Oxford University Press's
Academic Insights for the Thinking World

English, Chinese, and all, all, all

I think I should clarify my position on the well-known similarities between and among some languages. In the comment on the March gleanings (April 1, 2020), our correspondent pointed to a work by Professor Tsung-tung Chang on the genetic relationship between Indo-European and Chinese. I have been aware of this work for a long time, but, since I am not a specialist in Chinese linguistics and do not know the language, I never mentioned my skeptical attitude toward it in print or in my lectures, the more so as Tsung-tung Chang (1931-2000) can no longer answer me. But a comment in the blog calls for a response.

I think it is proper to refer to Tsung-tung Chang as simply Chang, because the Zürich Sinologist Professor Wolfgang Behr, the author of an excellent article, partly devoted to Chang’s work, does so. Chang, who was a professor in Frankfurt, put together a voluminous dictionary (over 1500 pages) comparing Indo-European and Chinese. I assume that the work’s language is German. That dictionary has not been published, but a representative specimen of the etymologies is available online in the essay titled “Indo-European Vocabulary in Old Chinese…”; I am familiar only with this extract. As his starting point, Chang used the reconstructed roots listed in the dictionary of Indo-European by Julius Pokorny. All etymologists consult this standard work, though the roots one finds there are being constantly discussed, improved, and even rejected. (It is enough to compare this dictionary with its predecessor, known as Walde-Pokorny.) Yet the great compendium remains the most authoritative one we have. For instance, the etymologies of most words in The American Heritage Dictionary of the English Language depend entirely on Pokorny. Chang discovered that hundreds of Chinese words resemble very closely, almost to a letter, Pokorny’s roots and concluded that they were of Indo-European origin. The two families turned out to be related.

A map of ancient China. See feature image credit.

Since I can say nothing about the origin of Chinese words, I’ll note only some general problems with Chang’s reconstruction. Curiously, while comparing almost any two languages, we notice countless similarities. The oldest European etymologists tried to derive the words of their languages from Hebrew or Aramaic, the language of Paradise, which Adam and Eve allegedly spoke. It is amazing how successful they were! Other old comparisons produced less spectacular results, but attempts to trace the words of English and German to Greek also yielded ostensibly noteworthy results.

Then several amateurs appeared who showed to their satisfaction that most words of the modern European languages (or of one language, for instance, English) go back to Dutch, Irish, Arabic (the number of look-alikes between Arabic and English is indeed amazing), or Russian. The Russian “monomaniac” of this type is still active, and it took a serious effort to stop a modern adventurer from reviving the Hebrew hypothesis. He, too, worked with English, and his dictionary is a joy to read (if you appreciate specious humor in historical linguistics). We should disregard the nationalistic impulses behind some of such efforts (they are wicked and not worthy of refutation) and admit that, wherever we may cast our net, the similarities will indeed be numerous, even spectacular. Characteristically, when such authors were not misguided amateurs or doctors (some of the craziest hypotheses on etymology have been advanced by medical men), they were usually not professional linguists, but historians, archeologists, specialists in paleography, or art history. They usually knew a good deal about languages but lacked professional training in the terrain where they ventured to tread. Nor was Chang a historical linguist, and his knowledge of Indo-European and the methods of reconstruction did not go too far. Pokorny’s dictionary is the juice of Indo-European wisdom, but one is expected to know the fruits that have been squeezed to produce it. Sorry for speaking in ornate metaphors.

The images of ancient gods. First image: Muzha (deity) by Anandajoti Bhikku. CC by 2.0 via Flickr. Second image: Marble head of a god, probably Zeus. Public domain via Picryl.

The massive similarities between languages can be explained in several ways. 1) If the oldest words of all languages were monosyllabic (and such are almost all roots in Pokorny’s dictionary), then the number of sound complexes like bid, gob, am, put, etc. is rather limited, and the coincidences are inevitable. 2) Sound-imitative words, whose role in the emergence of speech should not be underestimated, are bound to be similar in all languages. Professor Victor Mair, an American Sinologist and a supporter of Chang’s conclusions, cited the closeness between the reconstructed Indo-European and the Chinese name of the cow. Perhaps the Indo-European word for “cattle” was indeed gwōu-; most likely, it was onomatopoeic. Even the Sumerian noun of approximately the same meaning sounded as gu, and Sumerian is neither an Indo-European nor a Semitic language; its origin has not been discovered. (The Sumerians lived in Mesopotamia.)

Hensleigh Wedgwood, a prominent etymologist of the pre-Skeat era, succeeded in stringing together numerous look-alikes from unrelated languages, and quite often comparativists don’t know what to do with them (therefore, they choose to ignore his material). The same should be said about the words vaguely called expressive. Wilhelm Oehl, the Swiss scholar who wrote extensively about what he called “primitive word creation” (elementare Wortschöpfung), is familiar to the readers of this blog. His far-reaching conclusions are interesting but should be used with caution. 3) Massive borrowing is always a possibility. 4) Finally, perhaps all languages are related, so that there is no mystery in the similarities we observe. I have more than once mentioned Alfredo Trombetti and other precursors of the Nostratic hypothesis.

Rub-a-dub-dub: the beginning of language?Image by Fil.Al. CC by 2.0 via Flickr.
The Old High German “Hildebrandslied.” From the Landes- und Murhardsche Bibliothek, Kassel, Germany, 2° Ms. theol.54, Bl. 1r. Public domain via Wikimedia Commons.
It follows that responsible conclusions about the relations between the oldest languages should be drawn with extreme caution. Above, I ignored the ridiculous ideas that trace all the words of all languages to a very small number of roots, sometimes as few as four. Those conclusions also appear persuasive to many. (Incidentally, the famous Marxist Nikolay Marr, the one-time ruler of Soviet linguistics, was an archeologist and began his career by promoting the indefensible idea that Armenian and Georgian are related.) Apparently, Chang did not realize the complexity of the problem he attacked. Nor does he seem to have been aware of the works by several older scholars who had also noticed and discussed the numerous similarities between the words of Indo-European and Chinese. A historical linguist should take those convergencies seriously but beware of drawing rash conclusions.

Chang acquainted himself with Old High German and was struck by the presence of the words that resembled those of his native language. Among other things, he cited such Modern German phrases as alt und jung “old (people) and young” and concluded that Germanic had once had adjectives devoid of endings. But the forms in such idioms are the product of many centuries of development. He wrote that Southern Germanic might have emerged later under the influence of Altaic tribes, while the system of Northern Germanic had been much poorer. Obviously, he never looked at Runic inscriptions. He also suggested that the oldest forms of Germanic were very simple and lacked declensions and that the monuments of the early epoch at our disposal should not be trusted, because all of them are translations of the biblical texts. Strangely, he missed Old Germanic poetry. It is just such bold hypotheses (“Germanic emerging under the influence of Altaic”; “words in Old Germanic having no endings”) that will compromise any hypothesis.

Chang’s lists contain many animal names, the items that should be treated with great caution in the reconstruction of the ties between and among languages. Next to them, we see such abstract concepts as “tempt,” “defy,” and their likes. This looks odd, to say the least. My point is not to criticize Chang, but to appeal for an unhurried and informed analysis of an important problem that will easily repulse all cavalier attacks.

Feature image credit: Three kingdoms of Ancient China. CC BY-SA 4.0 via Wikimedia Commons.

Recent Comments

  1. Yves Rehbein

    > Chang acquainted himself with Old High German and was struck by the presence of the words that resembled those of his native language.

    –Pareidolia – Wikipedia–

    “Pareidolia is the tendency for incorrect perception of a stimulus as an object, pattern or meaning known to the observer, such as seeing shapes in clouds, seeing faces …”

    This is even funnier with pictograms and letters.

    > sound-imitative words, whose role in the emergence of speech should not be underestimated, are bound to be similar in all languages.

    That statement is unfounded in various, ways, a good rula of thumb, maybe.

    This also goes for characters. 牛 (“ngau”) ‘cow’ looks remarkably similar to the Egyptian and Proto-Semitic depictions of a side-ways cow-heads, but make no mistake, that is not the same cow! Anyway, the older forms do not so much; images of the oracle bone script: https://en.wiktionary.org/wiki/牛

    Whereas Sumerian cuniform logographs already looked uniquely distinct.

    The advent of cattal slavery is almost as far from Sumerian writing as Sumerian is from today (twice,or at least one and a half; while Sumerian is about 5.2 kya with earlier pottery, the spread of cattle domestication is difficult to trace, for a linguist).

  2. Constantinos Ragazas

    Anatoly,

    ” My point is not to criticize Chang, but to appeal for an unhurried and informed analysis…”

    Perhaps you can follow through with that and discuss just those words in Chang’s list with most convincing arguments.

    Kostas

  3. Rudy Troike

    Anatoly,
    You should have pointed out that comparing even the forms of a protolanguage with a modern language is scientifically inappropriate and naive. Any comparisons of Indo-European with Sino-Tibetan languages should be with proto-Sino-Tibetan, on which there is extensive research and literature. Otherwise one is comparing apples and oranges. If there are any loans from an Indo-European language into Old Chinese (NOT modern Mandarin), it would be from Tocharian, the burials of probable speakers of which have famously been found in western China. More recent loans could have come via the Silk Road (e.g. from Italian) or through Mongolian contacts with Russian or even German, thanks to Genghis Khan et al.
    –Rudy

  4. Constantinos Ragazas

    Rudy you write,

    “If there are any loans from an Indo-European language into Old Chinese (NOT modern Mandarin), it would be from Tocharian, the burials of probable speakers of which have famously been found in western China. More recent loans could have come via the Silk Road (e.g. from Italian) or through Mongolian contacts with Russian or even German, thanks to Genghis Khan et al.”

    Or from Alexander the Great incursions into China and subsequent cultural exchanges.

    Consider this:

    “Ancient Greeks artists could have travelled to China 1,500 years before Marco Polo’s historic trip to the east and helped design the famous Terracotta Army, according to new research.

    The startling claim is based on two key pieces of evidence: European DNA discovered at sites in China’s Xinjiang province from the time of the First Emperor in the Third Century BC and the sudden appearance of life-sized statues.”

    https://www.independent.co.uk/news/science/china-terracotta-army-ancient-greek-sculptors-alexander-the-great-marco-polo-a7357606.html

    It would be interesting if Anatoly posts what words in Chinese have Indo-European roots.

    Now that is a post I look forward to!

    Kostas

Leave a Comment

Your email address will not be published. Required fields are marked *