Content uploaded by Radek Čech
Author content
All content in this area was uploaded by Radek Čech on Mar 26, 2018
Content may be subject to copyright.
51
Glottometrics 40, 2018, 51-62
On the Development of Old Czech (En)clitics
Kosek Pave
Abstract. The presented study deals with the historical development of Czech (en)clitics (AuxP).
Based on the data from the previous research (Kosek, 2015a,b, 2017), it focuses on the development of
one group the Czech (en)clitics on the preterite auxiliary forms. In the article, three hypotheses are
formulated and then tested on the data gained from selected parts of historical Czech Bible trans-
lations. The suggest that there were two significant word order positions of historical Czech
(en)clitics: 1. the post-initial position, i.e. after first word / phrase, 2. the contact position, i.e. an
(en)clitic is located immediately before (pre-verbal position) or after (post-verbal position) its
syntactically or morphologically superordinate item (the post-verbal position is the more frequent
variant of the both variants of the contact positions). Since the time when the oldest analyzed text was
translated, the post-initial position has had the status of the basic word order position of the Czech
(en)clitic, while the contact position has had the status of a stylistically, pragmatically or textually
motivated position. It seems that the contact position begins to retreat only in 19th century and hence
the definitive historical change of Czech auxiliary (en)clitics in the sole second position clitics was
realized not before 19th or 20th century.
Keywords: Czech, enclitics
1. Introduction
The development of language characteristics is determined by various kinds of mechanisms
influencing human linguistic behaviour. Over time, some mechanisms grow stronger and
finally acquire the status of a grammatical rule, while other mechanisms gradually weaken
and can even disappear. In this study, we attempt to model the development of the word order
of Czech preterite auxiliary (en)clitics (AuxP) of the type nesl jsem in Old Czech.
We hypothesize that this development is systematically determined by a hierarchy of
mechanisms and also strength constant but varies in
different periods. As for the hierarchy of mechanisms, there seem to be three strong mechan-
isms which can be expressed in the form of stochastic rules for the word order of (en)clitics:
R1: if an (en)clitic appears in a clause, use it after the initial phrase of the clause (i.e.
in post-initial position)1:
(1) a. [ ]1 se2 [starý strom]3 [rázem]4 [skácel]2
in gardenLOC.M.SG REFLACC oldNOM.M.SG treeNOM.M.SG suddenly fall PART.PRET.ACT.M.SG
R2: if rule (R1) is not applied, use the (en)clitic in a postposition of a verb;
(1) b. [ ]1 [skácel se]2 [starý strom]3 [rázem]4
1 For an illustration of this phenomena, we use Czech examples with the reflexive pronominal
(en)clitic "se" that were quoted by Ertl (1924), or by their known variations in historical Czech;
enclitics in these examples are underlined.
52
in gardenLOC.M.SG fall PART.PRET.ACT.M.SG REFLACC oldNOM.M.SG treeNOM.M.SG suddenly
R3: if rule (R2) is not applied, use the (en)clitic in a preposition of a verb.
(1) c. [ ]1 [rázem]2 [se skácel]3 [starý strom]4
in gardenLOC.M.SG suddenly REFLACC fall PART.PRET.ACT.M.SG oldNOM.M.SG treeNOM.M.SG
All other positions of (en)clitics are a result of miscellaneous factors, such as functional
sentence perspective or rhythmical factors in poetry-like texts; however, these miscellaneous
factors are not strong and systematic enough to be detected in our model. Needless to say,
some word order positions can represent mere fluctuations which occur in any dynamic
system.
The idea of the hierarchy of the above-mentioned rules is
(2015a,b, 2017) description of the development of auxiliary (en)clitics. ies
(2015a,b, 2017) also show that the distribution of (en)clitics in different word order positions
has changed during the historical development of the language and it is influenced by text
type and style2. However, Kosek did not use any statistical tests to observe whether the
differences can be interpreted either as chance or as a consequence of pragmatic factors (i.e.,
time and style). To gain a deeper insight into the issue of (en)clitic properties and their
development, we postulated the following hypotheses:
H1: there are significant differences among the distributions of word order positions in
different historical periods;
H2: there is an increasing proportion of the post-initial position during the course of
the historical development;
H3: there are significant differences among distributions of word order position in
different styles.
The reasons for these hypotheses are presented in detail in Section 2. Further, the
results of statistical testing are used to interpret the relations between the hierarchy of
stochastic rules R1 R3 and pragmatic factors.
The article is organized as follows. Section 2 presents the main characteristics of
(en)clitics in historical Czech. Section 3 describes the language material, and the methodology
is introduced in Section 4. Section 5 presents and discusses the results of the study, and
Section 6 presents the conclusions of the article.
2. (En)clitics in historical Czech
(En)clitics are traditionally defined as expressions which never appear independently and
which are attached to a preceding clause element (word or phrase). The Old Czech enclitics
can be divided into two groups:
1. The archaic (en)clitic particles (e) (i), which are characterized by
different grammatical functions, for example the question marker (li/le/l), the focus marker
( ) or the indefinite determiner (-s(i)).
2. The pronominal and verbal (en)clitics: a) pronominal forms mi , si
, ti ; ho , mu (> se)
2 In the Czech language of the Baroque period, there are obvious differences in the distributions of
(en)clitic positions in different text types, in, e.g. in Bible translations, chronicles, sermons, chap-
books, etc.
On the Development of Old Czech (En)clitics
53
(ACC); b) preterite auxiliary forms (AuxP) Sg 1Ps (nesl) jsem , 2Ps (nesl) ; c)
conditional auxiliary forms (AuxC) Sg 1Ps (nesl) bych , 2Ps (nesl) .
The two groups display different properties as far as word order is concerned (for
more detail see the encyclopedia by Karlík et al., 2017, specifically the entry
klitik, or Kosek, 2017); in this paper we therefore focus solely on (en)clitics from the second
group. This group of (en)clitics is distinctly heterogeneous because it contains forms that
differ in their origin and grammatical status. The pronominal and verbal (en)clitics occur in
historical Czech in several word order positions. These positions are demonstrated by above
quoted Ertl's (1924) examples or their variants with the reflexive pronominal (en)clitic se that
show all in historical Czech known clausal positions of enclitic:
1. in the post-initial position shown in examples (1a.) (1e.),
2. in the contact or verbal adjacent position shown in examples (1b.), (1f.) (1i.),
3. in the position in the middle of a clause without contact with its syntactically superordinate
item shown in example (1j.).
(1) d. [starý strom]1 se [rázem]2 [skácel]3 [ ]4
An old
oldNOM.M.SG treeNOM.M.SG REFLACC suddenly fall PART.PRET.ACT.M.SG in gardenLOC.M.SG
e. [starý]1 se3 [strom]1 [rázem]2 [skácel]3 [v ]4
f. [starý strom]1 [skácel se]2 [ ]3 [rázem]4
g. [starý strom]1 [ ]2 [rázem]3 [skácel se]4
h. [starý strom]1 [rázem]2 [se skácel]3 [ ]4
i. [starý strom]1 [ ]2 [rázem]3 [se skácel]4
j. [starý strom]1 [rázem]2 se [v ]3 [skácel]4
There are two different competing theoretical concepts of the post-initial position in Old
Czech (similar to other Slavic languages, e.g. Serbian and Croatian):
1) position after the first modified phrase (so-
demonstrated by example (1d.);
2) position after the first stressed word in a sentence (so- ern
1995) demonstrated by example (1e.).
A similar distinction may be found within the contact position:
1) postposition3 of an (en)clitic after the syntactically or morphologically superordinate
item4 (the syntactically superordinate phrase is placed in the second position in the
sentence after a modified initial phrase) shown in examples (1b.), (1f.), (1g.);
2) anteposition5 of an (en)clitic before the syntactically superordinate item (the syntact-
ically superordinate item is positioned at the end of a phrase) demonstrated by
examples (1c.) (1h.), (1i.).
There is a third word order pattern of Older Czech clitics: a clitic separated from its super-
ordinate item could appear deep in the middle of a clause (in our analysis this is termed
clause-medial isolated / non-contact position). This position is also demonstrated by example
(1j.).
As our research of the historical development of the Czech pronominal (en)clitics is
only in its early stages, we have not yet collected a sufficient volume of data. For this reason,
we tested the hypotheses from Section 1 on data obtained by Kosek s previous research into
3 This position is also called the post-verbal position.
4 For the sake of simplicity, we classify words which are grammatically/syntactically superior to
pronominal and verbal (en)clitics as clitic regents (Toman, 2004). A similar approach was taken by
dependency.
5 This position is also called the pre-verbal position.
54
the development of the preterite auxiliary (AuxP) in historical Czech Bible translations
(Kosek 2015 a,b, 2017). As we mentioned above, AuxP is also a stable (en)clitic and as such
it can be tested.
To test the hypotheses, we consider only the three positions which represent the
strongest tendency in the word order of (en)clitics.6 Specifically, we distinguish the post-
initial position, i.e. the position after the first phrase of the clause, as shown in the following
example (2) from the oldest Czech Bible translation Bible olomoucká (the enclitic is
underlined; the list of abbreviations of the analyzed Czech Bible translations can be found in
Table 1):
(2) [ ] jest 187
He took our diseases
heNOM.M.SG+FOC beAUX.PRET.3ps.SG ourACC.F.PL diseasesACC.F.PL takePART.PRET.ACT.M.SG.,
the non-postinitial postverbal position demonstrated by example (3) (the verb is bolded):
(3) [Tehdy][ ] ,...
en great storm developed
then greatNOM.F.SG stormNOM.F.SG developeAORIST.ACT.3ps.SG in seaLOC.F.SG,
and the non-post-initial pre-verbal position, illustrated by example (4):
(4) [ ][my a duchovníci][ ] postíme
Why do we and the Pharisees fast often
why weNOM.PL and PhariseesNOM.M.PL often REFLACC fastPRES.3ps.SG.
Table 1
List of the abbreviations of the analyzed historical and new Czech Bible translations
(Kyas, ed., 1981; 1985; 1988; Kyas et al., eds, 1996;
et al., eds., 2009).
BiblOl
Bible olomoucká (ibid.)
BiblMlyn
(the third quarter of the 15th c.). NK v Praze
(Sg. XVII.A.10), Available at Manuscriptorium
http://www.manuscriptorium.com.
BiblBen
. 1506. Venice.
BiblMel
. 4
th
edition, Prague 1570.
BiblKral
. Kralice 1593.
BiblSvat
. Prague 1677
Available at http://www.obohu.cz/bible/.
All other positions occur only very rarely, so we pooled them together in the category
6 A very detailed analysis of these phenomena is presented in Kosek (2015a,b, 2017).
7 A complete translation of the Old Czech examples would lengthen this paper to an unacceptable
extent; for this reason, we generally cite one example of a particular phenomenon, with a simple gloss
The English Bible-translations have been taken from the New English Translation (NET Bible)
(http://www.bible.org/netbible/index.htm) or from the Clementine Vulgate
(http://vulsearch.sourceforge.net).
On the Development of Old Czech (En)clitics
55
Table 2
The distribution of positions of (en)clitics in different translations of the Gospel of Matthew.
The translations are ordered chronologically.
date post-initial
position
pre-verbal
position
post-verbal
position others
14th c. 292 8 34 3
BiblOl 1417 197 9 22 1
BiblMlyn 15th c. 322 8 108 0
BiblBen 1506 462 7 97 1
BiblMel 15704 148 1 25 1
BiblKral 1594 133 1 27 0
BiblSvat 1677 115 0 19 0
1994/2007 142 0 0 0
Table 3
The distribution of positions of (en)clitics in selected books of Bible olomoucká.
post-initial
position
pre-verbal
position
post-verbal
position others
BiblOl Mt
8
197 9 22 1
BiblOl Lk 396 8 36 3
BiblOl Rev 129 2 28 0
BiblOl Gn 434 7 45 0
BiblOl Is 293 12 71 1
BiblOl Sir 99 12 31 0
Figure 1. Proportions of positions of (en)clitics in different translations of the Gospel of
Matthew. The translations are ordered chronologically.
8 See Section 3 for the abbreviations of the Bible books analyzed.
56
Figure 2. Proportions of positions of (en)clitics in selected books of Bible olomoucká. The
books are ranked in descending order of post-initial position.
3. Language material
As has been mentioned above, the development of (en)clitic word order is investigated using
data which served as the material for research of the preterite auxiliary (AuxP) in historical
Czech Bibles (Kosek, 2015a,b, 2017). As the Czech language possesses a long literary
tradition and hence a large number of textual sources, the number of texts to be investigated
needs to be reduced. It appears very convenient to choose a text type that was adapted into
Czech in the very early stages of Czech literary history and one which (being a relatively
stable textual formation) has remained present throughout the history of Czech literature until
today, i.e. Bible translations. The complete text of the Bible was first translated into Czech
during the second half of the 14th century, and it was repeatedly re-translated/adapted (Kyas,
1997; Vintr, 2008) during the following centuries. Since the number of text variants of Old
Czech Bibles is large, the Bible text is extensive, and the data are annotated manually, the
corpus of texts to be investigated has to be reduced to samples consisting of:
1. parts of the New Testament and the Old Testament (Gospel of Matthew Mt, Gospel
of Luke Lk, Book of Revelation Rev, Book of Genesis Gn, Book of Isaiah Is, Book of Sirach
Sir9) from one of the oldest complete Czech Bible translations (Bible olomoucká, from
141710),
2. the Gospel of Matthew from Bible translations originating in different historical
periods ( from the end of the 14th century, Bible olomoucká, Bible mlyná-
from the last quarter of the 15th century, Bible benátská from the beginning of the 16th
9 As the books of the Old Testament differ from the books of the New Testament in their extent, we
reduced our analysis to the following chapters: Gn 1 28, Is 14 40, Sir 1 29.
10 Bible olomoucká is a younger copy of a pre-text that had been written in the middle of 14th century
(Kyas, 1997; Vintr, 2008).
On the Development of Old Czech (En)clitics
57
century, Bible Melantrichova from the second half of the 16th century, Bible kralická from the
turn of the 17th century, Bible svatováclavská from the turn of the 18th century), see Table 3.
The chosen texts present all documented developmental stages of the Czech language from
the oldest Bible translation (e.g. from the 14th century) to the present day. In order to obtain a
complete picture of the development of (en)clitics, we also included among the texts for
analysis the Study Bible Translation), which
represents a translation into the modern Czech language (it does not contain apparent
archaisms).
3. Methodology
The data presented in Tables 1 and 2 were used for testing the hypotheses. They are a typical
example of categorical data (Agresti, 2013). The hypothesis on the homogeneity (in this
context, the homogeneity corresponds to equal proportions of different (en)clitic positions in
all translations) of such data is most often tested by the -test (Snedecor Cochran, 1989).
As only the asymptotic as opposed to exact distribution of the test statistic is known, the
expected frequencies cannot be too small (otherwise results would not be reliable). Our
observed frequencies are (very) low for some (en)clitic positions, and in addition there is no
general consensus on the minimum acceptable values of the expected frequencies. Therefore,
we used simulated p-values (Ross 2006). Consequently, we do not present degrees of freedom
(this notion, relevant if the computation of p-values is based on the asymptotic distribution of
the test statistic, has no sense for simulated p-values). All computations were performed in the
statistical software environment R.11
4. Results
According to hypothesis H1, there should be significant differences among the distributions
of word order position in different historical periods. To test this hypothesis, we used the data
presented in Table 1, and we found significant differences among the distributions ( =
95.092, p-value < 0.001). To obtain a deeper insight into the development of word order, we
tested differences between distributions in pairs of chronologically subsequent texts12.
BiblOl was tested first, followed by the
difference between BiblOl and BiblMlyn and so on. The results are presented in Table 4.
Table 4
Results of testing of differences between distributions of pairs of chronologically subsequent
texts. Values represent adjusted p-values of the chi-squared test (adjusted by the Benjamini-
Hochberg-Yekutieli procedure). Bolded values denote a and
N means that the test cannot be applied (because of the lack of the data for some positions).
BiblOl BiblMlyn
BiblBen
BiblMel
BiblKral
BiblSvat
>0.999
BiblOl 0.0057
BiblMlyn 0.0371
BiblBen >0.999
11 www.r-project.org
12 As we test several hypotheses simultaneously, p-values must be adjusted (see Hochberg
Tamhane, 1987, in general, and Benjamini Yekutieli. 2001, for the procedure selected).
58
BiblMel >0.999
BiblKral N
BiblSvat N
The results show that significant differences are mainly caused by the specificity of BiblMlyn
it differs significantly from both the preceding BiblOl and the following BiblBen. There is
an obvious increase in the proportion of the post-verbal position (see Figure 1) in the analyzed
text from this Bible. This increase can be seen as a consequence of the loss of simple past
forms (aorist and imperfect). These past forms disappeared in the 15th century, and the
translators of BiblBen (and also BiblMlyn) may have used the non-typical post-verbal
positions of AuxP to preserve the ceremonial character of the Biblical language instead of
using the extinct forms of the aorist and imperfect (Kyas 1997, p. 132 133). Further, another
specific position among the Bible translations is occupied by
where the post-initial position is realized exclusively. Differences between the other chron-
ologically subsequent translations are not significant; thus, no evident developmental
tendency can be claimed. This finding, however, falsifies our second hypothesis. In other
words, we cannot see the development of the distribution of the (en)clitics as a gradual
development culminating in the contemporary situation. Instead, the results seem to reveal
that the complete predominance of the post-initial position is not an outcome of a long-term
development but rather a consequence of a relatively abrupt shift. The most likely explanation
is that the contemporary situation is a result of a change that happened in Modern Czech (i.e.
during the 20th century Ertl, 1924, p. 266 267; Avgustinova Oliva, 1997, p. 26; Toman,
2004, p. 74; Kosek, 2011, p. 320).
Finally, we hypothesize that there should be significant differences among distribu-
tions of word order position in different styles. The data presented in Table 2 were used for
the hypothesis testing, and we found significant differences among the distributions ( =
74.895, p-value < 0.001). To obtain a deeper insight into the impact of the style on the
distribution of (en)clitics, we tested differences between distributions in all pairs of texts; the
results of the tests are presented in Table 5 and Figure 3.
Table 5
Results of testing of differences between distributions of all pairs of chosen texts of BiblOl.
Values represent adjusted p-values of the chi-squared test (adjusted by the Benjamini-
Hochberg-Yekutieli procedure). Bolded values denote a significant differen and
N means that the test cannot be applied (because of the lack of the data for some positions).
BiblOl
Mt
BiblOl
Lk
BiblOl
Rev
BiblOl
Gn
BiblOl Is
BiblOl
Lk
>0.999
BiblOl
Rev
0.1512 0.0521
BiblOl
Gn
0.2357 >0.999 N
BiblOl Is 0.0992 0.0062 >0.999 0.0062
BiblOl
Sir
0.0093 0.0062 N N 0.1512
On the Development of Old Czech (En)clitics
59
The results show that the style has some impact on the distribution of (en)clitics.
Roughly half of the pairs differ significantly in this respect. A closer look at Table 5
reveals the extraordinary position of the text Sir. This result is not surprising because Sir is a
wisdom book, which mainly contains advices and instructions. As such, the text of Sir
approaches the form of maximas or aphorisms.
5. Conclusion
The presented study focuses on the development of Czech preterite auxiliary (en)clitics
(AuxP). Based on the data from the previous research (Kosek, 2015a,b, 2017), three hypo-
theses were formulated in Section 1. These hypotheses were statistically tested. The tests
corroborated hypotheses H1 there are significant differences among the distributions of
word order position ) and H3 (
among the distributions of word order position in ). On the contrary,
hypothesis H2 ( -initial position during the
) was rejected.
Nevertheless, these results need to be taken critically and subjected to a further
investigation on language material from other historical Czech Bible translations or other
types of texts:
1. The differences among the distributions of word order positions of AuxP (en)clitics
are evident especially between the historical Czech Bible translations on the one hand
and the modern biblical translation ( ) on the other. This
fact implies that the definitive historical change of Czech auxiliary (en)clitics in the
sole second position clitics was not realized until the modern Czech period (i.e. in the
19th and the 20th century), which was observed by several scholars in the past (see
Section 4 for more details). However, these findings have not been explored
sufficiently thoroughly yet, and they deserve a special attention in the future.
2. Differences among distributions of word order positions of AuxP (en)clitics in the
examined biblical translations were partially influenced by the historical change of
simple past forms of aorist and imperfect. However, the development of the Czech
pronominal (en)clitics is not influenced by such historical change. Hence, the
forthcoming research will examine how much these differences in the word order of
AuxP have been influenced by factors other than the developmental trends of word
order of Czech (en)clitics.
3. The analysis of the selected biblical books both from the New and Old Testament
suggested differences among distributions of word order positions of AuxP in
(en)clitics caused by style. The forthcoming research should look for further mani-
festations of stylistic differences in word order of historical Czech (en)clitics.
One of challenges for a future research is to analyze the pronominal (en)clitics using the
same methodological procedure, with the aim to gain knowledge of a more general
behaviour of (en)clitics in Czech.
Acknowledgement
This study was supported by the project Development of the Czech pronominal (en)clitics
( -02545S).
Pavel Kosek
Masaryk University, Department of Czech Language (Brno), kosek@phil.muni.cz
60
Radek
University of Ostrava, Department of Czech Language (Ostrava), cechradek@gmail.com
Olga Navrátilová
Masaryk University, Department of Czech Language (Brno), olganav@mail.muni.cz
Masaryk University, Department of Czech Language (Brno) / Comenius University,
Department of Applied Mathematics and Statistics (Bratislava), jmacutek@yahoo.com
References
Agresti, A. (2013). Categorical Data Analysis. Hoboken (NJ): Wiley.
Avgustinova, T. Oliva, K. (1997). On the nature of the Wackernagel position in Czech. In:
Junghanns, U. Zybatow, G. (eds.), Formale Slavistik: 25-47., Frankfurt a. M: Vervuert
Verlag.
Benjamini, Y. Yekutieli, D. (2001). The control of the false discovery rate in multiple
testing under dependency. The Annals of Statistics, 29, 1165 1188.
Ertl, V. (1924) 8(9), 257 268; 8(10),
293 309.
Franks, S. King, T.H. (2000). A Handbook of Slavic Clitics. Oxford: Oxford University
Press.
Gebauer, J. (1929). Historická m (Tráv , ed.).
Praha: Nákladem
Gebauer, J. (1958) [1906]. sování.
2nd ed
Grepl, M. Karlík, P. (1998). Praha: Votobia.
Halpern, A. (1995). On the Placement and Morphology of Clitics. Stanford: CSLI.
Havránek, B. (ed.) (1968).
Praha: Academia.
Hochberg, Y. Tamhane, A.C. (1987). Multiple Comparison Procedures. New York:
Wiley.
Jakobson, R. (1971). Les enclitiques slaves. In: Selected Writings. Volume II: Word and
Language: 16-22.. The Hague: Mouton &. Co.
Junghanns, U. (2002). Klitische Elemente im Tschechischen: eine kritische Bestand-
aufnahme. In: Daiber, T. (ed.), Linguistische Beiträge zur Slavistik IX: 117-150. München:
Sagner.
Karlík, P. Nekula, M. Pleskalová, J. (eds.). (2017). Nový .
Praha: Nakladatelství LN.
Kosek, P. (2011). Enklitika . Brno: Host.
Kosek, P. (2015a). Development of Word Order of Preterite Auxiliary Clitics in Old Czech
Bibles. In: Ziková, M. Caha, P. ), Slavic Languages in the Per-
spective of Formal Grammar. Proceedings of FDSL 10.5, Brno 2014. Frankfurt a. M.,
Berlin, Bern, Bruxell: 178-198. P. Lang.
Kosek, P. (2015b)
Linguistica Brunensia, 63(2),
Kosek, P. (2017). Wortstellung des Präteritum-Auxiliars in der alttschechischen Olmützer
Bibel. Die Welt der Slaven, 62(1),
On the Development of Old Czech (En)clitics
61
Kosta, P. Zimmerling, A. (2013). Slavic Clitic Systems in a Typological Perspective. In:
Schürcks, L. Giannakidou, A. Etxeberria, U. (eds.), Nominal Constructions in Slavic
and Beyond: 439-486. Boston: de Gruyter.
Kyas, V. (1997).
Kyas, V. (ed.) (1981).
Praha: Academia.
Kyas, V. (ed.) (1985).
-
Skut Praha: Academia.
Kyas, V. (ed.). (1988).
. Praha: Academia.
Kyas, V. Kyasová, V. , J. (eds.). (1996).
Sirachovec. Padeborn: F. Schöningh.
Lamprecht, A. Bauer, J. (1986). . Praha: SPN.
(2002)
: Hladká, Z. Karlík, P. (eds.), univerzália a
specifika, 4: 325-327. Brno: Nakladatelství LN.
Lenertová, D. (2004). Czech pronominal clitics. Journal of Slavic Linguistics, 12(1 2), p.
135 171.
Merhaut, L. et al. (eds.) (2008). , Part 4/II U
1882 1887. Praha: Academia.
Migdalski, K. (2009). On two types of Wackernagel cliticization in Slavic. In: Reich, J. et al.
(eds.), Formal Approaches to Slavic Linguistics: 147-162. Ann Arbor: Michigan Slavic
Publishers.
Pancheva, R. (2005). The rise and fall of second-position clitics. Natural Language and
Linguistic Theory, 23(1), 103 167.
. et al. (eds.) (2009).
- 2.
kniha Makabejská. Praha: Academia.
- (1996). The placement of Serbo-Croatian clitics: A prosodic Approach.
In: Halpern, A. Zwicky, A. (eds.), Approaching second: second position clitics and
related phenomena:329-445. Stanford: CSLI.
Ross, S.M. (2006). Simulation. Burlington (MA): Academic Press.
Snedecor, G.W. Cochran, W.G. (1989). Statistical Methods. Ames (IA): Iowa State
University Press.
Toman, J. (2004) : Hladká, Z. Karlík, P. (eds.),
univerzália a specifika 5: 73-79. Brno: MU.
(1935). Historická . Praha: Melantrich.
(1956). . Praha: SPN.
Vintr, J.
, Part 4/II U 1882 1887. Academia.
(2009). (1500 1620). Praha:
Karolinum.
Internet sources
Stluka, M. DIAKORP: Diachronní korpus, version 6. Ústav
http://www.korpus.cz
(accessed 30 June 2017).
62
Manuskriptorium. Available at http://www.manuscriptorium.com/cs (accessed 30 June
2017).
New English Translation (NET Bible). Available at
http://www.bible.org/netbible/index.htm) (accessed 30 June 2017).
[on-
http://vokabular.ujc.cas.cz (accessed 30 June 2017).
