A Note on the Origin of the T-Stem in Greek Abstracts Ending in -MA Author(s): Catalin Anghelina Source: Glotta, Bd. 86 (2010), pp. 1-6 Published by: Vandenhoeck & Ruprecht (GmbH & Co. KG) Stable URL: http://www.jstor.org/stable/41219874 Accessed: 28-09-2016 03:04 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact
[email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms
Vandenhoeck & Ruprecht (GmbH & Co. KG) is collaborating with JSTOR to digitize, preserve and extend access to Glotta
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
A Note on the Origin of the T-Stem in Greek Abstracts Ending in -MA By Catalin Anghelina, Columbus/Ohio
The issue of the origin of t in Greek neuter abstracts ending
in -|ш < */-mn/, type ovo|o,a, ôvó|iaxoç < *Л1зпотп/, has not been satisfactorily solved yet. The t in these words is a Greek internal matter, since all the other РШ cognates do not display a dental theme: cf. Lat. nõmen, nõminis' Skt. nãma, nãmnas, Hitt. la-a-ma-an etc.1 In addition, this un-etymological dental theme exists in all dialects of Greek and, therefore, must be considered as having originated in Common Greek or in the Ш dialect that gave birth to Greek. Besides, since different dialects show different outcomes for n (in Mycenaean, */n/ > a, e.g. Inst.sg. e-ka-
ma-te 'support' from e'xM^ but also */n/ > o, e.g. a-mo-ta = арцоха < */harmnta/ 'chariot'; cf. also the variants Мус. pe-mol ре-ma = аяерца), this /-insertion must have occurred when the syllabic n was still in place. In fact, the heteroclitic nouns might be a good indication of this. These nouns present a case similar to the nouns treated here, cf. N.sg. */yekwr/ vs. G.sg. */yekwn-t-os/
> Gk. /hëpatos/, displaying an un-etymological dental stem. The etymology of the oblique cases here shows that the ¿-insertion occurred before */n/ > a.
1 These forms display different ablaut grades of both the stem and the suffix in different IE languages. This type of neuter nouns was proterodynamic in PIE, e.g. N.sg. */h3nómn/, G.sg. */h3nméns/, and its paradigm was leveled differently in the daughter languages; cf. A. Sihier, A Comparative Grammar of Greek and Latin, Oxford, 279, p. 321. Whatever their refashioning was, it is important to notice that, aside Greek, there is no other language in which this type of stem ends in t. Glotta 86, 1-6, ISSN 0017-1298 © Vandenhoeck & Ruprecht GmbH & Co. KG, Göttingen 201 1
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
2 Catalin Angheiina
The most important theories re those of Kieckers, Kretschmer, For Kieckers the t has its origi цеХд,, (léXixoç. Although not imp
whole class of suffixes was c
Kretschmer argued for a spreadi -wnt (xapíeiç, %apfevxoç < */kh */khariwnt/),4 which include th
also assumed that the analogy w
*/n/ : /onoma/ : /onomatos/ vs.
Mycenaean data, however, whe
against this solution. The analogy
vocalized io a or о in differen
proportion must have been /khar : /X/. This solution is reasonable even in this case, it is difficult worked for the ř-insertion in the heteroclitic nouns. An addition-
nai problem with this solution is, in my view, the spreading from a class of words (adjectives) to another (nouns).
Brugmann and Risch viewed the origin of t in the mixing of two stems: those in */-mn/ and those in */-mn-to-/. This theory assumes that the extended stems would have had the same
meaning with their base, cf. Lat. strãmen-strãmentum (Skt. šromatam). The plural forms of this type would have ended in -цата, -цатсоу, which were interpreted as dental themes and applied to the singular forms. The main problem with this solution is that, unlike Latin, Greek does not show any traces of a neuter noun ending in the N. sg. in -axov. In addition, the suffix
2 E. Kieckers, IF 41/1923, p. 184 (also E. Schwyzer, Griechische Gram-
matik, München 1953, p. 520); P. Kretschmer, KZ 31/1892, p. 346; K. Brugmann, Morphologische Untersuchungen 2, Leipzig 1879, p. 220ff.; E. Risch, Wortbildung der homerischen Sprache 2, Berlin 1974, p. 49ff.; N. Oettinger, 'Die Dentalerweiterung von л-Stämmen und Heteroklitika...', Serta Indogermanica, Innsbruck 1982.
"Cf. Sihlern. l,p. 297.
The Greek neuter form xaptev was refashioned after the masculine.
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
A Note on the Origin oft the T-Stem in Greek Abstracts Ending in -MA 3
*/-to-/ seems to have built derivatives that had a meaning (adjectival) different from their base: cf. Skt. himá, hímã 'cold' and 'winter' respectively vs. hemantá- 'pertaining to winter'. It may be then that strömen and strãmentum were not originally
identical semantically. Finally, again, this solution cannot explain this athematic type for the heteroclitic nouns.5 Oettinger6 sees the origin of t in the re-segmentation of the
ablatival suffix -tas, which can still be seen, as a relic, in Sanskrit, Latin or even Greek.7 This solution is possible. However, its drawback, in my view, lies in the fact that in all these languages the function of -tas is adverbial, whereas in Greek the same suffix would have both shaped a whole paradigm and preserved some frozen adverbial forms such as évxóç and ектос. In addition, it is not clear why this suffix consistently affected the paradigms of the neuters only.8
The solution I am proposing below takes into consideration linguistic facts within Greek rather than within PIE. Moreover, the development of this intrusive t is seen as having occurred within the paradigm itself of these words.
Greek displays an interesting phenomenon of epenthesis, which can be described like this: an insertion of a stop homorga-
nic with the preceding nasal occurs in words where a nasal is followed by a resonant, e.g. ávfjp, ávôpóç < */anrós/; äjißpoxoc < */ámrotos/.9 Here the transition between the nasal stop and the rhotic continuant is phonetically realized through a stop homorganic with the previous nasal stop. The considerations above that the ¿-insertion must have
occurred when the */n/ was still present leads to the conclusion that the process could occur in the following way: */h3nomnos/ 5 Cf. Oettinger n.2, p. 243. 0 He follows A. Fick, BB 5, 1880, p. 183ff. Cf. Lat. intus, fimditus; Gk. éviòç and ектос; Skt. šírsatás etc. Oettinger argues for a spreading from neuters such as ôgòuxxto.ç 'from
home' or ютос 'from ear', which were presumably used more often. This
seems to me a desperate solution. This kind of epenthesis after nasals is quite common, e.g. in Romance:
Lat. similare > *simlare > Fr. semblen Lat. ingenerare > *ingenrare > Fr.
engendrer etc.
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
4 Catalin Angheiina
> */onomntos/.10 But the endin of stems did not begin with a vo ending was -si, which was attach */li3nomnsi/. It is worthwhile
similar to the one described abo is followed by a continuant. Th consonant following the nasal: in in the second, a dental fricative. However, in both cases the second consonant is a continuant. Since the place of articulation for the syllabic n is dental as well,11 it may be then that in this
latter case a dental stop developed between the nasal and the dental fricative in the same way it developed in /anrós/.12
Thus, this process of epenthesis could have occurred in the following way: */Ьзпотгш/ > *Л1зпотп^/.13 From the dative plural, the existence of an affricate led to a re-segmentation process, through which a new dental stem emerged, which could be
10 Since this type was proterodynamic in PIE, the locative plural must have been * /hanmensu/ (see n.l); the Greek dative plural form /onomasi/ < */h3nomnsi/ shows that the paradigm was leveled while the syllabic nasal was still present. The form */h3nomnos/ shows that this type has the endings of
the hysterodynamic declension; cf. Skt. G. nãmnas < */h3nomnos/; cf. J. Schindler, 'Zum Ablaut der neutralen s-Stämme des Indogermanischen', in
Flexion und Wortbildung (ed. H. Rix), Wiesbaden, 1975, p. 263-4. In this case, the air is also released through the nose only, while the tongue is being kept on the teeth. A syllabic n requires the previous unreleased stop be homorganic with it. Therefore, in order to pronounce /mn/, one must articulate an orally unreleased dental after Imi or otherwise /n/ would
assimilate to the place of articulation of the previous consonant; cf. P. Ladefoged, A Course in Phonetics, New York, 2001, p. 49-51. This phenol-
menon is called 'nasal plosion'. The fact that I si is a dental may have also contributed to this. The process does not apply to approximants: */laknya/ > /lakaina/ 'Laconian woman'. Similarly, the denominative verb òvoumvco < */onomnyõ/ is not formed using the dental stem; cf. also яерас, лерахос < */perntos/ vs. rcepaivco < */pernyõ/,
where it is clear that the /-stem was not used for the verb.
The result was directly an affricate, which eventually ended as a fricative. A similar process may have been the passage from ti to si, where one can posit an intermediate stage 'si. This can explain why the dative plural has the form in /-masi/ and not in */-massi/ (as from */-matsi/). Another explana-
tion may be that */-ss/ eventual ended up as /s/ after nasal: cf. D.pl.
*/pherontsi/ > */pheronssi/ > /pherousi/.
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
A Note on the Origin oft the T-Stem in Greek Abstracts Ending in -MA 5
used throughout the paradigm.14 Later, -at- < */-nt-/ could be
used as a unit, and new analogical stems could appear, e.g., in yóvu, yóvaxoç, where t was not originally present, as Homeric examples such as youvóç show. It is also interesting to see what happened with other stems in
n, where the same process could have taken place. This brings us to stems ending in -en or -on.
In a word like cppf|v there is no trace of a ř-stem in its paradigm. This can be explained by noticing that the paradigm of this word behaves differently: the dative plural cppaai is the only form in which the suffix has zero ablaut; all the other forms have the full grade. This might be the reason why the analogy
could not work. The oblique cases remained (ppevóç, (ppeví etc. even if, originally, */phrnsi/ could become */phrntsi/ through epenthesis. The same ablaut issues occur with the stems ending in -aw of the type ôaíjxcov. The dative plural had zero ablaut and, therefore, the spreading of t could not occur from here.15
The heteroclitic nouns could, in principle, behave like the abstracts in */-mn/, since here we also deal with a nasal stem, e.g., */yekwr/n-/. In this type, the insertion of t is old as well, occurring when */n/ was still present, cf. G.sg. */yekwn-os/ >
*/yekwn-t-os/. The origin of t here then could also be in an epenthesis that initially affected the dative plural. Thus, the ¿-stems of both the Greek neuter abstracts in -ma
and the heteroclitic nouns may originate in an epenthesis in the
14 German displays a similar phenomenon. In adverbs formed with the suffix -lieh, e.g. eigentlich, ordentlich, gelegentlich etc, t in the suffix is not etymological, but only a transitional sound 'Gleitlaut'. These words have the
interesting feature that in casual speech the n before the resonant is pronounced syllabic [aigntliç] (in formal speech the pronunciation is with
schwa [aigantliç]). The fact that it is the syllabicity of n that produced the glide and not the sequence n + / is shown by word like peinlich, in which n is not syllabic, and, therefore, the epenthesis did not occur. The dative plural for ôaíuíov was originally */daimasi/ < */daimnsi/; on the other hand, */daimonsi/ would have given */daimousi/; this shows that the D.pl. was refashioned; cf. Sihler n.l, p. 294.
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
6 Catalin Angheiina
dative plural, from where it sp hypothesis eliminates the difficu of this "intrusive" ř, which wou
words in some IE languages and the Greek one.
16 I wonder whether the same process could not occur in the nominative/
accusative plural, which were cases used presumably more often in speech and where the ending was possibly a laryngeal (pharyngeal fricative):
*/h3nomnh2/ > */h3nomn-t-h2/ > */onomnta/; this would fit the fact that these nouns are neuter. The existence of /nh2/ in Greek, however, is uncertain; cf.
Sihlern.l,p. 105.
This content downloaded from 140.254.87.149 on Wed, 28 Sep 2016 03:04:06 UTC All use subject to http://about.jstor.org/terms
