ܒܪܢܓ brng rice
This word has often been reborrowed: see the oldest loanword wrwnzʾ, and rwzʾ, ʾwrwzʾ, rzʾ, rʾzʾ
MP brinǰ [blnc] rice (CPD 19); ManMP brynz /brinz/ rice (Durkin-Meisterernst 2004, 111); Arm. LW brinj (Hübschmann AG 124, no. 121; Bolognesi 1960, 49; Henning 1963, 196 and n. 12, holds that the Arm. loanword presupposes a MIr. form with -nz-, probably attested as brynz and *gurinz; see also Henning 1940, 39, 45; Durkin-Meisterernst 2004, 167); NP birinǰ, gurinǰ id. (Steingass 179 and 1085; Horn 48). The rendering with final -g in Syr. is irregular: Ir. nǰ, Pers. nz should be rendered with Aram. nz, with or without the assimilation of the nasal (cf. Telegdi 203 f.): see the regular treatment in plzʾ plizzā (below, s.v.), from ManParth. plync /plinǰ/ bronze, MP brinǰ id. (see § 11.5). Probably, it is an attempt to represent approximatively the foreign phoneme /ǰ/ in a learned loanword; the same regularly happens with more recent loanwords from Arabic, or from Persian through Arabic, having final -ǰ, which is normally rendered with Syr. -g (see § 11.3.4). Even if I suppose that Syr. brng is a learned loanword, I do not think that it was borrowed from the MP written form, namely «blnc»: in this case, the form of the Syr. loanword would be completely unexplainable. On the contrary, Gignoux 1998–1999, 197 reconstructs an unattested MP *bring, and with regard to Syr. brng observes: “This form is interesting because of the final consonant -g, for the word in Pahlavi is written blnc (Great Bundahišn 117, 2), which is to read brinj [...]. The -c must represent -z however, if we compare with Greek βριζα-. As Old Ir. ganza- became ganj, one should have brinj coming from *brinz, very close to Greek. Anyway that does not explain the Syriac form, but since rice was unknown in Sasanian Iran, the word could have different spellings” ● Med 215, 5; 239, 12; BB 432, 14 ◆ LS 97b; Lagarde GA 24, 54; Id. GA p. 224: Gr. ὄρυζα ← Iran. non-SW ← Ind. vrīhi; PS Suppl. 64, where the Syr. word is vocalized as barang