Posted on February 25, 2009

A Fresh Look at the Lives of Civil War Soldiers Reveals the High Price of Diversity

David Glenn, Chronicle of Higher Education, February 27, 2009,


In Heroes and Cowards: The Social Face of War (Princeton University Press), Dora L. Costa and Matthew E. Kahn report that military units were more cohesive if they were composed of men who looked, voted, and worshiped like one another. Diverse units, meanwhile, did not fare as well.

The book is likely to draw attention from a wide variety of scholars–historians, political scientists, sociologists–because it brings a new kind of evidence to a longstanding debate about diversity and social cohesion that goes far beyond the Civil War. Scholars of education finance, for example, sometimes talk about a “Florida effect”: The typical property-tax payer in Florida is elderly and white, but the typical public-school student is Latino. In states where taxpayers are more similar to students, citizens tend to be more willing to invest public dollars in education. Ms. Costa and Mr. Kahn say that they do not support segregation of any kind–but that it is crucial to understand the costs associated with heterogeneity.

The Cost of Diversity


One of the book’s central questions is why more Union Army soldiers didn’t desert. (The Union Army’s desertion rate was about 9 percent.) More than 300,000 of them died during the Civil War–a fatality rate far higher, per capita, than in any other American war. Given that high risk of death, many historians have wondered why more soldiers didn’t flee.

The answer was almost certainly not the legal risk: Fewer than 400 Union soldiers were executed for desertion, and nearly half of the captured deserters were not formally punished at all.

In their statistical analyses, which cover more than 35,000 white soldiers and more than 5,000 black soldiers, Ms. Costa and Mr. Kahn demonstrate that certain individual traits were linked to desertion. All else equal, soldiers were less likely to desert if they were born in the United States or Germany than if they were born in Ireland, England, or elsewhere. Soldiers were less likely to desert if they were literate and had high incomes. And soldiers were less likely to desert if they were farmers than if they were artisans or laborers.

But Ms. Costa and Mr. Kahn also propose–and this is the heart of their book’s argument–that the soldiers’ social environments had effects above and beyond those personal characteristics. Holding all of the individual traits constant, white soldiers were less likely to desert if they fought alongside soldiers who were similar to them in terms of occupation, region, ethnicity, and religion. In African-American companies, soldiers were less likely to desert if they fought alongside soldiers from the same region. (Over all, black soldiers were slightly less likely to desert than were whites.) After the war, deserters were less likely to go home if their home counties were strongly pro-war (as measured by votes in the 1864 presidential election).


A Broad Scope of Data


But he [Thomas E. Rodgers, an instructor of history at the University of Southern Indiana] isn’t sure that the authors fully understand the context of Civil War recruitment. For one thing, he says, they might be too quick to assume that new soldiers actually had roots in the county where they enlisted and shared its general opinion of the war. “When you look at the 19th century in general,” he says, “young men moved all the time.”


But the two economists insist that on its own terms, their Civil War study tells a powerful story: Social networks matter. In a forthcoming paper in the journal Demography, they report that in Union veterans’ old age, their health was worse if they had experienced a high amount of battle stress during the war. That isn’t surprising. The paper’s startling finding is that among veterans whose military companies had been highly cohesive, the effect disappeared. Something about the experience of close kinship appears to have insulated the men from some of the long-term costs of warfare.

Lifetime Benefits

Not all the news about diversity in Heroes and Cowards is negative. Among African-American soldiers, recently freed slaves seem to have experienced long-term benefits if they served in diverse units with large numbers of Northern black freemen. The higher the proportion of freemen, the more likely the former slaves were to move north and become literate later in life.


In a previous paper on economic diversity and community life, Ms. Costa and Mr. Kahn criticized the political scientist Robert D. Putnam’s much-debated 2000 book, Bowling Alone: The Collapse and Revival of American Community. Mr. Putnam, they said, had exaggerated the decline of American civic life and had looked in the wrong places for explanations. According to Ms. Costa and Mr. Kahn, we shouldn’t blame television or overwork or urban design. The real explanation is the rise in social heterogeneity, which makes people slower to form bonds of trust.


For good or ill, the instant trust that comes from regional and ethnic ties can be a powerful force. In the Andersonville prison camp in South Carolina in 1864, imprisoned Union soldiers tried desperately to find unity against “raiders”–marauding bands of fellow prisoners who stole food and brutalized their victims.

According to a war memoir quoted by Ms. Costa and Mr. Kahn, the raiders were finally defeated by men from the Midwest who “spoke the same dialect, read the same newspapers, had studied McGuffey’s Readers, Mitchell’s Geography, and Ray’s Arithmetics at school, admired the same men and generally held the same opinions on any given subject. It was never difficult to get them to act in unison.”

[Editor’s Note: A Chronicle account is necessary to read the complete article.]