How do you see multilingual communication evolving over the next decade?

The internet is becoming increasingly multilingual, but we 'ain't seen nothin' yet.’

I see it increasing, for two reasons. First, the internet is becoming increasingly multilingual, as more languages come online – and we 'ain't seen nothin' yet,' as some very big language groups (e.g., in Africa and South America) have yet to gain routine access. The proportion of web pages in English is less than 50% now. Second, the rise of English as a global language has led to a reaction which has made language communities more alert to the dangers posed to minority languages by international and global languages.

It isn't only English, of course; in South and Central America, Spanish and Portuguese are the threatening languages. But the result has been that international bodies (e.g., UNESCO, the EU) have put together statements about the protection of minority languages, linguistic rights, and so on. There is even pressure slowly growing for a UN Declaration of Linguistic Rights. Locally, there are also many fresh initiatives aimed at supporting multilingual communities.

How will politics, economics and technology influence its evolution?

All three are important. Politics, in the form of the statements I just mentioned, which help form a climate of opinion, and which give moral backing to the efforts of local groups, espcially in parts of the world where initiatives are threatened (as with Brazilian Indian communities). Also, as multilingualism can be expensive, politicians are the ones who can establish priorities and authorize spending.

Politics also plays an important role in creating the conditions for multilingualism to thrive. For example, EU expansion has enabled many more language groups to move around the region. There are now more than 350 languages spoken in London alone.

The cost of multiligualism must be weighed against the hidden costs of what happens if ethnicity is not respected.

The economics of multilingualism is critical. Yes, it's expensive to maintain several languages, but the cost has to be weighed against the hidden costs of what happens if ethnicity is not respected. The world is full of reports of costly riots, strikes, marches and activism of all kinds over the issue of language. The impact on productivity, tourism and the like is well-known.

Computer technology is enabling things to happen that were inconceivable a decade ago. Chatrooms can now be used as virtual speech communities. Access to language texts is cheap and easy (if your community is online, of course). Translation is more readily available, and in a generation (but not sooner) automatic translation will begin to produce some respectable results. The crucial point is that the technology is 'cool' in the eyes of young people – and these are the critical carriers of language to the next generation.

How can we leverage language policy and education to build and sustain developing economies?

It may seem like stating the obvious, but first there needs to be a language policy (many countries do not have one), and it needs to be well worked out.

It is not a question of language sustaining economy, but the other way round.

There is only the beginnings of a coherent language policy in the U.K., for example, with the National Curriculum in English still being slowly implemented, and government changing its mind about foreign language teaching in schools. Good teacher training is critical to ensure that the policy principles become established in the curriculum – again, something which is only partially the case in the U.K. There are still few published materials.

But once a policy is in place, as in the area I live, Wales, then the relationship with the conomy is plain to see. It is not a question of language sustaining economy, but the other way round. A good economy is needed to enable a language to thrive. Without jobs and a reasonable quality of life, people simply are not interested in language preservation, and who can blame them? But once their economic needs are catered for, then they welcome language diversity and all it stands for. A clear example is Catalunya, where the economy-first principle has been very successful.

What are the cross-cultural repercussions of Friendster-type technology? Most cultures in the world, outside of the U.S., have been living this type of networking naturally for centuries and centuries. How will it impact non-U.S. cultures?

I don't know much about Friendster, though I guess it's like Friends Reunited and various blogging communities that have come to be established over here. I'm not sure. It's too soon to say, as the technology is so recent. The main effect I've seen on language so far is that it is fostering the development of new styles of communication, especially in writing. Normally, written language is edited before being seen in public. Not so in these new contexts. That means that people feel comfortable introducing elements of language that reflect their identities, such as mixed dialect and mixed language forms. There are some amazing written lanaguge styles out there now.

At one time, you predicted that a “world standard spoken English” would eventually appear. Do you still believe that this is possible, or will humans the world over continue to be endlessly creative, putting their own stamp on this “unruly family of dialects, creoles, and patois?”

English is the first language to be spoken by more people as a 2nd language than as a 1st.

Both will happen. That's the point: these two trends are not mutually exclusive. It's perfectly possible to maintain the role of a standard language as a lingua franca and, at the same time, maintain local languages – the standard guarantees intelligibility; the local expresses identity.

There are already signs of English becoming a diglossic language (like Arabic, Greek, German), in which people control two types of language to be used in different circumstances. I already use standard English in public and a local dialect of English at home. This will simply extend internationally, it seems to me. I will use my local dialect at home, my British dialect in the U.K., and my international dialect when travelling abroad. That would be a triglossic situation, in fact.

In some parts of the world the two levels are really clear – such as Singapore, with standard English and Singlish both playing a role. All the signs are that identity languages and dialects will continue to grow. The more they do, of course, the more we need a standard to enable international (or even intranational) communication to work.

Linguistically speaking, it's a whole new world, with non-native speakers of English now outnumbering native speakers 3 to 1. I must point out that there are no real historical precedents for what happens to a language when it is spoken by so many people in so many places – there's never before been a language that's been spoken by more people as a second than a first. A quarter of the world's population uses English now. In Asia alone, the number of English-users has topped 350 million – roughly the combined populations of the United States, Britain and Canada. There are more Chinese children studying English – about 100 million – than there are Britons. (See Not the Queen’s English, 7 March 2005, Newsweek.)

On the other hand, that means the burden is shared, to some extent. No one 'owns' English now, so what happens to it is on the shoulders of all of us.

Editor’s Note: For more on this subject, read English as a Global Language, by Professor Crystal.

As China continues to gain power worldwide, how will the role of English change?

A language becomes a world language for one reason only, and that is the power of the people who speak it.

It does indeed all depend on power. A language becomes a world language for one reason only, and that is the power of the people who speak it. So the question is as much one for political futurologists as for linguists.

However, all the signs are that China is putting its weight behind English (especially in the run-up to the Olympics), recognizing that it is the established global lingua franca, so it seems unlikely that Chinese
will much affect the status of English in the foreseeable future.

Power comes in different forms, of course – political, military, economic, cultural, technological, religious... If we look at India (which currently has more speakers of English than the U.S. and U.K. combined), we see the possibility of IT power growing, based on the country’s call-centers and other IT developments. If the next Google comes from India, then suddenly Indian English will gain in prestige.

According to a recent article in Newsweek, 80% of the electronically stored information in the world is available only in English. How can we improve access to information that is often only available in English? How do you do this in your own work in reference publishing?

Worldwide, 80% of electronically stored information is available only in English.

Well, over the years, the database which produced the Cambridge Encyclopedia (and currently the Penguin Encyclopedia) family of reference books has produced several translations and cultural adaptations. For example, we did a Greek version some years ago, which required that we translate the English text, as well as increase the proportion of coverage devoted to Greece. That extra material was provided by Greek specialists, and was written originally in Greek, then translated into English so that it could be added to the database.

It is, you will perceive, quite an expensive process, but we have done it several times now. It is, moreover, an ongoing process. Our database is updated daily for current affairs. So a translation is not a 'once and for all' event. All translations need to be maintained, if they are to stay current and be useful. The web will make this process easier, in due course.

The other point is to make sure that the kind of English used is as accessible as possible to speakers of other languages. All the entries in my database are edited to a high-school language level, so we avoid the sometimes horrendously complicated polysyllabic entries I have encountered in some encyclopedias.

How does your Global Data Model/”Sense Engine” fit into the concept of the semantic web?

The so-called semantic web is actually a misapplication of the term “semantics” from linguistics.

It doesn't, directly. The so-called semantic web is actually a misapplication of the term semantics from linguistics, where it means the meaning-system of a language (as expressed in vocabulary, discourse, sentence structure, etc.). The people who thought it up have made the term equivalent to knowledge in the broadest sense, i.e., given it an interpretation in conceptual terms. They have every right to do this, of course, but personally, I doubt that the system, as currently advocated, can be given the kind of formalization it needs to enable it to achieve its aims.

My sense engine is a classification scheme that allows groups to organize large quantities of diverse data in a coherent, accessible and efficient manner. It is grounded in a linguistic conception of the handling of meaning and is essentially a means of relating a language's lexicon to a real-world encylopedic classification. In layperson’s terms, the engine is a technology that enables the creation of products, which are, in turn, incorporated into search engines and corporate information networks.

The original name Global Data Model reflected the intention to have the lexical (semantic) analysis applied to all areas of knowledge – people, places, objects, concepts, etc. – and this is actually the name that is used in the U.S. and U.K. patents. But I use the clearer term sense engine these days.

Editor’s Note: Crystal Reference provides content to some of the world's top publishers. Its most recent collaboration has been with Penguin Books for the New Penguin Encyclopedia, Factfinder and Concise Encyclopedia. Its content is also widely used on some of the world's leading Internet sites, from biography.com to websterworld.com, and to enhance content services for the mobile internet.

Can the model be applied across languages and cultures?

About 75% of a language's lexicon is culture-neutral.

Yes, it can, but it does have to be applied. The encyclopedic dimension is fairly universal – trees are trees, whichever part of the world we are in.

The linguistic side is more complicated. About 75% of a language's lexicon is culture-neutral, and the analysis we have carried out for English would thus be capable of translation into another language quite quickly. But the remaining 25% would require careful work.

Is linguistic diversity really as important as you claim? People have sometimes responded negatively to your argument by claiming that you are comparing apples and oranges when you compare linguistic diversity to biodiversity.

The comparison with biodiversity is not a crucial part of the linguistic diversity argument.

There are two issues here. The comparison with biodiversity is not a crucial part of the linguistic diversity argument. If you don't like it, you don't have to use it. Having said that, an awful lot of people do see merit
in the parallel, and I personally think there is a useful correspondence which can be drawn between the way the human race has evolved by adapting itself to different physical and biological environments, and the way the human mind has evolved by developing an array of languages.

The chief arguments in favor of linguistic diversity are four:

• we learn from the visions of others, as expressed in their languages;
• language diversity expresses identity in a way no other aspect of human behavior can;
• language diversity uniquely expresses the history of a community;
• and language diversity gives us insight into what it is to be human.

None of these depend on a parallel with biodiversity.

Sometimes, what we might learn from a language is eminently practical, as when we discover new medical treatments from the folk medicine of an indigenous people. Sometimes it is intellectual, as when the links between languages tell us something about the movements of early civilizations. Sometimes it is literary: every language has its equivalent - even if only in oral form - of Chaucer, Wordsworth and Dickens. And of course, very often it is linguistic: we learn something new about language itself – the behavior that makes us truly human, and without which there would be no talk at all.

The world is a mosaic of visions, expressed through language. Ezra Pound summed up the core intellectual argument: "The sum of human wisdom is not contained in any one language, and no single language is capable of expressing all forms and degrees of human comprehension."

Editor’s Note: For more background on this debate, see (1) Professor Crystal’s book, Language Death, (2) Professor Crystal’s presentation, Creating a World of Languages, at Forum Barcelona 2004, (3) Death Sentence, (4) Words About Language, and (5) Let Languages Rest in Peace.

What cognitive processes in the brain of my neighbor’s 4-year-old multilingual daughter allow her to speak perfect U.S. English to me, perfect Swiss French to my husband, “broken” French to my Turkish brother-in-law (who is learning French), Spanish with the housekeeper and Dutch with her small friends?

Multilingualism is the normal human condition. Three-quarters of the human race is bilingual, and probably half is trilingual.

Why do people find this so surprising? Only people reared in a monolingual environment would do so. In fact, multilingualism is the normal human condition. Three-quarters of the human race is naturally bilingual, and probably half is trilingual. The brain of the developing child is perfectly capable of handling many languages with un-self-conscious ease.

You have to remember that, from the child's point of view, there is no adult perspective: you and I know that these are different languages; the young child does not. They are simply different ways of talking. Awareness of language names doesn't emerge until the fourth year. The principle of accommodation accounts for the differential ability (in the broken French case): people quickly adapt to the behavior of anyone with whom they are in rapport. There is evidence of babies doing this in the first year of life – for example, babbling at a higher pitch when vocalizing to their mother, and lower to their father.

Editor’s Note: For more on this subject, read Professor Crystal’s book, Listen to Your Child: A Parent’s Guide to Children’s Language.

How can those of us involved in high tech be more sensitive to the language-impaired as we develop tools and services to manage and manipulate knowledge and information, and access to them?

The problem in managing language impairment is not in technology, but in our shortage of personnel.

The problem in managing language impairment is not in technology but in personnel – the shortage of them. This is a political issue. We need more speech pathologists, remedial language teachers, and so on. The waiting lists in some parts of the world are terribly long.

But, having said that, I do think we need more focus on the application of technology to this domain, especially at the research level. I don't think we have a cadre of PhDs who have specialized in (for want of a term) technological language pathology. Is there a university department anywhere where this is a primary focus? It would include hardware and software development, ergonomics, computer processing, internet activity and a great deal more.

Central language labs would process language, instead of blood, samples.

When I was developing clinical linguistics in the 1970s with various colleagues, a pressing need was to help clinicians by saving them time in doing their detailed, case-by-case language analyses. We devised a set of profiling procedures which seemed to help, but they were still time-consuming. One or two people experimented with computer applications, which were also successful. However, my impression is that very few people were competent to do this, and I'm not sure that the situation is greatly improved today.

I would still like to see the development of a central language pathological laboratory system – like the path labs in conventional medicine. These labs would process language samples, not blood samples. Speech and language pathologists could send in their data samples to have them processed. I first suggested this in the 1970s, but the idea was thought to be too expensive. Maybe the climate is better today?

What are the next big challenges in the field of (applied) linguistics?

In my book, The Language Revolution (2004), I identify three.

• Managing the consequences of having a global language.
• Preserving linguistic diversity, both by documentation and (where practicable) revitalization.
• And managing the consequences of the internet, which has brought us a fourth medium of communication (alongside speech, writing and signing).

Editor’s Note: For more on this topic, read Language and the Internet , by Professor Crystal.

---

Professor David Crystal is one of the world's foremost experts on language and on reference publishing and has written more than 100 books. He currently writes, edits, lectures and broadcasts from his home in North Wales in the U.K., where he is also chairman of Crystal Reference Systems. Crystal was awarded the Order of the British Empire (OBE) for services to the English language in 1995 and is now Honorary Professor of Linguistics at the University of Wales.