There is considerable discussion in the literature on Bislama (and related languages) about what the best linguistic characterization of i (as in i luk) is. Meyerhoff (2000a) considers the various analyses that have been proposed and argues from the distribution of single pronoun, focused pronouns and null subjects that i (and oli) are not pronominal. Meyerhoff (2000a, b) calls them "agreement", another reasonable alternative is "predicate marker" (if it is understood that the form of the marker is related to the person and number of the subject).
Source: Meyerhoff nd