Assessment of language learning requires linguistic expertise
Linguistics as an area of study has grown in many directions. While linguists are working on archaeological projects to trace historical details of languages, their counterparts in another area are engaged in writing algorithms for different languages so that machines can process cross-linguistic data. Many linguists are reading brain waves to discover how language is represented in human brains; groups of other linguists are spending hours trying to elicit responses from children so that language development patterns of children can be identified.
A particular direction of research on child language development involves working with children with language difficulties. This group of children may have difficulties associated with Autism Spectrum Disorder (ASD), Down Syndrome (DS), Attention Deficit Hyperactivity Disorder (ADHD), Specific Language Impairment (SLI), and developmental delay. Depending on the nature of the condition, intervention services for these children need to address their language-related problems which require a clear understanding of the language progress they are making.
There are assessment tools that can inform us about that. The assessment tools need to be separately designed for each language, keeping the features of that language in mind, and understandably so. Languages are distinct in their features, and these reflect in the developmental patterns of children’s language. For example, for many years in child language research, it was claimed that children’s particular challenge lies in showing tense marking on verbs (the view is still around). This claim was mainly based on English child data that revealed that children with language disorders commonly missed the third person -s, and the past -ed markers, and produced the bare verb forms.
However, research conducted on languages that were typologically different from English (eg, Italian, Turkish) showed that children speaking those languages did not produce bare forms of verbs. Similar findings were reported when Bangla-speaking children were tested in recent studies. For example, young Bangla-speaking normally-developing children as well as children with language difficulties showed that producing a tense marker on verbs is not a universal problem. Bangla-speaking children almost always produced verb forms with tense markings, even the very young ones.
However, they made errors in that they often produced simplified forms of the verb forms that were required; instead of “ghash khachchilo,” children often said “ghash khachche.” Bare forms of verbs such as “kha” were rarely produced where “tensed” forms were required. Based on data from cross-linguistic research, linguists and therapists now suggest that children’s language development patterns are heavily dependent on the parameters of their languages.
When children with language difficulties are tested, they are assessed in comparison with normally-developing children who speak the same language. Note that, while developmental patterns across languages are tied through some common principles, specific linguistic markers to look out for can be very different across languages.
This is the reason why language assessment tools need to be designed with direct references to the linguistic features of the target language as well as the developmental picture of the normally-developing peers. This particular issue is often overlooked while designing the assessment tools in less-researched languages. In a situation where there is a lack of resources and awareness, often tools are designed by simply translating the tests developed in English by making superficial adjustments.
In fact, it is often ignored that the assessment of language development requires linguistic expertise. Just because a therapist or a doctor is a native speaker, (s)he is not able to necessarily and accurately interpret why certain markers are missing or substituted in the language.
Assessing children and apathy
There seems to be a general lack of acknowledgement that a technical understanding is required in using language in different fields. How many people believe that teaching a language requires skills beyond simply knowing the language?
When the same insensitivity (and apathy) is applied in assessing child language development, we end up with resources that are not likely to be theoretically sound. In a context where children with a range of developmental difficulties are assessed, of course, there are “more serious” dimensions to investigate.
However, when we attempt to determine the linguistic aspects, people with relevant expertise need to be brought on board; the specific language needs to be closely examined. This is why, ready-made assessment tools cannot be simply “imported” from other languages. We need our own tests that can validly identify the level of language development of our children.
Asifa Sultana is Child Language Researcher and Associate Professor, Brac University. Email: [email protected]