Accounting for Sonority Violations: the Case of Georgian Consonant Sequencing

Georgian is one of several languages which allows consonant sequences not predicted by the Sonority Sequencing Principle (SSP). The goal of the present study is to formulate a new generalization regarding consonant sequencing in Georgian, one not necessarily relying on the SSP. This generalization refers to observed preferences for combinations of certain feature specifications in the sequences. Support for these preferences is found in the acoustic properties of the sequences, which may indicate that they best satisfy requirements for cue preservation, while also being easier to articulate. Data from two native speakers, one male and one female, are analyzed. The results support the hypothesis that homogeneity of laryngeal features within a sequence, and the front to back order of the places of articulation present some articulatory advantages which may explain why they are preferred in the language.