Adelaar Willem, Muysken Pieter. The languages of the Andes
Подождите немного. Документ загружается.
564 6 The languages of Tierra del Fuego
(22) ˇsewla ˇswat’en ma:t’e-k’-ʔʔo
ˇ
Sewla S sash make-RD-DV
‘
ˇ
Sewla makes a sash.’
(Fern´andez Garay 1998c: 409)
In clause-second postion we often find a subject marker. In (22) it is preceded by the
lexical subject, but in (23) by a time adverb:
(23) maʔʔ ˇs e-t-ʔʔo:mk’e-ˇs-k’
now S 1S.SG-3O.SG-know-PD-RD
‘Now I know her.’
(Fern´andez Garay 1998c: 342)
In imperatives, the clitics occur after the verb.
Returning briefly to the issue of the relationship between Selk
nam, Tehuelche and
G¨un¨una Yajich within the Chonan family, the very brief typological comparison made
here is not conclusive. The phonological inventories of the languages are not dissimilar
(especially taking into account that Casamiquela may well be making more distinc-
tions than warranted by phonological oppositions), and the languages have complex
morphologies involving both prefixing and suffixing. The word orders of the three lan-
guages do not correspond, but this may be a matter of external syntactic influences, an
issue to which we return below.
6.4.2 Chono and Kawesqar
From the data available in Bausani (1975) the sound pattern of Chono can be re-
constructed roughly as follows. Syllables are often closed, but need not be. There
are only a few consonant clusters, but vowel clusters are frequent. From the Spanish
transcription, which may of course have been highly inadequate, we arrive at the
tentative sound inventory in table 6.7. The following sample forms, in the origi-
nal orthography and with Bausani’s tentative translation if available, illustrate these
patterns:
(24) quentaumet ‘exist’ zuquena ‘true’
jeyeulam ‘do’+‘good’ vla ‘for the sake of’
¨ue ˜nec ‘one’ jaguaitau
cot ‘son’ agic
sap ‘father’ eyuic
About Chono morphology, little can be said, except that it does not appear to be a
highly agglutinative language. Question particles, etc, appear to be enclitic. Sometimes
they are written as part of the preceding word, sometimes they appear separately. There
appears to be one reduplicated form: lam=lam ‘very good’.
6.4 Linguistic features 565
Table 6.7 Tentative sound inventory of Chono (based on the materials in Bausani 1975)
Labial Dental Alveolar Palatal Velar Glottal
Voiceless stops p t ˇck
Voiced stops b g
Fricative f z (?)
*
sxh
Nasal m n n
y
ŋ
Lateral l l
y
Glide w y
Front Central Back
High i u
Mid e o
a
Diphthongs: aw, ew, ow, ay, yu, wa, we, wi
*
For the symbol <z> various possible pronunciations are imaginable: [θ], as in Mapuche, [z], [t
s
]
or [s].
Chono, as far as it can be established from Bausani’s tentative deciphering of the
catechism manuscript, was OV and had postpositions or postnominal case markers:
(25) lam jeyeu lam toquieu?
good do good wish
‘beneficient and benevolent?’
(Bausani 1975: 108)
(26) acha-tau met Dios?
heaven-L be.there God
‘Is God in heaven?’
(Bausani 1975: 108)
Viegas Barros (to appear a) presents a highly informed listing of the lexicon and of
individual endings plausibly attributable to Chono, on the basis of the text in Bausani
(1975), place names and travellers’ accounts.
The sounds of Kawesqar are presented in table 6.8. However, in the Kawesqar alphabet
adopted in Aguilera (1984a, b) a number of sounds do not occur. These are put in brackets
in the array in table 6.8. Consonant clusters may occur in both syllable-initial and -final
position: awspena ‘tired’, yetapaks ‘weave’, and fte ‘fear’ illustrate this.
Clairis (1987) does not give an explicit description of Kawesqar morphology, but
from his description it appears there are (a) free forms; (b) suffixes; (c) enclitic
566 6 The languages of Tierra del Fuego
Table 6.8 Phonemes of Kawesqar (based on Clairis 1987 and Aguilera’s website
http://www.kawesqar.uchile.cl 2002)
*
Labial Dental Palatal Velar/Uvular Glottal
Plain stops p t ˇc ∼ c (k) q
Glottalised stops p’ t’ ˇc’ ∼ c’ k’ ∼ q’
Aspirated stops (p
h
)(t
h
)(k
h
∼ q
h
)
Fricatives f s x ∼ x
.
h
Nasals m n
Lateral l
Vibrant r ∼ rr
Glides w y
Front Central Back
High (i) (u)
Mid e o
Low (æ) a
*
Clairis treats the uvular and velar stops (k, q)asrealisations of a single phoneme, whereas Aguilera
considers this distinction to be contrastive. Aguilera treats the aspirated stops as allophones of
their non-aspirated counterparts. Clairis identifies a three-vowel system (a, e, o), whereas
Aguilera recognises six vowels.
elements; (d) optionally enclitic elements (e.g. person markers). In addition, there is
some reduplication (aswalaq=aswalaq ‘the day after tomorrow’), and there may be
compounding.
Aguilera (1997: 275) shows that the language contains a number of time and as-
pect suffixes. Time is organised on a continuous axis, marked by the following verbal
suffixes:
(27) -seku´e ‘future’ (seqwe in Clairis 1987
1
)
-yen´ak ‘present’
-pas ‘inmediate past’
-afq´at ‘recent past’
-h´oraras ‘remote past’
-h´oyok ‘narrative or mythical past’
Kawesqar, from the sentences presented in Clairis (1987), frequently has an SOV
order with a postverbal auxiliary (SOV Aux), although there is considerable word-order
1
Clairis (1987, part III) does not accept Aguilera’s distinction between velar k and uvular q and
uses q in both cases.
6.4 Linguistic features 567
freedom:
(28) manteqa qyexena
[cf. Spanish manteca]
butter want
‘I want butter.’
(Clairis 1987: 456)
(29) aswalaq ce qoteyo...yetas seqwe
tomorrow 1.SG again weave F
‘Tomorrow I am going to weave again.’
(Clairis 1987: 461)
(30) caw-nowaq cefalay-q
h
ar cefalay seqwe yemase
2-C drink-N drink F boat
‘Let us go and drink wine together in the boat.’
(Clairis 1987: 464)
In addition, it seems to be postpositional:
(31) qalpon alewe
[cf. Spanish galp´on]
dormitory inside
‘inside the dormitory’
(Clairis 1987: 459)
Head nouns occur at the end of the nominal complex:
(32) nawareno-s qwaseq
Navarino-G board
‘on board the Navarino’
(Clairis 1987: 459)
6.4.3 Yahgan
We will now turn to one particular Fuegian language, Yahgan, and try to describe it in
more detail, basing our descriptions on the main sources available for this language:
Adam (1884–5), Bridges ([1879] 1933, 1894), Gusinde (1926, 1937), Golbert de
Goodbar (1977, 1978, 1985), and Poblete and Salas (1999). Adam has reconstructed
aspects of the grammar of Yahgan from various printed sources, without any fieldwork
of his own. We referred to Bridges’s monumental work in the introduction to this chapter.
In addition to his dictionary, he wrote a brief but illuminating grammatical sketch of
the language. In his enormously detailed monograph on the Yahgan (1937), Gusinde
describes the life and culture of the group in nearly 1,500 pages, but relatively little
space is devoted to their language. In an earlier article (1926) he just describes the
phonological system of Yahgan and the other Fuegian languages. Golbert de Goodbar
worked in 1973 with the one surviving speaker of the language in Ushuaia, Argentina,
someone who had grown up and lived on a Protestant mission, and whose speech was
consequently influenced by English. Her articles deal with basic syntax and with verbal
and nominal morphology. Poblete and Salas (1999), finally, have worked with the two
surviving speakers on Navarino Island.
568 6 The languages of Tierra del Fuego
Table 6.9 Phoneme inventory of Yahgan (based on Golbert de Goodbar 1977 and
Poblete and Salas 1999)
*
Labial Dental Palatal Retroflex Velar Laryngeal
Obstruents p t ˇc(ˇc
.
)k(ʔ)
Fricatives f s ˇsx
Nasals m n
Lateral l
Vibrants r
Glides w y (ɹ)
Front Central Back
High i u
Mid e ə o
Low (æ) a (ɑ)
*
The three consonants that only occur in Poblete and Salas are placed between parentheses. Notice
that
ŋ
in some of Golbert de Goodbar’s examples is not represented in the phoneme inventory.
Her unspecified lateral
l has not been included either. It does not seem to coincide with the
voiceless lateral (l
)ofthe earlier sources Adam and Bridges.
For Golbert de Goodbar
ɑ
and
ə
are allophones of a single phoneme, the former occurring under
the accent, the latter in non-accentuated position. Poblete and Salas treat the vowels æ and
ə
as
phonemically distinct. Apparently, their fronted low vowel æ corresponds to the back low vowel
ɑ
of Golbert de Goodbar. All authors coincide in that the place of the accent in Yahgan is not
phonologically predictable.
The sound inventory of contemporary Yahgan is presented in table 6.9, which is
based on Golbert de Goodbar (1977) and Poblete and Salas (1999). The phonological
description of Golbert de Goodbar (1977) and Poblete and Salas (1999) is roughly the
same, except that the latter source mentions an additional distinctive vowel, as well as
a glottal stop and a retroflex series (
ˇc
.
,
ɹ
). By contrast, it lacks a special lateral sound
represented as
l,which was noted by Golbert de Goodbar.
2
In Adam (1885), however, a more complex inventory is given, particularly for the con-
sonants, as shown in table 6.10. In this table a series of preaspirated glides is postulated:
h
w,
h
y,
h
r,inaddition to dental fricatives and voiced stops. Adam’s description cor-
responds roughly to that of Bridges (1894), so it may well be that the language has
been simplified phonologically in the century intervening between the earlier studies
and Golbert de Goodbar’s work. The fact that Gusinde (1926) describes an intermediary
2
Golbert de Goodbar (1977) reports that it was not possible to determine the phonetic character-
istics of Yahgan
l, due to the fact that her consultant missed her teeth.
6.4 Linguistic features 569
Table 6.10 Phoneme inventory of Yahgan (based on Adam 1885)
*
Labial Interdental Alveolar Palatal Retroflex Velar Laryngeal
Voiceless obstruents p t ˇck
Voiced obstruents b d d
ˇz
g
Voiceless fricatives f θ sˇsxh
Voiced fricatives v δ z
Nasals m n ŋ
Pre-aspirated nasal
h
n
Lateral l
Voiceless lateral l
Vibrants r
Pre-aspirated vibrant
h
r
Glides w y ɹ
Pre-aspirated glides
h
w
h
y
Front Central Back
Short Long Short Long Short Long
High i i: u u:
Mid e e: ə oo:[ɔ:]
Low a a:
Diphthongs: aw, ow, ay, oy
*
The t
s
present in some of Bridges’s examples is not represented. Adam records two different
palatal sibilants (written sh and sch, respectively). The corresponding phonetic difference is
not known.
phonological system in which the voiced stops occur, but the dental fricatives and the
coarticulated velar fricatives do not, supports this hypothesis, which must remain tenta-
tive until a full-scale lexical analysis is undertaken.
In what remains we will follow Golbert de Goodbar’s analysis and transcription. The
opposition in length in the vowel system is allophonic, according to her, conditioned
by stress. She describes syllable structure as being quite simple. According to Golbert
de Goodbar onsets consist of at most two consonants, of which the second must be
a sonorant. However, the data presented show several clusters involving sounds other
than sonorants, and Bridges (1894: 54) shows sequences of complex consonants in his
data, as in tstwi:a: ‘paint brush’. Codas consist of one consonant at most. There may be
extra-long vowels in stressed positions. Bridges (1894: 54) claims that the Yahgan stress
system is rather irregular, but that in bisyllabic words penultimate stress is prevalent,
while in longer words (very frequent) ante-penultimate stress is prevalent. Poblete and
Salas (1999), however, claim that penultimate stress is prevalent throughout. There is a
570 6 The languages of Tierra del Fuego
complex set of morphophonemic alternations evident when words are combined. Bridges
(1894: 55) gives the following pairs of sounds:
(33) rˇs əəra ‘cry’ / kuˇsmu:ta ‘she is crying’
t,d
h
r ata ‘take’ / annu: gu
h
r ‘he has taken’
pf a:pu: ‘pull up by the roots’ / əəfga:mata ‘pull up the wrong one’
gx ta:gu: ‘give’ / təəxləəbata ‘to give several’
kx yi:ku: ‘scrape’ / kəənna yixgaye:te: ‘who is scraping?’
As in Kawesqar, some lexical elements belong to several grammatical categories,
but for the rest there is a clear opposition between verbs and nouns. Adjectives often
resemble nouns, and both classes may be marked with the predicate marker -a(ki):
(34) a. ur´atur l´ıf wɑɑl´ıˇc-´as´ɑɑxup´ay
3
[cf. English leaf; Spanish
some plant good-PD illness for valer ‘be worth’]
‘Some plants are good for illness.’ (Golbert de Goodbar 1977: 24)
b. par´ıkan-´oyna yɑɑˇs-´a
break-IM.2S hand-PD
‘Break it with your hands.’
(Golbert de Goodbar 1977: 7)
While generally predicate phrases are marked with verbal inflection, this predication
marker -a(ki) either marks secondary predication, as in (35), or appears in basic pred-
icative constructions on the predicate, as in (36):
(35) antɑɑp-´aki ha-wɑɑˇst´ak-oan at´ama
meat-PD 1.SG-make-F meal
‘I am going to prepare a meal with meat.’
(Golbert de Goodbar 1977: 7)
(36) h´ıpay wɑɑl´ıˇcyɑɑh´ɑɑ(ki)
1.D good boy.PD
‘We two are good boys.’
(Golbert de Goodbar 1977: 7)
Nominal compounds are right-headed:
(37) tala-glas
[cf. English glass]
eye-glass
‘spectacles’
(Guerra Eissmann 1995: 273)
Yahgan verbal morphology, as we will see, contains compounding, derivational pre-
fixes, proclitic pronouns and enclitic tense markers. Nominal morphology is mostly
3
In Golbert de Goodbar’s example sentences the symbol ɑ stands for a low back vowel when
stressed; elsewhere it is [ə].
6.4 Linguistic features 571
limited to a few case suffixes. One of the most striking features of Yahgan is its verbal
morphology. ‘Singular, dual, and plural verbs are a specialty in Yahgan, and simplify
its syntax. The dual verbs are an inflection of the singular, but the plural verbs are to a
very large extent totally different from their singular forms’, Bridges (1894: 67) writes.
What is remarkable about the suppletive pattern is that it seems ergative in nature in that
objects pattern with intransitive subjects. Consider the paradigm in (38):
(38) singular dual plural
ta:gu: ta:gu:pay yatu: ‘to give 1, 2, more things’
ata ata:pay tu:mi:na ‘to take 1, 2, more things’
atəəpi atəəpipay wa:gu:pi ‘to put 1, 2, more things on board’
---------------------------------------------
apəəna apəəna:pay ma:maya ‘1, 2, more die’
kəəna kəəna:pay a:lu: ‘1, 2, more are aboard, are on the
water’
ka:taka ka:taka:pay u:tuˇsu: ‘1, 2, more go on foot’
(Bridges 1894: 68)
With transitive verbs the number of the object determines the shape of the verb, with
intransitives the number of the subject.
Another feature of Yahgan verbal morphology worth mentioning, and which resembles
Selk
nam and many Algonquian and northwest-coast languages of North America, is
verb compounding and verb classification, which has led to an enormous lexicon in the
language. Bridges (1894) gives a number of examples; we will illustrate the process with
the derivatives of the verb kwissa ‘to pull’:
(39) kwisseta ‘to pull along, to draw on to some place’
kwissakaya ‘to pull up’
kwissa:mana:tsikari ‘to pull out’
kwissa:muˇci ‘to drag in’
kwissa:teka ‘to drag one thing to another, so as to be on it’
kwisso:ara:gu: ‘to drag ashore’
kwisso:anari ‘to pull into the water’
kwissa:pu:ku: ‘to pull into the fire’
kwisso:ana ‘to drag past’
kwissa:ku:ˇci ‘to pull into a boat any living object’
kwissa:təəpi ‘to hoist on board’
kwissu:aka:na ‘to pull down, as trees down a steep bank’
572 6 The languages of Tierra del Fuego
kwissu:ala ‘to pull out, undo, as a piece of knitting’
kwissayaˇsa / kwissa:tayaˇsa ‘to pull over and cove’
kwissu:ara:pu ‘to pull up, as a boat up a creek, by a line’
kwissu:ispe:ata ‘to pull awry, so as to be no longer straight’
kwissəmma / kwissəəmmata ‘to rend, tear, pull to pieces, as a garment’
kwissu:a:turi ‘to pull down, as, for instance, the higher
yards and masts’
kwissu:unna ‘to draw, as a horse does a cart’
‘and so on, ad libitum’
4
(Bridges 1894: 70)
Bruce Chatwin gives the following characterisation in his In Patagonia (1983: 128–9)
of the Yahgan derivations, to support his view of Man as the Essential Wanderer:
The Yahgan tongue – and by inference all language – proceeds as a system
of navigation. Named things are fixed points, aligned and compared, which
allow the speaker to plot the next move.
Bridges gives many examples of the way meanings are formed, through often complex
chains of metaphorisation. The root yi:nara ‘to gnaw’ yields yenuˇsyella ‘to continue
gnawing’, which in turns gives rise to ˇci:nuˇsyella ‘to leave unconsumed by gnawing, as
dogs the skeleton of an animal’. The latter form then comes to mean simply ‘skeleton’
(Bridges 1894: 70).
The verbs of motion in Yahgan can be prefixed and, in many discourse contexts, must
be prefixed with directional particles. These include the following:
(40) kəə-, ka:g- ‘upward’, ‘the upper end of the wigwam’, ‘westward’
ku:-, kwi- ‘westward’
ku:t-, ku:ta:- ‘southward’, ‘to go out, as on a bowsprit, or to the edge
of a cliff or branch to do anything’, ‘to go out from
shore, whatever the direction’
ku:p-, ku:pa- ‘downward’
ma:n-, mana- ‘outward’
ma-, ma:t- ‘northward’, ‘to go or come ashore’, ‘to remove from
off the fire to its confines, as the hob’
məət-, məəta- ‘eastward’, ‘inward’, ‘to get to do anything well from
custom’
(Bridges 1894: 71)
4
The forms taken from Bridges (1894) that are listed in (39) and elsewhere are based on a tentative
interpretation of his rather idiosyncratic orthography and have to be considered with reserve.
6.4 Linguistic features 573
Adam (1885: 26) notes that some of these prefixes may also be suffixed to noun phrases:
(41) əəkəə
h
r-ku
house-west
‘to the west of the house’
(Adam 1885: 26)
It is clear from Bridges’s description that these elements may also be used referentially:
(42) a. hawa-məə-ˇci
1.SG-east-side
‘the person to the east of me’
b. si-məə-ˇci
2.SG-east-side
‘the person to the east of you’
(Bridges 1894: 59)
From a syntactic perspective, the Yahgan inflection system can be characterised as
follows:
(43) ( )
Modals, such as k´ay ‘can’, k´uru ‘want’ and m´ay ‘habitual’, are never inflected for person
and tense, and their complement is infinitive:
(44) h´ıpay k´ay ´una p´okˇsu y´enɑɑ m´o
la
1.D can walk fast L day
‘We can walk fast during the day.’
(Golbert de Goodbar 1977: 40)
(45) h´ay k´uru m´uru ´anˇci w ´ɑɑrˇs
1.SG want climb that tree
‘I want to climb that tree.’
(Golbert de Goodbar 1977: 41)
(46) ´ɑɑkɑɑrk´ɑɑmi-a m´ay il´ana
house wood-PD CU build
‘Houses are generally built from wood.’
(Golbert de Goodbar 1977: 41)
The modal t´eynik(ɑ), marking external obligation, can either function as an impersonal
verb, taking the clause as its complement, as in (47), or it can occupy the customary
modal position, as in (48):
(47) l´ɑɑkɑɑxt´ɑɑrɑɑ t´eynik h´ıpay ´una ´utɑɑˇs
night during it.is.necessary 1.D walk slowly
‘At night we must walk slowly.’
(Golbert de Goodbar 1977: 55)