By Anatoly Liberman
As always, I am grateful for comments. So far, I have received the most responses to my post on the death of the adverb. One of our correspondents notes that students in creative writing tend to put adverbs at the beginning of the sentence, as in “Happily, she met her boyfriend at the mall.” Sorrowfully, I have also noticed this mannerism, and not only in students’ stories. The adverb, nearly wiped out by morphology in American English, is doing “just fine” in syntax. For instance, the authors of scholarly publications love the word undoubtedly; they use it when arguments are weak and doubt exists. Obviously, certainly, and definitely serve the same purpose. Actually has become the bane of our life.
Stephen Goranson, to whom I owe numerous valuable suggestions, points out that punk turned up in an 1819 text (I was unaware of any 19-th century citations of this word). The citation confirms my belief that for a long time punk was dormant in the language of the underworld before it reemerged in the 20th century (for how could it be otherwise?). Another correspondent sent us his observations on gig: “Gig ‘the continuous line from trouser fly to shirt collar’—first heard this one in 1984 in San Diego where I was in ‘Boot Camp’ for the US Navy. Common usage in boot camp is ‘Sailor! Square away that gig line.’ The intended result is that the person addressed will check in a mirror and ensure that the trouser fly, belt buckle, shirt placket, and tie are in a vertical line. Also from my Navy days, gig in the sense of being caught out on some infraction of the rules. I was gigged for having dirt on my cover (hat).” In discussing the origin of lollygag, I noted that words beginning with and ending in the same consonant (gag, dud, bib), by their form, conjure up images of things lacking dignity. The following slangy meanings of gig have been recorded (in addition to those listed in my previous post): “vagina; rectum” (up your gig) and “trick, swindle.” In The Random House Dictionary of American Slang, I find “state of affairs” and one of the senses known to our correspondent, namely “demerit; to place on disciplinary report.” But “continuous line” is not given there. Is this sense a gross extension of “rectum”? In any case, with gig “vagina” and “demerit” we are far from “revolving object.” Gig is a sound complex that allows the speaker to endow it with almost any meaning. The Latin for “giant” was gigas. Was it not a baby word, the name of a Roman boogie? Nurses perhaps said to little children: “Sleep or the gigas will come and snatch you” (and it was a big gigas, for a in the second syllable was long, as in Engl. father). The child would fall asleep, and the nurse would giggle in satisfaction and go about her business. Latin Titan (pronounced teetahn) is another suspicious reduplication of the same type.
Latin and Pseudo-Latin
Mr. Goranson has also asked me to return to the vexing etymology of element. In the older set of gleanings, I said only that the origin of Latin elementum had not been discovered, that its derivation from L-M-N had been abandoned, and that modern researchers look on this word as a borrowing from some unknown language. But more or less by chance I have my own ideas on this problem. Not being a Classical scholar, I realize the danger of encroaching on the turf of my neighbors in the area of philology and have to explain why I dare do so. My area of expertise is Germanic, and for many years I have been studying the literature on the origin of the words book (it is unclear whether it has anything to do with beech and the wooden tablets the old speakers of Germanic used for painting or scratching letters) and rune. In connection with book and rune, I read many works on the invention of alphabets and became aware of elementum.
The earliest meaning of Latin elementum was “letter”; the word translated Greek stoikheion “sequence; step; ground, base.” My database features 18 articles and reviews devoted to element, 16 of them written before 1901. No new conjectures on the origin of this word seem to have been offered in the 20th century. In the past, etymologists kept returning to two suggestions: either elementum is connected with Latin alo “nourish” (we see it in the root of Engl. alimentary) and its problematic cognate olo or it is a word made up of the letters l, m, n. (I will leave out a host of fanciful guesses.) However, a does not alternate with e in Latin, while olo has not been attested (the alternation e ~ o would be trivial: compare bene ~ bonum or toga from tegere “cover”). It is easy to see that the Roman alphabet falls into two parts: the a-b-c part and the l-m-n part. Yet no one could explain why l, m, n, rather than a, b, c, were chosen for coining the word el-em-en-tum. By contrast, abecedarium was a common name for an abc-book.
I believe that both “camps” were partly right. It is true that the verb olo had no independent existence, but—for the reasons that need not bother us here—after prefixes, alo changed to olo. The best-known example is Latin adolesco “grow” (one is nourished and, as a result, grows), from which we have adolescent. The prefixed verbs were so common that the Romans extracted, or abstracted, olo from them, and this illegitimate verb came into being; olesco “I grow” has been recorded. Alimenta (neuter plural) meant “food, sustenance”; in everyday speech, it may have had the variant *olimenta (the asterisk designates unattested, reconstructed forms). The alphabet is a rigid system in which any element generates, “nourishes” the next one and lets the sequence “grow.” This is why we “know something from A to Z” and believe that “if you said A, you must say B.” The many fruitless attempts to trace elementum to its putative Indo-European source or to discover its Greek cognates have shown that this word has no prehistory. To put it differently, elementum was not derived from an identifiable root with the help of a suffix but coined whole to gloss Greek stoikheon. Its structure is unusual in that the second e violates the rules of Latin phonetics. Some grammarian must have taken alimenta ~ *olimenta and changed it into elementa (the singular elementum postdates the plural). But even if olimenta did not exist, alimenta could have been used for the same purpose, for ali– would have been understood as a variant of oli– anyway. The new coinage could not but be an immediate success. First, since o and e alternated in a natural way, the path from *olimentum to elementum was short and everybody understood the joke: a noun appeared that designated a “self-nourishing sequence,” and this is what an alphabet is; second, the word, although recognizable as Latin, had a slightly odd phonetic shape (a feature typical of slang) but was easy to remember; and it turned out to be an elegant word, with three short e’s in a row; finally, it contained a pun on l, m, n. Later, elementum came to designate the basic unit of anything, not only of the alphabet. Thus, elementum is indeed connected with alo ~ *olo, but not derived from them in a regular way, and it is an artificial word alluding to but not made up of l, m, n. The idea that elementum is a loan from an unknown language takes us nowhere. Among other things, it does not explain why such a useful term of grammar had to be borrowed, and borrowed late, from some mysterious neighbor.
Now comes a piece of supporting evidence. I believe that not only the Romans but also those who came into contact with them realized that elementa meant something like “self-generating sequence.” At the time elementa surfaced in Latin the Scandinavians began using the Runic alphabet. Its origin remains a riddle, but, most probably, the runes were brought to the north from the Mediterranean basin. Rune masters held both the script and the alphabet in high esteem. The alphabet appears in the same order in several inscriptions. This order is amazing: the first six letters were F, U, TH, A, R, K (there was a special sign for TH), and modern scholars call it futhark—a word like QWERTY for the keyboard on our typewriters and now computers. But qwer means nothing, while futh means… “vulva.” No one could miss this fact. Imagine our alphabet beginning so: C, U, N, T, A, B. I suppose that those who learned the art of writing from their southern teachers were informed why the Romans called their alphabet elementa and adapted the idea to their needs in the most striking way. They prefaced the sequence with a word meaning “vulva” (for what can generate anything better than the futh?) and then began the alphabet in good earnest, with A. I am sure that serious scholars will ridicule my etymology, but I don’t begrudge them their happy hour and will be the first to join in the amusement. Laughing at oneself, they say, is the subtlest kind of humor.
What is the origin of the conjuring formula hocus pocus, the gibberish repeated by jugglers all over Europe when they perform their tricks? Something unites elementum and hocus pocus: both are late coinages, though the first is genuine Latin, whereas the other only pretends to be such. Hocus pocus was first recorded in England, in a slightly different form, in 1624. From England it spread to the Netherlands and Germany and allegedly to other countries. Whatever its origin, hocus pocus belongs with words like helter-skelter and higgledy-piggledy. As far as I can judge, in reduplications beginning with an h, the base word is usually the second, while the first is added only to provide the rhyme and therefore has no etymology of its own. However, this regularity is far from being law. The German variants of hocus pocus were at one time ox box, hogges und pogges, and okos bokos. Those who sought for the solution in the second element cited Latin poculum “bowl, chalice,” English poke, and the name Puck. The names of other demons and a few Welsh words have also been pressed into service. Hensleigh Wedgwood, the most influential etymologist before Skeat, thought of huk, puk, an onomatopoeic group, and noted the similarity between hocus and Old Engl. hux (or husc) “deception,” a probable but not certain etymon (source) of hoax. Reference to poke makes sense, for conjurers carried bags with the tools of their art (rabbits, as we know, are also produced out of the bag). The most authoritative modern dictionaries follow a foundational work “on the language of gods and ghosts” and suggest that hocus pocus is an alternation of hax pax max Deus adimax, a 16th-century pseudo-Latin magical formula coined by vagrant students. Hax pax max was written on wafers to avert fever and other diseases. It is unclear to what extent that rigmarole was able to engender the conjuring formula, and the difference between English o and a in hax, etc. is usually explained away too quickly.
John Tillotson, the archbishop of Canterbury, said: “In all probability those common juggling words of hocus-pocus are nothing else but a corruption of hoc est corpus [“this is (my) body”], by way of ridiculous imitation of the priests of the Church of Rome in their trick of Transubstantiation.” His etymology has been repeated and contested many times. Such a jeer by street jugglers, who must have tried to please as many spectators as possible, seems unlikely. Also, the words hoc est corpus are usually pronounced slowly and do not sound like hocus pocus. However, the Scandinavian variant of the formula hokus-pokus filiotus (so Ernest Weekley) does make one think of Latin filius “son” and, consequently, of a blasphemous perversion of the sacramental formula, but in Dutch the concluding words were Pilatus pas. Although here reference to a religious context is unmistakable, mention of Pilate would not have offended anyone on the streets. Hocus pocus resembles hotchpotch and the obsolete conjuring formula hicius docius. It was probably coined as a nonsense word. Charms, curses, and all kinds of incantations depend for their effect on obscure language, and a juggler would do everything in his power to confuse his public. Several formulas were current at the same and must have influenced one another. Nor did they gain in clarity as time went on.
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.”