I've been thinking about the concept of tokenism and why we find it problematic, given that virtually everyone who complains about tokenism thinks there is some good in having representation by those who are underrepresented in a particular sphere. What makes the difference between the cases where we find it unproblematic to try to get people more represented and those where we consider it tokenism? A few considerations come to mind:
The most obvious cases of tokenism are when someone just wants to appear forward-thinking and progressive by selecting people of an underrepresented group without really being concerned at all about the underlying ethical issues. If a college's admissions literature and website are littered with pictures of non-white students when such students are only 1% of the college's population, we might cry foul and wonder why they think they can pretend the school is more diverse than it is just to make themselves look good.
We should be careful here, of course. An institution might not be doing this just to look good. They might be thinking that portraying the student body in such an inaccurate manner will help attract students in those very groups, and they might have good motivations for wanting such a change. But it still seems wrong in such a case, even though it's not merely to generate a false view of the school to garner a better reputation. The dishonesty in the portrayal seems like a kind of tokenism. We might select people out of underrepresented groups to make it look like our institution is better than it is, or we might do so for purer motives, namely to try to make it better than it is, but either way the dishonesty of portraying it that way seems to fall under our concept of tokenism.
So is it basically a kind of dishonesty that makes something a case of tokenism? I don't think it's as simply as that. Consider a TV show that has their one token black in a mainly white cast. That black character might display all the stereotypes of black characters, in which case it might be criticized for stereotyping. On the other hand, it might display no such stereotypes, in which case it might be accused of sanitizing the character to make them more white-friendly. You might then think the critics are unfair. You can't win, no matter what you do? Actually, I don't think that's the problem. I think the no-win situation is set up because you don't have enough black characters both (a) on TV in general and (b) on the show in question. Even having two black characters, one of each type, is better than having one token black who fits either mold. The solution seems to me to be to have a diversity of black characters, some of whom display some stereotypical characteristics but who nonetheless are real characters, some of whom display fewer stereotypical characteristics but who nonetheless are real characters. What saves the day for a show that might be accused of tokenism is to have a variety of real characters showing a diversity of real-life traits from real people. Portray them so that the audience cares about them. Portray real inner conflict, hard choices, and so on. Make your characters of color as interesting and developed as all the other characters, and have enough of them across the variety of TV shows that we create, and you're a lot less susceptible to be accused of tokenism.
What does that suggest about what tokenism is? It's not just plain honesty, because there's plenty of room in there for trying to have as many characters as you can that don't fit well with the actual percentages of which black people have which traits. You don't need to have your black characters have children out of wedlock at exactly the rate that happens among black people in real life. You don't need to have them like hip-hop at the same percentages. You don't need to have them attending college or being incarcerated at the same rates. You need some level of honesty there to the point where you're not ignoring realities in society too much, but you can steer stereotypes by having lots of counter-stereotypical characters, and of course a lot of what you can do will be affected by what kind of show it is. Game of Thrones won't have anyone listening to hip-hop or being incarcerated in American prisons. The core problem seems to be, rather, that tokenism doesn't care about the people or characters enough to do much more than trot them out for the appearance. A character on a superhero show who is a token black might be stereotypical or might not be, but we won't care about the character very much, because the person isn't fleshed out very much. Tokens in college promotional literature are there for the appearance, and in a sense so are the undeveloped characters who are there just to have representation.
Now how does this relate to the use of tokenism-language in the context of affirmative action? Some conservative critics of affirmative action see it as harmful to those it's intended to help, partly because it isn't concerned with their success in college but just wants to have diversity as an element of its student body. It isn't concerned with finding students who will be as prepared to succeed, because it's more interested in showing off its diverse composition. In that sense, it would be like the case of admissions literature. But this isn't the only way to conceive of affirmative action. Even with the diversity rationale, one can be engaged with affirmative action policies in order to promote diversity, where there's a further goal for that diversity, and that can be to promote further racial justice for the sake of those who would be benefited by their being such racial policies. That motivation strikes me as not tokenist, even though the actions would seem to have roughly the same outcome with either motivation. So tokenism is not just about consequences. It's about why you engage in the actions you engage in to begin with.
I can imagine a student group at a college, maybe a religious or political group, that wants to seek more diversity. They might undertake efforts to promote their group among groups that are not well represented in their group at present. They might change their methods or approach to be more culturally acceptable to such groups. They might change their focus to include things people in those groups would care about. Is this tokenism? It seems to me that the answer depends on why they're doing it. If they want the people they're targeting merely because they want it to be true that their group is more diverse, I think it is tokenism. If they want them to be present because they think they themselves will be enriched by the experience, and the newcomers will benefit as well, then it seems to me not to be tokenism.
The same goes for inclusion in an academic conference or in high governmental positions. If a president seriously would like cabinet or judicial nominees to come from underrepresented groups, as both the last two presidents have (at least at times) shown concern for, then the crucial question is why. Is it to make the party or the administration look good, or is it out of a genuine concern for having diversity in that sphere of government? If I tried to put a conference together, and someone pointed out that none of the invited speakers were women, I might try to remedy that. Am I remedying it because I committed a faux pas and am embarrassed, or am I doing it because I think we all benefit by having more women presenting at philosophy conferences and because I think we have a systematic implicit bias against thinking first of women when thinking of the movers and shakers in a discipline like philosophy? The former might be tokenism. The latter seems not to be. But the actions are exactly the same.
This is a first attempt to think through this carefully. A number of questions remain in my mind. Are there any examples of what seems like tokenism that doesn't fit the kind of thing I'm saying here? Are there any examples that don't seem like tokenism that do have some of the characteristics I've been trying to identify tokenism with? It may well be that there's more complexity to what we typically call tokenism, and it might be that I'll need to figure out what to do when there are disagreements over what counts as tokenism. There's also the possible complication of whether tokenism is always wrong. Are there cases that we would call tokenism where we wouldn't find it morally problematic, or is it a term like 'racism' or 'murder' where we'd only use the term if we thought there was something problematic going on?