How Science Can (And Can’t) Fix Politics

Progressives tend to place a lot of political faith in science. The results of the scientific method tells us that climate change is real, that austerity hurts the poor without boosting economic growth, and that so-called “fetal pain” anti-abortion bills are grounded in nonsense. That scientific research has important political implications is an essential part of the progressive worldview.

But it’d be a grave mistake to assume that because science is politically useful, it can tell us everything we need to know about political life. Can research tell us whether it’s right to tax the rich to feed the poor? What experiment could prove that LGBT Americans deserve equal rights to civil marriage?

Delineating the sorts of questions science can and can’t answer, then, is important to understanding what we do and don’t know about political life. My basic view is that science can powerfully inform and shape, but not resolve, the sorts of moral questions upon which political life crucially depends — a point which the remainder of these comments will attempt to illuminate, though they will not come close to proving it conclusively.

In The New Republic, Steven Pinker has a very interesting essay defending the social and intellectual role of science against charges of “scientism.” Scientism, broadly speaking, is the when scientific arguments tread on intellectual territory — religion or ethics, for instance — on which they don’t belong. Pinker’s defense is to basically deny that scientism in this bad sense exists and argue that science has something important to say about virtually all domains of human life:

[The mindset of science] is, rather, indispensable in all areas of human concern, including politics, the arts, and the search for meaning, purpose, and morality…The term “scientism” is anything but clear, more of a boo-word than a label for any coherent doctrine. Sometimes it is equated with lunatic positions, such as that “science is all that matters” or that “scientists should be entrusted to solve all problems.”

Pinker’s right that, in various different forms, science has useful things to say about all of those things. But he’s wrong to deny that “scientism” is a problem, particularly with respect to politics and ethics. Pinker even mentions one of the worst offenders in passing, Sam Harris, whose 2010 book on morality is an Ozymandian monument to the limits of experimental science. Harris argued that scientific research can prove what’s morally and politically right, because morality must be about maximizing total improvements in “human well-being” and science can quantify what’s good for people. As the deluge of negative reviews at the time pointed out, there’s no scientific way to prove his basic premise — essentially, that utilitarianism is the only conceivably rational ethical system. That’s the domain of philosophy, and Harris’ argument skates over the vast ocean of philosophical reasons to be skeptical of his bold claim.


The Harris case illuminates why Pinker’s dismissal of “scientism” is far too quick. There’s a temptation to assume the tools of science provide an ever-expanding toolbox that can eventually pry open the universe’s secrets. We’ve made a lot of scientific progress! But some moral/political problems (“Is a fetus a person with a right to life? Is the United States obligated to spend money on fighting AIDS in sub-Saharan Africa?”) depend on more concerns questions (“what kinds of life has rights not to be killed? Do national borders define the scope of our moral obligations?”) that cannot be subject to resolution through empirical investigation. But because science’s track record is so good in its own domain, imperialists like Harris will always be there to plant science’s flag in lands it does not rightfully own. That is what scientism is, and denying that it exists as a class of error does neither science nor morality any favors.

In that sense, then, I agree with Ross Douthat’s critique of Pinker’s essay, which pegs scientism as class of thinking that is “empirically overconfident, intellectually unsubtle, and deeply incurious about the ways in which human beings can rationally disagree.” But Douthat’s attack on Pinker’s view of how science informs morality errs too far in the opposite direction. He sarcastically summarizes Pinker as saying:

You see, because we do not try witches, we must be utilitarians! Because we know the universe has no purpose, we must imbue it with the purposes of a (non-species-ist) liberal cosmopolitanism! Because of science, we know that modern civilization has no dialectic or destiny … so we must pursue its “unfulfilled promises” and accept its “moral imperatives” instead!

Douthat picked some spectacularly terrible examples of scientifically-informed moral thinking to mock. Take “non-species-ist” — by which he means the idea that morality shouldn’t discriminate against non-humans simply because they’re non-humans. Our moral view about the moral status of animals is critically tied to what we know scientifically about animals. In the 17th century, Descartes posited that animals were “automatons,” responding to inputs and outputs in the roughly the same way that computers do today without feeling or consciousness. Such creatures, obviously, didn’t deserve any rights or moral protections. Subsequent research on animal biology and psychology has blown Descartes’ picture to bits — a conference of leading scientists declared that the evidence that animals were conscious beings was “unequivocal.” If animals feel pain and consciously experience the world, then a central historical premise of the moral case against animal rights falls apart. The results of diligent scientific research really does “militate,” to use Pinker’s word, in favor of a moral system that grants stronger protections to animals.

The animal rights case illustrates the proper way to think about the role of science in political morality. The use of moral arguments to defend certain political arrangements depends crucially on empirical premises — the moral justification for sexist gender roles, for instance, depended heavily on theories about women being hyper-emotional and irrational. Scientific research can challenge that sort of false assumption or even resolve reasonable empirical disagreements, clarifying what is and isn’t true about the world so we can proceed to have a more targeted moral argument about politics.


Sometimes, those scientific results ineluctably point towards a particular moral conclusion: if it’s true that climate change will kill hundreds of millions of people in the medium term, then pretty much everyone with a working moral core should be inclined towards doing something about it. That can also be true at the level of more abstract moral principles, as the gender example illustrates — dissipating myths about female frailty helped advance the case for moral equality of persons in general by debunking the idea that some people as a class were “naturally suited” to certain social roles. You can believe that facts can favor particular moral theories without engaging in “scientism.”