Adelaar Willem, Muysken Pieter. The languages of the Andes
Подождите немного. Документ загружается.
334 3 The Inca Sphere
(259) chelu-ng-¯o
130
chelu-ng-¨an-¯o
hawk-G-AJ hawk-G-PL-AJ
‘belonging to the hawk’ ‘belonging to the hawks’
(Middendorf 1892: 53)
Instead of -ei-ˆo, Carrera often has -ii-ˆo. This is frequently the case after roots ending
in palatal consonants; e.g. ¸ciœiz/ssi¨as ‘word’, genitive ¸ciœiz-ii-o (Altieri 1939: 69).
131
However, genitives in -ii-ˆo were also recorded with roots not ending in a palatal conso-
nant (e.g. far/farr ‘celebration’, genitive far-`ıi-o), whereas -ei-ˆo has been found after a
palatal consonant; e.g. eiz/eis ‘child (relative)’, genitive eizi-ei-o (eis-i-¯o in Middendorf
1892: 56).
Middendorf also gives a number of exceptions and irregular forms, such as rak-ei-¯o
from rak ‘mountain-lion’, pe-ng-¯o from pei ‘grass’ and pojod-ei-¯o from poj ‘spleen’.
132
The word col/koj for ‘horse’ (originally ‘llama’) has two irregular options for the genitive:
col-ui-ˆo and col-ung-ˆo (Altieri 1939: 15). The interrogative pronouns ei˜n ‘who’ and ech
‘what’ have special genitive forms: i˜n-ˆo/i˜n-¯o and ich-ˆo/ich-¯o (Altieri 1939: 22).
As we anticipated, the element -ˆo/-¯o, traditionally described as a component of all
genitive endings, is absent from genitive nouns modifying another noun in a possessive
construction, in genitive nouns referring to the agent of a passive construction and before
postpositions (except benefactive -pœn/-p¨an, see above). Middendorf (1892: 52) points
out that -ˆo was not only stressed and long, but that it was also pronounced separately. It
may be assumed that -ˆo wasagrammatical element of its own, indicating, among other
things, the predicative character of a genitive noun. The ‘short’ endings of the genitive
are -œr/-¨ar, -e and -ng, respectively.
(260) mœi˜n ef-e cɥɥilpi-ss
I.G father-G blanket-RL
‘my father’s blanket’
(Villareal 1921: 71)
(261) ˜ni-ng j¯a
sea-G water
‘the water of the sea’
(Middendorf 1892: 55)
(262) mo an ang aio ˜nofn-¨ar ef-ei-¯o
this house be that man-G house-G-AJ
‘This house belongs to that man’s father.’
(Middendorf 1892: 56)
(263) ´chuvet-¨ar rr¯an-¨ad.o
snake-G bite-SN
‘bitten by a snake’
(Middendorf 1892: 55)
130
In Carrera Daza (Altieri 1939: 83) the word for ‘hawk’ is recorded as c
ɥ
elˆu.
131
The diacritic on the ending -ˆo is often left out in Carrera Daza’s grammar.
132
In Carrera Daza (Altieri 1939: 81) pol, polod-ei-o.
3.4 The Mochica language 335
(264) nech-¨ar-nik
river-G-L
‘in(to) the river’
(Middendorf 1892: 96)
As we anticipated, a number of kinship terms which form their genitive in -ei-ˆo/
-ei-¯o also have a special case ending -en to indicate the agent in a passive construction
(265). The agentive form of ei˜n ‘who’ is i˜n-in (Altieri 1939: 22) or i˜n-en (Middendorf
1892: 133).
(265) uxllur-en ei˜n xllip-quem [xllip-co ‘call’, ‘address’]
nephew-GA 1S.SG speak-T.PS
‘I am called by my nephew.’
(Altieri 1939: 16)
The formation of relational substantives, also referred to in the literature as the ‘sec-
ond nominative’ (Middendorf 1892: 56), often involves a suffix -s or -ss,asinc
ɥ
ilpi-ss;
cf. (260). The agentive nominalisation in -(V)pœc/-(V)p¨ak can be made relational by
adding -œss/-¨ass, e.g. chi-co-pœc-œss ‘someone’s creator’ (from chi ‘to be’ and -co/
-ko ‘transitiviser’; cf. Altieri 1939: 14). Of many bisyllabic substantives which end in
-Vc/-Vk,arelational counterpart is obtained by changing the final stop into -r; e.g.
˜ne˜n-uc/˜ne˜n-uk ‘toy’ (from ˜nei˜n ‘to play’), relational ˜ne˜n-ur (Middendorf 1892: 57;
Villareal 1921: 33). A third possibility to form relationals is by the addition of -Vd,
e.g. col-œd/koj-¨ad from col/koj ‘horse’.
With typically possessed nouns, the relational form may be the more basic one,
whereas the absolute form is more marked. Such nouns often have the ending -quic/-kik
in the absolute, which is either absent, or replaced by -Vng in the relational.
(266) a. ef-kik ‘father’ (absolute) ef ‘father’ (relational)
b. pol-kik ‘heart’ (absolute) pol-¨ang ‘heart’ (relational)
133
(Middendorf 1892: 57)
The relational and absolute forms of substantives need not be etymologically related.
This appears to be the case with the word mecherrœc/mecherr¨ak ‘woman’ (absolute) in
relation to ssonœng/sson¨ang ‘wife’ (relational).
Adjectives in Mochica precede the substantive they modify. In that case they are often
followed by a suffix -o (-io after vowels), which is not to be confounded with the marker
-ˆo/-¯o of the extended genitive. Carrera Daza’s grammar also contains many cases of -o
with adjectives in a predicative position (267). When the adjective acts as a modifier, the
plural marker -œn/-¨an is attached to the modifying adjective rather than to the modified
133
Villareal (1921: 37) translates the word polquic as ‘stomach’ or ‘will’.
336 3 The Inca Sphere
substantive, in which case it precedes -o (268). When the substantive is in the genitive
case, the plural marker remains on the substantive (269).
(267) mœich eix ˆutzh-o
we 1S.PL tall-AR
‘We are tall.’
(Altieri 1939: 32)
(268) ¯uts-¨an-o nep¨at
big-PL-AR tree
‘high trees’
(Middendorf 1892: 65)
(269) ¯uts-o nep¨at-¨an-¨ar-¯o
big-AR tree-PL-G-AJ
‘belonging to high trees’
(Middendorf 1892: 65)
The suffix -o is also found after substantive roots that are used attributively before
another noun.
(270) mecherrœc-o ˜nai ˜n
woman-AR bird
‘a female bird’
(Villareal 1921: 33)
Adjectives can be turned into abstract nouns by adding the suffix -œss/-¨ass,asin
pe˜n-œss ‘goodness’ from pe˜n ‘good’. Such nouns are always relational.
The verbal system of Mochica presents a rather hybrid picture, in which suffixes,
prefixes, as well as adverb-like elements play a part. The preterit is formed by the
addition of a suffix -(V)da- (-da- after vowels), to which the personal reference mark-
ers are attached. When the personal reference markers are moved to a position which
precedes the verb (see above), the resulting preterit stem ends in -(V)d, e.g. met-ed
(bring-PA). The future tense of the verb met ‘to bring’ is formed by means of a prefix t-
followed by (partially reduced) personal reference markers and subsequently by the root.
In seventeenth-century Mochica the second-person plural marker was split into a prefix
and a suffix part (Altieri 1939: 34).
134
Both the preterit and the future are represented in
table 3.20.
A remote past tense can be expressed by adding the suffix -top,but only with a
third-person subject.
(271) ssiu-ng kap¨ak j ˚um-top
that-G upon die-RM
‘On that (the Cross) He died.’
(Middendorf 1892: 158)
134
In nineteenth-century Mochica this was no longer the case, as can be seen in the example
t-¨as-chi tem (F-2-PL love) ‘you (plural) love’ (Middendorf 1892: 80).
3.4 The Mochica language 337
Table 3.20 Mochica preterit and future tenses (Altieri 1939)
Preterit Future
1 pers. sing. met-eda-i˜n ‘I brought.’ t-i˜n-met ‘I shall bring.’
1 pers. plur. met-eda-ix ‘We brought.’ t-ix-met ‘We shall bring.’
2 pers. sing. met-eda-z ‘You brought.’ t-œz-met ‘You shall bring.’
2 pers. plur. met-eda-z-chi ‘You brought.’ t-œz-met-chi ‘You shall bring.’
3 pers. sing. met-eda-ng ‘He brought.’ t-œng-met ‘He shall bring.’
3 pers. plur. met-ed-œn-ang ‘They brought.’ t-œng-met-œn ‘They shall bring.’
Another paradigm which may be formed synthetically is the desiderative. It is formed
by inserting -ma- between the root and the personal reference markers, as in met-ma-i˜n
‘may I bring!’ As an alternative, Middendorf mentions an analytic construction with the
element mang indicating desiderative, which is used in the same position as the elements
e, fe and ang (see above); e.g. moi˜n mang tem ‘may I love!’
The imperative ending for second person is -an (plural -an-chi). When the verb root
ends in a vowel (e.g. xllipco/
’
ipko ‘to call’), the ending is -n (xllipco-n/
’
ipko-n).
(272) met-an mullu
bring-2S.IM egg
‘Bring eggs!’
(Middendorf 1892: 149)
(273) met-an-chi pei
bring-2.IM-PL grass
‘Bring (plural) grass!’
(Middendorf 1892: 150)
It is also possible to locate an element an before the verb root. The resulting construc-
tion has an imperative meaning and can be translated as ‘Come and ...!’ (274). Although
this preverbal element an has been interpreted as an instance of the imperative marker
(Middendorf 1892: 140), there seems to be no reason to assume that the two markers an
share a common origin. As a matter of fact, an is also found before the so-called ‘supine’
ending in -(V)d (see below). In that case the interpretation of the resulting construction
is ‘Go and ...!’ (275).
(274) an funo
2S.IM eat
‘Come and eat!’
(Villareal 1921: 105)
(275) an funo-d
2S.IM eat-SP
‘Go and eat!’
(Villareal 1921: 103)
338 3 The Inca Sphere
The adverbial elements ca/ka and pi˜n can be postponed to a verb form, either alone
or in combination, in order to add additional shades of tense. The element pi˜n indicates
transposition of an event to the past (276), whereas ca indicates directedness towards
the future (277).
(276) met-ei˜npi˜n met-eda-ix pi ˜n
bring-1S.SG PA bring-PA-1S.PL PA
‘I was bringing.’ ‘We had brought.’
(Villareal 1921: 51)
(277) chi-da-i˜nca
be-PA-1S.SG F
‘I would have been.’
(Altieri 1939: 31)
The adverbial element chœm/ch¨am indicates obligation (278). It must not be con-
founded with the affix -c
ɥ
œm/-´ch¨am,which is said to indicate progressive aspect in
combination with polite respect, as illustrated in (279).
(278) chi-jx chœm
135
be-1S.PL OB
‘We must be.’
(Altieri 1939: 30)
(279) ciad-a-cɥɥœm-ang
136
sleep-EU-PR.CS-3S
‘You are asleep.’
(Villareal 1921: 90)
The passive is formed morphologically by adding to the verb root either one of the
suffixes -œr/-¨ar∼ -er or -œm/-¨am ∼ -em.With non-derived verb bases both endings are
used indistinctly, although -Vr is usually presented as the first option.
137
Passive verbs
are conjugated as any other verb. The choice of the vowel appears to be free, although
the ‘impure’ vowel is preferred when the passive suffix occurs in word-final position,
e.g. in (281).
(280) zoc-œrr-ei˜n pong-er tœp-œr-ei˜n lactu-ng-er
throw-PS-1S.SG stone-IS beat-PS-1S.SG hide-EU-IS
‘They throw at me with stones and beat me with hides.’
(lit.: ‘I am being thrown at with stones and beaten with hides.’)
138
135
The sequence ijx is an orthographic variant of iix.
136
In this example a euphonic vowel (copying the root vowel) is inserted.
137
A further alternative is -(V)p,asinai-ep ‘to be made’.
138
This example sentence was left untranslated by Carrera Daza and considered untranslatable by
Villareal (1921: 98). It could be interpreted with the help of the verb lexicon in Middendorf
(1892: 102). There is no ready explanation for the semantic question involved by ‘beating with
hides’. Possibly, a leather whip was meant.
3.4 The Mochica language 339
(281) mœi˜nemet-œr ca lena pei˜n pol-er mœi ˜n sson¨ang
I.G be bring-PS F along good heart
139
-IS I.G wife
‘I would be pleased to bring along my wife.’ (lit.: ‘By me
would be brought along with pleasure my wife.’)
(Villareal 1921: 98)
Verbal derivational morphology is weakly developed in Mochica. However, at least
two suffixes appear to be productive, ‘transitiviser’-co-/-ko- and ‘applicative’ -c-/-k-
(also -ec-/-ek- or -œc-/-¨ak-). The transitiviser, which may or may not have a causative
interpretation, is often found in combination with passive, in which case the resulting
portmanteau marker is -quem-/-kem-.
(282) fai˜n-ko
lie-T
‘to cheat someone’
(Middendorf 1892: 148)
(283) funo-kem-ei˜n
eat-T.PS-1S.SG
‘I am being fed.’
(Middendorf 1892: 148)
The applicative suffix may cause ablaut in a root to which it is attached, as in the case
of met ‘to bring’.
(284) mit-c-an moi˜n xllac
bring-AP-IM.2S I fish
‘Bring me some fish!’
(Villareal 1921: 30)
(285) xllipqu-ec-an mœi˜n eiz
call-AP-IM.2S I.G son
‘Call my son for me!’
(Villareal 1921: 30)
Nominalisations in Mochica include an agentive, a stative and an instrumental nom-
inalisation, as well as an abstract verbal noun. The agentive is formed by the addition
of the affixes -(V)pœc/-(V)p¨ak. The connective initial vowel of the suffix appears after
consonant-final stems. In Carrera it is a (286), less frequently œ (especially in roots with
a suffix extension). Middendorf (1892: 141) provides a more complex picture including
cases of harmony between the root vowel and the connective vowel (287a), as well as
absence of the latter (287b).
139
The form pol-er is derived from pol-quic/pol-kik ‘heart’ with elimination of the absolute affix
-quic/-kik.
340 3 The Inca Sphere
(286) a. fel ‘sit’ fil-apœc ‘one who sits’
(Villareal 1921: 22)
b. ai ‘make’ ai-apoec ‘the creator’ (Villareal 1921: 9)
(287) a. pui ‘ascend’ pui-up¨ak ‘one who ascends’ (Middendorf 1892: 140)
b. fol ‘breed’ fol-p¨ak ‘one who breeds’ (Middendorf 1892: 140)
The agentive nominalisation is frequently followed by an affix -o,which may be the
same as the one found with adjectives (see above). The resulting form is interpreted as
‘being in the habit of ...ing’ and functions simultaneously as a verb stem. The main
vowelofthe agentive suffix can then be suppressed:
(288) ai-ap¨ak-o ‘in the habit of making’
(Middendorf 1892: 113)
ai-ap(¨a)k-o-i ˜n ‘I am in the habit of making.’ (Middendorf 1892: 113)
Forms resulting from stative nominalisation, usually called ‘participles’ in the lit-
erature, refer to accomplished events. They have passive meaning when the verb
base is transitive but active meaning when it is intransitive. Stative nominalisation
is indicated by the ending -(V)d-o. Its connective initial vowel is found after conso-
nants and tends to harmonise with the root vowel, although there are several cases
where such harmony is not found. No harmony is observed in roots with an inter-
nal vowel i,where the connective vowel can be either a or œ/¨a. Middendorf (1892:
142) points to the fact that stative participles are formally related to the preterit stem
(ending -(V)d-a, see above), where he finds a similar inconsistency in the choice
of the connective vowel. As a matter of fact, the final vowel of the preterit end-
ing -(V)d-a- is often replaced by o (289). The articulated character of the ending
-(V)d-o is shown by the fact that a plural marker can be inserted between the two
components (290).
(289) met-ed.o met-ed.a-i˜n ∼ met-ed.o-i ˜n
bring-SN bring-PA-1S.SG
‘brought’ ‘I brought.’
(Villareal 1921: 52–3)
(290) j˚um-¨ad-¨an-o
die-SN-PL-SN
‘those who have died’
(Middendorf 1892: 144)
Stative participles can be used in combination with verbs ‘to be’ in order to indicate
an accomplished event. The homophony (and possible identity) of the participle and
preterit forms can lead to ambiguity in the interpretation of such constructions in relation
to periphrastic verb forms. With transitive verb stems, a periphrastic construction will
be interpreted as active if the verb is in the preterit, whereas it will be passive if the verb
is in its participle form.
3.4 The Mochica language 341
(291) moi˜netem-ed.o moi ˜netem-ed.o
Ibelove-PA I be love-SN
‘I have loved.’ ‘I have been loved.’
(Middendorf 1892: 136)
Stative participles play an important role in the formation of relative clauses. Carrera
Daza mentions sentence (292) as an example of how relative clause constructions should
look like in Mochica.
(292) œnta-zta f(e) queix Limac tœ-d.ˆo˜nofœn
not-NE be return Lima go-SN man
‘The man who went to Lima has not yet returned.’
140
(Altieri 1939: 19)
Instrumental nominalisation consists in adding a suffix -ic/-ik or -uc/-uk to a
consonant- final verb root. The relational form corresponding to these nominalisations
ends in -ir/-ur (cf. the case of ˜ne˜n-uc/˜ne˜n-ur ‘toy’, which was mentioned before). In
some cases, both vowel options co-occur yielding different meanings (293):
(293) man-ik ‘cup’, ‘drinking-vessel’ man-uk ‘dining-room’
[from man ‘eat’, ‘drink’]
(Middendorf 1892: 109)
Abstract events are referred to by forms in -i(z)¸cœc/-iss¨ak. Middendorf (1892: 110)
gives examples of the use of -i(z)¸cœc/-iss¨ak both with verb roots and with adjectives
followed by the affix -o and the verb chi- ‘to be’. The relational counterpart of this affix
likewise ends in -r: -i(z)¸cœr/-iss¨ar.
(294) a. j˚um-iss¨ak ‘death’ from j˚um ‘die’
[<*lœm]
b. t¨arr¨ak-o chi-ss¨ak ‘slowness’ from t¨arr¨ak ‘slow’ [< *tœrrœc]
(295) œnta f(e) ezta i˜n aj-i¸cœr œzta mo ef-quic
not be NE who.G make-N.RL NE this father-A
‘This Father is of no one’s making.’
141
(Altieri 1939: 87)
Mochica is rich in non-finite subordinate verb forms, traditionally referred to as
gerunds and supines. The ending -nœm/-n¨am, with occasional ablaut in the root (e.g.
mit-nœm from met ‘to bring’) indicates a purpose ‘in order to’; cf. examples (245)–
(246) above. The ending -(V)scœf/ -(V)sk¨af indicates an event that has been completed
before the main event (‘after ...ing’); -(V)lœc/ -(V)l¨ak and -(V)ssœc/-(V)ss¨ak indicate
140
In Carrera Daza’s text the first part of this sentence is subdivided as œntaz taf queix.Itmust have
incited Villareal (1921: 39) to include a non-existing verb tafqueix ‘to return’ in his word list.
In reality, we are dealing with a contraction of two negative markers œnta and ezta, followed by
an abbreviated form of fe ‘to be’ (cf. Middendorf 1892: 117). The verb ‘to return’ is queix/keˇs
(Middendorf 1892: 87).
141
I˜n is the (short) genitive of ei˜n ‘who’; aj-isanalternative way of writing for ai- ‘to make’.
342 3 The Inca Sphere
simultaneousness (‘while ...ing’). Finally, -(f)uno, -(f)unta and -(f)un can indicate a
negative subordination (‘without ...ing’).
142
(296) mœi˜nefang lœm-œd.o, tzhang Limac chi-lœc
I.G father.RL be die-SN you Lima be-GR
‘My father died while you were in Lima.’
(Villareal 1921: 102)
Particularly intriguing because of their often irregular formation and their ability to
express the active–passive distinction are the supines that indicate the complement of
verbs of motion (including the hortative particle an ‘go and ...!’; see above). The verb
met ‘to carry’ has an active supine tet and a passive supine tinipœd/tinip¨ad. Most other
verbs form their active supine in -(œ)d/-(¨a)d, -(V)p-œd/-(V)p-¨ad being a frequent option
for the passive. Both the syntax and the limited choice of the postconsonantal vowel
suggest that there is no close relation with the preterit and the stative participle in these
cases.
(297) an tin-ip-œd ˜nof faichca
go.and bring-PS-SP we.AC firewood
‘Go and have us brought firewood!’
(Villareal 1921: 103)
(298) xllon-quic ang ta tet
food-A 3S come bring.SP
‘He has come to bring (you) food.’
(Altieri 1939: 60)
Mochica has a system of numeral classifiers which refer to tens and, to a lesser extent,
hundreds of a specific class of objects. There are no numeral classifiers for units. The
first four numerals have free forms (including genitives), as well as bound forms which
are used in combination with the classifiers or as multipliers of other numerals.
Even though there is a special word for ‘ten’, as shown in table 3.21, the usual way
to count in tens was by combining the bound forms of the numerals (or full forms when
bound forms are lacking) with a numeral classifier referring to ten units of a particular
subclass of nouns. For instance, pong was used for (tens of) people, animals and reeds,
and ssop for (tens of) coins or time units (day, year). The classifier c
ɥ
oquixll/choki
’
was
used for (tens of) fruits and ears of maize.
(299) ¸coc-pong cɥɥelˆu
four-CL.10 hawk
‘forty hawks’
(Altieri 1939: 83)
(300) na-ssop xllaxll
one-CL.10 money
‘ten reales’
(Altieri 1939: 82)
142
The forms with initial f occur after o (e.g. funo-funta ‘without eating’).
3.4 The Mochica language 343
Table 3.21 Numerals 1 to 10 in Mochica (Altieri 1939: 82)
Nominative Genitive Bound
‘one’ onœc onc-œr-o na-
‘two’ aput* apt-ur-o pac-
‘three’ ¸copœt ¸copt-œr-o ¸coc-
‘four’ nopœt nopt-œr-o noc-
‘five’ exllmœtzh exllmœtzh-œr-o
‘six’ tzhaxlltzha tzhaxlltzha-ng-o
‘seven’ ˜nite ˜nite-ng-o
‘eight’ langœss langœss-œr-o
‘nine’ tap tap-œi-o
‘ten’ ¸ciœc
ɥ
¸ciœc
ɥ
-œi-o
*
The Altieri edition mentions the form atput (genitive apt-ur-o)
alongside aput. All the other sources only have aput.
(301) pak-choki
’
mang
two-CL.10 maize
‘twenty ears of maize’
(Middendorf 1892: 130)
The word for ‘hundred’ palœc/pal¨ak could be combined with the bound forms of the
numerals, yielding na-palœc ‘100’, pac-palœc ‘200’, etc. In order to count ‘hundreds of
fruits or crops’ there was a special numeral classifier chiœng/chi¨ang.
(302) pak-chi¨ang ´chun
two-CL.100 gourd
‘two hundred gourds’
In addition to the classifiers for tens and hundreds there are also classifiers for pairs.
The classifier luc/luk was used for counting fruits and crops, whereas felœp/fel ¨ap was
used for domestic fowl and vessels. The word cœss was recorded by Carrera Daza
as a classifier for counting time in tens of days, e.g. exllmœtzh cœss ‘fifty days’
(Altieri 1939: 84). In Middendorf’s time the word k¨ass apparently had lost its deci-
mal meaning. It was still combinable with bound numerals but with the meaning of
‘day’ rather than ‘ten days’. Other lexical items referring to time, such as ˇsi ‘month’
and f¯ur ‘year’, could also be combined with the bound numerals (e.g. nok-f¯ur ‘four
years’).
The Mochica numeral system also included a word for ‘thousand’ cunˆo/kuno. Com-
plex numerals involving the addition of units to tens, tens to hundreds, etc., were con-
structed by means of the connector allo, e.g. ˜nite palœc allo na-pong allo onœc ‘711’
(Altieri 1939: 83).