CC BY
15
DOI: 10.30842/ielcp230690152642
Petr Kocharov
Institute for Linguistic Studies, RAS, St. Petersburg. peter.kocharov@gmail.com
Andrey Shatskov
Institute for Linguistic Studies, RAS, St. Petersburg. andrey.shatskov@gmail.com
CAUSATIVIZATION OF PHENOMENON-BASED EXPERIENTIAL VERBS AND NARTEN-LENGTHENING1
There is an ongoing discussion concerning the origin and function of the so-called Narten present. The lengthened grade is either associated with particular roots or is assumed to have an imperfectivising function allowing to derive durative present stems from telic verbs with root aorists. An alternative explanation is suggested in the present article, according to which the Narten present could be a morphological type that expressed increase of the valency of base verbs with a generalized full grade (e.g. media tantum verbs). Such valency-increasing model could develop at a relatively recent stage of PIE by analogy to the active singular forms with the full grade opposed to the middle singular forms with the zero grade of the productive amphydynamic present. Given that the causation of agentive verbs was typically expressed by the morphological causative in PIE, the original domain of the Narten present might have been primarily based on anticausative (inchoative) and experiential verbs. This assumption is supported by the Narten presents derived from *dek- 'to see', *kwei- 'to observe' and *steu- 'to be manifested'.
Keywords: Proto-Indo-European, Narten present, acrostatic present, reconstruction, lengthened grade, verbal morphology.
П. А. Кочаров
Институт лингвистических исследований РАН, Санкт-Петербург.
peter.kocharov@gmail.com
А. В. Шацков
Институт лингвистических исследований РАН, Санкт-Петербург.
andrey. shatskov@gmail. com
Каузативизация экспериенциальных глаголов и продленная ступень Нартен-презенсов
Функция и происхождение акростатических основ настоящего времени (так называемых Нартен-презенсов) остаются предметом дискуссий. Продленная ступень в этих основах объясняется либо как морфологический признак лексемы, либо как деривационный показатель, выражающий дуративность основ настоящего времени, образованных от предельных глаголов с корневым аористом. В данной
1 This research has been supported by the Russian foundation for basic research (project № 19-012-00581)
статье предлагается альтернативное объяснение, согласно которому основы настоящего времени с продленной ступенью корня выражали переходные пары отложительных глаголов с полной ступенью корня. Такая морфологическая модель могла развиться на относительно позднем этапе праиндоевропейского по аналогии с активными формами единственного числа с полной ступенью корня, противопоставленными медиальным формам с нулевой ступенью корня в составе продуктивного амфидинамического презенса. Учитывая, что переходные пары от агентивных глаголов образовывались в праиндоевропей-ском с помощью морфологического каузатива, развитие Нартен-пре-зенса могло быть связано в первую очередь с антикаузативными и экспериенциальными глаголами. Это предположение поддерживается рассмотренными в статье Нартен-презенсами, образованными от *dek- 'видеть', *kwei- 'наблюдать' и *steu- 'проявляться'.
Ключевые слова: праиндоевропейский, основы настоящего времени, Нартен-презенс, реконструкция, продленная ступень, глагольная морфология
According to standard textbooks, PIE had two types of the athematic root presents. One of them had the full grade of the root in the active singular and the zero grade in the dual, plural and all forms of the middle. A more rare type, named Narten present after its discoverer Johanna Narten (1968), had the lengthened grade in the active singular and the full grade in the remaining forms of the present (cf. Fortson 2010: 96). With respect to the position of the accent these two types are defined as amphidynamic (the accent moves from the root to the ending in the forms with the zero grade) and acrostatic (the accent remains on the root throughout the paradigm).
Two opposing views have been put forward to explain the function of the acrostatic type. Schindler (1994) associated it with particular roots rather than with a specific grammatical meaning. A grammatical explanation has been suggested by Kümmel (1998) who pointed out a number of verbs, for which an amphidynamic root aorist can be reconstructed next to an acrostatic root present. According to him, the aorist and present stems are opposed to each other in such cases by the contrastive values of the [± durative] aspectual feature. In such verbs, the ablaut pattern with the lengthened grade marked imperfective aspect in contrast to the amphidynamic pattern of the aorist, cf.:
aor. *dekldek- 'to perceive' : prs. *dek/dek- 'to keep noticing'
aor. *h1edlh1d- 'to bite' : prs. *hjzdlhxed- 'to eat'
aor. *keHslkHs- 'to order' : prs. *keHs/keHs- 'to instruct'
aor. *kremh2/krmh2- 'to get asleep' : prs. *kremh2/kremh2- 'to sleep' aor. 'to make a step' : prs. remH/kw)remH- 'to walk'
aor. iu- 'to set in motion' : prs. ieu/k(wieu- 'to move'
aor. *uelh1/ulh1- 'to make a choice' : prs. *uelh1/uelh1- 'to wish' aor. *tetk/tetk- 'to craft' (telic) : prs. *tetk/tetk- 'to craft' (durative)
The formal oppositions within the active and mediopassive parts of the paradigm, as implied by the aspectual hypothesis, are shown in Table 1.
Table 1. The aspectual hypothesis of the Narten present
Aorist [-durative] Present [+dirative]
Act. *CeC- (sg.) / CC- *CeC- (sg.) / CeC-
Mp. *CC- *CeC-
Melchert (2014) argued that both acrostatic and amphidynamic presents could be derived from an atelic root and pointed out two examples: 1) atelic *h1es/h1s- 'to be' next to *hjes/hjes- 'to abide, sit' (Hitt. eszi); 2) atelic *uek/uk- 'to wish' (Ved. vasti) next to *uek/uek- 'to demand' (Hitt. wek-). Note, however, that plene spellings in the Hittite verbs es- and wek- do not necessarily indicate the lengthened grade, see Kloekhorst 2014: 158f., 168ff. Besides, Norbuis (2021: 235ff.) argues that the direction of derivation was opposite in case of *hes- and that the meaning 'to be' must be secondary given that cross-linguistically body posture verbs often develop into copulas. With that, the aspectual hypothesis remains open for further discussion.
In what follows we suggest an alternative explanation of the Narten type. We propose that rather than being an aspectual marker the Narten present was one of the morphological expressions of valency-increasing alternations.
It has been noticed that the zero-grade athematic stems with middle inflections, that constitute part of the productive amphidynamic pattern, are found in verbs, where the middle has an oppositional value with respect to the active forms; the middle voice typically marks intransitivizing alternation of basically transitive active verbs, cf. PIE act. *mleuh2-ti > Skt. braviti 'speaks') / mid. *mluh2-toi > Skt. bruve 'is called'; see Villanueva Svensson 2012
with further evidence from Indo-Iranian and Greek. Such pairs can be described as a type of causal / noncausal alternation2, in which the causal member is semantically unmarked.
By contrast, semantically unmarked noncausal verbs could have the full grade athematic presents with the middle (the full grade media tantum verbs) or active endings, cf. cf. also PIE *kei-or 'to lie' (Hitt. kitta, Skt. saye, Gr. Ksixai 'to lie') and *ues-tor 'to wear' (Hitt. westa, Skt. vaste, Gr. soxai), on the one hand, and PIE *h1es-ti 'to be' and *h1ei-ti 'to go', on the other hand. It was probably the full grade of the singular and not the voice that originally marked the default meaning in the amphydynamic presents (see Inglese 2020 on the inner-PIE grammaticalization of the oppositional voice). The same marking principle may be postulated for the root aorists, cf. *hie-sed-t intr. 's/he sat down'.
Transitivity pairs, the unmarked meaning of which was expressed by the noncausal member, expressed the causal member by productive transitivtizing morphology including reduplication, nasal presents and *eie/o-causatives (see most recently Covini 2016, Luraghi 2022 on the transitivizing patterns of PIE with further references). We suggest that the lengthened grade of the Narten present have developed as a marginal expression of the causal semantics at a relatively late stage of PIE, when the grammaticalization of the oppositional voice and the corresponding ablaut pattern featuring the contrastive full grade of the singular active gained productivity. Villanueva Svensson (2012) claims that the full grades of many media tantum verbs "go back to Narten presents". In fact, the direction of derivation may have been the opposite: some Narten presents may have been derived from media tantum verbs, which expressed the unmarked meaning of a transitivity alternation.
A systematic revision of the evidence that has been suggested in support of the reconstruction of the Narten present goes beyond the scope of the present paper. It is limited to only a few preliminary observations in support of the suggested hypothesis.
On the formal side, the markedness of the singular active with the lengthened grade could have been triggered by the opposition of
2 The terms "causal" / "noncausal" are applied in the present paper to the two members of a transitivity alternation as defined in Haspelmath 2016. Accordingly, the noncausal member may refer to an intransitive or transitive verb that allows for causalitization.
the zero-grade middle forms and full-grade active forms in the singular of the productive amphidynamic presents, cf. Table 2. A secondary nature of the Narten present (and presumably its relatively recent age) in comparison to the amphidynamic presents expressing the oppositional voice is supported by the fact that the Narten type gained moderate productivity only in some of the IE branches, Indo-Iranian in particular.
Table 2. Analogical formation of the lengthened grade in the active singular of the Narten present
Amphidynamic Acrostatic
Mp. *CC- [noncausal] *CeC- [noncausal]
Act. *CeC- [causal] ^ *Ce C- [causal]
On the semantic side, the "transitivizing" hypothesis of the Narten present requires that it be reconstructed for verbs that allow for a valency-increasing alternation. Moreover, given that the causal pairs to agentive base verbs were typically expressed with the help of the morphological causative in PIE (cf. *kieu- 'to move' ^ caus. *kiou-eie/o- 'to set in motion'; see LIV2: 394f.), the rise of the Narten present may be associated primarily with the anticausative or/and experiential verbs lower on the spontaneity scale (see Haspelmath 2016).
One particular type of such alternation may be illustrated by three experiential verbs that could express the Experiencer-based / Phenomenon-based polysemy by the same root (see Benedetti 2012 on the Experiencer-based / Phenomenon-based alternation in ancient IE languages, Indo-Iranian in particular) and are commonly reconstructed with the Narten present.
1) PIE *dek- 'to perceive' (Arm. tesanem 'to see', Gk. Ssxo^ai 'to expect', etc. LIV2: 109-112) / *dek- ' to make noticeable, remarkable' (Skt. dasti 'to venerate'). The semantic relation between the base and derived stems may be described as that between an Experiencer-based transitive verb of perception and a causative derived from its corresponding intransitive Phenomenon-based equivalent:
*dek- 'X notices Y' (Experiencer-based) / 'Y is/becomes noticed/noticeable (by X)' (Phenomenon-based) ^ *dek- 'Z makes Y noticed/noticeable (by X)'3 In contrast to the morphological causative *dok-eie/o- (cf. Lat. doceo 'to teach'), derived from the Experiencer-based meaning ('Z makes X notice Y'), the Narten present may be considered as the causal pair to the noncausal Phenomenon-based verb.
Comparable causative derivation may underlie the classical example of the Narten present, PIE *steu-/*steu- 'to make manifested, known; praise', if it was indeed derived from Phenomenon-based *steu- 'to be(come) manifested'. Kümmel reconstructs a full-grade stative stem *steu- on the basis of Hitt. istuwari 'to be(come) known' and Vedic stave 'to be praised' (LIV2: 600). Inglese (2020: 298f.) shows that the Hittite verb indicates a spontaneous change-of-state event, whereas its counterparts in Greek (oxewai 'to declare, be manifest') and Vedic (stave) rather point to a stative meaning. According to Inglese (ibid.) the root was originally a medium tantum. Regardless of the aspectual reconstruction (stative/nonstative), there is an agreement that the original meaning of the verb was non-agentive and its generalised full grade might have been associated with the middle inflections at the stage when the analogical lengthened grade was formed. The grammatical semantics of the active forms with the lengthened grade, derived from a Phenomenon-based experiential meaning, seems to neatly correspond to that of *dek-.
A yet another similar example is offered by *kwei-/*kwei- 'to pay attention to smb., to venerate smb.' (Ved. cayati 'to venerate'; see Cheung 2008: 28f. with refs.). If PIE *kwei- had the base Experiencer-based meaning 'to observe smth.' (see Kocharov 2016 on the perfecto-present semantics of the root) and a corresponding Phenomenon-based meaning 'to be(come) visible, remarkable', the causativization of the latter would, again, yield *'to make smb. remarkable' > 'to pay attention, to venerate'.
The three aforementioned examples, illustrating one specific type of lexico-syntactic alternation, provide some positive evidence
3 See Tichy 1976 with a comparable semantic interpretation and the discussion of competing causative expressions derived from the Phenomenon-based meaning of *dek- (cf. Hom. 3pl. 5ei5éxaxai 'to welcome'; but see Haróarsson 1993: 62-65 who interprets the derived meaning of 5ei5éxaxai as intensive).
in favour of the suggested hypothesis of the rise of the lengthened forms of the active singular by analogy to the contrastive full grade of the oppositional active voice. These preliminary observations invite for a further detailed examination of other types of valency alternations, which may have been associated with the Narten present.
Bibliography
Benedetti, M. 2010: Valency alternations with perception verbs in Indo-European languages. In: Melchert H. C. (ed.). The Indo-European Verb. Proceedings of the Conference of the Society for Indo-European Studies, Los Angeles 13-15 September 2010. Wiesbaden: Reichert Verlag, 1-6. Cheung, J. 2008: Etymological dictionary of the Iranian verb. Leiden, Boston.
Covini, A. 2016 Formazioni causative nelle lingue indoeuropee di piu antica attestazione. Tesi di dottorato. Universita per stranieri di Siena, Universität zu Köln. Fortson, B. W. 2010: Indo-european language and culture: an
introduction. Chichester, U. K.; Malden, MA: Wiley-Blackwell. Haröarson, J. A. 1993: Studien zum urindogermanischen Wurzelaorist und dessen Vertretung im Indoiranischen und Griechischen. Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft. Haspelmath, M. 2016: Universals of causative and anticausative verb formation and the spontaneity scale. Lingua Posnaniensis 58/2 (2016): 33-63.
Inglese, G. 2020: The Hittite Middle Voice. Synchrony, Diachrony,
Typology. Leiden, Boston: Brill. Kloekhorst, A. 2014: Accent in Hittite A Study in Plene Spelling, Consonant Gradation, Clitics, and Metrics. (StBoT 56). Wiesbaden: Harrassowitz.
Kocharov, P. 2016: The issue of lexical and grammatical distance of PIE extended roots (*kwei-, *kweis-, *kweit-). Indo-European Linguistics and Classical Philology 20 (2016): 456-465. Kümmel, M. J. 1998: Wurzelpräsens neben Wurzelaorist im
Indogermanischen. Historische Sprachforschung 111(2): 191-208. LIV2 —Rix, H., Kümmel, M., Zehnder, Th., Lipp, R., Schirmer, B. 2001: Lexikon der indogermansichen Verben. Die Wurzeln und ihre Primärstammbildungen. 2 erweit. und verbess. Aufl. bearb. von M. Kümmel und H. Rix. Wiesbaden: Reichert. Luraghi, S. 2022: Basic valency orientation, the anticausative alternation, and voice in PIE. In: M. Malzahn, H. Fellner, Th.-S. Illes (eds.), Zurück zur Wurzel - Struktur, Funktion und Semantik der Wurzel im Indogermanischen. Akten der Tagung der Indogermanischen Gesellschaft vom 13. bis 16. September 2016 in Wien. Wiesbaden: Reichert, 185-196. Melchert, H. C. 2014: "Narten formations" versus "Narten roots." Indogermanische Forschungen 119(1): 251-258.
Narten J. 1968: "Zum 'proterodynamischen' Wurzelpräsens". In: J. C. Heesterman, G. H. Shokker & V. I. Subramoniam (eds.).
Pratidanam. Indian, Iranian and Indo-European studies presented to Franciscus Bernardus Jacobus Kuiper on his sixtieth birthday. The Hague: Mouton, 9-19. Norbruis, S. 2021: Indo-European origins of Anatolian morphology and semantics: innovations and archaisms in Hittite, Luwian and Lycian. PhD thesis. Leiden University. Schindler, J. 1994: Alte und neue Fragen zum indogermanischen Nomen. In: J. E. Rasmussen, B. Nielsen (eds.), In honorem Holger Pedersen: Kolloquium der Indogermanischen Gesellschaft vom 25. bis 28. März 1993 in Kopenhagen. Wiesbaden: Reichert, 397-400. Tichy, E. 1976: Gr. 5ei5exaxo und idg. *dskti, dektoi. Glotta 54 (1976): 71-84.
Villanueva Svensson, M. 2012: Ablaut of the middle root athematic presents in Indo-European. In: H. C. Melchert (ed.). The Indo-European verb. Proceedings of the Conference of the Society for Indo-European Studies, Los Angeles 13-15 September 2010. Wiesbaden: Reichert, 2012: 333-342.
