/ 2 June 2013

Maths still best taught in English

Graphic: John McCann
Graphic: John McCann

In one of the courses I lecture to postgraduate in-service teachers on language and communication in mathematics classrooms, a student posed this question to me: "Would you send your child to an isiZulu-medium school?" My students were surprised when I responded with an emphatic "no". 

Let me explain. In a country where most classes are multilingual, it is pertinent to ask whether multilingualism does or can contribute ­positively to the learning of mathematics. My approach to this issue goes beyond making an African indigenous language a compulsory subject in all schools. 

I am an advocate of mother-tongue instruction. This stems from both research and my personal experience of having done my first degree in French — a language with which I had first come into contact only some months prior to my first year and which I was still learning.

When the use of other languages in the teaching and learning of mathematics is debated, it is not uncommon for people to ask: Why don't we stick to using English? Has English failed? Why do we want to go through the trouble of teaching maths in an indigenous language that is not fully developed to cater for all the mathematics vocabularies in English? 

Yet, before the 16th century, these same questions were being asked about the use of English for teaching and learning mathematics because at the time mathematics was done only in Greek and Latin, which enjoyed privileged status in scholarship. It was only in the 16th century that the first mathematics book was written in English.

Extensive research on mother-tongue teaching and learning shows the cognitive advantages that accrue to such practice. I will cite merely two examples. In Ethiopia, the number of years in which the mother tongue is used in teaching and learning varies, according to region, between zero and eight years. Research carried out between 2000 and 2008 revealed that the regions that perform best generally are those that spend more years using their mother tongue as the medium of instruction, and the poorest were those who spend the least number of years doing so.

Second, in Papua New Guinea research revealed that bilingual students with proficiency in both the mother tongue and English outperformed students who were proficient in only the mother tongue or English — and this despite the fact that the monolingual students attended better-resourced schools. Bilingual students with low competence in both languages performed very poorly. 

These findings resonate with the "threshold hypothesis" postulated by the educationist James Cummins. The threshold theory stipulates that those aspects of bilingualism that might positively influence cognitive growth are unlikely to come into effect until the child has attained a certain minimum or threshold level of competence in his or her second language. Cummins also suggested that cognitively beneficial bilingualism can be achieved only when the learner's first language is adequately developed. 

The two examples described above, and many more similar findings all over the world, are clear indications that students' proficiency in both the  mother tongue and the language that serves as the medium of instruction does make a difference to their academic attainment. 

So why would I not send my child to an isiZulu-medium school? My reasons are simple.

First, most teachers are trained in English. In the foundation phase in most public schools, where the medium of instruction is the common indigenous language of the area, there is an underlying assumption that these educators will recontextualise what they have learnt in English into a different linguistic context at the end of their qualification. 

This is not a straightforward matter. And even when foundation-phase teachers are trained using the indigenous language with which they teach at the end of their qualifications (as with some universities), problems centred on technical lexicons abound. 

Second, there has not so far been a concerted and deliberate effort to co-ordinate the development of the indigenous languages to the point where they become languages of scholarship. 

Any development of the lexicon of a language involves — necessarily — the coinage of new words as much as the borrowing of words from other languages. If there is no centralised body tasked with overseeing standardisation of the developed terms, the upshot is that different groups of people will coin different words for the same concepts.

This is currently the situation with the proliferation of multilingual mathematics dictionaries in South Africa. Although I do not believe in the mere importation of practices that have worked elsewhere, at the same time I think there are lessons to be learnt from contexts such as New Zealand, where an indigenous language — Maori — has been developed for use and is used in teaching and learning up to the final year of high school in all subjects including mathematics. For me, this development of indigenous languages as languages of scholarship needs to come first, before the introduction and use of any of them as mediums of instruction for training teachers in the foundation phase. 

Third, negative attitudes to the use of the mother tongue as a medium of instruction need to be combated. In South Africa, this is particularly sensitive given the fact that during the apartheid era mother-tongue schooling for Africans was favoured for quite different reasons than those advanced by educationists and researchers. It was used as a tool of oppression and marginalisation of blacks, which has contributed in no small measure to giving mother-tongue education — and, to a large extent, the use of African languages — a negative connotation, because of the tendency to equate it with the ravages of Bantu education. 

I contend that any attempt to use the mother tongue as a medium of instruction, or doing it as a subject in school, should not simply be for political reasons. 

It should arise from the inherent advantage that accrues from the development of proficiency in both the mother tongue and English, and proficient teaching of English must accompany the use of the mother tongue as medium of instruction. 

But the use of an indigenous language as a medium of instruction comes at a price and, until this price is paid and until the issues I have raised here are dealt with, I am afraid I would have to go for second-best — and send my child to an English-medium school right from grade R.

Dr Anthony Essien is a lecturer in mathematics education in the school of education, University of the Witwatersrand