At the end of March, Senator David Leyonhjelm announced he would not support any legislation proposed by Australia’s health minister until the country’s current ban on nicotine e-cigarettes was lifted.

In Australia’s political system, this amounts to a single senator—one with a crucial vote—holding the health portfolio to ransom. It’s serious. It also provides a clear and useful case study of how a politician does ethical reckoning in response to scientific evidence.

In last month’s column, I introduced readers to Hume’s Guillotine—that is, it’s impossible to derive a moral “ought” from a scientific “is”—and promised some worked examples from two books: David Leyonhjelm’s Freedom’s Salesman and Chris Snowdon’s Killjoys. If you haven’t read that piece, please do so now, then make your way back here.

The global debate over nicotine e-cigarettes (“vaping”) illustrates every element of the is-ought conundrum I outlined last month. E-cigarettes represent a spontaneous, leaderless, and market-driven solution to the problem of smoking addiction. At least initially, they were not widely understood—invented as they were by a Chinese pharmacist who desperately wanted to quit. The response from the public health establishment in most countries has been confusing and confused, and in Australia’s case notably authoritarian. E-cigarette regulation is also enormously varied across jurisdictions and sometimes downright contradictory. Snowdon makes a compelling case that the European Union’s particularly cack-handed regulatory effort contributed to the Brexit vote.

Neither Freedom’s Salesman nor Killjoys are about e-cigarettes—although they do feature strongly in both books. Freedom’s Salesman is a selection from Leyonhjelm’s extensive list of publications and speeches, and covers the issues a politician elected to represent Australia’s most populous state would be expected to address: tax, welfare, conservation, infrastructure, national security, public health, renewables, the works. Killjoys is a focussed attack on paternalism, particularly public health paternalism—something I’ve addressed previously for the Cato Institute. Crucially, Leyonhjelm often makes use of Snowdon’s arguments, and Snowdon—although British—has appeared as an expert witness before Australian parliamentary inquiries.

Both men are classical liberals. They have obvious (and admitted) ideological abstractions on which they call in the absence of complete data: free markets and individual liberty. However, they also have a resonant respect for scientific evidence and do not devalue scientific truths. Snowdon is an economist and Leyonhjelm—before he was elected in 2013—was a veterinarian and agribusiness consultant. Some of the most compelling writing in both books is when they make use of specialist knowledge from their respective disciplines. Leyonhjelm brings his expertise to bear on vaccination, GM crops, and animal welfare. Snowdon is devastating when assessing the shoddy economic modelling produced by public health bodies and overblown claims about the power of advertising from media commentators.

When arguing for legalising e-cigarettes with nicotine (Leyonhjelm) or against the silly labelling mandated by the EU’s Tobacco Products Directive (Snowdon), both men draw on scientific expertise from government and private sector public health organisations. They use Public Health England data showing that e-cigarettes are a safe and effective way to quit. They deploy research by the Royal College of Physicians contrasting the lethality of combustible cigarettes with the relative harmlessness of e-cigarette vapour.

However—consistent with the ethical reckoning he must do as a politician—Leyonhjelm also makes use of ideological heuristics. He invokes John Stuart Mill’s “harm principle,” pointing out not only are e-cigarettes less harmful than combustible cigarettes, but they also eliminate the problem of passive smoking (“second hand smoke”).

He also argues people’s choices matter not because they are right but because they are theirs. This claim is ancient. You’ll find it in Roman jurists: it was one of the defining characteristics of freedom as opposed to slavery. By this logic, people are free to do what pleases them, because they find it pleasurable, and those pleasurable choices that don’t harm others also should not matter to others.

It may be tempting—now we have scientific knowledge available to us that Roman jurists and John Stuart Mill alike did not—to abandon the ideological abstractions that form part of every political tradition. Here, it is Snowdon who highlights the importance of moral heuristics unsupported by empirical evidence in the development of public policy, and his book is worth reading for this alone.

In a long section describing various public policy interventions that failed (often because they were based on science that was later debunked) he illustrates the danger of “making a law” because something seems to be true. Sometimes, the sanest policy prescription is simply, “do nothing.” “Doing nothing” makes a lot of people uncomfortable, but Snowdon’s litany of public health policy foul-ups caused by people’s wanting to “do something” is instructive and often horrifying.

Britain taxed diesel fuel at lower rates for decades, for example. This shifted market preferences such that many people bought diesel cars on the basis their fuel was “cleaner.” We now know (because science marches on) that nitrogen oxides (NO_x) in diesel fumes are dangerous to human health and ruin the built environment. In the UK, it was the Volkswagen emissions scandal (which involved engines producing NO_x pollutants at up to 40 times above permitted rates) that forced the issue into public consciousness.

Another case study concerns government dietary advice. Starting in the 1960s, governments across the developed world campaigned against (and even taxed) high-fat diets, with predictable results. Food processors and manufacturers cut fat content and lavished their “low fat” products with sugar. Later—because science marches on—we learned this government fat-shaming was nonsense on stilts. Sugar is now considered far worse for you than fat, and its government-encouraged profusion may have contributed to the obesity crisis, particularly among the poor.

Here, two simple ideological heuristics—one from Leyonhjelm’s classical liberalism and one from another political tradition, democratic socialism, are more useful than “evidence based policy.” The classical liberal heuristic, as mentioned above, is “maybe it’s better to do nothing and leave people alone.” The democratic socialist heuristic is, “don’t impose heavy taxes on things mainly poor people buy, it’s regressive.” While scientific evidence can indeed introduce honesty to moral choices and prevent the overuse of ideological shorthand, this is no reason to ignore the underlying principles articulated by those political parties competing for our favour at election-time.

Leyonhjelm makes another valuable point as an aside in one of his pieces on vaccination. When scientific experts are wrong repeatedly (as they were with diesel fuel and dietary advice), those failures can call successes into question. This has happened—with alarming consequences—to vaccines and GM foods, and forms the basis of pseudoscientific “anti-vaxx” and “frankenfood” claims. Non-scientists—even well-educated non-scientists—are often unaware of how the strength of evidence supporting different scientific findings varies, and struggle to draw meaningful distinctions between “science says vaccinate your children” and “science says eat less fat.” One only has to read articles littered with casual misuse of words like “hypothesis” and “theory” to appreciate this.

Evidence for the effectiveness of vaccines is overwhelming. Evidence for the safety of GM crops is overwhelming. By contrast, what constitutes a “healthy diet” or “clean fuel” is far woollier and new studies complicating the picture emerge almost on a daily basis. If science is to inform public policy, then it simply must be “true” to a very high standard—“beyond reasonable doubt” rather than “on the balance of probabilities,” to borrow two terms of art from the law.

So much for problems that emerge when politicians and civil servants hitch their cart to scientific claims later falsified. Vaulting Hume’s Guillotine even when scientific claims turn out to be largely true can also be terrifying, and once again simplifying ideological heuristics may save us from grievous moral error.

Here, Snowdon’s most devastating case study concerns eugenics, testament to the danger of scientists discovering a true thing, then persuading everyone else—and particularly lawyers—that certain policies automatically follow. Tellingly, state-backed eugenics was never legislated in Britain—despite widespread support—but it was enacted almost everywhere else, from Nazi Germany to the United States to India, China, and Scandinavia.

First, however, you may be shocked at the phrase “scientists discovering a true thing” to describe the basis of eugenics. Let me explain.

Although many of the detailed claims made by eugenicists were false or misleading, many of their broad claims were correct. The extent to which we are products of our biological and genetic ancestry is apparent. IQ is real and heritable and has a significant and measurable effect on educational outcomes and some effect on income. Social status is heritable over multiple generations and resistant to attempts to improve social mobility. Assortative mating is marvellous for educated people—especially educated women—but tanks what little social mobility is left over after genetics has done its work.

Most of this was known to eugenicists of the late 19^th and early 20^th century. Indeed, the science underlying IQ is quite old, and the mass testing of US Army recruits provided the sort of vast data sets that can prove really useful to a statistician or analyst. I suspect our current inability to have an adult conversation about these things—and our ongoing infatuation with thoroughly debunked “blank-slatism”—is thanks to the sheer horror of eugenics.

Progressives, today’s most ardent blank-slatists, need to grasp the historical nettle. They must honestly assess whether their problem with evolutionary psychology is that they genuinely believe its claims are false, or if instead they’re just embarrassed that the last time they got their hands on similar information they proceeded to misuse it in spectacular fashion.

By the same token, for evolutionary psychologists to have their discipline taken seriously requires not only a franker engagement with this dreadful history but a disciplined refusal to make any policy prescriptions at the same time they provide scientific findings. Watching various biologists bang on about how important it is, say, for educated couples to have more children makes civil servants (regardless of politics) cringe inwardly and throw our hands up in despair.

Because eugenics was overwhelmingly a progressive cause, when it was defeated, as in Britain, it took a determined alliance of Christian conservatives (like G. K. Chesterton) and classical liberals to do the job. Notably, these people did not attempt to rebut its scientific claims—unlike those who don’t like the implications of climate modelling or evolutionary psychology now do. Instead, they made arguments with clear roots in ideological heuristics. Here is Chesterton:

The Eugenists [sic] would probably answer all my examples by taking the case of marrying into a family with consumption (or some such disease which they are fairly sure is hereditary) and asking whether such cases at least are not clear cases for a Eugenic intervention. Permit me to point out to them that they once more make a confusion of thought. The sickness or soundness of a consumptive may be a clear and calculable matter. The happiness or unhappiness of a consumptive is quite another matter, and is not calculable at all.

What is the good of telling people that if they marry for love, they may be punished by being the parents of Keats or the parents of Stevenson? Keats died young; but he had more pleasure in a minute than a Eugenist gets in a month. Stevenson had lung-trouble; and it may, for all I know, have been perceptible to the Eugenic eye even a generation before. But who would perform that illegal operation: the stopping of Stevenson?

It can be discomforting—if one has lit upon an important scientific truth—to see one’s hard-won evidence done down by political rhetoricians in an endless game of scissors, paper, rock. However, we live in liberal democracies, with polities that require the transmutation of science into policy by force of law, mediated always through the ballot box.

This reality demands not only a resonant respect for scientific scholarship, but also an acknowledgment it is rare that scientific knowledge can be combined with a widely accepted moral principle to produce a single, uncontested policy prescription. Killjoys and Freedom’s Salesman show how people from one political tradition manage their engagement with Hume’s Guillotine. You do not have to agree with Leyonhjelm’s or Snowdon’s politics to see how the reasoning works. You do, however, have to reject the belief that, if only facts were clear and proven, playground morals would show us the way.