edge (n.)

In some models of NON-LINEAR PHONOLOGY, the everyday sense of this term is used to identify phonological effects which apply at the margins of a linguistic unit, such as at the beginning or end of a WORD or SYLLABLE. For example, some models talk about edge effects, where a given SEGMENT behaves as though it bears the FEATURE [+F] with regard to segments on one side and [−F] with regard to those on the other side, as in the case of PRE-NASALIZED STOPS. Segments which do not display these effects are then said to show ‘anti-edge effects’, as in the analysis of AFFRICATES, which behave as stops with respect to following segments and/or as FRICATIVES with respect to preceding segments. The edge-marking parameter is cited in some approaches to METRICAL structure: this places a parenthesis at one edge of a sequence of marks (a left parenthesis to the left of the leftmost element in a STRING, or a right parenthesis to the right of the rightmost element). The phrase edge prominence constraint states that an edge CONSTITUENT will be more prominent than that of a constituent not located at an edge. In the analysis of REDUPLICATION in PROSODIC MORPHOLOGY, phonological constraints suggest that the two components (the base form and the reduplicant) must share an edge element – initial in PREFIXING reduplication, and final in SUFFIXING reduplication.

In later versions of the MINIMALIST PROGRAMME, those parts of a PHASE which allow SYNTACTIC operations to apply. A DERIVATION proceeds phase by phase, and once it has reached a higher phase, a lower phase becomes inaccessible. However, elements on the edge of the lower phase (such as SPECIFIERS) remain accessible. The phase edge has been described as an ‘escape hatch’ through which elements can be accessed.