In Defence of Knowing What You're Talking About

The backlash against expertise has been building for years, and it is not difficult to understand why. Experts told us that Iraq possessed weapons of mass destruction. Experts assured us that the financial system was sound in 2007. Experts promised that the pandemic would be brief, that inflation would be transitory, that certain technologies were decades away from maturity. In each case, the confident pronouncements of credentialed professionals turned out to be wrong, sometimes catastrophically so, and the people who suffered the consequences were not the experts themselves but the ordinary citizens who had trusted their judgment.

The failures are real, and any defence of expertise that does not begin by acknowledging them is not worth making. Experts failed, publicly, on some of the most consequential questions of the twenty-first century. The intelligence analysts who assessed Iraq's weapons programmes were not amateurs; they were highly trained professionals working within institutions designed to produce reliable assessments, and they got it badly wrong. The economists who failed to see the financial crisis coming were not outsiders; they were the people running the models, advising the regulators, and setting the policies that allowed the crisis to develop. The public health officials who misjudged the pandemic's trajectory were not uninformed; they were the world's leading epidemiologists. The problem was not a lack of expertise. It was that the expertise, in these specific and highly visible cases, was insufficient to the complexity of the questions being asked.

But the response to these failures — the growing tendency to treat all expert knowledge as suspect, to equate credentialed opinion with mere opinion, and to regard the rejection of expertise as a form of democratic empowerment — is worse than the failures themselves. It is, in fact, a category error of the most dangerous kind. The correct response to an expert being wrong about a specific question is to demand better expertise — better data, better models, more rigorous analysis, more intellectual humility. The incorrect response is to conclude that expertise itself is worthless and that the untrained intuition of the layperson is an adequate substitute. It is not.

Expertise is, at its core, probabilistic rather than oracular. This distinction is critical and almost universally misunderstood. An epidemiologist does not — cannot — tell you exactly how a pandemic will unfold. What they can do is assign probabilities to different scenarios, identify the variables that will determine which scenario materialises, and recommend actions that improve the odds of a better outcome. When the public expects certainty and receives probability, the result is disappointment, and disappointment curdles into distrust. But the fault lies not with the expert, who offered what expertise can honestly offer, but with the expectation, which demanded what expertise cannot provide.

The alternative to expertise is not, as its critics sometimes imply, the wisdom of the crowd or the common sense of ordinary people. On technical questions — the efficacy of a vaccine, the trajectory of the climate, the stability of a financial system, the structural integrity of a bridge — there is no meaningful 'common sense' to appeal to. These are questions that require specialised knowledge, acquired through years of training and practice, and the notion that a sufficiently large group of untrained people can arrive at a better answer than a smaller group of trained ones is not democratic. It is innumerate. Democracy is a system for making collective decisions about values and priorities. It is not a system for determining the tensile strength of steel.

History provides abundant examples of what happens when expertise is dismissed in favour of ideology, intuition, or political convenience. The Soviet Union's embrace of Lysenkoism — the rejection of mainstream genetics in favour of a politically congenial alternative — set Soviet agriculture back by decades and contributed to famines that killed millions. The dismissal of engineering expertise in the design of the space shuttle Challenger's O-ring seals, overridden by managers who preferred their own judgment, killed seven astronauts. The rejection of epidemiological expertise during the early stages of the AIDS crisis, driven by political squeamishness about the populations most affected, cost hundreds of thousands of lives that could have been saved by earlier action. In each case, the expert consensus was imperfect but directionally correct, and the decision to ignore it produced consequences that dwarfed the imperfections.

None of this is an argument that experts should be immune from challenge. They should not. Experts are human beings with biases, blind spots, institutional incentives, and a documented tendency toward overconfidence in their own judgments. The healthiest intellectual ecosystems are those in which expert claims are routinely subjected to scrutiny — by other experts, by informed critics, and by the evidence itself. Peer review, replication, open data, and the adversarial process of scientific debate all exist precisely because the expert community recognises, at least in principle, that any individual expert or any individual study may be wrong. The system is designed to catch and correct errors over time, and it does so imperfectly but far better than any available alternative.

The critical distinction — and the one that the anti-expert movement consistently fails to make — is between challenging expertise and dismissing it. Challenging expertise means engaging with the evidence, identifying the weaknesses in an argument, proposing alternative hypotheses, and subjecting those alternatives to the same rigorous scrutiny. It is a constructive process that strengthens knowledge. Dismissing expertise means ignoring the evidence entirely, substituting intuition or ideology for analysis, and treating the credentials of the expert as a reason to distrust rather than to engage. It is a destructive process that degrades public discourse and, when it influences policy, degrades public welfare.

The backlash against expertise is understandable. The failures were real, the consequences were severe, and the experts themselves have not always been sufficiently humble about the limits of their knowledge. But the answer to imperfect expertise is not the absence of expertise. It is better expertise: more transparent, more accountable, more honest about uncertainty, and more willing to say 'I do not know' when the honest answer is that nobody does. The alternative — a world in which every claim to specialised knowledge is met with reflexive suspicion and the loudest voice wins regardless of its command of the evidence — is not a more democratic world. It is a more dangerous one.