Adelaar Willem, Muysken Pieter. The languages of the Andes

Подождите немного. Документ загружается.

584 6 The languages of Tierra del Fuego

9. P´ain u´aitshk¨au´ait¨a

{pa:y(n) w-ayt-ˇs-k’e w-ayte}

consequently RF-get.lost-PD-RD.MS RF-get.lost

‘So then they disappeared, they disappeared.’

10. K’¨an ¨uknk’on d´ıos k¨a- ´ut¨ashke k’a

Ruya g¨a-ut’ ¨ak¨anu;

{qenkenk’en dios k-e:wet’e-ˇs-k’e k’aroʔ ...ge:wt ʔay k’e-n naon}

Sun god 3O.MS-send-PD-RD.MS hawk...land L look.around-IF BN

‘The Sun God then sent the hawk to look around on earth,’

11. g´omshkent tu´al k’´a

{k’om-ˇs-k’n t-wa:l k’aroʔ}

not-PD-RD.NU 3S-come.back hawk

‘but the hawk did not come back.’

12. g´omshkent x¨an ¨a-uk; tsh´ait x´atkan k’´or.

{k’om-ˇs-k’n t-xeʔn ʔe:w-k’ ˇc’ay-t(o) xa:t’e-k’n k’o:r}

not-PD-RD.NU 3S-ﬂy upward-AL, be.big-AV eat-RD.NU grass

‘He could not ﬂy any more; he had eaten too much grass.’

13. K’¨an ¨uknk’on k¨a- ´ut¨ashken k´amk¨om;

{qenkenk’en k-e:wet’e-ˇs-k’n kamk’om}

sun 3O.FE-send-PD-RD.FE dove

‘Then the Sun God sent the dove.’

14. k’´or ´an¨ashk ¨a-uk k´amk¨om.

{k’o:r ʔa:ne-ˇs-k’ ʔe:w-k’ kamk’om}

grass bring-PD-RD upward-AL dove

‘The dove brought (a leaf of) grass upwards.’

15. G¨a-ut ´arshk¨an

{ge:wt ʔa:r-ˇs-k’n}

land be.dry-PD-RD.NU

‘Then the land was dry.’

The Spanish presence

Manuel Tenorio, over sixty years old, and a Quechua peasant from central Ecuador,

recounts how he learned Spanish as a boy, on the job at the hacienda and during visits

to the market town. When a white man came by on horseback, Manuel was obliged to

say the greeting alabado sacramento ‘blessed (be the) Sacrament’ or alabado evangelio

‘blessed (be the) Gospel’, and was answered back alabado hijo ‘blessed (be the) Son’

(perhaps also ‘blessed, sonny’). However, as a Quechua speaker he had great difﬁculties

with the vowels [e] and [o] (often pronouncing them as [i] and [u], respectively), and

the punishment for mispronunciation was ﬁerce, so he used to dive into the irrigation

channel when he heard a horse approaching. Almost at the end of his life, in spite of

years in the capital as a construction worker, his Spanish still shows many traces of his

Indian background.

The Spaniards have imposed their language in all their former colonies in the New

World. Not only the immigrants from Europe speak Spanish, but the descendants of the

African slaves and Chinese coolies, and a large part of the native Amerindian population

as well. Nonetheless, the transplanted forms of the language have undergone many

modiﬁcations, as in the speech of Manuel Tenorio, and the relationship between Spanish

and the dominated Amerindian languages is a complex one.

In this chapter we discuss various aspects of the relation between Spanish and the native

languages of the Andes. We ﬁrst look at Andean Spanish from a number of perspectives,

and then at various dimensions of language planning and bilingual education.

Other immigrant languages, such as Mennonite German and Japanese in Bolivia,

Chinese in Peru, and Lebanese Arabic in different Andean countries, will not be dis-

cussed: they are characteristic of the New World as a whole rather than of the Andean

nations.

7.1 Characteristics of Andean Spanish

Andean Spanish can be approached from many perspectives. We will brieﬂy discuss the

relation of this variety to the dialects of the metropolis, before turning to issues such as

the inﬂuence it may have undergone from Quechua and other Amerindian languages.

586 7 The Spanish presence

7.1.1 Demography and Iberian dialectal origins

The question of which Iberian dialects contributed most to New World Spanish is a

vexed one, but a modicum of consensus emerged in the 1970s, mainly due to work by

Boyd-Bowman (e.g. 1973). There are several main issues:

(a) The varieties of Spanish in the New World are much more homogeneous than the

ones in Spain itself, in spite of the much wider geographical spread, the lack of inter-

regional communication, the potential diversifying inﬂuence of Amerindian substratum

languages, etc. This is part of a general phenomenon noticed by the anthropologist

George Foster in his Culture and Conquest (1960), namely that export varieties of a

culture represent only a limited subset of the cases of variation present in the source

culture. Boyd-Bowman (1973) points to the fact that in the early period, 1490–1520,

there may have been extensive dialect levelling in the Antilles, and that the mainland

was populated from the islands. Also, the Canary Islands may have been a site for

dialect levelling intermediate between Spain and the New World. From other studies

of export varieties of European languages and from creole studies we know that of-

ten the early period in colonisation is decisive for the establishment of new linguistic

norms. The levelling process may already have been started in Seville, the Spanish port

through which the emigration stream was funnelled, and this brings us to the second

issue.

(b) A classic problem has always been that Latin American Spanish resembles the

southern varieties of peninsular Spanish, particularly Andalucian Spanish, more than the

standard in the sixteenth century, which was surely that of the court in Toledo and later

Madrid, i.e. Castilian. We have already pointed to the role of Seville as the Atlantic port,

but Boyd-Bowman’s main contribution lies in establishing that indeed the main body

of emigrants to the New World was from the south. For the whole period 1493–1579,

35.8 per cent of the settlers came from Andalucia and 16.9 per cent from neighbouring

Extremadura. The two Castiles together only provided 29.5 per cent.

For the early period, the ﬁgures are even more striking. Thus between 1493 and 1519

the Seville province provided six times as many (1,259 of a total of 5,481) emigrants

as the province of Toledo (208), and the nearby provinces Huelva, Badajoz and C´aceres

provided another 1,174. If we look at women settlers, who may have played a crucial

role in creating the new linguistic norms, between 1509 and 1519, when the New World

society was being given shape, over half (of 308 total) were from Seville.

This type of ﬁgure is fairly constant for the different Andean regions, although it is

noted by Boyd-Bowman that Colombia and Venezuela had a somewhat higher proportion

of settlers from the two Castiles than Peru.

areas tend to resemble Andalucian speech more directly than the highland areas, actually

throughout Latin America. Boyd-Bowman (1973: 67) tentatively explains this in terms

7.1 Characteristics of Andean Spanish 587

of ‘a vast maritime empire the ports of which were linked by sea to Seville (and to

each other) along trade routes controlled and maintained predominantly by Andalucian

sailors and merchants’. The ensuing network may have had a more direct and enduring

inﬂuence on the coastal areas around the ports than on the highland regions.

7.1.2 Linguistic features

It has been said many times before, but systematic and reliable studies on varieties of

South American Spanish are few and far between; cf. the critical evaluation by Lipski

(1994) of a number of studies, as well as Fontanella de Weinberg (1976). The discussion

of speciﬁc Andalucian dialect features has focused on phonology. The most striking

feature, of course, shared by Latin America and southern Spain is the absence of the

[θ]/[¸s] contrast (‘ceceo’ versus ‘seseo’) so characteristic of Castilian. Other features,

such as aspiration and loss of word-ﬁnal /s/, loss or weakening of intervocalic /d/, and

merger of /l

/ and /y/ (‘ye´ısmo’) can also be mentioned, but are characteristic of the

coastal areas, not of the highlands. In addition, these features are not realised uniformly

either in southern Spain or in Latin America, and in many places need more detailed

study.

The most immediately striking difference in the domain of morphosyntax concerns

the second-person pronouns. Very roughly, the present situation in some places is as

follows:

(1) Spain South America

2.SG familiar: t´u 2.SG familiar: vos

2.SG polite: usted 2.SG intimate: t´u

2.PL familiar: vosotros/as 2.SG polite: usted

2.PL polite: ustedes 2.PL ustedes

Here ‘intimate’ refers to respectful address in a close relationship, and with ‘familiar’

no particular respect is implied.

The form vos was present in older peninsular Spanish as a polite second-person-

singular pronoun, but has largely disappeared in Spain and in parts of Latin America,

such as coastal Peru and Ecuador. In some areas (e.g. Argentina), t´u has been replaced

almost entirely by vos, and there the original verbal inﬂections associated with the

latter pronoun are preserved as well. Elsewhere, vos is used with the verbal inﬂections

associated with t´u.Inareas where both t´u and vos occur, the former is used in more

formal contexts and the latter is often considered a sign of lack of a proper education.

In parts of Colombia, ﬁnally, usted, the polite second-person-singular form elsewhere,

is often called for in non-intimate but familiar contexts.

There are a great many lexical differences both between the varieties of Spanish in

the different countries of South America, and between this continent and Spain. These

588 7 The Spanish presence

Table 7.1 Major isoglosses in the Andean areas of Latin American Spanish (based on

Zamora Munn´e and Guitart 1982)

/s/ /x/ vos

IV. Colombia except the coast + – ±

V. Paciﬁc coast of Ecuador and Colombia – – ±

VI. Coast of Peru, except the extreme south – – –

VII. Rest of Ecuador and Peru, west and centre of Bolivia, ++±

northwest Argentina

VIII. Chile – +±

need not concern us at the moment, although we will return to items borrowed from

Amerindian languages.

Zamora Munn´e and Guitart (1982) provide a fairly balanced and comprehensive

account of some of the major dialectal divisions in Latin American Spanish. They

base themselves on three diagnostic features, which constitute major isoglosses: (a) the

retention (+)versus loss or aspiration (−)ofﬁnal /s/; (b) the velar (+)orglottal (−)

realisation of /x/; (c) general use of vos (+), of t´u (−), of both (±). For the regions that

concern us most, they distinguish the ﬁve major dialect areas (with their numbering)

listed in table 7.1. Surely other or ﬁner divisions are imaginable, but the dialect splits

outlined in table 7.1 correspond fairly well to global impressions of speakers of varieties

of Andean Spanish.

It should be stressed that thorough descriptive work of Andean Spanish dialects is

uneven in quality and coverage (Bolivian Spanish, for instance, has been barely in-

vestigated); in addition, different studies have focused on widely different features. A

classical study remains Kany’s (1945) survey of regional grammatical features in the

different varieties of Latin American Spanish, based on written sources. Lipski (1994)

provides an excellent country-by-country account of available studies. Perl and

Schwegler (1998) survey the evidence for African inﬂuences on Ibero-American lan-

guage varieties.

Colombian Spanish is fairly well described, starting with the pioneering work of

Ruﬁno Jos´e Cuervo (1867) on the speech of Bogot´a. More recent studies include, for in-

stance, de Granda’s interesting study of Afro-Hispanic speech varieties of the Colombian

Paciﬁc coast area (1977).

Present-day Ecuadorian Spanish has been described most completely by Toscano

Mateus (1953). There are considerable lexical and phonological differences between

Coastal and Highland Spanish, on the one hand, and between educated urban Spanish

and different rural varieties on the other. On the whole Ecuadorian Spanish conforms to

the generalisation that holds for all varieties of South American Spanish: many archaic

7.2 Amerindian substratum inﬂuence 589

Peninsular features have been preserved, and the dominant dialectal inﬂuence has been

the Andalucian dialects of southern Spain.

The major source for Peruvian Spanish remains Benvenutto Murrieta (1936), although

numerous articles and monographs have appeared on individual phenomena. Ana Mar´ıa

Escobar (2000) surveys much current research on Peruvian Spanish, paying particular

attention to semantic factors, and Rivarola (1985a, 1985b, 1986, 1987) has investigated

the origins and development of Peruvian Andean Spanish. Another good general source

is Klee (1996).

Reliable information about Bolivian Spanish is rather limited. Building on an earlier

study by Kany (1947), there is a brief but detailed phonetic study by Gordon (1980),

showing that Bolivian Spanish cannot be simply subsumed under the varieties of neigh-

bouring countries, dialectologically. Further, work by Herrero (1969) and Laprade (1981)

argues that different varieties of Bolivian Spanish have undergone Quechua and Aymara

substrate inﬂuence, respectively. The complex lexical interactions between Bolivian

Quechua and Spanish are studied in Van Hout and Muysken (1994). Alb´o (1988b)

documents the ‘llapuni’ Spanish of highland migrants to the tropical Santa Cruz area.

For Chile the best recent source remains Oroz (1966). A dialect atlas of southern

Chile was compiled under the direction of Guillermo Araya (Araya et al. 1973). An

interesting study of mining vocabulary in northern Chile, which reveals the presence

of several Quechua and Aymara words, is Parada et al. (1976). Mining vocabulary,

incidentally, has also been studied, albeit in a rather amateurish way, in Bolivia, but

this extremely rich subject, revelatory of the shaping of a mixed Amerindian–Hispanic

technology and culture, still awaits a more deﬁnitive study.

Quechua inﬂuence on the Spanish of northwestern Argentina has recently been studied

by de Granda (1995, 1996, 1997a, b).

7.2 Amerindian substratum inﬂuence

The possibility of various Amerindian languages inﬂuencing Andean Spanish has been

a long-standing issue in Hispanicist scholarship, with important cultural implications as

well. When the new republics gained independence from Spain in the 1830s, an important

link to the old metropolis, but also something linking the young nations themselves, was

the Spanish language. This led to a set of contradictory attitudes. On the one hand, the

unity of Latin America, the old dream of Sim´on Bol´ıvar, was best maintained through a

uniﬁed Spanish, and thus directly a continuation of the metropolitan norm. If each new

nation were left to its own devices, it was felt, the Spanish of the Americas would fragment

just as Vulgar Latin had in the Dark Ages. On the other hand, the Latin Americans felt

a need to express, in their language, both a distance from Spain and the many particular

features of the New World and of the area they inhabited and were trying to forge into

a country.

590 7 The Spanish presence

Table 7.2 Types of speakers of Spanish that may

show inﬂuence from Quechua

(A) Quechua speakers learning Spanish

(B) Stable Quechua–Spanish bilinguals

(D) Spanish monolinguals living in bilingual regions

(E) Spanish monolinguals living in bilingual countries

It is against this background that we must see the debates about possible Amerindian

inﬂuences on Andean Spanish (for a general account of the role of language ideology

see Woolard and Schieffelin 1994). But there is more. The Spanish colonial empire was

founded on the assumption that what is Hispanic has inherent value, and what is native

is worthless if not downright ungodly. It is not until the indigenista movements of the

1920s that the Amerindian heritage was recognised as worthwhile, not just for the Indian

peasants themselves, but for the countries in general.

There have been discussions in the literature about various cases of Amerindian lan-

guage inﬂuence on Spanish. Araucanian inﬂuence on Chilean Spanish, was proposed

by Lenz (1893) (cf. chapter 5). Similarly, several studies have focused on Aymara inﬂu-

ence on the Spanish of the Bolivian altiplano (Hardman 1981). Nonetheless, the main

debate has focused on Quechua inﬂuence, and this is what we will concentrate on here,

mentioning Aymara only in passing. Given the many structural and semantic parallels

between these two languages (cf. chapter 3), it would be surprising if the inﬂuence of

Quechua and Aymara on Spanish in southern Peru and Bolivia had not led to mutual

reinforcement.

When we consider Quechua inﬂuence on Spanish, the picture is quite complex, since

much depends on which variety of Spanish is meant: the Spanish of the whole central

Andes, of the areas where Quechua is or was spoken, of lower-class mestizos in those

areas, or of Quechua–Spanish bilinguals? Taking the whole central Andes is too wide,

yet taking just bilinguals is too narrow, we believe. There are also considerable regional

and national differences. Finally, it depends what aspect of the language is considered.

Schematically, the Spanish of the different speaker types is shown in table 7.2. On the

whole, these categories are arranged implicationally: if a Quechua feature occurs with

speakers of type (D), it will also occur in (B), but not necessarily vice-versa. However,

we will see that in some cases Quechua inﬂuence exerts itself in the Spanish speech

of stable bilinguals, while the same feature is not present in the speech of incipient

bilinguals.

Lexically, the inﬂuence of Quechua on Spanish has been slight, limited to the domains

of agriculture, food, ﬂora and fauna, the household and clothing, musical folklore,

7.2 Amerindian substratum inﬂuence 591

Table 7.3 Sprachbund phenomena in the pronunciation of liquids and vibrants in

different varieties of Quechua and Spanish in Ecuador

Quechua Spanish Quechua Spanish

North [ˇz] [ˇz] [aˇzi] ‘good’ [poˇzo] ‘chicken’

South [l

][l

] [al

i] [pol

North [ˇz

], [ˇr] [ˇz

iŋri], [ˇrinri] ‘ear’ [peˇz

o], [peˇro] ‘dog’

South [rr] [rr] [rriŋri] [perro]

popular religion and illnesses. The vast majority of borrowed elements are nouns, such

as llama, alpaca and chacra ‘ﬁeld’. There are a handful of derived verbs, such as yapar

‘to add a bit (in a sales transaction)’, chimbar ‘to cross’, pircar ‘build a wall of stones

and adobe’ and chancar ‘to ﬂatten’, an occasional adverb such as pite pite ‘bit by bit’,

and exclamatives such as achachay ‘brrr, cold’ (Buesa Oliver 1965).

Afew Quechua words have gained acceptance beyond Spanish, such as English jerky

from charki ‘dried meat’ and Louisiana French la-gnappe ‘small added bit’ from the

above-mentioned yapa. Mej´ıas (1980) has shown that there were Quechua lexical items

used in Spanish from the sixteenth century onwards; their number gradually increased, at

least until the eighteenth century. Lexical inﬂuence is the only type of Quechua inﬂuence

of type (E) in the scheme above. Of course, it varies a great deal in terms of region, type

of speaker, setting and domain, etc.

The lexical inﬂuence of Aymara has been somewhat more limited in its diffusion. In

La Paz Spanish many Aymara words are used, such as chusca ‘square carrying cloth’,

imilla ‘girl’, colla ‘inhabitant of the altiplano’ and cala ‘stone’ (Laprade 1981), but

these are not known much outside the region. An important issue mentioned by Laprade

concerns lexical semantic inﬂuence exercised by Amerindian languages on Spanish. A

case in point is pie,which in metropolitan Spanish means ‘foot’, but under the inﬂuence

of Aymara kayu has come to refer to most of the leg. This same phenomenon merits

detailed investigation in the case of Quechua inﬂuence on Spanish as well.

On the phonological level, the one case of Quechua inﬂuence on Spanish reported by

all authors is the pronunciation of /o/ as [u] and of /e/ as [i]. More precisely, we encounter

fusion of /e/ and /i/ into something varying between [e] and [i], and of /o/ and /u/ into

something varying between [o] and [u]. Quechua-speakers may pronounce pero ‘but’ as

[piru], but also cubicar ‘make into a cube’ as [kobekar], due to hypercorrection. This

inﬂuence, though general in the whole Quechua-speaking Andes, is limited, however,

to (A) and (B). Even lower-class mestizo monolinguals do not have it.

In Ecuador, there are two interesting sprachbund or convergence phenomena, both

involving liquids, as shown in table 7.3. In the northern part of the Ecuadorian high-

lands the palatalised [l

], which is widely found in Andean Spanish and in Quechua,

592 7 The Spanish presence

is pronounced as a voiced palatal sibilant in the local dialects of both languages, and

the Spanish trilled [rr] (with its Quechua equivalent in word-initial position) is a voiced

retroﬂex fricative or an assibilated retroﬂex vibrant. Notice that it cannot be plausibly

claimed that Quechua inﬂuenced Spanish here, since the more conservative varieties

of Quechua (here represented by the south of Ecuador) do not possess the features in

question.

Syntactically and semantically, Quechua may have had some inﬂuence on the mono-

lingual highland varieties of rural Spanish (C and D), in addition to stamping the bilingual

varieties. If we try to imagine how this inﬂuence was exerted, the most plausible scenario

is one of second-language learning by Quechua speakers in a sociolinguistically com-

plex environment. The particular stratiﬁcation of variable elements within the Spanish

target-language speech community affects the process of acquisition of these elements.

This stratiﬁcation is crucial because it may reﬂect, in part, stages of interrupted or in-

complete L2 acquisition at an earlier point in time. As time goes on, the products of

intermediate and advanced interlanguage grammars (A and B) are incorporated into

the native speech community (C and D), but most often as vernacular, non-standard

forms. In a synchronic perspective, then, native speakers of the target vernacular end up

producing outputs which seem like interlanguage outputs. The particular interlanguage

features which come to be adopted as non-standard features in the Spanish target speech

community serve as models, at a later stage, for new learners. L´opez and Jung (1989)

show how non-standard features are present in primary school in the speech of the local

teachers with bilingual backgrounds.

The process of selection and adoption of these features, however, may be gov-

erned by factors such as superﬁcial resemblance to native language features. This

selection and adoption may result in a situation in which two speech communities

may have different vernacular varieties, while they result from the interaction and

contact of the same L1 and L2. We will illustrate this by comparing the acquisi-

tion of Spanish by Quechua speakers in different regions of the Andes. Quite dif-

ferent forms of Spanish vernacular have emerged in the different contexts, making

a simple prediction on the basis of a contrastive analysis of Quechua and Spanish

implausible.

The existing literature on non-standard Spanish, as used in regions with Quechua–

Spanish bilingualism, is by now quite extensive, but little of it is based on ac-

countable and rigorous data. Most studies deal with Peru. We will take the situ-

ation of Peruvian Andean Spanish as our frame of reference and contrast it with

what is known of other areas. The main grammatical constructions and semantic fea-

tures mentioned are listed in table 7.4. We will discuss the features in table 7.4 one

by one.

7.2 Amerindian substratum inﬂuence 593

Table 7.4 Features claimed to be due to Quechua in different

varieties of Spanish

a. Double possessives

b. Clitic usage

c. Object–verb sequences

d. Irregular gender and number concord, and irregular verb morphology

e. The redundant use of the preposition en ‘in’ in locative expressions

f. The frequent use of gerunds

g. Sentential particles and discourse organisation

h. Semantics of tense, notably of the sudden discovery tense

A. Double possessives are a good case to start with. They occur in examples such as

(Lozano 1975):

(2) Era su amiga de Juan.

it.was 3P girlfriend of John

‘It was John’s girlfriend.’ (Type I)

(3) Se quem´o del joven su pantal´on.

RF it.burned of.the young.man 3P trousers

‘The young man’s trousers burned.’ (Type II)

Note that in the ﬁrst example the possessor noun, Juan, follows the possessed noun,

amiga,while in the second one the order is the reverse. In both, there is a possessor noun

as well as the possessive pronoun su. Lozano (1975), based on material from Ayacucho,

south-central Peru, is not very explicit about his data base, and describes different aspects

of Ayacucho rural Spanish deviant from Lima standard Spanish, focusing on the question

of possible inﬂuence from Quechua. Lozano does not report on the frequencies of the

constructions described or on the issue of whether they typify monolingual vernacular

or bilingual Spanish. He concludes (1975: 304) that all cases are traceable, at least

indirectly, to Quechua interference. However, the ways in which the innovations have

been formed are varied and complex.

While Lozano (1975) makes strong claims concerning the inﬂuence of Quechua on

bilingual Spanish, Puente (1978) would like to limit this inﬂuence to subordinate bilin-

guals, more proﬁcient in Quechua than in Spanish. Puente’s work is based on observations

from Huancayo, central Peru, and Ayacucho, south-central Peru. Puente claims that type

II sentences are more frequent with compound bilinguals.

Rodr´ıguez Garrido (1982) argues that type I, with a following possessor noun, is

characteristic of Peruvian Spanish in general, and an archaic construction. It is absent