LINGUIST List 24.3284|
Fri Aug 16 2013
Review: Language Documentation; Saramaccan: McWhorter & Good (2012)
Editor for this issue: Rajiv Rao
From: Anne-Sophie Bally <bally.anne-sophieuqam.ca>
Subject: A Grammar of Saramaccan Creole
E-mail this message to a friend
Discuss this message
Book announced at http://linguistlist.org/issues/23/23-4434.html
AUTHOR: John H McWhorter
AUTHOR: Jeff Good
TITLE: A Grammar of Saramaccan Creole
SERIES TITLE: Mouton Grammar Library [MGL] 56
PUBLISHER: De Gruyter Mouton
REVIEWER: Anne-Sophie Bally, Université du Québec à Montréal
This book belongs to the ‘Mouton Grammar Library’ series and presents a
description of Saramaccan Creole. This Creole language is mainly spoken in
Suriname, where it was created during the seventeenth-century. Its genesis is
the result of contact between West African language speakers and English and
Portuguese speakers. Dutch contributed to a lesser extent to the Saramaccan
lexicon. The present work is a synchronic grammar, thus, questions about
Saramaccan creation are left aside. It addresses a broad audience: all
linguists and not just creolists should find the data interesting, especially
those interested in prosodic phonology, verb serialization, tense, mood and
aspect (TMA) systems or adjectives. It assembles over a thousand examples,
which is a large corpus reflecting substantial fieldwork, and as such, is a
truly valuable linguistic tool. Among the examples, we find ungrammatical
sentences; this is noteworthy since ungrammatical sentences accompanying
grammatical sentences allow linguists to know, for instance, if an item has a
specific syntactic position or not, or if certain items cannot appear
The book contains 17 chapters which cover mainly topics in
phonetics/phonology, morphology and syntax. Half of these chapters contain
less than ten pages. The focus is on phonetics and phonology, verb
characteristics and sentence structures.
Chapters 1 and 2 deal with phonology: the first chapter is about segmental
phonology, whereas the second one is about prosodic phonology. Chapter 1
begins with an inventory of the consonants and vowels found in Saramaccan.
Among these consonants, it is worth noting that Saramaccan has a series of
prenasalized stops. As for the vowels, Saramaccan makes a distinction between
long and short forms and has a large set of vowel combinations (i.e. the
possibility of having two vowels in a row). Interestingly enough, the authors
do not label them as “diphthongs” since there is a lack of evidence for such a
statement. Following this description of segmental inventory, we learn that
Saramaccan only permits vowel (V) and consonant + vowel (CV) syllable
structure (and maybe an exceptional shape, CVm, with a coda m).
Chapter 2 (about prosodic phonology) is divided in two main sections:
word-level phonology and phrasal prosody. One of the main points the authors
talk about is a process they call “high-tone plateauing”, which consists of
high-tone spreading and occurs in specific environments like compounds or
subject-verb (when the verb is preceded by a verbal particle) syntactic
environments. They also describe how complex this process is when the sentence
structure involves a serial verb construction (SVC).
The third chapter presents the extremely simple morphology of Saramaccan, in
addition to all the morphophonemic changes that occur in this language (mostly
with pronouns, when they appear after particular prepositions and a small set
of verbs). The main derivational process in this language is reduplication,
which can be total or partial. This process is productive and has different
functions, depending on the grammatical category of the reduplicated word. For
example, a reduplicated dynamic transitive verb receives a resultative
interpretation and is used as an adjective. Reduplication also occurs to
convey an intensified meaning (mainly with adjectives) or to express a high
number of entities (for the most part, with nouns). A compounding mechanism
(which is most of the time right-headed) is described briefly at the end of
Chapter 4 is about the noun phrase (NP) and the items within it (i.e.
determiners and items expressing demonstration and possession, relative
clauses and quantifiers). Section 4.4 presents relative clauses according to
an accessibility hierarchy, underlining the fact that subjects, objects and
indirect objects are relativized easily, but that a resumptive pronoun is
required to form relative clauses from obliques. In addition, it is
particularly difficult and unnatural to formulate relative clauses from
possessors. Within this chapter, a brief section is dedicated to noun
coordination, which is indicated with markers that only appear inside the NP
Chapter 5 briefly discusses personal pronouns. A table compiles the tonic,
subject and oblique forms. The authors analyze two pronouns as having clitic
status because of the morphophonemic changes that happen to them in specific
environments. The ways of expressing reflexivity and reciprocity are described
at the end of the chapter.
Chapter 6 deals with adjectives (also called “property items”), which are used
either predicatively or attributively in this language. There are many
arguments in favor of these items acting as predicates. When used as
predicates, adjectives can be preceded by TMA markers. They also are subject
to predicate cleft and can occur in serial verb constructions. Adjectives
describing a property can undergo reduplication, which generally yields a
counterexpectational meaning. Then, a section of this chapter presents another
kind of reduplicational phenomenon already mentioned in Chapter 3: dynamic
verbs reduplicating to get resultative adjectives. Finally, Chapter 6 ends
with a section about comparative constructions and a short paragraph about
Chapter 7 groups together all the grammatical items appearing before the verb,
especially the negation marker and preverbal TMA particles (i.e. bi and ó,
which are tense markers indicating, respectively, past and future, and tá and
ló, which are imperfective and habitual aspect markers, respectively). Other
items or constructions used to express aspect, like durative, completive and
continuative aspects, but which are not preverbal TMA particles, are presented
in Sections 7.3.5 to 7.3.8. Modality is mostly rendered through the verbs músu
(‘must’) and sá (‘can’). Musú appears as a deontic marker, but also as an
epistemic marker expressing probability. Sá is only an epistemic marker
indicating ability and possibility. Chapter 7 concludes with a section
presenting the order of occurrence of these markers in sentences.
Chapter 8 describes verb serialization, a distinguishing feature of the
language. In the first section, syntactic tests confirming the existence of
verb serialization are presented. Then, a classification of serial semantics
is listed. This kind of constructions is frequently used to express
directional meanings, where the second verb in the series indicates the
direction (e.g. ‘back’, ‘around’, ‘out’, etc.) of one of the verb arguments.
The following section enumerates serial verb constructions which are already,
or on the edge of, being grammaticalized. Thus, the verb dá (‘give’) in a SVC
is nowadays more like a dative and benefactive marker, and the verb táa
(‘say’) is better analyzed as a complementizer following verbs of utterance or
verbs of cognition. In the next section, it is shown that some verbs are not
grammaticalized, but rather undergo a change of meaning when they appear in a
SVC. For example, the verb léi means ‘to show’, but when it occurs as the
second verb in a SVC, it means ‘to show how to’.
Coordination and subordination are described in Chapter 9. Saramaccan does not
have a coordinating conjunction equivalent to ‘and’ in English. Instead,
Saramaccan uses several strategies, like SVCs, to indicate a sequence of
events or sequential markers comparable to English’s ‘then’ or ‘and now’.
However, Saramaccan has two conjunctions indicating disjunction and exclusion
. The second section deals with subordination and is divided between finite
and nonfinite complementation. Different strategies to form complex sentences
in Saramaccan are presented. The formation of adverbial complement clauses
generally implies the use of a complementizer, but also the use of temporal
and locational markers. Note that this chapter gathers types of complex
clauses with different meanings; for example, we learn strategies to get
factive or conditional clauses, most of which occur within complements and are
further described in other chapters on grammar. Suitably, the authors provide
the reader with helpful references to other sections.
Passive and imperative sentences are introduced in Chapter 10.Strictly
speaking, there is no passive voice in Saramaccan. However, it is possible to
express passive meaning in restricted contexts with the help of the verbs dá
(‘give’) or kó (‘come’) occurring in SVCs. The second section presents a list
of four verbs that allow for double-object construction (i.e. that are
ditransitive), and a list of verbs expressing causativity. Chapter 10 closes
with imperative constructions, which simply require the use of a bare verb.
Chapter 11 explains question formation, i.e., open-ended and closed-ended
questions and indirect questions. Within the section about open-ended
questions, we find a number of examples illustrating the inventory of
Saramaccan wh-words. We also learn that preposition stranding does not exist
in Saramaccan, and thus, prepositions are fronted in question formation.
The properties of the two BE-verbs dε and da are presented in Chapter 12,
which shows how their distribution differs: dε is almost a true BE-verb,
expressing locative, class equative and existential meanings, while da is a
presentative and equative item that lacks genuine verbal behaviour. Finally,
the authors include a section that sums up the few contexts where copula can
In Chapter 13, we learn how time and space are expressed with the help of
nouns, adverbs and verbs. Since Saramaccan has a set limited to three
prepositions, it uses spatial indicators that occur after located nouns to
express spatial conceptualizations like ‘under’, ‘behind’ or ‘inside’. These
spatial indicators seem to be nouns. In addition to these items, Saramaccan
uses a three-level distinction to express proximity, with the help of deictic
adverbials meaning ‘here’, ‘there’ and ‘yonder’. A section then summarizes how
verbs like púu (‘to pull’), túwε (‘to throw’) or kó (‘to come’) play a role in
the expression of direction, essentially in SVCs.
This leads the authors to talk about adverbial modification in Chapter 14. It
is worth noting that there are ideophones in Saramaccan; some of them are
presented in this chapter. This chapter is more like a corpus that brings
together sentences containing intensifiers, time adverbials, adverbs of
quantity and adverbs of frequency. The authors note that the adverb position
is generally clause initial or clause final.
Chapter 15 deals with pragmatics, showing which markers or strategies are used
to highlight important or new information in discourse. The authors recall
that Saramaccan is primarily an oral language. Thus, topic-comment
constructions are more frequent than subject-predicate ones. The first section
explains how to put contrastive focus on a verb, on its arguments or on
adjuncts. It is worth noting that Saramaccan shares the contrastive focus
marker wε with Fongbe, which was one of the most influential substrate
languages at the time of Saramaccan genesis. As for presenting new
information, Saramaccan has two markers (nearly equivalent to the English
‘then’) that characterize a non-past and a past sequence of events. At the end
of the chapter, six pragmatic-marking adverbs are listed; they provide the
hearer with information about what the speaker thinks of a proposition.
Chapter 16 deals with the lexicon; it identifies a small set of cardinal
numbers and describes all the basic vocabulary expressing time on the clock
and the calendar (days and months).
Finally, Chapter 17 briefly explains that there is recognized dialectal
variation between two regions in Suriname. There is also a less documented
free variation phenomenon awaiting further research.
The book closes with a 100-word list, a folktale transcription and a
conversational passage, plus a three-and-a-half-page reference list (which I
will discuss in my evaluation).
This book is a great starting point for anybody who would like to be familiar
with Saramaccan Creole. The fact that the grammar is theoretically neutral and
synchronic is welcome; because of its particularly fast and isolated genesis,
scholars working on this language usually track the origin of functional
categories or specific syntactic constructions. Here, the authors present the
language as it is now spoken, without comparing it with its contributing
The first part of my evaluation will raise some general issues about editorial
choices made by the authors. More specific points I am familiar with will be
evaluated in the second part.
My first editorial criticism of this book is that the authors too often
describe Saramaccan in parallel with English, which is one of the superstrate
languages. For example, they observe that “[t]he Saramaccan consonant l has a
comparable realization to English l” (p. 12) or that “[t]he basic constituent
order of the Saramaccan noun phrase is that of English: determiner/
/quantifier – adjective – noun, with relative clauses occurring postnominally”
(p. 76). It is not clear to me why the authors compare Saramaccan with
English. Is it because of the contribution of English to the formation of
Saramaccan? Or is there a pedagogical aim (because of the international status
of English) to provide the reader with a better level of comprehension of the
linguistic data? This should have been made clear at the beginning of the
The second problem is related to the selected methodology for data collection.
In the introduction, the authors recognize that “this grammar is founded upon
a deficiency: almost all of the corpus was collected from emigrants in
Amsterdam, the Bay Area of California and New York City” (p. xvi). Thus, the
corpus constructed during the authors’ fieldwork does not reflect the grammar
of Saramaccan people, but rather the grammar of a handful of informants who
have not lived in Suriname for many years. Although I know that the authors
worked with trustworthy informants -- I have myself worked with some of them
-- a grammar of a given language should mainly contain data produced in the
place where it is spoken. This provides a reliable source not only for
synchronic, but also diachronic variation, especially for scholars who would
want to retrace the origin of specific items.
Moreover, a large part of the corpus presented in the book duplicates data
that had been previously elicited by scholars in Suriname with different
informants. This brings me to the bibliographical choices. A really
questionable aspect of this book is its bibliography and, more generally, the
way it deals with references. Saramaccan is one of the most documented Creole
languages, and there are far more references available than the ones presented
at the end of the book, in three and a half pages. But in his introduction,
McWhorter acknowledges that “a number of academic scholars have specialized in
Saramaccan to varying extent” (p. xv). He then cites 13 scholars who have
contributed to the study of this language. However, when we consult the
bibliography, we realize that five of the cited authors do not appear in the
bibliography, including McWhorter himself. Since the book is addressed to
linguists who want to know more about this language, it would have been good
to present an extensive bibliography, even if the contents of referenced
articles or books are not theoretically neutral. For example, we would have
expected to find references to Lefebvre and Loranger (2006, 2008) and Aboh
(2006) in the complementizer chapter, considering their extensive work on this
matter. As I said above, books of the Mouton Grammar Library collection are a
great starting point, but they should also let linguists extend their
knowledge about research that has been done on a specific matter. A section
containing an exhaustive bibliography at the end of each chapter would have
been really welcome.
My last comment about the book’s organization concerns the way chapters are
grouped together. A logical order would have been to introduce functional
items and the way they combine to form phrases, followed by information about
sentence structures. Instead, we find an eclectic organization of chapters.
Adverbs are presented at the end of the book, seven chapters after the one
about the verb, despite the fact that the verb chapter contains information
about negatives and tense particles, which could arguably be analysed as
I now turn to the evaluation of more specific points. First, I would like to
address the quality of Chapters 1 and 2, depicting Saramaccan phonology. This
awaited survey was a gap to fill in the Saramaccan literature. These chapters
contain not only a good description of phonological specificities, but also a
reliable analysis. Based on this analysis, morphosyntactic phenomena (Chapter
3) and SVCs (Chapter 8) receive a more powerful explanation when seen from the
standpoint of plateauing effects. It is regrettable, however, that the authors
chose to present prosodic phonology phenomena from the sole perspective of
tonal plateauing. This clear-cut theoretical choice hardly appears “theory
neutral” and seems inconsistent with the objectivity claimed by the book.
There is no mention of other explanations, such as the ones proposed in Kramer
(2001), which qualify these prosodic phenomena as “tone sandhi”, and later, in
Kramer (2004), as “high-tone spreading”.
My second comment bears on the way items are classified. The grouping of items
cannot be atheoretical; putting a word in one category rather than another
entails a theoretical choice. Sometimes the authors decide to put an item in a
category without justifying their choice, while in other cases they seem to
put an item in two different categories. For example, there is a debate among
creolists about the future marker ó. For some (e.g. Van de Vate, 2011), it is
a mood marker, whereas here, it is presented as a tense marker, without any
explanation or demonstration of why it clearly belongs to this category. A
hesitant portrait is drawn with the complementizer táa, which is presented as
an item occurring “in serial conjunction with verbs of utterance and cognition
to serve as a complementizer” (p. 143). It is unclear to me, as a naïve
linguist, whether this item is a verb in a series or a complementizer; the
authors choose to put it in the verb serialization chapter, but classify it as
a complementizer. Yet, the following chapter deals with complementation, and
again, they speak of “factive sentential complements formed with the serial
verb and complementizer táa” (p. 152). In the absence of evidence for the item
belonging to one part of speech rather than another, the authors should have
presented more clearly the existing debate about categorization in Saramaccan.
Third, I would like to talk about the content of Chapter 7. The description of
tense and aspect interaction with verbs, according to Aktionsart (or aspectual
classes of verbs), is incomplete; the authors consider only the distinction
between stative and dynamic verbs as having to be taken into account. However,
the division of dynamic verbs in three categories proposed by Vendler (1967)
is quite relevant, as has been shown in Bally (2004, 2011) and in Van de Vate
(2011). The interaction between TMA markers and aspectual classes of verbs is
more complex than what the authors describe.
Finally, I will underline the fact that in many chapters, the information is
valid and reliable, but incomplete. For instance, in Section 4.4, about
relative clauses, the authors say that dí is the relativizer used when the
antecedent is singular, and déé is the one used when the antecedent is plural.
This is true, but insufficient, since dí can also serve as a relativizer with
a plural antecedent (Bally, 2011). In Section 4.1, about determiners, the
number description is not precise enough. According to the authors, bare nouns
are possible in Saramaccan “when the noun is non-referential (refers to no
real-world entity)” (p. 76). This is incongruous with their example (6),
(6) I bi bisí hε̃́pi.
2S PAST wear shirt
‘You put on a shirt.’ (as advice or description of a generic event)
In this example, it is not descriptively adequate to say that there is no
real-world entity corresponding to the shirt. In fact, it is the identity of
the shirt that is unknown, but its existence is real. It has been shown in
Bally (2009, 2011) that bare nouns in Saramaccan are used in many contexts;
they appear with non-count and countable nouns. With countable nouns, they
express Generic, Kind or Existential readings. This is due to the fact that
Saramaccan has General Number (as presented in Corbetts, 2000, typology). It
allows NPs to not obligatorily contain a determiner. When a determiner occurs,
this is because the referent is accessible to all participants. This is
slightly different from the authors’ description of the NP structure.
Thus, my overall evaluation of this grammar is that it is a great starting
point for linguists who want to know more about Saramaka, especially those who
would like to have a global portrait of the complex tonal system and the way
high-tone plateauing/spreading occurs in definite environments. The book
contains a large and reliable corpus, however, one should be aware of the lack
of references; this grammar does not acknowledge the abundant existing
research on this language. The reader should not see this grammar as an
accurate reflection of the totality of existing research about Saramaccan, but
rather as a good, yet non-exhaustive portrait of this language.
Aboh, E. O. (2006). Complementation in Saramaccan and Gungbe: The Case of
C-type Particles. “Natural Language and Linguistic Theory”, 24, 1-55.
Bally, A.-S. (2004). “L'interprétation aspectuo-temporelle des énoncés en
saramaccan.” (Unpublished master's thesis). Université du Québec à Montréal,
Montréal, Québec, Canada.
Bally, A.-S. (2009, January). “Definiteness and Number in Saramaka.” Paper
presented at the scientific meeting of the 83rd Annual Meeting of the
Linguistic Society of America, San Francisco, CA.
Bally, A.-S. (2011). “Structure nominale et expression du temps, du mode et de
l'aspect en saramaka : analyse synchronique et diachronique.” (Doctoral
dissertation, Université du Québec à Montréal, 2011). Retrieved from
Corbett, G. (2000). “Number”. Cambridge: Cambridge University Press.
Kramer, M. G. (2002). Substrate transfer in Saramaccan Creole. (Doctoral
dissertation, University of California, Berkeley, 2002).
Kramer, M. G. (2004).High tone spread in Saramaccan serial verb constructions.
“Journal of Portuguese Linguistics.” 3(2), 31-53.
Lefebvre, C., & Loranger, V. (2006). On the properties of Saramaccan fu.
“Journal of Pidgin and Creole Languages,” 21(2), 275-335.
Lefebvre, C., & Loranger, V. (2008). A Diachronic and Synchronic Account of
the Multifunctionality of Saramaccan táa. “Linguistics”, 46(6), 1167–1228.
Van de Vate, M. (2011). “Tense, Aspect and Modality in a Radical Creole: The
Case of Saamaka.” (Doctoral dissertation, University of Tromso, 2011).
Retrieved from http://ling.auf.net/lingbuzz/001378
Vendler, Z. (1967). “Linguistics in Philosophy”. Ithaca: N.Y. Cornell
University Press .
ABOUT THE REVIEWER
Anne-Sophie Bally is a lecturer at Université du Québec à Montréal, Montréal,
and Université du Québec à Trois-Rivières, Trois-Rivières, Québec, Canada. She
wrote her thesis about the nominal and the verbal structures in Saramaccan and
their genesis, from a neo-saussurean perspective. Her interests lie in
semantics and syntax, but also in sociolinguistics and languages in contact
Read more issues|LINGUIST home page|Top of issue
Page Updated: 16-Aug-2013
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.