Stereotypes are ways of making extremely primitive and simple differentiations. Differentiations of gender, race, class, social status - so ordinary social life is very much built upon a whole repertoire of stereotypes we carry around. And those are immediately laminated onto people, and it isn't just visual.

Objectivism is basically the same thing as faith-based science or for that matter faith-based foreign policy, where you start out with the assumption "We are good, they are evil," or "We know what is good and right and we know what is wrong," so all questions are settled in advance by a set of ideological prejudices.

Obviously invidious and prejudicial stereotypes need to be deconstructed and overcome, but it's not that they can be destroyed. I think that would be an illusion to think that we can somehow get rid of these basic search templates that allow us to sort out our social lives and to sort out the material world as well.

I think its a fundamental feature of images that they move from one medium to another. And this has become hyper-evident in our time with the computer, which is a kind of master-medium also and allows us to transfer data of all kinds from one platform to another, turning sounds into sights or language into image. The computer has made something that is very old evident in a new way.

Stereotypes, I want to say, have to be thought of not just as these invidious, bad things that we could get rid of, but as images that we cannot get rid of, that we have to live with.

Distinction between species and specimen is very much like the distinction between images and actual pictures, or, you know, objects that have a definite material identity. The classifications, the categories, the stereotypes, and the images are on one side, and the material pictures, statues, texts, and so forth are on the other.

The great moment I think in human consciousness is when you realize that the object in front of you is perhaps not nameable or is new, it does not fit a stereotype, and so you need to reconfigure your whole structure of knowledge to account for it.