Etymologists constantly lament their fate: either the word to be explained was not recorded early enough (a typical case with vulgarisms and slang) or it is isolated (some monster like catawampus), or it has a sufficient number of allied forms but they are so similar that their semantic history cannot be traced (this circumstance—to give one example—handicaps the research into the origin of dwarf: like their English cognate, German Zwerg, Swedish dverg, etc., denote a short person, but how this combinations of sounds acquired its meaning is hard to decide ). The list is long enough for composing a full-fledged elegy “The Etymologist’s Complaint.” By contrast, an overwhelming amount of material may also pose problems. A case in point is the children’s verse of the eena—meena—mina—mo type. It has been recorded in numerous countries from east to west—naturally, in different form, but the first word is more or less the same everywhere. Although eena and its kin resemble one, they are too far from it to be qualified as its “garbled” or “corrupted” variants. Nor are meena, mina, and mo less obscure than eena. Why some human or beast has to be caught by the toe is equally puzzling but will not concern us. I will touch only on the English word eena and its analogues.
Our first sojourn will be in the Yorkshire Dales, where sheep are (or were at the end of the 19th century) scored as follows: yain, taien, tethera, (m)ethera, pi(m)p (the first five numerals). Another variant is eina, peina, para, pattera, pith. The origin of those numerals has been subjected to a long and fruitful discussion, whose main result is that we observe here a relic of an ancient British (Celtic) system of counting. Since to modern speakers yain, taien, etc. are meaningless words, their form is unstable and tends to vary from region to region. Some of the lists are mere gibberish, with English words replacing the original numerals, and rhyming words invented by informants. Complications arise when we cross the ocean and discover a similar string of numerals in use among the native population in North America, for example, een, teen, tother, fither, pimp, with the variant eeny, teeny, tuthery, fethery, fip. The American (Wawena) numerals were published in 1867. According to an informed opinion, those scores should be regarded as tally-marks rather than numbers; they were used in counting by fives, tens, or twenties. Presumably, they were “brought to New England by English colonists and used by them in dealing with the Indians in counting fish, beaver skins, and other articles of traffic. When the memory of their origin was lost, the Anglo-Americans believed them to be Indian numerals, and the Indians probably believed them to be good English.”
According to other hypotheses, the home of the phrase eena, meena, mina, mo is French Canadian or a language spoken on an island off the West Coast of Africa. Both hypotheses are fanciful. One should never tire of repeating that the idea of borrowing has value only when the way of penetration is known. In the world of words, tales, and customs, seeming convergences abound. Some words, plots, and rituals often have close analogue in different regions. It is easy and tempting to suggest borrowing. Positing loans without identifying intermediaries is a waste of time. The Celtic origin of sheep scoring is nearly certain. Incidentally, not only sheep and not only in Yorkshire are counted this way. The unresolved question is the connection between the American list and eena, meena, mina, mo. The resemblance between the rhyme and eina, peina, para, pattera; yain, taien, tethera, (m)ethera; eeny, teeny, tuthery, fethery is minimal. Only the first word is almost the same, and, as pointed out, it is such everywhere. Far from the English speaking world, Russian girls begin their games with the words eniki, beniki (eni-, pronounced like Engl. any). The second words (meena and beniki) must have been coined as rhyming partners of eena and beniki, but where did the first ones come from? The Old English for one was pronounced an (with a as in father), in Middle English it changed to on (with o as in British pawn); neither can be called a phonetic variant of eena. Equally slight is the similarity between eniki (after we subtract the meaningless ending -ki) and Russian odin “one” (stress on the second syllable; its older form began with ted- or yed-). Referring eeny to some “ancient British numeral” like eina suggests itself (obviously, a reasonable solution), but why wasn’t the entire sequence reproduced? Why only the first word? Children’s folklore often preserves remnants of ancient incantations, but no such incantation has been found. As far as we can judge, no magic formula ever began with eeny or eniki. Russian etymological dictionaries do not discuss eniki; the OED calls eena a nonsense word. To be sure, all is nonsense that we do not understand. Folklorists believe that the English counting out rhyme is relatively recent. If this is true, the emergence of eeny and its closeness to words like eniki makes the riddle of its origin even harder.
Anatoly Liberman is the author of Word Origins…And How We Know Them. His column on word origins, The Oxford Etymologist, appears here each Wednesday. Send your etymology question to firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”