Politicians, partisans, and pundits were surprised and traumatized by the election of Donald Trump as President of the United States. Anger at experts seems to have contributed significantly to his victory (Easterly 2016). Brexit was led in part by Michael Gove, who exclaimed, “I think the people in this country have had enough of experts” (Lowe 2016). Whatever one’s opinion of Trump or the European Union, ordinary people in Western democracies have cause to be angry with experts. The Flint water crisis is an example.
On April 25, 2014, the city of Flint, Michigan changed its municipal water supply in a manner that produced impotable brown water (Adewunmi 2017). “Flint water customers were needlessly and tragically exposed to toxic levels of lead and other hazards” (Flint 2016, p. 1). “Flint residents began to complain about its odor, taste and appearance” (Flint 2016, p. 16). “On 1 October 2015, 524 days after the switch to the Flint River, the Genesee County Health Department declared a public health emergency and urged Flint residents to refrain from drinking the water” (Adewunmi 2017). On January 24, 2017, Michigan state environment officials indicated that lead levels in Flint water no longer exceeded the federal limit (Unattributed 2017). The journalist Bim Adewunmi says, however, “By the time I left Flint on 22 February this year [2017], the water was still not safe enough to drink directly out of the faucet, according to the politicians and charity workers I spoke to, and the residents’ feelings on the matter had remained at a simmer about it” (Adewunmi 2017).
In March 2016 an official Michigan task force investigating the Flint water crisis found that the Michigan Department of Environmental Quality (MDEQ) “bears primary responsibility for the water contamination in Flint” (Flint 2016, p. 6). The report found that the Michigan Department of Health and Human Services and the United States Environmental Protection Agency (EPA) shared a substantial portion of the blame (Flint 2016, p. 1). The task force’s report chronicles the actions of Dr. Mona Hanna-Attisha and Marc Edwards to produce change, in part by documenting important facts such as elevated lead levels in the blood of Flint children. Noting that “the majority of Flint’s residents are black,” Adewunmi (2017) records the opinion of many that anti-black racism contributed significantly to the crisis. In this tragedy, state experts charged with ensuring water quality instead allowed a malodorous, polluted, and toxic liquid to flow into Flint homes and poison its people. Unfortunately, the Flint water crisis is but one of many examples of expert failure.
In 2009 two judges in Pennsylvania, Mark A. Ciavarella Jr. and Michael T. Conahan, pled guilty to fraud and tax charges in a scheme to imprison children for money (Chen 2009). The case was dubbed “kids for cash.” These two judges were experts. They were experts in the law giving their opinions of the guilt or innocence of children coming before them and deciding what punishments were just. They took $2.6 million in “kickbacks” (Urbina 2009) from two private detention centers to which they sent convicted juveniles. To get their kickbacks, they sent children to jail. In what is presumably one of the more egregious cases, thirteen-year-old DayQuawn Johnson, who had no previous legal troubles, “was sent to a detention center for several days in 2006 for failing to appear at a hearing as a witness to a fight, even though his family had never been notified about the hearing and he had already told school officials that he had not seen anything” (Urbina 2009). Ciavarella and Conahan sent children into detention facilities at twice the state average and seem (at least in the case of Ciavarella) to have declined to even advise them or their parents of the children’s right to an attorney (Chen 2009; Urbina 2009). Their scheme went on for years before they were finally caught and arrested.
Social work provides further examples of experts causing harm to ordinary persons. In the United States, social services can be intrusive and arbitrary. In 2014 a woman in South Carolina was jailed for letting her nine-year-old daughter play unsupervised in a public park that was “so popular that at any given time there are about 40 kids frolicking” (Skenazy 2014).
Another woman reports that her children, “between the ages of 10 and 5,” were taken from her after she was widowed. She chose to leave them unsupervised in the house for “a few hours” at a time while she attended college classes (Friedersdorf 2014). She says officials entered her home and removed her children without attempting to reach her, even though her children knew where she was: “Over the two years during which the case dragged on, my kids were subjected to, according to them, sexual molestation (which was never investigated) and physical abuse within the foster care system. They were separated from each other many times, moved around frequently, and attended multiple schools.” She was subject to arbitrary conditions for the return of her children. She reports, for example,
I was required to allow CPS [Child Protective Services] workers into my home to conduct a thorough “white glove” type inspection. According to the court order, if any of the workers felt anything was amiss, the return of custody would be delayed or denied. I was told to sweep cobwebs and scrub the oven to their satisfaction, which I did, obsequiously.
Her experience led her to conclude that the system “was not about protection, but power” (Friedersdorf 2014).
After Kiarre Harris acquired the legal right to home-school her children, Child Protective Services (CPS) arrived at her door with uniformed police in tow. If early reportage is correct, the CPS workers told Harris they had a court order to remove the children, but were unable to supply the order when Harris asked to see it. She declined to surrender her children without seeing the order. She was arrested for obstructing a court order and her children were moved to foster care (Buehler 2017; Riley 2017). One report (Williams and Lankes 2017) cautioned that “there could be more to the situation” and that Harris “has a history of domestic violence, including using a knife.” Only much later in the article, however, was the substance of this supposed knife-wielding violence revealed: “In 2012, a woman complained that Harris kicked, scratched and gauged her vehicle with a knife. That complaint was classified as criminal mischief. It’s not clear if that was the domestic dispute referenced in the CPS petition.” The “domestic violence” that Williams and Lankes (2017) soberly note seems to have been nothing more than an unsubstantiated claim that Harris keyed someone’s car or otherwise damaged it. So far, The Buffalo News has been unable to find more damning evidence of past crimes or irregularities in Harris’s life.
Even if facts clearly unfavorable to Harris should eventually emerge, it seems unlikely that her arrest and the precipitous removal of the children were necessary or appropriate. It seems far more likely that the removal has been harmful to the children. In this as in other cases, CPS seems to have been arbitrary, imperious, and oppressive. In a post on social media Harris claims that one court document includes the vague remark, “Respondent seems to have a problem with whatever school the children are attending.” She says the same document remarks, “Respondent recently posted a comment on social media ridiculing the school system and people who attend school or graduate from school” (http://thefreethoughtproject.com/mother-arrested-homeschooling-children/.) The quoted contents of the court document seem irrelevant to the charge of neglect, and they seem to reflect more concern for the interests of the public school system than for the children’s welfare.
Home-schooling advocates have identified the pattern seemingly at work in this and other cases. It is, they say, “common for the [affected school] district to not file families’ homeschool notices, then report them to CPS.” Once the child stops coming to class, the school marks them as absent. “Although the child is not ‘absent’ and is being instructed elsewhere, often the school will continue to mark them so, which is why Child Protective Service gets involved” (Hudson 2017).
A British report on family courts (Ireland 2012) that gained national attention at the time of its release found grave deficiencies in the UK system of social services. The report was funded by the Family Justice Council, which is described in it as “an independent body, funded by the Ministry of Justice.” It summarized the evaluations of the psychological assessments submitted to Family Court in 126 cases. Admittedly, the report gives the opinions of some experts of the job done by other experts. Much of it is the credentialed casting aspersions on the non-credentialed. Nevertheless, some of the findings provide a clear indication that state experts have played an arbitrary, obnoxious, and intrusive role in the lives of many residents of the United Kingdom. The arbitrary nature of the psychological assessments being made is suggested by the report’s finding that more than 40 percent of the case findings reviewed failed to adhere to required procedural norms (Ireland 2012, p. 21). Thus, the representatives of the state failed to adhere to state-mandated procedural norms more than 40 percent of the time. “Key findings focus on the fifth of psychologists who, by any agreed standards, were not qualified to provide a psychological opinion, coupled with nearly all expert witnesses not maintaining a clinical practice but seeming to have become full time ‘professional’ expert witnesses” (Ireland 2012, p. 30).
One news report (Reid 2012) provides several horror stories supporting the view that family-court practice in the United Kingdom is abusive and harmful. In one case, “after a woman was found by a psychologist to be a ‘competent mother,’ the social workers are said to have insisted on commissioning a second expert’s report. It agreed with the first. They then commissioned a third, which finally found that the mother had a ‘borderline personality disorder.’ All three of her children were taken away for adoption.” Reid notes two institutional facts that seem to explain why “No other country in Western Europe removes so many children from their parents.” First, there were financial incentives to remove children. “The last Labour government set adoption targets and rewarded local councils with hundreds of thousands of pounds if they reached them.” Although these targets had been “scrapped” by the time Reid wrote, “Social workers now [in 2012] get praise and promotion if they raise adoption numbers. David Cameron is also demanding more adoptions – and that they are fast-tracked” (Reid 2012). Second is secrecy. The 1989 Children Act “introduced a blanket secrecy in the family courts,” thereby encouraging “a lack of public scrutiny in the child protection system” (Reid 2012).
People have suffered from expert error and abuse with state health and environmental experts, state schools, state-controlled or regulated healthcare, and the criminal justice system. Joan C. Williams (2016) is probably right to say that the “white working class” in America “resents professionals,” including lawyers, professors, and teachers, in part because “professionals order them around every day.” Health economics expert Jonathan Gruber, sometimes dubbed the “Obamacare architect,” famously said that the “stupidity of the American voter” was essential to the passage of “Obamacare,” i.e., the Affordable Care Act of 2010 (Roy 2014). Experts advised the US government to send young Americans to die in Iraq because that nation had weapons of mass destruction, when in fact no such weapons were there. Monetary policy expert Alan Greenspan, former head of the Federal Reserve System, reported his “shocked disbelief” over the Great Recession in testimony before Congress. He confessed that the crisis had revealed a “flaw” in his model of capitalism (Greenspan 2008). America’s economic experts were unable to prevent economic crisis. During and after the crisis, many large organizations received bailouts while many ordinary Americans were left with underwater mortgages, unemployment, or both.
The evils of eugenics are an important example of expert error and abuse. As I will note again in Chapter 4, we cannot view such evils as entirely in our past. Ellis (2008) has explicitly called for a “eugenic approach” to fighting crime (p. 258) that would dictate the “chemical castration” of “young postpubertal males at high risk of offending” (p. 255). The experts will tell us which young men are at risk of offending in the future and castrate them as a preventive measure. The Center for Investigative Reporting (Johnson 2013) has found that “Doctors under contract with the California Department of Corrections and Rehabilitation sterilized nearly 150 female inmates from 2006 to 2010 without required state approvals.” They seem to have been more likely to pressure inmates thought to be at risk of re-offending: “Former inmates and prisoner advocates maintain that prison medical staff coerced the women, targeting those deemed likely to return to prison in the future.” A former inmate who worked in the infirmary of Valley State Prison in 2007 “said she often overheard medical staff asking inmates who had served multiple prison terms to agree to be sterilized.” At least as recently as 2010, medical experts in California prisons pressured women to accept tubal ligation as a preventive measure when they thought such women to be at risk of offending again in the future.
It seems fair to conclude that for many people the “problem of experts” discussed in this book is urgent and concrete.
I have noted the destructive role of experts in the lives of ordinary people. I oppose the rule of experts, in which monopoly experts decide for non-experts. I sympathize with the people over the experts, technocrats, and elites. Such sympathies may seem populistic. I fear populism, however, and value pluralistic democracy. Because my sympathies might seem populistic to at least some readers, I should probably explain why, in my view, populism and the rule of experts, at least in their more extreme forms, are equally inconsistent with pluralistic democracy.
Populist rhetoric is often, though not always, anti-expert (Kenneally 2009, de la Torre 2013). Boyte (2012, p. 300) is probably right to say that “Populism challenges not only concentrations of wealth and power, but also the culturally uprooted, individualized, rationalist thinking characteristic of professional systems, left and right.” Populism is usually a revolt against the “elites,” and that term is usually construed to include state experts and technocrats. We have seen Michael Gove disparage experts. The official site of the French Front National has warned against placing “the destiny of the people in the hands of unelected experts” (Front National 2016). The founder of Italy’s “5 Star Movement” has sharply criticized “supposed ‘experts’” in “economics, finance, or labor” who would presume to speak for the movement. The party’s platform, he said, would be “developed online” by “all of its members” and it would be “a space where everyone really counts for one” (Grillo 2013).
Mudde (2004) defines populism as “an ideology that considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ versus ‘the corrupt elite,’ and which argues that politics should be an expression of the volonteé geéneérale (general will) of the people” (p. 543). Populism, Mudde explains, “has two opposites: elitism and pluralism” (2004, p. 543). Elitism “wants politics to be an expression of the views of the moral elite, instead of the amoral people. Pluralism, on the other hand, rejects the homogeneity of both populism and elitism, seeing society as a heterogeneous collection of groups and individuals with often fundamentally different views and wishes” (pp. 543–4).
Bickerton and Accetti (2015, pp. 187–8) describe “populism and technocracy” as “increasingly … the two organizing poles of politics in contemporary Western democracies.” They note, however, that both poles are opposed to “party democracy,” which they define as “a political regime based on two key features: the mediation of political conflicts through the institution of political parties and the idea that the specific conception of the common good that ought to prevail and therefore be translated into public policy is the one that is constructed through the democratic procedures of parliamentary deliberation and electoral competition.” Thus, “despite their ostensible opposition, there is also a significant and hitherto unstudied degree of convergence between populism and technocracy consisting in their shared opposition to party democracy.”
I will argue in Chapters 6 and 7 that knowledge is often dispersed, emergent, and tacit. (It is often, I will say, “synecological, evolutionary, exosomatic, constitutive, and tacit.”) This view of knowledge is consistent with pluralistic (or “party”) democracy. Knowledge is dispersed. Each of us has at best a partial view of the truth. Plural perspectives are thus inevitable and good. In a pluralist democracy, competing partial perspectives on the truth have at least a chance to be heard and to influence political choices. Decisions in a political system – be it populist, elitist, or something else – that override or ignore plural perspectives will be based on knowledge that is at best limited, partial, biased. If knowledge were uniform, explicit, and hierarchical, then we might consider whether it could be best to determine which system of knowledge is the true one upon which all political decision making should be based. In this case, some might seek wisdom in the experts while others might turn to a party or leader embodying popular wisdom, and there would be no “neutral” way to adjudicate the dispute between them. If my more egalitarian view of knowledge is correct, however, then plural democracy is more likely to be the least worst system of political decision making. Thus, my sympathy for ordinary people against elites, experts, and technocrats is not, after all, populistic.
Fear of populism is justified. But we should recognize that the rule of experts is also an “escape from democracy” (Levy and Peart 2017). If we are to preserve pluralistic democracy, all of us in the scribbling professions of scholarship, journalism, and policy analysis should recognize that experts often harass, harry, and harm ordinary people. Examples are legion. I have given a few in this chapter. Popular anger with and repudiation of experts should not be dismissed as irrational fear or ignorant anti-intellectualism. It is all too well justified. There is a problem of experts and it matters.
In this volume, I address the problem of experts. I offer an economic theory of experts. My theory is “economic” because it adopts the economic point of view (Kirzner 1976). It is not a theory of the “economic aspects” or “economic consequences” of experts or expertise. It is a theory of experts on all fours with the theories of philosophers such as Mannheim (1936) and Foucault (1980), science and technology scholars such as Turner (2001) and Collins and Evans (2002), and sociologists such as Berger and Luckmann (1966) and Merton (1976).
In my theory, an expert is anyone paid for their opinion. Here, “opinion” means only the message the expert chooses to deliver, whether or not they sincerely believe the message to be true. If you are paid for your opinion, you are an expert. If you are not paid for your opinion, you are not an expert. More precisely, if you are paid for your opinion, you occupy, in that contractual relation, the role of “expert.” Thus, “expert” is a contractual role rather than a subset of persons. As I will attempt to show, this definition of “expert” creates a class of economic models that is distinct (though not disjoint) from other classes of economic models, including principal-agent models, asymmetric information models, and credence-goods models. Usually, an expert is defined by their expertise. By such a definition, however, everyone is an expert in something because we all occupy different places in the division of labor and, therefore, the division of knowledge. It thus becomes unclear who is supposed to be an expert and who a non-expert. My definition in terms of contractual relations seems to get around that problem. It also avoids the question of whether you are “really” an expert if your expertise is false or deficient. The economic theory of experts developed in this volume does not require us to judge whose expertise is legitimate or scientific or in some other way sufficiently certified or elevated to “count.”
I begin with the nature and history of the problem, which I discuss in Chapters 2–4. There is a large literature on the problem spanning many fields, including philosophy, law, sociology, science and technology studies, economics, forensic science, and eugenics. This literature has not, however, been clearly delineated in the past. While I have not attempted a proper survey, I have attempted to delineate the literature, to identify the main themes of it, and to characterize what I believe to be the four main general theoretical positions one may take. To anticipate, one may take a broadly favorable view of experts or a broadly skeptical view. And one may view non-experts as having in some way the potential to choose competently among expert opinions or, alternatively, one may view non-experts as lacking the potential for such competence. These two broad perspectives on experts and two broad perspectives on non-experts create four general theoretical postures one might adopt toward experts. The great majority of theorists seem to fit reasonably well into one of these four broad categories, notwithstanding the variety of theories to be found in the literature. In Chapter 2 I discuss these four broad categories for the theory of experts and provide exemplars for each.
In Chapter 3 I review two important episodes in the history of the problem. The first is the emergence of Socratic philosophy and its development with Plato, Aristotle, and the Academy. In this tradition, philosophers are experts. The second episode is a mostly nineteenth-century Anglo-American literature on expert witnesses in the law. I will argue that in both literatures the expert is often viewed as both epistemically and morally superior to non-experts. They should be obeyed. Such lionization of experts and expertise is common today as well and is, in my view, inappropriate and unfortunate.
All such arguments seem to find their original in Socratic philosophy. This origin was recently invoked by one defender of experts against populism, British celebrity and physics expert Brian Cox. Commenting on Gove’s disdain for experts, he has said, “It’s entirely wrong, and it’s the road back to the cave” (Aitkenhead 2016). With this clear allusion to Socrates’ cave, Cox is telling us that it is unphilosophical to challenge the experts. He goes on to suggest that experts are superior, being unsullied by parochial interests: “Being an expert does not mean that you are someone with a vested interest in something; it means you spend your life studying something. You’re not necessarily right – but you’re more likely to be right than someone who’s not spent their life studying it” (Aitkenhead 2016). As we shall see in Chapter 3, this view of experts as better and wiser is clearly expressed in the Socratic tradition of philosophy, and again in the railings of nineteenth-century “men of science” against the challenges and supposed indignities they experienced when testifying in court.
Finally, in Chapter 4 I review several recurrent themes in the theory of experts and discuss how they have been addressed in the past. These common themes are power, ethics, reflexivity, the well-informed citizen, democratic control of experts, discussion, and market structure. I have tried to give at least some indication of what choices or strategies might be available for addressing each theme within the context of a theory of experts. Part I of this volume provides, then, a kind of map of the territory occupied by the literature on experts.
The economic point of view I adopt in this volume is, I think, easily misunderstood. I have, therefore, included a discussion of important supporting concepts from economics. This discussion is found in Chapters 5–7. My theory of experts builds on a theory of the co-evolution of the division of labor and the division of knowledge. Vital to this theory is the idea that the division of labor and division of knowledge are not planned. They emerge unintendedly from the dispersed actions of many people who have not all somehow pre-coordinated their plans. The system was not planned, but it somehow coheres and functions anyway. This notion of “spontaneous order” may seem quite strange. For this reason, I suppose, it is easily misinterpreted. It may seem to be a kind of scientific mysticism, to “reify” markets, or to be in some other way absurd or mysterious. I have tried to dispel this sense of strangeness in part through a purposefully silly example of spectators standing up together in a sports stadium. My willfully silly example shows, I hope, that there is nothing absurd or mysterious in the idea of spontaneous order. The idea is surprising, but not strange.
I also consider more serious examples of spontaneous order, including the division of labor. We should not think of the division of labor as driven, somehow, by a grand purpose. It embodies no unitary hierarchy of values. The division of labor has no purpose and serves no particular hierarchy of ends. It is, rather, the emergent and unplanned result of a variety of persons pursuing a variety of potentially inconsistent goals. We can get along, so many of us so well, precisely because we do not have to agree on values. Believers buy Bibles from atheists and the system bumps along tolerably well, all things considered.
In Chapter 5 I also consider the perhaps more fraught ideas of “competition” and “competitive” markets. I have called my approach to the problem of experts an “economic theory of experts.” It may not be surprising, therefore, that I take a comparative institutional approach in which expert error and abuse are more likely when experts have monopoly power and less likely in a “competitive” market for expert opinion. I put the word “competitive” in scare quotes, however, because it easily creates misunderstanding. It may seem to invoke the incoherent idea of a market in which “anything goes” and there are, somehow, “no rules.” As I attempt to show in Chapter 5, any such notion of a rules-free market is incoherent. The “free market” of economic theory is always “regulated” by some set of rules, and the questions are what rule sets have what consequences, whether a given set of rules can be improved, and how best to determine or update the different rules governing different markets. The words “competition” and “competitive” can easily suggest something very different than I have in mind, but I have no satisfactory substitute for them. Therefore, I will use them and hope that the clarification I have attempted in Chapter 5 will minimize misunderstandings.
There can be a problem of experts only if different people know different things. Thus, the division of knowledge must be at the center of a theory of experts. The first clear statement of a general problem of experts, that of Berger and Luckmann (1966), built on a clear statement of the division of knowledge. “I require not only the advice of experts, but the prior advice of experts on experts. The social distribution of knowledge thus begins with the simple fact that I do not know everything known to my fellowmen, and vice versa, and culminates in exceedingly complex and esoteric systems of expertise” (p. 46). One’s theory of experts is likely to go wrong unless it embodies and reflects a theory of social knowledge that is at least approximately correct. What is the nature of the knowledge that guides action in society? How is such knowledge produced and distributed? And so on. In Chapters 6 and 7 I argue that knowledge is often dispersed, emergent, and tacit. I will impute this view to Hayek (1937), although I have also learned from Mandeville (1729), who adopted a more radically skeptical and egalitarian view of knowledge than Hayek. If my broadly Hayekian (or perhaps Mandevillean) view of knowledge is about right, then knowledge is not hierarchical, unitary, explicit, and bookish. It is, instead, generally emergent from practice, often tacit, and embodied in our norms, habits, practices, and traditions. The human knowledge that sustains the division of labor in society is better represented as a vast web of Wittgensteinian language games than as some more organized and codified entity such as Diderot’s Encyclopedia. In Chapter 6 I review the history of thought on the division of knowledge from Plato’s Apology to Mandeville’s Fable. In Chapter 7 I continue the history to the present and suggest that the notion of dispersed knowledge is still not widely understood.
I describe my theory of experts in Chapters 8–11. I call the theory “information choice theory” for two reasons. First, it underlines the fact that experts are economic actors who must choose what information to convey. Second, it highlights the relationship to the economic theory of “public choice.” Like public choice theory, information choice theory assumes people are the same in all their roles in life. Thus, experts are neither more nor less honest or selfish than non-experts. I identify three key motivational assumptions for an economic theory of experts. First, experts seek to maximize utility. Thus, the information-sharing choices of experts are not necessarily truthful. Second, cognition is limited and erring. Third, incentives influence the distribution of expert errors. These motivational assumptions, together with a broadly Hayekian view of dispersed knowledge, support a theory of experts and expert failure in which “competition” tends to outperform monopoly. Expert failure is more likely when experts choose for their clients than when the clients choose for themselves. And (again) expert failure is more likely when experts have an epistemic monopoly than when experts must compete with one another, although the details of the competitive structure matter, as we shall see. Finally, in Chapter 11 I discuss how to design piecemeal institutional change to improve markets for expert advice.
My theory of experts points to an epistemic critique of what is variously called the “military-industrial complex,” the “national-security state,” or the “deep state.” For reasons I explain in Chapter 12, I will adopt the label “entangled deep state.” I sketch this critique in Chapter 12. My critique of the deep state formed no part of my original intent in writing this book, but it emerged organically from my efforts and belongs, I think, in the volume. Unfortunately, my program for designing incremental reform does not seem to suggest how to reform the entangled deep state. Coincidentally, events in the early days of the Trump administration have caused the term “deep state” to be used more frequently in the popular press. It has become a cliché in very short order. Instead of “military-industrial complex,” we may now read of the dangers of the American “deep state.”
The “entangled deep state” is a threat to pluralistic democracy. It is thus an issue of great importance. My critique, however, will be epistemic. Rather than showing how the entangled deep state contradicts or threatens my values, I will show that it produces expert failure.
In contrasting “competition” with “monopoly” in the market for expert opinion, I take a comparative-institutions approach to the problem. I ask which institutional arrangements tend to encourage and enable expert error and abuse and which institutional arrangements tend to discourage and diminish expert error and abuse. We have already seen an example of the importance of institutions. We have seen that expert abuses in British family courts are at least partly attributable to the combination of secrecy of court proceedings and financial incentives for the removal of children. The institutional structure of social services in the United Kingdom leads to needlessly high rates of expert error and abuse in the form of arbitrary and inappropriate removal of children from their families. A different institutional arrangement could produce fewer miscarriages of justice. By emphasizing comparative institutional analysis, my theory is structurally similar to the works of many predecessors, including William Easterly and Vincent Ostrom.
Easterly (2013) rightly rails against the “tyranny of experts” in development. As I shall say again in Chapter 4, he repudiates the “technocratic illusion,” which he defines as “the belief that poverty is a purely technical problem amenable to such technical solutions as fertilizers, antibiotics, or nutritional supplements” (p. 6). Easterly says, “The economists who advocate the technocratic approach have a terrible naïveté about power – that as restraints on power are loosened or even removed, that same power will remain benevolent of its own accord” (p. 6). Poverty is about rights, not fertilizer. “The technocratic illusion is that poverty results from a shortage of expertise, whereas poverty is really about a shortage of rights” (p. 7). Easterly thus adopts a comparative institutional approach to the problem of experts. Where experts have power in a technocratic model, there is scope for abuse and bad results easily emerge. If, instead, the poor are given civil and political rights, better outcomes will likely emerge. Top-down planning cannot match undirected economic development.
Ostrom (1989) adopts a comparative institutional approach to public administration. He criticizes the administrative state. Strauss (1984, p. 583) identifies four “significant features of modern administrative government,” and I take those features to define the “administrative state” Ostrom opposed. The gist of it is that the US federal government includes a complex bureaucracy that is at least somewhat protected from democratic process and exists as a separate – though not homogeneous or unified – power base. The President is “neither dominant nor powerless” (Strauss 1984, p. 583) in their relations with the administrative agencies of the federal government.
The terms “administrative state” and “deep state” identify distinct, but overlapping, phenomena. The term “administrative state” has generally been used in the context of regulatory agencies and has not usually included the military or the intelligence agencies. Moreover, the “deep state” has usually been construed to include nominally private actors such as defense contractors. The two terms have been distinct, but overlapping. They have recently come into popular use with vague, inconsistent, and overlapping meanings.
Ostrom’s critique of the administrative state challenges the traditional view in public administration that sees the state as a unitary actor. In the traditional and still dominant view put forward by Wilson (1887), Ostrom explains, “There will always be a single dominant center of power in any system of government; and the government will be controlled by that single center of power” (1989, p. 24). In the Wilsonian vision Ostrom identifies, strict hierarchy is the only possible form of government. “So far as administrative functions are concerned,” Wilson explains, “all governments have a strong structural likeness; more than that, if they are to be uniformly useful and efficient, they must have a strong structural likeness … Monarchies and democracies, radically different as they are in other respects, have in reality much the same business to look to” (1887, p. 218). In this view of things, trained experts must be protected from democratic pressures. Wilson avers that “administration lies outside the proper sphere of politics. Administrative questions are not political questions” (1887, p. 210). Wilson acknowledges some role for democracy, but the core of governance is technical decision making that only experts are competent to perform: “Although politics sets the tasks for administration, it should not be suffered to manipulate its offices” (Wilson 1887, p. 210). The principle of central bank independence illustrates the idea that the technical decision making of experts should be protected from democratic interference. Today, this sort of thinking informs not only politics and “administration,” but also law, medicine, journalism, and education.
Against this hierarchical view, Ostrom cites the basic institutional fact of “overlapping jurisdictions and fragmented authority” (1989, p. 106) characterizing American federalism. If “Individuals who exercise the prerogatives of government are no more nor no less corruptible than their fellow citizens” (1989, p. 98), then such fragmentation may produce better results than the unitary government imagined by Wilson. Ostrom defends polyarchy against hierarchy. In so doing he proposes a radical change in perspective for the theory of public administration.
My theory of experts expresses a vision similar to that of Ostrom (1989). As I will show, my views on experts are congruent with those of other scholars upon whom I have drawn, including Levy and Peart (2017), Turner (2001), and Berger and Luckmann (1966). It is nevertheless true, I think, that the dominant view in the literature on experts is more hierarchical than that of Ostrom (1989), as illustrated, perhaps, by Cole (2010) and Mnookin et al. (2011). Like Ostrom, I propose moving away from the mistaken idea that only hierarchy is coherent and workable and toward the idea that polyarchy generally produces better results. I reject hierarchical views such as that of Wilson (1887) in favor of polyarchy.
As with Ostrom, my preference for polyarchy over hierarchy is a conclusion of the analysis and not its starting point. Moreover, the preference for polyarchy follows only if we add a norm of beneficence to my value-free analysis. If the theory is more or less right, however, no very specific normative assumption is required to infer the desirability of polyarchy. A general good will toward other humans is all that is required. If I am mistaken to prefer polyarchy to hierarchy, the error is to be sought, presumably, in my value-free scientific analysis rather than my morals.
Unfortunately, it is possible to be beneficent toward some and malicious toward others. All men are created equal, but we can have perfectly polycentric political order with liberty for some and tyranny for others. The United States during slavery and South Africa under apartheid are salient examples. One may point out that restricting the liberty of some narrows society and reduces the general level of benefits it provides even to those not in the tyrannized group. But powerful members of the tyrannizing group may derive special benefits from oppressing others. Large slaveholders in the American South grew rich by stealing the labor of the persons they owned (Fogel and Engerman 1974). We must choose between the “love of domination and tyrannizing” (Smith 1982, p. 186) and the material and spiritual benefits of amity and free exchange among equals.
My analysis in this book is more thoroughly epistemic than Ostrom’s, but the theory of public administration and the theory of experts both address the question of hierarchy vs. polyarchy. Do we get better results with hierarchy or polyarchy? How best can we coordinate the dispersed knowledge of many persons to produce good social results? The Wilsonian vision of expert rule through hierarchical administration is based on a hierarchical vision of knowledge. Knowledge is uniform, explicit, and integrated in the Wilsonian vision. To coordinate human actions, it holds, we must bring them into harmony with the hierarchy of knowledge, and such knowledge is embodied in experts. But if knowledge is not hierarchical, if it emerges from practice, sometimes in tacit form, then only polyarchy can coordinate action well. If knowledge is dispersed and heterogeneous, administrative hierarchy and the rule of experts will give us bad outcomes.
Like Ostrom, I wish to challenge the administrative state. In Chapter 12 I will note that the administrative state is inconsistent with the rule of law. Unfortunately, the term “rule of law” is often used to mean something like “vigorous enforcement of legislation” or “cracking down” on supposed bad actors. In this volume, however, it is a term of art in law with a very different meaning. The “rule of law” is central to the liberal ideal of liberty. The leading interpreter of the rule of law, A. V. Dicey, described it by saying: “It means, in the first place, the absolute supremacy or predominance of regular law as opposed to the influence of arbitrary power, and excludes the existence of arbitrariness, of prerogative, or even of wide discretionary authority on the part of the government” (1982: 120). (See Fallon 1997 on the meaning of “rule of law” and Epstein 2008 on “Why the modern administrative state is inconsistent with the rule of law.”) Mises (1966) invoked the rule of law when contrasting hegemonic and contractual bonds in society. “In the hegemonic state there is neither right nor law; there are only directives and regulations which the director may change daily and apply with what discrimination he pleases and which the wards must obey” (p. 199).
The administrative state abrogates the rule of law and is in this sense lawless. It should be dismantled. But it is impossible to predict the consequences of constitutional innovations (Devins et al. 2015). Thus, any dismantling of the administrative state should be piecemeal. It is not clear how to unwind it, and precipitate measures may replace a bad thing with something worse. In particular, if the administrative state is replaced by party rule we will have moved even further away from pluralistic democracy.
Party rule would strip away the necessity of supporting policy decisions with rational analysis. In the administrative state, rational support for the experts’ decisions is often more ceremonial than substantive. Nevertheless, the façade of objectivity and neutrality does provide at least some check on arbitrariness and inconsistency. Under party rule, the arbitrariness and inconsistency of state decision makers would likely be even worse than under the current administrative state. Decisions would be overtly political. They would reflect supposed judgments that this person or group is loyal or good whereas that group or individual is disloyal or bad, that this way of proceeding reflects the inner nature of the nation or its people whereas other ways are foreign and bad. Such “arguments” are built on a supposed fealty to the ruling party and its values. Such justifications of arbitrary decisions are immune to rational criticism. They allow similar cases to be treated in dissimilar ways. Anyone challenging the wisdom or legitimacy of a given policy choice thereby reveals themself to be evil or a servant of foreign interests. Populist movements seem to carry with them the danger of replacing the administrative state with party rule.
It is hard to guess whether “deconstruction of the administrative state” (Rucker and Costa 2017) in America would produce party rule or increased liberty. Gloomy prognosticators would do well to remember the embarrassing spectacle of Stuart Hall (1979), who mistakenly warned of the “authoritarian populism” of Thatcher’s government, which “has entailed a striking weakening of democratic forms” (p. 15). Hall went so far as to suggest that Thatcher’s government was a kind of semi-fascism, which, “unlike classical fascism, has retained most (though not all) of the formal representative institution in place” (p. 15). Hall’s harrowing alarm may now seem exaggerated.
If Chicken Littleism is a consummation devoutly to be unwished, some considerations are nevertheless discomforting. As the American public grows more aware of abuses of both the administrative state and the deep state, the legitimacy of traditional constitutional structures may be lost. These traditional structures include legislative and judicial checks on executive power. In such an environment, limits on state power may be further reduced, and a move to one-party rule is not unthinkable.
The danger I am warning of is not the product of one political party or one elected official. America’s two major political parties have both given greater support to elite privilege and state interests than to the rule of law or the interests of the people (Greenwald 2011). They have both cooperated in restricting political competition, thereby reducing political pluralism (Miller 1999, 2006). Political pluralism and the rule of law have therefore become increasingly alien concepts in American politics. The problem is not that this politician or that party seeks evil. The problem is that the rule of experts is incompatible with the rule of law and pluralistic democracy. The rule of experts progressively weakens the rule of law and pluralistic democracy, making it ever more likely that any democratic overthrow of the system will exclude as unknown and unimaginable both pluralism and the rule of law.
As administrations and parties change, the risk of a turn to party rule persists unabated. We are not doomed to travel the road to serfdom (Hayek 1944). We can choose to strengthen the rule of law, embrace democratic pluralism, and repudiate party rule. Until we do, however, the dangers inherent in the twin evils of the administrative state and the entangled deep state will persist. Those dangers arise from the logic of the situation, rather than one political party or one elected official.
My interest in the problem of experts emerged from my interest in forensic science and the problem of forensic-science error. That starting point may have helped my thinking. The problem of experts in forensic science presents itself free of some potentially distracting issues present in other areas such as climate science and monetary policy. The problem of expert error in forensic science cannot be attributed to Wilsonian progressivism. All parties, whether “left,” “right,” or something else, agree that forensic-science error is a bad thing. The fact that forensic science has not been associated with any political party or program may have helped me to see the problem of experts relatively unclouded by ideological bias. It also freed me from the necessity of judging the theories putatively used by experts. Forensic scientists judge matters of fact that are independent of the theory they use to evaluate evidence. Whether the defendant shot the victim is not a matter of perspective or theoretical framework. Jones did or did not shoot Smith. This theory-independence allowed me to consider error and the influences on error without pretending to second-guess the expert’s theory. I think this fact may have helped me to see in clearer focus the influence of social context, including incentives, on the decisions of experts. Finally, forensic science has, unfortunately, been a rich source of examples of expert failure.
Although my interest in experts arose from the relatively narrow context of forensic science, my analysis of it applies broadly. I noted above that it points to an analysis and critique of the entangled deep state. The “intelligence community” is a group of experts who seem clearly to be defending their power and prerogatives as experts. Whether these experts are cynical or sincere, their knowledge claims and their corresponding claims to power and autonomy are an undemocratic invitation to expert failure.
It may be that my more radical view of knowledge adds support to what I have called “Humean status quo bias” (Koppl 2009; Devins et al. 2015). Vernon Smith (2014) has noted that “the language ‘regulation’ or ‘deregulation’” is “unfortunate language.” I expand on this important point in Chapter 5. Economists of the “Austrian” school have warned of the “perils of regulation” (Kirzner 1985). I strongly agree that regulation has the very sort of perils Kirzner warns of. The epistemics of this volume strengthens Kirzner’s warning. But deregulation too has epistemic perils. Beware the perils of deregulation. We may call a general resistance to political and institutional change “Humean status quo bias.” Hume remarked that “a regard to liberty, though a laudable passion, ought commonly to be subordinate to a reverence for established government.” My position is similar to that of Charles Lindblom (1959), who argued in favor of “muddling through.” Devins et al. (2015, 2016) take up a similar position “against design.”
I have noted that liberty does not imply equality. But the reverse entailment is valid: Equality implies liberty. The cliché that we may be equally slaves under collectivism is wrong. We cannot all be equal unless we are all equally free. It is well understood, I think, that in a controlled system the powerful are more likely to serve their own interests than the general interest. In that case, however, we will not have equality. There will be no equality of income, status, or rights between those who command and those who must obey. Something similar is true of a “regulated” system even under the most perfect democracy. If the system is “regulated,” then state experts will, ex hypothesi, decide many things. Non-experts will have to obey the decisions of experts. And if these experts cannot avoid the “synecological bias” I describe in Chapter 10, then many of those expert decisions will be bad for the non-experts on whose behalf (putatively) they were made. Given synecological bias, the experts’ decisions will be bad even if they have zeal and good will. My discussion of expert failure in Chapter 10 may help to make the point that monopoly experts are not on a footing of equality with non-experts. Only when each of us is free of the rule of experts can we be equal. If you get to choose for me, we are not equal.
Mises (1966, p. 196) has expressed the point forcefully:
Where and as far as cooperation is based on contract, the logical relation between the cooperating individuals is symmetrical. They are all parties to interpersonal exchange contracts. John has the same relation to Tom as Tom has to John. Where and as far as cooperation is based on command and subordination, there is the man who commands and there are those who obey his orders. The logical relation between these two classes of men is asymmetrical. There is a director and there are people under his care. The director alone chooses and directs; the others – the wards – are mere pawns in his actions.
Under the rule of experts, knowledge is imposed on the system. Knowledge should instead emerge from the system. If knowledge is imposed on the system, it is imposed by someone who imposes upon and therefore dominates others. The persons imposed upon are not in a relation of equality with those imposing a knowledge scheme on society. The view of emergent knowledge I develop in this volume shows that we need not impose a unitary scheme of knowledge on society. We can let knowledge emerge and flourish without attempting to control or systematize it. If we are to be free, we must let knowledge emerge freely. And we cannot be free unless we are free of the domination and tyrannization of those who would impose a uniform system of knowledge on others. In other words, we cannot be free unless we are equal.
The disposition to impose knowledge on the system arises from the very act of studying society. In examining human society, we may easily forget that we too are humans in society. We see society as an anthill and people as ants. We gaze down upon the anthill as if we were higher beings. Alfred Schutz described what we might call the “anthill problem”: The theoretical perspective requires us to imagine ourselves above the system even though we live within the system. I don’t think Schutz meant his description of the anthill problem as a warning against hubris among social scientists, but he should have. Schutz compared the agents in our models (“personal ideal types”) to “puppets” whose strings the theorist pulls. The puppet’s “destiny is regulated and determined beforehand by his creator, the social scientist, and in such a perfect pre-established harmony as Leibnitz imagined the world created by God.” The very act of theorizing society puts you in a spurious godlike position. “What counts is the point of view from which the scientist envisages the social world” (Schutz 1943, pp. 144–5). The permanent and ineradicable crisis of social science is the theorist’s dual role as godlike observer and equal participant.