58 RICHARD J. FILE-MURIEL

(1990) and suggest that the assumption that the European language
was the target language in most plantation slave-societies may be
flawed. Furthermore, if the European language cannot be definitely
established as the target language, the term 'imperfect learning' does
not adequately describe the process of creole genesis. 3. I will look at
work done on Palenquero (Schwegler 1997, 1999, Patifio Roselli 1999,
Lipski and Schwegler 1993) based on the creativist model of
pidginization (Baker 1990) and attempt to answer the following
questions: What was the TL for the community of Palenque? Is
imperfect learning of some pre-existing language an appropriate
description for the formation of Palenquero?l


1. Literature review

 Ferguson & DeBose (1977), henceforth F&D, see pidginization as a
process that accepts normal language as input and produces a reduced,
hybridized, and unstable variety of language as output. This unstable
variety is identified as Broken Language (BL) when used by non-native
speakers and Simplified Register (SR) when used by native speakers.
SRs (e.g. baby talk, Foreigner Talk, etc.) are used by native speakers to
address those thought to have less-than-normal competence in the
language. These registers show simplification processes, which are
modifications intended to make utterances easier to perceive,
understand, or produce by omitting material, reducing irregularity, and
making sound-meaning correspondences more transparent. BL is used
by non-native speakers to address native speakers and/or non-native
speakers who do not share the same natural language. A pidgin, then,
is the output that results from the communicative interaction between
native and non-native speakers in a particular contact situation.
 F&D note that the modifications typical of SR and the imperfect
approximations that characterize BL are similar in many respects, and
it is often impossible to distinguish between texts of these two registers
without diachronic or variational information about the reference
languagess. While the characteristics of SR have been claimed to
represent universal tendencies (i.e. speakers of different languages


 I would like to thank J. Clancy Clements and Don Walicek for their comments
on previous versions of this paper. I claim all responsibility for any of its
shortcomings.