spacer spacer spacer meaningless logo

Trying to make necessary and sufficient conditions for gaslighting out of what we're given in "Allies Behaving Badly".

The summary (not claimed to be complete) is: "In short, gaslighting is when a hearer tells a speaker that the speaker’s claim isn’t that serious, or they’re overreacting, or they’re being too sensitive, or they’re not interpreting events properly."

My worry, personally, is that McKinnon seems to have a picture of what gaslighting is that applies clearly to the cases she's thinking about, but that her picture doesn't hand us necessary and sufficient conditions on a plate, and that that will turn out to be a problem if the word gets used in other types of scenarios.


An attempt:

  • Some testimony happens by someone we'll call "the speaker".

  • The speaker independently (?) suffers an identity prejudice at the hands of a putative gaslighter ("the credibility deficit ... would have to be due to an identity prejudice", p.3).

  • The putative gaslighter mentions a harm "or injustice" (interpreted how broadly?) they've received (or can it be that someone else has received, say due to being similar to the speaker in some way?).

  • The gaslighter says something negative about the speaker's testimony. Can this be anything negative? One case is saying that the speaker is "not interpeting events properly". What if it's epistemically negative but something the speaker really wanted to be told? I imagine that in the cases McKinnon has in mind whether the speaker wants to be told negative things is irrelevant, but what about in question time in seminars? What about a lawyer advising a client? And surely there are many other cases in which someone wants to be told when they've misinterpreted something, and in which it's also in their interests to be told. And surely that is sometimes a good thing, even if the speaker is suffering an identity prejudice. Lawyers shouldn't honestly advise their clients?


How to stop the definition being too broad?

(Main example: a lawyer advising a client about the client's testimony.)

Presumably the term is meant to be always derogatory, so we want to rule out cases in which it's good to question someone's testimony. How to do that? Add "excluding cases in which the denial of testimony is a good thing"?!

Is the truth of what the gaslighter says relevant? Or their epistemic warrant for saying it? I'm guessing it's not meant to be.

How about the gaslighter's intent? I don't THINK McKinnon wants it to be relevant, but from p.5 onwards they start to use Polhaus Jr's characterisation of it as "willful hermeneutic ignorance".

I don't know the answers to these questions, but maybe if we incorporate harm in a Haslangerish way that'll be good enough.


How to stop the definition being too narrow?

(My main example: someone criticising a spouse's testimony with bad intent but with no reliance on their spouse's identity weaknesses - e.g., the speaker might be a rich upper-class straight cis male.)

"Gaslighting" IS (perhaps unfortunately) used in a way that has nothing to do with prejudice against a class of people. For example, the OED glosses "gaslight" as "manipulate (someone) by psychological means into doubting their own sanity". And I'm sure this is a common usage in discussions of relationships, and the doubted person doesn't have to be female or part of any other marginalised group. The OED gives examples which support this, including Starsky and Hutch!

I think this is a big problem with McKinnon's use of the word, unless we say that it has a separate (technical?) meaning in ... in what? Trans* theory? Feminist theory? Social theory? But if we make the field in which it has a technical meaning too broad, doesn't it become implausible that it actually has a technical meaning in that field?

Maybe that's the solution: "gaslight" has a technical meaning in social theory. If no, this seems to be very different from what we've tried in other cases of tricky definitions. We haven't said that "gender" has a technical meaning in social theory. And I think we haven't said that for good reasons, because we want "gender" to be useable in political action. If a word has a technical meaning and a popular meaning, using it in political action is going to be super tricky, because the sociolects will overlap: groups of people will be using both meanings at each other. No? I actually think there must be ways round this problem, but what I want to know is whether we have to work out what those ways are, or whether, on the other hand, I'm thinking about the necessary and sufficient conditions all wrong.

Jason, 2017