On Linguistics as a Science and Linguists as Scientists

(Randy Allen Harris, The Linguistics Wars (1993))

Towards the end of ch. 9, while discussing what it was about the generative semantics enterprise that caused it to fall apart in the mid- to late-70s, Harris says, "The real damage was done by the closely related traits that Lakoff calls honesty and data-love." (236)

On "honesty," Harris notes that "the generativists were clearly explicit about their shortcomings in a way very few scientific collectives are." (236)  He contrasts the generative semanticists' tendency to be almost confessional in their shortcomings with Chomsky's rhetorical ability to admit the contingency of his proposals while projecting confidence that they (or some modification of them in the future) will turn out right.  Thus, "Chomsky has the impressive rhetorical talent of offering ideas which are at once tentative and fully endorsed, of appearing to take the if out of his arguments while nevertheless keeping it safely around." (237)  Harris uses John Robert Ross as an example of GS' tendency:

Ross's arguments are . . . multilayered, tortuous, and subtle, very sensitive to fluctuations in the data. They contain dozens of threads, looping around one another in the main text and annotated with lengthy, contorted discussions, some of which offer counter-examples, some of which offer alternative analyses, virtually none of which offer any support for the annotated point. They consist of long catalogs of subarguments—some strong, some weak, the weak ones always painstakingly flagged as such—interspersed with declarations of mystification and awe. . . . If honest is the appropriate descriptor for Ross's style, and it does seem right, it is painfully honest. The effect is confessional, in a way that perhaps makes one empathetic to the difficulties of his program, but hardly kindles interest in joining it. The effect is also cacophonous, with argument countering argument until it is all but impossible to hear a clear line of thought rising above the noise. The following assessment is telling: "It must be seriously open to doubt whether there is a coherent point of view to communicate." (237-8)

The argument is that the honesty of GS linguists about the contingency of their work, about its possible  flaws, about the myriad open questions their research left, was one of its fatal characteristics.  "Naturally," Harris is saying to us, "linguists who were looking for a direction to take in their research chose Chomsky/interpretive semantics over GS, at least in part because Chomsky/IS provided more than a mere possibility of open questions and recalcitrant data."

Ending this argument with a flourish a few paragraphs later, Harris poses this astonishing rhetorical question: "What could a working linguist do with a paper full of facts whose raison d'etre was that there was no conceivable explanation for them in current theory?" (238) 

I don't know, maybe try to work out either a conceivable explanation in the current theory, or a more thorough explanation for why the current theory can't explain the data, or sketch some ideas on what a theory that could explain the data would look like, or. . .?  

Instead, the suggestion is that linguists don't have any use for all that recalcitrant data that undermines their theories and muddies up the whole enterprise of theorizing. And that, to me, is astonishing, if true, not least because linguistics has pretensions of being a science.