Sanskrit dhīra-₂ ‘steady, brave, energetic’ (original) (raw)

Ha! Linguistic Studies in Honor of Mark R. Hale

Ha! Linguistic Studies in Honor of Mark R. Hale

Laura Grestenberger, Charles Reiss, Hannes A. Fellner and Gabriel Z. Pantillon (eds.)

Bibliografische Information der Deutschen Nationalbibliothek

Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.dnb.de abrufbar.
© 2022 Dr. Ludwig Reichert Verlag Wiesbaden

ISBN: 978-3-7520-0606-3 (Print)
eISBN: 978-3-7520-0085-6 (E-Book)
https://doi.org/10.29091/9783752000856
www.reichert-verlag.de
Das Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar.
Das gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Speicherung und Verarbeitung in elektronischen Systemen.
Gedruckt auf säurefreiem Papier (alterungsbeständig pH 7 -, neutral)
Printed in Germany

Contents

Preface … IX
Bibliography of Mark R. Hale … XI
List of Contributors … XVI
Korlin Bruhn, Bridget Samuels & Bert Vaux Phonological Knowledge and Perceptual Epenthesis1
Thórhallur EYthórsson
Accent Placement and Word Formation in Tocharian B: Resolving an Apparent Paradox … 31
Hannes A. Fellner
No Deviation from the Party(-ciple) Line … 43
Benjamin W. Fortson IV
The Genesis of the Greek Future Deponents … 53
David Goldstein
There’s No Escaping Phylogenetics … 71
Laura Grestenberger
Periphrastic Perfects in Greek and Sanskrit … 93
Dieter Gunkel & Kevin M. Ryan
Vedic Sanskrit Vocatives in -an: The Case for Restoring Two Endings … 117
Alice C. Harris
Tmesis in Aluan: Precursors of Endoclisis … 135
Patrick Honeybone
Unnecessary Asterisks and Realism in Reconstruction: Underspecified is Still Real … 153
Stephanie W. Jamison
Stray Remarks on Nominal Relative Clauses in Vedic and Old Iranian: Proto- proto-izafe … 171
JAY H. JASANOFF
Hitt. tamâšzi ‘(op)presses’ … 183
Madelyn Kissock
Pseudo Harmony in Telugu … 189
JARED S. KLEIN
Adversative Conjunction in Gothic II: alja and sweepauh … 203

Bernhard Koller
Tocharian A Indefinites as Wh-Words … 217
JaKlin KornFilt
Silent Subjects in Turkish: pro and PRO, Arb(itrary) and Not … 233
Melanie Malzahn
Tocharian sâl- ‘fly, throw’-Unsafe at Any Speed … 249
H. Craig Melchert
Non-focused “Fronted” Constituents in Hittite … 263
Alexander Nikolaev
Sanskrit dhîra-2 ‘steady, brave, energetic’ … 277
Alan J. Nussbaum
Classical Latin iûdicâre and Corcolle iouosdiCA-:
Can You Get Here from There? … 285
Georges-Jean Pinault
Starry Dawns in Vedic Time … 299
Markus A. Pöchtrager
Why e/o in Proto-Indo-European? … 311
Charles Reiss
Plastics … 327
Jochem Schindler †\dagger
Zur Theorie der Doppelpossessiva … 331
Sarah G. Thomason
Safe and Unsafe Language Contact … 339
Höskuldur ThRÄinsson
Airports and Islands: The Icelandic Gender System and Some Standardization and Reformation Attempts … 351
Michelle Troberg & John Whitman
Syntactic Glosses and Historical Syntax … 369
Brent Vine
Latin glacies ‘ice’ … 395
Michael Weiss
A Venetic Sound Change … 401
Kazuhiko Yoshida
Some Diachronic Remarks on the Hittite Enclitic Particle -ya® … 413

Sanskrit dhīra-2 ‘steady, brave, energetic’*

Alexander Nikolaev

Besides the well-documented Vedic word dhīra-1 ‘wise, characterized by insight’ ( dhıˉ−d h \bar{\imath}- ), there is another word dhīra-2 that makes its first appearance in Classical Sanskrit (BhagGīta 14.24+14.24+ ) and refers to a different set of personal qualities: ‘steady, firm, brave, energetic, courageous; hero’. 1{ }^{1} The standard dictionaries lemmatize the two words separately, with good reason, since their meanings cannot be easily aligned with each other. As Otto Böhtlingk 2{ }^{2} wrote to Rudolf Roth from St.Petersburg on July 23, 1861: “dhīra- macht mir viel Kopfbrechens. Aus der Bedeutung klug, verständig kann ich die von beständig nicht recht entwickeln”. 3{ }^{3}

What is the origin of dhīra-2? For obvious semantic reasons Böhtlingk and Roth, followed by Monier-Williams, Macdonell and other lexicographers in modern times, 4{ }^{4} wanted to derive dhīra-2 from the root dhr‾d h \underline{r} - ‘to uphold’, but this is impossible: dhr‾d h \underline{r} is an anit root (cf. ipv. didhṛtam, verbal adj. dhṛtá-, dhartár-, dhárman-, etc.) and an analogical model for adj. dhīra- to be created to match the forms of dhr‾d h \underline{r} - does not present itself. 5{ }^{5} This paper explores the possibility that dhīra-1 and dhīra-2 are

[1]

1. • I would like to thank Roberto Batisti, Stefan Höfler, Jared Klein, Sasha Lubotsky, Norbert Oettinger, Georges-Jean Pinault, Massimo Poetto, Kevin Ryan, Michael Weiss, and the editors for their comments on earlier versions of this paper. All views expressed in this paper and all errors are mine only.
     1 E.g., Rām. 2.110.21: dhīrá samupásta tapodhanām “[Sītā] waited in steadfast attendance upon the ascetic”; MBhār. 1.43.31.3 dhairyam ālambya vāmorūr hṛdayena pravepatá “but having gathered her courage, the beautiful one [spoke] with trembling heart”; Bhartṛhari’s Śatakatraya 1.81.2: na niścitārthād viramanti dhīrāh “the resolute ones do not abandon the set goal ever”; Bhavabhuti’s Uttararāmacarita 6.75: dhīroddhatā namayativa gatir dharitrīm “his brave and haughty stride causes earth itself to bow”. dhīra-2 is amply reflected in Middle and Modern Indo-Aryan languages, e.g., Pkt. dhīra ‘bold’, dhīra ‘courage’ ( < dhairya-), dhīraī ‘is courageous’, Pā. dhīra ‘firm’, Kāśmirī dīr ‘steady, constant’, Hindi dhīr ‘steady, firm’ (PED 341; CDIAL 390).
     There is a remote possibility that dhīra-2 is already in the RV in the problematic verse 5.50 .4 cd nṛmáná vīrāpastiyo | ārnā dhīreva sānitā “the manly minded [= Indra], with a dwelling full of heroes, (will) win the floods like dhīrá (acc.neut.pl.)”: dhīra-1 ‘wise’ only refers to animates and ‘wise’ does not yield a satisfactory sense here anyway, but ‘firm’ or ‘vigorous’ (referring to things that Indra has won) may be contextually more plausible. However, the passage is very unclear; see Jamison, RVC ad loc. for further discussion.
     2 Not yet von Böhtlingk.
     3 Brückner and Zeller 2007: 302.
     4 Two homophonous words dhīra-1 and dhīra-2 were also accepted by Gonda (1936: 169).
     5 According to Wüst (1966: 169-70), Indo-Iranian nominal stems in -īra- tend to pair up with verbal stems in -āraya-: if correct, this observation could have provided an explanation for dhīrá- made to match dhāraya-. However, all of Wüst’s examples involve dubious etymologies and there is no evidence that the speakers may have drawn secondary associations between these words: kṣirá- ‘milk’ (Iran. * xsˉıˉra−x \bar{s} \bar{\imath} r a- ) is unrelated to kṣāraya- ‘let flow’ (Av. yžar- ‘to flow’), jīrá’active’ ( ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 28: …w} i h_{x}-r a ̲́- ) is unrelated to jārayánmakha- ‘awakening the bountiful’ ( <∗h1ger−<{ }^{*} h_{1} g e r- ), and virá- ‘man, hero’ is unrelated to YAv. paiti … vāraiieiti 'he destroys’7 (Yt. 10.27), translated by ↩︎

historically unrelated and proposes a new etymological solution for dhīra-2 based on a hitherto unrecognized sound law.

It is well known that there is no geminate rr in Sanskrit. 6{ }^{6} Whenever a sequence −rr-r r - is expected in external sandhi or in composition, the first rr is dropped with a compensatory lengthening of the preceding vowel, if the latter is short (Pāṇ. 6.3.111; Ṛk-Prātīśākhya 4.29), 7{ }^{7} cf. RV 4.7.8b ubhé ant aˉ‾\overline{\bar{a}} ródasi (for antár ródasi) saṃcikitván ‘between the two world-halves, observing them both together’ or AVŚ 1.24.4 sādhaya púná rúpáni (for púnar rúpáni) kalpaya ‘make the forms right again’. 8{ }^{8} The rule applies both to underlying rr and to r<sr<s in Ruki context, cf. RV 1.140.9 mätü rihán ‘licking mother’s (garment)’ from mätüs + rihán or AB 4.21.1+ dúrohaṇá’difficult mounting’9 from duṣ- + róhaṇa-. 10{ }^{10} All consonants in Sanskrit may optionally be doubled in clusters (e.g., brahmma- ‘Brahman’, apahnnute ‘denies’, daddhy atra ‘curds here’, etc.), but rr constitutes the sole-and conspicuous-exception (Pāṇ. 8.4.46-52). The sequence −rr-r r - is thus prohibited in Sanskrit both in derived contexts 11{ }^{11} and morpheme-internally, 12{ }^{12} just as it is in many other languages (including

Gershevitch 1959: 182 as ‘diverts’ and by Hintze 1994: 27 as ‘envelops’.
According to AiGr 2.2:857 (followed by EWAia 3.278), dhīra-2 is etymologically identical with dhıˉra−1d h \bar{\imath} r a-1 which was reanalyzed in post-Vedic times as a derivative of dhr‾d h \underline{r}-. While this is not impossible, note that this would presuppose a split into two words, not a plain semantic change, since dhīra-1 ‘wise’ continues being used in later Sanskrit, as a quick search of Digital Corpus of Sanskrit readily confirms.
6 One apparent exception is the compound vārrāśs- (m.) ‘ocean’ (scil. ‘water-heap’) cited by Rājā Rādhākānta Deva 1825-57/4: 353, allegedly from a Purāṇa; I have not been able to affirm the form.
7 See Whitney 1889: 61; AiGr 1.335; Macdonell 1910: 72; Bloch 1951: 44; Renou 1952: 101; Allen 1972: 70; Kavitskaya 2011: 91-2. The standard example, widely cited in modern literature from Whitney, is puná ramate ‘he is pleased again’ from punar + ramate; as far as I can tell, it was adopted by Whitney from Laghusiddhāntakaumudī 112, a 17th 17^{\text {th }} cent. commentary on Pāṇini’s Aṣṭādhyāy ıˉ\bar{\imath}.
8 Instances like RV 10.190.2(+) ahorátrá- ‘day and night’ (for expectable ahá- < ahar-) or RV 8.31.9 ūdho (for ūdhá < ūdhar) romaśám ‘hairy udder’ are due to analogy: the reason for the praśrita treatment here is that rr-stems ahar- and uˉdhar\bar{u} d h a r - were reanalyzed as neuter ss-stems on the basis of the fact that the distinction between several allomorphs of rr - and ss-stems was neutralized in external sandhi (e.g., the pause form -ah in nom.-acc. sg.), Macdonell 1910: 72 n. 4. Further evidence for this reanalysis comes from instr. pl. ahobhih ( 2 x RV ) showing that aharwas treated like an ss-stem.
9 Dúrohaṇá- is a mode of recitation in which the stanza is recited first by pāda, then by half-verse, then by three pādas, and then in a reverse order (cf. ĀśŚS 8.2.12-13).
10 The only example of word-internal -ṣr- in a derived morphological context is 3 pl . aor. ájussran ‘they delighted’ (RV 1.71.1): normally, -ṣr- is dissimilated into -ṣr- as in tisrás ’ 3 ’ (nom.pl. fem. of tri-) or usrás (gen.sg. of uṣár- ‘dawn’), see Hale 1998.
11 Unsurprisingly, the sequence is not found in internal sandhi either: there is only one set of rr initial endings in the language, namely, 3 pl . -re/-rate, -ran/-ram and ipv. -rám, but there are no such forms in Vedic made from roots ending in −r-r : for instance, the root sˉıˉ\bar{s} \bar{\imath} - ‘to smash’ makes aor. pass. (a)śári, but in the plural the past passive participle is used instead, as in RV 1.174.6 tváyá śúrtáh ‘they were shattered by you’, not tváyá ṭaśṛran (where, admittedly, the cluster would have been −rr-r r - and not −rr-r r-).
12 Sasha Lubotsky points out to me that cases of distant dissimilation of r…rr \ldots r, e.g., intens. *árar(root r‾\underline{r}-) > álar- t { }^{\text {t }} ‘be on the rise’ (AiGr 1.221) or * durhṛ̣ná- ‘rage’ (root hr‾h \underline{r}-) > durháṇá- (Narten 1982: 140) may be related to the phenomenon under discussion.

Dravidian) in which rhotics are not geminable while otherwise double consonants are allowed. 13{ }^{13}

Rather than seeing here an areal phenomenon, I propose to look for a diachronic explanation for this restriction, namely, a sound change that simplified the sequence *-rr- in the history of Sanskrit. Admittedly, this sequence must have been quite rare: Proto-Indo-European generally avoided geminates. 14{ }^{14} However, Proto-IndoAryan ∗−rr15{ }^{*}-r r{ }^{15} can come from other sources, namely, PIE ∗−rl{ }^{*}-r l - or ∗−lr16{ }^{*}-l r{ }^{16} as well as PIE *-LHL- after the application of the Saussure’s effect and the creation of “long vocalic resonants”. There is no reason to doubt that ∗CVLhx{ }^{*} C V L h_{x}-Lo- or ∗CLhx{ }^{*} C L h_{x}-Lo- were valid formations in PIE (cf. * g¨helh2\ddot{g}^{h} e l h_{2}-ro- >> OIr. galar ‘sickness, disease’, Hitt. kallar(a) ‘unfavorable, baleful’); the expected outcomes of * CoL⁡(hx)\operatorname{CoL}\left(h_{x}\right)-Lo- and * CLhxC L h_{x}-Lo- in Sanskrit would be * Carra- and * Cıˉ/uˉrra−17C \bar{\imath} / \bar{u} r r a-{ }^{17} but we never find such forms. It is therefore reasonable to assume that whenever new ∗−rr{ }^{*}-r r - sequences emerged after the loss of laryngeals, these were eliminated. 18{ }^{18} A nice parallel to this presumed development may be seen in the elimination of the secondary ∗{ }^{*}-mm- that arose in the word for ‘mouth’ after the application of the Saussure’s effect: *stómh h1−mn>h_{1}-m n> *stómmn >> *stómn >> Gk. ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 15: \sigma \tau o ̲́ \mu \alpha, Hitt. ištaman-. 19{ }^{19} I therefore submit that the lautgesetzlich outcome of PIE * CLhx−LVC L h_{x}-L V - in Sanskrit was * Cıˉ/uˉrV−>∗Cıˉ/uˉrVC \bar{\imath} / \bar{u} r V->{ }^{*} C \bar{\imath} / \bar{u} r V - with degemination of two rhotics: this theory is borne out by the synchronic facts of Vedic sandhi discussed above and can be supported by several etymologies. 20{ }^{20}

13 The languages in question are Wolof, Biblical Hebrew and Tamil (see Ryan 2019: 125-6) and the Romanesco dialect of Italian (see Palermo 1993). A specific phonetic basis for this restriction in Sanskrit is hard to pin down, since we do not know how exactly Skt. /r/ was realized phonetically: Indian grammatical tradition variously describes this sound as dental (dantya), alveolar (dantamūlīya ‘tooth-rootic’) or retroflex (mūrdhanya ‘cerebral’), see Allen 1953: 54-5. Kobayashi (2004: 99) put forth a phonetic reason for the restriction on -rr- based on a doctrine that Sanskrit /r/ was an alveolar flap [r], but see the criticism by Ryan (2017: 303).
14 A general statement to this effect is given by Byrd (2018: 2071): “heteromorphemic geminates were strictly banned in the proto-language”. This idea finds further support in Jochem Schindler’s observation that there are no adjectives in *-ro- made from roots in final *-r (apud Mayrhofer 1986: 121 n. 99): PIE speakers apparently avoided the derivation that would lead to a dispreferred structure. (So-called “expressive gemination” is a special case, see Watkins 2013).
15 I am going to proceed on the admittedly conservative assumption that PIE ∗l{ }^{*} l merged with PIE ∗r{ }^{*} r at least in those dialects of Indo-Aryan which underlie our Vedic texts and from which the examples discussed in this paper have been drawn.
16 At least the former appears to have been a possible cluster in PIE: individual IE languages either show no restriction on the sequence -rl- (cf. Luw. hūtarlānni- ‘(little) slave’ or Lith. gurlûs ‘weak’ from gûrti ‘crumble, fall apart’) or eliminate it before our eyes, as it were, cf. Lyd. serli-/selli’authority’ (<s⟩=[f]\left(<\mathrm{s}\rangle=[\mathrm{f}]\right. ), OIr. sell ‘eye, iris of the eye, glance’ <∗<{ }^{*} stillo- << PCelt. *stirlo- (but Lat. stélla ‘star’ is inconclusive since it may go back to *stérelā- with syncope), Toch. A aˉraˉl\bar{a} r \bar{a} l ‘end’, perl. aˉrlaˉ(<∗aˉraˉle\bar{a} r l \bar{a}\left(<{ }^{*} \bar{a} r \bar{a} l e\right. with epenthesis, root aˉr\bar{a} r - ‘to cease’), or Arm. molar ‘erroneous’ << *mol-ro-.
17 For the phonological development ∗CLhx−Co−>∗Cıˉ/uˉrCa−{ }^{*} C L h_{x}-C o->{ }^{*} C \bar{\imath} / \bar{u} r C a- cf. tīrná- ‘one who has crossed’ ( << ∗{ }^{*} tṛh ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 11: h_{2}-n o ̲́-) or gürtí- ‘praise’ (<∗gwrhx−ti−)\left(<{ }^{*} g^{w} r h_{x}-t i-\right), Pinault 1987-1988: 329; Clayton 2018.
18 There are no geminate consonants in Old Iranian (see e.g., Kümmel 2014: 208, 210-11). The proposed degemination in ∗Cıˉ/uˉrra{ }^{*} C \bar{\imath} / \bar{u} r r a - is predicated on the loss of the laryngeal in the sequence ∗CLhxLV{ }^{*} C L h_{x} L V (which happened independently in Proto-Indo-Aryan and Proto-Iranian), while the simplification of the geminate in a hypothetical *Carra- <∗Col⁡(hx)<{ }^{*} \operatorname{Col}\left(h_{x}\right)-ro- would have taken place at an earlier point in time after merger of ∗l{ }^{*} l with ∗r{ }^{*} r.
19 See Neri 2005: 212 n. 49.
20 The examples, naturally, range from merely possible to unprovable and outlandish; here is a selection: (1) mūrā- ‘stupid’ <∗<{ }^{*} mūr-ra- <∗mr‾hx<{ }^{*} m \underline{r} h_{x}-lo-, cf. Hitt. marla(nt)- ‘stupid’, OIr. mer

We can return to dhīra-2 ‘steady, brave, energetic’ which can now be provided with an Indo-European pedigree even though the late date of the word may ultimately invite caution. My proposal is to derive dhīra-2 from * dhīr-rá- (with degemination) << * dhlh1d^{h} l h_{1}-ro-, identical with Arm. dalar ‘green, fresh’ and Gk. ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 45: …epsilon \rho o ̲́ s ‘blooming, fresh’ (often used of people in the meaning ‘vigorous, robust’). 21{ }^{21} The common preform for these much-discussed words used to be reconstructed either as ∗dhalh1* d^{h} a l h_{1}-ro- or as * dhh2lh1d^{h} h_{2} l h_{1}-ro- 22{ }^{22}, but recently a scholarly consensus seems to have emerged that the root in question should be reconstructed as ∗dhelh1* d^{h} e l h_{1} - with regular ablaut. 23{ }^{23} If Arm. dalar and Gk. ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 45: …epsilon \rho o ̲́ s go back to a zero-grade ∗dhlh1* d^{h} l h_{1}-ro- (subsequently remodeled in Greek 24{ }^{24} ), one may hypothesize that the outcome of this form in Sanskrit would be * dhīrra-, whence dhīra-. Greek and Armenian words provide an attractive comparandum to Skt. dhīra-2 under the theory that its original meaning ‘fresh, energetic’ in developed to ‘vigorous, strong’ and then to ‘steady, firm, well-controlling’, possibly under the influence of the root dhrd h r -.
‘demented’ (<∗mer(hx)\left(<* m e r\left(h_{x}\right)\right.-ó-), Gk. (Ion.) ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 19: … \omega \rho o ̲́ s ‘stupid, obtuse, foolish’ (<∗moˉr(hx)−\left(<* m \bar{o} r\left(h_{x}\right)-\right. - −)\left.-\right)-(Nussbaum 2015); (2) ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 13: \bar{u} r a ̲́-\right. ‘thigh’ (possibly remade from * ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 13: \bar{u} r a ̲́- after hāh uˉ−\bar{u}- ‘arm’) <∗(<{ }^{*}( u ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 17: … \bar{u} r-r a ̲́-< * (hx)u‾lhx−Lo−\left(h_{x}\right) \underline{u} l h_{x}-L o-, cf. Hitt. walla- ‘thigh’ and perhaps Lat. vulva ( <∗yelyaˉ<∗yelayaˉ)<{ }^{*} y e l y \bar{a}<{ }^{*} y e l a y \bar{a}) if the original meaning of the Latin word was ‘external female genitals’ and not ‘womb’; (3) k uˉla\bar{u} l a - ‘slope, hillside’ either <∗kh‾x−Lo−<{ }^{*} k \underline{h}_{x}-L o- ‘Abhang’ (Lith. kárti ‘hangs’) or *klh 3{ }_{3}-Lo- ‘elevation’ (see Vine 2006: 507); (4) ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 26: …ar{\imath} r a ̲́-, an epithet of fire, śīráśocis- ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 40: …ine{h}_{x}-L o ̲́- (cf. Ved. śrāyant- ‘burning’); (5) kirí- (if the meaning ‘praiser’ is correct) <∗kr‾H−Li−←∗kr‾H−Lo−<{ }^{*} k \underline{r} H-L i-\leftarrow{ }^{*} k \underline{r} H-L o- (cf. Ved. kirrtí- ‘praise’); (6) mūla’root of a plant’ <∗mlhx<{ }^{*} m l h_{x}-lo- (cf. Gk. μωˉλ∪\mu \bar{\omega} \lambda \cup ‘mythical herb’). I intend to discuss these and other possible examples in detail elsewhere.
It would be reasonable to suppose a parallel development of ∗Ce/oL−LV−{ }^{*} C e / o L-L V- and ∗CoL⁡(hx)−LV−{ }^{*} \operatorname{CoL}\left(h_{x}\right)-L V- to Skt. Cūr VV - with degemination and compensatory lengthening (similar to the treatment in external sandhi, e.g., punar ramate >> puná ramate), but plausible examples are going to be hard to discover, since a Sanskrit word of the structure * Cūr V-, e.g., Cūra- can always be analyzed as a reflex of a plain thematic stem (quasi PIE *CoL-o- or *C cˉL−o−\bar{c} L-o- ), rather than a stem with a liquid suffix (quasi PIE * Ce/oL-Lo-).
21 Cf. Il. 3.26 ϑαλερoiτ′αζηα\vartheta \alpha \lambda \varepsilon \rho o i \tau^{\prime} \alpha \zeta \eta \alpha ‘vigorous young men’ or Hes. Th. 136 ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 45: …epsilon \rho o ̲́ v \delta^{\pri… ‘he hated his vigorous father’.
22 See Klingenschmitt 1982: 172; Mayrhofer 1986: 127 n. 118 and Clackson 1994: 118-20 with ample references to earlier literature.
23 See Hackstein 2002: 221; Driessen 2005: 58-60; LIV Add. s.v. * dhelh1d^{h} e l h_{1}-; Schumacher and Matzinger 2013: 968-9. An ee-grade of this root is found in Arm. det ‘herb, medicine’, as well as in Alb. djalē ‘young man’ ( <∗della−<TPdhelh1no−<{ }^{*} d e l l a-<{ }^{T P} d^{h} e l h_{1} n o- (n.) ‘offspring’, see Neri and Demiraj 2021). Zero-grade formations are found in Germanic (OHG tola (f.) ‘cluster or bunch of grapes’ << Proto-Gmc. * duló-) and Celtic (Welsh dail ‘leaves, foliage’ <∗dhlh1−ih27<{ }^{*} d^{h} l h_{1}-i h_{2}{ }^{7} ). An o-grade * dolh1d o l h_{1} - is found in OIr. duilne ‘foliage’ ( <TPdoliˉnijaˉ<{ }^{T P} d o l \bar{i} n i j \bar{a} ), Gaul. - δo∪λα\delta o \cup \lambda \alpha, and perhaps in Gk. ϑλ^λλααλα^δo∪ς\vartheta \hat{\lambda} \lambda \lambda \alpha \quad \alpha \lambda \hat{\alpha} \delta o \cup \varsigma, rˉ1φω^λλα\bar{r}_{1} \varphi \hat{\omega} \lambda \lambda \alpha and Lat. folium ‘leaf’.
24 Under this theory the /a/ in Gk. ParseError: KaTeX parse error: Expected 'EOF', got '́' at position 45: …epsilon \rho o ̲́ s (possibly from * ϑελερos<∗dhlh1\vartheta \varepsilon \lambda \varepsilon \rho o s<{ }^{*} d^{h} l h_{1}-ro- with later accent shift by analogy to oxytone ro-adjectives) would have to be imported from the nasal-infixed verbal stem ϑα^λλω\vartheta \hat{\alpha} \lambda \lambda \omega and/or the uu-stem adj. * ϑαλω^ς\vartheta \alpha \lambda \hat{\omega} \varsigma (fem. ϑα^λεια\vartheta \hat{\alpha} \lambda \varepsilon \iota \alpha ‘flourishing, rich’); Arm. dalar may be a regular reflex of ∗dhlh1* d^{h} l h_{1}-ro- (see Beekes 2003: 194; Kocharov 2018: 132).

Abbreviations

References

Allen, W. Sidney. 1953. Phonetics in Ancient India. Oxford: Oxford University Press.
Allen, W. Sidney. 1972. Sandhi. The Theoretical, Phonetic, and Historical Bases of WordJunction in Sanskrit. Second printing. The Hague: Mouton.
Beekes, Robert Stephen Paul. 2003. Historical phonology of Classical Armenian. In Armeniaca: Comparative Notes, ed. Frederik Kortlandt, 133-211. Ann Arbor: Caravan Books.
Bloch, Jules. 1951. La prononciation de rr en sanskrit. Bulletin de l’École française d’ExtrêmeOrient 44(1):43-45.
Brückner, Heidrun, and Gabriele Zeller, ed. 2007. Otto Böhtlingk an Rudolf Roth: Briefe zum Petersburger Wörterbuch 1852-1885. Bearbeitet von Agnes Stache-Weiske. Wiesbaden: Harrassowitz.
Byrd, Andrew Miles. 2018. The phonology of Proto-Indo-European. In Handbook of Comparative and Historical Indo-European Linguistics, vol. 3, ed. Jared S. Klein, Brian D. Joseph, and Matthias Fritz, 2056-2078. Handbücher zur Sprach- und Kommunikationswissenschaft 41/3. Berlin: de Gruyter.
Clackson, James. 1994. The Linguistic Relationship between Armenian and Greek. Oxford: Blackwell.
Clayton, John. 2018. Rounding of Indo-Iranian * R(H)R(H). Paper presented at the 30th 30^{\text {th }} Annual UCLA Indo-European Conference, Los Angeles, November 9, 2018.
Driessen, Michiel C. 2005. On the etymology of Lat. fulvus. In Sprachkontakt und Sprachwandel. Akten der XI. Fachtagung der Indogermanischen Gesellschaft,17. bis 23. September 2000, Halle an der Saale, ed. Gerhard Meiser and Olav Hackstein, 39-64. Wiesbaden: Reichert.
Gershevitch, Ilya. 1959. The Avestan Hymn to Mithra. Cambridge: Cambridge University Press.

Gonda, Jan. 1936. Zur Homonymie im Altindischen. Acta Orientalia 14:161-202. Reprinted in Selected Studies, vol. 3: Sanskrit. Grammatical and Philological Studies, 1-42. Leiden: Brill, 1975.
Hackstein, Olav. 2002. Die Sprachform der homerischen Epen: Faktoren morphologischer Variabilität in literarischen Frühformen. Tradition, Sprachwandel, Sprachliche Anachronismen. Wiesbaden: Reichert.
Hale, Mark. 1998. Post-lexical RUKI and the tisrā-Rule. In Mīr Curad: Studies in Honor of Calvert Watkins, ed. Jay Jasanoff, H. Craig Melchert, and Lisi Oliver, 213-222. Innsbrucker Beiträge zur Sprachwissenschaft 92. Innsbruck: Institut für Sprachen und Literaturen der Universität Innsbruck.
Hintze, Almut. 1994. Der Zamyād-Yašt: Edition, Übersetzung, Kommentar. Wiesbaden: Reichert.
Kavitskaya, Darya. 2011. Compensatory Lengthening: Phonetics, Phonology, Diachrony. New York: Routledge.
Klingenschmitt, Gert. 1982. Das altarmenische Verbum. Wiesbaden: Reichert.
Kobayashi, Masato. 2004. Historical Phonology of Old Indo-Aryan Consonants. Tokyo: Research Institute for Languages and Cultures of Asia and Africa.
Kocharov, Petr. 2018. A comment on the vocalization of word-initial and medial laryngeals in Armenian. In Farnah: Indo-Iranian and Indo-European Studies in Honor of Sasha Lubotsky, ed. Lucien van Beek, Alwin Kloekhorst, Guus Kroonen, Michaël Peyrot, and Tijmen Pronk, 123-136. Ann Arbor: Beech Stave.
Kümmel, Martin Joachim. 2014. Syllable- and word-related developments in earlier IndoIranian. In Syllable and Word Languages, ed. Javier Caro Reina and Renata Szczepaniak, 204-221. Berlin: de Gruyter.
Macdonell, Arthur Anthony. 1910. Vedic Grammar. Strassburg: Trübner.
Mayrhofer, Manfred. 1986. Indogermanische Grammatik. Bd. 1, 2. Halbband: Lautlehre [Segmentale Phonologie des Indogermanischen]. Heidelberg: Winter.
Narten, Johanna. 1982. Die vedischen Präsensstämme hrnāyá-, hrnīyá- und Verwandtes. Münchener Studien zur Sprachwissenschaft 41:139-149. Reprinted in Albino, Marcos and Matthias Fritz (ed.), Kleine Schriften, 263-269. Wiesbaden: Reichert, 1996.
Neri, Sergio. 2005. Riflessioni sull’apofonia radicale di proto-germanico *namō" ‘nome’. Historische Sprachforschung 118:201-250.
Neri, Sergio, and Bardhyl Demiraj. 2021. djāl eˉ−i.DPEWA:\bar{e}-i . \quad D P E W A: Digitales PhilologischEtymologisches Wörterbuch des Altalbanischen (15.-18. Jh.). https://www.dpwa.gwi. uni-muenchen.de/dictionary/?lemmaid=14739.
Nussbaum, Alan. 2015. Greek vōxap ‘lethargy’ and other weighty matters. Paper presented at the GSAS Workshop on Indo-European and Historical Linguistics, Harvard University, April 25, 2015.
Palermo, Massimo. 1993. Note sullo scempiamento di rr nel romanesco pre-belliano. Studi Linguistici Italiani 19:227-235.
Pinault, Georges-Jean. 1987-1988. Védique jir̀rvi-/ jívri-. Indologica Taurinensia 14:313-338. Reprinted in Vedica, vol. 1: Etymologica, 1-21. Paris: Les Cent Chemins, 2019.
Rājā Rādhākānta Deva. 1825-1857. Śabdakalpadrumah. 5 vols. Calcutta: Varanasi. Reprint 1967, Delhi: Motilal Banarsidass.
Renou, Louis. 1952. Grammaire de la langue védique. Lyon: IAC.
Ryan, Kevin M. 2017. Attenuated spreading in Sanskrit retroflex harmony. Linguistic Inquiry 48(2):299-340.
Ryan, Kevin M. 2019. Prosodic Weight: Categories and Continua. Oxford: Oxford University Press.
Schumacher, Stefan, and Joachim Matzinger. 2013. Die Verben des Altalbanischen: Belegwörterbuch, Vorgeschichte und Etymologie. Wiesbaden: Harrassowitz.

Vine, Brent. 2006. Autour de sud-picénien qolofitûr: étymologie et poétique. In La langue poétique indo-européenne. Actes du Colloque de travail de la Société des Études IndoEuropéennes, Paris, 22-24 octobre 2003, ed. Georges-Jean Pinault and Daniel Petit, 499517. Leuven: Peeters.

Watkins, Calvert. 2013. Aspects of the “expressive dimension” in Indo-European: Toward a comparative grammar of speech registers. In Proceedings of the 24th 24^{\text {th }} Annual UCLA Indo-European Conference, Los Angeles, October 26th 26^{\text {th }} and 27th ,201227^{\text {th }}, 2012, ed. Stephanie W. Jamison, H. Craig Melchert, and Brent Vine, 243-254. Bremen: Hempen.
Whitney, William Dwight. 1889. Sanskrit Grammar. 2nd 2^{\text {nd }} ed. Cambridge, MA: Harvard University Press.
Wüst, Walther. 1966. Altpersische Studien: Sprach- und kulturgeschichtliche Beiträge zum Glossar der Achämeniden-Inschriften. München: Kitzinger.