“Intellectuals hate progress,” Steven Pinker states in his latest book, Enlightenment Now: The Case for Reason, Science, Humanism, and Progress.

Coming from a top intellectual himself–a distinguished professor of psychology at Harvard University and the author of several best-sellers, including The Blank Slate and The Better Angels of Our Nature: Why Violence Has Declined–this assertion could seem odd. But Pinker explains that it is not progress itself that intellectuals have an aversion to, but the idea of it.

In Enlightenment Now, Pinker makes the case that, despite all we think we know about the poor state of the world–from what we hear from the media, our friends, our political parties–the world is actually improving on a number of fronts. Pinker’s approach is an extension of his take on violence, which he explored at length in The Better Angels of Our Nature. In Enlightenment Now, he applies a scientific perspective to trace 15 separate measures of progress, in chapters such as “safety,” “democracy,” and “the environment” to assess where we stand.

The result, according to Pinker, is that we are better off than ever. “Not only have we become less violent overall,” he told me, “but we have reduced global poverty, illiteracy, disease, and hunger.”

I spoke with Pinker about why he sees a dark side of political correctness, the differences between men and women, and his optimism about artificial intelligence, among other topics. Here is our conversation, edited for length and clarity.

Hope Reese: As you did your research for Enlightenment Now, what did you learn that was unexpected?

Steven Pinker: I had not realized that every measure of safety had shown vast leaps–on the order of 95-to-99 percent reductions. Car accidents, pedestrian accidents, even lightning strikes–we are 99 percent less likely to be killed by a bolt of lightning than our ancestors were a century ago. Extreme poverty–the ability to feed oneself and one’s family–had declined to less than 10 percent of the world’s population, and there was a goal to eliminate it altogether by 2030. The world is very close to parity in girls’ education.

Do “reason” and “science” have to go hand in hand? Can science be biased? What about intuition?

They are completely related. But it’s important not to confuse “rational” with “infallible.” The entire ethic of science is that infallibility is impossible. That scientists are human. That the norms, such as peer review, open debate, empirical testing, are important precisely because scientists, like all humans, are prone to error and bias and selective attention and confirmation bias. The valorization of science is not the valorization of scientists. Science is so valuable because it’s based on the idea that all humans, including scientists, are fallible.

On climate change, the science nearly unequivocally points in one direction–that we are causing irreparable harm to our planet. Why are you optimistic that things can change? 

If the world continues with business as usual, there’s no reason for optimism. The question is: Is the world committed to business as usual? Or will changes be made in policy and technology that divert the disastrous outcomes? The main way to persuade the deniers is not to have them pay attention to the science, but to depoliticize the issue as much as possible and to prevent climate change from being a litmus test of which political ideology you belong to. There’s no evidence that deniers of climate change are any less scientifically literate than people who accept man-made climate change.

A lot of people who accept the reality are scientifically clueless, but they know it is a belief they ought to embrace if they want to be members in good standing in the social and political cliques that they value. Likewise on the side of the deniers. We do need people associated with the political right and libertarian right to publicly acknowledge the scientific case. And we need to decouple the factual hypothesis of man-made climate change from particular remedies for it that rankle people––such as abolishing capitalism, returning to a technologically primitive lifestyle, undergoing sacrifices.

You’ve been critical of the negative spin the media puts on news events. But what’s the harm in pointing out where we could be doing better as a society?

There’s no harm in pointing out where we could be doing better. The harm is giving the impression that we’re spiraling into disaster. That society is constantly deteriorating.

There are three harms. One is lack of appreciation of efforts in the past that have led to the progress we enjoy in the present. This leads to a cynicism about institutions of liberal democracy, of the rule of law, and science, and of the benefit of a free press. The second, complementary danger, is fatalism–the belief that we are doomed no matter what we do, so we might as well enjoy life while we can. The third danger is radicalism–the belief that society is so corrupt and decadent and doomed, that the our only recourse is to destroy the institutions around us. To burn the empire to the ground, to drain the swamp, to smash the machine––to lurch to some radical alternative that can only be better than the hellish dystopia we’re in.

One could argue that the rhetoric of Trump’s presidential campaign, focusing on the dangerous state America is in, helped get him elected. What’s the psychology behind that?

We see that in lurid detail in his infamous “American Carnage” inauguration address, which was largely written by the now-departed Steve Bannon. Trump campaigned on the premise that the country was in a state of crisis and deterioration. He was abetted by coverage of relatively minor events that got saturated coverage–like terrorist attacks and rampant police shootings–all of which are dwarfed by the unreported violence of day-to-day street crime, which has drastically decreased over the last 25 years. And abetted, ironically, by some of his fiercest opponents on the left, who portray current American society as a hellhole of inequality and racism. Indeed, one of the strongest predictors of votes for Trump was a belief that society is deteriorating, that everything is getting worse.

You point out religion as a way of thinking that falls on the flip side of rationality. And you also write that politics have become “a form of religion.” What do you mean?

When political ideologies become dogmas, where people feel that they do not need evidence to understand how the world works, that government policy always makes society worse, or society better–for example, that capital punishment deters murder, or doesn’t–then one can see the religious mindset affecting political thinking. But even more than ideology, we have seen that raw tribalism can distort reason in the public sphere. We see the astonishing flip-flops of Republican voters in the last election, such as sympathy to Russia, or hostility toward free trade–it’s about being loyal to one or another team, and embracing whatever position the standard-bearers endorse.

Right now, we’re having a hard time as a country agreeing on facts. Yet you wrote on Twitter that we are not in a “post-truth” era. Why not?

One can’t argue that we’re in a post-truth era because you could say, well, “is that statement true?” There are lies. There always have been lies. There are conspiracy theories. There always have been. There’s been a recent technology-enhanced effort to spread misinformation and conspiracy theories, but it’s hardly as if this is a new development in human history. There have been wars started based on false information–like the Spanish-American War, the escalation of the war in Vietnam, the Iraq War in 2003.

So you don’t think there’s something particularly dangerous about cyberwar that can spread misinformation and put democracy at risk?

I don’t want to criticize the media across the board, because part of the problem is the wholesale attack on the press. But there is a bad habit of journalists treating isolated events as signs of sea change or revolution. That’s one of the recurring fallacies of journalism. Since there’s a terrorist attack, society is becoming more violent. Because there is racist graffiti, there’s an increase in hate crimes. The idea that because there are some Macedonian sites spreading misinformation, we’re in a post-truth era, is an example of the bad habit of journalists of confusing an event with a revolution.

There is the very worrying spreading of lies. And there’s also an unprecedented apparatus of fact-checking. Organizations like PolitiFact and Snopes and Wikipedia. And an increased desire on the part of news-readers to have politicians fact-checked––an ethic we did not see 15 years ago.

Where is the line between free speech and political correctness? Are there certain things that should no longer be said? Ideas that should not be promoted?

There’s a history of American jurisprudence on legitimate constraints on free speech, like libel, extortion—a dozen leading restrictions. But they do not extend to the expression of popular opinions. The idea of free speech does not commit us to providing a platform for kooks and liars and people who have no respect for truth-seeking and testing and rationality.

But many of the liberal outbursts on campuses have not been directed at demagogues or fanatics, but rather at scholars who made a reasoned case for their positions. They may be mistaken, but they are certainly worth hearing. The reason for the opposition isn’t that there’s an absence of attention to argument and data, but rather that the conclusions are unpalatable, according to the prevailing politics of the institution.

You said recently that there are certain subjects that have become taboo–such as recognizing differences between men and women. What do you see as fundamental differences that we shouldn’t be afraid to discuss?

There are not that many fundamental differences between men and women. The only one that’s truly fundamental is sexuality. Among men, there’s a far greater desire for sex with a variety of partners than among women–including the use of various aggressive tactics to try to achieve a variety of partners. In the recent scandals, there have been very few episodes of female supervisors emerging from the shower naked to shock their employees, or masturbating in their presence. The difference between the sexes is quite stark.

For the other differences, they are not fundamental–they are quantitative. There appears to be a statistical difference in life priorities. Professional power and status are given more weight, on average, by men. Family life and improving human welfare are given greater weight, on average, by women.

There are also some psychological differences that may be minor, on average, but produce differences in extremes. Men tend to be represented in very high and low ends of ability. They are overrepresented in retardation and autism and many disorders, and also overrepresented in the highest ends of mathematical achievement and spatial ability. Women are better, on average, than men in verbal fluency and reading emotions from faces and behavior—reading between the lines, more generally. These are far smaller differences than those in life priorities and sexuality. Men are also far overrepresented in violence and jealousy.

How many of these differences can we attribute to how men and women are socialized?

The existence of differences is a distinct question than where the differences come from. You can look at differences across cultures–there are certain sex differences, like a preference for pink versus blue, that we know can be reversed in other cultures or historical periods. But there are others, such as violence and sexuality, that are pretty much universal across cultures. We can look at other species–if there are differences that we see in humans that aren’t common in other primates, that leads us to suspect that they may not be a matter of a particular culture. We can look at exotic cases of girls raised as boys, or vice versa, to see if, despite the societally-assigned gender roles, they still exhibit “male” or “female” traits. And we can look at practices of socialization and culture and see if “male” traits are encouraged or valued.

I expect there is not a lot of social prestige that comes from a man masturbating in front of a woman. We don’t tend to valorize those people, but men do it anyway–which suggests that the reason they do that is not because they are socially rewarded. Likewise, for infidelity, even with the sexual revolution, there’s been virtually no change in acceptance of cheating on a spouse, yet we have reason to believe this is far more tempting to men than to women.

You wrote in the New Republic that the humanities have failed to “define a progressive agenda” and issues within the discipline have been “self-inflicted.” What did you mean by that?

This was not applying to the humanities in general, but to the mainstream that has signed on to post-modernism as an overarching philosophy. In that ideology, there is no difference between liberal democracy and fascism. The institutions of modernity are responsible for horrors like racism and genocide, the pursuit of objectivity and truth are delusional, all statements are ultimately paradoxical–those beliefs sap the energy toward understanding our problems and trying to ameliorate them. As such, they don’t offer a progressive agenda for those who might want to improve society, but lead either to the sterile practice of just piling criticism on top of criticism or to radical alternatives, such as tearing down institutions, because they believe that anything will be better than what we have today.

Bill Gates–who called your book his “favorite book of all time”–argues that you’re “a bit too optimistic about artificial intelligence.” Can you explain why you aren’t worried?

I think there is a challenge that comes from automation–that as jobs are performed by machines, we should be concerned whether the economy is nimble enough to create new occupations to employ people as old occupations become obsolete. And about whether the ability to have a good life depends on the ability to have a good job. Should we look to policies like a universal basic income that can cushion people from the displacement that automation will bring?

The other fear that has been expressed about intelligent systems–that AI systems will take over–that is, displace humans and maximize their own power–or, by accident, will bring on catastrophes because they will take the goals we program into them too literally, and harm people as a byproduct, I think are fanciful.

But what about those who advocate for “safety” in AI–that is, designing AI systems that we can understand?

There’s the ethic of safety engineering–that you don’t give a system control over machinery that can harm humans until you have tested for safety and built in safeguards. I don’t see any evidence that, somehow, engineers have completely forgotten that that is one of the responsibilities of being an engineer. And they won’t start, idiotically, handing control to systems that have not been tested beforehand.

I don’t think it’s very likely that Tesla will put a self-driving car on the market, that if you program to take the shortest route to the airport will take it literally, and drive a straight line, mowing down pedestrians on the sidewalk. That would be such an idiotic way to design a car. It would never occur to Tesla to sell a product that would do that.