A mini-review of Bryan Caplan’s The Case Against Education

“It was not until years afterward that I came upon Tolstoy’s phrase “the snare of preparation,” which he insists we spread before the feet of young people, hopelessly entangling them in a curious inactivity at the very period of life when they are longing to construct the world anew and to conform it to their own ideals.” -Jane Addams, Twenty Years at Hull House

Bryan Caplan has long inspired me. We don’t share a political ideology, but his writing on child-rearing has often come at exactly the right moment for me. (His Selfish Reasons to Have More Kids helped me overcome a brief antinatalism phase, for instance.) His work on borders and immigration is less groundbreaking, but no less true, and in his forthcoming fun comic on the topic he and Zach Weinersmith will bring scholarly rigor and friendly advocacy to new heights. He’s obviously right that immigration restrictions are immoral and self-defeating—but no one is listening in this new age of nationalism. His latest book has fewer concrete ethical consequences—but it deploys evidence from educational psychology that has long puzzled me in service of a policy argument that has almost no chance of uptake, and so cements my view of Bryan as a careful and provocative scholar doing his best to tell the truth even when no one will listen.

Mini-Review

The argument in The Case Against Education is simple: most people don’t learn much of value to employers in their college educations. This is possibly also true even for some parts of K-12 schooling. Education instead is largely a mix of experience high-ability people would seek out on their own and an opportunity to distinguish oneself from other applicants in the resume rat race. The bulk of the book is a response to the various objections that are now forming in your mind.

You’d have to be pretty nerdy to be reading this, so the first step for evaluating the argument is to use a bit of empathy: forget your own experience in school, except the bad parts. I hated high school, but I loved college so much I took it as a career. Even then, I don’t remember a good deal of what I studied outside of my chosen field. And many of my fellow students were much less enthusiastic. So ask yourself:

  1. How much high school Spanish do you remember?
  2. Do you remember the titles—let alone the plots—of all the books you read in 11th grade English?
  3. What is ionization energy?
  4. Remember calculus? Can you solve a parametric equation today?

Perhaps you can answer half of these questions today without Google. That’s not a lot of retention. Whenever I get stuck in conversations on planes with people about the one philosophy class they took in college, they tend not remember much of the content. (“The cave, right?! Brains in vats? Veil of ignorance…. I hated that class.”)

Caplan summarizes well-established but little-known work in educational psychology on learning transfer which seems to show that mostly students don’t learn or retain much. Instead, a lot of education seems to combine three things, in some combination: an accumulation of habits, skills, and knowledge that we can call “human capital,” a costly and difficult signal that distinguishes us to employers, and a kind of consumption that is distinctive of high ability and high-income people.

I won’t say much about signaling as such: for Caplan, education provides future workers with an opportunity to create truthful, hard to fake resumes that demonstrate intelligence, conscientiousness, and conformity. On his view, the time you spent acing classes you’ll never need proves you’re willing to play the game better than any personal statement could ever do. Of course that’s part of it… but how much?

The human capital model is the one we’re all thinking about when we recommend education. Education, we want to believe, makes you smarter, more capable, more knowledgeable, and more effective. Caplan seems to think that this is a relatively small part of what is going on in education. In the book he sometimes says human capital is 20% of education’s contribution to income, though he’ll also say it is 11% of the effect of education.

That’s because education is also fun, and especially fun for people who tend to earn high incomes because they are intelligent, curious, and conscientious. In that sense, education is like other high-class consumption goods: eating good food or taking fancy vacations, for instance.  In fact, the “fun” part of education rivals the signaling element. (He estimates ‘ability bias’ accounts for 45%, and signaling for 44%.) I’ve known many smart, curious people who retire from a successful career and go back to school. They’re not in school to learn and become more effective workers, but rather because education can be an intrinsic good with no instrumental value.

This is likely the case my progressive friends would make: you don’t study philosophy to be a better nurse or accountant or medical doctors—though there are ways that the critical thinking skills you learn may help you—you study philosophy because you’ve got questions about the nature of the universe, existence, death, justice, beauty, and truth. And the smarter and more successful you’ve been, the more you can enjoy learning about philosophy and literature. It’s an end-in-itself. Caplan seems to think that education as a high-ability consumption like backpacking in Europe or kite-surfing in the Caribbean—for kids wealthy enough to afford it on their own or retired adults looking to reflect on it all, but not for that time in your life when you’re trying to figure out your place in the economy.

I think we progressives should take Caplan’s argument seriously. But in some ways we already do: we’ve all read and shared articles like these: “Why American Colleges are Becoming a Force of Inequality,” and “Schools that accept ‘no excuses’ from students are not helping them.” Progressives are coming around to the idea that higher education is not a great leveler, and the segregated K-12 schools are increasingly a pipeline to prison rather than jobs for the least advantaged.

Our counterarguments often play up underfunding of state flagship universities, and so progressives often seek to double down on higher education with Bernie Sanders-style free college guarantees and increased spending. But at the same, we are increasingly aware of efforts to make schooling more regimented, disciplinary, and prison-like. We see that African-American and poor students are being shuttled towards “no excuses” schools while white and wealthy students find get play-based curricula, experiential learning, and above all a kind of caring and loving environment. Those experiences should tell us something.

Look forward to some future posts (or maybe someone will ask me for a real review) using my favorite sources: Michel Foucault, Paolo Freire, Pierre Bourdieu, Elizabeth Anderson, and John Dewey. But I put Jane Addams there at the top for a reason: it’s not just libertarians but one of the founders of progressive pragmatism who holds this view.

A review wouldn’t be complete without some criticisms: Caplan quotes Richard Arum and Jospia Roksa only once, and ignores their findings that the right kind of liberal arts education can increase critical thinking, problem solving, and analytic writing skills. He believes that this can only work for eager students, which are in short supply, and that most of the results of the Collegiate Learning Assessment can be confounded with IQ. His emphasis on IQ means that he also hasn’t properly evaluated the Foucaultian argument that schools produce large amounts of social conformity and conscientiousness, rather than merely measuring it. Finally, there is plenty of evidence that education plays an important signaling role for historically oppressed groups (women, African-Americans, and the formerly incarcerated). In fact, Georgetown’s Center on Education and the Workforce just published this study, which is being reported widely with headlines like this: “Women need one more degree than men to earn the same average salary.”

Still, these objections don’t overcome the overall problems with education as it is currently practiced. Very often we see policy justification switches like the following: when the evidence from Quebec and Tennesse on early childhood education began to countermand the Abecedarian Project’s consensus view that universal pre-K could benefit poor children, advocates switched their arguments from the benefits to children to benefits to mothers’ employment. This kind of motte and bailey argument doesn’t have to be a total fallacy, since after all a policy can have many possible promising effects, some of which end up being disproven. But it’s more evidence against schooling as the accumulation of individual human capital.

(previously: What are the ruling ideas today? Is ‘College For All’; among them?Academically Adrift’s Methodological ShipwreckFor Education, Against Credentialism)

An Ostrom Reader

Lexington Press has recently finished publishing a four volume collection of the work of Elinor Ostrom and her husband Vincent–before that I do not believe the work has been gathered anyplace easily accessible. Since the price is astronomical–though well worth it for the serious scholar or scholarly library, I’m sure–I’d love to have a single-volume reader that collects the most important pieces, while perhaps leaving some of the more detail-oriented empirical and modeling work behind.

Perhaps one reason no such “Portable Ostrom” collection exists is that her work has been widely pirated online–claimed by the commons if you will–a fact that made the links below easier to find. Here are some things I might include in such a reader:

Ostrom frequently plagiarized herself and many of the links above have repeated passages and arguments. She thought that the public needed access to certain information about governance and skills at self-organization that we don’t teach in school, and that mainstream economics has actively undermined. She felt an obligation–which is now ours–to find some method for expressing these insights in less technical and more accessible ways.

Provoking pedagogically-effective discussion in college courses, with an example using Danielle Allen’s Cuz

Today is the first day of classes in my seventeenth year of teaching. I have taught a lot over those years–sometimes as much as a 5/5/1 (5 courses in Fall, 5 in Spring, and one over the summer.) My sense from that time is that the value of a philosophy course is largely not derived from excellent lectures on my part–but rather from an engaged seminar discussion. This is sometimes called “Socratic” but I happen to think that Socrates provided a terrible model for contemporary faculty.

Still, I think students learn more from what they do and say and write in the classroom than from what I do, say, and write. The kind of reading, note-taking, and preparation I do to give a lecture helps me understand material deeply–and it’s precisely that kind of reading and preparation that I want my students to cultivate themselves. In that spirit, I have developed a kind of “in-class” presentation which is both how I think of my own best classes, and also allows students to easily step into the role of “guiding discussion” themselves.

During the semester each student takes responsibility for a “provocation,” a written and oral project whereby they start off the class. This works best in small seminars under 15, but it can scale up to 30 with careful management. Each class period a student takes responsibility for kicking off our discussion of the reading with a short paper that briefly summarizes the argument, pulls a choice textual selection for discussion, and asks a provocative question or two, and then explains why this question meets three critera: (1) it is personally interesting to the student, (2) difficult to answer because it turns on a deep philosophical disagreement/confusion or rests on tricky empirical issues, and (3) important for directing further study and/or its answers will have implications for other relevant questions.

I always make sure to model these provocations for students myself, and indeed this afternoon I’ll be doing so using an article by Danielle Allen:


Danielle Allen’s “The Life of a South Central Statistic” is an excerpt from her book Cuz, which describes her cousin Michael Allen who was incarcerated as an adolescent for a string of robberies and thefts. Danielle Allen describes how Michael was locked up under the then-new three strikes policy in California (which also enhanced sentencing for carjacking) and how prison changed him—and how the relationships he formed there eventually led to his murder. Though he worked as a firefighter while incarcerated his criminal record kept him from taking firefighting up as a career upon release, and he fell into the drug trade. Though she lays some blame at the feet of the California legislature for meting out such a harsh sentence, Danielle Allen also describes the violence of organized drug trafficking as a “para-state” with twice the resources of the CIA operating in American cities to exploit and kill men like her cousin.

One of the more striking passages in the article is this one:

“California’s legislators had given up on the idea of rehabilitation in prison, even for juveniles. This is a point that critics of the penal system make all the time. Here is what they don’t say: legislators had also given up on retribution. Anger drives retribution. When the punishment fits the crime, retribution is achieved, and anger is sated; it softens. This is what makes it anger, not hatred, a distinction recognized by philosophers all the way back to antiquity. Retribution limits how much punishment you can impose.

The legislators who voted to try as adults sixteen-year-olds, and then fourteen-year-olds, were not interested in retribution. They had become deterrence theorists. They were designing sentences not for people but for a thing: the aggregate level of crime. They wanted to reduce that level, regardless of what constituted justice for any individual involved. The target of Michael’s sentence was not a bright fifteen-year-old boy with a mild proclivity for theft but the thousands of carjackings that occurred in Los Angeles. Deterrence dehumanizes. It directs at the individual the full hatred that society understandably has for an aggregate phenomenon. But no individual should bear that kind of responsibility.”

In the quoted paragraphs above, Danielle Allen seems to suggest that the political morality of deterrence is worse than revenge. Is the purpose of criminal punishment to prevent crime? Does this treat a person like an aggregate–a statistic–as she suggests?

This fascinates me because I am tempted to believe that the only reasonable use of state violence to punish is to deter worse behavior, but such efforts are often accused of dehumanizing the perpetrator. Yet revenge seems more dehumanizing, doesn’t it? Perhaps this is difficult to answer because the manifold justifications for punishment all speak to us at different times in terms of different crimes: when we see the individual harm to a victim we are much more likely to demand the satisfaction of our anger in revenge—but when we think about the ways that a deterrence theory might prevent some crimes from even happening it seems better than having more crime and more retribution for those crimes! I wonder whether there are techniques that could be used to combine these theories: perhaps there are ways that revenge is itself deterring—for instance it signals that crimes are unacceptable. But still there is more to deterrence than renaming revenge: for instance it might be the case that some crimes are difficult to prevent, while other crimes—which cause less harm overall—can be prevented best with really graphically shameful punishments. (For instance, perhaps slumlords are best deterred by being required to stand shamefully in front of their badly maintained buildings holding a sign indicating their violations.) There’s a lot of further study warranted here—and plenty of room for both empirical assessment and more principled philosophical exploration of the related themes.


This provocation barely touches the surface of the interesting themes raised by the article and Allen’s book. But it’s enough to get a conversation started, and I usually come prepared with four to six passages and questions like this for an hour-long class. Quite often I find that even students who are randomly assigned to provoke on some topic develop a semester-long fixation on the themes that arose during their provocation–just because the deep thinking and preparation required to write this short assignment and share it with others gives them a sort of endowment effect with those issues. Here are some more provocations on Allen:

  • Michael was technically a “violent” criminal but his victims weren’t really hurt. He was also a teenager, and perhaps less culpable than an adult in a similar situation. What should we make of his age in assessing his culpability?
  • Michael’s lover–and murderer–was a trans woman named Bree and there are all sorts of issues raised by her time in a men’s facility in California. Should Bree have been housed with women? What would have happened to Michael then?
  • Michael had a loving and supportive mother but her struggles with abusive partners may have contributed to his fate. Could she have done anything differently? And how do our public policies exacerbate these circumstances?
  • Some of Michael’s difficulties upon release are closely tied to the stigmas he faced during reentry. But others are tied to the fact that he fell in love with Bree while incarcerated–they are the results of the deliberate decisions of an adult man struggling to manage social expectations, economic needs, and an obviously abusive relationship with someone who he loved helplessly. What should we make of his story?
  • I find Allen’s discussion of the para-state endlessly fascinating and I wonder whether this is something that prison abolitionists should spend more time working on. Why does she name it a “para-state” and what should we say about the violence that arises from it? Does she partly exonerate the United States for its racist, mass incarcerating policies thereby?

Ideology and Education

Thomas Edsall has a good review of some recent research on polarization in the New York Times today:

The strength of a voter’s identity as a Democrat or Republican drives political engagement more than personal gain. Better educated voters more readily form “identity centric” political commitments to their party of choice, which goes a long way toward explaining the strength of liberal convictions among more affluent Democrats.

This is a striking chart:

 

 

 

The Enduring Appeal of Perversity Arguments and Unintended Consequences Warnings

James Forman, Jr. won the Pulitzer Prize last week for his book Locking Up Our OwnIt is well-deserved. That book–and his earlier work wrangling with Michelle Alexander’s The New Jim Crow–shows the ways that we have arrived at the wicked problem of mass incarceration through something much harder to disdain than evil scheming by distant elites. We did it to ourselves, and African-American political leaders did it with the full support of their African-American constituents.

Forman’s argument is sometimes conflated with that of Naomi Murakawa, who argued that it was liberals, not conservatives, who created mass incarceration by emphasizing the importance of safety over all other civil rights goals. According to Murakawa, when Lyndon Johnson championed the the 1968 Safe Streets Act which swelled the flow of federal dollars–and federally procured military equipment–into local law enforcement, it wasn’t just a capitulation to conservatives, but:

“part of a long-term liberal agenda, one that reflected a belief that federally subsidized police recruitment and training could become racially fair.” (73)

That is, according to Murakawa, Democrats didn’t adopt law-and-order rhetoric to respond to the policy entrepreneurship from Republicans that threatened to swamp them–they explicitly preferred more coercive and punitive state institutions so long as the men and women wielding the riot gear were racially diverse. By the mid-nineties, Bill Clinton wasn’t passing the 1994 Crime Bill because he got dragged there by Republicans, the Democrats lead the way–in fact the Senate version was sponsored by none other than Joe Biden. Murakawa’s conclusion is damning:

In the end, the Big House may serve racial conservativism, but it was built on the rock of racial liberalism. Liberal law-and-order promised to deliver freedom from racial violence by way of the civil rights carceral state, with professionalized police and prison guards less likely to provoke Watts and Attica. Despite all their differences, Truman’s first essential right of 1947, Johnson’s police professionalization, Kennedy’s sentencing reform, and even Biden’s death penalty proposals landed on a shared metric: criminal justice was racially fair to the extent that it ushered each individual through an ordered, rights-laden machine. Routinized administration of race-neutral laws would mean that racial disparate outcomes would be seen, if at all, as individually particularized and therefore not racially motivated.” (151)

James Forman’s book is quite different. Where Murakawa places most of the blame squarely on white Democrats, Forman places his lens on Black politicians in DC, and finds a very different dynamic. From the start, the story of the rise of racialized mass incarceration is a tragic story of reasonable and well-intentioned Black leaders fighting white supremacy and Black disadvantage with reason and evidence. They made deliberate choices that were well-justified and supported by their constituents. And incrementally, they made things worse.

DC’s leaders saw drugs like heroin as a scourge and heroin dealers as race traitors. They saw violent crime rising, and guns playing a major role. And they wanted Black police–because those were good jobs and because Black police officers wouldn’t be tempted to engage in racist practices. So they punished drug dealers and ultimately drug users. They punished violent crime and gun possession. And they did it with a Black-led and majority-Black police force. But still they ended up creating a majority Black prison population in our (I live in DC too) Black-led and Black-staffed prisons and jails.

Forman does the hardest thing in criminology and law: he adopts the stereoscopic vision that can see both from the perspective of those who fear crime and those who bear the costs of policing and prisons.* Black District residents know what it’s like to fear that their family and neighbors will fall into drug addiction or be the victims of gun crime. They also know what it’s like to fear that their family and neighbors will be terrorized by the police or have their lives derailed by imprisonment. And Forman is able to square those stories: see the victim’s fear and rage with one eye and the perpetrator’s circumstances and his community’s losses with the other. Alone, either perspective gives a flat, two-dimensional image, but together you get depth: three-dimensions of a wicked problem where values are always at stake but a way forward is possible.

The difference between Forman and Murakawa is that where Forman wants to tell a careful story about wicked problems and their double-binds, Murakawa seems to want to show up liberals (including Black liberal elites) as self-undermining and doomed to failure. This is Afropessimism at its best and worst: any efforts at racial equity are perverse and doomed to failure. I find such arguments deeply challenging when they come from non-white authors, which is why it’s important to me to think seriously about what a perversity argument is doing.

A perversity argument is any argument that claims that when we try to do a thing we believe is important, we will fail and make it worse, falling further behind as we try to move ahead. The actual use of perversity and unintended consequences arguments are often justified by some of the available evidence, as well as some of the speculative hypotheses: try to make someone love you and they will feel manipulated; create a minimum wage to help the poor and you’ll increase unemployment; try to reform the criminal justice system and you’ll just make it stronger and more pervasive; tell someone they’re wrong and they’ll sink even deeper into their error; try to engage in affirmative action to reverse racial discrimination and you’ll entrench stereotypes of inferiority.

Margo Schlanger’s 2015 review of Murakawa’s The First Civil Right has stuck with me for a while in part because I still occasionally hear people pushing the Murakawa line that Democrats and liberals are primarily responsible for mass incarceration, and thus can’t be trusted to reverse it. Schlanger has a great reading of Albert Hirschman’s work on reactionaries, radicals, and academics and our enduring love of perversity arguments:

Indeed, perversity arguments are appealing not only to reactionaries and the left-of-liberal left but to academics, irregardless of ideology. As Hirschman says, a perversity argument “is, at first blush, a daring intellectual maneuver. The structure of the argument is admirably simple, whereas the claim being made is rather extreme.” Perversity arguments are counter-intuitive, attention-grabbing. These are attractive characteristics for someone trying to stand out in a crowd of monographs. And sure enough, the attack on liberalism as perversely harming the disempowered has become quite fashionable in criminal justice in particular. Bill Stuntz is its most well-known (and least radical) author, but structurally similar claims have sprouted up all over, usually from the far left. These are arguments that prison conditions litigation causes an increase in incarceration, Miranda rights cause increased arrests, and so on. The claims are empirical—A caused B—but the arguments are usually a combination of ideological and hypothetical.

Perversity arguments feel smart and daring. They make you feel like you’ve seen a secret truth. But they also work to disempower and disengage. They paralyze us with fear, uncertainty, and doubt. Every step in the minefield of unintended consequences and backlashes is probably doomed, so the only safe thing to do is stand still. From the perspective of perversity helping hurts, loving hates, attacking strengthens, and truth-seekers lie. Nothing is what it seems, and everything must be viewed through a hermeneutics of suspicion that ends with a kind of paralysis or status quo preference.

But at the same time… sometimes everything is not what it seems. Sometimes our well-intentioned efforts do make things worse. If Forman is right, DC’s leaders were facing real crime problems in need of real solutions, and they built a tidy mass incarcerated city without ever seeking to do so. And his chapters on DC’s responses to gun violence, especially, strike me as importantly relevant to current discussions of gun control in the wake of the Parkland shootings.


*Another book that manages the “stereoscopic view” well is Danielle Allen’s Cuz.