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

  • Policing the content of linguistic examples*
  • Paul M. Postal

1. Background

Inherent in the linguistic enterprise is the choice and citation of example sentences in various natural languages. While the purpose is to illustrate and/or support certain claims, any cited example inevitably has a content which in most cases is independent of the reasons for its citation. For instance, a study of English that clauses might exemplify that they are barred from prepositional phrases, schematically:

(1) *X talked P (= about/of/ . . .) that S.

A real example requires choice of specific instantiations for the nominal variable X and the clausal variable S. Such decisions are irrelevant to the point at issue; so no linguistic ground, other than trivial ones (examples should be short enough to be grasped), constrains an author’s choice for such instantiations. Despite this, the Linguistic Society of America (LSA) in its LSA guidelines for nonsexist usage1 has seen fit to issue a code of suggestions which was intended to, and to the extent that it is followed, does, constrain authors’ and speakers’ freedom to instantiate variables like X and S in 1. The code made explicit that example sentence regulation represented a key domain where it was meant to function.

The existence of this code has been defended and justified in an article in this journal (Macaulay & Brice 1997), a work that, moreover, complains about its historical evolution which, in these authors’ view, has wrongly weakened it. Bergvall (1996:436) claims to have been ‘disturbed’ to find there had been controversy over the guidelines. These brief remarks argue for an opposed position and, if they contribute even slightly to an understanding of why there should be such controversy, I will have wasted neither my time nor that of the editors.

Adoption of the LSA guidelines for nonsexist usage was a nontrivial mistake which should be undone and the society should officially incorporate a resolution to never institute anything resembling it. At least four reasons justify rejecting such codes: (i) they represent an intrusion of parochial social and political goals in a society explicitly not devoted to social or political outcomes; (ii) to adopt a code that embodies one social goal but ignores multitudes of others no less (un)worthy is arbitrary and unfair; (iii) such codes represent a step, albeit limited, toward censorship and constraints on freedom of speech or at least the appearance of such, inappropriate in the United States; and (iv) it is impossible to specify explicit principles to accomplish the putative goals of codes like the current LSA one. I take up these points successively, on occasion in connection with related issues. [End Page 182]

2. Social and political goals

The explicit goal of the LSA, repeated in every issue of Language, is to further the scientific study of natural language; furthering social or political agendas is not mentioned. So there is no justification for an LSA code embodying the social goal of eliminating sexist language any more than for one of eliminating the use of fur, or of preventing abortion. Multiple social and political issues have some link to language, but that in no way justifies politicization of the society to take official stances on them. Saying this expresses no opinion about the merit or lack thereof of any social or political goal but merely demarcates the LSA as one social space where their intrusion is inappropriate. One notes that a characteristic of totalitarian societies is the politicization of everything.

I suggest that the LSA recognize the following as the only appropriately codifiable criteria for determining the airing and publication of linguistic examples in the United States:


    a. Validity: Do the examples support the point they are intended to?

    b. Legality: Do the examples violate any law, national, state, or local?

Beyond 2, it is none of the society’s affair, nor that of its editors in their official capacities, nor that of meeting referees, chairpersons, and so forth, to in any way police the examples chosen by authors and presenters. That job should be left to individual linguists, whose examples should have whatever content they choose regardless of how offensive others may find them. While one wants...


Additional Information

Print ISSN
pp. 182-188
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.