I was reading about the death (or otherwise) of sexism and racism on Feministing and one of the comments mentioned a hardware store guy assuming she didn't know what a screwdriver was.
This got me thinking again about stereotypes. As is said, stereotypes have to come from somewhere. Well many of them anyway.
Where I live, there are 3 groups of drivers over-represented in the group that induces exasperation from me: white van drivers, green P plate drivers and asian drivers. This makes an asian, green P plater driving a white van the scariest thing on the road. Or not.
The problem is that human beings are spectacularly good pattern matchers. So good,in fact, that we often see ones that aren't there. Our ability to infer inductively is almost certainly a huge factor in our pre-historic survival and also continues to drive our search for knowledge now. We all work with the logic 1,2,3...many...all. Which works exceptionally well with poisonous red berries, tasty black berries and the laws of physics - not so well with other human beings. At least not that last step. 1,2,3...many is actually pretty valid, especially when based on culture rather than genetic race.
The "isms" come in when you make the leap from many to all, and our normal language doesn't differentiate well enough. When I say "white van drivers, green P platers and asians are bad drivers" I mean "many" in the area I live. I believe they are valid statements with those limitations, others may be able to prove to me that I have been selectively observing, but that is another matter. If I move from many to all, I make the conclusion about this specific driver being bad on account of fitting into one of those groups. That is a Bad Idea.
However, it is very difficult to tell people not to make these generalisations. We are strongly programmed to do it, and more importantly, it is an effective strategy. Assuming the observation is correct, I definitely benefit from being more alert when I see a white van. I have avoided several accidents because I am hypervigilant when I see a white van or a green P plate (less so with asian drivers, I know from observation that the correlation is not as strong as with the other two groups). Of course this is why P platers have to wear them, so that the rest of can compensate for them on the road. It is a necessary part of becoming a driver.
So how do we avoid the isms? I think we have to embrace our pattern matching and stereotyping behaviour, and then we can give them the analysis they deserve. Stop pretending that stereotypes have no use at all, and instead accept that they help, but really pull them apart and see their limitations. Show people how easily they can be misled by selective observation (you only notice the coincidence, not the non-coincidence) and even try to refine the stereotypes. Australians don't have kangaroos in their backyards, but we are lazy and alcoholic as well as easy going and tolerant*, at the stereotype level.
As a person who used to have to try to guess how technical the person at the other end of the telephone was based on their gender, and if I was lucky their job title, I can tell you that the poor hardware guy was between a rock and a hard place. You can easily cop as much flack for assuming contrary to the stereotype as with it. And saying "don't assume anything" doesn't help, because I had people get offended that I asked. I reckon up to date stereotypes with no stigma would be the best option, you can be upfront about matching or not matching the stereotype and we could all avoid a lot of awkwardness.
* We were tolerant pre-Howard, and I believe we will be again, especially if Alan Jones ever leaves the air waves. My mother and I were wondering why "Shooting Alan Jones" wasn't one of the options on the "If I had 30 days to live" list. :)