Thanks to Vincent Chow, Benjamin Kolb and Yoka Krämer for their useful comments!

Effective Altruists have the ambition to affect the course of human civilisation in the very long run. We want technology to contribute to the flourishing of sentience rather than to its demise. The trajectory of human civilization will not be a matter only of technical solutions but will crucially depend on the collective wisdom that will bear on humanity’s actions in the face of the existential challenges to come. In this post, I am arguing for giving more consideration to education as a key way of improving the collective rationality and hence the collective decisions of societies. I am advocating to look into or advocate for changes in the contents of education. I am not advocating for societies to spend more on education. Concretely, I am thinking about a much larger focus on philosophical, psychological, political, microeconomic and statistical education in high school, which, for lack of a better word, I will abbreviate as rational education.

Previous discussion

This post is not the first one on the topic in the EA forum. This post provides some evidence on the benefits of philosophical education. This post is about "epistemic progress" but does not focus specifically on education targeted at the general population. This post presents some experiences with teaching EA ideas in high school. Changing the contents of education to foster rationality otherwise does not seem to be a frequently considered cause. I did not find much discussion that argued against it. In the current post, I would like to strengthen the case for looking into this further, providing some arguments for the crucial importance of raising the level of rationality of the broader public. I am not providing more than speculation on the impact education can have on rationality.

• Education as a vehicle for spreading values: A clarification
  There is some past discussion about spreading values. I am not primarily arguing here that one should lobby for education to directly propagate particular values. Exposure to ethics could be beneficial by leading to more reflected values or by inducing individuals to give more consideration to moral arguments - without directly promoting one particular set of values. For example, I think that many might agree that the welfare of future generations is not worth less than that of present humans as long as the practical implications of such views on people’s own lives are not too salient. Exposure to explicit moral deliberation could align individuals’ actions and practical views with the idealistic values they might have upon reflection. In this vein, rationality and morality (by which I mean the extent to which people are guided by explicit values rather than emotion, instincts and impulses) may not be orthogonal.

A worsening mismatch between the power and wisdom of humanity

Humanity’s advancing technological capabilities creates a dangerous discrepancy between its power and its maturity. Evolution has not prepared mankind for the modern problems of a technologically advanced and globalized world. The sphere of influence of a modern human has expanded much more than her moral circle as a result of technology and globalization: Individuals can take actions or decisions which affect people in distant locations in space and even time, but evolution hasn’t endowed with the moral instincts or the intellect adequate to such responsibility. The immature handling of climate change may only be the first example of humanity’s technological might outgrowing its wisdom and ability of acting collectively. If technological progress keeps creating existential problems, it is hard to see how humanity will handle these in the long run at its current level of rationality. Future problems could be trickier than climate change, with less time, or more ambiguity, or larger sacrifices to be made, making it harder to convince everyone of its urgency. 

Policy-makers will be making ever more consequential decisions. How will we find the right path as a society in which most people don’t even know the distinction between consequentialism and deontology and where people rarely think in terms of expected value? This imbalance may become a constant source of risk to the development of civilization. Therefore, unless one thinks that the way to go is to try to move from democracy to technocracy (which may be very hard to achieve), the rationality of common citizens becomes a central consideration.

Why it matters

To some extent, the solutions to humanity's great challenges are technical in nature. For example, cheap sustainable energy technologies make it much likelier that we will end climate change. AI safety technologies will reduce the probability of advanced unaligned AI becoming a problem. But to a large extent, the future will be determined by the ability of society to make collective decisions of high quality. Many political decisions that decide about the fate of humanity are likely to be salient to the wider public because they require trading off interests of some citizens against the interests of the future and are therefore (at least in democracies) strongly influenced by public opinion rather than expert opinion. The impact of general ideas from academia or EA will remain limited until they are discussed by a significant fraction of the population because many of the big decisions with the big impacts are not made by academics and intellectuals. They are made by career politicians in the executive or legislature who are largely driven by public opinion because big decisions often involve high stakes and the right choice is not in everyone’s interest.

Many decisions that would benefit the future require a sacrifice in the present. Currently, public discourse hardly features explicit considerations of values and tradeoffs. The consequence is that discussions on climate change mitigation end as soon as they touch on people’s current habits and living standards – although this implies selfish value systems that, when stated explicitly, many people would not defend. Education can supersede selfishness probably only to a limited extent but at least it can make it harder to hide selfishness behind dubious arguments. Besides, education can expand students’ horizons and make them aware of impactful career options that are aligned with their self-interest, thereby doing good without requiring sacrifice.

Some EAs see promise in improving the rationality of politicians. I share this view because politicians’ decisions probably depend on their own views to a large extent wherever policy has low visibility. But many of the most consequential decisions are highly visible to the public and the incentives faced by politicians (or which politicians will be in power) will be largely decided by public opinion and public rationality, not by the leader's opinion and rationality.

How curricula could be improved

Some examples of things I would like to see taught to many more students:

• Philosophy
  • Ethics: Deontology versus consequentialism, how to deal with uncertainty, the moral weight of other species, the moral weight of future sentience
  • Some basic practical epistemology (e.g. conspiracy theories with real world examples and analysis why not to believe in them, how to come up with theories and beliefs) and logic
• Psychology
  • Common biases and fallacies
  • Debating (e.g. recognising dishonest conversational tactics, and not how to be most persuasive!)
• Microeconomics
  • some basic microeconomic concepts such as efficiency, moral hazard, adverse selection and basic game theory (prisonners’ dilemma)
• Statistics
  • Basic statistics: expected value, variance, distributions
  • Bayes’ Theorem
  • Some statistics that is fundamental to all applied sciences: Statistical testing and significance, publication bias, basic statistical modelling, overfitting, underfitting, bias-variance tradeoff
• Politics and history could maybe receive more attention.

Some of these things are already taught to some high school students – but only in passing and without truly making them relevant to one’s own experience of the real world. Couldn't it be quite impactful if these subjects were taught as core subjects for many years in a way that directly relates their contents to how to form positive and normative beliefs?

Can education affect rationality?

Education interventions seem to have a low standing in EA. 80,000 presents 5 reasons not to go into education. But these apply mostly to education interventions whose aim is to improve the economic outlook of students or simply to improve the academic performance at conventional objectives of basic education (such as reading and maths). The argument that education is neglected (because it already receives a lot of public funding) is not applicable either since I am not arguing for more resources to go to education in general but to consider advocating for shifting these resources from, say, science education to philosophical, political, etc. education.

One might be pessimistic about how many people would be receptive to the abstract concepts of moral philosophy. One might say that the moral progress that has apparently been made in recent history can be reduced to the economic progress that put people in a position to worry not only about themselves. Nonetheless, it seems implausible to me that, if philosophy, psychology etc. were taught as core subjects for the entire high school period or longer, they would not affect some people’s thinking. I believe that current education already affects people’s values and thinking a lot. Wouldn’t norms and common beliefs in a society where children receive no political or moral education at all be quite different from what they are now?

What rational education can achieve certainly deserves some proper research. I am just making the case that this research would be valuable from an EA/longtermist perspective. The aforementioned post already cites some studies documenting the impact of philosophical education.

Tractability of lobbying for curriculum change

Since I am not an expert I don’t know how tractable lobbying for more rational education would be. I will only note that, if the the benefits educational reform  are large, this sets the bar for tractability lower. If curriculum changes are long-lasting in case of success (i.e. don’t require constant lobbying to maintain for a while), initial lobbying may lead to better collective decisions in the face of upcoming existential challenges for a long time. 

A few counterarguments

A reason against prioritizing curriculum change even when granting the points above would be the immediacy of other problems. If one thinks that there is a high and immediate existential risk which can be solved by targeted efforts, such efforts should maybe be prioritized over (many) other causes. However, it’s not plausible to me that such considerations would entirely rule out investigation and advocacy on this topic since some people are likely much better placed to pursue this rather than efforts to fight the very immediate problem.

It is also possible that the rationality of policymaking is currently more bottlenecked by the voters’ incentives to rationally reflect on their views rather than their ability to do so. (On the other hand, incentives are themselves a matter of values, and education could instill in voters a sense of the responsibility to consider public matters more carefully.)

Education could also become irrelevant if humanity soon handed over decision-making to an aligned AI or if genetic enhancement would soon become tractable.