Oxford University Press's
Academic Insights for the Thinking World

The Oxford Etymologist Looks at Race, Class and Sex (but not Gender), or, Beating a Willing Horse

By Anatoly Liberman

For a long time, I have been meaning to join the mainstream and address, if not the importance of race in our life (and is there anything more important?), then at least the derivation of race. Why I am doing it now will become clear from my April “Gleanings.” (This is said to keep the readers of my blog in awful suspense for a whole week.) The English word was borrowed from French, and the speakers of many other European languages also added race to their vocabulary through French mediation, but its distant origin has been a matter of debate for more than a century. Only one fact seemed to be certain, namely that despite the success of race the story began with Italian razza. The debate raged only over whether, in Italian, razza was native (that is, going back to some Latin noun) or whether the Italians had taken it over from somewhere else. As a rule, the etymology of such intractable words remains unsolved forever, but in this case sufficient clarity seems to have been achieved for language historians to be able to give a satisfactory answer. A list of conflicting hypotheses is typical of the state of the art.

There was an old German noun reiza “straight line, stroke, mark” that Eugen Diez, the founder of comparative Romance philology, took for the etymon of razza ~ race. The semantic fit is good (compare the inner form of the noun lineage and phrases like line of descent; Diez also adduced a good parallel from French and Walloon), and even the vowels can perhaps be made to match, but the chronology militates against this reconstruction, for the German word went out of use rather early, while razza, with this spelling, surfaced in Italian only in the 13th century. The non-attested Langobardian raiza, meaning the same as reiza, can be dismissed as a lame attempt to save the situation, though it was offered by an outstanding specialist in Romance etymology (Langobardian is a Germanic language at one time spoken in the north of Italy; hence the name of the province Lombardy). Incidentally, all the conjectures being surveyed here stem from specialists, so that none of them is silly.

Another candidate for the etymon of razza was Latin radix “root.” Once again the semantic match is excellent. The phonetic difficulties are not insurmountable, as evidenced by the English noun race “root,” (compare Shakespeare’s a race of ginger) that does go back to radix or rather to radicem, its accusative (stress on the second syllable). Besides, Old and Middle French had racine “race,” obviously from radix. According to perhaps the most ingenious suggestion, razza is a clipped form of Latin generatio “generation” through naraccia (allegedly, from naraccia to una narazza and una razza). Clipped forms have always existed, and in English, which favors monosyllables, they play an outstanding role, but this method of producing words was little used in the past.

A rather bizarre derivation of race from Slavic raz “a blow” (a blow leaves its mark or imprint) died almost without issue, though Skeat’s predecessor Hensleigh Wedgwood, who used to string look-alikes from many languages, looked on this connection as possible. Arabic ra’s “head,” Basque arraca ~ arraza “male pedigree animal,” and Romance (unattested) raptiare “to breed falcons” (from raptiare to the verb racer and further to the noun), have also been tried. The Arabic etymon became well-known to the German readership thanks to the efforts of Eugen Oberhummer. The obsession with race in the Nazi time needs no explanation, but Oberhummer’s three articles (one was published in Austria, and two appeared in Germany) had no evil overtones, except that a reference to a Jewish scholar was withheld; however, one of them graced the infamous journal for Rassenkunde, literally “race-lore, the study of race.”

All the suggestions mentioned above share an important weakness. They are, to a varying degree, plausible from a linguistic point of view but have no foundation in the history of the concept. To decide how a word for “lineage, pedigree” originated, we should know who needed it, that is, in what circumstances it was coined and how it spread from one layer of speakers to another. An abstract reconstruction, based only on the interplay of sound and meaning, is bound to remain a shot in the dark. Only two hypotheses go beyond intelligent guessing. Latin ratio “account, calculation; reason” also meant “order, law, system, way,” and already in the early Middle Ages it could designate “type, kind, species.” Italian razza developed, as it seemed, from ratio and, with phonetic variants, became a household word in the rest of Europe. This etymology ran into several chronological difficulties, but, on the whole, it made sense. Leo Spitzer, whose name turns up every time I touch on a Romance subject, did not invent it, but he was its main proponent, and the most influential scholars, with few exceptions, agreed with him. (Those who disagreed defended much less attractive derivations.) It did not escape the proponents of this theory that the French word race was often applied to horses, and Spitzer accounted for the popularity of the word by referring to the role of horsemanship.

The most significant breakthrough happened in 1959, when Gianfranco Contini published an article in which he showed that an old Italian author had used razza while translating the French noun haras “stud” (to anticipate the natural question: most probably, haras has nothing to do with the English verb harass). He concluded that the etymon of race was French haras, which lost its initial h (as always), and that Italian razza, far from being the etymon of race, was an adaptation of the French noun. I’ll skip the morphological complications that have been dealt with rather well and mention only one fact. A chance gloss in a translated text would not have gone far enough to explain a swift adoption of race by the French (after all, it could have been a case of folk etymology, almost a pun, with the French author being seduced by the similarity of the two forms), but subsequent research showed how race ~ razza progressed in Italian and French, and there is no reason to doubt its results. All the pieces of the puzzle suddenly fell into place. It could have been expected that race would emerge not as a bookish creation but as a term of cattle or horse breeding (whatever the etymology of haras may be) and that it would be applied to humans later. Indeed, in Dante’s Italy razza was used only about animals; for people the word schiatta existed. Both Italian schiatta “stock, descent, lineage” (to say nothing of razza) and French haras “stud” have continued into the present (compare di nobile schiatta “of noble descent”). As we have seen, the true connection had been suspected early enough: Arabic ra’s, Basque arraza, and Spitzer’s reference to chivalry and horsemanship should not be overlooked, but in all those theories horses, falcons, and so forth played an accidental role, whereas they should have been the focus of the investigation.

So now we know how race came into being, but a short addition may not be out of place. New words, whether native coinages or borrowings, have a better chance of survival if, once they surface, they find support from other words. Perhaps ratio would not have yielded race on its own, but race, from haras, met a powerful ally, for ratio, which became its homonym, also meant “species” (among other things). Even generatio, a more distant ally, may have helped race to stay. Accidental and folk etymological ties play a significant role in the life of words. (Engl. race “a rush forward” is not related to race “pedigree,” horse races notwithstanding.)

Class and sex were also promised in the title. They were used as a lure, I am sorry to say. Like race, both are Romance words. Sexus, from Latin via French, is probably akin to the verb secare “to cut” (compare section). Class is a direct borrowing of Latin classis. The sense “division of persons or things” goes back to the 17th century. In the 16th century, class “division of the Romans; a group in a university” occurred in Scots. Latin classis meant “fleet; muster of citizens.” Its origin is obscure, and the explanations in older dictionaries and compendia should be treated with great caution.

The race may be to the swiftest, but in etymological studies the hedgehog has every advantage over the hare, if reference to the famous tale from the Grimms is allowed.

Anatoly_libermanAnatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to [email protected]; he’ll do his best to avoid responding with “origin unknown.”

Recent Comments

  1. […] week, my post was devoted to the origin of race. Here I think we are in a better situation than with pet. Contini’s reconstruction is not an […]

  2. Counsel

    I don’t think your race (or anyone’s race for that matter) is that important. I’ll judge you (and everyone else) on how you treat others and on your thoughts and actions.

    To me, race is irrelevant. I don’t care what you are, and I don’t think most younger kids do either. That is why “we” don’t see things in the limited color-wheel as older folks. No disrespect intended…

  3. Steve Sailer


    Tracing race to “stud” helps explain why a large fraction of Shakespeare’s uses of “race” seem to mix some combination of lineage, breed, class, sex, and horses. This is not to say that Shakespeare was an etymologist, but he had a rich understanding of how sex and race are related, an insight that seems to have largely been lost in modern America — and because America now dominates thinking about race, the world.

  4. […] On the derivation of the word ‘race’ read this. […]

Comments are closed.