In the Grey Matter section of the Sunday Review in the New York Times, Cornell Professors Wendy M. Williams and Stephen J. Ceci published an article entitled “Charles Murray’s ‘Provocative’ Talk.” In it, they described a small ad hoc study that they conducted to test whether the words of Charles Murray are objectively offensive and thus deserving of the level of resistance to his lecture at Middlebury College (see here).
In their study, the authors took a transcript of Murray’s actual talk and sent it without attribution to 70 college professors with a request to rate the words on a 9 point scale from very conservative to very liberal. They found that although “American college professors are overwhelmingly liberal,” those surveyed found Murray’s words to be “middle of the road” with an average score of about 5. Williams and Ceci interpret this finding as indicating that the protest over Murray’s invitation to speak was objectively ill-informed and unjustified.
This argument is deeply and fundamentally flawed. We often see similar tricks played when someone reads an excerpt from the Constitution or Mein Kampf and asks for an opinion about it – before the gotcha reveal when they identify the authorship.
One major study flaw is the premise that words stand alone. Context matters and the meaning and intent of words can only be fully assessed with due consideration of the person making the statement. Authorship is an essential part of that greater context. If PT Barnum claimed he had a Yeti in his house, I would have received it with tremendous skepticism. If Carl Sagan made the exact same claim, I would have been very excited about the potential of an important new anthropological discovery.
The reality is that Charles Murray has a long history of promoting what many consider to be highly destructive public policy research and analysis that has undermined valuable social programs and has attacked and divided us along gender and racial differences. For example, his statement that “We believe that human happiness requires freedom and that freedom requires limited government,” may sound perfectly reasonable to 70 of our professional contacts if unattributed. Coming from a known liberal speaker, this could be meant to affirm that we should not be forced to live in an overly-policed state. However, coming from Charles Murray it is clear that his intent is to promote the dismantling of social assistance programs. The same statement might mean something even more extreme if David Duke had said it.
Based on the work of Williams and Ceci one might argue that we should remove all bias in approving speakers by using a blinded, unidentified process in which presenters are approved or rejected based solely on the text of their planned presentation. That would be extremely foolish. The reality is that the larger views and history of any speaker plays an essential role in how we should interpret their statements. Reasonable but isolated statements can conceal a larger and very different agenda that is only apparent if we know the source.
I have no doubt that the authors would respond by saying that intellectually unbiased people should be willing to hear any reasonable speaker and make this assessment for themselves, without forced censorship. However, surely they would also agree that there is some limit beyond which a speaker would not be acceptable even to them. But reasonable people can reasonably disagree about where this fuzzy boundary should lie – and that boundary must consider not only the message but the messenger as well.
Clearly a determinative number of alumni, faculty, and students at Middlebury judged that the lifetime body of work by Charles Murray, as well as his very clear lifelong mission, crossed that fuzzy line for them. Williams and Ceci may disagree on their placement of this line and that is legitimate and fair debate. But it is not legitimate and fair to conduct what amounts to a gotcha stunt under the guise of objective science to prove that these people’s determination in this instance is illegitimate and irrational.
All that Williams and Ceci may have actually shown is that, without attribution, college professors don’t assume the worst or the best. They may merely fill the void with their own middle-of-the-road interpretation of unattributed quotations.