Creole studies have advanced competing hypotheses about the structure and origins of creoles--languages based on two or more prior languages--most prominently contact with a colonial language, substrate influence, and universal linguistic processes. The Spanish-based creole Palenquero is spoken in the Afro-Hispanic community of San Basilio de Palenque, Colombia. This project examines variably occurring forms in Palenquero in order to test hypotheses about their functions and provenance. In the domain of the noun phrase, the focus of study is the demonstrative-article 'e' as the result of contact with Spanish, a reflex from the substrate Kikongo, or the product of language-internal demonstrative-to-article grammaticalization. In parallel, in the domain of verbal categories, this research investigates whether tense-aspect expressions function as grammatical markers and their possible sources.

Since creole forms may share surface similarity to those of presumed donor languages, adjudicating between hypotheses about their functions and origins requires comparison of their distribution patterns. The Variationist Comparative Method will be employed. The data will be naturalistic speech gathered in sociolinguistic interviews with 50 Palenquero speakers, which will be recorded and transcribed. 1,500 tokens for each functional domain will be extracted, coded for factors operationalizing hypotheses, and submitted to multivariate analysis. The constraints on variant forms will be compared with constraints on apparently similar forms in Spanish and Kikongo and in light of predictions from grammaticalization theory.

At the level of the community, this study will offer linguistic descriptions which can serve in the development of educational materials for the local schools. Moreover, innovative in this project is the training of community members to revise transcriptions, which provides an opportunity for involving community members in the scientific study of their language. More generally, this study contributes to the documentation of Afro-Hispanic language varieties, which have been socially stigmatized, and to demonstrating their systematicity in the structure of linguistic variation.

Project Report

Creole non-exceptionalism: Variable tense-aspect marking in Palenquero Are creole languages, whose grammars derive from two or more languages, fundamentally different from other languages? This project examines variably occurring verbal forms in Palenquero, a creole language spoken together with Spanish in the Afro-Hispanic community of San Basilio de Palenque, Colombia, the oldest surviving maroon settlement and often referred to as the ‘first free people in the Americas.’ A speech corpus for systematic quantitative analysis The corpus of Palenquero speech (Smith 2014) constituted by this project comprises audio-aligned transcripts for 10 hours of speech from sociolinguistic interviews and conversations (Labov 1984) with 30 community members. A transcription protocol was developed and implemented in collaboration with two community-member assistants. Such principled and exhaustive transcription affords an archive of this endangered language and a broadly usable corpus permitting accountable analysis. Are pre-verbal tense-aspect particles grammatical markers? Palenquero verbs sometimes appear with pre-verbal particles, such as asé in (1), which according to the literature expresses habitual aspect, signaling a regularly repeated or customary action. However, verbs also appear in habitual contexts with an alternative pre-verbal particle sabé or as zero (bare) forms. Is a preverbal particle such as asé, then, a true grammatical ‘marker’ of habitual aspect? (1) Majaná asé salí ahora sin pemiso ri ma tatá ku ma mamá ané. ‘Kids go out these days without the permission of their father and mother.’ Quantitative analysis of 2,544 tokens confirms form-function asymmetry, in other words, for any given form and the function expressed by it, that very same function need not be expressed by that form, but may also be expressed in other ways (Sankoff 1988). Thus, in past temporal reference, although the pre-verbal particle asé virtually always expresses habitual meaning, past habitual contexts do not always display asé: past habitual meaning itself is expressed by asé approximately two-thirds of the time. In present temporal reference, habitual contexts exhibit asé even less often, a little over one-third of the time; just as often, the verb is a zero form. Put another way, two-thirds of the time present habitual meaning is not expressed by asé at all. The distribution of the variant forms is not random, however. For example, in past temporal reference contexts asé is favored with non-stative verbs, such as kumé (‘eat’), but sabé with stative verbs, such as tené (‘have’). Pre-verbal particles, then, qualify as grammatical markers insofar as there is both an overwhelming association between the form and the function it expresses, and conversely, between a function and a certain form relative to other forms that may express it, as assessed by the relative frequencies of variants according to linguistic contexts (Poplack & Tagliamonte 2006). Thus, asé can be said to probabilistically ‘mark’ habitual in the past with non-stative verbs, but not in the present. Does the Palenquero tense-aspect system conform to cross-linguistic tendencies? Does Palenquero conform to patterns in the development of tense and aspect markers in the world’s languages? To ascertain this, the variable use of tense-aspect forms was compared in present and past temporal reference, in light of the predictions of grammaticalization theory, which hypothesizes cross-linguistic evolutionary paths for grammatical expressions. Cross-linguistically, habituals tend to have overt expression in past but not in present temporal reference contexts. The explanation is that habitual expressions grammaticalize, or develop, in the past before the present, because habitual is not part of the default meaning of past and so must be flagged with overt material. In contrast, habitual is part of the default meaning of present, and as such is usually zero coded in the world’s languages (Bybee 1994). Comparison of the distributions of Palenquero pre-verbal particles and suffixes revealed several patterns that are consistent with grammaticalization. The habitual pre-verbal particles tend to have more phonetic bulk in the past (aseba and sabeba) than in the present (asé and sabé), as predicted, and they are more frequent in the past than in the present, where more zero expression of habitual was also observed (Figure 1). It is also known that progressives—signaling an ongoing action in progress—may further grammaticalize to include habitual meaning, but that habituals do not evolve into progressives. In Palenquero, the progressive pre-verbal particle ta appeared in habitual contexts, but neither habitual pre-verbal particle (asé or sabé) was found in progressive contexts, again in consonance with cross-linguistic unidirectional grammaticalization paths. Impacts This project enabled systematic quantitative analysis of natural speech community data, which counters the stigmatization of Palenquero by showing the systematicity of this creole language’s pre-verbal particles, whose distributions, furthermore, adhere to cross-linguistic grammatical patterns (Smith 2014). Such linguistic analysis puts science to the service of language and community revitalization. Smith, Hiram L. 2014. Patterns of Variable Tense-Aspect Marking in Palenquero Creole. Pennsylvania State University dissertation.

National Science Foundation (NSF)
Division of Behavioral and Cognitive Sciences (BCS)
Standard Grant (Standard)
Application #
Program Officer
William Badecker
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Pennsylvania State University
University Park
United States
Zip Code