Xinka Lexicography and Morphology

Research Year:  2000
Culture:  Xinka
Chronology:  Contemporary
Location:  Santa Rosa, Guatemala

Table of Contents

Introduction
Planification and Implementation of the Project
Research Results
The Linguistic Situation in the Xinka Area
The Data
Xinka Lexicography
Xinka Morphology
Dissemination of Research Results
Concluding Remarks
Sources Cited

Introduction

Here I wish to present my project on "Xinka lexicography and morphology" which I have been able to carry out last year owing to the generous funding from the Foundation for the Advancement of Mesoamerican Studies, Inc. (FAMSI). The project pursued the aim of documenting contemporary Xinka language knowledge and analysing it subsequently with regard to the language’s lexical and morphological typology. Purpose and relevance of this research shall be argued in the following.

In the field of Mesoamerican studies, the culture and languages of the Xinka in Southeastern Guatemala have never been subject to very intensive ethnographic or linguistic research. Their language not affiliated and their cultural behaviour distinct from other Mesoamerican ethnic groups, the Xinka are of unknown origin. Formerly inhabiting an area from the Pacific coast to the Southeastern highlands covering the present-day department of Santa Rosa as well as parts of Jutiapa and Jalapa, the Xinka bordered on settlements of Ch’orti and Poqom Maya as well as Pipil groups, which attributes a certain significance to Xinka presence in the Mesoamerican cultural area.

In this settlement area, Lyle Campbell and Terrence Kaufman have distinguished the variants of Guazacapán, Chiquimulilla, Jumaytepeque, and Yupiltepeque as four different Xinka languages, which form one extremely moribund language family with the Xinka of Yupiltepeque being considered extinct since the beginning of the 20^th century (Campbell, 1972:187). Concerning the status of extinction of the other variants there have been deviating reports covering a spectrum from total extinction (Kaufman, 1997: personal communication) to more than 300 speakers (Museo Ixchel, 1996).

The linguistic documentation of the Xinka languages is scant. The only linguistic data accessible for academic research are a few concise vocabularies or word lists as well as short and insufficient grammatical descriptions. There is, furthermore, a colonial grammar with the title "Arte de la lengua szinca" written by the priest Manuel Maldonado de Matos around 1770 (forthcoming: Sachse, 2001). Together with the "Estudios lingüísticos" of Eustorgio Calderón (1908) and Otto Schumann’s thesis on the "Xinca of Guazacapán" (1967) this latin style description is the most extensive source on a Xinka language which is available heretofore. Neither of the mentioned sources of linguistic data on Xinka yields substantial information on the language’s grammatical structure. Xinka morphology and it’s linguistic typology have not been subject to intensive research.

The project "Xinka lexicography and morphology" was aimed at acquiring better information on the structure of Xinka languages by documenting the language knowledge of present-day speakers. Research was, therefore, focussed on the compilation of lexical data and the morphological structure of Xinka, hoping for new insights on the languages’ affiliation and on the impact other languages have had on Xinka (e.g. loan words, structural borrowings) (see Campbell, 1972; Kaufman, 1977). Furthermore, searching for present-day Xinka speakers and working with these informants of rather advanced age was expected to clarify the linguistic situation of Xinka concerning it’s state of extinction.

Submitted 02/20/2001 by:
Frauke Sachse
University of Bonn
xinka@gmx.net

Planification and Implementation of the Project

The project was designed as a joint venture with the COPXIG (Consejo del Pueblo Xinka de Guatemala), the official representative body of the Xinka community of Santa Rosa. Having already started to collect language data themselves the COPXIG was very much interested in this kind of cooperation. The role of the COPXIG in the project was searching for speakers, establishing contact with potential informants, and accompanying me during fieldwork. In exchange for this help the COPXIG receives the analysed data and the results on the grammatical structure of the language that has been lost to them.

The cooperation with the COPXIG was thought to have several advantages: Due to the connections and ties the members of the COPXIG have with influential people in many rural communities of the research area, the search for informants would be rather efficient. As a certain unwillingness of the people to communicate about their language and culture had been observed, it was hoped that potential speakers would be less reluctant to deal and work with the COPXIG rather than with a foreign researcher and that this would assure the project of the people’s approval.

In the course of the project two periods of field research were planned. The first campaign was carried out in February/March, the second field stay was programmed at the beginning of the dry season in October/November. In the interim the documented material was revised and partially analysed. Questions and ideas that arose as first results were re-checked with the informants during the second period of fieldwork.

The COPXIG was meant to establish contact with potential informants prior to my arrival, but locating Xinka speakers turned out to be more difficult than first expected: The sad fact that two old women the COPXIG wanted me to work with had already died when I arrived – one in November 1999, the other in January 2000 – limited the number of disposed informants. Another speaker whom we did not meet in the first period of fieldwork and who was prepared to work with us during the second field campaign had died two weeks prior to my arrival. According to the COPXIG several dozen of speakers were known in Guazacapán and Chiquimulilla alone. Some of them had been contacted by COPXIG members and had arranged for interviews. When I started the interviews in February, however, I became suddenly aware of the fact that the majority of elders who are considered knowledgeable of the idioma de antes are only able to reproduce a very limited amount of words, and most of it quite deficiently.

The misconception of the term "hablante" affected the research project heavily and we shifted our focus and concentrated instead on the search for informants. Absolutely no speakers could be pinpointed in Chiquimulilla and the search was concentrated on the municipio of Guazacapán alone.

During fieldwork I stayed with members of the COPXIG in Chiquimulilla commuting to surrounding villages to search for and interview informants. The entire linguistic fieldwork, i.e. interviewing, was carried out in the neighbouring municipio of Guazacapán which can be easily reached by public transport. For additional excursions, which were arranged to extent the search for informants to more distant villages like St. María Ixhuatán, a privately owned pick-up (with driver) was rented.

The first period of field research (February/March):

During the first period of fieldwork we found and worked with eight informants. Only three of these can be regarded as speakers. A fourth one has contributed to lexicographic research by remembering lexical items from a wide range but has no command of speaking the language fluently and, therefore, cannot be considered a speaker. The remaining four informants have only knowledge of a few, single lexical items and do not show any language competence either.

As regards to the three fluent speakers, only two of them actually provided valuable information: An old woman of more than 90 years of age was already encountered in a state of mental disarray which made it impossible to interview her. Occupied with his agriculture and reluctant to share his language knowledge with foreigners, the second speaker, who is 72 years old, allowed us not more than one hour of interviewing per week. Only with the third informant, who is 88 years of age and in a threatening health condition, it was possible to work regularly and effectively.

The second period of field research (October/November):

In the second period of fieldwork a certain amount of time was dedicated to re-checking the forms and phrases documented during the first field stay with the two major informants. Since the COPXIG had encountered difficulties in locating more informants, even more time was invested in searching for further speakers. Although four more speakers could be located in the area of Guazacapán, none of them was willing or able to cooperate in the project. Further excursions into the surroundings did not provide us with more informations about speakers.

Research Results

Research has not been completed and will continue. In the following a description will be given as to the immediate results of the project rather than to specific aspects of morphology and the Xinka lexicon, since the latter are still preliminary.

The Linguistic Situation in the Xinka Area

The project on "Xinka lexicography and morphology" yields as one result information on the linguistic situation in Santa Rosa. As it is still possible to find some speakers we can correct the sad assertion of Xinka being totally extinct (see Fernandez 1938, Tujab 1987), although the official Guatemalan figures or the COPXIG’s estimates of about a hundred active speakers are to be considered an enormous exaggeration. This confusion has to be attributed to the misunderstanding of the term "hablante", which has lead to data stating an excessive numbers of speakers: thus an unpublished linguistic survey carried out by the Proyecto Lingüístico Francisco Marroquín in 1997 mentions as much as 9% of the total population of Guazacapán as speakers of Xinka (PLFM 1999: personal communication).

In fact, the linguistic situation is more discouraging: The few speakers we managed to find and work with in the course of this project were either very reluctant to share their knowledge, or they tried to commercialize it. The language knowledge of existing speakers has to be rated as extremely deficient and the documented data vary enormously in certain aspects. It needs to be mentioned that Xinka is not used for daily communication anymore. None of the speakers, including the individuals who did not agree to cooperate, has affirmed to practice conversation with any of the other speakers – each of them asserted not to know any further speakers. Moreover, all of them have stated to have learned their ability to talk "en lengua" from their grandparents or other elders, but not from their parents who already employed Spanish as the language of daily communication. From this we can conclude that there do not seem to be any speakers left who speak Xinka as their maternal language which supports the idea of Xinka being a dead rather than a threatened language.

The Data

Given the fact that all of my informants are already of a very advanced age and get tired of giving information very soon (i.e. generally after one hour), the interviews we managed to attain are fairly concise. Since there are no speakers who still actively communicate in Xinka among themselves, the methodological frame of documentation was limited to interviews and elicitation alone. Due to the linguistic situation and the degree of language competence of the speakers, it was neither possible to document the language in it’s cultural context, nor was it achieved to tape record longer coherent narrations. When explaining more complex topics the two informants we regarded as speakers usually switched back to Spanish.

In total, 40 hours of tape recorded interviews can be presented as the project’s substantial result. Regarding lexicography, a useful amount of lexical items was documented (see below). Lexical information was elicited in form of simple phrases. The informants also uttered entire sentences up to short prayers and very concise stories. In comparison to the data on Xinka that have been available for study up now (s.a.), the newly documentated material forms a valuable corpus for a linguistic analysis of Xinka morphology and morphosyntax which will exceed the information on the language structure accessible to date.

Xinka Lexicography

Since it would not suffice to add just another list of lexical items to the lexical data on Xinka that are already known, the study of the Xinka lexicon was aimed at comparing the newly documented data with the lexical entries of those word lists and vocabularies which have been available thus far (e.g. Calderón 1908; Maldonado de Matos ~1770; McQuown 1948; Schumann 1967). The lexical data documented in the course of the project do not exceed hitherto known data in number. However, the project data show that the information about certain words varies depending on the respective speaker, and that these forms may deviate from older data.

To present a rather simple example for the divergence, we may look at the varying lexical information in Xinka given for the spanish word "asiento" or "silla":

Informant 1	wapa
Informant 2	wapa
Informant 3	wapan

In this case, the lexical data show variation only in the last consonant which raises the question of whether wapa is a nominal word stem and the divergence in the last consonant can be attributed to morphological variation. Indeed, the suffix –n appears in the context of the first person singular possessive of inalienably possessed nouns (see below) and it is likely that the speaker referred directly to ’his seat’ by contesting with "mi silla" during the interview. However, the suffix - cannot be attributed to any other morphem identified thus far, but one could assume that it is the mispronounced suffix of the third person singular possessive of inalienably possessed nouns –j conveying the meaning "su silla". Although - has not been identified in the context of a possessive suffix, the project data suggest – at first glance – a morphological significance for the final consonant.

But if we compare the new data with the lexical entries for "silla" and "banco" in the vocabularies of Maldonado de Matos (~1770), McQuown (1948), and Schumann (1967)

Maldonado	guapa
McQuown	nuwapa
Schumann	wapár

this argumentation fails, since (1.) - appears repeatedly in final position and can therefore not be interpreted as a mistaken form of –j, (2.) the likely word stem wapa can be inflected with the prefix nu-, which is the first person singular possessive of alienably possessed nouns, and (3.) the suffix –r cannot be verified in any other context as it usually only appears in final position with loans (see Campbell 1972). Therefore, we can neither attribute a convincing morphological significance to the deviating final consonants, nor can we make a statement as to whether wapa is an alienably or inalienably possessed noun.

In the compilation of a Xinka lexicon we have to take into account that in general such divergences can be caused by several possible conditions. Divergence in the forms can as much be an indicator for regional (dialectal) variation as for a lack of reliability of the data, i.e. it can either be attributed to a declining language competence of the speakers and thus can be understood as the natural result of the language situation, or it can result from the documented data not being utterly representative. Therefore, a potential reconstruction of the Xinka lexem for "silla" requires a more profound analysis of the processes responsible for the deviations.

The present example reveals not only how much Xinka lexicography is dependent on morphological analysis and lexical reconstruction, but also shows the difficulty in the evaluation of the project data.

Xinka Morphology

Regarding Xinka morphosyntactic structure, the same methodological concerns that we pointed out for the reconstruction of the Xinka lexicon apply to the analysis of morphology. As a preliminary and by far not sufficient result we can hold the following features to be typologically relevant characteristics (at least) for the Xinka of Guazacapán:

Personal inflection of the verb is conditioned by the valence of the verb, i.e. whether it is an intransitive or a transitive verb, and by the tense/aspect of the action described. To mark the subject or agent of the verb, Xinka employs prefixes and suffixes, while the patient or object is represented by a free morpheme (pronoun) which follows the inflected verb (1.). The use of personal prefixes and suffixes depends basically on tense/aspect: In transitive verbs the agent is marked with a prefix to state a verbal action in the present/incompletive (2.), while suffixes are employed to mark the agent in the past/completive (1.).

(1.)	hin ka-ka nin neg hallar-2s 1s	no me hallaste
(2.)	an-pua tamatz’e’   1s-hacer    matate	yo hago matate

Intransitive verbs, on the other hand, mark the subject independently from tense/aspect by means of the free pronoun which is identical with the marker of the patient in a transitive phrase (3., 4.). The ergative pattern of grammatical relations resulting from this distribution, however, cannot be regarded as a significant feature of Xinka. If we compare the results from the project data with the information on the verbal system that can be reconstructed from the Maldonado grammar, we find intransitive verbs to be inflected with prefixes only - irrespective of tense/aspect (5., 6.).

(3.)	tiki    nin dormir  1s	yo duermo
(4.)	tiki-a      nin dormir-Past  1s	yo dormí
(5.)	an-màra 1s-descansar	yo descanso (Maldonado)
(6.)	an-màra 1s-descansar	yo descansé (Maldonado)

In transitive verbs the category of tense/aspect is thus marked indirectly with personal inflection. There is serveral degrees of "past" marked by temporal suffixes or particles following the inflected verb.

Concerning the morphology of nouns, nouns are inflected for the categories of possessive and plural. The plural is only inflected with animate, or more specifically human related nouns. Xinka is a head-marking language. To form the possessive, Xinka employs the same prefixes and suffixes used in the formation of the transitive verb. Suffixes mark nouns which are inalienably possessed (7.), while prefixes indicate possession in all other nouns (8.).

The project data show even more divergence regarding morphosyntactic structure than regarding lexical information. These divergences should not only be understood as signs of language death and limited reliability of the data, but do also reveal certain patterns which might be typologically relevant to language loss.

Dissemination of Research Results

The data collected in the project on "Xinka lexicography and morphology" form the basis of my doctoral research which is focussed on preparing an in-depth description of Xinka which includes all available language materials. The complete data and results of the linguistic analysis will, therefore, be disseminated in form of a PhD thesis.

Each of the two campaigns of field research was concluded with a workshop for the COPXIG in order to comply with the agreement of cooperation. The first workshop held in Guatemala City was dedicated to the discussion and decision of an official alphabet for the Xinkan languages. In order to be allowed financial support from the government’s Ministry of Education for the preparation of educational materials which could be used for teaching Xinka culture and "language" to the pupils of primary schools in Santa Rosa, the COPXIG needed to present recent linguistic research as well as an official alphabet that would be used in the preparation of the primer.

The alphabet has been based on the documented data as well as on the phoneme system and phonological rules of Xinka as issued by Campbell (1972:187) and will be published in Revalorización de la cultura xinka a través de la escuela – una guía documental que valora las culturas que conviven en el departamento de Santa Rosa (Dirección Departamental de Educación de Santa Rosa, 2001).

The focus of the second workshop held in Antigua was Xinka morphology and especially the formation of the verb. On behalf of the Ministry of Education, the COPXIG is currently preparing a primer for first year pupils containing a few basic terms in Xinka; some of the findings discussed during the second workshop will be used in this educational material. Independently from the preparation of the thesis, results on the grammatical structure of Xinka will be communicated to the COPXIG.

Concluding Remarks

In the course of this project it was achieved to document the language knowledge of some of the very few Xinka speakers still existing, who are already at very advanced age and might not be available for research in a very short time from now. Research will continue in March. It is planned to carry on re-checking data and searching for further informants. FAMSI will be provided with updates on any progress and results related to the project which is presented here. I wish to thank FAMSI, also on behalf of the COPXIG, for funding this project.

Sources Cited