*ā́tr-/-n-

Comparative evidence drawn from Luwian, Lycian and Carian allows for a Proto-Luwic reconstruction. The etymological connection between Hierglyphic Luwian /atr(i)-/ and Lycian atra-/atla- was suggested by Hawkins 1989:190 n. 7 and accepted by Melchert 1994a:372, Hajnal 1995a:110 n. 99, van den Hout 2002a:174 and Kloekhorst 2006a:104. For the connection with Carian otr-, see under the respective lemma.The previous etymological connection with Hitt. ēšri- (n.) (Oettinger 1986a:28 n. 30) is outdated because of the unsurmountable phonological difficulties that it entails. The connection of Carruba 1995a:150f. with Greek ἐταῖρος is likewise outdated, cf. EDG:473f.

The place of accentuation of the Proto-Luwic pre-form of the Luwian, Lycian and Carian cognates can only be seen in Car. otr-, where the vowel goes back to an accented vowel (Adiego 2007a:258). However, the Anatolian preform cannot have been only barytone, since a lenition of the alveolar would have otherwise occurred, *ā́t-r- > *ā́d-r-. Since none of the three languages concerned shows the lenited outcome, one must posit two accent variants. The Luwic evidence speaks for the existence of a paradigm, where analogical pressure would have restored the non-lenited variant from *ā́d-r- to *ā́t-r-, thus Car. otr-.The sequence *-éh₁- becomes Proto-Anatolian *æ̅́ and causes lenition of a following obstruent (Melchert 1994a:56, 67f.).

While the Luwian outcome of *ā́tr- belongs to the paradigm of mutated stems, the Lycian one belongs to the a-stems (< *-eh₂-). Although neuter stems in Luwian do not undergo i-mutation, the reconstruction of an original neuter stem cannot be excluded, if one assumes that a change in gender happened first. The reason for such a development to a secondary common (or animate) gender may have been caused by the shift of the semantics, i.e. (a kind of organ) > ‘person, self’. Therefore, even if a Proto-Luwic common gender stem is reconstructed here, an original Proto-Anatolian neuter heteroclitic stem is very likely considering the alternation found in Lycian as well as the n-variant in Luw. /attan(i)-/ (see under reconstruction). For the Proto-Indo-European perspective see Part 2 below.

[D.S.]

The basic meaning of PIE *h₁éh₁tr̥/*h₁(e)h₁tén- n. is difficult to determine. The individual languages reflect meanings such as ‘vein, sinew, intestines, (lower) abdomen, stomach, heart, soul, self’. It seems possible that the basic meaning was something like ‘the interior’ or – more concrete – ‘intestines’ (cf. IEW:344). Due to further semantic specialization in the individual languages the word was then used to designate, on the one hand, various inner organs such as ‘abdomen’, ‘vein’, etc. In Anatolian and Greek, on the other hand, more figurative meanings such as ‘mind, soul, self’ vel sim. can be found.

According to Humbach & Faiss 2016a:31–33, PIE *h₁éh₁tr̥ (in their notation: “*ētr̥”) is also preserved in Av. ātar- m. ‘fire’. In their view, the meaning ‘fire’ developed secondarily from ‘heart, soul, self’: “Eine wesentliche Rolle spielt bei diesem Vorgang sicherlich das Bild vom Feuer als dem Wesen der Seele, dem Selbst und der Potenz der angesprochenen Gottheit, es ist die immaterielle Manifestation der Präsenz Ahura Mazdās” (Humbach & Faiss 2016a:31f.). To bolster their view Humbach & Faiss 2016a:107 refer in particular to Y 36.3: ātarš vōi mazdā̊ ahurahiiā ahī mainiiuš vōi ahiiā spǝ̄ništō ahī “you are indeed the Fire of Ahura Mazdā, you are indeed his most Beneficent Spirit”. While the simple form Iran. *ātr̥-/*āθr- received the secondary meaning ‘fire’, the primary meaning ‘heart, soul, self’, according to Humbach & Faiss 2016a:32f., could still be preserved in the second member of the Avestan compounds x^vāθra- n. ‘wealth, happiness, paradise’ and dužāθra- n. ‘unhappiness, misfortune’ if these are to be analyzed as hu- + -āθr-a- and duž- + -āθr-a-, with -āθr-a- meaning something like “materieller oder/und seelischer Zustand” (Humbach & Faiss 2016a:32f.). As far as the root and primary meaning of Av. āt-ar- are concerned, Humbach & Faiss 2016a propose an etymological connection with Ved. āt-mán- m. ‘soul, self’ (RV+): “Das führt zu der parallelen Segmentierung griech. ēt-r und avest. āt-r- versus altind. āt-mán-, alle drei mit ēt bzw. āt als Träger der ältesten fassbaren Bedeutung der etymologischen Gruppe” (Humbach & Faiss 2016a:33). However, it is not certain that Av. ātar- forms an equation with Gk. ἦτορ. Semantically, it seems more straightforward to trace Av. ātar- ‘fire’ back to a preform *h₂éh₁-tr̥-, a neuter tr-stem formed to the root PIE *h₂eh₁ ‘to be hot’ (cf. LIV²:257 n. 1).For the original neuter gender of Av. ātar-, cf. ALF:152: “Es handelt sich wohl um die Maskulin-Flexion eines ursprünglichen Neutrums *ātr̥-”. Cf. also Humbach & Faiss 2016a:32. However, the compound member Av. -āθra- contained in x^v-āθra- and duž-āθra- might indeed be a reflex of a form *-h₁e/oh₁tr-o- and contain the noun *h₁éh₁tr̥-, which is attested in the second member of compounds in Germanic and Celtic as well (cf. above sub 1. Transmission).

Morphologically, the forms attested suggest the reconstruction of a PIE proterokinetic r/n-stem *h₁éh₁tr̥/*h₁(e)h₁tén- vel sim. In Anatolian, there are still traces of both the r- and the n-stem – cf., e.g., HLuw. /atr(i)-/ ‘person, self’ beside /attan(i)-/ ‘soul, self’. Outside Anatolian, however, there are only traces of the r-stem *h₁éh₁tr̥, which is best preserved in Gk. ἦτορ ‘the interior, ‘heart’’, with Aeolic °ορ < PIE *°r̥ (GEW I:645 with lit.). In Greek, there is also an o-stem ἦτρoν ‘abdomen (esp. the lower part of it)’ < *h₁éh₁tr-o-. This might be a thematization of the type seen in Gk. δάκρυ n. ‘tear’ : δάκρυ-ον n. ‘id.’, the latter being a secondary thematic singular backformed to the athematic neuter plural δάκρυ-α (then perhaps sg. ἦτορ : pl. *ἦτρ-α → sg. ἦτρ-oν). However, the semantic difference between ἦτορ (‘the interior, ‘heart’, soul, etc.’ < *‘intestines’?) and ἦτρoν (‘abdomen’) could also suggest a ‘real’ derivation. In this case, ἦτρoν might have been derived at a stage when ἦτορ still had its original meaning: i.e., ἦτορ n. *‘intestines’ (whence later ‘the interior, ‘heart’, mind, soul, etc.’) → ἦτρoν n. ‘abdomen’, with ἦτρ-o- being originally perhaps a possessive or genitival adjective ‘having/containing the intestines’ or ‘pertaining to the intestines’ vel sim. For a similar derivation, cf. perhaps Gk. ποιμήν m. ‘herdsman, shepherd’ → ποίμν-η f. ‘flock, herd’, literally perhaps *‘what belongs to the shepherd’ vel sim.And then maybe the substantivized feminine of an o-stem adj. *ποιμν-ó- ‘belonging to the shepherd’ vel sim.

The difference þ : đ in PGerm. *ēþrō- : *ēđrō- points to two accent variants: PGerm. *ḗþrō- : *ēđrṓ-, “and this may point,” according to EDPG:120, “to an originally mobile paradigm, e.g. sg. *h₁éh₁t-r, pl. *h₁h₁t-ṓr.” As for the root etymology, Kroonen (EDPG:120) connects PGerm. *ēþrō- : *ēđrō- ‘vein, rivulet’ with PGerm. *ēþman- : *ēđman- m. ‘breath’, which is related to Ved. ātmán- m. ‘breath, soul, self’ (RV+; cf. EWAia I:164f.); cf. also Humbach & Faiss 2016a:31–33. Such a connection suggests a morphological analysis PGerm. *ēþ-rō- : *ēđ-rō-, PGerm. *ēþ-man- : *ēđ-man-, Ved. āt-mán-, with a root PGerm. *ēþ/đ, Ved. āt < PIE *h₁eh₁t. The exact meaning of this root, however, remains unclear.

[T.S.]

Hom. ἦτορ n. ‘the interior, ‘heart’, mind, etc.’For the meaning, cf. GEW I:645 with lit., LfgrE II:943 (“Inneres, ‘Herz’, Gemüt”). (Il.+), always nom./acc. sg., but dat. sg. ἤτορι in Simon. 37.6 codd. Ath.; cf. also the compound μεγαλ-ήτωρ, -ορος ‘great-hearted, magnanimous’ (Il.+) and the derivative Ion.-Att. ἦτρoν n. ‘abdomen (esp. the lower part of it)’For the word formation, cf. Schwyzer 1939a:461 and see below sub 3. Morphological Reconstruction.

PGerm. *ēþrō-, *ēđrō- f. ‘vein, rivulet’: cf., e.g., ON ǽđr, gen. ǽđar f. ‘vein’Cf. EWAhd I:55 (with further lit.): “da man das ausl. -r in aisl. ǽđr als nom. sg.-Endung mißdeutete, so wurden die übrigen Kasus zu einer reduzierten Stammform ǽđ-, entsprechend den iō- bzw. i- (und ō-)Stämmen, gebildet, also gen.sg. ǽđar, dat.acc. ǽđi, nom.pl. ǽđir, ǽđar”. Cf. also IEW:344, EDPG:120., OE ǣdre, ǣder f. ‘river; vein, artery; nerve, sinew’, pl. also ‘kidneys’, OFris. eddere, ēdere f. ‘vein’, OHG ād(a)ra, MHG āder, ādre f. ‘vein, sinew, muscle’, pl. ‘intestines, bowels’Cf. EWAhd I:54–57 s.v. âdra., OS (only in compounds) wintathren ‘arteriae’, ināthiri ‘intestines’, hartinnēth(e)re ‘viscera, vitalia’, hertāthere ‘fibrae’, etc.For these and further attested forms, cf., e.g., IEW:344, EWAhd I:54–57, EDPG:120. 

OIr. inathar m. ‘entrails, intestines, bowels’< *en-ōtro- < *en-h₁oh₁tr-o-, according to IEW:344, EDPC:115; differently Kroonen, EDPG:120, who traces OIr. inathar as well as MCorn. inederen ‘extum’ back to a preform “*h₁en-utro-” and separates them from the PIE stem *h₁eh₁tr-.