There is no real equivalence between Donald Trump's gleeful and repeated encouragement of his supporters chanting "lock her up" in reference to Hillary Clinton's use of a private email server and recent statements about Trump by House Speaker Nancy Pelosi (“I don’t want to see him impeached. I want to see him in prison") and by Senator Kamala Harris (in a Harris administration, DOJ "would have no choice" but to prosecute Trump for obstruction). Pelosi made her remark in private and as a way to tamp down impeachment, while Harris made her statement in response to a question. Moreover, as the Mueller report documents, there is substantial evidence that Trump committed crimes that would subject him to indictment and conviction were it not for the DOJ policy against indicting a sitting president. By contrast, Clinton's email server use was reckless but not the sort of act that typically warrants prosecution.
Nonetheless, in a Lawfare column on Wednesday, Ben Wittes argued that Democrats should stop talking about prosecuting Trump, because (quoting Paul Rosenzweig) "you don’t protect norms by violating norms." Although I disagree with Wittes's implication (which may not have been intended) of equivalence, I agree with his advice to Democrats to stop talking about prosecuting Trump, because doing so very easily lends itself to the appearance that they share Trump's willingness to convert the US into the sort of country in which new governments routinely prosecute their predecessors for real and imagined crimes--thus undercutting the incentive of leaders who are defeated at the polls to relinquish power. It's true that the Trump-supporting critics of the Pelosi and Harris statements are shameless hypocrites, but any advice for US politicians operating in the real world must take account of the fact that shameless hypocrites litter the media landscape.
I also agree with Wittes on another point: that the NPR reporter's question that prompted Senator Harris to say that the DOJ would need to prosecute Trump contained embedded within it an implicit question about whether the next president ought to pardon Trump. Here I want to channel Alexander Hamilton, whose analysis in Federalist 74 may provide some useful context.
Among the reasons Wittes gives for why Democrats should not currently commit to (or against) prosecuting or pardoning Trump is the fact that the prosecution decision might look different in January 2021 from how it appears today. He highlights a crucial factor:
The matter will look very different if [Trump] loses in a landslide and leaves office graciously (for him) than it will if he loses narrowly, refuses to accept the results of the election, and resists leaving office. One important prudential factor in the question of his prosecution, after all, might reasonably be the extent to which criminal process against the former president would serve to protect against an ongoing threat to democracy—rather than, say, merely to serve as retroactive accountability for past conduct.I read Wittes to be saying that prosecution would be more appropriate if Trump refuses to accept the results of the election than if he follows the precedent of every prior losing president (with the appropriate caveats about the elections of 1800, 1876, and 2000) by conceding graciously. Assuming the issue arises after Trump has left office (even if only after a struggle), that may be right. However, we can imagine a scenario in which Trump's refusal to accept a loss at the polls cuts the other way: a Democratic President-elect might want to promise a hitherto-recalcitrant Trump a pardon in exchange for Trump's agreement to encourage his supporters to accept the result.
Consider Hamilton's observation. In Federalist 74, he responds to an objection to the Constitution's vesting of the pardon power in a single individual. He focuses, as did the critics to whom he responds, on pardons for treason:
when the sedition had proceeded from causes which had inflamed the resentments of the major party, they might often be found obstinate and inexorable, when policy demanded a conduct of forbearance and clemency. But the principal argument for reposing the power of pardoning in this case to the [President] is this: in seasons of insurrection or rebellion, there are often critical moments, when a welltimed offer of pardon to the insurgents or rebels may restore the tranquillity of the commonwealth; and which, if suffered to pass unimproved, it may never be possible afterwards to recall. The dilatory process of convening the legislature, or one of its branches, for the purpose of obtaining its sanction to the measure, would frequently be the occasion of letting slip the golden opportunity.Unfortunately, it is possible to envision a scenario in which Hamilton's proposal would be literally applicable. Suppose that the 2020 presidential election is close enough for Trump and his allies to claim that some combination of imagined voter fraud and other irregularities in one or more swing states mean that he, rather than the Democratic candidate, is the rightful victor. Suppose further that the courts reject his challenges but not without dissents from one or more judges and justices. Trump supporters, spurred on by the right-wing media and Trump himself, resist, initially peacefully but as time goes by with increasing violence. Prior to Inauguration Day 2021, the Democratic President-elect could promise pardons to Trump and perhaps to some of his supporters as well, but such a promise is not legally enforceable, and thus it is easy to imagine that it would not satisfy Trump. To ensure domestic tranquility, immediately following his or her inauguration, the new President could strike a deal with Trump for one or more pardons in exchange for his calling off the resistance.
Would that be corrupt? Sure, but if Hamilton is our guide, this is exactly the sort of corruption that the pardon power was intended to embody. We are accustomed to thinking of pardons as principally serving a mercy function to mitigate harsh justice. But as the experience in countries transitioning to democracy reveals, some sort of amnesty for leaders and members of the ancien regime can also serve to promote democracy.
Would the use of the pardon power in this way undercut accountability? Absolutely. Promising Trump a pardon in exchange for calling off the insurrection would be the lesser of two evils and would, in a sense, be acceding to a kind of political extortion. But it would be better than civil war.
Finally, I note that I was reluctant even to broach this subject, for fear of giving people ideas that then create two possible incentives. First, if Trump really does fear prosecution and thinks that his best shot at a pardon is to "buy" it using the threat or reality of insurrection by his supporters, he could be led to encourage such insurrection. Second, if Trump thinks that a future Democratic president will pardon him if push comes to shove, he will feel free to break the law.
Those are real risks, but I judge them to pose less total danger than: (a) the risk of civil war; and (b) the risk that loose talk from Democrats about prosecuting Trump is already incentivizing Trump to refuse to acknowledge defeat in 2020 should it come to that. As for the worry that the prospect of a pardon will lead Trump to think he can break the law with impunity: (a) he could not count on a pardon from the next president until he actually had it in hand; and (b) he already seems to think he can break the law with impunity.