Pub. Date:
University of California Press
The Missing Spanish Creoles: Recovering the Birth of Plantation Contact Languages / Edition 1

The Missing Spanish Creoles: Recovering the Birth of Plantation Contact Languages / Edition 1

by John McWhorterJohn McWhorter


Current price is , Original price is $63.0. You

Temporarily Out of Stock Online

Please check back later for updated availability.


John McWhorter challenges an enduring paradigm among linguists in this provocative exploration of the origins of plantation creoles. Using a wealth of data—linguistic, sociolinguistic, historical—he proposes that the "limited access model" of creole genesis is seriously flawed. That model maintains that plantation creole languages emerged because African slaves greatly outnumbered whites on colonial plantations. Having little access to the slaveholders' European languages, the slaves were forced to build a new language from what fragments they did acquire. Not so, says McWhorter, who posits that plantation creole originated in West African trade settlements, in interactions between white traders and slaves, some of whom were eventually transported overseas.

The evidence that most New World creoles were imports traceable to West Africa strongly suggests that the well-established limited access model for plantation creole needs revision. In forcing a reexamination of this basic tenet, McWhorter's book will undoubtedly cause controversy. At the same time, it makes available a vast amount of data that will be a valuable resource for further explorations of genesis theory.

Product Details

ISBN-13: 9780520219991
Publisher: University of California Press
Publication date: 07/03/2000
Edition description: First Edition
Pages: 292
Product dimensions: 6.00(w) x 9.00(h) x 1.00(d)

About the Author

John H. McWhorter is Associate Professor of Linguistics at the University of California, Berkeley, and the author of Towards a New Model of Creole Genesis (1997), The Word on the Street: Fact and Fable about American English (1998), and Spreading the Word: Language and Dialect in America (1999).

Read an Excerpt



University of California

Copyright © 2000 Regents of the University of California
All right reserved.

ISBN: 0-520-21999-6

Chapter One

Where Are the Spanish Creoles?


Forty years of articles, books, dissertations, and presentations have enshrined limited access to a lexifier language as a driving force behind the emergence of plantation creoles. As noted in the previous chapter, while there is great variety among the genesis theories proposed in the field, all of them share the limited access conception as a pivotal component. It is important to recall, however, that the limited access mechanism has never been observed, and documentation sheds only the dimmest of light on the facts. Unlike, say, an interpretation of the causes of World War I, the limited access hypothesis springs not from an examination of empirical documentation, since this would be impossible, but simply from a natural interpretation of the fact that creoles are so often spoken in former plantation colonies, leading to the supposition that something about plantations created the creoles. It is a thoroughly plausible induction that demographic disproportion was the key.

However, what is plausible is not always true. Truth can be identified only via systematic testing, and the limited access conception has yet to be tested per se. To truly test it, creolists would have to search out as many plantation contexts as possible where demographic disproportion developed along the lines typical of European plantation colonies, and to ascertain that creoles have emerged in all or most such contexts.

When we actually test the limited access model in this fashion, we find that while superficially plausible when reconstructed for former English, French, Portuguese, or Dutch colonies, the model founders when applied to Spanish colonies. This is a first indication that a large-scale revision of creole genesis theory is necessary.



Creole genesis work on Spanish colonies has generally assumed that under the Spaniards, massive African labor crews were a phenomenon of nineteenth-century Cuba and Puerto Rico. The supposition appears to be that until then, while other powers had developed full-scale plantation economies by the late 1600s, the Spanish were using Africans mostly as domestic help and on small farms.

However, the restriction of this work to the Spanish Caribbean islands is odd, given that it is general knowledge that Spanish colonization of the New World extended much further than these islands. In fact, the Cuban and Puerto Rican explosions of the 1800s were merely the last act in a long tradition of large-scale exploitation of African slave labor by the Spanish. From the early seventeenth century on, the Spanish had gathered massive African plantation and mining crews in their mainland colonies, while their island colonies were still subsisting on small-scale farming. The mainland societies were very much of a kind with the plantation colonies which would emerge later in the century under England, France, and Holland.

Importantly, creoles simply are not spoken in these mainland Spanish settings, contrary to what all leading models of creole genesis would predict. For example, starting in the late seventeenth century, the Spanish began importing massive numbers of West Africans who spoke a wide variety of languages into the Pacific lowlands of northwestern Colombia to work their mines. This context shortly became one which, according to the limited access model, was a canonical breeding ground for a contact language of extreme structural reduction.

In the Chocó region, for example, there were no fewer than 5,828 black slaves by 1778, while there were only about 175 whites-a mere 3 percent of the total population (West 1957: 100, 108). Slaves had little sustained contact with whites. The slaves were organized into large teams, or cuadrillas, each formally supervised by a white overseer but actually directed by a black capitanejo (131-2). Cuadrillas typically consisted of two hundred blacks or more, with ones as large as 567 reported (115-6). One of the most numerically precise hypotheses regarding creole genesis (Bickerton 1981: 4) specifies one in five as the minimum ratio of speakers to learners necessary to produce a creole sharply divergent from its lexifier language. It is significant, then, that in the Chocó, even in the very smallest cuadrillas, the proportion of whites would have been, on a day-to-day basis, roughly 3.3 percent, and in most cuadrillas the proportion would have been virtually negligible. Furthermore, slaves were forbidden to communicate with what freed blacks there were (139-40), eliminating the latter as possible sources of Spanish input.

Some creolists might guess that the absence of a creole in the Chocó might be due to there having been a long initial period during which whites and blacks worked in equal numbers, the blacks being thereby able to acquire relatively full Spanish and then pass this on to the larger influxes of blacks later on. For example, Chaudenson (1979, 1992) and Baker and Corne (1982) have observed that a long period of this type prevented the emergence of a French creole on Réunion as opposed to Mauritius, where blacks came to outnumber whites quite quickly.

However, in the Chocó, there was no period of numerical parity between black and white. The nature of mining is such that relatively large numbers of slaves were needed from the outset, and they were immediately engaged in work arrangements ensuring little contact with whites. One of the earliest cuadrillas, for example, was established with forty slaves and was increased to sixty-five later that year (Restrepo 1886: 77-8). Sharp disproportion of black to white was not only established at the outset, but also increased by leaps and bounds throughout the 1700s: there were 600 slaves in the Chocó in 1704, 2000 in 1724, and 7088 by 1782 (Sharp 1976: 21-2).

More to the point, the slaves never worked alongside whites as they would have in the English or French Caribbean, but instead alongside Native Americans, who were second-language speakers themselves. Furthermore, the Native Americans had neither lived in intimate domestic conditions with their masters nor been a long-term, stable presence as had the early slaves in Réunion. For one, they were used only for the first fifteen years or so (Sharp 1976: 119-20). In addition, Africans were imported not to supplement and be trained by the Native Americans but to replace them, the Indians tending to die of European disease or escape. In sum, the Indians were second-language Spanish speakers, had had only distant and negative relations with the Spaniards, and had worked the mines for only a brief period during which they exhibited a high turnover rate. The Indians that Africans encountered were thus likely to be recent and miserable recruits unlikely to remain in service for long-they could have transmitted only fragments of Spanish to Africans at the very most.

Today, the descendants of the Chocó slaves live in the same lowlands where their ancestors toiled under the Spanish, subsisting via small-scale mining. Whites, having retreated to the urban centers after the slaves were emancipated, are a negligible presence in the lowlands (e.g., 8 percent by the 1950s; West 1957: 108-9). Relations between blacks and whites are, unsurprisingly, edgy and distant (Rout 1976: 243-9). In short, we could conceive of no situation more likely to yield a creole: vast numbers of Africans of groups ranging from Senegal down to Angola (Sharp 1976: 114-5), in massive disproportion to whites, with few blacks engaged outside of the trade at hand, all in a difficult-to-access region which the blacks still inhabit in virtual isolation.

Yet the Spanish of black Chocoanos is essentially a typical Latin American dialect of Spanish, easily comprehensible to speakers of standard Spanish varieties:

(1) Esa gente som muy amoroso. Dijen que ... dijeron que
that people COP very nice they-say that they-say-PAST that
volbían sí ... cuando le de su gana a ello
they-return-IMP yes when to-them give their desire to them
Those people are really nice. They say that ... they said that they
would come back ... when they felt like it.

(Schwegler 1991a: 99)

While displaying certain phonological and morphological reductions, as well as African lexical borrowings (not shown above), this dialect clearly lacks the radical grammatical restructuring in creoles such as Sranan Creole English, Haitian Creole French, and São Tomense Creole Portuguese.

To be sure, pidgins and creoles cannot be defined on the basis of specific constructions, since any commonly found in pidgins and creoles can also be found in regular languages, thus invalidating them as diagnostic of pidginization or creolization per se. However, it is uncontroversial that pidgins (and their creole descendants) can indeed be defined by two developmental processes in their past: marked structural reduction and heavy morphosyntactic interference from native languages (Hymes 1971: 70-1; see also Chapter 5, Section

Along these lines, then, what is remarkable about Chocó Spanish is that, in contrast to Sranan or Haitian, inflectional morphology is robust, and structural transfer from African languages is minimal (but see Schwegler 1991a and 1996a for certain parallels). Thus there is no need to deny the African heritage of Chocó Spanish, nor that it displays a certain degree of paradigmatic leveling. However, this variety classifies more as a Spanish dialect, retaining some traces of second-language acquisition, than as an example of the extreme reduction and transfer typical of Sranan, Haitian, and others.

Su ch an assessment cannot be based upon a formal line of demarcation between "dialect" and "pidgin/creole": to require this would be to set up a straw man. Studies such as that of Thomason and Kaufman (1988) have long demonstrated that contact-induced restructuring operates on a cline. However, such clines do not invalidate the usefulness of a perceptual distinction between "dialect" and "creole," anymore than we would balk at distinguishing a puppy from a dog. The cline acknowledged, few would disagree that Chocó Spanish falls on the "dialect" end.

What is important is that creole theory predicts that the Chocó context would have generated not a second-language dialect diverging only slightly from the local standard, but a more radically reduced, pidginized register, with much higher levels of structural interference from West African languages. In short, the modern situation in the Chocó is a striking counterexample to current creole genesis theory, all strains of which would predict a Spanish creole in this region.


Can we possibly ascribe the Chocó situation to a mere fluke, leaving the limited access hypothesis intact? If this were the only such situation, perhaps we could. In fact, however, the Chocó is nothing less than an unremarkable example of a regular pattern in Spanish America.

For example, when Jesuit missionaries settled in the Chota Valley of Ecuador in the seventeenth century, they established massive sugar plantations worked by Africans. Creolists have considered sugar plantations to be a prime context for the development of creole languages because of the vast manpower which sugar cultivation required. Significantly, then, La Concepción hacienda in the Chota Valley, for example, had no fewer than 380 slaves in 1776, the Cuajara 268, and so on (Coronel Feijóo 1991: 88). Slavery was not abolished until 1852. Today, descendants of these slaves live "a life apart" from the surrounding society, separated from the nearest city by a mountain, not marrying out, and considered an exotic local curiosity (Lipski 1986a: 156-9). Again, current genesis theory predicts a creole here.

Yet the black Choteños speak a dialect only marginally distinct from the local standard, typified by occasional, but by no means regular, lapses of gender and number concord (haciendas vecino "neighboring haciendas"), prepositional substitutions (cerca con la Concepción "near la Concepción" instead of cercade), article omissions (porque ø próximo pueblo puede ser Salina "because the next town may be Salina"; Lipski 1986a: 172). Such things leave the fundamental Spanish grammar intact, including, as in the Chocó, robust inflectional paradigms (see Schwegler 1996b, however, for some evidence of marginal West African structural transfer).

Once again, there was no initial period of parity between black and white which could explain the absence of a creole here. As in the Chocó, the original intention was to use Native Americans rather than Africans, but even they were brought to the plantations in large numbers at the outset. Unlike the English and French, who first devoted the small farms of their New World colonies exclusively to tobacco, coffee, or indigo, the Jesuits bought large swatches of land and devoted them to several products at a time: cotton, livestock, cacao, and plantains as well as sugar (Coronel Feijóo 1991: 63). Thus at one point two Jesuit haciendas were sharing some ninety Indian laborers (85)-clearly a different situation from the intimate interracial contact among a dozen or so whites and Malagasies on small farms in early Réunion. Even this phase, however, lasted a mere twenty years or so after the first haciendas were purchased in 1614. Consequently, when Africans were imported to gradually replace the Indians, it was immediately in the large numbers necessary to harvest and process sugar cane (e.g., a shipment of 114 in 1637 [86]), and by 1780, the eight Jesuit plantations were worked by no less than 2,615 slaves (88).

Two flukes? No-we find yet another example in vestigial, isolated Afro-Mexican communities in Veracruz, descended from African slaves who were imported at the transformation to sugar cultivation in the 1500s, Indian labor having proven unsuitable to the cultivation of other crops (Carroll 1991: 62-5). African labor forces were as enormous as elsewhere in the Caribbean, an example being the two hundred Africans working the Santísima Trinidad plantation in 1608 (65). Yet in the 1950s, the local speech in these Afro-Mexican enclaves was little different from vernacular dialects elsewhere in Mexico (Aguirre Beltran 1958: 201), as shown in the following sample (with departures from standard indicated in parentheses):

(2) Ese plan tubo (estuvo) bien hecho ... pero si el gobierno
that plan was well done but if the government

atiende (la) lej, ba a causa (causar) gran doló (dolor).


Excerpted from THE MISSING SPANISH CREOLES by JOHN H. McWHORTER Copyright © 2000 by Regents of the University of California . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Table of Contents

1. Introduction

2. Where Are the Spanish Creoles?
2.1. Introduction
2.2. Colombia, Ecuador, Peru, Venezuela, and Mexico
2.3. "There Are Spanish Creoles": Papiamentu and Palenquero
2.4. "There Were Spanish Creoles": Bozal Spanish and the "Extinct Pan-Hispanic Creole"
2.5. "There Will Turn Out to Be Spanish Creoles"
2.6. Accommodating the Theory to the Data: Sociétés d'Habitations versus Plantations
2.7. The Spanish as Kinder, Gentler Colonizers
2.8. "Nothing Is at Issue": The "Case-by-Case" Argument
2.9. Conclusion

3. The Atlantic English-Based Creoles: Sisters Under the Skin
3.1. Introduction
3.2. Methodology
3.3. The Features
3.4. A Closer Look
3.5. Implications
3.6. Sociohistorical Evidence
3.7. Summary

4. The Creationist at a Cocktail Party: Afrogenesis and the Atlantic English-Based Creoles
4.1. Introduction
4.2. Dating the Emergence of Sranan
4.3. A Theoretical Anomaly
4.4. Barbados?
4.5. West African Trade Settlements
4.6. The Cormantin Castle
4.7. Linguistic Evidence for the Cormantin Scenario
4.8. Preserving the Paradigm
4.9. Hancock's Domestic Hypothesis
4.10. Conclusion

5. Off the Plantation for Good: The French-Based Creoles
5.1. Introduction
5.2. Linguistic Data
5.3. Sociohistorical Evidence
5.5. Exploring Other Perspectives
5.6. The Portuguese Creoles
5.7. The Dutch Question
5.8. Conclusion

6. Synthesis
6.1. Geocentrism and Creole Studies
6.2. The Afrogenesis Hypothesis: Fundamental Outline
6.3. The Afrogenesis Hypothesis: Elaboration
6.4. The Afrogenesis Hypothesis: Problems Become Predictions
6.5. The Afrogenesis Hypothesis: Changing the Lens
6.6. The Case-by-Case Argument
6.7. The Reality of the Paradigm

7. Conclusion
7.1. The Middle Ground
7.2. The Domain of the Afrogenesis Hypothesis
7.3. Standards of Evaluation
7.4. Curtain



Customer Reviews