Somali wordhood and its relationship to prosodic structure

Two patterns of tone lowering in Somali

“…This debate opposes those researchers who claim that Somali is fundamentally accentual (cf. [13], [14], and [1]) and those who point out the essentially tonal nature of Somali word prominence ( [15]- [17]). Although the most part of [1]'s approach will serve as a working basis throughout the paper, the data analysis will lead us to reconsider this issue.…”

“…As for NOM, in addition to segmental suffixes used in some word classes, the prevailing mark that distinguishes it from all other cases is the loss of the TA (Maxamed, inan-i "girl.f"). Most authors ( [1], [7], [11], [12], [14]) classically account for this loss with an accent deletion rule (*°).…”

Somali as a tone language

“…After the application of accent shift and deleting rules, the surviving accents receive invariant high tones. Thereafter, the tonal (or pitch) accent analysis was adopted in its broad outlines by [5]- [13]. However, in recent papers, Hyman himself ( [14]) and [15] have called into question the tonal accent analysis of Somali, along with that of other languages, such as Japanese or Basque, for instance.…”

“…As seen, Hω exists independently from * and does not always require an * to surface. Likewise, in a paper addressing Somali wordwood and its relationship to prosodic structure, [13] crucially distinguish the tonal aspects from the accentual behaviors of words.…”

“…[1] was a first attempt to give a principled account of these alternations in a 'classical' generative framework using ordered rules. [5], [6], [8], [10]- [13] considered accents as fundamental units as well, even though tones could endorse a more important role. Now, how can these alternations be accounted for without any accent?…”