To understand why the current conservative crackup so confounds and confuses the Republican establishment, you have to recognize that the party is facing two separate revolts taking place simultaneously: one led by Ted Cruz, the other by Donald Trump. The first is well described by E. J. Dionne in his important new book, “Why the Right Went Wrong.” For six decades, he explains, conservatives have promised their voters that they were going to roll back big government. In the 1950s and early 1960s, they ran against the New Deal (Social Security). Then they railed against the Great Society (Medicare). Today it is Obamacare.

But they never actually did anything. Despite nominating Barry Goldwater, electing Richard Nixon, Ronald Reagan and two George Bushes, despite a congressional revolution led by Newt Gingrich, these programs endured and new ones were created.

The simple reason for this, of course, is that while Americans might oppose it in theory, in practice they like the welfare state. And the bulk of government spending is on the middle class, not the poor. Social Security and Medicare take up more than twice as much of the federal budget as all nondefense discretionary spending put together. One middle-class tax exemption for employer-based health care costs the federal government more than three times the total for the food stamps program.

Whatever the reality, Republicans kept promising something to their base but never delivered on it. This has led to what Dionne calls the “great betrayal.” Party activists are enraged that they have been hoodwinked, and they view the Washington establishment as a bunch of corrupt compromisers. They want someone who will finally deliver on the promise of repeal and rollback.

Enter Cruz. How did a one-term senator, despised within his party both in Washington and Texas, get so far so fast? By promising to take on those party elites and finally throttle big government. Cruz declares that he will repeal Obamacare, abolish the Internal Revenue Service and replace the income tax with a 10 percent flat rate and enact a constitutional amendment to balance the budget which would mean hundreds of billions of dollars of spending cuts.

Trump’s supporters, on the other hand, are old-fashioned economic liberals. In a powerful analysis, drawing on the most recent and thorough survey data from the Rand Corp., Michael Tesler shows that the Trump voter is very different from the Cruz voter. “Cruz outperforms Trump by about 15 percentage points among the most economically conservative Republicans,” he writes. “But Cruz loses to Trump by over 30 points among the quarter of Republicans who hold progressive positions on health care, taxes, the minimum wage and unions.” Trump is well aware of this fact, which explains why he has repeatedly promised not to touch Social Security and Medicare, spoken fondly of the Canadian single-payer system, denounces high CEO salaries, promises to build infrastructure, and is against free-trade deals.

Trump’s voters reflect an entirely different revolt. Ever since the 1960s, a part of America’s white middle and working class has felt uncomfortable with the changes afoot in the country. They were uneasy with the social revolutions of the 1960s, dismayed by black protests and urban violence, and enraged by the increasing tide of immigrants, many of them Hispanic. In recent years, they have expressed hostility toward Muslims. It is this group of Americans, many of them registered Democrats and independents, who make up the core of support for Donald Trump. (Obviously there are overlaps between the two candidates’ supporters, but the divergences are striking.)

In his analysis, Tesler shows that, statistically, “Trump performs best among Americans who express more resentment toward African-Americans and immigrants and who tend to evaluate whites more favorably than minority groups.”

The New York Times’ Nate Cohn points out that Trump’s support geographically is almost the opposite of that of the last major populist businessman to run for president, Ross Perot. Perot did well in the West and New England, but poorly in the South and industrial North. Trump’s support follows a different but familiar pattern. Cohn writes: “It is similar to a map of the tendency toward racism by region.” To be clear, there are many people who back Trump for reasons entirely unrelated to race, religion or ethnicity, but the correlations shown by scholars are striking.

Could these revolts have been prevented? Perhaps, if the Republican Party had been honest with its voters and explained that the welfare state was here to stay, that free markets need government regulation, and that the empowerment of minorities and women was inevitable and beneficial. Its role was to manage these changes so that they develop organically, are not excessive, and preserve America’s enduring values. But that is the role for a party that is genuinely conservative, rather than radical.