robert bellah and people of color in habits of the heart

Habits of the Heart is simply a great book. Period. It’s not only a classic statement on American character, it’s also the first major book that employs a “cultural toolkit” framework, as developed by Swidler, Bellah, Tipton, and others. Still, that does’t mean that it’s without limitations. This post is a strong criticism of Habits‘ research methodology and how these problems lead to incorrect conclusions.

When I teach Habits, usually to graduate students, a common criticism is that the book only reflects the lived experiences of White Americans. One student said that after reading the book, you come away with the impression that the book is really about urban yoga fanatics.

I see what they mean. The book’s data is incredibly biased. In the preface to the first edition, the authors basically throw away standard social science data collection techniques. Each author did field work in a “community,” which is not specified. There is literally no discussion of how the field work was conducted (how long? auto-ethography? participant observation? field site selection?). Each author chose a “representative form” of public life, such as love and marriage.

They also offer therapy (page xliii) as a “increasingly important” aspect of middle class life. Wow! There is no argument or information presented about how common therapy is. Furthermore, there is a massive selection bias. If one of the issues you address is coping and pragmatic responses to particular life situations, then selecting therapy participants biases you towards a very specific kind of person. And don’t bother looking for descriptions of how interviews are conducted, or what the differences between populations might be.

When we read about data collection, it gets worse. Sample quote from the 1st edition:

“We do not claim that we have talked to average Americans or a representative sample. We have read a great many surveys and community studies, enough to know that those to whom we talked are not aberrant.” (page xliv)

Which studies? None mentioned. How did they measure the difference? No details, either.

Ok, now let’s get to racial differences. If you search the text for discussions of Blacks, you get very few, and only in reference to segregation or the Civil Rights movement (e.g., page 203). For a book about how people think about individualism, it is shocking to have so little discussion of how race may affect how people think about freedom and autonomy.

Someone drew  my attention to a 2007 Sociology of Religion article by Bellah where he answers critics. You can read it here. What he says on page 190 is that (a) he claims there is no difference and that (b) he addressed any differences in The Broken Covenant.

Let’s examine each point: (a) The critics are correct and Bellah is wrong. If you sample 200 people and interview them (see pages xlii-xliv), you will get about 30 Blacks – not enough statistical power to make any firm inference. It might be the case the he doesn’t understand statistical inference. With sample sizes that small, you simply will have a tough time picking up effects. But he admits he doesn’t have a representative sample to start with! Frankly, this is a mess.

(b) Bellah is wrong again. The Broken Covenant is a historical review of civil religion in America. To his credit, he does talk about race, a few times. But it is not an empirical examination of how Blacks and Whites deal with civil religion. There is nothing that I could find in this book that would lead me to believe that Whites and Blacks experience civic life in just about the same way. Heck, there are passages which suggest the opposite! A central message of The Broken Covenant is that civic religion has often come up short in America, which would suggest that some people feel left out.

Let me wrap up with a theoretical argument. One of the major innovations in the study of race and ethnicity is the application of habitus theories. This comes out with Bonilla-Silva and the “race without racism” school and it also comes out in more recent books like Emirbayer and Desmond’s treatment of race. If we understand habitus as being aligned with structures of inequality, our theoretical expectation is that whites and blacks would have very different situational responses to everyday problems. This theory may be wrong and maybe Bellah et al. might be right, but they simply don’t have the data to prove the null is true. Race (probably) matters.

Bottom line: Habit’s is commendable for many reasons, but research methodology is not one and it leads to some dodgy inferences.

50+ chapters of grad skool advice goodness: Grad Skool Rulz ($4.44 – cheap!!!!)/Theory for the Working Sociologist (discount code: ROJAS – 30% off!!)/From Black Power/Party in the Street / Read Contexts Magazine– It’s Awesome! 


Written by fabiorojas

April 25, 2018 at 4:40 am

%d bloggers like this: