MARKING: THE VIEW FROM L1 LITHUANIAN
INETA DABAŠINSKIENĖ1
Abstract. The present paper examines the development of Differential Object Marking (DOM) in Lithuanian as L1. The analysis is based on the corpus data of two Lithuanian children, a girl and a boy, collected during their age period of 1;7 till 2;6.
Both children start to mark direct objects with the grammatically correct accusative or genitive case rather early, at the age of 1;9–1;10. The number of erroneous productions attested during the observed period is very low. The two most typical errors were overgeneralization of the accusative case in the contexts where the genitive is appropriate and the use of an inappropriate case form after a negated verb.
In negative constructions children tend to use the accusative case instead of the genitive, which is also eligible for marking the direct object. The data are interpreted as indicating that DOM in L1 Lithuanian is acquired early in spite of the complexity of the system.
Keywords: Lithuanian, acquisition of DOM, independent partitive genitive (IPG), genitive of negation, accusative.
1. INTRODUCTION
The phenomenon of differential object marking (DOM) exists in many languages (Bossong, 1991, Comrie, 1979), but the recent cross-linguistic interest in this area has been stirred by Aissen’s (2003) study. DOM denotes a morpho-syntactic distinction between overtly marked and non-marked direct objects, where the variation between marking and non-marking is mainly based on semantic, pragmatic and discourse factors. Certain features, such as animacy, specificity and definiteness of the direct object, are relevant to its marking, but their scope differs across languages (Leonetti 2004). Learners of DOM languages must grasp the contexts in which such marking is obligatory, or optional, and/or unacceptable.
Previous L1 acquisition research acknowledges the fact that semantically, pragmatically and grammatically complex structures are difficult for a child to acquire.
However, despite anticipated obstacles, already at the age of 2;0 children are able to identify coherent, recurring patterns and start to apply rules creatively even though many rules are overgeneralized (Marchman and Bates 1994, Smoczyńska 1986, Dressler 1997).
At this early age the basic language-specific properties of each system are productively mastered. Children are able to comprehend and to produce patterns of a higher degree of
1Vytautas Magnus University, [email protected] RRL, LX, 4, p. 369–382, Bucureşti, 2015
complexity; they also develop abilities to understand less frequent, less transparent and less salient linguistic structures (Dressler 1997).
As DOM is a semantically, pragmatically and grammatically complex structure, it is not implausible to assume that it might be difficult to acquire. DOM has recently received much attention in L2 learning studies (Silva-Corvalán 1994, Montrul 2004, Montrul and Bowles 2009, Guijarro-Fuentes and Marinis 2009). However, research in first language acquisition is still very scarce. The only exception is the study on the acquisition of DOM in Spanish by Rodríguez-Mondoñedo (2008). This study analyzes the use of direct objects by Spanish monolingual children drawing on data from longitudinal corpora of available in the CHILDES database and checks the acquisition predictions made by Aissen (2003), which are advanced in the Optimality Theory (OT). According to this approach, the child will entertain a number of grammars different from the target grammar before acquiring the final ranking of constraints. This predicts that children will make mistakes in their performance. The unexpected finding of the study was that during the course of the acquisition of DOM the Spanish children demonstrated virtually error-free performance.
These results are different from those reported for L2 Spanish, which show that DOM is a vulnerable domain.
The controversial conclusions of the studies on the (L2 or L1) acquisition of DOM, the vast majority coming from Spanish, suggest that cross-linguistic research is necessary if one aims at understanding how children acquire DOM.
The goal of the present study is to investigate the acquisition of DOM in L1 Lithuanian. Extending the study to languages other than Spanish might shed new light on the acquisition of DOM, in particular on the way in which language specific properties can interfere with the acquisition of object marking.
2. DIFFERENTIAL OBJECT MARKING IN LITHUANIAN
2.1. The accusative/genitive alternation
Discussions of the Lithuanian accusative/genitive alternations as DOM have emerged only recently (Aleksandravičiūtė 2013, Seržant 2014) in the context of the widely- discussed phenomena of Differential Object and Differential Subject Marking (DOM/
DSM) (Bossong 1991, Aissen 2003) as well as the independent partitive genitive (IPG), which represents a subtype of the DOM and DSM (Seržant 2014, Kagan 2005, 2010, Neidle 1988, Partee et al. 2012).
Lithuanian is a DOM language, and the choice of case depends on syntactic and semantic factors. Among the semantic factors which determine case choice, definiteness is central: objects are marked with the accusative case when they denote a definite quantity.
Quantitatively indefinite objects are marked with the partitive genitive. Negation is another important factor. In Lithuanian the ‘genitive of negation’ rule requires that in positive statements the direct object of transitive verbs be in the accusative case, whereas in the corresponding negated sentences the direct object must be in the genitive case.
Following traditional Lithuanian grammar (Ambrazas 1997), examples of the accusative/genitive alternations are provided below.
a. If a transitive verb is used with negation, the object must be assigned genitive case:
(1) Matau namą.
see house-ACC.SG
‘I see a house.’
(2) Nematau namo.
NEG.see house-ACC.SG ‘I don’t see a house.’
b. The direct object of an infinitive which is the complement of a negated modal or finite verb is marked with the genitive:
(3) Jis mėgsta rašyti laiškus.
he likes to write letters-ACC.PL ‘He likes to write letters.’
(4) Jis nemėgsta rašyti laiškų.
he NEG.like to write letters-GEN.PL ‘He doesn’t like to write letters.’
c. The accusative case may be retained in a negative sentence, especially if there are other words placed between the finite verb and the infinitive:
(5) Nepamiršk man kitą dieną parašyti laiško /laišką.
NEG.forget me-DAT next day to write letter-GEN.SG /ACC.SG ‘Don’t forget to write me a letter on the next day.’
d. If a direct object refers to an indefinite quantity, the genitive case is used instead of the accusative (6); the accusative is always used for definite quantity (7):
(6) Atnešiau knygų.
brought- PST.1SG books-GEN.PL ‘(I) brought some books.’
(7) Atnešiau knygas.
brought -PST.1SG the books-ACC.PL
‘(I) brought the books.’ (Ambrazas 1997: 486–487, 668–669).
2.2. The independent partitive genitive
As noticed by Seržant (2014) with respect to the accusative/genitive alternations in Lithuanian, “cross-linguistically this phenomenon may be governed not only by the NP internal properties, but also by the tense or aspectual properties of the verb phrase (Lestrade and de Hoop 2011). This is exactly what is found with the independent partitive genitive (IPG) in Lithuanian: “the alternation with the structural case is governed not only by indefiniteness of the reference and quantity, but may also depend on the aspectual interpretation of the verb” (Seržant 2014: 259). This suggestion by Seržant is worth mentioning here as he provides examples and interpretations in the context of the
independent partitive genitive (IPG). In (8) below the direct object bandel-ę (roll-ACC.SG) is assigned accusative case, since “singular discrete entities cannot be marked by the IPG in Lithuanian by definition” (Seržant 2014: 260), i.e. in Lithuanian there is no case alternation with singular discrete reference. But the first object NP “allows both IPG and the functionally (but not morphologically) unmarked accusative. One of the functions of the IPG is to cause its NP to be marked as quantificational, with an inherent indefinite quantity and, consequently, indefinite reference” (Seržant 2014: 260–261).
(8) Nupirkau pien-o ir bandel-ę.
buy-PST.1SG milk-GEN.SG and roll-ACC.SG ‘I bought some milk and a roll.’
The assumption of an implicit quantifier evoked by the IPG finds support from those verbs that require their arguments to be quantified. Thus, the IPG is triggered by verbal prefixes (analysed as quantifiers in Seržant 2014) such as pri-, at-, per- with the meaning of
‘a lot’ and už- ‘a little bit’ (in the relevant meaning of these prefixes). These verbs often occur in combination with the reflexive suffix/prefix -si-, e.g. už-si-kąsti duonos ‘to eat bread (IPG) a little bit’, per-si-valgyti obuolių ‘overeat apples (IPG)’ (Seržant 2014, Ambrazas 2007).
As for the aspectual interpretation of the verb, Seržant claims that the IPG disambiguates morphologically ambiguous verbs since Lithuanian does not have morphological means to encode viewpoint aspect; it can only be disambiguated contextually (Arkadiev 2011, Seržant 2014). Seržant (2014: 281) stresses that in Lithuanian the IPG may be used with both telic and atelic predicates. Its use “seems to be independent of the choice of the actionality type”. For example, compare the verb ger-ti ‘to drink’ with telicity overtly marked in (9) by means of the prefix iš-, to the morphologically unmarked gėrė with no such marking in (10):
(9) Jis iš-gėrė vanden-s/ vanden-į 3.NOM.SG.M TEL-drink.PST.3 water-GEN.SG/ water-ACC.SG ‘He drank up (some) water / (the) water.’
(10) Jis gėrė vanden-s/ vanden-į 3.NOM.SG.M drink.PST.3 water-GEN.SG/ water-ACC.SG ‘He drank (some) water / (the) water.’
The examples demonstrate the complexity of the Lithuanian DOM system as in many cases the problem of ambiguity has to be solved on both levels, the level of the object (indefiniteness, quantity) and the level of the sentence (tense, aspect, negation).
2.3. Predictions for the acquisition of DOM in L1 Lithuanian
When acquiring the Lithuanian DOM system, children must identify the role of definiteness and negation. They have to learn that direct objects which denote indefinite quantity require the genitive and those which denote definite quantity require the accusative.
But in negative sentences, the genitive is required with all the objects, irrespective of
definiteness. The child has to understand that there are two types of genitive of negation with direct object constructions. The first one is the negative counterpart of the direct object construction which takes the form of ‘a verb + accusative object’, e.g., pasiekiu lempą ‘(I) reach the lamp-ACC’, pažįstu kaimyną ‘(I) know the neighbour-ACC’, duok knygutę ‘give the book-ACC’. The negative version requires the obligatory use of the genitive, i.e. ‘negative verb + genitive’, e.g., nepasiekiu lempos ‘(I) NEG.reach the lamp-GEN’, nepažįstu kaimyno
‘(I don’t) NEG.know the neighbour-GEN’, neduok knygutės ‘NEG.give the book-GEN’.
The second type is the direct object construction with the genitive and it consists of
‘a verb (specific) + genitive’, for example, noriu arbatytės ‘(I) want some tea-GEN’, reikia meškiuko ‘(I) need a teddy-bear-GEN’. This type preserves the genitive with the negated verb, i.e., ‘a NEG.verb + Genitive’, e.g., nenoriu arbatytės ‘(I) NEG.want tea-GEN’, nereikia meškiuko ‘NEG.need a teddy bear-GEN’. For this type of construction a child does not need to change case marking, therefore it should be easier to acquire. On the other hand, the first construction might present problems because a child has not yet acquired the syntactic rule that the negated verb always governs the genitive case. Because of this fact children, while using negative verbs, retain the case appearing in the positive constructions, namely, the accusative.
Assigning the correct genitive case also interferes with the use of the genitive with those verbs which require a genitive direct object, irrespective of definiteness. We therefore predict that learning when the genitive case has to be used with direct objects may be subject to delayed acquisition.
The main aim of the present paper is to study the acquisition of DOM in Lithuanian.
More specifically, we will look at both correct and erroneous productions of different DOM contexts in order to explore specific constraints of this category. We predict that the children will produce a variety of DOM contexts from very early; however, errors will occur during the course of acquisition, especially in relation to genitive case assignment.
3. THE ACQUISITION OF DOM IN L1 LITHUANIAN 3.1. Data and participants
The present investigation is based on the longitudinal data of two Lithuanian children, a girl Rūta and a boy Elvijus, who were both recorded in various everyday situations (e.g. playing, book reading, eating) interacting mostly with their mothers at their homes. Both children are first-born children in middle-class families living in big cities of Lithuania (Vilnius and Kaunas). All parents have university degrees and use standard Lithuanian. Data collection for the girl started in November 1993 and continued up to 1996, when she was about 3;5. Recordings were made three or four times per week and they last for about fifteen minutes each. For the present study we have chosen to analyse Rūta’s speech covering the period from 1;7 to 2;6, containing 35 hours of recordings. Data collection for Elvijus started in 2006, when the boy was 1;6 and continued until 2;6. This corpus consists of 20 hours of recordings, most of which were made three or four times per week and lasted about fifteen minutes each. The data have been transcribed and coded using an adapted Lithuanian version of CHILDES (MacWhinney 2000). The transcripts
were coded for morphological analysis and double-checked. Adult utterances were transcribed orthographically. The child’s utterances, in addition, were transcribed phonetically. Contextual notes were inserted where necessary.
For the present study, every DOM context was coded according to semantic (definiteness) and grammatical properties (negation). The corpus data are summarized in Table 1:
Table 1
Longitudinal corpus: age range, number of utterances and MLU Age Elvijus
Nr of utterances Elvijus
MLU Rūta
Nr of utterances Rūta MLU
1;7 1072 2.190 294 1.524 1;8 1247 1.315 1035 1.409 1;9 1588 1.205 2521 1.363 1;10 2089 1.333 2597 1.442 1;11 2009 1.366 1565 1.901 2;0 1370 1.699 1817 1.925 2;1 1726 2.092 1726 2.290 2;2 1138 2.234 1767 2.539 2;3 912 2.511 1851 2.553 2;4 1403 2.385 1853 2.666
2;5 1752 2.704
2;6 1000 3.246
3.2. Results
The results section is divided into two subsections. First, we will start the discussion with the presentation of a general view on the acquisition of direct object marking (accusative and genitive case) and its use by children. In the second part, we will discuss contexts of accusative and genitive alternations and the genitive of negation.
3.2.1. The acquisition of direct object marking
The data analysed in this study show (Table 2) that the genitive and the accusative case are already used from the first recordings, namely from 1;7. It is important to point out that the frequency of both case forms at the very beginning of the recordings in the boy’s data is much higher because his onset of morphological development starts two months earlier (he is defined as “an early talker”). The results show that from approximately 1;9 the use of the genitive and of the accusative in both children’s corpora is similar: the accusative is used almost twice more frequently than the genitive. Starting from 2;0, the use of the object genitive, especially the genitive of negation, becomes prominent in the girl’s corpus. The data show that at 1;10 both children already start using the constructions of negation more intensively.
Table 2
Results. The distribution of direct object marking:
Accusative, Genitive and Genitive of negation
Age
Rūt Elv
Acc Gen Neg Gen Acc Gen Neg Gen
1;7 3 0 105 31
1;8 42 17 93 57 3
1;9 107 60 137 53
1;10 134 48 11 149 42 6
1;11 102 47 13 118 67 12
2;0 110 130 45 174 78 8
2;1 214 66 21 122 75 9
2;2 229 76 28 131 31 10
2;3 171 90 23 156 56 7
2;4 125 49 26 133 68 7
2;5 122 52 23
2;6 69 49 7
As for the input results, they are also similar: the mothers tend to use the accusative more frequently than the genitive to mark direct objects (shown in Table 3). The girl’s input demonstrates a much higher frequency of the genitive of negation, however, the girl’s production shows numbers which are almost twice smaller. Elvijus’ mother, on the other hand, uses significantly fewer negative constructions, and a similar pattern is observed in the boy’s data as well.
Table 3
The distribution of direct object marking in the input: Accusative, Genitive and Genitive of negation
Input Rūt Elv
Age
of children Acc Gen Gen neg Acc Gen Gen neg
1;7 16 19 1 408 125
1;8 75 45 5 164 83 13
1;9 228 136 33 142 35 17
1;10 253 88 48 113 40 11
1;11 116 50 30 157 80 19
2;0 143 80 45 154 68 16
Table 3 (Continued)
Input Rūt Elv Age
of children Acc Gen Gen neg Acc Gen Gen neg
2;1 167 55 57 93 47 7
2;2 183 38 39 110 44 8
2;3 194 88 57 71 33 7
2;4 129 42 36 124 55 8
2;5 167 49 35
2;6 36 24 4
The accusative is the default case to mark the direct object in Lithuanian. The children start to distinguish the direct object quite early; however, the formal expression of this function at the early stage of language acquisition takes grammatically incorrect forms.
3.2.2. The acquisition of DOM: the partitive Genitive
The genitive object is attested in both children’s speech since the beginning of our recordings. The frequency of occurrence of this particular object in the children’s speech increases from 1;8–1;9 onwards, and no incorrect usage was attested in the use of the genitive form. The definiteness and quantity parameters do not challenge children, not even at this early stage:
(11) Adult: Ką pirksim? ‘What are we going to buy?’
Child: Duonytės. ‘(some) bread-GEN.’
(Rūta 1;8) (12) Adult: Ko nori? ‘What do you want?’
Child: Pieno. ‘(some) milk-GEN.’
(Rūta 1;9) (13) Mother: Eikime! ‘Let’s go!’
Elvijus: Gejam alaus [= geriam alaus] ‘(Let’s) drink (some) beer-GEN.’
(Elvijus 2;0) 3.2.3. The use of the Genitive with some “special” verbs
However, case marking is not always target-like. In our data we have found some instances of object marking errors: in constructions that require obligatory genitive marking the accusative case was used. This is especially common in some combinations of nouns with the verbs reikia ‘need, must’ and noriu ‘want’, e.g.:
(14) a. *man eikia aikastį [= man reikia laikraščio]
‘I need a newspaper-ACC.’ (Rūta 2;1)
b.* mesiukai reikia ziustą [= meškiukui reikia žiursto]
‘The teddy-bear needs an apron-ACC.’ (Rūta 2;5) c. *noju makaroną [= noriu makaronų]
‘I want macaroni-ACC.SG.’ (Elvijus 2;2) d. *arbatą noriu [= arbatos noriu]
‘I want tea-ACC.’ (Elvijus 2;0) The above examples with reikia ‘must/need’ and noriu ‘want’ demonstrate that it is difficult for children to perceive the syntactic properties of these verbs: problems arise in distinguishing between the classes of verbs which govern either the genitive or the accusative.
The following example reveals the young child’s ability to grasp this distinction of the meaning even at a very early period.
(15) Adult: Ko nori? ‘What-GEN do you want?’
Child: Gybo, duok gybą [= grybo, duok grybą]
‘A mushroom-GEN, give a mushroom-ACC.’ (Rūta 1;9) From the age of 2;0 the correct genitive marking of objects becomes more frequent, and children start using more verbs, e.g. want, need and to be afraid of, which require the genitive case, as illustrated by the examples in (16) below:
(16) a. Miego nori kiškis.
‘Sleep-GEN the rabbit wants.’ (Rūta 2;1)
b. Noriu vėliavos.
‘(I) want the flag-GEN.’ (Rūta 2;1)
c. Kepuytės eikia [= kepurytės reikia]
‘Need the cap-GEN.’ (Rūta 2;1)
d. Kamuoliuko eik.
‘Ball-GEN need.’ (Rūta 2;1)
e. bijau kaimyno
‘(I’m) afraid of the neighbour-GEN.’ (Rūta 2;1) 3.2.4. The Genitive of negation
The investigated data suggest that the use of the grammatically correct genitive of negation causes some problems for both children (this phenomenon was also observed by Smoczyńska 1985 in the acquisition of case marking in Polish). Both children started using the genitive of negation at about 1;10 but it took more time for the girl to master the correct usage (Table 4).
The genitive of negation appears during the period of two-word utterances (because the genitive has to be combined with a verb). It is worth mentioning that the use of the genitive of negation presented certain difficulties for both children, but more for the girl, as demonstrated by the erroneous utterances found in the corpus (Table 4).
The corpus data show that the two mothers use constructions with the genitive of negation quite differently. The girl’s mother uses more of these constructions as they appear in her speech with negative commands, such as nelipk ‘don’t climb’, neimk ‘don’t take’, nedaryk
‘don’t do’, negalima ‘you must not’, etc. Elvijus’ mother uses fewer genitives of negation, and the frequency decreases particularly from 2;2 onwards.
The type of construction with the genitive in both positive and negative contexts are quite numerous and usually occur in a correct form:
(17) a. nejeikia suniuko [= nereikia šuniuko]
‘I don’t need the dog-GEN.’ (Rūta 2;2) b. nebijau gyvatės
‘I’m not afraid of the snake-GEN.’ (Rūta 2;4) c. nejeikia kojinės
‘I don’t need the sock-GEN.’ (Rūta 2;2) We can conclude that this type of construction is relatively easy for children to acquire as it does not require any shift of the form in negative contexts.
However, when children encounter situations where the rule of negation must be applied and the genitive form should be used instead of the accusative, errors are recorded, as in the following examples (most of the instances are observed after the age of 2;0):
(18) a. *petelį nekauda [= pirštelio neskauda]
finger-ACC NEG.hurt
‘The finger does not hurt.’ (Rūta 1;11) b. *neatims ziedą Ūtytės [=neatims žiedo Rūtytės]
NEG.take ring-ACC from Rūta
‘(Someone) won’t take away Rūta’s ring.’ (Rūta 2;1) c. *lempą nepasiekiu [=lempos nepasiekiu]
lamp-ACC NEG.reach
‘(I can’t ) reach the lamp.’ (Rūta 2;1) d. * mociutei ankytę nekauda [= močiutei rankytės neskauda]
for mother hand-ACC NEG.hurt
‘Mother’s hand doesn’t hurt.’ (Rūta 2;1)
The overgeneralized form of the nominative instead of the genitive was attested in an early stage (before the age of 2;0) in negative constructions as well:
(19) a. nemoku sudėt kėdutė [= nemoku sudėt kėdutės-GEN]
(I) NEG.know fold chair-NOM
‘(I) don’t know how to fold the chair.’ (Rūta 1;11)
b. nepagavau maza katytė [= nepagavau mažos katytės-GEN]
(I) NEG.caught the little cat-NOM
‘(I) didn’t catch the little cat.’ (Rūta 1;11)
Despite the fact that some erroneous utterances were attested, the data show correct usage as well. The first constructions with the genitive of negation to mark an object appear in the children’s speech at around 1;9. Later the number of the genitives of negation increases considerably, especially the correct uses of the construction, for instance:
(20) a. neliesk andenuko [= vandenuko]
‘Don’t touch the water-GEN.’ (Rūta 2;0)
b. mesiuko netuju [= meškiuko neturiu]
‘I haven’t got a teddy-bear-GEN.’ (Rūta 2;0) c. neduosiu kamuoliuko
‘I am not going to give the ball-GEN.’ (Rūta 2;0) d. kojyčių neskauda
‘The legs-GEN don’t hurt.’ (Rūta 2;0)
e. neimsiu mašinos
‘I shall not take the car-GEN.’ (Rūta 2;4)
Table 4
Errors: overgeneralized DOM forms
Rūt Elv
Age Acc instead of Gen Gen instead of Acc Acc instead of Gen Gen instead of Acc 1;7
1;8 1
1;9
1;10 4
1;11 2
2;0 5
2;1 10 2 2
2;2 7 1 3
2;3 4 1 1
2;4 6 1
2;5 5
2;6
The overgeneralized genitive, although not to the same extent as in the case of the overgeneralized accusative or nominative, occurs in both children’s data:
(21) a. kamuoliuko galima mesti [= kamuoliuką galima mesti]
‘The ball-GEN is possible to throw.’ (Elvijus 2;3) b. jeikia kepurytės dėti [= reikia kepurytę dėti]
‘It is necessary to put the cap-GEN on.’ (Rūta 2;1) From 2;4 onwards the number of incorrect utterances decreases, especially in Elvijus’
data, and till the end of the research period very few erroneous structures are found.
To sum up, the construction, which at the beginning seemed difficult for the two children to acquire, was fully mastered within a very short period. The overall number of
erroneous utterances was very low: the erroneous accusatives (the accusative case was used instead of the genitive) in the girl’s data amounted to 2.5% and to only 0.8% in the boy’s data.
The number of incorrect genitives (the genitive case used instead of the accusative) was also very low: 2.4% in the girl’s data and 0.6% in the boy’s data.
4. CONCLUSIONS
Even though the Lithuanian DOM system is complex, there were no serious difficulties for the children to correctly produce the required case marked forms. Despite a number of similarities observed in the early acquisition across different languages, every language (and every child, for that matter) exhibits some idiosyncratic properties. It should be stressed in the first place that despite the complex nature of DOM and despite the case marking errors attested in the two corpora (with the genitive or the accusative), the data clearly show, at the same time, that children were producing correct forms as well. The most typical errors were (1) overgeneralization of the accusative and (2) the use of an inappropriate case form after a negated verb. In negative constructions, the children tended to use the accusative case instead of the genitive to mark direct objects; few instances of the nominative instead of the genitive were recorded.
The use of the accusative case in genitive contexts was especially frequent in both corpora. It is important to point out that the observed overgeneralizations in the data of the Lithuanian children’s performance were not accidental. The evidence for this situation comes from the genitive of negation construction (see also Mačiukaitė 2006). As the presented examples show, the partitive genitive posed almost no difficulty for the children:
in Lithuanian the partitive construction is always marked with the genitive, and there are no possibilities for alternations (unless specific pragmatic contexts are invoked, but these were not observed in the data). Therefore, this might be the reason for a child to acquire it early and easily. However, the genitive of negation does present some challenges. In Lithuanian affirmative clauses complements of transitive verbs take the accusative case. Negated verbs require genitive case assignment to their direct objects. This is a language-specific constraint that the child has to acquire. As noted above, both children were overgeneralizing the use of the accusative and sometimes of the nominative in the “genitive of negation” contexts. The specificity of this construction caused erroneous production and it took longer for the children (especially the girl) to figure out some language-specific rules; however, as our data show, the process of overcoming language-specific constraints was not very long. The study shows that the process of DOM acquisition was gradual: the children started from positive contexts and learnt the rule for using the accusative; at the same time the children showed mastery of the partitive genitive at a very early stage. The negative contexts that require the genitive case were more difficult for children to acquire, and more errors were attested in the data.
The remaining question to be resolved concerns identifying the factors which facilitate the early mastery of this complex phenomenon. This is especially surprising if we take into consideration (as discussed in section 2) the point that the rules for DOM vary and are not straightforward. Answering this question requires further studies and data analysis of the use of DOM by older children as more ambiguous DOM contexts might appear at an older age.
REFERENCES
Aissen, J., 2003, “Differential object marking: Iconicity vs. economy”, Natural Language and Linguistic Theory, 21, 3, 435–448.
Aleksandravičiūtė, A., 2013, “The semantic effects of the subject Genitiveof negation in Lithuanian”, Baltic Linguistics, 4, 9–38.
Ambrazas V. (ed), 1997, Lithuanian Grammar, Vilnius, Baltos lankos.
Arkadiev P., 2011, “Aspect and actionality in Lithuanian on a typological background”, in: D. Petit, C. Le Feuvre, H. Menantaud (eds), Langues baltiques, langues slaves, Paris, CNRS Editions, 61–92.
Bossong, G., 1991, “Differential object marking in Romance and beyond”, in: D. Wanner, D. Kibbee (eds), New Analyses in Romance Linguistics, Amsterdam, John Benjamins, 143–170.
Comrie, B., 1979, “Definite and animate direct objects: A natural class”, Linguistica silesiana, 3, 13–21.
Dressler, W. U., (ed), 1997, Studies in Pre- and Protomorphology, Wien, Verlag der Österreichischen Akademie der Wissenschaften.
Guijarro-Fuentes, P., T. Marinis, 2009, “The acquisition of the personal preposition a by Catalan- Spanish and English-Spanish bilinguals”, in: J. Collentine, M. García, B. Lafford, F. M.
Marín (eds), Selected Proceedings of the 11th Hispanic Linguistics Symposium, Cascadilla Press, 81–92.
Kagan, O., 2005, Genitive case: A modal account, Ms. Jerusalem.
Kagan, O., 2010, “Genitive objects, existence and individuation”, Russian Linguistics, 34, 1, 17–39.
Leonetti, M., 2004, “Specificity and differential object marking in Spanish, Catalan Journal of Linguistics, 3, 75–114.
Lestrade, S., H. de Hoop, “On case and tense: the role of grounding in differential case marking”.
Unpublished manuscript, Radboud University Nijmegen.
MacWhinney, B., 2000, The CHILDES Project: Tools for Analyzing Talk, Transcription, Format and Programs, Mahwah, Vol. I, NJ, Lawrence Erlbaum Associates.
Mačiukaitė, S., 2006, “Making the case: Acquisition of Lithuanian Genitive”, Žodis ir kontekstas, 48–58.
Marchman, V. A., E. Bates, 1994, “Continuity in lexical and morphological development: a test of the critical mass hypothesis”, Journal of Child Language, 21, 2, 339–366.
Montrul, S., 2004, “Subject and object expression in Spanish heritage speakers: A case of morphosyntactic convergence”, Bilingualism: Language and Cognition, 7, 2, 125–142.
Montrul, S., M. Bowles, 2009, “Back to basics: Differential Object Marking under incomplete acquisition in Spanish heritage speakers”, Bilingualism: Language and Cognition, 12, 3, 363–
383.
Neidle, C., 1988, The Role of Case in Russian Syntax, Dordrecht, Kluwer.
Partee, B. H., V. Borschev, E. Paducheva, Y. Testelets, I. Yanovich, 2012, “The role of verb semantics in Genitive alternations: Genitive of negation and Genitive of intensionality” in:
A. Grønn, A. Pazelskaya (eds), The Russian Verb, Oslo Studies in Language, 4, 1, 1–29.
Rodríguez Mondoñedo, M., 2008, “The acquisition of differential object marking in Spanish”, Probus, 20, 1, 111–145.
Seržant I., 2014, “The independent partitive Genitivein Lithuanian”, in: A. Holvoet, N. Nau (eds), Grammatical Relations and their Non-Canonical Encoding in Baltic, 257–299.
Silva-Corvalán C., 1994, Language Contact and Change, Oxford, Oxford University Press.
Smoczyñska, M., 1985, “The acquisition of Polish”, in: D. I. Slobin (ed), The Crosslinguistic Study of Language Acquisition, Vol. 1, Hillsdale, NJ, Erlbaum, 595–686.
