Policymaker Ignorance as an Empirical Problem
The knowledge relevant to policymaking is publicly available
The problem of policymaker ignorance is amenable to a degree of empirical analysis, unlike the forever vexed problem of whether policymakers are more motivated to pursue goals in their own interests rather than objectives in their constituents’ interests (or, what might sometimes be the same thing, goals that constituents demand), a problem I have dubbed “the problem of policymaker incentives.” We can never get inside the heads of policymakers to determine their reasons for acting, their motives, motivations, or incentives, but we can determine, up to a point, what policymakers do and do not know, and whether their knowledge is adequate to the deliberate realization of some policy goal. There is no need to “get inside the heads” of policymakers to get a rough grasp on their knowledge.
This is because the knowledge available at any given time to be applied in the pursuit of policy goals, whether constituent-minded or not, is publicly available. To be ignorant in the sense that is relevant to the problem of policymaker ignorance is not necessarily to be stupid or miseducated, it is not to be cognitively faltering, as Biden obviously is, or crude, cruel, and insensitive to facts, as Trump clearly is, it is to lack knowledge of the causal structure of society adequate to bring about particular states of affairs.[1]
It is the business of the social and behavioral sciences to theorize about the causal structure of society and to gather data concerning the current state of the causes of social phenomena. The knowledge of the required social-causal sort that exists at any given time is not hidden away in a private vault or befogged in the mystical obscurantist writings of a secret society, or accessible only to the privileged members of some conspiratorial sect, it is available to anyone who would care to seek it out in the many emissions of the various social and behavioral sciences.[2] Of course, this is not to say that all such emissions are relevant for policy purposes. The point is that if knowledge of the requisite causal-social kind exists at all, it must come from the social and behavioral sciences. Given that these sciences operate entirely in the open, if such knowledge exists, it is publicly available.
None of this is to deny that governments and the policymakers that constitute them tend to be secretive, but it is to doubt the policy relevance of the secrets that governments try to obscure from public view.
To get a grasp on whether, to what extent, and in what ways policymakers might be ignorant regarding some potential policy goal, it is necessary to compare the knowledge required to deliberately realize the goal with the relevant knowledge available to policymakers.
Determining the knowledge requirements of a particular policy goal requires some hypothesizing about, in particular, 1) theories of relevant causal structures, 2) data concerning the state of relevant causes, and 3) practical capacities required to control events, adequately to deliberately realize the goal.
The next step of the analysis involves comparing these hypothesized knowledge requirements with the public knowledge available to policymakers.
If it is discovered that the publicly available knowledge falls short of the required knowledge, then the next step of the analysis involves further hypothesizing about the consequences of this shortfall, in particular, about the possibilities for spontaneous realization of the goal despite the nature and extent of policymaker ignorance. What are the prospects that policymakers will learn the missing knowledge? In what ways might the goal be spontaneously realized and what are the prospects for such spontaneous realization despite policymakers’ relevant epistemic burdens?
The relevant question for present purposes is whether the secrets that governments keep are very likely to close any gap that might exist between the knowledge required to deliberately realize some policy goal and the public knowledge available to policymakers. Could such secrets ever be adequate to compensate for the apparent ignorance of some set of policymakers with regard to some potential policy goal?
There are reasons to answer this question in the negative.
First, such secrets are not likely to take the form of theories of society’s causal structure that are unknown to the broader social-scientific community. How could governments develop such secret theories in the first place? How could they keep the social scientists involved in the development of such secret theories from discussing them with others? Most mysteriously, how could governments surreptitiously test these secret theories in the ways and to the extent required to have any confidence in their policy implications? Are we to imagine clandestine government-funded social science think tanks and laboratories the employees of which devotedly hold to their non-disclosure agreements?
Second, such secrets are surely not of the form of special powers or unique capacities unavailable to non-policymakers. Policymakers are not conjurers or wizards. They are not expert players of multidimensional chess. They are normal, cognitively and practically limited, human beings, like everyone else. Moreover, government is not some magical machine capable of superhuman feats. Governments try to realize goals through the emphatically unmagical mechanisms of legislation, implementation, and administration. No one who has spent an afternoon at their local Department of Motor Vehicle office can think much of the powers and capacities of government bureaucracy.
So, if they typically cannot take the forms of either clandestine theories or preternatural powers, governmental secrets must most often consist of data – facts – that are obscured to an extent from public view. Without theories to explain such facts and without the capacities necessary to implement and administer a related policy to deliberately realize a relevant goal, such facts in isolation are likely to be of limited instrumental value for policy purposes.
Consider a concrete example recently in the news, the “Chinese spy balloon” fiasco. According to President Biden, only the first of the four objects that entered American airspace over the course of a few days in early February was a surveillance balloon operated by the Chinese government. The other three objects shot down by the American military were more likely owned by private companies or research institutions than the Chinese, or any other, foreign government.
I think it’s safe to assume, as some have suggested, that the Biden administration knows more than it is letting on about this series of events, that the American government is trying to safeguard its secret knowledge of various relevant facts. But, even if this is right, without a related theory to explain these facts and the capacity to do something about them, such secret knowledge would seem effectively useless for policy purposes.
Governments keep secrets, obviously, but it's not obvious that they know what to do or are capable of doing much with them.
[1] This being said, I think it is reasonable to regard stupidity, miseducation, senility, insensitivity to facts, etc., as markers of the kind of lack of social-causal knowledge that constitutes policymaker ignorance. I tend to think that one should never put faith in policymakers’ ability to purposefully bring about states of affairs in the social world. One should positively believe in policymakers’ ability to bring about some particular state of affairs only if the empirical evidence shows that the same or similar policymakers have realized the same, or a similar, state of affairs in the past. Absence such evidence, though it may seem intuitive, instinctive, or reasonable to place a degree of confidence in policymakers who are smart, well-educated, cognitively astute, and sensitive to facts, there can never be a reason – absent the required empirical evidence – to positively believe that stupid, poorly educated, mentally enfeebled, or unprincipled policymakers possess the causal knowledge required to realize some particular state of social affairs.
[2] Social-scientific knowledge may not be the only kind of knowledge required to make effective policies. Depending on the particular social problem to be addressed through policy, knowledge from the humanities, arts, and natural sciences (not to mention, perhaps just plain ol’ “common sense”) may also be required. Of course, this doesn’t alter the point about the public availability of any knowledge that might be applied to the making of effective policies.