*ne-g̑nh₁-o-

The reconstruction of the word for ‘brother’ (Hitt. negna-, and with the regular Luwic loss of *g/g̑, CLuw. *nān(i)- and Lyc. nẽne/i-) must also take into consideration the Hittite word for ‘sister’, i.e. nega- c., which has led scholars to posit various ideas (Neumann 1974b:281 “*naniga- > niga-”, HEG N:297-301 with further literature, EDHIL:601 “*nego- → *neg-no-”), also in pappanika- ‘paternal sister’ (HED N:105f.). An overview of the research literature can be found in Steer 2015a:43-45. Considering first the word for ‘brother’, it is highly tempting to see the PIE root *g̑enh₁- ‘to be born’ (thus also Steer 2015a:45) in the zero-grade, i.e. in ne-gn-a-, which is often used in other Indo-European branches to build words for kinship and siblings (HEG N: 302, HED N:108 and the Proto-Indo-European lemma below). The zero-grade of this root, i.e. *gnh₁-, would also be attested in Palaic wāšukini-/wāšuniki-. Unfortunately, the Hittite word for ‘sister’ neka- cannot be readily derived from this pre-form, whereas the Luwian (and Lycian) word with the compounded nāna-šr(i)- can be easily accommodated (see CLuw. *ašr(i)-). On the other hand, if the Hittites reinterpreted negna- ‘brother’ synchronically as neg-na-, they might have secondarily built neg-a- ‘sister’ via back-formation. This assumption would allow us to keep the Indo-European etymology. Another approach that also keeps the same etymology is presented by Steer 2015a:51-54, who argues that Hitt. nega- c. reflects the generalization of the nominative singular of an old n-stem paradigm *nega(n)- with reinterpretation as an a-stem, whose credibility is diminished by the fact that this did not happen in e.g. the word for ‘eagle’, i.e. ḫāra(n)- c. In any case, he further derives negna- ‘brother’ from the original word for ‘sister’ with the possessive suffix *-ó-, i.e. nega- ‘sister’ → negn-a- ‘belonging to the sister, brother’.

Contra Neumann 1991b:63f., negna- cannot go back to *ni-g̑nh₁-o-, because the -i- in the preverb would have to be reflected in Luwian and Lycian as -i- (thus also Steer 2015a:48). The Luwic evidence speaks for *ne-. Therefore, a better hypothesis would be the one of Dunkel (LIPP II:551), who suggests seeing the particle *ne ‘near to, beside’, thus *ne-g̑nh₁-o- ‘born beside, sibling’ → ‘brother’. On the other hand, Steer 2015a:48-52 suggests a more concrete derivation starting with the prefix *en- ‘in’, presenting a plausible morphological chain with various ablaut patterns; cf. below.

[D.S.]

A second compound member *°-g̑n(h₁)‑o‑ ‘born’ is well attested in many branches of the Indo-European family. According to Steer 2015a:43–60, Hittite nega- c. ‘sister’, negna- c. ‘brother’, Lycian A nẽne/i- c. ‘brother’, Cuneiform Luwian *nān(i)- ‘brother’ have an allomorph of *en-, *n̥- ‘in’ as their first member (see below). In this regard, it may be significant that although a variety of kinship terms are composed of semantically equivalent members (Ogam Irish INIGENA f. ‘daughter’, Old Irish ingen f. ‘daughter, girl’ (*eni-genā-), Ancient Greek ἔγγονος m., f. ‘grandson, granddaughter’, ἐγγόνη f. ‘grandaughter’ (*en-gono-, *en-gonā-), ἶνις m., f. ‘son, daughter’ (*en-gni-); see Steer 2015a:47), Anatolian is the only branch in which such a formation ‘born within (sc. the same house, family)’ would have come to signify ‘brother’ and ‘sister’. For this semantic reason (and also a formal one; see below) it might be preferable to identify the first member *ne-° with the adverb *ne ‘near, close’ (LIPP II:551). Despite the weak evidence for the latter, such a compound would explain the meaning ‘brother’ more satisfyingly (cf. Greek κασίγνητος ‘brother’ with first member *km̥tí ‘with’; LIPP II:426).

The loss of root-final *h₁ (perhaps also *h₂ and *h₃) in the second member of a compound is regular (cf. Mayrhofer 1986a:129; Schrijver 1991a:328–330; Kümmel 2007a:334f.) and has been known as the “νεογνός rule” (cf., e.g., Zair 2012a:255).

Steer 2015a:43–60 reconstructs a compound *n̥-g̑enh₁- ‘born within’ that was substantivized to an amphikinetic *né-g̑onh₁- ‘the one born within (sc. the family)’ > Hittite nega- c. ‘sister’. It is far from clear, however, that *en-° ‘in’ had a compositional allomorph *n̥- (the only good example being Greek ἀκαρός ‘brain’ < *n̥-k̑r̥h₂-ó- ‘being inside the head (κάρᾱ)’ – note the absence of the νεογνός rule with *h₂), and even less clear that such a *n̥-° was upgraded to *né-° (and not simply *én-°) in “amphikinesis” (i.e., internally derived amphinetic substantivizations based on hysterokinetic adjectives), if such an “amphikinesis” was even possible with compounds in the first placenote that compounds cannot be “hysterokinetic” in the true sense of the word, as this terminology was devised for simplex athematic nouns and adjectives.

It is also unexpected that the word for ‘sister’ then served as the derivational basis for the word for ‘brother’. In view of these difficulties, it might be preferable to see ‘brother’ (in the interpretation mentioned above at the end of II. Semantic Reconstruction) as the basic term and ‘sister’ as somehow secondarily back-formed from it within Hittite (perhaps in children’s speech), though the exact mechanisms remain unclear.

[S.H.]

Hittite negna- c. ‘brother’, nega- c. ‘sister’, Lycian A nẽne/i- c. ‘brother’, Cuneiform Luwian *nān(i)- ‘brother’, Hieroglyphic Luwian /nanasr(i)-/ c. ‘sister’.See EDHIL:601; Steer 2015a:43–60.

Gaṛhwālī nauno ‘son’, naunī ‘daughter’ < *nava-jña-.Cf., e.g., EWAia I:568.

Young Avestan ā-sna- ‘innate’cf., e.g., EWAia I:568., perhaps Proto-Iranian *gā̆u̯aźna- ‘deer’ < *gā̆u̯a-źna- ‘of the bovid kind’ (YAv. gauuasna-, NPers. gavazn, etc.)thus Michael Weiss, p.c..

Ancient Greek νεο-γνός ‘new-born’ (h. Hom, Hdt., Hp., A., E., etc.).Cf., e.g., GEW I:307f. s.v. γίγνομαι.

Latin benignus ‘generous, kind’ (Pl.+; *du̯eni-g̑n(h₁)-o- ‘born good, good by nature’), malignus ‘unkind, ungenerous’ (Pl.+), prīvignus ‘stepson’ (< ‘born separately’), singulus ‘one at a time’ (Pl.+; dissimilated from *sem-gno- < *sm̥-g̑n(h₁)-o- ‘born at the same time’), bignae ‘twins’ (Paul. Fest.; < *du̯i‑g̑n(h₁)‑o‑).Cf., e.g., WH I:600; Nussbaum 2017b; Weiss 2020a:388.

Gaulish -gnos, -cnos in names (Aicio-gnus, Alle-cnus, Biti-cnos, Cintu-cnus, Diuuo-gna, Uelu-gnus, etc.), andogna f. ‘indigenous’, anandogna f. ‘foreign’, Ogam Irish -AGNI gen. sg. (e.g., COIMAGNI), Old Irish -án, -én (diminutive suffix).Cf., e.g., Delamarre 2003a:178; Schaffner 2005a:383.

Gothic niu-klahs ‘childish’ (dissim. from *niu-kna-ha-), ainakls ‘single’ (*aina-kna-).Cf., e.g., Schaffner 2005a:383.

(?) Phrygian ουεγνω < *su̯e-g̑n(h₁)‑o‑ ‘self-begotten’.Cf., e.g., NIL:139.