Social engineering: being ‘libertarian’ won’t magically make it work


Social engineering is widely rejected by libertarians and classical liberals. Like all coercion, it shuts down criticism by forcing people not to try alternatives to the decision being ‘engineered’ – either by prohibiting them altogether or by making other decisions more expensive. As Hayek wrote in the context of competition between different social standards and forms of organisation: “It is only when such exclusive rights are conferred on the presumption of superior knowledge of particular individuals or groups that the process ceases to be experimental and beliefs that happen to be prevalent at a given time may become an obstacle to the advancement of knowledge.” (1)

The tax break for married couples, recently ruled out by David Cameron in the March budget, would have been a salient example. Advocates of this sort of social engineering are likely to point to the slew of advantages correlated with the practices they want to encourage, especially in the case of marriage or religion.

Libertarians might respond in one of a number of ways. They might argue, in Hayekian vein, that it is wrong to assume the superior knowledge of policymakers about how people should live their lives. They might reject the very premise of a tax break, opposing tax and therefore tax breaks as such. Or they might point out to our conservative social engineers that the correlation of social advantages to a given practice does not imply a causal relationship.

I fear that most of these responses are unlikely to satisfy our social engineer. It is rare nowadays that political debate can take place on an elevated enough footing to give voice to arguments about natural rights and the legitimacy of taxation itself. And while study after study correlates certain practices with social advantages, the fact that this is a fallacious form of inference—that by dint of correlation, baby names might as well cause housing bubbles—suggests that no one is really inferring causation from correlation. There is a reason marriage or religion is picked out as the causative factor, to the exclusion of anything else. There is an underlying theory about the value of those things in themselves—and that is what would need to be refuted, not the declared fallacious argument.

Another shortcoming of the libertarian response is that it generally does not rule out non-coercive social engineering. The libertarian political scientist Charles Murray advocates just this, arguing that members of the ‘new upper class’ in America should “voice their disapproval” of people who have children out of wedlock. As well as highlighting the correlation of marriage with stable families, higher earnings and other advantages, people who want to promote marriage in this way have some idea of the traditional significance of the institution. Traditionally, marriage is a promise to be honest, caring and loyal to someone, and the importance of such promises when made in marriage vows is greater than a mere verbal contract. Hence, marriage is more conducive to better and longer relationships. So, what is wrong with private groups campaigning for it?

Consider that it is only by tradition that marriage has such significance. As a fellow Backbencher writer lamented yesterday, there is nothing inherent in it that inevitably causes people to be more loyal or honest to each other.  I for one will not rule out the possibility that there is value in marriage as a tradition. But traditions arise spontaneously. They are a sophisticated interplay of inexplicit ideas. Misunderstanding is common enough when we are trying to convey explicit theories. If I persuade you of an idea, with an argument, it will not take the same shape in your mind as it does in mine; it will be your own unique ‘version’ or interpretation of the idea. It is impossible to predict the content of this interpretation, as it involves the growth of knowledge—which is unpredictable by definition—for if we were able to predict it, we would already have that knowledge.

The same is true of attempts to promote traditions. But there are further difficulties in this case. To continue with the example of marriage: there is the act of getting married on the one hand, and the significance of that act on the other. We can persuade people to marry, and we can even do this with explicit arguments, but this does not mean people will interpret the act as having the same significance with which it is traditionally associated. It is impossible for a deliberately propagated tradition to be the same as the one that had been adopted spontaneously. As such, there is no way to ensure that people who are persuaded by our voluntaryist-social-engineers will interpret marriage in such a way that it has the same benefits as the tradition spontaneously conceived. Perhaps it will work. But, just as likely, perhaps they will be blasé about it. Or perhaps they will interpret it as boxing them into a corner, and it will become a source of resentment in their lives. Crucially, if these ideas are inexplicit (as traditional knowledge generally is) there is no way for ‘wrong’ interpretations of marriage to be corrected.

Misunderstanding is to social engineering as the law of unintended consequences is to statist intervention; and in both cases, mechanisms for error correction are their saving grace. Social engineering should hence confine itself to explicit moral arguments; only then can the inevitable misunderstandings be identified and corrected. Endeavours to promote practices that are merely linked by tradition with other advantages are quite as futile (if not quite as harmful) as social engineering by governments.


1. Hayek, F.A. The Constitution of Liberty. New York: Routledge (2007), p. 33.


Liberty Fitz-Claridge is a UCL undergraduate who also writes in a personal capacity at The Taints of Liberty.


Please enter your comment!
Please enter your name here