Mar 31, 2017
What's good is bad
Epistemic Status Reasonably confident. Limits of knowability are discussed throughout, and research cited.

Unconventional ideas are alluring, no? I often find myself suspicious of the pleasure derived from having a core belief debunked, afraid of having capitulated too quickly. That possibility feels like fodder for a strawman of economists or the news media, who are often claimed to confuse the latest piece of evidence for the most true. In modern times, the ability to say ‘I was wrong’ gives one the intellectual upper hand, so it is tempting to make the claim prematurely. Thus, when I heard the following statement - on the potential to enhance human intelligence - I was cautious.

Clever people often think that being more clever is what everybody wants, and that that would be a good thing, but I’m not sure that it is. I think if you look around us, and we think [about] what would benefit humanity, … it’s more basic things that are important.

Though imprecise and qualification-rich, the implication that ‘being more clever’ might not be wise is deeply confronting. Education, along with healthcare and environmental conservation, is part of the trinity of services that are usually untouchable in their consensus of social benefit. One-size-fits-all schooling may be inefficient, and tertiary degrees may be nothing but signalling, but what could be detrimental about the pure act of learning?

The quote belongs to sociologist Anne Kerr, and comes from a debate4 with philosopher Nick Bostrom over whether to artificially enhance human intelligence, should we ever have the drug or implant or massage therapy that would enable us to do so. Admittedly this is a more industrial context for ‘becoming clever’ than my immediate concern, but the two scales of learning lie on a continuum not easily divided. Let me be more systematic.

Conventional thinking views education as an act of improvement. By definition, improvements are good. There are only three criticisms of improvements of which I can conceive.

Below, I will use these three classes as a framework for analysing the arguments against ‘being more clever’. Also note the distinction between schooling (institutionalised learning) and education, which is more internal. Learning is synonymous with education, but not with schooling, which doesn’t always succeed in it’s goal of bringing about education. The difference is important, because the terms are often used interchangeably. Here they are not.

While language is rarely used so precisely, the prevailing assumption of goodness is presumably attached to education rather than schooling, which is only good insofar as it brings education about. It isn’t surprising, then, that all three of the above criticisms have been made of institutionalised education. To fill you in:

The topic of discussion is not whether schooling, but education, can be a bad thing. This poses a problem. On a societal scale, education can only be measured indirectly—through schooling. It’s difficult to discern how much knowledge and know-how resides in a group of people, but easy to count the number of PhDs. With no way to directly measure ‘learning’ and only contrived, reductionist methods of quantifying ‘social benefit’, is it possible to isolate the effects of education? I don’t think so.

There are myriad reasons why education should bring about no end of social benefit. More able people mean that every organisation is better organised and run. Cleverer citizens mean a more focused, more accountable and better prioritised government. Intelligence gives us the ability to grapple with more nuanced social issues, and makes it more difficult to ignore arguments that, while inconvenient, form valid opposition. With knowledge comes greater awareness of the fortuitousness of our positions in society, and a sense of responsibility to help those people and ecosystems less fortunate. So on and forth.

Yet there is clearly a disconnect between the speculated benefits and our ability to observe them, which limits how empirical any discussion on the topic can be. That acknowledged, I shall attempt to steelman the case against education.

Opportunity Cost

Let me introduce you to someone, who I’ll call Geoff. Geoff works as a doctor on a remote, jungled archipelago. He is the sole medical professional available to 20,000 people, and does his best with scant access to medicines, sterile facilities or self-cloning machines. There are no specialists to whom he can handball patients. You’ve probably seen a TV documentary about someone like him, with a journalist and camera trailing along on patient rounds. Filming, that is, from a second dinghy they brought in for the shoot.

Though fully qualified, Geoff’s marks in an Australian medical school were mediocre. But that has no consequence here. The majority of Geoff’s time is spent stressing the importance of hand-washing, and dispensing dysentery tablets. You don’t have to be an exceptional person to do exceptional good.

No matter which context we choose to learn in—how well-intentioned or outcome-oriented our approach to learning is—the undertaking will use time that could be spent implementing knowledge we already have. This is a flaw in any justification for altruistic study. I suspect the Pareto principle might be at work: learning 80% more stuff only enables you to add 20% more value, or similar.

Certainly there’s something off about wanting to do your social service from the comfort of a world class university. Personally, I often feel as though a task has to operate at or above some threshold level of abstraction before it seems important enough for me to engage in. This isn’t a redeeming trait. Transcendental thinking is fun, but intellectual humility is needed.

On the scale of societies, education and research can be viewed as benefiting future populations, while implementation of existing ideas benefits those alive today. If you think that future populations will be better off than the current one, it ethically makes sense to prioritise the alleviation of today’s suffering over tomorrow’s. But the two courses of action are not mutually exclusive. It’s an unanswerable question whether a bias towards the intellectual is, through opportunity cost, damaging society. I’m open to being convinced that through some mechanism of societal evolution we are striking more or less the right balance between knowledge progression and knowledge implementation.


Admittedly these first two classes of criticism—Scope and Opportunity Cost—have a degree of overlap; both speak to a notion of trade-off. As we have seen, one such trade-off is that of improvements in the future against improvements in the now. Another is the natural inequality that arises when demographic groups have differing access to the resources required for learning: time, instruction and inclination. With the proliferation of the web, quality didactic tools are in most cases easy to find, but time and inclination tend to be more readily available to the rich than the poor.

Unhelpfully, economics presents us with two contradictory frameworks concerned with differing desires to learn. The first is that of time-inconsistent behaviour, which posits that people heavily discount the future utility they would gain from beneficial investments of time (such as education) that they could undertake in the present5. On the other hand, the notion of rational ignorance undermines the claim that citizens should educate themselves on affairs of government to more accurately represent their interests when voting; the time-cost of learning might simply outweigh any resultant payoff. While conceived in a political context, the basic principle could apply to any flavour of education. If nothing else, these two theories show that there are cogent arguments for spending both more and less time learning, with the ‘rational’ outcome depending on context. The more time an individual requires to satisfy basic needs, however, the less important hours spent on education will seem.

Note also that many of the personal benefits are rooted in a form of social relativism. This suggests that in spite of the personal rewards of education, it may be a futile rat-race for society to collectively partake in. Schooling’s undisputedly high financial returns, which occur even in countries with high-cost colleges like the US6, are probably due only to the scarcity of such qualifications. Similarly, the more we know and the more things we know how to do, the easier it is to find satisfying justifications for our own [false] superiority over those who don’t share the privilege. Thus the further down the road of learning we are, the more biased we will be towards continuing the journey, and in doing so may be perpetuating a social construct which merely reinforces existing economic inequalities.


For those of us who intrinsically enjoy learning the benefits are easy to argue. It allows us to weave more coherent narratives about the world, reducing uncertainty. It enables us to put daily inconveniences into perspective, and weather them more cheerfully. It increases the amount of change we are able to personally effect, supporting self-worth. Not everyone gets this, though.

And it is foolish to claim that education is always self-gratifying; one can equally argue that a glut of information leads to fettered decision-making, or that the most common subjects of inquiry are unpleasant. It’s a cliché: the philomath, having achieved omniscience, becomes crippled by a constant awareness of humanity’s bottomless imperfections. When I emailed Anne Kerr about her quote from the debate, she replied

[Regarding] how people who are considered or consider themselves to be intelligent operate, I think there is a lack of evidence that (i) they are any happier or content than other people (and thereby in a position to bring social benefits to other because they are happier/content) and (ii) that intelligent people are intrinsically more likely to act in a way which benefits society by virtue of their insights into what society needs or how society might benefit from their actions.

Absence of evidence is not evidence of absence, but it is suggestive, and I find the counter-argument unconvincing. The original discussion bypassed the rub, but on querying Bostrom sent through a paper of his: The Reversal Test: Eliminating the Status Quo Bias in Applied Ethics. The test (in my own words) is as follows.

When James asserts that us becoming smarter would make society worse off, you should reverse the situation in question and ask him if us becoming dumber would also make society worse off. James will likely agree, at which point the onus is on him to justify why we are currently at the optimal level of intelligence.

If sidestepping the responsibility of proof is the best one can do, one’s line of argument can’t be all that strong. Even if this is an indictment on the provability of one’s assertion and not on its veracity, one is still guilty of misrepresenting one’s level of certainty. Hence, ‘unconvincing’.

That said, I too am guilty - of a pretense of rationality. Whatever revelations have been unearthed in this discussion, I’m not sure that I could ever abandon learning to any meaningful degree. There are some truths—be they boring, uncomfortable or inconvenient—of which people simply don’t want to know. A reflexive demonstration: I originally pitched this piece to my university’s student magazine, because surely students of all people would want to be made aware of the possible downsides to education? The editor disagreed.

  1. Taken from this slidedeck by economist Bryan Caplan, who has an upcoming book titled The Case Against Education. He uses the term education like I use the term schooling. 

  2. See, for instance, the corpus of sociologist Richard Breen. Start with: Breen, R. and Chung, I., 2015. Income inequality and education. Sociological Science, 2, pp.454-477. 

  3. Ram, R., 1989. Can educational expansion reduce income inequality in less-developed countries?. Economics of Education Review, 8(2), pp.185-195. 

  4. Audio from the debate, which was held at LSE, can be downloaded here

  5. The experiments of Kahnemann and Tversky show that not only are we time-inconsistent, but are inconsistently-time-inconsistent. See the Wikipedia article on Discounted Utility

  6. Full transparency: I haven’t read extensively on this subject, but am relying on Bryan Caplan’s characterisation of the literature in this talk, which implies that almost every cost-benefit analysis ever performed has concluded that schooling is monetarily worth it in the long-run. 

Thank you to Anne Kerr, Nick Bostrom and David Stuckler for their correspondence on this topic.