Among the many things about which philosophers disagree, one of the debates I find interesting is the debate between actualists and possibilists. Roughly, possibilists believe you should engage in the best possible action you could, whereas actualists think you should do the best thing you will actually do given your imperfections, even if it’s technically possible for you to do better. To try to see the disagreement play out, consider this scenario.

I’m engaged in a tennis match with Bob, and Bob handily beats me. I’m a hothead with a terrible temper, and I really want to smack Bob over the head with my tennis racquet, but of course that would be bad to do. Let’s say there are three possible ways things can go.

In the best scenario, I approach the net, shake hands with Bob and congratulate him on a good game like a good sport. A less than ideal scenario is that I storm off the tennis court in a huff. And the worst case scenario is that I go over to Bob and whack him over the head with my tennis racquet. Let’s say that I know myself and my temper well enough to be sure that if I go near Bob right now, I will give in to my anger and whack him over the head. It’s metaphysically possible for me to not do this, but in practice this is what I will in fact do. Should I approach the net?

The possibilist would say that since the best thing I could possibly do would be to walk up to the net and shake hands like a good sport, I should approach the net. The actualist says that, given the facts of my personality and weakness, the best thing I will actually do is walk off the court in a huff, so I should not approach the net.

This debate often plays out in discussions of utilitarian and consequentialist ethics. Suppose a philosopher named Seter Pinger concludes that if you don’t take the highest paying  job you can find, work as many hours as you can before you collapse, and donate every penny beyond what you need to provide yourself with the barest of subsistence, then you’re morally no better than a serial killer. And let’s suppose that given certain plausible features of human psychology, if you demand people live up to this standard, they’ll end up feeling overwhelmed and just not donate to charity at all. However, if you instead argue people live up to a more moderate standard, like taking the Giving What We Can pledge and donating 10% of their  income to effective charities, the actual result of this will be more money given and more lives saved. If Pinger is a possibilist, he will push people to work like madmen and live like monks. If Pinger is an actualist, he will push people to take the aforementioned pledge.

Though he doesn’t use this terminology, Scott Alexander would seem to describe himself as an actualist in this post. He accepts that much of what goes on in the meat industry is morally unacceptable. He also says he “tried being vegetarian for a long time” but that he found it “really hard” and that he “kept giving up” on it. But then, rather than being vegetarian, he decided to follow what he called “a more lax rule,” namely, “I can’t eat any animal besides fish at home, but I can have meat (other than chicken) at restaurants. I’ve mostly been able to keep that rule, and now I’m eating a lot less meat than I did before.”

A possibilist would say Alexander should give up meat altogether, whereas an actualist would say Alexander should stick to his more lax rule. In a very actualist vein, Alexander says “if I am right that this is the strictest rule I can keep, then I’m not sure who it benefits to remind me that I am scum. Deny me the right to feel okay when I do my half-hearted attempt at virtue, and I will just make no attempt at virtue, and this will be worse for me and worse for animals.”

This divide strikes me as being very similar to a difference in how people consider what the government should do – there is a possibilist and actualist divide here too. For example, I once wrote about how Bernie Sanders claimed that if the government levied a $100 billion tax on Bill Gates, the government “could end homelessness and provide safe drinking water to everyone in this country” and Gates “would still be a multibillionaire.” Sanders is talking very much like a possibilist here – he claims that since the best results the government could possibly achieve with $100 billion would be very good, the government in fact should take that $100 billion.

My criticism of his claim, on the other hand, was to take something more like the actualist line. After all, I said, “if Sanders is right about the cost of ending homelessness, the federal government could completely end all homelessness in America with just 1.7% of what the federal government already spends in a single year.” Yet I notice that homelessness has not been eliminated.

It’s worth noting that Sanders didn’t claim that the federal government could end homelessness and provide clean drinking water to everyone at a cost of $100 billion per year. He claimed that both issues could be completely solved with a one time cost of $100 billion. So, by Sanders’ lights, the government could possibly have already ended homelessness scores of times over with its vast resources, but has not actually done so for various reasons. Yet at the same time, he thinks the government taking another $100 billion in taxes should be evaluated, not on the basis of what real-world experience shows the government will actually do, but on what he thinks is the best thing the government might possibly do, according to his ideal standard.

In another post, Scott Alexander evaluates the prospect of taxing billionaires to try to produce good results, where he also takes something very much like the actualist perspective:

Two of the billionaires whose philanthropy I most respect, Dustin Moskovitz and Cari Tuna, have done a lot of work on criminal justice reform. The organizations they fund determined that many innocent people are languishing in jail for months because they don’t have enough money to pay bail; others are pleading guilty to crimes they didn’t commit because they have to get out of jail in time to get to work or care for their children, even if it gives them a criminal record. They funded a short-term effort to help these people afford bail, and a long-term effort to reform the bail system. One of the charities they donate to, The Bronx Freedom Fund, found that 92% of suspects without bail assistance will plead guilty and get a criminal record. But if given enough bail assistance to make it to trial, over half would have all charges dropped. This is exactly the kind of fighting-mass-incarceration and stopping-the-cycle-of-poverty work everyone says we need, and it works really well. I have donated to this charity myself, but obviously I can only give a tiny fraction of what Moskovitz and Tuna manage.

If Moskovitz and Tuna’s money instead flowed to the government, would it accomplish the same goal in some kind of more democratic, more publicly-guided way? No. It would go to locking these people up, paying for more prosecutors to trick them into pleading guilty, more prison guards to abuse and harass them. The government already spends $100 billion – seven times Tuna and Moskovitz’s combined fortunes – on maintaining the carceral state each year. This utterly dwarfs any trickle of money it spends on undoing the harms of the carceral state, even supposing such a trickle exists. Kicking Tuna and Moskovitz out of the picture isn’t going to cause bail reform to happen in some civically-responsible manner. It’s just going to ensure that all the money goes to making the problem worse – instead of the current situation where the overwhelming majority of money goes to making the problem worse but a tiny amount also going to making it better.

It seems to me that there is likely a strong overlap with how much one finds the actualist line of thought persuasive, and their proclivity to view public policy decisions through the lens of concepts like public choice economics, or to evaluate economic regulation with the theory of regulatory capture as opposed to the public-interest theory of regulation. Just like James Buchanan described public choice as evaluating politics without romance, actualist philosophers think behavior should be guided by a similarly unromantic view of human nature.