Journal of Pidgin and Creole Languages - Volume 7, Issue 2, 1992

A unified theory is developed in this paper to account in a comprehensive way for variable tense marking patterns in the serial verb constructions found in Saramaccan, a creole language of the central interior of Suriname. The disparate patterns include verb initial marking, tense copy, and the problematic single marking on any noninitial verb in a serial string. It is argued that regardless of which tense pattern occurs in a given serial structure, there is an essential typological unity which is best explained through the interrelated processes of scope and spreading. Moreover, because Saramaccan exhausts the logical possibilities for such marking, the analysis presented here should be applicable to other serial tense marking instantiations, a claim which is discussed in relation to Ijo and Fante in latter sections of the paper.

Previous research (including the recent monographs of Friedemann & Patino Rosselli 1983 and Megenney 1986) states that Palenquero (henceforth PAL) has but a single future particle — tan — and lacks overt irrealis markers to express conditionals. This paper takes issue with earlier claims by showing that this creole language in fact has an additional irrealis marker — ake or its variants k(e) — whose principal function is to signal future and conditional.After the presentation of data, attention is focused on the syntax and origin of aké. In the course of the discussion it will become apparent that the PAL facts speak against Bickerton's well-known hypothesis about the prototypical creole tense-modality-aspect system, which is said to order preverbal morphemes as follows: tense (± anterior) + modality (±irrealis) + aspect (±durative). The final section offers a preliminary investigation into PAL modal distinctions (certain future versus probable future) which casts strong doubts on earlier assertions that PAL lacks contrastive mood differentiations. The article concludes with suggestions for further research.

Afro-Hispanic or bozal Spanish, from the sixteenth century to the early twentieth century, exhibited numerous cases of "epenthetic" nasal consonants, exemplified by Punto Rico < Puerto Rico; limbre < libre 'free'; pincueso < pescuezo 'neck'; and monosyllabic clitics such as lon < lo(s), lan < la(s), and so on. The present study, based on a comparison of Afro-Hispanic (AH) language data from a wide range of regions and time periods, provides alternative models for spontaneous nasalization. The first involves vowel nasalization, analyzed as the linking of a free (nasal) autosegment to the first available vowel of relevant words; Spanish speakers in turn reinterpreted the nasal vowels as a nasal consonant homorganic to the preceding consonant. Cases of apparent word-final nasal epenthesis, invariably involving phrase-internal clitics, resulted from prenasalization of following word-initial obstruents, a well-documented process in Afro-Iberian linguistic contacts. The preference for voiced obstruents to pre-nasalize is attributed to the lack of the normal fricative pronunciation of /b/, /d/, and /g/ in AH speech. In general, Spanish voiced obstruents are pronounced as stops only following nasals. The stop pronunciation of Pol, /d/, and /g/ by AH speakers was reinterpreted as an additional Root node, to which a floating (nasal) autosegment could be linked. AH nasalization generally seems to stem from Africans' underspecification of Spanish vowels and consonants, resulting from the precarious conditions under which Spanish was learned by speakers of various African languages.