Skip to main content
SpringerLink
Log in

Paradigmatic structure in the tonal inflection of Amuzgo

  • Published:
Morphology Aims and scope Submit manuscript

Abstract

The tonal inflection of verbs of the Amuzgo language of San Pedro Amuzgos (Oto-Manguean, Mexico) displays a great degree of allomorphy. When faced with allomorphy of this sort, the inflectional class model often reveals an internal logic in a system, but in the case of Amuzgo organizing the inflection into tonal classes results instead in a system which is seemingly chaotic, and somewhat impractical for descriptive purposes. In order to make sense of the apparent chaos, in this paper I pursue an alternative view of the data based on characterizing verbs firstly according to their paradigmatic structure with regard to tonal inflection and then characterizing tonal exponents by way of default and implicative rules of exponence which allow us to comprehend the core of this inflectional system. Having identified this core, I then show how verbs relate to each other on a continuum of morphological complexity.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. The 1sg forms of the verbs in (e) and (f) end in an echo vowel with T3 (see Sect. 2.2).

  2. SP-Amuzgo tones are written with numbers following Smith-Stark and Tapia García (2002); the lower the number, the lower the tone (i.e. 1 represents the lowest level tone). When I refer to tones in the text, I will represent them as T1, T3, T5, etc. Two numbers on a single syllable indicate a contour tone (e.g. T53 represents a high-mid descending tone).

  3. My sample comprises only verbal lexemes with (different and distinct) monosyllabic stems. All the verbs in the sample have semantics that allow them to be inflected in all persons (i.e. they can all have a human subject).

  4. New verbs are introduced into the language by means of a light verb construction involving the verb -ts’aa ‘do/make’.

  5. For convenience, I subsume clusivity of 1pl under the umbrella of person/number.

  6. The enclitics I use for third person in Table 2 indicate a human subject, e.g. \({=}\tilde{e}^{5}\) {3sg.hum}. For the singular, we also have: \({=}\tilde{e}^{3}\) as an honorific for humans and deities; =oʔ5/ = ãʔ5 for an animal; and =ãʔ1 for an inanimate subject. These semantic distinctions are neutralized in the plural where there are only two plural enclitics: \({=}\tilde{e}^{3}\) and \({=}\tilde{a}^{34}\) {3pl}; the first is used in a neutral register, the second is honorific.

  7. Abbreviations: excl: exclusive; hum: human; impf: imperfect; incl: inclusive; nc: number class; pc: paradigmatic class; pl: plural; sg: singular; ta: tonal arrangement; tc: tonal class; tmc: tonal macro-class.

  8. An echo vowel is a mora with a specific, invariable, tone that is realized by a copy of the last vowel of the base with which the mora is associated; in verbs, the base for this mora is the stem. I have used angle brackets to represent both echo vowels and lowered vowels, because neither of them are clearly segmentable.

  9. For the singular, apart from \({=}hu^{5}\) for a human, we have: \({=}ho^{ n53}\) as an honorific for humans and deities; =oʔ5 for an animal; and = a n ʔ 1 for an inanimate subject. Again, the semantic distinctions of the singular are neutralized in the plural where there are only two plural enclitics: \({=}ho^{5}\) and \({=}a^{ n34}\) {3pl}; the first is used in a neutral register, the second is honorific.

  10. While making a clear distinction between affixes and clitics has no bearing on my proposal here, it is worth stating the basic arguments on which this distinction is based. Pronominal clitics of third person alternate (never co-occur) with overt subject NPs. As for 2pl, I have treated =oʔ3 as a verbal enclitic, and not an affix, because of its syntactic behaviour in verbal serializations involving two inflected verbs. To illustrate this, consider how the associated motion meaning “go and come back” is expressed in Table 3, which consists of the verb ‘go’ (exhibiting stem suppletion) plus its complement verb ‘return’, which occurs in the irrealis marked by \(n^{5}\mbox{-}\). Both verbs are inflected for person/number by internal changes, but note that =oʔ3 is absent from the stem of the matrix verb ‘go’, occurring only after the verb ‘return’. I take the fact that =oʔ3 occurs after the second verb as an indication that it is an enclitic that takes the verbal complex created in the serialization as its syntactic host.

  11. I interpret this lack of tonal changes as a consequence of the fact that inactive verbs are not subject to person/number inflection.

  12. Based on the data from four native speakers (the oldest of whom was 60 years of age in 2019 and the youngest 25), Hernández Hernández (2019) gives a very different picture of the tonal system of SP-Amuzgo. She argues that there are only seven tones instead of eight. She further postulates that there are five level tones (super high, extra high, high, mid, and low) and only two contour tones (mid-high and mid-low). This system is puzzlingly different from the one that has been consistently reported by Tapia García for years, on whose data this analysis is based. Given that Fermín Tapia García is more than 20 years older than Hernández Hernández’s oldest speaker, the disparity may well reflect intergenerational changes. Hernández Hernández’s analysis does not attempt to account for tonal correlations between the two systems, so it is not apparent to what extent the differences in categorization reflect diachronic changes nor to what extent such changes may have affected the state of the verbal inflection as I am describing it here. But meaningful correlations need to exist, because if speakers from different generations were not able to perceive them, it would make it impossible for them to understand each other. As far as I have been able to establish from Hernández Hernández’s data, her extra high tone corresponds to Tapia García’s T53, while her super high is rendered as T51 by Tapia García. In view of Hernández Hernández’s (2019) data and in want of more correlations of this type, for the time being the label SP-Amuzgo should be taken to represent the variety of SP-Amuzgo spoken by Fermín Tapia García and his generation.

  13. Kim (2016) proposes that Tapia García’s (1999) T51 could be better treated as a case of T51. I adopt Kim’s representation of this tone here.

  14. We might refer to this as tonal syncretism, but we could alternatively talk about this verb having invariant tone for all cells except 1pl.excl; this is what I argue for in Sect. 5.

  15. In SP-Amuzgo the morphological derivation of a causative verb from an intransitive verb is no longer productive. There are also transitive verbs in this group which lack a semantically transparent lexical correlate in an intransitive verb.

  16. The formatives of the \(ba^{5} +\) class and the baʔ1+ class also exhibit tonal changes in the singular forms.

  17. Kim (2016) argues convincingly that T51 and T1 are the surface outcomes of a morphophonological tone lowering rule that involves a change from T53 to T51 and from T3 to T1 when the glottal suffix for 2sg is attached to a stem that already ends in a glottal.

  18. I have used the online software at https://www.cs.uky.edu/~raphael/linguistics/analyze.html created by Raphael Finkel and Gregory Stump at the University of Kentucky as a computational tool to produce quantitative views of the internal structure of inflectional classes including entropy measures across cells. As far as my knowledge goes, the results output by this tool are not indicative of any sort of order in the system. Readers wishing to carry out the exercise themselves are advised to input the data in Table 7, replacing the numbers for each tone value with a specific letter (e.g. T53 > ‘a’, T51 > ‘b’, etc.).

  19. Such a system would raise important questions about learnability which I cannot address here, but which would doubtless constitute a most relevant topic for language acquisition research.

  20. As a causative verb, the verb -tz+kʔua53 ‘confound’ in Table 13 has a complex stem by means of two stem formatives. As we have seen in Tables 9 and 11 above, stem formatives in complex stems carry tone when syllabic, but such tones are fixed by their formative class. Inflectional tone appears on the lexical stem. In the verb -tz+kʔua53 ‘confound’, the tones we are concerned with are T53 and T51.

  21. Recall that Tables 5 and 6 show that we find all tone values in the forms of the 3sg and pl cells; this could be taken as independent evidence that the stems in these forms exhibit lexical tone.

  22. A label such as ‘A/I’ reflects, for example, a reduction of two tone classes ‘A’ and ‘I’, which had been proposed earlier on as two independent classes. Kim’s (2016) specific goal is to propose that for most instances of T51 in 2sg forms, one is dealing with a disguised T53, so her purpose is to reduce the inflectional complexity of the verbal system by equating many instances of T51 and T53 as representing the same inflectional exponent.

  23. The absence of such classes in my sample may be due to the fact that I am using different verbs, and it should bear no relevance for the current exercise.

  24. The four categories are meant as an idealized model to measure inflectional deviations in general. The model is meant to be superimposed on Amuzgo data here, but it could be equally extended to understand deviations in other similarly complex inflectional systems.

  25. When there is only one verb in a correlation that happens to not abide by the default, I take it to represent its own idiosyncratic pattern no matter what the percentage is, and I rank it as D2, accordingly.

  26. An anonymous reviewer commented that the proposed implicative rules of exponence for 1sg, covering just 12% of verbs on top of the 73% that already get the default tonal inflection, could be seen as a potentially disappointing result. While I still think that increasing ‘regularity’ by 12% is not at all negligible for a complex system like Amuzgo’s, in Sect. 7 I show that, when calculating the core of tonal inflection as a whole, the virtue of implicative rules adds 20% more verbs to the core.

  27. In a way, the different paradigmatic classes could be seen as having different degrees of morphological complexity specific to their types, but this is not an issue I deal with here. I consider that membership to a paradigmatic class is lexically specified (although causative verbs, for example, are more prone to belong to Class VII). This means that when it comes to inflecting a verb, speakers need to first know whether a verb is, for example, Class I or Class VIII. Once that information is dealt with, each verb should be judged as equal, each with its own idiosyncratic inflectional properties, so that for inflectional purposes using a default exponent to realize a given value for the types that require it would place such verbs at the same level as verbs with no inflectional tone.

  28. The ones attested are: {Ø/Ø/Ø}; {Ø/Ø/D0}; {Ø/Ø/D2}; {Ø/D0/Ø}; {Ø/D0/D0}; {Ø/D1/Ø}; {Ø/D1/D0}; {Ø/D2/Ø}; {Ø/D2/D0}; {Ø/D2/D2}; {D0/D0/Ø}; {D0/D0/D0}; {D1/Ø/Ø}; {D1/Ø/D2}; {D1/D0/Ø}; {D2/Ø/Ø}; {D2/Ø/D0}; {D2/Ø/D2}; {D2/D1/D2}; {D2/D2/Ø}; {D2/D2/D0}; and {D2/D2/D2}.

  29. It could be argued that this needn’t be the case, because verbs of Class VIII could just as well have abided by defaults. However, it remains a descriptive fact about SP-Amuzgo’s tonal inflection that verbs of this paradigmatic class are the ones that display the greatest number of idiosyncrasies.

  30. Together with the efforts of linguists like Yuni Kim, who make the data accessible to the larger community of linguists and language users.

References

  • Ackerman, F., & Malouf, R. (2013). Morphological organization: the low conditional entropy conjecture. Language, 89(3), 429–464.

    Article  Google Scholar 

  • Ackerman, F., Blevins, J., & Malouf, R. (2009). Parts and wholes: implicative patterns in inflectional paradigms. In J. Blevins & J. Blevins (Eds.), Analogy in grammar: form and acquisition (pp. 54–82). Cambridge: Cambridge University Press.

    Chapter  Google Scholar 

  • Ackerman, F., Stump, G. T., & Webelhuth, G. (2011). Lexicalism, periphrasis, and implicative morphology. In R. D. Borsley & K. Börjars (Eds.), Non-transformational syntax: formal and explicit models of grammar (pp. 325–358). Oxford: Blackwell.

    Chapter  Google Scholar 

  • Albright, A., & Fuss, E. (2012). Syncretism. In J. Trommer (Ed.), The morphology and phonology of exponence (pp. 236–288). Oxford: Oxford University Press.

    Chapter  Google Scholar 

  • Andrews, A. D. (1990). Unification and morphological blocking. Natural Language and Linguistic Theory, 8, 507–558.

    Article  Google Scholar 

  • Apostol Polanco, J. (2014). Clases flexivas verbales en el amuzgo de Xochistlahuaca, Guerrero. Unpublished MA thesis at CIESAS, Mexico.

  • Aronoff, M. (1994). Morphology by itself: stems and inflectional classes. Cambridge: MIT Press.

    Google Scholar 

  • Baerman, M. (2005). Directionality and (un)natural classes in syncretism. Language, 80, 807–824.

    Article  Google Scholar 

  • Baerman, M., Brown, D., & Corbett, G. C. (2005). The syntax-morphology interface: a study of syncretism. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Baerman, M., Brown, D., & Greville, G. C. (2009). Morphological complexity: a typological perspective. Paper read at the conference ‘How do we cite words in action: interdisciplinary approaches to understanding word processing and storage’, Pisa, 11-14 October.

  • Baerman, M., Brown, D., & Greville, G. C. (2015). Understanding and measuring morphological complexity: an introduction. In M. Baerman, D. Brown, & G. G. Corbett (Eds.), Understanding and measuring morphological complexity (pp. 3–11). Oxford: Oxford University Press.

    Chapter  Google Scholar 

  • Beniamine, S. (2018). Classifications flexionnelles: Étude quantitative des structures de paradigmes. Unpublished PhD Thesis at Paris Diderot University.

  • Beniamine, S., Bonami, O., & Sagot, B. (2017). Inferring inflection classes with description length. Journal of Language Modelling, 5(3), 465–525.

    Article  Google Scholar 

  • Bonami, O. (2012). Discovering implicative morphology. Paper read at Les Décembrettes 8: Colloque international de morphologie, 6-7 December. University of Bourdeaux.

  • Bonami, O. (2014). La structure fine des paradigmes de flexion. Études de morphologie descriptive, théorique et formelle. Mémoire d’habilitation à diriger des recherches, Tome I: document de synthèse. Paris Diderot University.

  • Bonami, O., & Beniamine, S. (2015). Implicative structure and joint predictiveness. In V. Pirelli, C. Marzi, & M. Ferro (Eds.), Word structure and word usage. Proceedings of the NetWordS final conference.

    Google Scholar 

  • Bonami, O., & Beniamine, S. (2016). Joint predictiveness in inflectional paradigms. Word Structure, 9(2), 156–192.

    Article  Google Scholar 

  • Boyé, G., & Schalchli, G. (2016). The status of paradigms. In A. Hippisley & G. Stump (Eds.), The Cambridge handbook of morphology (pp. 206–234). Cambridge: Cambridge University Press.

    Chapter  Google Scholar 

  • Brown, D. (2016). Defaults and overrides in morphological description. In A. Hippisley & G. Stump (Eds.), The Cambridge handbook of morphology (pp. 272–296). Cambridge: Cambridge University Press.

    Chapter  Google Scholar 

  • Brown, D., & Hippisley, A. (2012). Network morphology: a defaults based theory of word structure. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Buck, M. J. (2000). Gramática amuzga de San Pedro Amuzgos, Oaxaca. In C. Stewart & R. Stewart (Eds.), Diccionario amuzgo de San Pedro Amuzgos, Oaxaca (pp. 361–480). Mexico City: SIL Mexico.

    Google Scholar 

  • Buck, M. J. (2018). Gramática del amuzgo de Xochistlahuaca. Gramáticas de lenguas indígenas de México: Vol. 16. Mexico City: Instituto Lingüístico de Verano.

    Google Scholar 

  • Carstairs, A. (1983). Paradigm economy. Journal of Linguistics, 19, 115–128.

    Article  Google Scholar 

  • Corbett, G. G. (2015). Morphosyntactic complexity: a typology of lexical splits. Language, 91(1), 145–193.

    Article  Google Scholar 

  • Corbett, G. G., & Fraser, N. (1993). Network morphology: a DATR account of Russian nominal inflection. Journal of Linguistics, 29, 113–142.

    Article  Google Scholar 

  • De Jesús García, I. (2019). Fonología del amuzgo de Cochoapan, Guerrero. Unpublished MA dissertation in Indoamerican Linguistics. CIESAS, Mexico City.

  • DoBui, B. (2018). Grammaire de l’amuzgo de Xochistlahuaca, une langue otomangue orientale: Documentation d’une variété amuzgoane de langue en danger. Unpublished PhD Thesis at Paris-Sorbonne University.

  • Feist, T., Palancar, E. L., & Tapia, F. (2015). Oto-Manguean Inflectional Class Database: San Pedro Amuzgos Amuzgo. University of Surrey. https://doi.org/10.15126/SMG.28/1.04.

  • Finkel, R., & Stump, G. T. (2007). Principal parts and morphological typology. Morphology, 17, 39–75.

    Article  Google Scholar 

  • Finkel, R., & Stump, G. T. (2009). Principal parts and degrees of paradigmatic transparency. In J. Blevins & J. Blevins (Eds.), Analogy in grammar: form and acquisition (pp. 13–54). Cambridge: Cambridge University Press.

    Google Scholar 

  • Hernández Hernández, N. (2019). El sistema tonal en el amuzgo de San Pedro Amuzgos: interacción entre el tono de la base nominal y los clíticos. Unpublished MA dissertation in Indoamerican Linguistics. CIESAS, Mexico City.

  • Herrera Zendejas, E. (2014). Mapa fónico de las lenguas mexicanas: formas sonoras 1 y 2. Mexico City: El Colegio de México (Chap. 5).

    Google Scholar 

  • INEGI (2010). Censo de población y vivienda. Mexico: Instituto Nacional de Estadística y Geografía. https://www.inegi.org.mx/app/indicadores/?ag=20300#divFV6207019034.

    Google Scholar 

  • Kim, Y. (2016). Tonal overwriting and inflectional exponence in Amuzgo. In E. L. Palancar & J.-L. Léonard (Eds.), Tone and inflection: new facts and new perspectives (pp. 199–224). Berlin: Mouton de Gruyter.

    Google Scholar 

  • Kim, Y. (2019). A morphophonological parameter hierarchy for Amuzgo glottalization classes. In E. L. Palancar, T. Feist, & M. Baerman (Eds.), Amerindia: Vol. 41. Inflectional class complexity in the Oto-Manguean languages (pp. 247–278).

    Google Scholar 

  • Müller, G. (2007). Notes on paradigm economy. Morphology, 17(1), 1–38.

    Article  Google Scholar 

  • Palancar, E. L. (2016). A typology of tone and inflection: a view from the Oto-Manguean languages from Mexico. In E. L. Palancar & J.-L. Léonard (Eds.), Tone and inflection: new facts and new perspectives (pp. 109–140). Berlin: Mouton de Gruyter.

    Chapter  Google Scholar 

  • Palancar, E. L., & Feist, T. (2015). Agreeing with subjects in number: the rare split of Amuzgo inflection. Linguistic Typology, 19(3), 337–383.

    Article  Google Scholar 

  • Sagot, B. (2013). Comparing complexity measures. Paper read at computational approaches to morphological complexity, Paris, France.

  • Sagot, B., & Walther, G. (2011). Non-canonical inflection: data, formalisation and complexity measures. In C. Mahlow & M. Piotrowski (Eds.), Systems and frameworks in computational morphology (pp. 23–45). Zurich: Springer.

    Chapter  Google Scholar 

  • Sims, A., & Parker, J. (2016). How inflection classes work: on the informativity of implicative structure. Word Structure, 9(2), 215–239.

    Article  Google Scholar 

  • Smith-Stark, T. C., & Tapia García, F. (2002). El amuzgo como lengua activa. In P. Levy (Ed.), Del cora al maya yucateco: estudios lingüísticos sobre algunas lenguas mexicanas (pp. 81–129). Mexico City: Universidad Nacional Autónoma de México, Insituto de Investigaciones Filológicas.

    Google Scholar 

  • Stump, G. T. (2016). Inflectional paradigms: content and form at the syntax-morphology interface. Cambridge studies in linguistics: Vol. 149. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Stump, G. T., & Finkel, R. (2013). Morphological typology: from word to paradigm. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Stump, G. T., & Finkel, R. (2015). Contrasting modes of representation for inflectional systems: some implications for computing morphological complexity. In M. Baerman, D. Brown, & G. G. Corbett (Eds.), Understanding and measuring morphological complexity (pp. 119–140). Oxford: Oxford University Press.

    Google Scholar 

  • Tapia García, L. F. (1999). Tzon ’tzíkindyi jñò ndá: Tzjón Noà yo jñò tzko. Diccionario amuzgo-español, el amuzgo de San Pedro Amuzgos, Oaxaca. Mexico City: CIESAS & Plaza y Valdés.

    Google Scholar 

  • Walther, G. (2013). Sur la canonicité en morphologie: perspective empirique, formelle et computationnelle. Unpublished PhD Thesis at Paris Diderot University.

  • Wichmann, S. (2019). Power-laws and preferential attachment in linguistic morphology: evidence from meʔpá verb classes. In E. L. Palancar, T. Feist, & M. Baerman (Eds.), Amerindia: Vol. 41. Inflectional class complexity in the Oto-Manguean languages (pp. 19–38).

    Google Scholar 

  • Woodbury, A. (2019). Conjugational double-classification: The separate life cycles of prefix classes vs tone ablaut classes in aspect/mood inflection in the Chatino languages of Oaxaca, Mexico. In E. L. Palancar, T. Feist, & M. Baerman (Eds.), Inflectional class complexity in the Oto-Manguean languages: Vol. 41. Amerindia (pp. 75–120). .

    Google Scholar 

Download references

Acknowledgements

This work has been evolving thanks in part to the comments I received from numerous colleagues at different events where I presented my progress and I am grateful to all of them. My special thanks go to Olivier Bonami, Gilles Boyé and Géraldine Walther, who helped me to better understand the implications of my metrics, and to Yuni Kim, who helped me to understand the intricacies of the morphophonology of Amuzgo. I am immensely grateful to the two anonymous reviewers and to the editor in charge of my article in Morphology for all their invaluable comments that allowed me to improve this proposal, to Fermín Tapia García for sharing his precious Amuzgo data with all of us, and to Tim Feist for proof-reading my English on several occasions and for reading the final manuscript with an unparalleled degree of attention. All errors remain my own responsibility.

Author information

Authors and Affiliations

  1. SeDyL, CNRS, Paris-Villejuif, France

    Enrique L. Palancar

Authors
  1. Enrique L. Palancar
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Enrique L. Palancar.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A

Appendix B

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Palancar, E.L. Paradigmatic structure in the tonal inflection of Amuzgo. Morphology 31, 45–82 (2021). https://doi.org/10.1007/s11525-021-09373-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11525-021-09373-x

Keywords

Search

Navigation