My suspicion regarding the source of *r2 in Pre-Altaic seems to be confirmed

I just came across this parenthesized snippet from page 75 of Street/Miller, Altaic Elements in Old Japanese (1975) (see link) which discusses the evidence for causative markers in Proto-Altaic:

"(This looks like pA *-r₁-/*-r₂i-, conceivably from pre-Altaic *-r-/*-ri- [i.e. *-r(i)-] before the phonemic split of the pre-Altaic liquid.)"

I find this fascinating because it seems to confirm something that I had reasoned long ago concerning the source of the phoneme *r₂ in the Altaic plural marker. I figured that if the original plural common to Proto-Indo-European (PIE), Proto-Aegean, Proto-Uralic and Proto-Altaic (PA) was *-it, then presumably PIE and PA share the isogloss of Sibilantization of word-final *-t. After becoming *-is, the ending would have rhotacized in Pre-Altaic to *-ir before being palatalized to *-(i)r₂. So it appears very strongly that the source of all instances of Proto-Altaic *r₂ lies in the early palatalization originating from a neighbouring *i in a Pre-Altaic stage.