Adelaar Willem, Muysken Pieter. The languages of the Andes
Подождите немного. Документ загружается.
304 3 The Inca Sphere
or goal of motion in emphatic use. Obviously, additional research is needed to obtain a
full understanding of accusative marking in Jaqaru.
(196) k
h
ul
y
-ha want-ˇs-i-s-na uht-q
h
-tna
{k
h
ul
y
(u)-ha want(a)-(i)ˇs(i)-i-s(a)-na
94
uht(a)-q
h
(a)-t(a)na}
log-AC carry.on.the.shoulder-RF-SU-4S-SU come-RP-4S
‘Carrying a tree trunk we come back.’
(Hardman 1966: 93)
The instrumental-coordinative case is indicated by two different markers, -(w)ˇsqa and
-mina. The former indicates association with the subject, whereas the latter is used for
association with the object. The longer form -wˇsqa is attached to roots, whereas -ˇsqa is
found after other suffixes.
(197) hayt’-w-utu mam-ˇca-
ŋŋ-mina-wa
{hayt’(a)-w-utu mam(a)-ˇca-
ŋ(a)-mina-wa}
leave-PV-3S.1O mother-DL-1P-IS-AF
‘He left me alone with my mother.’
(Hardman 1966: 93)
(198) miˇs-uru atuqu tiŋŋku-w-i kuntri-wˇsqa
95
{miˇs-uru atuqu tiŋku-w-i kunt(i)ri-wˇsqa}
one-day fox meet-PV-3S condor-IS
‘One day fox met condor.’
(Farf´an 1952: 79)
In addition to the regular case markers, Jaqaru uses a number of spatial nouns, which
form a compound with the substantive root to which they are attached. These spatial
nouns can, but need not, be followed by a case marker.
(199) uˇc
.
u
ŋŋs-nuri-t-qa hanwa-w-i
{uˇc
.
u
ŋs(a)-nuri-t(
h
a)-qa han(a)wa-w-i}
hole-inside-AB-TO appear-PV-3S
‘It appeared from inside a hole.’
(Hardman 1966: 86)
As in Aymara, the Jaqaru verb has a considerable number of derivational affixes
referring to space, direction or manner. They normally suppress a preceding vowel.
96
Some of them, such as -kata-‘extending action, motion across’, -kipa- ‘turning mo-
tion’, -ˇsu- ‘outward motion’, ‘completion’ and -t’a- ‘at once’ have cognate equivalents
in Aymara. Others, such as -kusu- ‘as one goes’, -q
h
asa- ‘holding, maintaining’ and
94
Instead of wanta-, Belleza Castro’s dictionary (1995) lists wantu-; cf. Quechua wantu- ‘to carry
on a litter’.
95
Since the distinction between the velar and alveolar nasals is contrastive in Jaqaru, we use the
symbol
ŋ
for any constant velar nasal, even in environments where assimilation plays a role.
96
The only exceptions are the suffixes that begin with a sequence -rC-.
3.3 The Aymaran language family 305
-q
h
ul
y
u- ‘beginning’ are particular to Jaqaru. The four main directions are indicated by
the affixes -uru- ‘inward motion’, -ˇsu- ‘outward motion’, -pta- ‘upward motion’ and
-naca- ‘downward motion’.
(200) hayr-kus-k-i-wa
{hayr(a)-kus(u)-k-i-wa}
dance-as.he.goes-SM-3S-AF
‘He dances as he goes (while going).’
(Belleza Castro 1995: 95)
(201) ir-q
h
as-k-i-wa
{ir(a)-q
h
as(a)-k-i-wa}
carry-holding-SM-3S-AF
‘He is holding it.’
(Hardman 1966: 51)
(202) kuntiri-qa
97
say-pta-w-ata
{kuntiri-qa say(a)-pta-w-ata}
condor-TO stand-UW-PV-RM.HS.3S
‘The condor got on his feet.’
(Farf´an 1952: 81)
The valency-changing suffixes -iˇsi- ‘reflexive, reciprocal’ and -ya- ‘causative’
98
are
cognate with their Aymara equivalents. Remarkable is the absence in Jaqaru of a ventive
(‘hither’) marker, considering its omnipresent and obligatory character in Aymara and
in the surrounding Quechuan dialects.
99
Plural marking is optional and emphatic. It can be indicated by the suffix -rqaya- and
may refer either to the subject, or to the object.
(203) nura-rqay-ma
{nura-rqay(a)-ma}
work-PL-2S.IM
‘Work (you folks)!’
(Belleza Castro 1995: 158)
(204) hayˇc-k
h
a-rqay-sa-ŋŋa-na
{hayˇc(a)-k
h
a-rqay(a)-sa-ŋa-(a)na}
beat-M-PL-PO-1S-PA
‘I could have killed them.’
(Belleza Castro 1995: 158)
97
Hardman (1966: 105, 1983a: 170) explains that the behaviour as to vowel suppression of an
element preceding the suffixes -qa and -wa is determined by the characteristics of the element
itself; compare the effect of -qa in example (193). Clearly, more information is needed concerning
the apparently unpredictable behaviour of nominal roots before these suffixes. The absence of
vowel suppression in (202) may possibly also be attributed to Farf´an’s unsystematic orthography.
98
The dictionary of Belleza Castro (1995) contains several cases of causative derivation in -aya-
with suppression of the preceding root-final vowel if it is not -a-.Insome cases, forms in -ya-
and in -aya- coexist, e.g. naki-ya- ‘to make (it) burn’, but nak´a-y-k-iri ‘arsonist’.
99
Cerr´on-Palomino (2000) notes the use of a marker -ni- (as in Aymara) in combination with the
future tense in Jaqaru. This coincides with our own preliminary observations.
306 3 The Inca Sphere
The formation of the indicative mood in Jaqaru involves the use of the affixes -k- and
-w-,which respectively indicate simultaneousness and previousness.
100
Both markers
suppress a preceding vowel except for -w-when it is itself followed by a consonant. The
use of these affixes is interrelated with that of a marker -q
h
(a)- referring to repetition
or restitution. The latter can either precede the simultaneous event affix or follow the
previous event affix. In combination with a verb not otherwise marked for tense, subordi-
nation or nominalisation, the presence of -k-isinterpreted as present tense, whereas -w-
refers to past tense. There is also the option of an unmarked form which neither contains
-k- nor -w-.Itisnot very frequent, except in verbalised expressions and in verbs with
the repetition marker -q
h
(a)-; cf. Hardman (1983a: 91) and example (196) above.
The ‘transitional’ verbal endings that indicate person of subject and object in Jaqaru
are formally related to their Aymara counterparts, but they present an additional option.
The ending -uˇsta is reserved for a subject in the second person acting upon an object
in the fourth person ‘you–us’. (Conversely, 4S.2O does not exist.) One might expect
this 2S.4O combination to be logically excluded given the fact that the fourth person
has been interpreted as the sum of the first and second persons (cf. section 3.3.4). In a
discussion of the personal reference systems of the Aymaran languages, Hardman (1975:
448) contrasts this characteristic Jaqaru combination with the situation in Aymara, where
her consultants considered it a ‘semantic impossibility.’
101
The ending -uˇsta is related
to Aymara -ista,which is used to indicate the transition 2S.1O (‘you–me’). The latter
combination is expressed in Jaqaru by means of an ending -uta, possibly a cognate of -ita
of the Aymara imperative paradigm. Also worth mentioning is the ending -ima,which
indicates a first-person subject and a second-person object (Aymara -sma). The Jaqaru
non-future indicative personal reference paradigm is represented in table 3.14.
In addition to the unmarked tense just introduced, Jaqaru has two remote tenses, a
near-remote past and a far-remote past. The distinction between the two is again reflected
by the presence of the markers -k- (in the near remote) and -w- (in the far remote). If the
subject is third person and there is no morphologically encoded object, the ending can
either be -ana or -ata. The latter is used for information from hearsay and in combinations
with the repetitive marker -q
h
(a)- (only allowed after -w-). If the subject is either not third
person, or it is part of an explicit subject–object transition, the endings in table 3.14 are
used with the following additions. An element -Vh- with a harmonically variable vowel
is inserted after the markers -k- or -w-, and before the personal endings. This element is
obligatory if the personal ending begins with a consonant (1S, 2S, 4S and 3S.2O), and
optional if it begins with a vowel (1S.2O, 2S.1O, 2S.4O, 3S.1O, 3S.4O). In contrast, the
100
The common denominations for these two markers are ‘present’ and ‘past’ (Hardman 1966,
1983a; Belleza Castro 1995). These denominations do not seem adequate because both markers
can occur in other temporal contexts as well.
101
Speakers of Cauqui do accept the combination, but a formal difference between 2S.4O and
2S.1O is maintained only in the future tense (Hardman 1975).
3.3 The Aymaran language family 307
Table 3.14 Subject and subject–object
endings of the unmarked tense in
Jaqaru (Hardman 1966, 1983a)
1 pers. subject (3 pers. object) -t
h
a
2 pers. subject (3 pers. object) -ta
3 pers. subject (3 pers. object) -i
4 pers. subject (3 pers. object) -tana
1 pers. subject + 2 pers. object -ima
2 pers. subject + 1 pers. object -uta
2 pers. subject + 4 pers. object -uˇsta
3 pers. subject + 1 pers. object -utu
3 pers. subject + 2 pers. object -tama
3 pers. subject + 4 pers. object -uˇstu
latter five personal endings must be followed by the suffix -ana,which retains its initial
vowel before a pause but loses it when followed by other suffixes. Retention of the initial
vowelof-ana means that the final vowel of the personal endings is suppressed, thus
eliminating the contrast between second- and third-person subjects in the transitions at
issue.
102
(205) il
y
-w-ih-ima-n-wa
{il
y
(a)-w-ih-ima-(a)n(a)-wa}
see-PV-RM-1S.2O-RM-AF
‘I remember that I looked after you.’
103
(Hardman 1983a: 97)
(206) iˇsa-w il
y
-k
h
-k-ah-tam-t
y
i
{iˇsa-w(a) il
y
(a)-k
h
(a)-k-ah-tam(a)-t
y
i}
not-AF see-suddenly-SM-RM-3S.2O-NE
‘He/she had not seen you.’
(Hardman 1983a: 97)
As in Aymara and Quechua, the future tense has special endings which cannot be pre-
dicted from their non-future counterparts. When contrasting it with the Sitajara paradigm
presented in section 3.3.6 (table 3.9), one observes that the future paradigm in Jaqaru
has m where conservative Aymara has
ŋ
,except for the first-person marker, which is -
ŋ
a
in both languages.
The non-transitional subject endings for first, third and fourth persons cannot
be analysed any further. The fourth-person ending is furthermore identical with its
non-future counterpart. As for the other endings, some generalisations can be made.
102
This unusually complicated piece of grammatical description is based on Hardman (1983a:
96–9), which differs substantially from the earlier Hardman (1966: 57–9).
103
According to Hardman, the element -Vh- adds the notion of personal remembrance in those
cases where its presence is optional.
308 3 The Inca Sphere
Table 3.15 Subject and subject–object
endings of the future tense in Jaqaru
(Hardman 1966, 1983a)
1 pers. subject (3 pers. object) -
ŋ
a
2 pers. subject (3 pers. object) -mata
3 pers. subject (3 pers. object) -ni
4 pers. subject (3 pers. object) -tana
1 pers. subject + 2 pers. object -mama
2 pers. subject + 1 pers. object -utumata
2 pers. subject + 4 pers. object -uˇstumata
3 pers. subject + 1 pers. object -utuni
3 pers. subject + 2 pers. object -matama
3 pers. subject + 4 pers. object -uˇstuni
Third-person subject endings that do not involve a second-person object are formed
by adding the 3S future ending -ni to the corresponding non-future endings. The
non-transitional second-person subject ending and the transitional endings involving
a second-person object are characterised by the insertion of an element -ma- before the
corresponding non-future ending (with loss of -i- in the case of 1S.2O -ima > -mama).
Finally, the transitional endings involving a second-person subject are formed by adding
the non-transitional second-person ending -mata to the corresponding non-future end-
ings in which the final -a- is replaced by -u-.
It should be observed that the longer future endings are subject to all sorts of vowel
suppression, so that they rarely, if ever, occur in their full form. For instance, suffix-initial
-u- is only preserved after the suffix -k
h
(a)- ‘suddenly’, otherwise -ˇstumata,-ˇstuni,
-tumata and -tuni are found (with suppression of the preceding vowel). The fourth-
person ending is -tna before a pause, but -tan-or-tana- elsewhere. The first-person
suffix -
ŋ
a preserves the preceding vowel, possibly as a disambiguating device vis-`a-vis
the possessive ending; cf. example (194). It retains its own vowel only before a pause.
The third-person ending -ni always retains its full shape, preserves the preceding vowel
before a pause, but eliminates it elsewhere.
(207) il
y
-ˇstuni-wa
{il
y
(a)-(u)ˇstuni-wa}
see-3S.4O.F-AF
‘He will see us.’
(Hardman 1983a: 101)
(208) aru-ni ar-ni-wa
{aru-ni}{ar(u)-ni-wa}
call-3S.F call-3S.F-AF
‘He will call him.’ ‘He will definitely call him.’
(Hardman 1983a: 100)
3.3 The Aymaran language family 309
The imperative paradigm presents the complication of having distinct forms for use
in negative and affirmative constructions whenever the ending involves a second-person
subject. In a negative construction the endings of the non-future indicative are used:
(209) han hayt’-ta-t
y
i
{han(i) hayt’(a)-ta-t
y
i}
don’t leave-2S-NE
‘Don’t leave him!’
(Hardman 1983a: 103)
The affirmative imperative ending for (non-transitional) second-person subject is -ma.
It suppresses the preceding vowel before a pause, but preserves it, losing at the same
time its own vowel, before other affixes. The other 2S endings can be derived from the
future endings by eliminating final -ta. The 3S endings in -ni can be derived from their
future counterparts by replacing -ni with the (vowel-suppressing) ending -p
h
a; 3S-2O
has the same ending as its future equivalent (-matama). The 4S ending -tana is kept apart
from its non-imperative counterparts by preserving a preceding vowel. All trisyllabic
endings lose their middle vowel (-matma,-utma, -uˇstp
h
a, etc.) before a pause, but an
initial vowel u is generally preserved.
(210) pur-ˇsu-q il
y
-uˇstma
{pur(i)-(u)ˇsu-q(a) il
y
(a)-uˇstma}
arrive-SU.SS-TO see-2S.4O.IM
‘Come and see us when you arrive!’
(Hardman 1983a: 104)
The Jaqaru potential is based on the equivalent of the Aymara -sa- paradigm; cf. section
3.3.6. (table 3.11). The non-transitional subject endings are -sa (1S), -sama (2S), -sp
h
a
(3S) and -sana (4S). The transitional forms involving a second-person subject can be
derived from the future forms by substituting -sama for -mata. The transitional forms
that involve a third-person subject ending -ni in the future tense can be derived from
the latter by substituting -sp
h
a for -ni. Interestingly, the combinations 1S.2O and 3S.2O
have unanalysable endings: -ˇstama and -masama, respectively. Preceding vowels are
preserved before -sp
h
a, -sama and -sana but suppressed before the other endings. The
past tense of the potential (cf. Aymara) can be derived from it by adding -(a)na in
combination with a few formal adjustments. The subject endings of this paradigm are
-sa
ŋ
a-na (1S), -sama-na (2S), -sapa-na (3S) and -sana (4S), the last one being identical
to the non-past form. In both potential paradigms suffix-initial -u- is seldom effectively
present.
104
104
Hardman (1983a: 106) observes that this has only been found to be the case after the root
at
y
ama- ‘to warn’, which is then reduced to at
y
m- (e.g. at
y
m-uˇstusama-na ‘you should have
warned us’).
310 3 The Inca Sphere
Jaqaru has a well-developed system of verbal subordinate markers involving switch-
reference. The switch-reference system comprises a derived subject (‘same subject’)
subordination marker -uˇsu,aswell as a full set of markers referring to specific persons
and combinations of persons that can be used to indicate a different subject in the
subordinate verb with respect to the superordinate verb. Object marking is possible but
not obligatory. The combination of a third-person subject and a second-person object
(3S.2O) lacks a specific morphological expression.
The subordinate forms with subject marking are formed according to a pattern -i-P-
na,inwhich P represents the nominal possessive marker, except that the third-person
morpheme -pa occurs without aspiration. (This pattern coincides in principle with the
Morocomarca Aymara paradigm discussed in section 3.3.6.) First- and fourth-person
objects can be indicated by substituting -utu- and -uˇstu-, respectively, for the element -i-.
The 1S.2O transition is indicated by the ending -imamana. All subordination endings
are subject to different types of vowel suppression, in particular those affecting their
initial vowel.
Switch-reference is clearly relevant when the third-person subject ending -i-pa-na
is contrasted with the same subject form in -uˇsu.Itisthe situation where the switch-
reference mechanism attains its maximum functionality in a discourse. However, if the
verbs in the construction share a non-third-person subject, both subordinate forms, either
with or without person-of-subject marking, can be used. The use of the derived subject
marker with referentially identical third-person subjects is exemplified in (211). For a
case of its use with second persons see (210) above. Example (212) shows that, when the
subjects are identical but not third person, explicit subject marking in the subordinate
verb is also possible.
(211) at
h
ataˇsk-uˇsu im-iri ma-k-i
{at
h
ataˇsk(a)-uˇsu im(a)-iri ma-k-i}
seed receive-SU.SS sow-AG go-SM-3S
‘After receiving the seed, he goes sowing.’
(Belleza Castro 1995: 169)
(212) misa-s(a) iˇsp-i-s-na uht-q
h
a-tn(a) ak-
h
a
{misa-s(a) iˇs(a)p(a)-i-s(a)-na uht(a)-q
h
a-t(a)n(a) ak(a)-ha}
Mass-4P hear-SU.4S.SU come-RP-4S here-AC
‘After hearing our Mass, we came back here.’
(Hardman 1983a: 120)
The Jaqaru subordination markers can be combined with the marker -k-toindicate
simultaneousness (213). The element -k-ata- indicates simultaneousness in a remote
unwitnessed past (214).
3.3 The Aymaran language family 311
(213) hayr-k-i-
ŋŋa-n-qa
{hayr(a)-k-i-
ŋa-n(a)-qa}
dance-SM-SU-1S-SU-TO
‘while I am dancing...’
(Belleza Castro 1995:129)
(214) uk-nur-n(a) hayr-k-ata-p-na nak-ˇsu-w-ata
{uk(a)-nur(i)-n(a) hayr(a)-k-ata-(i)-p(a)-na nak(i)-ˇsu-w-ata}
that-inside-L dance-SM-RM-SU-3P-SU burn- OW-PV-3S.RM.HS
‘He burned them all, while they were dancing inside.’
(Hardman 1983a: 122)
In Jaqaru it is possible to express an infinitive-like nominalisation function of the verb
by using its root without any specific ending. The verbs ma- ‘to go’ and sa- ‘to say’ have
extended forms (maha, saha) for this function.
(215) kumpari, na-ps hum-hama-w hac mun-k-t
h
a
{kumpari, na-ps(a) hum(a)-hama-w(a) hac(a) mun(a)-k-t
h
a}
compadre I-AD you-CP-AF sing.N want-SM-1S
‘Compadre,
105
I also want to sing like you.’ (Hardman 1966: 116)
As in Aymara, the agentive nominaliser ends in -iri.Asinthe case of the third-person
subject marker -i, the final vowel of the morpheme preceding the affix is lost. (Compare
Aymara where in such a case u is retained, but other vowels are lost.)
106
In its main
function of indicating a subject-centred deverbal noun, the agentive marker is preceded
by the marker of simultaneousness -k- when referring to a specific activity that is taking
place or has taken place. If the event is either non-specific, or still to occur, -k- is absent.
(216) niwni-k-ir-mna il
y
-utu
{niwni-k-ir(i)-mna il
y
(a)-utu}
steal-SM-AG-HS see-3S.1O
‘They say they saw me stealing.’
(Hardman 1966: 78)
(217) niwn-iri haqi
{niwn(i)-iri haqi}
steal-AG person
‘a thievish person’, ‘a thief’
(Belleza Castro 1995: 122)
The marker of stative nominalisation is -ta,asinAymara (218). When explicitly
referring to present tense or to an event simultaneous with the action denoted by the
105
Compadre (Spanish): the godfather of one’s child.
106
The verbs ma- ‘to go’ and sa- ‘to say’ either form their agentives on the basis of the extended
root (mah-iri, sah-iri), or two special forms, ma-li and sa-li, are used.
312 3 The Inca Sphere
main verb, the sequence -k-ata is used instead. It can be analysed as containing the marker
-k- of simultaneousness followed by an element -(a)ta,which is probably identical with
-ta (219).
(218) ima-ta
sow-SN
‘what has been sown’
(Belleza Castro 1995: 70)
(219) miˇsi-w-k-ata-p-t
h
awal-naqa-ya-rqay-k-i
{miˇsi-w-k-ata-p(
h
a)-t
h
awal(a)-naqa-ya-rqay(a)-k-i}
cat-CV-SM-SN-3P-AB run-DD-CA-PL-PR-3S
‘Because he is a cat, they persecute him.’
(Belleza Castro 1995: 90)
For non-realised events the nominaliser -nuˇsu is used. Purposive clauses are indicated
byacombination of -nuˇs(u) and -taki (‘benefactive case’).
(220) pal-nuˇsu
{pal(u)-nuˇsu}
eat-FN
‘something to eat’, ‘food’
(Belleza Castro 1995: 131)
(221) pal-nuˇs-p-taki
{pal(u)-nuˇs(u)-p(
h
a)-taki}
eat-FN-3P-B
‘in order for him to eat’
(Belleza Castro 1995: 124)
Copula verbalisation in Jaqaru is realised by means of a segmental suffix -w(a)- (cf.
Hardman 1983a: 177–9). This suffix always has the shape -w- except before certain
suffixes that require a preceding vowel (for instance, third-person subject future-tense
-ni). An illustration has been given in (219). Example (222) contains a sequence of the
verbaliser -w(a)- and the marker of previousness -w-; (223) illustrates the use of the
copula verbaliser with a noun marked for instrumental case.
(222) amru-(w-)w-ata
{am(u)ru-w(a)-w-ata}
good-CV-PV-3S.RM.HS
‘It turned out to have been good.’
(Belleza Castro 1995: 187)
(223) ipi-m-ˇsqa-w-k-ta
{ipi-m(a)-(w)ˇsqa-w(a)-k-ta}
paternal.aunt-2P-IS-CV-SM-2S
‘You are with your paternal aunt.’
(Belleza Castro 1995: 187)
3.3 The Aymaran language family 313
Belleza Castro reports the existence, as in Aymara, of a verbaliser -ka- used after
the truncated form of the locative case marker -n(a) in order to indicate location:
(224) punu-n-ka-ni-wa
107
{punu-n(a)-ka-ni-wa}
Puno-L-LV-3S.F-AF
‘He will be in Puno.’
(Belleza Castro 1995: 122)
Negation in Jaqaru is indicated in a way similar to that in Aymara, with the differ-
ence that there are three negative adverbs, instead of one. The most common negative
adverb iˇsa is found in declarative sentences (206) and in combinations with adjectives
(iˇsa amuru ‘not good’, iˇsa hiw-iri ‘immortal’). The element hani (cf. the Aymara gen-
eral negation marker hani)isreserved for prohibitive expressions (209). Full negative
sentences furthermore must contain the independent suffix -t
y
i (cf. Aymara -ti), which
is also used to mark polar questions. Subordinate clauses take the negation marker
mat
y
i. The notion ‘without’ can be indicated morphologically with nouns by means
of the suffix -wiˇsi and with subordinate same-subject verbs by means of the suffix
-maya.
(225) kasra-w-t
h
a mat
y
i tat-p-psa mun-k-utu-p-na
{kas(a)ra-w-t
h
a mat
y
i tat(a)-p(
h
a)-psa mun(a)-k-utu-p(a)-na}
marry-PV-1S not father-3P-AD want-SM-1O-3S-SU
‘I married her although her father did not want me.’
(Belleza Castro 1995: 107)
(226) pal-maya ma-w-q
h
-i
{pal(u)-maya ma-w-q
h
(a)-i}
eat-NE.SU go-PV-RP-3S
‘He left without eating.’
(Belleza Castro 1995: 108)
Jaqaru has a rich array of independent affixes. The affixes -qa ‘topic marker’, -t
y
i
‘interrogative-negative’ and -wa ‘affirmative’ are similar in use to the cognate affixes in
Aymara (see the preceding pages for examples). The interrogative marker -sa is used
with interrogative pronouns to emphasise that they are used as interrogatives. This is
also one of the functions of -sa in Aymara. Note that all aspirated consonants preceding
107
The verbaliser -ka- should be kept apart from the marker of simultaneousness -k-. When
combined, both are realised as -k- (as a result of vowel suppression) so that only one suffix
may appear to be present, e.g. ika-n-(k-)k-t
h
a ‘I am in Ica’ (example from Belleza Castro 1995:
122).