PIE *h2ebVl-
Gloss: ‘apple’ (item 42 in Bjørn 2017)

Attestations: (?)Osc. Abella (toponym); OIr. ubull; ON epli; Lith. obuolỹs; OCS (j)ablŭko; (?)Thr. dinopula, synopula; (?)Pashaei wālī, (?)Kashmiri ambari-trel ‘small kind of a.’, (?)Pashto maṇá

Notes: The Thracian forms may well be connected as literal ‘dog-apple’, a perfect cognate of Lithuanian šun-obuolỹs (Markey 1988: 51), but there is no evidence for a PIE term (ibid. passim.). An Indo-Iranian strain may be included, but requires some irregular developments to fit with the European stock (Mallory & Adams 1997: 25), especially if the more erratic Iranian strain continuing *marnā– (Morgenstierne 1927: 45) is heeded. Here, like elsewhere, the reconstruction of *h2e– rather than *a– is based on assumptions of PIE that were evidently resolved by the ultimate dispersal of the dialect continuum. [Note: See also Kroonen 2016 and Fenwick 2016 for the root *smh2l-]

External comparanda:
Uralic: Fenno-Volgaic *omɜrɜ or*omena 

Discussion: Apples were only domesticated in the 1st millennium BC (Zohary 1990:39), and the item is a clear Wanderwort and traces its entrance to a handful of IE languages long after the dissolution of PIE (Campbell 1990: 163f., cf. also S. Starostin 2009: 93f.).
[Add. See Vennemann 2003: 466-468 for an Atlantic (~Semitic) origin ultimately related to PSem. *ˀabal- ‘genitals’]
[Add: Disregarding the Iranian forms, Wodtko et al. al but rejects the wanderwort hypothesis with reference to a widespread hysterokinetic paradigm unlikely to have been adopted independently in three our four IE branches (2008: 264 fn.1)]


PIE *w(ó)ih1-no
Gloss: ‘wine’ (item 133 in Bjørn 2017)

Attestations: Hit. wiyana-, HLuw. wi(y)ana; Lat. vīnum; Alb. vẽnë; Myc. wo-no, Gr. οἶνος; Arm. gini

Notes: An internal IE derivation has been proposed for this root, rather transparent even, from *ueih1– ‘to wind, twist’ with o-grade (de Vaan 2008: 680), and great antiquity is, indeed, conceivable with relative wide distribution, including Anatolian. Forms in Germanic, Baltic, Slavic, and Celtic forms are usually assumed to be through Latin, but the relative simplicity of the root renders any clear separation difficult, even of the cognates included above. The absence in the Central Asian branches of Indo-Iranian and Tocharian is noteworthy, see further under discussion. For a thorough treatment of the reconstruction and the overall likelihood of internal derivation, see Beekes (1987b).

External comparanda:
Semitic: *wajn
Kartvelian: *ɣwin- (Nichols 1997: 143 and Smitherman 2012: 505)
Kartvelian: Georgian-Zan *wenaq ‘vineyard’

The point of dispersal of the technical aspects of winemaking is of some pertinence to the current inquiry as viticulture may be traced to the Caucasus in the 6th millennium BCE (Fortson 2010: 42), but there need not be a contradiction when Zohary concludes that grapes, along with olives and dates, were only domesticated between 3700-3.500 BCE in modern day Israel and Jordan (1990: 39), since the fermentation of grapes, obviously, could have been done with wild species. The important lesson here, of course, is that the entire region, from the Caucasus mountains in the northeast, to the eastern Mediterranean in the west, were areas of possible early winemaking. Favoring a Semitic origin, Dolgopolsky notes that the root-final consonantal cluster is otherwise uncommon for PIE (1987: 16), but the donor language can hardly be Semitic since the internal cognates are only relevant in the western languages in immediate contact with the Mediterranean trade network, while the southern forms are considered secondarily borrowed (Lipiński 2001: 573). There is no trace of the root in any other Afro-Asiatic branch, which precludes a more ancient origin within that particular language family. Proof is an elusive phenomenon in historical comparative linguistics, so a strong conclusion like the one Dolgopolsky reaches on this term, i.e. that PIE and Semitic necessarily were in close proximity to one another (1993: 244), is bordering on tendentious. Diakonoff prefers a late borrowing from Mycenaean (1990: 59). The same term appears to be continued also in Hattic windu– (Chirikba 1996: 427), but due to the problematic nature of the language, the consequences of its presence here are difficult to evaluate. In Kartvelian Fähnrich (2007: 486) prefers an internal derivation, not much unlike the formation in PIE, to the verb *ɣun- ‘wind, bend’. The traditional explanation for these forms is a later loan from early Armenian that developed g- from PIE *w- via *gw- (Matasović 2012: 288 and Gippert 1995: 117ff.) and thus in formal accordance with the Kartvelian forms, but this still fails to account for the internal verbal connection. Consequently, both a loan from PIE → Semitic and Kartvelian, and, in the opposite direction, a loan from Kartvelian → PIE and Semitic, with subsequent folk etymological reinterpretation, are both viable trajectories. Otherwise a wholesale borrowing of the entire complex, i.e. with verbal root and nominal derivations, is required, but this scenario is hardly conceivable. A third option is a calque that coincidentally resembles the external proto-type. The fact that Fähnrich accepts the loan etymology for Georgian-Zan ‘vineyard’ (2007: 159) thus introduces another layer of confusion; Dolgopolsky similarly deduces that the Kartvelian forms were borrowed directly from PIE *-ah2 due to the *-q that he considers a reflex of the PIE laryngeal (1989: 12f.). A peculiarity of the IE distribution is its complete absence in the eastern and likely also northern branches, and thus, conversely, presence only in the languages historically attested in the Mediterranean region in contiguous geographic relation with Kartvelian, Hattic, and Semitic. All things considered, a Kartvelian provenance seems more likely, emanating from the Caucasus throughout the region. [Add. See now also Gorton (2017) for an adamant defense of Proto-Indo-European origins. I remain skeptical, but welcome the level of detail brought forth in the article. See also Pereltsvaig & Lewis (2015: 193-194) for a positive assessment of a Kartvelian provenance.]


PIE *ker̂ -(ker̂ -)
Gloss: ‘chickpea’ (item 67 in Bjørn 2017)

Attestations: Lat. cicer; Alb. thjer ‘lentil’; (?)Gr. κριός; Mac. κίκκεροι; Arm. sisəṙn

Notes: De Vaan rejects the Greek comparandum as a chance resemblance and tentatively posits the verbal root *kerh3– (2008: 113). The widespread reduplication points to an old phenomenon (Greppin 1981: 6f.), and the Greek form, if connected, may thus be a simplified variant. The distribution is very centralized in the circum-Pontic area, especially the Balkans, which may provide lexical evidence of an ancient Armenian presence in that particular region (cf. Solta 1960: 331f.).

External comparanda:
NE Caucasian: *qarhV

Discussion: Dolgopolsky bases his loan trajectory (PIE → NC Caucasian) on paleobotanical arguments (1989: 16), but the linguistic side certainly does not warrant a reconstruction for PIE. A similar root is visible in NW Caucasian with the meaning ‘pea’, and a pan-North Caucasian phenomenon is substantiated by similar items in all branches, of which Abkhaz and Kryts show reduplication (Mikić & Vishnyakova 2012). A further comparandum is suggested in Burushaski gərk ‘peas’ and thus shows hallmark signs of a wanderwort scenario. A reconstruction for PIE looks more like an example of biased science than a valid inference of the scattered and inconsequent IE data, unfortunately prompting an indefensible conclusion on directionality.


PIE *semh1r-
Gloss: ‘clover’ (not in Bjørn 2017)

?Gaul. ui-sumarus, OIr. seamar; Icel. smæra

Notes: The data is taken directly from Guus Kroonen’s presentation on the agricultural substrate in Europe, available here (PDF). Not reconstructable to PIE proper, and the forms are not completely compatible either, thus Proto-Celtic *semmar and Proto-Germanic *smēr-; with extensive contact between early speakers of Celtic and Germanic the item may have entered one and passed on to the other.

External comparanda:
Kartvelian: *s-m-q-r > Georg. samqura 

The argument is best explored through Kroonen’s slides (see link under Notes above). Most salient is the proposed Kartvelian etymology from sam ‘three’ and qur ‘ear’, and thus comparable to Latin trifolium. For geographical and distributional reasons, the transfer is highly unlikely to have been  directly from Kartvelian to Germanic/Celtic. Rather, a transfer may have occurred along with some of the goat words that also appear to have unique ties between European languages and the Caucasus. The connection seems plausible, probably in the semantic sphere of agriculture rather than a natural term.


PIE *gwher-nu-
Gloss: ‘millstone’ (item 28 in Bjørn 2017)

(?)Toch. B kärweñe ‘stone’; Welsh breuan; Goth. qaírnus ‘mill’; OPrus. girnoywis; OCS žrŭny; Arm. erkan; Ved. grávan- ‘stone for pressing soma’

The noun is traditionally derived from *gwher– ‘heavy’, and despite some controversy to the provenance of Tocharian it seems best reflected within this bulk (Adams 2013:176); this evidence is accompanied by the question whether ‘millstone’ is a narrowing of original ‘stone’ (Winter 1998: 351) or a broadening of the inherited ‘millstone’. If, indeed, Tocharian represents the second (known) branching of ancient PIE, the former hypothesis seems natural, and consequently preferable.

External comparanda:
Semitic: *gúrn-u ‘threshing floor’ > e.g. Ugaritic grn (examples in Tyloch 1975: 57)
North Caucasian: *χIwĕrV ‘mill, m.’ > e.g. Ingush ħajra ‘mill’, aha ‘to mill’

The Semitic loan etymology is defended by Dolgopolsky (1987: 16 and 1989: 6) and Takács (1997: 374) for their phonetic similarity, while Mallory & Adams (1997: 474) and Diakonoff (1985: 128f.) ascribe the similarity to sheer chance. The latter addresses the semantic gap between an IE ‘millstone’ and a purported Semitic ‘threshing floor’, which may only be bridged through the dialect semantics of Arabic ‘mortar’ (Gamkrelidze & Ivanov 1995: 770f.), but without similar variation in the ancient Semitic languages, this meaning is most probably innovated and without consequence for the present inquiry. Thematically the implement, or facility, belong to the agricultural package and should, if viable, be considered within that same context. All things considered, the Semitic comparandum requires a difficult semantic drift, especially for a purported technical loan, that renders chance resemblance the favorable option. Otherwise semantically attractive is the case for a North Caucasian connection, where a native verbal root seems to constitute the derivational basis for the implement (S. Starostin 2009: 96f.), which is a point of criticism with Matasović (2012: 290). Such a connection would, however, render the PIE system folk-etymologically associated with the root for ‘heavy’ and possibly separate Tocharian from the stock, rendering a transfer into late or late middle PIE the most likely scenario.


PIE *ĝherd
Gloss: ‘pear’ (item 25 in Bjørn 2017)

(?)Alb. dardhë; Gr. ἂχεδρος ‘wild p.’, ἀχράς ‘id.’; Mac. ἀγέρδα ‘id.’

Limited distribution to a very compact geographical region, the Balkan, has very weak bearings on PIE etymology, even more so if Orel’s reservations on Albanian are heeded (1998: 56). Furnée considers it a particular pre-Greek substrate word (1972 :127).

External comparanda:
NW Caucasian: *q(w)a
NE Caucasian: *qcör (Dolgopolsky 1989: 15)
Kartvelian: *msxal– (Fähnrich 2007: 296f.)

Nikolayev & Starostin, scrutinizing the North Caucasian connections, further compare the Kartvelian forms, although they concede the formal difficulties the inclusion entails (NCED s.v. ‘pear’). The linguistic travels of the fruit may be compared to that of the apple (item 42) that nonetheless appear to have wider distribution within IE. If connected, Dolgopolsky’s suggestion of a loan from North Caucasian *qcōrV (NCED *qHǖre, s.v.) or one of its descendants into a dialectically diversifying PIE (1989: 15) best fit the picture (cf. also S. Starostin 2009: 88f.), although the geographic discontinuity poses a serious obstacle (cf. Matasović 2012: 290). Caucasian connections for the Balkan languages are also pertinent in the discussion of *ĝhor-io- ‘pig’ (item 18).

Note: See now also Fenwick (2017) “An Indo-European origin of Kartvelian names for two maloid fruits”. Iran and the Caucasus, vol. 21: 310-323. []


PIE *dhoHn
Gloss: ‘grain’ (item 14 in Bjørn 2017)

(?)Hit. dannas ‘kind of food’, (?)Luw. tannas ‘id.’; Toch. B tāno; Lith.  dúona ‘bread’; Ved. dhāna-, Av. dāna.

Adams considers the earlier suggestion to derive it from the verbal root *dheh1– ‘put, place’, but concedes it as semantically strained (2013: 303). The identity of the Anatolian forms with the rest of the IE stock is tentatively suggested by Watkins (1965: 121), somewhat substantiated by the meaning in Lithuanian, but other, more immediate, alternatives likely reject the comparison (cf. Tischler 1991: 99).

External comparanda:
Semitic: *dúχn– ‘millet’ (Dolgopolsky 1987: 15, 1989: 5)

Orel & Stolbova connects the Semitic root to Central Chadic *dwan– ‘corn’ (1995: 166), adding credibility to an Afro-Asiatic origin. Diakonoff remains sceptical pending further knowledge of the importance of ‘millet’ in the diet (1985: 127), and the comparison may further be questioned by the semantic drift required to fit the IE attestations. The connection is attractive in light of the spread of agriculture, but remains inconclusive.


PIE *dheĝh-om
Gloss: ‘earth’ (item 10 in Bjørn 2017)

Hit. tēkan, HLuw. takam- (; Toch. A tkaṃ, Toch. B (t)keṃ; Lat. humus; OIr. ; Goth. guma ‘man’; Lith. žẽmė; OCS zemlja; Alb. dhe; Gr. χθών; Ved. kṣás, Av. zam-.

This item harks back to the oldest layers of PIE with representation in most branches, including the decisive ancient Anatolian and Tocharian that further help establish the original sequence of the obstruents (D-G) that underwent metathesis sometime in late middle PIE (to G-D). This item has been mentioned as an argument in favor of Tocharian as the second branch to leave the PIE dialect continuum (cf. Kretschmer 1931).

External comparanda:
Kartvelian: *diɣwam- ‘fertile soil’ > Georgian diɣvam– ‘black earth, sufficiency’ and Svan diɣwam ‘damp low place with f.’.
Kartvelian (alternative): *diqa- ‘clay’ > e.g. Georgian tixa, (?)Svan gim

Klimov proposes the connection with ‘fertile soil’ (1998: 41), which semantically seems a viable loan item due to its specialized meaning and formal triconsonantal quasiidentity (D-G-M), although the vocalism seems somewhat discontinuous; the labial
treatment of the IE velar similarly needs to be explained. These considerations would establish PIE as the provider of the term. Alternatively, Gamkrelidze & Ivanov suggest that the meaning ‘clay’ in Kartvelian may be related to PIE ‘earth’, and that the Svan form represent a simplified stem *ĝhem– < *dhĝhem– (1995: 774) to illuminate internal PIE development as well, but note that Klimov does not include Svan in this cognate set (1998: 72), reducing the reconstruction to Georgian-Zan, while Fähnrich treats it along with Mingrelian gim-e ‘below’ (2007: 107). This Svan form is, moreover, not an exact semantic parallelism, but the connotations in PIE with inherent association with ‘human’, e.g., at least some European languages, Goth. guma and Latin homō (cf. Mallory & Adams 1997: 366), could mend the comparison somewhat. The sequence of the obstruents in either Kartvelian comparandum would surely establish the loan to an early stratum of PIE, i.e. before the metathesis sometime after the presumed split of Tocharian. Kaiser & Shevoroshkin reject the hypothesis on formal grounds (1986: 368), but, all things considered, the connection with ‘black soil’ appears most attractive if the similarity is not merely due to chance resemblance.


PIE *bhar-(s-)
Gloss: ‘barley’ (item 2 in Bjørn 2017)

Lat. fār; (?)OIr. bairgen ‘bread, loaf’; Goth. bariz-eins ‘of b.’; OCS brašĭno ‘food’, Rus. bor ‘millet’; Alb. bar ‘grass’; (?)Gr. Περσεφόνη ‘Persephone (? = the grain-slayer)’ (PN)

It is noteworthy that a bare stem may also exist in Celtic and Slavic (Russian bor ‘millet’) next to the somewhat more prolific extensions. The inclusion of the Greek deity is highly dubious (cf. Chantraine 1968: 889), and, even if accepted, would not introduce significant new evidence to the picture already painted by the more secure attestations. Lehmann proposes that the lexeme be internally derived from a verbal root (1986: 62), but the a-vocalism and the external comparanda treated immediately below demand that the possibility of foreign influence, at least, be entertained (cf. de Vaan 2008: 201f.).

External comparanda:
NE Caucasian: *bVrcị̌nV
NW Caucasian: possibly Adyghe ‘grain’, Abkhaz ‘id.’
Semitic: *barr-/burr- ‘cereal, wheat’

Some objections have been raised to the Semitic loan hypothesis, e.g. by Mallory & Adams who consider a borrowing ‘unlikely’ based on the morphology (1997: 51), while Diakonoff’s rejection of the claim, based on its putative isolation in Semitic (1985: 126f.), has become mute in light of the fact that Orel & Stolbova (1995: 56) connect it to an almost ubiquitous Afro-Asiatic root, *bar-/bur– ‘grain, cereal’, which certainly advocates for Semitic (or related) origins. Gamkrelidze & Ivanov insist that the IE branches representing the item were in direct contact with speakers of a Semitic proto-language (1995: 770), but, as several other instances suggest (e.g. *ghaid– ‘goat’, item 21), the proposition of an agricultural
substrate, possibly even related to Semitic (§ 2.5), blunts the urgency of the claim. Note, too, that the semantic shift from ‘wheat’ or ‘cereal’ speaks against direct contacts. Dolgopolsky introduces the Caucasian comparanda, and questions a Semitic provenance on the basis of its simpler stem that lacks the *-s (1989: 15f.), but, as shown above, a similar IE variant could represent the original state only secondarily derived. The North Caucasian comparanda seems to reflect a higher complexity than either of PIE and Semitic, possibly hinting at greater antiquity, but, more likely, a window to PIE phonetics may be encountered here, seeing that there is a decent argument in the proposition that the desinence *-inV in North East Caucasian reflects the PIE derivational suffix *-in-o-, cf. Slavic *boršĭno– ‘flour’, Latin farīna ‘id.’, and probably also Goth. barizeins ‘of barley’ (ibid., cf. also S. Starostin 2009: 91 and further Matasović 2012: 291). This comparison is also favorable due to identical semantics. A tentative history of the term can thus be schematized as Afro-Asiatic > Semitic & Old European substrate → middle or late PIE → North (East) Caucasian.


PIE *bha-bh/k-̂
Gloss: ‘bean’ (item 1 in Bjørn 2017)

Lat. faba; OHG bōna; OPrus. babo; OCS bobŭ; Alb. bathë; Gr. φακός

Formal inconsistencies and limited distribution suggest a regionalism.

External comparanda:

Old-European: Unattested

This item shows all the hallmark traits of a regional borrowing, quite possibly a loan from an agricultural substrate, to which, e.g., *ghaid– ‘goat’ (item 21) may be compared, cf. Kroonen (2013: xviii,55).