Lately I’ve been thinking back to something that John Kerry told The Atlantic’s editor in chief, Jeffrey Goldberg, earlier this year. Asked about the importance of the Middle East to the United States, Kerry answered entirely about the Islamic State.

“Imagine what would happen if we don’t stand and fight [ISIS],” he said:

If we didn’t do that, you could have allies and friends of ours fall. You could have a massive migration into Europe that destroys Europe, leads to the pure destruction of Europe, ends the European project, and everyone runs for cover and you’ve got the 1930s all over again, with nationalism and fascism and other things breaking out. Of course we have an interest in this, a huge interest in this.

The 1930s all over again—Kerry was laying out a prediction in April, but it sounds a little more like description now. Even if America’s current dunderheaded demagogue loses the presidential election, the European project already falters in the United Kingdom, and Russia rumbles with revanchism. Fueled now (as then) by an ailing global economy, far-right nationalism seems ascendant worldwide. It’s hard not to think of the 1930s as the catastrophe which presaged our contemporary tragicomedy.

I write and report on climate change, not a pursuit that usually encourages optimism, but watching all this unfold with the atmosphere in mind has been particularly bleak. For the past few months in particular, I’ve been thinking: Wow, this is all happening way earlier than I thought it would.

Spend enough time with some of the worst-case climate scenarios, and you may start to assume, as I did, that a major demagogue would contest the presidency in the next century. I figured that the catastrophic consequences of planetary warming would all but ensure the necessary conditions for such a leader, and I imagined their support coming from a movement motivated by ethnonationalism, economic stagnation, and hatred of immigrants and refugees. I pictured, in other words, something not so far from Trump 2016.

I just assumed it wouldn’t pop up until 2040.

This kind of worry is speculative—very speculative—but it is not ungrounded. A large body of scholarship suggests that climate change could exert grave effects on international politics this century. Planet-wide warming will dry out regions of the world already riven with ethnic and political strife, all the while impoverishing and destabilizing the Western powers that backstop global order. A recent study even argues that climate-triggered environmental shocks will exacerbate the very divisions that authoritarians have historically sought to exploit.

So to now watch a demagogue contest the presidency, running a campaign that appeals to racism and xenophobia, has felt less like the sudden apparition of an unfathomable nightmare and more like the early realization of a seasonal forecast. You can hear the long-predicted gusts, the rain pounding on the roof and the groaning thunder. It’s all just happening four decades earlier than the weather person said.

So I want to propose a new way of understanding Donald Trump. He not only represents a white racial backlash, and he has not only opened the way for an American extension of the European far right. Insofar as his supporters are drawn to him by a sense of global calamity, and insofar as his rhetoric singles out the refugee as yet another black and brown intruder trying to violate the nation’s cherished borders, Trump is the first demagogue of the Anthropocene.

We should take Trump at his word when he calls Syrian refugees “one of the great Trojan horses,” or when his son bizarrely describes them as Skittles that “will kill you.” In Europe, Trump’s far-right kin have long blurred the differences between legal immigration, Islamist terrorism, and the refugees fleeting the Syrian War. After the Paris attacks last year, one leader of the French far-right National Front said, “Today, we can see that immigration has become favorable terrain for the development of Islamism.”

This xenophobia is grounded in real-life trends. I will focus on two in particular: moribund economic growth and the mass migration of non-white people. Both will likely intensify as the planet warms. (A third vital trend—the political and cultural upheaval of the U.S. racial hierarchy—will not vary with climate change.)

First, climate change could easily worsen the inequality that has already hollowed out the Western middle class. A recent analysis in Nature projected that the effects of climate change will reduce the average person’s income by 23 percent by the end of the century. The U.S. Environmental Protection Agency predicts that unmitigated global warming could cost the American economy $200 billion this century. (Some climate researchers think the EPA undercounts these estimates.)

Future consumers will not register these costs so cleanly, though—there will not be a single climate-change debit exacted on everyone’s budgets at year’s end. Instead, the costs will seep in through many sources: storm damage, higher power rates, real-estate depreciation, unreliable and expensive food. Climate change could get laundered, in other words, becoming just one more symptom of a stagnant and unequal economy. As quality of life declines, and insurance premiums rise, people could feel that they’re being robbed by an aloof elite.

They won’t even be wrong. It’s just that due to the chemistry of climate change, many members of that elite will have died 30 or 50 years prior.

Yet the second trend—the combination of mass migration and racist backlash—could push even more polities toward authoritarianism. Migration is also harder to predict than inequality: Wars and exoduses are not as easy to model as flood damage and agricultural yields.

But academics are trying. Jürgen Scheffran, a professor of geography at the University of Hamburg, has been investigating whether climate change makes armed conflict more likely for more than a decade. In 2012, he worked on a team that analyzed all 27 empirical studies investigating the link between war and climate change.

“Sixteen found a significant link between climate and conflict, six did not find a link, and five found an ambiguous relationship,” he told me. He described these numbers as inconclusive. Trying to prove that climate change is linked to war, he said, would be like trying to prove that smoking causes cancer with only one available case study.

“That alone was complicated to prove over time,” he said. “There were millions of cases of individuals who were smoking, and millions who got cancer, and you can develop a correlation between these two phenomena.”

“But there is only one world, and not a million worlds, in which the temperature is rising, and you cannot associate a single event—like a single hurricane or a single conflict—to climate change. It’s a statistical problem, and we don’t have enough data yet,” he said.

He sketched the basic dispute in the field: One set of researchers, whose most prominent advocates work at the Peace Research Institute of Oslo, contend that there is not a meaningful link between conflict and climate change. Another school of thought, which centers around researchers in the Bay Area, say that warming is already driving conflicts worldwide.

Partly it depends on whom you ask. Malin Mobjörk, a senior researcher at the Stockholm International Peace Research Institute, recently described a “growing consensus” in the literature that climate change can raise the risk of violence. And the U.S. Department of Defense already considers global warming a “threat multiplier” for national security. It expects hotter temperatures and acidified oceans to destabilize governments and worsen infectious pandemics.

Indeed, climate change may already be driving mass migrations. Last year, the Democratic presidential candidate Martin O’Malley was mocked for suggesting that a climate-change-intensified drought in the Levant—the worst drought in 900 years—helped incite the Syrian Civil War, thus kickstarting the Islamic State. The evidence tentatively supports him. Since the outbreak of the conflict, some scholars have recognized that this drought pushed once-prosperous farmers into Syria’s cities. Many became unemployed and destitute, aggravating internal divisions in the run-up to the war.

Scheffran underlined these climate connections but declined to emphasize them. “The Syrian War has so many complex interrelated issues—and most of them are political and economic—that the drought is just one contributing factor to the instability in the region,” he said.

This basic disagreement is what makes Carl-Friedrich Schleussner’s research so compelling. Schleussner, a researcher at the Potsdam Institute for Climate Impact Research, trained as a climate scientist. When he started studying armed conflict, he was surprised to see that many studies focused on changing rainfall or gradual temperature increases.

“What we found surprising about this debate was people were purely focusing on temperature indices or precipitation indices,” he told me. “In our analysis, it’s all about the exogenous shock. We were all interested in, to what extent does a big event like a flooding or a drought undermine society, or trigger a conflict outbreak?”

He and his colleagues suspected that most people can adjust over time to slower changes in their local environment. What they wanted to know what whether conflict could be stirred by a flood, drought, or wildfire—a sudden act of god that destroys deep wealth with little warning.

To get at these questions, they avoided using a meteorological database at all. Instead, they opted for an index of the economic damage wrought by natural disasters. They used, in fact, one of the largest databases of this type: the natural-disaster-cost data assembled by Munich Re, a enormous global reinsurance company. (Reinsurance firms sell insurance to consumer insurance companies, and therefore must monitor the changing costs of natural disasters.) Munich Re’s tally encompasses over 18,000 “climate-related events” between 1980 and 2010. Schleussner and his colleagues were the first to use it in climate research.

They were not disappointed. Heatwaves, droughts, and other climate-related exogenous shocks do correlate to conflict outbreak—but only in countries primed for conflict by ethnic division. In the 30-year period, nearly a quarter of all ethnic-fueled armed conflict coincided with a climate-related calamity. By contrast, in the set of all countries, war only correlated to climatic disaster about 9 percent of the time.

“We cannot find any evidence for a generalizable trigger relationship, but we do find evidence for some risk enhancement,” Schleussner told me. In other words,  climate disaster will not cause a war, but it can influence whether one begins.

Schleussner demurred when I asked him to think through some scenarios. Right now, his conclusions don’t suggest how climatic disasters become armed conflict. He also does not believe that every ethnically divided country will crack up as the planet warms. “Out of the 50 countries that can be classified as ethnically factionalized, there’s also Canada. And Canada is not that conflict-prone, apart from hockey, right?” he said.

But his paper does detail particular areas of concern—and both of them border Europe. Models predict that northern Africa and the Levant, both already drought-prone, will dry out significantly over the course of the century. On the phone, Schleussner also cited southern Africa and south-central Asia as regions to watch. (It’s no coincidence that some of the largest, longest wars this century have occurred in those places.)

Schleussner and his colleagues also allude to a nightmare scenario in the paper itself, though they couch it in clinical language: “Further destabilization of Northern Africa and the Levant may have widespread effects by triggering migration flows to neighboring countries and remote migrant destinations such as the European Union.”

In other words, a drought-and-flood-fueled armed conflict near the Mediterranean Basin could send people toward Western Europe in the hundreds of millions. This is the “1930s all over again” scenario that Kerry mentioned, the one playing out in miniature right now, made all the worse through the aggravation of a climate-changed world.

Never mind armed conflict. Could disastrous environmental upheaval produce mass migration all by itself?

The consensus here is more muddied. “It’s very difficult to predict anything about migration, generally speaking,” says Cristina Bradatan, a professor of sociology at Texas Tech University who studies, well, migration. “And when we talk about climate-change migration, that is forced migration—which is the worst situation from the point of view of projection. So if there is a consensus, it’s that it’s very difficult to know what will happen.”

She also resisted commenting on the nightmare scenario. “In Europe, there is all this fear that there will be a huge mass migration from Africa due to environmental changes,” she said. The literature makes that seem unlikely: Most poor people have neither the means nor the international connections needed to actually flee a country, she said. So if their homes become untenable, they instead move to the nearest safe place—which is often the nearest city.

“I wouldn’t say that there would be a mass migration to Europe, but I would expect to see a large number of people being displaced within Africa,” she said.

No matter what, any international mass migration will produce systemic strain. If people do cross a national border while seeking refuge, existing international law will not have a mechanism to understand them.

“When I would teach my students this, I would say: There is literally, in legal parlance, no such thing as an environmental refugee,” says Edward Carr. “To meet the international standard for refugee, a changing environment is not a forcing. It doesn’t count.”

Carr is a professor of international development and the environment at Clark University, and he sits on the Intergovernmental Panel on Climate Change. He believes governments must focus on reducing the risk of migration crises in the first place, partly because defining and creating a safety net for environmental refugees will prove so difficult.

“The reason this is a problem, and so challenging to deal with politically, is that this is what they call the slow-onset disaster,” he said. “Rarely will we see situations where a place just goes bone-dry for 10 years, and everyone has to move or die. That’s not the scenario.”

He went on. “As I used to pose to [students]: When would you attribute the decision to move to changes in the climate? Does a place have to be dry for five years? For 10 years? Does someone have to have three children die, and then they decide to move? There’s no line for this.”

Climate change could push Western politics toward demagoguery and authoritarianism in two ways, then. First, it could devastate agricultural yields and raise food prices; destroy coastal real estate and wash away family wealth; transform old commodities into luxury goods. Second, it could create a wave of migration—likely from conflict, but possibly from environmental ruination—that stresses international reception systems and risks fomenting regional resource disputes.

In effect, it could erode people’s sense of security, pushing them toward authoritarianism.

If this model of authoritarian response seems simplistic, that’s because it is. Economic strife and mass migration have produced far-right authoritarianism in the past, so I assume they could in the future. Empirically speaking, financial crises especially seem to cause a flight to the extremes. But they do not guarantee anything, and I’ve focused on the questions of contributors—mass migration and conflict—because they are easier to predict than politics.

Yet in doing so I’ve committed the trend-tracker’s fallacy. Like the CEO in the 1950s who predicted that America would see flying cars and three-day workweeks by the year 1999, I’ve assumed that every ongoing trend line can be extrapolated out indefinitely. They can’t. The actual future will be far stranger.

Yet history will still be constrained by demography, ideology, and atmospheric chemistry. “There are certainly plenty of viable scenarios where people could move into places and you get an ethnonationalist backlash. We’re seeing it!” Carr said at one point. “So clearly it could happen again. And I absolutely am convinced that, in the long run, the effects of climate change will be problematic and destabilizing for many of us in many places.”

Trump is, in essence, a double case—a preview of what’s to come and a way to practice dealing with it. He represents a test that the leaders of a major American political party are failing, and that the electorate may only narrowly pass. He is showing us how ill-prepared the United States is for post-climate demagoguery, and he gives us an opportunity to improve our societal immune response.

How might we do that? His rise also suggests a number of defense mechanisms. Obviously, the first is that climate change must be mitigated with all deliberate speed. But he also suggests certain cultural mechanisms. Some Americans may favor more restrictive immigration policies, but—in order to withstand against future waves of mass migration (and humanely deal with the victims of climate change)—racist fears must be unhooked from immigration restrictionism. In other words, as a matter of survival against future authoritarians, white supremacy must be rejected and defeated.

And there is a third method of fighting back against Trump and his ilk. Carr doesn’t think it makes sense to improve the response in receiving countries; I am less convinced. After all, he also told me that reception to migrants in the U.S. depends greatly on regional cultures. Central Massachusetts, where he lives now, welcomed about 10,000 Ghanian refugees in the 1970s, and it never entered a period of mass white-nationalist revulsion. The United States still welcomes and integrates immigrants faster than European countries.

Those regional cultures can still be improved and strengthened. In April, a poll conducted by The Atlantic and the Public Religion Research Institute found that the voters most likely to vote for Donald Trump were civically disengaged—they did not go to church, or volunteer at school or Girl Scouts, or join a book club. These Americans were also more likely to be financially insecure and less likely to be well educated.

When journalists write about how you can avert the worst of climate change, they focus primarily on technological means. The environmentally anxious are encouraged to give up industrial beef, to buy carbon credits, to install rooftop solar panels. An entrepreneurial neighborhood might be told to build community solar. A well-off consumer might be asked to splurge on an electric car.

The only social or political act that most of these explainers will propose is this: You should vote for candidates who understand climate science and who will act to reduce greenhouse gas emissions. (In the United States, and in no other developed country, that only describes the candidates of one major political party.)

It makes sense to put voting  in that list because, really, all those technological actions resemble voting—they are all essentially just types of harm reduction. At worst, skipping beef or buying carbon credits is an ethically valuable but economically worthless gesture; at best, it modestly helps avoid a much worse outcome.

But Trump’s success in the primary among the civically disintegrated suggests another way forward. Improving the United States’s immune response to authoritarian leadership—a response that could be repeatedly tested in the century to come—can follow from weaving its civic fabric ever tighter. I don’t know what this will look like, exactly, for every person. But here are some places to start: Volunteer. Run for local or state office. Give to charity (whether due to religion or effective altruism). Organize at work. Join a church or a community choir or the local library staff. Make your hometown a better place for refugees to settle. Raise a child well.

These may seem inconsequential, tasks unrelated to the final goal of restricting how much carbon dioxide enters the environment. And, admittedly, they are. But climate realists have always split their work between mitigation—that is, trying to keep the climate from getting worse—and adaptation—trying to protect what we already have. As more warming gets baked into the biosphere, as seas rise and livelihoods fall, these prosaic steps will become vital forms of adaptation.

Climate mitigation is a worthy goal in itself. It is all the more important when understood as one more type of long-term anti-fascism.