Adelaar Willem, Muysken Pieter. The languages of the Andes
Подождите немного. Документ загружается.
454 4 Languages of the eastern slopes
Table 4.7 Phoneme inventory of Huao (based on Saint and
K. L. Pike 1962)
∗
Labial Dental Palatal Velar
Voiceless stops p t (ˇc) k
Implosive click (p
<
)
Voiced stops b d ∼ ry∼ d
y
g
Nasals m n n
y
ŋ
Implosive nasal (m
<
)
Glide w
Oral Nasal
Front Central/back Front Central/back
High i ˜ı
Mid e o ˜e˜o
Low æ a ˜æ˜a
∗
The extra-systemic sounds ˇc, p
<
and m
<
are limited to ono-
matopoeic expressions and exclamations. All sounds are subject
to allophonic variation. The phonemes o and o˜ vary between
back and central and can take the character of a glide ([γ ], [γ˜])
in non-nuclear position.
4.9 Small families and supposed language isolates in Ecuador
In addition to these larger and smaller language families there are a number of families
with only a few members and language isolates. We will discuss these starting with the
remaining Ecuadorian languages.
In the Ecuadorian oriente several small tribes are found along the Colombian border.
They are the linguistically isolated Cof´an and western Tucanoan tribes already men-
tioned in chapter 2. In the Ecuadorian lowland province of Pastaza, another linguistic
isolate is found. It is Huao, the language of a tribe alternatively referred to as Auca,
Huaorani, Sabela or Auishiri. This tribe for long remained hostile to outside contacts.
Most linguistic work on this language has been done by the Summer Institute linguist
Catherine Peeke (1973, 1979), and Rachel Saint has collected a number of Huao texts
(cf. E. G. Pike and Saint 1988). The phoneme inventory has been described by Saint
and K. L. Pike (1962) as in table 4.7. There are three marginal phonemes: /ˇc/, used in
onomatopoetic descriptions; /p
<
/, a labial inverse oral click, used as emphatic negative;
and /m
<
/, inverse pulmonic nasal. There are no syllable-final consonants and no conso-
nant clusters, while there are numerous vowel clusters. Nasalisation plays an important
role in the language.
4.9 Small families and supposed isolates in Ecuador 455
There is fairly extensive verb suffixing, as in:
(115) apæ˜ -ne-ga-d˜an(i)-˜ı-pa
speak-CL:mouth-RM-3S.PL-IL-AF
‘They spoke long ago.’
(Pike and Saint 1988: 119)
There is evidence of noun classification; the examples given all involve body parts, as
-ne- ‘mouth’ above, -miœ˜-‘tail’ and -po- ‘finger’:
(116) bo-t˜o ta:-miæ˜ -kæ-bo-˜ı-pa
1.SG-PU cut-CL:tail-IC-1S.SG-IL-AF
‘I am going to cut off his tail.’
(Pike and Saint 1988: 125)
(117) deye ˜ı-ma˜ıw˜e˜e-n˜e˜e-po-k˜a-ta-pa
spider.monkey this-like long.ago-L exist-CL:finger-3S.SG-PA-AF
‘Long ago the spider monkey had four fingers and a thumb.’
(Pike and Saint 1988: 129)
These forms also illustrate a number of other features of the language: the contrast
between past and remote-past tenses, the presence of aspect categories such as inceptive,
and of evidentials, including an inferential marker, the use of person markers, and case
suffixes.
The suffixal person markers differentiate for number (singular, dual and plural), gender
(animate, inanimate) and (feminine) honorific. The second- and third-person dual and
plural forms do not only refer to number, but are also used to denote specific relations
in the kinship system (Peeke 1973: 40). Note that all stop symbols in Peeke’s inventory
of person markers reproduced below have to be pronounced as nasals when preceded
by a nasal vowel. So -d˜adi is either pronounced [d˜ani], or [n˜ani]; -k˜a is either [k˜a] or
[ŋ˜a], etc.
(118) singular dual plural
1 -bo -b¯oda -b˜odi (exclusive)
-b˜o (inclusive)
2 -bi -b˜ıda -b˜ıdi
3 animate -k˜a -da -d˜adi
3 inanimate ∅∅
The feminine honorific forms are designed to address mothers and women of an equiv-
alent rank in relation to the speaker. They distinguish between second and third person,
which are -b˜ı and -d˜a, respectively.
456 4 Languages of the eastern slopes
The case markers include:
(119) -ba˜ı ‘like’
-d˜e ‘in’
-d˜o ‘toward another, to’
-ke ‘limit, until’
The case markers in (119) that have voiced initials are subject to nasalisation as
described for the person markers above.
There is no marking for subjects and objects, which are differentiated through the
(fairly strict) SOV constituent order. An indication of the complexities of the interaction
of person marking and subordination is given in (120). (Note that -d˜ani refers to second-
person plural in permissive complement clauses such as illustrated in the following
example.)
(120) m˜a-n˜ani a-te “p˜on˜o-m˜ıni kæ-m˜o-e-d˜ani” ˜a-n
y
˜o-ŋŋ˜a-te, ba ba ˜a-n˜ani
have-3S.PL see-SU give-2S.PL eat-1O.PL.EX-PM-2S.PL say-PR-3S-SU
no no say-3S.PL
{When, seeing they had, the (others) said: you give (you permit) us to
eat, they said no no.}
‘Since they had lots of corn, when the others asked: “Can we have some
of your corn to eat”, they refused.’
(Pike and Saint 1988: 115)
The subordination marker -te generally marks identical subjects, but not always.
4.10 Small families and supposed language isolates in Peru
In the northern half of the Peruvian lowlands, the linguistic situation is characterised
by the presence of several language isolates and small families. On the eastern slopes
of the Andes two families are on the verge of extinction, Chol´on (discussed in detail
below in section 4.11) and Muniche. Muniche is spoken in the village of Muniches on
the Paranapura river (department of Loreto) (Gibson 1996). The unclassified and nearly
extinct Taushiro language of the Tigre river (department of Loreto) has been mentioned
in connection with Zaparoan (cf. Fabre 2001: 1007). Kaufman (1994) gives evidence for
a possible genetic relation between Taushiro, the extinct Mayna or Omurano language
(once spoken on the Urituyacu river in Loreto) and Candoshi (see below). Another
unclassified and possibly extinct language of the Peruvian lowlands is Au(i)shiri or
Tekiraka, once spoken near lake Vacacocha in the Curaray river area (Tessmann 1930).
In spite of the similarity in the name, the vocabulary given by Tessmann indicates no
relationship with Auishiri, one of the alternative designations of the Huaorani people
(see above section 4.9).
4.10 Small families and supposed isolates in Peru 457
Further north, Candoshi, Shapra or Murato is the last member of a small family; other
members were contacted during the sixteenth century (cf. chapter 3: sections 3.9.1 and
3.9.2). It has often been classified with the neighbouring Jivaroan languages, but David
Payne (1990) argues that this is a misclassification due to borrowing. Shapra also refers
to a variety of Candoshi spoken by a rather isolated subgroup. The Murato were first
contacted in 1744, but resisted missionary efforts of the Jesuits in 1748 (Chirif and Mora
1977).
Further east in Loreto, Urarina or Simacu is a language isolate with a few thousand
speakers. The group is now also known by its own denomination Kach´a. Based on the
sentences given in ILV (1979: 139–42), we can draw the following preliminary sketch
of the language. In noun phrases, adjectives (actually stative verbs) follow nouns:
(121) aka helaher˜ı: ‘cold water’
aka aharot˜ı: ‘hot water’
Possessors and demonstratives precede the noun, however:
(122) a. i: kra:
‘your name’
i: lureri
‘your house’
b. ka: nenaha kra:
this settlement name
‘the name of this settlement’
c. ka: nkekra:
this river name
‘the name of this river’
We have the order modifier–modified:
(123) atawari tahe
‘chicken egg’
The object precedes the verb:
(124) a. kan ama- Atalaya-
1.SG take-IM Atalaya-AL
‘Take me to Atalaya!’
b. karay abio:n sar--ˇce
for.me airplane push-IM-PL
‘Please push the airplane!’
458 4 Languages of the eastern slopes
Information status is marked with clause-final particles, such as the question clitic -na
in (125a):
(125) a. kna-a-na
hurt-3S-Q
‘Does it hurt?’
b. kna-to-a
hurt-VE-3S
‘It hurts.’
c. kna-i
hurt-3S.NE
‘It doesn’t hurt.’
Question words are fronted. They include:
(126) d
z
a ‘what?’, ‘who?’
d
z
a-bana ‘when?’
d
z
a-el˜o ‘in which direction?’
d
z
a-hia ‘along where?’
d
z
a-toan-˜ey ‘how?’ [tona ‘compare’, -˜ey ‘gerund’]
d
z
‘where?’
There is evidence of a demonstrative paradigm intersecting with the question-word
system (-hia indicates location in a wide sense).
(127) d
z
a-hia ‘whereabout?’
ta-hia ‘over there’
ka-hia ‘here’
It is difficult to gain a good picture of the morphology, except that it is suffixal. There
are a great many suffixes and about twelve suffix positions.
Cajas Rojas (1990) deals with the complex interaction between (progressive) nasali-
sation and nasal weakening in Urarina. The language is currently being studied by Knut
Olawsky.
15
The Zaparoan family, mentioned above, and the Peba–Yagua family fill part of the
space separating the Amazon river from the Colombian and Ecuadorian border. Peba–
Yagua is now represented by a single language, Yagua. Yameo, another language of
the same family, became extinct in the 1960s. Doris Payne (1986) and Payne and Payne
(1990) contain extensive data on Yagua. Near the place where Peru meets with Colombia
and Brazil the isolate Ticuna is spoken (cf. section 2.20).
15
We thank Knut Olawsky for his revision of the Urarina examples given in this section.
4.10 Small families and supposed isolates in Peru 459
There are two languages in the southern Peruvian lowlands which have been mis-
classified as Arawakan: Culina and Harakmbut. The Culina live on both sides of the
Brazilian border. There is a larger Culina population in Brazil. Culina belongs to the
Araw´afamily; the Culina constituted a warrior group and were first sighted in 1869 by
the English explorer Chandless; in the 1890s they suffered intensive exploitation and
by the end of the 1940s they established themselves in the Alto Pur´us (Chirif and Mora
1977).
Harakmbut (spoken in Madre de Dios) has long been considered a language isolate
known by several names. Adelaar (2000), however, has argued a link with the Katukina
language family in Brazil. There are several dialects which fall into two large clus-
ters (Helberg Ch´avez 1990: 227–8). Toyoeri and Huachipaeri form one cluster, while
the other is formed by Sapiteri, Arasaeri and Amarakaeri, which is the best known
and has the largest number of speakers. It has been studied by Helberg Ch´avez (1984,
1990). After an initial incursion by the Inca Tupac Yupanqui, a new attempt to conquer
Madre de Dios was made by the Spaniards in 1566. The Huachipaeri managed to attack
highland haciendas from time to time in the colonial period. Towards the end of the nine-
teenth century the Toyoeri were decimated by the infamous rubber-baron Fitzcarraldo,
an event which resulted in an internecine war between the survivors and other Harakmbut
subtribes (Gray 1996). In 1950 there was successful renewed contact with Dominican
missionaries.
Helberg Ch´avez (1990) provides detailed information about the highly complex struc-
ture of the Amarakaeri verb. Verbs can have both prefixes and suffixes, and a number of se-
mantic categories can be marked on the verb: transitivity, causativity, intensive/extensive,
and person, tense/mood/aspect, and polarity. The personal markers, pronominal and pre-
fixal, are (Helberg Ch´avez 1990: 243–4):
(128) 1.SG
n
doʔʔ i
h
-
2.SG ˜on iʔʔ-
3.SG k˜en oʔʔ-
1.PL oro oʔʔ-
2.PL opd
nm
bo-/m˜o-
3.PL k˜en-˜o-my ˜onʔʔ-
Twoexamples can serve to illustrate the complexities of Harakmbut verbal morphology,
which includes a complex system of classifiers marking shape:
(129) i
h
-waʔʔ-pet-
n
da-piʔʔ ˜ı
h
-n˜o-p˜o-˜e-˜ı
1S.SG-go-PO-IT-DU 1.SG-CL:centre-CL:round.form-know-1.SG
‘I think I will go a long time from now.’
(Helberg Ch´avez 1990: 239)
460 4 Languages of the eastern slopes
The classifier system has been described in some detail by Hart (1963): morphemes
referring to body parts are often combined with morphemes denoting more abstract
spatial features. These can occur in nouns, but also in adjectives and verbs. An example
of a nominal complex is (130), containing a nominalising classifier prefix:
(130) wa-pa-pi-k-ti-pi
N-CL:rod-CL:stick-CL:head-CL:extension-CL:stick
‘shin’
(Hart 1963: 3)
The data in ILV (1973) allow us to draw some preliminary typological conclusions as
well. Compounds have the order modifier–modified, and the same holds for adjective–
noun and determiner–noun order:
(131) a. baka ˜onʔʔw˜eh cow milk ‘cows’ milk’
b. karg
ŋŋ
w˜eʔʔ˜ey cold water ‘cold water’
c. ˜ın wa-km˜aiʔʔ this N-CL:pill ‘this pill’
Case is expressed through suffixes:
(132) ˜ın-te this-L ‘here’
iʔʔ-m˜aiʔʔ-po 2S.SG-drink-IS ‘while you drink...’
4.11 Chol´on
We now will provide a sketch of Chol´on, a language of the upper Huallaga valley in
northern Peru, north of the town of Tingo Mar´ıa. There was also a language closely
related to Chol´on, Hibito. Only a few elderly people remembered Chol´on until recently
(reportedly in Si´on, department of San Mart´ın). Both Chol´on and Hibito are now probably
extinct. Chol´on is discussed in some detail below. This language has been documented
in a number of sources:
– Pedro de la Mata’s (1748) Arte de la lengua cholona. The manuscript was
purchased by the British Library in 1863 and has been – rather unfaithfully
and only partly – transcribed by Julio Tello (1923: 690–750). We went back
to a filmed copy of the British Museum text, which has been prepared for
publication by Astrid Alexander-Bakkerus.
– Mart´ınez Compa˜n´on’s (1985 [1782–90]) comparative word list of 43 items
gathered from speakers in the same mission areas (conversiones), partic-
ularly the conversi´on de Hivitos. The list includes several other languages
besides Chol´on and ‘Hivito’ (see section 3.9.2).
– G¨unter Tessmann’s (1930) 30-item word list from his large comparative
study of northeastern Peru, based on data provided by Harvey Baessler
from an old man living in Pachiza.
4.11 Chol´on 461
4.11.1 The Chol´on lexicon and relationship with Hibito
There is some dispute about the genetic relationship between Chol´on and Hibito. Most
scholars, and most convincingly Rivet (1949), have argued that the two languages are
closely related, but Torero (1986: 533) casts doubt on this claim. In our view, the lex-
ical correspondences strongly support the relationship. Pedro de la Mata, at the end of
his Chol´on grammar, gives a list of both the Chol´on- and the Hibito-speaking towns,
presumably intending his grammar to be used in both. Consider the following items,
taken from Pedro de la Mata (PM), Mart´ınez Compa˜n´on (MC) and Tessmann (T), which
surely belong to the core vocabulary of both languages. This makes Torero’s suggestion
that shared lexicon has resulted from extensive language contact less plausible:
(133) Chol´on Hibito
tree, stick mech (PM), mees= ˜ngup (MC), mixs (MC),
me
¯
ˇs (T) mitˇs (T)
water cot (PM), qu˜ot (MC), k¨o
˘
ta (T) cachi (MC), otˇsj (T)
daughter ˜nu (PM), -˜nu (MC) ˜noo (MC)
son pul (PM), -pul (MC) pool (MC)
fruit queniya (MC) llagna (MC)
woman ila (PM), yla (MC), hil´a (T) etlec (MC), ud ¯u,
alu (T)
man nun (PM), num (MC), l¯un(o) (T) nuum (MC),
n¯um, nun (T)
father nguch ‘his/her father’ (PM) cotc (MC)
mother nguetz, ˜neetz, n˜guech queec (MC)
‘his/her mother’ (PM)
die ˜ngoli=cho (MC), calgesquim (MC)
col- (PM)
black zal, tzal, chal (PM) uts´almana (T)
bone chel (PM, MC) chepce (MC)
one an- (PM), ans- (T) ets´ı (T)
two ip- (PM), hips- (T) optˇs¯e (T)
three is- (PM), hies- (T) ´utsi (T)
house yip, (t)zip (PM), h¯ıp(o) (T) ¯ıp (T)
earth pei, pey (PM), lluspey (MC), puts
16
(T)
p¯eij (T)
head setch (PM), mutsitˇs´e (T) s´otˇsa (T)
stone ta (PM), t¯a (T) tˇs¯e (T)
16
Compare Culli pus ∼ pos ‘earth’.
462 4 Languages of the eastern slopes
Sound correspondences may include:
(134) Chol´on Hibito
ta ˇce, ˇci
c, ˇcc
oa
u o:, u:
There are also a number of items which do not correspond, but part of the non-
correspondence may simply be due to the elicitation procedure. Of Hibito, no
more than fifty words are known: those recorded by Mart´ınez Compa˜n´on and by
Tessmann.
Avery interesting case involves the Chol´on words for ‘father’ and ‘mother’, which
are regularly inflected with person prefixes except for the suppletive third-person form,
as Pedro de la Mata notes:
(135) ‘father’ ‘mother’
a-pa a-paŋŋ 1P.SG
mi-pa mi-paŋŋ 2P.SG.MS
pi-pa pi-paŋŋ 2P.SG.FE
ŋŋuˇc ŋŋeˇc 3P.SG
The last two forms, however, can be regularly derived from *
ŋ
-kuˇc and *
ŋ
-keˇc,which
presumably contain the older forms for ‘father’ and ‘mother’, respectively, if we take
the Hibito evidence into account.
4.11.2 Gender-determined language use
Certain forms of Chol´on language use are determined by the gender of the (singular)
addressee. Consider first the contrast between the following examples:
(136) a. inˇcam-ma
what-MS
‘What do you say or want (man)?’
b. inˇcam-pa
what-FE
‘What do you say or want (woman)?’
The same contrast with m for masculine forms and p for feminine forms we find with
second-person-singular prefixes:
4.11 Chol´on 463
(137) a- 1P.SG a-kcok ‘my box’
[kacok ‘box’]
mi- 2P.SG.MS mi-kcok ‘your (masc.) box’
pi- 2P.SG.FE pi-kcok ‘your (fem.) box’
We find p for feminine also in the contrasting address forms: there is an -ey ending for
men, and -pey for women.
4.11.3 Chol´on phonology
On the basis of the historical sources it is not easy to get a clear idea of Chol´on phonology.
Pedro de la Mata developed elaborate transcription conventions in his Arte,but does
not indicate with which sounds his diacritics correspond. Tessmann, however, uses a
much more uniform and precise system, so that we have a reasonable general idea. The
sound inventory is roughly as in table 4.8. Vowel length is not distinctive, according
to Pedro de la Mata, and there is no contrastive stress. Stress is word final, except on
certain verb tenses, where we have stress on the penultimate syllable. For the notation
of Chol´on we use a spelling based on a reconstruction of the Chol´on sound system
developed by Alexander-Bakkerus (in preparation).
17
Doubled consonants, which are
usually presented as alternatives for simple consonants in de la Mata’s grammar, are
written as single consonants.
There is evidence of harmony for high vowels in the prefixes, as can be seen in the
following contrasts:
(138) ki-coc ‘our guinea pig’ ku-cuˇc ‘our alfalfa’
[yoc ‘guinea-pig’,
i-coc ‘their guinea pig’ u-cuˇc ‘their alfalfa’ yuˇc ‘alfalfa’]
Syllables tend to be simple, as will become clear from the examples below.
4.11.4 The principal grammatical features of Chol´on
As far as the morphology of this language is concerned, a large part of Chol´on mor-
phology is clearly suffixal and agglutinative, but person markers are prefixal. There is
some compounding as well. There is a complex set of morphophonological adjustment
rules for the third-person prefix, as will be shown below.
Chol´on has a number of word classes, in addition to nouns and verbs, which are the
two dominant categories in the language and to which we will return below.
17
Much of the morphological analysis underlying the interpretation of the example sentences in
this section is also based on Alexander-Bakkerus’s unpublished work.