Writing good non-fiction books
Can academics do it? Is political science lagging?
For a long time, I’ve thought about which academic books make it big - that is, they actually sell a decent number of copies and are read by non-academics. Everyone loves to criticize Malcolm Gladwell for distorting the research he summarizes. So why not cut out the middleman and produce popular works based on research?
Many of us have certainly tried and a few have succeeded. As a practicing political scientist, it occurred to me that my profession hasn’t done a great job at this, despite the fact that there is lots of public interest in politics. My sense is that people do buy books about politics, just not ones written by political scientists. Or at least I had trouble thinking of many.
(Levitsky and Ziblatt’s How Democracies Die might be the exception, but I worry that their book succeeded mainly by appealing to a hunger for anti-Trump books rather than a hunger to learn about politics.)
In any case, I tried to answer these questions by analyzing Tyler Cowen’s annual lists of best non-fiction (though I’d note that he is not selecting on sales or popularity). Here are a few trends that I spotted. I’ll comment more on them at the end.
Academics in general are the authors of some of the best non-fiction works, at least in Cowen’s view.
In terms of subject matter, history is the leader. Political science definitely lags at least in its academic form.
Biographies (coded separately from subject matter) also score well.
Regionally, the less developed world seems to lag.
Cowen’s best non-fiction
Again, my goal was to figure out what makes a good work of non-fiction and whether political science lags behind in producing it.
I decided to start with Tyler Cowen’s year-end lists of the best non-fiction. As many know, Tyler is a voracious reader with broad tastes. As far as I can tell, his lists go back to 2011 and continue to the present. Altogether he has recommended 235 books. You can find all of them along with my categorizations here. I hope to conduct the same exercise with some other sources in the future.
Subject matter
I started by characterizing the subject matter of each book. The categories I created were a bit arbitrary, especially the lines between politics, current events, and society, but I think they capture some differences. Figure 1 shows the number of books in each category.
Figure 1: Subject matter of Cowen’s best non-fiction
A surprise for me was how many history books he recommended. I counted 57 or about a quarter of the total. This was considerably more than even economics (34), his own field.
What makes history so appealing? Are historians better writers? Or storytellers? (Cowen is suspicious of stories.)
I’d guess that the key is an aspect of books that Cowen often praises: informational density. History books tend to include lots of facts (though presumably they differ in how novel and interesting these facts are). By contrast, social science books tend to work more on a theme and variations approach - a big argument followed by various tests, nuances, and applications.
Probably the social science style is better suited to articles, while history works better as books. (Cowen’s blog often cites articles by social scientists but relatively few by historians.) This corresponds with publication norms in both fields. It also jibes with the advice I received in political science grad school which was to read the first and last chapters of books and skip the case studies in the middle. (Do history grad students get the same advice?)
Political science in non-fiction
Though history led Cowen’s list in terms of subject matter and economics came in second, politics was the third most common field with 28 books. (It might have done better if I had added some of the books that I coded as current events and society.)
But most of these books weren’t written by political scientists. By my count, less than half. The others tended to be written by historians or pundits or were biographies of political figures (Lincoln, Thatcher, Bismarck, Stalin...). By contrast, a good proportion of the economics books were written by academic economists.
If you are curious, here are his choices written by academic political scientists (links are in the google spreadsheet above):1
Thane Gustafson, The Bridge: Natural Gas in a Redivided Europe.
Eric Kaufmann, Whiteshift: Populism, Immigration, and the Future of White Majorities.
Dennis C. Rasmussen, The Infidel and the Professor: David Hume, Adam Smith, and the Friendship That Shaped Modern Thought.
James C. Scott, Against the Grain: A Deep History of the Earliest States.
William Taubmann, Gorbachev: His Life and Times.
Daniel W. Drezner, The Ideas Industry: How Pessimists, Partisans, and Plutocrats are Transforming the Marketplace of Ideas.
David B. Roberts, Qatar: Securing the Global Ambitions of a City-State.
Arthur M. Melzer, Philosophy Between the Lines: The Lost History of Esoteric Writing.
Daniel Drezner, The System Worked
Mark Lawrence Schrad, Vodka Politics: Alcohol, Autocracy, and the Secret History of the Russian State.
Daniel Treisman, The Return: Russia’s Journey from Gorbachev to Medvedev.
And even these choices are somewhat off the beaten track. Besides Scott and Treisman, they are not the most acclaimed scholars. Two of the books are political theory (Rasmussen, Melzer) and another three are basically histories (Scott, Taubmann, Schrad), both marginalized parts of the discipline. Few of them look like or are based on what passes for standard political science. Treisman and Kaufmann would be closest on this score as both books are based on quantitative and game theoretic analyses.
Maybe also noteworthy is that four of these books were written by scholars of Russia, a regional specialty that has traditionally been a bit disconnected from the rest of political science despite its importance. Though note what Paul Poast says about Russia and international security:
Academics
The low number of political scientists is not due to the fact that academics in general have trouble writing good non-fiction (at least according to Cowen). Most of Cowen’s choices were written by academics. I counted 137, just short of 60%, written by PhD holders (and 122 of those were primarily academics) versus 98 by those with other degrees.
Being in academia isn’t stopping political scientists (or others) from writing good books. Again, however, it was academic historians who seemed to be doing this the best.
Biography
Another surprise was how many of Cowen’s choices were biographies/memoirs, something I coded separately from subject matter. (I coded a biography of Euler, for example, as science for subject matter.) I counted 53, close to a quarter of all his choices. Again, I don’t tend to read many biographies, but maybe I should.
My worry about biographies is that they seem to be selecting on the dependent variable - only successful people get books written about them. It is hard to make attributions of their success without comparisons to others who might share their characteristics but did not achieve success, though that may not be the main point of biographies.
Regions
Finally, I checked the regional focus of Cowen’s choices (for ones that were clearly based on a place). Europe led with 65 books, the US with 53, and Asia with 37.
The regions that were left out were Latin America (only 5 books) and Africa (only 2). These trends shouldn’t hurt political scientists who also focus a lot on Europe and the US.
In closing (for now)
As I suspected, political science doesn’t seem to be producing a lot of successful non-fiction, at least in Cowen’s view. Why? The best I can come up with is that our books lack the informational density of histories and biographies. Maybe we should put more “stuff” in our books. Something like Treisman’s The Return which alternates between narrative and analysis. This should apply to economists as well but their books might have more practical use and they tend to have crisper conclusions. Or perhaps we should not worry about book writing so much. I’ll keep thinking about this.
A few final takes based on Cowen’s choices:
Perhaps I should be reading more histories and biographies.
Academics in general can write good non-fiction.
We could probably use more good non-fiction books about Latin America and Africa.
My next step is to consider a few more indicators of success in publishing by academics, perhaps NYT best-seller lists. I’m happy to hear some advice on where to look.
UPDATE: I’ve added Cowen’s 2022 choices to the spreadsheet. They mostly follow the trends outlined above. Of the 28 new books, 15 were written by PhDs, 14 focused on Europe (of which 2 were on Russia), 10 were about the arts, 8 about history, and 5 about economics. And unfortunately, no political scientists!
There were also a couple of others with PhDs in political science who don’t really practice political science, in particular, Charles Murray and Anatol Lieven.
I think there's something pretty general about Tyler's information consumption that he prefers a lot granular, contingent, and detailed stories that have a good dose of an overall framework, but not necessarily where a clear theory that is being tested. You see this in his preferences for travel, his distance from the sort of Top-5-journal econ crowd, his preference for literary fiction and movies, and even in his preference of Plato over Aristotle.
While he is very much trained in and finds himself in the build-a-theory-and-test-it tradition and finds a lot of value in that, I think in a lot of his reading, viewing, listening, traveling he is trying to correct the excesses of that tradition by constantly exposing himself to complex, messy, loosely organized details. This is why he is among public intellectuals very good at avoiding at the pitfalls of monism and professional deformation, i.e., the pitfalls of seeing the world only the lens of one's preferred theory.
>Maybe we should put more “stuff” in our books.
I would strenuous vote "no" on this. The worst thing about writing by historians is the tendency to include voluminous, barely relevant detail. Facts which do not play a role in establishing (or challenging) the author's thesis should no be included.
And, on another note, I am skeptical that there are 53 biographies that are more worth reading than, say, The Deadly Ethnic Riot, or some of Bruce Bueno de Mesquita's stuff, to name just a couple. Cowen's list just isn't very credible.