Anti-racism is work that I believe to be incredibly important and almost vital for our society. That said, it's a confusing pile of uncomfortable, guilt, accusations, blame, and competing political philosophies. And the biggest flaw of all is uncertainty in action, which leads to "awareness-raising" and very, very little else.
Awareness by itself does nothing, and awareness that self-propagates is a vicious little thing. And that's how AR work functions when used among those who clearly benefit from privilege. Telling a white person who makes 100k or greater a year that they benefit from an unfair privilege because of their race doesn't go over well, and it only gets harder the more distant a person becomes from obvious signs of wealth and privilege. The resistance that has built around Affirmative Action for decades has a lot to do with those on the low socio-economic end being really upset that another group of low socio-economic folk get an advantage they don't. As they see it, poor's poor, and race doesn't appear to be a big enough factor to warrant special treatment. (As an aside, this is part of why the individualistic and self-made impulses are so strong among the people who would most benefit from welfare state policies - aid has historically been seen as unfair, so it makes more sense to do away with unfair aid instead of expanding aid to everyone).
So awareness by itself is flawed, and leaves either guilt or resistance. Awareness is supposed to be uncomfortable, and guilt/resistance are both reactions to uncomfortableness that cannot be processed. But there's a point to awareness, or it wouldn't be a consistent agent of change. Awareness, if done properly, changes behavior. So then awareness has done a good job changing language, changing some behaviors, and getting rid of the most overt signs of racism. But Anti-Racist activism focuses a lot on institutional racism, which isn't something that can be changed except by either institutional reform or widespread changes public opinion. Institutions are incredibly hard to change by guilting individual members, and structural factors that hinge on race are easily dismissed because they are obscured by the many factors that correlate with race (wealth, education level) and which can be explained as a lack of drive or initiative. Racism, without overt signs, is really hard to change on an individual basis, and almost impossible on an institutional basis.
AR work is incredibly tricky, and when it fails, it fails big. "Whiteness" is not a comfortable topic, not a casual state, and it is such a weird mixing of legacy (early New Englanders, slave importers) and assimilation (Swedes, Irish, Italians et al becoming white after a generation of immigrant experience) that "whiteness" has a wide range of internal interpretations. The "assimilated into privilege" makes the formula much more complicated than "our ancestors are guilt for your suffering", and it's the assimilation that makes all races in the US White versus (racename).
So what's to be done? This dichotomy of white vs other leads to notions like colorblindness, where everyone just doesn't have race be a part of their identity, or at least not an obvious huge defining part of their identity, like it has for many white Americans. Sure, when asked I'll say my race is Welsh, British, and Scottish, but those terms are almost meaningless for me in the US because the unstated "white" identity is so overriding. The classic line goes "much like fish in water, white people in American society are oblivious to the privilege they exist in all the time". The problem with the colorblind solution is that it assumes everyone else will assimilate into that same privilege, and that the only thing stopping them is a race that is part of their identity. Here Anti-Racism doesn't offer a solution so much as destroy a flawed one.
Another common approach is to look at the flawed socio-economics in this nation, and assume that, given the strong correlation of race to class, the problems can all be dealt with by programs that ignore race and just look at economic status. I'm somewhat partial to this plan, and it's informed my attitudes towards changing affirmative action (go meritocracy! don't make the law vulnerable to angry white men!), but that's not a perfect solution. This, too, ignores race and legacies of racial injustice for a classist slant - it would rob affirmative action of the notion that it corrects past wrongs and make it about more equal distribution of wealth, which changes the whole thing. And it treats race as class, which it isn't; it correlates, but that is all. Dynamics of race are different than dynamics of wealth, and while both are unjust and need to be changed, it ignores race to treat it as class, and it confounds things because in the US notions of class are linked to personal performance, and poverty is often seen as an individual failing. While it may address similar ends, race as class isn't the way to go.
Now we're back at square zero, it seems. To affect change in people, AR work has to make them uncomfortable enough to want to radically change their behavior for good. When that fails, people become reactionaries and resistant and are less open to notions of racism as endemic of systems; they'll abstain from meaningful work against organizations with flawed practices because they don't want to confront the guilt of being racist, or of being blind to a system that is racist. But that's if being made uncomfortable fails; should being uncomfortable work, people will now find themselves with a drive to change things (I know I did!), and with very few means of driving that change. If you spread awareness, you risk making a whole bunch of people guilty and resistant or guilty and agitated by how little there is that they can do. Or people could try and engineer individual ways out like colorblindness, which doesn't work and is offensive (in that whole "denying an aspect of humanity" sort of way). People may turn to class politics and understandings, but that is only part of the racial dynamic. When it fails, Anti-Racism work leaves a lot of people bitter and uncomfortable. When it succeeds, it can turn people into agents of change but it doesn't provide many valuable outlets for that energy, and it may burn people out before critical mass to change institutions (by having anti-racists in sufficient number running them) can work. The picture, as it can undoubtedly be painted, is bleak.
There's hope and there's some solace in all of this, though. Firstly, Anti-Racism is a young discipline and movement, which traces its origins back to the early 1990s (as far as I can tell). One of the benefits of spreading awareness rapidly since then is that, while total numbers are small, there are plenty of anti-racist activists working to figure out what to do with the knowledge that anti-racism has given them. This allows anti-racism to be a tool used in other debates, adding it to the side of human dignity. An example: knowledge of drug usage trends + knowledge of disparity in drug sentencing (500x the penalty for black drug crack versus the penalty for white drug cocaine) allows one to look at a flawed system and say, in addition to previous flaws, that it is racist in execution and should be changed. That's a powerful asset. The solace in all of this is that anti-racist work is moral work, it's valuable work, and it wouldn't be done if there weren't a pressing need for it. It's a flawed system currently, but as a moral imperative it is solid. It's why I keep up with AR work despite the criticisms. We'll hit upon the right solution eventually, and until then we can be gadflies who add minds to the collective brainpower trying to engineer a solution. To those skeptical, you have every right to be so, but take solace in the fact that we'll get their. The drive is too strong and the stakes are too high for us to not make it work.