*i̯ek^u̯r̥/n-

Following a join of two Cuneiform Luwian fragments, which sheds light on the meaning of Luw. ikkuwar/n- ‘liver’, Sasseville 2021a:562f. connects it to PIE *i̯ek^u̯r̥/n- ‘liver’. The PIE accentual ablaut can no longer be seen in Luwian, which must have generalized either the strong or the weak stem. Both the phonological sequence *#i̯é- and *#i̯ḗ- would yield Luw. #ī-. For the unconvincing possibility of interpreting the general initial plene spelling ī- as the reflex of the initial *h₁, see Simon 2010b:261. However, the latter would trigger the lenition of the following labio-velar. The fortis status of -kkw- speaks at least for the generalization of the variant *#i̯ék^u̯- (as opposed to *#i̯ḗg^u̯-). It is also possible that there was a zero-grade of the root, i.e. *i̯ék^u̯/*ik^u̯én-, but this cannot be demonstrated by the synchronic evidence either. In the end, the forms attested so far for the Luwian word for ‘liver’ do not contribute to the original accentual paradigm. Therefore, a PAnat. pre-form can be reconstructed as such *i̯ek^u̯r̥/n-.

[D.S.]

PIE *(H)i̯ē̆́/ók^u̯-r/n- n. means primarily ‘liver’ (NIL:392, EIEC:356). Derivatives based on the r/n-stem noun, furthermore, exhibit the secondary meanings ‘roe, spawn, caviar’ (in Celtic and Balto-Slavic) and ‘calf’ (anat.; in Balto-Slavic). “Semantische Vermittlung kann über ‘leberartige, weichliche Masse’ erfolgt sein”, as argued in ALEW I:391 with lit., at least as far as the meanings ‘roe, spawn, caviar’ are concerned, but ‘calf’ might be explained via the same semantic path, with a different semantic specialization (ALEW I ibid.).

The PIE word for ‘liver’ was a neuter r/n-stem. Its exact reconstruction, however, remains a matter of debate: PIE *(H)i̯ē̆́/ók^u̯-r/n- vel sim. For a brief overview and discussion of several different reconstructions proposed so far (e.g., *(H)i̯ḗk^u̯-r̥/*(H)i̯ék^u̯-n-, *i̯ék^u̯-r̥/*ik^u̯-n-´, *i̯ḗk^u̯-r̥/*i̯ók^u̯-n-, et al.), cf. NIL:393 n. 1 with lit. CLuw. ikkuwar/n- n. ‘liver’ (nom. sg. ikkuwar, ergative ikkun-ant(i)-), which can now be added to the transmission, points to a PAnat. preform *i̯ék^u̯r̥/n- (see sub PAnat. Reconstruction). The Luwian form thus seems to require the reconstruction of an e-full grade *(H)i̯ek^u̯° for the PIE paradigm of ‘liver’, as do Ved. yákr̥t, Lat. iecur as well as Lith. dial. (j)ẽknos f. pl. ‘liver’ and Latv. dial. jȩknas f. pl. ‘id.’. It appears that Gk. ἧπαρ will then tip the scales when it comes to the reconstruction of the PIE paradigm: if its ē-vowel is inherited, it seems best to reconstruct a PIE ḗ/é-acrostatic paradigm *(H)i̯ḗk^u̯-r̥/*(H)i̯ék^u̯-n- n. ‘liver’; if the Greek ē-vowel, however, is a secondary innovation (e.g. due to an analogy to κῆρ ‘heart’, ἦτορ ‘heart’, and σπλήν ‘spleen’, which also contain η),Cf., e.g., Kloekhorst 2014a:142 with older lit.an ó/é-acrostatic paradigm *(H)i̯ók^u̯-r̥/*(H)i̯ék^u̯-n- can alternatively be reconstructed. Both paradigms would also account for YAv. yākarǝ n. ‘liver’, the long ā-vowel of which at first glance points to a preform *(H)i̯ḗk^u̯-r̥ or *(H)i̯ók^u̯-r̥ (cf. Neri 2003a:26, NIL:393 n. 6). However, it seems not even clear whether the form yākarǝ (attested in Frahang-ī ōim 189) can be taken seriously, as one of the two manuscripts transmitting the Frahang-ī ōim attests a form yakarǝ with short a in the first syllable, which is in line with the attestations of all other Iranian languages that point to PIran. *i̯akar- (de Vaan 2003a:68). Yet even if we opt for the lectio difficilior yākarǝ, its long ā need not be inherited from the protolanguage, but could also be due to a secondary lengthening Av. a > ā, which in Avestan does indeed occur quite frequently in initial syllables (ALF:56, de Vaan 2003a:69). As a consequence, de Vaan 2003a:69 comes to the conclusion that, “The cost of positing Avestan *yākarǝ would be quite high, since nowhere else in Iranian or Indic do we find an ablaut grade *yākr̥ of the word for ‘liver’.” In any case, it seems that the long vowel of YAv. yākarǝ cannot be considered a reliable piece of evidence. – Lat. iocineris gen. sg. ‘of the liver’ with o-grade ioc° (whence finally also nom./acc. sg. iocur alongside iecur) might be metathesized from iecinoris (Klingenschmitt 2005a:332f., Meiser 1998a:142, NIL:394 n. 10) and is thus not necessarily indicative of a PIE o-grade form *(H)i̯ok^u̯°. An o-grade root, however, seems to be reflected by Lith. dial. (j)ãknos f. pl. ‘liver’, Latv. aknas f. pl. ‘id.’, and OPr. ⁺jagno f. ‘id.’ (besides e-grade Lith. dial. (j)ẽknos f. pl. ‘liver’, Latv. dial. jȩknas f. pl. ‘id.’), which point to pre-PBalt. *(H)i̯ok^u̯-n-eh₂- (NIL:392). However, the o-grade could very well reflect a Baltic innovation rather than a PIE inheritance. Petit 2004a:118 comes to the conclusion “que le degré *o résulte de la création secondaire d’un néo-singulier *i̯ok^w-r/n-, d’après un modèle général (type *u̯od-r/n- ‘eau’).” – Summing up the preceding brief discussion, it seems best for the time being to reconstruct the PIE paradigm for ‘liver’ as either acrostatic *(H)i̯ḗk^u̯-r̥/*(H)i̯ék^u̯-n-With a potential loc. sg. **(H)ik^u̯-én if it had the ablaut structure R(∅)-S(é) (R = root, S = suffix) as in the case of the ó/é-acrostatic nouns, cf. PIE *u̯ód-r̥/**u̯éd-n- ‘water’ with loc. sg. *ud-én (Ved. udán), according to Schindler 1994a:397.(if Gk. ἧπαρ reflects an inherited lengthened grade) or proterokinetic *(H)i̯ék^u̯-r̥/*(H)ik^u̯-én-With a potential loc. sg. **(H)ik^u̯-ē̆́n. A proterokinetic paradigm *(H)i̯ék^u̯-r̥/*(H)ik^u̯-én-, loc. sg. **(H)ik^u̯-ḗn would be comparable with, e.g., PIE *mén-ti-/*mn̥-téi̯- ‘mind’, loc. sg. *mn̥-tḗi̯ (for the loc. sg., cf. Klingenschmitt 2005a:319 with n. 30, followed by Schaffner 2001a:437 with n. 8).(if Gk. ἧπαρ owes its η to an analogy). These reconstructions are based on the premises that all o-grade forms reflected by Latin and Baltic are of secondary origin and that YAv. yākarǝ is not a reliable form.

[T.S.]

Ved. yákar- (yakn-) n. ‘liver’ (nom./acc. sg. yákr̥t [AV], abl. sg. yaknás [RV], instr. sg. yaknā́ [VS]); YAv. yākarǝ nom./acc. sg. n. ‘liver’, MPers. /ǰagar/ ‘id.’, NPers. ǰigar ‘id.’, etc.Cf., e.g., EWAia II:391, NIL:392, 393f. n. 2, 3, 6, 7.

Gk. ἧπαρ, -ατος n. ‘liver’ (Il.+)Cf., e.g., GEW I:639, NIL:392f. with n. 5, EDG:522.

Lat. iecur n. ‘liver’ (Plaut.+), also iocur n. ‘id.’ (Liv.+), gen. sg. iecinis, iecinoris, iocineris, iocinoris.Cf., e.g., EDLIL:296, NIL:392, 393 n. 4, 394 n. 10, 11.

MIr. iuchair f., usually pl. (e.g., acc. pl. iuchra) ‘roe, spawn’Cf., e.g., NIL:392, 394 n. 13, EDPC:171f.

OPr. ⁺jagno (lagno) ‘liver’, Lith. dial. (j)ẽknos f. pl. ‘id.’, dial. (j)ãknos f. pl. ‘id.’, Latv. dial. jȩknas f. pl. ‘id.’, aknas f. pl. ‘id.’; Lith. ìkras m. ‘fish-egg, (anat.) calf’, pl. ‘roe, spawn, caviar’, Latv. ikri m. pl. ‘roe, spawn, caviar, (anat.) calves’Cf., e.g., NIL:392f., 394 n. 8, 9, 394f. n. 14, EDBIL:199 s.v. ikras, 211 s.v. jeknos, ALEW I:391 s.v. ìkras, 413f. s.v. jẽknos.

CS ikra f. ‘roe’, Russ. ikrá f. ‘roe, spawn, caviar, (anat.) calf’, SCr. ȉkra f. ‘roe’, etc.Cf., e.g., NIL:393, EDSIL:210.