/sark(u)-/, sa5+ra/i-ku-
‘to make shoes (?)’
- Language
- Hieroglyphic Luwian
- Grammar
- verb
- ID
- 957
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/sark(u)-/, sa5+ra/i-ku-
‘to make shoes (?)’1.1.1Transmission
The lexeme is a hapax legomenon, attested once on a lead strip from Tabal dating to the mid- or late 8th century BCE.
[AH.B.]
1.1.2Forms
| inf. | /sarkuna/ | sa5+ra/i-ku-na | KULULU lead strip 2, §3.19 (Tabal) |
1.1.3Graphic Features
1.1.3.1Phonological Interpretation
The attested spelling can represent a number of phonological shapes, and its current interpretation is based on the suggested cognate link to Hittite šarkue/a-(mi) ‘to put on shoes’ and its word family (see Hawkins 2000a:513 and HEG S:909). The interpretation of the ending is secure, as infinitives in /-una/ are frequent, but see discussion of stem below.
[AH.B.]
1.1.4Meaning
The meaning was first suggested tentatively by Hawkins 2000a:513, who also mentions a possible cognate link with Hittite šarkue/a-(mi) ‘to put on shoes’ (followed by HEG S:909). In the context of the lead strip, the infinitive /sarkuna/ represents the purpose for which 14 sheep are brought to a certain town, followed by mention of a further allocation of 200 sheep to a /sarkunalla-/ ‘shoe maker?’.
Outdated interpretations of /sarkuna/ include ‘tribute’ (Meriggi & Poetto 1982a:108).
[AH.B.]
1.1.5Stem
Because the lexeme is a hapax legomenon, the shape of its stem is not entirely clear. It may be /sark-/ or /sarku-/, the latter representing a possible neo-root; cf. Sasseville 2021a:487.
[AH.B.]
1.1.6Derivatives
2. Reconstruction
2.1Reconstruction ‘shoe’
The HLuw. verb /sark(u)-/ ‘to make shoes(?)’ (inf. sa5+ra/i-ku-na) must be related to Hitt. šarku-(mi), šarkue/a-(mi) ‘to put on footwear’. The Hieroglyphic Luwian verb could be u-extended as in Hittite, which would be hidden before the infinitive ending /-una/, but this must be left open until further attestations are found. Based on the Hittite cognates, an amphikinetic root verb *sérk-u-/sr̥k-u-´ (šar-ku-zi, šar-ku-ut-ta) and a deradical verb in *-i̯e/o-, i.e. *sErk-u-i̯e/o- (šar-ku-i-ia-zi, šar-ku-ia-ad-du) (Oettinger 1979a:330, 335, EDHIL:735), can be posited tentatively for Proto-Anatolian. To which of the two formations the Luwian verb belongs remains undecided until further attestations are found.
Considering the similar looking root šark(u)- ‘to lift’, the semantic shift from ‘lifting’ to ‘making/wearing shoes’ has been interpreted in various ways. On the one hand, the sole of the shoe makes a person taller, but on the other hand, the action of putting on shoes (or boot-like shoes) requires a movement of pulling upwards (as explained by Ehelolf apud Sommer & Falkenstein 1938a:86, followed by Oettinger 1979a:335). Therefore, the original semantics of ‘lifting’ seemed a priori attractive, but rejected in HEG S:907 and EDHIL:735 and see below under Proto-Indo-European Etymology. If one takes the Hittite verb šarni(n)k-(mi) ‘to give compensation’ to belong etymologically here (thus EDHIL:736f.), one would have to assume a strong semantic shift; cf. under Proto-Indo-European Etymology.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
The neo-root PAnat. *sarku- may be connected with a u-stem adjective of unknown meaning. Although Hittite attests an adjective šarku- ‘eminent’ and Palaic provides verbal evidence for such a stem (cf. Pal. šarku- ‘to lift, raise’ [lemma link]), it is advisable to separate these from the lemma under investigation on semantic grounds (but cf. above sub PAnat. Reconstruction). While the Greek name Ἕρκυν(ν)α may be compared phonologically with PAnat. *sarku-, perhaps even with HLuw. /sarkunalla-/ ‘shoemaker’ (cf. below), its original meaning is obscure. Therefore, PAnat. *sarku- remains isolated and is thus not reconstructable for PIE.
The only word for ‘shoe’ which may be reconstructed for the proto-language is Core IE *kr̥h1pi- (EIEC:514f.) and is ultimately derived from the root PIE 2.*(s)ker- ‘to shear, scratch, cut off’ (LIV²:556f.). Thus it refers to the preparation of the material (esp. leather) used to produce such shoes. PAnat. *sarku- similarly may be traced back to a verbal action related to shoemaking, as the lemmata HLuw. /sark(u)-/ ‘to make shoes’ (cf. above) and /sarkunalla-/ ‘shoemaker’ indicate, and not to the putting on of shoes.For this Benennungsmotiv, cf. Gk. ὑπό-δημα ‘sandal’ from Gk. ὑπό ‘below’ and δέω ‘to bind’ or Ved. upā-nah- ‘shoe, sandal’ from Ved. upa ‘below’ and nah- ‘to tie, bind’ (Buck 1949a:428–430). However, words for shoes in particular and for items of clothing in general tend to be loans, cf. Buck 1949a ibid. et passim.The meaning of PAnat. *sarku- may thus tentatively be reconstructed as ‘±to make shoes’.
3.1.1Root etymology
PAnat. *sarku- ‘±to make shoes’ may be connected with Hitt. šarni(n)k-(mi) ‘to give compensation’ as well as Lat. sarcīre ‘to mend, repair’. These, together with their cognates (cf. sub Transmission), are usually traced back to a root PIE *serk- meaning ‘to repair, compensate (LIV²:536); to make good (García Ramón 2019a)’.On the question whether the root PAnat. *sark- ‘eminent’ [lemma link] belongs here, cf. ibid. and, e.g., García Ramón 2019a and EDHIL:734f., 736f.However, if PAnat. *sarku- ‘±to make shoes’ as well as some non-Anatolian material (e.g., Lat. sarcīna ‘bundle’, TochA sark ‘series’) is taken into account, the semantics may be concretized as ‘to weave’ or, more general, ‘to tie/sew together’ vel sim.Cf. similarly IEW:912 (“Flechtwerk, einhegen?”), EIEC:629 s.v. WALL (“to construct or repair a wall”); differently EIEC:108 s.v. CIRCLE (“make a circle, complete; make restitutions”).On the one hand, these techniques are used to mend woven/wattled material, to darn, hence ‘to repair, compensate’. On the other hand, they denote kinds of tailoring and shoemaking, viz. the weaving of bast fibres (→ sole, sandal)Otten 1971c:125 mentions seals that show herringbone patterns symbolizing shoe soles (sandals?). Cf. also Russ. lápot’ ‘bast-shoe’ besides lápit’ ‘to mend, repair’ (EIEC:110), which may resemble PAnat. *sarku- ‘shoe’ besides Lat. sarcīre ‘to mend, repair’ (and Hitt. šarni(n)k-(mi) ‘to give compensation’, etc.).or the sewing of leather (→ shoe, crakow).Otten 1971c:124 shows that crakows are known early on in Hittite tradition, whereas sandals are attested only later. Cf. also Lat. sūtor ‘showmaker’ < ‘sewer’ from Lat. suere ‘to sew’, etc.Thus, the underlying root is reconstructed as PIE *serk- ‘to tie/sew together’ (vel sim.), which may also be a predecessor of Hitt. šarku-(mi), šarkue/a-(mi) ‘to put on footwear’ via ‘to lace (a shoe)’ or ‘to strap (a sandal)’.
3.2Morphological Reconstruction
Although the Greek river/personal/divine name Ἕρκυν(ν)α may be equated with PLuw. *sarkuna-, on phonological grounds a putative derivational base of HLuw. /sarkunalla-/ ‘shoemaker’, it has to be excluded from the comparison because the name escapes a semantic analysis (*serku-no- ‘related to shoes’?). Therefore, PAnat. *sarku- ‘±to make shoes’ is not traced back to the proto-language. Morphologically, PAnat. *sarku- may theoretically be interpreted as a secondary root extracted from a u‑present pre-PAnat. *sérk-u-/*sr̥k-u-´, attested as Hitt. šarku-(mi) ‘to put on footwear’ and perhaps HLuw. /sark(u)-/ ‘to make shoes(?)’.Cf. similar neo-roots like Ved. dhanv- ‘to run, flow’ (LIV²:145 n. 3), jīv- ‘to live’ (LIV²:215 n.6), PToch. *śau- ‘to live’ (LIV²:216 n. 10), PIE ?*melh2u̯- ‘to grind’ (LIV²:433 with n. 1 s.v. ?*melh2u̯-), PIE *leu̯h3- ‘to wash’ (Melchert 2011a:130, cf. CLuw. lūwa- [lemma link].This u-present is not reconstructable for PIE due to the lack of non-Anatolian cognates.
3.2.1Root etymology
The root underlying PAnat. *sarku- ‘±to make shoes’ is reconstructed by Kloekhorst (EDHIL:735) as PIE *serku̯-, for which there is, however, no good evidence (cf. Beekes apud EDHIL:735).But cf. Meissner 2013a:17 on Kloekhorst’s comparandum Gk. ἁρπίς, -ῖδος ‘(kind of shoe)’. This may rather be connected with Gk. ἅρπη ‘sickle; (kind of predator bird)’ ← ‘sickle; to sickle, cut’ (cf. PIE *ser-p- ‘sickle’, IEW:911f.; on the semantics, cf. above) or with Gk. ἕρπω ‘to sneak’ (cf. PIE *serp- ‘to crawl, sneak’, LIV²:536) as *‘sneaker’ or *‘means to sneak’.Conversely, the connection with Hitt. šarni(n)k-(mi) ‘to give compensation’ and Lat. sarcīre ‘to mend, repair’ (contra EDHIL:735) is much more plausible on semantic grounds (cf. above). Therefore, the root is not likely to be reconstructed with a labiovelar, leading PAnat. *sarku- to reflect PIE *serk-u-. While in LIV²:536 the exact quality of the auslauting (not labialized) velar is left undecided, it is now possible to specify it as a plain velar on the basis of its reflex in certain languages like Luwian or Albanian. The other phonemes of the PIE root are unequivocal. The root can thus be reconstructed as PIE *serk- ‘to tie/sew together’.
[A.O.]
4. Transmission
Gk. river/personal/divine name Ἕρκυν(ν)α epithet of Demeter (at Lebadea, Boeotia), Ἑρκύνια ‘(Gk.) ἑορτὴ Δήμητρος, festival of Demeter’ (hapax, Hsch. ε 5953)The myth of Herkyn(n)a seems to be an aitiology meant to explain the cult of Demeter Herkyn(n)a and ultimately the river name Herkyn(n)a. The (folk‑)etymological element in the myth is the implied connection of Herkyn(n)a and the (caught and) escaped goose with Gk. ἕρκος, which can also denote a ‘catching net (for birds)’ (Od.22.469+), cf. LSJ:690 s.v. ἕρκος 5.
GEW III:93, LSJ:690
PIE
(?)PIE *sórk-o- ‘what is tied together’, cf. Gk. (?)ὅρκος m. ‘(the object by which one swears); oath (< restriction, obligation?)’ (Il.+) (GEW II:418f., GEW III:163, DELG:820f., EDG:1103f., LSJ:1252, LSJSuppl:231), Alb. gjarkëz ‘peritoneum’< *‘that which surrounds’, according to Adams (DTB²:766, EIEC:108, 629). Differently Orel (AED:130), who regards it as a “dialectal plural of qark” ‘circle’, which itself is a loan from Lat. circus ‘circle’ (cf. AED:354). Moreover, taking the meaning of the root into account that is presented here, it may originally have meant ‘histoid, tissue, web’.(AED:130), TochA sark ‘series’, TochB serke ‘series, cycle’TochB ṣärk- ‘to be good (vel sim.)’ rather belongs to PAnat. *sark- ‘eminent’ [lemma link].(DTB²:765f.)On (?)TochA särśi ‘backbone?’, TochB sark* ‘back (of the body)’, cf. PAnat. *sark- ‘eminent’ [lemma link].
Greek
Gk. ἕρκος n. ‘fence, enclosure’ (Il.+)
GEW I:561, GEW III:93, DELG:372f., EDG:460f., LSJ:690, 1252
Italic
Lat. sarcīre ‘to mend, repair; to restore’ (Plaut.+), sarcīna ‘bundle (of things tied together)’ (Plaut.+), adv. sar(c)tē ‘(Lat.) integrō, unbrokenly’ (hapax, Fest. 322M), (?)Umbr. adv. sarsite ‘whole, altogether’ (IgT VIb.11)
WH II:478f., EM:594f., EDLIL:539, Untermann 2000a:656f., Heidermanns 1999a:248 §145, OLD²:1864f., 1866On the radical -a- in Lat. (and maybe already in PItal.), cf. the references in LIV²:536 s.v. *serk/k̑- n. 1. EM:595 consider it to be expectable “dans un terme technique ; cf. sarpō, etc.”
Tocharian
TochB com. sarki-mpa ‘warp, (Germ.) Kettfaden, Kette, (Fr.) fil de chaîne’DTB²:740 seems to prefer the meaning ‘woof, weft’ (cf. also, e.g., Schmidt 1974a:207f. [“Einschlag”], Peyrot 2013a:708 on THT 3 b5), which is the yarn threaded through the warp. However, since sarki* is attested exclusively in the comitative, it rather denotes the fixed set of yarns through which the weft is threaded, thus ‘warp’.
DTB²:740f., CEToM s.v. sarkimpa