The Romance of Words (4th ed.)

he derives from Parma, which has given its name to parmesan cheese. On the word cockney[144] he waxes anecdotic, always a fatal thing in an etymologist—

Molière often makes fun of the etymologists of his time and has rather unfairly caricatured, as Vadius in Les Femmes savantes, the great scholar Gilles Ménage, whose Dictionnaire étymologique, published in 1650, was long a standard work. Molière's mockery and the fantastic nature of some of Ménage's etymologies have combined to make him a butt for the ignorant, but it may be doubted whether any modern scholar, using the same implements, could have done better work. For Ménage the one source of the Romance languages was classical Latin, and every word had to be traced to a Latin word of suitable form or sense. Thus Fr. haricot[145] is connected by him with Lat. faba, a bean, via the conjectural "forms" *fabarius, *fabaricus, *fabaricotus, *faricotus, *haricotus, a method to which no problem is insoluble.[146] He suggests that Fr. geindre, or gindre,[147] baker's man, comes from Lat. gener, son-in-law, because the baker's man always marries the baker's daughter; but this practice, common though it may be, is not of sufficiently unfailing regularity to constitute a philological law. Perhaps his greatest achievement was the derivation of Span. alfana,[148] a mare, from Lat. equus, a horse, which inspired a well-known epigram—

These examples show that respect for Ménage need not prevent his work from being a source of innocent merriment. But the above epigram loses some of its point for modern philologists, to whom equations that look equally fantastic, e.g. Eng. wheel and Gk. κύκλος,[149] are matters of elementary knowledge. On the other hand, a close resemblance between words of languages that are not nearly related is proof presumptive, and almost positive, that the words are quite unconnected. The resemblance between Eng. nut and Ger. Nuss is the resemblance of first cousins, but the resemblance of both to Lat. nux is accidental. Even in the case of languages that are near akin, it is not safe to jump to conclusions. The Greek cousin of Lat. deus is not θεός, God, but Ζεύς, Jupiter.

An etymology that has anything to do with a person or an anecdote is to be regarded with suspicion. For both we want contemporary evidence, and, in the case of an anecdote, we never, to the best of my knowledge, get it. In Chapter III. are a number of instances of words formed according to authentic evidence from names of persons. But the old-fashioned etymologist will not be denied his little story. Thus, in explanation of spencer (p. 40), I find in a manual of popular information of the last century,[150] that—

This is what Pooh-Bah calls "corroborative detail intended to give artistic verisimilitude to a bald and unconvincing narrative." From the same authority we learn that—

and that—

The history of chouse exemplifies the same tendency. There is no doubt that it comes from a Turkish word meaning interpreter, spelt chaus in Hakluyt and chiaus by Ben Jonson. The borrowing is parallel to that of cozen (p. 110), interpreters having a reputation little superior to that of horse-dealers. But a century and a half after the introduction of the word we come across a circumstantial story of a Turkish chiaus who swindled some London merchants of a large sum in 1609, the year before Jonson used the word in the Alchemist. "Corroborative detail" again. The story may be true, but there is not an atom of evidence for it, and Skinner, who suggests the correct derivation in his Etymologicon (1671), does not mention it. Until contemporary evidence is adduced, the story must be regarded as one of those fables which have been invented in dozens by early etymologists, and which are perpetuated in popular works of reference. It is an article of faith in Yorkshire that the coarse material called mungo owes its name to the inventor of the machine used in its fabrication, who, when it stuck at a first trial, exclaimed with resolution, "It mun go."

Many stories have been composed après coup to explain the American hoodlum and the Australian larrikin, which are both older than our hooligan (see p. 12). The origin of hoodlum is quite obscure. The story believed in Australia with regard to larrikin is that an Irish policeman, giving evidence of the arrest of a rough, explained that the accused was a-larrikin' (larking) in the street, and this was misunderstood by a reporter. But there appears to be not the slightest foundation for this story. The word is perhaps a diminutive of the common Irish name Larry, also immortalised in the stirring ballad—

As I write, there is a correspondence going on in the Nottingham papers as to the origin of the nickname Bendigo, borne by a local bruiser and evangelist. According to one account, he was one of triplets, whom a jocular friend of the family nicknamed Shadrach, Meschach, and Abed-Nego, the last of which was the future celebrity. It is at any rate certain that his first challenge (Bell's Life, 1835) was signed "Abed-Nego of Nottingham." The rival theory is that, when he was playing in the streets and his father appeared in the offing, his companions used to warn him by crying "Bendy go!" This theory disregards the assertion of the "oldest inhabitant" that the great man was never called Bendy, and the fact, familiar to any observer of the local dialect, that, even if he had been so called, the form of warning would have been, "Look aht, Bendy, yer daddy's a-coomen."

In the Supplement to Littré there is an article on domino, in which he points out that investigation must start from the phrase faire domino (see p. 102). He also quotes an absurd anecdote from a local magazine, which professes to come from a "vieille chronique." Littré naturally wants to know what chronicle. In Scheler's Dictionnaire étymologique (Brussels, 1888), it is "proved," by means of the same story elaborated, "que c'est là la véritable origine du mot dont nous parlons."

In Brewer's Dictionary of Phrase and Fable, s.v. sirloin, we read that "it is generally said that James I. or Charles II. knighted the loin of beef, but Henry VIII. had done so already." This sounds like a determination to get at the root of things, but does not go far enough. The word is found in the 15th century, and Fr. surlonge, from which it comes, in the 14th. It is compounded of sur, over, and longe, a derivative of Lat. lumbus, loin. The belief in the knightly origin of the sirloin was so strong that we find it playfully called the baronet (Tom Jones, iv. 10). Hence, no doubt, the name baron of beef for the double sirloin. Tram is persistently connected with a Mr Outram, who flourished about 1800. This is another case of intelligent anticipation, for the word is found in 1555. It means log or beam, and was probably first applied to a log-road laid across bad ground, what is called in America a "corduroy" road. On the other hand, the obvious and simple derivation of beef-eater, i.e. a man who is in the enviable position of being sure of his daily allowance,[153] has been obscured by the invention of an imaginary Fr. *beaufetier, waiter at the side-board. Professor Skeat attributes the success of this myth to its inclusion in Mrs Markham's History of England. But the most indestructible of all these superstitions is connected with the word cabal. It comes from a Hebrew word meaning hidden mystery, and is found in the chief Romance languages. The word is of frequent occurrence in English long before the date of Charles II.'s acrostic ministry,[154] though its modern meaning has naturally been affected by this historic connection.

Even anecdotic etymologies accepted by the most cautious modern authorities do not always inspire complete confidence. Martinet is supposed to come from the name of a well-known French officer who re-organised the French infantry about 1670. But we find it used by Wycherley in 1676, about forty years before Martinet's death. Moreover this application of the name is unknown in French, which has, however, a word martinet meaning a kind of cat-o'-nine-tails. In English martinet means the leech-line of a sail, hence, possibly, rope's end, and Wycherley applies the term to a brutal sea-captain. The most renowned of carriers is probably Hobson, of Cambridge. He was sung by Milton, and bequeathed to the town Hobson's conduit which cleanses the Cambridge gutters. To him is also ascribed the phrase Hobson's choice, from his custom of refusing to let out his horses except in strict rotation. But we find a merchant venturer, living in Japan, using "Hodgson's choice" fourteen years before the carrier left this world and became a legendary figure—

The most obvious etymology needs to be proved up to the hilt, and the process is rich in surprises. Cambridge appears to be the bridge over the Cam. But the river's older name, which it preserves above the town, is the Granta, and Bede calls the town itself Grantacester. Camden, in his Britannia (trad. Holland, 1637), notes that the county was called "in the English Saxon" Grentbrigseyre, and comments on the double name of the river. Nor can he "easily beleeve that Grant was turned into Cam; for this might seeme a deflexion some what too hardly streined, wherein all the letters but one are quite swallowed up." Grantabrigge became, by dissimilation (see p. 57), Gantabrigge, Cantabrigge (cf. Cantab), Cantbrigge, and, by assimilation (see p. 56), Cambridge, the river being rechristened from the name of the town.

A beggar is not etymologically one who begs, or a cadger one who cadges. In each case the verb is evolved from the noun. About the year 1200 Lambert le Bègue, the Stammerer, is said to have founded a religious order in Belgium. The monks were called after him in medieval Latin beghardi and the nuns beghinæ. The Old Fr. begard passed into Anglo-French with the meaning of mendicant and gave our beggar. From béguine we get biggin, a sort of cap—

Cadger, or rather its Scottish form cadgear, a pedlar, occurs about one hundred and fifty years earlier than the verb to cadge. We find, noted as foreign words, in 16th-century Dutch, the words cagie, a basket carried on the back, and cagiaerd, one who carries such a basket. These must be of French origin, and come, like the obsolete Eng. cadge,[155] a panier, from cage, for the history of which see p. 109. Cadger is used in Scottish of an itinerant fish merchant with his goods carried in paniers by a pony—

Tobacco does not take its name from the island of Tobago, but from the native name of the tube through which the Caribs smoked it.

The traditional derivation of vaunt is from Fr. vanter, and this from a late Lat. vanitare, to talk emptily, used by St Augustine. This looks very simple, but the real history of these words is most complicated. In Mid. English we regularly find avaunt, which comes from Old Fr. avanter, to put forward, from avant, before. This gets mixed up during the Tudor period with another vaunt from Fr. vanter, to extol, the derivation of which can only be settled when its earliest form is ascertained. At present we find venter as early as vanter, and this would represent Lat. venditare (frequentative of vendere, to sell), to push one's goods, "to do anything before men to set forth himselfe and have a prayse; to vaunt; to crake; to brag" (Cooper).

A sound etymology must fulfil three conditions. It must not violate the recognised laws of sound change. The development of meaning must be clearly traced. This must start from the earliest or fundamental sense of the word. It goes without saying that in modern corruptions we are sometimes faced by cases which it would be difficult to explain phonetically (see p. 136). There are, in fact, besides the general phonetic and semantic laws, a number of obscure and accidental influences at work which are not yet codified. As we have seen (p. 188), complete apparent dissimilarity of sound and sense need not prevent two words from being originally one[156]; but we have to trace them both back until dissimilarity becomes first similarity and then identity.

The word peruse meant originally to wear out, Old Fr. par-user. In the 16th century it means to sort or sift, especially herbs, and hence to scrutinise a document, etc. But between the earliest meaning and that of sifting there is a gap which no ingenuity can bridge, and, until this is done, we are not justified in regarding the modern peruse as identical with the earlier.[157]

The maxim of Jakob Grimm, "von den Wörtern zu den Sachen," is too often neglected. In dealing with the etymology of a word which is the name of an object or of an action, we must first find out exactly what the original object looked like or how the original action was performed. The etymologist must either be an antiquary or must know where to go for sound antiquarian information. I will illustrate this by three words denoting objects used by medieval or Elizabethan fighting men.

A fencing foil is sometimes vaguely referred to the verb foil, to baffle, with which it has no connection. The Fr. feuille, leaf, is also invoked, and compared with Fr. fleuret, a foil, the idea being that the name was given to the "button" at the point. Now the earliest foils and fleurets were not buttoned; first, because they were pointless, and secondly, because the point was not used in early fencing. It was not until gunpowder began to bring about the disuse of heavy armour that anybody ever dreamt of thrusting. The earliest fencing was hacking with sword and buckler, and the early foil was a rough sword-blade quite unlike the implement we now use. Fleuret meant in Old French a sword-blade not yet polished and hilted, and we find it used, as we do Eng. foil, of an apology for a sword carried by a gallant very much down at heel. As late as Cotgrave we find floret, "a foile; a sword with the edge rebated." Therefore foil is the same as Fr. feuille,[158] which in Old French meant sword-blade, and is still used for the blade of a saw; but the name has nothing to do with what did not adorn the tip. It is natural that Fr. feuille should be applied, like Eng. leaf, blade, to anything flat (cf. Ger. Blatt, leaf), and we find in 16th-century Dutch the borrowed word folie, used in the three senses of leaf, metal plate, broadsword, which is conclusive.

We find frequent allusions in the 16th and 17th centuries to a weapon called a petronel, a flint-lock fire-arm intermediate in size between an arquebus and a pistol. It occurs several times in Scott—

On the strength of a French form, poitrinal, it has been connected with Fr. poitrine, chest, and various explanations are given. The earliest is that of the famous Huguenot surgeon Ambroise Paré, who speaks of the "mousquets poitrinals, que l'on ne couche en joue, à cause de leur calibre gros et court, mais qui se tirent de la poitrine." I cannot help thinking that, if the learned author had attempted this method of discharging an early fire-arm, his anatomical experience, wide as it was, would have been considerably enlarged. Minsheu (1617) describes a petronell as "a horseman's peece first used in the Pyrenean mountaines, which hanged them alwayes at their breast, readie to shoote, as they doe now at the horse's breast." This information is derived from Claude Fauchet, whose interesting Antiquités françoises et gauloises was published in 1579. Phillips, in his New World of Words (1678) tells us that this "kind of harquebuse, or horseman's piece, is so called, because it is to aim at a horse's brest, as it were poictronel." When we turn from fiction to fact, we find that the oldest French name was pétrinal, explained by Cotgrave as "a petronell, or horse-man's peece." It was occasionally corrupted, perhaps owing to the way in which the weapon was slung, into poitrinal. This corruption would be facilitated by the 16th-century pronunciation of oi (peitrine). The French word is borrowed either from Ital. petronello, pietronello, "a petronell" (Florio), or from Span. pedreñal, "a petronall, a horse-man's peece, ita dict. quod silice petra incenditur" (Minsheu, Spanish Dictionary, 1623). Thus Minsheu knew the origin of the word, though he had put the fiction in his earlier work. We find other forms in Italian and Spanish, but they all go back to Ital. pietra, petra, or Span. piedra, pedra, stone, flint. The usual Spanish word for flint is pedernal. Our word, as its form shows, came direct from Italian.[159] The new weapon was named from its chief feature; cf. Ger. Flinte, "a light gun, a hand-gun, pop-gun, arquebuss, fire-arm, fusil or fusee"[160] (Ludwig). The substitution of the flint-lock for the old match-lock brought about a re-naming of European fire-arms, and, as this substitution was first effected in the cavalry, petronel acquired the special meaning of horse-pistol. It is curious that, while we find practically all the French and Italian fire-arm names in 17th-century German, a natural result of the Thirty Years' War, petronel does not appear to be recorded. The reason is probably that the Germans had their own name, viz., Schnapphahn, snap-cock, the English form of which, snaphaunce, seems also to have prevailed over petronel. Cotgrave has arquebuse à fusil, "a snaphaunce," and explains fusil as "a fire-steele for a tinder-box." This is medieval Lat. focile, from focus, fire, etc.

The most general name for a helmet up to about 1450 was basnet, or bacinet. This, as its name implies (see p. 156), was a basin-shaped steel cap worn by fighting men of all ranks. The knights and nobles wore it under their great ornamental helms.[161] The basnet itself was perfectly plain. About the end of the 16th century the usual English helmets were the burgonet and morion.[162] These were often very decorative, as may be seen by a visit to any collection of old armour. Spenser speaks of a "guilt engraven morion" (Faerie Queene, vii. 7). Between the basnet and these reigned the salet or salade, on which Jack Cade puns execrably—

It comes, through Fr. salade, from Ital. celata, "a scull, a helmet, a morion, a sallat, a headpiece" (Florio). The etymologists of the 17th century, familiar with the appearance of "guilt engraven morions," connected it with Lat. cælare, to engrave, and this derivation has been repeated ever since without examination. Now in the Tower of London Armoury is a large collection of salets, and these, with the exception of one or two late German specimens from the ornate period, are plain steel caps of the simplest form and design. The salet was, in fact, the basnet slightly modified, worn by the rank and file of 15th-century armies, and probably, like the basnet, worn under the knight's tilting helm. There is no Italian verb celare, to engrave, but there is a very common verb celare, to conceal. A steel cap was also called in Italian secreta, "a thinne steele cap, or close skull, worne under a hat" (Florio), and in Old French segrette, "an yron skull, or cap of fence" (Cotgrave). Both words are confirmed by Duez, who, in his Italian-French Dictionary (1660), has secreta, "une secrette, ou segrette, un morion, une bourguignotte, armure de teste pour les picquiers." Ergo, the salet belongs to Lat. celare, to hide, secrete.

We now caulk a ship by forcing oakum into the seams. Hence the verb to caulk is explained as coming from Mid. Eng. cauken, to tread, Old Fr. cauquer, caucher, Lat. calcare, from calx, heel. This makes the process somewhat acrobatic, although this is not, philologically, a very serious objection. But we caulk the ship or the seams, not the oakum. Primitive caulking consisted in plastering a wicker coracle with clay. The earliest caulker on record is Noah, who pitched[163] his ark within and without with pitch. In the Vulgate (Genesis, vi. 14), the pitch is called bitumen and the verb is linere, "to daub, besmear, etc." Next in chronological order comes the mother of Moses, who "took for him an ark of bulrushes, and daubed it with slime and with pitch" (Exodus, ii. 3), bitumine ac pice in the Vulgate. Bitumen, or mineral pitch, was regularly applied to this purpose, even by Elizabethan seamen. Failing this, anything sticky and unctuous was used, e.g., clay or lime. Lime now means usually calcium oxide, but its original sense is anything viscous; cf. Ger. Leim, glue, and our bird-lime. The oldest example of the verb to caulk is about 1500. In Mid. English we find to lime used instead, e.g., in reference to the ark—

and—

Our caulk is in medieval Latin calcare, and this represents a rare Latin verb calicare, to plaster with lime, from calx, lime. Almost every language which has a nautical vocabulary uses for our caulk a verb related to Fr. calfater. This is of Spanish or Portuguese origin. The Portuguese word is calafetar, from cal, lime, and afeitar, to put in order, trim, etc.

The readiness of lexicographers to copy from each other sometimes leads to ludicrous results. The origin of the word curmudgeon is quite unknown; but, when Dr Johnson was at work on his dictionary, he received from an unknown correspondent the suggestion that it was a corruption of Fr. cœur méchant, wicked heart. Accordingly we find in his dictionary, "It is a vitious manner of pronouncing cœur méchant, Fr. an unknown correspondent." John Ash, LL.D., who published a very complete dictionary in 1775, gives the derivation "from the French cœur, unknown, and méchant, a correspondent," an achievement which, says Todd, "will always excite both in foreigners and natives a harmless smile!"

It is thus that "ghost-words" come into existence. Every considerable English dictionary, from Spelman's Glossarium (1664) onward, has the entry abacot, "a cap of state, wrought up into the shape of two crowns, worn formerly by English kings." This "word" will no longer appear in dictionaries, the editor of the New English Dictionary having laid this particular ghost.[164] Abacot seems to be a misprint or misunderstanding for a bicocket, a kind of horned head-dress. It corresponds to an Old Fr. bicoquet and Span. bicoquete, cap, the derivation of which is uncertain. Of somewhat later date is brooch, "a painting all in one colour," which likewise occurs in all dictionaries of the 18th and 19th centuries. This is due to Miège (French Dict. 1688) misunderstanding Cotgrave. There is a Fr. camaïeu, a derivative of cameo, which has two meanings, viz., a cameo brooch, and a monochrome painting with a cameo effect. Miège appears to have taken the second meaning to be explanatory of the first, hence his entry—brooch, "camayeu, ouvrage de peinture qui n'est que d'une couleur." In Manwayring's Seaman's Dictionary (1644), the old word carvel, applied to a special build of ship, is misprinted carnell, and this we find persisting, not only in the compilations of such writers as Bailey, Ash, etc., but even in technical dictionaries of the 18th century "by officers who serv'd several years at sea and land." The Anglo-Saxon name for the kestrel (see p. 100) was stangella, stone-yeller (cf. nightingale), which appears later as stonegall and staniel. In the 16th century we find the curious spelling steingall, e.g., Cooper explains tinnunculus as "a kistrel, or a kastrell; a steyngall." In Cotgrave we find it printed fleingall, a form which recurs in several later dictionaries of the 17th century. Hence, somewhere between Cooper and Cotgrave, an ornithologist or lexicographer must have misprinted fleingall for ſteingall by the common mistake of fl for ſt, and the ghost-word persists into the 18th century.

The difficulty of the etymologist's task is exemplified by the complete mystery which often enshrouds a word of comparatively recent appearance. A well-known example is the word Huguenot, for which fifteen different etymologies have been proposed. We first find it used in 1550, and by 1572 the French word-hunter Tabourot, generally known as des Accords, has quite a number of theories on the subject. He is worth quoting in full—

The only serious etymology is Ger. Eidgenoss, oath companion, which agrees pretty well with the earliest recorded Swiss-French form, eiguenot, in Bonivard's Chronique de Genève.

The engineering term culvert first appears about 1800, and there is not the slightest clue to its origin. The victorious march of the ugly word swank has been one of the linguistic phenomena of recent years. There is a dialect word swank, to strut, which may be related to the common Scottish word swankie, a strapping youth—

But, in spite of the many conjectures, plausible or otherwise, which have been made, neither the etymology of swank nor its sudden inroad into the modern language are at present explained. The word ogre, first used by Perrault in his Contes de Fées (1697), has occasioned much grave and learned speculation. Perhaps the philologists of the future may theorise as sapiently as to the origin of jabberwock and bandersnatch.

INDEX