Longtermism is controversial. It has many good-faith critics and, particularly in recent weeks, a great many bad-faith ones. In a nutshell, longtermism holds that people in the future matter no less than people in the present, and that we have a moral obligation to help them to the extent that we can.
Longtermism's bad-faith critics like to draw tenuous links between caring about future people and various kinds of sinister beliefs or consequences. I won't describe the bad-faith objections in detail. I think these kinds of guilt-by-association arguments are hard to caricature since, for most thoughtful readers, these pieces caricature themselves.
The good-faith critics present a tougher attack. If you can't shake the feeling that current people matter too, then strong longtermism — a focus on the future, perhaps to the neglect of the present — is probably not for you. Fair play to these critics: strong longtermism is a bullet that many people do not want to bite.
The toughest objection comes from the final clause in the opening paragraph of this post. We're obligated to help future people to the extent that we can. But can we at all? The future is a big place. Who are we to think we can so much as touch it?
From our vantage point in the present, nearly any concern about the future beyond two or three years from now appears not just speculative but closer to fictional. There are so many paths that history can take, and trying to act for the benefit of the future seems to entail knowing the twists and turns ahead.
I think this is the wrong way to think about planning for the future. First, although most of the people who longtermists care about will live in the future, the threats facing them exist much closer to the present. That is to say, it's current events that will shape—or, tragically, prevent—future lives. Second, acting for the sake of future people who live down one of history's more unpleasant byways requires not prediction, but planning. Most people who buy fire insurance will never have a fire. Longtermist action is an insurance policy for the survival and flourishing of humanity.
So planning for the future doesn't entail knowing how the future is going to go or being able to affect it with any sort of confidence. Planning for the future is preparation. There are a lot of ways things can go wrong, and an ounce of prevention is worth a pound of cure. Catastrophes are really only catastrophic if they're unplanned-for, and they're typically unplanned-for because they're seen as unlikely. The best way to avoid catastrophe in general is to put some effort into planning for many unlikely catastrophes.
In fact, it is possible to usefully prepare for the problems of the future. We know this because our ancestors had the tools to see the problems we and our children and grandchildren will face. And because they failed to help us.
Around the turn of the last century, a catastrophe that might have seemed particularly unlikely to most people was the prospect that the Earth could become unlivable due to extreme climate change. But there were people who foresaw it. In 1896, the Swedish chemist Svante Arrhenius published one of the first climate models, estimating that a doubling in atmospheric CO2 levels would result in roughly six degrees C of additional warming. Arrhenius noted that human combustion of CO2 could lead to warming, but thought that the effects would ultimately be beneficial.
In 1938, the English engineer Guy Callendar demonstrated that the Earth's land temperature had been warming over the previous 50 years. He, too, thought the effects would be mostly positive. He was mostly ignored anyway. But by the 1950s, reputable scientists had started to be concerned.
It took another five decades before policymakers came anywhere close to taking the threat seriously. By that time, we were perilously close to severe and irreversible warming. We still are. But one bright spot is the rapid success of renewables over the past few decades decades.
The progress we've made in climate, such as it is, is largely down to cost reductions for wind power, solar energy, and power storage, which have all gotten cheaper, and therefore much more plentiful. No matter what your opinion is about other parts of a comprehensive climate strategy — carbon capture and storage, nuclear energy, geothermal — you probably agree that increased renewables capacity is an important part of the solution.
Such a solution is clearly needed, since climate change could cost as many as 83 million excess deaths by 2100. The faster decarbonization happens, the fewer people will die over the coming century.
Let's run with the solar example. What generated a 99% drop in the cost of photovoltaics over the past four decades? Recent work has found that government policies that stimulated market demand for renewables were responsible for about 60% of the decline in price of solar panels. Crucially, that price decline is responsible for solar's rapid expansion, with a twenty-fold increase in installed capacity in the past decade alone.
Price declines in solar started in earnest in the 1970s, partially as a result of the use of solar panels to power satellites in orbit. But they had existed for much longer:
Above is a picture one of the first functional solar panels — sitting on a New York City roof in 1884. This technology has existed for a long time. It just took nearly a century for it to become a thing, and another half-century for it to enter widespread usage.
Imagine, now, that you're a longtermist circa 1900. You care about future people, like those who will live around the year 2000. You even care about people a century further out. You want to buy an insurance policy for all of these people and their descendants; you want to plan for a horrible future that may or may not come to pass.
Though Arrhenius and the scientists who immediately followed him weren't hip to the possibility of warming-induced cataclysm, people were drawing the link between fossil fuel burning and disaster within a decade. If you had asked Arrhenius for the probability that warming could actually be bad, what might he have said? 1%? 10%? If warming could be net-bad, then the warming that Arrhenius had forecast could be very bad indeed. Even a small probability of this outcome would have been enough to warrant a long-term insurance policy.
I think a turn-of-the-century longtermist would have done well to agitate for increased R&D on photovoltaics. A wealthy longtermist might have funded this research themselves. Ultimately, this has been the final consequence of the federal policies that have created demand for solar panels more recently: increased demand has paid for more research into more efficient solar panels. But it was helped along at first by government R&D: funding for solar research kicked off price declines starting in 1974.
Around 1900, there were already big research labs that did this sort of thing, and philanthropists to fund them all over the world. The Royal Society had already been funding independent science since 1660, publishing Isaac Newton's calculus and backing James Cook's 1769 journey to track the transit of Venus. Industrial labs like the one operated by General Electric were soon to develop the electric fan and the tungsten light bulb filament, and philanthropists like Andrew Carnegie were in the process of setting up foundations that would eventually fund world-changing science. A Victorian longtermist would have faced an embarrassment of altruistic riches: he could have can gone to work at GE, suctioned himself to a robber baron, or lobbied the Royal Society to invest in photovoltaics.
Crucial developments in solar technology development realistically could have have happened earlier. Bell Labs eventually made an efficiency breakthrough when its scientists replaced the selenium used in the 1884-era solar panels with the silicon we use today. This breakthrough was possible at the turn of the century, around the time that the author of the first organic chemistry textbook was doing pioneering work on silicon polymers that would lead to the development of synthetic rubber.
You may think that 1900 is too early for our hypothetical past longtermists to have gotten a handle on the challenge posed by climate change. Even so, a late start in the 1950s, when a canny longtermist could have lobbied for a trickle of the exploding federal R&D budget, is still better than the world we ended up with. If unmitigated climate change will cost 80 million lives over the course of the century, then even speeding the decarbonization trajectory up by a single year is worth a million lives, give or take.
That's basically the point of the whole thing: because the future is big, and because a lot of people will live in it, small tweaks in the present have massive impacts on their lives. That's a lot of power over future people and, vitally, we have that power whether or not we take active steps to direct it. We are doing things that affect the lives of future people every day. The least we can do is buy them some insurance.