Added data from Roberts' 2014 Comprehensive Etymological Dictionary of the Spanish Language to the Iberian Peninsula Languages page of the Substrate Dictionary. About twenty lexical items in all. Why he specifically fingers Iberian as the source language of many of these words is beyond me. In many of the cases, the simple, confounding answer is 'We don't know.' Some things to note about Roberts work:
- Often leaves source roots unreconstructed.
??? > argaña "wheat/grain beard" ~ árgoma "furze"
??? > balsa "raft" ~ Catalan balsa "place where brambles grow" ~ Spanish barzal "id."
Sometimes reconstructs certain items but doesn't reconstruct the final proto-form.
??? > *teutia ( > Mozarabic tauta "esparto") ~ ??? (> Spanish atacho "id." ~ retozo "frolic")
??? > *calappacu "tortoise" ( > galápago "tortoise") ~ carapacho "carapace", "crustacean shell"
Largely agrees with Dworkin (2012)...
...except focuses on the second element of substratal proto-form to arroyo "brook" ~ Latin carrugus "canal" where Dworkin focused on the first element. (I.e. Dworkin's focus is on reconstructing *(k)arr- while Roberts focuses on *-ruk-. The third explanation, which ties these words to hypothetical *(h)arus (Correo 2013), we can safely ignore.
I have no idea what Roberts argues under entry *ribiccu. Your guess is as good as mine. If you can figure it out, please send me your explanation in a message on the contact page.