Adelaar Willem, Muysken Pieter. The languages of the Andes
Подождите немного. Документ загружается.
154 2 The Chibcha Sphere
suggesting that a third-person-singular subject marker can be zero as well. A num-
ber of examples presented by Howard (1977a) contain unambiguous evidence of the
fact that person-of-object is encoded in the verb form. Combinations of third-person
markers are illustrated in (248) and (249). Note that the number of the subject marker
is determined by the total number of participants.
(248) i-mo-xawiy´ana
3O.SG-3S.PL-say
‘They said to him.’
(Howard 1977a: 58)
(249) i-bo-xawiy´ana
3O.SG-3S.D-say
‘He said to him.’
(Howard 1977a: 59)
A first-person-singular object is characterised by the presence of a prefix ˇs
.
-, as in
(250) and (251).
(250) ˇs
.
-mo-ˇc-c-ob´a
1O.SG-3S.PL-F-PR-kill
78
‘They are going to kill me.’ (Howard 1977a: 58)
(251) ˇs
.
-ko-ˇc-at-oben´a-ye
1O.SG-2S.SG-F-UF-have.power-CN
‘You will not be able (to move) me.’
(Howard 1977a: 59)
In addition, the sequence k-bo- indicates first-person subject combined with second-
person object.
(252) k-bo-ˇc-c-ob´a
2O.SG-1S.D-F-PR-kill
‘I shall kill you’
(Howard 1977a: 58)
Nominal case is represented by some twenty-two different suffixes or postpositions
(Mel´endez Lozano 2000). Subject and object remain unmarked. Possessive constructions
are formed by means of the genitive marker -be. The genitive expression precedes its
head.
(253) ´aˇc
.
-be bc-tait´a
I-G big-father
‘my grandfather’
(Mongu´ıS´anchez 1981: 29)
78
The identification of the affixes in the examples remains provisional, considering the lack of
an overall grammatical study. The prefix ˇc(a)- ‘intention’ (Mel´endez Lozano 2000) is here
glossed as F ‘future’; the prefix -at- is glossed as ‘unfulfilled’ in Howard (1977b: 276), -ye as a
‘continuative’ (cf. Howard 1977a: 9). The interpretation of -c- as ‘progressive’ is also tentative.
2.19 Esmeralde˜no 155
Kams´a has a copula verb ‘to be’, as illustrated in (254). Number is indicated both on
the subject and on the nominal predicate.
(254) kem ˇs
.
ˇs
.
´o-
ŋŋ
g m-
n
dm´nobon´-
ŋŋ
g
this infant-PL 3S-be fat-PL
‘These little children are fat.’
(Mongu´ıS´anchez in Mel´endez Lozano 2000c)
Nouns are marked for number. There are distinct suffixes for singular, dual and plural.
These are singular -(n)a or -(n)´a, dual -(a)ta, and plural -(
)
ŋ
ga.
79
Some nouns do not
take a singulative suffix (Juajibioy Chindoy and Wheeler 1973, in Mel´endez Lozano
2000c). The distinction is also reflected in the free personal pronouns, e.g. in aˇc
.
(e) ‘I’,
b
n
d´ata ‘we (dual)’, s
n
d´ata ‘we (dual inclusive)’, b´
ŋ
ga ‘we (plural)’.
2.19 Esmeralde˜no
Esmeralde˜no or Atacame,alanguage of the Ecuadorian coastal region survived until
the second half of the nineteenth century. Together with Cha
palaachi (see section 2.17
Barbacoan), it is the only language of the Ecuadorian coast of which we have some
knowledge. All other languages, including the language of the island of Pun´a, have
disappeared long ago without leaving any documentation of importance. In the nine-
teenth century the Esmeralde˜no language was spoken in the western part of Esmeraldas
province on the lower course of the Esmeraldas river valley (called Chinto in Esmer-
alde˜no). The possibility that it may have been a remnant of a language of wider extension
in the Ecuadorian Pacific region cannot be excluded.
W. B. Stevenson, the British secretary to the president of the Audiencia in Quito, visited
the area at the beginning of the nineteenth century (Stevenson 1825). He reports that
the language was in use in the coastal port of Esmeraldas, a town with a predominantly
negroid population. According to local tradition, a shipload of African slaves had landed
in the area after a shipwreck, killed the native men and settled with the women. Thanks to
these women, the language was preserved. Stevenson also observes that the inhabitants
of the neighbouring town of Atacames spoke Spanish. For a more historical account
of African–Amerindian relations in the Ecuadorian Pacific area, which date back to
the sixteenth century, see Phelan (1967). The zambos (of mixed Indian–black origin) of
Esmeraldas managed to maintain a factual independence throughout the colonial period.
The only data on the Esmeralde˜no language were collected by J. M. Pallares in 1877.
They were published in Wolf (1892), and subsequently reproduced and discussed in
Seler (1902: 49–64) and in Jij´on y Caama˜no (1941: 424–539). Constenla Uma˜na (1991:
85–7) notes a number of typological particularities. Object and subject tend to follow
the verb; the genitive follows its head; adjectives can occur both before and after the
79
The ending -
ŋ
g
is obviously a variant of -
ŋ
ga.
156 2 The Chibcha Sphere
noun they modify. He observes that these syntactic features are unusual for the region
and suggestive of contact with Mesoamerica. A similar influence has been proposed by
archaeologists with regard to the local pre-Columbian culture of La Tolita (Willey 1971:
295; cf. Constenla Uma˜na 1991: 87).
According to the available data, the Esmeralde˜no vowel system consisted of five vow-
els (a, e, i, o, u). However, Jij´on y Caama˜no draws attention to the great amount of
vocalic variation, which may point at the existence of a three-vowel system (a, i, u),
rather than a five-vowel system. He suggests that the language may have had nasal
vowels and reports that there was a special vowel sound, written `a`a or ´a´a,which
was limited to the ending of the past participle (e.g. in yat´a´ale ‘finished’). The ac-
cent is often indicated with a diacritic (by Seler in particular), suggesting that stress was
contrastive.
The consonant inventory comprised voiceless and voiced stops (p, t, k; b, d, g); a
voiceless affricate (ch); voiceless and voiced fricatives (f, s, sh, j, h; v); nasals (m, n);
vibrants (r, rr); laterals (l, ll); and glides (w, y). The interpretation of these symbols
remains uncertain. It is doubtful if the symbol v represented a separate sound, because it
frequently alternates with b.Acontrast between h and j is also unlikely, as they appear to
be in complementary distribution. The symbol h is found (very seldom) in word-initial
position, whereas j mainly appears inside a word. Internal clusters of two consonants
(including geminates) are frequent, often as a result of vowel suppression. The palatal
lateral ll [l
y
]isoften found in syllable-final positions (as in Quechua), e.g. in allki ‘pain’.
Jij´on y Caama˜no’s morphological analysis of the Esmeralde˜no language has to be used
with great care because of methodological deficiencies. (Many forms are overanalysed.)
He has, nonetheless, managed to expose a number of interesting facts concerning this
little known language and its relations to neighbouring languages.
The Esmeralde˜no language uses both prefixes and suffixes. Personal reference and
case are indicated by means of suffixes. The suffixes -s(a) and -va refer to first and
second person, respectively. They indicate a possessor with nouns, and either a subject
or an object with verbs. In some verb forms subject and object markers can occur in
combination. When -sa indicates nominal possession a preceding vowel is normally lost
(except with monosyllabic roots); the same may occur with long verb bases (e.g. bases
of more than two syllables). The suffix -sa can be stressed, but this does not seem to be
always the case. A suffix referring to a third-person possessor is -e or -´e (Seler 1902:
61; Jij´on y Caama˜no 1941: 450).
(255) mil-s´a
[mil-e ‘(someone’s) heart/stomach’]
heart/stomach-1P
‘my heart’, ‘my stomach’
(Jij´on y Caama˜no 1941: 539)
2.19 Esmeralde˜no 157
(256) mul-va
[mula ‘eye’, ‘face’]
eye-2P
‘your eye’
(Jij´on y Caama˜no 1941: 450)
(257) ene-s´a
eat-1S
‘I eat.’
(Jij´on y Caama˜no 1941: 509)
(258) peli-va
row-2S
‘You are rowing.’
(Jij´on y Caama˜no 1941: 524)
(259) pisko-v´a-s
sell-2S-1O
‘Sell it to me!’
(Jij´on y Caama˜no 1941: 524)
The form in (259) appears to have an imperative interpretation. Regular imperatives
are indicated with special endings, e.g. -ma for second person singular and -aja for first
person plural.
(260) kuli-ma
rise-2S.SG.IM
‘Stand up!’
(Jij´on y Caama˜no 1941: 509)
(261) naka atarai-ti-aja [Spanish atarraya ‘casting-net’]
let.us.go net-V-1P.PL.IM
‘Let us go and throw out the nets!’
(Jij´on y Caama˜no 1941: 502)
Jij´on y Caama˜no observes that many verb forms are apparently used ‘impersonally’
as they do not contain any indication of personal reference at all. He also notes the
absence of free personal pronouns. This may be due, of course, to the deficient nature of
the material. A very frequent suffix, which occurs both with verbs and with adjectives,
is -le. Its function may be to indicate a state or situation.
80
(262) akolinshe-le
be.indebted-ST
‘I am indebted.’
(Jij´on y Caama˜no 1941: 502)
(263) uba-le
die-ST
‘He/she died’, ‘dead’
(Jij´on y Caama˜no 1941: 505)
80
Instead of akulinshe-le (∼akolinshe-le) one also finds akulinshe-le-ne ‘I am indebted’. It is used
with both first- and with third-person subjects; Seler (1902: 61) suggests that -len´e indicated
present tense.
158 2 The Chibcha Sphere
(264) ´uvve kar´o-le
water red-ST
‘red-coloured water’
(Seler 1902: 55; Jij´on y Caama˜no 1941: 510)
Some adjectives obligatorily contain the suffix -le. This is the case, in particular, of
agroup of adjectives that have the canonical shape vi-/bi- ...-le,asinvi-k´o-le ‘dirty’,
vi-sh´u-le ‘cold’, vi-se-le ‘good’, bi-ga-le ‘bad’.
Negation is indicated by means of a prefix ba-; the frame bal-...-ka indicates negative
possession, as in (265):
(265) bal-di-ka
NE-hand-OW
‘one-handed’ (lit. ‘having no hand’,
‘with a hand missing’)
(Jij´on y Caama˜no 1941: 428)
Esmeralde˜no nouns frequently contain classifying prefixes which refer to shape. Par-
ticularly common are the prefixes mu- ‘indivisible bulky object’, ra- ‘protruding ele-
ment’, ta- ‘long object’, and vi(l)- ‘skin’, ‘wrapping’. These elements are either used
to reinforce basic semantic concepts, or to distinguish between related concepts. Words
referring to protruding body parts are normally preceded by ra-, e.g. ra-rapo ‘hair’,
ra-rap-s´a ‘my hair’, ra-ak-s´a (Seler: re-ac-s´a)‘my ear’, ra-an-s´a ‘my tongue’, ra-au-s´a
‘my nose’ (Seler: re-au-s´a). A contrastive use of the classifying prefixes can be observed
in the following pairs (Jij´on y Caama˜no 1941: 435–6).
(266) ra-tuna vil-tuna
protruding.element-mouth.area wrapping-mouth.area
81
‘beard’ ‘lips’
(267) di-sa ta-di-sa (Seler: ta-d´ı-ssa)
hand-1P long.object-hand-1P
‘my hand’ ‘my arm’
(268) ta-kel-s´amu-kil-sa
long.object-bone-1P bulky.object-bone-1P
‘my back’ ‘my bone’
Case relations are indicated in a heterogeneous way. Both case suffixes and preposi-
tional elements are used; in (269) and (270) the suffix -ra refers to location (‘in’), but tun
81
The attested word for ‘my mouth’ is vil-to-s´a (Seler 1902: 54).
2.19 Esmeralde˜no 159
is translated as ‘in front of’. The prepositional status of tun is related to the possessed–
possessor order noted above. (The case marker -ra is similar to the postvocalic allomorph
of directional -ta in Awa Pit, cf. section 2.17.)
(269) ama t´ushe qu´ıam-ra
there be house-L
‘He is there in the house’.
(Seler 1902: 61)
(270) tun kian-sa
front house-1P
‘in front of my house’
(Jij´on y Caama˜no 1941: 517)
Cases of noun incorporation have been attested:
(271) nuts-tate-aja
carry-pole-1S.PL.IM
‘Let us carry poles!’
(Jij´on y Caama˜no 1941: 481)
The genetic relations of the Esmeralde˜no language have been the object of some spec-
ulation. Seler (1902) suggested a relationship with the Yaruro linguistic isolate of the
Venezuelan lowlands, emphasising that the evidence was insufficient. This suggestion
was followed by Loukotka (1968), who brought both languages together in a Paleo–
Chibchan branch within his Chibchan family. In 1941 Jij´on y Caama˜no had already
compared Seler’s evidence and found no more than two good lexical look-alikes, uwi
‘water’ and taha ‘foot’ (Yaruro ta), as well as a number of vaguely resembling lexical
pairs. Although Jij´on y Caama˜no was seldom reluctant to accept highly controversial
proposals of cognation, he found the evidence too weak in this case. Instead, he com-
pared Esmeralde˜no words with their equivalents in Chibchan languages and a series of
Colombian and Ecuadorian languages which he thought related to Chibchan. A short
inspection of the proposed shared lexicon shows that the resemblances are convincing
only for so far as Cha
palaachi and Tsafiki, the two Barbacoan neighbours of Esmer-
alde˜no, are concerned. The similarities with these languages, particularly with Tsafiki,
are astonishing and indicate that intensive borrowing must have occurred. (A genetic
origin for these similarities is less likely because Esmeralde˜no is otherwise very different
from Barbacoan.) The shared lexicon mainly consists of plant and animal names, but it
also includes some basic lexicon (‘hand’, ‘firewood’). In (272) a number of shared or
possibly related items in Esmeralde˜no and Tsafiki are presented; the items in (273) are
found in the three languages.
160 2 The Chibcha Sphere
(272) Esmeralde˜no Tsafiki
amane am´an, amana ‘now’
bule bol´ı ‘gourd’
di t´ede ‘hand’
[cf. Tsafiki ned´e ‘foot’]
duka du ‘hill’, ‘mountain’
kinke k´ınki ‘tick’
matra m´atara ‘iguana’
nata n´a
h
ta ‘shad (a fish: Sp. s´abalo)’
para p´ara ‘wild boar’
pep-le pe
h
p´e ‘fan (for fire)’
vara b´aro ‘a bird of prey (Sp. gallinazo)’
ware w´alan ‘macaw’
(273) Esmeralde˜no Tsafiki Cha
palaachi
chula ˇciw´ıla ˇcil
y
a ‘pineapple’
d´ototu‘earth’
kuve kuw´a kuwa ‘cotton’
muripe mol´omulu ‘bean (Sp. frijol)’
ta-p´ake pa
h
k´ı pahki ‘bamboo species
(Sp. guadua)’
piama pin´ı pin
y
e ‘snake’
sheve s´abe sabe ‘rubber’
[cf. Spanish jebe]
walpa w´alpa wal
y
apa ‘chicken’ [Quechua wal
y
pa]
The word for ‘bamboo’ ta-p´ake is remarkable because it contains the classifying
prefix for long objects in addition to a shared root. If the Esmeralde˜no word for ‘tree’
t´a(k)te is analysed in the same way, its second element may be compared to te,which
means ‘firewood’ in both Tsafiki and Cha
palaachi. Conversely, the Esmeralde˜no word
for ‘firewood’ chite can be compared to Tsafiki ˇcid´e and Cha
palaachi ˇci ‘wood’, ‘tree’;
in addition, it contains an element te,which may be the same as the one found in
t´a(k)te. Apart from the inevitable wal
y
pa, Quechua loans were almost non-existent in
Esmeralde˜no.
82
Of particular interest is the word sheve ‘rubber’, which is found as jebe
in modern Andean Spanish. The evidence seems to indicate that this very common word
was borrowed from an Ecuadorian coastal language, be it Esmeralde˜no or not.
The past linguistic contacts of Esmeralde˜no deserve more attention than they have
received so far, particularly, in the light of processes of creolisation. Given the historical
background, the possibility of African influences should be taken into consideration.
82
Interestingly, all Quechua dialects north of Ca˜nar (as well as Awa Pit) have reflexes of atal
y
pa
(< *atawal
y
pa) for ‘chicken’. The fact that the languages of the Ecuadorian Pacific area have
reflexes of wal
y
pa suggests a different Quechua loan source.
2.20 The eastern Colombian lowlands 161
Finally, a few interesting lexical coincidences can still be mentioned, namely, the words
for ‘fire’ mu (cf. Culli mu) and ‘to die’ uba (cf. Kams´a-oba-na). Rivet (1942) mentions
a small number of lexical similarities with Yurumangu´ı.
2.20 Overview of the languages of the eastern Colombian lowlands
In the northwest of South America, as well as further south, the contrast between the
Andean cordilleras and highlands and the eastern lowlands with their tropical climate
is no less than dramatic. Roughly speaking, this is reflected in the distribution of ethnic
groups and languages as migrations tended to be confined to either one of the two regions.
In the eastern lowlands of Colombia, watered by the Amazon and Orinoco rivers and their
numerous tributaries, natural borders play a less important role than in the Andes, so that
long distance migrations were not unusual. As a result, many lowland languages belong
to families of considerable extension, and some individual languages are separated from
their closest relatives by a distance of thousands of kilometres. A striking example is
provided by a group of Cariban speakers which inhabited the department of Caquet´ain
southeastern Colombia (now in Amazonas and Guaviare). The closest living relatives
of their language, called Carijona (also Guaque, Hianacoto or Umaua, possibly clan
names; cf. Durbin and Seijas 1973c), are the languages of the Trio and the Akuriyo in the
border region of Brazil and Surinam (Meira 2000). Nevertheless, it would be a mistake to
view the division between the Andes and the eastern lowlands as impermeable. It is likely
that groups of Andean origin were forced into the pre-Andean foothills and lowlands on
several occasions, and, vice versa, the occupation of Andean river valleys, such as the
Magdalena valley, by Amazonian peoples has been attested beyond reasonable doubt.
In the preceding sections we have discussed several lowland languages which appear
to have an Andean origin, namely, Uw Cuwa (section 2.10), Kams´a (section 2.18), and
possibly Andaqu´ı (section 2.16). One small family and an isolate, which are located
near the Andes without substantial eastward extensions, are further candidates for such
a background. Additional research is needed in order to establish their possible genetic
connections. The Betoi family (cf. Constenla Uma˜na 1991; Zamponi 1996) consisted of
several closely related languages or dialects (Airico, Betoi, Ele, Jirara, Lolaca, Situfa,
etc.), located in the Colombian departments of Arauca and Casanare and in the Venezue-
lan state of Apure. There is a small group of Spanish-speaking survivors. The Cof´an
or A
i-ngae language is spoken by a strong ethnic group inhabiting a border area in the
western part of the Colombian department of Putumayo and the Ecuadorian province
of Sucumb´ıos (Borman 1976; Constenla Uma˜na 1991). The Cof´an number well over a
thousand, but their language still lacks a grammatical description.
A further linguistic isolate, located in the Sierra de la Macarena in the department of
Meta, is Tinigua (cf. Tobar Ortiz 2000). It is the last surviving language of a small family,
which further comprised Pamigua and Majigua. The language had long been considered
162 2 The Chibcha Sphere
lost until two aged speakers were discovered about 1990. According to tradition, the
Tinigua migrated into their present territory from the Yar´ıriver in the department of
Caquet´a, where they were neighbours of the Huitotoan peoples until the beginning of
the twentieth century.
The Guahiboan language family is mainly confined to the savannah-like area known
as the Colombian llanos with extensions into neighbouring Venezuela. With more than
20,000 people the Guahibo or Sikuani are the largest native group in eastern Colom-
bia. Their territory covers most of the department of Vichada and parts of the depart-
ments of Arauca, Casanare, Guain´ıa and Meta. Some 5,000 Sikuani live in Venezuela
(Ardila 2000). The Guahiboan family further includes three smaller languages: Cuiba,
Guayabero and Hitn¨u or Macaguane. Guayabero is the most divergent language with
respect to the other three; it is spoken along the Guaviare river in the departments of
Meta and Guaviare. Sikuani and Cuiba form a dialect continuum (Queixal´os 1993). The
Guahiboan family has been strongly influenced by neighbouring Arawakan languages.
However, a genetic relationship, as assumed by Loukotka (1968), is unlikely.
The Arawakan languages once had an important presence in eastern Colombia. The
Achagua people inhabited the Meta region in the Colombian llanos.Agrammar and
vocabulary of the Achagua language was published during the colonial period (Neira
and Ribero 1762). Today a group of a few hundred speakers survive along the Meta
river, mainly between Puerto L´opez and Puerto Gait´an (department of Meta). Achagua
is closely related to Piapoco (c. 4,000 speakers). The Piapoco live near the banks of
the Guaviare and Meta rivers and in the area between these two rivers (departments of
Meta and Vichada). The extinct Maipure language of Vichada, known from colonial
documents, for a time gave its name to (part of) the Arawakan family (Payne 1993).
More Arawakan languages are found in the Amazonian southeastern part of Colombia.
The most important one from a numerical point of view is Curripaco in the department
of Guain´ıa (7,000 speakers). In neighbouring Brazil this language is known as Baniva
do I¸cana.Arelated but distinct language is Baniva del Guain´ıa (G´omez-Imbert 2000;
Landaburu 2000b). It is closely related to the Yavitero language in Venezuela, which
has recently become extinct. Mosonyi (2000) notes specific lexical similarities between
Baniva–Yavitero and the Guajiro language (see section 2.12). The Cabiyar´ı and Tariana
languages in the department of Vaup´es (the latter mainly in Brazil) are surrounded by
Tucanoan languages and are subject to language shift. Yucuna is spoken in Amazonas
on the lower Caquet´a. Finally, the language of the Res´ıgaro,who were all but extermi-
nated during the rubber boom, was still spoken in the 1970s by individuals in a mixed
community of Bora and Huitoto in the Peruvian–Colombian border area (Allin 1976).
Two languages of the Saliban family, Piaroa and S´aliba, are spoken in the northern
llanos of Colombia. They are clearly, though not closely related. The Piaroa, who live
in a border area in the department of Vichada, are much more numerous in Venezuela
2.20 The eastern Colombian lowlands 163
(c. 10,000). The S´aliba colonised the valley of the Meta river during colonial times. Their
language acted as a general medium of communication in the Jesuit missions of the Meta,
Orinoco and Vichada valleys during the seventeenth century (Triana y Antorveza 1987:
172–3). A linguistic isolate of the Venezuelan llanos with occasional extensions into
Colombia is Yaruro or Pum´e.Two other isolates from that area, Guamo and Otomaco,
have long been extinct.
The language family known as Mak´u–Puinave is found in the Amazonian southeast
(Guain´ıa, Guaviare and Vaup´es). The Puinave language, which has several thousand
speakers (mostly in Guain´ıa), is mainly known from word lists (cf. Landaburu 1998).
The name Mak´u is applied to a number of small tribes, some of which continue to
live in isolation. The Nukak of Guaviare are such a group. The speakers of Kakua and
Hupda (or Yuhupde) can be found in the Vaup´es region, in an area mainly occupied by
Tucanoan peoples. The Kakua act as servants to several Tucanoan groups (Mel´endez
Lozano 2000d). Martins and Martins (1999) consider Kakua and Nukak to be variants of
the same language. The Hupda–Yuhupde language extends into adjacent areas of Brazil.
The Mak´uan languages are tonal. They have a system of contours of nasal consonants
which can be pre- and postoralised. This is reminiscent of the Ge languages in Brazil
(e.g. Kaingang) and the Harakmbut language in Peru, although there seem to be no
lexical correspondences.
The Tucanoan languages are spoken in a large discontinuous section of the Amazonian
region, extending from the foothills of the Andes well into Brazil. Traditionally, two
subfamilies are distinguished: Western Tucanoan and Eastern Tucanoan. Among the
Western Tucanoan languages Siona acquired the status of a lengua general during the
colonial period. It was used as a missionary language for the missions of the Putumayo
river (Triana and Antorveza 1987: 171–2). Today only a few hundred speakers remain,
mainly on the Putumayo, where it forms the border between Colombia, Ecuador and
Peru.Inaddition to Siona, Wheeler (2000) lists the languages Angutero (or Pioje),
Koreguaje–Tama, Makaguaje, Orej´on, Secoya (Ecuadorian and Peruvian varieties) and
Tetete as Western Tucanoan. All these languages are very closely related, and some
differ at the dialect rather than at language level. The Koreguaje and Tama together form
a somewhat divergent group and are difficult to separate linguistically. They occupy
areas on the Orteguaza and Cagu´an rivers adjacent to the Upper Magdalena highlands
(see section 2.16), where many of them were transferred as workers during the colonial
period. Koreguaje is considered to be a tonal language (Rodr´ıguez Gonz´alez 2000).
The Eastern Tucanoan languages occupy a major portion of the department of Vaup´es
and a small part of Amazonas. More groups are found in northwestern Brazil. A
unique feature of this area is that its population is almost entirely indigenous. Eastern
Tucanoan peoples are exogamic and marry outside their own ethnolinguistic group.
There has been extensive work on the Vaup´es region with respect to language contact and