Strong evidence that the roots of binding constraints are pragmatic from Cole et al. (2015)

Cole, Hermon, and Yanti’s (2015) new paper is an extremely important contribution that is likely to have a powerful impact on debates that focus on where grammatical constraints in languages come from. The authors compare Traditional Jambi Malay (TJM) with a dialect of Jambi Malay spoken in Jambi City (JCM). TJM is an example of a language in which the longer forms involve the addition of an intensifier or emphatic, which serves to indicate the pragmatically marked nature of the coreference (König & Siemund 2000; Levinson 2000). In particular, a longer pronominal form in TJM (which consists of the shorter form together with an emphatic marker) tends to be translatable as a reflexive (e.g., ‘himself’) that is coreferential with an antecedent in the same clause, while the shorter form tends to be translatable as a simple pronoun (e.g., ‘him’ ); but importantly, these interpretations are only tendencies and the alternative interpretations are also possible. For example, the authors provide the examples below, noting that the short form dio can be interpreted as either herself, or as him/her as in (1), and the long form, dio deweʔ, can lbe interpreted as either herself or him in (2).

(1) yanti pikir [arna cinto dio]
Yanti think [Arna love 3[short form] ]
‘Yanti thinks Arna loves Arna/Yanti/him/her.’ (Cole et al. 2015: 11, example 29)

(2) mamaʔ nik cinto dio deweʔ
mother Nick  love     3 intensifier [long form]
‘Nick’s mother loves herself/ (only) him.’ (Cole et al. 2015: 11, example 26)

Neighboring JCM, on the other hand, has grammaticalized these tendencies into the familiar binding constraints. That is, the long and short form anaphors are not ambiguous in JCM. Strikingly, speakers of the two dialects are unaware of the distinction. As the authors note, “What we claim is of special interest about Jambi anaphora is that Jambi anaphora may constitute a live snapshot of Binding Theoretic reflexives and pronouns in the process of grammaticalization as a distinct system of pronouns and reflexives. We will claim that the anaphoric forms in TJM are instances of a functionally- based system in the stage prior (perhaps immediately prior) to grammaticalization.” They argue, cogently, that if one is based on innate grammatical knowledge in the form of some sort of Universal Grammar, and the other is semantic/pragmatic, then it remains to be explained how speakers barely notice the difference.

The authors are well-respected experts on binding theory and have in the past defended a mainstream generative grammar approach to anaphora (Cole & Hermon 1998a,b; 2005). Thus the careful and nuanced fieldwork presented in this paper serves as an unusually serious challenge to the UG orthodoxy as it concerns constraints on anaphora, and is very much worthy of the field’s attention.


Cole, P., Hermon, G. and Yanti (2015). Grammar of Binding in the languages of the world: Innate or learned?. Cognition, 141, 138-160.

Cole, P., & Hermon, G. (1998a). Long distance reflexives in Singapore Malay: An apparent typological anomaly. Linguistic Typology 2(1), 57–98.

Cole, P., & Hermon, G. (1998b). VP ellipsis and Malay reflexives. In A.Z. Wyner (Ed.), IATL 5: The proceedings of the thirteenth annual conference of the Israeli Association for Theoretical Linguistics (pp. 39–54).

Cole, P., Hermon, G., & Huang, C.-T. J. (2006). Long-distance binding in Asian languages. In M. Everaert & H. V. Riemsdijk (eds.). The Blackwell companion to syntax.

König, E., & Siemund, P. (2000). The development of complex reflexives and intensifiers in English. Diachronica 17(1), 39–84.

Levinson, S. C. (2000). Presumptive meanings: The theory of generalized conversational implicature. Cambridge, MA: MIT Press

Cite this blog post
adelegoldberg (2015, July 13). Strong evidence that the roots of binding constraints are pragmatic from Cole et al. (2015). Diversity Linguistics Comment. Retrieved June 13, 2024, from

3 thoughts on “Strong evidence that the roots of binding constraints are pragmatic from Cole et al. (2015)

  1. Pingback: Is Universal Grammar ready for retirement? – Philosophy of Linguistics

  2. That’s great that we agree that languages are learned on the basis of the input and possibly only domain-general processes (as far as we know so far)! So then we’re all interested in human languages and how languages are learned (and processed). I guess I’m unclear why anyone then ever refers to “Universal *Grammar*”. I mean once we agree not to *assume* that universal (absolutes or tendencies) are about *grammar* rather than factors related to attention, memory (working and long term), meaningful categories (e.g., causer, undergoer), categorization, social cognition, why not retire the term?

  3. I think that the last time anybody seriously believed that something as specific as, say, `principles A and B of the binding theory are an innate part of UG´ was some time in the late eighties. I am not sure who or what the target of criticism is here? I am a minimalist syn-sem person. Everyone I know believes that there are no actual innate things of that level of specificity. Grammaticalized pronoun anaphoric systems emerge from the categories we are cognitively predisposed to attend to, plus e.g. drives to generalize, pressures brought to bear on the system by locality of processing and working memory. When grammaticalized systems emerge in languages, they tend to end up quite similar to each other in many cases. This will be due to innate developmental and representationa/categorizational l factors, some of which may be language specific who knows, but many of which will not, in the context of building a systematic open ended symbolic system (which brings its own design pressures). I am spelling this out because I do think that if linguists are going to have arguments about things, we should have arguments about points of substantive disagreement instead of talking about UG in exactly the same way as Chomsky used it in 1981, and which no one actually thinks.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.