Abstract
Maltese verbal morphology has attracted the attention of many due to the co-existence of a Semitic system alongside a separate inflectional pattern reserved for its substantial stock of borrowed items. While this fact has led to split morphological analyses across the lexicon, this study provides a novel unified analysis of the system, using a surface-based approach that focuses on stem-alteration patterns internal to inflectional verbal paradigms, derived out of a variety of stem-indexing possibilities. It is shown how on this account all Maltese verbs can be categorised within eighteen stem pattern classes, with each class displaying additional paradigmatic stem-space organisations. These patterns also serve as a tool with which to measure the degree of morphological complexity, understood in this study as morphologically-triggered arbitrary stem-alterations that correlate with non-canonical paradigmatic behaviours. It is the interwoven complex of phonological, morphophonological and morphological factors (including additional lexical idiosyncrasies), that yields the attested array of output patterns. These factors interact with the base form, which provides three important variables: (i) the verb’s phonological structure, (ii) its stem-vowel(s) and (iii) information about the verb-form’s consonants. A number of paradigmatic dependencies are also shown to exist, projecting the paradigm as a structure of interdependent networks. In most cases the collective information that results is sufficient to deduce the stem-alternation patterns. Other paradigmatic behaviours that result out of morphological-conditioning require the listing of principal parts. By eschewing analyses in terms of consonantal-roots as opposed to stems, this study allows for a unified analysis of the entire Maltese verbal lexicon, including Semitic and non-Semitic verbs, and ‘regular’ and ‘irregular’ verbs alike. On this analysis verbs are classified on the basis of the patterns of stem-alterations displayed in their paradigms, resulting in non-affixal inflectional classes, in contrast to the proposed invariant system of affixation. Analysing such alternations furnishes additional evidence for the salience of the aspectual sub-paradigmatic split in the language, and shows the intricate relation between stems and affixes. The resulting account also reveals a number of parallels between the stem-alternation patterns displayed by argument-structure changing verbs and the pattern changes observed across derivationally-related verbs. It also highlights how the majority of non-Semitic verbs display the same pattern of alternations as the historically gh-final Semitic verbs; a fact that has remained unnoticed, given the traditional morphological split that has long characterised the study and analyses of Semitic vs. non-Semitic verbs in the language.