Adelaar Willem, Muysken Pieter. The languages of the Andes
Подождите немного. Документ загружается.
284 3 The Inca Sphere
Table 3.9 Subject and subject–object endings for the
future tense in La Paz and Sitajara Aymara (Yapita
1991; Briggs 1993: 198)
La Paz Sitajara
1 pers. subject (3 pers. object) -: -
ŋ
a, -:
2 pers. subject (3 pers. object) -:ta -
ŋ
a:ta
3 pers. subject (3 pers. object) -ni -ni
4 pers. subject (3 pers. object) -n
y
ani -n
y
ani
1 pers. subject + 2 pers. object -:ma -ma:(ma)
2 pers. subject + 1 pers. object -ita:ta -ita
ŋ
a:ta
3 pers. subject + 1 pers. object -itani -itani
3 pers. subject + 2 pers. object -:tam -
ŋ
ata:ma
3 pers. subject + 4 pers. object -istani -stani
The endings of the future tense are quite different from those of the unmarked and
remote past tenses. In table 3.9 the endings of the future-tense paradigm are represented in
two dialect forms: La Paz Aymara and the conservative Sitajara dialect of Tacna, Peru.
All instances of vowel length in La Paz Aymara (length is represented segmentally)
appear to reflect sequences that once contained a nasal, still present in Sitajara.
The future-tense paradigm given by Bertonio (1603b) is similar to that of Sitajara
with the difference that the velar nasal
ŋ
is replaced by j [h]. Instead of -n
y
ani, Bertonio
has -tana,aform still found in several of the southern and western Aymara dialects. It
should be observed that the first-person future subject ending is etymologically identical
to the first-person possessive ending, whereas the second-person possessive ending is
etymologically identical to the second-person future object ending. The 1O and 4O
endings can easily be segmented into elements -ita- and -(i)sta-, respectively, with the
corresponding subject endings following them.
As in Quechua, the Aymara imperative paradigm comprises a full set of endings en-
coding all the options involving a second- or a third-person subject. Besides, future-tense
forms can be used in a hortative sense, thus compensating for the absence of specific first-
and fourth-person-subject endings. The imperative markers for second- and third-person
subject, variably vowel-retaining -m(a) and vowel suppressing -p
(h)
a(na), are subject to
some dialectal variation but are clearly reminiscent of the corresponding possessive
markers. The combination 2S.1O has a special ending, -ita; the 3S.2O combination is
indicated by a special ending -:tpa(n) or by the corresponding future-tense form (Briggs
1993: 200–4). The La Paz Aymara paradigm for second- and third-person-subject forms
of ˇcura-isrepresented in table 3.10.
Again like Quechua (cf. section 3.2.6), Aymara has a potential mood, which is divided
into two tenses: present potential and past potential. These are referred to in Hardman
3.3 The Aymaran language family 285
Table 3.10 Aymara subject and subject–object paradigm for the
imperative mood (Yapita 1991)
2 pers. subject (3 pers. object) ˇcura-m ‘Give it to him/her!’
3 pers. subject (3 pers. object) ˇcur-pa ‘Let him/her give it to him/her!’
2 pers. subject + 1 pers. object ˇcur-ita ‘Give it to me!’
3 pers. subject + 1 pers. object ˇcur-itpa ‘Let him/her give it to me!’
3 pers. subject + 2 pers. object ˇcura-:tpa(n) ‘Let him/her give it to you!’
3 pers. subject + 4 pers. object ˇcur-istpa ‘Let him/her give it to us (incl.)!’
et al. (1988) and in Briggs (1993) as the ‘desiderative tense’ (tiempo desiderativo) and
the ‘reproacher tense’ (tiempo reprochador), respectively. The formation of the present
potential is quite complicated. It appears to be the result of a merger of two earlier
paradigms, which are still partly kept apart in Bertonio (cf. Briggs 1993: 205–10). One
of these paradigms was characterised by the presence of an element -irik-or-irih-, the
other one by the element -s(a)-. Some dialects have -irik-s-or-iri-s- (La Paz) rather
than -irik-/-irih-. The personal endings of the -irik- paradigm coincide with those of the
unmarked tense. This can be seen in Bertonio and in some of the modern dialects (e.g.
Sitajara), where the -irik- paradigm has been preserved almost completely.
The -s(a)- element is followed by specific personal endings: -ma for second-person
subject, -p
(h)
a(na) for third-person subject (cf. the imperative paradigm), and -na for
fourth-person subject. A first-person-subject form in -a (ending -sa) has been recorded
in Ebbing (1965: 124) and Cerr´on-Palomino (1994a: 113), but most modern dialects
now seem to prefer a form from the -irik- paradigm (-irista in La Paz Aymara). The
endings which encode a first-person or a fourth-person object are formed by inserting
the elements -ita- and -ista- before the -s(a)- affix. The second-person object is expressed
by -irik- forms.
The behaviour of -s(a)- in relation to the preceding vowel is that of a vowel-preserving
affix, except in the sequence -sna (4S). The suffixes -sma (2S) and -spa (3S) are vowel-
preserving even when following the object markers -ita- and -ista-. The fact that -sma
(2S) does not suppress a preceding vowel helps to avoid confusion with the 1S.2O ending
of the unmarked tense (ˇcursma ‘I give it to you’, ˇcurasma ‘You could give it to him/her’).
The paradigm of the past potential (or ‘reproacher tense’) is derived from the present
potential paradigm. In all forms except first-person subject, the suffix -(a):na of the
near-remote past tense is added to the corresponding present potential form.
75
At the
same time, the vowel of the suffix -s(a)- is recovered. In the 4S form one element -sa-is
75
Cerr´on-Palomino (1995a: 145) points at the similarity between the presence of the suffix -(a:)na,
restricted to third-person subject in most of present-day Aymara, and the use of the third-person
subject auxiliary verb kar(q)a in the past potential of Quechua.
286 3 The Inca Sphere
Table 3.11 Subject and subject–object endings for the present and
past potential mood in La Paz Aymara (Yapita 1991)
Present potential Past potential
1 pers. subject (3 pers. object) -irista -iriska:ta
2 pers. subject (3 pers. object) -sma -sama:na
3 pers. subject (3 pers. object) -spa -sapa:na
4 pers. subject (3 pers. object) -sna -sa:na
1 pers. subject + 2 pers. object -irisma -iriskasama:na
2 pers. subject + 1 pers. object -itasma -itasama:na
3 pers. subject + 1 pers. object -itaspa -itasapa:na
3 pers. subject + 2 pers. object -iristam -iriskatama:na
3 pers. subject + 4 pers. object -(i)staspa -(i)stasapa:na
suppressed. In many dialects (including La Paz), the sequence -irik-or-iri-s- must be
followed by an element -ka-inthe past potential.
The present and past potential endings for La Paz Aymara are listed in table 3.11.
Some examples are:
(164) k
h
a: anu-w hala-ni, aˇc-ha-rak-sta-s-pa
{k
h
a(y)a anu-w(a) hala-ni-Ø, aˇc(u)-ha-rak(i)-(i)sta-s-pa}
that dog-AF run-H-3S, hold.with.teeth-LS-AN-4O-PO-3S
‘That dog comes running, it may bite us!’
(Yapita 1991: 84)
(165) ma: iki-n
y
apa-si-ni-ki-sa-m-a:na
{ma(y)a iki-n
y
(a)-Ø apa-si-ni-ki-sa-m(a)-a:na}
one sleep-FN-Z carry-RF-H-DL-PO-2S-PA
‘You just could have brought yourself a bed.’
(Yapita 1991: 52)
Two further distinctions that are expressed in the Aymara verbal system belong to the
domain of evidentiality. A suffix -paˇca-, called ‘inferential’ (inferencial)inHardman
et al. (1988), is used for referring to events of which knowledge is obtained by de-
duction, rather than by observation; a suffix -ˇci-, referred to as the ‘non-involver’
(no-involucrador), indicates conjecture.
76
Both suffixes are vowel suppressing; -paˇca-
can be inserted before the endings of the unmarked, the future and the near-remote tenses;
-ˇci- can be combined with any tense of the indicative and optative moods (Hardman et al.
1988: 158, 165). After the suffixes -paˇca- and -ˇci-, the 3S ending of the unmarked tense
consists in a zero marker. All other endings are those of the corresponding tenses and
moods, allowing for occasional morphophonemic adjustments. The suffix -paˇca- retains
76
Cerr´on-Palomino (1994a: 114–15) points at the fact that this suffix could be historically related
to the conjectural independent suffix -ˇc
.
i in Central Peruvian Quechua.
3.3 The Aymaran language family 287
its final vowel except before i;-ˇci-isoptionally reduced to -s- before a vowel-suppressing
consonant, while it becomes -ˇc- before i.
77
(166) ˇcur-paˇca-wa
{ˇcur(a)-paˇca-Ø-wa}
give-IP-3S-AF
‘He/she must have given it.’
(Hardman et al. 1988: 149)
(167) ina-s sar-ˇci
{ina-s(a) sar(a)-ˇci-Ø}
maybe-AD go-DU-3S
‘Maybe he went.’
(Hardman et al. 1988: 150)
(168) ina-s k
h
iti-r ˇcur-s-ta
{ina-s(a) k
h
iti-r(u) ˇcur(a)-ˇc(i)-ta}
maybe-AD who-AL give-DU-1S
‘Maybe I gave it to someone.’
(Yapita 1991: 35)
(169) k
h
iti-ru-w ˇcur-paˇca:-ta
{k
h
iti-ru-w(a) ˇcur(a)-paˇca:-ta}
who-AL-AF give-IP-2S
‘I must have given it to someone.’
(Yapita 1991: 35)
Verbal subordination in adverbial clauses can be indicated by the vowel-preserving
suffixes -sa and -sina. The descriptions vary as to the semantic distinction (if any)
between these two suffixes, which would have to do with evidentiality (see, for instance,
Briggs 1993: 286). The switch-reference dimension does not play a role, except in those
dialects which preserve the subordinator -ipana,which indicates that the subjects of the
subordinate and the superordinate verb are different. Originally, -ipana was the third-
person form of a fuller paradigm including four subject distinctions (1S, 2S, 3S, 4S).
According to Briggs (1993: 288), this paradigm, which is also mentioned in Bertonio,
is in full use only in the Aymara dialect of northern Potos´ı (Morocomarca), where it has
the following shape: 1S -in
y
ana,2S-imana,3S-ipana,4S-isana. The varying central
elements (-n
y
a-,-ma-, -pa-, -sa-) coincide with the possessive personal endings of the
nouns. The full subject paradigm of the switch-reference marker is also preserved in the
related Jaqaru language, which may indicate that Proto-Aymaran had an elaborate switch-
reference system. In contrast to Quechua, Aymara subordinate verbs do not encode
person of object.
77
Hardman et al. (1988) and Yapita (1991) often do not coincide in the notation of vowel length in
the Aymara verbal paradigms. When referring to -paˇca-,Yapita always indicates length on the
final vowel a,except before -ya:- of the near-remote past tense, e.g. 3S future tense -paˇca:ni in
Yapita (1991: 34), but -paˇcani in Hardman et al. (1988: 165).
288 3 The Inca Sphere
Table 3.12 Nominalising affixes in Aymara
-iri ‘agentive’
-n
y
a ‘infinitive’, ‘non-realised event’
-ta ‘stative’, ‘realised event’
-(:)wi ‘place or time’, ‘realised event’
The role of nominalisation in Aymara is similar to that in Quechua. The principal
nominalising affixes in Aymara are shown in table 3.12.
The -iri agentive is subject-centred and tenseless, as is its Quechua equivalent in -q.
In (170) ik-k-ir-paˇc is a nominalised verb form used as a relative clause.
(170) waw ik-k-ir-paˇciˇc-su-ni-way-i
{waw(a)-Ø ik(i)-k(a)-ir(i)-Ø-paˇc(a) iˇc(u)-su-ni-way(a)-i}
child-Z sleep-IA-AG-Z-thus carry.a.child-OW-H-DT-3S
‘He went to pick up the child, asleep as she was.’
(Yapita 1991: 115)
Like in Quechua, the agentive can indicate the purpose of a motion verb, as in the
following example from Sitajara:
(171) um way-t-iri-w sara-
ŋŋ
{um(a)-Ø way(u)-t(a)-iri-w(a) sara-ŋ(a)}
78
water-Z carry.by.handle-UW-AG-AF go-1S.F
‘I shall go and fetch water.’
(Briggs 1993: 270)
The (vowel-preserving) nominalisation in -n
y
a has a much wider use than the Quechua
infinitive in -y, with which it otherwise shares many characteristics. It may refer to an
event in abstracto without any restrictions of time, participant identity, etc. It can also
express the complement of a modal verb, in which case its final vowel suffers zero
complement suppression as any nominal direct object would.
(172) ˇc’un
y
u-ˇc-iri-r un
y
-ha-n
y
mun-irista
{ˇc’un
y
u-ˇc(a)-iri-r(u) un
y
(a)-ha-n
y
(a)-Ø mun(a)-irista}
chu˜no-FA-AG-AL see-LS-IF-Z want-1S.PO
‘I should like to see the one who prepares chu˜no.’
79
(Briggs 1988: 193)
78
The loss of the final vowel in sara-
ŋ
a remains unexplained.
79
Frozen and soaked potatoes. The form ˇc’un
y
u-ˇc-iri-r can be replaced by its zero complement
correspondent ˇc’un
y
u-ˇc-ir,inwhich case it refers to the process of chu ˜no-making, rather than
to the person who makes it.
3.3 The Aymaran language family 289
Nominalisation in -n
y
a further covers the abstract functions of the future-instrumental
nominalisation in -na which is found in Quechua.
80
This includes the expression of future
or desired events, the indication of purpose, which also requires the use of the benefactive
case marker -taki, and the expression of obligation.
(173) naya-x
.
hupa-n iskuyla-r sara-n
y
a-p mun-ta
{naya-x
.
(a) hupa-n(a) iskuyla-r(u) sara-n
y
a-p(a)-Ø mun(a)-ta}
I-TO he-G school-AL go-IF-3P-Z want-1S
‘I want him to go to school.’ (lit.: ‘I want his going to school.’)
(Yapita 1991: 122)
(174) yati-qa-n
y
a-taki-w hut-ta
81
{yati-qa-n
y
a-taki-w(a) hut(a)-ta}
know-DW-IF-B-AF come-1S
‘I have come to learn.’
(Briggs 1993: 279)
(175) sara-n
y
a-ha-wa
go-IF-1P-AF
‘I must go.’ (lit.: ‘It is my obligation of going.’)
(Yapita 1991: 72)
As an alternative to the construction with the possessive marker, exemplified in (175),
it is also possible to reverbalise the infinitive by means of vowel lengthening. The person
to whom the obligation applies is then expressed by the subject marker (176). Still
another, more common, alternative involves the addition of the ownership suffix -ni to
the infinitive before reverbalisation takes place (177).
(176) naya-x
.
ˇcura-n
y
a-:-t-wa
{naya-x
.
(a) ˇcura-n
y
a-:-t(a)-wa}
I-TO give-IF-CV-1S-AF
‘I must give it to him/her.’
(Briggs 1993: 275)
(177) naya-x
.
apa-n
y
a-ni-:-t-wa
{naya-x
.
(a) apa-n
y
a-ni-:-t(a)-wa}
I-TO carry-IF-OS-CV-1S-AF
‘I must take it away.’
(Briggs 1988: 258)
Stative nominalisation is indicated by vowel-preserving -ta.Itisused for referring
to any concrete entity affected or defined by a previous action. Both object-centred
(transitive) and subject-centred (intransitive) examples are found.
80
The array of functions represented by -n
y
a in Aymara is reminiscent of that of -na in Ecuadorian
Quechua.
81
The verb huta- ‘to come’ is used without the ventive suffix -ni-inLaPaz Aymara. Bertonio
(1612a, II: 169) observes that huta- has a limited morphological valence, but mentions the
existence of huta-ni-tha ‘to come to where we are’.
290 3 The Inca Sphere
(178) p
h
aya-ta-m naya-: manq’-t’a-si-:
{p
h
aya-ta-m(a)-Ø naya-ya manq’(a)-t’a-si-:}
cook-SN-2P-Z I-AT eat-M-RF-1S.F
‘I shall eat what you have cooked.’
(Yapita 1993: 122)
(179) manq’a-ta-:-ta-ti, hani-ˇca
eat-SN-CV-2S-IR, not-IR
‘Have you eaten, or not?
(Cotari et al. 1978, part I: 220)
Nominalisation in -(:)wi is used for referring to a realised event, or to the place, time
or occasion of an event. It is often found in place names (Candaravi, Ilavi). Briggs (1993:
282) observes that the nominalising suffix -(:)wi is losing ground to -ta, particularly in
its non-local functions. There is no particular rule predicting the presence of the long
vowel; Briggs (1993) gives both -wi and -:wi as competing possibilities.
(180) naya-x
.
hupa-n ut-ha-:wi-p un
y
-h-t-wa
{naya-x
.
(a) hupa-n(a) ut(a)-ha-:wi-p(a)-Ø un
y
(a)-h(a)-t(a)-wa}
I-TO he-G stay-LS-LN-G-Z see-LS-1S-AF
‘I know the place where he lives.’
(Yapita 1991: 122)
(181) sar-naqa-wi-ni-:-n
y
a
{sar(a)-naqa-wi-ni-:-n
y
a}
go-DD-LN-OS-CV-IF
‘to have culture’, ‘to have good behaviour’
82
(Hardman et al. 1988: 267)
Aymara differs from Quechua in the inventory and behaviour of its independent
suffixes. The evidentiality distinctions are expressed within the verbal paradigm, rather
than by independent suffixes. The independent suffix -wa,which expresses affirmation
or personal witness, is reminiscent of -mi in Quechua, but has no hearsay or conjectural
counterparts. However, -wa does interact with distinctions that are expressed in the verb.
It is not found with imperatives, nor does it co-occur with the non-involver forms in -ˇci
(see above). In sentences containing a non-involver form there is, however, the optional
presence of an independent suffix -ˇci, itself restricted to the sequences -ˇci-m
83
or -ˇci-x
.
a
(Hardman et al. 1988: 287). As a matter of fact, the semantic value of non-involver -ˇci
(and, for that matter, the independent suffix -ˇci)isnearly identical to that of the Quechua
conjectural suffix -ˇc
.
i/-ˇca.
Several independent suffixes, viz. -ki,-puni (∼ -pini) and -raki, can be inserted within
the verb, as well as follow it. In fact, in Hardman et al. (1988), the term ‘independent
suffixes’ (sufijos independientes)isreserved for this particular category, the remaining
82
Sar-naqa- (lit. ‘to walk back and forth’) conveys the meaning of ‘behaviour’, ‘lifestyle’, etc.
83
The suffix -m,ofhighly restricted use, is probably related to the Jaqaru hearsay marker -mna.A
suffix -mna in Aymara was recorded by Bertonio (cf. Briggs 1993: 257).
3.3 The Aymaran language family 291
ones being termed ‘sentential suffixes’ (sufijos oracionales). The suffix -puni (∼ -pini),
like its Quechua homonym, can be interpreted as ‘always’, ‘definitely’; the suffix
-raki is translated as ‘even’, ‘more’, ‘else’, or ‘take care not to’. In (182) both -puni
and -raki are followed by the vowel lengthening that indicates the notion of a copula
‘to be’:
(182) inklisa-st kuna ˇc’ama-puni-raki-:-spa-sti
{inklisa-st(i) kuna ˇc’ama-puni-raki-:-spa-sti}
English-IR what difficult-EM-AD-CV-3S.PO-IR
‘And English, how difficult exactly would it be?’
(Hardman et al. 1988: 284)
In (183) -raki precedes the person and tense ending:
(183) naya-x
.
aymar yati.ˇc-t’a-raki-:ma
{naya-x
.
(a) aymar(a)-Ø yati.ˇc(a)-t’a-raki-:ma}
I-TO Aymara-Z teach-M-AD-1S.2O.F
‘I shall also teach you Aymara.’
(Yapita 1991: 98)
The limitative suffix -ki, equivalent to Quechua -l
y
a (‘just’, ‘only’), shares with the
two preceding affixes the ability to occur inside verbalisations, or be inserted within a
verb form; see example (165).
Aymara is rich in independent suffixes with an interrogative function. The suffix -ti
is used in negations, cf. (142), and in polar questions, cf. (179). It is the equivalent
of Quechua -ˇcu.Ifthe question consists of several alternatives, the second and further
alternatives are presented with -ˇca; see again (179). The topic, and sometimes also the
verb, in a pivotal question is marked with -sti; cf. (182).
Interrogative expressions containing WH-question words are followed by -sa (cf.
Quechua -taq); see (140) and (143) for examples. However, the same expressions, also
with -sa, can be found in negative sentences, where they indicate absolute negation (cf.
Quechua -pas,-pis,-si).
(184) hani-w makina-s kuna-s ut-ha-p-k-itu-ti
{hani-w(a) makina-s(a) kuna-s(a) ut(a)-ha-p-k(a)-itu-ti}
not-AF machine-AD what-AD exist-LS-PL-AN-3S.1O-NE
‘We have no machines, none whatever.’
(Hardman et al. 1988: 285)
The neutral function of -sa is to convey the meaning ‘too’, ‘also’. It is frequently
found in enumerations.
Aymara has several independent affixes that indicate emphasis, politeness or attenu-
ation. The topic marker x
.
a is also used as an attenuator (Hardman et al. 1988: 280), for
292 3 The Inca Sphere
instance, in a request for understanding on the part of the hearer. Its use is extremely
frequent, and its presence is not always easily explained. The suffix -ya (usually -y
or -:)isalso used for attenuation, for instance with imperatives with the meaning of
‘please!’.
Complex sentences in Aymara are often constructed by juxtaposition of non-
subordinate verbal clauses.
84
The relation between the clauses can be made explicit
by the presence of independent affixes, the form of the verb in the first clause, and by
the presence of lexical elements. Yapita (1991) gives a number of interesting examples
of such constructions, some of which are represented in (185)–(187). The equivalent
of a relative clause can be constructed by means of a full verb containing the non-
completive suffix -ka-, followed by a demonstrative pronoun uka ‘that’, which, in turn,
is grammatically integrated into the main clause.
85
(185) hupa q
h
ul
y
-t-k-i uka yap sara-:
{hupa q
h
ul
y
(i)-t(a)-k(a)-i uka yap(u)-Ø sara-:}
he plough-UW-AN-3S that field-Z go-1S.F
‘I shall go to the field that he has ploughed.’
(Yapita 1991: 123)
(186) hupa ut-h-k-i uka uta-kama-w sara-n
y
ani
{hupa ut(a)-h(a)-k(a)-i uka uta-kama-w(a) sara-n
y
ani}
he stay-LS-AN-3S that house-LI-AF go-4S.F
‘We shall walk until the house where he lives.’
(Yapita 1991: 123)
The topic marker -x
.
a is used after verbs in the potential tense and in the unmarked
non-involver tense to express a condition (‘if’).
(187) huta-n
y
a-puni-:-ˇci-x
.
a, ina-s mariyanu-x
.
uta-r q
h
aru:ru-x
.
hut-ˇci-ni
{huta-n
y
a-puni-:-ˇci-Ø-x
.
a, ina-s(a) mariyanu-x
.
(a) uta-r(u) q
h
ar(a)-uru-x
.
(a)
hut(a)-ˇci-ni}
come-IF-EM-CV-DU-3S-TO maybe-AD Mariano-TO house-AL
tomorrow-day-TO come-DU-3S.F
‘If it really should be urgent to come, Mariano may possibly come to
the house tomorrow.’
(Yapita 1991: 123)
As in Quechua, direct speech plays a major role in Aymara discourse. Direct speech
is obligatorily followed by a form of the verb ‘to say’ sa-, either in its subordinate form
84
Forafuller treatment of Aymara complex sentences see Dedenbach and Yapita (1994:
126–50).
85
The use of uka as a relativiser is reminiscent of similar uses of ˇcay-qa in Cuzco Quechua and
hina in Puno Quechua.
3.3 The Aymaran language family 293
sa-sa (188) or in one of its finite forms (189). The construction sa-sa sa- is parallel in
use and structure to Quechua IIC ni-spa ni-.
(188) hupa-x
.
apa-si-m sa-sa-w s-i
{hupa-x
.
(a) apa-si-m(a) sa-sa-w(a) s(a)-i}
he/she-TO carry-RF-2S.IM say-SU-AF say-3S
‘He/she said to me: Take it away!’
(Hardman et al. 1988: 314)
(189) kul
y
aka-ha-x
.
huta-m s-itu-wa
{kul
y
aka-ha-x
.
(a) huta-m(a) s(a)-itu-wa}
sister-1P-TO come-2S.IM say-3S.1O-AF
‘My sister told me to come.’ (lit.: ‘My sister said to me: Come!’)
(Yapita 1991: 98)
The verb sa- ‘to say’, the equivalent of Quechua ni-, is the only fully conjugated verb
in Aymara that has a monosyllabic root.
86
It is subject to the usual vowel-suppression
rules, which in this case generate consonant clusters in root-initial position. Some dialects
preserve these complex initials, but others (in particular, those of La Paz and Juli) tend
to modify the root by adding a prefixed element hi-orsi-, whenever an initial consonant
cluster obtains (e.g. hista, sista instead of sta ‘I/you say’); in Bertonio (1603b) the
prefixed element is i-. The dialect of Jopoqueri (central Oruro, Bolivia) uses a root with
a long vowel, sa:-, to avoid initial clusters. See Briggs (1993: 229–38) for a detailed
discussion.
3.3.7 Aymara lexicon
Many of the observations that were made with regard to the lexicon of present-day
Quechua (cf. section 3.2.7) are valid for present-day Aymara as well. Although the
sixteenth-century native lexicon has been preserved remarkably well, some of the most
frequent items in present-day Aymara speech are of Spanish origin. These have been
integrated so well that most speakers will not immediately recognise them as borrowed
words. For instance, the word parla- ‘to speak’ was probably derived from an archaic or
regional (Catalan?) Romance term, which is not in use in present-day standard Spanish.
The word for ‘good’, wali,was probably derived from the Spanish expression vale ‘it is
OK’, ‘it holds’. The phonotactic rules of Aymara tend to change the shape of borrowed
words considerably (e.g. hawasa ‘beans’ from Spanish habas).
The remark that the number of lexical roots, especially verbs, is limited also holds
for Aymara. However, the derivational morphology of Aymara is considerably richer
still than that of Quechua, in particular, in the domain of spatial categories. Dictionaries
contain long lists of verb roots with derivational extensions that have acquired idiomatic
86
The cognate root of sa-inJaqaru, saha-, consists of two syllables.