Saturday 26 March 2011

The language of science

Science doesn’t have a language, of course. People have languages. Since people socialise in specific groups, from which they get their linguistic uses, mindsets and other behavioural habits, doing science in different languages may sometimes look like doing different sciences.

I should perhaps explain what I mean by science. Common meanings of this word have it as synonymous with natural science or hard science (interesting qualifiers, these two, especially when we think of what their antonyms might evoke). That is, the word refers to a particular object of study. To me, science is a how, not a what: it is the art of looking for patterns, and for ways of explaining them – have a look at Chapter 1 of The Language of Language, for more on this. 

Science is not the domain of the unintelligible either. Some of us do choose to sound obscure in order to sound learned, which is quite useful to send readerships and/or audiences to instant sleep, so no one can tell the difference. To me, using language clearly is the best evidence that you know what you’re talking about.

Nor does science need to be conveyed in printed form. Our ancestors were scientific in the ways they went about their business, long before they decided to invent printed language. So are our contemporaries who have no idea that literacy matters, and so are small children. Cooking dinner is doing science, because it involves a theory and empirical verification, and so is taking toys apart to see how they work.

Since science draws on research questions, findings, analyses and argumentation expressed in particular languages, two things follow. First, science reflects the resources available to the language in which we talk about it. I’ve offered a few thoughts on our understanding of science through language, on the one hand, and through languages, on the other, in a review of Roy Harris’s book The Semantics of Science that deals with precisely this topic.

Second, whoever commands a language that happens to be used to talk about science calls the shots of science-talk. Any language can be used for this purpose, of course, because users of any language can do what users of successive “languages of science”, Arabic, Greek, Latin, French, English, have done to make their languages fit their needs: borrow, adapt and invent whatever linguistic resources come in handy for science-talk. The reason why so few languages, out of several thousand, have historically acquired recognition as “languages of science” is the same that has businessmen looking like carbon copies of one another (or photocopies of one another, for those born a little later than me) wherever we turn to, out of richly diverse, and certainly more comfortable, dress codes out there: human beings love to have codes to abide by.

Abiding by the current code of using English to talk about science is sometimes misinterpreted to mean that the only science that matters speaks English. Believing that this is so, and that English-bound ways of doing science are the only respectable ways of doing it, can have regrettable effects, as Frans de Waal vividly reports in Seeing Through Cultural Bias in Science. The articles collected in the 2008 AILA Review retell similar stories of Linguistic inequality in scientific communication today.

The issue is not just that, here too, we’re facing an uneven playing field akin to the one that Viv Edwards described in a previous post. It’s principally that the unevenness cuts both ways: those of us who cannot access the current language of science lose as much as those of us who cannot access science but through it.

Global languages, however, like the ones we may use to deal with science – or business, or finance –, are not all-purpose languages, for most of their users: they serve specific global(ising) purposes. For most of us, a global language is an adopted language. And when we human beings adopt something, we adapt it: we may love following codes, but we absolutely adore breaking them. My next post has a few thoughts about why I find it difficult to envisage a one-language-fits-all global future.

© MCF 2011

Next post: Going global, full monolingualism ahead? Saturday 2nd April 2011.

Saturday 19 March 2011

Learning to read in a multilingual society: challenges for Africa
=Guest post=

by Viv Edwards

As any parent of a multilingual – or potentially multilingual – child knows, speaking more than one language is one thing and being literate in more than one language is quite another. But have you ever thought what it might be like if very little written material was available in the languages you want to nurture? This is precisely the case for millions of children growing up in Africa. The legacy of the colonial era means that, in many countries, children are educated through the medium of a European language – most often English or French – even though they may seldom hear or need to use that language outside school.

However, change is afoot. Many countries are recognizing the importance of a sound foundation in mother tongues for future educational success. Often, the use of mother tongues is limited to the first two or three years of school, but in South Africa the aim is to extend bilingual education to at least the first six years – drawing on the evidence of international research that the best academic results are achieved when children continue to use their mother tongue alongside the language of wider communication for a sustained period.

What, though, if reading materials in the mother tongue are very few in number – and of dubious quality and appeal? The publishing industry in South Africa is world class but is dominated by books in English and, to a lesser extent, Afrikaans. However, the 1994 Constitution which followed the dismantling of Apartheid recognizes nine other official languages – isiNdebele, isiXhosa, isiZulu, Sepedi, Sesotho, Setswana, SiSwati, Tshivenda and Xitsonga. The challenge then, is for publishers to provide the same range and quality of books in the other African languages as in English and Afrikaans.

Things are seldom straightforward. For a wide variety of reasons – the historical importance of oral culture, lack of disposable income and low levels of literacy – very many Africans don’t engage with books and those who do often prefer reading in English. This means that publishers are dependent almost exclusively on the schools market. Unfortunately, for the time being at least, official support for bilingual education remains more at the level of rhetoric than reality, which means that publishers often can’t be sure of a large enough market to break even, let alone make a profit.

Predictably, more books are published in the “big” languages like isiZulu and isiXhosa than in languages spoken by smaller numbers of speakers. Small, of course, is relative. Tshivenda, the “smallest” of the official languages, is spoken by over a million people. This compares with the 3.6 million speakers of English who have access to by far the largest number of books in circulation.

In spite of the uneven playground, the number of good quality children’s books in African languages is growing, though often driven more by NGOs than commercial publishers. One very exciting initiative is the Stories Across Africa project. Coordinated in South Africa but involving organizations and individuals committed to good quality books across the continent, they have produced a set of very appealing 16 Little Hands books, already translated into 26 African languages. Their appeal is huge: children like to see – and hear – themselves in the books they read.  

Reading in the book club in a Cape Town township.
Photo: Viv Edwards

Anyone wishing to be persuaded of their appeal should spend an hour at a community-run reading club for children!

Interested in finding out more?

Viv Edwards is Director of the National Centre for Language and Literacy, University of Reading, and the author of Learning to be Literate. Multilingual Perspectives.

© Viv Edwards 2011

Next post: The language of science. Saturday 26th March 2011.

Wednesday 16 March 2011

First, main, best

If the title of this post reminds of the Olympic motto, it should: it’s pure competition out there. Which language, in the singular, tops the lot? Despite lingering suspicion that all languages of a multilingual should preferably stand on equivalent footing, ranking languages ex-aequo is not a preferred option.

Let’s see. First, first. What comes first comes before anything else, in time or space, so your first language is the one you have after you have no language. People can have more than one first language, of course, if they’re brought up multilingually from birth. But speaking of several first languages generates confusion when “first language” is represented as “L1”, which might have been a usable way of naming the L in question when everyone was assumed to start life in monolingual worlds. Since this is not so, and since confusing labels about language matters have a habit of persevering, you can then have several el-ones, which looks and sounds funny because if something is found to be suitably represented by the single unit “1”, then it cannot be suitably represented by several unit “ones”.

The confusion increases when multi-el-oners go to school and learn new languages. These languages, whether one or more, become their el-twos, because they are all second languages. The confusion peaks when the word first in “first language” is taken to mean not ‘first’ but ‘main’, in the sense that, say, a First Lady is a Main Lady, or ‘best’, in the Olympic sense that whoever comes first is best.

Main, then. This is variously taken to be the language you spontaneously use when initiating an exchange, the language you dream or swear in, or the language you do your maths in, for example. The problems are that when you’re initiating talk, you’re preparing to talk to someone, who also has at least one language, so you choose language not on instinct but according to what you know about your interlocutor’s language(s); that you dream and swear in whatever languages became relevant for what prompted your dreams and swearwords; and that maths is something that you learn, usually in school, from someone who must speak to you in some language, which then becomes your maths language.

Next, best. Really tricky, this one. Best for what? Or does “best” mean something like ‘best quality’, in which case do we mean quality grammar, or vocabulary, or fluency, which, again, must be best for specific purposes? For example, English is my best language for work, and also the main one and the first one that comes to me when I think about work matters, but I wouldn’t honestly know how to answer a simple yes/no question about whether English is my best language in absolute terms.

I believe that we need to shed the absolute nature of terms like first, main and best, when talking about our languages, and start thinking about them in terms of their uses rather than their properties. We’ve known this ever since the Polish anthropologist Bronisław Malinowski wrote, all the way back in 1923, that:

“A statement, spoken in real life, is never detached from the situation in which it has been uttered. For each verbal statement by a human being has the aim and function of expressing some thought or feeling actual at that moment and in that situation, and necessary for some reason or other to be made known to another person or persons – in order either to serve purposes of common action, or to establish ties of purely social communion, or else to deliver the speaker of violent feelings or passions.”

This is what we have our languages for, and each one will naturally shift from best to subsidiary to first and then back again, depending on which uses we go on making of them. But what happens, then, when we want to use our languages, and develop our skills in them, and find little or no support for this in our community? The next post, a guest post, gives us some answers to this.

© MCF 2011

Next post: =Guest post= Learning to read in a multilingual society: challenges for Africa, by Viv Edwards. Saturday 19th March 2011.

Wednesday 9 March 2011

Language percentages

Amounts of language can be measured. We can measure number of words, for example – assuming that we all know what a word is, of course. (Have you tried asking yourself this question, by the way?) We can measure length of utterances, counted in words or in assumedly less dodgy units that linguists call morphemes (do a Look Inside! search for morpheme here). We can measure number of utterances, phonemes, metaphors, syntactic constructions, and/or sequences of all of these.

We then compare our observations about these amounts to amounts that are available from standardised averages or norm-referenced databases, to reach conclusions about the relative state of health of an individual’s linguistic repertoire. Some people even like to do this to support claims about rich(er) and poor(er) languages, or sophisticated and primitive ones, and label their users accordingly, on the understanding that languages own things and that their users are owned by them (in more than one sense of owned).

Measuring amounts of language is different from measuring amounts of languages, that I touched upon previously, but both issues end up raising issues to do with multilingualism. One recurring question I get from worried parents in multilingual families, second only to the question of which single language they should model in order to secure proper child multilingualism-to-be, concerns how much language they should use. How many and how much thus top the list of perplexities about multilingual interaction.

The suggestions that the queriers themselves submit to my appreciation speak of the bewilderment that we’ve allowed to run loose about being multilingual. Should they hire tutors, nannies, or both, strictly sworn-in to scheduled and exclusive use, to the child, of one of the languages of the parents, so that the parents can in turn pledge themselves to use another of their languages (one each), in order to nurture healthy multilingualism at home? Should the parents themselves switch languages (those who dare defy the OPOLicy, that is) and, if so, who should speak what at what times of day, or days of the week, to avoid confusion in developing brains? Shouldn’t they stick (those who daren’t) really, really to one language each, and so expose their children to really, really health-inducing parental dialogues where one parent always says things in language X and the other always responds in language Y?

OK, some people do tend to get a bit worked up about the latest health-related fads they read about in the local newspaper. We should all currently rinse our nasal cavities with specially crafted implements and saline solutions offered in a range of prices that serve plebeian and aristocratic noses, for example, or think twice about antioxidant contents before brewing our daily cup of tea. Or hire monolingual tutors and nannies to talk to multilingual children in the languages that multilingual parents should not talk to them in. But these questions also speak of the anguish, the guilt, the daily burden of making language decisions before daring to open one’s mouth, that we can but glimpse through them. I’ve had parents begging me for help about what to do when they wake up in the morning, every morning, with no idea about which language to use to their own children.

All this may not be surprising, in fact, in view of comments, from lay and specialist sources alike, to the effect that, for example, bilingual children get only half the input in each language. By extension, we may reason that trilingual children get ⅓, quadrilingual children ¼, and so on. Who will want to be blamed for providing incomplete linguistic input to their children? Aren’t incomplete languages the very essence of semilingualism, and isn’t multilingualism but a fancy name for semilingualism? Fractional maths is also popular in other multilingual settings, those involving language learning in later life, where amounts of exposure to language(s) and whole vs. partial language input also come into question. I will have quite a lot to say about this in coming posts, but what both settings have in common is first, that the percentile gold, 100, is represented by one language, all of the time; and second, consequently, that no matter how much language is used in multilingual settings, “how much” never means ‘enough’.

It is therefore interesting to think a while about why questions concerning best languages, or main languages, also crop up so often in discussions of multilingualism. Upon which of the language fractions of a multilingual should this honour be bestowed, how and why? I’ll try to work this out next time.

© MCF 2011

Next post: First, main, best. Wednesday 16th March 2011.

Wednesday 2 March 2011

Bilingual phonological development is like driving in traffic
=Guest post=

by Brian A. Goldstein

Phonological development of bilingual children is a little like driving in traffic to and from work – sometimes you accelerate, sometimes you decelerate, and sometimes you just idle. I prefer acceleration, but I know I can’t do so all the time. I know there will be times when I have to slow and times when I just stop and wait. Bilingual phonological development is much of the same. Whether you are able to notice acceleration, deceleration, or idling in bilingual phonological development depends on what you’re looking at and when you’re looking at it.

These notions of acceleration, deceleration, and idling are not mine. They’re hypotheses from Paradis and Genesee (1996). A couple of preparatory notes. First, their original terms were acceleration, delay, and transfer. In a recent paper (Fabiano-Smith and Goldstein, 2010), Leah Fabiano-Smith and I changed delay to deceleration. We did so because in our field (speech-language pathology), the term delay is a clinical one indicating a disorder, usually requiring intervention to ameliorate the problem. So, we did not want to conflate the two terms. We also thought it was a nice parallel to acceleration.

Second, their third hypothesis was transfer – the bidirectional influence of one language on another. We addressed transfer in our 2010 paper. I’m not going to address that issue but perhaps will in a future post (if Madalena wishes me to invade her space again and if you all find some interest in bilingual phonology – I hope you will let me know). Instead I’ll focus on acceleration, idling (to continue the driving in traffic metaphor), and deceleration. So, here goes.

Because I like to accelerate, I’ll start with that one. The mythology that bilingual language development (phonological, in this case) is slower than monolingual development is simply that – mythology. There is evidence that phonological skills in bilinguals are accelerated (i.e. more advanced) compared to those of monolinguals (e.g. Lleó, Kuchenbrandt, Kehoe, and Trujillo, 2003). In that study, coda (i.e. syllable final) consonants were acquired in German-Spanish bilinguals before monolingual Spanish-speaking children. That said, there is also evidence that phonological skills in bilinguals are decelerated (i.e. less advanced) compared to those of monolinguals (e.g. Gildersleeve-Neumann, Kester, Davis, and Peña, 2008), in that consonant accuracy was lower in the bilingual children.

So, there is evidence of acceleration and deceleration in the phonological development of bilingual children. Bilingual children, however, spend a great deal of time idling – just like me on my afternoon drive from work to my house 17 miles away. That is, the phonological skills of bilinguals are mostly commensurate with those of monolingual children and thus are within the “normal” range of skills exhibited by monolingual children.

Evidence for this position (which we pitched as a variation of the acceleration hypothesis in our 2010 paper) comes from a variety of studies examining phonological skills in (mostly) 3-, 4-, and 5-year-old bilingual children. Such studies have looked at Spanish-English bilingual children (e.g. Fabiano-Smith & Goldstein, 2010; Goldstein, Fabiano, and Washington, 2005); Russian-English bilinguals (e.g. Gildersleeve-Neumann and Wright, 2010); and Italian-English speaking bilingual children (Holm and Dodd, 1999). It should be noted that in the case of the Italian-English speaking bilingual children in Holm and Dodd, the children had a speech problem indicating that even those bilingual children with speech problems can show similar skills to monolinguals with speech problems.

So what these results from these studies tell us is that we should expect to find evidence for acceleration, deceleration, and idling in the same kids. Such results shouldn’t be that surprising. After all, don’t we all experience each of those conditions every day in our commutes to and from work?
Brian A. Goldstein is Dean of the School of Nursing and Health Sciences and Professor of Speech-Language-Hearing Sciences at La Salle University, Philadelphia, PA, USA.

© Brian A. Goldstein 2011

Next post: Language percentages. Wednesday 9th March 2011.


Related Posts Plugin for WordPress, Blogger...