By Anatoly Liberman
All modern dictionaries state that the adverb ajar goes back to the phrase on char, literally “on the turn” (= “in the act of turning”). This is, most probably, a correct derivation. However, such unanimity among even the most authoritative recent sources should be taken with caution because reference books tend to copy from one another. Recycling a plausible opinion again and again produces an illusion of solidity in an area notorious for debatable results. That is why it is so interesting to read books published before Skeat’s dictionary (1882) and the OED came out. After their appearance, the lines of English etymological research hardened and few people took the trouble of questioning the giants’ conclusions.
First, let us look at the accepted explanation. The etymon of ajar is said to be on char (1510; cf. at char, 1708; Jonathan Swift). Char(e) is related to Old Engl. cierran “to turn.” Its root can be detected in charwoman and chore. Today we remember the latter word only with the sense “an unpleasant task” (“a job of work”), and it usually occurs in the plural (chores). Its meaning accords well with that of charwoman, even though the cause of the variation char ~ chore remains unclear (but see below).
In its present form ajar surfaced only in 1786. For this reason, it doesn’t occur in pre-nineteenth-century dictionaries; it is absent even from the first editions of Webster’s (1828). Later editions listed the word without any derivation, but in 1864 C.A. F. Mahn revised Webster’s etymologies, explained ajar as a + jar (a self-evident move), and cited its Dutch synonym akerre (sic), but he also referred to jar “harsh sound.” In his opinion, the connecting link between “harsh sound” and “partly open” was provided by Shakespeare’s usage: jar “a vibration of the pendulum of a clock.” However, jar “turn” and jar “vibrate” are different words, and if ajar is related to Dutch op een kier (this is the modern gloss for ajar), it has nothing to do with a discordant noise. To be sure, when a door is ajar it often vibrates (and while vibrating creaks) and turns on its hinges, but only one of the two verbs, namely “to turn,” is akin to Old Engl. cierran, Dutch keren, and German kehren. The other jar is a sound imitative verb (or so it seems).
Along the way from on char to ajar, ch must have undergone voicing. James A. H. Murray, while editing the letter A for the first edition of the OED, suggested that the association with ajar “in a jarring state, out of harmony, at odds” caused the confusion. In a way, it is a variation on Mahn’s idea. The confusion is quite possible, the more so as the Holstein phrase in de Kirr indicates both being not tightly closed and creaking (and see what is said above), but the story of a door at war with itself does not end here.
Later scholars tried to find a phonetic analog of the change. According to a well-known rule, s sometimes alternates with z, depending on the place of stress in a word; compare ’exhibition (ks) ~ex’hibit (gz), ’execute (ks) ~ ex’ecutor (gz), and so forth. In such examples (note the value of x in the word example), s acquires voice when it stands after an unstressed syllable. It has been surmised that in the string on char, in which ch also stands after an unstressed syllable, it yielded j for the same reason. Even though events may have developed this way, the question remains open or, if I may take such liberties, ajar. Initial ch– hardly ever became j– in the history of English. The only two instances may be jowl in cheek by jowl and jaw; neither is certain. To complicate matters, the local variant jarwoman “an occasional assistant in the kitchen” exists. One should also bear in mind that Engl. j– often has a strong expressive value in both word initial and word final position, as in jig, jog, job, budge, nudge, grudge, and many others. To account for its emergence, one does not always need a phonetic law.
I would like to point to an amazing number of similar sounding synonyms of ajar in the Germanic languages. The closest of them is Scots ajee (also spelled agee). It means the same as ajar but has a different root. Gee (jee) is a call to a horse. It also functions as a verb and means (at least in some British dialects) “to move (to one side), stir, alter one’s position,” as in fowls go gee (or chee, as it was generally pronounced in Kent!) when they go to roost; my evidence is dated to 1875. Agee is less limited in its use than ajar, for in Scots one can also look “agye,” that is, aside, while in English only doors and windows can be (or stand) ajar.
Then there is British dialectal ashore (with the variants ashard and ashare) with early antecedents. In reference to a door it means “ajar” (this ashore isn’t allied to ashore “on a shore”). If you voice sh– in ashore, you will get ajore, a form not too different from ajar, especially in light of the fact that Engl. char alternates with chore. In Low (northern) German, we encounter enkarich, corresponding to Dutch aankerre (their southern German cognate is achar), while Frisian has in ‘t kier. As we have seen, the Old English verb related to char was kierran. Middle English k regularly became ch before i. Once again try to voice ch in the German, Dutch, and Frisian forms cited above and you will get something like English dialectal ashore with j in place of ch. This ashore may be related to Modern Engl. askew (a word of French origin). In any case, it isn’t a cognate of ajar “on the turn.” Of the several Icelandic phrases corresponding to ajar I’ll mention only á gaur, with its variant á gaul. Kentish chee ~ gee, Engl. dialectal jarwoman, and the Scandinavian g-forms weaken the idea that jar for char in ajar is the product of final stress.
As long as we analyze the words featured here one by one, each seems to have an acceptable etymology, but the group in its entirety gives us pause. Is it possible that on char (which must have passed through the stage achar before becoming ajar), agee (ajee), ashore (ashare), and á gaur are quite unconnected? And whence this fixation on half-open doors? Outside Germanic, it is hard to find even one word with such a meaning. Why was it so important to have a multitude of synonyms for this situation? Who used them with such insistence? Perhaps, at least initially, the word(s) referred to doors that refused to close well, and phrases with the senses “askew, aslant, on the turn, on the go, off the beam,” to mention some of them, served the purpose well. In English, such doors were also “out of harmony” and this circumstance may have contributed to the survival of ajar. Carpenters were employed to make repairs, and the term for “half-open” (or rather, “not closing properly”) became part of their lingua franca, a common language of wandering handymen. The initial term may have been coined in Germany or elsewhere and, once adopted, changed in every land by folk etymology.
I realize how shaky my hypothesis is, but, in my opinion, it is almost unbelievable that so many look-alikes designating a marginal state of a door and a window should have sprung up independently and acquired a similar form. Even if what I suggested here cannot be accepted as the basis of a viable etymology, people may start thinking anew about the origin of the enigmatic English adverb ajar. (The alliteration at the end of the last sentence — all the words there begin with a vowel — is intentional. I needed a final flourish for a poetic reinforcement of a crazy idea.)
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 [email protected]; he’ll do his best to avoid responding with “origin unknown.”
Image credit: Vermeer’s Young Woman with a Water Pitcher, ca. 1662. Metropolitan Museum of Art, New York. Photo by dmadeo via Wikimedia Commons.