Winning! It’s the White House watchword when it comes to the US armed forces. “We will give our military the tools you need to prevent war and, if required, to fight war and only do one thing—you know what that is? Win! Win!” President Donald Trump exclaimed earlier this year while standing aboard the new aircraft carrier USS Gerald R. Ford.

Since World War II, however, neither preventing nor winning wars have been among America’s strong suits. The nation has instead been embroiled in serial conflicts and interventions in which victories have been remarkably scarce, a trend that has only accelerated in the post-9/11 era. From Afghanistan to Iraq, Somalia to the Philippines, Libya to Yemen, military investments—in lives and tax dollars—have been costly and enduring victories essentially nonexistent.

But Amadou Sanogo is something of a rare all-American military success story, even if he isn’t American and his success was fleeting. Sanogo learned English in Texas, received instruction from US Marines in Virginia, took his intelligence training in Arizona, and underwent Army infantry officer basic training in Georgia. Back home in his native Mali, the young army officer was reportedly much admired for his sojourn, studies, and training in the United States.

In March 2012, Sanogo put his popularity and skills to use when he led a coup that overthrew Mali’s elected government. “America is [a] great country with a fantastic army. I tried to put all the things I learned there into practice here,” he told Der Spiegel during his tenure as Mali’s military strongman. (He eventually lost his grip on power, was arrested, and in 2016 went on trial for “complicity in kidnapping and assassination.”)

Since 9/11, the United States has spent more than $250 billion training foreign military and police personnel like Sanogo. Year after year, a sprawling network of US programs provides 200,000 of these soldiers and security officers with assistance and support. In 2015, almost 80,000 of them, hailing from 154 countries, received what’s formally known as Foreign Military Training (FMT).

The stated goals of two key FMT programs—International Military Education and Training (IMET) and the Combating Terrorism Fellowship Program (CTFP)—include promoting “international peace and security” and increasing the awareness among foreign military personnel of “internationally recognized human rights.” In reality, these programs focus on strengthening US partner and proxy forces globally, though there’s scant evidence that they actually succeed in that goal. A study published in July, analyzing data from 1970 to 2009, finds that FMT programs are, however, effective at imparting skills integral to at least one specific type of armed undertaking. “We find a robust relationship between U.S. training of foreign militaries and military-backed coup attempts,” wrote Jonathan Caverley of the US Naval War College and Jesse Savage of Trinity College Dublin in the Journal of Peace Research.

Bad Actors

Through nearly 200 separate programs, the State Department and the Department of Defense (DoD) engage in what’s called “security cooperation,” “building partner capacity,” and other assistance to foreign forces. In 2001, the DoD administered about 17 percent of security-assistance funding. By 2015, that figure had jumped to approximately 60 percent. The Combating Terrorism Fellowship Program, a post-9/11 creation indicative of this growth, is mostly run through the DoD and focuses on training mid- and senior-level defense officials from allied militaries in the tenets of counterterrorism. The State Department, by contrast, is the driving force behind the older and larger IMET program, though the Defense Department implements the training.

Under IMET, foreign personnel—like Sanogo—travel to the United States to take classes and undergo instruction at military schools and bases. “IMET is designed to help foreign militaries bolster their relationships with the United States, learn about US military equipment, improve military professionalism, and instill democratic values in their members,” wrote Joshua Kurlantzick in a 2016 Council on Foreign Relations memorandum aimed at reforming the program.

However, in an investigation published earlier this year, Lauren Chadwick of the Center for Public Integrity found that, according to official US government documents, at least 17 high-ranking foreigners—including five generals—trained through IMET between 1985 and 2010 were later accused and in some cases convicted of criminal and human rights abuses. An open-source study by the nonprofit Center for International Policy found another 33 US-trained foreign military officers who later committed human rights abuses. And experts suggest that the total number of criminal US trainees is likely to be far higher, since IMET is the only one of a sprawling collection of security-assistance programs that requires official reports on human rights abusers.

In their Journal of Peace Research study, Caverley and Savage kept the spotlight on IMET because the program “explicitly focuses on promoting norms of civilian control” of the military. Indeed, it’s a truism of US military assistance programs that they instill democratic values and respect for international norms. Yet the list of US-trained coup-makers—from Isaac Zida of Burkina Faso, Haiti’s Philippe Biamby, and Yahya Jammeh of The Gambia to Egypt’s Abdel-Fattah el-Sisi, Mohammad Zia-ul-Haq of Pakistan, and the IMET-educated leaders of the 2009 coup in Honduras, not to mention Mali’s Amadou Sanogo—suggests an embrace of something other than democratic values and good governance. “We didn’t spend, probably, the requisite time focusing on values, ethics, and military ethos,” then chief of US Africa Command, Carter Ham, said of Sanogo following his coup. “I believe that we focused exclusively on tactical and technical [training].”

Caverly warns that Washington should also be cautious about exporting its own foreign and domestic policy imperatives, given that recent administrations have left the Defense Department flush with funding and the State Department’s coffers so bare that generals are forced to beg on its behalf. “Put more succinctly,” he explained, “you need to build up multiple groups within civil society to complement and sometimes counterbalance an empowered military.”

Caverley and Savage identified 275 military-backed coups that occurred worldwide between 1970 and 2009. In 165 of them, members of that country’s armed forces had received some IMET or CTFP training the year before the coup. If you add up all the years of such instruction for all those countries, it tops out at 3,274 “country years.” In 165 instances, a takeover attempt was carried out the next year. “That’s 5 percent, which is very high, since coups happen rarely,” Caverley told TomDispatch. “The ratio for country-years with no US training is 110 out of 4101, or 2.7 percent.”

While US training didn’t carry the day in The Gambia in 2014 (as it had in 1994 when US military-police-training alumnus Yahya Jammeh seized power), it is nonetheless linked with victorious juntas. “Successful coups are strongly associated with IMET training and spending,” Caverley and Savage noted. According to their findings, American trainees succeeded in overthrowing their governments in 72 of the 165 coup attempts.

Train Wreck

There is significant evidence that the sprawling patchwork of America’s military training programs for foreign forces is hopelessly broken. In 2013, a State Department advisory board found that American security aid had no coherent means of evaluation and no cohesive strategy. It compared the “baffling” array of programs to “a philanthropic grant-making process by an assemblage of different foundations with different agendas.”

A 2014 RAND analysis of US security cooperation (SC) found “no statistically significant correlation between SC and change in countries’ fragility in Africa or the Middle East.” A 2015 report from US Special Operations Command’s Joint Special Operations University noted that efforts at building partner capacity have “in the past consumed vast resources for little return.” That same year, an analysis by the Congressional Research Service concluded that “despite the increasing emphasis on, and centrality of, [building partner capacity] in national security strategy and military operations, the assumption that building foreign security forces will have tangible U.S. national security benefits remains a relatively untested proposition.”

“There are no standard guidelines for determining the goals of [counterterrorism] security assistance programs, particularly partner capacity-building training programs, or for assessing how these programs fit into broader US foreign policy objectives,” reads a 2016 Center for a New American Security report. “And there are few metrics for measuring the effectiveness of these programs once they are being implemented.” And in his 2016 report on IMET for the Council on Foreign Relations, Kurlantzick noted that the effort is deeply in need of reform. “The program,” he wrote, “contains no system for tracking which foreign military officers attended IMET…. [a]dditionally, the program is not effectively promoting democracy and respect for civilian command of armed forces.”

Studies aside, the failures of US training efforts across the Greater Middle East have been obvious for years. From the collapse of the US-built Iraqi army in the face of small numbers of Islamic State militants to a stillborn effort to create a new armed force for Libya, a $500 million failed effort to train and equip Syrian rebels, and an often incompetent, ghost soldier–filled, desertion-prone army in Afghanistan, large-scale American initiatives to build and bolster foreign forces have crashed and burned repeatedly.

One thing stateside US training does seem to do, according to Caverley and Savage, is increase “human capital”—that is, foreign trainees’ professional skills like small unit tactics and strategic planning as well as intangibles like increased prestige in their home countries. And unlike other forms of American aid that allow regimes to shuttle state resources toward insulating the government from coups by doing anything from bribing potential rivals to fostering parallel security forces (like presidential guards), FMT affords no such outlet. “If you give assets to a group with guns and a strong corporate identity within a country lacking well-developed institutions and norms, you create the potential for political imbalance,” Caverley told TomDispatch. “An extreme example of that imbalance is an attempt to take over the entire government.”

Strength and Numbers

The United States has a troubled past when it comes to working with foreign militaries. From Latin America to Southeast Asia, Washington has a long history of protecting, backing, and fostering forces implicated in atrocities. Within the last several months alone, reports have surfaced about US-trained or -aided forces from the United Arab Emirates, Syria, Cameroon, and Iraq torturing or executing prisoners.

Some US-trained figures like Isaac Zida in Burkina Faso and Amadou Sanogo in Mali have experienced only short-term successes in overthrowing their country’s governments. Others like The Gambia’s Yahya Jammeh (who went into exile in January after 22 years in power) and Egypt’s president—and former US Army War College student—Abdel Fattah el-Sisi have had far more lasting tenures as strongmen in their homelands.

Any foreign military training provided by the United States, write Caverley and Savage, “corresponds to a doubling of the probability of a military-backed coup attempt in the recipient country.” And the more money the United States spends or the more soldiers it trains via IMET, the higher the risk of a coup d’état.

In 2014, the United States resumed IMET support for Mali—it had been suspended for a year following the insurrection—and even increased that funding by a modest $30,000. That West African nation has, however, never recovered from the coup crisis of 2012 and, half a decade later, remains wracked by an insurgency that Sanogo, his successors, and a French- and US-backed military campaign have been unable to defeat. As the militant groups in Mali have grown and metastasized, the United States has continued to pour money into training local military personnel. In 2012, the year Amadou Sanogo seized power, the United States spent $69,000 in IMET funds on training Malian officers in the States. Last year, the figure reached $738,000.

For the better part of two decades from Afghanistan to Iraq, Yemen to Pakistan, Somalia to Syria, US drone strikes, commando raids, large-scale occupations and other military interventions have led to small-scale tactical triumphs and long-term stalemates (not to mention death and destruction). Training efforts in and military aid to those and other nations—from Mali to South Sudan, Libya to the Philippines—have been plagued by setbacks, fiascos, and failures.

President Trump has promised the military “tools” necessary to “prevent” and “win” wars. By that he means “resources, personnel training and equipment…the finest equipment in the world.” Caverley and Savage’s research suggests that the Pentagon could benefit far more from analytical tools to shed light on programs that cost hundreds of billions of dollars and deliver counterproductive results—programs, that is, where the only “wins” are achieved by the likes of Yahya Jammeh of The Gambia and Egypt’s Abdel-Fattah el-Sisi.

“Warfighters focus on training other warfighters. Full stop. Any second order effects, like coups, are not the primary consideration for the training,” Caverley explains. “That’s why security cooperation work by the US military, like its more violent operations, needs to be put in a strategic context that is largely lacking in this current administration, but was not much in evidence in other administrations either.”