The article deals with the analysis of phenomena of language contact between Spanish and Quechua, found in the Doctrina Christiana y Catecismo para instruccion de los Indios (1584). These phenomena include primarily loanwords, loan blends, shifts of meaning and morphosyntactic calques, encountered throughout the Quechua version of the Doctrina Christiana y Catecismo, a profound ecclesiastical work, which influenced greatly the process of evangelization of the Andes. In addition, the article concerns other issues, like the early adaptation of Quechua to writing and phonological conundrum associated with this process, origins of the verb iñiy and the use of Quechua evidential markers in dcc. The analysis is intended to understand better the influence of evangelization on Quechua language, both in terms of diachronic linguistics and corpus planning.
Among various texts written in indigenous languages of America, which we have today at our disposal, the missionary literature of the colonial epoch occupies a special place. Being generally indigenous by form, this sort of literature introduced terms and ideas that were unknown to the native people of America before the arrival of Europeans at the end of the fifteenth century; hence the authors and translators of first Christian texts had to surmount two barriers, linguistic and cognitive, that separated them from their target audience. As vernacular tools were not sufficient to explain the new world-view, they needed to be optimized, which eventually led to deliberate changes done to a target language, and therefore to the development of complex and diverse phenomena of language contact.
The Doctrina Christiana y Catecismo para instruccion de los Indios, published in 1584, was a principal result of the work of the Third Council of Lima (1582–83), whose main purpose was to put in order the process of evangelization of the indigenous people living under its jurisdiction. Written in three languages, i.e. Spanish, Quechua and Aymara, the Doctrina Christiana was the first official and at that moment the most exhaustive compilation of ecclesiastical literature in Andean vernaculars. Since the early years of Spanish rule various Christian texts in Quechua had been circulating across the Andes, but few of them acquired approval of the diocese of Lima, established in 1541 (Estenssoro-Fuchs 1998: 48–53). A number of the members of mendicant orders, Jesuits and even laymen of certain position (as in the case of Juan de Betanzos), who came to learn Quechua or another native language, felt obliged to bring “the word of God” to the Indians, and independently worked out sermons and brief catechisms that would help them to preach Christianity among the native population. There were no unified translations for a great number of religious concepts, which could not exist in Quechua before the arrival of Europeans, and the texts of cartillas varied significantly. Not only the likelihood of an inferior translation that could lead to misunderstanding of the principles of faith, but also the divergence of the texts was a subject of concern for the Church in the early colonial Peru. According to the principles, elaborated and imposed by the Council of Trent (1545–1563) no variability of the Doctrine, performed either in written or in oral form, was allowed, and this principle must have been extended to all religious literature produced in vernacular languages as well.
The Third Council of Lima was aimed to resolve these problems and prepare one document (or more exactly, a compilation of documents) that would serve as a main tool for preaching in parishes throughout the whole Andean region. A number of church officials, ecclesiastical scholars and interpreters were summoned to realize that task. The Quechua part of the Doctrina Christiana y Catecismo (henceforth referred to as “dcc”) was prepared by a team of four clerics, headed by Juan de Balboa, a Quechua chair at the University of San Marcos and also the canon of the Lima cathedral. The others included Alonso Martínez, the canon of the Cuzco cathedral, and two mestizo interpreters, Francisco Carrasco from Cuzco and Bartolomé de Santiago from Arequipa. There were other priests of high rank who put their signature under the Quechua text of dcc, but it was unlikely that they took any part in the translation. dcc was first printed in Lima in 1584 and then distributed to the dioceses of Peru, which at that period incorporated all Spanish colonies in South America, with a strict directive to the parishes not to use any other version of the doctrine and/or catechism, or even a handwritten copy of the same book (Durston, 2007: 101–102).
This article examines the evidence of the early influence of Spanish on Quechua, as found in the Quechua corpus of dcc. To begin with, Section 2 defines the variety of Quechua used by the authors and analyzes some remarkable features of the orthographic accommodation of the language to the Latin alphabet. Section 3 includes a description of direct loanwords from Spanish to Quechua as well as loanblends. Then in Section 4 I will turn to loan shifts, i.e., shifts in the meanings of traditional Quechua words due to semantic borrowings from Spanish, and neologisms, which were formed from Quechua elements as a result of language contact. Another type of phenomena involves morphosyntactic calques from Spanish to Quechua, which are discussed in Section 5. In this section I will analyze defects in translation from Spanish to Quechua that occur in the text, in order to distinguish such cases from calques and also find out what caused them. Finally, some preliminary conclusions will be given in Section 6.
2 Dialect, Phonology, and Orthography
The variety of Quechua cartillas before the Third Council revealed not only the multitude of their authors and different levels of their knowledge of the vernacular, but also the dialectal diversity of the Quechua language itself. Quechuan is now defined as a language family, divided into two main groups, Quechua I, which corresponds to the dialectal continuum in central Peruvian Andes, and Quechua ii. The latter consists of three subgroups, namely Quechua iia (a number of dialects in the north of Peru), iib (presumably Ecuadorian varieties) and iic (Southern Quechua, which includes the dialects of Ayacucho, Cuzco, Bolivia and northwest Argentina). One of the principal tasks that was laid upon the Quechua experts involved in the translation of dcc, was to develop a text that could be intelligible for the speakers of the majority of Quechua varieties.
The Quechua variety spoken in Cuzco (now referred to as the Cuzco-Collao dialect) was taken as a basis for the language of dcc for various reasons. Firstly, it was already widely accepted that it was the dialect of Cuzco which served as the Lengua General de los Incas, i.e., the administrative language of the Inca Empire before the conquest. The political and cultural importance of the city both in pre-conquest and early colonial times undoubtedly reinforced that view. Secondly, during the period preceding the Third Council, the city of Cuzco was also notable for the scale of the pastoral activity conducted in Quechua, including translation and distribution of sermons and catechisms (Durston, 2007: 72).
At the same time, the Quechua of Cuzco contained some specific phonological features in comparison to other, more northern varieties. This refers in the first place to ejective and aspirated counterparts for a number of consonants,1 i.e., [p], [t], [č], [k] and uvular [q]. Initially there was no agreement between the translators as to how and where these sounds should have been represented in writing, and finally it was stated that it would be better not to feature them at all (see dcc, Annotaciones y Scolios: 75).
In general, the translators tended not to overuse words and expressions that seemed to them rare and geographically/dialectally specific, attesting even slightly unordinary lexical and morphological features in the glossary at the end of the book. Such simplified language corresponded to the initial purpose of unification, and also could ease the work for parish priests, for whom Quechua was not a mother tongue.
The accommodation of native American languages to the Latin alphabet in the early colonial era was heavily influenced by the contemporary Castilian phonology, so the Quechua orthography of dcc provides valuable evidence regarding the phonological structure of both languages, Quechua and Spanish. Firstly, the writers do not distinguish between velar [k] and uvular [q], a contrast encountered in most of Quechua dialects. Thus, considering the fact that the ejective and aspirated consonants (characteristic for Quechua ii-C) did not acquire any form of designation in dcc either, the letter c alone in the Quechua text could correspond to 6 different consonants, which in certain circumstances could serve for the formation of 4–5 different glosses. A natural reason for this merger was an absence of such a contrast in Spanish, similarly to other European languages that were familiar to the authors of dcc. Although it does not mean that Spanish learners of Quechua could not distinguish between the two sounds, a representation of [q] could pose a certain problem for them. However, we can see that the authors of dcc frequently use o and e before or after c [q], unlike c [k], which is followed only by u or i. Indeed, Quechua usually does not distinguish [o]/[u] and [e]/[i] phonologically. In fact, it is the uvular consonant that makes the vowels, which in other sequences correspond more to European [u] and [i], sound like [o] and [e]. As a result, the absence in Spanish of the consonantal contrast found in Quechua, as well as the absence in Quechua of the vowel distinction found in Spanish, combined with each other and revealed in writing, help us to reconstruct the original phonetic system as represented in the text.
The representation of Quechua sibilants in dcc deserves a special attention. While the modern Cuzco-Collao dialect contains only one sibilant [s], in the sixteenth century it distinguished two sibilants, similarly to the dialects of central Peru and Ecuador. However, while in Quechua I the opposition (which is still maintained) lay between [s] and [ʃ], early colonial authors unanimously assert that the Quechua of Cuzco lacked x [ʃ] (Landerman, 1982: 209). The Quechua text of dcc systematically distinguishes between ç and s, and it is obvious that ç could not designate [ʃ] in this case.
The main problem in this case is that in the same period Spanish was experiencing a significant phonological change, resulting among others in a full-scale merger of sibilants. While the official orthography norms at the turn of the sixteenth century included six sibilants (presumably apical [s], dorsal [s], [ʃ] and their voiced counterparts), only the unvoiced ones continued into the seventeenth century, with [ʃ] developing into a velar [x] (Núñez Méndez, 2012: 66–68). In addition, in the sixteenth century the use of sibilants in Andalusia, and in Seville in particular, not only significantly differed from that of the rest of Spain, but itself exhibited a kind of phonological mess (Kiddle, 1977: 331–332). As is well-known, the norms of pronunciation of this region eventually extended to Latin American Spanish. Landerman (1982) proposed a resolution of this conundrum by comparing the standards for writing Spanish, Nahuatl and Aymara in the same epoch with those for Southern Quechua. According to him, ç represented a dorsal [s], which remained in the dialect of Seville and was designated by the same letter in that period, i.e., the 2nd half of the sixteenth century, while s was an apical [s]. If we accept this view, the relationship between the Quechua I [ʃ] and Cuzco Quechua dorsal [s] (probably alveolo-palatal) becomes more obvious; the alternation ç─h, mentioned in Annotaciones y Scolios (dcc: 75), therefore can be connected to the phonological change that originated within the dialectal continuum of Central Quechua.2
Practically, the uniform character of the written Quechua of dcc significantly contributed to the growing prevalence of Quechua ii-C over other varieties of Quechua and langauges of other families, spoken in Central Andes. Tagging the Quechua of Cuzco as “pure” and other dialects, precisely those of Quechua I as “barbaric”, the authors of dcc reinforced the notion of their worthlessness. In this view, the publication, distribution and propagation of dcc was not only a part of colonial religious policy, but also of a language policy, the consequences of which can be observed nowadays: in the dominant status of Quechua ii-C, its spread onto large parts of modern Bolivia and ongoing extinction of Central Quechua dialects.
3 Loanwords and Loanblends
With regard to the use of Spanish loanwords, dcc virtually represents a new stage of development of the pastoral literature in Quechua. The authors of cartillas written during the period of “the first evangelization”, which continued up to the 1560s, tried to find analogies for Christian religious concepts primarily within the vernacular lexicon. For example, Juan de Betanzos was the first to propose the use of Viracocha (in phonemicized spelling wiraqucha, a name of one of the major Andean deities) for ‘God’. Such an approach probably had its source in a view that the Incas were conscious of the Christian virtues and acquired the notion of God by “natural reasons” before the arrival of Spaniards. This view was advocated by Bartolomé de Las Casas and was also widespread among the mendicant orders, which acted in the colonies of the New World in the mid-sixteenth century. Hence, it justified the incorporation of traditional Andean elements into the Catholic liturgy and provided a more tolerant attitude towards religious syncretism in general (Estenssoro-Fuchs 1998: 196–197).
However, owing to the increased ascendancy of the secular church and the Jesuit order (stipulated by the full establishment of the rule of the Spanish crown in Peru), the approach towards the evangelization of the native people changed drastically. The writings of the Jesuits of the period preceding the Third Council of Lima, particularly those of fray José de Acosta, severely condemned the attempts to incorporate the Inca folklore, rites and feasts above all, into the pastoral activity and denied any connection between the religious views of the Incas and Christianity. In accordance with these views, de Acosta also insisted that Christian terminology must be used as loanwords in the Quechua catechism, an attitude that was later accepted by the Third Council (Durston, 2007: 83–84). A list of the most found loanwords in dcc is present in Table 1.
A considerable part of the loanwords encountered in dcc corresponds to the terms directly associated with Christian beliefs and practices, which had no analogues in pre-conquest Andean culture. Such loanwords usually include names of institutions (yglesia), rites (misa, communion, baptismo) or titles (archangel, apostol). According to the dichotomy of Myers-Scotton (2002: 239), these loanwords lie within the category of so-called cultural borrowings, adopted in order to designate the phenomena that are new for the speakers of recipient language. But if the cultural borrowings of dcc can be recognized rather easily, a presence of core borrowings (a type, which corresponds to objects and concepts already designated in a recipient language) in the text is more problematic to confirm.
As stated above, dcc initially represented a new stage in the development of pastoral Quechua, when the church policy turned in favor of the usage of loanwords. However, the question remained, which terms must have been considered untranslatable, i.e. indispensable for doctrinal needs, and which could be replaced by native lexemes. Before dcc there had been multiple attempts to translate culturally specific terms into Quechua, and some of these translations had already struck root in the language. Viracocha for ‘God’ was seemingly abandoned even in the period of the first evangelization, but other concepts, such as ‘saint’, ‘virgin’, ‘angel’, expressed by loanwords in dcc, were once translated the by means of traditional Quechua terms. In Santo-Tomás’s Platica para todos los indios (1560), which contains the first surviving Christian text in Quechua, ‘saint’ is interpreted as Diospa yanan (literally ‘God’s servant’), while ‘angel’ is translated as alli zupay (‘good spirit’, for discussion of the ambiguity of the word zupay see the next section).3
A closer examination of the text of dcc reveals that the authors sometimes chose to use loanwords precisely when the meaning of a Quechua item differed only slightly from that of a Spanish item. However, for a cleric such semantic difference must have been obvious and even crucial. While Spanish-Quechua dictionaries of the early colonial period provide vernacular terms for a girl who is not married or has not had sexual intercourse, the loanword virgen was nonetheless put in use by the translators of dcc, as in the text it was applied solely to St. Mary, and the Quechua terms might depreciate the sacred nature of her virginity in the eyes of the indigenous neophytes (Durston, 2007: 214). A similar pattern can be traced in the case of cruz. It is known that the native Quechua name for the cross was chakatana from chakatay, ‘to cross’ (gh: 465), and it was very likely a pre-conquest word; nonetheless, it was decided to use the more symbolic word cruz as denomination for the most significant material symbol in Christianity.
A more sophisticated example is provided by the word anima ‘soul, spirit’. It is assumed that Quechua sunqu, ‘heart’, originally had metaphorical connotations,4 and some other Quechua interpretations of anima/alma also could be employed here, according to the Lexicon of Santo-Tomás (Durston, 2007: 213–214). Nonetheless, the translators preferred anima, probably in order to make the notion of a soul religiously specific.5 Firstly, it is used to denote the souls of the saints, then it refers to the spiritual works of mercy, as opposed to the corporal works, and finally it is used as a denominator for “the enemies of a soul”. This also makes us assume that it was important for the translators to utilize the Latin loanword to emphasize the contrast of the spiritual and the corporal, one of the basic principles of the Christian worldview. Any Quechua-stemmed interpretation could hardly serve the required criterion, given that the traditional Andean worldview ostensibly lacked this principle.
Another example of a ‘near-core’ borrowing is the word gracia ‘grace’, found primarily in the sermons addressed to St. Mary. Since the main Christian virtues named in dcc have proper interpretations in Quechua, similarly to the capital sins, it seems that gracia here has a particularly contextual value, as a property that is immanent to God. Therefore possible Quechua interpretations of the concept of grace, as in the case of anima, were likely considered inappropriate. Yet another example shows us that even in cases when the Quechua lexicon had a close equivalent of a Spanish term, the translators preferred to use a loanword. This is the verb ayunar ‘to fast’, conjugated in the text as a regular Quechua verb (this is also typical for other loaned verbs, for example, jurar). Southern Quechua had the word çaciy, translated as ayunar both in the dictionaries of Santo-Tomás and González Holguín, which by all appearances originally corresponded to the ritual fasting in the Inca tradition. However, for the Incas the fast presented a rather arduous form of abstention,6 which was not implemented by the Catholic Church at that period. The association of çaciy with other “pagan” practices also could affect the choice of the translators.
Thus, while from the first sight a number of loanwords in the text can formally correspond to the notion of core borrowing, in fact they conceal certain semantic differences from analogous Quechua terms, the differences, which in the context of dcc acquire particular importance. A linguist may consider them as core borrowings, but in the eyes of a Jesuit translator they apparently were cultural.
The loanblends, encountered in the text of dcc (see Table 2), can be differentiated into two different types. The first type corresponds to “hybrid compounds”, described and classified by Weinreich (1974 : 52). For example, the verbs confesakuy and baptizakuy contain Spanish roots and the Quechua mediopassive suffix -ku. Apparently, those, who first used this blend, identified the suffix -ku with Spanish se, translating Spanish reflexive verbs by choosing the marker, which seemed to have the same function as se. However, in fact the Quechua -ku bears more complex semantic nature, as most often it marks an action directed towards an actor him/her/itself (either transitive or intransitive) or performed by an actor for his/her/its own benefit (Cusihuaman, 1976: 212; Parker, 1969: 71–72). It is a question, if these features could fully correlate with the concept of confession as ecclesiastical practice, which implies address to another person, namely to priest.
The second type of loanblends includes two phrases that nonetheless serve for the translation of single concepts found in the parallel Spanish text: sanctokta ruray (literally ‘to make saints’) for Spanish “sanctificar”, and Dios kay (literally ‘being God’) for “divinidad”. It should be noted that there are also occasions when Dios and the verb kay (‘to be’) together form syntactic constructions, but do not correspond to a new concept of divine nature, which is employed only when there is a need to explain the dual nature of Jesus, indicating an opposition between his ‘divinity’ and his ‘humanity’ (runa kay). The use of ruray, a verb, which generally corresponds to material work, in sanctokta ruray, also shows that the loanblend was likely compounded by Spanish speaker.7
In general, the overview of the loanwords and loanblends found in the analyzed text demonstrates how the purposes of dcc determined the semantic nature of the terms loaned from Spanish and affected the regularity of their use. The propagation of the new religion naturally led to a great influx of cultural innovations, which inevitably resulted in the innovations of linguistic character. At the same time, the loanwords used by the authors of dcc were not merely associated with Christianity, but were aimed to emphasize the difference between it and the pre-conquest native beliefs, and thus the key concepts of the doctrine were expressed by loanwords, even if the translators could easily render them by using Quechua lexicon. In addition, the form by which the loanwords and loanblends are written in the Quechua text, exactly in the same manner as in the Spanish version, indicates that they had not passed a process of assimilation and probably still represented rather fresh novelties. Thus we can assume that dcc played a certain role in their spread and absorption by the Quechua lexicon.
4 Shifts of Meaning and Neologisms
Shifts of meaning or “loanshifts”, according to the terminology of Haugen (1950: 215), occur when lexical units of a recipient language change their semantic content as a result of the influence of a donor language. The main problem in the examination of this phenomenon in early colonial Quechua is a lack of factual evidence, based on which we could firmly state whether we deal with a meaning change in a specific case or not. Quechua was not a written language before the European invasion, and the earliest documents describing its lexicon and grammar are dated only some two decades before the publication of dcc. The most prominent sources, on which this part of the analysis is based, are the dictionary of fray Domingo de Santo-Tomás (1560) and that of Diego Gonçález-Holguín (1607), one of which predates dcc and the other one, being more substantial and more accurately compiled, postdates it. Furthermore, our knowledge of the Inca history and culture also represents a quite valuable source of information.
In Table 3, there is a list of the most common loanshifts found in dcc. The term “core meaning” in our case can only mean that we have evidence of another meaning of a discussed item, which corresponds more to the pre-conquest cultural reality of Central Andes. Therefore, “borrowed meaning” is the one, which is attested by the term in dcc carries signs of the Christian world-view.
Firstly, we shall start from the more radical shifts of meanings and then proceed to the slighter ones, less discernible, but at the same time requiring a special attention. One of the examples of semantic substitution is the word çupay. In dcc it plainly means ‘devil’, named among other “Enemies of the Soul”. However, in the dictionary of Santo-Tomás (1560: 12) the word does not carry a negative connotation in itself; in fact, it serves for the designation of ‘angel bueno’ (‘alli çupay’) y ‘angel malo’ (‘mana alli çupay’). This is an indication that the original zupay are more likely associated with supernatural powers of the Andean religion, probably with some kind of ‘shadow’ spirits.9 Thus the initially neutral lexico-semantic concept comes to obtain a clearly negative meaning in dcc; we can also assume that the partition to ‘good’ and ‘bad’ zupay disappeared in pastoral Quechua before the Third Council. Zupay then maintained the reference to pre-conquest religious practices of the Incas, but this reference now carried a clearly negative sense (Salomon and Urioste, 1991: 103).
Other key Quechua terms encountered in dcc, i.e., hucha (‘sin’) and its derivative, huchallikuq (‘sinner’), also likely developed from a substitution of meaning. There are several indications that hucha originally corresponded not to the concept of sin but to that of obligation, and was one of the terms describing the complex system of reciprocal labor in the Inca empire. Among such indications, it is necessary to note one of positions of Inca officials, cited by Guaman Poma as hucha kipuq, ‘accountant of hucha’ (gp: 111), and also a pre-conquest annual rite named qapaq hucha (‘the great hucha’), renowned for its imperial scale and lavish sacrifice (including human). The shift of meaning could occur as a result of the well-known paraphrase of Pater Noster, where ‘debts’ and ‘debtors’ imply sins and sinners. Thus, after the literal translation of Pater Noster into Quechua was made and then diffused, the terms hucha and huchalliku- could acquire the negative connotation.10 Considering the cornerstone role of the prayer in Christianity, it is logical to assume that its translation preceded the dictionary of Santo-Tomás, where hucha is already associated with sin (Santo-Tomás, 1560: 84, 138).
It is also likely that the verb pampachay acquired a new meaning of the absolution of sins as a result of the same process, since it also had a meaning of ‘settle the difference’ and ‘relieve of obligations’ (gh: 276). Here, unlike in the case of hucha, as we can see from the sources postdating dcc, the term was not deprived of its traditional semantic content. Furthermore, pampachay along with other lexemes frequently used in dcc represents an item, whose original meaning was not replaced by a borrowed one, but rather was extended to include the latter. Such extension attached a spiritual, immaterial connotation to the terms which generally corresponded to material objects and activities, so that they could serve for a clearer rendition of the principles of Christian doctrine.
Firstly, it is worth-mentioning the word aycha, which in dcc corresponds to ‘flesh’ as opposed to ‘soul’; aychanchik (‘our flesh’) is even listed among other “Enemies of the Soul”. This is an obvious calque from Spanish carne, as it can refer both to flesh/meat as food and to the human body with possible metaphorical connotations. In contrast, aycha in Quechua has the plain meaning of flesh, basically that of animal; moreover, the opposition between soul and body as between ‘good’ and ‘evil’ has not been attested in the Inca culture.
Next, there is a verb waqaychay, which in Quechua generally means ‘to guard something’ (physically), while in dcc it is used to translate the concept of observing God’s commandments, thus creating a calque from the Spanish guardar. It is also possible that the use of the verb qispi(chi)y in the sense of ‘liberation from sins’ embodies a similar calque. Another example of a similar extension of meaning is the verb watiqay, encountered in Pater Noster. According to González Holguín, this verb generally meant ‘to deceive’ or ‘to waylay’ (gh: 187), but in dcc it corresponds more to the concept of spiritual temptation. Since the traditional Andean system of beliefs does not give us any evidence of such a concept, we can assume with a certain confidence that a shift of meaning took place here as well.
The verb muchay represents a rare example of the adaptation of a term, which previously expressed a concept associated with Inca religious practices, for designation of Christian ones. There is evidence that originally the root mucha corresponded to a special manner, by which the Incas showed reverence to their deities, namely by blowing them a kiss (Estenssoro-Fuchs 1998: 92–93). Although the translators preserved the meaning of worshipping (for example with the reference to St. Mary), they also employed the verb to translate the concept of prayer. In addition, on several occasions they adjusted the suffix -pu to render the notion of saints praying to God for the souls of other Christians.11
Among the loan shifts encountered in dcc, one more case should be examined more closely, as it concerns not merely a shift of meaning, but also involves a shift in the whole system of the Andean cosmovision. This is the concept of three worlds, pachakuna, as they are mentioned in dcc. The translators use it while referring to the Christian/Greco-Roman vision of the universe as divided into three worlds: upper world, heavens, associated with paradise; this world, earth; and lower, subterranean world, associated with hell. In the Quechua corpus of dcc they are interpreted as hanaq pacha, kay pacha and uku pacha, respectively (Fig. 1).12 The adjustment of the plural ending -kuna to hanaq pacha or uku pacha is obviously a morphological calque from Spanish (cf. cielo s , inferno s). However, various sources confirm that the notions of hanaq pacha, kay pacha and uku pacha indeed existed in pre-conquest Quechua (see, for example, gp: 54, 285; Lozada Pereira, 2007: 104–105, 118–120), together forming a picture of the universe as it was perceived by the Incas. At the same time, it is assumed that the Inca universe consisted not of three but of four pacha (Fig. 2), including a part called qaylla pacha (‘world beyond the limits’). Furthermore, kay pacha (‘this world’) and uku pacha (‘inner world’) were also united in one entity named hurin pacha (‘lower world’), situated to the opposite of hanan pacha (‘upper world’), which consisted of hanaq pacha and qaylla pacha, respectively (Szemiński, 1997: 153).
In addition, the Andean worldview originally did not contain negative connotations of kay pacha as opposed to hanaq pacha; in this sense, the expression kay llulla pacha (‘this deceptive world’), featured among “Enemies of the Soul”, can be considered a lexico-syntactic calque.
Another phenomenon of ambiguous nature can be found in the paragraph of Las Virtudes Theologales. There we face two Quechua words, suyana and munana, which denominate two of the three “Theological Virtues”: Hope and Charity. The verb suyay indeed corresponds to Spanish esperar, but precisely in the sense of ‘to wait, expect’ (Soto Ruiz, 2012: 170), and it is doubtful that it ever had the meaning of ‘hope’ in Quechua. The other root, muna-, always carried the meaning of ‘love’, ‘desire’, and thus could naturally be employed for the designation of Charity in its classical sense. However, the substantivizing suffix -n(q)a refers not merely to action, but “to action in its potential state” (Parker, 1969: 55), and on the other hand can refer to an instrument or a place, by which or where the action is done. This correlates to the meaning of munana found in the dictionary of González Holguín, where it corresponds to ‘will’ (gh: 249).13 We can notice that -na here functions as an instrumental marker, since munana means something that makes one want. In this case it turns to be that it was not suyana (since hope can mean potency to wait in the same manner), but munana that underwent the meaning change.
The only more or less clear neologism encountered in dcc is the verb iñiy, ‘believe’, and its derivatives, as for example, Iñinqanchik (literally ‘what we shall/must believe in’), which serves as a denomination of Catholic faith in general, and of the Apostles’ Creed in particular. All the words with the stem iñi- are now ingrained in the Quechua lexicon exclusively in the sense applied to them by the authors of dcc. This stem was formed from the verbal stem ñi-, which means ‘say, tell’, and i-, which originally was an affirmative interjection and also appears in dcc as independent unit. It indicates that the core meaning of i ñiy was more like ‘agree’, what is confirmed by its description in the glossary (dcc: 78).
Thus, in this case the neologism was created on the basis of two processes: shift of meaning and grammaticalization.14 I would like to propose a following hypothesis about how i-ñiy could turn to mean ‘believe’. It is clear that in English and Spanish ‘believe’ and ‘creer’ can communicate two somehow different meanings, i.e. ‘believe to’/’creer a’, which means belief to some kind of information or an agent of this information, and ‘believe in’/’creer en’, which corresponds to religious or universal belief. Those who sought the Quechua analogue for ‘creer’, in order to explain the Christian theology to Andean natives, could not probably find a verb, which would correlate with ‘creer’ in its universal sense, yet through the interaction with Quechua speakers they could come upon the expression i ñiy, which generally corresponded to the first, trivial meaning of ‘creer’, for example, qamta i ñinim (‘I agree with / believe to you’) or ñisqaykikta i ñinim (‘I agree with / believe to what you said’). As a result, the first translators of the Apostles’ Creed likely calqued the other meaning of ‘creer’ onto i ñiy and also started to refer to it as a single word, thus actually producing a new verb.
The loan shifts of dcc, as well as the verb iñiy, are remarkable for the regularity with which they appear in the text. Each of the Quechua terms with evangelical connotation is strictly bound to the analogous Spanish term, so there is every reason to assume that the use of such terms was well planned, and a change of meaning, even if it was realized by the translators, in each case was considered legitimate enough to keep practicing it. In the same manner as in the case of loanwords, dcc visibly contributed to the introduction of new meanings for traditional terms, and sometimes to the elimination of their old semantic content.
5 Morphosyntactic Calques, Evidentiality and Ambiguous Phenomena
As many other Amerindian languages, Quechua is a synthetic, strongly suffixing language, and its morphology is used to communicate a large number of meanings, which in modern European languages, particularly in Spanish, are expressed analytically. Therefore, when we refer to the influence of the language contact with Spanish either on Quechua morphology or syntax, we should be aware of a rather vague boundary between them. Precisely, if we describe some specific contact phenomenon, we can easily notice how, while originating from a syntactic construction in Spanish, it becomes a morphological one in Quechua, while still remaining a calque. Thus it is more appropriate to refer to a whole set of such phenomena as morphosyntactic calques, examples of which are demonstrated in Table 4.
By all means, our notion of traditional Quechua syntax is limited even more than the notion of its pre-conquest lexicon. Therefore, in order to attest a calque with enough confidence, it is necessary to compare Quechua phrases, which contain some visible deviations from the regular syntax, to Spanish ones from the parallel text and see if we can find there a similar construction. Otherwise, a “suspicious” Quechua phrase can result from an improper handling of the language by translator/scribe, or on the contrary, can represent a rare kind of traditional construction and thus have a value in itself.
The most widespread type of morphosyntactic calque in dcc emerges from false identification of Quechua case marker. While Spanish widely uses prepositions to govern objects, Quechua in the same situations uses case suffixes. Therefore, a calque apparently develops when translator, interpreting a phrase, tried to choose a Quechua case suffix which would functionally correspond to a Spanish preposition. However, it was not taken into consideration that these units are identical in their functions only partially, so that an equality between Quechua and Spanish marker within one particular context led to a false extrapolation of this equality on every case possible.15
For example, in ‘Por la Señal’ the opening phrase is translated as:
The Quechua suffix -rayku functions mainly as causative; since the Spanish preposition por also marks causative governing in some situations, the translators erroneously drew an analogy between the two units. A similar misinterpretation occurs in the next clause:
Here the Spanish en naturally does not designate location, but serves rather as a marker of instrumental construction; nonetheless, the translator(s) mistakenly identified en with the locative marker pi.
There are other cases of the calqued use of locative, where traditional Quechua employs ‘zero’ case marker, mainly with modifiers of time. For example, in Los Mandamientos de la Yglesia we encounter constructions such as Domingocuna
pi, which literally mean “on Sundays, on holidays”. Here the locative -pi- is applied by the analogy with Spanish en
Domingos y fiestas. The forms punchaw
pi (“on the day”) o pacha
pi (“at the time”), especially in the phrases recounting biblical events, are also a common place.
Interestingly, the calqued use of case and number most frequently occurs in combination with other phenomena of language contact, i.e. loanwords, neologisms or shifts of meaning. It gives an impression that lexico-semantic change also motivates a change of morphosyntactic character, as for example in the case of hanaq pacha and uku pacha, which first turned to mean ‘paradise’ and ‘hell’, and then also acquired plural number, used in Spanish for cielos/infernos.
There are also a few examples of a calqued word order in dcc. While in Spanish the most common word order is svo, Quechua has an overwhelming preference for sov. Generally, the translators managed to follow the basic word order in Quechua, but sometimes deviations occurred. There are two constructions with svo order, in the beginning of ‘El Credo’ (a) and in ‘La Confesion General’ (b):
It is worth-mentioning that in both cases the clauses contain an enumeration of objects, at least three in line. Although Quechua usually does not alter the basic word order in such a situation, in the eyes of translators sov order might be not so obligatory, when multiple objects were present in a clause. Besides, the svo order of the opening phrase of ‘El Credo’, particularly the opening Iñinim, could have a symbolic importance for the authors of dcc as both Latin and Spanish versions of this principal Christian prayer begin with Credo/Creo, and the syntax here could be considered as an element of the doctrine.
A few words must be also said about the possible influence on the use of Quechua grammatical evidentials in dcc. The Quechua texts of the book contain only the evidential -m/ -mi, which corresponds to the most evident sort of information, usually implying that the events were directly witnessed by a speaker (in speech). Regarding the context of dcc, it is obvious that much of information rendered there could not be witnessed personally by the authors, so there is a good reason to suspect here improper treatment of Quechua evidentials. However, a discussion on this subject in Durston (2007: 185–187) shows that the translators did not necessarily violate the Quechua narrative norms. In any case, it is clear that the approach of the authors to the use of evidentials with exclusive preference for evidential -m/-mi was dictated by ideological reasons. The ecclesiastical dogma simply could not be referred as something heard from another source (-s/-si), all the more as something probable (-ch/-chá); from the point of view of clerical translators, everything related in the Doctrine was an absolute truth, and there was a strong need to emphasize that truth before the neophytes.
There are also some cases, which involve words and phrases that were simply translated into Quechua incorrectly or inaccurately. Although such cases can hardly be valuable for linguistic science by themselves, they also require our attention as some of them can be treated by mistake as language contact phenomena of the types described above. Furthermore, as already mentioned, seeming mistakes might reveal to us some hidden aspects of the Quechua language. Here we will refer only to the most remarkable cases, which stand out from other errors and possible misprints.
There is a term, encountered in the Quechua version of dcc on two occasions, which directly corresponds to misericordia or bondad (‘mercy’) from the Spanish text, namely wakchaykuya. Since it is written down and functions as one word, and at the same time consists of two roots (wakcha- ‘orphan’, kuya- ‘care’), which is an extremely rare phenomenon in Quechua, one can first take this term for neologism, by analogy with iñiy. However, in Platica of dcc and in other sources, for example, in Christian prayers cited by Guaman Poma (gp: 834, 835), we also encounter the form wakchay kuyaq (‘merciful’), which has a similar meaning, but at the same time is constructed more in accordance with Quechua morphosyntax. Literally, it means ‘one who cares for orphans’ and probably was in use as an epithet in pre-conquest period. A meaning of the suffix -y in this case is unclear, but it might be identified with the intensifier -á/-ya (Parker, 1969: 84–85). Therefore, it is obvious that the expression somehow changed its meaning by extending ‘care for orphans’ to ‘mercy’ in general, yet it would be more correct to regard wakchaykuya as erroneous writing of wakchay kuyaq, and not as a separate neologism, especially if we take into account the fact that, unlike iñiy, wakchaykuya does not appear as an independent word in other colonial Quechua sources.
The analysis of morphosyntactic phenomena of language contact in the case of early colonial Quechua represents a more challenging activity than the analysis of its merely lexical aspects. The main problem is that we have at our disposal a considerable number of dictionaries, grammars and pastoral texts of the end of the sixteenth and the beginning of the seventeenth century, but at the same time a much smaller number of texts written by indigenous authors on non-evangelical subjects. Thus we can hardly acquire a full understanding of traditional Quechua syntax before the influence of Spanish took place. It is also possible that the adaptation of Quechua to writing undermined the initial richness of its syntax and morphology, not to mention the standardization of the language. Therefore, we also should be attentive to ‘errors’ found in the Quechua text, since such ambiguous cases can familiarize us with features that were obliterated from the language in the course of its history.
The above examination of the Quechua corpus of the Doctrina Christiana y Catechismo provides us with a full range of language contact phenomena. The evidence discussed here demonstrates the multifaceted influence of the Spanish language on the evangelical Quechua of dcc, including phonological, lexical and morphosyntactic changes which only obtained a standardized written form in that period. The Quechua corpus of dcc undoubtedly has a crucial significance for the historians of the language. This is particularly true with respect to the ‘Anotaciones o Scolios de la lingua Quichua and Aimara’ and the vocabulary in the end of the book, which explains the most sophisticated and uncommon words and phrases used in the text. Hence there is a great amount of unique material that can deepen our understanding of the Quechua language before it was affected by the European influence, specifically to achieve better comprehension of its phonological and grammatical features, and to enrich our knowledge of its lexicon and phraseology.
The contents of the ‘Anotaciones o Scolios’ show, that the translators were well-informed about the peculiarities of Quechua phonology and grammar in comparison to Spanish and Latin. Nonetheless, it did not prevent them from allowing obvious morphosyntactic calques or sometimes even inadequate translations (the latter can be also explained by a possible lack of coordination of the work of the team or carelessness on the part of one of the translators). The stylistic features of certain fragments of dcc indicate that they may have been borrowed from the translations made prior to the Third Council of Lima. On the other hand, it is also possible that different members of the team of translators worked on different parts of dcc. For example, the Quechua version of Platica is distinguished by richer stylistics and lack of morphosyntactic calques, so we can assume that it was combined by a native Quechua speaker.
However, in general the regularity of most of the cases described above, especially loanwords and meaning changes, is impressive. Almost any term in the Quechua text that carries signs of European influence, either overt or covert, strictly corresponds to an analogous term in the Spanish text. Such orderliness was stipulated by the very character of the work, and since no lexical or stylistic liberties were tolerated in the case of the Spanish text, the same was to apply to the Quechua version of dcc. It should have implied a more than accurate handling of the written text and above all the use of religious terms, which were new for the native population of the Andes. Thus, it can be assumed that the translators were aware of most of the semantic changes introduced to Quechua as a result of the translation. Therefore, the influence of Spanish on the Quechua language of dcc turns to be more a product of deliberate action rather than a reflection of naturally occurring processes of language contact.
James Lockhart (1998: 42), while comparing the Quechua experience of language contact with Spanish to those of Nahuatl and Maya, noted that Quechua somehow possesses evidence only of the third, the latest stage of linguistic change, which includes borrowing on every possible level (lexical, semantic and syntactic). The two other stages, with more moderate influence of a donor language, characteristic for Nahuatl and Maya, remained virtually unattested. It is worth-mentioning that Lockhart in his comparison referred primarily to the written corpus created by indigenous authors, while in the case of Quechua such corpus was almost undisclosed till the beginning of the seventeenth century. Indeed, unlike in Mexico, the written form of Quechua developed solely on behalf of the colonists. However, the period of The First Evangelization, when colonial authors strived to interpret Christian concepts using indigenous lexicon, they visibly embodied the characteristic features of Lockhart’s “Stage I”, i.e. attempts to interpret the cultural innovations by native lexicon; moreover, the properties of the Quechua corpus of dcc (very few loanblends and neologisms, few loaned verbs, no borrowing of auxiliary words or morphemes) correspond more to “Stage 2” features than to those of “Stage 3” (Lockhart, 1998: 34). This shows us that the stages of linguistic change can be represented not only by the indigenous response to contact with a European language, but also by the attitude of Europeans towards a vernacular.
For the authors of dcc the ecclesiastical purity of the ultimate Quechua text was of prevalent importance, so the authenticity and richness of the vernacular in this case were somehow pushed into the background. The main purpose of the translators and their supervisors was to standardize Quechua and adapt it to the needs of unified pastoral activity, which inevitably led to the change of the language as a whole. A number of innovations that were reinforced or put forward by the team of translators, are now considered as inherent elements of the modern Quechua. Thus, the Quechua corpus of dcc should be regarded not merely as an evidence of language contact between Spanish and Quechua, but in a certain manner as a moving force of that contact, as a carefully adjusted act of language planning, guided by representatives of higher, Spanish-speaking society.
Hence the missionary purpose of dcc determined not only the character of the phenomena of language contact analyzed here, but also secured their incorporation into Quechua. It is a pity that the Aymara version of dcc still awaits its researchers, considering how scarce are the sources that could provide us with a notion of Aymara language in that remote period when, alongside Quechua, it was first adapted to writing. As regards the Quechua corpus of the early colonial period it is necessary to cherish and scrupulously examine every piece of it not only investigaing and classifying the consequences of its contact with Spanish, but also attempting to reconstruct the language in its pre-conquest state, of which we do not have any direct evidence.
I wish to deeply thank the persons who read the draft version of this article and helped me with their priceless comments: Prof. Jan Szemiński from the Hebrew University of Jerusalem, Dr. Justyna Olko from the University of Warsaw and Dr. Marcin Kilarski from the Adam Mickiewicz University in Poznań.
Research reported in this article was originally supported by a grant from the Foundation for Polish Science within the Focus Program for the research project nr F3/09/P/2013 “Language Encounters between the Old and New Worlds. Language as the medium of cross-cultural transfers in early New Spain”, and was then continued with the support of funding from the European Research Council, under the European Union's Seventh Framework Programme (FP7/2007–2013)/erc grant agreement n° 312795.
Doctrina Christiana y Catecismo para la instrucción de los Indio, 1584
Gonçalez Holguín  1952
Guaman Poma de Ayala  2004
Doctrina Christiana y Catecismo para instrucción de los Indios y de las de mas personas que han de ser enseñadas en nuestra sancta Fé. 1584. Lima: Antonio Ricardo primero.
Estenssoro-Fuchs Juan Carlos. 1998. Del paganismo a la santidad: la incorporación de los indios del Perú al catolicismo 1532–1750. Translation by Gabriela Ramos. Lima: Insttituto Francés de Estudios Andinos.
RuizSotoClodoaldo. 2012. Runasimi-kastillanu-inlis llamkaymanaq qullqa: Ayakuchu-chanka. I rakta / Diccionario funcional quechuacastellano-inglés: Ayacucho-chanka . Vol. I / Quechua-Spanish-English functional dictionary: Ayacucho-Chanka. Vol. I. Lima: Lluvia Editores.
1 This feature likely developed as a result of the contact with Aymara languages (Mannheim, 1991: 178).
2 Adelaar and Muysken (2004: 202–203) characterize this change as velarization of apical or alveodental [s], although the phonological divergence between Jauja and Wanka dialects (both related to Quechua I) reveals precisely the alternation from dorsal [s] in Jauja to [h] in Wanka (Mannheim, 1991: 161–162).
3 The examples are taken from the transcription of Quechua and Spanish texts provided in the analysis of Platica by G. Taylor (2001).
4 See, for example, the phraseological richness of the word in the dictionary of Gonçalez Holguín (gh: 328–331).
5 It is interesting that anima was also one of the first religious terms borrowed from Spanish into Nahuatl (Karttunen and Lockhart, 1976: 57).
6 See, for example, a description made by Felipe Guaman Poma de Ayala in his El Nueva Corónica y Buen Gobierno (gp: 190).
7 In addition, when there is a need to create a verb from a noun X, Quechua usually employs certain verbalizing suffixes, such as -cha (‘to convert smth/smb into X’ > X-cha-y).
8 Terms are given in the phonemicized writing of Quechua.
9 See also the variability of the meanings of zupay, zupan and derivative phrases in the dictionary of Gonçález Holguín (gh: 88).
10 At the same time, some scholars assume that huchalliku-, unlike hucha, was a neologism (Harrison, 2014: 20).
11 “/-pU/ indicates action performed on behalf of or at the expense of someone other than the actor.” (Parker, 1969: 71).
12 In dcc they appear most frequently written as hanacpacha, cay pacha and ucupacha.
13 “…la voluntad, que es potencia” (gh: 249).
14 Middendorf (1970: 214) considered the combinations of the verb ñiy with a number of interjections as compound verbs, but it is doubtful whether we can consider the pre-conquest i-ñiy as compound verb and not as set expression. The authors of dcc in their Quechua glossary (dcc: 78) compare iñiy with other verbs/expressions of the similar meaning, all of which contain ñini (‘Í say’), but only one, ariñini, is written as a single word, while two others, checan ñini and sullullmi ñini, are written separately, the latter even containing an evidential before ñini.
15 This misconception apparently results from the strategy of ‘word-to-word’ (in this case ‘word-to-morpheme’) translation, openly criticised by the translators themselves (dcc: 74).