It is ironic when proponents of diversity don't support diversity of opinions, hypotheses or data when they are contrary to their own beliefs or perspectives.
I am guilty of this too. For example, a paper was recently published in the prestigious Journal of Personality and Social Psychology that purported to empirically demonstrate the existence of Psi (psychic phenomena). The research, and the author Daryl Bem, were immediately attacked, and some argued that the paper should never have been published. However, I agree with a colleague who noted that "nobody gains by censoring dialogue." I must admit that my first reaction to Bem's study on Psi was that it was probably wrong, and that it should not have been published (I'm not a fan of Psi). However, it helped me to identify my own biases of immediately classifying research on Psi as false pseudoscience. Several critiques of Bem's work were published, along with Bem's rebuttal. Reading these made me re-think my perspectives. Even if it turns out that Bem's research was flawed, better to have explored all aspects of the issue than to have censored Bem's paper because to some (including me) it was "obviously wrong" and distasteful.
Kanazawa's blog would likely not be as controversial if he simply included more qualifiers in his posts ("it could be..." "one hypothesis..." "this needs further research..."). Most scientists don't need many such qualifiers when we read scientific articles because they are a given to us (we tend to be a pretty skeptical bunch), and I read his blog with that perspective. Laypersons, however, may indeed need many such qualifiers to help them distinguish what are tentative hypotheses or initial data analyses from well established scientific facts. And, as noted by Robert Kurzban in this post, some have taken their disdain for one post, or one blogger, to tarnish an entire discipline. (Also see the post by Scott Barry Kaufman and the post by Gad Saad.)
One point that has yet to be noted -- Kanazawa's post may have helped to start a dialogue that ultimately may be particularly valuable to black women. First, he was the messenger of bad news (let's not shoot him), but, news about data that (if correct) needs to be heard. Social psychologists have demonstrated the huge impact physical attractiveness has on many aspects of peoples' lives. If black women suffer further discrimination because, for whatever reason, they are still generally seen as less attractive, the first step to dealing with this inequity is to acknowledge it.
Had his post been censored from the outset (and assuming the effect is real), we may not have seen this problem -- we may have simply assumed that it was a non-issue. So Kanazawa's post alerts us to a issue of particular concern to a group of people that have already experienced a great deal of discrimination. If there are current cultural factors that are further exacerbating this, our consciousness needs to be raised, and these issues need to be confronted if cultural biases and perceptions are to be challenged, and, hopefully, changed over time.
See my more recent post on this topic: The Two Professions that Get to Tell the Truth: Scientists and Humorists
Exposing and opposing censorship: backfire dynamics in freedom-of-speech struggles
Black Women are Not (Rated) Less Attractive!: Our Independent Analysis of the Add Health Dataset by fellow Psych Today blogger Dr. Scott Barry Kaufman
Lewis, M. N. (2011). Who is the fairest of them all? Race, attractiveness and skin color sexual dimorphism. Personality and Individual Differences, 50(2), 159-162.
Here is a link, and another, to some additional relevant data from the dating site Okcupid.com.
Kanazawa's post -- "If the truth offends, it's our job to offend. Scientists' only responsibility is the truth."
Additional PT posts about censorship: Gad Saad's post, Douglas Kenrick's first and second posts, and George Ford's post.