You are weird. I am very weird. And the worst part is, we don’t really recognize how weird we are.
That’s the basic argument of Donald Kinder and Nathan Kalmoe’s Neither Liberal nor Conservative: Ideological Innocence in the American Public. Their study begins with a famous paper by political scientist Philip Converse titled “The Nature of Belief Systems in Mass Publics.” The nature of those belief systems, Converse concluded, was that they really weren’t systems at all. The overwhelming majority of Americans were free of anything that resembled coherent liberal or conservative ideologies — indeed, only “about 17 percent of the public could both assign the terms ‘liberal’ and ‘conservative’ correctly to the parties and say something sensible about what the terms meant.”
Which isn’t to say that voters didn’t have opinions, much less party and group loyalties. They did, and they do. But the internally coherent (or at least semi-coherent) ideological frameworks that drive the activities of politicians, pundits, and other political actors are foreign to most voters.
Converse’s basic findings have been replicated in a number of different studies done over the past 50 years, and Kinder and Kalmoe extend on them here. In a telling bit of research, they scoured massive election surveys to see what bearing self-reported ideology had on policy opinions on issues ranging from LGBTQ rights to health care to foreign aid to Social Security. The answer, across years ranging from 1992 to 2009, was basically none — “ideological differences,” they reported, “have little influence over opinion on immigration, affirmative action, capital punishment, gun control, Social Security, health insurance, the deficit, foreign aid, tax reform, and the war on terrorism.”
There were two glaring exceptions: LGBTQ rights and abortion. But the exceptions were so stark that Kinder and Kalmoe wondered if they were missing something, and they had a theory of what it might be: religion. So they ran the data again, “adding measures of faith, religiosity (the degree to which Americans take their faith seriously), and group sentiments to the model.” Once they did that, the effect of ideology all but disappeared.
So this, then, is the bottom line: Most voters aren’t ideologues, and even accounting for that, most ideologues aren’t particularly ideological.
This can seem like a strange finding if you live in Washington, DC, or you’re a political obsessive who surrounds yourself with other political obsessives. Once you’re used to seeing politics through an ideological lens, it becomes easy to forget that others don’t. And that helps illuminate an important question in American politics: how Trump, with all his ideological heterodoxies, took over the Republican Party, and why so few political professionals saw it coming.
Party > ideology
One consistent finding in Kinder and Kalmoe’s research is that party identification bests ideological identification. Most people are a Republican or a Democrat before they are a conservative or a liberal. And most people will stick with their party long after they’ve abandoned their ideology.
This table tells the story. It measures the persistence of both ideological and party affiliation across different time periods, on different surveys, and among voters of differing information levels — so it’s asking, in other words, how often people switch parties, or switch how they describe their political philosophy, over time. For most voters, party identification is far more stable than ideological identification. Only among the very best-informed voters are the two even close:
The instability of voters’ ideological identities is really striking here. If you asked an average voter in 2000 whether they were liberal, moderate, conservative, or none of the above, their answer was only 63 percent predictive of what they’d tell you two years later. For voters with very little political knowledge, ideological identity is so fragile it’s probably not even worth calling it an identity. If you are a diehard liberal or conservative who hasn’t changed your views in 20 years, look at this table and reflect on just how unusual you are.
These findings pose a profound challenge to traditional models of politics. In theory, ideology comes first and party comes second. We decide whether we’re for single-payer health care, or same-sex marriage, or abortion restriction, and then we choose the party that most closely fits our ideas. You’re a liberal and so you become a Democrat; you’re a conservative and so you become a Republican.
The truth, it seems, is closer to the reverse: We choose our party for a variety of reasons — chief among them being the preferences of our family members, core groups, and community — and then we sign on to their platforms. In this telling, write Kinder and Kalmoe, “ideological identification is primarily an effect, not a cause, of a person’s political views.”
This theory makes a prediction: If party identification is stronger than ideological identification, then as parties change their ideological identities, their loyalists will change with them, rather than abandoning them. And that’s a lot closer to what we see. The exception is high-information voters, who keep their party identification and ideological identification linked.
“One enduring lesson to carry forward is an appreciation for the deep divide between elites and publics,” write Kinder and Kalmoe. For elites, politics is driven by ideology, and that seems like the most natural thing in the world. But it’s not, and it’s hard for highly ideological actors to appreciate just how weird they really are.
Which brings us to Trump.
How a confused ex-Democrat took over the Republican Party
Donald Trump is an ex-Democrat who flirted with running on the Reform Party line in 2000 and only became a Republican recently. Back in 2008, Trump said Hillary Clinton would “make a good president or good vice president,” and refused to answer whether he would vote for her or for a Republican if she won the nomination. "I know her very well,” Trump said. “She's very talented. And she has a husband that I also like very much.”
As befits someone whose conversion to the Republican Party was relatively recent, Trump speaks conservatism with an accent, when he speaks it at all. In the past, Trump had praised single-payer health care systems and declared himself “very pro-choice.” During the 2016 election, he broke with conservative dogma and swore to protect Social Security, Medicare, and Medicaid from cuts; to make sure everyone had health insurance paid for by the government; to raise taxes on rich guys like himself — and, for good measure, he praised Planned Parenthood and slammed the Iraq War.
Real conservatives noticed. In their “Against Trump” editorial, National Review, as close to a guardian of conservative ideology as exists in American life, warned that “Trump is a philosophically unmoored political opportunist who would trash the broad conservative ideological consensus within the GOP.”
Trump’s ideological heterodoxies were a key reason pundits assumed he would eventually be wiped out in the Republican primaries. Many believed Republicanism was conservatism, and so a non-conservative could never win over Republican voters. But party trumps ideology. Republicanism is Republicanism, and for most voters, it is based more on group attachments and resentments than it is on ideology. These were the voters Trump understood and political elites didn’t, and he understood them because he is one of them: His group allegiances were tribal even as his ideology was flexible.
Trump was far better than Marco Rubio or Jeb Bush or Ted Cruz at expressing his distaste for Democrats, for immigrants, for Black Lives Matter protesters, for condescending cosmopolitans, for President Obama. That Rubio and Bush and Cruz were better at expressing their fealty to conservative ideology didn’t much matter. Henry Adams once wrote that “politics, as a practice, whatever its professions, had always been the systematic organization of hatreds,” and Trump was masterful at organizing those hatreds.
Trump was easy for political professionals to underestimate because they are ideologues who are surrounded by ideologues, and so they naturally came to see a coherent ideology as a prerequisite for a successful politician. And there was, and is, truth to that: Most politicians really are highly ideological, and they use their power over the party’s machinery to beat back or convert those who would seek to lead their party without joining in their ideological crusade.
But Trump, because of his celebrity, his money, and his media savvy, was able to campaign without party support. And that let him show that you don’t need to be a consistent conservative to appeal to Republican voters, because most of them aren’t consistent conservatives either — there’s much more to politics than ideology, even if political professionals likes to pretend otherwise.