By Anatoly Liberman
A word, some scholars say, can have several etymologies. This is a misleading formulation. Various factors contribute to a word’s meaning and form. All of them should be taken into account and become part of the piece of information we call etymology, because words are like human beings. Someone we know had two parents and inherited their traits, along with those of many generations of his ancestors, then grew up in an environment that partly reinforced and partly suppressed those traits, changed his habits under the influence of his domineering wife, and took her last name, to spite his parents. He has recently celebrated his 100th birthday. Words too come from a certain source, begin to interact with their neighbors (some mean nearly the same, and the newcomer either tries to stay away from them or drives them out of existence; others sound like it, and their closeness affects its meaning or stylistic coloring), grow old and dull or join a disreputable gang, and perhaps die. Each event deserves the attention of a language historian, but it is better not to speak of the multiple etymologies of one word.
In other cases, two or three sources look like a word’s probable etymons. Only one of them was its true parent, but we have no way of recognizing it. Both situations seem to be relevant to the history of akimbo. Among the conjectures about its origin some are reasonable. It is also possible that, regardless of the real etymon of akimbo, the word may have succumbed to the lures of folk etymology, a process that usually obliterates ancestral traits. This is the reason the most cautious dictionaries say “origin unknown.” But theirs is not the ignorance born of the lack of evidence. It is akin to the dilemma that faced Buridan’s ass, which, being placed between two equally appetizing stacks of hay, starved to death, unable to choose the best one. Those who can visualize the position called “with hands (or arms) akimbo” will agree that invoking the image of that unhappy animal could not be more apt.
There is the Italian phrase a sghembo “awry, aslope,” and it has been proposed as the etymon of the English word. Several factors weaken this idea. Someone who suggests borrowing should show in what circumstances the lending language shared its resources and why people from another country decided to accept the gift. If these conditions are not met, the hypothesis has no merit. We know why English took over a multitude of musical terms from Italian, but why akimbo? Were Italians famous for having their “hands on the hips and the elbows turned outward,” to quote an admirable dictionary definition? The worst thing about this etymology is that the Italian phrase has nothing to do with the position of the arms. Consequently, the English are supposed to have borrowed a sghembo and endowed it with a sense remote from the original one. As we will see, this argument will also prove deadly for another attempt to trace akimbo to a foreign source. An etymology killed with such heavy artillery may not need a few additional bullets, but we cannot help observing that Italian gh designates “hard g” (as in Engl. get), whereas akimbo has k. The parallel form a schembo (sch = sk), was dialectal, so that its popularity among English-speakers could not have been significant at any time.
Akimbo surfaced as in kenebowe (1400). More than two centuries later the variants a kenbol(l) ~ a kenbold appeared. For their sake, and perhaps not without some regrets, we will leave Italy for Scandinavia. The Icelandic words kimbill, kimpill, and kimbli “bundle of hay; hillock,” once compared with akimbo, exist. According to some old dictionaries, they mean “the handle of a pot or jug,” but they do not. Their root is related to Engl. comb and was used in Germanic for coining the names of fastenings, barrel staves, and so forth. However, similar words (kimble, kemmel, and many others), designating various vessels (not handles), are current in modern British English and Swedish dialects. For this reason, Ernest Weekley set up Middle Engl. kimbo “pot ear, pitcher handle.” The metaphor, from a pitcher with two handles to a person with hands akimbo, is perfect and widespread. In kenebowe may have been a conscious translation of the French phrase en anses “on the handles,” as Weekley says, but why is it so different from present day Engl. akimbo, especially if we remember that Middle Engl. kimbo has been reconstructed rather than recorded and that 17th century authors knew kembol(l). What happened to final -l? Weekley did not provide an answer to those questions. Akembol could not develop from in kenebowe in a natural way. More likely, it was a product of folk etymology, perhaps indeed under the influence of the names of pots and jugs.
A third putative source of akimbo is Gaelic cam “bent, crooked”; the English adverb kim-kam “all awry, all askew” has been attested. Since -bowe in kenebowe means “bend” and is identical with -bow in elbow and rainbow, kimbo, from ken-bow ~ kin-bow ~ kinbo, emerges in this reconstruction as “bent bend,” a tautological compound (both of its parts mean the same), like many others in the Indo-European languages. Compare Engl. courtyard, pathway, etc. and numerous place names, which, when deciphered, yield “white white water,” “hill-hill,” and so forth. While reading the entry akimbo in Skeat’s dictionary, I discovered, much to my surprise, his passing statement on the popularity of such compounds, as though this fact were the most obvious thing in the world. It is not, and few researchers are aware of them. The suggestion that just one component of akimbo is Celtic has little to recommend it. In sum, akimbo would be easy to explain, if its earliest form were not kenebowe. Lost among Italian, Gaelic, Icelandic, and English, we will return to Scandinavia.
Another form that allegedly might generate akimbo is Icelandic kengboginn “bent into a crook.” British dialectal kingbow looks like a variant of it. This etymology is given in most dictionaries as final. A late 14th century English word could have been borrowed from Scandinavian, but Italian a sghembo hastens to take its revenge. Kengboginn never meant “akimbo,” and a change from “bent, crooked” to such a highly specific meaning (“with one’s hands on the hips”) is suspect. Also, keng- in kengboginn, like kimble, bears little resemblance to kene- (-bowe, is not incompatible with -boginn, however). Once again we wish there were no kenebowe.
At first blush, kene- in kenebowe is the adjective keen. If so, in kenebowe must be understood as “in keen bow,” that is, “in a sharp bend, at an acute angle, presenting a sharp elbow” (such are the glosses in The Century Dictionary). In Middle English, keen “sharp-pointed” “was in common use as applied to the point of a spear, pike, dagger, goad, thorn, hook, anchor, etc., or to the edge of a knife, sword, ax, etc.… In its earliest use, and often later, the term connotes a bold or defiant attitude, involving, perhaps, an allusion to keen in its other common Middle English sense of ‘bold’,” The quotation is from the same dictionary, which calls all the previous explanations erroneous.
Skeat defended the kengboginn etymology and kept repeating that Middle Engl. kene was not used to denote “sharp” in such a context. He never elaborated on his phrase in such a context. Despite Skeat’s objection, the etymology of kenebowe defended in The Century Dictionary seems to be the least implausible of all, assuming that the first vowel of kenebowe was long; the vowel in keen undoubtedly was. This is not too bold an assumption, for kene- with short e has no meaning. Later, this e must have been shortened (a usual process in trisyllabic words, to which we owe short o in holiday, as opposed to long o in holy, for example), and the change destroyed the tie between kene- and keen. The second e was shed—another common process in Middle English. The new form (let us spell it kenbow) began to resemble words for vessels with two handles and in kennebowe became akingbow, akingbo, akimbo, and so forth. In the disguised compound akimbow, the idea of a bow also disappeared (even an association with elbow did not save it); hence the spelling -bo. The influence of Gaelic cam need not be invoked in the history of akimbo.
Faced with many hypotheses, none of which should be dismissed as untenable, we are still not quite sure where akimbo came from, but “origin unknown” would be an unnecessarily harsh verdict. In 1909, the first edition of Webster’s New International opted for keen-, the second (in 1934) cited kingboginn, and the third (1961) gave the earliest form (kenebowe) and stopped. This is what I call the progress of the science of etymology.
Anatoly 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@example.com; he’ll do his best to avoid responding with “origin unknown.”