Early last year, the ABC newsmagazine show 20/20 aired a segment called “Scandal in the Peace Corps.” It told the story of a young volunteer named Kate Puzey, who was teaching English in a small village in the African country of Benin at the time of her violent death in 2009. Shortly before her murder, the twenty-four-year-old Puzey sent an e-mail to the agency’s country director saying she had reason to believe that a male Beninese Peace Corps employee based in her village was raping and molesting young children. Aware that the suspected rapist’s brother worked in the same office as the director, Puzey asked that her message remain confidential. It appears, however, that news of her whistle-blowing somehow made its way back to the accused, with bloody results. On March 11, 2009, Puzey was found dead on the porch of her village home, her throat slit. A Beninese criminal investigation is still pending.

Puzey’s suspected killer is a minor and shadowy presence in the 20/20 report, the bulk of which scrutinizes the Peace Corps’s handling of her case. According to Puzey’s parents, the agency gave them no clue about the circumstances surrounding their daughter’s murder; they say agency officials did not even personally visit the family to extend their condolences until after ABC started its investigation. When interviewed by 20/20, a high-level Peace Corps official declined to comment about the events leading up to Puzey’s death, and on whether the agency bears some culpability, deferring instead to the unfinished Beninese investigation. This prompted ABC correspondent Brian Ross to accuse the agency of “stonewalling.” Finally, the segment closed with an interview of several women who said they had been raped or sexually assaulted during their Peace Corps service, only to be hushed, ignored, or treated with less than basic compassion by the agency after their ordeals.

It’s not at all hard to fathom why a show like 20/20 would take up Puzey’s story. It is, after all, the tale of a young woman’s act of bravery, and how it resulted in her untimely death. But the segment also rested on something else: the sheer man-bites-dog quality of anything sordid coming out of the Peace Corps. “It is one of the most iconic and respected organizations in the world,” said 20/20’s anchor, Elizabeth Vargas, as she introduced the segment. “Its very name embodies the best ideals of service.”

Ironically, that sainted reputation may lie at the root of the kinds of failures that 20/20 calls out—namely, the agency’s propensity to be evasive, opaque, or even callous when things go wrong. Since the beginning, the Peace Corps has been treated as a kind of Agency on a Hill, and its volunteers as exemplars of national conscientiousness. (“They are the greatest advertisement for the American system of government that there is in the world,” said Sargent Shriver, the agency’s founding director. “They are worth a thousand Coca-Cola signs.”) And likewise since the beginning, the agency has had a tendency to plug its ears against anything that contradicts that glowing narrative, even in cases when confronting such inconvenient evidence would be in its long-term self-interest.

In the early years, under Shriver, the agency had a bulwark against such complacency in the form of Charles Peters’s Office of Evaluation. (After his time in the Peace Corps, Peters went on to found a small political magazine called, ahem, the Washington Monthly.) “My job in the government was to rub the collective noses of my agency’s top officials in what they were doing wrong and why our programs in the field weren’t working,” Peters recalled in these pages last year. The agency’s lily-white reputation, Peters wrote, “had the effect of making it even harder for the top officials to face my news that things weren’t quite as good as they seemed.”

If the Peace Corps has a tendency to believe its own PR— and with such generally flattering press, who wouldn’t?—it is in part because the agency’s good reputation is well founded. In September of last year, the National Peace Corps Association released the largest survey ever of volunteers who had finished their terms in the field; 90 percent of those who responded rated their Peace Corps experience as excellent or very good, while 98 percent said they would recommend service to their close family. I was a Peace Corps volunteer myself, and during my service I met volunteers working in Namibia, Botswana, Zambia, Lesotho, Chad, and Nicaragua. The vast majority were extremely positive about their assigned country and their service.

But the Peace Corps’s perennial shortcoming—almost its tragic flaw—is its extreme reluctance to admit mistakes when they occur, and its less-than-strenuous efforts to root out screwups, breakdowns, and dysfunction when they are in the making. While not inherently a security issue, cases of violence are this tendency’s most extreme manifestation. A more typical result, as Peters noted repeatedly as far back as the 1960s, is that volunteers wind up stuck in a strange country without adequate preparation, meaningful work to do, or a suitable place to live—which is pretty much the experience I had working as a volunteer in South Africa from July 2009 to August 2011.

In 2008, I graduated from Reed College with a degree in chemistry, ready to make my way in the world but caught up short by an economy in free fall. As many Americans have done in similar circumstances, I turned to the Peace Corps. After filling out dozens of pages of paperwork, getting several medical evaluations, and choosing my geographical preference, my recruiter called in May 2009. She informed me that I would be deployed to teach math and science in South Africa in July and said that I should “get more teaching experience” in the meantime. With just a matter of weeks before my departure, I wasn’t exactly sure how that was supposed to work. So I put my trust in the Peace Corps’s vaunted volunteer training program.

I shouldn’t have. At the training center in a South African village called Marapyane, logistical problems plagued us from start to finish; often my fellow volunteers and I would wait hours for a session to start or for transport to arrive. The language course in Setswana was comparable to a semester of high school instruction—and that’s being generous. Not a single volunteer in my group even approached fluency, though a few achieved basic competence, mostly through their own efforts. The teacher training was conducted primarily by current South Africa volunteers, and did not prepare me for what I would be facing. I was afforded precisely two hours of practice teaching.

In training, I picked up on a theme that would mark much of my service: it wasn’t that the Peace Corps itself did not have high standards; it was that the people directly responsible for setting me on my feet had learned to skate by under the agency’s radar—or there was little evidence of a radar at all. For example, at the end of our language training, a mandatory audio recording was made of our final verbal language test, presumably for review in the D.C. headquarters. Our examiners cheated by showing us the questions written out in English on the sly while they spoke them in Setswana. It’s not hard to see why they did so; if they had not, I estimate that three-quarters of my group would have failed—which would have brought undesired attention from headquarters.

Of course, there were good moments. During training, I stayed with a poor family consisting of a grandmother, her son, and her two grandchildren, a boy and a girl. During my first couple of days at their house, the grandmother was away at a funeral and the kids were in Pretoria with their mother. I spent the time reading and awkwardly hanging out with the children’s emaciated and AIDS-stricken uncle, who spent most of the time sleeping. I felt lonely and ill at ease, but when the rest of the family returned to meet me for the first time, that changed instantly. The kids were literally jumping with excitement to have me as a guest. That night they put on an old tape of gospel music and the four of us danced in the living room, the little girl riding on my shoulders.

But even fostering such moments of happy cultural exchange requires some vigilance on the part of the Peace Corps, I would soon learn. After training, I moved to my permanent site, a tiny village in the Northern Cape Province, about 370 miles west of Pretoria, hopeful things would take a turn for the better. They didn’t. Peace Corps officials had placed me in what I think I can call, with little fear of exaggeration, a stupendously inappropriate home. The patriarch of the family had hanged himself in the family shop in 1979 after he beat his wife so badly he thought he had killed her, and the remaining members still seemed scarred. They had turned their home into a tavern, which was alternately empty or full of reeling drunks begging for money. Worse, the local government disbursed the government pension from their house, meaning that once a month a truck full of cash and men packing automatic weapons would spend a couple of hours in my front yard. Needless to say, this violated basic Peace Corps safety guidelines.

The local school where I was assigned to work was no less fraught with problems. The school principal—the man who had filled out the paperwork for me to live in his village for two years—was supposed to teach math, science, and English in grades seven to nine. He did not step into the classroom to teach once in an entire quarter. It didn’t matter much, though, because he retired three months after I arrived, at the end of 2009.

Most of the ninth-grade students could not get through a simple English conversation, or figure out, say, six times four without a calculator. The teacher for the second and third grades refused to teach any English, flouting a mandate in South African law. Though corporal punishment is illegal in South Africa, the teachers regularly beat their students. When the kids learned that I would not do the same, they quickly stopped listening to me. It made classroom management a nightmare.

I grimly hung on to my job through 2010, slowly losing my morale and sense of commitment. After the school’s education department promoted one of the teachers to the job of acting principal, she tried to institute some reforms. This ignited a spate of bitter, resentful political wrangling as the teachers fought hammer and tong to avoid accountability. I began to spend more and more time away from my village and the classroom. I spent hundreds of hours wandering the bush, going through audiobooks by the terabyte. I helped other volunteers with their projects; I remain particularly proud of helping a friend refurbish a computer lab in a neighboring village from scratch, then fixing the server by myself after it broke down.

That neighboring village, only three miles away, provided ample evidence that the Peace Corps was capable of selecting and establishing far better volunteer sites than mine. The primary school in this neighboring village was excellent; it employed one of the best teachers of small children I have ever seen. By sixth grade, that school’s students could speak better English than many of the teachers at mine. It was much easier to find meaningful work in such an environment. And perhaps even more importantly, the village’s Peace Corps host family comprised some of the kindest, most generous people I met in South Africa; I was probably closer to them than to my own hosts. But this only goes to show, again, that even the “cultural exchange” component of Peace Corps service requires good choices from the agency.

At the start of 2011, well into my second year, the education department hired a new, permanent principal—a big improvement on the first one, though he could hardly have been worse—but by then I was so disillusioned and disgusted with teaching that I decided to concentrate on something, anything, else. (Even today, the thought of classroom teaching arouses a feeling in me of powerful revulsion.) I decided to revise and typeset the Peace Corps grammar manual for Setswana, and to make one last big effort to give something back to the agency: I signed up to help with the training of the new group of volunteers arriving in early 2011.

If I was so dissatisfied with my own training, I thought, I should at least do something to try to improve things. I participated in a weeklong “training of trainers” beforehand, and helped lead a week’s worth of sessions for the new volunteers. While not a disaster, it still did not go as well as I had hoped. Frustrated, I wrote a post on my personal blog that laid out what I thought was wrong with the process. I didn’t mince words. When Peace Corps officials at the South Africa post got wind of the post, they kicked me out of training.

And so, a few months later, I finished my service and came home.

How might my experience have been avoided or improved? For one, I could have switched sites. In the first few months, my supervisor asked if I wanted to move, but I declined. It was a vexing decision. To change my site—though it probably would have been wise—would have been a sharp insult to my host family that I wasn’t willing to deliver. It is easy, given the deprivations that volunteers are expected to endure, to develop a macho, “I can take it” attitude, and slog on in hopeless situations. I could have been more honest with myself. But a good supervisor probably would have recognized that continuing to work in such a situation would ultimately crush my morale—as indeed it did—and would have encouraged me to relocate.

As it was, however, my supervisor mostly ignored me. I was promised a visit from an official “within the first 4 to 8 months of service and between 14 and 18 months of service.” Instead I was visited once in twenty-three months, and I was almost never called. Budget cuts may have had something to do with this—the Peace Corps was enduring a withering round of them at the time—and again, I could have been more proactive about communicating. But the fact of the agency’s neglect stands. Whatever the reason for it, this sense that I had been forgotten only exacerbated my growing cynicism and apathy.

Most fundamentally, if the Peace Corps had followed its own stated rules, I would not have been placed at my site in the first place. My supervisor can’t have done much research on it before I arrived; even thirty minutes of investigation would have shown how dysfunctional the school was and how many rules my host family was breaking. The family—God bless them—would have cheerfully explained all the details of my housing situation if asked; as far I can tell, they had no idea there were any rules to break. And though I admit it might have been unprofessional to publish my criticisms of training online, the reaction I got is telling: I was dismissed, and my criticisms were largely ignored.

But as noted above, blunt criticism and oversight was critical for the Peace Corps’s early development. In Stanley Meisler’s excellent history of the Peace Corps, When the World Calls, he tells how Peters, as evaluation chief, discovered that many of the first groups of volunteers didn’t have meaningful work at their Peace Corps sites. “It was painful to see the idealism of the Volunteers squandered as they sat there with nothing to do,” Peters wrote in an evaluation. Meisler details how idealists at the D.C. headquarters fought with Peters over the “numbers game”: the idealists pushed for the maximum number of volunteers, while Peters pushed back, as the Peace Corps often had not laid the logistical groundwork to ensure that the volunteers had good sites and meaningful work.

Current top agency administrators would no doubt protest that they have vigorous, extensive, in-house oversight. But they do not. It is true that there is still a division labeled “Evaluation,” but it is only a shadow of its former self, having come under the knife during the Nixon administration after Peters left the agency in 1968. In June 2010, the agency produced for the first time an overall report on the Peace Corps, written by this evaluation division. Though it contains many reasonable recommendations, it is basically a public relations document, slickly produced and written like a corporate press release: “The Peace Corps at fifty is ready for a strong new beginning— rooted in the vibrant past of those early years, yet ready to harness twenty-first century American intellectual power, innovation and commitment to results.”

The Peace Corps’s Office of the Inspector General provides the closest thing the agency has to meaningful oversight. Its reports and congressional testimony, which are posted online, are far more incisive and clear-sighted than the aforementioned report. Unlike the usual IG model, which only investigates problems after they occur, they proactively evaluate a few posts a year. But this is still a far cry from the Peters days. He had his team evaluate every post once a year, and the reports often ran over a hundred pages. Expanding the OIG to the old standard and independence would require a bit of extra money, but the amounts involved are a rounding error in the federal budget—for the 2012 fiscal year, the office requested $5.3 million. To give some perspective, the entire Peace Corps budget is around $375 million (and historically much less), which is only a little more than what the U.S. spent in Iraq every day for the past eight and a half years.

Nevertheless, wringing more funds out of Congress does not seem to be an option at this point. Republicans have been cynically using the budget deficit to slash programs they don’t like; they cut $25 million from the Peace Corps allocation in the last quarter of the 2011 fiscal year, after most of the year’s money had been spent. I can personally testify that this caused all manner of chaos as posts scrambled to pinch pennies.

Even if the Peace Corps must accept some funding cuts, however, keeping the quality of sites as high as possible should be a top priority for the agency—even if that means fewer volunteers and more evaluators. As Peters insisted in his fights over the “numbers game,” bad sites are bad both for the volunteer and for the agency as a whole. My site was shuttered after I left, and though I can’t be sure if it was due to budget cuts or to the bold, all-caps, underlined, twenty- four-point-font complaint I filed before I left, I am confident that the South Africa post as a whole is better for it.

The Peace Corps is well worth improving. It still represents the best of America, plays a crucial but underappreciated role in our efforts at public diplomacy, and provides us with a critical dose of international awareness in a global age. And the problem is not that the agency is incapable of competent management; it’s that the management is uneven. At the beginning of every volunteer’s service, he or she must complete a community survey—a look at the resources and needs of the volunteer’s permanent site. In South Africa, the community survey consisted of a large batch of paperwork, and my supervisor said not one word about it after I turned mine in. Some of my peers never even filled one out. In Nicaragua, however, the community survey takes the form of an extensive presentation that volunteers must deliver in front of their entire village and a Peace Corps representative, in Spanish, and if it is not accepted the volunteer is sent home. In South Africa there was no penalty for failure even on our rigged language exams, while in Nicaragua failure means dismissal, which results in nearly everyone studying hard and passing. Where I had only a single visit from a supervisor in two years, in Nicaragua volunteers are visited frequently and closely monitored.

The tools and techniques required to make a great post are already known. What the agency sorely lacks is someone— or, better yet, a small office of someones—whose job is to find out whether every post is using these tools and techniques, and who will complain bluntly, in public if necessary, if one isn’t. The Peace Corps administration will likely fight like mad to prevent it, but the restoration of aggressive and regular oversight to all parts of the agency is the only thing that will bring the Corps up to a high standard—and help it earn back the stellar reputation it now stands at risk of losing.

Ryan Cooper

Follow Ryan on Twitter @ryanlcooper. Ryan Cooper is a national correspondent at The Week. His work has appeared in The Washington Post, The New Republic, and The Nation.