component elements of English is very different. M. Müller quotes a calculation which makes the classical element about 68 per cent, the Teutonic about 30, and miscellaneous about 2 (Science of Language, 8th ed. i, 89). See Skeat, Principles of Eng. Etymology, ii, 15 seq., and infra p. 25.]
[4] [What here follows should be compared with the fuller and more accurate lists of words borrowed from foreign sources given by Prof. Skeat in his larger Etymolog. Dictionary, 759 seq.; and more completely in his Principles of Eng. Etymology, 2nd ser. 294–440.]
[5] Yet see J. Grimm, Deutsche Mythologie, p. 985.
[6] The word hardly deserves to be called English, yet in Pope’s time it had made some progress toward naturalization. Of a real or pretended polyglottist, who might thus have served as an universal interpreter, he says:
“Pity you was not druggerman at Babel”.
‘Truckman’, or more commonly ‘truchman’, familiar to all readers of our early literature, is only another form of this, one which probably has come to us through ‘turcimanno’, the Italian form of the word. [See my Folk and their Word-Lore, p. 19].
[7] [‘Tulip’, at first spelt tulipan, is really the same word as turban (tulipant just above), which the flower was thought to resemble (Persian dulband).]
[8] [Ultimately from the Arabic zabād (N.E.D.).]
[9] [Apparently to be traced to the Persian shim-shír or sham-shír (“lion’s-nail”), a crooked sword (Skeat).]
[10] [Rather through the French from low Latin satinus or setinus, a fabric made of seta, silk. But Yule holds that it may be from Zayton or Zaitun (in Fokien, China), an important emporium of Western trade in the Middle Ages (Hobson-Jobson, 602).]
[11] [Probably intended for cacao, which is Mexican. Cocoa, the nut, is from Portuguese coco.]
[12] See Washington Irving, Life and Voyages of Columbus, b. 8, c. 9.
[13] [It is from the Haytian Hurakan, the storm-god (The Folk and their Word-Lore, 90).]
[14] [From old Russian mammot, whence modern Russian mamant.]
[15] [‘Assagai’ is from the Arabic az- (al-) zaghāyah, ‘the zagāyah’, a Berber name for a lance (N.E.D.).]
[16] [This puts the cart before the horse. ‘Fetish’ is really the Portuguese word feitiço, artificial, made-up, factitious (Latin factitius), applied to African amulets or idols.]
[17] [‘Domino’ is Spanish rather than Italian (Skeat, Principles, ii, 312).]
[18] [‘Harlequin’ appears to be an older word in French than in Italian (ibid.).]
[19] On the question whether this ought to have been included among the Arabic, see Diez, Wörterbuch d. Roman. Sprachen, p. 10.
[20] Not in our dictionaries; but a kind of coasting vessel well known to seafaring men, the Spanish ‘urca’; thus in Oldys’ Life of Raleigh: “Their galleons, galleasses, gallies, urcas, and zabras were miserably shattered”.
[21] [A valuable list of such doublets is given by Prof. Skeat in his large Etymological Dictionary, p. 772 seq.]
[22] This particular instance of double adoption, of ‘dimorphism’ as Latham calls it, ‘dittology’ as Heyse, recurs in Italian, ‘bestemmiare’ and ‘biasimare’; and in Spanish, ‘blasfemar’ and ‘lastimar’.
[23] [‘Doit’, a small coin (Dutch duit) has no relation to, ‘digit’. Was the author thinking of old French doit, a finger, from Latin digitus?]
[24] Somewhat different from this, yet itself also curious, is the passing of an Anglo-Saxon word in two different forms into English, and continuing in both; thus ‘desk’ and ‘dish’, both the Anglo-Saxon ‘disc’ [a loan-word from Latin discus, Greek diskos] the German ‘tisch’; ‘beech’ and ‘book’, both the Anglo-Saxon ‘boc’, our first books being beechen tablets (see Grimm, Wörterbuch, s. vv. ‘Buch’, ‘Buche’); ‘girdle’ and ‘kirtle’; both of them corresponding to the German ‘gürtel’; already in Anglo-Saxon a double spelling, ‘gyrdel’, ‘cyrtel’, had prepared for the double words; so too ‘haunch’ and ‘hinge’; ‘lady’ and ‘lofty’ [these last three instances are not doublets at all, being quite unrelated; see Skeat, s. vv.]; ‘shirt’, and ‘skirt’; ‘black’ and ‘bleak’; ‘pond’ and ‘pound’; ‘deck’ and ‘thatch’; ‘deal’ and ‘dole’; ‘weald’ and ‘wood’†; ‘dew’ and ‘thaw’†; ‘wayward’ and ‘awkward’†; ‘dune’ and ‘down’; ‘hood’ and ‘hat’†; ‘ghost’ and ‘gust’†; ‘evil’ and ‘ill’†; ‘mouth’ and ‘moth’†; ‘hedge’ and ‘hay’.
[All these suggested doublets which I have obelized must be dismissed as untenable.]
[25] We have in the same way double adoptions from the Greek, one direct, at least as regards the forms; one modified by its passage through some other language; thus, ‘adamant’ and ‘diamond’; ‘monastery’ and ‘minster’; ‘scandal’ and ‘slander’; ‘theriac’ and ‘treacle’; ‘asphodel’ and ‘daffodil’; ‘presbyter’ and ‘priest’.
[26] The French itself has also a double adoption, or as perhaps we should more accurately call it there, a double formation, from the Latin, and such as quite bears out what has been said above: one going far back in the history of the language, the other belonging to a later and more literary period; on which subject there are some admirable remarks by Génin, Récréations Philologiques, vol. i. pp. 162–66; and see Fuchs, Die Roman. Sprachen, p. 125. Thus from ‘separare’ is derived ‘sevrer’, to separate the child from its