When I joined Ehud Olmert’s government on June 18, 2007, as minister of defense, it was almost three months since planning of the destruction of the Syrian reactor in Deir az-Zor had begun (in late March). I was aware of this activity, having been briefed in late April about the reactor’s existence by Olmert, Mossad head Meir Dagan, and IDF head of intelligence Amos Yadlin. Asked for my opinion on what should be done, I answered on the spot: “We must destroy it.” This issue was the reason for my insistence on entering the defense ministry as soon as possible. I assumed that the Israel Defense Forces (IDF) was deep into preparations to execute an operation, and I believed I could contribute to the operation’s success.

On my first day at the ministry (I had already served as defense minister alongside my premiership, 1999-2001), I convened a “status of operation” discussion with participation of all relevant operational arms—Intelligence, Mossad, and Air Force (IAF), as well as experts on nuclear reactors. The two operational plans for the reactor’s destruction on which the Air Force and others had been working were presented to me in full detail. The prevailing view in the room, as well as the conventional wisdom in the Prime Minister’s Office, was that of an urgent, immediate need to implement the plan, preferably within a week or two. It was also perceived as critical to proceed swiftly lest our awareness of the reactor’s existence became public, which would significantly complicate its destruction, and before the reactor became “hot” and rendered the operation impractical. There was a general unanimity regarding the need “to destroy the reactor and avoid a wider clash with Syria.” To my surprise, I found that both plans, quickly nearing “D-day,” failed to meet these requirements.

The first plan envisaged a massive air attack that might surely destroy the reactor but would involve a direct engagement with the Syrian air force and air defense. Such an attack could not conceivably be denied the morning after and carried a significant risk of triggering a wide clash with Syria and possible deterioration to full-fledged confrontation with Hezbollah in Lebanon as well. I called this plan “Fat Shkedi” (Maj. Gen. Eliezer Shkedi was the then-IAF commander). The second plan, prepared in the past for another mission, was an extremely “low signature” operation that would not trigger a major clash but could not assure—beyond serious doubt—the destruction of the reactor.

I pointedly asked again: “How much time do we have before the reactor becomes hot?” The answer was: “Around three months.” “Will we know for sure if and when our window starts closing, even if this happens earlier than predicted?” I asked. The answer came: “Yes, absolutely.” I summarized as follows:

I then redefined the limiting parameters more clearly: I directed all concerned actors to start working in this vein.

Keenly aware of the risks attending my directive, I ordered that “Fat Shkedi” be brought ASAP to operational completion so that it could be executed on very short notice as a hedge against the risk of a possible leak. Moreover, being unable to ensure that despite our efforts to avoid such an eventuality we would not find ourselves in a wide clash in the north, I instructed the Northern and Home Commands to increase and deepen preparations for a wide-scale confrontation.

I also asked Washington for precision munitions, spare parts, and other necessary means in the event war ensued. In order to avoid these preparations from being leaked, all these activities had to be done under a thick veil of secrecy through a variety of explanations and disguises. Indeed, it worked. On a Friday night in August, without leaks or media footprint, a U.S. vessel in the port of Ashdod unloaded 35,000 tons of munitions and spare parts necessary for possible deterioration to a full-scale war—the equivalent of 230 heavy transport airplanes carrying about 150 tons each.

The first person to come with an idea for a better plan was Shkedi himself. About a week after the above discussion, during a visit to an IAF base where preparations for “Fat Shkedi” were presented to us, he asked me to have a cup of tea with him. There, across the table, he took a triangular paper napkin, opened it flat, and drew on it a sketch of a “surgical air raid” on the reactor, which had a dramatically lower signature than the one in preparation. I asked how long it would take to have it prepared, and he said, “It could have taken a month, but since I have first to polish and ready the ‘Fat’ plan, it might take a little bit more.”

“Very good,” I said. “Start preparing it immediately and, later on, bring it for my approval.”

“Why didn’t you go this way in the first place?” I wondered.

“It was originally presented as a mission to be executed immediately within the shortest time possible, and probably carried out if a leak started developing, even before preparation had been fully completed,” said Shkedi. I dabbed the new plan “Lean Shkedi” (also meaning lean and mean).

The second person to approach me with an alternative idea was Head of Intelligence Yadlin, a former IAF senior commander who participated as a young F-16 pilot in the destruction of Iraq’s Osirak nuclear reactor in 1981. He came with a totally different idea than “Lean Shkedi,” and a somewhat more complicated plan, which involved a very low footprint that could ensure both the reactor’s destruction and a very high chance of avoiding a wider confrontation. I asked how long it would take to prepare the plan, to which Yadlin replied, “Probably two months.” “Very good,” I told him. “Start preparing immediately and bring it for my approval later.”

In the coming weeks, I led an intensive series of discussions regarding all aspects of the operation. These involved first and foremost the needs of the budding new “low signature” plans, “Lean Shkedi” and “Yadlin"; a detailed intelligence analysis of possible Syrian reactions and our possible responses; the IDF’s preparation for a possible major confrontation that could escalate to full scale war; diplomatic containment of Syria in the immediate wake of the operation, and initial thoughts about the “day after,” including implications for the struggle against the Iranian nuclear program, the risks of rising questions regarding Israeli strategic capabilities, etc. Naturally, some of these discussions were followed by similar consultations with the prime minister and the inner cabinet. On July 26, I issued a set of directives to the IDF to be ready for “a possible war in the north (Syria and Hezbollah),” and on August 3, I ordered the completion of preparations for the two “low signature” operations by September 1 as a first target date.

One has to bear in mind that, in Israel, most of these subjects fall under the responsibility of the minister of defense, unlike in the United States where the president is the commander in chief with the secretary of defense and chairman of the joint chiefs of staffs acting during operations as advisers to the commander in chief rather than as two links in the “chain of command” that goes directly from the president to the commanding generals. In Israel, the government as a collective, or its inner security cabinet, are the equivalent of the commander in chief. The prime minister is the most important member, yet he is formally just “first among equals.” The minister of defense is the person responsible for the IDF on behalf of the government and/or the inner cabinet. He is in the “chain of command” representing the government, and according to the Basic Law of the IDF, “the chief-of-staff, the top uniformed person, is subordinated to the minister of defense, reports to him and is also the next link in the ‘chain of command.’”

Many inaccurate, at times even distorted, stories and urban legends, many of them critical in tone, have been spread over the years regarding my abrupt and forceful intervention in the course of events. However, in the so-called “bottom-line test"—the test of reality—the picture that transpired is clear and definitely positive. When we sat down on September 5, 2007, to decide how to destroy the Syrian reactor, the two leading operational plans were exactly the two low signature plans that resulted from my intervention rather than those that were originally planned. And the story did not leak, nor did the reactor become “hot.” Not to mention that the operation carried out that night was extremely successful and did not lead to any clash at all.

There was tension in the air in every discussion on the prime ministerial or cabinet level. And there were many of them, several times every week. I could feel the eagerness, a hasty and impatient desire to stop this nerve-wrecking, slow advance and “just do it.” I even felt an underlying theme, well articulated in a recent Middle East Quarterly article,[1] that, for some improper reasons, I tried to delay or even to dodge the operation.

None of these thoughts had ever crossed my mind. We needed the cool-headed approach attending my experience in order to ensure the operation’s complete success: the reactor’s assured destruction, minimizing the risks of wider confrontation, and preparedness for the worst-case scenario of escalation to a full-fledged war. I just was confident that my opponents were wrong.

Thus, somewhat humorously, I made it a rule that whenever I had to talk I would say:

During one of these frustrating debates, a senior minister with a deep security background told me: In the first eight days of August, three cabinet discussions brought tensions to a head. Basically the prime minister tried to forge a majority in the cabinet, together with military and intelligence officers, who all, except for myself, Deputy Prime Minister Shaul Mofaz, and Minister of Internal Security Avi Dichter, supported an immediate attack.

The attempt to reach a majority or even consensus is legitimate, but the chosen way was not. PM Olmert basically asked the IDF’s top echelons to present an opinion and recommendations supporting an immediate attack. But he arranged it as a bypass of the basic procedural rules in Israel, where, as mentioned above, the position of the defense bodies should be first approved by the minister of defense. Of course, officers can have different views from their minister and should be allowed to express them and try to convince the cabinet. But in the Israeli constitutional framework, officers cannot bring to the cabinet “recommendations for action” that were never presented to the minister of defense who is their direct superior in the chain of command. So, to some people’s surprise, I ordered the presentation to be halted when the recommendations began to be read. The prime minister, somewhat oddly, decided to read them to the forum himself with Yadlin expressing his view as well. Yadlin could do this because, according to the same law, the IDF head of intelligence is responsible for “national net assessment” and, as such, reports directly to the government in this regard—and only in this regard. The atmosphere got heated, but the somewhat “tricky track” did not work.

In another meeting a few days later, I presented to the cabinet the ministry of defense’s position. A fierce debate ensued where the prime minister and several others expressed anxieties about a “doomsday scenario” whereby a leak of the IDF planning combined with a hot reactor would generate an irreversible rush toward a Syrian bomb, followed by apocalyptic pictures of panicky Israeli citizens fleeing abroad and the Jewish state seemingly hovering on the verge of collapse. I strongly rejected this exaggeration and insisted on the following:

• We have to, and we will, destroy the reactor once the low signature plans are ready.
• We still have some time and will know for sure when we have run out of time.
• We have the “Fat Shkedi” plan as an insurance policy in the event of a leak.
• We must complete our preparations for a possible escalation to war despite our desire to avoid this eventuality.

By way of cooling down the sense of panic in the room, I noted that we were lucky to have discovered the reactor before it turned hot:

In a third meeting a few days later, we were subjected to a long exposition by PM Olmert detailing in somewhat legalistic language the development of the project from day one and his arguments for an early attack. I kept disagreeing, claiming that we had to use all the time at our disposal to be as best prepared as we could under the circumstances, and then strike while trying to avoid being dragged into a wide scale confrontation.

These tense cabinet meetings were followed by exchange of letters between me and the prime minister that actually ironed out much of the apparent dispute, ending around mid-August with the prime minister realizing that dialogue and understanding were the right way to reach the necessary balanced solution to legitimate disputes, rather than corridor manipulations and “power games.”

Toward the end of July, we came to the prime minister’s residence for the weekly summary of developments and final approval of a sensitive operational step to be executed over the weekend with regard to the “Yadlin option.” To our surprise, we were told that the step probably had to be cancelled for reasons that cannot be detailed here. I argued that there was no need to halt any step and recommended withholding a decision for some time while checking if the problems had not been overestimated. A few hours later, it became clear that I was right.

On another Friday noon meeting, sometime in early August, Mossad Head Dagan brought a serious piece of information that according to his feeling had to compel us to overcome all hesitations and strike immediately:

Hence, he argued that we were facing an immediate risk of leakage that had to be preempted. I did not buy this alarm. Telling the forum that I happened to know the British intelligence services quite well from my experience as head of intelligence, chief of staff, prime minister, and minister of defense, if I had to assess where the bigger risk of a leak lay—from the British services or from this room—I would point inside. Indeed, the British never leaked the secret.

A week later, Dagan warned again of a possible leak, this time from U.S. sources, suggesting we accelerate a decision. I wondered who the possible leaker was but did not get an answer. I said that if I had more information about the publication where the story was going to be published I could have a better sense where the leak was coming from. No further information came, and again, no leak was published.

Towards the end of August, preparations for the two low signature plans were almost completed with “Lean Shkedi” fully finalized and the “Yadlin plan” almost there. Provisions for war were at an advanced stage. All in all, this signified a miss of my original target date of September 1 by at most a few days. The prime minister and I then discussed the legal aspects of such a decision with the attorney general and made sure that the cabinet was authorized to make the general decision to destroy the reactor and delegate the choice of the concrete plan and the timing of its execution to the prime minister, myself, and Minister of Foreign Affairs Tzipi Livni. This was needed since the dramatic nature of such a decision made it prone to be leaked in short time. The intention was to remain vague regarding the exact way and timing and then, immediately after the cabinet’s decision, to meet briefly, one by one, with the heads of Mossad, Intelligence, and the IDF chief of staff to hear their recommendations and immediately make the final, formal decision to execute it the same night.

Around September 1, we agreed to convene the cabinet on September 5 for a final decision. The previous evening, following consultation with Chief of Staff Gabi Ashkenazy, I informed Olmert that my recommendation the next day would be to use “Lean Shkedi” as the preferred option. Both plans were viable and ready, but the “Yadlin plan,” which I liked very much, was clearly less likely to gain consensus. It was clear to me that Olmert thought the same. The cabinet meeting began around noon and ended in late afternoon-early evening. It was a relatively focused and short meeting. The participants were acquainted with the different options and somewhat relieved by the absence of the usual tension in the air. In a way, it was simply a ritual, however important, with results known in advance. When the vote came, only Dichter, a former head of the Secret Service (Shabak), abstained.

Immediately after the ministers departed, we continued the process as agreed upon in advance. Olmert, myself, and Livni remained in the room and decided to execute “Lean Shkedi” that very night, meaning that airplanes had to take off in a few hours. The chief of staff and his people, as well as the many hundreds in IAF squadrons and in Intelligence who participated in the operation knew in advance to be ready to execute it that very night. Many thousands in other parts of the armed forces without concrete knowledge of what was happening felt the uniqueness of that evening.

Before leaving my Tel Aviv residence on the 31^st floor of a high-rise building, I looked outside into the city’s fading night view and suddenly realized that I was watching the F-15s or F-16s flying near my home towards the reactor, some hundred feet underneath my window. Two hours later, from the underground IAF headquarters, we watched the operation. It went as smoothly as we could hope. I could not stop thinking of the violent confrontation that might have ensued under the “Fat Shkedi” alternative. What a difference. After more than an hour-and-a-half, the fighter planes pulled up and released their munitions. A minute or so later came the report: “Accept: Arizona.” The reactor had been successfully destroyed. A partial sigh of relief.

More than an hour later, all planes landed safely. According to foreign reports, Bashar Assad was informed of the reactor’s destruction sometime soon after the event. It took another twelve hours to confirm our intelligence assessment that a low signature plan would give the Syrian president sufficient leeway to keep the attack under wraps by avoiding any clash with us. The fact that the circle of people in Syria who knew about the attack was extremely narrow also helped. A great mission accomplished.

The operation succeeded because despite all our disagreements, we had a strong unity of purpose that brought us all together: The devoted Mossad operatives under Dagan’s extremely creative leadership who brought the original proof of the reactor’s existence; the IDF’s intelligence analysts and operational units under Yadlin, a gifted and effective leader; Chief of Staff Ashkenazy, the tireless IDF commander who coordinated the military activities between the IAF and prepared the army for a potential full scale war; Ministry of Defense Director Gen. (res.) Pinhas Buchris, a most capable out-of-the-box thinker, who together with his dedicated subordinates safeguarded the massive supply line needed for the worst-case scenario of war; and, of course, IAF Commander Shkedi, a great commander and great man, with his top teams of pilots and aircrews who did the actual planning, preparation, training, and execution of the operation. Certain credit is due to our internal teams—generals Herzog and Dangot in my office, Turbovitz and Turgeman in the prime minister’s office. A special credit must also be given to Prime Minister Olmert who bore the supreme responsibility from day one, never lost sight of the need to destroy the reactor, and took upon himself the burden of the most sensitive contacts with the U.S. government on all levels. I genuinely salute them all, without giving up any of the above criticism. This is the way operational capabilities and national standards of confronting challenges are created: by combining mutual respect with honest, critical, and at times painful discussions of what really happened and what has to be corrected.

It is in this respect that I asked myself time and again what created and fed the continued emotional tension around this operation. I assume that other participants have their own views on the same questions. I can hardly fault Olmert for writing, while in prison, a personal, bitter autobiography on his entire life (including the story of the reactor’s destruction). Others, myself included, might also have certain personal biases but it is important not to try deliberately to mislead any future student of the case.

Looking back on my entry into the picture, my first memory is of the intense, emotionally loaded, impatient drive for immediate action shared by most people in the room, reinforced by genuine concern lest the operation be torpedoed by a premature leak. Was this due to the overriding influence of PM Olmert on the one hand, and the relative passivity of the outgoing defense minister, Amir Peretz (who had no operational experience), on the other? Or was it due to being on the verge of execution after intensive months of planning, only to be interrupted all of a sudden by a new minister who started raising profound questions, as if the whole planning process had to be restarted all over again? Having commanded this group’s members for many years (during my IDF service and as PM and defense minister), I knew all participants in the room much better than both the prime minister or the outgoing minister of defense, which might be somewhat frustrating for them. Whatever the reasons for the tension, I considered preparation of the two low-signature plans an absolute necessity and ensured that they be pushed all the way to successful execution on September 6, 2007.

There was, however, another side to the ledger. Upon assuming my post, I noticed a certain disturbing similarity between what had unfolded in this project during the past three months and the way that the second Lebanon war had been opened and conducted.

A year earlier, on July 12, 2006, an Israeli patrol along the Lebanese border had been attacked with two soldiers killed and two abducted. A short time afterward, a tank that crossed the border in search of the missing soldiers stumbled on a big explosive charge and another five soldiers were killed. It was clear to all that this aggression called for a tough response, yet one that needed to be made in a calm and cool-headed fashion: What were the operation’s goals and how to achieve them? What was the “exit strategy”? What would Hezbollah’s likely response be, and what were the broader implications of our chosen options? That is what professional standards would dictate.

But nothing of the sort actually happened. Instead, within a few hours, an IAF contingency plan for the destruction of all known Hezbollah missiles and rockets was grabbed from the shelf—where it had been for the preceding six years—and executed the next morning. Dubbed “Specific Weight” and prepared as a surprise opening gambit of an all-out war with Hezbollah, the plan was completed in 2001 and was continuously updated and checked in training exercise every two years or so. Neither I nor Prime Minister Ariel Sharon after me used it even when soldiers were abducted along the border or terror attacks from Lebanon killed several Israelis. Rather, we kept it for the event of a full-fledged war. Yet, on July 13, 2006, Israel found itself in a war that the government did not plan, did not want, and did not prepare for. In fact, when the cabinet met that evening and made the decisions for the next morning, including the execution of “Specific Weight,” it did not know it was initiating a war. Thus, reservists, who constitute the main fighting body of the ground forces, were not mobilized and an emergency situation was not announced. The economy was not put on war footing, and objectives for the war/campaign were not defined. That is not the way for a war to be started, and Israel paid the price for this rush decision in the ensuing thirty-four days. When, after the war, I asked one government minister and most involved generals what happened, how did this lapse of judgment come to pass, the most common answer was,

One of them added: History never repeats itself, and the two cases differ on many levels. But there is a strong similarity in the hasty adoption of a contingency plan that was prepared for something else, together with a hyperactive plan (“Fat Shkedi”), and rush to execute it without an orderly process of considered examination of all things in advance. Systematic thinking and analysis should precede action. Not follow it.

I was cautious not to explicitly express this observation until at the somewhat panicky cabinet meeting on the first week of August, when Deputy PM Mofaz, a former minister of defense and IDF chief of staff, suddenly erupted: “Are you crazy? Is it a replay of the last Lebanon war?”

His question remained unanswered. I can understand the huge pressures that probably influenced Olmert’s judgement. Not only did he face the burden of leading the nascent operation, but he had simultaneously to handle a criminal investigation that was to cloud the rest of his premiership (and eventually force him out of office) and an official commission of inquiry of the Lebanon war, headed by Supreme Court Justice Eliyahu Winograd. In these pressuring circumstances, his inner circle fabricated the charge that “Barak was postponing the reactor’s destruction in anticipation of his downfall.” Complete nonsense. There were too many weighty reasons for my actions, and reality proved them to be fully vindicated.

---

[1] Ori Wertman, “When Israel Destroyed Syria’s Nuclear Reactor: The Inside Story,” Middle East Quarterly, Spring 2022.