Deciding what a language is has never been easy, although we go on talking about languages as if we knew what they are. This is partly because we’ve become habituated to having language names used all around us, but mostly because we’ve become inured to the fact that names serve as convenient labels to help us talk about imaginary and abstract concepts, among other things. We can also talk about unicorns and hypotenuses, for example.
Labels like Arabic or Kimbundu do not designate ‘objects’: they are a handy notation to refer to ways of using language which are each found to be intelligible to other language users. The criterion of cross-(un)intelligibility is regularly invoked as an equally handy way of attempting to define one particular language as opposed to another – more on which in a coming post. Let me just add two things here: in some cases, the name we use for a language in fact refers to its printed form, for example when we talk about “the Chinese language”; and in many other cases, language names have become indistinguishable from nationality names, which further complicates the matter of defining what “a language” is.
If we can’t pinpoint language boundaries in any precise way, it then makes little sense to say that languages can suffer injury to their integrity. We nevertheless go on talking about languages as if they owned territorial rights to what can be done with them. Difficulties in delimiting languages relate to both historical and geographical factors. Whether in monolingual or multilingual communities, grandparents and grandchildren, say, or African and Asian users of the “same” language don’t use it in the same way. There are virtually as many ways of using each of these abstract entities called “languages” as there are individuals – sociolinguists even devised a label for this observation, idiolect. This is one of the reasons why no one has ever produced a “full” grammar of any language, and this is the reason for the scare quotes in the title of this post.
Whatever we use, we necessarily change, because using things means making them ours, so they can serve our needs. “Things” like languages are no exception. For sequential multilinguals, for example, those of us who learn our languages at different times in life, one common observation in the literature is that our earlier language(s) affect the way in which we use our later one(s). This has been called “first language interference”, and means that we use our younger languages in ways that remind of our use of our older ones (for reasons why this may be so, regardless of learning order, see a previous post).
But research shows that the converse is also true. The Norwegian-American linguist Einar Haugen provided us with the first comprehensive report of how a later language (English) interfered with an earlier one (Norwegian), in his 1953 book The Norwegian Language in America. A Study in Bilingual Behavior. More recently, and particularly concerning accent, two other studies reported similar findings. In an article titled Gestural drift in a bilingual speaker of Brazilian Portuguese and English, Michele L. Sancier and Carol A. Fowler observed interference from English in the production of consonants of a native Portuguese-speaking adult learner. My own report on my children’s learning of English, a language which they acquired in school, Prosodic mixes: strategies in multilingual language acquisition, showed use of English prosody in their home languages.
Given these observations, it may make better sense to conclude that using our languages cannot threaten their “integrity”. Speaking in metaphors about abstract entities (like saying that “languages affect languages”) is just that, speaking in metaphors about abstract entities. In an earlier publication, The analysis of linguistic borrowing (1950), Haugen also reacted to the unhelpful use of such imagery. Discussing shortcomings in the metaphorical use of the word mixes to describe what multilinguals do with their languages, he had this to say:
“A further inaccuracy is introduced if the resulting language is called ‘mixed’ or ‘hybrid’. It implies that there are other languages which are ‘pure’, but these are scarcely any more observable than a ‘pure race’ in ethnology.”
Using our languages is what we are supposed to do with them. We decide. In the introductory chapter to my book Multilingual Norms, I argue that multilingualism is not about what we do to the languages, it’s about what we do with them. If you want to read more about this, you can download the full chapter, titled Multilingualism, language norms and multilingual contexts.
But doing things with languages is surely true of monolingualism too. Next time, I’ll have a look at why we associate threats to the “integrity” of particular languages more with multilingual than with monolingual users of them.
© MCF 2012
Next post: Being multilingual in a single language? Saturday 19th May 2012.