Today’s guest post is authored by Sean Trainor, a historian of the early American republic with an interest in the intersection of labor, popular culture, and the body. He is a PhD candidate in History and Women’s Studies and Pennsylvania State University, where his dissertation examines the history of men’s grooming in the urban United States between the turn of the nineteenth century and the American Civil War.
A few weeks ago, I finished compiling a database, long in the works, containing the names and addresses of all of the barbers in the cities of Boston, Cincinnati, and New Orleans between 1800 and 1860. Thrilling, I know, but the project has broader implications for historians interested in the intersection of quantitative and cultural history which, if you’ll bear with a brief exposition, I’ll discuss below.
My goal for the database: to compare the number of barbers in each of these cities with their total populations over time. My hope: to find a declining trend line in each locale, suggesting a nationwide decline in demand for barbers’ services. Such a finding, I believed, would help substantiate some of the biggest claims of my study of men’s grooming in the nineteenth-century U.S.: that the barbershop was in decline between 1800 and 1860, victim to American men’s growing fondness for hirsute chins, a swelling anxiety among white customers regarding barbershops’ black proprietors, and the increasing availability of cheap razors and other grooming tools.
In typical fashion, however, history proved far messier. While the number of barbers relative to the populations they served declined fairly consistently between 1800 and 1840, the same figure increased dramatically over the next twenty years. Indeed, at the very moment when American men were ostensibly “fling[ing] away the razor,” as one disgruntled shaver put it, and abandoning the barbershop for good, demand for barbers’ services actually seems to have increased.
What was going on? Perhaps the apparent discrepancy had something to do with the rococo facial hair stylings of the 1840s, 50s, and 60s. While an average man might learn to shave himself with a modicum of safety and comfort—even using a dangerous nineteenth-century straight razor—he would have struggled to maintain a waxed, well-coifed ‘imperial’ or sculpt his hair after the fashion of Union General Ambrose E. Burnside. No doubt many of these men turned to tonsorial professionals for help.
Strangely enough, this resurgence failed to register in the realm of public discourse. An anonymous writer in an 1860 issue of The Albion, for instance, spoke for many when he declared “[h]ow are the barbers fallen! … [T]he red and white pole is as scarce as good Madiera”—a casualty of “the present generation,” who had “ruin[ed] the barber’s trade by selfishly shaving themselves.”
Why did the growing demand for barbers’ services fail to register culturally? Why the persistent myth of tonsorial decline? Most likely because the men who did the bulk of shaving, waxing, and curling in the 1840s and 1850s no longer fit the cultural image of a barber. Barbers, many antebellum Americans seemed to believe, were supposed to be men of color—proprietors of upscale shops located in the ground floors of hotels or along well-travelled city streets. But after 1840, the tonsorial ranks were increasingly dominated by German immigrants occupying dingy basements or alleyway shops. Indeed, while the total number of barbers, according to historian Douglas W. Bristol, Jr., increased sharply in Baltimore and Philadelphia between 1850 and 1860, the number of black barbers, as a percentage of the whole, declined dramatically. Barbers, in other words, were hiding in plain sight; elite commentators just didn’t know what (or rather whom) to look for.
Why should this story matter to a broader community of historians? Like Sara Damiano’s Junto post last fall, it speaks to the challenges and rewards of integrating quantitative methods and cultural history. If we are, in fact, in the midst of a quantitative revival, as Caitlin Rosenthal recently suggested, I think my struggles with data may prove useful to others.
Now, I make no claim to mastery of quantitative analysis. I am well aware of the shortcomings of my sources—including city directories and the U.S. Census (the problems of which are discussed in Seth Rockman’s Scraping By and a recent H-SHEAR thread, respectively)—as well as my limited grasp of statistical methods. My analyses are therefore simple, my claims modest. But I have gained one piece of portable insight from my forays in number crunching: that while quantification cannot transcend or replace other modes of analysis, it can often help highlight fruitful areas of inquiry. The story often gets most interesting, I find, when words, images, and numbers conflict.
Such is the case in my analysis of barbers. I would have been pleased to find that my quantitative study backed up contemporary assertions that barbering was in a state of decline. But instead, I got something far richer: a knotted mass of evidence that speaks to the complex ways in which perceptions of urban space are inflected with ideas of race, ethnicity, class, and gender.
 “Fling Away the Razor,” American Phrenological Journal 23 (February 1856): 32-3.
 “Barbers,” The Albion 38 (June 30, 1860): 303.
Great post, Sean. I was discussing this topic a few weeks ago in relation to my own dissertation research, which focuses on British historical memory in eighteenth-century colonial America. Part of my secondary reading covers history of the book literature on historical works, libraries, the book trade and book selling, etc…. In that field, there were amazing studies done throughout the 1970s and 1980s by historians of the book and printing that often included significant quantitative research and data. In the Introduction to a brief work on books in colonial Philadelphia, Edwin Wolf II wrote: “I have not drawn general conclusions from the evidence, merely set it down. The historical significance . . . I leave to others.” This seems to me to be a relatively common approach among the hardest of the hardcore historians of the book in this period. For himself, Wolf expected (or at least hoped) that cultural historians would use the data he had collected to make the cultural connections between books and politics, society, and economy. Unfortunately, that doesn’t seem to have happened on any significant scale and I can’t help but wonder if, over the 1980s-2000s, it had anything to do with an aversion to quantitative studies and data on the part of early American cultural historians. And, if so, why? Nevertheless, in my specific field of inquiry (as in others like yours), I, too, suspect that there is much to be gained by creating new intersections between quantitative and cultural early American history that will validate the former and enrich the latter.
Thanks for the response, Michael. I’ve always suspected / been led to understand that cultural-turn historians’ aversion to quantification was primarily political – i.e. rooted in a Foucauldian / po-mo critique that saw statistical / quantitative analysis as a way of knowing closely linked to corporate, bureaucratic, populationist ways of knowing (a critique to which I am not unsympathetic). I also suspect that, prior to the cultural / linguistic turn, it was a bit too easy for many scholars to overlook the shortcomings of the sources themselves – in other words, to get lost in the finer points of quantitative and statistical methodologies while overlooking the fact that city directories, the U.S. Census, and other sources that often served as quantitative fodder were wickedly problematic. Perhaps that’s unfair to both quantifiers and their critics. I’d love to hear back from someone who was privy to those arguments / conversations (on either side).
I’m not particularly privy to either side of the discussions, but I want to add a practical element to the cultural turn that you mention, Sean. Quantitative analysis is incredibly labor-intensive, even if you’re insensitive to the issues raised in evidence such as Census data. The ratio of labor to reward is completely out of whack compared to cultural history or other less quantitative modes of analysis.
Or, as one of my graduate school professors put it, the best dissertations maximize the historiographic impact while minimizing the labor involved, but most dissertations end up the reverse.
As you well know and demonstrate above with respect to barbers, it’s just a lot of work. I have trouble, for example, convincing students of just how many weeks or months it can take to produce a graph such as yours.
In any case I don’t think my point negates what you’re saying about cultural trends, and I very much enjoyed the post.
Good point, Joseph! Although it’s rather funny that quantification had its heyday during a period when the job market was nearly as bad as it is at present (which presumably put downward pressure, then as now, on time to completion), while the anti-quant revolt occurred during a comparatively friendly job market of the late 80s and early 90s. This doesn’t negate your point either, but it is curious.
Reblogged this on DailyHistory.org and commented:
Sean Trainor has posted an article at the Junto where he writes about how quantitative analysis changed the way he viewed his historical research topic. Trainor originally wanted to study the decline in men’s barbershops between 1800-1860. Throughout this period, his sources suggested that barbershops were slowly disappearing because of the introduction of cheap razors and the growing popularity of beards. But his project changed after sought to calculate the number of barbershops using quantitive analysis. He quickly realized that his original hypothesis flawed. While barbershops declined between 1800 and 1840, he noticed a meaningful increase in the number barbershops in after 1840. Instead of looking solely at the decline of barbershops he realized that he needed to update his thesis and re-conceptualize his project.
One of the potentially most exciting and frightening parts of any research project is when you realize that your initial assumptions were wrong. Trainor describes in the post what happened when he went through this process. It is a great post.
Thanks for the reblog, Sandvick! Well put.