José Antonio Flores Farfán

Introduction: On typological change in Mesoamerica

De Gruyter | Published online: August 4, 2018

1 Background

The Sprachbund known as Mesoamerica comprises a wealth of mega-linguistic diversity. Spread throughout a vast territory, Mesoamerica not only occupies what today we call Mexico. It ranges in its northern frontier from Arid America (roughly the regions linked to desert areas to the North of Mexico), passing through the Mexican central plateau (in and around where Mexico City is today), to reach Central America, around the limits of El Salvador and Honduras (see Map 1). Historically, previous to the Spanish invasion in the sixteenth century, it is estimated that the Mesoamerican territory encompassed at least 100 and up to 300 languages (see Suárez 1983; versus the upper figure close to what is reckoned in contemporary sources: e.g. INALI, cf. Garza Cuarón and Lastra 1991). Yet the Mesoamerican linguistic landscape is of course very different today. In passing, it is interesting to notice that these figures are related to the complex question of what is understood as a language, where it ends, and where a dialect or another language starts. A good example is provided thinking of the 365 “variants” listed today by the INALI (Instituto Nacional de Lenguas Indígenas: The National Institute of Indigenous Languages), which politically are conceived as “languages”, against the 68 linguistic “clusters” recognized by the same official institution. This could be rephrased as 68 languages, with 365 dialects. Actually, this is a very multifaceted issue. It involves several sociolinguistic constraints, which cannot be reduced to a single criterion, such as the conventional approach to define a language versus a dialect, intelligibility. In practice this constitutes a heated debate, more of a political and ideological one. For instance, some linguists consider Nahuatl, one of the major languages of Mesoamerica, as several languages, while other actors, especially high educated Nahuatl speakers, conceive it as a single language. This will also become clear in the present volume, especially although not exclusively considering the contributions on Nahuatl and Mazatec (see infra).

Map 1: Limits of Mesoamerica in Prehispanic times (Wikimedia 2011. Map of Mesoamerican civilizations cultural area.

Map 1:

Limits of Mesoamerica in Prehispanic times (Wikimedia 2011. Map of Mesoamerican civilizations cultural area.

Historically, there were several linguae francae in prehispanic Mexico. The most general language in pre-colonial Mesoamerica was Nahuatl, the language of the so-called Aztecs or Mexicas. They became the most dominant group in Mesoamerica about 300 years before Spanish invasion, themselves also invaders before the process of European colonization. The Spaniards recovered the status of Nahuatl lingua franca for their own purposes of evangelization and control of the New Spain, even expanding Nahuatl to regions where it was not previously present. Other wide-spread tongues were probably at least Maya Yucatec and Zapotec, in their own regions (the Yucatec Peninsula and Oaxaca, respectively).

Quantitatively speaking, Mesoamerica hosts the most numerous populations of indigenous people in the Americas today, with approximately 15 to 20 million people of indigenous origin, depending on the source one relies on. This is in any case the highest total population speaking an indigenous tongue in absolute terms in the American Continent; mostly in Mexico, which represents over 10% of the total Mexican population, amounting up to almost 130 million people. In terms of qualitative typological diversity, Mesoamerica includes several different types of languages (ranging from languages with very complex, diverse morpho-phonological templates, tonal languages, as well as a continuum of polysynthetic/agglutinative versus analytic ones, with several possible combinations, among other characteristic traits, such as ergativity in e.g. Mayan languages).

According to official figures from INALI, the only institute of its type in Latin America, linguistic diversity in Mexico encompasses 11 linguistic families (see Map 2). This refers to languages that exist in Mexico as “national” languages, the nomenclature utilized by the Mexican state to refer to all these endangered tongues – an ideology derived from the nationalistic chauvinism of the Mexican state. Four of these languages are so called isolates (Cmique Itoom or Seri, Ikoots or Huave, Purepecha or Tarascan and Tequistlatec or Chontal). Yet the 68 languages reckoned by INALI with their variants, reaching according to the same source 365, do not cover the enormous linguistic diversity of the country. For instance, there is a Creole language (Afro-Seminole) in Mexico, not considered by INALI, together with other threatened languages, such as Romani. The only language that was, ironically, just recently added by INALI to its catalog is Spanish, the only official, fully standardized major language of the country. In practice, INALI understands “variant” as language, and that is the reason of the 365 figure. Most of the indigenous population is bilingual (or multilingual, including an unknown number of speakers of several indigenous languages, together with English, which is starting to play a role in the Mexican linguistic ecology, due to high rates of migration to the USA, a topic to my knowledge still uninvestigated). Mexico, and generally Mesoamerica, still has diverse rates of monolinguals in indigenous languages. Depending on the region one refers to, monolinguals can also reach relatively high figures, especially when compared to other countries such as the USA (e.g. about 30% percent in Tzeltal and Tzotzil, major languages from Chiapas, a Mexican state bordering Guatemala). In overall terms monolinguals comprise at least 1% of the total indigenous Mexican population, reflecting a very complex linguistic ecology involving both indexes of vitality of some Mesoamerican languages, even when they are all endangered, several reaching the brink of extinction, often times in pervasive, extreme conditions of isolation and marginality.

Map 2: Language families of Mexico according to INALI (Instituto Nacional de Lenguas Indígenas: National Institute of Indigenous Languages, Compare to Ethnologue’s map.

Map 2:

Language families of Mexico according to INALI (Instituto Nacional de Lenguas Indígenas: National Institute of Indigenous Languages, Compare to Ethnologue’s map.

In contrast to INALI, Ethnologue reckons several other languages belonging to other linguistic families, with a total of 14 (including what they label languages of immigration: Basque, Chinese, English, Japanese, Q’anjob’al; plus a sign language, see Map 3), as well as the already mentioned Creole spoken by the Mascogos (English Muskoguees) in the Mexican state of Coahuila. Speaking in general of Mexico, Ethnologue states:

The number of individual languages listed for Mexico is 292. Of these, 287 are living and 5 are extinct. Of the living languages, 282 are indigenous and 5 are non-indigenous. Furthermore, 1 is institutional, 87 are developing, 76 are vigorous, 90 are in trouble, and 33 are dying. (

Map 3: Ethnologue’s general map of Mexico’s language families (

Map 3:

Ethnologue’s general map of Mexico’s language families (

Even when INALI speaks of 365 variants, which as suggested in practice are considered languages in the general Law of Linguistic Rights which accompanied the creation of the institute over a decade ago, Ethnologue’s overt figure of 292 languages indicates that the latter exacerbates linguistic differences, speaking of languages instead of what often times, from other points of view (e.g. speakers themselves), are considered modalities of the same language or “dialects” – a term which in Mexico in general has a negative connotation among laymen, and is derogative in an ideology penetrating indigenous communities themselves, something that would speak in favor of INALI’s nomenclature. INALI does not recognize extinct languages, and does not integrate, as Ethnologue does, based on Fishman’s (1991) well known GIDS, scales of endangerment/vitality, even when providing a category and developing a project on “lenguas en riesgo” – languages at risk. In practice recall that all Mesoamerican languages are endangered, although to varying and very different degrees.

Other clear divergences with INALI’s Mexican linguistic profile also includes the status of Cmique Itoom (Seri), which Ethnologue suggests belonging to the Hokan family, whereas for INALI, as stated, it is an isolate. Even when both sources provide an idea of the complex linguistic landscape of the country, neither of these classifications is entirely precise, totally accurate or complete, speaking of the complexity of sociolinguistic profiles and classification criteria, at times contradicting each other. For instance, Kickapoo, the only language from the Algonquian stock in Mexico, a neighboring language of the (Mascogo) Afro Seminol community, is not included as a language of immigration by Ethnologue, whereas for INALI Kickapoo is a national language. In turn, while the Afro-Seminole is not considered by INALI, Kickapoo is. Even when both groups arrived escaping from slavery in the USA at roughly the same time, in the nineteenth century, which should have prompted Ethnologue to consider them as languages of immigration (as it does with other indigenous Guatemalan languages e.g. Mam, Q’anjob’al or the Germanic language Plautdietsch), it does not. This produces an arbitrary, unsystematic listing. There are other missing languages in INALI’s list too. These include, among others, Plautdietsch, Romani, etc., as well as the Yucatec Maya sign language (which Ethnologue recognizes). There is at least another sign language, unrecognized by either source: Tzotzil sign language. Other very different tongues such as Arab, Hebrew, Korean, or even Yiddish are not taken into account in either catalog. Awaiting for a systematic research project, all this invites a whole program on the historical as well as the contemporary typology of all these and specifically in Mesoamerican languages in the face of multiple contacts, especially although not exclusively with Spanish and even among indigenous languages, outstanding samples of which are included in the publication at hand.

2 The present volume

The set of papers included in this volume offers novel approaches to a series of issues related to typological change in Mesoamerica. It includes contributions on different Mesoamerican linguistic families, e.g. Otomanguean: Mazatec (Léonard and Fulcrand); several different languages belonging to diverse stocks considered by Robbers and Hober (e.g. Maya Yucatec versus Nahuatl) and Uto-Aztecan: Nahuatl (Stolz; Olko et al.).

The volume is a contribution to the understanding of typological change both as processes of contact induced change (Robbers and Hober; Olko et al.) and also as internally, independent driven phenomena (Stolz; Léonard and Fulcrand), based on historical as well as contemporary sources.

A case in point of the novelty of the papers included is Stolz’s paper “On classifiers and their absence in Classical and Colonial Nahuatl”. Among several issues, it invites revisiting the approach to what have traditionally been called “classifiers” in Nahuatl or Mexicano, as speakers generally call the language in different regions scattered around Mexico, and which encompasses over a million speakers. After an extensive review of the literature on the topic, Stolz argues for a different approach, the optionality hypothesis, together with both a qualitative and a basic quantitative analysis of key colonial sources, and concludes with the Labovian cumulative principle, reminding us that the more we know about a specific topic, the more we need to investigate it. Stolz’s paper thoughtfully provokes an array of different quests, including descriptive (e.g. should we talk about “classifiers”, “mensuratives”, or both?), empirical (e.g. further research is needed on the pragmatics of Nahuatl numerals both historically and in Modern varieties of the language), and of course theoretically, posing queries such as:

  1. What is the status, if at all, of optionality to explain numerals in Nahuatl?

  2. What is the role of variability in understanding the typology of languages, especially Nahuatl?

  3. Which is the best approach to engage the complexity of quantification in general and specifically of numerals with respect to other facets of the language, especially pragmatics?

Of course all these questions can be applied to other languages with classifiers and even other tongues, making Stolz’s paper a model for similar research on numerals and quantifiers in general. Incidentally, Stolz’s piece even suggests recasting the colonial record of Nahuatl ethno-mathematics for contemporary revitalization purposes.

Based on a large-scale typological cross-linguistic approach, Robbers and Hober’s work, “Verb-framed spatial deixis in Mesoamerican languages and the increasing complexity of source constructions via Spanish de”, touches upon several Mesoamerican languages. Encompassing 4 different linguistic families (Uto-Aztecan, Maya, Mixe-Zoque, and even an Araucanian isolate, Mapundungun), the authors examine spatial deixis both in interrogative and declarative forms. They review abundant corpora, including Bible translations and other contemporary sources, which highlight different typological characteristics of typological marking. These include basic and multiple markings, location, goal and source, polysynthetic versus analytic traits, etc., as well as other meaningful oppositions, at times triggered through a specific and apparently trivial particle, the Spanish preposition de. The paper adequately oscillates between typological continua of very different languages (e.g. Maya Yucatec versus Nahuatl, a more analytic versus a more polysynthetic language) exploring different types of possible contact-induced changes triggered by de. Surprisingly, albeit de is presented as potentially enabling an outstanding impact in typological change, the authors find that spatial relations are zero-coded in most of their samples and that directional meaning is encoded in verb meaning solely in the Mesoamerican languages discussed. A change to overt marking may happen via de, as in Sierra Popoluca in their sample. The role of de is documented by other scholars such as Hill and Hill (2004), which, among other facts, has prompted Nahuatl to change from a polysynthetic to an analytic language, making highly Hispanized varieties of contemporary Nahuatl more Spanish-like as compared to old or monolingual Nahuatl (cf. Flores Farfán 2012).

Regarding the typological traits in their discussion, Robbers and Hober emphasize the critical issue of indifferent or zero-marking of spatial relations in several of these languages. For instance, the authors consider the role that other (e.g. syntactic) resources play to deploy spatial deixis in Mesoamerican languages, especially when compared to other languages of the world, which in contrast overtly and even pervasively mark spatial deixis morphologically.

This as well as all the contributions of the present volume invites a series of reflections on critical methodological points. For instance, regarding the construction of representative data to understand typological change, the following issues are at stake:

  1. What is the role of elicitation and other research instruments in the construction of corpora versus so called spontaneous speech?

  2. How do we construct robust samples to reach representative conclusions taking into account the wide range of data variability due to a wealth of sociolinguistic factors in our research?

  3. How does our own identity as researchers trigger significant differences in the construction of corpora?

The impact of the written versus the oral code in constructing our samples, as well as the weight of different genres in their complex conformation and understanding, is also a case in point. For instance, it is remarkable to witness that in Bible translations there is almost no presence of Spanish particles or in general of loanwords, specifically prepositions. This makes it very challenging to grasp the role of transfers such as de in (induced) typological change, a point that becomes apparent as a clinching argument at the end of Robbers and Hober’s paper too. Moreover, the paper suggests that purism is often an ever-present issue in the construction of our data, inhibiting the appearance of such particles, and in this sense, paradoxically after all, an effect of contact. This is especially clear in the case of the Bible sample of Guerrero Nahuatl analyzed. De, as well as several other Spanish prepositions, has played an outstanding role in typologically induced change in languages such as Nahuatl, driving the language from a polysynthetic language to a more analytic one (Spanish, see Flores Farfán 2012). In passing, it is worthwhile noticing that Mesoamerican languages, maybe more than any other languages of the Americas, have integrated a wide set of particles as de in their contemporary repertories.

In summary, Robbers and Hober stress the fact that spatial marking is still zero-coded in the Mesoamerican languages considered for their study. Yet even when directional markings are morphologically inherent to the verb, the Spanish de virtually represents the possibility of overt marking of spatial deixis, as shown by Sierra Popoluca in their study, as well as in other scholars’ findings, especially Nahuatl (e.g. Hill and Hill 2004; Flores Farfán 2012). More research on the topic is required, especially recasting oral, extemporaneous speech corpora in these and several other Mesoamerican languages.

The language spoken in the ritual mushroom ceremonies by the famous María Sabina in Huautla de Jiménez, is the subject of Léonard and Fulcrand’s study, “Inflectional class shifts in the Mazatec diasystem: Innovation, contact and metatypes”. Elaborating on a number of authors, among others Haspelmath’s graphs for ordering typological traits, which they incorporate in their dialectological study, Leonard and Fulcrand make a significant contribution to Mazatec linguistics and sociolinguistics. As compared to other highly studied languages such as Nahuatl and Yucatec Maya, the authors tackle an under-documented language, expanding our knowledge of the social dialectology of Mazatec and highlighting the inextricable link of the social and linguistic realms.

Going beyond the phonological level, which has traditionally been the canonical way of constructing isoglosses, the authors reach conclusions not proposed before for this and probably any other Mesoamerican language. This includes unraveling processes of Mazatec compartmentalization and diversification, in the face of Mazatec unification and dialectological leveling, specifically in morphological verbal expressions. This provocative paper invites revisiting dialectology with typology in mind, and vice versa, with very useful concepts such as (sub) conflation, together with other diagnostic tools, such as developing modules to understand types of change stemming from the dynamics of Mazatec sociolinguistics. As a result, the authors bring about connections between dialectological differences linked to more prestige varieties, comparing rural versus urban settings, such as the multifaceted contacts between different Puebla and Oaxaca Mazatec regions. A wealth of very interesting findings linked to different linguistic taxonomies, at the same time present phenomena which deserve further investigation, such as the alluded impact of several variables in corpora construction. For instance, what is the role of idiolects in representing dialectological variability? This suggests an entire research program for this and other languages, as the role of Spanish or resistance to it in Mazatec social dialectology.

According to Ethnologue, Mazatec (“deer people”, which, as in many other cases, was a name imposed by Aztec invaders) received a series of different names, suggesting at least nine different languages. In contrast, Léonard and Fulcrand articulate a holistic multifaceted original way of studying this language, speaking of levels and continua of intelligibility, rather than different languages, a matter of heated debate, not only in Mazatec linguistics, as has been alluded to in the background of the present introduction. Based on a diasystemic ecolinguistic framework, which invites connecting dialectology to typology and noticeably vice versa, the authors study Mazatec inflectional complex verb morphology giving place to different verbal taxonomies. All this allows the authors to work on a unique model that sheds new light on the language’s diversification in terms of a variety of understudied dialects, emphasizing the contact situation within Mazatec itself. By turning their research into a study of dialects in contact and the interesting ways they converge and diverge, independently of the role of Spanish in Mazatec variability, this study also suggests the role of the dynamics of sociolinguistic accommodation in typological research, not limiting itself to the customary approach in dialectological studies, intelligibility.

Léonard and Fulcrand’s paper even invites issues of applied linguistics, such as the usefulness of these types of studies in the development of educational materials.

Recasting the well-established tradition of previous research by Nahuatl scholars on historical materials and direct fieldwork, Olko and her coauthors Borges and Sullivan provide a systematic, rich and comprehensive review of Nahuatl contact history in their paper “Convergence as the driving force of typological change in Nahuatl”. Developing a systematic in depth, long-term study of Nahuatl contact history, the authors concentrate on the reconstruction of several contact-induced changes triggered by Spanish. These series of contact induced changes in colonial as well as contemporary dialects include:

  1. The transformation of the grammatical category of animacy, which is apparently becoming less and less productive and probably obsolescent or at least reanalyzed and/or reconceptualized

  2. The emergence of prepositions as part of the lexicon, stemming from relational nouns, together with the blurring of distinctions between comitative and instrumental forms

  3. Typological changes in existential predicative possession

  4. The role of Spanish syntactic structure on the typology of Nahuatl, producing a relatively free word order in Nahuatl, a pattern mirroring Spanish, at least to a certain extent

  5. The role of Spanish syntactic structure on the drift of the typology of Nahuatl from a polysynthetic to an analytic language in highly Hispanized varieties

The history of Nahuatl from the very first evidence of contact with Spanish until today has been the subject of several studies (e.g. Lockhart 1992; Hill and Hill 1986). This paper is not only one of the most complete reviews of Nahuatl-Spanish in this respect. Additionally, the ground-breaking contribution of Olko et al. includes reviewing typological innovations in less studied, peripheral (e.g. Huastec Nahuatl) as well as more central varieties (e.g. Tlaxcala), which opens up understanding geographical and social variability due to Nahuatl-Spanish contact. Tracing the continuing paths of several morphosyntactic features that produce typological change, the authors convincingly argue that crucial innovations in Nahuatl in the colonial period are mutually borrowed from Spanish and/or motivated by internal drifts, which progressively become prevalent due to intense contact with Spanish structures. As an overall conclusion Olko et al. demonstrate that together these two processes have driven typological change in Nahuatl, a fact that confirms the crucial understanding of typological change as multiple convergent processes and probably the best, most realistic approach to engage in contact linguistics.

In sum, the present volume offers an array of findings regarding typological change in different, selected Mesoamerican languages. It ranges from the study of spatial relationships in a variety of sources and linguistic families, to in depth studies of specific aspects such as Nahuatl classifiers or the state of the art of Nahuatl contact history, as well as approaching typology and dialectology in understudied languages (Mazatec). I hope that such challenging contributions will invite more research on these and related topics, especially by young, emergent scholars.

A word of gratitude is due to all the people who made this volume possible. To all the peer reviewers of the papers, who with their comments increased the already high quality of the present contributions, Alexandra Aikenvald, Mario Chávez Peón, Francisco Barriga Puente, and Jean Leo Léonard. For their patience and always-willing collaborative attitude, the authors of the present volume are also gratefully acknowledged. I especially wish to thank Thomas Stolz for his generous support not only in forming this volume, but for his continuing inspiration to study these and other fascinating topics over the years. Many thanks to Cornelia Stroh too, who has been crucial in the final production of the present volume. Last but not least, many thanks to Elsie Rockwell and Jane H. Hill, who helped me with the final version of the present introduction.


Flores Farfán, José Antonio. 2012. Cuatreros Somos y Toindioma Hablamos. (first edition 1999). México: CIESAS. Search in Google Scholar

Fishman, Joshua. 1991. Reversing language shift. Clevedon: Multilingual Matters. Search in Google Scholar

Garza Cuarón, Beatriz & Yolanda Lastra. 1991. Endangered languages in Mexico. In Robert Robins & Henry Uhlenbeck (eds.), Endangered languages, 93–134. Oxford & New York: Oxford University Press. Search in Google Scholar

Hill, Jane H. & Ken Hill. 1986. Speaking Mexicano. Tucson: University of Arizona Press. Search in Google Scholar

Hill, Jane H. & Ken Hill. 2004. Word order type change and the penetration of de in Modern Nahuatl. STUF/Sprachtypologie und Universalienforschung 57(1). 23–48. (Special issue „Nahuatl in typological perspective“, edited by José Antonio Flores Farfán). Search in Google Scholar

Lockhart, James. 1992. The Nahuas. Stanford: Stanford University Press. Search in Google Scholar

Suárez, Jorge A. 1983. The Mesoamerican Indian languages. Cambridge: Cambridge University Press. Search in Google Scholar

Websites Search in Google Scholar Search in Google Scholar Search in Google Scholar Search in Google Scholar

Published Online: 2018-08-04
Published in Print: 2018-08-28

© 2018 Walter de Gruyter GmbH, Berlin/Boston