Oxford University Press's
Academic Insights for the Thinking World

Monthly Gleanings: May 2007


By Anatoly Liberman

Many thanks to our correspondents for questions and comments. First, the future of my etymological dictionary. Work on the dictionary and a bibliography of English etymology is going well, with occasional unexpected but pleasant complications. At the moment, I am mostly interested in words beginning with the letter B, but one day, while reading an old book on Dutch etymology, I understood where Engl. yeoman came from. I am sorry that yeoman begins with the second letter from the end rather than from the beginning of the alphabet, but the way of research are unpredictable: one often has to enjoy the dessert before an appetizer.

Second, an ironic response from Australia reminded me that I should not speak only about British and America English, because English is also a native language of people living in other climes. I agree and apologize but cannot change my ways. British versus American is a conventional division into two major groups. Otherwise, I would have to refer to Canadian, Australian, and South African English, let alone English in India and in several other countries. (Specialists invented the ugly word Englishes.)

Third, it was interesting to read the comments from Kent about the pronunciation of some place names: root’tm (Wrotham), mep’m (Meopham), and trosley (Trottiscliffe). The moment one realizes that Wrotham and Meopham are Wrot-ham and Meop-ham, the current pronunciation does not cause too much surprise (both names turned up in Daniel Jones’s English Pronouncing Dictionary, and, of course, in the form given in the letter), but trosley, with its cliff- and the whole middle section gone, is tricky. If I can be allowed to return the favor, I wish to inform our correspondent that I have recently spoken to my students about the lathes of Kent and compared them with the rapes of Sussex and the ridings of Yorkshire. Lathe “administrative district” is a word hardly anyone except historians knows here. Its oldest meaning is, most likely, “landed property.” The letter to the blog ended with a characteristic statement: “I remember feeling a sense of outrage when the council decided to erect a sign using the ‘new’ spelling and felt that it smacked of laziness.” This is the usual reaction of all literate people.

Finally, I have nothing to say about the connection between loo (the theme of my recent post) and toodle-oo. My greatest fear is that one day somebody will definitively trace loo is German lu-lu “piss-piss,” though I have no idea how such a feat can be accomplished. Synonyms are countless in this area (nessy, netty, biffy, boggart ~ bog ~ bogs, and what not); some of them baffle etymologists no less that loo.

As on previous occasions, I will incorporate into these gleanings several questions I received during my talk show on MPR (the latest one took place in May). The main topic was hyphenated words of the shilly-shally and nitty-gritty type. (Next time the focus will be on words combining opposite meanings like Latin altus “high”/“deep” and Engl. fast “moving quickly”/“staying in one place”). Fuddy-duddy and British dialectal (Cumberland) duddy-fuddiel “ragged fellow” (duddy-fuddiel is cited in the OED). All the dictionaries I have consulted follow the Supplement to the OED and say “origin unknown,” but the solution may be not far to seek. In northern British dialects and Scots, fud means “buttocks” or “scut,” that is, “short, erect tail of a hare, rabbit, etc.” Thus, fuddy, like ars(e)y, is devoid of polite overtones. Nor does dud evoke pleasant associations, so that fuddy-duddy turns out to be a contemptible thing or person. When such a word is coined, it usually has a wide range of meanings (simply a term of abuse), so that it will fit equally well an old fogy and a tatterdemalion. Transposing the parts changes neither its general sense nor its overall shape. My database contains only one discussion of fuddy-duddy in an article. There may be some overkill in the author’s conclusions, but what he (John P. Hughes, 1954) says is so interesting and amusing that I will quote him at length: “From Roger nicknames we have a ‘Doddie’ for a dehorned cow or bull (whence to dod) and a hodmandod or dodman, which is a snail or its empty shell, or a scarecrow, or a clumsy or stupid person. Either of these may be responsible for dud, hoddy-doddy, dotty (in the sense of ‘foolish’; the nickname is from Dor’t’y), and, doubtless [!] fuddy-duddy (though it has been suggested, probably with more wit than science, that the latter American colloquialism designates a holder of the Ph.D. and D.D.).” Hughes also derives dodder and dither from the same “root.” (A parenthetical remark: beware of the word doubtless in any argument, for it is used only when things are doubtful.)

Pom-pom is sound imitative, like boom-boom and pooh-pooh, to cite two other onomatopoeias rendering the sound of firearms. For dilly-dally see my post on dildo. Oaky-dokey looks like a humorous rhyming jingle based on OK. Willy-nilly goes back to the phrase wil I nil I “I am willing, I am unwilling” (hence the meaning “whether one likes it or not”). Nil is a contraction of will with a negation (Old English already had the form nyllan). Hanky- panky is obscure. The British dialectal word (Staffordshire) ‘anky-pranky has the same first element, but the second is either a folk etymological alteration of panky or a distant relative of the 16th-century princum-prancum “trick.” Although, according to the OED, hanky-panky is based on hocus-pocus ~ hoky-poky, the similarity is slight. The influence of hand (as in sleight of hand; Ernst Weekley’s suggestion) is unlikely. I have been able to formulate the following tendency: when we deal with a reduplicating word whose first component begins with an h, the determining part of the compound is usually the second one. For example, discover the origin of mugger and add hugger to it. Therefore (assuming that my rule is reliable), if we could trace panky to some plausible etymon (source), the riddle would be solved. Punk, a loanword from Scandinavian—it originally meant “trash,” and in 16th-century English it meant “prostitute”–is relatively close, but the connections between it and panky, as The Oxford Dictionary of English Etymology says in such cases, “cannot be made out.”

The beginnings of topsy-turvy (153) are also lost, but topsy must have been derived from top-s. This compound was around as early as 1530. The earliest spellings are topsy tervy and topsy tyrvy, but nothing follows from this fact since all three spellings (with u, e, and y) reflect the same pronunciation (compare Modern Engl. hurl, sterling, and girl, with y being a gratuitous variant of i, as in Smith ~ Smyth). The somewhat later forms topset and topside need no be the original ones: they look more like a folk etymological alteration of up-so-down, which preceded upside down. Then there were the phrases toppe over terve and topside-turfway. Middle English had the verb terve(n) “roll over, wallow” (which continued as turf into the Elizabethan epoch and is current as turve in the modern eastern and southeastern British dialects), a plausible etymon of turvy. Conjectures about the origin of topsy-turvy, including quite a few absurd ones, are numerous, for example, “[t]the expression is derived from a way in which turf for fuel is placed to dry on its being cut.” Especially long-lived was the derivation of topsy-turvy from the nonexistent phrase topsi’ to’er way (a characteristic note going back to the end of 1852: “I have always understood this to be a corruption of ‘topside t’other way’, and I still think so”; times change, but homegrown etymologists do not).

Flim-flam “nonsense,” another 16th-century word, poses the familiar difficulty: Do we have to explain the derivation of only one element or of both? Knowing the history of one part is, in principle, sufficient. For example, if flim has been attested, flam could have been added according to the formula: first a syllable with a closed vowel, then a syllable with an open one, as in tit for tat and shilly-shally. Both flim and flam existed: flam meant “mock,” and flim is an Old Icelandic word for “lampoon, libel” (a legal term, mainly applied to mocking verses). The two words were related. Flim, if it had any currency in English, may have engendered flam or the other way around (flam has been attested). In the topsy-turvy world of reduplicating compounds, everything is possible. So uncertain is the material before us that I cannot think of a better plan than finishing my story with a few lines about wishy-washy, an 18th-century word (as far as we can judge). Its base is washy, but it was a happy idea to add wishy to it. The initial meaning of wishy-washy was “weak and insipid,” but today the word suggests the blurry contours of the quality it describes and contains a hint of an unfulfilled desire (“lacking in strength and purpose; indecisive”).

Questions and comments sent during the summer months are welcome. Read the next gleanings on July 25.

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 [email protected]; he’ll do his best to avoid responding with “origin unknown.”

Recent Comments

  1. John Cowan

    Altus does not combine two contradictory meanings: its unitary meaning is ‘large in the vertical dimension’, which in English we call high if we are conceptually at the bottom, as in high mountain, or deep if we are conceptually at the top, as in deep valley.

  2. […] an old post devoted to the role of luck and serendipity in etymological research, I mentioned the word yeoman, […]

Comments are closed.