In lieu of an abstract, here is a brief excerpt of the content:

  • Letters to Language
  • Cilene Rodrigues

Language accepts letters from readers that briefly and succinctly respond to or comment upon either material published previously in the journal or issues deemed of importance to the field. The editor reserves the right to edit letters as needed. Brief replies from relevant parties are included as warranted.

Response to a book review

April 5, 2020

To the Editor:

In their review of the book Recursion across domains (ed. by Luiz Amaral, Marcus Maia, Andrew Nevins, and Tom Roeper, Cambridge University Press, 2018), Daniel Everett and Edward Gibson (E&G; Language 95(4).777–90, 2019) concluded that the chapters devoted to Pirahã do not meet ‘high scientific standards’ (p. 780). Their review is unusual in criticizing not only the argumentation and claims of these chapters, but also the very competence of the authors to conduct their research, as well as the competence of a local assistant thanked by the authors, mistakenly described as the ‘primary Pirahã consultant’ (p. 781).

I coauthored two of the chapters criticized in their review, Chs. 6 and 15. For space reasons, I focus on their criticism of the first of those papers. There is much to be said about the rest of their review, but I believe that the failings of their response to Ch. 6 satisfactorily exemplify issues that pervade the review. I am writing this communication not because their review was negative, but because it contains misrepresentations and falsehoods, which should be corrected.

I first briefly summarize the evidence and argument that Ch. 6 presents in favor of structural embedding. I then consider E&G’s critique, ending with a brief response to their comments concerning our competence, the competence of our sources, and their criticism of our data.


Word order and control: Our argument for syntactic embedding was straightforward, based on fieldwork data that showed obligatory control, a phenomenon found internal to complex sentences in the world’s languages, and not found across discourse.

The canonical word order of Pirahã is SOV (Everett 1983:201, ex. 3; A lingua Pirahã e a teoria da sintaxe, Universidade Estadual de Campinas dissertation). However, in constructions that one might take to show a clause embedded as a complement of a higher verb, Pirahã generally places that clause after the verb, yielding a structure that one might analyze as SVO (Everett 2005:629, ex. 24; ‘Cultural constraints on grammar and cognition in Pirahã: Another look at the design features of human language’, Current Anthropology 46(4).621–46). Everett 2005, however, took this as evidence that such structures do not actually involve clausal complementation, but paratactic conjunction, the putative clausal object actually constituting an independent sentence.

We presented the following arguments against E’s (2005) claim:

  1. a. The Pirahã counterpart of sentences like ‘I want to study’ shows the phenomenon of obligatory control (OC). When the subject of ‘study’ is unpronounced, it must be understood as controlled by the subject of ‘want’. Our consultant permitted no other interpretation.

  2. b. In such structures, both SOV and SVO are acceptable (Everett 2005:117, ex. 13, 14). The first of these orders is clearly not amenable to reanalysis as paratactic conjunction.

  3. c. Main-clause material may follow the complement clause in control structures with SVO order. A temporal adverb following a complement clause in an SVO OC structure can only modify the main clause (Everett 2005: 119, ex. 17) (possibly because the embedded clause is too small to support ad verbial modification). If the putative complement clause were actually an independent sentence, we would not expect that it could be followed by an adverb modifying an earlier independent sentence.

  4. d. SOV order is not restricted to environments of control. The desiderative verb sogabagai may also take a full clausal complement with an overt subject. In this case, SOV order is restricted; only the embedded predicate can precede the desiderative verb. The subject of the embedded clause follows the higher verb, resulting in a subject-predicate split. Once again, we would not expect an independent sentence to be followed by the subject of the sentence that preceded it.


E&G’s critique: According to E&G, Rodrigues et al. ‘missed some crucial...


Additional Information

Print ISSN
pp. 221-223
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.