Wh movement in Russian agrammatism
In press in Journal of Neurolinguistics
Comprehension and production of movement-derived sentences
by Russian speakers with agrammatic aphasia
Naama Friedmann , Julia Reznick, Dina Dolinski-Nuger, Katya Soboleva
Tel Aviv University
(1) [Which flower] did you draw t ?
Wh-movement is a type of movement that involves the movement of a phrase to the
beginning of the clause (to spec-CP). Wh-movement is crucial for the derivation of relative
clauses (such as I saw the flower that you drew), topicalization structures in which the object
This study explored the way Russian speakers with agrammatism understand and repeat
movement-derived sentences, and examined whether they use morpho-syntactic cues to assist
comprehension. The comprehension of three Russian-speaking individuals with agrammatism
was tested, and their performance was compared to 15 matched control participants. In
addition, the repetition ability of one of the participants was assessed. The study included
topicalization structures, relative clauses, and SVO sentences. The individuals with
agrammatism performed at chance level on topicalization structures and object relative
clauses, whereas their comprehension of SVO sentences and subject relatives was
significantly better and above chance. Their comprehension of topicalization structures was
poor although all sentences included morphological cues of case marking on the topicalized
object and on the subject. Case and gender morphology on the relative pronoun did not lead
to better comprehension of object relative clauses compared to relative clauses in which
gender inflection could not be used as a cue for interpretation. The repetition task indicated a
considerable difficulty in repetition of sentences that include syntactic movement to high
nodes of the syntactic tree.
moves to the beginning of the sentence (This flower, the girl drew), and Wh-questions
(example 1).
Studies of the comprehension of phrasal movement, including Wh-movement, found, in
various languages, that individuals with agrammatic aphasia can understand simple sentences
without syntactic movement, but have difficulties in the comprehension of semantically
reversible movement-derived non-canonical sentences, such as passive sentences, object
relative clauses, object cleft, topicalization, and object which questions (Friedmann &
Shapiro, 2003; Grodzinsky, 1989, 2000, 2006; see Grodzinsky, Piñango, Zurif, & Drai, 1999
for a review, see also O’Grady & Lee, 2005 for a discussion of the definition of canonicity).
These findings led Grodzinsky (1990) to suggest the Trace Deletion Hypothesis (TDH),
according to which the difficulty in comprehension in agrammatism is caused by a deficit in
1. Introduction
the representation of movement traces, which results in inability to assign a thematic role to
Many studies indicate that individuals with agrammatic aphasia show difficulties in the
the trace and from there to the moved phrase.
comprehension and production of structures derived by syntactic movement. The way
Nevertheless, studies of agrammatic comprehension found that when the sentence in not
Russian individuals with agrammatic aphasia understand and produce movement-derived
semantically reversible, it is understood even if it is derived by movement. Furthermore,
sentences has not been investigated until now. Russian has a relatively free word order, and
individuals with agrammatism consistently perform well in movement-derived sentences that
rich inflection and case morphology that provides cues to healthy speakers about the syntactic
keep the canonical structure of the constituents, such as subject relative clauses, subject
role of the constituents in sentences. This characteristic of Russian permits the examination of
clefts, and subject questions. If indeed they have a problem in movement, how come they still
the ability of individuals with agrammatism to use morphological cues for comprehension of
understand movement-derived sentences that are non-reversible or canonical? Since the
movement-derived sentences. This research focuses on two related issues in the context of
pioneering studies of agrammatism (Caramazza & Zurif, 1976; Zurif & Caramazza, 1976), it
agrammatism in Russian: the comprehension and production of movement-derived sentences,
has become clear that individuals with agrammatic aphasia are able to use cues for the
and the role of morpho-syntactic cues in the comprehension of these sentences.
comprehension of movement-derived sentences. One such cue is their lexical-semantic
Syntactic movement is a movement of a constituent from its base-generated position to some
knowledge and world knowledge. Sentences that are semantically irreversible (such as This is
other position in the sentence. When a constituent moves it leaves a trace (t) in its original
the apple that the boy ate), allow comprehension on the basis of the comprehension of the
position. The verb assigns the thematic role to the trace and the thematic role is transferred
words apple, boy, eat, even when the syntactic structure of the sentence is beyond the
from the trace to the moved constituent via a thematic chain (see example 1).
syntactic ability of the hearer. An additional cue used by individuals with agrammatism is the
We thank Inna Lavro, Lola Karsenti, Aviah Gvion, Michal Biran, Hagar Levy-Eitan, Irena Botwinik, and
Esther Ruigendijk for their helpful comments and discussions of this research and manuscript. This research was
supported by the Israel Science Foundation (grant No. 1296/06, Friedmann).
order of sentential constituents is agent, verb, theme. Again, even if a sentence is derived by
order of constituents in the sentence. In English and many other languages, the canonical
2
Wh movement in Russian agrammatism
3
Wh movement in Russian agrammatism
syntactic movement, individuals with agrammatism can still understand it if it keeps the
layer is the highest layer in the sentential hierarchy.
canonical order of constituents. This is why, although both subject relatives and object
The agrammatic deficit in production can thus be explained by an inability to construct the
relatives are derived by Wh-movement, individuals with agrammatism understand subject
syntactic tree up to its highest nodes (Tree Pruning Hypothesis, Friedmann, 2001, 2002,
relatives, which keep the order agent, verb, theme (This is the grandmother that
2006b). According to some analyses, this generalization, regarding deficits in the high nodes
kisses the
girl). However, in object relatives the noun phrase that moves is the object of the embedded
of the syntactic tree, accounts also for the deficit in the comprehension of sentences derived
sentence, and because it is the theme of the action and it moves to a position before the agent,
by Wh-movement such as object relatives and topicalization structures (Friedmann, 2006a,
a non-canonical order is created (as in the sentence This is the grandmother that the girl
2008; Luzzatti & Guasti, 2000).
kisses ). Thus, individuals with agrammatism fail to understand object relatives, but perform
Hence, to derive predictions with respect to agrammatic comprehension and production of
relatively well in tests of subject relatives. A similar account explains the differences in
movement-derived sentences in a given language, it is necessary to examine several
comprehension of subject and object Wh-questions, and subject and object clefts (Dick et al.,
variables: whether there is movement in the sentence, whether the sentence maintains the
2001; Friedmann & Shapiro, 2003; Grodzinsky, 1989, 2000; Hickok & Avrutin, 1996; see
canonical order of constituents, whether it is semantically reversible, to which node the
Grodzinsky et al., 1999 for a review). A somewhat similar approach suggests that the
constituent moves, and which nodes are involved in the sentence.
comprehension of canonical (or “isomorphic”) orders is less taxing for the syntactic processor
than noncanonical orders, and thus yield better performance (O’Grady & Lee’s, 2005
1.1. Movement and morphology in agrammatism in Slavic languages
Isomorphic Mapping Hypothesis). In contrast to semantic reversibility and canonicity cues,
Until now, as far as we know, no study has explored the comprehension of movement-
some studies indicate that agrammatic speakers cannot use morpho-syntactic cues to assist
derived sentences in Russian agrammatism. Some information comes from other Slavic
their interpretation of movement-derived sentences (see Burchert, Friedmann, & de Bleser,
languages. Lukatela, Shankweiler, and Crain (1995) studied the comprehension of relative
2003 for German and Hebrew; Friedmann & Shapiro, 2003 for Hebrew).
clauses in 7 Serbo-Croatian speakers with agrammatism. Subject relatives were compared to
Agrammatic speech production is characterized by difficulties in structures that involve the
object relatives. On the group level, the comprehension of center-embedded object relatives
high nodes of the syntactic tree. These difficulties are manifested in errors of tense inflection,
was at chance, whereas center-embedded subject relatives were above chance. On the
in difficulty in the production of verb movement to second sentential position, in inability to
individual level, only 3 of the 7 participants performed at chance on center-embedded object
produce embedded sentences (with or without movement). Agrammatic speech is also
relatives. In addition, their participants’ performance on both right-branching subject- and
characterized by difficulty in producing topicalization structures, Wh-questions, and even
object relatives was above chance, however, these sentences were not semantically reversible
yes/no questions in languages that require movement to higher nodes in this type of questions
in the relative clause, and therefore the results are hard to interpret. Thus, the results of the
(Friedmann, 2001, 2002, 2006b; Ruigendijk, Kouwenberg, & Friedmann, 2004). When
study are not conclusive with respect to the ability of Serbo-Croatian speakers with
speakers produce or comprehend sentences, they represent them in syntactic trees (Chomsky,
agrammatism to understand movement-derived sentences, although they do indicate that at
1995; Pollock, 1989). It is widely accepted within the GB framework that the phrasal
least some of the participants have difficulties with such sentences. Another study in Serbo-
architecture of the sentence consists of three main structural layers (Chomsky, 1986, 1995;
Croatian examined the ability of individuals with agrammatism to judge ungrammatical
Rizzi, 1997). The basic model includes three layers: (1) the lexical layer VP (verb phrase),
movement-derived sentences with a filled gap position (Milekic, BoškoviN, Crain, &
which contains the subject, the verb and the object, (2) the inflectional layer IP (inflectional
Shankweiler, 1995). The agrammatic participants in this study had difficulties rejecting these
phrase), which is responsible for verb inflections and for the agreement between the verb and
ungrammatical sentences, in comparison to the control subjects. Several studies show that
the subject and (3) the complementizer CP-layer, which is responsible for embedding and for
Slavic-speaking agrammatic aphasics are sensitive to the morphology of case and gender
constituents that move to beginning of the sentence such as Wh-morphemes, verbs that move
agreement, and transitive verb subcategorization (Akhutina et al., 2001; Lukatela, Crain, &
to second sentential position, and auxiliaries in yes/no questions in some languages. The CP-
Shankweiler, 1988; Milekic et al., 1995; Smith & Bates, 1987; but see Luria, 1975 for
4
Wh movement in Russian agrammatism
5
Wh movement in Russian agrammatism
findings on difficulties in grammaticality judgment of sentences with ungrammaticality is
that Russian-speaking agrammatic aphasics did not omit case morphology when such
related to case agreement, tense and gender). Smith and Mimica (1984) tested the role of
omission creates a bare noun stem that is an impossible word, but rather substituted case
morpho-syntactic cues in agrammatic comprehension, and found that although speakers of
morphology (Grodzinsky, 1990, 2000; Ruigendijk, 2002).
Serbo-Croatian with agrammatism are relatively sensitive to case morphology, their ability to
use these cues for comprehension is poorer than that of the control participants. Smith and
1.2. Russian syntax: some relevant aspects
Bates (1987) reported the same with respect to gender morphology.
Russian is an SVO3 language (Bailyn, 2002, 2004) with a relatively free word order. For
Researchers who examined the spontaneous speech of Russian-speaking agrammatic
example, all possible word order permutations created by movement from the SVO sentence
aphasics found that agrammatic speech is characterized by omission of verbs, prepositions,
(3a) create grammatical sentences (sentences 3b-f), all with the same basic meaning, but with
personal pronouns, and objects, word order errors, preference to produce verbs in the present
different discourse functions (Bailyn, 2002; King, 1995; Rudnev, 1968; Timberlake, 1993).
tense, frequent substitutions of various cases to nominative case (which is the default case in
(3)
Russian, Babyonyshev, 1993), and few errors in gender, number, and tense morphology1
(Akhutina, 1991 in Ruigendijk, 2002; Tsvetkova & Glozman, 1975). Luria provided a similar
a. Devochka kormit sobaku
girl-NOM feeding dog-ACC
"The girl is feeding the dog"
description of agrammatic speech and difficulty with case production. Example (2) shows a
b. devochka sobaku kormit
SOV
target sentence and two agrammatic attempts to repeat it (in Luria, 1982, p. 223)2.
c. sobaku devochka kormit
OSV
(2) Target: Mal’chik
udaril sobaku
boy-NOM-sg hit
dog-ACC-sg
“The boy hits the dog”
d. sobaku kormit devochka
OVS
e. kormit sobaku devochka
VOS
f. kormit devochka sobaku
VSO
Because word order in Russian is relatively free, healthy Russian-speaking individuals
Response: Mal’chik … sobaka
boy-NOM-sg dog-NOM-sg
“boy… dog”
Mal’chik … sobaka … udarit’
boy-NOM-sg dog-NOM-sg hit-INF
“boy… dog… to hit”
According to Ruigendijk (2002, Ruigendijk & Baauw, in press; Ruigendijk & Friedmann,
2008), case itself is not damaged in agrammatism, including Russian agrammatism, but the
realization of case morphology depends on the existence of case assigners in the sentence,
and hence might be incorrect when the relevant case assigner is missing. It was also found
understand the syntactic function of sentential constituents on the basis of case morphology
and sometimes verb inflection. The verb agrees with the subject (in person and number in the
present and future tense, and in number and gender in past tense), and this enables the
identification of the subject. The main cue Russian speakers use to identify the syntactic
function of sentential constituents is case marking, particularly when the word order in the
sentence deviates from the canonical SVO order4. Case is a mechanism that specifies the
syntactic relationship between, for example, a verb and the subject and object. It marks the
function of each noun phrase in the sentence. The subject receives nominative case, and the
direct object usually receives accusative case. The Russian case system includes six main
cases: nominative, accusative, genitive, dative, instrumental and prepositional case
(Botwinik-Rotem, 2008)5. The morphological realization of the case system is determined
1
The passive form is also produced very rarely by the agrammatic speakers, but this finding is difficult to
interpret because no information from individuals without impairment is available for comparison.
We use the following transliteration rules and abbreviations: P=a R=b S=v T=g U=d V=e ë=jo X=zh Y=z
Z=i [=j \=k ]=l ^=m _=n `=o a=p b=r c=s d=t e=u f=f g=kh h=c i=ch j=sh k=shh l=’’ m=y n=’
o=eh p=ju q=ja. NOM=nominative case, ACC=accusative case, neutr=neutral, sg=singular, masc=masculine,
fem=feminine, 1st=declension class I, 2nd=declension class II. Because Russian does not distinguish between the
and a, we used the in our translations.
2
3
But see King (1995) for the suggestion that Russian is a VSO language.
When morphology does not distinguish between the subject and the object, for example, when both appear in
an ambiguous nominative/accusative form, the tendency is to interpret sentence of the type NP-V-NP as SVO
(OVS interpretation is also possible in specific contexts) (from Rudnev, 1968).
5
Two additional cases exist, partitive and locative, but they have distinct markings only for a limited number of
lexical items, and are identical in most occurrences to the genitive and prepositional cases, respectively.
4
6
Wh movement in Russian agrammatism
7
Wh movement in Russian agrammatism
according to the declension class to which the noun belongs. Example (4) shows the way
movements, but because all of them agree that the target is at or above IP, this will not be
singular animate nouns from two declension classes (class I and class II) receive the two
relevant to the current study6.
cases relevant to this research, nominative and accusative.
This research examined whether agrammatic aphasics understand topicalization sentences,
whether there is a difference between topicalization with and without verb movement, and
(4) a.
whether morphological cues (case marking on the noun and on the modifier) assist their
Suffix
Case
nominative
accusative
b. Devochka
class I
a
u
class II
-
kormit sobaku
girl-NOM-1st feeding dog-ACC-1st
"The girl is feeding the dog"
c. Mal'chik-
kormit klouna
sentence comprehension.
1.2.2. Relative clauses
Another structure derived by Wh-movement is the relative clause. We focus on headed
relative clauses, which are characterized by a nominal head that is realized phonetically, and
by the relative pronoun kotoryj ("which"), see example (7). This is the common way to form
relative clauses in Russian (Borras & Christian, 1971; Peshkovskij, 1956; Rudnev, 1968;
boy-NOM-2nd feeding clown-ACC-2nd
Smirnitsky, 1975; Spencer, 1993; see Pugh, 2006 for a report of relative clauses in modern
"The boy is feeding the clown"
Russian).
Below we shortly describe the Russian structures that are tested in this study: topicalization
(7)
structures and relative clauses and their syntactic analyses.
Ehto
devochka
kotoraja
risuet malchika
this-neutr girl-NOM-sg who-NOM-fem-sg draws
boy-ACC
"This is the girl who draws the boy"
1.2.1. OSV and OVS topicalization structures
OSV and OVS topicalization structures are derived by movement of the object from the basic
An important property of headed relative clauses in Russian is that the (singular) relative
SVO structure. In both structures the object moves to the beginning of the sentence, before
pronoun is inflected for case, number, and gender (and it sometimes shows also animate /
the subject and the verb. In OVS sentences the verb also moves to a position before the
inanimate property). The case of the relative pronoun matches the case of the trace in the
subject (see examples 5, 6, both meaning "The girl is feeding the boy").
relative clause. The gender and number inflection of the relative pronoun match the gender
(5)
theme are singular and animate, there is one-to-one matching between the morphological
and the number of the head of relative clause. Crucially, in sentences where the agent and
(6)
OSV
Mal'chikai devochka kormit ti
suffix of the relative pronoun and the morpho-syntactic categories of case and gender of the
boy-ACC girl-NOM feeding
moved constituent, indicating the gender of the relative head and the case of the moved
OVS
element within the relative clause, as seen in (8).
Mal'chikai kormitv devochka tv ti
boy-ACC feeding girl-NOM
6
All analyses of topicalization in Russian agree that topicalization structures are derived by
object movement to the position before the subject in the left edge of IP or above IP, and that
the verb raises in OVS sentences. The analyses differ with respect to the target of these
According to Bailyn (2002, 2004), OSV is derived by object adjunction to IP or above IP, and according to
King (1995), OSV is derived by object adjunction to IP. OVS includes verb raising to I and object movement to
Spec-IP (Bailyn, 2002, 2003, 2004; King, 1995), or its adjunction to IP (King, 1995), or to a location above IP
(Spec-RefP according to Erechko, 2002; Spec-TopP according to Ionin, 2001). Lavine and Freidin (2001)
assume that the topicalized object is located in Spec-CP or adjuncted to IP; Junghanns and Zybatow (1997)
assume object adjunction to AgrsP. (Also see Sekerina, 1997 for a comparative analysis of major approaches to
word order variations in Russian.)
8
Wh movement in Russian agrammatism
9
Wh movement in Russian agrammatism
(8)
Gender
Case
(10) Ehto
malchiki
[kotorogo ti] j
devochka risuet tj
masculine
feminine
this-neutr boy-NOM-sg who-ACC-masc-sg girl-NOM draws
nominative
kotoryj
kotoraja
"This is the boy who the girl draws"
accusative
kotorogo
kotoruju
Very few studies within the generative paradigm framework undertook the mission of
analyzing headed relative clauses in Slavic languages in general and in Russian in particular.
The two prevalent analyses for relative clauses in other languages were implemented by
different researchers for Russian as well. One analysis assumes that the constituent that
moves from within the embedded clause is an operator (empty or a Wh-phrase), which
moves to Spec-CP (Haegeman, 1994, see example (9)). The operator gets its interpretation
according to the predication rule, which assigns identical index to the operator and to the
head of the relative clause (Williams, 1980; see Browning, 1987 for details of this
mechanism). This analysis of relative clauses can also be implemented to Slavic languages
(see Spencer, 1993 for Russian; Szczegielniak, 2005a, 2005b for Russian and Polish,
assuming movement to Spec-TopP; Lavine, 2003 for Ukrainian).
Thus, under both analyses, the movement that derives the relative clause involves CP.
According to one approach it moves to CP, according to the other, it moves to CP and
continues outside of CP. Furthermore, both approaches require the involvement of CP in the
agreement between relative clause head and relative pronoun.
The order between the subject and the verb in the relative clause is free, with a certain
preference to VS order (according to judgments of 15 native speakers of Russian) (see
example (11)). There is no analysis in the literature for this VS structure in relative clauses in
Russian, but analyses of VS order in main clauses suggest that the verb moves to I (Bailyn,
2004, King, 1995; but see also Kallestinova & Slabakova, 2007)8.
(11) Ehto
malchik
kotorogo
risuet devochka
this-neutr boy-NOM-sg who-ACC-masc-sg draws girl-NOM
"This is the boy who the girl draws"
Object relatives are derived by Wh-movement to CP that creates a non-canonical order, and
(9)
Ehto
malchikj
kotorogoj
devochka risuet tj
this-neutr boy-NOM-sg who-ACC-masc-sg girl-NOM draws
"This is the boy who the girl draws"
Another analysis of relative clauses is the head-raising analysis. According to this approach,
rooted in Vergnaud (1974) and Kayne (1994), the head of the relative clause itself, rather
than the operator, moves from the embedded sentence. Citko (2004) and Szucsich (2003)
present the following versions of this analysis for Slavic languages: Wh-DP (which includes
the relative pronoun as head D0 and the relative clause head as its complement) moves from
the position of the complement of the verb to Spec-CP, and then the NP further raises from
Spec-CP to the position of the relative clause head, which is external to the CP (if the same
analysis applies for Russian, it would have the form as in example (10), adapted to Russian
from the Polish analysis in Citko, 2004) 7.
7
The case of the relative head may differ from the case of the relative pronoun, which is the case of the trace in
the relative clause. Bianchi (2000a, 2000b) suggests that case deletion precedes case assignment during the last
derivation step (but also see Borsley, 1997, 2006).
thus they are expected to be impaired in agrammatism. The agreement between the relative
head and the relative pronoun also requires the CP node. If individuals with agrammatism are
able to get to CP node, they are expected to be able to use the case and inflection cues on the
relative pronoun to identify the agent and the theme, even if the sentence is derived by
movement. If they cannot access CP, they should not be able to use the relative pronoun
morphological cues.
Taking into account the variety of movement-derived structures in Russian on the one hand,
and the morphological richness on the other hand, it would be interesting to examine how
Russian individuals with agrammatism understand movement-derived sentences with
morphological cues. Specifically, the question is whether morphological cues such as
agreement in case and gender between the relative clause head and the relative pronoun in
relative clauses, as well as case marking in noun phrases in topicalization sentences, may help
8
An argument for V-C movement in Russian relative clauses is the following: unlike the relative pronoun
kotoruju, which resides in spec-CP, the complementizer chto resides in C. The ungrammaticality of V-C
movement in relative clauses with chto (*Vot ehta devushka, chto vljubljon ja, This is the girl that love I)
indicates that the verb in VS relative clauses (with kotoruju) moves to C. When C is occupied by chto, the verb
cannot move to C, and the sentence is therefore ungrammatical. We thank Lola Karsenti for this argument.
10
Wh movement in Russian agrammatism
individuals with agrammatism comprehend the thematic relations in movement-derived
sentences. It is also interesting whether they understand better sentences in which the
morphological cues can assist comprehension (for example, feminine gender inflection when
there is one feminine NP in the sentence, and one masculine), compared to sentences in
11
Wh movement in Russian agrammatism
Table 1
Background information on the agrammatic and control participants
Group
n
Agrammatic
which the inflection cannot be used as a cue (for example, when all the NPs in the sentence
are of the same gender).
Participants
Age
TA
52
Gender Years in
Israel
female
10
Education
(years)
at least 10
Hand
right
Months
post onset
3
AB
16
female
6
9
right
13
PS
52
male
10
at least 8
right
5
Because both topicalization sentences and relative clauses include movement to higher nodes
of the syntactic tree, it is also interesting to examine whether Russian agrammatic aphasics
are able to produce these sentences.
2. Method
Control
5
Matched to TA
5
Matched to AB
5
Matched to PS
2.1. Participants
The participants were three native speakers of Russian with agrammatic aphasia, and 15
45-61
M=55
17-26
M=20
41-57
M=48
females
females
males
6-14
M=9
5-15
M=8
6-15
M=11
The control group included 15 Russian-speaking individuals without language impairment
native speakers of Russian without language impairment. All three individuals with
who lived in Israel. Five control participants were matched to each participant with
agrammatism lived in Israel, but their main language of communication was Russian. AB
agrammatism in age, sex, and number of years in Israel (see bottom of Table 1).
spoke Hebrew as a second language, TA and PS spoke almost only Russian and had very
limited knowledge of Hebrew. Because 14% of the population in Israel were born in Russian-
2.2. Comprehension
speaking countries, Russian speakers who immigrated to Israel can continue speaking
Russian exclusively or almost exclusively without the need to use Hebrew. They can listen to
2.2.1. Material
Russian TV and radio and read Russian newspapers, and speak Russian in their social
The comprehension task included 200 semantically reversible sentences: 30 simple SVO
settings with only limited exposure to Hebrew.
sentences; 60 topicalization sentences: 30 in OSV order and 30 in OVS order; and 110
The speech of the agrammatic participants was characterized by considerable production
relative clauses: 54 subject relatives and 56 object relatives. Half of the object relatives
difficulties: non-fluent speech, limited production of verbs and complex sentences, and errors
included VS order in the embedded clause, and half – SV order. To test for the effect of a
in inflectional morphology. Their comprehension on the single word level was unimpaired.
morphological gender cue, half of the relative clauses included figures of different gender,
They were diagnosed with Broca's agrammatic aphasia using the Russian version of ILAT
and half had two figures with the same gender. Thus – all relative clauses included gender
(Schechter, 1965) and an adaptation of the syntactic test battery BAFLA (Friedmann, 1998).
inflection but only in half of them this inflection could be used to identify thematic relations.
All agrammatic participants had right hemiplegia (see the top part of Table 1 for background
In relative clauses with different genders, half of the relative pronouns were masculine and
information on the agrammatic participants).
half were feminine. All sentences were reversible and included a transitive action verb, and
TA had an ischemic stroke that resulted in left MCA territory infarction, PS had a stroke that
all the NPs were singular. The sentences were presented in random order. Table 2 includes
resulted in a lesion in the posterior part of her left frontal lobe, and AB sustained a
examples of each type of sentence.
penetrating head trauma in a terror attack, which led to left frontal fractures and extensive
ischemia in the left hemisphere, including severe damages in language areas.
12
Wh movement in Russian agrammatism
13
Wh movement in Russian agrammatism
Table 2
correctly described the sentence. The task was performed without time limit, and the
Examples of target sentences in the sentence-picture matching task
sentences were repeated as many times as the participant requested.
SVO
OSV
OVS
Same gender
Different gender
Same gender VS
Same gender SV
Different gender VS
Different gender SV
Simple sentence
devochka risuet mal'chika
girl-NOM drawing boy-ACC
"The girl is drawing the boy."
Topicalization
ehtu
zhenshhinu devochka risuet
this-ACC-fem-sg woman-ACC girl-NOM drawing
"This woman, the girl is drawing (her).”
ehtu
zhenshhinu risuet devochka
this-ACC-fem-sg woman-ACC drawing girl-NOM
"This woman, the girl is drawing (her).”
Subject relative
ehto
devochka kotoraja
risuet
zhenshhinu
this-neutr girl-NOM who-NOM-fem-sg drawing woman-ACC
"This is the girl who-she is drawing the woman."
ehto
devochka kotoraja
risuet
mal'chika
this-neutr girl-NOM who-NOM-fem-sg drawing boy-ACC
"This is the girl who-she is drawing the boy."
Object relative
ehto
devochka kotoruju
risuet zhenshhina
this-neutr girl-NOM who-ACC-fem-sg drawing woman-NOM
"This is the girl whom-her the woman is drawing."
ehto
devochka kotoruju
zhenshhina risuet
this-neutr girl-NOM who-ACC-fem-sg woman-NOM drawing
"This is the girl whom-her the woman is drawing."
ehto
mal'chik kotorogo
risuet
devochka
this-neutr boy-NOM who-ACC-masc-sg drawing girl-NOM
"This is the boy whom-him the girl is drawing."
ehto
mal'chik kotorogo
devochka risuet
this-neutr boy-NOM who-ACC-masc-sg girl-NOM drawing
"This is the boy whom-him the girl is drawing."
2.2.2. Procedure
Comprehension was assessed using a binary sentence-picture matching task. The participant
heard a sentence read by a native speaker of Russian, and saw two pictures on the same page,
one above the other. In one picture the roles matched the sentence, and in the other picture
the roles were reversed (Figure 1). The participant was asked to point to the picture that
Figure 1. An example of a picture pair used in the comprehension task
2.3. Repetition
2.3.1. Material
The repetition test included 38 semantically-reversible sentences: 6 SVO sentences, 6 OSV
sentences, 5 OVS sentences, 9 subject relatives, and 12 object relatives. Five of the object
relatives included a VS order in the embedded clause, 7 object relatives were in SV order. All
the sentences included a verb in the present tense, and singular NPs.
2.3.2. Procedure
Only TA participated in the repetition task. The sentences were read to her by a native
Russian speaker, and she was requested to repeat each sentence as accurately as possible. The
task was performed without time limit, and the sentences were repeated as many times as she
requested.
2.4. Data analysis – comprehension and production
The results were analyzed on the group level as well as for each individual participant. We
only showed the group results when all the individuals had the same tendencies in their
performance. To examine the effect of sentence structure, we compared the performance in
different structures using Chi square or Fisher's exact test (according to the number of items
14
Wh movement in Russian agrammatism
15
Wh movement in Russian agrammatism
in the comparison) on the group level and the individual level. Chi squared comparisons
Table 3
appear in the tables with v2 and p values, Fisher’s exact results are reported with the p value.
Comprehension of SVO and topicalization sentences of the orders OSV and OVS. Percentage
To compare the performance of each experimental participant to the control group we used
correct and number of correct responses out of the total number of sentences of each type9
Crawford and Howell’s (1998) t-test. For the comparison to chance level we used the
binomial distribution. An alpha level of 0.05 was used.
3. Results
Participant
SVOtopicalization
OSV-OVS
SVO
OSV
OVS
TA
75% (21/28)
54% (14/26)
46% (13/28)
v2 = 4.75, p = .03
v2 = 0.3, p = .59
AB
86% (25/29)
68% (19/28)
69% (20/29)
Fisher’s p = .06
v2 = 0.008, p = .93
PS
67% (20/30)
40% (12/30)
57% (16/28)
v2 = 2.7, p = .1
v2 = 1.7, p = .19
Total
76% (66/87)
54% (45/84)
58% (49/85)
2
2 = 10.04, p = .002
22 = 0.28, p = .59
3.1. Comprehension
3.1.1. Control group
The control participants performed between 95% and 100% correct across all sentence
structures. The performance of the three sub-groups, each matched to a different agrammatic
participant, was compared using one-way ANOVAs, per each structure. No significant
difference was found between the sub-groups in any of the tested structures, and therefore the
control group was treated as a homogeneous group, and the performance of the individuals
with agrammatism was compared to the whole group of 15 control participants.
With respect to performance relative to chance level, all three participants performed above
chance level on the SVO sentences without movement, and two, TA and PS, performed at
chance level on the topicalized OSV and OVS structures (p < .05). AB's performance was
above chance also in the topicalized structures, but she was still considerably impaired on the
comprehension of the topicalization sentences in comparison to the SVO sentences.
Analyzing the performance of the three participants together, they performed significantly
above chance in the comprehension of SVO sentences, but at chance on the topicalized OSV
and OVS sentences.
3.1.2. Individuals with agrammatism
The control participants performed between 95% and 100% correct on SVO, OSV and OVS
3.1.2.1. SVO, OSV, and OVS sentences
sentences. Their average performance was 100% in SVO, and 99.5% in each topicalization
The performance of each participant and of the group in each type of structure, and the
structure. In each structure, the performance of each participant with agrammatism was
statistical comparisons are presented in Table 3. The individuals with agrammatism, as a
significantly poorer than that of the control group, t(14) > 22.72, p < .0001.
group, performed significantly better on the comprehension of SVO sentences than on the
topicalization sentences in either OSV and OVS orders (v2 = 9.32, p = .002, and v2 = 6.44, p
3.1.2.2. Subject relatives
= .01, respectively), and significantly better on SVO than on both topicalization structures
The comprehension of subject relatives was significantly above chance, both for subject
together. No significant difference was found between the two topicalization structures
relatives with a morphological gender cue, and for subject relatives without this
(Table 3). Each participant showed the group pattern. For TA the comprehension of SVO was
morphological cue. It was also significantly above chance for the two types together.
significantly better than the comprehension of the topicalization structures, for PS the
Comprehension was above chance for the group level and for each participant with
comprehension of SVO sentences was only significantly better than OSV sentences
agrammatism.
comprehension (v2 = 4.29, p = .04), and for AB the difference between SVO and the two
The performance on subject relatives with a morphological gender cue did not differ from
topicalization structures was just below significance. None of the agrammatic participants
subject relatives without this cue. Namely, both sentence types included gender inflection,
showed any difference between the two topicalization structures.
but it could be used to identify the thematic roles in the sentence only when the two NPs in
9
The small difference in the total number of sentences per participant resulted from some experimental
problems.
16
Wh movement in Russian agrammatism
17
Wh movement in Russian agrammatism
the sentence were of different genders. Nevertheless, there was no indication of better
Table 6
comprehension of the sentences with the gender cue - this held on the group level, as well as
Comprehension of object relatives with and without a gender cue
for each participant with agrammatism (Table 4). Therefore, we treated subject relatives of
Participant
Same gender
Different gender
TA
48% (12/25)
Table 4
AB
Comprehension of subject relatives
both types together for comparison to the other structures.
52% (13/25)
Total
object relatives
50% (25/50)
Same-different
v2 = 0.08, p = .78
71% (17/24)
60% (15/25)
65% (32/49)
v2 = 0.63, p = .43
PS
44% (12/27)
58% (15/26)
51% (27/53)
v2 = 0.93, p = .33
Total
54% (41/76)
57% (43/76)
55% (84/152)
22 = 0.11, p = .74
Same
Different
Total
gender
gender
subject relatives
Same-different
TA
86% (19/22)
70% (16/23)
78% (35/45)
Fisher’s p = .16
AB
78% (18/23)
75% (18/24)
77% (36/47)
Fisher’s p = .53
the comprehension of object relatives in which this cue could not be used, both on the group
PS
72% (18/25)
83% (19/23)
77% (37/48)
Fisher’s p = .19
level, and on the individual level (Table 6). Therefore, we treated both types of object
Total
79% (55/70)
76% (53/70)
77% (108/140)
22 = 0.16, p = .69
relatives as one group in further analyses. On the group level, comprehension of object
Participant
The control group performance on subject relatives ranged between 97.5% and 100% correct,
with an average performance of 99.7%. The performance of each participant with
agrammatism was significantly poorer than that of the control group, t(14) > 23.35,
p < .0001.
The comprehension of object relatives with the morphological gender cue did not differ from
relatives was at chance. TA and PS also performed at chance level, and AB performed above
chance.
The control group performed between 97.5% and 100% on object relatives, with an average
performance of 99.8%. The object relative comprehension of each participant with
agrammatism was significantly poorer than that of the control group, t(14) > 48.14,
p < .0001.
3.1.2.3. Object relatives
Object relatives with VS order did not differ from object relatives with SV order on the group
level, and on the individual level (Table 5). Therefore, we treated both types of object
3.1.2.4. A comparison between subject relatives and object relatives
relatives as one group in further analyses.
The comprehension of subject relatives was significantly better than that of object relatives
for the group, and individually for TA and PS, AB comprehended subject relatives better than
Table 5
object relatives, but this difference did not reach significance (see Table 7).
Comprehension of VS and SV object relatives
Participant
TA
VS
58% (14/24)
SV
42% (11/26)
VS-SV
2
v = 1.28, p = .26
2
Table 7
Comprehension of subject and object relatives
AB
67% (16/24)
64% (16/25)
v = 0.04, p = .84
PS
50% (7/14)
42% (5/12)
Fisher’s p = .49
TA
78% (35/45)
50% (25/50)
v = 6.57, p = .01
Total
60% (37/62)
51% (32/63)
22 = 1, p = .32
AB
77% (36/47)
65% (32/49)
v2 = 1.48, p = .22
PS
77% (37/48)
51% (27/53)
v2 = 7.41, p = .006
77% (108/140)
55% (84/152)
The comprehension of object relatives was at chance, both for object relatives with
morphological cue of gender and for object relatives without this cue. This result was valid
both on the group level and on the individual level (except AB for the condition without
morphological cue in which she performed above chance) (Table 6).
Participant
Total
Subject relative
Object relative
Subject-object
2
22 = 15.49, p < .0001
18
Wh movement in Russian agrammatism
19
Wh movement in Russian agrammatism
No significant difference was found on the group level between the comprehension of
Table 8
topicalized sentences (OSV and OVS together) (56%) and that of object relatives (55%),
TA: types of repetitions of movement-derived sentences
2
2
v = 0.004, p = .95. These structures did not differ on the individual level either, v x 0.12,
Structure of incorrect response
p y .73.
Target structure
repetition of the movement-derived sentences. Her repetition on the SVO sentences was
ungrammatical
SVS OVO other
Correct
response
OSV
5
0
-
-
-
-
4
1
OVS
5
2
-
-
-
1
1
1
3.2. Repetition
TA repeated correctly all the SVO sentences, but showed considerable difficulties in the
grammatical
SVO OVS VSO
Total
flawless, with 6 correct repetitions of the 6 sentences, but OSV and OVS sentences yielded
Subject relative
9
0
5
-
1
1
-
2
only 0/5 and 2/5 correct repetitions respectively. She could not repeat correctly any of the
Object relative
12
2
3
3
-
1
-
3
subject relatives (0/9), and repeated only 2 of the 12 object relatives correctly.
Total
13%
48%
39%
Her repetition of SVO sentences was significantly better than OSV sentences (p = .008),
subject relatives (p < .001), and object relatives (p = .001). The difference between SVO and
OVS was just below significance (p = .06). Importantly, no significant difference was found
between the number of correct repetitions of subject relatives and object relatives (p = .36).
(12) Examples of repetition errors
a. OSV target
response
the control group on all of the other structures (p < .0001).
In order to grant to TA the maximal credit for production abilities, in the analysis of
repetition errors the first utterance that was sentence-like was included, and there was no
b. Subject relative target
response
were incorrect and more than half of her responses were ungrammatical. Table 8 presents the
types of responses she made in repetition of movement-derived sentences. The case produced
with each NP is an important indication of the syntactic role that TA gave to each NP in the
c. Object relative target
the accusative case are marked in O. Most of the grammatical productions were characterized
Ehtogo
klouna
obnimaet medvezhjonka
this-ACC-masc clown-ACC hugs teddy-bear-ACC
"[This clown]-ACC hugs teddy-bear-ACC."
Ehto
devochka kotoraja
tolkaet mal'chika
Kotoraja…
devochka tolkaet mal'chika
who-NOM-fem girl-NOM pushes boy-ACC
"Who… girl pushes boy."
Ehto
mal'chik kotorogo
tolkaet devochka
this-neutr boy-NOM who-ACC-masc pushes girl-NOM
"This is the boy who the girl pushes."
sentence. In Table 8, arguments in the nominative case are marked with S, and arguments in
by SVO order (67% of all the grammatical productions). The ungrammatical productions
medvezhjonok obnimaet
this-neutr girl-NOM who-NOM-fem pushes boy-ACC
"This is the girl who pushes the boy."
reference to initial groping that included word parts, single words, or single phrases when she
produced a more target-like response at any trial. Even using this count, 87% of her responses
klouna
this-ACC-masc clown-ACC teddy-bear-NOM hugs
"This clown, the teddy-bear hugs (him)."
The five control participants matched to TA repeated all of the structures 100% correct. Thus,
TA repeated SVO sentences similarly to the control group, but was significantly poorer than
Ehtogo
response
Ehto
mal'chik kotorogo…
ehto
kotorogo…
this-neutr boy-NOM who-ACC-masc this-neutr who-ACC-masc
included structures with two arguments in the same case: OVO and SVS structures, which
"This is the boy who… this who…"
constituted 53% of the ungrammatical repetitions, see examples (12a) and (12c), respectively.
(the response continues) …mal'chik tolkaet devochka
boy-NOM pushes girl-NOM
"The girl-NOM pushes the boy-NOM"
For the subject relatives, 5 out of the 6 incorrect but grammatical repetitions preserved the
20
Wh movement in Russian agrammatism
21
Wh movement in Russian agrammatism
thematic roles in the original sentence, compared to only 3 of 6 grammatical repetitions of
reports on agrammatic comprehension in many other languages: individuals with
object relatives. None of the incorrect repetitions included a well-formed relative clause
agrammatism typically experience difficulties understanding semantically reversible non-
structure with a relative pronoun.
canonical sentences derived by Wh-movement, and succeed in the comprehension of simple
TA did not produce any bare noun stem without case morphology that would create an
sentences without movement and movement-derived sentences that preserve the canonical
impossible word in Russian. The only case inflections she used in her repetitions were
order (Grodzinsky, 1989, 2000; Schwartz, Saffran, & Marin, 1980; Zurif & Caramazza, 1976;
nominative and accusative case (these were the only cases in the target sentences). In
see Grodzinsky et al., 1999 for a review). A common account for the relatively good
addition, TA showed no tendency to omit noun phrases: significantly more structures with
performance on right-branching subject relatives is that although these sentences are also
two arguments were produced compared to structures in which one argument was omitted,
derived by Wh-movement, they can be comprehended on the basis of the canonical order of
v2 = 21.41, p < .0001. When she produced the verb, it was produced in the present tense,
their constituents: the agent is first and the theme is second (Grodzinsky, 1989, 1990, 2000).
Such a strategy can also account for the good performance in subject relatives in Russian.
singular, like in the target sentences.
This non-syntactic strategy cannot assist performance in topicalization and object relatives, in
4. Discussion
which the theme precedes the agent.
This study explored how Russian-speaking individuals with agrammatism comprehend and
Why was the comprehension of topicalization structures and object relatives impaired? Both
produce sentences derived by Wh-movement. We had two main research goals. One was to
include Wh-movement and are non-canonical. But they also share another property: both
examine a topic that was already extensively tested, the comprehension and production of
require the high nodes in the syntactic tree. The derivation of Russian relative clauses
movement-derived sentences, in a language that has never been tested before, Russian, and to
requires the CP node, and for OSV and OVS structures there are analyses according to which
see whether the pattern in Russian is similar to that found in other languages. The other goal
the object moves up to IP (Bailyn, 2002, 2003, 2004; King, 1995)10 and analyses according to
was to use the richness of Russian morphology to examine whether morpho-syntactic cues
which the movement extends above IP (Erechko, 2002; Ionin, 2001). Thus, the failure to
can be used by individuals with agrammatism to improve comprehension of these sentences.
understand these structures is consistent both with the theory that suggests that movement of
The main findings were that the comprehension of semantically reversible non-canonical
noun phrases is impaired in agrammatism (TDH, Grodzinsky, 1990, 2000) and with a theory
movement-derived sentences and the production of sentences that involve movement to CP
according to which the deficit stems from the inaccessibility of the high nodes of the
are severely compromised in Russian, as they are in agrammatism in other languages.
syntactic tree (Friedmann, 2006a, 2008; Luzzatti & Guasti, 2000). The trace deletion
Morphological cues of case and gender did not assist the comprehension of these structures.
hypothesis suggests impairment in the representation of traces of (certain types of) NPs, this,
4.1. Comprehension of Wh movement-derived sentences
The tree pruning hypothesis suggests that there is a deficit in accessing the high nodes of the
in turn, can be the source of the impaired comprehension of movement-derived structures;
The main finding with respect to the comprehension of Russian-speaking individuals with
syntactic tree. Because these structures that contain Wh-movement involve the high nodes of
agrammatism was that the comprehension of semantically reversible non-canonical sentences
the tree (specifically, CP), if this theory is extended to comprehension, the deficit in
derived by Wh-movement was severely compromised. The comprehension of topicalization
accessing the high syntactic nodes can be the reason of the impaired comprehension of Wh-
structures and of object relatives was at chance, whereas the comprehension of SVO, a
movement derived structures. The deficit is also consistent with a combination of the two
structure without Wh-movement, was above chance, and so were the subject relatives, which
accounts.
are derived by Wh-movement but keep the canonical order of constituents. The
comprehension of the topicalized sentences and of the object relatives was significantly
poorer than that of SVO and of subject relatives. These findings are in line with previous
10
Bailyn (2002) does not reject the possibility that OSV derivation includes object movement above IP.
22
Wh movement in Russian agrammatism
23
Wh movement in Russian agrammatism
4.2. Does the increase in general complexity contribute to the difficulty in comprehension?
4.3. Does morphology assist comprehension?
Thus, the comprehension of semantically reversible non-canonical structures derived by Wh-
Another research question was whether morphological cues are used by individuals with
movement is impaired in Russian agrammatism, either because movement itself is impaired,
agrammatism to understand movement-derived sentences. Russian, being a morphologically
or because the landing site of this movement, the high nodes of the tree, is inaccessible.
rich language, allowed us to test two types of morpho-syntactic cues that can provide
The study was also designed to examine whether the movement of the verb across the subject
information about the syntactic role of constituents in sentence. One cue is the overt marking
further hampers the comprehension of sentences that already contain Wh-movement. For this
of case – noun phrases in Russian are marked for case, so that in topicalization sentences, for
aim we compared topicalization structures in VS order to topicalization structures in SV
example, case information can suggest a very straightforward identification of the array of
order, and object relatives in SV order in the relative clause to object relatives in VS order.
thematic roles of the participants in the sentence: the agent is marked by a nominative case,
The results of these comparisons were straightforward: the comprehension of the VS
and the theme by an accusative case (in the current study there was also a modifier before the
structures did not differ from the SV structures for any of the participants and for the group,
noun in the accusative case that appeared in the same case). The other type of morphological
either for the topicalization sentences or for the object relatives. These results suggest that
cue that we tested was the relative pronoun in relative clauses. The relative pronoun carries
when the sentence is derived by Wh-movement, the addition of verb movement does not
both case and gender information (and number, which was not manipulated in this study),
further impair the comprehension of the sentence (although there are indications for a deficit
which can be used to identify the thematic roles in the relative clause. A relative pronoun
in verb movement comprehension in agrammatism, Friedmann, Gvion, Biran, &
inflected for accusative and feminine indicates that the relative head is feminine, and that it is
Novogrodsky, 2006). These findings join existing evidence (for example, Friedmann &
the theme in the relative clause. Thus, the case cue can be used to identify the thematic roles
Shapiro, 2003; Shapiro, Gordon, Hack, & Killackey, 1993) illustrating that given a
in the sentence, and the gender inflection can be used to identify the moved element, when
movement of an argument, verb movement does not have an additive contribution to the
the NPs in the sentence differ with respect to gender. Despite this, the participants were
impairment.
unable to use these morphological cues for the interpretation of the thematic roles in non-
Thus, the non-additive nature of increased complexity, and the lack of significant difference
canonical movement-derived sentences. In the topicalization sentences, case morphology on
in the comprehension of structures that differ on complexity, are not consistent with the
the subject and moved object did not lead to correct interpretation of the sentences, nor was
general complexity or resource capacity accounts that rely on the purported additive nature of
the relative pronoun case and gender morphology used to interpret the thematic roles in
syntactic operations (e.g., Caplan, Baker, & Dehaut, 1985; Caplan, Waters, DeDe, Michaud,
object relatives, as indicated by the lack of difference between the sentences with and without
& Reddy, 2007; Haarmann, Just, & Carpenter, 1997). For additional discussion of processing
the morphological cues.
capacity explanation for the agrammatic difficulty in comprehension see also Friedmann and
These results are in line with previous reports in other languages that showed that
Gvion, 2003, and Gvion and Friedmann, 2007.
morphological cues do not help agrammatic interpretation of movement-derived sentences.
An additional support for the conclusion that increased complexity does not result in
For example, Burchert et al. (2003) tested case and number agreement in German, and gender
decreased performance when both structures include Wh-movement and non-canonical order,
agreement in Hebrew. They found no improvement in the comprehension of relative clauses
comes from the comparison of the comprehension of relative clauses and topicalization
and topicalization structures when these cues were present in the sentence. Friedmann and
structures. No difference was found between comprehension of the topicalization structures
Shapiro (2003) reported that case marking on the topicalized object in Hebrew did not assist
and the object relatives, although the relative clauses can be characterized as more complex
agrammatic comprehension of topicalized sentences.
than topicalized structures because they include, in addition to the movement to CP, also
Notice that although the finding in the current study and in previous studies are similar, the
matching of the relative pronoun and the relative head, and embedding.
source for the lack of use of morphological cues might be different. Whereas gender and
number verb inflection cues, which were tested in Hebrew and German (Burchert et al.,
2003), might be accessible but not used for interpretation, morphological cues on the relative
24
Wh movement in Russian agrammatism
25
Wh movement in Russian agrammatism
pronoun, and case marking on the topicalized object (on the noun and on the modifier), the
4.4. Repetition of movement-derived sentences
cues tested in the current study, might not be accessible, due to the position of the relative
Although only one individual with agrammatism participated in the repetition experiment,
pronoun and the topicalized object on the syntactic tree. Namely, whereas it might be that
several results of this experiment are suggestive regarding the nature of the agrammatic
gender and number cues are simply not part of the automatic syntactic mechanism for the
deficit. Firstly, sentences without syntactic movement (SVO sentences) were repeated
identification of thematic roles in input sentence, case cues are part of the syntactic
flawlessly, and significantly better than the movement-derived sentences (OSV, OVS, subject
mechanism, but they cannot be used by individuals with agrammatism, because this
relative, object relative), which were notably impaired. Close to 90% of the repetitions on the
information is concealed in high syntactic nodes they cannot access. Thus, if morphological
structures with movement were incorrect, and more than half of them constituted
inflections of case and gender are part of the relative pronoun in relative clauses, which
ungrammatical productions. Most of the grammatical incorrect responses had an SVO
resides in CP, and CP is inaccessible to individuals with agrammatism, this explains why the
structure.
morphological cues cannot be accessed and utilized for the comprehension of relative
According to the Tree Pruning Hypothesis (Friedmann, 2001, 2006b; Friedmann &
clauses. Similarly, in topicalization sentences, when the case and gender morphology cues are
Grodzinsky, 1997), individuals with agrammatic aphasia cannot construct the syntactic tree
part of the object, which has moved to CP, these cues are not available to individuals with
up to its higher nodes, but they can construct a partial tree, which includes only lower nodes.
agrammatism who have impaired access to CP, and hence, cannot assist their comprehension
An inability to construct the higher nodes is realized in production difficulties of the various
of topicalization.
morpho-syntactic structures that involve these higher nodes. TA's performance pattern is
Note, however, that although morphology was not used to identify the thematic roles in the
consistent with this account: she succeeded in the production of SVO sentences, which can be
sentences, the chance performance on the topicalization and object relative structures
produced using only the lower part of the syntactic tree, but had difficulty producing
indicates some preserved sensitivity to morphology. Had the individuals with agrammatism
topicalization and relative clause structures, which require the high tree nodes11.
not been sensitive to morpho-syntax at all, and had they interpreted sentences solely based on
One interesting comparison that might shed light on the syntactic deficit in agrammatism is
linear order, they would have been expected to perform below chance on the topicalization
the comparison between comprehension and production of subject relatives. Whereas the
and object relative sentences, assigning the theme an agent role, and the agent – a theme role.
performance in the comprehension task of subject relatives was relatively good, the
The finding that they performed at chance level rather than below chance indicates that they
production of this structure was severely compromised. Especially if one wants to consider a
did retain some sensitivity to case morphology which kept them from consistently reversing
unified account for comprehension and production, this has to be explained. The explanation
the roles (see Friedmann & Shapiro, 2003 for a similar discussion). This is in a way similar to
falls immediately from the consideration of the strategies that can be applied by agrammatic
the findings from Serbo-Croatian indicating that the agrammatic speakers were sensitive to
participants in these tasks. The performance in the sentence-picture matching comprehension
morphology, but could not use it for sentence interpretation.
task can be based, for subject relatives, on a non-syntactic strategy that relies on canonical
Thus, the findings from the comprehension experiment indicate impaired comprehension of
structure. Because the first noun phrase in subject relatives is the agent, the performance on
semantically reversible non-canonical movement-derived sentences that involve high
subject relatives can be good even though they involve syntactic movement. Therefore, the
syntactic nodes. This comprehension impairment is consistent both with a deficit in Wh-
performance on subject relatives in comprehension tasks is usually better than that of other
movement and with a deficit in high syntactic nodes. The rich morphology in Russian, which
movement-derived structures, particularly object relatives, and this was the case in the current
includes cues about the syntactic role of constituents in sentences, does not assist agrammatic
comprehension.
11
The production deficit in agrammatism also affects structures that need high tree nodes and do not involve
Wh-movement, such as sentences with sentential complements. Notice, that the agrammatic deficit is not related
to any bi-clausal structure: individuals with agrammatism can produce bi-clausal sentences as long as they do
not involve the high syntactic nodes: they produce coordinated sentences (Friedmann, 2006b; Menn & Obler,
1990) and can produce and repeat embedded sentences that do not involve the high nodes, such as untensed
embeddings (Friedmann, 2001; Ni, Shankweiler, Harris, & Fulbright, 1997) and reduced relatives (Friedmann,
2001; 2006b).
26
Wh movement in Russian agrammatism
27
Wh movement in Russian agrammatism
experiment too. Another type of support for the correct assignment of thematic roles in
4.5. Brain function and the syntactic deficit
subject relatives comes from the pattern of errors in the repetition of subject relatives and
The participants in the present study were diagnosed with Broca's agrammatic aphasia. This
object relatives in the current study. Whereas the participant failed to repeat both, she kept
impairment is typically associated with lesions that include Broca's area (left IFG, including
the target agent and theme order and their cases in most of her repetitions of subject relatives,
Brodmann areas 44 and 45) (e.g., Alexander, Naeser, & Palumbo, 1990; Zurif, 1995). The
but only in half of her grammatical repetitions of object relatives.
left IFG has been consistently linked to syntactic processing, across a wide range of
In contrast, the production of subject relatives requires the high nodes of the syntactic tree,
approaches and methods (Berndt & Caramazza, 1980; Caplan, Alpert, & Waters, 1998, 1999;
just like object relatives, and no strategy can rescue the production of the relative pronoun in
Constable et al., 2004; Damasio & Damasio, 1989; Dapretto & Bookheimer, 1999; Embick,
such sentences. An impaired access to the high node of the syntactic tree would thus cause
Marantz, Miyashita, O’Neil, & Sakai, 2000; Friedmann & Shapiro, 2003; Grodzinsky, 2000,
subject relatives to be poorly produced, similarly to other structures that require the high
2006; Love, Swinney, Walenski, & Zurif, 2008; Luke, Liu, Wai, Wan, & Tan, 2002;
syntactic nodes, like object relatives, and this is indeed what we found in the current study.
Stromswold, Caplan, Alpert, & Rauch, 1996; Swinney & Zurif, 1995; Zurif, 1995).
The findings of this research do not determine whether the difficulty TA had in production is
Specifically, brain imaging studies have reported consistent activations in Broca's area in
related to the IP layer or only to the CP layer. It is impossible to determine whether TA could
response to sentences that are derived by syntactic movement, such as relative clauses and
access the IP node or not, because the sentences she produced in SVO order could have been
topicalization, structures that have been tested in the current study as well, and which load
produced in a node lower than IP (Friedmann & Shapiro, 2003). In future research it would
onto the CP layer (Ben-Shachar, Hendler, Kahn, Ben-Bashat, & Grodzinsky, 2003; Ben-
be interesting to examine individuals who are only impaired in accessing CP, but who can
Shachar, Palti, & Grodzinsky, 2004; Constable et al., 2004; Grodzinsky & Friederici, 2006;
access IP, and test whether they can produce topicalization sentences or not. This would help
Grodzinsky & Santi, 2008; Just, Carpenter, Keller, Eddy, & Thulborn, 1996; Stromswold et
determine between linguistic theories which analyze topicalization in Russian as involving
al., 1996). Imaging studies have also found activations in left IFG (in BA 45) in response to
movement to CP and theories according to which topicalization includes movement only up
structures that involve CP embedding, even in sentences that do not involve movement
to IP.
(Shetreet, Friedmann, & Hadar, in press). Our current results from agrammatism in Russian
One additional point that could be seen in the way TA repeated the sentences was that she
are completely consistent with these two functions ascribed to the left IFG: the structures that
never produced bare noun stems without case morphology, when such bare nouns would be
our participants found difficult to comprehend and to repeat were structures that involved
ungrammatical. This finding is in line with earlier reports in the literature (Grodzinsky, 1984;
syntactic movement and the CP node. The inability to use morphological cues on the relative
Ruigendijk, 2002) showing that individuals with agrammatism obey constraints on well-
pronoun is consistent with left IFG (possibly specifically BA 45) being responsible for the
formedness of words, and are aware of the requirement that each noun phrase in Russian has
processing of the CP level.
to carry case. TA did produce case errors in the repetition of topicalization sentences. Some
The processing of the CP level that is ascribed to the left IFG might be captured in terms of
of these responses included two noun phrases in the same case (in most occurrences,
functional nodes and their hierarchical order on the syntactic tree. According to such
accusative case – OVO). A similar pattern of errors was also reported by Ruigendijk (2002).
perception, the brain areas implicated in agrammatic aphasia are involved in the structuring
The source of these errors might be that the participant heard the movement-derived
and projection of the syntactic tree. It might be that a lesion in the left IFG restricts the
topicalization sentence, did not understand the thematic relations in it, and therefore made
number of phrasal nodes that can be projected in the syntactic tree, leading to inability to
case errors in repeating it.
access CP, and possibly also TP below it. Thus, these results, by using the linguistic toolbox
to characterize the agrammatic deficit, can bear on the function of the left IFG in language
processing. These data can also be looked at from the other direction – from brain to
linguistics - as providing evidence from neuropsychology to the neurological reality of the
syntactic tree.
28
Wh movement in Russian agrammatism
4.6. Clinical implications
These research findings have also implications for diagnosis and treatment of Russianspeaking individuals with agrammatism. The main finding was that Russian-speaking
individuals with agrammatism have a difficulty in the comprehension of semantically
reversible non-canonical movement-derived sentences. In this respect, results from treatment
studies in English which were conducted by Shapiro et al. (see for example Shapiro &
Thompson, 2006) are encouraging. These studies indicated that it is possible to improve
comprehension of movement-derived sentences by explicit instruction of movement
structures. Thus, it seems necessary to try to apply a similar treatment to individuals with
agrammatism who speak Russian. Other treatment directions suggest the treatment of high
nodes in the syntactic tree, to induce improvement in the production of relative clauses and
topicalization structures (Friedmann, Wenkert-Olenik, & Gil, 2000; Thompson, Shapiro,
Kiran, & Sobecks, 2003). Given the difficulty in the production of such structures, it might be
useful to apply such treatment for Russian-speaking individuals with agrammatism too.
Finally, the inability of the participants to utilize the morphological richness of Russian to
interpret movement-derived sentences, in the presence of indications that at least some of the
morphological abilities such as gender agreement inflection are intact (Friedmann, 2001,
2006b) and the certain sensitivity to morphology that was detected in the current study (in
that there was no consistent reversal of roles) suggest another treatment direction. Individuals
with agrammatism can be explicitly taught to use the morphological information encoded in
the inflection of the verb, the modifier, and the relative pronoun to assist the comprehension
of meaning relations in sentences.
References
Akhutina, T., Kurgansky, A., Kurganskaya, M., Polinsky, M., Polonskaya, N., Larina, O.,
Bates, E., & Appelbaum, M. (2001). Processing of grammatical gender in normal and
aphasic speakers of Russian. Cortex, 37, 295-326.
Alexander, M., Naeser, M. A., & Palumbo, C. L. (1990). Broca's area aphasias: Aphasia after
lesions including the frontal operculum. Neurology, 40, 353-362.
Babyonyshev, M. (1993). Acquisition of the Russian case system. MIT Working Papers in
Linguistics, 19, 1-43.
Bailyn, J. F. (2002). Inversion, dislocation and optionality in Russian. In G. Zybatow, U.
Junghanns, G. Mehlhorn, & L. Szucsich (Eds.), Current issues in formal Slavic
Linguistics (FDSL 3) (pp. 280-293). Frankfurt/Main: Peter Lang.
Bailyn, J. F. (2003). A (purely) derivational account of Russian scrambling. In E. W. Browne,
J.-Y. Kim, B. H. Partee, & R. A. Rothstein (Eds.), Formal approaches to Slavic
Linguistics 11 (pp. 41-62). Ann Arbor: Michigan Slavic Publications.
Bailyn, J. F. (2004). Generalized inversion. Natural Language and Linguistic Theory, 22, 1-
29
Wh movement in Russian agrammatism
49.
Ben-Shachar, M., Hendler, T., Kahn, I., Ben-Bashat, D., & Grodzinsky, Y. (2003). The
neural reality of syntactic transformations: evidence from fMRI. Psychological Science,
14, 433-440.
Ben-Shachar, M., Palti, D., & Grodzinsky, Y. (2004). Neural correlates of syntactic
movement: converging evidence from two fMRI experiments. NeuroImage, 21, 13201336.
Berndt, R. S., & Caramazza, A. (1980). A redefinition of the syndrome of Broca’s aphasia:
implications for neuropsychological model of language. Applied Psycholinguistics, 1,
225-278.
Bianchi, V. (2000a). Some issues in the syntax of relative determiners. In A. Alexiadou, P.
Law, A. Meinunger, & C. Wilder (Eds.), The syntax of relative clauses (pp. 53-81).
Amsterdam: John Benjamins.
Bianchi, V. (2000b). The raising analysis of relative clauses: A reply to Borsley. Linguistic
Inquiry, 31, 123-140.
Borras, F. M., & Christian, R. F. (1971). Russian syntax: Aspects of modern Russian syntax
and vocabulary. Oxford: Clarendon Press.
Borsley, R. D. (1997). Relative clauses and the theory of phrase structure. Linguistic Inquiry,
28, 629-647.
Borsley, R. D. (2006). More on the raising analysis of relative clauses. Unpublished Ms., the
University of Essex. Retrieved from http://privatewww.essex.ac.uk/~rborsley/relatives.pdf.
Botwinik-Rotem, I. (2008). Indirect object and obligatory prepositions. In G. Hatav (Ed.),
Theoretical Hebrew Linguistics (pp. 217-251). Jerusalem: Magnes Press. (in Hebrew).
Browning, M. (1987). Null operator constructions. PhD Dissertation, MIT.
Burchert, F., Friedmann, N., & de Bleser, R. (2003). Morphology does not help
comprehension in agrammatism: A study of German and Hebrew. Brain and Language,
87, 52.
Caplan, D., Alpert, N., & Waters, G. (1998). Effects of syntactic structure and propositional
number on patterns of regional cerebral blood flow. Journal of Cognitive Neuroscience,
10, 541-552.
Caplan, D., Alpert, N., & Waters, G. (1999). PET studies of syntactic processing with
auditory sentence presentation. NeuroImage, 9, 343-351.
Caplan, D., Baker, C., & Dehaut, F. (1985). Syntactic determinants of sentence
comprehension in aphasia. Cognition, 21, 117-175.
Caplan, D., Waters, G., DeDe, G., Michaud, J., & Reddy, A. (2007). A study of syntactic
processing in aphasia I: Behavioral (psycholinguistic) aspects. Brain and Language,
101, 103-150.
Caramazza, A., & Zurif, E. B. (1976). Dissociation of algorithmic and heuristic processes in
sentences comprehension: Evidence from aphasia. Brain and Language, 3, 572-582.
Chomsky, N. (1986). Knowledge of language: Its nature, origin, and use. New York:
Praeger.
Chomsky, N. (1995). The minimalist program. Cambridge, MA: MIT Press
Citko, B. (2004). On headed, headless, and light-headed relatives. Natural Language and
Linguistic Theory, 22, 95-126.
Constable, R. T., Pugh, K. R., Berroya, E., Mencl, W. E., Westerveld, M., Ni, W., &
Shankweiler, D. (2004). Sentence complexity and input modality effects in sentence
comprehension: an fMRI study. NeuroImage, 22, 11-21.
Crawford, J. R., & Howell, D. C. (1998). Comparing an individual’s test score against norms
derived from small samples. The Clinical Neuropsychologist, 12, 482-486.
Damasio, H., & Damasio, A. R. (1989). Lesion analysis in neuropsychology. New York:
30
Wh movement in Russian agrammatism
31
Oxford University Press.
Dapretto, M., & Bookheimer, S. Y. (1999). Form and content: Dissociating syntax and
semantics in sentence comprehension. Neuron, 24, 427-432.
Dick, F., Bates, E., Wulfeck, B., Utman, J. A., Dronkers, N., & Gernsbacher, M. A. (2001).
Language deficits, localization and grammar: Evidence for a distributive model of
language breakdown in aphasics patients and neurologically intact individuals.
Psychological Review, 108, 759-788.
Embick, D., Marantz, A., Miyashita, Y., O’Neil, W., & Sakai, K. L. (2000). A syntactic
specialization for Broca’s area. Proceedings of the National Academy of Sciences of the
United States of America, 97, 6150-6154.
Erechko, A. (2002). Subject-verb inversion in Russian. Ms., ConSole XI, Padua, Italy.
Friedmann, N. (1998). BAFLA – Friedmann's Battery for Agrammatism. Tel Aviv, Israel: Tel
Aviv University.
Friedmann, N. (2001). Agrammatism and psychological reality of the syntactic tree. Journal
of Psycholinguistic Research, 30, 71-90.
Friedmann, N. (2002). Question production in agrammatism: The tree pruning hypothesis.
Brain and Language, 80, 160-187.
Friedmann, N. (2006a). Generalizations on variations in comprehension and production: A
future source of variation and a possible account. Brain and Language, 96, 151-153.
Friedmann, N. (2006b). Speech production in Broca's agrammatic aphasia: Syntactic tree
pruning. In Y. Grodzinsky, & K. Amunts (Eds.), Broca’s region (pp. 63-82). New
York: Oxford University Press.
Friedmann, N. (2008). Traceless relatives: Agrammatic comprehension of relative clauses
with resumptive pronouns. Journal of Neurolinguistics, 21, 138-149.
Friedmann, N., & Grodzinsky, Y. (1997). Tense and Agreement in agrammatic production:
Pruning the syntactic tree. Brain and Language, 56, 397-425.
Friedmann, N., & Gvion, A. (2003). Sentence comprehension and working memory
limitation: A dissociation between semantic and phonological encoding. Brain and
Language, 86, 23-39.
Friedmann, N., Gvion, A., Biran, M., & Novogrodsky, R. (2006). Do people with agrammatic
aphasia understand verb movement? Aphasiology, 20, 136-153.
Friedmann, N., & Shapiro, L. P. (2003). Agrammatic comprehension of simple active
sentences with moved constituents: Hebrew OSV and OVS structures. Journal of
Speech, Language, and Hearing Research, 46, 288-297.
Friedmann, N., Wenkert-Olenik, D., & Gil, M. (2000). From theory to practice: Treatment of
agrammatic production in Hebrew based on the Tree Pruning Hypothesis. Journal of
Neurolinguistics, 13, 250-254.
Grodzinsky, Y. (1984). The syntactic characterization of agrammatism. Cognition, 16, 99-120.
Grodzinsky, Y. (1989). Agrammatic comprehension of relative clauses. Brain and Language,
37, 480-499.
Grodzinsky, Y. (1990). Theoretical perspectives on language deficits. Cambridge, MA: MIT Press.
Grodzinsky, Y. (2000). The neurology of syntax: Language use without Broca's area.
Behavioral and Brain Sciences, 23, 1-71.
Grodzinsky, Y. (2006). A blueprint for a brain map of syntax. In Y. Grodzinsky, & K.
Amunts, (Eds.), Broca's Region (pp. 83-107). New York: Oxford University Press.
Grodzinsky, Y., & Friederici, A. D. (2006). Neuroimaging of syntax and syntactic
processing. Current Opinion in Neurobiology, 16, 240-246.
Grodzinsky, Y., Piñango, M., Zurif, E., & Drai, D. (1999). The critical role of group studies
in neuropsychology: Comprehension regularities in Broca's aphasia. Brain and
Language, 67, 134-147.
Wh movement in Russian agrammatism
Grodzinsky, Y., & Santi, A. (2008). The battle for Broca's region. Trends in Cognitive
Science, 12, 474-480.
Gvion, A., & Friedmann, N. (2007). Is phonological working memory involved in sentence
comprehension? The difference between phonological and semantic reactivation. In Y.
Falk (Ed.), Proceedings of the 19th IATL conference.
Haarmann, H. J., Just, M. A., & Carpenter, P. A. (1997). Aphasic sentence comprehension as
a resource deficit: A computational approach. Brain and Language, 59, 76-120.
Haegeman, L. (1994). Introduction to government and binding theory. Oxford: Blackwell.
Hickok, G., & Avrutin, S. (1996). Comprehension of Wh-question in two Broca's Aphasics.
Brain and Language, 52, 314-327.
Ionin, T. (2001). The one girl who was kissed by every boy: Scope, scrambling and discourse
function in Russian. Proceedings of ConSole X, Leiden, the Netherlands.
Junghanns, U., & Zybatow, G. (1997). Syntax and information structure of Russian clauses.
In E. W. Browne et al. (Eds.), Annual Workshop on Formal Approaches to Slavic
Linguistics. The Cornell Meeting 1995 ( Michigan Slavic Materials; 39) (pp. 289-319).
Ann Arbor: Michigan Slavic Publications.
Just, M. A., Carpenter, P. A., Keller, T. A., Eddy, W. F., & Thulborn, K. R. (1996). Brain
activation modulated by sentence comprehension. Science, 274, 114-116.
Kallestinova, E., & Slabakova, R. (2007, May). Does the verb moves in Russian? Presented
at the 16th annual meeting of the Formal Approaches to Slavic Linguistic, Stony Brook
University.
Kayne, R. S. (1994). The antisymmetry of syntax. Cambridge, Massachusetts: MIT Press.
King, T. (1995). Configuring topic and focus in Russian. Stanford: CSLI Publication.
Lavine, J. E. (2003). Resumption in Slavic: Phases, cyclicity and cases. In E. W. Browne, J.Y. Kim, B. H. Partee, & R. A. Rothstein (Eds.), Formal Approaches to Slavic
Linguistics, 11 (pp. 355-372). Ann Arbor: Michigan Slavic Publications.
Lavine, J. E., & Freidin, R. (2001). The subject of defective T(ense) in Slavic. Journal of
Slavic Linguistics, 9, 253-290.
Love, T., Swinney, D., Walenski, M., & Zurif, E. (2008). How left inferior frontal cortex
participates in syntactic processing: Evidence from aphasia. Brain and Language, 107,
203-219.
Lukatela, K., Crain, S., & Shankweiler, D. (1988). Sensitivity to inflectional morphology in
agrammatism: Investigation of a highly inflected language. Brain and Language, 33, 1-15.
Lukatela, K., Shankweiler, D., & Crain, S. (1995). Syntactic processing in agrammatic
aphasia by speakers of a Slavic language. Brain and Language, 49, 50-76.
Luke, K. K., Liu, H. L., Wai, Y. Y., Wan, Y. L., & Tan, L. H. (2002). The functional
anatomy of syntactic and semantic processing in language comprehension. Human
Brain Mapping, 16, 133-145.
Luria, A. R. (1975). Two kinds of disorders in the comprehension of grammatical
construction. Linguistics, 154/155, 47-56.
Luria, A. R. (1982). Language and cognition [Iazyk i soznanie] (J. V. Wertsch, Trans.). New
York: Wiley. (Original work published 1979).
Luzzatti, C., & Guasti, M.-T. (2000). Agrammatism, syntactic theory, and the lexicon:
Broca’s area and the development of linguistic ability in the human brain. Behavioral
and Brain Sciences, 23, 41-42.
Menn, L., & Obler, L. (Eds.) (1990). Agrammatic aphasia: A cross-language narrative
sourcebook. Philadelphia: John Benjamin.
Milekic, S., BoškoviN, Z., Crain, S., & Shankweiler, D. (1995). Comprehension of nonlexical
categories in agrammatism. Journal of Psycholinguistic Research, 24, 299-311.
Ni, W., Shankweiler, D., Harris, K. S., & Fulbright, R. K. (1997). Production and
32
Wh movement in Russian agrammatism
comprehension of relative clause syntax in nonfluent aphasia: A coordinated study.
Brain and Language, 60, 93-95.
O’Grady, W., & Lee, M. (2005). A mapping theory of agrammatic comprehension deficits.
Brain and Language, 92, 91-100.
Peshkovskij, A. M. (1956). Russkij sintaksis v nauchnom osveshhenii. [The Russian syntax in
the light of science]. Moskva: Ministerstvo Prosveshhenija RSFSR. (in Russian).
Pollock, J. Y. (1989). Verb movement, universal grammar, and structure of IP. Linguistic
Inquiry, 20, 365-424.
Pugh, S. M. (2006). Relative constructions in contemporary popular Russian prose.
Reflections on a changing language. Russian Linguistics, 30, 213-266.
Rizzi, L. (1997). The fine structure of the left periphery. In L. Haegeman (Ed.),
Elements of Grammar (pp. 281-337). Dordrecht: Kluwer.
Rudnev, A. G. (1968). Sintaksis sovremennogo russkogo jazyka [The syntax of modern
Russian]. Moskva: Vysshaja Shkola. (in Russian).
Ruigendijk, E. (2002). Case assignment in agrammatism: A cross-linguistic study. Doctoral
dissertation. Groningen University, Groningen, The Netherlands.
Ruigendijk, E., & Baauw, S. (in press). The production of determiners and pronouns in Dutch
agrammatism: Syntactic and pragmatic aspects. Language and Brain, 9. (in Hebrew).
Ruigendijk, E., & Friedmann, N. (2008). On the relation between Structural Case,
determiners, and verbs in agrammatism: A Study of Hebrew and Dutch. Aphasiology,
22, 948-969.
Ruigendijk, E., Kouwenberg, M., & Friedmann, N. (2004). Question production in Dutch
agrammatism. Brain and Language, 91, 116-117.
Schechter, Y. (1965). ILAT - Israeli Loewenstein Aphasia Test. Loewenstein Hospital,
Rehabilitation Center.
Schwartz, M., Saffran, E., & Marin O. (1980). The word-order problem in agrammatism: I.
Comprehension. Brain and Language, 10, 249-262.
Sekerina, I. (1997). The syntax and processing of split scrambling constructions in Russian.
Unpublished doctoral dissertation, CUNY Graduate School and University Center.
Shapiro, L. P., Gordon, B., Hack, N., & Killackey, J. (1993). Verb-argument structure
processing in complex sentences in Broca's and Wernicke's aphasia. Brain and
Language, 45, 423-447.
Shapiro, L. P., & Thompson, C. K. (2006). Treating language deficits in Broca’s aphasia. In
Y. Grodzinsky, & K. Amunts (Eds.), Broca’s region (pp. 119-134). New York: Oxford
University Press.
Shetreet, E., Friedmann, N., & Hadar, U. (in press). An fMRI study of syntactic layers:
Sentential and lexical aspects of embedding. NeuroImage.
Smirnitsky, A. I. (1975). Essentials of Russian grammar. Moscow: Higher School Publishing
House.
Smith, S., & Bates, E. (1987). Accessibility of case and gender contrast for agent-object
assignment in Broca's aphasics and fluent anomics. Brain and Language, 30, 8-32.
Smith, S. D., & Mimica, I. (1984). Agrammatism in a case-inflected language:
Comprehension of agent-object relation. Brain and Language, 21, 274-290.
Spencer, D. (1993). Aspects of the syntax of relative clauses in colloquial and standard
Russian. Doctoral dissertation. University of Toronto.
Stromswold, K., Caplan, D., Alpert, N., & Rauch, S. (1996). Localization of syntactic
comprehension by positron emission tomography. Brain and Language, 52, 452-473.
Swinney, D., & Zurif, E. (1995). Syntactic processing in aphasia. Brain and Language, 50,
225-239.
Szczegielniak, A. (2005a). Two types of relative clauses in Slavic – Evidence from
33
Wh movement in Russian agrammatism
reconstruction and ellipsis. In Proceedings of the Western Conference on Linguistics
2004 (pp. 373-384). University of Southern California, CA.
Szczegielniak, A. (2005b). Two types of resumptive pronouns in Polish relative clauses.
Linguistic Variation Yearbook, 5, 165-185.
Szucsich, L. (2003). The structure of relative clauses in Slavic. In P. Kosta, J. B•aszczak, L.
Geist, & J. Frasek (Eds.), Investigations into formal Slavic linguistics, vol. II, 697-714.
Thompson, C. K., Shapiro, L. P., Kiran, S., & Sobecks, J. (2003). The role of syntactic
complexity in treatment of sentence deficit in agrammatic aphasia: The complexity
account of treatment efficacy (CATE). Journal of Speech, Language, and Hearing
Research, 46, 591-607.
Timberlake, A. (1993). Russian. In B. Comrie, & G. G. Corbett (Eds.), The Slavonic
languages (pp. 827-886). New York: Routledge Language Family Descriptions.
Tsvetkova, L. S., & Glozman, J. M. (1975). A neurolinguistic analysis of expressive
agrammatism in different forms of aphasia. Linguistics, 154/155, 61-76.
Vergnaud, J. R. (1974). French relative clauses. Doctoral Dissertation, MIT.
Williams, E. (1980). Predication. Linguistic Inquiry, 11, 203-238.
Zurif, E. (1995). Brain regions of relevance to syntactic processing (second ed.). In L.
Gleitman & M. Liberman (Eds.), An invitation to cognitive science (Vol. 1).
Cambridge: MIT Press.
Zurif, E., & Caramazza, A. (1976). Psycholinguistic structures in aphasia: Studies in syntax
and semantics. In H. Whitaker & H. A. Whitaker (Eds.), Studies in neurolinguistics
(Vol. I) (pp. 260-292). New York: Academic Press.
34