When Margaret Thatcher died, there was a debate in the UK parliament on whether Prime Minister’s questions should be suspended to allow MPs to pay tributes instead. In other words, should the professional and democratic obligation of UK legislators to hold to account the government of the day be waived in favour of self-aggrandising, sycophantic, verbal masturbation? The veteran Labour MP Dennis Skinner was savage. Listing all the pressing concerns on which he would have wished to demand action from the Prime Minister, he noted that the luxury of simply downing tools and mourning had never been available to his former mining comrades. When a miner died on the job, the co-workers who carried his cadaver out of the pit were allowed to go home, and everyone else had to return to work. This was due a deal that had been struck by the union that would allow for some payment to his widow. In the commons chamber, a state funeral, no less, for their former colleague assured, Conservative MPs practised their Churchillian eulogies, safe in the knowledge that they were perfectly within their rights to take the day off legislating. They admonished those like Skinner who declined to conform, but he ploughed on regardless. "Why the difference?", he demanded. In my view, it’s one of the best speeches he ever made. The unarguable glaring differences in how working class and ruling class people are treated was laid bare. The absurdity of Tory MPs’ calls for “taste” only served to make his argument for him, exposing the hypocrisy of those who would always deny their poorer constituents the luxury of mourning, but could not fathom that someone would want to deny them that right. As Skinner reached his crescendo, he hammered it home repeatedly: "What's it all about?" he asked. "It's all about class."
Well, not always. In this post, I am going to take issue with Michael Lind's thesis that populist movements can be explained in terms of a class war between a managerial class which is liberal, university-educated, pro-trade, pro-immigration, and an anti-establishment working class. Instead, I want to argue that this is better understood as a culture war: essentially an orthogonal diversion from class war. Furthermore, those who, like Lind, would support the idea of "countervailing power" to strengthen working class communities against the elite cannot do so while maintaining any kind of truce with populist movements. In other words, any genuine working-class movement can only succeed if it is also uncompromisingly anti-racist.
Why am I going after Lind? Only because he is the latest person to advance this kind of argument. He is also to some extent a soft target. It would be easy to make an ad hominem attack on him based on his soggy justification for Trump's response to Charlottesville in the Newyorker: he couldn't possibly have intended to give succour to anti-semites because he is politically close to Netanyahu and has Jewish family. Or his lazy and frankly boring attacks on political correctness as "distinguishing the college-educated from the vulgar majority below them", as if, in the face of all evidence to the contrary, working class communities are in some way incapable of being inclusive of LGBTQ people and ethnic minorities. However, his account is not especially original. There are echoes of it in the "Blue Labour" project which sought to re-connect the Labour party with its working class traditions and in the process triangulate with the right on issues of culture. This ended with "controls on immigration" mugs and defeat in the 2015 general election, but is now experiencing a revival, albeit euphemistically, as people talk about reconnecting with the so-called "red wall" which fell to the Conservatives in the 2019 general election, with the supposedly left-wing leadership candidate Rebecca Long-Bailey conspicuously failing to challenge voters who blamed NHS waiting times on immigration in a video posted on her own website. Labour MPs John Trickett and Ian Lavery pin the loss of Northern voters to the Conservatives on a working class revolt against the neo-liberal policies of the Blair era. In the US, there are plenty who have taken the line that the Trump movement in particular and right-wing populism in general can be understood as a working-class movement. In other words, while it is Lind which has provoked this rant, I am more interested in taking issue with his argument about the nature of populist movements.
Disentangling class, race, education and anti-establishment sentiment in the Trump vote
So in what sense does Trumpism represent a working-class movement? Ostensibly, the case is not too difficult to make. His positions on social security, infrastructure, and trade appeared to be aimed at winning support amongst the post-industrial parts of the US ("flyover country") and were conspicuously to the left of the traditional Republican platform, if not especially radical; he certainly appeared to be anti-establishment, as evidenced by the Republican party doing all it could to prevent his nomination. But Lind wants to go further, pointing out that his racism with respect to Mexican immigration ("they're bringing drugs; they're bringing crime") was founded in the economic concerns of the US working class that they were facing competition from immigrants. There are certainly echoes of how Brexit has been understood by its proponents as anti-establishment (Farage's claim that Brexit was a victory for "ordinary people") and again, ostensibly it is, with both major parties, the Treasury, the IMF, and most of the stock market being opposed to it at the time.
However, there are some conceptual distinctions which need to be ironed out in order for this to be discussed with any precision. The first is that anti-establishment politics does not need to be class politics. Class either refers to the economic incentives conferred by one's relationship to the labour market, or it means nothing specific. The most complete operationalisation of the concept of class that I'm aware of is the Erikson-Goldthorpe scheme, which breaks class down into seven categories based on the type of job and associated benefits: professional, semi-professional, self-employed, routine nonmanual, foremen, skilled manual, semi-skilled manual, unskilled manual. The crucial insight of this is that class goes beyond one's income or wealth. It refers to how sensitive one's economic wellbeing is to economic shocks and particular policies. So, for example, although a primary school teacher and a trainee barista might have a similar income, they cannot be considered to be in the same social class because the first has high job security and the second does not; the first has access to a defined benefit pension, sick pay, and annual leave, while the second is unlikely to; politically, the first has an interest in expansion of the public sector, and the second does not; the second has an interest in expansion of the insurance arm of the welfare state, and the first does not. These are the distinctions that Skinner might note and then fairly judge to be "all about class". Note, therefore, what class is not. It is not the same thing as education, support for the establishment, income, wealth. If this sounds like pedantry, I hope to show that it is important later on.
Surprisingly few researchers in the US appear to have studied class voting appropriately, but the general social survey of 2018 includes questions on occupation class and race, as well as vote choice. Morgan and Lee investigate turnout by class using the 2016 survey, and do so by sorting occupations into the Erikson-Goldthorpe classes. They detail how they map each occupation to a class in an online supplement, and I have applied the same methodology to the 2018 data in order to work out how vote choice correlates with class. To abstract from turnout issues, I investigate only amongst those who voted either for Trump or Clinton. And I also exclude military and those who did not answer the question about occupation from the dataset. This leaves us with 2235 people from the original dataset of 2348. The results (percentage voting for Trump) by class and race are plotted below.
Looking at the distribution of votes by social class, it’s hard to see this as a straight-forwardly “working class” movement. The working class might be roughly defined as encompassing unskilled and skilled manual and routine non-manual. It’s certainly true that skilled manual were amongst the most likely to vote for Trump, but so were self-employed non-professionals. Routine non-manual and unskilled manual voters on the other hand were amongst the least likely to vote for Trump. Look how much easier it is to make the argument that this reflects racial voting! In every social class, white people are more likely to vote for Trump than non-white people. In every social class except higher-grade professionals, white people were more likely to vote for Trump than Clinton. As Paul Mason put it immediately after the election “millions of middle-class and educated US citizens reached into their soul and found there, after all its conceits were stripped away, a grinning white supremacist”.
The CNN exit poll can also tell us a lot about the character and nature of Trump’s voters. Although as I have just said, income is not a good proxy for class, it's notable that the poorest Americans were not most likely to vote for Trump. It's also notable that the wealthiest Americans were not less likely to vote for Trump. And union households were as likely to vote for Trump. The numbers are close, and there appears to not be a lot here. Turn to social issues, though, and it becomes much clearer. Only 14% of LGBTQ people in this survey voted for Trump. 65% of people who were concerned by Trump's treatment of women (as a reminder, he boasted about serial sexual assaults) voted for Clinton. 73% of people who believed there was not a problem in how the Federal government treated black Americans voted for Trump. A majority of Christians and particularly evangelical Christians voted for Trump. And a strong majority of non-whites voted for Clinton. None of these can be considered class-related, and all of them appear to be stronger predictors of vote choice than class: that is, while the Trump vote might be seen as anti-establishment (a kick against a cultural elite perceived to be too anti-sexist, anti-racist and accepting of sexual minorities) it can hardly be considered class politics. More precisely, being white, religiously conservative, and unconcerned about the equality of women seem to be the main determinants of a Trump vote.
This brings me to education, and a further distinction between class voting and class politics. If we ignore the fact that there are non-white people in the US, it's certainly true that white people with no college-level education were more likely to vote for Trump than white people with a college degree. First, though, note that both were more likely to vote for Trump than Clinton, while non-White people with no college degree were less likely to vote for Trump. So even if I concede this, the Trump vote is better explained by race than class. But secondly, as I have already said, education and class are not the same thing. As Oesch describes here, it is totally possible for differing educational experiences which give rise to different value sets to lead to certain values being disproportionately represented in certain occupations. Specifically, in the US case, if going to university exposes people to more liberal values, and these people then become disproportionately represented in, say, the legal profession, then it might give rise to class voting. Does this make it class politics? Not really. Class politics occurs not just when sociological correlations give rise to particular but when parties deliberately mobilise on the basis of class (e.g. see Evans here). That is, you might get class politics without class voting, or class voting without class politics. I don't see how Trump can be considered to be mobilising the working class when his language deliberately, repeatedly alienates the non-white element of it.
So to summarise, these are some distinctions to be made. Class refers to the economic incentives that variable exposure to the labour market give rise to. Class voting occurs when this is correlated with vote choice. Class politics occurs when these groups are mobilised on this basis. As the above demonstrates, the Trump movement was not an example of either class politics or class voting. It much closer to being identity politics, where whites and religious conservatives were mobilised into a coalition against liberal values. This is shown both by the fact that these were the stronger predictors of vote choice and by the difference in opinions between Trump and Clinton voters.
Is there any (working) class politics in the Trump movement? Analysing the marginal Trump voter
If I was an advocate of Lind's view, I would currently be arguing the following: yes Trump combines class politics and identity politics, and yes part of the Trump vote was driven by social issues which were motivated by anti-establishment sentiment as well as class issues, but you can't deny that a significant part of Trump's appeal lay in his attempts to appeal to the economic interests of his base. Okay then, let's look at the policy preferences of Trump's voters (using the same dataset). Most thought that the Federal government did too much. So on this classic issue of class (do you think there should be a welfare state to protect against the worst excesses of the market?) working class economic interests cut against voting for Trump. It's not clear prima facie whether support for increased deportations should be read as opposition to competition for jobs, or a dislike of non-white people. The strongest case that can be made is that Trump voters were motivated by economic concerns which were distinctively class-based comes on international trade. Most people who thought that international trade destroyed jobs supported Trump. So the best that can be said is that class politics in the Trump coalition was limited to the white working class interest in less trade and more deportations. That’s a very thin form of class politics which ignores a large part of the domestic working class and the vast majority of the foreign working class.
Suppose we stick with this, then. Here's the argument: international trade theory (e.g. the Hecksher-Ohlin model) tells us that as economies open up to international trade, the owners of the less relatively abundant factor of production lose out. In the US, this is unskilled labour. Hence, trade policy is an issue which falls along class lines. Unskilled labour therefore has an interest in voting for Trump. Trump combines his anti-establishment, conservative, social appeal and otherwise liberal economic policy with a targeted pitch to unskilled workers through his stance on trade. Those workers who are white and are not sufficiently concerned about his racism or his stance on other economic issues are therefore tempted to vote for him based on these economic interests. It's just about plausible. It does make the argument into a shadow of what it was. Rather than saying that the Trump movement is a working class movement, it says that while it might have been primarily an orthodox conservative movement with a highly anti-establishment flavour to it on social issues, there was a section of his support which did support Trump because of his appeal to that particular social class. To do justice to the argument, they might further add that this section of Trump's support were the "swing" voters: those who might have voted for Clinton but were won over by Trump's economic proposals. To fully explode this argument, I propose to demonstrate that not only was Trump's average supporter not motivated by concern for the economic wellbeing of the working class, but also that his marginal supporter was not either. In other words, however much it might be comforting to the left to think that the white working class deserted them to vote for Trump based on their economic interests, this is not an argument that can be sustained.
To answer the question about the average white working class Trump voter, we can turn to the Atlantic-PRRI study based on surveys in the run up to the 2016 election. They surveyed 3043 white working-class Americans in September-October 2016, asking around 120 questions about their social and economic position and political opinions. The results paint an extremely interesting picture of white working-class life and opinion in the US. The dire overall situation is apparent from questions on personal finances (26% would find it very difficult or impossible to meet an unexpected $400 bill), and mental health (38% of the sample either suffered from depression or lived with someone with depression; the corresponding figures for alcoholism and drug addiction were 12% and 8%). This tallies with the “deaths of despair” theory of Case and Deaton, who note that the opioid epidemic appeared to be curiously concentrated amongst white Americans.
Of interest for our purposes here is a logit model constructed by Cox, Lienesch and Jones which attempted to explain what drove certain voters into the arms of Trump versus others. (A logit model tells us which factors influence the probability of a particular event happening. In this case, what opinions make it more or less likely that a voter will plan to vote for Trump.) If Lind is correct, we might expect to see that those white working-class Americans who are most economically dispossessed, or the most economically populist, were most likely to be considering a Trump vote. However, this is not their finding. Of the factors which were the statistically significant (i.e. the results are numerically distinguishable from those that would arise randomly with reasonable probability), the order of size was: identifying with the Republican party, agreement with the view that America was in danger of losing its culture and identity, believing that illegal immigrants should be deported, believing that going to college was an unnecessary gamble, and (this only at a 10% significance level) fragile personal finances. This final factor pointed in the opposite direction, however. Having poor personal finances made a person less likely to be considering a vote for Trump. Note what didn’t make a difference: gender, age, downward social mobility, adverse health conditions, religion, views on gender.
How does this square with the previous evidence? Previously we saw that opinions on gender and one’s religion had an effect on one’s likelihood of voting for Trump. It could be that this pattern does not hold amongst the white working class, but partially it may be the inclusion of party identification in the model. If being a Republican is correlated with being an evangelical Christian or conservative views on women then this could obscure the independent impact of these. Fortunately, we can test this because the PRRI provides the part of dataset they use. I was able to recreate this model following their methodology as closely as possible. They do not say exactly what model they use- I used a binomial regression. I also followed their methodology in excluding all voters who said they did not know who they were planning to vote for, but also all voters who said they did not know their current social class or the social class of the family they had grown up in, because there is no obvious way to use this in calculating whether there had been downward social mobility. The results will differ partially because they appear to have access to data collected after the election in addition to prior to the election, whereas the PRRI data only have information on people intending to vote for Trump. However, I can get passably similar odds ratios, with the exception that the coefficient on the poor financial situation is not significant at all in my regression. Now what happens when we exclude the Republican dummy and re-run the model? The results are in the table below:
The odds ratio refers to how the probability of voting for Trump increases with each of these factors. For example, someone who is over 65 is only 25% as likely to vote for Trump as someone who is under 65. This might seem surprising, but bear in mind that this holds everything else constant. The reason the young tend to be less likely to vote for Trump is because they are less conservative on average. In this group, holding views constant, the young are more likely to vote for Trump. This is actually consistent with Cox, Lienesch and Jones’ finding that younger white working-class people are more likely to identify with the Republican party. The p-value is the probability of observing values as extreme or more extreme in a random sample, assuming that in the whole population of white working-class people, there is no effect at all. That is, the smaller the p-value, the less likely that we have observed what we observed purely through sampling error. Normally we consider p-values of less than 5% to be “statistically significant”. These are bold in the table.
With this analysis we can be confident that views on social issues were much more important than views on economics in determining votes for Trump, even amongst the white working class. There is no evidence at all that being worse off made someone more likely to vote for Trump, even when we exclude the confounding effects of race by considering only white people. However, conservative views on Muslims, race, concern for traditional American culture and religion were predictors of a vote for Trump. One saving grace for Lind’s argument might be that agreement with deporting illegal immigrants might be motivated by concern about competition for jobs, but if this view is motivated by economic concerns, we would expect support for deportations to be correlated with downward social mobility or poor financial status. It is actually negatively correlated with the former and mildly positive correlated with the latter (6.2%) while it is 17% correlated with agreement with the view that discrimination against white people is a serious problem.
So even amongst Trump’s “base”, the white working class, support for Trump was not primarily motivated by economic concerns; that is, it may have been class voting but it was not class politics. “Fine,” you might say. “white working class Trump voters are on average more conservative. That doesn’t prove that the Trump movement is not class politics. Obviously the conservatives voted for Trump. What about those who switched to Trump. Maybe they were motivated by his economic policies.” It’s here that I fulfil my promise to look at the marginal Trump voter. What made someone who voted for Obama switch to voting for Trump?
Using the PRRI dataset, I have limited my consideration just to those voters who voted for Obama in 2012. Then amongst these voters, I compare opinions between those to stuck with the Democrats and voted for Clinton, and those who switched to Trump. The first thing to note is that the vast majority of Obama voters voted for Clinton. 74% of Obama voters in this sample also voted for Clinton, with just 8% switching. I will leave it to readers to decide whether this supports or refutes the widespread concern with the Obama-to-Trump demographic. In the graph below, I present the differences which were statistically significant (again defined by whether the difference in the proportion is large enough that it is unlikely to have arisen because of sampling error).
(The questions were as follows: whether LGBTQ people should be allowed to marry; whether the minimum wage should be increased; whether taxes should be raised on the wealthy; whether small business owners should be allowed to discriminate against LGBTQ people; whether discrimination against white people is a significant problem; whether the American way of life is under threat; whether society is too soft and feminine; whether God has granted the US a special role in history; and whether the US in in danger of losing its identity.)
So what can we conclude from this? Recall that if Trumpism is a working-class movement, at the very least we would expect the marginal voter, that is, the voter who voted for Obama and Trump, to be concerned about economic policies which might benefit the working class, or otherwise motivated by economic issues. But these voters were in fact on the whole less concerned with redistributing wealth and less keen on higher minimum wages. They were however significantly more conservative on issues of LGBTQ rights and racial equality, and were more culturally insecure. Note what was not significant: Trump voters were not more economically insecure, nor were they more likely to have suffered psychological distress. Crucially, they were not statistically significantly more likely to think that the economy was biased in favour of the wealthy. This heavily undercuts the argument that Trump’s election had anything to do with economic populism. All in all, those voters who were tempted to switch from Obama to Trump were more socially and economically conservative. There is no evidence that class politics played any role in driving Obama voters into the arms of Trump.
To summarise the evidence presented so far: there is limited evidence that the Trump election exhibited class voting in favour of Trump. The working class overall backed Clinton. The only way that we can support the view that the working class propelled Trump to the White House is if we look only at white Americans. But in this case, race is a much stronger predictor of vote than class. Turning to class politics, there is even less evidence that Trumpism was primarily tapping into the economic concerns of Trump’s working-class voters. Trump’s voters as a whole were not more economically populist than Clinton voters, except on the narrow issue of trade. Neither were (specifically) Trump’s supposed base- working class white voters- more economically populist or economically disadvantaged than their Clinton-voting counterparts. Neither was the marginal Trump voter (who switched from Obama to Trump). However, in each of the three groups considered here (voters as a whole, white working class voters, and white working class voters who voted for Obama) Trump voters were markedly more conservative on social issues than Clinton voters. There is very limited evidence that the Trumpism is a working-class movement. Overall, the evidence is much stronger that Trump, while (perhaps) being anti-establishment, predominantly drew his support from a coalition of white supremacists, religious conservatives, and traditional Republican (economically conservative) voters.
In case anybody thinks I'm ducking the Brexit question, similar arguments can be made about the question of the UK's membership of the EU. Brexit voters were, on average, wealthier than remain voters. Their concerns were disproportionately related to immigration, sovereignty and national identity, while on economics, "Labour Remain voters are no different to Labour Leave voters on the ‘left-right’ scale, similarly Conservative Remain voters are not different to Conservative Leave voters on this scale. In contrast, there are large differences between Labour Leave and Remain voters on the liberal-authoritarian scale and between Conservative Leave and Remain voters on this scale. Labour Remain voters are by far the most liberal group, whilst Leave voters are equally illiberal regardless of whether they voted Conservative or Labour in 2017" according to Surridge. Notably, what should have been a technical issue about trade and administrative institutions became transformed into a proxy for social values. My favourite example of this is how the LGBTQ community lines up on the question. Logically, Brexit has a tangential relationship to questions of sexual freedom. In practice, there is a stark separation, because Brexit became a lightning rod for a particular kind of illiberal populism. Was Brexit to some extent anti-establishment? Certainly. Was it a working-class movement? Not really.
From class war to culture war
If you've followed and agreed with my argument so far, then you accept that the populist right is better understood as a form of anti-establishment identity politics. The right has created a culture war in order to supplant class war. Instead of economic interests ranged against each other, political movements are arranged now with social conservatives on one side and those they deride as "snowflake liberals" on the other: those who are in favour of multi-racial communities, women's and sexual liberation. What does this mean for the left? There are some on the left who would like to have some sympathy with the populist right here. The argument goes like this: working class movements are anti-establishment; the populist right is anti-establishment. But as (I hope) I have demonstrated above, it does not follow that the populist right is a working-class movement. Might the populist right nevertheless be a useful ally in the left's anti-establishment crusade? A kind of Nazi-Soviet pact for the twenty-first century?
I want to argue a few things here: firstly, the populist right is not straightforwardly anti-establishment and may in fact be an extremely unhelpful diversion; secondly, that vacating the socially liberal space and trying to redirect attention to economic issues is not, by itself, a winning response, and thirdly, that a successful left-wing response must involve a more robust defence of liberal values than we have seen so far.
We can start by questioning the claim that the populist right is necessarily anti-establishment. It is true that many “establishment” institutions (the Republican party, broadsheet newspapers, the World Trade Organisation, Hollywood, the UN) tended to be opposed to Trumpism during his rise to power. The reaction from certain sections of the media in particular made it easy for Trump to paint himself as anti-establishment. Witness, for example, the pearl clutching at the mention of a wall on the US’ Southern border, from the same people who continue brush over Obama’s shocking record on deportations. One gets the impression that much of what the establishment finds so unappetising about a rhetorically challenged walking bad hair day they were happy to excuse of an eloquent, Harvard-educated lawyer. A prime example of the scorn middle class journalists have been willing to pour on the President is the amount of space and time that was given to the fact that, in the middle of a government shut-down, Trump had ordered large amounts of fast food to the white house when visited by Clemson college football team. The storm on social media was distinctly snobbish, sneering at food choices which are seen as normal for many Americans. On the international stage, the establishment has made it clear that they regard Trump as a risible character. It’s certainly the case that in his behaviour and in some of the opinions he expresses, Trump has broken taboos and slain scared cows.
Something smells wrong about this story, though. For one thing, if Trump is so anti-establishment, why the closing of ranks within the Republican party with regard to legislation, supreme court nominations and the ill-fated impeachment? Why were so few Republicans so appalled by the actuality of his Presidency that they actually voted against his policies? This applies to the repeal of the affordable care act, where in the event, only three Republican senators actually voted against Trump’s policy, and funding for border security, where the Republican party stuck with their president in the face of a government shutdown in an attempt to secure funding for a border wall. The major legislative achievement of the Trump presidency was a traditional Republican policy: large, unaffordable tax cuts for the wealthy and corporations. If Trump is so offensive to the establishment, why was Mitt Romney the only Republican senator to vote for his impeachment in the senate? As a thought experiment, can one imagine the Democratic party being so indulgent to a hypothetical Sanders Presidency? Unlikely. Indeed, a key argument against Sanders’ proposals during the 2020 primary was that he would struggle to gain Democratic legislative support for them. Perhaps the Republican party is itself primarily an anti-establishment party? This seems highly unconvincing, given the power it has historically wielded in both the legislature and the executive.
If Trumpism is anti-establishment, why has his administration been so stuffed full of oil barons and arms dealers, and Robert Mercer-funded Steve Bannon? Why indeed did he draw funding from Mercer and the Koch brothers? How did Fox news end up making its peace with him? Are these groups really the likely bedfellows of an anti-establishment leader? The most obvious answer is that to explain Trumpism we need to drop the cartoon depiction of the political structure in which there is “an establishment”. This would not come as a particular surprise to sociologists, who have explored since Dahl’s seminal book Who Governs? the idea that power might be fragmented into difference spheres (i.e. the interest groups who dominate the media might not be the same as the interest groups who dominate the financial sector or political institutions) or perhaps that there might be multiple elites (see pp. 51). The fight of the 2016 election can be seen as a fight not between establishment and anti-establishment, but between two different elements of the establishment. This is certainly Mason’s argument. He argues that Clinton gained the support of those within the establishment who were comfortable with neoliberalism when it was tempered by the increase in regulation which was required to maintain public consent after the global financial crisis: publicly-owned companies, cultural elites, and (tacitly) global institutions such as the IMF. Trump, on the other hand, gained the support another section of the economic elite: those who benefit from chaos. This group is typically made up of private companies, who are less concerned about fluctuations in earnings, and less beholden to the exigences of good governance. These are the “disaster capitalists” who thrive from a strong man who is willing to smash through regulatory barriers, bring the full power of the state behind the necessity of creating and opening up new markets, and in particular temporarily suspend the constraints that climate change legislation was placing on their pursuit of short term profit. In Mason’s account, this group had decided that they could not accept the post-GFC settlement and sought to supplant it.
No doubt about it, a section of the elite backed Trump. There is part of this account which feels incomplete, however. Why pick a guy who was openly racist? Why the truck with the woman-baiting? I think the most sensible way to interpret this is to see Trumpism as the creation of a right-wing cultural movement in order to smash working class solidarity and divert anger from economic mismanagement towards vulnerable groups. This is not a new tactic for the right. There is a rich literature showing that ethnic fractionalisation and sparser welfare state provision tend to go together. In Alesina and Glaeser’s account, however, they detail the history of working class movements in the US and how they have tended to be superseded and punctured by rightwing populism. Racial competition is not some organic outcome of having different races coexisting in the same country, but rather is deliberately stoked in order to divert anger away from the unfair economic settlement and drive a wedge in any working class coalition. The ways in which nativism stunted the development of class consciousness is evident in cases such as Sacco and Vanzetti, but there is also a more wholistic sense in which racial segregation and prejudice, both de jure and de facto makes a unified working class movement very much harder. Wallerstein (pp.78) makes a crucial distinction between racism and xenophobia precisely on this basis: while xenophobia is just fear and suspicion of strangers (and therefore might plausibly be a natural human reaction to different ethnic groups) racism is an ideology which justifies an unequal division of the economic pie. Indeed, there is an argument that the invention of racism itself in its modern form was a response of Virginian landowners to concerns that black slaves and indentured white workers would combine to challenge their working conditions. (see Vanessa Williamson here).
In this sense, Trumpism, rather than being straightforwardly anti-establishment, is drawing from the age-old play book of stoking identity politics in order to forestall an authentically leftist response to the global financial crisis and the policy response. There is an ecosystem of right-wing culture war propogandists, Breitbart, Fox news, the Cato institute, the Spectator, often funded by economic elites, which is steadily radicalising white people in both the US and Europe. This is aided by useful idiots and pseudo-intellectuals like Jordan Peterson and Gad Saad, so that traditional social conservatism becomes fused with a reductionist masculinism to create a generally reactionary ideology. The distinction between the decentralised network of individuals supplying the anecdotes that support this politics and the powerful organs which supply the theoretical framework is that between content and narrative. For example, Turning point US targeted all of their facebook ads at people over the age of 24, despite their stated mission statement being to combat the rise in leftism on university campuses. Why? Because the primary aim of these organisations is to embed narratives. Once a section of the public has been propogandised into believing that “cultural Marxists” are taking over their country via universities and the professional classes, new evidence can easily be slotted into place. I believe this is the context in which Trumpism has grown, and Brexit was allowed to morph into a lightning rod for an all-consuming rightist ideology.
In response to this culture war, the left has three main options: deflect to economic issues, triangulate, and fight back. Triangulation is predominantly the response of the “Blue Labour” tribe, represented in the recent labour leadership election to some extent by Lisa Nandy. Relatedly is the “deflect to economics” approach, which was primarily the rhetorical response of the Sanders campaign when asked about issues of race and gender. Both of these are informed by the idea that the left is weak on social issues but can attract strong support on economic issues, and that if the cultural or social issues can be neutralised then the left can win on economic issues. The problem with this approach is that social conservatism, if left unchallenged, is fundamentally corrosive to the left’s project. For the left to succeed it must unify women’s liberation, racial equality, and the workers’ struggle against artificial scarcity into a unified critique of capitalist society as a whole. And for it to even partially succeed, for example, for it to secure better working conditions, it has to convince workers that the main challenge at hand is the oppression of the working class and not, for example, the need to protect white men from sexual assault allegations or protect people from predatory trans people in public toilets. History teaches us that cultural conservatism and the international workers’ movement cannot coexist. In other words, before there can be fertile ground for a leftist political movement, the left must undertake a de-radicalisation campaign. It has to out-argue, out-persuade, and out-mobilise the radical right.
So how does one win a culture war? Well partly, the same way any political conflict is won: by building a coalition of interests on your side, arming them with ideas and policies which motivate them, and by winning arguments. The answers to the first two are fairly clear. The left should be a coalition of marginalised groups, including the working class, answering the call for liberation. The problem is how to win arguments. It’s my view that the left has focused too much on the economic arguments. Programmes of economics education, including excellent content like this video have played some role in shifting the centre of political gravity to the left on economics. But if I’m right, there needs to be a similar project which takes on the festering socially conservative movement, which makes liberal values including challenging racism, women’s liberation, and trans-rights the new “common sense”. This involves building and disseminating a convincing counter-narrative. The challenge is that the right’s narrative- that the politically correct left is stifling freedom of speech, looking down on ordinary people, and unfairly attacking white men- has now bedded in amongst a lot of the electorate. On the hand, this is a trick the left has pulled off many times before, resulting in the civil rights movement, women’s liberation, and the gay rights movement in the twentieth century. All the left has to do is to re-discover its interest in communicating simple stories about equality in these domains.
Conclusion
As the centre left has experienced a series of electoral setback from 2016, the argument that this was due to working class voters deserting their traditional parties because of a lack of solid economic offering was a comforting narrative to the left. If this were true, all that would have been required was for left-wing parties to dust themselves down and develop a radical economic programme, and these voters would return home at subsequent elections. This was more or less the basis of the 2019 Labour party platform, and Sanders’ and Warren’s bids for the US presidency in 2020. In the event, against weak opposition, all three of these campaigns failed dismally.
My argument here is that this is because they completely mis-diagnosed the problem. The collapse of the centre left is not due to economic dissatisfaction, but rather due to the emergence of a violent strain of rightism. This was nurtured in online forums and billionaire-funded propaganda platforms in order to radicalise a section of the electorate, causing them to reject a more equal society like a battered immune system rejects a new organ. Trumpism draws into the existing Republican coalition of low-tax conservatives and religious extremists a new strain of disaffected white masculinity, now armed with a new-found hatred of ethnic minorities, liberated women, and trans-people. Similar forces are at work in the UK, and Brexit was a lightning rod for them. For as long as this poison is allowed to flow freely through the veins of the political system- more or less unchallenged, if not actively encouraged, by the traditional parties of the left- its tendency towards division will throttle at birth any attempt to construct broad-based working class political coalition. For this reason, it cannot be negotiated with, deflected from, or otherwise ignored. Like fascism, it must be smashed, with arguments rooted in reason, compassion, and common sense. This is now a necessary condition for laying the groundwork for any left-wing political movement.