Suppose that you believe, as I do, that the threat to American democracy posed by Donald Trump and his Republican enablers and imitators is by far the most important issue confronting us today. One implication of this view is that opinion leaders – politicians, pundits, academics, journalists – should make every effort to forestall disaster by educating people about the risk of democratic backsliding, and trying to bring people together to preserve free and fair elections.
Ross Douthat appears to disagree with this assessment. He argues that those of us who are panicked about the state of democracy overstate the risk of backsliding. In particular, he thinks that we* exaggerate the risk that Trump will steal the 2024 election. He points out that Trump tried to subvert the 2020 election but failed due to his own incompetence and the resistance of many government officials, including his own vice president. He argues (plausibly) that Trump would be less able to steal an election when he is not the sitting President. He concludes that the threat of democratic backsliding is, in the end, not categorically different than other problems or dangers:
. . . My thesis is that Trump is an adventurer of few consistent principles rather than a Hitler, that we’ve seen enough from watching him in power to understand his weaknesses and incapacities, and that his threat to constitutional norms is one of many percolating dangers in the United States today, not a singular danger that should organize all other political choices and suspend all other disagreements.
Note that the final clause suggests that Never Trump Republicans are wrong: if you are (say) a social conservative, or a libertarian who wants lower taxes and opposes climate legislation, you can go ahead and vote for Trump and other Republicans despite the threat that they pose to democracy. Douthat did not support Trump, but he thinks we should be open to that possibility: he appears to be encouraging people to be “Maybe Trumpers”.
What should we make of this?
One problem is that Douthat claims that those of who believe that democratic failure is a threat of singular importance are overestimating the likelihood of a democratic failure. Even if this is true – and I will raise doubts about it below – any threat or risk is characterized by a frequency and a severity. If a democratic failure in the United States would be catastrophic, then the fact that the chance of this happening may be lower than some people think would not show that democratic failure is just another risk, to be traded off against the pursuit of other goals in the ordinary manner. I believe that a failure of democracy in the United States could indeed be catastrophic, but I will not argue for this point here. My point is simply that Douthat ignores the possibility that democratic failure is a distinctly awful outcome.**
But let us put this aside and focus simply on the likelihood that Trump will substantially undermine democratic governance. Douthat understates the probability of democratic backsliding in several ways. First, he focuses on the risk that Trump steals the 2024 election. But there is also a risk that Trump or some other Trumpy Republican wins outright in 2024. This would pose a serious risk to the continuation of democratic government, and it is another reason for people like Douthat to shout from the rooftops that keeping Republicans from gaining control of the government must be our over-riding priority.
Second, by taking an ex-post view of the 2020 election, Douthat overlooks or downplays factors that could have been different in 2020, and that might be different in 2024. This makes the failure of Trump’s effort to steal the election seem more likely than it was in 2020. Notably, in 2020, none of the states where votes were “in dispute” was large enough to tip the election from Biden to Trump. This meant that partisan state legislators and election officials and judges had less incentive to violate the letter and spirit of election laws and procedures. Similarly, to throw the 2020 election to Trump, the Supreme Court would have had to incinerate its reputation with a whole series of slanted decisions in several states all favoring Trump. If the electoral college winner could be shifted by changing the outcome in one or two states, either because the vote was close or because the vote in a large state was in dispute, the incentives to undermine election results would be much stronger.
Finally, the fact that large numbers of Republican voters and politicians are prepared to support election subversion was not nearly as clear in 2020 as it is today. The result is that one important barrier to electoral subversion has been removed, and Republican legislators and interest groups are actively working to undermine non-partisan election administration.
Douthat is a smart and reasonable person. I sincerely hope that he reconsiders his approach to this issue. The critical question is what can we do to minimize the risk to our democracy? not are we over- or under-estimating the exact magnitude of the risk?
[*Douthat is explicitly responding to Robert Kagan’s recent essay in the Washington Post, and to Barton Gellman’s pre-election piece in the Atlantic, which I fretted about here. I discussed a similar argument by Douthat here.]
[**Of course, there are worse outcomes than democratic failure, such as nuclear holocaust, but Douthat isn’t arguing that we should support Trump to reduce the risk of nuclear annihilation.]