« Oracle v. Google - Round I jury verdict (or not) | Main | Cassidy on Suggestions for Improving Legal Education »

Monday, May 07, 2012

Principles for Productive Conversations About Race

During the last presidential election cycle, then-candidate Obama called on Americans to talk about race.  Yet today, it's remarkably difficult to find examples of conversations about race that advance understanding rather than leaving participants frustrated and more entrenched in their positions than before.

In the course of teaching Constitutional Law this semester, I've given some thought to the characteristics of useful conversations about race.  I tend to think that the most helpful guidelines are procedural rather than substantive.  Here's a non-exhaustive list:

1. Strive for empathy. Before the word "empathy" became politically charged rhetoric, it referred to the entirely reasonable exercise of putting yourself in someone else's shoes. What does it feel like to be an economically disadvantaged white man who lost a job to an applicant of color with lower test scores and less work experience? What does it feel like to be a person of color whose daily experience of racism is discounted as exaggerated or imaginary? Engaging in good-faith efforts to understand the racial frustrations of those not similarly situated to you furthers a more nuanced conversation regarding racial policy.

2. Don't cite to your friends.  There are various permutations of this argumentative flaw.  One of the more common involves sheltering controversial views behind non-white people.  In its most basic form, the claim is:  "I have non-white friends; therefore, what I'm saying can't be racist."  Some have dubbed this the "some of my best friends" defense.  Perhaps proving the point, others advise making friends with "intelligent and well-socialized blacks" as an "amulet against accusations of prejudice."  A slightly more advanced version of the "non-white friends" argument goes something like this: "My non-white friend agrees with this argument. Therefore I'm not racist, the argument I'm making can't be a racist argument, and moreover my argument is correct."  Put simply, this isn't useful.  The fact that one non-white person, somewhere, once upon a time, agreed with a particular argument doesn't make that argument correct.

3. Don't assume homogeneity. Within and among various racial groups, there is a wide range of opinion on race-related issues ranging from affirmative action to employment discrimination. And people who agree on the big picture often draw quite nuanced distinctions when it comes to the details. Assuming that people of a particular race agree about everything amounts to stereotyping. Figure out what other people really think, and respond to that.

4. Don't substitute anecdotes for arguments.  The fact that you "know a few school teachers" is an unhelpful basis for drawing conclusions about access to technology among inner-city schoolchildren nationwide.  Examples, of course, are different from anecdotes -- specific examples can be helpful in illustrating abstract points.

5. Don't cite to outliers. Observing that Barack Obama is the President does little to further an argument that blacks, as a group, are not subject to systemic disadvantage. There may be support for that claim, but isolated examples don't provide compelling support for a generalization about forty million black Americans.

6. Consider intersectionality. Most of us are privileged in some ways and disadvantaged in others. Treating race in isolation ignores these various alternative axes of advantage and disempowerment. And considering intersectionality also furthers #1, above. Understanding what it's like to be disadvantaged on the basis of sexual orientation may further an understanding of what it's like to be disadvantaged on the basis of race.  Of course, there are important differences, which I don't mean to understate.  But certain aspects of the experience of disempowerment may translate across categories and, again, further our understanding of one another's experiences.

7. Avoid offensive rhetoric.   This should really be too obvious to say, but recent history suggests that it's not.  What purpose is served by referring to Elizabeth Warren's "$1.7 million dollar wigwam in Cambridge"?  To the extent you have a colorable argument, relying on this sort of language will ensure that no one will pay attention to the substance of what you have to say.

8. Question your starting assumptions.  I could give a lot of examples, but here's one that I discuss with my students.  In the affirmative action debate, a lot of people begin with the premise that "merit" is objectively ascertainable from grades and SAT scores. Do you know the correlation between SAT scores and college grades? If you don't, you should educate yourself before you engage in a discussion about how affirmative action results in the admission of "less qualified" candidates.  To get you started, consider this multi-year study involving nearly 78,000 students within the University of California system.

I would be interested in hearing others' ideas for facilitating better conversations about race, both in and beyond the classroom.

Posted by Nancy Leong on May 7, 2012 at 08:50 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Principles for Productive Conversations About Race:


I'm grateful to those who offered helpful comments, particularly AnonAnonAnon and Interested Student.

@Dan, thanks so much for mentioning this very useful article by Richard Delgado, which I also engage around page 72 in my forthcoming article Racial Capitalism, available here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2009877. For those interested, the full cite to the article Dan mentions is Richard Delgado, Rodrigo’s Book of Manners: How to Conduct a Conversation on Race — Standing, Imperial Scholarship, and Beyond, 86 Geo. L.J. 1051 (1998).

@Allegory of the Caveman, regarding your statement: "how can a (non-racist) white person know whether racism exists in everyday life if he or she can't draw conclusions based on the experiences of minorities he or she knows?" Here are a few thoughts. First, to clarify, the point of the second principle I listed above is that "My black friend thinks X" is not an analytically sound way of advancing an argument regarding broad legal or policy decisions that affect millions of people. Second, as to how a white person might know that racism exists without referring to the experiences of non-white associates, I suggest that you familiarize yourself with the fields known as "psychology" and "sociology," within which researchers have spent decades investigating bias and racism. I would be happy to offer more specific reading suggestions should you choose to shed your protective veil of pseudonymity. Finally, nothing I've said above is meant to discourage people of various races from sharing personal experiences of racism, which can provide useful examples of abstract arguments (see #4 above). My point, though, is that the most productive conversations tend to rely on empirical evidence and sound logic rather than secondhand citation to other people's views.

Posted by: Nancy Leong | May 9, 2012 9:52:49 AM

And I don't think "a lot of people begin with the premise that 'merit' is objectively ascertainable from grades and SAT scores." They're just upset that "merit" under certain admission policies means also having been born a certain color.

Posted by: Allegory of the Caveman | May 8, 2012 10:05:46 PM

"Rules" 2, 4 and 5 are pretty much the same thing in practice.

And how can a (non-racist) white person know whether racism exists in everyday life if he or she can't draw conclusions based on the experiences of minorities he or she knows?

Posted by: Allegory of the Caveman | May 8, 2012 9:59:37 PM

What a wonderful way to present this concept. Thank you very much for sharing these ideas.

Posted by: Nareissa Smith | May 8, 2012 7:08:21 PM

Professor Leong,

This is an excellent post, and your rules–if followed–would certainly improve the debate. While you touch on this in Rule #6, I believe that we must all remember that discussions related to racial issues cannot occur in a vacuum. Too often it seems that we narrow discussions of important issues down to their racial components. While race certainly plays a role in many social issues, ignoring other components of these issues only brings into contrast the extreme positions that parties take on race related issues. In short, what I am saying is that race plays a pivotal role in society, but must be viewed in the context of the larger social issues that facilitate this role. I have been party to many conversations where an individual will assert that if he or she was a minority they would still enjoy the success they do today, or conversely, that if he or she were not a minority than they would be more successful. Apart from the fact that these statements further inaccurate and overbroad stereotypes, such assertions are harmful because they assume that racial issues can be viewed independent of other societal issues, such as economic status, access to quality education, access to business/social networks, and availability of quality mentoring. I believe that it is important that as we address the racial issues that still in many ways divide this country, we recognize that myriad of factors that contribute to enhancing or suppressing certain racial, ethnic, and social groups. Only by viewing racial issues in this larger context can we truly appreciate the challenges, and have a truly effective dialogue moving forward.

Posted by: An Interested Student | May 8, 2012 4:57:48 PM

Dear PublicSpiritedPig,

I’m afraid that you’ve both misapplied your suggested rule and run afoul of it.

The study Professor Leong linked to examined admissions in the U.C. System, which includes schools with a range of selectivity. The 25%-75% SAT range for U.C. Merced, for example, is 1350-1740. You don’t explain where you came up with the suggestion that the study only examined schools where “everybody’s SAT score is 1700+.” But by my count, only three of the U.C. Schools have a 25% SAT score above 1700 (Berkley, UCLA, San Diego). I do declare, rule #9 has been violated!

The study also uncovered the fact that at the U.C. schools, SAT I scores provided ZERO explanatory power of undergraduate GPA once socioeconomic factors are held constant. In other words, the SAT I predicts that students from poorer families with less educated parents will on average perform slightly less well than their peers from wealthy families with better educated parents. The SAT I provides no information about which students from different backgrounds will perform better (see chart on page 11). This is a less-flagrant violation of rule #9, but declaring that the study only shows that the SAT II better predicts undergraduate GPA ignores this portion of the study, and that strikes me as “cherry-picking data.”

As for your misapplication of the rule, unlike you, I’m not sure exactly what Professor Leong thinks the study means. She doesn’t say. But the study is cited after she suggests that “[i]n the affirmative action debate, a lot of people begin with the premise that ‘merit’ is objectively ascertainable from grades and SAT scores.” The study clearly provides evidence that this “assumption” is worthy of consideration. Strikingly, the study found that even combining all traditional metrics of merit – high school GPA, SAT I score, and SAT II score – we can still predict only a little over 20% of the variation in undergraduate GPA of U.C. students during the studied time period. It's true, of course, that the U.C. Schools are more selective then some other institutions. But the debate over affirmative action largely focuses on selective institutions. This study suggests that in considering admissions policies at elite institutions, we should consider whether “’merit’ is objectively ascertainable.”

Posted by: longtimewatcher | May 8, 2012 2:41:40 PM

Nancy, have you seen Richard Delgado's review of Farber and Sherry's book about this topic way back? It has implications not only for conversations about race but (legal) scholarship more generally.
See 86 Geo. LJ 1051 (1998), entitled something like: How to Conduct a Conversation on Race.

Posted by: Dan Markel | May 8, 2012 11:56:50 AM

Yes, we know that the correlation between HSGPA and SAT and undergraduate performance is strong. You cite to a study that shows that within a narrow range of high-performing SAT/HSGPA candidates (those in the UC system, see fn. 2) the SAT II is a better predictor. As the study itself shows, the less selective the institution, the more powerful the SAT effect is. When everybody's SAT score is 1700+, there's less variance. It IS interesting that the SAT II is better, even in a similarly restricted range -- which is the point of the study. It doesn't show what I think you think it does. In an institution where the range of SAT scores is 1700-1800, the SAT will offer less prediction than in an institution where the SAT range is 800-1800.

So I'd add a 9th point:

9. Avoid cherry-picking data.

Posted by: PublicSpiritedPig | May 8, 2012 11:28:32 AM

Great post! If we followed these deceptively simple rules, perhaps we could begin to address the cultural phenomenon that maintains race-based social stratification.

Posted by: Anonymous | May 7, 2012 11:14:32 PM

8. Assume others are arguing in good faith. Nothing shuts down debate faster than accusing someone of being racist or arguing in bad faith. Take the arguments of others as being made in good faith, even if they are arguments that make you uncomfortable. If you think others are misguided, respond to them on the merits rather than by attacking them as persons.

Posted by: Anon Anon Anon | May 7, 2012 10:39:49 PM

The comments to this entry are closed.