1

Arawak Linguistic and Cultural Identity through Time:

Contact, Colonialism, and Creolization.
In, Comparative Arawakan Histories, (ed). F. Santos-Granero and J. Hill.
University of Illinois Press
Neil L. Whitehead

ATo imagine a language is to imagine a form of life.@

Ludwig Wittgenstein, Philosophical Investigations

This chapter is concerned with the basis of linguistic classifications, the particular history
of how the linguistic classification >Arawakan= worked culturally in the region of northeastern
South America during the colonial period, and the pitfalls that process presents to the uncritical
identification of socio-cultural relatedness on the basis of such categories. The papers collected
here show convincingly that such pitfalls can be negotiated and that there are many reasons for
seeking to identify the long term historical trajectories among linguistically related groups. This
issue has been particularly sensitive within the study of indigenous South America because
models of historical evolution have tended to take a de-historicized view of linguistic
 2

relatedness, assuming that such relatedness was itself supra-historical and so a given rather than
a matter to be investigated (Greenberg 1987, Loukotka 1968, Rouse 1948a, 1948b, 1992). The
papers collected here depart from such models by demonstrating the meaning of linguistic
relatedness through attention to the archaeology, history and ethnography of Arawakan speakers.
In this way they have broken that mold of glottochronological approaches to historical linguistic
relatedness by emphasizing social and cultural historical trajectories over rates of linguistic
change. The two phenomena are of course closely related but the ground breaking aspect of these
studies lies in their attention to processes which produce glottochronological change, rather than
seeing that change as evidence of historical relatedness in itself.
The emphasis on linguistic over historical relatedness really begins with the classification
of languages by the colonial regimes throughout South America, the Caribbean and beyond. This
was a powerful political tool since to identify a language was to simultaneously >invent= a new
culture. Thus, it was thought that the intellectual capacities and cultural proclivities of a culture
stemmed from the workings and complexities of that language (Kroskrity 2000). As a matter of
intellectual history it needs to be noted that the concept of Alanguage@ precedes that of Aculture@
and that to a large degree the pre-nineteenth century notion of a Alanguage@ was equivalent to the
modern notion of Aculture@. Given this it should come as no surprise to find that the
Aidentification@ of indigenous languages in South America and the Caribbean was a highly
political process. Moreover, since communication with colonial subjects was key to the success
of the colonial project, gaining competency in native languages was a principal concern for
colonial regimes. In this context missionary evangelism, centered on verbal communication of
the gospel and textual ordering of indigenous speech, was pragmatically relevant to the colonial
project as a whole. Nowhere is this more evident than in the initial contacts with indigenous
American cultures in South America and the Caribbean (Whitehead 1999a, 1999b) and it is the
purpose of this paper to examine how that moment came to exercise an influence on the
subsequent linguistic ethnology of the whole region, and even beyond.
The Columbian Encounter and the Politics of Language.
It was Columbus himself who made the first and fundamental politico-linguistic
distinctions with regard to the native population of the Americas and our subsequent failure to
 3

understand our own cultural prejudices with regard to ideas of Aculture@ and Alanguage@ have
served to perpetuate those distinctions and allowed them to become encrusted with
glottochronological and historical linguistic theory (Whitehead 1995b). This has resulted in a
rather confusing picture as to the ethnic identities and cultural relations that once pertained
amongst the native peoples of Amazonia and the Caribbean.
Most obvious amongst these confusions is the question as to the ethnic and cultural
nature of so-called "Island Carib" society, since it would appear that these people were neither
Cariban (linguistically) -their natal language being Arawakan, nor islanders (exclusively) -as
there is evidence that they were also settled extensively on the mainland, in the coastal area
between the Orinoco and Amazon rivers (Whitehead 1995a). This paradoxical situation directly
results form the initial ethnographic judgement made by Columbus and confirmed by other
contemporaries, that there were two principal groupings of native peoples, one "tractable"
(guatiao, aruaca) and the other "savage" (caribe, caniba). Although not a linguistic
classification this ethnological scheme came to directly inform colonial policy, and so was also
self-fulfilling (see Sued-Badillo 1995). Consequently, subsequent ethno-linguistic studies, as
with the missionaries discussed below, reflected precisely these changes in native society
induced by the consequences of colonial policy, reconfirming the initial discriminations and
definitions of the colonizers. Also contributing to the perpetuation of this dualism was the
ethnological substitution of the mainland aruacas (Lokono), for the guatiao of the islands, as
the latter were destroyed or dispersed in the occupation of the islands in the sixteenth century1.
This dualism was not simply a colonial projection, nor was it a purely linguistic judgement, but
reflected real divisions in the native population. How such divisions functioned politically,
linguistically and culturally is still a matter of controversy as new historical and archaeological
evidence continues to emerge.
Modern anthropological approaches to the archaeology, history, linguistics and
ethnography of the northern region of the South American continent and the Caribbean islands
took these colonial schema as their starting point, especially as seventeenth century native

1
The guatiao came to be known as the Taíno in the nineteenth century, following the
terminology coined by the antiquarian C. F. Rafinesque (1836: I, 215-59). It is this term that
 4

testimony as to their own cultural origins was itself already partly expressive of these dualistic
cultural schema, as a direct result of Spanish colonial policies of ethnic discrimination and
slavery of those designated caribes (see discussion below).
The analysis resulting from this set of assumptions was given its classic statement by Irving
Rouse (1948a, 1948b) in his essays on "Arawak" and "Carib" for the Handbook of South
American Indians and even in more recent publications (Rouse 1986, 1992), it is still maintained
that "Island Carib" origins are linguistically and historically extraneous to the islands
themselves. Thus the character of their society, as well as its political and military conflicts with
other peoples in the Caribbean, is held to have resulted from a pre-Columbian military invasion
and occupation of the Lesser Antilles by the "mainland Carib" (i.e. Kariña), as a result of which
the Arawakan (i.e. Igñeri, guatiao) men of these islands were killed and cannibalized, while the
women of these vanquished men were taken as concubines by the Kariña war-parties.
The linguist Douglas Taylor (1977) also maintained that the explanation of the different
speech modes of the "Island Carib" (i.e. a natal language = Igñeri, and several jargons or pidgins
used exclusively by men) were the result of this pre-Columbian conquest by a group of Kariña
speakers of the Igñeri, using the example of Norman French supplanting Saxon English as his
model for linguistic replacement. Certainly autodenominations within these gendered speech
modes differed, Karipuna being used within the natal language, and Kalinago in the male
jargons. Taylor further argued that the natal language of the Kariña fell into disuse as the
offspring of the Kariña conquerors and their captive Karipuna wives evolved a new society,
although the "fact" of this past conquest continued to be expressed in the gender polarity of the
"female" Karipuna and "male" Kaliñago speech modes.
Luridly attractive though this tale may, other explanations of these speech patterns are
equally possible and actually more plausible. For example, given both the frequent
communication between the islands and mainland, which presumably facilitated this Aconquest@

today is still used to suggest a profound cultural cleavage in the aboriginal population.
 5

in the first place, as well as the fact that Kariña lived alongside Karipuna on the islands as well
as the mainland, the pidgin-Kariña used by the Karipuna men could have easily had other origins
(Whitehead 1988), not least since that pidgin was used with a Arawakan syntax (Hoff 1995: 49-
50). Most probably, as the historian Sued-Badillo (1978) has also suggested, a political and
economic adaptation and alliance to the emergent Kariña polity of the sixteenth century
(Whitehead 1990a) resulted in the name 'Carib` often being applied, by indigene and colonial
alike, without regard to strictly linguistic or cultural considerations; just as the Spanish used the
term caribe to designate any and all wild or fierce Amerindians (see Whitehead 1988). French
usage of the terms Galibi and Caraïbe to designate the difference between island and mainland
ethnic groups was therefore more precise than the English Carib or Spanish caribe and it is
significant to note that the Jesuit linguist Raymond Breton (1665:105) also refers to Caraïbes
insulaires, implying that they were present on the continent as well, since he does not confuse
them with the Galibi.
Further evidence of these close social and political relationships was the use of a Kariña
pidgin, or even Kariña itself, by other Amerindian groups as a lingua geral2 in the Antillean
-Amazonian corridor (Barrère 1743, Biet 1664, Boyer 1654, Pelleprat 1655). Moreover, gender
polarity in speech, as well as the use of special male jargons, is noted both from Kariña itself
(Chrétien 1725) and from Arawakan languages, like Palikur (Grenand 1987) and Lokono
(Stæhelin 1913 II-2:170), as well as from the Tupian (Maghæles 1527:33), whose speakers had
further notable cultural homologies with the native peoples of the islands. Given this complexity
and variety in indigenous linguistic practice the burden of explanation seems rather to fall on
those who insist that there was a 'conquest` by Kariña-speakers, since, if this was indeed the
case, why didn=t the natal Karipuna (or Igñeri) language die out, given the facility with which
contacts with Kariña-speakers could be maintained? In any case the first modern efforts to give
the conquest theory a scientistic footing -by attempting to correlate the data of archaeology with
that of linguistics (Rouse & Taylor 1956) -produced contradictory results as to the time-depth of
a Karipuna (or Igñeri) presence in the Lesser Antilles, which remain unresolved. Accordingly it

2. Comparison with the formation and usages of neêhengatú, a Tupi based pidgin would seem to
be particularly appropriate.
 6

is necessary to examine the theoretical origins of this situation through an appraisal of the
ethnological and anthropological judgements of Columbus and his contemporaries3, discussion
of the linguistic theories that informed later missionary accounts of Arawakan and Cariban
languages, and an assessment of how that has affected current anthropological thinking.
Most recent work on Columbus's interpretations and inferences about the native
Caribbean, are keen to emphasize the extent to which the ethnological categories he uses derive
from his own cultural expectations (see Greenblatt 1991). Thus the expectation of encountering
Asia leads Columbus to construct the caniba as soldiers of the "Great Khan", the expectation of
encountering human monstrosity leads him to note the existence of people with tails or without
hair, and, most notoriously, by the second voyage, the expectation of anthropophagism, deriving
from Columbus, leads Chanca into interpreting funerary customs on Guadeloupe as evidence of
anthropophagism (i.e. cannibal-ism).
Nonetheless, whatever the intellectual origins of these categorical anticipations it is legitimate
to ask what elements in the resulting interpretations derive from the unique experience of the
Caribbean encounter. In particular the contradictory and confusing way in which the term Carib
and similar terms, such as caniba, canima, canibales are used in the texts is generally held to be
expressive of Columbus's own confusion and inability to understand what was being told to him
-which of course it is. However, this does not mean that this uncertainty may not also reflect the
complex and contradictory nature of native socio-political reality, although the manner of its
refraction through the Columbian lens is certainly difficult to reconstruct.
Equally, the Columbian presentation of the caribe as fierce and warlike, wild and man-eating,

3. In a relatively brief presentation such as this it will be necessary to concentrate on a few key
texts; the Journal and Letter of Columbus, the Letter of Chanca, and the Life of the Admiral by
Hernando Colon - for a more extensive discussion see Whitehead 1995a.
 7

although most often thought to derive from the need to justify the colonial ambitions of the
Spanish -which it certainly later came to do -in the first instance may be seen as actually
reflecting the opinions of the ruling elite of Aitij / Bohío (Hispaniola). Columbus's adoption of
their viewpoint manifestly leads him and others into a number of contradictory propositions
within their texts, especially as regards the timidity, civility and lack of anthropophagy of those
who are not caribe.
For Irving Rouse (1948a/b, 1986) these confusions are due to the unreliability of the historical
data in general and the scheme of "fierce Carib" and "timid Arawak" is chosen from a number of
possibilities that the ethnographic observations of Columbus actually permit. The reasons for this
choice are many and are not properly part of this paper, but the fact that the idea of a group of
men advancing through the islands eating enemy men and copulating with their women is so
powerfully resonant for our own culture may be the most relevant consideration here, rather than
native Caribbean behavior in 1492. In any case, as was indicated above, both native testimony as
to conflict between the AIsland Carib@ and the AArawak@ (Lokono) in the seventeenth century, as
well as the work of seventeenth century missionaries in the field of linguistics, have been
misunderstood as directly verifying the "conquest theory".
However, the extent to which the Aconquest theory@ also relies on a misreading of Columbian
texts is nicely illustrated from the well known Journal entry for November 23rd, 1492. At this
point Columbus is sailing off Colba (Cuba) towards Bohío in the company of some Amerindian
captives. We read;
" ... those Indians he was carrying with him... said... that on it [Bohío] there were
people... called canibales, of whom they showed great fear. And when they saw that he
was taking this course, he says they could not speak, because these people would eat
them, and are well armed. The admiral says that he well believes there was something in
this, but that since they were well armed they must be people with reason; and he
believed that they must have captured some of them and because they did not return to
their lands they would say that they ate them. They believed the same thing about the
Christians and about the admiral the first time some of them saw them.
(Hulme & Whitehead 1992:18)
 8

There are a number of features in this passage that could well stand as an example of how the
Columbian texts have been poorly analyzed in anthropological readings. Firstly, the
identification of the Spanish, as rapacious conquerors, with the canibales, is most striking, and
often commented upon, as is the empathetic treatment of the political consumption of those
captured (see also Whitehead 1990b). Secondly, the link between military capability and being
gente de razon is an explicit anthropological principle to be found throughout the Columbian
texts. Its significance is illumined by this identity of Carib and Spaniard; the Spanish of course
having just completed their own Reconquista. However, since these observations and
interpretations relate to the heartland of "non-Carib" settlement -Bohío -they have been ignored
or suppressed in the analyses of subsequent commentators, as in the later Columbian texts, rather
than being treated as evidence of the inadequacy of the resulting dualistic ethnographic schema.
Similar contrasts in the ethnographic observations of the Letter and Journal emerge concerning
the diversity of language and custom present in the islands, material culture and the
identification of cannibalism with the caribes (Hulme & Whitehead 1992:12,13,15,21,26) .
Indeed Columbus is quite explicit in his Letter that;
In all the islands I saw no great diversity in the appearance of the people or in
their manners or language; on the contrary they all understand one another, which
is a very curious thing..
(Hulme & Whitehead 1992:13)
Nevertheless, by the second voyage we find that Columbus is making greater discriminations
and notices some lexical differences between those he suspects of being caribes and others in the
islands; although this is a long way from being the profound cultural difference that is implied by
the conquest theory since we are told that his native interpreters A... understood more, although
they found differences between the languages4 because of the great distances between the lands.@

4. It should also be emphasized that the use of the term lenguaje did not necessarily carry the
sole meaning of Alanguage@ in its modern linguistic sense but would have meant a manner of
speech, or dialect.
 9

(Hulme & Whitehead 1992:25). Las Casas (Historia Apologetica, chap.197) also tells us that
there were three languages spoken on Bohío which were not mutually intelligible, thus further
emphasising how deceptive an appearance of linguistic homogeneity may have been.
However, such ambiguities were not an idle question of scholarly dispute but intimately
connected to the pragmatics of conquest. Consequently, subsequent accounts attempt to resolve
issues of variation in dialect as well as appearance, for the caribes are described by Columbus in
the Journal as wearing black body-dye and long hair tied with parrot feathers (Hulme &
Whitehead 1992:25). Chanca's "official" anthropology, incorporating Columbus's first
ethnography, achieves this by the consistent application of a political decision to use the
caciques of Bohío, not the soldiers of the el Gran Can, as a bridgehead into the regional native
polity. Accordingly the ambiguities and uncertainties surrounding the identity of caribes within
the ethnoscape of the sixteenth century Caribbean are resolved by casting caribes in the role of
ferocious man-eaters and guatiao or aruacas as tractable and pliant. Thus, for Chanca, the
recovery of human long-bones on Turuqueira (Guadeloupe) is linked to cannibalism (Hulme &
Whitehead 1992:32), but on Bohío the recovery of human heads is linked to funerary rites (Gil &
Varela 1984:168-9). More generally the caribe's cannibalism of the natives of Burequen (Puerto
Rico) and the other islands is given continual emphasis, although it is also briefly noted that;
"... if by chance they [of Burequen] are able to capture those who come to raid
them they also eat them, just as those of Caribe do to them." (Hulme & Whitehead
1992:36)
This residual ambivalence as to the nature of the caribes, as well as its manner of resolution
within Chanca's text, is then fully revealed in his closing remarks on Turuqueira. Chanca writes
first that;
"These people seemed to us more polished than those who live in the other islands. [...]
They had much cotton [...] and many cotton cloths, so well made that they lose nothing
by comparison with those of our own country... [but later adds that]... The way of life of
these caribe people is bestial." (Hulme & Whitehead 1992:33)
Such an analytical distinction, if not an actual contradiction, must clearly derive from the
political purposes of the text.
 10

The political factors that had informed Chanca's anthropology changed over the next 20
years or so, not least due to the extinction of the native elites of Bohío and Burequen. As a result,
and since Chanca's anthropology had been given legal force through Queen Isabella's
proclamation of 1503 which rendered all cannibals who resisted the Spanish liable to
enslavement, it was necessary to conduct a second ethnographic exercise -in one sense, precisely
because of the ambiguity between the status of cannibal (i.e. eater of human flesh), and that of
caribe (i.e. native resistant to the Spanish) that the proclamation itself implied.
To this end the licenciado, Rodrigo Figueroa, was dispatched by Charles V in 1518, to
determine the exact locations where caribes were to be found. However, the ethnographic
criteria for their identification had simplified under the political necessities of colonial
establishment, as foreshadowed in the proclamation of Isabella, and mere opposition or
intractability towards the Spanish, rather than anthropophagic customs, was deemed sufficient to
consider a given population as caribe. At no time, however, was any kind of dialect or other
linguistic feature suggested as a way of achieving this discrimination. It should thus be very
evident that it was the politics of colonialism that determined the ethnological agenda, and so, in
turn, the creation of the ethnographic observations and linguistic descriptions that were thought
to verify it.
However, these colonial linguistic and ethnological texts were not composed of seamless
arguments and perfect data sets but were often mere accumulations of unsorted observation and
secondary testimony. As a result such texts also contain many indications for other kinds of
interpretations of the native Caribbean and, when combined with later sources and the data from
archaeology, may be used to provide a more adequate and more complete interpretation of the
situation encountered by Columbus, and in particular the significance of the terms carib / caniba,
and aruaca / guatiao.
In short, the social interdependency and cultural similarity of caribe and aruaca is a
possibility that was still ignored within earlier anthropological schema which all relied on the
assumption that the caribe were invasive or external to a primordial "Arawakan" or "Taíno"
cultural context. Yet evidence of social continuity underlying an ethnic and cultural interchange
between caribe and aruaca is present, as we have seen, in the early Columbian documentation;
 11

particularly in regard of that behavior considered definitional of the caribe -anthropophagy.

Thus, aside from the ambivalence of Columbus and Chanca we learn that the natives of Bohío;
"... les pagan [los caribes] en la misma moneda, pues descuartizan a un canibal ante los
ojos de la demas, lo asan, lo desgarran a rabiosas dentelladas y lo devoran."
(Anghiera 1530: II, 9-12)
While Hernando Colon (1947) stated that Caonabo, one of the principal chiefs of Bohío,
was himself a caribe and a stranger. Traces of such cultural homology also seem to be reflected
in the way in which the much abused term taino has registered in the speech of the Karipuna.
Thus, Taylor (1946) gives the orthographic form ni'tinao -formal friend (ws)5 or progenitor
(ms/ws), Raymond Breton (1665:454, 1666:19,315) giving the form ne'tegnon -and nitino /
neteno -husband=s father, husband=s mother or daughter=s husband (ws).
Mutuality in the ethnic definition of caribe and aruaca is also clearly implied by
evidence from the myth cycles of native Bohío, as recorded by Pané (1496) and Oviedo (1535).
Thus, during the journey of Guayahona, their mythical progenitor, in search of the mystic alloy
guanin, he traveled to the lands south and east of Bohío -that is the Lesser Antilles and the
mainland -taking with him their women and children. At the isla de guanin golden objects were
collected but the women and children lost, providing a symbolic alternative to the gastronomic
context in which most commentators, from Chanca onwards, have evaluated the claims by the
ruling caciques as to their "consumption" by the caribes. So too, by initiating the exchange cycle
of women for guanin, Guayahona may also be said to represent the first caribe cacique of Bohío,
becoming an ideological model for the authority of Caonabo (see above) and thus providing a
myth-charter for the chieftains of Bohío and legitimizing their marriage exchanges, or
marriages-by-capture, with the caribes who controlled access to guanin (Whitehead 1996a,
1998a: 70-90).

5
That is; ws - woman speaking, ms - man speaking
 12

It is thus evident that European fascination with the consumption of human flesh, as in
the case of Columbus, led to a total identification between "caribism" and "cannibalism"; but, as
has been argued above, it is evident from the Columbian texts that there were a variety of
orthographically related terms (i.e. caniba, caribe, canima, caribal) in usage in the Antilles,
which it can be argued had two referents, not just one. One pole of reference amongst these
terms, deriving from the politics of the ruling elite of Bohío, was the meaning of "mainlanders /
enemy people from the south", as de Goeje (1939) suggests, and as is indeed the contextual
sense, since the form caniba occurs alongside, not just as an alternative to, the term caribe in the
Columbian sources. Taylor (1946) in particular gave much attention to the derivation of such
terms but only as ethnic designations and did not consider the second, supra-ethnic, pole of
reference, orthographically represented by caribe (or caraïbe in the later French sources), and
for which there is a wealth of evidence from the mainland through the widespread use of the
terms caraybe / caraïbe / karai as Tupian spiritual honorifics or Cariban designations of a
martial prowess, associated with the possession of related anthropophagic rituals.
Missionary Linguists and the Cultural Inscription of Language.
If then the earliest reports belie later interpretations it remains to examine how explicit
consideration of native language by the missionaries of the seventeenth and eighteenth centuries
consolidated an erroneous ethnological dualism in the Caribbean and northern South America, in
which the Arawakan AIsland Carib@ came to stand as an icon of Acaribness@ (Whitehead 1995b,
Trouillot 1991). The missionaries brought a variety of different ideas to the task of conversion
and the evaluation and recording of the speech practices of linguistic communities effectively set
the agenda for evangelism. Thus, those with the capacity for rational understanding and spiritual
enlightenment were separated from those whose primitive and undeveloped speech required
military chastisement rather than spiritual suasion. In the word of one Jesuit missionary Athey do
not hear the Voice of the Gospel where they have not first heard the echo of gunfire@.
For instance, Raymond Breton, a Jesuit missionary to Dominica states that the Caraïbe:
"have no words to express the power of the soul, such as the will, the understanding, nor
that which concerns religion, [or] civility. They have no honorific terms like Our Lord.
They express however some acts of the understanding and of the will, such as to
 13

remember, to wish." (Hulme and Whitehead 1992: 110)

However, a later account, written by a lay Protestant traveller, Charles Cesar de Rochefort, notes
that the Caraïbe word for rainbow is "God's plume of feathers" (Hulme and Whitehead 1992:
122) and emphasizes the complexity and creativity of the Karipuna language. In short, the
cultural positioning of the reporter had a fundamental influence on the nature of linguistic
representation. Accordingly, I will briefly discuss Cesar de Rochefort, Raymond Breton, and the
accounts of two other Catholics, the missionary Jean Baptiste du Tertre, and a layman Sieur de la
Borde, who wrote from a Jesuit mission, in terms both of the influence that the French
Enlightenment had on their analyses of native language and the way that their analyses further
influenced French Enlightenment thought. The contrast between Catholicism and Protestantism
in their approaches to language is also relevant and reminds us that linguistic description was not
the simple recording of Anatural@ facts, but a complex argument about Amoral@ capacities.
For seventeenth century thinkers language was an important indication of the capacity for
ACivility@, APolity@ and AReligion@ 6 which set human beings above animals and corresponded to
the historical level of development of society as a whole. In this way analysis of native
languages was integral to the development of colonial and missionary policies. Breton (quoted
above) asserted that the Caraïbe did not have words that would enable or reveal cultural
development and so by implication provided justification the French colonial project in
Dominica.
In religious debates of the era concerning the evolution of human society and the role of
divine creation, understanding the origin of language was as relevant for the doctrine of natural
law as it was for Biblical criticism7. In turn many Enlightenment philosophers were profoundly
influenced by the work of the missionaries. Indeed Jean-Jacques Rousseau used du Tertre's
characterization of the Caribs as "noble savages" as a point of departure in his writings on human
nature and society (Hulme & Whitehead 1992:128) and the influence of both missionary

6
During the fifteenth century the evangelization of Brazil French missionaries likewise
formulaically judged many indigenous groups to be Asin loi, sin foi, sin roi@ and so the more
difficult to convert (Whitehead 1993b, see also Bono 1995).
7
Ricken (1994: 140) notes, "every... theory of the origin of language also contains
considerations of the origins of society …but also as it pertained to the nature of the human
 14

linguistic judgments, and the philosophical assumptions of Enlightenment thinkers, are still very
much present in modern anthropology, as we have seen.
Raymond Breton
By contrast with Rousseau the views of Raymond Breton (Hulme & Whitehead
1992:107-116) on the origins of Karipuna society were fully consistent with his negative views
of their language. Breton concedes that the Karipuna are not monstrous cannibals, as was the
Columbian representation, but he does see them as truly sauvage, lacking strict marital laws, and
so apt to practice incestuous relations. They lack the capacity for human affection and merely
mate out of instinct and a desire to reproduce. Crucially, he claims that there is a separate
language for men and women, and this is reflected in his massive two volume Dictionnnaire
(Auxerre 1665-6) which paradoxically expresses a supposed absence of linguistic complexity,
via an excess of lexical notation and cultural explanation. The Dictionnnaire also systematically
favors male speech forms over female in the representation of the Karipuna speech community.
Breton thus firmly, but incorrectly, inscribes the notion that male speech-forms, referred to as
Kalinago8, constituted a distinct language. Breton=s account of the Karipuna language of course
reflects European colonial thought and the cultural construction of the colonized. Indeed, even as
the AIsland Caribs@ provide Rousseau with an icon of noble savagery, so too they function as the
Awild man@ of the European imaginary. Bartra (1994: 124) neatly summates the attitudes
encapsulated in this icon of the colonized and its connection to theories of language and
civilization;
"The wild man did not have language, but took words by storm in order to express the
murmurings of another world, the signals that nature gave to society. The wild man spoke
words that did not have literal meaning, but were eloquent in communicating sensations
that civilized language could not express. His words were devoid of sense, but expressed

species at the beginning of human history."

8
Lexically this jargon was based on Cariban Kariña but employed an Arawakan syntax,
consistent with the natal language, Karipuna, of its male users (Hoff 1995).
 15

feelings."
In just the same way Caliban, in Shakespeare=s The Tempest, is a ghostly reminder of the reality
of Karipuna survival in a colonized Caribbean and the notions of linguistic superiority that
underpinned that colonization;
A ... I pitied thee,
Took pains to make thee speak, taught thee each hour
One thing or other. When thou didst not, savage,
Know thine own meaning, but wouldst gabble like
A thing most brutish, I endow=d thy purposes
With words that made them known.@
(Act I, Scene II, lines 353-8)

Jean Baptiste du Tertre

A similar depiction, consonant with the same general linguistic analysis, emerges in the
account of fellow Jesuit Jean Baptiste du Tertre (Hulme & Whitehead 1992:128-137), who
wistfully stresses the rude superiority of the Karipuna over the civilized nature of the Europeans,
which isolates them from their simpler and gentler natures;
"So, at the very word Savage, most people imagine in their mind's eye the kind of men
who are barbarous, cruel, inhuman, without reason, deformed, as big as giants, as hairy as
bears: in a word, monsters rather than reasonable men; although in truth our Savages are
Savages in name only, just like the plants and fruits which nature produces without
cultivation in the forests and wildernesses, which, although we call them wild, still
possess the true virtues in their properties of strength and complete vigor, which we often
corrupt by our artifice, and change so much when we plant in our gardens." (Hulme and
Whitehead 1992: 129)
Thus du Tertre, while accepting Breton's ideas about the origins of Karipuna culture and
language, sees the ACaraïbes@ as exemplifying a noble simplicity, and he stresses the difficulty in
learning their language, that it is "impoverished and imperfect" (Hulme and Whitehead 1992:
137). Breton is accordingly congratulated for having made their conversion more possible
 16

through his notation of their speech and the main obstacle to evangelism becomes the poor
treatment they have learned to expect from the Europeans. This contrast between Breton's and du
Tertre's account is interesting for the way in which it highlights shifting missionary attitudes, but
notice that the ethnology, captured in the linguistic judgements of Breton, goes unchallenged. Du
Tertre writes;
"They have good reasoning, and a mind as subtle as could be found among people who
have no smattering of letters at all, and who have never been refined and polished by the
human sciences, which often, while refining our minds, fill them for us with malice; and I
can say in all truth that if our Savages are more ignorant than us, so they are much less
vicious, even indeed that almost all the malice they do know is taught them by us
French."
Hulme and Whitehead 1992: 130

A linguistic incapacity, whatever its origins, thus still remains the key trait of “primitive@
society.
Sieur de la Borde
While de la Borde was not a missionary, he was influenced by the Jesuit missionaries
with whom he worked, especially Father Simon. He was either part of the French military and
naval presence or a functionary of the local administration. De la Borde shares with Breton the
idea that the Caraïbes are savages, with no trace of the nostalgia for simplicity that du Tertre
shows. Nonetheless de la Borde does provide an important description of the myths and spiritual
beliefs of the Caraïbes, although he treats most of these beliefs as primitive superstitions, he
does acknowledge that the Caraïbes are capable of forming ideas of spirituality and divinity,
albeit regarding the devil and evil spirts. He writes;
"Their language is very destitute: they can only express what is obvious. They are so
materialist that they do not have a term to designate the workings of the spirit, and if the
beasts were able to speak I would want to give them no other language than that of the
Caraibes. They have not one word to explain matters of religion, of justice, and of what
pertains to the virtues, the sciences, and a great number of other things about which they
 17

have no notion. They are not able to converse, as I have said elsewhere."
(Hulme and Whitehead 1992: 153)
However, it is again a contradiction and irony that de la Borde provides detailed descriptions of
the complex spiritual beliefs of the Caraïbes, only to suggest that they are linguistically
impoverished. Nonetheless, he precedes this passage with the following comments;
AAlthough there is some difference between the language of the men & that of the
women, as I have said in the chapter on their origin, nevertheless they understand one
another. The old men [also] have a jargon when they are dealing with some plan of war,
which the young do not understand at all.@ (Hulme and Whitehead 1992: 153)
Borde also refers to a copious linguistic study made by one Father Simon that Awill be useful to
those who might plan to acquire some awards in the conversion of these infidel peoples@ and one
wonders if this lost work might have given a very different view of gender, age and the linguistic
practices of the Caraïbes given these few tantalizing remarks.
Charles de Rochefort
Charles de Rochefort provides an account of the Caraïbes which notably differs from that
of Breton, du Tertre and de la Borde. This contrast is certainly connected to the fact that he was a
Protestant in the service of a particularly anti-clerical governor in Dominica and his discussion
of the Caraïbe speech practice is therefore revealing. He immediately stresses the unity of
human speech practices noting that, "The Caribbians have an ancient and natural language, such
as is wholly peculiar to them, as every nation hath that which is proper to it." (Hulme and
Whitehead 1992: 118) and emphasizes that;
"What advantage soever the Europeans may imagine they have over the Caribbians,
either as to the natural faculties of the mind, or the easiness of the pronunciation of their
own language, in order to the more easie attainment of theirs, yet hath it been found by
experience, that the Caribbians do sooner learn ours than we do theirs." (Hulme and
Whitehead 1992: 119)
Rochefort actually offers a quite detailed account of the Karipuna language and consonant with
the idea that attitudes to language are part of a wider cultural interpretation, also challenges the
established theory, so often advanced to explain gender differences in speech, of a Carib
 18

invasion from the mainland. Certainly his work was controversial in its own day because of the
explicit challenge it made to Jesuit views of the Karipuna and he was accused of having
plagiarized the work of Du Tertre. Subsequent commentators, no doubt because of the
ethnographic authority of du Tertre=s=s own work, due to its association with that of Raymond
Breton, have also largely accepted du Tertre=s published accusation against Rochefort.
Nevertheless, Rochefort=s account is accurate and intelligent for the way in which it recognizes a
plurality of influences in the linguistic repertoire of the Caraïbes. He also illustrates the
complexities and idiomatic uses of Karipuna speech, noting that besides their Aancient and
natural language@;
A... they have fram=d another bastard-speech, which is intermixt with several words taken
out of foreign languages by he commerce they have had with the Europeans... among
themselves they always make use of their ancient and natural language...@ (Hulme and
Whitehead 1992:118)
In short we have here clear testimony as to both the propensity for the formation of creolized or
pidgin languages by the Karipuna due to the presence of the Europeans, as well as gender and
age differences in the use of similar jargons formulated via interactions with other indigenous
peoples. None of these complexities have been adequately recognized in the missionary or later
accounts.
The Enlightenment and Linguistic Representation
As is anticipated in the account of Rochefort, Enlightenment philosophies increasing
emphasized the idea that language was not a pre-ordained product of divine intervention but the
result of human experience and custom and as such open to human manipulation (see Ricken
1994, Bono 1995). Principal among the proponents of this view were John Locke in An Essay
Concerning Human Understanding (1690) and Etienne Condillac Lettres Philosophiques sur les
Anglais (1734). For Locke there were no innate ideas and all human thought and classification
had its origin with sensory experience, understood as both sensation and reflection, or memory.
In short, God did not invent language but placed humanity in the world with a capacity for such,
and in this way modern theory, supplanting God by the inheritable cognitive and motor abilities
which support speech behavior, remains embedded in Enlightenment analysis. This line of
 19

reasoning was a radical departure with the Cartesian and pre-Cartesian traditions of seeing
humans differing from animals through the possession of a faculty of raison, of which language
was the prime symptom. For Descartes animals were mere automatons, lacking raison and so the
ability to use or learn language. For Locke, however, both humans and animals show cognitive
activities that develop on the basis of sense perception, yet only for humans do these reach such
a level of abstraction that they become expressed in words.
This sensualist philosophy was further developed by Condillac (Ricken 1994:80) who
places the origins of language and thought in a phylogenetic, or evolutionary and historical,
perspective, rather than the ontogenetic relationship pictured by Locke. Again this debate is still
current in modern linguistics as the resurgent interest in the materialist theories of Vygotskii
(1986, 1994) illustrate.
The entanglement of linguistic philosophy and ethnological observation is extensively
and overtly developed in the writings of Jean-Jacques Rousseau (1959), who broadly adopted
Condillac=s sensualist philosophy but added to that a distinct historical sense of the conflicts and
differences that arise as a result of the contradiction between the social nature of human beings
and the inequalities of their social existence. Rousseau therefore explicitly links language theory
and anthropological theory through sensualist philosophy. The result is the re-discovery of the
primitive as a subject free from the constraints and inequalities of the civilized, and expressing a
unique, untranslatable and even impenetrable, cultural outlook. It is arguable that we have yet to
divest ourselves of such notions as recent debates on cultural commensurability and
comparability would suggest (Obeyesekere 1992, Sahlins 1995). Moreover such ideas are still
relevant to anthropological theory since advocates of linguistic relativity in cognition supplant
the Lockean notion of Ainnate ideas@ with the Whorfian argument that anthropological linguistics
would be another way through which the culture and mentality of a particular linguistic
community could be uniquely revealed - as in the well known example of the supposed absence
of recognizable temporal terms in the Hopi language (Whorf 1962:58).
However, for these reasons ideas about the origins and development of language are not
just matters for linguistic description and analysis but reverberate in current anthropological
theory in a number of ways. In archaeology the assumption of a close Afit@ between language and
 20

culture is necessary for the idea that linguistic groups represent historical (archaeological)
cultures (Lathrap 1970). This is not to suggest that there are never continuities and historical
equivalences between a speech community and a socio-cultural group (Loukotka 1968); but it
does mean that these have to be demonstrated before glottochronology can be used to substitute
for history or other kinds of temporal sequencing (Renfrew 1987, Whitehead 1993a). This much
is clearly shown by the divergence between linguistics and archaeology over the time-depth of
AIsland Carib@ occupation in the Caribbean discussed above and in the utter failure of attempts to
distinguish the ACarib conquest@ as a discrete style emerging in the ceramic sequence for the
Lesser Antilles (Boomert 1985: 30-33). Moreover, the wider implications of the over-
identification of language with a cultural Aworld-view@ become evident in the work of Greenberg
(1966, 1987) who, with the geneticist Cavalli-Sforza et al. (1996), has recently grouped most of
the world=s languages into just eighteen primal groupings. On the basis of genetic similarities
among modern speakers from these groupings, these language distributions are also held to be
expressive of a Arace@ history.
Certainly, the idea of a close the integration of language and culture has also been often
contested and has led repeatedly to the formation of theories concerning the role language plays
in the development of specific representational and cognitive modes within a given linguistic
community. However, what should have become very clear from a consideration of the case of
the AIsland Carib@ is that while a language is a Wittgensteinian Aform of life@, a cultural
phenomenon, it is also an historical one, and this fundamentally affects the character of its
development and so the relevance and validity of any comparative exercise.
This history of the Karipuna and the way it is reflected in linguistic usage through time
makes the search for an Arawakan cultural-linguistic substrate that might function to identify
AArawakan@ peoples in the historical past appear quite pointless. The Arawakan Karipuna have
been Acaribe@ for so long that even today ethnologists are unable to quite let go of the idea that
they are Caribs in some sense - for indeed that is indeed the opinion of their modern descendants,
the Garifuna, themselves. The story of the Garifuna of Belize is therefore instructive as to the
meaning and colonial origins of the categories of AArawak@ and ACarib@, the creolization of an
Arawakan language, and the confusion this causes to an anthropology still dependent on the
 21

dualism of the colonial past and wedded to the idea of language as a cultural substrate that
produces social continuity through time.
The Garifuna are the descendants of African slaves who fled to St.Vincent from the sugar
plantations of Barbados. The wreck of a slave ship off St Vincent in 1635 greatly augmented the
black population who were integrated into Karipuna society, as they had been throughout the
previous century as well (see Hulme & Whitehead 1992:38-44). Over the next 150 years the
ABlack Caribs@, as they were known, grew in political significance within the colonial rivalries of
the French and British for control of the Lesser Antilles. The Carib War by the British against
these Karipuna communities in 1795 lasted three years with the result that the British deported
the entire ABlack Carib@ community to an island off Honduras from where they gradually
migrated to the mainland of Honduras and Belize. Their communities have survived into the
present day and still speak the Garifuna language, unlike their fellow AAmerindian@ or ARed@
Caribs in St. Vincent and Dominica who retain only a few words and phrases of Karipuna.
For the eighteenth and nineteenth century colonial regimes of this region, however, these
caribes were quite different from the aruacas who were retroactively identified with the lost
populations of the Greater Antilles. In fact the term aruacas historically referred to the Lokono9,
settled from the Amazon north to the Essequibo along the Atlantic coast and into the uplands at
the head of the Demerara, Berbice, and Corentyn rivers. The Lokono quickly allied with the
Spanish who were attempting to settle the Orinoco and Guyana coast in the sixteenth century,
since they received Spanish military assistance in occupying rivers to the north of the Essequibo,
including the Pomeroon, Orinoco and parts of the Caribbean coast of Venezuela. Here the
Lokono drove out the existing population comprised of Kariña, Warao, Yao, Nepoyo and
Suppoyo. The Lokono were also given black slaves by the Spanish to work the tobacco
plantations they had pioneered in the lower Orinoco. These events were the origin of a lasting

9
Aruaca derives from the Lokono word for manioc flour, aru, that was their principal item of
trade to the Spanish, just as the name Pomeroon derives from baurooma, a ball of such flour
(Bennett 1989), reflecting the strategic nature of that river in the trade with the Spanish.
 22

military exchange between the Lokono and the Kariña, who in turn made use of Dutch and
French allies in opposition to the aruaca occupation of the Essequibo and Orinoco regions
(Whitehead 1988).
The Karipuna played into this situation as allies of the Kariña (hence their honorific in
the men=s jargon Kalina-go) and as war and trade partners of the Lokono. The tradition of raid
for women and guanin (gold work) between the Karipuna and Lokono was thus expressive of
their basic cultural similarities; Loquo was the first man in both Karipuna as well as Lokono
myths of origin, and the sources of the magic metal guanin lay in an exchange of women for this
substance with the mythical ancestor Guahayona. In this way Lokono and Karipuna conflicts
and exchanges in the seventeenth and eighteenth centuries reproduced the military and ritual
exchanges in the fifteenth and sixteenth century that were described by Columbus. However,
none of this was understood (or at least it was ignored) by earlier commentators who saw in the
tales and practices of Karipuna and Lokono raiding another aspect of a supposed manichean
struggle between Arawak and Carib across the whole of northern South America. In this way as
the ethnologists of the eighteenth and nineteenth centuries moved to classify and delineate major
cultural and linguistic relationships, the scheme of AArawak vs. Carib@ seemed a ready made
heuristic device. This model then attracted further confirmation as a specifically linguistic style
of comparison deriving from the work of missionary evangelists encrusted this distinction with
further evidence - notwithstanding the gross anomalies this created in describing and interpreting
perhaps the best documented and most studied Arawakan population in the whole of the
Americas - the Island Caribs. It thus transpires that the category AArawak@ is no less historically
and culturally complex than its twin ACarib@ and the Karipuna utterly transgress such ethnic,
cultural and linguistic boundaries.
The urge to group such cultural complexity and variety into finite categories has its
intellectual roots in the western scientific project as a whole, but the immediate historical
impulse to such an approach to cultural and linguistic typology was the colonial conquest itself.
As we have seen the role of missionary evangelists in both constructing languages from the
speech behavior of the native population, as well as their role in providing ethnological context
for colonial policy resulted in a perfect identification of linguistic and ethnic identity. Of course
 23

such keen observers were not unaware of the anomalies this produced in practice and the Jesuit
missionaries of Orinoco were fully aware, and utterly frustrated, at the tendency of non-Cariban
speaking groups to become caribe for the same political and economic reasons that the Karipuna
did (Whitehead 1998b). Nonetheless, they pursued policies of settling evangelized populations in
villages that were mono-lingual, thus directly acting to produce that Afit@ of culture, society and
language that was a theoretical desiderata of linguistic theories of the time.
In the absence of this missionary infrastructure, as in the Dutch, French, and British
Guianas, an implicit system of ethnic ranking achieved the same effect. ACarib@ groups were
treated as wild but fierce mercenaries and were used to hunt down escaped black slaves and to
provide a buffer against Spanish expansion beyond the Orinoco basin. AArawaks@ were used to
guard the immediate plantation and to provide servants in the planter=s household. They were
also courted and co-opted by the missionaries as evangelical agents among the hinterland
peoples, just as they had acted as military intelligence for the Spanish of the sixteenth century10.
By underwriting and promoting a strong identification of language and political attitude the
permeability of ethnic boundaries, clearly evident from the history of the Karipuna, was
curtailed. Well-defined ethno-linguistic groups - something that was no less the object of
Anationalist@ policies in Europe of the nineteenth century - enabled better administrative control
of the native population (Whitehead 1999c). As a result, by the end of the nineteenth century,
European national political loyalties also spread amongst the Amerindians producing indigenous
groups calling themselves ASpanish Arawaks@ and British Arawaks@, who then acted as the
slavers and evangelists of their own and neighboring peoples (Whitehead 1990a, 1990b). It
therefore would appear that the correlation between linguistic groups and socio-cultural ones is
uncertain at best, for speech communities may be riven by political, economic and ideological
divisions that in practice outweigh the notional ties of sentiment and cultural similarity that
common speech modes would seem to imply.
This created no few problems for the linguists of the nineteenth century who, working
from the missionary materials gathered in the widespread evangelization of native populations in

10
In particular the native evangelist Jeptha, a Lokono from the Berbice river, provided the
Moravians in Surinam with a continent-wide digest of the location of various ethnic groups and
 24

the eighteenth century, were unable to properly classify the Karipuna population. Im Thurn
(1883) was the first to attempt to resolve this situation by designating the Kariña as ATrue
Caribs@ and the Karipuna as AIsland Caribs@. This was partly done not just from the linguistic
evidence but also via a general identification of cannibalism with the presence of ACaribs@.
William Brett (1868) having overseen the opening of some shell-mounds in the Pomeroon
Barima river in north-west Guyana, interpreted the skeletal material uncovered to be the detritus
of a cannibal feasts. He further inferred that the feasts must have been conducted by the local
Kariña in conjunction with their ACarib@ allies from the islands, given the estuarine position of
the site. In fact such skeletal evidence is related to a much more ancient occupation of the region
and is funerary in origin (Williams 1981). In this way Brett and Im Thurn perfectly recapitulate
the false ethnological inferences made by Chanca in his fifteenth century account of Guadeloupe
(discussed above) and so provided a revitalized basis for the persistence of the old Arawak /
Carib dualism.
Other attempts to classify Arawakan languages moved to a new level with the work of
Daniel Brinton (1871, 1891). Brinton (1871) demonstrated the stability of the Lokono lexicon
through comparison with sixteenth century materials and made reference to the work done on the
Lokono language by missionaries in Surinam (see Crevaux 1882, Quandt 1807, Stæhelin 1913).
In his search for linguistic affiliates to the Lokono language and with the aim of identifying an
Arawakan family of languages Brinton considered historical sources mentioning the term
AArawak@ which suggested connections with western Venezuela and the Amazon north bank.
However, it was in the Caribbean that he felt the closest connection would lie and so he
attempted to reconstruct elements of the Igñeri or AIsland Arawack@ language as well as that of
the Greater Antilles, though he chose not to term this ATaíno@. Again, with regard to the story of
ACarib conquest@, Brinton wrote (1871:1);
AFrom the earliest times they [Arawaks] have borne an excellent character. Hospitable,

their associated political relationships with each other (Stæhelin 1913, II-2:174-5)
 25

peace-loving, quick to accept the humbler arts of civilization and the simpler precepts of
Christianity, they have ever offered a strong contrast to their neighbor, the cruel and
warlike Caribs.@
Precisely because of his credulity with regard to this colonial scheme Brinton never attempted a
comparison between Karipuna and Lokono lexicons and so did not even consider including the
Karipuna in an AArawakan@ language grouping.
This same framework of historical and linguistic interpretation was unfortunately again
adopted by de Goeje (1939) who had already done much to expand the recording of Lokono (de
Goeje 1928). Still considering a ghost-language, Igñeri, to have been the aboriginal language of
the Karipuna - i.e. before the supposed ACarib conquest@ - he convincingly demonstrated
continuities and relationships between Lokono, female word forms from Karipuna and the
Alanguage@ of the Greater Antilles that he called ATaíno@. However, he did take the suggestion
made by Adam (1878) who had noticed that the male speech forms in Karipuna were close to
Kariña, and those of the women were close to Lokono. He also realized that the still extant
Garifuna were a source of further information on these linguistic relationships and included
materials form the ACaribe du Honduras@ for comparative purposes, and as an example of
AMaipuran Arawak@. However, although gender difference in lexical items was certainly
apparent from these comparisons, in fact his tables (1939:3) actually show that in all but four out
of the nine categories of lexical comparison, words forms in common between men and women
exceed those that were distinct.
As already indicated, it is not surprising therefore that when the linguist Douglas Taylor
and archaeologist Irving Rouse published a joint article (Taylor & Rouse 1955) on the peopling
of the Caribbean, AWe found ourselves in complete disagreement@ (Rouse 1985: 18). On the one
hand Rouse thought that the ceramic evidence showed that there had indeed been a movement
from the mainland to the islands in late pre-history which he assigned to the ACarib conquest@.
However, Taylor had already recognized the inconsistencies in this position, especially the
identification of ATaíno@ with AIgñeri@. This seems to imply that the Antilles were peopled by two
distinct migrations of different Arawakan tribes.... In this case, it seems unnecessary to assume
than any Aconquest@ or fighting took place.@ (1955:108-9)
 26

Taylor also suggested that Karipuna was part of a ANu-Arawak@ family, which following
Mason=s suggestion that this grouping be so named for the invariable presence of nu as first
person pronoun, included the Campa and Amuesha11. Thus Karipuna origins were still seen as
extraneous to the islands but their linguistic affiliations and the fallacy of a ACarib conquest@
theory was beginning to be recognized. In conjunction with the linguist Berend Hoff, Taylor was
finally realized that the Kariña elements in the men=s speech actually were assimilated using an
Arawakan syntax (Taylor & Hoff 1980). However, this important finding was not integrated in
archaeological understanding and Irving Rouse (1985), though now recognizing the Arawakan
nature of AIsland Carib@, prefers to classify it along with ATaíno@ as a separate AWest Indian@
branch of Anorthern@ Arawakan. Rouse (1985, 1986) also now accepts that Aimmigration@ into the
islands best explains the nature of the ceramic evidence, but the idea of a conquest to explain
gendered speech modes remains despite the many cogent archaeological reasons for rejecting it
(Boomert 1995).
In short the Karipuna have continued to challenge conventional forms of linguistic and
cultural classification and this suggests that our categories of classification are simply inadequate
to the complexity and dynamism of indigenous linguistic practices - just as the linguistic
exogamy of Tukanoan groups in the western Amazon confounded historical linguists into
suggesting a compression of previously dispersed populations, instead of appreciating the way in
which language was manipulated as a cultural and ethnic marker by native people, themselves

11
Taylor noted that Karipuna was such a language and so unlike Lokono, Goajiro or
reconstructed Taíno which have a prefixed marker of first person singular with apical stop (T-
dA, L-dA, G-tA) not nasal as for A...Island Carib, Campa and probably the majority of Arawakan
languages@ (1954:154).
 27

rarely mono-lingual anyway (Reichel-Dolmatoff 1996). As Sorensen realized in his analysis of

multilingualism in northwest Amazonia; 'A linguistic theory limited to one language / one group
situations is [itself] inadequate to explain ... actual linguistic competence` (1972, 91). A point
strongly reiterated by Colson as regards groups of northeast Amazonia (1983a, 11; 1983b).
Conclusion
I have tried to show how the category AArawak / aruaca@, originally political as well as
linguistic in its meaning, subtly evolved into a colonial cultural classification that in turn
constrained the development of both historical and ethnographic understanding of the indigenous
people in the Caribbean and north-eastern South America. This suggests that a linguistic
connection or relatedness by itself does directly not translate into social and cultural propinquity
but are produced by processes of historical transculturation, such as occurred in the case if the
>Island Caribs=. This implies that the relationship between language and the rest of culture is a
matter for historical investigation, through archaeology, linguistics and historiography such as is
carried out in the papers collected here. The evidence of the comparative Arawakan histories
presented through the case studies in this volume show many such relationships. The substantive
comparisons that emerge from this volume proceed by reference, not to the mere presence of
linguistic similarity, but also to the cultural products of shared historical circumstance, such as
ritual discourse. For example, the Karipuna areyto, a ritual forum for male and female oratory
about the past and its continued presence in a landscape of mythic significance (see discussion of
guanin and Guayahona above), is clearly analogous to the ways in which musicality, enchanted
landscapes, and supra-ethnic sodalities have produced and defined ethnic consciousness in
multiple contexts, both AArawakan@ and otherwise - as in Reichel-Dolmatoff=s (1996) discussion
of the Yuruparí myth of the Tukano. This may not uniquely define AArawakans@ as opposed to
others, but such long term cultural features do demonstrate a substantive historical aspect to
Arawakan identity. Similarly, a wide range of evidence presented in the following chapters
strongly indicates long term continuities and similarities in the local socio-cultural practices of
Arawakan speakers. This is particularly important where the archeology and history (see
Heckenberger, Vidal and Zucchi this volume) produces striking analogies with contemporary or
recent ethnographic description of the ritual use of landscape and the practice of social hierarchy
 28

(see Hvalkof, Passes, Pollock, Gow and Wright this volume). However, very distinct kinds of
historical and socio-cultural experience are also present among Arawakan speakers, as is shown
by the contrasting social and military orientations of, say, the Matsigenka and Piro (Gow,
Rosengren - this volume) to the Palikur or Karipuna (Passes, Whitehead - this volume). This
suggests that we can already demonstrate strong local or regional historical and cultural
relatedness among Arawakan groups and that an even broader relatedness is to be expected.
Moreover, notions of >Arawakaness= do not emerge only from contemplation of the peoples
discussed in this volume, but take shape from the similar relatedness of other language families,
such as the Tupi-Guaraní. However, five hundred years of colonial conquest has badly damaged
our ability to reconstruct the historical and cultural interactions of many peoples and that process
itself has marked modern indigenous consciousness of history and cultural identity (Hill 1996). It
has been the aim of this chapter to show how that process has to be carefully thought through, as
in the case of the >Island Caribs=, in all the local and regional contexts that we encounter
Arawakan speakers. However, what the papers here clearly indicate is that, despite these
obstacles, Arawakans share a substantive cultural repertoire that has proved highly resilient to
such external intrusions, producing a distinct historical trajectory that is still being played out. In
this way the identification of the nature of that Arawakan historicity has become integral to all
future archaeological, historical and ethnographic understanding, not just of Arawaks, but of
indigenous South America overall.

Bibliography
Adam, Lucien
1878 Du parler des hommes et du parler des femmes dans la langue caraïbe. Mémoires de
l=Academie de Stanislas. Paris.
Anghiera, Pietro Martire d',
1530 De Orbo novo. Compluti, apud Michele d'Eguia.
Barrère, P.
1743 Nouvelle Relation de la France Equinoxiale. Paris.
Bartra, R.
 29

1994 Wild men in the looking glass : the mythic origins of European otherness. Ann Arbor :
University of Michigan Press.
Basso, Ellen B. (ed),
1977 Carib-speaking Indians : culture, society, and language. Tucson : University of Arizona
Press.
Bennett, John
1989 An Arawak-English Dictionary. Archaeology and Anthropology 6 (1,2)
Biet, A.
1664 Voyage de la France Equinoxiale en l'isle Cayenne. Paris: F. Clouzier.
Bono, James J.
1995 The Word of God and the Languages of Man, Interpreting Nature in Early Modern
Science and Medicine. 1: Ficino to Descartes. Madison: University of Wisconsin Press.
Boomert, Aad
1995 Island Carib Archaeology. In: N. L. Whitehead (ed.) Wolves from the Sea. Readings in
the Archaeology and Anthropology of the Island Carib, pp.23-36. Leiden: KITLV Press.
Boyer, C.
1654 Veritable Relation de [...] l'Amérique Occidentale. Paris.
Breton, R.
1665 Dictionnaire caraïbe-françois. Auxerre: Giles Bouquet.
1666 Dictionnaire françois-caraïbe. Auxerre: Giles Bouquet.
Brett, William Henry
1868 The Indian tribes of Guiana : their condition and habits. With researches into their
past history, superstitions, legends, antiquities, languages. London : Bell and Daldy.
Brinton, D. G.
1871 The Arawack Language of Guiana in its Linguistic and Ethnological Relations. American
Philosophical Society Transaction 14:427-44.
1891 The American Race. A linguistic Classification and Ethnographic Description of the
Native Tribes of North ans South America. New York: N. D. C. Hodges.
Cavalli-Sforza, L. L., Paolo Menozzi, & Alberto Piazza.
 30

1996 The history and geography of human genes. Princeton, N.J. : Princeton University Press,
Chrétien, J.
1725 Lettre inédite sur les Galibis. [Extrait du Mss. 847, Biblioteque Méjane]. Chantilly:
Archives des jésuites des Paris.
Colson, A.
1983a A comparative survey of contributions. Antropologica 59/62:9-38.
1983b The spatial component in the political structure of Carib speakers (Guiana Highlands).
Antropologica 59/62:73-124.
Crevaux, Jules, P. Sagot, & L. Adam
1882 AArawakisch-Deutches Wörterbuch@, in Grammaires et vocabulaires roucouyenne,
arrouague, piapoco et d'autres langues de la région des Guyanes. Paris : Maisonneuve,
1882.
de Goeje, C.H.
1928 The Arawak Language of Guiana. Bijdragen :1-47,246-267
1939 Nouvel Examen des langues des Antilles. Journal de la Société des Américanistes
31:1-120.
Gil, J. & C. Varela (eds.)
1984 Cartas de particulares a Colón y relaciones coetáneas. Alianza Editorial: Madrid.
Greenberg, Joseph H.
1966 The Languages of Africa. Bloomington; Indiana University Research Center in
Anthropology, Folklore, and Linguistics ; 25.
1987 Language in the Americas. Stanford University Press
Greenblatt, S. J.
1991 Marvelous Possessions. The Wonder of the New World. Oxford: Clarendon Press
Grenand, F. & P.
1987 La Côte d'Amapa, de la Bouche Amazone à la Baie d'Oyapock à travers la tradition
orale Palikur. Boletim Museu Goeldi (sér. Antropol.) 3(1):1-77.
Hill, J (ed.),
1996 History, Power and Identity, Ethnogenesis in the Americas, 1492-1992. Iowa City:
 31

University of Iowa Press.

Hoff, Berend
1995 Language, contact, war and Amerindian historical tradition. The special case of the
Island Carib, in N. L. Whitehead (ed.) Wolves from the Sea. Readings in the Archaeology
and Anthropology of the Island Carib, pp. 37-60. Leiden: KITLV Press.
Hulme, P. & N. L. Whitehead,
1992 Wild Majesty. Encounters with Caribs from Columbus to the Present Day. Oxford:
Clarendon Press.
Im Thurn, E. F.
1883 Among the Indians of Guiana. London: Kegan Paul.
Kroskrity, Paul V. (Ed)
2000 Regimes of Language. Ideologies, Polities, and Identities. Santa Fe: SAR Press.
Lathrap, D.
1970 The Upper Amazon. London: Thames and Hudson.
Loukotka, Cesmir
1968 Classification of South American Indian Languages. Los Angeles: Latin American
Center, UCLA.
Maghæles, P de
1576 Historia da provincia santa Cruz a que vulgarmete chamamos Brasil. Lisbon. [New
York: Cortes Society]
Obeyesekere, G.
1992 The Apotheosis of Captain Cook. European Myth-making in the Pacific. New Jersey:
Princeton University Press.
Oviedo, G. F. de
1535 Historia general y natural de las Indias, islas y Tierra Firme. Sevilla: Juan Cromberger.
Pané, Ramón
1999 An Account of the Antiquities of the Indians. (Ed. J. J. Arrom). Durham: Duke University
Press.
Pelleprat, P.
 32

1655 Relation des missions des pères jésuites. Paris.

Quandt, C.
1807 Nachricht von Suriname. Amsterdam: S. Emmering.
Rafinesque, C. S.
1836 The American Nations; or, Outlines of their General History, Ancient and Modern.
Philadelphia.
Reichel-Dolmatoff, Gerardo
1996 Yuruparí. Studies of an Amazonian Foundation Myth. Cambridge (Mass.): Harvard
University Press.
Renfrew, C.
1987 Archaeology and Language. London: Jonathan Cape.
Ricken, Ulrich
1994 Linguistics, Anthropology and Philosophy in the French Enlightenment. London:
Routledge.
Rouse, I.
1948a 'The Arawak`, in: J. Steward (ed.) Handbook of South American Indians (vol.4).,
pp.507-546. Washington: Smithsonian Institution.
1948b 'The Carib`, in: J. Steward (ed.) Handbook of South American Indians (vol.4).,
pp.547-565. Washington: Smithsonian Institution.
1985 Arawakan phylogeny, Caribbean chronology, and their implications for the study of
population movement. Antropologica 63-64: 9-21.
1986 Migrations in Prehistory: Inferring Population Movement from Cultural Remains. New
Haven: Yale University Press.
1992 The Tainos: Rise & Decline of the People Who Greeted Columbus. New Haven: Yale
University Press.
Rousseau, J.-J.
1959 Discours sur l'origine et les fondements de l'inegalité parmi les hommes. Paris:
Bibliotèque de la Pléiade.
Sahlins, M.
 33

1995 How Natives Think. About Captain Cook For Example. Chicago: University of Chicago
Press.
Sorensen, A. P.
1972 Multilingualism in the Northwest Amazon. In Sociolinguistics. eds. J. Pride & J. Holmes,
pp.78-93. Harmondsworth: Penguin Books.
Stæhelin, F. (ed)
1913 Die Mission der Brudergemeine in Suriname und Berbice im achtzehnten Jahrhundert.
Paramaribo-Hernhutt.
Sued-Badillo, Jalil
1995 Searching for Ethnicity: An Economic Contextualization. In: N. L. Whitehead (ed.)
Wolves from the Sea. Readings in the Archaeology and Anthropology of the Island Carib,
pp.61-90. Leiden: KITLV Press.
Taylor, D.
1946 'Kinship and Social Structure of the Island Carib`, Southwestern Journal of Anthropology
2:180-212.
1954 'Diachronic note on the Carib contribution to Island Carib`, International Journal of
American Linguistics 20:28-33.
1956 'Languages and Ghost-Languages of the West Indies`, International Journal of American
Linguistics 22:180-183.
1958 'The Place of Island Carib Within the Arawakan Family`, International Journal of
American Linguistics 24(2):153-156.
1977 Languages of the West Indies. Baltimore: John Hopkins University Press.
Taylor, D. & B. Hoff
1980 The Linguistic Repertory of the Island-Carib in the Seventeenth Century: The Men=s
Language - A Carib Pidgin?, International Journal of American Linguistics 46:301-312.
Taylor, D. and I. Rouse,
1956 'Linguistic and Archaeological Time Depth in the West Indies`' International Journal of
American Linguistics 21:105-115.
Tertre, J. B. du
 34

1667 Histoire générale des Antilles habitées par les François. [4 vols.] Paris: Thomas Jolly.99
Trouillot, M-R.
1991 'Anthropology and the Savage Slot. The Poetics and Politics of Otherness`, in R. G. Fox
(ed.), Recapturing Anthropology, pp.17-44. Santa Fe: SAR Press.
Vygotskii, L. S.
1986 Thought and language. (edited by Alex Kozulin) Cambridge, Mass. : MIT Press.
1994 The Vygotsky reader. (edited by René van der Veer and Jaan Valsiner). Oxford, UK ;
Cambridge, Mass. : Blackwell.
Whitehead, N. L.
1988 Lords of the Tiger Spirit. A History of the Caribs in Colonial Venezuela and Guyana,
1498-1820. [KITLV Caribbean Studies Series, 10]. Dordrecht/Providence: Foris.
1990a 'The Snake Warriors -Sons of the Tiger's Teeth. A Descriptive Analysis of Carib Warfare,
1500-1820`, in J. Haas (ed.) The Anthropology of War, pp.146-170. Cambridge:
Cambridge University Press.
1990b 'Carib Ethnic Soldiering in Venezuela, the Guianas and Antilles, 1492-1820`,
Ethnohistory 37(4):357-385.
1993a Historical Discontinuity and Ethnic Transformation in Native Amazonia and Guayana,
1500-1900. L'Homme 28:289-309.
1993b Native American Cultures along the Atlantic Littoral of South America, 1499 -1650,
Proceedings of the British Academy 81:197-231.
1995a Ethnic Plurality and Cultural Continuity in the Native Caribbean. Remarks and
Uncertainties as to Data and Theory, In: N. L. Whitehead (ed.) Wolves from the Sea.
Readings in the Archaeology and Anthropology of the Island Carib. Leiden: KITLV
Press.
1995b The Island Carib as Anthropological Icon. In: N. L. Whitehead (ed.) Wolves from the
Sea. Readings in the Archaeology and Anthropology of the Island Carib. Leiden: KITLV
Press.
1996a The Mazaruni Dragon. Golden Metals and Elite Exchanges in the Caribbean, Orinoco,
and the Amazon. In, C. H. Langebaek & F. C-Arroyo (eds.) Chieftains, Power & Trade:
 35

Regional Interaction in the Intermediate Area of the Americas, pp. 107-132.

1996b Ethnogenesis and Ethnocide in the settlement of Surinam, In: J. Hill (ed.), History,
Power and Identity, Ethnogenesis in the Americas, 1492-1992, pp. 20-35. Iowa City:
University of Iowa Press.
1998a The Discoverie of the Large, Rich and Bewtiful Empire of Guiana by Sir Walter Ralegh.
(Edited, annotated and transcribed). American Exploration & Travel Series Vol. 71,
Norman: Oklahoma University Press.
1998b Colonial Chieftains of the Lower Orinoco and Guayana Coast. In: E. Redmond (ed.)
Chiefdoms and Chieftaincy in the Americas, pp.150-163. University Press of Florida.
Foreword. In: E. Redmond (ed.) Chiefdoms and Chieftaincy in the Americas, pp. ix-x.
University Press of Florida.
1998c Who is Carib? What is Carib Society? Ethnology, Ethnicity and History. Paper prepared
for the workshop on ACarib Societies@, Georgetown, September 1998
1999a The Crises and Transformations of Invaded Societies (1492-1580) - The Caribbean, In: F.
Salomon & S. Schwartz (eds.), The Cambridge History of Native American Peoples,
vol.III, Chapter 10. Cambridge University Press.
1999b Lowland Peoples Confront Colonial Regimes in Northern South America, 1550-1900,
(in) F. Salomon & S. Schwartz (eds.), The Cambridge History of Native American
Peoples, vol.III, Chapter 14. Cambridge University Press.
2000 'Tribes Make States and States Make Tribes: Warfare and the Creation of Colonial Tribe
and State in Northeastern South America`, in: R. B. Ferguson and N. L. Whitehead
(eds.), War in the Tribal Zone; Expanding States and Indigenous Warfare, pp.127-150.
(2nd edition) Santa Fe: School of American Research Press.
Whorf, B.
1962 Language, Thought and Reality. Cambridge (Mass.): Technology Press of Massachusetts
Institute of Technology.
Williams, D.
1981 Excavation of the Barabina Shell Mound, North-West District. Archaeology and
Anthropology 4(1): 14-34