On Anatolian Verbal Prosody: Synchronic Evidence for (Non-) Default Accentuation

On Hittite Verbal Prosody: Synchronic Evidence for (Non-)Default Accentuation Anthony D. Yates University of California, Los Angeles adyates@ucla.edu Introduction §1 §1.1 Toward a synchronic approach: Over the last thirty-five years, Anatolian scholarship has greatly advanced our understanding of how the Anatolian languages continue word-level prosodic patterns inherited from Proto-Indo-European (PIE); relatively less attention, however, has been paid to the synchronic principles by which a word’s accentuation is determined. · Understanding how PIE prosodic patterns are continued into Anatolian depends crucially on determining the attested position of the ictus in (esp.) Hittite, where the primary diagnostic is plene writing. Toward this end, significant contributions by (e.g) Hart (1980), Carruba (1981), Kimball (1983, 1999), Melchert (1984, 1992, 1994), and most recently Kloekhorst (2008, 2014a). §1.2 Explaining innovations? This focus on inheritance has led to difficulty in explaining certain cases where Anatolian seems to depart from PIE accentual patterns, the accentual peak (or ictus) surfacing in an unexpected position. §1.3 “Retraction” in Hittite: In a number of (Old) Hittite forms, the ictus appears to be “retracted” with respect to its PIE position, falling instead on the leftmost syllable, e.g. (1): (1) a. b. c. d. hūnikzi [Xó:nikts <i] ‘batters’ < PIE *h2 u-né-g-ti ˇ nı̄ninkzi [nı́:niNkts <i] ‘mobilizes’ < PIE *ni-né-k-ti hullizzi [Xú(:)l:its -né-h1 -ti <i] ‘fights’ < PIE *h2 wl ˇ ˚ zinnizzi [ts ı́(:)n:its i] ‘finishes’ < PIE *ti-né-h 1 -ti < < e. f. g. h. h -né-h -ti dwarnizzi [twárnits 1 <i] ‘breaks’ < PIE *d wr ˚ terippzi [té(:)ripts i] ‘plows’ < PIE *trép-ti < pūnušzi [pú:nusts <i] ‘interrogates’ < PIE *pnéuH-s-ti teri– [té(:)ri–] ‘3’ < PIE *trı́– §1.4 “Retraction” as phonological default: In Yates (2014a), it was argued that these forms owe their innovative accentual pattern to the application of Anatolian Default Accentuation — stated informally in (2) — and thereby provide diachronic evidence for the synchronic operation of this default phonological principle in (Proto-)Anatolian: (2) Anatolian Default Accentuation (ADA): If a word has no underlyingly accented morpheme, the leftmost syllable bears the ictus. · For cogent arguments that diachronic prosodic change may be diagnostic of a principle of default accentuation, see Probert (2006:137–43). Lundquist (2014, 2015) has demonstrated that PIE *–tı́– stems show the same “reversion to default” in the (pre)history of Vedic and Greek. Barber (1997:131) identifies similar phenomena in the history of English. §1.5 Synchronic evidence for ADA: Primarily diachronic claims remain to be supported with clear synchronic evidence for ADA. Aims are thus: (i) Show that regular prosodic patterns in Anatolian athematic verbs derive synchronically from the interaction of ADA and the unaccented property of their verbal roots (§2). (ii) Demonstrate a systematic prosodic contrast between unaccented and accented roots, and assess its implications for Anatolian accentuation, including how ictus is determined when multiple accented morphemes are present (§3). (iii) Discuss the implications of Anatolian synchronic accentuation for PIE (§4). A.D. Yates Š 2 AOS 225, 13–16 March 2015 The operation of ADA §2 §2.1 Unaccented roots: For (1f) teripzi and (1g) pūnušzi, Yates (2014a) proposed that unaccentedness is characteristic of Anatolian verbal roots that form simplex athematic verbs according to productive patterns in the mi– and hi-conjugations — thus both */trep–/ and */pneuss–/, which belong to this category. ˇ §2.2 Accentuation of athematic verbs: Productive inflectional patterns in Hittite simplex athematic verbs of the mi- and hi-conjugations involve alternating ictus between root (singular) and inflectional endings (plural); theseˇgenerally accepted surface alternations may be accounted for by the interaction of: (i) unaccented roots; (ii) accented plural endings; (iii) unaccented singular endings; and (iv) ADA. · In the mi-conjugation, Hitt. ē/a and ē/∅ directly continue universally reconstructed PIE *é/∅ ablaut with mobile ictus. Following Jasanoff (2003, 2013), *ó/é ablaut is original in *h2 e-presents, but per Melchert (2013), Hittite preserves this pattern only exceptionally in roots of the shape *TRe(n)T–, e.g. k(a)rāp–/k(a)rēp–, h(a)mank–/h(a)mink– (contra Kloekhorst 2012, ˇ 2014b). The productive pattern in Hittite — and very likely, Proto-Anatolian ˇ— is ā/a, which continues both secondary *ó/∅ ablaut in *h2 e-presents and *ó/ó (exc. 3rd pl. *∅) in *h2 e-aorists. §2.3 Accounting for the data: These assumptions will correctly derive ictus alternations in “ordinary” Anatolian simplex athematic verbs: ADA assigns root (= leftmost) ictus in the singular, while accented inflectional endings receive ictus in plural — e.g. (3) vs. (4) in mi -verbs; (5) vs. (6) in hi-verbs: ˇ (3) mi -conjugation strong stem: a. PA */gw en – ti/ b. PA */es – ti/ (4) *gw énti *ésti Hitt. kuēnzi ‘kills’ > Hitt. ēšzi ‘is’ *gw nénti *asénti > Hitt. kunanzi ‘they kill’ Hitt. asanzi ‘they are’ hi -conjugation strong stem: ˇ a. PA */doh – ei/ → *tóh-ei “ “ b. PA */konk– ei/ → *kónk-ei > Hitt. dāi ‘takes’ > Hitt. kānki ‘hangs’ (6) hi -conjugation weak stem: ˇ a. PA */doh – téni/ → *t@ h-téni “ “ k-énti b. PA */konk – énti/ → *kn ˚ > Hitt. dattēni ‘you (pl.) take’ > Hitt. kankanzi ‘they hang’ (e.g. KBo 6.2 i 3) > (e.g. KBo 17.74 ii 29) mi -conjugation weak stem: a. PA */gw en – énti/ b. PA */es – énti/ (5) → → → → > (e.g. KBo 16.71 Ro i 12, 14) (e.g. KUB 36.104 iv 7´) (e.g. KBo 6.2 i 8) (KBo 17.2 i 7) (e.g. KUB 36.106 Vs. 8) (e.g. KBo 17.74 ii 27) §2.4 Default accent and PA */trep–/, */pneus:–/: If correct, the leftmost ictus of the strong stem of PA *térep– and *púnuss– follow straightforwardly from the normal application of ADA, i.e. (7): (7) a. PA */trep -- ti/ b. PA */pneus: -- ti/ → → *térepti > Hitt. terippzi ‘plows’ *púnusti > Hitt. pūnušzi ‘interrogates’ (VBoT 58 i 30)) (e.g. KUB 14.15 ii 12 (1s. pret.)) · Hitt. teripp– has subsequently generalized the strong stem in the plural, viz. surface [té(:)rip:–] became treated as new underlying /térip:–/; per Melchert (2013:140), the original (post-epenthesis) pattern was *térepti : *trp-énti (pace Kloekhorst 2012, 2014b). ˚ In (7), it is trivially assumed that The same may hold for pūnušš– in view of 3rd pl. pūnuššanzi (KBo 20.5 iii 7; OH/OS). epenthesis is a synchronic phenomenon in PA; the ictus of the output forms would be identical if the epenthetic vowel had become underlying. §2.5 Mechanical or motivated? While the forms in (3–7) are consistent with the assumptions stated in §2.2, it remains to be demonstrated that unaccented roots — and not some other feature — are the crucial factor in determining their accentual patterns. §2.6 A prosodic contrast: If unaccentedness is decisive, the “ordinary” accentual patterns in (3–6) might be expected to contrast synchronically with the prosodic behavior of mi- and hi-verbs based on ˇ accented roots; typologically, this type of contrast is paralleled in Cupeño (Takic, Uto-Aztecan), where minimal pairs illustrate the opposition between (8) unaccented roots and (9) accented roots: A.D. Yates Š 3 AOS 225, 13–16 March 2015 (8) (9) Unaccented roots in Cupeño: a. /max - qá/ → max-qá ‘giving’ b. /w@n - qá/ → w@n-qá ‘put (it)’ ( ‘give’ + Pres.Sing) (‘put’ + Pres.Sing) Accented roots in Cupeño: a. /Páyu - qá/ b. /mı́yax – qá/ Páyu-qa ‘(he) wants’ mı́ya-qa ‘(it) happens’ → → (‘want’ + Pres.Sing.) (‘be’ + Pres.Sing.) · In Cupeño, accented and unaccented roots are systematically opposed (Alderete 2001a,b; cf. Hill 2005). Accented roots always receive primary stress (i.e. ictus) on the accented syllable, whereas in unaccented roots, stress is determined by a complex interaction between the lexically-specified properties of morphemes and phonological principles. In the absence of accented morphemes, Cupeño (like Anatolian) has default leftmost stress, e.g. /max - @m/ → máx-@m ‘Give!’. §2.7 A synchronic contrast: Such a synchronic contrast is in fact apparent in a small set of Hittite verbs exhibiting fixed root ictus in the plural, e.g. (10) (cf. (3–6)): (10) a. wekanzi [wé(:)gants <i] b. ārranzi [á:rrants i] < c. ānšanzi [á:nsants <i] : wek– ‘demand’ : arr– ‘wash’ (KUB 9.28 iv 8 / KBo 11.45 iv 19) : ans– ‘wipe’ (KBo 23.23 Vs. 77 / KBo 19.163 i 23; iv 4) (KBo 19.133 6) §2.8 A minimal difference? Within Anatolian, the verbs in (10) with fixed ictus are morphologically identical to the verbs in (3–6) with mobile ictus; it may be hypothesized, then, that these synchronic prosodic contrasts should be ascribed to differences in the accentual status of their verbal roots (cf. §3). · Provided that their surface allomorphy can be derived as in §2.3 and §3.3, both (3–6) and (10) are ordinary mi- and hi-verbs. ˇ The prosody of (un)accented verbal roots in Anatolian §3 §3.1 A systematic contrast: Hittite imperfectives in –škē– and –nt– participles standardly bear ictus on the derivational suffix, e.g. (11–12); however, the same verbal roots whose derivatives bear “exceptional” fixed root ictus in their plural forms in (10) also exhibit fixed root ictus in these morphological categories, i.e. (13–14): (11) Hittite –ške– presents: : : : dā– ‘take’ eku– ‘drink’ dai– ‘place’ (12) a. daškēmi [t@sk:é:mi] b. akkuškēwani [@k:usk:é:w@ni] c. zikkēt [ts <ik:é:t] Hittite –nt– participles: a. ašānt– [@sá:nt–] b. arānt– [@rá:nt–] c. lagānt– [l@gá:nt–] : : : es– ‘be’ ar– ‘arrive’ lag– ‘incline; fall’ (13) (KUB 36.110 Rs 7) (KBo 22.2 Vs 3) (e.g. KUB 21.5 iii 36 ) (e.g. KBo 21.22 Vs 15) (KUB 33.68 ii 4) “Exceptional” Hittite –ške– presents: a. wekiškizzi [wé(:)giskits <i] b. āršikitta [á:rsikitta] / ārreškizzi [á:rriskits <i] c. ānšikizzi [á:nsikits <i] / ānaškizzi [á:nskizzi] (14) (KBo 17.3 iv 10) (KBo 10.5 iii 4*) (KUB 9.28 iv 8 / KBo 11.45 iv 19) (KBo 23.23 Vs. 77 / KBo 19.163 i 23; iv 4) “Exceptional” Hittite –nt– participles: a. wekantan [wé(:)gantan] b. ārranza [á:rrants <] c. ānšanza [á:nsants <] (KUB 4.3 Vs 16) (KBo 21.57 iii 8) (KBo 16.97 Vs. 35) · Significantly, note that both archaic and renewed iterative-inchoative stems in (13b–13c) show initial ictus, which suggests that the principles operative in determining their ictus remain stable in Hittite. On the accentuation and vocalism of –škē– “imperfectives” (Melchert 1998), see Yoshida (2010:386–7); it is generally assumed that this accentual pattern continues the “ (e.g. Fortson 2010:98–9). The Anatolian –nt– participle and its development are discussed in Melchert PIE situation, i.e. –ské– (1994:146–7). For the phonology of (11b) and (synchronically renewed) (11c), see Kavitskaya (2001). A.D. Yates Š 4 AOS 225, 13–16 March 2015 §3.2 Accented roots: This systematic contrast across morphological categories is most plausibly explained as the result of differences in the accentual status of their verbal roots — specifically, that the “exceptional” roots in (10) and (13–14) are underlyingly accented, i.e. /wé(:)g–/ ‘demand’, /árr–/ ‘wash’, /áns–/ ‘wipe’ vs. (e.g.) /kw en–/ ‘kill’, /es–/ ‘be’, /ka:nk:–/ ‘hang’. §3.3 Implementing the contrast: The prosodic contrast in Hittite between derivatives of accented and unaccented roots would thus be a consequence of different underlying structure — e.g. fixed (15) /wé(:)g–/ vs. mobile (16): (15) (16) a. /wé(:)g – ánzi/ → wekanzi [wé(:)gants <i] ‘they demand’ wekiškizzi [wé(:)giskits <i] ‘demands’ b. /wé(:)g – ské – zi/ → c. /wé(:)g – á:nt – an/ → wekantan [wé(:)gantan] ‘demanded’ a. /es – ánzi/ → b. /ep – ské – t/ c. /es – á:nt – s / → → ašanzi [@sánts <i] ‘they are’ appišket [@p:iské(:)t] ‘took’ ašānza [@sá:nts <] ‘being’ (KBo 19.133 6) (KBo 10.5 iii 4*) (KUB 4.3 Vs 16) (e.g. KUB 36.104 iv 7´) (KBo 4.14 iii 26) (e.g. KBo 21.5 iii 36) · On deriving a-vocalism in (16), see Yates (2014b). §3.4 Consequences for Hittite accentuation: Two implications of this analysis for synchronic ictus assignment in Hittite: §3.4.1 Leftmost wins? The consistent leftmost ictus observed in (15) suggests a principle according to which, when multiple accented morphemes combine, the leftmost surfaces with the ictus, i.e. (17): (17) If a word has one or more underlyingly accented morphemes, the leftmost accented syllable bears the ictus. · In a constraint-based approach, it may be assumed that ADA is driven by an alignment constraint such as Leftmost: “The ictus must be aligned with the left edge of a prosodic word.” This approach in fact makes an empirical prediction in the case of multiple accent resolution, viz. that the ictus will fall on the leftmost accented syllable. This prediction is consonant with the observed “leftmost wins” pattern in complex derivation. §3.4.2 Establishing ADA: While ADA does not apply to accented roots as in (18a), it must be responsible for ictus assignment in the strong stem of unaccented verbal roots, which contain no accented morpheme. The derivation in (18b) thus establishes ADA as part of the synchronic grammar of Hittite. (18) a. Hitt. /wé(:)g – zi / → b. Hitt. /es – zi/ → wēkzi [wé:kts <i] ‘demands’ ēšzi [é:sts (via ADA) <i] ‘is’ (e.g. KUB 29.1 i 27; 32.137 ii 13, 14) (e.g. KBo 17.74 ii 29) Diachronic and comparative aspects of Anatolian accentuation §4 §4.1 Reconstructing PA accentuation: If the Hittite pattern of multiple accent resolution in (17) can be reconstructed for PA, it will yield — in conjunction with (2) ADA — the general principle in (19) for synchronic ictus assignment in (Proto-)Anatolian: (19) Anatolian Accentuation Principle (AAP): If a word has no underlyingly accented morpheme, the leftmost syllable bears the ictus. If a word has one or more underlyingly accented morphemes, the leftmost accented syllable bears the ictus. · Per Yates (2014a), ictus “retraction” phenomena driven by ADA are reconstructible for PA, hence the phonological principle itself. While the reconstruction of (17) for PA seems likely on the basis of Hittite alone, it remains to be confirmed by identifiable effects in the other Anatolian languages. §4.2 AAP in the Hittite verbal system: As shown in §§2–3, the AAP accounts for the systematic prosodic differences exhibited by accented and unaccented roots across several verbal categories (simplex weak stem; imperfective; participle). A.D. Yates Š 5 AOS 225, 13–16 March 2015 §4.3 (Un)accented roots — how? If regular ictus alternations in mi- and hi-conjugations are a ˇ and accented roots consequence of root unaccentedness, then the vast majority of roots must be unaccented, rare — how did this situation arise? For a possible answer, see the Appendix (§5). §4.4 AAP in comparative perspective: The AAP finds very close parallel in Vedic, where Kiparsky (e.g. 2010, forthcoming) has argued that ictus assignment is governed by the BAP in (20): (20) Basic Accentuation Principle (BAP): If a word has more than one inherently accented syllable, the leftmost of these gets the ictus. If a word has no inherently accented syllable, the leftmost syllable gets the ictus. (based on Kiparsky and Halle (1977:209) and Kiparsky (2010:6)) §4.5 Implications for PIE accentuation: This shared prosodic feature of Anatolian and Vedic is unlikely to be due to chance; rather, it must reflect inheritance from PIE, where ictus assignment was governed by similar principles. A new question then arises: what are the properties of the inputs to the PIE grammar that yield reconstructed prosodic patterns? Appendix — On the development of accented roots §5 §5.1 Whence accentedness? Insight into development of synchronically accented roots may be offered by Hitt. /wé(:)g–/. Following Kümmel (1998) and Melchert (2014), there are no “Narten roots,” only “Narten formations” — i.e. a derived morphological category — which according to Sandell (2014) reflect underlying reduplicated forms. This category must ultimately be the source of Hitt. wēk– (cf. Melchert 2014; pace Kloekhorst 2008, 2014a). “ is fundamentally atelic, naturally forming the root present directly continued in Ved. vásti/uśmási · Per Melchert (2014), PIE *wek– ˙ ‘wish’; Hitt. wēk– reflects the derived imperfective with inceptive (or Anfangsterminativ) sense. Building upon and ˙extending earlier proposals (esp. Schumacher 2005), Sandell (2014) contends that the characteristic long vowel of this class is derived via compensatory lengthening of the */e/ vowel of the reduplicant — i.e. */Té - TT - ti/ → *T´ ēT-ti — when syllabification of the root-initial segment is blocked by high-ranking phonotactic constraints governing licit onsets (cf. Byrd 2010; Sandell and Byrd 2014) and OCP-σ (Zukoff 2014). §5.2 “Narten presents” in PIE and PA: It is standardly held that “Narten presents” had *´ē/é inflection (cf. Narten 1968). Sandell (2014) has suggested that *ReT– roots had uniform paradigmatic *´ē. Whatever is original, the invariant lenited velar of Hitt. wēk– directly reflects only PA *´ē. · By Sandell’s (2014) analysis, compensatory lengthening of the reduplicant would occur even in the pre-vocalic 3rd pl., since syllabification of the root-initial segment of *ReT– roots would be blocked by OCP-σ and the SSP (Clements 1990). §5.3 Reanalysis in PA: When “Narten derivation” was lost in (pre-)PA, this category would have been formally distinguished from athematic root presents only by its distinctive prosodic patterns; at this stage, these patterns may have been preserved by restructuring — surface *wē“g– was reanalyzed as underlying */wé:“ g–/, viz. with lexicalization of accentual properties and vowel length. “ The historical development of Hitt. wēk– may thus be schematized §5.4 The diachrony of PIE *wēk: as in (21); neither the rules of the grammar nor surface forms (modulo sound change) are altered, only the properties of the inputs to the grammar: (21) G1 . . . > G2 . . . > G3 “ – mi/ a. G1 : PIE */wé – wk b. G2 : PA *w´ē“gmi c. G3 : Hitt. /wé(:)g – mi/ → ← → “ *w´ēkmi */wé:“g – mi/ wēkmi (KBo 3.38 Vs. 32.34; KUB 34.53 Rs 8) · On restructuring in phonological change generally, see Kiparsky (1982); for recent work on the selection of surface forms as new underlying representations, cf. Albright (2002, 2008). AOS 225, 13–16 March 2015 A.D. Yates Š 6 References Albright, Adam C. 2002. The Identification of Bases in Morphological Paradigms. Ph.D. diss., University of Californa, Los Angeles. ———. 2008. Explaining universal tendencies and language particulars in analogical change. In Jeff Good (ed.), Language Universals and Language Change. Oxford / New York: Oxford University Press. Alderete, John. 2001a. Morphologically Governed Accent in Optimality Theory. New York: Routledge. Alderete, John D. 2001b. Root-Controlled Accent in Cupeño. Natural Language & Linguistic Theory 19(3).455–502. Barber, Charles L. 1997. Early Modern English, 2 edn. Edinburgh: Edinburgh University Press. Byrd, Andrew M. 2010. Reconstructing Indo-European Syllabification. Ph.D. diss., University of Californa, Los Angeles. Carruba, Onofrio. 1981. Pleneschreibung und Betonung im Hethitischen. Zeitschrift für vergleichende Sprachwissenschaft 95.232–248. Clements, George N. 1990. The role of the sonority cycle in core syllabification. In J. Kingston and M.E. Beckman (eds.), Papers in Laboratory Phonology I: Between the grammar and physics of speech, 282–333. Cambridge, UK / New York: Cambridge University Press. Fortson, Benjamin W. 2010. Indo-European Language and Culture, 2 edn. Oxford, U.K. / Malden, MA: WileyBlackwell. Hart, Gillian R. 1980. Some observations on plene-writing in Hittite. Bulletin of the School of Oriental and African Studies 43.1–17. Hill, Jane H. 2005. A Grammar of Cupeño. Los Angeles / Berkeley, CA: University of California Press. Jasanoff, Jay H. 2003. Hittite and the Indo-European Verb. Oxford University Press. ———. 2013. The Tocharian Subjunctive and Preterite in *–a–. In Adam I. Cooper, Jeremy Rau and Michael Weiss (eds.), Multi Nominis Grammaticus: Studies in Classical and Indo-European linguistics in honor of Alan J. Nussbaum on the occasion of his sixty-fifth birthday, 105–120. Beech Stave Press. Kavitskaya, Darya. 2001. Hittite Vowel Epenthesis and the Sonority Hierarchy. Diachronica 18(2).267–269. Kimball, Sara. 1983. Hittite Plene Writing. Ph.D. diss., University of Pennsylvania. ———. 1999. Hittite Historical Phonology. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck. Kiparsky, Paul. 1982. Explanation in Phonology. Dordrecht: Foris. ———. 2010. Compositional vs. Paradigmatic Approaches to Accent and Ablaut (revised version at http://www.stanford.edu/∼kiparsky/). In Stephanie W. Jamison, H. Craig Melchert and Brent Vine (eds.), Proceedings of the 21st Annual UCLA Indo-European Conference, Los Angeles, October 30-31, 2009, 137–181 [1–39]. Bremen: Hempen. ———. forthcoming. Accent and Ablaut. In Handbook of Indo-European Studies. Oxford / New York: Oxford University Press. Kiparsky, Paul, and Morris Halle. 1977. Towards a Reconstruction of the Indo-European Accent. In Larry Hyman (ed.), Studies in Stress and Accent, 209–238. Los Angeles: University of Southern California Press. Kloekhorst, Alwin. 2008. Etymological Dictionary of the Hittite Inherited Lexicon. Leiden / Boston: Brill. ———. 2012. Hittite “ā/e”-ablauting Verbs. In H. Craig Melchert (ed.), The Indo-European Verb: Proceedings of the Conference of the Society for Indo-European Studies, Los Angeles 13-15 September 2010, 151–160. Reichert. ———. 2014a. Accent in Hittite: A Study in Plene Spelling, Consonant Gradation, Clitics, and Metrics. Wiesbaden: Harrassowitz. ———. 2014b. Once more on Hittite ā/e-ablauting hi-verbs. Indogermanische Forschungen 119.55–77. ˇ Kümmel, Martin J. 1998. Wurzelpräsens neben Wurzelaorist im Indogermanischen. Historische Sprachforschung 111.191–208. AOS 225, 13–16 March 2015 A.D. Yates Š 7 Lundquist, Jesse. 2014. Vedic -tı́- Abstracts and the Reconstruction of Proterokinetic *-tı́- Stems in PIE. Paper presented at the 26th Annual UCLA Indo-European Conference, Los Angeles, 24-25 October 2014. ———. 2015. Greek Nouns in –σις: History and Prehistory. Paper presented at the 146th Annual Meeting of the Society for Classical Studies. Melchert, H. Craig. 1984. Studies in Hittite Historical Phonology. Göttingen: Vandenhöck & Ruprecht. ———. 1992. Hittite Vocalism. In Onofrio Carruba (ed.), Per una Gramatica Ittita – Towards a Hittite Grammar, 181–196. Pavia: Gianni Iuculano. ———. 1994. Anatolian Historical Phonology. Amsterdam / Atlanta: Rodopi. ———. 1998. Aspects of Verbal Aspect in Hittite. In Acts of the IIIrd International Congress of Hittitology, Çorum, 16–22 September 1996, 413–418. Ankara. ———. 2013. Ablaut Patterns in the Hittite hi-Conjugation. In Stephanie W. Jamison, H. Craig Melchert and Brent ˇ UCLA Indo-European Conference, Los Angeles, 26-27 October 2012. Vine (eds.), Proceedings of the 24th Annual Bremen: Hempen. ———. 2014. “Narten formations” versus “Narten roots”. Indogermanische Forschungen 119.251–258. Narten, Johanna. 1968. Zum ‘proterodynamischen’ Wurzelpräsens. In J.C. Heesterman, G.H. Schokker and V.I. Subramoniam (eds.), Pratidānam: Indian, Iranian and Indo-European Studies Presented to Franciscus Bernardus Jacobus Kuiper on his Sixtieth Birthday, 9–19. The Hague / Paris: Mouton. Probert, Philomen. 2006. Ancient Greek Accentuation: Synchronic Patterns, Frequency Effects, and Prehistory. Oxford Classical Monographs. Oxford / New York: Oxford University Press. Sandell, Ryan. 2014. The Phonological Origins of Indo-European Long-vowel (“Narten”) Presents. Paper presented at the 26th Annual UCLA Indo-European Conference, Los Angeles, 24–25 October 2014. Sandell, Ryan, and Andrew M. Byrd. 2014. In Defense of Szemerényi’s Law. Paper presented at the 33rd Annual East Coast Indo-European Conference, Blacksburg, VA, 5–7 June 2014. Schumacher, Stefan. 2005. ‘Langvokalische Perfekta’ in indogermanischen Einzelsprachen und ihr grundsprachlicher Hintergrund. In Gerhard Meiser and Olav Hackstein (eds.), Sprachkontakt und Sprachwandel: Akten der XI. Fachtagung der indogermanischen Gesellschaft, Halle an der Saale, 17. bis 23 September 2000, 591–626. Reichert. Yates, Anthony D. 2014a. Accent ‘Retraction’ in Hittite: Toward a Unified Phonological Account. Paper presented at the 26th Annual UCLA Indo-European Conference, Los Angeles, 24–25 October 2014. ———. 2014b. On Proto-Anatolian Verbal Ablaut: The Hittite ašanzi-Type Reexamined. Paper presented at the Kyoto-UCLA Indo-European Workshop, Kyoto, 24-26 March 2014. Yoshida, Kazuhiko. 2010. Observations on the Prehistory of Hittite ie/a-Verbs. In Ronald Kim, Norbert Oettinger, “ Elizabeth Rieken and Michael Weiss (eds.), Ex Anatolia Lux: Anatolian and Indo-European Studies in honor of H. Craig Melchert on the Occasion of his Sixty-Fifth Birthday, 385–393. Ann Arbor, MI / New York: Beech Stave Press. Zukoff, Sam. 2014. On the Origins of Attic Reduplication. In Stephanie W. Jamison, H. Craig Melchert and Brent Vine (eds.), Proceedings of the 25th Annual UCLA Indo-European Conference, Los Angeles, 25–26 October 2014. Bremen: Hempen.