The issue of the phonological characteristics of the linguistic units represented by the letters վ (v), ւ (ṷ), and ու (oṷ) in Old Armenian has been a topic of scholarly discussion since the late 19th century, particularly in the works of H. Hübschmann. He argued that all three letters reflected the same phoneme (i.e., phone – V.P.), with a key difference. Hübschmann claimed that վ (v) was a simple fricative consonant, whereas ւ (ṷ) and ու (oṷ), had a dual function. Depending on their position they could function either as consonants [v] or as vowels [u]. From a phonological perspective, this means that the letters վ (v), ւ (ṷ), and ու (oṷ) were merely different positional variants – allophones – of the same phoneme. Most Armenian linguists who have explored this topic in one way or another have, conceptually, accepted Hübschmann’s paradigm (A. Meillet, N. Marr, H. Acharyan, S. Ghazaryan, E. Tumanyan, H. Muradyan) with slight variations. This is especially true in the case of ու (oṷ): all the scholars mentioned above considered Old Armenian ու (oṷ) to be a simple vowel [u] represented by two letters. Only H. Pedersen argued that Old Armenian ու (oṷ) was a diphthong. S. Ghazaryan, A. Abrahamyan, E. Tumanyan, and V. Hambardzumyan also supported this view. E. Aghayan was the first Armenian linguist who, in the 1960s, analyzed the phonemic system of Old Armenian from a phonological perspective drawing on N. Trubetzkoy’s principles of functional phonology, which is based on the concepts of distinctive features and binary oppositions. Nonetheless, even Aghayan supported Hübschmann’s concept on the phonological features of these sounds. Aghayan believed that the actual phoneme among the three was <Ու> (oṷ) – a sonorant, while ու (oṷ), վ (v), and ւ (ṷ) were its positional variants, i.e. sub-phonemes. Guided by the principles of functional phonology, we have demonstrated that: 1) վ (v) and ւ (ṷ) were integral members of the phonemic system of Old Armenian, despite some positional limitations. More specifically, a word could not begin with ւ (ṷ), and վ (v) could only appear at the end of a word if it followed the vowel ո (o) and the resulting sound was not the diphthongoid ու (oṷ) (compare: բով “a furnace for melting metal” vs. բու “owl”). Otherwise, in post-vocalic positions – both medial and final – ւ (ṷ) was the usual form, not վ (v): 2) ու (oṷ) represented not a simple vowel, but a diphthongic structure, more specifically a diphthongoid. This can be demonstrated by the fact that the semivowel ւ (ṷ), like the semivowel յ (i̭), only appeared as part of diphthongs and could alternate independently with both consonants and the semivowel յ (i̭), forming phonological oppositions (contrastive units), i.e., phonological oppositions. (Compare: բաւ “border, edge” ~ բան “speech, saying” ~ բառ “word”; գոյն” shade, color” ~ գուն-(ել) “to color” etc.). Moreover, phonological oppositions existed even between the vocalic components of the same diphthongoid (compare: նաւ “ship” ~ նու “bride”; չու “journey” ~ չեւ “not yet, still absent” etc.). This suggests that the connection between the vocalic and semivocalic elements of diphthongs was weak, and they could also function independently.
