- In La gémination consonantique d'origine expressive dans les langues germaniques, 1937, p. 59 ff. A. Martinet maintains, against Meillet, that the agreement of Greek (á) and Gothic (atta) does not prove the existence of a primitive IE *atta and is inclined to believe that the several IE languages made their own (sometimes identical) choice from among the existing variants, i.e. ata/āta/atta. In want of sufficient evidence it is, of course, extremely unsafe to draw dogmatic conclusions regarding the form of this IE word. It should be pointed out, however, that the OSlav. form оίbϲb cannot prove an IE *ata, since owing to the Slavonic rule of open syllables, *atta would have become *oto- in any event, just as *pokuto- gave potЪ, *nokuti- gave notv, etc. The same rule of ‘open syllables’ seems to account for the absence of gemination in the non-Indo-European languages quoted by Martinet (Hung, atya, Turk ata, Basque aita) which for this reason cannot help us in elucidating the Indo-European problem. However this may be, our Germanic evidence (on which see further on in the text) only allows to infer the existence of a normal (as opposed to an affective) lexical word *atta and for our present purposes that is sufficient.
- I do not want to dwell on the peculiar forms reconstructed for our group by Walde-Pokorny, and repeated by Pokorny in his new dictionary. But there is certainly no justification for positing beside Gmc. *apala- a form *ōpela-, with a puzzling e, instead of *ōpala-, the lengthened grade derivative of *apala-.
- Behagel devoted a special study to ōdal, SBayr. Akad. 1935. Concluding that there is no connection between this word and adal ‘Geschlecht’, he puts the blunt question: “… ist man eigentlich verpflichtet, an der etymologischen Verknüpfung der beiden Wurzeln festzuhalten?” I hope that my discussion will prove this to be the case, and in view of the clear morphological pattern it is rather disheartening to see where a narrow “Einzelsprachphilologie” can lead.
- The latter was suggested by Grimm (Dt. Wb. s. v. Adel) who construed a German adan/uod and thought it possible to connect this with Skt. at- ‘ire’ because this “leicht in crēscere übergeht.” This view has been perpetuated in the Oxford Dictionary s.v. Alhel.
- It is not impossible that Tokh. A ātäl ‘Mann’ belongs to our group (see Pokorny, l.c., p. 71); in this case the basic meaning must have been ‘offspring’.
- This would seem to be the ‘received’ explanation of the first three words, see Walde-Hofmann, s.vv. In the case of sine, both ending and root vowel are disputed (cf. ibidem s.v.). I am of opinion that it simply represents IE *seni which became *sini and eventually sine. As to cilium, it must not be passed over in silence that it may be an abstraction from supercilium (as admitted in the 3d edition of Ernout-Meillet), in which case it can have no bearing on our discussion.
- In view of the distinct morphological and semantic advantages of E. Leumann's explanation, it seems difficult to understand why J. B. Hofmann should prefer to trace similis < *semilis and òµαλó to IE *semelos/somelos (/d ed., II 539). Although the mechanical application of phonetic rules is not violated, yet there seems to be no explanation for the presence of a reduced vowel between *sem- and the “suffix” l. This strange circumstance receives full light only if we adopt E. Leumann's view.
- I am thus in agreement with Walde-Pokorny (II 257) who connected µεγαλо- with χθαµαλόѕ. But whereas they explained µ𝛆𝛄α𝛌о- as a crossing between µέ𝛄α and the unexplained χθαµαλόѕ, my explanation sheds light on both forms and also on 𝛅µα𝛌óѕ.—Of the two suggestions usually proposed for ON mjok ‘sehr’, it seems that the one equating it with Skt. mahi (from -ə, Bezzenberger in BB 3, 174) can now be ruled out (see T. Burrow, Transactions of the Philological Society 1949, 45) since the Hittite i-stem mekkiš renders it highly probable that the Skt. form is the ntr. of the same i-stem adjective; on the other hand, Brugmann's derivation (Morph. Untersuch. II 175) of mjok from IE *megri can now be connected with the various Greek formations in -a (τά𝛞α, 𝛌íлα etc.) which have plausibly been derived by Benveniste from IE -ṇ (Origines de la formation des noms, 1935, 89). I am thus led to think that Greek µέ𝛄α, too, should be traced to IE *meg-ṇ, and adverbial formation to the root *meg-. In this case the, in any case highly irregular and vexed, form µ̥𝛄αѕ can be explained as having been formed to µ̥𝛄α on the model of 𝛑ο𝛌ύ: 𝛑ο𝛌ύѕ; thus the unavailing efforts to explain the, from the structural point of view disturbing, forms µέ𝛄αѕ, -av with the help of a laryngeal would become superfluous.
- Mezger, I.e., evaluates in the same sense many other instances discussed above (e.g., ň𝛌ικ∈ѕ, οµα𝛌οѕ) and finds an ‘augmentative’ l in µε𝛄α𝛌ο- and 𝛑ο𝛌𝛌ο- < 𝛑ο𝛌υ𝛌ο-; with the latter view he follows suit to Schwyzer who (Griech. Gramm. 484) assumed a ‘diminutive’ -𝛌ο- for 𝛘θαµα𝛌óѕ and µϵ𝛄α𝛌ο- 𝛑ο𝛌υ𝛌ο- although he thought that for the latter instances an ‘augmentative’ meaning was also admissible. I cannot help thinking that this theory of “adaptation” is intrinsically less probable than the alternative explanation employed in the text above, especially as the view of suffixes originating from autonomous (but often obsolete) second compositional members is supported by a large body of evidence in the Indo-European languages as well as in other, unrelated, language groups.
- Feist (o.c., s.v. wiprus) posits an IE *wetrus which is highly improbable. I would suggest that Gmc. *weðrus is a Gmc. innovation, the transfer from the o-stems being motivated by other names of animals belonging to the u-stems.
- I am indebted to Prof. Martinet for various suggestions concerning both the substance and the form of my paper which I have gratefully made use of.
- Choose new content alerts to be informed about new research of interest to you
- Easy remote access to your institution's subscriptions on any device, from any location
- Save your searches and schedule alerts to send you new results
- Export your search results into a .csv file to support your research