This is a lightly edited text of a talk I gave at St. Anne’s College Oxford on 21 September 2019

We live in an age of anxiety – which may well be the default setting for all human societies, but which feels to many of us to be something new.

The reasons for the anxiety are many and potentially existential, beginning and possibly ending with the climate emergency. But there’s something else going on too: a loss of confidence that something we once took for granted — that western liberal democracies were securely entrenched – may not be true.

The global order is facing what some analysts have called a “democratic recession” – that is, for the first time since the 1980s the advance of democracy has gone into reverse.

To be fair, that’s a recession from a very high tide – there was a long wave of liberalisation beginning with the collapse of dictatorships in Latin America in the 1980s, and accelerating with the end of the Cold War.

But there’s one big difference today that makes the fate of liberal democratic regimes potentially more perilous than ever: the United States for the first time ever in its history is no longer, in its formal leadership at least, playing the democracy-promotion leadership role it once did.

Of course, the United States hardly has an unblemished record of democracy-promotion – in the nineteenth century it bullied and subjugated other peoples and territories just like the European powers, and in the Cold War covertly backed some brutal dictatorships. All that is true, and yet… rhetorically and ideologically if the US stood for anything it was “freedom and democracy”.

Today, however, there are plenty of commentators warning that something once unthinkable is now happening – American democracy is collapsing. The Economist magazine’s intelligence unit in its annual audit has reclassified the United States as a “flawed democracy” – marking it as substantially less democratic than Uruguay, Chile and even (bizarrely) the United Kingdom.

Why is this happening? One answer is just a syllable long: Trump. There is no precedent for having President with such open admiration for authoritarian leadership

But this isn’t just about Trump.

Opinion surveys show a large and rising minority of US voters display authoritarian traits. For example, a recent paper by political scientists Stephen V. Miller and Nicholas T. Davis based on survey data from the first decade of the twenty-first century found white voters were increasingly likely to support the idea of a ‘strong man’ leader.

In fairness, one might argue that there are signs of democratic health in US too. The types of Americans long excluded from the halls of power are entering them in greater numbers than ever before. This Congress has more gender balanced and is more racially diverse than ever. An openly gay man is running for president and his sexuality is barely an issue.

But the increasing inclusivity of American democracy over the last 50 years prompted a backlash among people willing to compromise their commitment to democracy in order to preserve their status or values they believe to be under threat.

I’d suggest that the underlying causes of the American democratic crisis are:

first, increasing economic inequality;

second, racial and ethnic tension,

which, third, have driven a partisan polarisation unprecedented in modern times.

And finally, in the background and insufficiently acknowledged are on-going consequences of the end of the Cold War.

Let me take each of these in turn.

First, the implications of decreasing economic security for ‘middle class’ and working class Americans.

The sociologist Seymour Martin Lipset warned 50 years ago in his study of democratic failure in the early twentieth century that “a society divided between a large impoverished mass and a small favoured elite would result either in oligarchy … or in tyranny”.

That’s not yet an accurate description of the United States, but since 2000 income has risen five times as fast for highest as for lowest earners, and as Sean Reardon from Stanford and others have shown, the achievement gap in education between rich and poor is 30-40 percent higher among those born in 2001 compared to those born in 1976.

In the US as in the rest of the western world, it is no longer the case that people can expect to be wealthier than their parents – one study showed that only 40% of millennials (born after 1980) were wealthier than their parents had been at the age of 30 compared to. 80% of baby boomers who were wealthier than their parents at the same age. That’s an amazing drop with big consequences for people’s faith in the system.

All the studies of the 2016 election I have read conclude that a deep feeling of powerlessness characterised Trump’s voters. According to Yascha Mounk’s analysis of 2016 data from the American National Election Studies, those who voted for Trump in the Republican primaries were more likely than those supporting his rivals to say that they “don’t have any say about what the government does,” that “public officials don’t care much what people like me think,” and that “most politicians care only about the interests of the rich and powerful.”

And it may well be that those voters were objectively correct to feel that way. In 2014 two political scientists, Martin Gilens and Ben Page, made an interesting attempt to quantify the power of economic elites to influence policy. In comparing the capacity of ‘ordinary citizens’, ‘pressure groups’ and ‘economic elites’ to influence policy-making Gilens and Page concluded that

“When the preferences of economic elites and the stands of organized interest groups are controlled for, the preferences of the average American appear to have only a minuscule, near-zero, statistically non-significant impact upon public policy.”

That’s pretty strong stuff for the country that’s supposed to have government of, for and by the people.

The last time large numbers of white Americans felt that the system wasn’t delivering for them was after the 1929 crash. In the 30s, fascism was far more appealing to Americans than we often remember, as, of course, was communism. But an important difference between the 1930s and today was that back then, the political system was capable of responding. Franklin Roosevelt worked with Congress to transform US government, accruing vast new authority for the Federal government in the process. One can approve or disapprove of what the New Deal did and tried to do but no one can deny its far-reaching effects. The political system, in other words, offered a means of addressing problems.

It doesn’t seem to now. The major piece of legislation passed by Congress was the Affordable Care Act in 2010. Even when the Republicans controlled both Congress and the White House from 2016 to 2018 they accomplished little other than a regressive tax cut.

Most of the changes that have affected US life in the last sixty years have come not from elected politicians, but from the Supreme Court – it has ordered the desegregation of schools, ended the death penalty (then restored it), struck down anti-abortion laws, decriminalised homosexuality and allowed same-sex marriage, struck down gun-control legislation and campaign-finance legislation and determined whether millions of people get health insurance or can be deported for not having proper immigration status. You may agree or disagree with any of these changes but the fact is that the central role the Court now plays in Americans’ lives has, in my judgement, made US politics less transparently responsive to popular pressure.

And then there’s race. Racial inequality has always driven American politics but it has manifested itself in profoundly different ways at different points in history.

Since the African American rights revolution of the 1950s and 60s, a hopeful narrative has been possible – with de jure racial injustice overthrown, perhaps a postracial future might lie ahead.

An increasingly visible middle class of people of colour and, above all, Obama’s election was taken by some as evidence that America’s racial divisions could be transcended. White Americans can now dismiss the problem of racial inequality without (in their minds) any association with the overt racism of the old southern segregationists. According to Pew surveys the vast majority of white Americans think racism is an individual problem – just a few bad apples. Black Americans know different – that racism is structural. The attainment gap between whites and people of colour, the huge racial disparity in the prison population, the disproportionate disfranchisement of non-whites – all these are evidence of this truth.

But racial politics in America today is no longer, as it once was, a matter of black and white. In retrospect perhaps the most far-reaching piece of legislation during Lyndon Johnson’s administration was not the 1964 Civil Rights Act but the 1965 repeal of the previous race-based restrictions on immigration. The huge surge of Asian and Latin American immigrants since the 60s poses what some white Americans now think of as an existential challenge to their idea of what America is. According to most projections the United States will be majority non-white by 2043. It is already the case that a majority of American children are non-white.

How does this affect the stability of the political order? Well, historically, the marker of the stability of a regime is the acceptance of the legitimacy of political institutions, and that, in turn rested on a high degree of acceptance of who was part of the political community.

 Whenever there was serious contest over that question – whose voice matters – the legitimacy of the system was called into question and instability resulted.

Throughout American history when white people faced a serious challenge to their dominance – cultural, economic or political – they have proved perfectly willing to abandon liberal values, as white southerners were in the Jim Crow era. One quite convincing way of explaining the rise of Donald Trump is that southern politics has gone national – the white majority, under threat, want to redraw the boundaries of who is in the political nation – hence disfranchisement and the clamp down on undocumented immigrants.

In different ways, economic insecurity and racial polarisation have both fed into the most dramatic ideological sorting of the two political parties.

Partisan polarisation was the great fear of the Founding Fathers that the republic would be destroyed by “faction” – by which they meant a powerful party that saw themselves as the only legitimate embodiment of the nation … and their opponents, therefore, as enemies.

Notwithstanding the Founders’ fastidiousness about this, the US has had political parties since the 1790s but for most of that time parties were coalitions with large areas of ideological and programmatic overlap.

In the 1960s and 70s there were still plenty of self-identified conservatives in the Democratic party, especially from the South, and plenty of liberal Republicans from the northeast. No longer. The ideological sorting of the parties was complete by the end of the twentieth century and has continued apace ever since.

Conflict seems irresolvable because politics is increasingly conducted by isolated, self-referential groups with sharply polarised identities. The rise of the internet is part and parcel of this development. Stuck in Facebook echo-chambers, avoiding information that challenges their preconceived narratives. A Reuters Institute (Oxford University) study in 2017 unsurprisingly found Americans’ news consumption was the most polarised in the western world.

Unlike in the polarised 60s, there is no Walter Cronkite figure to stand above the fray; no trusted national broadcasters respected by both sides of the political divide.

One might think, rationally, that the response to a divided electorate and the mixed messages of elections would be a bipartisan path, finding shared priorities. But that is impossible for the same reasons that create the problem in the first place.

And then there’s the impact of the ending of the Cold War. This, I think, is an often-under-appreciated element in American political development.

Between Pearl Harbor and the fall of the Berlin Wall, presidents were able to assemble bipartisan coalitions behind their foreign policy positions, and the US’s role in the world was generally accepted by Americans to underpin their own domestic prosperity.

But beginning in the 1980s, more and more people began to associate globalisation with declining economic security for average Americans. The scale of US global leadership was seen as an economic burden that might be discarded.

There had always been voices making that case, but so long as there was a clear and present danger from the Soviet Union there was a broad consensus uniting most moderate Democrats and Republicans – and that bipartisanship was reassuring not only to allies but also to voters who were then more likely to accept the legitimacy of leadership in Washington even if it was from a party they had not themselves supported.

The Cold War, in other words, provided clear boundaries to American politics. With those boundaries gone, the internal fissures have burst open.

I began this lecture by talking about the crisis of American democracy, but more precisely what is really under threat is actually a particular kind of plural, liberal democratic polity.

In the twentieth century, democracy was conventionally juxtaposed against totalitarian regimes of the left and right; in the twenty-first century, that polarity no longer makes so much sense. Where once “liberal” and “democratic” were almost synonyms, now the threat is that the US will slip into an authoritarian-lite illiberal democracy – look at Turkey or Hungary for examples — with a kleptocratic elite, and increasingly corrupted civil institutions – albeit one legitemised by elections.

Some scholars have suggested that the US has experienced three or four successive “regimes” or “political orders” in the past.

By political order, I mean the constellation of rules, institutions, practices, and assumptions that define how political actors operate and which – crucially – depend on a broad agreement about the boundaries of the political community.

So, in the years immediately after the American Revolution there was a brief period of rule by elites in which the franchise and the power of electoral politics deliberately limited.

That old order gave way in the early nineteenth century into a regime defined by vigorous, populist democratic politics but in which only white men could take part.

The issue of slavery, which profoundly divided white men, broke the political system in the 1860s and the amendments passed after it were a revolutionary challenge to the whites-only political order.

Yet that order, based on mass parties, survived until a wholly new polity was born in the twentieth century with the rise of what has been described as the “administrative state” or the “New Deal Order” —  in which policy-making was increasingly ceded to experts in arms-length agencies and the courts.

That New Deal rested on the foundations of a sense of collective responsibility, a willingness to trust government to act for the public good, and was bolstered by rising living standards and broad consensus about who Americans were and what their role was in the world. All those foundations have now been dismantled.

The US has reinvented itself before. Maybe even now it has the capacity, even in the absence of a Soviet-style external threat, to create a shared national project – a real governing coalition of the kind that has existed only for brief periods in American history.

It will not be possible unless a majority feel that once again they are getting a fair deal from the system.

I sometimes wonder if the reason I became a historian is that when you study the past you have the comfort of knowing what happened next. I wish I knew now what the next decades will bring for the US and the global order in which it’s so tied. We’re now at a crisis point in the transition from that old order to something else. Like the 1770s, the 1850s or the 1930s, we’re living at a time in which old certainties are crumbling into dust. I suppose whether that’s exciting or alarming depends as much on individual temperament as anything else.