CC BY
28
Velka Popova
COGNITIVE ASPECTS OF THE BULGARIAN VERB ACQUISITION
The purpose of this research is to trace the developmental steps in the acquisition of verb in Bulgarian and to examine the validity of the Hypotheses on early form-function mapping in first language acquisition.
0. Introduction
In the current work an attempt has been made to represent the main aspects of the acquisition of the VERB grammatical category in the ontogenesis, in the context of the concept that the early language development is a step-by-step process of differentiation and further specialization of the primary global cognitive system on the base of a figurative-basic opposition, determined by the perception symptoms.
The researcher's attention is focused on the earliest stages of the formation of the verbal lexicon in the ontogenesis. An attempt has been made to outline the universal and the language-specific aspects during this process. The main steps in the formation of the children's lexicon have been traced in the context of the overall language development, i.e. - there has been reported the relation between the morphology and the other components, on the one hand, and the relation between the verbs morphology appearing and the rest of morphological subsystems, on the other.
The complete corpus of the material analyzed contains the records of the spontaneous speech of six Bulgarian children (Stefka, Diana, Alexandra, Ivelin, Stefani, Bogomila), marked in the relevant longitudes in CHAT-format of the CLAN program as: *STE, *DIA, *ALE, *IVE, *TEF, *BOG [cf. Попова, Попов 2004]. In the center of the researcher's attention, however, there will be predominantly the data from the longitudinal observations of two of the children explored (*ALE and *TEF), because they had been systematically recorded for several months (there have been analyzed totally 7474 utterances: 3871 utterances of *ALE from the period 1;01-2;04 and 3603 utterances of *TEF from the period 1;03-2;05) and this would enable us to outline the main trends, characterizing the early acquisition of the verbs lexicon on the background of the overall process of the language development. The statistic processing of the data has been accomplished with the help of the resources of the CLAN program package. The data from the other four persons explored will be quoted only for comparison and illustration due to their incomplete and episodic character.
1. Emergence of word class
The origin of the first words in the ontogenesis ranges from 8- to 18- month-age in the individual children's cases. They function as components of the one-word-utterances and the primary two-word-utterances, but they do not have any morphological characterizations yet. So, immediately there emerges the subject about the essence and the classification of these primary language units. Their grammatical amorphousness, which is determining for the whole pre-grammatical stage, cannot be connected at all with the idea of availability of lexico-grammatical categories, though most of the researchers use terms «noun», «adjective», «verb», etc. when describing these earliest stages of development. It must be noted, however, that in this case only work ideas are spoken of, because in «the pre-morphological phase no system of grammatical morphology has yet become dissociated from a general cognitive system that handles, inter alia, words of whatever form (including morphological forms), i.e. pre- and at least early pro-tomorphology are part of the lexicon» [Dressler 1997: 11]. The beginning of the acquisition of the basic oppositions between the categories of words and within the flective systems of children cannot be expected until the stage of the transition to the proto-morphology.
Within the framework of the proto-morphology, characterized by movement in the direction to morphological analysis and recognition of models, there has been observed quantitative enrichment of the morphological, lexical and syntax structures. Thus, in the cases of both children explored there has been registered the overcoming of the one-word utterance as well as increasing of the lexical diversity, accompanied by the process of accumulation of verb forms (in contrast to the pre- morphology, where we have one verb lemma per one form learnt by heart - cf. Table № 1 for ALE and for TEF).
The appearing of the verb morphology has been connected with such an increasing of the verb lexical forms, which at a certain moment leads to reaching the so-called «critical mass». Thus, the continuous accumulation of inflexion forms learnt by heart stops and there appears kind of a creative
mechanism for the morphological principles of sensible forming and recognition as well as morphological segmentation, this happening after a period of many attempts and guess-work. Actually, the transition from lexical processing to morphological models building is not an automatic result of the lexical information entered, but it is a process of active constructing by the child [cf. Karmiloff-Smith 1992]. Thus, among the increasing number of word forms, entering the lexicon of both the children during the transitional stage, the verb forms apparently have reached the critical mass degree both in keeping and in processing information. As a consequence there has been accomplished the separation of the verb in its own field; through the most frequently used form verb properties are given to these lexemes. In other words, in the speech production of the children there have already been recognized the forms which signalize the verb concept, as well as the ones that don't do so. This gives a good reason to consider that the transition from pre-morphology to proto-morphology is a crucial moment in the grammar ontogenesis, because just then the child passes from mechanic memorizing to productive usage of the morphological characteristics.
In the cases of the both children observed -ALE and TEF - there has been registered early appearing and expansion of the verb miniparadigms, which could be interpreted as an expression of the first form-function mapping. In this sense, they are a new step in the process of the acquisition of the morphology and the flective system of the Bulgarian language. It should be outlined, however, that these mini-paradigms are different from the paradigms of the adults. The form used often is not determined for person and number, as it is assumed to be in the adults' language. There has been observed a tendency
towards using the forms from the miniparadigms in a more general meaning, which shows that the child already correlates features, extracted from certain situations, with forms taken from the language perceived. Therefore, it can be suggested that the transference processes lead first to distinguishing between limited and unlimited essences and this is extrapolated respectively into the children's lexicon as initial global noun/verb differentiation, and later this can be expected also in the correlation of form and meaning in the further specialization of the verb grammar in the process of language acquisition.
2. Early establishing of the form-function relation
In the ontogenesis the form-function transference processes are first realized in the distinguishing of limited-unlimited essences, which is observed in the initial global differentiation in the early children's lexicon into two fields: verb and noun. The initial separation of word categories could be interpreted as the first proto-grammatical opposition [Bittner 2003]. In the cases of both girls this differentiation of the early children's lexicon is accompanied by the tendency towards domination of the nouns at the expense of the verbs.
The first contrastive forms of noun have been detected in the both children's cases even in the pre-grammatical stage. This early acquisition can be explained, on the one hand, by the fact that a noun represents prototypically clearly outlined objects [Ol-guin, Tomasello 1997], and on the other hand, in the formal aspect, noun is remarkable for the much greater simplicity of the paradigm in comparison with a verb. Thus, «for the Bulgarian language there exist morphological differences, which make the nouns «easier» for acquisition as formal-grammatical category, than the verbs: this is the limited ranging within the framework of the noun paradigm - four (for some nouns of masculine gender - five) word forms versus the dozens of paradigm variations of the verbs (even in the too limited entry data for the small children). A noun in singular without article satisfies most of the contexts, which a Bulgarian child faces at the beginning of the verbal communication. There doesn't exist however a single verb word form, which is analogically «universal» in functional aspect» [Стоянова 2002: 323-324] [translated by me - VP]. Even in this limited amount, in which they are in the entry data, the Bulgarian verbs are distinguished by their extreme diversity of paradigm variations, which turns into an obstacle for their acquisition. The situation gets considerably more complicated also due to the
Table № 1
ALE TEF
age lemmas types tokens age lemmas types tokens
1;01.29 7 8 43 1 ;03.11 2 2 5
1;05.06 26 31 92 1;06.22 5 5 10
1;06.21 33 42 87 1;08.0 14 17 62
1;07.20 45 49 87 1;11.04 56 74 167
1;08.22 70 90 203 1;11.25 39 57 140
1;09.19 66 97 154 2;0.09 37 57 174
1;10.12 59 88 174 2;0.23 45 53 120
1 ;11.14 69 98 188 2;01.16 43 59 134
2;0.13 34 49 71 2;02.16 38 62 147
2;02.24 90 131 270 2;03.23 85 128 387
2;03.12 115 199 358 2;04.21 57 81 172
2;04.09 81 119 273 2;05.25 72 106 257
availability of aspectual couples, which have as a referent one and the same event.
In the process of the language development there follows a further specialization in the field of the verb (imperfectives and perfectives), based on the «result-process» differentiation [Slobin 1985], predetermining also the children's early marking of various actions. Both children start using verbs even at the very beginning, those usually being present forms for marking the action, going on at the moment of speaking. They, however, cannot be interpreted as means of temporal deixis yet, because there has not been observed any opposition within the framework of the temporal-aspectual system yet.
The most important proofs of the availability of morphological analysis in the transitional stage are just the cases of over-generalization. They play an essential role in the first proto- grammatical differentiation, built over the idea of the verb as «a moving pic-ture» in opposition to the noun as «a static frame» [Tomasello 1992: 11]. The early acquisition of the verbs in the Bulgarian language is characterized by the extended use of the Present forms for 3rd pers., singular, which in the cases of both girls observed play the role of default- forms. This is confirmed not only by the data of the other four children, but also by the observations of other researchers [cf. Стоянова, Фетваджиева 1986: 32]. They are the evidence that reaching the critical quantity in the use and memorizing of verbs, the child interprets the most significant formal models as a signal for the fact that the referent of the linguistic unit is «unlimited» regarding other essences.
The over-generalization of the aoristic participle as the next important step in constituting the form-function relation is connected already with the field of the verb itself. These forms represent the situation as a completed, finished one, in opposition to the situation of speaking. Thus, in the field of verb, connected so far with referents, perceived as «unlimited» there has been differentiated a considerable sub-category of el-forms, which are connected with referents, perceived as 'bounded (whole)' / 'limited (entities)'.
3. Forming of the verb lexicon
The origination of the real verbs in the ontogenesis is preceded by a rather long period, in which the child not only becomes aware of the actions, processes and conditions of the complete situation and segments them, but also marks them verbally.
From the very beginning of the recording, both children use onomatopoeias, marking by them the situations completely and not - partially. For exam-
ple, when TEF (1;06) looks at the table as well as at the cake, plates and the utensils put over it, she says: «am-am». This am-am may practically mean anything, involved in the situation: the cake, the utensils, the participants, the process of eating. The mixed use of the onomatopoeias without any clear referential relation is typical. So, for example, lyu-lyu in the speech of TEF (1;06) can be observed regularly both in the sense of 'cradle' and 'to swing':
Context 1: Context 2: Context 3:
*BAB: How do you do in the cradle, Granny? *BAB: How do you swing? *TEF: Lyu-lyu.. *BAB: Where will we go to? *TEF: Lyu-lyu. *BAB: What is this? *TEF: Lyu-lyu.
*BAB: And what will we do there? *TEF: Ulka [:cradle]. *BAB: Where will you go with uncle? *TEF: Ulka [:cradle]. *BAB: What is this? *TEF: Ulka [:cradle].
From the examples indicated it immediately becomes clear that the parts of speech have not been differentiated yet, though in the child's production there are already formal possibilities for this (cf. the later realizations from the sort of Teti lyu-lyu, in which clearly differentiable is the predicative function of the onomatopoeia, versus uses from the sort of Mau-mau papa).
Later, the onomatopoeia is used more often particularly for marking the perceived subject or event (action). At this stage of the language development (before the verb appearing) the question of the adult determines to a certain degree the predicative function of the onomatopoeia and thus aids the forming of the verbs category. The use of the onomatopoeia is an important stage in the acquisition of the verb morphology, because it helps the kid: a) to learn how to make difference between events (actions) and subjects; b) to recognize different kinds of situations (with the help of the interlocutor's explanations) and to mark them initially by onomatopoeias, and then -also by the relevant verbs. The adults' strategy to repeat selectively the onomatopoeia with the relevant verbs and nouns can additionally catalyze the process of forming word categories. In this way, a moment is reached, when in the speech production of the child there has been clearly outlined the tendency towards dissociation of the global system into two fields, and namely - that of the noun and that of the verb. Thus, for example, in the longitudinal data of *TEF at the age 1;08.0, when there had already been made the first steps in this aspect, there are observed specific uses of onomatopoeias, in which there has been
clearly outlined the striving for formal representation of a fragment of the reality once as a complete (limited) essence, i.e. - as a noun, and in another context - as an unlimited essence, i.e. - as a verb.
Similar onomatopoeic realizations in the speech of *TEF are not the only proof of the availability of primary lexical paradigms during the age discussed. Together with them, the aspiration of the child to mark certain lexemes as verbs is also shown in the use of verb flections for naming actions, though very often this is accomplished incorrectly. Thus, in the child's utterances there are flections for 1P SG, 1P PL; 3P SG (Present); 3P SG (Aorist); 2P SG (imperative); as well as a form for 3P SG (perfec-toid). As an illustration, there will be quoted the verb lemmas (together with the formal types) and the onomatopoeias, which marked 'action' in the speech passage analyzed: 14 verb lemmas // 61 tokens // 6 types versus 11 onomatopoeias // 16 tokens (*TEF -1;08.0). The statistical analysis shows that at the age 1;08.0. In the presence of synonymous means for formal marking of a certain verbal concept, *TEF demonstrates an apparent preference for the verb forms at the expense of the onomatopoeic ones, which is illustrated in Table № 2.
Table №2
verbs onomat
Lemmas 14 11
Tokens 61 16
action 'feeding' 17 4
In the further development there has been observed the tendency towards functional limiting of the onomatopoeias and in a certain moment it has even resulted in encountering them only in predicative use (cf. for example *TEF- 1; 11.04: Granny, pish (= 'I will piss'); Bebito. Kot-kot (= 'jumps'); Dus-dus (= 'I joggle'); Rm-rm-rmy-rmy (= 'growls'); Botzi ('I stitch'); Am (= 'Eat!')). Some of them change also their phonetic appearance and start looking like verbs (for example - the botzi form instead of the usual so far botz / botz-botz onomatopoeia is used), and some of them have already totally lost their onomatopoeic character and are perceived mostly as elements of the relevant verbal paradigm (for example the chipi form, used in parallel with chipa, has been associated with 3P SG PRESENT form and could be explained by the peculiar competition of the language forms at the entry during the process of constituting the miniparadigms).
The role of onomatopoeia as «laying the foundation» for the verb appearing is undoubted. In both children's cases there has been observed a period of
accelerated use of onomatopoeias in predicative function. With the appearance of the verb (as well as their contrastive forms), the reverse tendency has been detected - in both children's records there has been registered decreased use of onomatopoeias (cf. for example Diagram № 1, which illustratively represents how even from their origination, the first verb forms become apparently preferred by *TEF). Thus, for example, in the case of *ALE the first significant decrease is from 15,4% (during the age of 1;01.29) to 5,3% (during the age of 1;05.06). After an unimportant increase at 1;07 to 7%, later this percent does not exceed 2%. The first decrease in the use of onomatopoeia is observed during the transition from pre- to proto-morphology, when the first proto-morphological signs appear; the second considerable decrease, observed at the age of 1;08, is accompanied by increase of a relative share of the utterances with verbs (more than 50%), which coincides with the beginning of the proto- morphological stage and the appearance of real mini-paradigms. This is illustrated in Diagram № 1.
Diagram № 1
From Alexandra's and Stefani's data analyzed it becomes clear, that during the earliest stages, the predicative functions can be expressed by onomatopoeias (including onomatopoeic reduplications), i.e. -by extra-grammatical means (for more details cf. Стоянова 2002). Little children have a rich repertoire of means for expressing predication, even before they have mastered the correct verb forms. But though even they have forms of specific verbs, expressing pragmatic or morpho-semantic sense, they may use in the same meaning non-verbs as well (called also predecessors of verbs), together with the onomatopoeias (cf. also: dolu (= down) in the meaning of 'take down!'; opa in the meaning of 'take up!'; na in the meaning of 'take!'/'give' (*ALE, 1;01.29)), peculiar fillers, which constitute the phonetic structure of the relevant verb being mastered from the sort of titit-
eni, tititani respectively in the meaning of 'sit down' and 'stand up' (*BOG, 1;11). During the transitional stage from pre- to proto-morphology, these verb predecessors are gradually replaced by the real verb lemmas, whose increase, however, is insignificant. This is an evidence of the instability of the verb system, in which pre-morphological and proto- morphological elements cohabit. The appearing of verbs is followed by dramatic changes in the whole grammar system: contrastive forms - conjugated forms for perfect and imperfect type (in the case of verbs) appear; there are observed changes in the structure of sentence (sentences with subjects and predicates) and children start an intensive movement towards generalization of grammatical rules. The dominating tendency to grammaticalization during this period apparently suppresses the lexicon's extension. Considerable and stable increase of the number of the verb lemmas is accomplished only after entering the proto-morphology stage, which is shown on the base of the material from *ALE in Diagram №2.
Diagram №2
The perfective (PFV) and imperfective (IPFV) verbs have been detected almost simultaneously in both the children's lexicon, the imperfective ones dominating initially, but gradually with the age, there has been noticed the reverse tendency (Table №3). Their early opposing can be accounted for by the fact that the imperfective verbs usually represent tangible acting process, and the perfective verbs - perceptible (or experienced) result of a concrete action, accomplished not too long ago.
Table №3
*ALE *TEF
Age IPFVj PFVl AgeH IPFVH PFVl
1;03 - 1;09| 69%! 31%! 1;03 - 1;09| 78%H 22%И
1;09 - 2;0l| 58%! 42%! 1;09 - 2;01| 52% 48%И
2;02 - 2;04 46% 54% 2;02 - 2;04| 40%И 60%И
During the transitional stage there has been made the start with a more active production of perfective verbs. In the cases of both the children early acquisition of the productive technique for perfectiva-tion with the help of the -n- suffix is observed, which probably could be explained by its high constructive iconicity and transparency. In this aspect, quite naturally, as it has been expected, the prefixation turns to be mastered later and this is synchronized with the universal principle, formulated by D. Slobin: «pay attention to the ends of words» [Slobin 1973: 191]. Together with this, the prefixation suggests also a much more complicated complex of semantic nuances and a lower degree of iconicity in comparison with the n-suffixation. In Table №4 there has been schematically shown the sequence in appearing of the prefixation and suffixation.
Table №4
*ALE *TEF
Age suffix-N- prefix Age suffix-N- prefix
1;03 -1;09 16% 5% 1;03 -1;09 0 0
1;09 -2;01 15% 26% 1;09 -2;01 11% 2%
2;02 -2;04 13% 26% 2;02 -2;04 18% 10%
As a result of a detailed analysis of the data, connected with the acquisition of the perfectivation through prefixing and through n-suffixation in the cases of both the children observed (3678 uses of verbs: 1907 in the case of *ALE and 1771 by *TEF), it becomes clear, that both the girls comparatively early master (and to a high degree) the diversity of possibilities for expressing perfectiveness, provided by the input. This is particularly clearly outlined in the case of Alexandra (early speaker), who within the framework of the period observed, has passed through the pre-morphological, transitional and proto- morphological stages and has entered the morphological stage. In her case, till the end of the observations, there have been mastered 13 (iz-, raz-, za-, na-, o-, ot-, po-, pod-, pre-, pro-, sa-, s-, u-) of the 18 Bulgarian prefixes for perfectivation (72%), there has been registered also a considerable lexical diversity (84 lemmas). In Stefani's case, as it can be seen in the table, the suffixation acquisition starts significantly later and there has not been accomplished to a sufficient degree the diversity of the target prefixes for perfectivation - during the period observed there had been registered only seven of them (iz-, na-, po-, pre-, raz-, s-, u-), and, namely, 39% accomplished at 14 lemmas. This significant difference between the results of the two children can be explained by the fact
that Stefani is a later speaker and within the framework of the period observed, she has not managed to enter the morphological stage of her language development.
In the process of evolution of the verb lexicon, it is important to note the appearing and the following displaying of the aspectual couples, including also the inflected cognate verbs for marking different ways of action. They could be given as a proof of the accomplishing of the next stage in the acquisition of the verb lexicon. In the cases of both the girls, the appearing of similar couples for representing an action from a different point of view, has been registered towards the end of the transition from pre- to proto-morphology (in Stefani's case - at the age of 1;11.04. And in Alexandra's case - at the age of 1;07.20). Till the end of the period observed, in the records of Alexandra there have been detected 44 different verb couples, and in Stefani's case - only 17.
The acquisition of the verb couples during the period observed passes with extremely small amount of mistakes. Before the appearing of prefixes and the -n- suffix, there has very often been observed nontarget use of non-prefixed imperfectives in perfective context. Later, with detecting verb couples, there have been met only single cases of non-target use of imperfective verb instead of perfective one and vice versa. In the data of both the children, significant is the presence of the productive model «perfective verb base + suffix for imperfectivation», which is the reflection of the strong limitation in the contemporary Bulgarian language of the tendency to change the basic vowel [KocnapTOBa 1969]. In the corpus of the verb couples the opposition has been built exclusively on the affixation base, while the suppletivism and the change of the root vowel are narrowed to single cases.
In order to generalize everything said so far, it is necessary to pay attention to the mistakes, which are made in the process of the acquisition of perfec-tives and imperfectives. Here, we must immediately note, that for the whole period observed their quantity is insignificant. The character of mistakes changes alongside with the development of the children's morphological abilities. At the beginning of our observations the cases of missing perfective affixes prevail. The children use imperfectives in perfective contexts and there has been noticed compensatory appearing of a certain subject in the utterance. Alongside with it we registered only occasional non-target uses of perfectives instead of imperfectives where a non-determined subject appears instead of a determined subject. At the end of the period observed the deviations registered have quite a different character.
In these cases we already speak about the over-generalization in the acquisition of the semantics of prefixes, which is a proof of the next stage in the formation of the language competence, when the child is already ready to accomplish a further semantic specialization within the framework of the verb vocabulary (expressing different ways of the action).
The appearing and the following display of aspectual couples, as well as of verb synonyms and cognate verbs to mark different ways of the action, is a proof not only for the further development of the verb lexicon, but probably of the verb grammar itself.
4. Conclusion
The forming of the Bulgarian verbal lexicon in the ontogenesis is detected even at the earliest stage, when the gradual differentiation of the general cognitive system and its further specialization in the field of verb is accomplished within the terms of opposition [±BOUNDED].
The corpus of longitudinal data analysis provides for the following main steps in the forming of the early verb lexicon, in morphotactic and mor-phosemantic aspects respectively:
I. Main morphotactic steps in the ontogenesis of the Bulgarian verb lexicon:
^ Zero start, characterized by unclear use of verbs of both types, learnt by heart, accompanied by quantity priority of the non-marked verbs (imperfec-tives) in both children's cases.
^ Appearing and detecting of perfective verbs, marked with -n-suffix to note single actions.
^ Appearing and detecting of perfective verbs marked with prefixes.
^ Detecting of different prospects of representing one and the same action in ontogenesis: early verb synonymy; appearing and detecting of aspectual couples.
II. Main steps in the detecting of the early aspectual semantics:
^ [± Continuance] of the action in the onomatopoeias stage.
^ [± Completeness] (effectiveness / procedure-ness) - the over-generalizations, the -n-suffixation.
^ Semantic variation - detecting and extension of the prefixation, passing through a stage of over-generalization of individual prefixes.
References
Коспартова М. Особености на детския говор // Български език. - 1969. - № 5.
Попова В., Попов Д. Програмата CLAN и приложението й в изследването на детската реч //
Сборник научни трудове, посветен на 100-годишнината от рождението на Джон Атана-сов. Том II. - Шумен: УИ «Епископ Константин Преславски», 2004.
Стоянова Ю., Фетваджиева С. Някои про-блеми на прагматиката в развитието на детската реч // Съпоставително езикознание. - 1986, №1.
Стоянова Ю. Ранните детски изказвания между прагматика, семантика и граматика // Езикът: история и съвременност. - Шумен: УИ «Епископ Константин Преславски», 2002.
Bittner D. The emergence of verb inflection in two German-speaking children // Development of verb inflection in first language acquisition. From first form to miniparadigms. - Berlin/New York: Mouton de Gruyter, 2003.
Dressier W.U. Introduction // Pre- and Proto-morphology in Language Acquisition, 7-14 (Papers and Studies in Contrastive Linguistics 33). - Pozna^ Adam Mickiewicz University, 1997.
Karmiloff-Smith A. Beyond modularity. A developmental perspective on cognitive science. -Cambridge, Mass, 1992.
Olguin R., Tomasello M. Twenty-five-month-old children do not have a grammatical category of verb. // Cognitive Development. - 1997. - Vol. 8.
Slobin D. Cognitive Prerequisits for the Development of Grammar // C.A. Ferguson, D. Slobin (eds.). Studies of Child Language Development. -New York, 1973.
Slobin D. Crosslinguistic evidence of the language-making capacity // D. Slobin (ed.). Crosslinguistic Study of Language Acquisition. - 1985. -Vol. I. Hillsdale: Lawrence Erlbaum.
Tomasello M. First verbs. A case study of early grammatical development. - Cambridge: Cambridge University Press., 1992.
ВелкаПопова
КОГНИТИВНЫЕ АСПЕКТЫ УСВОЕНИЯ БОЛГАРСКОГО ГЛАГОЛА
В ОНТОГЕНЕЗЕ
Целью данного исследования является выявление этапов формирования словаря и грамматики болгарского глагола, а также проверка гипотезы о раннем стабилизировании отношения форма-функция в процессе усвоения родного языка (Ьц).
