kappiša

All secure attestations of this word are found in tablets concerning the ḫišuwa- festival text, dated to the NH period.

[E.R.]

nom./acc. sg.	kap-pí-ša	KUB 25.48 rev. iv 9´	(CTH 628, NS)
		CHDS 5.14 rev.? iv 7´	(CTH 628, NS)
		KBo 25.163 rev. v 13´	(CTH 628, NS)
		KUB 40.102 obv. i 11´	(CTH 628, NS)
		CHDS 4. 42 5´	(CTH 628, NS)
	kap-*pí*-ša	KUB 32.128 obv. ii 17	(CTH 628, NS)
	kap-pí-ša[	KUB 25.42 obv. iii 8´	(CTH 628, NS)
	kap-pí-[ša]	KBo 38.16 rev. iv 22´	(CTH 628, NS)
	kap-pí-š[a?]	KBo 20.116 obv.? 10´	(CTH 628, NS)
	⸢kap-pí⸣-š[a]	KBo 25.163 rev. v 2´	(CTH 628, NS)
	ka[p-pí-ša]	KBo 15.49 obv. i 5´	(CTH 628, NS)
	[ka]p-⸢pí⸣-ša!	CHDS 5.14 rev.? v 12´	(CTH 628, NS)
	[kap]-⸢pí⸣-ša	KUB 25.48 rev. iv 30´	(CTH 628, NS)
nom./acc. sg.?	kap-pí-ša	VSNF 12.24 7´	(CTH 628, NS)
unknown	[ka]p-pí-ša	DAAM 1.17 obv. 5	(CTH 628, NS)
	kap-pí-[ša]	KBo 31.179 obv.? 2´	(CTH 628, NS)

The fragmentary attestation of 1 ga.kin.ag kap-p[(í) in Bo 3648 obv. 10 (see Taracha 2000a:28 n. 17, 66, 97, Groddek 2011b:124 n. 90) is completed by its duplicate Bo 3351 obv. i 16 and has hence been revealed not to contain kappiša but Hitt. kappi ʻsmallʼ (against HW 2nd ed. K:116, correct Groddek 2011b:124 n. 90).

[E.R.]

Checked against the photographs in the Mainzer Fotoarchiv.

[E.R.]

In the literature, two different meanings are offered, often treated separately. With the exception of KUB 40.102 obv. i 11´, all attestations of kappiša modify the Sumerogram ga.kin.ag ʻcheeseʼ, mostly preceded by the numeral 1. The word has thus been identified as a designation for a special kind of cheese and commonly left untranslated, either as a technical term or in the absence of a secure meaning (HEG A-K:493, HED K:63, HHw:72, HW 2nd ed. K:115-116, Taracha 2000a:116, CHD L-N:184, Tischler 2008a:79, Groddek 2011b:123-124). Taracha 2000a:97, however, opts for a more specific meaning ʻcheese bread or cheese cakeʼ, based on parallels in other lists of offerings.

In KUB 40.102 obv. i 11´ (and its duplicate CHDS 4.42 5´), kappiša precedes a list of (baked?) goods, itself quantified by the numeral 15. The relevant passage reads as follows: nu 15 kap-pí-ša [šà].ba 3 ku-gul-la-aš 3 ke-el-lu [3 za]-am-mu-ra-an 3 še-e-na-an 3 ma-ri-iš ⸢i⸣-ia-an na-at a-na dingir-lim ḫi-in-ik-zi “15 kappiša, including 3 disk-shaped (items), 3 kellu-items, [3 za]mmura-pastries, 3 figurines (and) 3 spear-shaped (items), (are) made. He offers them to the god” (KUB 40.102 obv. i 11´-14´, cf. CHD L-N:184, Haas 1992a:106-107, HW 2nd ed. K:116-117). Hence, it is safe to assume that kappiša here translates as the amount or measurement of the objects that follow.

Accordingly, it is commonly assigned the meanings ʻamount’, ʻmeasureʼ, ʻportionʼ, and ʻrationsʼ (Haas 1992a:105, Ünal 2007a:312, Görke 2010a:203, HED K:63). A different view is held by HW 2nd ed. K:115 translating it as ʻ(some sort of) pastriesʼ. Ünal 1996a:48 n. 106 (followed by HHw:72 and Tischler 2008a:79) associates kappiša with Hitt. kappi- ʻsmallʼ, but see under Stem. Inconclusive and without translation are Hoffner 1974a:171 and Groddek 2011b:124. The latter argues against any connection to Hitt. kappi- ʻsmallʼ, which, according to him, is based on a scribal error leading to a confusion of tur ʻsmallʼ and kappiša in KBo 25.163 rev. v 13´.

Assuming an etymological connection with Hitt. ^DUGkappi- c. ʻbowl (vel sim.)ʼ and Akkadian kappu ʻid.ʼ, Melchert (DCL:107) suggests the meaning ʻbowl (or similar container)ʼ. In this light, the putative use as a measure in KUB 40.102 obv. i 11´ and as a modifier of ga.kin.ag elsewhere can be connected with one another. This seems especially attractive since kappiša is attested only in the ḫišuwa- festival. As the modifier of ga.kin.ag, kappiša should probably be understood as an apposition ʻas bowl, in bowl-shape (vel sim.)ʼ, specifying either the type of cheese, which requires a term for some sort of container, or the amount fitting into one such container. Regarding the traditional interpretation ʻportions (vel sim.)ʼ in KUB 40.102 obv. i 11´, a basic meaning ʻbowl (vel sim.)ʼ is contextually equally appropriate. It is important to note that, with the exception of ^NINDAzammura/i-, the other objects listed in KUB 40.102 obv. i 11´-14´ do not necessarily have to represent models made of dough (contra HW 2nd ed. K, Groddek 2011b, Görke 2010a, DCL:s.v.), but the consistent absence of a determinative ^NINDA points to the objects proper. As suggested by Susanne Görke (pers. comm.), the passage could thus be translated as “15 bowls, including 3 (with) disk-shaped items, 3 (with) kellu-items, [3 (with) za]mmura-pastries, 3 (with) figurines (and) 3 (with) spear-shaped items, (are) made.” Compared with Hitt. ^DUGkappi- ʻbowlʼ however, which is attested multiple times in CTH 628 and is always accompanied by its respective determinative, we find that ^DUG is curiously missing for kappiša. This fact points to the use of kappiša as a measure, which had become lexicalized and thus obscured in its connection to the bare meaning ʻbowlʼ (cf. engl. cup). Nevertheless, given its status as a borrowed technical term (see under Origin), which appears to be unique to the Kizzuwatnean ḫišuwa- festival, it is also likely that no determinative was attributed by the scribe. With all these caveats in mind, the meaning ʻbowl (as measure)ʼ is cautiously chosen here because it has the advantage of covering all attestations.

[E.R.]

The lexeme kappiša following ga.kin.ag is analyzed as an a-stem adjective by several authors, either as the bare stem or a neuter plural form (Hoffner 1966b:29, Tischler 1982c:34, HEG A-K:493, HHw:72, Tischler 2008a:79).Tischler (HHw:72 and Tischler 2008a:79) wants to identify kappiša in connection with ga.kin.ag as the nom. sg. c. Hitt. kappiš ʻsmallʼ with the conjunction =ya attached, but in addition to the lack of the geminating effect, the sole conjunction in the middle of a list is unexpected. Other scholars argued for a Luwian lexeme kappi(t)-. Neu 1983b:305 (followed by Haas 1992a:107 and HED K:63) identifies the neuter particle =ša, which is attached to the nom. and acc. case forms in Luwian. Melchert (DCL:107) accepts this proposal and specifies kappiša as an underlying i(t)-stem kappi(t)- (see Starke 1990a:46 for the stem type).However, while we have to keep in mind that we are dealing with attestations in one text group only, it is unexpected that, in kappiša, this particle =ša is attested consistently, given the frequent variation of =ša with zero in the historical sources of the 2^nd millenium BCE. As an alternative, one could analyze the formation as a plurale tantum s-stem, but since numerals with collectives and pluralia tantum are specified by the Akkadograms -nūtu and -tapal, the interpretation of an alleged s-stem can be discarded. It is very likely that the participle iyan ʻmadeʼ at the end of the list in KUB 40.102 obv. i 14´ refers to kappiša, which makes a compelling case for a nom./acc. singular neuter form. This speaks in favour of an analysis as an it-stem. Accordingly, since neuter nouns show singular agreement with numerals in Luwian, the instances involving the number 4 and 15 would morphologically smoothly fit the attested formation. Thus, the table above with all listed forms adheres to this analysis.

[E.R.]

Contextually, it is possible to assign a common meaning to both types of occurrences. This fact favors the conclusion that we are dealing with a single lexeme. If we accept the analysis as a neuter i(t)-stem, the use of the secondary ending =ša clearly indicates a Luwian origin. Besides its function to derive concrete nouns and nomina qualitatis, the suffix -it- /-id-/ n. is frequently deployed to integrate Hurrian loanwords ending with -i/-e into Luwian (Starke 1990a:152). Therefore, it is very attractive to identify Luw. kappiša, being attested in the ḫišuwa- festival from Kizzuwatna, as a Hurrian loan. Most recently, this origin was tentatively suggested by Melchert (DCL:107). In this light, the traditional view that Hitt. ^DUGkappi- c. ʻbowl (vel sim.)ʼ is a direct loan from Akkadian kappu ʻid.ʼ and is connected to kappiša (thus HEG A-K:491, HED K:63, Haas 1992a:107, HW 2nd ed. K:105, 108) should be revised, too. Based on the suggestive i-stem and its attestations from only  the MH period onwards, it seems likely that Akkadian kappu was first borrowed into Hurrian as ^DUGkappi- ʻbowl, (precious) containerʼ (for the Hurrian form see Richter 2012a:186). The Hurrian form was subsequently adapted as kappiša (stem kappi(t)-) in Luwian, whereas its Hittite equivalent kappi- c. ʻbowl (vel sim.)ʼ can be either a Luwian loanword or a direct borrowing from Hurrian.The integration of Hurrian loanwords into Luwian as neuter i(t)-stems, and further from Luwian into Hittite as commune i-stems, is a common pattern, cf. Hurr. kešḫi ʻseat, chair, throneʼ → CLuw. ^GIŠkišḫi(t)- n. ʻid.ʼ → Hitt. ^GIŠki/ešḫi- c. ʻid.ʼ (Starke 1990a:215, DCL:112). – Since there is no pattern apparent yet that reliably identifies a Hittite stem as an Akkadian borrowing, words ending in -i should probably be analyzed as mediated into Hittite via Hurrian first. Additionally, there seem to be no secure direct Akkadian loanwords into Hittite except for tuppi ʻclay tabletʼ (Schwemer 2005-2006b). The suggestion made in HEG A-K:491 that the i-stem in Hitt. kappi- is influenced by the adjective kappi- ʻsmallʼ must be discarded accordingly and finds a more systematic explanation instead.

[E.R.]

*This lemma was written together with Melvin Pötzsch.