Alexander Ludskanov and Natural Translation

The fundamental tenet of the Natural Translation (NT) Hypothesis is that all bilinguals can translate. The first person to state that explicitly was the brilliant Bulgarian semiotician Alexander Ludskanov (A.L.) in 1967. He did it in a book in Bulgarian; the literal translation of its title would be Human and Machine Translation. He translated the book into French himself in 1969, and what he wrote was this:

Grâce à une certaine intuition et à une certaine habitude, chaque sujet bilingue traduit d'une manière ou d'une autre. Par conséquent, la science de la TO [traduction humaine] n'avait pas à s'occuper de la question comment apprendre à l'homme à traduire, mais de la question comment lui apprendre à agir d'une manière ou d'une autre pour obtenir des résultats correspondant à certains critères acceptés a priori.

Here’s my English translation:

By intuition and habit, all bilingual people can translate in some way or other. Consequently the fundamental question for the study of human translation is not how to teach people to translate, but how to teach them to behave in a way that will produce results conforming to certain well-established, accepted criteria.

He couldn’t have said it better; but does that mean he was an early proponent of research on NT? Not at all. First, if you re-read his second sentence, you’ll see that what he recommends is not the study of the spontaneous phenomenon, but of how natural translators can be trained to translate according to the norms of their society. Secondly, the passage I’ve quoted only occurs in a footnote. He certainly knew what he was saying, but the fact that he didn’t put it in the body of his text means, IMHO, that like several other early discoverers of NT, he didn’t appreciate its full significance.

Never mind. He clearly saw that NT exists and that it’s universal.

A.L. impressed everyone who heard him lecture by the convincing clarity of his arguments - his first degree was in law. Alas, he died in 1976 at only fifty. My wife and I knew him and his family personally, and their little old-world house in Sofia. It was thanks to him that we and his other friends of the International Committee on Computational Linguistics were able to slip behind the Iron Curtain and visit Bulgaria (Sofia, Varna, the Black Sea coast) in those far-off days of the Communist regime. We were surprised by the high level of translation activities and of intellectual life in what most people in the West thought of as a Balkan backwater. This was partly due to the Bulgarians' proximity to the linguists and scientists of the Soviet Union: A.L. himself had studied in Moscow. The Bulgarian government aped the Soviets, and A.L. received official support for his machine translation project because MT research was in vogue in the USSR. He was no supporter of the regime, but he’d managed to find a refuge in the Institute of Mathematics of the Bulgarian Academy of Science because, as he explained to me, “Mathematics is the only branch of learning that they haven’t found a way to politicize.”

There are a number of other interesting ideas in his book, but they aren’t directly relevant to this blog. What does it mean, however, to say that he was a semiotician? Semoticians study all kinds of sign systems (N.B. not just signs, but systems of signs). For them, languages are one kind of sign system, and there are others just as important. What would be such another? The genetic code, for example. For semioticians, therefore, a language translation is a conversion (or a series of conversions) of information-bearing signs, just as the transformation of DNA into RNA into protein is. (In this connection, see my July 25 post.) Semioticians think, as A.L. would have put it, “at a higher level of abstraction” than linguists.

To be continued.

References

Ludskanov, Ljudskanov, etc.: These are just variant transliterations of his Bulgarian name. Likewise Alexander, Alexandre, Aleksandar.

Aleksander L’udskanov. Prevezhdat chovekt i machinata. Sofia: Nauka i Izkustvo, 1967. 159 p. Published version of his doctoral thesis of 1964.

Alexandre Ljudskanov. Traduction humaine et traduction mécanique(Documents de linguistique quantitative 2 and 4). French translation by A.L. himself, Paris, Dunod, 1969, 2 fascicles.

The website of the International Committee on Computational Linguistics is at
http://nlp.shef.ac.uk/iccl/.

There's an article in Wikipedia on the Institute of Mathematics of the Bulgarian Academy of Sciences.