The Road to Hell (23 page)

The report calls CARE the “most efficient agency in the Somalia refugee program” and notes that 1,800 national and international staff were employed by the end of 1984.

As it turns out, however, this “evaluation” of the CARE project was written by the CARE branch office in Mogadishu. It was the very people who were collecting that million dollars in salary who decided that phasing
out their own jobs wasn't a priority. As I looked through UNHCR's files of evaluation reports, I realized that most of the reports proclaiming CARE's success were “self-evaluations” based on the “branch office's observation of the overall project implementation.”

I had a vision of CARE, UNHCR, and NRC, alone in the desert with some Somali refugees and tens of millions of dollars.

There was no mention in any of the documents about repatriation, and of the hundreds of meetings logged that year, none was about ending the project. Everything was about delivering food more efficiently, nothing about ending it altogether. It was as if ELU/CARE was just another business planning for a bright and distant future. Whatever it was, CARE headquarters in New York gladly took its 7.5 percent administrative fee of at least $600,000 a year, on the way to building the largest NGO in the world.

T
here was all sorts of correspondence, but mostly there were cash requisitions: $10,000 for shovels, $4,000 for office supplies, $8,000 for tarps; money given from UNHCR to the National Refugee Commission adding up to millions and millions of dollars year after year after year. I thought about those refugee camps in Somalia, and I tried to imagine a UN auditor trekking through the huts with a translator who was working for the NSS, trying to ascertain whether all those tarps and blankets and buckets on the requisition forms had ended up in the hands of refugees.

And of course there were far fewer refugees than were actually budgeted for. After several failed attempts at counting the refugees, a number was agreed upon—750,000—although the Somali government and NGOs continued to use numbers from 1 to 2 million, depending on whom they were trying to impress.

Then, as I looked through the files, I found information about the final attempt to count the refugees. It was a letter from a Dr. Murray Watson, on the letterhead of his company, RMR (Resource Management and Research), written on December 20, 1990, a month before the government of Siyaad Barre collapsed. The letter was addressed to the recently appointed UN High Commissioner for Refugees.

UNHCR has behaved incorrectly in its contractual relationships for political reasons and even now is hiding behind immunity to avoid facing its civil responsibilities to RMR.

UNHCR under the same temptation has, it would seem, the same sort of wish to use the “numbers game” for its own interests as does the Somali government,
Oxfam, and the world NGO community as a whole. For understanding why these bodies feel an exaggeration of the magnitude of impending famines etc. is justified, we cannot help but observe that so-called donor fatigue is more than just a reluctance to support the same emergency too frequently. It is also evidence of a growing public perception that there is a welloiled begging industry which can create dramatic images of suffering and six-figure forecasts of death with considerable skill more or less regardless of the actual situation. Our company is very experienced in getting at the numbers behind the myth, be they elephants, Somali wild ass, or Ogaden refugees. We did indeed, despite continuous interference from the UNHCR branch office, develop an extremely usable methodology for refugee estimation.

Unfortunately no UNHCR personnel who came into contact with the work were able to understand the intricacies of the method, nor did they seem to understand why it was necessary for RMR to treat UNHCR and NRC evenhandedly, providing neither party with special explanations of the implications of the results.

Watson was claiming that he hadn't been paid. His attempt to enumerate the refugees was stopped.

Dr. Murray Watson is one of those big bwana figures who seem to wander into every African quagmire. A consultant based in London and Nairobi, Watson describes himself as an ecologist. He is also a pilot with his own aircraft. In 1978 he began using aerial surveillance to count cattle, elephants, and other large animals. He didn't figure that counting people would be all that different when UNHCR asked for tenders on a contract to “re-enumerate” the refugees. Watson had close connections with Somalia and the Somali government. Some Somalis claim that he informed on Somali dissidents in Nairobi who were later collected by Kenyans and handed over to the Somali NSS. Watson denies it.

The Somali government wanted Watson to get the enumeration contract. The UN wanted to hire I. M. Lewis, the Oxford-based dean of the Somali studies universe. Watson got the job. It was a small contract, $300,000, including money to hire the people and the Land Cruisers. He immediately embarked on a training program for his enumerators. Almost without exception, the people he hired criticized his methods. UNHCR also insisted on having an observer present while he was carrying out his operation. The UN observers agreed with the assessment and cut off his funds before the operation was completed.

Watson, who has been described as both a genius and a kook, rises to his own defense. “No one had the capacity to understand the process, my
methodology,” he says. According to Watson, his design had to be carried out in secrecy. UNHCR and the NRC were full of leaks. If word of the methodology got out or if their schedule for counting refugees got out, he wouldn't be able to do it. The Somalis, and even the NGOs, would do everything in their power to skew the count. Watson's claim is supported by some of the documents I found in Geneva.

Before Watson and his team started the count, they went to Beledweyne to do some training. The rumor spread that they were going to start count' ing for real.

Among the papers I found in Geneva was a February 1988 letter from the local UNHCR representative to the NRC commissioner Mohamed Abdi Tarrah:

Dear Mr. Tarrah,

Just a short note to inform you the re-enumeration team has started their two'month training program for interviewers on 11 February in Beledweyne. The following observations have been made:

—Many town refugees have gone back to the camps. It was observed that the main street in Beledweyne was indeed much less crowded, especially in the evening!

—Many refugees are coming back from Mogadishu!

—ELU/CARE ration trucks which usually have passengers on their return to Mogadishu are quite empty these days!

—New houses are being built in the camps!

—Refugee workers in town rush home immediately after working hours! ELU/CARE refugee employees do not come to English class in the evening anymore!

Though the team made it clear that they were here for training only, rumors have gotten around that counting has started. All the inhabitants in the camps have been alerted and those elsewhere have been alerted about the reenumeration exercise.

The letter reconfirmed that many of the refugees were totally integrated into Somali society. The refugee camps were suburbs. People had jobs and businesses. The refugee supplies were subsidies, yet the NGOs and CARE continued to raise money to feed them.

Watson's method involved aerial photography combined with surprise raids on camps by his team of enumerators, who would take attendance in the refugee huts. Sometimes those raids were to take place at three or four
in the morning. If a child was missing from a home at three in the afternoon, a mother could have any number of excuses. If that same child was also missing at three in the morning, the child either didn't live in the camp or didn't exist.

“You couldn't just look at refugees and accuse them of lying, but if the child was missing on both visits and they couldn't come up with a thesis, then I would know that they were lying.”

Watson recalls that he had been warned by a UNHCR confidante in Geneva that he was “in for a rocky ride.” UNHCR no longer wanted or needed this count by early 1988. Someone would have to pull the plug on the refugee programs no matter what kind of numbers Watson came up with. UNHCR was having its vehicles stolen at a regular rate—in the north, the SNM was stealing them to be made into technicals
*
; in the south, bandits and soldiers were looting them. UNHCR had begun to regard its personnel as hostages. Civil authority in the country was beginning to break down.

Watson also accuses ELU/CARE of getting in the way of a count. They refused to service UNHCR vehicles that he was using for the survey. “When the thing had broken down, they had actually refused to provide fuel,” he said. In addition, the UN observers who accompanied Watson's teams criticized his methodology. “They wanted to sabotage any possibility I'd come up with the real numbers. They wanted to discredit the survey, degrading the role of objective science. These are things on which I cannot compromise.”

Watson says that it was during the training exercise in Beledweyne that he realized things were not going to work out. “This is when I realized how dishonest and incompetent UNHCR was. They'd been working in Somalia in this crazy operation for a long time and were not objective. I was astonished by the lack of honor and their cynicism. They thought everything Somalis did was devious, sinister, and wrong.”

“One UNHCR observer submitted a report on my method—which was not her job. It was her job to determine if procedures were being carried
out, to see if I was performing as a contractor, not a scientist. None of them was qualified to understand methodology.”

Watson, for example, developed a language key so he could show the refugees pictures of a chicken, and, based on their pronunciation, determine where the refugee originated from. The UNHCR observer didn't approve. “This Japanese lady produced a report saying Dr. Watson was trying to find out where refugees are coming from. A better way would be to have a questionnaire about local names of commissioners and towns, et cetera. I responded: Oo you know how much it would cost to produce that kind of database? And it would take all day to do that part of the survey. I have to develop something so we can interview a thousand refugees in an afternoon.* That was the sort of argument I was getting into.”

Watson says that his initial data showed that the actual numbers were 40 to 50 percent of the number of refugees officially on the books. Of UNHCR he says, “They're the most unpleasant, dishonest people I ever met in my life. Dealing with them was worse than the money I lost on the project.”

The real reason the UN ditched Watson may be found in this memo sent to Rome by Robert C. Chase, the UN World Food Program's operations director in Somalia, on January 12, 1989:

Reenumeration … finally commenced, but the results are not expected until Spring this year. They are in any case not likely to be impartial since the north of the country has been affected by civil war and access to refugee camps has been impossible since June. In any case, UNHCR, who have been monitoring the progress of the survey, are not at all convinced that the methodology finally applied guarantees reliability of the final result. Doubts about the numbers will persist.…

However, looking at events in 1988 … UNHCR feel the reenumeration will be less significant because there is no further reason for the refugees to remain in Somalia. WFP agrees with their views.

There are many duplicate ration cards and there is reason to believe that the numbers of beneficiaries on the ration cards have been exaggerated. Judging by the extent to which sales of donated food took place, it was equally obvious that food was being received by beneficiaries in excess of their basic requirements. As a result, it was accepted that the most painless method of reduction of food supplies, and therefore passing across the message that international relief is coming to an end, would be to reduce the quantities of food supplied to the refugees.

That would be much easier said than done. In 1988, following the government assault on Hargeysa—one UN memo called it a genocide—the
UN knew the end was near. The programs would have to be ended before the government of Siyaad Barre was ended. The problem was how to get out of the refugee situation gracefully before the government collapsed, and how to get out without upsetting the armed refugees who were being fed. A blizzard of a cable traffic followed between Somalia and UNHCR headquarters, as well as between Somalia and WFP headquarters in Rome. The UN wanted out but didn't have the courage to walk away.

According to Murray Watson, the NGOs agreed to shut down the refugee programs in the late 1980s for one reason: “The refugees ceased to attract funds. The event lost the power to pull in money. NGOs trade in the guilt industry. They could no longer sell hungry people.”

On October 6, 1988, Jean Pierre Hocke, the UN High Commissioner for Refugees, sent the following trial balloon in the direction of Dr. Mohamed Ali Hamud, minister of state for foreign affairs of the Somalia Democratic Republic:

The circumstances that caused the great majority of the refugees to flee no longer prevail. Their home areas are at peace and there have been other positive changes affecting their situation. For some time refugees have been returning to Ethiopia both spontaneously and in limited numbers with UNHCR's assistance. Thus, as a first stage in phasing out relief assistance, food supply in excess of the real individual needs must be eliminated. This is a precondition for durable solutions. For to make the phasing out of relief assistance conditional on attaining such solutions would be in effect to continue to offer the third option, open-ended relief assistance.

With our colleagues in the world food program, we have been examining various ways of reducing the amount of food now supplied. As is to be expected after providing free food for so long, this will not be easy, but we believe that a progressive reduction in the maximum number of beneficiaries per ration card is the most viable approach.