Why is there no “master key” to the closet hiding the origin of language and all the oldest words?
Historians deal with documents or, when no documents have been preserved, with oral tradition, which may or may not be reliable. The earliest epoch did not leave us any documents pertaining to the origin of language. We cannot even agree on what system of signals constitutes language. The “first” words are lost, and only the oldest recorded words have come down to us. Those are late, and the method of reconstructing their past depends on the state of the art and the researcher’s ingenuity. Some “primitive” words we know are simple (bow–wow, hush, buzz); others are short but not very simple (bad, put, big, dud, dude, eat; if, and). The mechanisms of word coining have hardly changed since the beginning of time. Consequently, though words come and go, the earliest products of language creativity may have been supplanted by similar or even identical formations. Yet each word needs an individual approach. The desired near-universal master key, when proposed, always turns out to be an illusion. This is especially true when someone comes up with a list of a limited number of roots from which all words allegedly sprang up or with one language as the source of all others.
The origin of Engl. stir “prison”
There is not much I can add to what is universally known about the etymology of this word. Let me only say that words for “prison” are often borrowings from foreign languages. Such are both Engl. prison and jail (or gaol as, for example, in Oscar Wilde’s The Ballad of Reading Gaol). Conversely, many are slang, and their origin is, predictably, obscure. Eric Partridge once remarked that a lot of nonsense had been written about the derivation of stir, but the literature on this word is poor and only two hypotheses on it compete, or rather competed in the past. The earliest citation of stir (1851) in a printed text goes back to Henry Mayhew, whose description of London life has lost none of its value. The word was noticed almost at once. The first item in my database on stir is dated to 1852 (in the indispensable bimonthly Notes and Queries). It soon made its way into all the dictionaries of British slang.
Monosyllabic synonyms for prison are rather numerous: compare jug, nick, can, pen, and the unexpectedly elegant Latin quod. This complicates our search. By an incredible coincidence, one of the Old English words glossed as “prison” was stȳr, a variant of stēor (both with long vowels). Yet the speakers of the oldest Germanic languages had no prisons. Crime was punishable by fines, mutilation, and outlawry. Engl. prison and gaol appeared in the language in the twelfth and the thirteenth century respectively. The New Testament, with its Roman background, often mentions prisons, and Bishop Wulfila, who in the fourth century translated the Bible into Gothic, used karkara (Latin carcer; compare Engl. incarcerate). He could find no native word for such an institution in his language. Old Engl. stēor, which has the root of Modern Engl. steer, meant “steering one toward a certain goal (not gaol!), guidance, discipline, correction; penalty, fine, punishment” and therefore “prison,” but it did not refer to the prototype of our jail. The idea that stir existed in the lingo of the underworld for a thousand years and surfaced only in the twentieth century is hard to accept.
The other derivation looks more realistic. It traces stir to some Romany word for “prison,” such as stariben or stardo. The fact that nearly the same word is known in Spanish lends credence to this etymology. As usual, “proof” is wanting, but etymological propositions can seldom, if ever, be “proved.” We are at best able to assess the degree of the verisimilitude of this or that hypothesis. For the moment, the Romany origin of Engl. stir is the best we have, and some dictionaries have accepted it, with the traditional hedging (perhaps, probably, and so forth). According to Eric Partridge, The Start, The Old Start “Newgate Prison” in London got its name from stir. I was unable to check the date of this name and will be grateful for a comment from an expert on the history of London. Since stir was known in 1851, it must have appeared in the language earlier, but we don’t know how much earlier. In any case, Bampfylde-Moore-Carew, the author of a book of his “Adventures” (1755), did not include it in his sizable description of “the Gipsies (sic) and their language.”
The way from tree to true and trough and the community of sn-words
The way is devious and may not lead us to the desired answer. The great Indo-European scholar Hermann Osthoff (1847-1909) wrote an article titled “Eiche und Treue” (“The Oak and Loyalty”), in which he attempted to show that the idea underlying the word truth is the immobility of a mighty tree (“tree” = “steady and constant”). Modern etymologists treat this connection without enthusiasm, but its appeal is obvious. The ancient root of tree is dru– “wood” (as in Sanskrit). Trough, an object made of wood, has the same root. I have a dim recollection that I have once discussed the tree-true link but cannot remember when. The same correspondent asked me about sn-words (snow, sneeze, snicker, sniff, snuffle, and snot) designating moist and wet things (snicker does not quite belong here). See the posts for May 1 and May 8, 2019 on sn-. Not improbably, snow also meant “wet and sticky substance.” Yet elsewhere, this word’s cognates occur without s- (this is an example of s-mobile, often referred to in this blog), and, if the “primitive” root began with n-, the association with wetness may be secondary.
Our correspondent asked me to publish my list of reformed spellings. I don’t have such a list (The Spelling Society does), but I have some ideas. Get rid of the letter q (replace it with kw) and of y, at least in the middle of words. Abolish some mute consonants. (I am sure someone pronounces k-thonic, but this pronunciation is a tribute to academic pedantry. Likewise, quite a few British specialists say my-thology and my-thic, because the Greek vowel in this word is long. Knock and gnaw will do well without k- and g-, and acquaintance will do equally well in the form akwaintens. Rough will lose none of its roughness if spelled as ruf ~ ruff (compare rub and gruff). Many double consonants are also a nuisance. Think of simetri instead of symmetry (compare cemetery). But my wants are few: I am ready to accept the mildest version of the reform, only to see the stone rolling. Yet for now I’d leave a few of the worst high-frequency offenders (you, have, and give, among them) intact.
Odds and ends
Eeny-meeny. I’ll be ready to discuss it with my opponent after he has read the entry on this “word” in my etymological dictionary.
Bazaar among the b-z words. It would be nice to explain bazaar as a busy place with a lot of bizarre things on display and everybody abuzz. But I am afraid of losing my way in the Oriental morass; too many languages are involved in the history of bazaar, and it seems that s, rather than z, was in the middle of the ancient form. Perhaps specialists in that branch of linguistics will help us here.
Greater royalist than the king. Dismal phrases like to only come once are now ubiquitous, but watch the heroic splitter from The Washington Post: “Amy R now… calling to not to follow the will of voters or making baseless allegations of fraud….” The writer also rather inelegantly mixes infinitives and gerunds. (The moral: don’t split in haste and reread what you write.)
Feature image in the public domain from pxfuel.