Why I had to quit the John Edwards campaign | Salon News Even before Donohue stepped in, various right-wing bloggers were obsessed with my gender and sexuality. As I noted at the time of my resignation, the majority of the hate mail I was receiving was from men, and almost all the e-mails made note of my gender or suggested that I would be a more pleasant woman if I wasn't so "angry." Bluntly put, I find it hard to believe that many men would end up being denounced on TV for using words like "fuck" or "cunt" on their blog and expect to receive piles of e-mail offering an opportunity to suck the sender's dick.

A Hard Right Punch - washingtonpost.com “They’ve attacked my husband relentlessly. There’s a strong sexist strain among my liberal critics, who think it isn’t possible I could have gotten anywhere without my Svengali husband, or some white man, embedding ideas in my head.” …

“Particularly when you’re a minority conservative,” she says, “you get a lot of ugly, hysterical, unhinged attacks, because you’re challenging so many liberal myths about what people of color should think.” [quoting Michelle Malkin]

It may not be too outlandish to say that I’m I’m obsessed with the idea of the ‘surface of discourse’ (title of a book by Michael Hoey). But encountering these two exchanges my obsession seems justified. Both women describe a situation that is very common. Because of their controversial public stances on contentious issues, they receive not just criticism but also abuse that references their gender. This is fairly common (people from any group with a marked feature are exposed to similar reactions) as is the next step.They both reflect that this kind abuse stems from their attacker’s broader stance and point of view that is opposed to theirs. But this can hardly be the case (as much as my liberal self would like to believe the opposite) - or at least on the case to the extent claimed. Obviously, abusive communications come from both camps even from the one that is, at least in part, explicitly opposed to gender- or race-based slurs. Linguistically or psychologically, this is not particularly puzzling. An angry person is likely to reach for the most derogatory linguistic device possible and for representatives of disenfranchised groups usually stem from negative stereotypes. The speaker is merely using the strongest possible terms for emphasis. Emphasis in language can sometimes take very surprising turns. My favorite example is the English tense system. The present simple, despite its name, is used to express repeated actions (“He often reads books.”) and the primary function of the present continuous (progressive) is to describe actions happening at the moment of speech (The water’s boiling over.) However, to express repeated action with a negative emphasis, the continuous present is your best choice (“He’s always coming round here and asking stupid questions.”) as is the present simple to emphasize ongoing action (for instance in sports commentary: “He passes, he shoots, he scooooores.”)

Is it possible that racial and gender-based epithet are used the same way for emphasis? Perhaps, it is the rules of ‘civilized’ conversation that stops them from coming out rather than a natural instinct of the ‘civilized’ person not to use them. Women, for instance, often use derogatory language about other women that draws on male-constructed stereotypes. The ‘Uncle Tom’ insult in the African community is not dissimilar.

Would it then be possible to say that a person can seriously use racist or sexist language and not ‘really’ be sexist or racist? Michael Richards or Jade Goody would certainly like to think so. That may indeed be one of the problems with racism. Its really pernicious forms (assymetric perception, incremental and cumulative discrimination, race-based frames) can and do exist independent of the racial slurs often associated with them. So it is conceivable that the degree of a person’s political inclination cannot be reliably measured by the language they use in emphatic contexts.

However, when these expressions are publicly negotiated (hypostesized) their underlying frames are profiled and linked directly to a supposed deeper level of a person’s psyche (one for which there are ample folk but few good expert theories). To return to Michael Hoey. We can reconstruct much of a text’s cohesion and even coherence based purely on surface features (he looked simply at repetition) but what does this newly-discovered unity tell us about the intellectual and conceptual unity that we suppose is what the surface was generated from?

Two more concepts might be helpful here: 1. logistics and transactional costs: to get your text/discourse from A to B certain things have to happen (such as time has to pass and energy has to be expended). And the creation of cohesive harmony is one of them. And often the conceptual devolves from the logistical. 2. Construction/cognitive grammar’s destruction of levels (and dynamic hypostasis): Instead of language being generated from underlying meanings, both meanings and form are simply two sides of the coin of constructions of different degrees of complexity and schematicity. In some constructions (or more accurately in the process of integration/blending), conceptual meaning is the primary integrator and others it is the formal meaning. (And usually it is some combination of the two.) So, in the case of an insult being uttered, it is possible to view the formal meaning as primary and propositional meaning as secondary. However, in the process of the insult being reported, the propositional meaning becomes much more prominent and it is impossible not to conclude that the given speaker was expressing a racist/sexist/etc. sentiment.

To conclude, discourse analysis (whether done by experts or naturally during the process of frame negotiation [or psychoanalysis]) is a wonderful tool for discovering trends in populations of texts but extremely unreliable for determining the conceptual underpinning of an individual text, particularly if it is intended for use as a tool for the discovery of ‘hidden meanings’.