The American people don’t like long wars with uncertain outcomes—and never have. That was true in 1953, when the U.S. accepted a stalemate and armistice with the Chinese-backed North Koreans, and it was true again in 1975, when the U.S. suffered an ignominious defeat and 58,000 dead at the hands of pajama-clad guerrillas and the North Vietnamese army. “Never fight a land war in Asia,” General Douglas MacArthur famously said, and for good reason: in both Korea and Vietnam, the enemy could be endlessly supplied and reinforced.

The solution, in both cases, was to either widen the war or leave. In Korea, MacArthur proposed expanding the war by taking on Chinese military sanctuaries in China (which got him fired), while in Vietnam, Richard Nixon ordered the invasion of Cambodia and mined North Vietnam’s harbors, an expansion of the war that sparked a genocide and merely postponed the inevitable. America’s adventures in Iraq and Afghanistan have been as unsatisfying. A troop surge retrieved America’s position in Iraq, though most military officers now view Baghdad as “a suburb of Tehran” (as a currently serving Army officer phrased it), while the U.S. has spent over $800 billion on a Kabul government whose writ extends to sixty percent of the country—or less.

Given this, it’s not surprising that opinion surveys showed that the majority of the U.S. military supported Donald Trump in the last election; Trump promised a rethink of America’s Iraq and Afghanistan’s adventures, while Clinton was derided as an interventionist, or in Pentagon parlance, “cruise missile liberal.” Trump had the edge over his opponent among both military voters and veterans, especially when it came to ISIS: “I would bomb the shit out of them” he said, a statement translated in the military community as “I would bomb the shit out of them—and get out.” A headline in The Military Times two months before the election said it all: “After 15 years of war, America’s military has about had it with ‘nation building.’”

As it turned out, the military weren’t the only ones who’d “had it with nation building”—so too did Donald Trump. Back in January 2013, two years before he was a candidate for president, Trump made it clear what he would do if he ever occupied the White House. “Let’s get out of Afghanistan,” he tweeted. “Our troops are being killed by the Afghanis we train and we waste billions there. Nonsense! Rebuild the USA.” Three days later, Trump was even more outspoken, explicitly endorsing Barack Obama’s Afghanistan strategy—which amounted to a troops surge, followed by a troop drawdown. “I agree with Pres. Obama on Afghanistan,” he wrote. “We should have a speedy withdrawal. Why should we keep wasting our money – rebuild the U.S.!”

Now, after addressing the American people Monday on his “new strategy in South Asia” (a purposeful trope used to signal his intention to shape a broader, regional policy), Trump appears to have embraced the military’s anti-nation building sentiments, while adopting a policy of “winning,” though without saying exactly how that would happen. The policy— which also includes not saying how many troops “winning” will take, or setting a timetable for victory—includes a pledge of help from America’s allies, and a new focus on Pakistan. Trump was also intent to signal that his new strategy (the war will be left in the hands of warfighters, he announced, and not “micro-managed from Washington”) is much different than the one adopted by his predecessors who, as he all but said, got it wrong.

In fact, though he would almost certainly deny it, what Trump has proposed is a reprise of what Barack Obama did in January of 2009.

Back then, one of Obama’s first decisions on Afghanistan was to assign Bruce Riedel, a 30-year CIA veteran and South Asia expert, to study the conflict and come up with ways to fight it. The following March, on Air Force One, Riedel briefed Obama on his conclusions. Afghanistan would be a big problem for a long time, he said, but the situation in the country was getting worse. The Kabul government was corrupt, its leaders were out-of-touch with the Afghan people and the Taliban and al-Qaeda were gaining strength. But even with that, Riedel added, the real problem wasn’t really Afghanistan, it was Pakistan. “That’s the real challenge,” Riedel said.

Obama agreed with Riedel’s sobering assessment and, on March 27, 2009, he announced his decision to the American people. “The future of Afghanistan is inextricably linked to the future of its neighbor Pakistan,” Obama said in a nationally televised address. “In the nearly eight years since 9/11, al-Qaeda and its extremist allies have moved across the border to the remote areas of the Pakistani frontier.” Put more simply (though Obama did not mention it), the same problem that the U.S. had faced in Korea, and again in Vietnam and Iraq—its failure to destroy the sanctuaries where its enemies could be reinforced and resupplied—it was now facing in Afghanistan. To deal with that problem, Obama appointed super-diplomat Richard Holbrooke to serve as a special envoy to the region (and to work with Centcom commander David Petraeus “to integrate our civilian and military efforts”), launched a drone war against Taliban and al-Qaeda bases in Pakistan, urged Congress to pass a $1.5 billion aid package to Pakistan that would make American strikes more palatable and then, the following May, replaced General David McKiernan, the U.S. commander in Afghanistan, with Stanley McChrystal.

It didn’t work.

In 2012, reporter and author Rajiv Chandrasekaran (whose book Little America: The War Within the War for Afghanistan remains the authoritative source on the Obama plan) concluded that while the Taliban was “pushed out of large stretches of southern Afghanistan,” and the “influx of U.S. resources accelerated the development of the Afghan security forces” the surge did not achieve its objectives. In effect, the Obama administration threw good money after bad: Afghan president Hamid Karzai never bought into the strategy, the Pakistanis failed to “meaningfully pursue” the Taliban and the Afghan army hung back—allowing the U.S. to do the fighting. What the U.S. should have done, Chandrasekaran wrote, was “go long.” Afghanistan is not a sprint, he concluded, but a marathon—and America “got winded too quickly.”

James Mattis and H.R. McMaster have digested these lessons, a senior Pentagon official told me just hours before Trump’s national address, and “have spent the last weeks trying to convince the president that the ‘three yards and a cloud of dust’ approach,” as he termed it, will work. Roughly translated, what that means is that in adopting a more modest increase in American troops, as McMaster and Mattis told Trump, the president would be signaling that while the U.S. was willing to help the Afghan government fight the Taliban, the numbers would not be significant enough to defeat them—that would have to be done by the Afghan Army. In truth, the McMaster-Mattis approach (what one senior Pentagon officer described as “doubling down on a war that is going nowhere”) has some support in the U.S. diplomatic community, and particularly among those civilians who have spent years working in the country.

Among these is David Sedney, a senior associate at the Center for Strategic and International Studies, who is the former acting president of the American University of Afghanistan and served as deputy assistant secretary of defense for Afghanistan, Pakistan and Central Asia. For Sedney, it’s the uncertainty of the American commitment that has been the problem. “We’ve been ambivalent about Afghanistan for the last fifteen years,” he told The American Conservative, “and this has given hope to the Taliban and Pakistan. The message that they’ve taken is that all they need do is wait the U.S. out. Bush focused on Iraq and Obama put in troops caps.” One of the keys, Sedney goes on to say, is that the U.S. “has failed to strengthen the Afghan state in fundamental ways, but the most important is to make a commitment and keep it. That’s the key.”

Sedney also has little use for the views retailed inside the White House by outside experts, like Frontier Services Group president Eric Prince, who advised the administration (in a Wall Street Journal op-ed back in May, and then in a personal meeting with McMaster) to increase the number of contractors in the country, thereby allowing for a drawdown in U.S. troops while also, as Prince argued, saving the U.S. money. While some Pentagon officials speculated as late as last week that secretary Mattis “was not as opposed to the Prince’s ideas as was originally thought,” more recent reports say that the idea “was dead on arrival in the Pentagon, almost from the minute it was mentioned.” Sedney dismisses the idea out of hand, citing his experience with his students in Kabul. “My students don’t want an American proconsul,” he says, “they want an Afghan government that knows how to do the job, and that should be our focus.”

But while Trump has apparently nixed Prince’s contractor idea (and it went unmentioned in his speech), Pentagon officials tell The American Conservative that he has quietly bought into claims that the U.S. can help revive the Afghan economy by exploiting the nation’s mineral resources. While Trump did not mention the program in his speech, and the claim remains debated in the White House, the president (a senior Pentagon civilian told TAC) “is intent to explore ways for this war to pay for itself”—which apparently includes a review of whether Afghanistan’s resources can be exploited sufficiently to put the Afghan government on a sound footing. Will it work?

“This was a good idea back in 2009,” one former Pentagon official says, “but it’s not going to work now.” A geologic survey conducted a decade ago shows that Afghanistan is rich in deposits of gold, silver, and platinum, as well as large quantities of uranium, zinc, bauxite, coal, natural gas and copper—a mother lode of natural resources that could proved Kabul with a badly needed budgetary windfall.

“It’s a pig in a poke,” a former Pentagon official who worked in Afghanistan on identifying the deposits told The American Conservative, “don’t believe a word of it.” The archaic “pig in a poke” phrase, which denotes that a buyer should beware of buying a pig that couldn’t be seen (because it was in a “poke,” or bag), denotes the common belief that while Afghanistan may contain the mineral deposits numerous mining surveys have identified, they remain elusive. Then too, as the former Pentagon official with whom we spoke says, the idea that American companies will realize a windfall on the mineral scheme (to which, as a businessman, Trump is particularly attracted), is simply not in reach.

“American companies no longer do the kind of mining that it would take,” this former Pentagon official says, “security is bad, and commodity prices have collapsed. Why would companies invest in mineral deposits in Afghanistan when they won’t make the same investments in Australia.” Which is to simply say that the Afghanistan problem is now, under Trump, what it was under George W. Bush and Barack Obama—an intransigent challenge whose resolution is dependent on fighting and winning a war against an enemy who can fight, retreat, resupply and reinforce and fight again. The key to that victory is now what it has always been: Pakistan. Trump, and McMaster and Mattis, realize this of course, which is why tonight the president focused on providing a strategy for “South Asia”—a phrase the defense secretary, in particular, has used over the last weeks.

“I have hope for Afghanistan,” CSIS’s Sedney says. “The Afghan military is fighting better than ever before. When I went to Kabul in 2002, Kabul looked like Dresden, but now it’s a vibrant city. Yes, the Taliban can kill people, but most Afghanis are moving ahead with their lives in spite of this. The problem is that, as we’ve seen over the last decade, a small minority can keep the country destabilized. That’s what we have to stop. We have to come up with a way of stopping that.”

In the wake of Trump’s address, credit for its opening paean was given to new White House chief of staff John Kelly, the retired Marine Corps general who, TAC was told, insisted that Trump use the speech to walk back the controversy of his remarks on Charlottesville—a suggestion that both McMaster and Mattis readily agreed to when Trump’s national security team met on Friday at Camp David. In the end, however, it was McMaster and Mattis who had the greatest influence on Trump’s thinking. “There was all this speculation that maybe, just maybe, the president would somehow come around to getting out,” the senior Pentagon civilian with whom we spoke said, “but that was never going to happen. Jim Mattis wouldn’t let it happen. You can see his fingerprints all over this.”

Another Pentagon observer had a much different take. “This is Joe Biden’s plan, all the way,” he said, referring to the then-Vice President’s recommendation to Obama back in 2009. “Biden said that we should increase counterterrorism operations, draw down U.S. forces in the provinces, increase pressure on Pakistan and make a deal with India. Obama said ‘no’ to the idea, but you can bet Mattis was listening. This is his plan all the way.”

Almost everyone at the Pentagon agrees, though key senior military officers who have been privy to James Mattis’s thinking over the last weeks (but who remain unconvinced by it) provide a cautionary, and nearly fatalistic, note. “This Trump plan, at least so far as I understand it, sounds a lot like the kind of plan we’ve come up with again and again since the end of World War Two,” a senior Pentagon officer says. “We’re going to surge troops, reform the government we support and put pressure on our allies. In this building [the Pentagon] there’s a hell of a lot of skepticism. And that’s because we all know what this new strategy really means – and what it means that the only way we can get out of Afghanistan is to get further in. You know, it seems to me that if there’s one thing we’ve learned, it’s that that doesn’t work.”