CIA HISTORICAL REVIEW PROGRAM
RELEASE AS SANITIZED
18 SEPT 95
No Foreign Dissem
A case study in political research: reconstruction of the 1964 revolution.
A reconstruction of the events of the Zanzibar revolution of January 1964 shows particularly well the usefulness of going back for an unhurried reexamination of a crisis after all the returns are in: it reaches conclusions about both events and causes quite different from what was generally believed at the time.1
Because the Zanzibar revolution was so unexpected and so quickly over, there was confusion and uncertainty about the most basic questions, even about who started it and why. The need for an immediate assessment on which to base U.S. policy toward the new regime, however, required that conclusions be drawn on the basis of the information then available, though this was recognized to be inadequate. With the passage of time, therefore, and the accumulation of a great mass of retrospective reporting on the coup, a second look at the evidence seemed called for. As it turns out, the new evidence justifies a new verdict.
Needless to say, Zanzibar had not been a major target of our intelligence effort. Moreover, scholars on the outside had not studied the politics of the island and so compensated for our lack of background information. Intelligence reporting before the coup had concentrated on the Arab political minority from which the party in power was drawn, not on the African majority. This reporting indicated that active plotting for the overthrow of the government was being done in the Umma Party by followers of the radical Arab leader Abdulrahman Mohammed Babu. There were no reports of such plotting centered on the Afro-Shirazi Party, the chief spokesman of the African majority. Thus we did not know then what has since been established from good sources—that in the Afro-Shirazi Party a radical group of African trade union leaders led by Abdulla Kassim Hanga was also making plans, independent of those of the Arab Babu and his Umma Party group, for a revolution. As early as the middle of 1963 a group of ASP leaders including Hanga had gone to Tanganyika to ask President Nyerere for money and arms in support of their projected uprising.
In any crisis such as the Zanzibar revolution, there are bound to be inadequacies in the information available on events of the moment. In this case, however, we were all but cut off from any word about what was going on, because the new leaders promptly sent all but a few select foreigners on the island back to the mainland. For days the main news out of Zanzibar came from a remarkable series of radio broadcasts to the people by one John Okello, a hitherto unknown who thus suddenly emerged as chief spokesman for the regime. The only thing known for certain about Okello was that he was an African. In his first broadcast, he announced that he had been the strong man behind the revolution and that the government was now "run by us, the army."
We, the army, have the strength of 99 million, 99 thousand ... Should anyone be stubborn and disobey orders, then I will take very strong measures, 88 times stronger than at present.
I was a very high ranking person in Kenya in the Mau Mau army which knows how to make weapons. I can easily make not less than 500 guns per day. Undoubtedly, I can make a bomb that can destroy an area of 3 square miles. I can make about 100 grenades in an hour.
As the days passed, Okello's boasts about his role in the revolution and his power grew ever more fantastic. On one occasion he declared, "I am above the government and cannot die." It was these broadcasts, more than anything else, that seemed at the time to establish the revolution and the new government as extremist and unstable if not irrational. This was a fundamental misconception.
The emergence of Okello to a position of prominence was the most curious and most confusing aspect of the situation. There is a natural tendency to assume some logical ordering behind events even in a revolution, in this case to assume that Okello must be playing some pre-arranged role. It is not surprising that we were therefore at a loss to explain his rise to power, since actually, as we shall see later, Okello was the personification of spontaneity; his role in the coup was the most unplanned and unanticipated aspect of the whole unplanned affair.
In the midst of a crisis the hard-pressed analyst has no time to do research into the situation or check the assumptions of the reporting. At the time of the Zanzibar revolution, there were reports that revolutionary President Karume, Babu, Hanga, and other Zanzibari nationalists (sometimes Okello was included) had all worked together "without regard to nominal party affiliation" in planning the revolt. It was only later that one could see the mistake in lumping the pro-Communist elements in the ASP with Babu and the radical Arabs. There could be little basis for a close working relationship between men like the pro-Soviet African Hanga, who had strong anti-Arab feelings, and the pro-Chinese Arab Babu. At the time, however, the misleading assumptions of the reporting were not seriously challenged but carried over into the early analyses of the situation.
The conclusions of the XXXXXXXX American intelligence communities on the basis of the information then available were the following:
The revolution was the work of the Umma Party led by Babu and aided by Okello.
-British Weekly Summary, 22 January 1984.
The prime movers of the revolution were Babu's followers. However, they do not represent the majority of Africans.
-Dept. of State memo, "The Communist Spectre Looms in Zanzibar," 13 January 1984.
The Umma Party has been plotting a revolt, and the spark which set it off was probably a raid in early January by the Zanzibar police on Umma headquarters. The man most responsible for the revolution was Babu.
-CIA memo, 5 February 1984.
The Western press must have left this same impression with the general public. Typical of its analyses were the following:
Who engineered the coup? The most likely answer is Babu, leader of the Umma Party, which has no seats in Parliament but has close connection with the trade unions.
-Keith Kyle, "Letter From Zanzibar," New Leader, 17 February 1984.
It is unlikely that the Afro-Shirazi Party, even in the explosive situation in Zanzibar, would have taken a revolutionary initiative. It was waiting for the 1988 elections. But in the meantime, Mohammed Babu, with the prestige of 15 months in prison for sedition ... had formed his own Umma Party. He and his associate in the ASP, Hanga, appear to have been the leading political figures behind the revolution.
-New Statesman, 17 January 1984.
The conclusion that Babu was prime mover in the coup was consistent with what was known about him. The leading Zanzibar nationalist, he was the most outspoken critic of the West on the island. He had been known for several years to have Chinese Communist contacts and he had just returned in mid-December from China, where he was suspected of having attended a course in the military tactics of revolution. When his Umma Party headquarters were searched in early January, the police had seized large quantities of documents, including a diary written in Peking with a full description of methods for overthrowing a government by violence. Further, it was reported that the Umma Party had brought a supply of arms and ammunition into Zanzibar over a three-month period at the end of the previous year.
It was also a reasonable conclusion, at the time, that Okello must have played some role in Babu's plan and that he must then be speaking with the authority and approval of Babu and the other leaders.
The Facts Reexamined
For about a year after the revolution we continued to receive good new retrospective reporting which contributed to our understanding of what had happened. At the end of this time the main facts could be reasonably well established, and it was possible to proceed with a reconstruction of what lay behind them.
The first step was to sort out fact from conjecture in the mass of conflicting reports on the revolution. This was something that most observers at the time hadn't done. Perhaps the most spectacular example of their failure to discriminate was the widely publicized report that Cuban nationals had fought on the side of the rebels. Journalists to whom U.S. and British refugees reported they had seen Spanish-speaking soldiers wearing Cuban-type uniforms jumped to the conclusion that a number of Cubans were involved in the coup, and from then on the Cubans' presence on Zanzibar was reported as fact. It was necessary to track down the source of the refugee report and find out what basis there had been for it.
It was first established that the refugees had talked merely of Spanish-speaking soldiers. Now we knew that a group of about 25 Zanzibaris, all Arabs, had received military training in Cuba in 1962, and our initial supposition was that these had during the course of the fighting spouted Spanish slogans they had learned in Cuba. Then from a very reliable source who had been in Zanzibar at the time of the revolution and was closely questioned on this point later, it was finally learned that one individual, a Pakistani, had been wandering around the island during the revolution in a Cuban outfit, sporting a Castro-type beard; he and no one else. His behavior, and that alone, had been responsible for the rumor that Cubans were involved in the coup.
Such interviews with persons who had been there at the time or had other fast-hand information were of fundamental importance in establishing the main facts. We questioned most, if not all, of the U.S. officials who had served in Zanzibar before, during, and just after the revolution, as well as those familiar with the Zanzibar scene through a tour of duty in Dar-es-Salaam on the mainland. This kind of investigating can be done from Washington only some time after the event, when most of those to be questioned have returned here in the course of reassignment or at least for debriefing sessions.
Perhaps the main reason why such a reconstruction as this is not done more often is that it takes a lot of time—time to collect all the reports, sift through, screen, and correlate them, and follow up the questions they raise in personal interviews. Time is one great advantage the researcher has over the harassed current intelligence analyst. The other advantage is a greater variety of sources. After a crisis a great deal of retrospective reporting is available from new as well as the old sources. In the case of Zanzibar, the enrichment in sources brought a great improvement over pre-revolutionary coverage. For one thing, it provided new information on the pre-revolutionary activities of the radical group in the ASP, establishing the crucial fact that not just Babu's Umma Party group but at least one if not two others were independently planning for a revolution.
The second phase in the process of reconstructing the Zanzibar revolution consisted of developing tentative hypotheses and testing them against the facts. At this stage the theory of Babu's responsibility for the coup—the generally accepted interpretation—fell to pieces. Lengthy examination turned up no credible evidence that he had played any significant part at all. He was not in Zanzibar when the fighting started in the early hours of 12 January, but returned to the island only that evening. The testimony of Israeli businessman Misbah Feinsilber, who owned and operated the boat which brought him back, is crucial here: it was Feinsilber's strong impression at the time that Babu was completely uninformed about the events of that day in Zanzibar. A Zanzibari attorney who knew Babu well was also convinced that Babu had no previous knowledge of the coup. During the course of the day's fighting some of Babu's followers eventually joined the ranks of the rebels, but they do not seem to have played any part in the initial attacks. Most significant of all is the fact that the supplies of arms and ammunition which the Umma Party had brought in and concealed in different spots on the island were not used.
Other hypotheses were tested in the same manner. The contention in several reports that Okello had actually planned the revolution himself was unconvincing in the absence of any evidence to that effect, and Okello's own performance as a national leader afterward raised doubts that he was capable of planning anything at all. It was much more reasonable to suppose that the reports reflected deliberate attempts by Okello and his followers to enhance his role in the revolution; these men were definitely known to have distorted other facts to fit their exalted picture of him as the savior of his country.
The possibility that the revolution was the result of the secret planning by Hanga and other members of the extreme left wing of the ASP proved to be the best working hypothesis. It was consistent with the most obvious and the most important fact about the Zanzibar revolution—that it was a movement by Africans, not Arabs, to put Africans in control of the government. The violence of the revolt and the bloodletting that followed the overthrow of the Arab Sultan was manifestly racial violence by Africans against Arabs.
There were still a few facts that were not consistent with this broad hypothesis. The third and last phase in the reconstruction process consisted of refining it to square with these. For one thing, Hanga's plan apparently called for a coup in March or April, rather than January; as of a few days before the revolution the planners were still proceeding on the basis of a D day in the spring. Several of the key planners actually expressed surprise that the revolution occurred so soon, not according to their plan. Apparently something bad precipitated events ahead of time.
A great majority of the reports trace the course of the revolution from the same starting point—an African fete held at ASP headquarters on the night of Saturday, 11 January. Gradually it became clear that the revolt must have grown out of spontaneous action by the Africans at this affair. It seems to have been triggered by a rumor that the government planned mass arrests of ASP leaders the next day. A number of disgruntled ex-members of the Zanzibar police force who knew where the police arms were stored and how to get access to them are known to have been at the fete; they appear to have urged the crowd to follow them—right then—in an attack on the government's strongholds. It is fairly well established that Hanga, Karume, and other ASP leaders did address the crowd that night at the Seamen's Union Club, whither it had moved from Party headquarters. Apparently, sensing the excited state of the Africans, they decided then and there to seize the opportunity to overthrow the government. The revolution that materialized on the morning of 12 January was thus actually not the one planned by either Babu and his Umma Party or Hanga and his ASP group. It was more nearly a spontaneous action.
Okello was a creature of this spontaneity. It is possible that he was at the African fete and somehow contributed to touching off the attack with wild talk of revolution, but not one report mentions him by name as having played any role in inciting the crowd. He does not appear on the scene that night until he emerges several hours later as the hero of the attack on the Ziwani police armory and is immediately accepted by the rebels as their new leader. It is now quite clear that Hanga, Karume, Babu, and the other leaders had had only chance contacts with him before the revolution; they had certainly not foreseen that he would come out of it with a large popular following and claim a share equal to their own in running the government. Afterward, as soon as they conveniently could, they rid themselves of him.
Only such an explanation of the Zanzibar revolution fits all the major known facts about it. It is recognized that the researcher has an unfair advantage over the current analyst in being able to test a number of hypotheses slowly and carefully against all the information that has gradually come to light over a period of time and only then commit himself to an interpretation. On the other hand, because of this advantage, he has a responsibility to get the answer right.
No Foreign Dissem