It would be unfair to call Donald Trump’s interaction with black voters a love-hate relationship, since there’s little evidence of African American enthusiasm for Trump. But the Republican campaign has pursued a Janus-like strategy on black voters—ostensibly courting them in public while privately seeking to depress turnout.

This tension is on display in the last 24 hours. On Wednesday, Trump delivered a speech in Charlotte, North Carolina, advertised as an “urban renewal agenda for America’s inner cities.” Trump told the audience, “It is my highest and greatest hope that the Republican Party can be the home in the future and forevermore for African Americans and the African American vote because I will produce, and I will get others to produce, and we know for a fact it doesn’t work with the Democrats and it certainly doesn’t work with Hillary.”

Yet on Thursday, BusinessWeek published a big cover story, based on exclusive access to the campaign, that revealed that Trump’s team has decided that winning over black voters is a lost cause:

Instead of expanding the electorate, [campaign chairman Steve] Bannon and his team are trying to shrink it. “We have three major voter suppression operations under way,” says a senior official. They’re aimed at three groups Clinton needs to win overwhelmingly: idealistic white liberals, young women, and African Americans.

The reporters, Joshua Green and Sasha Issenberg, offer some more detail on what that looks like:

On Oct. 24, Trump’s team began placing spots on select African American radio stations. In San Antonio, a young staffer showed off a South Park-style animation he’d created of Clinton delivering the “super predator” line (using audio from her original 1996 sound bite), as cartoon text popped up around her: “Hillary Thinks African Americans are Super Predators.” The animation will be delivered to certain African American voters through Facebook “dark posts”—nonpublic posts whose viewership the campaign controls so that, as [campaign digital guru Brad] Parscale puts it, “only the people we want to see it, see it.” The aim is to depress Clinton’s vote total. “We know because we’ve modeled this,” says the official. “It will dramatically affect her ability to turn these people out.”

This wasn’t entirely unknown—Monica Langley reported two weeks ago that Trump was aiming for depressed turnout. What’s incredible is that Trump’s advisers called it “voter suppression.” When you’re talking about “suppressing” black votes, it’s a good sign you’re not competing for them, and this is messaging malpractice, since it makes the work seem nefarious. That’s all the more true because Republicans around the country have spent the last decade instituting laws that make it more challenging to vote—measures that they say are necessary to prevent election fraud, but critics say actually amount to voter suppression.

In fact, trying to depress turnout is not that unusual. There’s lots of evidence that negative advertising is designed to depress turnout among certain, targeted groups. Certainly, the Clinton campaign has used negative ads—they’ve fired off a broadside of spots using Trump’s own words. If these ads convince Trump supporters to vote for Clinton, that’s great for her, but if they convince voters who might otherwise vote for the Republican that he’s just too toxic or mean or extreme, that’s just fine for her, too. It’s still one less vote for Trump. Obama tried a similar tactic in assailing Mitt Romney four years ago, as Ross Douthat points out.

This is, at least, the theory. Whether it works is a different question. The answer is probably no. A 2007 meta-analysis concluded that there is not “any reliable evidence that negative campaigning depresses voter turnout, though it does slightly lower feelings of political efficacy, trust in government, and possibly overall public mood.” (In a charming proof that your mother was right that you shouldn’t say anything unless you have something nice to say, another study found that while negative advertising was mostly useless, positive advertising was effective in running up margins where support was already strong.)

Even if Trump’s advisers are acting on a questionable theory, that theory does help explain some of Trump’s strange approach to African Americans. It confirms the suspicion of many observers, myself included, that Trump is more going through the motions of courting black voters more than actually trying to woo them. Many of his events aimed at African Americans have actually been in heavily white jurisdictions, in front of heavily white crowds.

His appearance in Charlotte Wednesday was no different. Although the city has become a symbol of racial tension since the shooting of Keith Lamont Scott by police in September, the audience at Trump’s invitation-only event was mostly white, The Charlotte Observer noted. His speech was full of many of the same hamfisted overtures that have been met with everything from eyerolls to outrage by actual black Americans, from the implication that most blacks are living in squalid, violent “inner cities” to his deployment of false or misleading statistics to butress that vision of squalor. He has also used outdated and distancing language to discuss African Americans and other minorities.

The result of this halfhearted—or rather, insincere—outreach has been that Trump’s polling among African Americans is bad. He once boasted he could win 25 percent of the black vote, far outpacing the recent Republican high-water mark (Gerald Ford’s 17 percent in 1976). But with the election near, Fox News finds him trailing Clinton by 77 points. CBS found Trump at just 4 percent, with Clinton at 85 percent. That puts her a good bit behind Barack Obama’s 2012 pace of 93 percent, but Trump is also still behind Mitt Romney’s 6 percent, according to exit polls. (In one farcical turn, a major tracking poll turned out to be wildly distorted by a single 19-year-old black Trump voter, whose standing was heavily weighted.)

The Trump theory, as laid out in the BusinessWeek article, holds that this doesn’t really matter. He doesn’t need black votes! (His advisors admit in the piece that their polling finds them trailing, and they say their path to victory is real but narrow.) Of course, his aides might be wrong. In addition to the political-science literature casting doubts on turnout depression, Clinton has spent months trying to mobilize African Americans, especially in places like North Carolina or Ohio, where black turnout is likely the difference between a Democratic victory or a Republican win. If black voters vote in droves in Cuyahoga County, home to Cleveland, it might block Trump’s path to the White House altogether, and Trump’s decision to not even contest the bloc would look like a mistake.

Meanwhile, the BusinessWeek story made Republicans who aren’t affiliated with Trump practically apoplectic. That’s because even if Trump doesn’t think he needs black voters, future Republican candidates will. Trump has made clear that he owes no particular allegiance to the Republican Party and has little interest in its fortunes without him, which allows him to be blithe about writing off the demographic. Other analysts, including Republicans, have been warning for years that the GOP cannot survive as a rump party of whites (and, increasingly, white men). But voter relationships have to be built over time; a bloc written off or alienated can take a generation or more to win back. By not just passing on the chance to reach out to African Americans but actually bragging about their efforts to keep them from the polls, the Trump campaign isn’t just wasting an opportunity for outreach to blacks, but may in fact be setting back Republican efforts for years to come.