Last year, Iran’s chief nuclear scientist was killed while traveling by car in the town of Absard, not far from Tehran. Neither his wife nor the twelve bodyguards traveling with him were harmed. The operation, according to Jacob Wallis Simons of London’s Jewish Chronicle, was orchestrated by the Mossad using a team of twenty people and a one-ton remote-controlled machine gun. While the details were impressive, no one was very surprised by the idea that Israel might track down and kill individuals who threaten its safety. The most famous instance, perhaps, was the mission to punish the perpetrators of the Munich Olympics massacre (the subject of Steven Spielberg’s movie Munich).

One of the authors of the New York Times’s detailed account of the Iranian scientist’s assassination was Ronen Bergman, an Israeli journalist who has been covering his country’s clandestine operations for many years. His book on the subject, Rise and Kill First: The Secret History of Israel’s Targeted Assassinations, appeared in 2018. From the moment this book was released, I wanted to read it from cover to cover, including the 78 pages of detailed endnotes. Like many Israelis, I am eager to learn more about the storied episodes in the history of my country’s intelligence services. As a philosopher who has written extensively on the ethics of warfare, including targeted killings—and as an occasional advisor to the IDF and co-author of its 1994 code of ethics—I have a professional interest in the subject matter.

I was also drawn to the book by its author, who has distinguished himself with the thoroughness, fairness, and honesty of his reporting, and his strict adherence to journalistic standards. Unlike so many others in his field, he has assiduously avoided the sort of dependency on his sources that can undermine objectivity. He also benefits from possessing two advanced degrees from Cambridge University, which lend an academic rigor to his reporting. What I found was engrossing, but disappointing.

In its 35 chapters, the book provides dozens of detailed accounts of instances in which the Jewish state took covert action to preempt its enemies. It begins with the uncovering and destruction of a nuclear reactor being built in Syria. In one of the intelligence-gathering missions involved, Bergman tells us, a Mossad agent chatted with the head of the Syrian nuclear agency at a hotel bar in Vienna, while others entered his room, opened the safe, photographed its contents, and fled just half a minute before he returned.

The deciphered images revealed a significant, obvious danger to the security of the state of Israel and its citizens. After an extensive exchange of information with the Americans, who were unwilling to act, then-Prime Minister Ehud Olmert (who appears to have spoken with Bergman about the decision) gave the order to destroy the reactor. On the morning of September 6, 2007, Operation Orchard commenced, and dozens of planes took off from the Ramat David air-force base and flew westward in what appeared—even to Syrian intelligence—to be a routine exercise. All but seven of the planes then turned to the south. The remaining seven headed north and, at a distance of 30 miles from the reactor site, launched 22 missiles at three targets within the compound, putting a permanent end to Assad’s nuclear aspirations.

Rise and Kill First is not concerned solely with the recent past, and describes incidents throughout Israeli history. Early in the book we learn of the shooting down of a military plane carrying senior Egyptian officers from Damascus to Cairo in October 1956, on the eve of the Suez Crisis. Many of us in Israel grew up with the tale of this feat, known as Operation Rooster, which, as everyone knew, contributed significantly to Israel’s victory by eliminating the entire Egyptian general staff before the fighting had even begun. Bergman tells the story vividly, incorporating conversations in real time between then-commander of the air force Dan Tolkovsky and the pilot flying the mission.

The account contains various noteworthy details, such as the fact that Israeli Air Force planes flew over Turkey during the mission, which was not known to me and is significant for experts in the subject. The most important revelation comes in an endnote, which states that it is not known who was in the plane that was shot down, or what impact the interception had on the functioning of the Egyptian general staff. In this way, Bergman sheds new light not only on events we were not familiar with, but also on episodes that we thought we knew well.

Taken as a whole, Rise and Kill First is gripping, clearly written, and filled with the information that will fascinate both uninitiated readers and those well-versed in Israeli history. But to evaluate it properly, it’s necessary to ask three sets of questions. First, is the information in it accurate? Second, how did the author manage to get information about highly classified operations from people whose job it is to keep secrets, and did these informants, or Bergman himself, endanger Israel by publishing it? Lastly, is Bergman’s account of the moral and prudential concerns involved in targeted killings persuasive? After trying to answer these questions, I will provide my own evaluation of some of the topics covered in this book.

Accuracy

From the aforementioned 78 pages of endnotes, it is clear that Bergman’s sources are mostly unclassified publications and the testimonies of scholars and other experts. But he also draws on an impressive array of human sources and documents not available to the general public. The bibliography enumerates several hundred individuals whom the author interviewed for the book (myself included) as well as another “350 interviewees who cannot be named,” noting that “the initials or code names of 163 of them appear in the endnotes.”

I have no reason to doubt the reliability of these unnamed figures, or Bergman’s account of additional secret sources. That is to say, if Bergman states fact X and attributes it to source Y, I am confident that he is accurately reporting what Y said. Yet there are some telling absences in Berman’s references. For instance: no source is given for the claim that Israeli planes flew over Turkey during the raid on the Syrian nuclear reactor. Even citing an anonymous source would have lent more credibility to such new pieces of information.

While reading the book, I wrote a list of similarly interesting or important pieces of information that lacked notation of any supporting source at all, even an anonymous one. I counted 25, although if I broadened my definitions of “interesting” or “important,” my list would contain many more. Here are two examples.

In 1997, Mossad agents attempted to kill Khaled Mashal, one of the leaders of Hamas, in an operation widely reported in the media. Their plan was to inject him with a toxic substance that would kill any person who came into contact with it without leaving a trace. According to Bergman, the poison used, levofentanyl, was selected in a process of consultation between the technology unit of the Mossad and the Israel Institute for Biological Research, a government agency based in Ness Ziona. But no supporting reference is provided for this last detail.

Or take the following claim, also made without giving a source, about Tamir Pardo, who was the head of the Mossad from 2011 until 2016:

Like his predecessors, Pardo refrained from risking Israeli operatives in killings undertaken in target countries, particularly in places as dangerous as Tehran. All the hits on Iranian soil were, in fact, implemented by members of that country’s underground opposition movements and/or members of the Kurdish, Baluchi, and Azerbaijani ethnic minorities who were hostile to the regime.

Notably, Pardo is not mentioned in the list of people the author interviewed in preparing the book.

Bergman’s failure to provide any evidence, or even anonymous sources, for these and similar assertions strewn throughout the book considerably erode the reader’s readiness to trust the work as a whole. My overall confidence in Bergman was not sufficient to allay these concerns about the book’s veracity.

Secrecy

While Pardo was not interviewed, at least not on the record, he did provide a blurb, which is among several printed in the book’s opening pages. After calling it “the most impressive book I have seen on the subject and the first one written using real research rather than fictional narrative,” the former Mossad chief adds: “Bergman’s ability to reach sources inside Western intelligence communities is amazing (and, I must say, also a little disturbing).”

That parenthetical comment brings us to our second topic, which is whether a book like this poses a danger to Israel’s security by revealing what is best left concealed. Intelligence and clandestine agencies go to extremes to preserve the secrecy of their operations, and especially their sources and methods. The Mossad expressly prohibited its operatives from being interviewed by Bergman, although, as he himself testifies, it only issued this directive after he had already spoken with many of them.

Similarly, in the book’s first endnote Bergman relates the response he received from one of the Mossad’s official historians, when he requested an interview: “Even if I were the last person in the intelligence establishment who [had] not yet made the pilgrimage to you, I would by no means cooperate with you. I despise whoever it was who gave you my phone number, just as I despise you.”

Journalists often roll their eyes at such tightlipped responses, and invoke “the public’s right to know,” but secrecy is vital for intelligence and clandestine operations. To the simple question: does this publication harm national security by exposing or compromising intelligence operations, it seems to me—based on my extensive experience with these matters—that the answer is “no.” No particular detail in this book raised a red flag for me.

The deeper question is whether researchers from the Iranian or Syrian intelligence agencies—or even from some terrorist organization—might be able to learn something from this book that it would be better for them not to know. I do not know the answer to that question.

But this brings us to a third question, which the author seeks to answer in the book’s prologue: why were so many individuals working for security agencies prepared to tell him about secret operations? He concludes, and I think correctly, that they want their own perspective to make it into the history books. That they gave in to this understandable desire, in my mind, reflects poorly on them, and on the Mossad’s organizational culture.

Ethics

Still, even if every detail in Rise and Kill First were accurate, and even if there were no reason to worry that it betrayed too much, it would still suffer from serious problems. These begin with the title, “Rise and Kill First,” a reference to a teaching found in the Talmud and halakhic literature: “If someone comes to kill you, rise and kill him first.” Traditional commentary on this statement limits the license to “kill him” to situations in which an individual has no other alternative in defending himself.

Without the first half of the quotation, and taken out of context, “rise and kill first” seems more like a celebration of violence, or an expression of prideful self-satisfaction in Israel’s power, than an ethical doctrine of self-defense. Extracting this phrase from its full context reflects, perhaps unintentionally, a flawed approach that permeates the entire book, as I will explain presently.

The book’s subtitle, which promises a history of “targeted assassinations,” fares little better under scrutiny. While the more usual “targeted killings” would have worked just as well, the word “assassination” conveys an ethical judgment. Fifteen years ago, I authored an article with Amos Yadlin, the former head of the IDF’s military intelligence, about the term “assassinations,” and we concluded that it has no uniform definition, despite its use in official documents. Thus U.S. Executive Order 12333—which forbids federal agencies from carrying them out—never specifies what constitutes an assassination. But we did detect a pattern: the use of the word generally carries a negative connotation.

Seeking a neutral definition, Yadlin and I arrived at the following:

An assassination is an act of killing a prominent person, selectively, intentionally, and for political or religious purposes.

Intent is a necessary, but not sufficient, component of this definition, as is the exclusion of killings that are military, rather than political or religious, in nature. Thus, the U.S. Navy SEALs’ killing of Osama bin Laden in 2011 would not constitute an “assassination” since he was killed as part of an extended war, for the purpose of national self-defense. Lee Harvey Oswald shooting John F. Kennedy, however, would qualify.

Bergman instead gives us a twofold definition of assassination: first there is “killing a specific individual in order to achieve a specific goal—saving the lives of people the target intends to kill, averting a dangerous act that he is about to perpetrate” and second, “removing a leader in order to change the course of history.” Since acts in the first category shouldn’t be considered assassinations so much as military operations, justifiable at least in theory under the laws of armed conflict, this definition only serves to cast moral aspersions on all of Israel’s targeted killings.

As for the second category, which refers to acts more properly considered assassinations, it too is highly imprecise. A leader is not assassinated in order to “change the course of history,” but, rather, to bring an end to a policy for which that figure is responsible, so that the target will be replaced by a leader who will implement a different one. The man who murdered Yitzḥak Rabin did not want to alter the trajectory of history, but to interrupt the peace process which the Rabin government had led and replace it with unambiguous Israeli rule over the so-called Greater Land of Israel. Of course, doing so would, in a sense “change the course of history,” but this sort of grandiose and abstract wording obscures the concrete and specific goals that so often motivate assassinations.

Moreover, assassination of a political figure carries a special status only if a clear distinction exists, both in principle and in practice, between the activities of the political leader and the activities of the armed forces loyal to him—such as an army, militia, or terrorist organization. In Israel’s environs, enemy leaders do not maintain this distinction, in principle or in practice. Typically, they are the driving force behind the planning and execution of military or terrorist operations against Israel, its citizens, and its residents. The danger emanates from their practical decisions, which are carried out under their supervision and by their subordinates. Assassination of a leader whose actions pose a substantial and immediate danger is, therefore, an act of self-defense, not different in kind from the targeted killings of military figures.

Not all political figures reside in the chain of command. For instance, a political leader may express a policy hostile to the state without advancing that policy through actions that would endanger it. With respect to such a leader, one may employ the distinction accepted in the doctrine of just warfare between decisions made by the political echelons and those made at military level.

Even when a state regards a leader as posing a substantial and immediate danger, it is still obligated to deliberate about the moral aspects of a targeted killing. Is the danger posed by the target great enough to justify the operation? Can it be carried out without causing unacceptable collateral damage? What about the risks to the operatives who will do the killing, and the intelligence sources who make it possible? Will the target simply be replaced by someone else who will act in the same manner? What about the other short- and long-term consequences? Are there alternatives to assassination?

Striking an enemy leader can be of great value with respect to deterrence. However, one must always first consider the moral question of necessity: is there another way? Or is assassination the only remaining method of self-defense? For instance, I do not know whether Israel mulled the possibility of assassinating Yasir Arafat. Under such circumstances, one consideration would have been whether it was possible to neutralize the danger he posed without assassinating him—for example, through meaningful negotiations.

Bergman’s book does not dwell on these complexities, and that brings us to the deepest problems of Rise and Kill First, which come not from what the book contains but from what it lacks.

Process

First, there is no description of the process by which Israel decides to carry out an “assassination.” Prior to reading the book, I sketched out for myself an idealized version of this process, and came up with fifteen discrete steps, beginning with identifying the danger posed by the target, and ending with assessments by security officials of likely enemy responses over the short and long term. These steps include both practical questions—e.g., can the target be located and killed—and ethical and legal ones about the operation’s justification. They also involve both discussion among military and intelligence officers and consultation with civilian officials. Bergman says nothing about this process, and largely ignores the deliberations that lay behind every operation he chronicles.

And that’s not all he ignores. The final sentence of the book’s prologue states that “it is appropriate . . . to study the high moral price that has been paid, and still is being paid, for the use” of Israel’s impressive operational abilities. As a philosopher specializing in military ethics, I couldn’t agree more. But Bergman does not fulfill this promise. Indeed, he offers no serious and thorough discussion of the moral, ethical, legal, or even prudential aspects of targeted killings, although the book touches on some of these issues in passing. In other words, by using the term “assassination,” and speaking of a “high moral price,” he sets the reader up to believe that there is something scandalous or condemnable about the ways Israel defends itself. In fact, Bergman offers no critical argument to justify the moral insinuation that many readers will take to be the book’s core contention.

Even when the book comes close to a point at which it would be natural to pause to analyze ethical considerations, it swiftly moves on. For example, it refers to the aforementioned Executive Order 12333, issued by Gerald Ford and prohibiting U.S. agencies from participating in assassinations, to explain why the U.S refused to participate in killing a particular terrorist leader, while willing to accept Israel’s decision to do so—without providing any historical or political context.

Since Bergman neglects these questions, I will try to provide some answers. Following his own classification, let’s begin with what he calls “killing a specific individual in order to achieve a specific goal—saving the lives of people the target intends to kill, averting a dangerous act that he is about to perpetrate.” What we have here is a characterization of action taken in clear self-defense, which is universally considered legal. And what is the “moral price” paid for the involvement of young people in operations of this kind, beyond the moral price paid by any people participating in deadly military operations?

To answer the moral question, it is helpful to examine it from three distinct angles: (1) The nature of the danger: is it immediate or distant? Is it substantial or merely suspected? (2) The necessity of the operation: could the danger be averted some other way, e.g., by capturing rather than killing the target? (3) Collateral damage: is the action against this person liable to result in injury to other people who are not dangerous since they are not involved either in his activities or other hostile activities? If such is the case, would the action taken against the dangerous individual be considered proportional? That is, how do the advantages that stem from killing the target weigh against the collateral damage in doing so?

Any coherent evaluation of the ethics of a particular targeted killing must take these three factors into careful consideration. In describing certain episodes, Bergman refers to such parameters, but he never addresses or describes them in any appropriately consistent, thorough, or detailed fashion. As a result, the non-expert reader comes away from each chapter having formed moral, ethical, and legal judgments based on prejudice or shallow commonplace assumptions, rather than substantive evaluation.

What would an evaluation based on these parameters look like? Let’s begin with the first. A straightforward example would be the case of a suicide bomber en route to a kindergarten. The danger is serious and immediate, and if a sniper’s bullet can end the terrorist’s life without detonating the bomb, the case is clear cut.

But immediacy becomes more complex in the case of a reactor intended to produce nuclear weapons for a patently hostile state. Since it could be years before those weapons are produced, the danger is not immediate in the conventional sense. But this is a superficial and inadequate understanding of immediacy. Once a nuclear reactor becomes active, destroying it would produce tremendous collateral damage. A precision airstrike on a not-yet-active reactor, by contrast, could in theory be carried out without any casualties, and without danger of nuclear fallout and so forth. In the Syrian case, if Israel did not act when it did, it would soon have become too late to eliminate the reactor without causing much harm.

The importance of the second parameter, the nature of the operation, is evident by looking at the time during which Israel carried out frequent targeted killings in the Gaza Strip, but none in the West Bank territories under the control of the Palestinian Authority. In every case in which a substantial and immediate danger to the citizens and residents of the state is uncovered, it must be asked whether it is possible to capture the source of this danger in a way that would neutralize it. If it is indeed possible to do so without endangering a large number of soldiers or Shin Bet operatives, then capture is unequivocally better than killing. This is so not only for evident moral and legal reasons, but also because a captured enemy can be a valuable source of intelligence, which can be used to protect against future dangers.

If a terrorist hides in the Gaza Strip, the entry of Israeli military forces to capture him would involve a complicated operation whose success is far from assured, given the difficulties posed by the urban landscape there. Significant losses would be anticipated both among the Israeli military force and among the target’s noncombatant neighbors. In such circumstances, carrying out a targeted killing would lead to fewer casualties, and so is undoubtedly preferable for moral, legal, and practical reasons.

The third parameter, collateral damage, is the most complex. In the laws of armed conflict, the well-established guiding principle is that of proportionality, which demands that a commander planning an operation weigh the military benefits of its success with the harm liable to result from it—especially with respect to the lives of people not involved in the fighting but located in proximity to where it is taking place.

This oft-misunderstood principle is applicable when there is fear that people uninvolved in fighting may be harmed, and outlines the conditions under which those causing this harm are justified. That is to say, the fact that noncombatants are, or will likely be, harmed does not, in itself, constitute evidence of prohibited conduct. Under no circumstances is it permissible to harm people uninvolved in fighting intentionally and in a premeditated fashion. However, such “collateral damage” is sometimes permitted if it is a necessary outcome of harming those who are involved in the fighting, in a situation in which the collateral damage is proportionate to the military objectives.

Contrary to what many assume, proportionality does not require that the number of killed on one side of a conflict be similar to the number killed on the other. Nor does it mean that, say, if the enemy attacks your civilians with Molotov cocktails, you must respond only with Molotov cocktails. As a little reflection should make clear, such principles would be absurd. Internationally accepted laws of war, rather, would forbid as disproportionate, say, flattening an entire village because of two or three snipers located therein. But they might deem proportionate destroying an apartment building from which rockets are being fired at civilians.

In a separate article that I wrote together with Amos Yadlin, we outlined the unique contours of battle against terrorists operating among people who are not involved in terrorist activity. Prior to its publication, our article was presented to commanders in the IDF and the Shin Bet, after which it was published in the Journal of Military Ethics, alongside responses from experts in military ethics from universities and military colleges in the United States and a university in Beirut. Almost all of our conclusions are accepted by the military agencies of democratic countries, and by many scholars. There is just one point that aroused broad controversy, which I will lay out now in brief.

Deliberations about proportionality place the anticipated military benefit of the planned operation, including protection of the lives of combat soldiers and their ability to carry out their mission, on one side of the scale, and on the other side the lives of noncombatants who have the misfortune of being terrorists’ neighbors. Our claim, then and now, is that the state is permitted to give precedence to the protection of the lives of its own soldiers over the prevention of harm to the life of a terrorist’s neighbor. This precedence arises out of the obligations of the state toward its soldiers, whose function is to protect the citizens and residents under their care, as opposed to the obligations of the state toward the terrorist’s neighbors. States tend not to acknowledge that they act in accordance with this order of precedence, but they usually do.

What does Bergman have to say about this very thorny problem? Not much, apparently, since his book does not offer sustained discussions of military ethics or the laws of warfare that have been conducted in Israel and influenced events related to targeted killings. Even if the book had remained strictly neutral, I would view its relative neglect of these discussions as a very serious deficiency. But it is much worse than neutral because it assumes answers to the very moral questions it fails to ask.

In the final analysis, Ronen Bergman’s book is fascinating in the facts it details; it is impressive in its sources; it is somewhat deficient in referencing dependable sources for factual claims; and it is weak in presenting the moral, military-ethics, and international-law background to the operations it depicts. A disapproving undertone can be detected throughout the book, but this undertone is never made explicit, or given adequate justification.

A reader interested in acquiring a deeper understanding of Israeli actions in the realm of targeted killings will thus not be able to make do with this book alone. Rise and Kill First paints Israel as morally deficient through implication. Without resorting to the falsehoods, half-truths, and moral inversions that have established the Jewish state in the eyes of much of the world as an evil enterprise that runs roughshod over international law, Bergman nudges his readers in this direction, and lets them draw their own unsympathetic conclusions.