
the diaclitic distribution is independent of the argument status of the clitics. The reflexives in
the a-examples are true arguments of the embedded infinitives, whereas they are not in the b-
20
I adopt the term diaclisis for split-clitic-constructions from Björn Hansen (p.c.). To my knowledge, the notion
was first introduced by Janse (1998) discussing clitics in Cappadocian Greek.
21
Uwe Junghanns (p.c.) pointed out to me that it is impossible to determine a priori, whether the pronominal
clitics in (22)(23) have in fact climbed or whether they are still positioned within the infinitival phrase (cf. also
Junghanns 2002a: 6768). The contrast is schematised in (i) and (ii).
(i) [α Vα CLβ [β Vβ ]] (CC/non-local)
(ii) [α Vα [β CLβ Vβ ]] (in situ/local)
Proclisis to the following infinitive would be indicative of clitic in situ placement. Proclisis is available in Czech
(cf. Toman 1996) and Polish (cf. Kraska-Szlenk 1995: 6264) for the set of clitics relevant here.
22
The Czech data are problematic for Bošković’s (2001) PF-filtering approach to clitic clustering, according to
which clitics in a second position clitic language are placed according to two parameters: first, initial positioning
in an intonation phrase (ιP), second, being suffixed to a prop. Bošković accounts for diaclisis in Polish assuming
that Polish clitics do not possess the second position requirement of being ιP-initial. The Czech facts cannot be
captured, since the approach predicts that split clitics are placed in distinct ιPs. However, there is only one ιP for
the relevant clause in Czech (22), cf. (i)(ii). I thank Martina Berrocal (p.c.) for the judgement (# marks a pause).
(i) [ιP Lehce si (*#) uměla ho představit ]
(ii) [ιP kdekdo se (*#) začal mi smáti ]
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
examples. Both Cz. smát(i) se ‘laugh’ and Pol. bać się ‘be afraid’ are reflexives tantum, i.e. the
appearance of the reflexive is lexically specified. This finding supplements Lenertová’s (2004:
138139) observation that Czech clitics do not need to cluster together, as she found that
conditional and reflexive/pronominal clitics may occur non-contiguously.
(21) […] i počelo1 mi2 je [ da se2 vrti2 u glavi ].
and start-PST.SG.N me-DAT AUX.3SG that REFL spin-PRS.3SG in head
‘… and I started to feel dizzy.’ (Ser; Jurkiewicz-Rohrbacher et al. 2017: 188)
(22) a. Lehce si2 uměla1 ho2 představit2, […] (Cz; ČNK syn v8)
easily REFL be-able-PST.SG.F him-ACC imagine-INF
‘She could easily imagine him, …’
b. […] kdekdo se2 začal1 mi2 smáti2.
almost-everybody REFL start-PST.SG.M me-DAT laugh-INF
‘… almost everybody started to laugh at me.’
(23) a. My się2 musimy1 go2 nauczyć2. (Pol; NKJP full)
we-NOM REFL must-PRS.1SG him-ACC teach-INF
‘We have to learn it.’
b. Już się2 zaczęli1 go2 bać2, […]
already REFL start-PST.PL.VIR him-ACC fear-INF
‘They already started to be afraid of him, …’
5. Clitic Climbing and Information Structure
The previous sections showed that a purely syntactic account in terms of restructuring cannot
account for CC in Czech and Polish. The corpus data provided in section 4 contradict the
predictions of the approach. Therefore, I agree with Dotlačil (2004: 87) in that CC does not
occur because of restructuring and that both should be regarded as independent phenomena.
If restructuring is not responsible for CC, then the question arises, what is. Which
alternatives are available, if one refrains from accounting for CC by syntax proper? Proposals
in terms of phonology, morphology, and syntax-prosody interaction have been put forward, but
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
were also criticized.
23
An explanatory account of CC has to shed light on the actuation or
causation of CC vs. clitic in situ positioning.
24
I propose to take a candidate into consideration
that repeatedly appears in the literature on clitics, but has been mostly neglected: information
structure. Stjepanović (2004: 206) considers the possibility that CC is an instance of object
shift, which has been reported to rely on information structural notions in Northern Germanic:
only objects having background status are shifted.
Junghanns (2002a: 8283) takes the farthest step towards information structure I am
aware of and proposes that information structure is the actual reason for CC, whereas syntax
merely restricts, which domains clitics can escape. Consequently, a clitic climbs, iff. it belongs
to the background of the whole sentence, else it remains in situ. The else-case covers utterances,
in which the clitic is part of a topicalized or focused constituent. The clitic itself does not bear
topic or focus, but is an element of a domain specified as either [+TOPIC] or [+FOCUS].
Accounting for the ban of CC across CP, Dotlačil (2004: 93, 98; 2007: 89) suggests that
clitics cannot escape CPss, because they cannot bear the discourse functions of topic or focus.
He observes that topicalized or focused constituents are able to escape CPs in Czech. If clitics
are hosted in such a topicalized or focused domain, they can cross a CP as a part of the respective
constituent. However, no CC occurs, as the clitics remain in their licensing domain.
I adopt Junghanns’s (2002a) proposal and paraphrase it tentatively with the short
notation in (26), which is based upon a basically syntactic description suggested to me by Hagen
Pitsch (p.c.).
25
The notation in (24) reads as follows: a clitic that is licenced in domain β (= CLβ)
is realized in β, if β bears topic or focus, otherwise CLβ is realized in domain α.
(24) [α CLβ [β CLβ ][+TOP, +FOC]]
23
For instance, Franks & King (2000: 287291, 293305) critically review purely phonological and syntactic
accounts of cliticization (incl. CC) and conclude that both types of approach face several problems in accounting
for clitic phenomena (cf. also the discussion in Bošković 2001: 36–80). However, we do not pursue a mixed
syntax-PF account like Franks & King’s (2000: ch. 11 & 12) and Franks’s (2010) PF-filtering approach. Another
mixed account is Halpern’s (1995) Prosodic Inversion, which is critically reviewed in detail by Bošković (2001:
11–36). On the other hand, Bošković‘s (2001) own intonational-phrase-based proposal has been criticized by
Lenertová (2004: 150151) and Golden (2008: passim) (cf. also footnote 22).
24
I refer to the traditional notion of explanation basing on causality, not to Chomsky’s (1965: 2526) concept of
explanatory adequacy. Instead, I allude to the actuation problem coined by Weinreich et al. (1968) towards the
background of historical linguistics and adopt it for the field of synchronic grammar research.
25
I provide Hagen Pitsch’s original suggestion in (i) for the sake of transparency.
(i) [α {CL[−TOPIC, −FOCUS]} [β {CL[+TOPIC, +FOCUS]} ]]
As the reader can see, I applied minor revisions. A different proposal that I might suggest are the implications in
(ii). Formula (ii.a) reads as follows: for every x, iff. x has property CL (= is a clitic) and x is element of the
information structural background, then x is realized in domain α and co-indexed with a gap e in domain β.
Implication (ii.b) specifies the else-case with clitic in situ realization.
(ii) a.x(CL(x) x [BACKGROUND] → [α xj [β ej ]])
b. x(CL(x) x {[TOPIC], [FOCUS]} → [α … [β x ]])
I leave elaborations on the formal descriptions for future research.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
6. Conclusion
The paper empirically tested the hypothesis that clitic climbing in Czech and Polish is
contingent upon a mono-clausal restructuring environment. In particular, I reviewed the
predicted correlates of the proposal that clitics escape defective infinitival complements, which
are bares VPs. Utilizing data from both the Czech National Corpus and the National Corpus of
Polish, it has been shown that (i.) clitics climb to passivized domains, i.e. not for accusative
case licencing, (ii.) binding phenomena and ambiguities in climbing constructions call for an
underlying subject (PRO) analysis of the infinitival domain, (iii.) embedded infinitives possess
temporal reference independent of the finite matrix verb, (iv.) co-dependent clitics do not
behave uniformly with respect to climbing and end up non-contiguously. In sum, the respective
predictions of the restructuring approach have been falsified. Clitic climbing thus cannot be
regarded as being restricted to mono-clausal structures, but occurs in what is considered an
underlyingly bi-clausal structure. This state of affairs yields the approach uneligible for clitic
climbing in Czech and Polish. More generally, as syntax proper does not provide us with an
explanatory account for the very existence of clitic climbing, alternatives have to be taken into
consideration seriously. Following Junghanns (2002a), I referred to information structure,
whereby clitics climb, if they are elements of the background of the entire clause, but remain in
situ, if they are elements of a topicalized or focused constituent. Admittedly, the present study
does not address, how to account for diaclisis in terms of information structure. This has to be
dealt with in future research. What is more, the resemblance between Czech and Polish clitic
distributions suggests that the typological peculiarity of Polish is not well-grounded. I propose
to revisit and refine the micro-typology of Slavic cliticization on a sound empirical basis.
References
Adam, Nina. 2019. Clitics at the Interfaces of Grammar: Defining the First Position” in
Czech.: Typology of Morphosyntactic Parameters 2.1, 1127. Online:
http://tmp.sc/application/files/5015/7781/8054/Adam-2019-2-1.pdf, last retrieved 13th
September 2021.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Aguado, Miquel & Grzegorz Dogil. 1989. Clitics in Lexical Phonology: Alleged
Counterevidence? Linguistische Berichte 120, 99116.
Aljović, Nadira. 2004. Cliticization Domains: Clitic Climbing in Romance and in Serbo-
Croatian. In: Actes des / Proceedings of JEL’2004. [DOMAIN[E]S]. Journées d’Études
Linguistiques 2004, 567 Mai 2004, Nantes, France, ed. by Olivier Crouzet, Hamida
Demirdache & Sophie Wauquier-Gravelines, 191197. Nantes: Université de Nantes.
Avgustinova, Tania & Karel Oliva. 1997. On the Nature of the Wackernagel Position in Czech.
In: Formale Slavistik, ed. by Uwe Junghanns & Gerhild Zybatow, 2547. Frankfurt am
Main: Vervuert.
Birnbaum, Henrik & Jerzy Molas. 72012. Das Polnische. In: Einführung in die slavischen
Sprachen, ed. by Peter Rehder, 145164. Darmstadt: Wissenschaftliche Buchgesellschaft.
Borsley, Robert D. & María Luisa Rivero. 1994. Clitic Auxiliaries and Incorporation in Polish.
Natural Language & Linguistic Theory 12, 373422.
Bošković, Željko. 2001. On the Nature of the Syntax-Phonology Interface: Cliticization and
Related Phenomena. Amsterdam: Elsevier.
Burzio, Luigi. 1986. Italian Syntax. A Government-Binding Approach. (= Studies in Natural
Language and Linguistic Theory 1.) Dordrecht: D. Reidel Publishing Company.
Cardinaletti, Anna & Michal Starke. 1999a. The Typology of Structural Deficiency: A Case
Study of the Three Classes of Pronouns. In: Clitics in the Languages of Europe, ed. by Henk
van Riemsdijk, 145233. (= Empirical Approaches to Language Typology 20.) Berlin/New
York: Mouton de Gruyter.
ČNK: Český Národní Korpus. https://www.korpus.cz/, last retrieved 1st June, 2021.
Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press.
Chomsky, Noam. 1981. Lectures on Government and Binding. The Pisa Lectures. Dordrecht:
Foris.
Culicover, Peter W. & Ray Jackendoff. 2001. Control is not Movement. Linguistic Inquiry
32(3), 493512.
Damerau, Norbert. 1992. Polnische Grammatik. 2nd unrevised edition. (= Sammlung Göschen
2808.) Berlin/New York: Walter de Gruyter.
Dimitrova-Vulchanova, Mila. 1999. Clitics in the Slavic Languages. In: Clitics in the
Languages of Europe, ed. by Henk van Riemsdijk, 83122. (= Empirical Approaches to
Language Typology 20.) Berlin/New York: Mouton de Gruyter.
Dotlačil, Jakub. 2004. The Syntax of Infinitives in Czech. MA thesis, University of Tromsø.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Dotlačil, Jakub. 2007. Why Clitics Cannot Climb out of CP: A Discourse Approach. In: Formal
Approaches to Slavic Linguistics: The Toronto Meeting 2006, ed. by Ulyana Savchenko,
Magdalena Goledzinowska & Richard Compton, 7693. Ann Arbor, MI: Michigan Slavic
Publications.
Dziwirek, Katarzyna. 1998. Reduced Construction in Universal Grammar: Evidence from the
Polish Object Control Construction. Natural Language & Linguistic Theory 16, 5399.
Franks, Steven. 2009. Clitics in Slavic. In: Die slavischen Sprachen: Ein internationales
Handbuch zu ihrer Struktur, ihrer Geschichte and ihrer Erforschung, ed. by Sebastian
Kempgen, Peter Kosta, Tilman Berger & Karl Gutschmidt, 725738. (= Handbücher zur
Sprach- and Kommunikationswissenschaft 32.) Berlin/New York: Walter de Gruyter.
Franks, Steven. 2010. Clitics in Slavic. Glossos 10: 1157. (Revised version of the position
paper from the workshop “Comparative Slavic Morphosyntax”. Spencer, Indiana, 1998.)
Online:
https://slaviccenters.duke.edu/sites/slaviccenters.duke.edu/files/media_items_files/10franks
.original.pdf, last retrieved 13th September, 2021.
Franks, Steven & Tracy H. King. 2000. A Handbook of Slavic Clitics. Oxford: University Press.
Fried, Mirjam. 1994. Second-Position Clitics in Czech: Syntactic or Phonological? Lingua 94,
155175.
Golden, Marija. 2003. Clitic Placement and Clitic Climbing in Slovenian. Sprachtypologie und
Universalienforschung [STUF Language Typology and Universals] 56, 208233.
Golden, Marija. 2008. 2nd Position Clitic Climbing and Restructuring. Formal Description of
Slavic Languages, ed. by Gerhild Zybatow, Luka Szucsich, Uwe Junghanns & Roland
Meyer, 308317. Frankfurt am Main: Peter Lang.
Halpern, Aaron. 1995. On the Placement and Morphology of Clitics. Stanfor, CA: CSLI
Publications.
Hana, Jiri. 2007. Czech Clitics in Higher Order Grammar. PhD thesis, The Ohio State
University.
Hansen, Björn; Zrinka Kolaković & Edyta Jurkiewicz-Rohrbacher. 2018. Clitic Climbing and
Stacked Infinitives in Bosnian, Croatian and Serbian A Corpus-Driven Study. In:
Grammar and Corpora 2016, ed. by Eric Fuß, Marek Konopka, Beata Trawiński & Ulrich
H. Waßner, 259268. Heidelberg: Heidelberg University Publishing.
Hornstein, Norbert. 1999. Movement and Control. Linguistic Inquiry 30 (1), 6996
Hnátková, Milena; Michal Křen, Pavel Procházka & Hana Skoumalová. 2014: The SYN-series
corpora of written Czech. In: Proceedings of the Ninth International Conference on
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Language Resources and Evaluation (LREC'14), 160164. Reykjavík: ELRA. ISBN 978-2-
9517408-8-4. Online: http://www.lrec-conf.org/proceedings/lrec2014/pdf/294_Paper.pdf,
last retrieved 13th September, 2021.
Janse, Mark. 1998. Cappadocian Clitics and the Syntax-Morphology Interface. In: Themes in
Greek Linguistics II, ed. by Brian D. Joseph, Geoffrey C. Horrocks & Irene Philippaki-
Warburton, 257281. (= Amsterdam Studies in the Theory and History of Linguistic Science
159.) Amsterdam/Philadelphia: John Benjamins.
Junghanns, Uwe. 2002a. Clitic Climbing im Tschechischen. In: Untersuchungen zur Syntax und
Informationsstruktur slavischer Deklarativsätze, ed. by Uwe Junghanns, 5790.
(= Linguistische Arbeitsberichte 80.) Leipzig: Institut für Linguistik der Universität Leipzig.
Junghanns, Uwe. 2002b. Klitische Elemente im Tschechischen: Eine kritische
Bestandsaufnahme. In: Linguistische Beiträge zur Slavistik. IX. JungslavistInnen-Treffen,
Halle a.d. Saale / Lutherstadt Wittenberg, 27. September 1. Oktober 2000, ed. by Thomas
Daiber, 117150. Munich: Otto Sagner.
Jurkiewicz-Rohrbacher, Edyta; Björn Hansen & Zrinka Kolakovic. 2017. Clitic Climbing,
Finiteness and the Raising-Control Distinction: A Corpus-Based Study. Jazykovedný časopis
68, 179190.
Kraska-Szlenk, Iwona. 1995. The Phonology of Stress in Polish. PhD thesis, University of
Illinois at Urbana-Champaign.
Křen, Michal; Václav Cvrček, Tomáš Čapka, Anna Čermáková, Milena Hnátková, Lucie
Chlumská, Tomáš Jelínek, Dominika Kováříková, Vladimír Petkevič, Pavel Procházka,
Hana Skoumalová, Michal Škrabal, Petr Truneček, Pavel Vondřička & Adrian Zasina:
Corpus SYN, version 8 from 12.12.2019. Ústav Českého národního korpusu FF UK, Prague
2019. Available online: http://www.korpus.cz.
Kupść, Anna. 2000. An HPSG Grammar of Polish Clitics. PhD thesis, Institute of Computer
Science, Polish Academy of Sciences and Université Paris 7.
Landau, Idan. 2003. Movement out of Control. Linguistic Inquiry 34(3), 471498.
Lenertová, Denisa. 2004. Czech Pronominal Clitics. Journal of Slavic Linguistics, 12(12),
135171.
NKJP: Narodowy Korpus Języka Polskiego. http://nkjp.pl/, last retrieved 1st June, 2021.
Petkevič, Vladimír. 2009. Postavení příklonek v české klauzi v korpusech současné češtiny. In:
After Half a Century of SlavonicNatural Language Processing, ed. by Dana Hlaváčková,
Aleš Horák, Klára Osolsobě & Pavel Rychlý, 185196. Brno: Tribun EU.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Progovac, Ljiljana. 1993. Locality and Subjunctive-like Complements in Serbo-Croatian.
Journal of Slavic Linguistics 1, 116144.
Progovac, Ljiljana. 1996. Clitics in Serbian/Croatian: Comp as the Second Position. In:
Approaching Second: Second Position Clitics and Related Phenomena, ed. by Aaron L.
Halpern & Arnold M. Zwicky, 411428. Stanford, CA: CSLI Publications.
Przepiórkowski, Adam; Mirosław Bańko, Rafał L. Górski & Barbara Lewandowska-
Tomaszczyk (eds). 2012. Narodowy Korpus Języka Polskiego. Warsaw: Wydawnictwo
Naukowe PWN.
Przepiórkowski, Adam & Alexandr Rosen. 2005. Czech and Polish Raising/Control with or
without Structure Sharing. Research in Language 3, 3366.
Rappaport, G. 1988. On the Relationship between Prosodic and Syntactic Properties of the
Pronouns in the Slavonic Languages. In: American contributions to the 10th International
Congress of Slavists, ed. by Alexander M. Schenker, 301327. Columbus, OH: Slavica.
Rezac, Milan. 2005. The syntax of clitic climbing in Czech. Clitic and affix combinations:
Theoretical perspectives, ed. by Lorie Heggie & Francisco Ordóñez, 103140. (= Linguistik
Aktuell/Linguistics Today 74.) Amsterdam: John Benjamins.
Rizzi, Luigi. 1982. A Restructuring Rule. In: Issues in Italian syntax, ed. by Luigi Rizzi, 148.
Dordrecht: Foris Publications.
Rosen, Alexandr. 2001. A constraint-based approach to Dependency Syntax applied to some
issues of Czech word order. PhD thesis, Univerzita Karlova v Praze.
Rosen, Alexandr. 2014. Haplology of Reflexive Clitics in Czech. In: Slavic and German in
Contact: Studies from Areal and Contrastive Linguistics, ed. by Elżbieta Kaczmarska &
Motoki Nomachi, 97116. (= Slavic Eurasian Studies 26.) Sapporo: Slavic-Eurasian
Research Center. Online: http://src-h.slav.hokudai.ac.jp/coe21/publish/no26_ses/26-05.pdf,
last retrieved 13th September, 2021.
Rothstein, Robert A. 1993. Polish. In: the Slavonic Languages, ed. by Bernard Comrie &
Greville G. Corbett, 686758. London/New York: Routledge.
Rubadeau, Patrice Marie. 1996. A Descriptive Study of Clitics in Four Slavic Languages: Serbo-
Croatian, Bulgarian, Polish, and Czech. PhD thesis, University of Michigan, Ann Arbor.
Skoumalová, Hana. 2005. Review of: Clitic and Affix Combinations: Theoretical Perspectives.
Online: https://linguistlist.org/issues/16/16-3131.html, last retrieved 30th September 2021.
Spencer, Andrew. 1991. Morphological Theory. An Introduction to Word Structure in
Generative Grammar. Oxford/Cambridge, MA: Basil Blackwell.
Spencer, Andrew & Ana R. Luís. 2012. Clitics: An Introduction. Cambridge: University Press.
© Irenäus Kulik. 2021. To appear in: Advances in formal Slavic linguistics 2021. Berlin: Language Science Press.
Stjepanović, Sandra. 2004. Clitic Climbing and Restructuring with “Finite Clause” and
Infinitive Complements. Journal of Slavic linguistics 12 (12), 173212.
Todorović, Nataša. 2012. The Indicative and Subjunctive da-Complements in Serbian. A
Syntactic-Semantic Approach. PhD thesis, University of Illinois at Chicago.
Toman, Jindřich. 1996. A Note on Clitics and Prosody. In: Approaching Second: Second
Position Clitics and Related Phenomena, ed. by Aaron L. Halpern & Arnold M. Zwicky
(eds), 505510. Stanford, CA: CSLI Publications.
Trávníček, František. 1959. K postavení stálých příklonek po přestávce uvnitř věty. Naše Řeč
42, 6576.
Urbańczyk, Stanisław. 1976. Zarys dialektologii polskiej. Warsaw: Państwowe Wydawnictwo
Naukowe.
Veselovská, Ludmila. 1995. Phrasal Movement and X0-Morphology. Word Order Parallels in
Czech and English Nominal and Verbal Projections. PhD thesis, Olomouc: Univerzita
Palackého.
Weinreich, Uriel; William Labov & Marvin I. Herzog. 1968. Empirical Foundations for a
Theory of Language Change. In: Directions for Historical Linguistics. A Symposium, ed. by
W. P. Lehmann & Yakov Malkiel, 95195. Austin/London: University of Texas Press.
Wurmbrand, Susanne. 2001. Infinitives. Restructuring and Clause Structure. (= Studies in
Generative Grammar 55.) Berlin/New York: Mouton de Gruyter.
Zwicky, Arnold M. 1977. On Clitics. Indiana University Linguistics Club, Bloomington, IN.
ResearchGate has not been able to resolve any citations for this publication.
Clitics at the Interfaces of Grammar: Defining the "First Position" in Czech
  • Nina Adam
Adam, Nina. 2019. Clitics at the Interfaces of Grammar: Defining the "First Position" in Czech.: Typology of Morphosyntactic Parameters 2.1, 11-27. Online: