Is the world getting better? The Harvard psychologist and polymath Steven Pinker wants you to think that it is. And there happens to be a considerable body of evidence to support Pinker’s claim that humanity has indeed made progress with respect to human well-being, not to mention scientific knowledge and technological capability.
In a recent Wall Street Journal article, Pinker outlines some of this evidence, arguing that the supposed “gloominess” found on the political “right” and “left” is not merely wrong but “decidedly un-American.” Yet I believe that Pinker’s proselytizing for progress is misleading at best and potentially dangerous at worst. Let’s begin with the following paragraph from his article:
Isn’t it good to be pessimistic, many activists ask — to rake the muck, afflict the comfortable, speak truth to power? The answer is no: It’s good to be accurate. … Indiscriminate pessimism can lead to fatalism: to wondering why we should throw time and money at a hopeless cause. And it can lead to radicalism: to calls to smash the machine, drain the swamp or empower a charismatic tyrant.
Second, I agree that “accuracy” should be the goal of any serious assessment of the human condition — of where we were, are, and might end up. But this is precisely why Pinker’s analysis is problematic: his claims aren’t incorrect, they’re incomplete, resulting in a distortedly sanguine picture of humanity’s progress.
The reality is that the contemporary world is also more dangerous than it’s ever been in recorded history — or perhaps the history of the human species, bracketing the Toba catastrophe that, according to some scientists, nearly killed off our ancestors. There is a growing swarm of problems facing humanity that are unprecedented in their scale and severity; some even have genuinely existential implications. It behooves people — the electorate, our political leaders, scientists, and so on — to construct as comprehensive a view of the world as possible, and one cannot do this without recognizing the obstacle course of civilizational hazards before us.
(The rest of this paper will draw from a forthcoming article of mine titled “Facing Disaster: The Great Challenges Framework.” For a far more detailed analysis of our evolving existential predicament, please see this paper.)
First, anthropogenic climate change is a historically unique problem that, as such, humanity has never before encountered. Thus, we have no track record of displaying the collective wisdom needed to overcome this immense danger. The hottest 18 years on record have all occurred since 2000 (with one exception), and the best current science warns of extreme weather events, megadroughts lasting decades, devastating coastal flooding, sea-level rise, melting glaciers and the polar icecaps, desertification, deforestation, food supply disruptions, infectious disease outbreaks, mass migrations, and heat waves that surpass the 95 degree wet-bulb threshold for human survivability, meaning that even if one were naked in the shade in front of a giant fan, death would follow. In fact, a large 2017 study notes that about 30 percent of the global population is exposed to “lethal heat events” for 20 or more days a year. But if greenhouse emissions continue to grow, approximately 74 percent will be exposed to this “deadly threshold,” and even if humanity drastically reduces its emissions, the percentage will still rise to about 50. As another study reports, between 20 and 30 percent of the planet will undergo aridification if the global mean temperature rises to 2 degrees Celsius.
Climate change is also a “conflict multiplier” that will — and already has — increase the probability of wars and even terrorism. And by the end of this century, unless drastic changes to greenhouse gas (GHG) emissions are made, a huge portion of the US will become largely uninhabited arid land.
But carbon dioxide, one of the primary GHGs causing climate change, is also causing ocean acidification. And studies show that this is occurring faster than ocean acidification occurred during the Permian-Triassic extinction event, also known as the “Great Dying,” during which 95 percent of all species alive on the planet kicked the bucket. Even more, the ocean’s gyros are full of plastic trash (see the Great Pacific Garbage Patch), about 90 percent of the world’s coral reefs are projected to die by 2050, there are some 500 hypoxic “dead zones” around the world due to human pollution, and one study estimates that, if business continues as usual, there will be virtually no more wild-caught seafood by 2048. Another study projects that there will be more plastic than fish, by weight, in the world’s oceans by 2050. (Again, see here.)
Even more, human activity has launched the biosphere into the sixth mass extinction. This conclusion holds on even the most optimistic assumptions of current and past extinction rates. Studies show that, for example, the global population of wild vertebrates has declined by an unthinkable 58 percent since 1970, and we will likely see a two-thirds decline by the middle of this century. Colony collapse disorder has devastated the populations of honey bees, and while we will likely need to produce more food in the next 50 years than we have produced in our species entire history, soil erosion is reducing the annual crop yield by 0.3 percent, meaning that “at this rate, we will have lost 10 percent of soil productivity by 2050” — about the same loss that global warming is expected to cause.
Consequently, we have crossed three “planetary boundaries” that make us vulnerable to sudden, irreversible, and catastrophic changes to the biosphere upon which we depend for our survival. As the authors of an ominous paper on this topic write:
anthropogenic pressures on the Earth System have reached a scale where abrupt global environmental change can no longer be excluded. … Transgressing one or more planetary boundaries may be deleterious or even catastrophic due to the risk of crossing thresholds that will trigger non-linear, abrupt environmental change within continental- to planetary-scale systems.
Making matters worse, there is preliminary evidence that higher carbon dioxide concentrations in the ambient atmosphere could literally make us “dumber.” We have now crossed the 400 parts per million (ppm) threshold and we could reach 1,000 ppm by the end of the century; in contrast, our ancestors evolved with concentrations between 180 and 280 ppm. According to two studies, even relatively small increases in carbon dioxide levels could have measurable negative effects on our ability to think. If this is the case, it’s a recipe for disaster, given the complexity and size of the global problems that children today will have to face. (Add to this the calculation by Harvard researcher David Bellinger that “Americans have collectively forfeited” 41 million IQ points “as a result of exposure to lead, mercury, and organophosphate pesticides.”)
But environmental degradation isn’t the biggest challenge before us. An even more worrisome trend is the distribution of unprecedented offensive capabilities across society that is being driven by certain emerging technologies. Consider emerging artifacts like CRISPR/Cas-9, digital-to-biological converters, base editing, USB-powered DNA sequencers, SILEX (separation of uranium isotopes by laser excitation), which enable humans to manipulate the world in truly extraordinary ways. But there are also anticipated future technologies like nanofactories, which could enable state and nonstate actors to manufacture huge arsenals of advanced weaponry; autonomous nanobots that could target specific people, races, or species; lethal autonomous weapons — e.g., “slaughterbots” — not to mention invisibility cloaks, robot soldiers, mind-reading technologies, mind-control technologies, laser weapons, self-guided bullets, etc.
The point is that many of these “dual-use” technologies are becoming not only more powerful than ever before, but more accessible as well. In other words, they are rapidly multiplying the total number of actors — states, terrorist groups, and even lone wolves with a death wish for humanity — capable of wreaking unprecedented harm on civilization. As I have written elsewhere, the malicious actors of tomorrow will have bulldozers rather than shovels to dig mass graves for their victims.
This is an extremely worrisome predicament. There are perhaps 300 million psychopaths in the world today — and as I have detailed elsewhere, some really would destroy the world if only they could. Meanwhile, religion — especially Islam — is growing worldwide, meaning that the number of religious terrorists will likely grow in proportion. According to the Global Terrorism Index, religious extremism is the number one driver of global terrorism today, and religious terrorism has proven to be far more lethal and indiscriminate than past forms of political terrorism.
This is a bad situation: terrorism isn’t going to disappear anytime soon, and it could even increase in the coming century (as a result of “demographic inflation” and the instability caused by climate change); at the very same time, such individuals will be empowered like never before to inflict genuinely existential harm on our species.
The distribution of offensive capabilities will also increasingly undercut the ability of states to maintain law and order, a point that I have made elsewhere. As Max Weber famously wrote, states must have a monopoly of legitimate force in order to govern effectively. But if individuals begin to wield a comparable degree of power as states, the social contract will collapse and, along with it, the modern state system.
Consider that al-Qaeda consisted of perhaps 500 individuals when they attacked New York in 2001, resulting in two massive wars that could cost $6 trillion in total and that have resulted in more than 110,000 Iraqi civilian deaths. Similarly, the 2016 Dyn cyberattack may have been perpetrated by a single “angry gamer.” This lone wolf interfered with an incredible number of major websites, including Airbnb, Amazon, BBC, The Boston Globe, CNN, Comcast, FiveThirtyEight, Fox News, The Guardian, iHeartRadio, Imgur, National Hockey League, Netflix, The New York Times, PayPal, Pinterest, Pixlr, Reddit, SoundCloud, Squarespace, Spotify, Starbucks, Storify, the Swedish Government, Tumblr, Twitter, Verizon Communications, Visa, Vox Media, Walgreens, The Wall Street Journal, Wired, Yelp, and Zillow (to name a few).
How can states prevent society from plummeting into Hobbesian anarchy when small groups and single individuals can wreak havoc that crosses national boundaries and rivals what states themselves are capable of?
Thinking about this situation more abstractly, John Sotos has recently crunched some numbers to show that the distribution of offensive capabilities could all but guarantee civilizational collapse. For example, a 1 in 100 chance that only a few hundred agents releasing a pandemic-level pathogen yields almost inevitable doom with 100 years or so. If the total number of people who can cause global-scale harm rises to 100,000, the probability of any one person releasing such a pathogen must be less than 1 in 10⁹ for civilization to survive a single millennium. In other words, low probabilities can add up pretty quickly as the total number of individuals capable of mass destruction increases.
So far, we haven’t even mentioned the dangers associated with artificial intelligence as it approaches human-level intelligence — and possibly surpasses it. Here Pinker holds some strange views, or so I would argue. For example, he writes:
AI dystopias project a parochial alpha-male psychology onto the concept of intelligence. They assume that superhumanly intelligent robots would develop goals like deposing their masters or taking over the world. But intelligence is the ability to deploy novel means to attain a goal; the goals are extraneous to the intelligence itself. Being smart is not the same as wanting something. … It’s telling that many of our techno-prophets don’t entertain the possibility that artificial intelligence will naturally develop along female lines: fully capable of solving problems, but with no desire to annihilate innocents or dominate the civilization (Pinker 2015).
This succumbs to what the cosmologist Max Tegmark refers to as one of the “top myths of advanced AI.” The concern isn’t that an artificially intelligent machine will become evil or malicious, motivated by an alpha-male drive to dominate. Literally no one is arguing this! Rather, the concern is that a machine of this sort could have goals programmed into it that are in some way — perhaps in completely unforeseen ways — misaligned with our human values and that it could be competent enough to achieve those goals. For example, we might program the machine to cure cancer and in response it converts the entire planet into a giant cancer research center. Oops! When it comes to systems that are better at solving problems in every cognitive domain than even the smartest humans, it is absolutely crucial that the system does what we intend rather than what we tell it to do. Yet how to ensure that this is the case — that is, how to align the value system of a superintelligence with our value system — turns out to be an incredibly difficult, and perhaps insoluble, problem.
This is why many experts on the topic believe that superintelligence could pose the greatest long-term threat to our survival. And this is why these experts believe that superintelligence deserves immediate attention. In fact, survey after survey reveals that the AI research community overwhelmingly expects human-level artificial intelligence to arrive before the end of the century, with a sizable portion believing that this could occur even sooner — in a few decades — followed after by superintelligence. So, we may not have much time to solve the value-alignment problem mentioned above, which is worrisome.
Does this mean that we should be pessimistic about the future? Here I would argue that the pessimism-optimism dichotomy misses the point. What matters is that humanity does everything it can to ensure that the future is as good as it can possibly be. But to do this, as Pinker himself suggests, one needs an accurate model of what exactly our species is up against. This requires acknowledging both the good and the bad — it means affirming the moral progress that we have made so far, enabled in large part by the Enlightenment, as well as admitting that our existential predicament today is almost certainly more existentially risky than ever before. Pinker notes that progress isn’t guaranteed to continue — but I would argue that the best way to continue progress is to be painfully realistic about the immense risks to our survival and prosperity caused by phenomena like environmental degradation, emerging technologies, and artificial intelligence. Only once humanity takes these risks seriously can we implement judicious policies to obviate them.
There is no question that Pinker — among the leading intellectuals in the world, and for good reason — has made a huge contribution to human knowledge with his book The Better Angels of Our Nature, and no doubt his forthcoming Enlightenment Now (which Bill Gates has called his “new favorite book of all time”). But I worry that Pinker’s form of “Enlightenment progressionism” could lead to complacency, to a false sense of security at this critical juncture in human history.