Here I will attempt to refine some remarks that I recently made on Twitter, arguing that libertarians ought to be wary of the general phenomenon of public policy “wonkishness” — which I’ll define very loosely as a concern with offering practical public policy reforms or proposals based on statistical and empirical evidence (the kind of thing that mainstream think tanks tend to do). I have nothing against think tanks per se, believing as I do that educating people in libertarian principles is the single most important thing we can do to advance freedom in practice. Nevertheless, I happen to find a particular brand of wonkishness quite objectionable from a fairly plumb line, uncontroversial libertarian perspective. Fundamentally, at its core, wonkishness is an example of (perhaps more accurately a symptom of) what Friedrich Hayek famously called “the fatal conceit,” the idea that policymakers can deliberately design and optimize societal institutions. Furthermore and relatedly, the policy wonk, as such, is defined by his overreliance on and faith in statistical information and economic modeling, another example of the planner’s conceit identified by Hayek.
Anyone who has spent any significant amount of time reading think tanks’ policy papers knows well that every side of every debate has its charts, statistics, and studies, that statistics are inherently the language of spin, susceptible to countless interpretations, open to being massaged and manipulated to fit with virtually any argument one wants to pursue. All of this is to say that, in attempting to appear scientific, policy wonks are in fact engaged in a rather perverse bastardization of science, assuming that their numbers and data sets are revealing truths that they couldn’t possibly reveal. We can’t even draw a clear, hard-and-fast line that comprehensively and one hundred percent accurately defines aggression or invasion. How indeed do our libertarian policy wonks think that they know which laws are good and which are bad from a libertarian viewpoint? Many cases, of course, are easy enough, but when we see self-described libertarian policy experts supporting the baleful Trans-Pacific Partnership, for example, we might wonder about the “scientific” methods of the world of wonkery. One of the reasons I was first attracted to libertarianism is that is seemed to me that, as a general philosophy, it wanted to avoid this kind of simplistic and deeply foolish appeal to statistics. At its best, it wants to avoid politics too.
Nevertheless, we can perhaps understand the wonkish impulse. Libertarians want to sit at the “cool table,” as it were, to be taken seriously by elites in the toxic, corrupt slough of public policy. But libertarians ought to know better — ought to know that statistics are an integral aspect of political dissembling, the attempt to make the practice of power appear legitimate and scientific, for the most vile attempts at tyranny always appeal to science. “The word ‘statistics,’” teaches Deirdre McCloskey, “was a coinage of German and Italian enthusiasts for state action in the early eighteenth century, pointing to a story of the state use of numbering. Then dawned the age of statistics, and everything from drug incarcerations and smoking deaths to the value of a life and the credit rating of Jane Q. Public are numbered.” Thus are statistical arguments rather like legal arguments, their movements ever a shrewd, distracting dance, games with numbers just like games with words. I’m often struck by just how easy it is for lawyers to devise cogent legal arguments for just about any position — strong legal reasoning as ever a cover for weak, philosophically untenable fundamentals. Likewise, if a court’s opinion made good arguments, then the dissent made better arguments, and vice versa, and on and on. It’s all so much wasted cleverness, as so often are mathematical and statistical exercises. Without guiding principles, the words and numbers don’t tend to matter or mean much. Anything could be made plausible and convincing, could move you to believe, even if only for a moment, that the other side just had to be wrong. And it’s important to remember that any reasonably educated person possesses the mental dexterity to engage in these kinds of intellectual pirouettes.
Often attributed to Mark Twain and Benjamin Disraeli, the following quote gives voice to my exasperation at the policy community’s absurd overreliance on statistics: “There are three kinds of lies: lies, damned lies, and statistics.” Weak arguments and statistics are natural allies, creating an appearance of scientific rigor through carefully executed adjustments. Insofar as it leaves the uninitiated and credulous dazzled, politics and its agents simply love this process of lying with artful half-truths. One of the key and central insights of libertarianism is that we simply cannot legislate a free society into being. As Stephen Pearl Andrews writes in The Science of Society, “Mankind legislate themselves into confusion by their effort to escape it.” Statistics are the language of modernist state control, too often begging the question by presuming that some group has a right to rule, and that therefore the only question remaining is how to rule smartly or scientifically (or at least make it seem smart and scientific). Libertarian wonks are playing into the state’s hands — and indeed perhaps many of them want to do just that; I don’t pretend to know. What I am rather more certain of is that libertarians who desire to exercise, or even to steer, governmental or legislative authority are, by definition, no longer really libertarians. Excommunication, though, is not the goal. Just be wary of wonkish libertarianism, which is really just political libertarianism.
 Andrews similarly writes, “[D]iversity reigns throughout every kingdom of nature, and mocks at all human attempts to make laws, or constitutions, or regulations, or governmental institutions of any sort, which shall work justly and harmoniously amidst the unforeseen contingencies of the future.”