The English-Learning and Languages Review Æ Homepage

Donald Kenrick as polyglot:
could he be replaced by a machine?

Erik V. Gunnemark

This article is chapter 12 of Scholarship and the gypsy struggle: commitment in Romani studies, edited by Thomas Acton. This volume is to be published on 14th December 2000 by the University of Hertfordshire Press, which has kindly given permission for chapter 12 to be reproduced here. Detailed information on the book can be found at festchrift or at herts.

Chapter 12:

Donald Kenrick as polyglot: could he be replaced by a machine?

Erik V. Gunnemark

According to the information available to the association Amici Linguarum, Donald Kenrick is probably now ‘Polyglot No.1’. In the age of information technology we have to ask whether this is still a model to which to aspire, or whether it is a talent superseded by the computer.

The Association Amici Linguarum was founded in 1964 as a Scandinavian association of people interested in languages and culture. In the late 1970s Donald Kenrick joined the circle collaborating with me on the production of a general directory of countries, peoples and their languages, and was listed as co-author of the first two editions, published under the title A Geolinguistic Handbook in 1983 and 1985. In 1985 the association began collaboration with the American Society for Geolinguistics and by the end of the twentieth century it had members in about thirty countries around the world, particularly in Europe and North America but also in Latin America and East Asia. A mixture of amateurs and professionals, it has thirty-five professors among its members but also writers, librarians, translators, journalists, teachers, doctors, diplomats, civil servants and restaurateurs. A few of the members describe themselves as linguists but the interest of the group is not so much in linguistics as in the languages themselves and the use to which multilingual competence can be put.

Every year that passes sees a greater accumulation of knowledge and experience within the ranks of Amici Linguarum. However, as well as deriving pleasure from this treasure-house, many members have been able to make use of it in a direct practical way in the course of their work in schools, universities, libraries and various types of enterprise, and as authors or translators.

There has been considerable correspondence within the Association about remarkable polyglots of the past, and many stories – or perhaps we should say, legends – exist, but I know of only two cases where information which may be considered factual and reliable exists about such ‘superpolyglots’ of the past: Pent Nurmekund (Gethin and Gunnemark 1996: 317-9) and Emil Krebs (Matzat 1999).

Dr Emil Krebs (1867-1930) was a professional interpreter and translator for the German Ministry for Foreign affairs. Matzat’s (1999) pamphlet concentrates on the period of the German colonies in Peking and Tsingdao (where they founded the famous Chinese brewery which thirsty scholars still have cause to praise), and shows that Krebs was quite a character. His obituaries in newspapers and journals in 1930 celebrated the fact that he could translate from over 100 languages and spoke around sixty of these. The exact number of languages given depends, of course, on what is counted as a separate language. In this paper we are erring on the cautious side by not counting as separate languages dialects which are generally considered to be ‘the same’ language.

Professor Pent Nurmekund (1905-97) was founder of the Oriental Department in the University of Tartu in Estonia. He could translate from about eighty languages and also spoke many of them.

Among living European polyglots Donald Kenrick is pre-eminent, able to translate from over sixty languages, and speaking around thirty of them, mostly fluently. Others include: Eugen M. Czerniawski in Moscow, who can translate from about forty-five languages and is a fluent speaker of nearly twenty of them, and Arvo Juutilainen from Helsinki in Finland, who can translate from over fifty languages and speaks a dozen of them, more or less fluently. The greatest American polyglot is probably Georges Schmidt who was born in Strasbourg in 1915, but moved to New York. In his heyday he could translate from about sixty languages and speak over twenty of them.

So far only one book on polyglottery has been published which I can recommend, that by Dr Dmitri L. Spivak (1989). He based his work on interviews with polyglots in Russia and other parts of the USSR. The most striking fact he claimed to discover was that most of these polyglots agreed that they did not know more than about seven foreign languages ‘completely’, in the sense of being able to speak them fluently and read new and differing texts without any difficulty. He presented this as ‘The Law of Seven.’ This is a very provisional and approximate ‘law’, of course, but an interesting point of departure for further research – when he and I are able to resume the collaboration we put on hold in 1993 because we had too much other work to do.

Another unexpected finding of Spivak’s (1989) research was that all the polyglots interviewed preferred to learn languages on their own. So, polyglottery may well be called ‘a profession for autodidacts.’ Furthermore, like most polyglots in other countries, (c.f. Gethin and Gunnemark, 1996: 50-1), the Russian ones regard today’s language-teaching industry as a glaring scandal. But should we be worried about this? Is the example of Kenrick irrelevant in the modern world ? Could he be replaced by a machine – not now, of course, but say in about another twenty years?

The Fata Morgana of machine translation: “Just wait for another twenty years!”

It is sometimes argued that machine translation will make polyglottery redundant but it seems to us that the development of information technology will make competent multilingual translators more and not less necessary. The history of machine translation has been one of recurrent enthusiasm followed by disillusion. The initial burst of enthusiasm in the mid-1950s predicted that within twenty years most scientific, technical and other non-fiction would be machine-translated. In the mid-1960s, however, the US Department of Defense pulled out of financing research. As this experience was forgotten however, advances in computing brought forward a new wave of enthusiasm in the mid-1970s, suggesting that, if not now, at some time in the next 20 years machine translation would be used to save time and money in all non-fiction fields. As time progressed the claims became more extravagant. At a translation conference in Lund in Sweden in 1985 Professor B. Sigurd predicted that – again within the next twenty years – machine-translated texts will be read in our newspapers like sympathetic voices. In the 1990s, with the spread of the Internet to millions of people without the knowledge and experience which are necessary to be able to distinguish between true and false statements, the danger of misinformation has become rather more serious. There is an acute need of an information counter-offensive.

In my view the claims of various corporations offering machine translation are dangerously overstated, but meet with relatively little protest, perhaps because even academics have relatively little understanding of how machine translation ‘works’. The necessary post-editing is not mentioned at all, or only in passing. In fact, the term ‘machine translation’ is actually misleading, for as a rule it has to be combined with post-editing, even for rough translations. For professional standards of translation, competent translators are still needed.

Sometimes pre-editing may be necessary or preferable; that is to say, if computer manuals or such are put into a very standard form, machine translation may give better results. It is possible that some time and money has been saved by giant corporations using such techniques but they would still be wise to have a professional review of the result! One should therefore always take a sceptical attitude to claims that money is saved by using machine translation instead of traditional translation.

We need people to make sense of the things other people say or write. The development of computing and the globalisation of information technology make polyglots like Donald Kenrick more and not less necessary. They can form their own assessments of what is meant by, and what the importance is, of information in another language. The priorities programmed into machines can only ever be a selection of human concerns. A broad human vision must stay in charge of technology. As a language teacher Donald Kenrick has been in the forefront of using new technology to promote language learning, but not to replace it. Information technology will make multi-lingualism easier, rather than redundant.

References

Gethin, A. and Gunnemark, E.V., 1996 The art and science of learning languages, Intellect/6 Thomas Street, Lawrence Hill, Bristol BS5 9JG

Matzat, William, 1999, Dr Emil Krebs, The Association for German Life in East Asia, Bonn.

Spivak, Dimitri L. 1989 Kak stat’ polyglotom, [How one becomes a polyglot], private, St.Petersburg.

Revised layout 20 April 2010

The English-Learning and Languages Review Æ Homepage