The case for collective suffix *-ar in Proto-Basque

Proto-Basque is famous for its tantalizing mysteries. One aspect of the Proto-Basque language was a reliquary of words that point to a collection or abstraction of items. The list of nouns with this putative *-ar suffix is small, small enough to leave most linguists with a sense of hesitation, but the examples are quite convincing:

*abar "branch," "branches," "kindling," "stripped branch"
*habe "tree"
*adar "horn" (< "horns?")
*ad- "horn"?
belar "grass"
*bel- "grass shoot" ?, "stem" ?
*bizar "beard" (< "whiskers")
*biz- "strand" ?, "whisker" ?
(h)ondar "residue," "sand"
ondo "bottom," "residue" (< Latin hondo "bottom")
*irar "legumous plant"
*ir- "legume"
*izar "star"
*iz- "star" (*iz- probably appears in izotz "frost" < *iz- ? + (h)otz "cold"; saying "star____" for frost and dew was an uncommon Basque compounding tradition - cf. izarritz "dew" < izar "star" + ihitz "frost")
negar "tears"
*neg "tear" (a relationship with Zuberoan regionalism ñíka is nearly impossible to exact)
*ohar "advice," "warning"
*o(h)a- ?
*sagar "apple"
*sag- "fruit" ? Basque lacks a native word for fruit
*zatar "rag"
*zat- ?

This list was first formed in any full sense of the word by Michelena (1961) and re-evaluated by Trask (2004). Michelena had a very favorable position on the possibility of *-ar while Trask is considerably more skeptical.

Linguists seem to have missed some of the likely relatives, as in the case of negar ~ ñíka and izar izotz, so the case for a *-ar suffix is actually much stronger than the academic literature currently suggests.

Michelena, L. "Hispánico antiguo y vasco". Archivum. Is. 8. 1958.

Trask, R. L. Etymological Dictionary of Basque. University of Sussex. 2008.