uine Doric form would not be surprising, but I have no citation for this.

3. The prevalence of oi over Toi. The latter is the only form. in Old Phocian, Locrian, Boeotian and the Doric dialects of the Peloponnesus. Nearly all the Aetolian and Achaean inscr. show oi, though the late Delphian retains the roi with more persistency (but oi also frequent; sometimes roi and oi in same inscr., e. g. Collitz, 1707).

4. The prevalence of iepós over iapós, though both are found. The genuine Northwest Greek form was certainly iapós, as in Old Phocian (Collitz, 2501, inscribed at Athens, has iapós, iapów, but also ἱερομνάμονες beside ἱαρομνάμονες, and ἱερομηνία. These forms in iepo- are probably due to Attic influence, as it is unlikely that the pure Phocian used both iapó- and iepó-), Boeotian, Elean and the Doric dialects.

5. The frequency of eis beside ev cum acc., sometimes on the same inscription, e. g. Collitz, 1411, 1415. In general is more common than eis in Northern Greece, but in Achaean inscriptions it is not found. The difference may be accounted for by the assumption that the genuine Achaean form was never év, as in Northwest Greek and Elean, but ès (less probably eis), as in Peloponnesian Doric. This is borne out by the contrast of Achaean σTe 'until' (Collitz, 1615) with Northwest Greek evre.

6. The occasional appearance of eiva and πpòs. These are included by Meister in his list of Ach.-Dor. characteristics (1. c., pp. 82-3), but they are far less common than eiμev and worì, and so belong to what may be called a second stratum of Attic.

7. To this stratum of later or occasional Atticisms belong : genitives like Barthéws (Collitz, 1416-17), Tóλews (Collitz, 4576, Laconia); forms with rr as áλarra (Collitz, 1410, 1636), TÉTTAPES (TETTάPOLS with Aetolian dative, Collitz, 1539, Phocis); éáv (e. g. Collitz, 1634, which also has e ka); ews av (Collitz, 1545, Phocis; 4516, Laconia, etc.); imperat. övтwv, part. ☎v (e. g. Collitz, 1410; on Delphian manumission decrees v and ev are used indiscriminately, e. g. 1757); imperat. -rwσav; eikoσi, etc., etc.

Hitherto we have not questioned the propriety of regarding all that goes under the name of Achaean-Doric as a single κown. Yet the language is far from uniform. Not only are there widely different degrees in the amount of Attic influence shown (contrast, for example, the semi-Attic e ka with the wholly, Attic éáv), but, aside from this, the groundwork is not quite the same in

the dialect spread through Aetolian and that spread through Achaean influence. Only the former has év cum acc. and the dative plural of consonant stems in -ous (pepóvтois, etc.). It would be better, then, to distinguish the Aetolian Kown and the Achaean KOL. It must be remembered also that Attic influence shows itself more or less in nearly all Doric dialects, from the 4th century on, and, except for the specific Aetolian features just mentioned, the result is not essentially different.

UNIVERSITY OF CHICAGO, July, 1900.

CARL DARLING BUCK.

IX.-ETYMOLOGY AND SLANG.

3

1) Latin studet.'-Personally, I get no satisfaction from that etymology of studet which compares it with Gk. oneúder 'hastens,' ' and writes an Aryan base PSTEUD-. As accurate a single definition as we can give to studet is 'aims at,' and if we further paraphrase this by 'strikes at, for,' we open up the possibility of correlating studet with tundit 'strikes.' Goth. stautan, Alban. štüń are s-derivatives of the same base. A close student is called in the vernacular of to-day a 'grind' or 'dig.' Progress in etymology for the immediate future will mean progress in semantics rather than the discovery of new phonetic correspondences. All written languages are stiff and stilted as compared with the vernaculars out of which they are sprung. It is in the field of the spoken vernacular, of language in the nude, that we must hope to find the clue to developments of meaning that have taken place in words now so clothed upon by the formalities of written language that the skeleton beneath is hardly to be divined. Only in the vernacular-in slang-is popular psychology freely at work, reshaping, ever anew, primitive semantic types.

2) Latin pudet.-After these preliminary remarks I need not apologize for rendering me pudet by our frequent vernacular phrase 'I could kick myself,' a phrase by which we express impatience at an act of folly. Not alien is the metaphor of Aeschylus, Prometheus 881: "the heart kicks at the mind for fear."

3) Latin repudium.—An underbred young woman of to-day kicks her suitors. In Latin, repudium is the (parental) rejection of a proposal of marriage. It is clear that I regard pud- in the words pudet and repudium as akin to the word pès 'foot.' We might 1 The older literature can be controlled by referring to Kuhn's Zeitschrift, 4, 34; 8, 450; 11, 91; 12, 409, 428; 13, 307.

2 So Wharton and Prellwitz, in their etymological vocabularies.

3 Cf. Meillet in Mém. Soc. Ling. 9, 154,-without development of the semantics.

4

Wharton defines pudet by 'is stricken,' Schweizer-Sidler (K. Z. 18, 303) by 'es schlägt nieder,' Bréal (Mém. Soc. Ling. 5, 31) by 'frapper.'

suppose pudet to have got its vowel-color from compounds, and even to have carried pudor along with it. There is another possibility, however, as the word now to be discussed shows.

4) nūdapice 'hops, dances.'-It is tempting to compare this verb with tripodat 'dances the three-step'; and it reminds one of the song of Ariel in Shakespeare's Tempest: "Foot it featly here and there"; or of George Meredith's "Now the youth footed swift to the dawn" (in The Day of the Daughter of Hades). If dapite is cognate with mous 'foot,' we shall have to account for the vocalization. Are we to suppose that PŎD- ever had a form půd- beside it? and further to imagine, perhaps, a base POUD-? If so, the riddle of nous is solved. We might apply to the bases POUD- POD- PĒD- PUD- the principle already suggested for SEID- SED- SOD- SID- (v. Brugmann, Grundriss, I2, $549 c.). The same principle of gradation allows us to correlate Lat. caupo 'huckster' with Gk. κáñŋλos, and Lat. capit with cupit (v. Wharton, Etym. Lat., s. v.), without coming into irreconcilable conflict with other explanations (v., e. g., Uhlenbeck, Altind. Wört., s. v. kúpyati).

5) Latin piget.-I would connect piget and pingit, the latter with a formal meaning of 'paints,' developed from a vernacular 'pricks, tattoos.' I put beside eum piget the phrase 'he is all cut up,' or, the same metaphor in elevated language, 'his conscience pricks him.' I also note our words 'tired' (: tear) and 'bored.' We can also explain in this way piger, with a formal meaning of 'slow, lazy,' but comparable with our vernacular substantive 'stick, stick-in-the-mud.' So also we reach an explanation for

1

6) Latin tardus.-Its formal meaning is 'slow,' but in our vernacular 'pokey.' It belongs with Skr. tṛndánti 'they pierce, poke.'

7) Latin paenitet.—It is an easy thing to mediate between pingit 'pricks' and pangit 'fastens.' A knife 'pricks' or 'sticks,' mucilage 'sticks' or 'fastens,' a pin 'pricks,' 'sticks,' and 'fastens.' If we note the perfect pegit (: pangit), we might suppose that paenitet-this orthography being best warranted historically, but without bearing on the etymology-comes from *pegnitet (perhaps

1 Moeller (K. Z. 24, 493) defines by es sticht mich,' and Wharton (1. c.) compares pungit ‘pricks.'

2 Literature in K. Z. 14, 146; 19, 406; B. B. 22, 124; Mém. Soc. Ling. 5, 429. 3 Of course, pungit ‘pricks' is a cognate. The bases ping- pang- pung- must be reconciled by phonetic or analogical manipulation.

even from pågnitet; cf. my explanations of fenum lena in Class. Rev. 11, 298); and so me paenitet may also be turned by 'it pricks my conscience.' A kindred metaphor is seen in the word 'remorse.' It is possible also, considering Germ. fast 'almost,' which is a cognate of our verb fastens 'binds,' to connect paene 'almost' with pegit 'he fastened.' I would explain penuria as from a desiderative formation with the meaning 'desire to fasten upon.'

In line with the semantic explanations offered for piget and paenitet we may proceed to the consideration of

8) Latin taedet.'-This I derive from TEGSD- or TEGD- (with g). I note Skr. tákşati 'cuts,' akin to Lat. texit 'weaves,' tangit 'touches, pricks'; cf. acu tetigisti 'hast touched with a needle'; and does not Vergil make Dido say in soliloquy (Aen. IV 596): infelix Dido, nunc te facta impia tangunt? We have perhaps a -d-extension of this root in Skr. tāḍáyati 'beats, (cuts with a stick-whip)'; at least we may regard the d as the product of -gd-, if idé 'I worshipped' be correctly correlated with vájate 'sacrifices' (v. Wackernagel, Altind. Gram., §145). We may also explain taeda 'torch' from the 'tips' of flame or as (pine) 'splinters.' This etymology will also account for taeter 'stinking.' Smells are not only 'pungent,' but they 'knock us down.'

The words pudet piget paenitet and taedet have all been explained by kindred metaphors. This is amply justified by their identical syntax. The frequency of the metaphor I have used for these explanations is too great in our own vernacular for any one to take exception to this frequency in Latin. To say nothing of the full-dress exhibitions of the metaphor in 'remorse,' 'pricks and stings of conscience,'-in 'it touches me, wounds me, annoys me, cuts me to the quick,' we have a plentiful array of vernacular phrases,-'it hacks me, makes me tired (: tear), bores me, beats my time,'-'I am all cut up, broken up,'-'I could kick myself, he has gone to pieces.' These phrases all indicate various sorts of mental confusion, hesitation, and dismay, likened, metaphorically, to the effects of piercing or striking.

AUSTIN, UNIVersity of Texas.

EDWIN W. FAY.

1 K. Z. 13, 307; 16, 198; 18, 303; 19, 80.