ArticlePDF Available

Morphological development of Bangla-speaking children: A pilot study

Authors:

Abstract and Figures

Verb inflections have often been found to be challenging for children to acquire. This paper explores the methodological specifications for eliciting a set of inflected forms in Bangla morphology that is of interest for examining children’s morphosyntactic development, based on a pilot study conducted with 20 monolingual Bangla-speaking children between ages two and four. The study identifies a preliminary picture of development with regard to the grammatical markers examined, and puts forward a set of guidelines specific to examining those markers in larger studies.
Content may be subject to copyright.
1
AN EXPLORATION OF THE MORPHOSYNTACTIC DEVELOPMENT OF
BANGLA-SPEAKING CHILDREN: FINDINGS OF A PILOT STUDY
ABSTRACT: Although verb inflections have often been found to be challenging for children, the
developmental patterns in Bangla still remains largely unexplored. The paper explores the methodological
specifications required to elicit a set of inflected forms in Bangla morphology that is considered of
interest for examining children’s morphosyntactic development. The paper arises from a pilot study
conducted with 20 monolingual Bangla-speaking children between age two and four. Based on children’s
responses, the study identifies a preliminary picture of development with regard to the grammatical
markers examined, and forwards a set of guidelines specific to examining those markers in larger studies.
KEYWORDS: language acquisition, verb morphology, Bangla, language elicitation technique
0. INTRODUCTION
Analysis of child language has been an area of curious investigations for linguists, psychologists
and speech-therapists. Due to the ongoing research into a range of language domains, such as
vocabulary, morphosyntax, and phonology, significant information is known today about
children’s language acquisition process. Historically, child language research began as diary
studies in the early 20th century (Ingram 1989), which today has evolved into elaborate and
systematic investigations that involve a variety of testing instruments. A milestone in child
language research was the Harvard project conducted by Roger Brown and his colleagues (e.g.
Brown 1973). The studies in the Harvard project forwarded important findings about the
language use in early childhood, with particular regard to the area of morphosyntactic
Published in The EFL Journal, 7(1), 73-
92. This is the authors copy.
2
development, which have been reinforced in later studies, and are still commonly considered
guidelines of language acquisition research.
Apart from being a window to the general cognitive development of children, child language
research contributes significantly to a range of associated linguistic studies such as
computational linguistics and neurolinguistics. Moreover, a linguistic examination of typical
development is crucial for assessing the progress of the children with language difficulties, and
for developing intervention activities accordingly.
Bangla child data has not been examined to the extent required for identifying a reasonably
detailed profile of typical development. The repercussions of this are manifold. The dearth of
research on Bangla child data adversely affects all child data-based research mentioned earlier.
Having no profile of typical development particularly affects speech and language therapy, since
an accurate account of children’s language difficulty cannot be identified in absence of a typical
profile. Also, Bangla being one of the largest spoken languages (Comrie 2005; Klaiman 2008),
Bangla-speaking people with language difficulties represent a significant number of population.
Therefore, the absence of first language data results in compromised intervention services for
this population, and at the same time therapy motivated from the first language research is likely
to help a significant number of people.
1. CROSSLINGUISTIC FINDINGS OF MORPHOSYNTACTIC DEVELOPMENT
Crosslinguistic investigations of child language acquisition not only enhance the literature of
acquisition research of the language itself, and thereby supplement the associated research in the
same language, they also refine the understanding of the universal principles of language
3
acquisition. Increasing number of new child data from a variety of world’s languages has been
contributing towards building the current knowledge of child language development.
The studies conducted on a range of typologically diverse languages have identified that
typological properties of the language is reflected on children’s language use. Children learning
an agglutinative language (e.g. Turkish, Tamil) are different from their fusional (e.g. English,
Italian) language-speaking counterparts with regard to the errors they make. Children exhibit
significantly higher accuracy scores when the target languages offer agglutinative properties
compared to when the target language is a fusional one (Acarlar & Johnston 2011; Raghavendra
& Leonard 1989). The explanation lies in the observation that agglutination allows for a degree
of linearity and transparency in the morphological paradigm (Aksu-Koҫ & Slobin 1985) and
results in enhanced evidence for the markers (Pinker 1984), which contribute towards the relative
ease of mastering the language. Also, morphological complexity of the target language has been
identified to contribute in children’s performance; children speaking morphologically rich
languages have been found to have a faster rate of acquisition (Xanthos et al. 2011), and their
errors in the target language are often non-finite inflected forms as opposed to the bare forms
observed in morphologically sparse languages (e.g. English, German) (Phillips 2010).
A common finding in child data is that children’s production the grammatical inflections is
regulated by the cognitive demands of the target forms. Morphological markers that are deemed
cognitively complex often obtain lower accuracy scores (e. g. Aksu-Koҫ & Slobin 1985). In the
same line is the finding, irrespective of the language typology, that children often tend to
substitute the target markers with other non-target forms in case of unavailability of the markers.
These forms are typically less specified with regard to the morphological embellishments (Tamil:
Lakshmanan 2006; Italian: Leonard, Caselli, & Devescovi 2002; Hebrew: Lustigman 2012).
4
Children’s extensive use of the less challenging forms appear to be in line with the processing
constraint-based account of language acquisition; children’s utterances are regulated by their
processing capacities, and in the face of limited facilities, their utterances result in a
compromised production of the language (see Leonard 2014 for a review).
2. ELICITATION TASK AS A RESEARCH TOOL
The early investigations of child data were primarily based on spontaneous language samples,
which presented data in a relatively unstructured format. Language samples are still popular due
to the ease of data collection, and for offering a direct and comprehensive portrayal of children’s
performances. However, structured research tools such as sentence elicitation tasks have also
gained popularity due to it the control it offers to the researcher. When specific language items
are warranted from the data, elicitation tasks are considered ideal. Spontaneous language
samples, because of the nature of the task, often fail to obtain enough and comparable evidence
of the target utterance use, which may make them ineffective for assessment. Therefore, it is
commonly recommended in research to collect converging evidence from spontaneous and
structured data for building specific as well as comprehensive impressions of children’s
performance (e.g. Eisenbeiss 2010).
Language elicitation tasks can be very effective, once a particular area of interest can be
identified in child language research. Structured probes that are designed so that a particular
response can be obtained from children are used widely because it allows for testing specific
items within a variety of linguistic contexts. Since the present study aims to test a specific
language item, i.e. use of a set of verb inflections, elicitation task was deemed appropriate for
5
employing in this study. It is acknowledged that a profile of children’s use of verb inflections
needs additional evidence from spontaneous language samples.
3. PRESENT STUDY
The present study, as a pilot, was conducted as a preliminary assessment of the research
parameters of developing an examination of morphosyntactic development of Bangla-speaking
children. With this view, the study aimed to, firstly, identify the testability of a set of
morphological forms in Bangla through elicited production task, i.e. to identify whether the
forms lend themselves to be elicited through language probes. Secondly, the aim of the study
was to assess the usefulness of a set of elicitation tasks that could be administered more
extensively, if found effective. Finally, the study was expected to obtain a preview of some of
the developmental trends among Bangla-speaking children.
Since verb morphology has commonly been found to be intriguing and insightful for
understanding the developmental issues, the present study was restricted to examining some of
the verb morphological forms, namely, the Present Progressive, the Present Perfect, the Past
Simple, the Past Progressive, and the Past Perfect forms in Bangla.
4. BRIEF DESCRIPTION OF BANGLA VERB MORPHOLOGY
Bangla finite verb forms are usually marked for person, tense and aspect markers. Verbs are not
marked for plurality or gender. Having agglutinative properties in the paradigm (Kar 2009), the
markers attach to the verb in an incremental fashion. Finite verb forms rarely occur in bare
6
forms; in the structurally simplest form, i.e. the Present Simple, the verb takes the person marker.
Therefore, usually the inflected finite verb forms contain two to four members. While the
absence of the tense and the aspect marker results in the Present Simple form, the presence of all
three verb inflections can be reflected in the Past Progressive form. Examples of verb
conjugation in Bangla are presented below:
(1) Niloy bikale mathe khele.
Niloy afternoon.loc field.loc play.3p
Niloy plays in the field in the afternoon.
(2) Bipasha tokhon boi porchhilo.
Bipasha then book read.asp.tense.3p
Bipasha was reading a book then.
5. METHODOLOGY
5.1 Participants
The pilot study was aimed at executing the tests on a young Bangla-speaking population. With
this goal, 20 typically-developing children between age two (± three months) to age four (± three
months) were recruited to participate in the study (Table 1). Parents of the children were
informally interviewed and they filled out a questionnaire with information about the child and
his/her language behaviour. This was designed so that children with any medical condition or
(/khel-/ + Ø + Ø + /-e/)
(root + Ø + Ø + person)
(/por-/+ /-chhi-/+ /-l-/+ /-o/)
(root+ progressive+ tense+ person)
7
speech, language and hearing difficulty (reported by parents) could be excluded from the pilot
study.
In order to understand the target competence of a native Bangla speaker, the revised method
(described later) was also executed with seven Bangla-speaking adults living in Bangladesh.
Their ages ranged from 21 years to 65 years.
Mean (N= 20)
Range (Max- Min)
Age (months)
37.60
28 (53- 25)
Table 1: Age Profile of Children
5.2 Tests and Procedures
In acquisition studies on a variety of languages, Leonard and his colleagues used language
probes extensively to elicit the target inflections (Bangla: Chakraborty & Leonard 2012;
Chinese: Fletcher, Leonard, Stokes, & Wong 2005; Finnish: Kunnari et al. 2011; Hungarian:
Leonard, Lukács, & Kas 2011). They created contexts that warranted obligatory production of
the target inflections with the help of pictures, toys or puppets, and through enactment. A fixed
set of stems were consistently used in all the contexts to examine the development of the target
inflections. Such production probes designed to assess children’s language development were
reported to be highly effective when a predetermined set of inflections were to be tested.
Following Leonard and his colleagues’ probes, elicitation tasks were designed that were
administered with a cross-sectional sample of children. The tests were specifically aimed at
8
eliciting verb inflections, i.e. tense, aspect and person markers. Situations were demonstrated
using age-appropriate toys, in response to which the children were expected to produce
sentences. These target situations demanded that the children produce sentences involving the
use of tense, aspect and person inflections with verbs. The test to elicit the five tense-aspect
forms contained 30 items (5X6) and the set for the three person markers contained 15 items
(3X5).
In order to elicit the target forms, different contexts were created with the help of culturally-
appropriate stuffed toys. The present progressive forms were elicited by showing a toy
performing some actions and by asking children what the toy was doing. Children were expected
to say, ‘Teddy is dancing/walking.’ In order to elicit the present perfect form, the toy was shown
to do an action which later stopped. After that children were asked, ‘What has Teddy done?
Children were expected to say, ‘Teddy has danced/walked.’ For the past progressive forms, the
toy was shown to do an action which got interrupted. This was followed by a question from the
examiner ‘What was Teddy doing?’. Correct answers required children to say, ‘Teddy was
dancing/walking.’ In the simple past situation, the toy was shown to be doing an action. After
stopping that action, children were told, ‘Now he is not walking anymore/ Now the dance is
over. What did he do?’ Children needed to say, ‘He danced/walked.’ There was a time lapse at
this stage when free-play happened between the child and the examiner. This was required to set
up the past perfect situation. After the brief play session, the examiner and the child returned to
the toy and the child was told, ‘Oh Teddy is so tired! What had he done?’ Children were
expected to say, ‘He had danced/walked’. These situations were repeated for all the selected
verbs.
9
The person marker task engaged the parent as well. Children were asked, ‘Now let’s play a
game. Can you do the things Teddy has done? I can do some of them. Look, I am dancing. Can
you do anything else?’ After children demonstrated an action, they were asked, ‘What are you
doing?’ The children were expected to say, I am flying/walking’. Then actions were
demonstrated by the examiner and the parent, and children were expected to say, ‘You are
flying/walking’ and ‘Mom/Dad is flying/walking’. This continued till utterances were elicited for
all selected verbs in all three forms.
Three criteria guided the selection of the verbs. They had to be demonstrable in the elicitation or
repetition. The verbs were all early-emerging according to the Cross Linguistic lexical Norms
(http://www.cdi-clex.org) and they all translated to simple base (one-word) verbs in Bangla.
5.3 General information about the set-up
The tests were conducted in the children’s homes with the participation of their parents. Due to
the breadth of the tests, the tests were conducted in two visits, and the elicitation test was
presented in blocks to avoid fatigue. All sessions were video recorded using the video feature of
a digital camera, Canon Powershot S5IS.
5.4 Scoring
For the sentence elicitation tests, each utterance with correct inflections was credited. The use of
wrong inflections in the target words (verbs) was scored as incorrect. However, errors in other
words of sentences were ignored for scoring purposes.
10
6. RESPONSES TO THE TASKS
The primary reasons for conducting the extensive pilot study was to receive feedback on the use
of elicitation task as a research instrument, to examine if the verb forms identified lent
themselves to being tested in such a format, and also to identify a preliminary overview of
children’s development of verb inflections with regard to a set of verb forms. The pilot study, in
these respects, was informative since useful feedback arose from children’s responses.
The elicitation tasks varied in its effectiveness. In some contexts it was possible to elicit the
target sentences through the task, while some probes did not elicit the desired response. For
example, after responding to six actions in progression (e.g. What is Dolly doing?-> Dolly is
dancing.) when the children saw that the doll had completed the action and were asked ‘What
has dolly done?’, they replied ‘Dolly is walking.’ This could be an artifact of the series of similar
tasks conducted. Therefore, it appeared that the elicitation task may have captured a
compromised picture of the children’s language performance. Secondly, it was difficult to
communicate the situations for the past progressive and the past perfect to children below age
three. Finally, the task set with all the structural contrasts was lengthy and compliance became an
issue with the children.
7. MODIFICATIONS AND THE NEW METHOD
Considering the challenges of administering structured probes with very young children
suggested that having a more conversational framework may be more feasible for collecting
children’s responses. However, employing unstructured tasks with children also had pitfalls.
Having a conversation that was not controlled for language items ran the risk of obtaining a large
11
amount of ‘irrelevant’ language. Therefore, a combination of structured probes and spontaneous
language samples seemed an effective method to serve the purpose of the research. Also, the
length issue observed made it inappropriate for use with very young children. Unless, in larger
studies, there is a provision for meeting the children several times, the scope of the study need to
be adjusted. The particular study for which the pilot was conducted had time restrictions due to
which the set of tasks to elicit person markers was removed.
The revised method employed conversations with children in five situations. These situations
were designed to control the language to be produced by the children. These contexts were
shaped to elicit responses containing five target verb forms: present progressive, present perfect,
past simple, past progressive and past perfect.
Figures 1 and 2. Sample pictures used to elicit the Present Progressive form
Present progressive. Children were shown a picture book (Bernthal & Full 2006) with pictures
of Bop (a cartoon character) and his friends doing some actions in school such as riding bicycles,
12
making sand castles, and playing with tea sets. The children were asked what those characters
were doing in the pictures (Figures 1 and 2).
Present perfect. Children were asked what they had done since morning. In Bangla the present
perfect form seems a more natural choice than the past simple to use in this situation. The
expected responses were such as ‘I have brushed my teeth’, ‘I have played with mummy’.
Past simple. Children were asked to tell a story they knew. Narrative is one mode which
typically employs past simple constructions in Bangla.
Past progressive. To elicit responses bearing the past progressive forms, children were told about
the examiner’s visit to a zoo. They were told, at the time of the visit, different animals were
engaged in different activities. For some animals, sentences were left incomplete and children
were asked ‘Can you say what the tiger was doing’. Children were expected to respond in
sentences like ‘The tiger was sleeping’ and ‘The cow was eating grass’.
Past perfect. Children were asked about their visit to a restaurant or an amusement park, or how
they celebrated a festival. The expected responses were such as ‘We had ridden the toy train
there’ and ‘We had eaten lots of sweets’.
A group of seven adults were also invited to do the same tasks in order to identify the extent to
which children’s language errors were due not to their ages, but to language input and usage.
Note that these adults were unrelated to the children who participated in the study.
13
8. RESULTS
Table 2 presents a general reflection of children’s success with the target forms. Raw scores were
calculated in percentages. The results indicated that the present perfect form was acquired with
highest mastery by children. Unlike in English (Brown 1973), the Bangla present progressive
form was not one of children’s early forms. In addition, the children had low accuracy rates in
the past progressive and the past perfect forms, as anticipated. Using these forms meaningfully
has certain cognitive prerequisites, the result of which likely to have reflected in the low scores.
Mean
Range (Maximum-Minimum)
66.87
92.86 (100-7.14)
92.79
27.78 (100-72.22)
54.66
100 (100-0)
23.99
83.33 (83.33-0)
27.97
100 (100-0)
Table 2: Descriptive Statistics of Children’s Accuracy (%) on Tasks (N= 20)
8.1 Present progressive
Although assumedly a frequently occurring form, the present progressive form studies here
obtained a moderate score of accuracy. An error analysis indentified a dominant pattern of
substitution by the present simple form. Interestingly, Bangla permits production of the present
simple form in place of the present progressive in day-to-day conversation, at least in the
linguistic context (████████████) of the present study. Therefore, this pattern of substitution
was expected in children’s language as well. However, this pattern was not restricted to any
14
particular age group. Parents were also found to often structure sentences in the present simple
form to very young children (observed with parents of the twin children of 2;1 years). Therefore,
it is possible that overt marking of progression is not a priority for the Bangla-speaking young
children. However, such substitutions were not observed for the adults except for one; the adults
did not replace the progressive with the simple form, when stimuli were given in progressive
forms. Figure 3 presents children’s performance on the Present progressive task.
Figure 3. Percent accuracy in the Present Progressive form
8.2 Present perfect
The mean accuracy of the present perfect form was over 92% for the children tested.
Nevertheless, there were some substitutions. Some older children (3;9 and 4;5) replaced the
target form with the past progressive or the past perfect form. Analysis of the inaccuracies made
by younger children (2;2 and 2;3) showed a different pattern. These children produced the aspect
15
markers correctly, but combined them with inappropriate person markers. Figure 4 demonstrates
children’s accuracy in the present perfect form.
Figure 4. Percent accuracy in the Present Perfect form
8.3 Past simple
Children’s performance on the past simple showed a consistent pattern. It was widely substituted
with the present perfect counterpart by children as well as adults (when tested). Figure 5 contains
children’s scores on the past simple task.
16
Figure 5. Percent accuracy in the Past Simple form
8.4 Past progressive
The task to elicit this form required listening to and comprehending a story and completing the
story with utterances containing the present progressive form. Due to task demands it was
difficult to communicate the situation to most children below 2;6. But the older children showed
increasingly better responses. A common substitute was the present progressive form. Children’s
accuracy scores are presented in Figure 6.
17
Figure 6. Percent accuracy in the Past Progressive form
8.5 Past perfect
The overall accuracy in this task was very low (Mean= 27.97). Only two children (3;2 and 4;2)
scored above 70% and eight of the twenty children had no success at all. This form was
commonly replaced by the present perfect form. Figure 7 shows children’s performance on this
task.
18
Figure 7. Percent accuracy in the Past Perfect form
9. DISCUSSION ON VERB INFLECTIONS
Based on the data, some preliminary judgements can be made about the markers. Unlike the
findings observed in English (Brown 1973), overt progressive marking on verbs is probably not
the earliest linguistic skill of a Bangla-speaking child. On the other hand, present perfect forms
seemed to be a child’s strength from a very early age. One reason why the present perfect form is
relatively accurate from an early age (as opposed to the present progressive form) may be the use
of the form in the past simple and the past perfect contexts, whereas positive evidence for the
present progressive form is reduced by use of the present simple in present progressive contexts.
Also, it is noteworthy that in most cases, the substitute form is structurally simpler than the target
form. Therefore, the findings forward strong evidence for the contributions of structural
complexity on the acquisition of the grammatical forms. Interpretations based on the structural
19
properties of the target form has often been observed in crosslinguistic research of child language
(e.g. Tamil: Lakshmanan 2006; Italian: Leonard, Caselli, & Devescovi 2002).
Children’s performance on the past progressive form appeared to be informative. However, not
all children in the target age group could participate in the task considering the cognitive demand
it posed. It is very likely that the task was at a higher developmental level than some of the
children were and their apparent failure at the task was largely due to their inability to
comprehend it. Leonard, Caselli and Devescovi (2002) reported a similar issue with their set of
probes for Italian verb inflections.
Accuracy scores on past simple and past perfect forms are intriguing> However, making any
judgement on the development of these markers from this set of tasks and results may be
premature. It emerged from children’s responses that the past simple form was difficult to strictly
elicit through probes because the usage demanded the present perfect form as the most natural
choice in a general past context. On the other hand, it seemed that the past perfect form was
warranted only when there were references to more than one action in a past context.
10. REFINEMENTS AND RECOMMENDATIONS
Based on the experience with the testing instruments, and the data obtained, some refinements
are recommended for the original research project which was being piloted through the present
study. It is believed that the recommendations would be useful to consider in any future study
conducted on Bangla morphological development in children. The recommendations are
described below.
20
The situational tasks in the pilot study were not controlled for the number of opportunities for
each target marker. As a result, some markers had very limited opportunities, especially with
very young children. For example, if there were only two opportunities for a marker and the
child produced one correctly, it might not be reasonable to conclude that the child had 50%
accuracy for that marker. Therefore, it was concluded that at least a certain number of
opportunities need to be ensured for each marker.
The situations used in the pilot study were designed primarily to generate the target tense and
aspect marker combinations. Hence the responses yielded were sometimes in the first person and
sometimes in the third person. The situations were refined for the actual study so that all the
responses generated a single person marker.
A major finding of the pilot study was that all the target verb forms did not lend themselves to
being tested with two-year-olds. One of the primary purposes of this pilot study was to ensure
the feasibility of the tests in the target context. The study was successful in terms of making such
revelations about some of the target markers. Unlike English, use of the Bangla past simple
forms is not clearly defined. Situations that necessitate production of the past simple forms in
Bangla are rare. The pilot study employed a story-telling situation, because, in storybooks,
narratives typically contain this form. But the pilot study revealed that 1) story-telling as a task
did not succeed with very young children; 2) very young children often told stories with the
exact expressions used in the book which did not reveal their linguistic skills; and 3) while
telling stories children commonly replaced the past simple form with the present perfect.
Substitution of the past simple form by the present perfect was also common among the group of
adults in this study. These findings indicated that a story-telling situation was inappropriate for
the current study.
21
While designing the elicitation tasks for the past perfect forms, time was considered to be the
only defining factor. Therefore, situations that required children to talk about something that
happened in the remote past were considered a fit for the study. But the pilot study revealed that
the use of past perfect in Bangla also depended on whether or not the impact of the referred
action still held; for example, ‘Why did you dirty the floor? I had mopped it awhile ago.’ Also,
unlike the other situations discussed before, the past perfect contexts in Bangla warrant the
understanding of three different points in time: speaking time, event time and reference time,
which is highly demanding for the two-year-olds. Aksu-Koc and Slobin (1985) made a similar
suggestion about the emergence of different past inflections. They found that children’s
acquisition of different past markers was governed by the markers’ relative cognitive
complexity. They reported that the past inflection that stood for an immediately-observable
change was acquired before the form referring to a general past event, and the past marker for
witnessed past events was acquired before the form for non-witnessed past events. This suggests
that for Bangla the past perfect form is likely to be a later-emerging marker.
Therefore, the fact that elicitation of the past simple and the past perfect forms posed challenges
suggested that these elicitation tasks may not be appropriate for examining these forms among
children between age two and four. Although there are methodological disparities involved, a
solution could to examine the development of these forms from spontaneous language samples.
Results of the present perfect task revealed that the children had very high accuracy scores on
this form. This was consistent with the finding that this form was often used by children as a
substitute of other verb forms marking perfectivity, i.e. the past simple and the past perfect
forms. Therefore, it was anticipated that, for the purpose of the main study, a task eliciting the
22
present perfect form might not be adequately informative. However, for developing a general
language profile of children this form needs be included.
The structured conversation used in the pilot study for eliciting the past progressive forms
revealed that children below age three were often not able to respond to the situation. This could
partly be due to children at that age not being cognitively mature enough to process such a
situation. If this is the case, then we can expect that the past progressive form will be available to
children only beyond age three. Therefore, the task was modified by employing some pairs of
pictures depicting different actions with which the target forms would be elicited. Another
important reason for changing the context was the possibility that children might produce the
same verb for different situations. For example, when asked, ‘What was the monkey doing?’ and
‘What was the tiger doing?’, children might respond with ‘It was sleeping’ in both situations. A
set of questions guided by unique picture pairs would be free from such overlaps.
The pilot study did not include the Present Simple form since this form in Bangla does not
contain tense and aspect markers, and is not structurally complex. Therefore, it was not
considered intriguing enough to be examined in acquisition studies. However, the structural
simplicity of the form itself can lend a justification for the inclusion of this form. Since the
Present Simple form in Bangla does not take any overt tense and aspect marker, it might present
a good test case to be compared against the other verb forms that take more inflections.
Considering these issues, it was recommended that the Present Simple form also needs to be
examined in future studies.
Apart from identifying children’s language development with regard to different morphological
markers, it is also important to evaluate other measures of language development, for example,
23
mean length of utterance (MLU), count of word type (controlled for sample length), and use of
bound morphemes in verbs. In addition, these measures can also be useful in interpreting the
scores of the elicitation tasks. Structured conversations are not considered ideal for calculating
these measures, since those conversations are primarily shaped to achieve other goals. Therefore,
a language sample, preferably 20-minute long, will be insightful to supplement the structured
probes used in the studies.
11. FINAL WORDS
The present study conducted with 20 children between age two and four explored some research
possibilities with regard to the language areas to be tested and also examined the effectiveness of
elicitation tasks as possible research instruments. Some refinements are forwarded based on
children’s responses to the tasks. Future studies are expected to benefit from this framework and
suggestions, which can expand themselves in terms of the number of participants and the
language areas to be examined.
REFERENCES
Acarlar, F., & Johnston, J. R. 2011. Acquisition of Turkish grammatical morphology by children
with developmental disorders. International Journal of Language & Communication
Disorders, 46(6), 728-738. doi:10.1111/j.1460-6984.2011.00035.x
24
Aksu-Koҫ, A. & Slobin, D. 1985. The acquisition of Turkish. In D. I. Slobin (Ed.) The
crosslinguistic study of language acquisition (pp. 839-878). Hillsdale, NJ: Lawrence
Erlbaum.
Bernthal, M. S. & Full, D. 2006. Let’s go to school. New Delhi: Sterling Publishers.
Brown, R. 1973. A first language: The early stages. Cambridge, MA: Harvard University Press.
Chakraborty, R., & Leonard, L. B. 2012. A brief research report on acquisition of verb
Inflections in Bengali-speaking children. Journal of Advanced Linguistic Studies, 1(1-2),
41-53.
Comrie, B. 2005. Language. Microsoft Encarta Online Encyclopedia, 2005. Microsoft
Corporation.
Cross Linguistic Lexical Norms. 2012. Retrieved from http://www.cdi-clex.org/.
Eisenbeiss, S. 2010. Production methods in language acquisition research. In E., Blom & S.,
Unsworth (Eds.), Experimental methods in language acquisition Research (pp. 11-34).
Amsterdam: John Benjamins.
Fletcher, P., Leonard, L. B., Stokes, S. F., & Wong, A. M.-Y. 2005. The expression of aspect in
Cantonese-speaking children with specific language impairment. Journal of Speech,
Language, and Hearing Research, 48(3), 621-634. doi:10.1044/1092-4388(2005/043)
Ingram, D. 1989. First language acquisition: Method, description and explanation. Cambridge:
Cambridge University Press.
Kar, S. 2009. The syllable structure of Bangla in Optimality Theory and its application to the
analysis of verbal inflectional paradigms in Distributed Morphology. (Doctoral
dissertation). Universität Tübingen, Germany.
25
Klaiman, M. H. 2008. Bengali. In B. Comrie (Ed.), The world's major languages (pp. 417-436).
London: Routledge.
Kunnari, S., Savinainen-Makkonen, T., Leonard, L. B., Makinen, L., Tolonen, A. K., Luotonen,
M., & Leinonnen, E. 2011. Children with specific language impairment in Finnish: the
use of tense and agreement inflections. Journal of Child Language, 38(5), 999-1027.
doi:10.1017/S0305000910000528
Lakshmanan, U. 2006. Assessing linguistic competence: verbal inflection in child Tamil.
Language Assessment Quarterly: An International Journal, 3(2), 171-205.
doi:10.1207/s15434311laq0302_5
Leonard, L. B. 2014. Children with specific language impairment (2nd ed.). Cambridge, MA:
MIT Press.
Leonard, L. B., Caselli, M. C., & Devescovi, A. 2002. Italian children's use of verb and noun
morphology during the preschool years. First Language, 22(3), 287-304.
doi:10.1177/014272370202206604
Leonard, L. B., Lukács, Á., & Kas, B. 2011. Tense and aspect in childhood language
impairment: Contributions from Hungarian. Applied Psycholinguistics, 33(2), 305-328.
doi:10.1017/S0142716411000361
Lustigman, L. 2012. Developing structural specification: Productivity in early Hebrew verb
usage. First Language, 33(1), 47-67. doi:10.1177/0142723711426828
Phillips, C. 2010. Syntax at age two: Cross-linguistic differences. Language Acquisition, 17(1-
2), 70-120. doi:10.1080/10489221003621167
Pinker, S. 1984. Language learnability and language development. Cambridge, MA: Harvard
University Press.
26
Raghavendra, P., & Leonard, L. B. 1989. The acquisition of agglutinating languages: converging
evidence from Tamil. Journal of Child Language, 16(2), 313-322.
doi:10.1017/S0305000900010436
Xanthos, A., Laaha, S., Gillis, S., Stephany, U., Aksu-Koc, A., Christofidou, A., … & Dressler,
W. U. 2011. On the role of morphological richness in the early development of noun and
verb inflection. First Language, 31(4), 461-479. doi:10.1177/0142723711409976
... Chakraborty and Leonard's conclusion were based on data from only half the original sample, as it turned out that 18 of the children could not reach criterion on the elicitation tasks they attempted. Sultana (2016), again using elicitation, found that verb forms requiring the tense marker without aspect marker, i.e. the past simple, were rarely used by children, and confirmed that forms taking only the aspect marker in absence of a tense marker, e.g. present perfect, were frequent (Sultana, 2016). ...
... Sultana (2016), again using elicitation, found that verb forms requiring the tense marker without aspect marker, i.e. the past simple, were rarely used by children, and confirmed that forms taking only the aspect marker in absence of a tense marker, e.g. present perfect, were frequent (Sultana, 2016). ...
... This group was presented with the same elicitation procedures as thetypically-developing(TD) children reported in Sultana et al. (2016). Mean percent correct scores in response to the elicitation probes were as follows: Present Simple, 53.29%; Present Progressive, 27.78%; Past Progressive, 8.89%. ...
Article
The study examines the nature of productivity within the verbal paradigm among typically-developing Bangla-speaking pre-school children. Spontaneous language samples of 30 typically-developing children aged 2 to 4 were collected. Productivity of the verb stems and inflections were calculated based on a specific criterion. The results revealed that the children followed an item-based approach, in terms of acquiring verb stems as well as inflections. The findings, we propose, can guide the future studies, especially those working on child language development in Bangla.
Article
Full-text available
Previous studies of children with language impairment (LI) reveal an insensitivity to aspect that may constitute part of the children's deficit. In this study, we examine aspect as well as tense in Hungarian-speaking children with LI. Twenty-one children with LI, 21 TD children matched for age, and 21 TD children matched for receptive vocabulary scores were tested on their comprehension and production of both imperfective and perfective verb forms in past tense contexts. Although the groups did not differ in their comprehension performance, the children with LI were less accurate than both comparison groups in producing both imperfective and perfective forms. Based on these results, it appears that children with LI have difficulties selecting the appropriate aspectual marking in past tense contexts.
Article
Full-text available
This study proposes a new methodology for determining the relationship between child-directed speech and child speech in early acquisition. It illustrates the use of this methodology in investigating the relationship between the morphological richness of child-directed speech and the speed of morphological development in child speech. Both variables are defined in terms of mean size of paradigm (MSP) and estimated in a set of longitudinal spontaneous speech corpora of nine children and their caretakers. The children are aged 1;3–3;0, acquiring nine different languages that vary in terms of morphological richness. The main result is that the degree of morphological richness in child-directed speech is positively related to the speed of development of noun and verb paradigms in child speech.
Chapter
“Specific language impairment” (SLI) is a term applied to children who show significant deficits in language learning ability but age-appropriate scores on non-verbal tests of intelligence, normal hearing, and no clear evidence of neurological impairment. Children who meet this definition are not identical in their characteristics, though some linguistic profiles are rather common. Boys outnumber girls, with a ratio of approximately 2.8 to 1 (Robinson, 1987). At age five years, the prevalence of SLI might be as high as 7% (Tomblin, 1996). This percentage is probably lower at older ages, due to the fact that some proportion of children with milder language difficulties achieve normal levels of ability within a few years, often with the help of intervention.
Article
The study investigates acquisition of verb inflections by four monolingual Hebrew-acquiring children from middle-class backgrounds, audio-recorded in longitudinal, weekly samples at a mean age-range of between 18 and 26 months. Productive use of inflectional morphology is shown to manifest increasing structural specification, as a function of children’s developing knowledge of morphology–syntax interfaces illustrated by Subject–Verb agreement. The verbs used by all four children evolve from non-specified stem-like forms via sporadic structural alternations to initial productivity in the use of largely under-specified inflected forms. This process of increasing structural specification is interpreted as reflecting paradigmatic/syntagmatic interfacings in developing grammatical productivity, such that eventually a full range of inflectional contrasts is manifested in all and only appropriate syntactic environments.
Article
Five groups of Italian-speaking children ages 2 to 7 years participated in tasks designed to assess their use of a range of grammatical morphemes. Present tense verb inflections and noun plural inflections reached ceiling levels by 4 years of age, whereas present tense copula forms and definite singular articles showed high accuracy levels by 5 years of age. Errors on present tense inflections were rarely infinitives. Instead, most errors could be characterized as ‘near misses’ – productions of forms that differed from the target by a single feature of person or number. Many of these one-feature errors were directional; first person forms were more likely to be replaced by third person forms than the reverse, and plural forms were more likely to be substituted by singular forms than the opposite pattern. Errors of this type are not handled by models of grammatical development that deal with broader constructs such as the availability or optionality of tense or finiteness.