By Anatoly Liberman
I am saying goodbye to the Harlem Shake. The miniseries began two weeks ago with dance, moved on to twerk and twerp, and now the turn of the verb shake has come round. Reference books say little about the origin of shake. They usually list a few cognates and produce the Germanic etymon skakan (both a’s were short). This form adds nothing to what we already know, because Old Engl. scacan ~ sceacan, Old Saxon scacan, and Old Icelandic skaka have been attested; their Old High German cognate existed too. All of them meant “shake” and were obviously related. Dictionaries frequently stop in the most interesting place. After being presented with the information that once upon a time the ancestors of the Anglo-Saxons, Scandinavians, and others had the verb skakan, we would like to know why the sound complex skak- meant what it did. Oh yes, quite possibly, Sanskrit khajati “agitate, churn” also belongs here, and a dubious Old Irish congener need not be ignored. Nothing like being informed that so many Indo-Europeans once shook in sync, but the main question remains unanswered.
This “main question” can seldom be answered. If we lack the means to show that a word is sound imitative or sound symbolic, we usually end up with hypothetical roots that never existed in isolation and whose earlier history we are unable to reconstruct. Where do we go from skakan? The first editors of OED saw no helpful forms in Germanic and only one outside it (Sanskrit khajati). James A.H. Murray and Henry Bradley, OED’s first editors, adhered to the admirable principle that less is more and discouraged idle speculation. This policy had the result that their etymologies aged remarkably well, but it also had a negative side effect: countless authors who copied from OED (or should we say plagiarized it?) seldom dared offer original conclusions and stayed with what they found in that work. As a result, English etymological lexicography stagnated. To be sure, there also was Skeat, but he lacked the authority of the greatest dictionary in the world. Apparently, what was good for OED was good enough for everybody else. (I may once have quoted the probably fictitious remark of an English speaking student who refused to waste time on learning French and said: “The language that was good for Jesus Christ is good enough for me.”)A word meaning shake can — even should — have a symbolic base because shaking is a physical activity that cries for a colorful name. And indeed, in several Slavic languages skok means “jump”; elsewhere, sk- also refers to jumping and moving fast. Such are Latin scateo “bubble, spring forth,” Greek skaíro “run; spring, hop,” and quite a few others, including a similar Sanskrit verb and a noun for “hare,” the greatest jumper of them all. Middle High German schehen (known today only with the prefix: geschehen “happen”; sch goes back to sk) meant “hurry”; its possible cognate is Modern German schicken “send.” Wilhelm Theodor Braune, a famous German scholar, whom I once mentioned in another connection and whose habit of calling himself sometimes Wilhelm and sometimes Theodor I rued, believed that shake and schicken are related. Hensleigh Wedgwood, a staunch supporter of the onomatopoeic origin of too many words, cited Engl. shock, shog “shake, jolt” (now chiefly dialectal, but some people will remember it from Shakespeare’s Henry V, where it means “move along”), and jog alongside shake. Jog—its earliest recorded sense is “prod, stab”— may be an expressive variant of shog; I have often had a chance to note that initial j- is endowed with an expressive value.
Shock has a more complicated history. In addition to shock1, as in shell-shocked and in this news shocked us, there are shock2 “a pile of sheaves of grain” and shock3 as in a shock of hair. The first shock came to English from French, but despite some doubts expressed in dictionaries it is, very probably, of Germanic origin. Its base can be seen in Middle High German schocken ~ schucken “to swing” (Old High German skokka meant “swinging,” and its phonetically unexpected modern continuation is Schaukel “a swing”). Words that traveled from Germanic to Romance and back to Germanic are numerous. Shock2 and shock3 are almost certainly of Germanic descent. They seem to be part of a loosely related sk-family. Nothing is known about the origin of Engl. shuck “husk.” But I have always wondered how the exclamation shucks! (an expression of sham modesty or disappointment) came about. Exclamations of this type (at-a-boy!, oh boy, and so forth) sometimes preserve the oldest sense of an etymologically obscure word. Could the story begin with a dialectal verb shuck “to discard or dismiss contemptuously a worthless thing (by shaking it off),” a synonym or doublet of chuck “throw with the hand” (unless chuck is a doublet of shuck), with a later development to the noun “worthless thing” and ‘’the inedible part of the grain”? Joseph Wright’s English Dialect Dictionary cites (among others) shuck “shake; shirk; yawn, stretch.”
Engl. shack “fall, as grain at harvest; grain fallen from the ear and eaten by hogs” and shack “rove around, as a stroller or beggar; worthless person; truant” are on record. In this volatile part of the vocabulary, given shack (of course, unrelated to shack “hut”) and shock, the existence of shuck is easy to imagine. Old Icelandic skokkr meant “a loose board in a boat”—another disposable object? In the entry shackle, Skeat cited Swedish skakel “loose shaft of a carriage” (Old Engl. sceacul meant “a loose bond”) and connected them with shake. This etymology has been called into question on account of the words’ semantic incompatibility, but in such cases everything depends on what common denominator the researcher chooses. Perhaps it was “throw away with a shake of the hand; get rid of.” Our family swells. Even skate (as in a pair of skates), which cannot be separated from French échasse “stilt,” seems to belong to it too.
Someone pursuing such creeping etymologies encounters many pitfalls. More and more words are drawn into the net, phonetic correspondences become weaker, and underground semantic passages provide arbitrary links from one point to another. Shake, shock, shack, shuck, chuck, shog, jog…. In my forthcoming posts I will explore some other similar cases, but this is what can be said by way of general explanation. Some words are clearly and nobly related, for example, Engl. three and Latin tres. Others are more like mushrooms growing on the same stump: no roots but unquestionable affinity. Perhaps the historical linguistics of the future will deal not only with roots but also with spores.
The greatest master of creeping etymologies was Jost Trier, a distinguished twentieth-century scholar. He purported to show that many seemingly unrelated words can be traced to some basic concept or activity, for example, “foliage,” “underwood,” “love,” “work in a community,” or “wall building.” He had many admirers and as many detractors. Some of his solutions are not only ingenious but probably correct (I say probably, because unclear etymologies are bound to remain unclear). Others are fanciful. Neither following him all the way nor dismissing him out of hand should be recommended. Each of his proposals has to be evaluated for its merits and demerits. But, in principle, jumping from one look-alike to another poses numerous dangers. It is a fact that sk- and sk-k enter into words for springing and shaking in many languages. It is less clear what made those groups fit for designating strong movement. Be that as it my, perhaps my proposal won’t strike anyone as too foolhardy: shake, from skakan, means what is does for good reason, even though at the moment this reason escapes us.
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 him care of firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”