The Cycle of Violence

Is the Israeli-Palestinian conflict a case of "an eye for an eye"?

Have you heard of the Jewish Avengers?

They’re not comic book heroes, I assure you. The Nakam (נקם, “revenge”; also known as Dam Yisrael Noter, “the blood of Israel avenges”) were a very real, very committed group of about fifty Holocaust survivors who wanted to implement “an eye for an eye” at the scale of “a nation for a nation”.

The group’s leader, poet and writer Abba Kovner—seen below, testifying at the trial of Adolf Eichmann—was a Holocaust survivor who had been held in the Vilna Ghetto, attempted to orchestrate an uprising against the Nazis, and failed before fleeing into the nearby forest and finding and joining Soviet partisans so he could fight back.

Abba Kovner testifying at the trial of Adolf Eichmann, May 4th, 1961

After the war came to a close, Kovner was justifiably upset. His family, friends, and community had been exterminated and Jews were turned into an endangered species across parts of Europe where they had resided for more than a thousand years. After the war, he toured around Europe and witnessed this destruction firsthand.

Following visits to the site of the Paneriai massacre and the Majdanek concentration camp, his blood had boiled over. He believed that the Holocaust could happen again if nothing was done. But what was he to do?

His answer was Biblical, and specifically, Exodus 21:23–27. His interpretation was not by Midrash either—it was one obtained through Peshat: to ensure the world knew Jews could not be attacked again, he would take six million German lives.

He set himself to work recruiting other Holocaust survivors who he believed could prove to be reliable, discrete, motivated, and stable co-conspirators. Armed with the bloodlust of his compatriots, he formulated two plans of action:

• Plan A: Infiltrate the water supply of Nuremberg and poison it to kill the largest possible number of Germans.

• Plan B: Scale down and focus on poisoning thousands of former officers of the Schutzstaffel.

Nuremberg was selected for its position as a Nazi Party stronghold before and during the second world war. Once they were in the city, Nakam found that organizing the effort was difficult because the bombing of the city had left few rooms to rent. Nonetheless, they made progress, and through the judicious use of bribes, they managed to install one Willek Schwerzreich at the municipal water company.

Willek managed to secure the details about the city’s water system and tabulated a list of places where poison could be introduced to cause the largest possible number of deaths. The plan was going swimmingly, and now all they had to do was obtain the required poison.

This task fell to Kovner, who decided he would obtain it from Yishuv. Kovner obtained passage to Mandatory Palestine disguised as a Jewish member of the British armed forces, and when he arrived he was interrogated by the Mossad LeAliyah Bet member Shaul Meirov, a man who would go on to become deputy defense minister for David Ben-Gurion in the Arab-Israeli War. After three days, Kovner was released and he sought out Moshe Sneh and Yisrael Galili to try to convince them to give him the poison required to implement his plan.

His request fell on deaf ears. The pair of Haganah chiefs had neither the desire nor the resources for his revenge plot, so Kovner was forced to move on to Plan B.

For this plan, Kovner still needed poison, but because of the delivery means and scale, this poison would be easier to manufacture. He informed European Nakam cells that they should hire a chemist named Yitzhak Ratner—a fellow Vilna Ghetto insurgent—to produce the poison, and because of the need for swift action, he also attempted to obtain it in Israel at the same time. Per Kovner’s account, the person who would end up helping him to obtain the poison was future president of Israel and then president of the World Zionist Organization, Chaim Weizmann.

Weizmann put Kovner into contact with Ephraim and Aharon Katzir, who were then working at the Hebrew University of Jerusalem. The brothers supported Kovner’s plan and used university resources to give him the poison he requested.

Poison in hand, Kovner proceeded to make his way back to Europe via Alexandria, Egypt. But, the forged papers Kovner was using stood out and he was promptly arrested after boarding a ship to France. Ironically, Kovner’s arrest didn’t seem to have been a result of his involvement with Nakam. Later, members of Nakam claimed Kovner was arrested after his plans were betrayed to the British authorities by Haganah members, but according to historian Dina Porat, the real reason seems to have been that Kovner was suspected of helping people to illegally immigrate to Mandatory Palestine. After a two-month stint in jail, Kovner abandoned his quest for revenge.

Kovner asked Nakam cells to disband, and most did, but by this point, the actions of the group’s remaining members were out of Kovner’s control. The Nuremberg cell was committed to poisoning SS officers and they wouldn’t be stopped.

Yitzhak Ratner set up a laboratory to develop a tasteless, odorless poison without immediate effects, and he ended up with an arsenic-based glue that he found was excellent for killing cats. After he had synthesized enough of the compound, it was time for the group members employed undercover at the Langwasser internment camp to go to work; or, more specifically, at the camp’s bread supplier, the Konsum-Genossenschaftsbäckerei.

The group members discovered that each Sunday, the German prisoners of the American-manned camp would eat black bread, and the Americans would eat white bread. Knowing this, they sought to execute their plan by poisoning all the black bread on a Saturday night. Unfortunately, on the Saturday they wanted to enact their plan, the bakery workers had gone on strike! As a result, the Nakam cell was only able to get around 3,000 loaves of bread painted instead of their original 14,000 target.

Ten days later, the New York Times reported that 2,283 German POWs were poisoned and 207 were hospitalized with serious illness. Ultimately, they all survived. Although poison was enough to kill cats, the amounts on the loaves weren’t large enough to have the same effect in humans. This is somewhat shocking, as a memo uncovered years later noted that Nakam had smuggled enough arsenic into the bakery to kill 60,000 people!

To this day, it remains a mystery as to why the poison failed to kill Nazis. The prevailing theory is that the plotters in their haste spread the poison too thinly. Another is that the Nazi prisoners immediately sensed something was off with the bread and therefore no one ingested enough of it to die.

After this incident, the Jewish Avengers were practically no more. The group disbanded to little fanfare and most of the members ended up in Israel, somehow or another. Nakam had failed in its mission and many members and collaborators seamlessly transitioned to the next major Jewish project: establishing the state of Israel. This transition wouldn’t have been possible without the work of others.

During and after Kovner’s Israel trip, the man who led the group in Europe was one Pasha Reichman. Kovner and Reichman should have rightly been considered extremists. They wanted to commit mass murder, and if they could have done it, they would have. But they ultimately did not.

The sentiment they were acting on was common enough at the time, but few people ended up joining Nakam or other groups with similar intent. However, the desire for immense retaliation is, in fact, commonplace. Today, you only need to open up Twitter to see it. This desire just rarely manifests as much of anything. People simply have too much else to do, they don’t want their daily lives consumed by violence, they aren’t that committed, extreme feeling is too fleeting and abstracted away from the real world, and some of the people who feel like getting revenge are wise enough to oppose the feeling philosophically.

One of the men opposed to revenge was David Ben-Gurion, the man who would become Israel’s first prime minister. Ben-Gurion and Reichman had the opportunity to meet in a displaced person camp in Germany and, while he was there, Reichman informed Ben-Gurion about his plans. It seemed Reichman hoped Ben-Gurion felt like he did, and that he wanted some sort of retaliatory action against the Germans. But he didn’t. Ben-Gurion informed Reichman that he wasn’t seeking revenge; instead, Ben-Gurion sought to make something: he wanted to establish the nation of Israel.

The strategy Ben-Gurion pursued was peaceful: he wanted to avoid the much-discussed “cycle of violence” presumed to explain what happens in Israel today. After all, they say an eye for an eye makes the whole world blind.

---

Is That It?

The cycle of violence is perhaps the most well-known metaphor for what’s happening in Israel and Palestine. If it is a strictly true description of the Israeli-Palestinian conflict, then the solution is simple: lay down arms! As they say, “big if true”. The immediate implication is that the conflict will end and peace can be achieved. People simply have to stop seeking revenge.

The trueness of the metaphor also makes it possible to assign blame: it belongs to whoever committed the first violent act. But this is not as simple as it sounds, because it’s debatable which act constitutes the first instance of violence in the cycle.

Starting From First Principles

We have knowledge of the first Muslim-Jewish conflicts. The prophet Muhammad encountered Jews when he fled Mecca with his followers towards the town of Yathrib, three days’ ride away.

Among the Jewish inhabitants of the area were numerous tribes including, most prominently, the Qurayzah, but the Qaynuqa’ and Nadir were also well-known. Like the other inhabitants of the region, the Jews in Yathrib were largely date farmers, though also engaging in numerous other occupations, skilled and unskilled. They were considerable participants in the region’s trade since Yathrib had ample water and vegetation to support people on their travels.

Muhammad arrived at Yathrib in the midst of a conflict between the Arab pagan Aws and Khazraj who had moved into Yathrib some years earlier as clients of the town’s Jewish clans. What happened next depends on who tells the story. Muslim readers likely already know what Yathrib is called today, and all literate Muslims know it after hearing that it was the destination of the Hijrah. Yathrib is Medina.

Knowing precisely what happened next is difficult because the majority of the people who chronicled what Muhammad did have been Muslims. As such, his chroniclers have naturally been biased towards depicting him as the perfect man he is believed to be. But all accounts recognize that Muhammad was a merely religious leader before the Hijrah and in Yathrib his role expanded, transforming him into a truly political leader. The warring tribes accepted the prophetic claims of the small number of Muslims who turned up in Yathrib and they converted en masse, giving form to a movement by a true prophet figure.

The importance of clan bloodlines, the Battle of Badr, the conflict with the authorities of the Ansar, the creation of the calendar dating from the Hijrah, the conduct promoted by the Constitution of Medina—there’s so much that could be discussed about Muhammad’s early ventures and, indeed, his entire life, but that’s beyond the scope of this piece. What we do need to recognize is that Muhammad was a clever statesman. He sought to marginalize all opponents to his rule and he did so with apparent ease. Just one year after the Battle of Badr, he began to expel the Jewish clans from Yathrib. Between then and 627, it’s not clear which Jewish clans were expelled but we know that many Jews were forced out.

Could this have been because of something the Jews did to Muhammad? Well, on his arrival, it seemed Muhammad believed he could turn the Jews with ease. They did share a monotheistic vision and history based on the Hebrew Bible, with only minor alterations on Muhammad’s part, after all. But as the chroniclers have attested, the answer is ‘no.’

Jews had no desire to convert to Islam. For Rabbinic Jews, prophecy ended when the temple fell and, as we know, it will not return until after it is reestablished. For the followers of a faith that, at the time, was older than Islam is today, there were only negative reasons to take prophetic claims about Muhammad seriously.

The behavior of Yathrib’s Jews deeply grated Muhammad. Jews ate kashruth (kosher), which made it difficult to partake in meals with Muslims. The extent to which this barred friendship between the two groups was so severe, in fact, that the Qur’an depicts kashruth as a punishment inflicted by God on Jews for their sins. Muhammad’s initial acceptance of Jewish practices evolved into the creation of distinctly Muslim ones. Yom Kippur became Ashura’, but Muslims preferred Ramadan; three prayers a day became five; facing Jerusalem became facing Mecca; Moses’ lawgiving was superseded by Abraham’s.¹ These changes coincide with the gradual breakdown of Muhammad’s relationship with the Jews.

But to cut to the chase, the first recorded violence between the Jews and Muslims took place immediately after the Battle of Badr, and it could plausibly be explained economically. The Qaynuqa’ were a particularly well-off group of Jews who tended to abstain from agricultural work more than the other Jewish clans. Their economic role was specialized and they held some of the best lands in the region as a result of their successes. The Muslims attacked this group for little discernible reason and Muslim records fail to provide anything close to a meaningful explanation for why this happened. After it did, the Qaynuqa’ were expelled from Yathrib.

The way this incident is described by Arabic sources is as a brutal display of the cycle of violence in action. It occurred in a time and place where the cycle of violence was also uniquely plausible because of the ubiquity of honor culture at the time. In their telling, a Jew pranked a Muslim woman by violating her modesty in the marketplace: he lifted her skirt. A Muslim man who witnessed the event immediately murdered the offending Jew. A group of Jews who witnessed that acted equally swiftly, killing the Muslim responsible for the murder then and there.

The reason this appears to be economically-motivated is that it does not fit with how the Constitution of Medina is generally understood to have worked. When something like this happened, the constitution was supposed to prevent further escalations. But despite Muhammad’s role as the mediator of disputes in Yathrib, his people took up arms and sought blood when he should have been law-, tradition-, and duty-bound to intervene to settle the dispute through the payment of blood money. Since the Jew initiated the event with a prank, it might be fair to say he was to blame, but the Muslim clearly took things too far, and it was his side which should have owed a blood debt.

Regardless, arbitration was never called for. Instead, Muhammad rallied the Muslims to besiege the holdings of the Qaynuqa’. After two weeks, the Qaynuqa’ were broken and ordered to leave, but being smart so as to avoid conflict with the city’s other Jews, Muhammad gave the leaving Qaynuqa’ the opportunity to collect their debts before they went.

From here, the records only become more lopsided and bloody as the Muslim position strengthened. For example, the Nadir poet Ka’b al-Ashraf joined the Quraysh and began to write poems insulting Muhammad. For this, the Muslims planned his death. Being a cripple, Ka’b used a cane and, while being escorted by some Muslims, he accidentally touched one with it, whereupon they slew him then and there.

After the Battle of Uhud, Muhammad sought to meet with the Nadir to receive blood money for a dispute involving some of them. The Nadir hosted the Muslims and wanted to resolve any disputes peacefully, but Muhammad had other plans. He slipped away during dinner and claimed the Archangel Gabriel alerted him to a Nadir assassination plot. For formulating this plan, the Nadir were guilty of a high sin, for which he would seek blood. So, shortly after this vision, he instructed the Nadir to leave Yathrib. If they went peacefully, they could maintain the ownership of their palms and receipts from the sale of dates. But otherwise, they had to go.

The exile of the Nadir had a very obvious political motivation because of their position as an ally of the premiere munafiq, Ibn Ubayy. Clearly Muhammad was attempting to consolidate power and to discredit his rivals, and he did so cynically, fabricating excuses to expel people who would not turn into his followers, or whom he considered in any way a threat, including those who remained neutral.

When the Meccans realized the threat of Islam and united to confront the Muslims at Yathrib, the Meccans found support from some of the Jews Muhammad had expelled. Within Yathrib, the Qurayzah refused to come to Muhammad’s aid, but they also didn’t bring the sword to bear against him, despite this being a perfect opportunity.

Before the siege began, a Persian convert to Islam instructed the Muslims in trench warfare, setting them to work building khandaq, or fortified trenches. In their holed-up positions, the Muslims sought anything that would ensure their victory, including assistance from the Qurayzah. However, as mentioned, this Jewish clan stayed neutral, neither helping nor hindering the prophet and his followers. This barely mattered because Muhammad led a unified force, and the Meccans led a number of squabbling tribes whose petty disagreements led them to prematurely break their siege, pack their bags, and return to Mecca. The Battle of the Trench, as it has come to be known, resulted in a force of 3,000 Muslims losing five or maybe six men and the assaulting force of at least twice as many, and perhaps three times as many Meccans losing just three. It was a flop, and Muhammad’s might was now clear to the Meccans.

For the crime of failing to assist, for which there was no precedent,² Muhammad sentenced each male adult of the Qurayzah to death. The battle that preceded this sentence is, I believe, an excellent means of understanding how unusual the punishment was. Tribes sought to minimize bloodshed because the complex networks of tribal affiliation and alliance were known to result in retaliatory action, making full-scale fighting a risk that everyone—Muslims included—wished to avoid. In its place, tribes preferred to engage in mubarazah, or individual, ritualized fights that would result in less bloodshed because they could only result in less blood being shed.

Modern biblical scholars³ have argued that the mere possibility of the Qurayzah turning on Muhammad caused him so much grief that he acted prematurely, with whatever pretext he could muster, in an effort to cynically snuff out any nearby groups he saw as threats. In pursuit of this goal, he led the Muslims on their first genocide, without a legitimate reason for his actions.

Another proffered explanation speaks to Muhammad’s obvious intelligence: rather than repeating his previous action by demanding exile, he exterminated the men and enslaved the women and children to prevent them from contacting other Jewish clans to form a coherent alliance against him. Altogether, they might have been able to muster a formidable and unified force against Muhammad had they been allowed to go, making them a clearer threat than the disorganized and generally more-poorly-equipped forces of the other Meccan tribes. So this makes sense.

Another perspective some Arabic sources later engaged in was the unsupported claim that Muhammad had developed a pact with the Qurayzah, for their mutual protection against any opponents of Yathrib. In electing not to come to Muhammad’s aid, they have been said to have unilaterally violated their commitments to Muhammad and the city’s other denizens. But this is absurd, because there is no precedent for such a pact; it stood against Muhammad’s role as mediator in the city; and the Qurayzah had no ability to enter into such a pact due to their status as clients of the Aws. Supposedly, the motivation for breaking the agreement was Huyyay ibn Akhtab—a member of the Nadir who himself was expelled from Yathrib and opposed Muhammad—convincing the Qurayzah of its necessity. But why? But how? But when? There are no reasonable attestations to support this explanation.

An overwhelming tendency in the various Arabic sources that attempted to chronicle Muhammad’s time in Yathrib has been to make some effort to distance Muhammad from the decision to commit a genocide against the Qurayzah. Some people have even tried to claim the genocide was less extensive based on speculative claims that Muhammad spared some males without any particular reason.

It doesn’t seem that Muhammad killed the Qurayzah out of a hatred for Jews. Rather, it’s more likely that his actions were politically motivated, given the Qurayzah’s strong socioeconomic and political positions. They were capable of acting as a group in concert, and were thus a threat to him, unlike the numerous other unaffiliated Jewish families who remained in Yathrib after these major clans were handled.⁴

A final political explanation with some plausibility posits that Muhammad needed an excuse to kill his opposition, so he gave the Aws Sa’d ibn Mu’adh a sort of final wish to execute the Jews who had remained neutral instead of coming to his aid. Sa’d would have thus been claiming it was the Jews who were responsible for his injuries, and requesting that Muhammad extract his blood debt from the Jews. This explanation is odd, as it suggests Muhammad was willing to break from law and tradition at a single man’s behest, but it is preferred by many scholars because it seems to absolve Muhammad from the responsibility for the genocide of the Qurayzah, by assigning it instead to Sa’d and the influence he had within Muhammad’s coalition.

Later, when Muhammad wished to conduct a full-scale assault against Mecca, few people came to his support, leading him to abandon the endeavor. Because of this, he agreed to the Truce of Hudaybiyah, a ten-year agreement between Meccans and Medinans with concessions allowing Muslims to make a pilgrimage to Mecca and freeing Muslims the Meccans had imprisoned or enslaved. This treaty left Muhammad unsatisfied, as it was an unusual failure on his part. He had to temporarily forswear continued conquest and victory, something he was not used to!

To regain his footing, Muhammad exploited a condition of the Truce: the Quraysh of Mecca agreed to end their alliance with the Jews of Khaybar. Muhammad immediately set off with his forces he had gathered for the abortive siege on Mecca and not six weeks later, the Muslim host had arrived in Khaybar.

The Jews were surprised. There was simply no reason to attack them, so why had he come? Regardless, they knew they had to flee to their fortresses and, after doing so, the siege set in. After six weeks, supplies were exhausted and the Jews of Khaybar acquiesced to Muhammad’s demands. They pledged to pay him tribute in the form of half of the produce they grew. Out of fear that they would also end up besieged and slaughtered, the Jews of Wadi al-Qura and Tayma also came to similar agreements with Muhammad shortly thereafter.

Arabic sources record that the Caliph ‘Umar ibn al-Khattab who succeeded Abu Bakr broke these tributary agreements and expelled Jews from Hejaz as a whole. This account is likely exaggerated, as indicated in the Cairo Geniza. ‘Umar’s expulsion of the Jews likely refers to Mecca and Medina, not the Hejaz as a whole.

Regardless of the extensiveness of al-Khattab’s ethnic cleansing of the Jews of western Arabia, after their loss at Khaybar, Jews never again barred Muslim conquest. But from this time, we do see things getting worse in Muslim depictions of Jews.

Consider the widespread belief that Muhammad was poisoned by a Jewish woman at Khaybar. The woman evidently sought revenge for the death of her male family in the slaughter of the Qurayzah. Muhammad allegedly succumbed to the poisoning, but believers in this story generally omit that the poison took four years to kill him. This story has also been permuted into a tale about Muhammad’s greatness, where a Jewish woman from the Nadir poisoned Muhammad, but the prophet detected the poison and spat it out, demanding an explanation, to which the woman claimed she was testing his prophetic status. After confirming Muhammad was, indeed, a prophet—as prophets should be able to detect poison—she married him and converted to Islam.

These accounts end up little more than a means of pushing a trope that Jews are untrustworthy into the domain of conventional wisdom. After all, what sort of creature would attempt to kill the prophet Muhammad? And who are we to support their continued existence in Arabia? So, for their slights, imagined or otherwise, they were slaughtered.

Starting Anew in Their Old home

One could argue it’s not fair to look all the way back. I agree with this perspective, as the modern conflict is clearly not just a continuation of some nearly-1,400 year-old one. It bears little resemblance to it, so we ought to look more recently.

The earliest modern violence seems to have been an 1882 accidental discharge incident in the middle of confused fighting over a gun. Arabs from Tzfat were invited to a wedding ceremony in Rosh Pinna and, as it was the Arabs’ custom to fire rifles in the air after a wedding, one of them grabbed the rifle off a settler who had no idea, they fought over the gun, and the settler shot the Arab.

Was this the start of the Israeli-Palestinian conflict?

Hardly. Not long after, the Arabs of Tzfat whipped up a group of some 200 men to get revenge, but their efforts amounted to little more than stonethrowing. The incident was resolved by the mukhtar of the Jaouni, who informed the Arabs they couldn’t resolve misunderstandings with murder. Instead, he got the Jews to agree to pay a settlement that practically bankrupted them.

Four years later, Jewish settlers demanded a group of Arabs leave land the Jews owned in Petah Tikva. This resulted in the Arabs attacking and robbing a Jewish man. The Jewish man had evidently stolen several of the mules the Arabs allowed to graze when they entered his pastures. Since it’s not clear which event came first, it’s unclear who provoked this. If the Jew confiscated the mules, he would likely have been in the right to do so given where they were grazing; if the Arabs attacked him first and he retaliated, well, then they were purely in the wrong. Since the settler ultimately refused to return the mules, a group of five dozen or so Arabs organized to attack Petah Tikva to reclaim the mules. They injured four Jews, mortally wounded a fifth—an elderly woman with a heart condition—vandalized the town, and stole more than their share of the town’s livestock. This ended unambiguously with the Arabs in the wrong, but that is the crux of it: it ended. It failed to inspire Jews to act violently in turn, to sustain the cycle of violence.

We needn’t belabor the point with more stories. Suffice it to say, violence continued, and until the state’s founding, the murders remained lopsided: Arabs killed more Jews than vice-versa. As Benny Morris noted, by 1908, thirteen Jews had been victims of Arabs: four for explicitly nationalistic reasons, with the rest mere victims of crime. The pace of violence increased and, over the following five years, twelve Jewish settlement guards would be killed by Arabs.

The Jews did not come violently. They were not expropriating land. In fact, this was viewed as antithetical to the mission of Zionism. As Morris wrote in his 2011 book Righteous Victims: A History of Zionist-Arab Conflict, 1881–2001:

Between 1878 and 1908 Jews purchased about four hundred thousand dunams out of a total land mass of 27 million dunams. The most prominent families—including Nashashibis, Husseinis, and al-‘Alamis of Jerusalem, Dajanis of Jaffa, ‘Abd al-Hadis of Nablus and Jenin, and Shawas of Gaza—sold land to the Jews. The major incentive was swiftly rising prices, caused largely by Zionist demand. Land prices in Palestine increased between 1910 and 1944 by as much as 5,000 percent.

Land purchase was the underpinning of Zionism. As Menachem Ussishkin put it in 1904, “Without ownership of the land, Eretz Yisrael will never become Jewish.” Purchasing was referred to in Zionist parlance as “redemption” or, indicatively, as “conquest” of the land. Land is acquired in the modern world by three methods, wrote Ussishkin, a Hovevei Zion leader: “By force—that is, by conquest in war, or in other words, by robbing land from its own;… by expropriation via governmental authority; or by purchase.” The Zionist movement, Ussishkin made clear, was limited to the third choice, “until at some point we become rulers.”

Eventually Jews did become the rulers of Israel, but between the end of the 19ᵗʰ-century and May 14ᵗʰ, 1948, the pattern of violence was decidedly factious.

The Arabs became increasingly incensed at the waves of Jewish immigration to the region as time went on, with particular ire towards the imposition of strict land ownership and demarcation, where herds had previously been allowed to roam and they had been allowed to plant crops in a largely poorly-regulated way. The imposition of these rules led a large number of Fellah to feel as though the Jews were dispossessing them from their traditional lands. Since many Jews settled on land technically owned by absentee landlords, the land’s tenant farmers felt they were being replaced. When the Nazis came to power in Germany, the Jewish population rapidly doubled, adding fuel to the flames.

But even before that, from 1920 on when the Grand Mufti Haj Amin al-Husseini—pictured below having a discussion with Adolf Hitler—came to power, he incited anti-Semitic violence on a regular basis. The violence he engaged in makes sense, as he was aware of and desired the extermination of the region’s Jews. He even visited a concentration camp on at least one occasion. The violence he precipitated only sometimes led to retaliation, and when the incidents involved more than a few victims, the response was rarely at the same scale as when it was meted out to his Jewish victims.

al-Husseini and Hitler in one of their several meetings

Equal retaliation was simply a limited occurrence for the Jews of the time, as in the 1936 Tulkarm shooting, where Arabs claiming to be martyrs with the goal of killing “all the Jews and Britons in Palestine” set up a roadblock, stopped a car full of three Jews, executed two of them, and injured the other. The roadblock was intended only to lead to the deaths of Jews, as supported by the fact that another driver at the roadblock shouted “I am a Christian German!” and was told to “Go ahead, for Hitler’s sake.”

Something important to note here is that the men who claimed to be martyrs against Jews and Britons in Palestine also claimed to be the compatriots of noted rebel leader Izz ad-Din Abd al-Qassam, the namesake of the modern Qassam rocket used by Hamas.

The following day, two Arab laborers were shot near Petah Tikva, allegedly by two Jews who were part of Irgun. Two days later, a funeral for one of the Jews who was shot in the roadblock was held in Tel Aviv and thousands of Jews showed up for it to demonstrate against British control and Arab violence. The British reported several cases of Jewish demonstrators attacking Arabs during the protests.

Not long after, books appeared with the picture of one of the Jewish victims on the cover, and the note that he was one of the “first victims”. Two days after the funeral, the Yafo Riots began due to a false rumor that the Jewish demonstrations had involved a number of Arab deaths that did not actually occur. Three days of rioting ensued, and during it, the British would kill two Arabs by firing into a crowd, while the Arab rioters would go on to kill fourteen Jews. Dozens of other Jews were injured, with the majority being victims of stabbings they were lucky to survive.

In response to the British suppression of the riots, the Arab Higher National Committee declared a general strike, and the 1936-39 Great Palestinian Revolt began.

The Arabs demanded their independence and the end of all Jewish immigration. They would not obtain it. Instead, the British and Jewish counterparties to this rebellion would lose 262 British and around 500 Jews, while inflicting losses numbering in excess of 5,000 on the Arab side. The outcome here mirrors the later, and better-known Israeli War of Independence: the conflict was largely not preceded by Jewish violence, but when it came time to fight, the Jews would fight to win, inflicting disproportionate casualties and ultimately winning the fight.

From here, we needn’t continue with stories and history, since what happens next is widely known. I’ll leave conclusions about who’s responsible given all the events leading up to the establishment of the state of Israel for readers to decide.

For now, I wish to argue that it’s all irrelevant.

---

Who Hurts Whom and Why?

For the modern-day, we have quantitative evidence that we can use to speak to the idea that the violence in Israel and Palestine can be described as tit-for-tat, back-and-forth—an eye for an eye.

In a particularly revealing paper on this very topic, Jaeger & Paserman constructed a daily series of fatal casualties in Israel and Palestine for the period between September 29, 2000 and January 15, 2005. In their data, there were 3,244 Palestinian fatalities versus 994 on the Israeli side. The monthly fatalities looked like this:

The authors theorized that there were three factors that dominated the link between violent incidents on either side of the conflict. In their own words, which I’ve separated into sections:

First, violence by one side can have an incapacitation effect, if it limits the other side’s capability to react. For example, Israeli targeted killings of key Palestinian leaders might reduce Palestinians’ ability to carry out further attacks against Israel; this is the stated Israeli rationale for such actions.

Second, violence can have a deterrent effect, when one side refrains from using violence in fear of the others side’s reactions.

Finally, violence by one side can lead to a reaction by the other side through a vengeance effect, to the extent that one side wishes to dispense retribution in response to the fatal casualties it suffers.

Speaking to these in their fullest forms would require modeling a dynamic game-theoretic equilibrium for violent behavior, and Jaeger & Paserman didn’t do that. Instead, what they did was estimate both side’s impulse response functions in a vector autoregressive (VAR) framework.

A VAR is a clever way to model whether the Israeli-Palestinian conflict is tit-for-tat. One common use of VARs is practically testing whether A reacts to the actions of B and B reacts to the actions of A through testing Granger causality: variable X does not Granger-cause variable Y if, after conditioning on the lagged values of Y, lagged values of X have no predictive power when it comes to the current value of Y.

Granger causality is a concept that has bothered a lot of people, because to give examples of Granger causality a true causal interpretation requires an exogeneity assumption. Jaeger & Paserman justified that assumption deftly:

While it is possible that there are factors that may induce a correlation between the regressors and the disturbance (e.g., an endogenous increase in Israeli preventive measures following an attack against Palestinians), it is the nature of the type of violence in this conflict that many of the realized fatalities are due to random elements. For example, did the suicide bomber enter a crowded or empty bus? Did the intended target of an assassination attempt sit in the front or the back of his car? Given these random factors, we suspect that endogeneity bias is unlikely to be quantitatively important for our results.

Another major assumption when attempting to treat Granger causality as an example of actual causality is that a time series may reflect forward-looking behavior, leading to reverse causation. But this is hardly worth worrying about when it comes to their results, as Jaeger & Paserman explained:

In the context of this conflict, it would be unreasonable to assume away any form of forward-looking behavior. We argue, however, that plausible forms of forward-looking behavior are likely to bias our estimates in a direction that is contrary to what we actually observe. For example, attempts by Palestinian radical groups to scuttle the peace process would lead to a finding of Palestinian violence Granger-reducing Israeli violence, which is the opposite of what we observe in practice. Similarly, if Israelis preemptively struck the Palestinians in anticipation of future Palestinian violence, we would be biased toward finding evidence of Israeli violence Granger-causing a Palestinian response, which is also in contrast to what we actually observe. The potential for mistakenly interpreted Granger causation as true causation would exist if, for example, the Palestinians engaged in preemptive strikes in anticipation of an Israeli attack (leading us to find that Palestinian violence Granger-causes Israeli violence). Because of the large imbalance in the military and intelligence capabilities of the two sides, this scenario seems, to us, fairly implausible. Thus, we view the potential for our results to be a product of reverse causation to be very small indeed.

With that out of the way, Jaeger & Paserman estimated the empirical impulse response functions based on fatalities. Since I do not expect this quantity to be familiar to everyone, briefly, the number on the y-axis is the number of excess deaths after an event compared to the series’ average. The number of excess deaths is estimated for a given day after an incident, with the days listed on the x-axis. So let’s compare empirical Israeli responses to being killed by Palestinians in terms of Palestinians killed with empirical Palestinian responses to being killed in terms of Israelis killed.

A few things are clear in this plot.

Firstly, there’s strong evidence against cointegration. To test this, I regressed one series on the other, extracted the residuals, and ran an augmented Dickey-Fuller test on them, with the result being a p-value of 0.081. The hypothesis test in this scenario is inverted from what the ADF tests for by default, so with n = 61 observations (0-60) per series, this is actually quite strong evidence against cointegration. The null is that the residual is stationary and so this p-value indicates that the residual is probably not stationary, and thus the two time series aren’t cointegrated.⁵ The original authors did not test this, but I believe it speaks more explicitly to their hypothesis than a simple visual comparison.

Secondly, the scales vary dramatically. The two response functions are extremely dissimilar: the number of excess Palestinian fatalities following the killing of an Israeli is elevated for 38 days, with significant elevation for the first ten days. The Palestinian response, on the other hand, never reaches statistical significance, but it is positive for 26 of the first 31 days. With an Israeli response that’s seventeen times larger for the first ten days and fifteen times larger across sixty days, it’s easy to see that the Israeli response is strongly retaliatory, whereas killings by Palestinians contain a larger random element.

Finally, this alone fails to answer the critical question why? For that, we need additional hypothesis testing, some of which Jaeger & Paserman did conduct. For instance, they checked whether Palestinians were simply incapable of responding due to Israeli terror countermeasures. As they noted, Israel, “anticipating a Palestinian reaction, may step up its preventive measures to thwart any possible Palestinian response. These can take the form of more frequent roadblocks, tighter restrictions on the movement of Palestinians within the Occupied Territories and from the Territories into Israel, increased presence and alertness of the Israeli security forces in crowded areas, as well as arrests of suspected Palestinian militants.”

If true, this suggests that Israel acts in a way that simply curtails tit-for-tat fatalities. Since Israel is strongly retaliatory, this is good for fatality numbers on both sides. So, how can we know?

Jaeger & Paserman reported that checkpoint closings were positively correlated with lagged Palestinian fatalities, “meaning that Israel does impose tighter restrictions on the movement of Palestinians after it has inflicted a high number of Palestinian fatalities…. We also find that lagged Israeli fatalities are strongly positively associated with closings, suggesting that to some extent Israel uses closings as a retaliatory measure against Palestinian violence.”

So, now that we know this, can we say that this measure of preventative actions on Israel’s part is why Palestinians are unable to realize a significant response to Israeli violence? Well, no. Controlling for this doesn’t change the results. But, one might argue, this variable is only an imperfect proxy for increased Israeli vigilance following an attack. However, it’s almost-certainly correlated with other forms of Israeli mitigation methods, which is probably why Jaeger & Paserman said “While our border closing variable may not be a perfect measure of Israeli vigilance, we suspect that any potential endogeneity bias in our results is quite small—and cannot account for the lack of an estimated Palestinian response to Israeli violence in our general specifications.”

As mentioned, Jaeger & Paserman’s analysis concerned fatalities, and if Palestinians responded in the context of a limited capability to inflict fatalities was limited, their analysis would, accordingly, fall apart. Thousands of instances of Palestinians being caught in the act or otherwise thwarted by the Israeli Defense Forces (IDF) have occurred over the years, so these sorts of en route interceptions might have been cases where, had they not happened, the cycle of violence would have revealed itself. But a supplementary analysis revealed that, even using failed Palestinian attacks or a combination of failed and successful attacks, the Palestinian response function was not consistent with retaliation.

So that’s all, right? Israel-Palestine is not a “tit-for-tat” conflict and we can abandon that model.

But this result failed to replicate!

Haushofer, Biletzki & Kanswisher performed the same analysis with extended data, accordingly greater power, and more diverse outcomes than killings alone. What they found was a somewhat different set of empirical response functions:

The residuals in this case are not cointegrated (p = 4.28e⁻¹²). To make it clear just how not cointegrated these groups are, if these processes were cointegrated, then we would have to accept that we flipped a fair coin thousands of times and landed on heads every single time. It’s as if we viewed back-to-back tail events from a Gaussian each and every day, for twenty days in a row.

Cointegration in time cleanly shows the lack of a stationary residual, and Granger causality assuming a linear relationship in terms of acts of violence. We might speculate that this relationship is nonlinear and wish to extract the relevant features to test that idea. Since there’s too little data to do feature extraction robustly, we can attempt to study wavelets, which are still dense in a decent feature space. To do this, we must don our Piketty hats and Kalman filter the data to gain power. This is probably fine since, firstly, wavelets are robust against filtering losses and, secondly, fortunately, violence isn’t so dense in time that you can obtain a reliable FFT from it alone.⁶ Kalman smoothing will weaken our result because it’s evident that the dynamics of the Israeli-Palestinian conflict are driven by exogenous shocks with larger magnitudes than drift and that the violence is also incapable of being explained by exogenous variables such as responses to violence by Palestinians.⁷

By running the test with this fatality data, we can see that the real p-value is 0.0834 and the imaginary p-value is 0.0984. Recall the need to implement a game-theoretic equilibrium model to understand retaliatory violence in full: this test provided weak evidence in favor of the idea that we cannot build a game theory of violence between Israel and Palestine in a manner that is consistent with the “cycle of violence”.

A more obvious thing this graph shows is that both groups had consistently significant responses to killings by the other group, so the lack of a consistent cycle is quantitative: there is retaliation, but it describes a quantitatively minor portion of the conflict’s aggregate dynamics.

This difference from Jaeger & Paserman’s work doesn’t mean they were wrong. Recall that Jaeger & Paserman used a 60-day window rather than a 20-day window. Plotting their data at twenty days alongside this extended data, the two sets of functions are actually nearly indistinguishable, which should be somewhat unsurprising because of source data overlap.

The problem with treating these as replications based on the overlap of their confidence intervals and the similarity of the forms of these response functions is that it’s easy to obtain a conclusion of “no difference”, as seen here, by simply being underpowered. An underpowered null is not at all impressive!

The mean magnitudes over this window may be a more useful point of comparison. So, for example, in Jaeger & Paserman’s dataset, the Israeli response was 0.2323 excess deaths versus a Palestinian response of 0.0194 excess deaths, while in Haushofer, Biletzki & Kanswisher’s dataset, the Israeli response was 0.2931 excess deaths versus a Palestinian response of 0.0321. The differences went in contrasting directions, but the estimates were all positive, and in three out of four series, consistently significant. But let’s not focus on this, because there’s something more operative afoot.

Jaeger & Paserman might have been onto something when they stated that perhaps the Palestinian response to killings comes in the form of retaliatory nonfatal violence. To test this, Haushofer, Biletzki & Kanswisher used data on Qassam rocket launches and found that, after a Palestinian was killed, the number of rockets that were shot off spiked! On the other hand, the Israeli response to Qassams in terms of fatalities inflicted on Palestinians was muted compared to their response to the killing of Israelis. Or in other words: Israel only really cares about lost lives, not nonfatal attacks.

So now we have clearer evidence for an asymmetry: when an Israeli is killed, Palestinians get killed as a result; when a Palestinian is killed, comparatively fewer Israelis get killed as a result, but a relatively large number of rockets fly.⁸ Remember: excess estimates are relative to the mean of the series. Armed with these estimates, we can define precisely how much of the Israeli-Palestinian conflict appears to be, in some sense, part of a cycle.

It’s not much.

As Haushofer, Biletzki & Kanswisher noted when they performed their magnitude estimations in their supplementary materials, “after a killing of Palestinians by Israelis, an extra 0.08 Qassam rockets are fired, which corresponds to a 6% increase over the 1.37 average daily rockets” which meant that since “Israeli attacks caused 4,478 Palestinian fatalities, each of which led to the firing of an extra 0.08 Qassam rockets, or 358 rockets in total… [retaliation accounted for] 10% of the 3,645 rockets that were fired in total.”⁹ This excerpt contained an error, however, because there were 4,874 Palestinians killed by Israelis in this timeframe, not 4,478, so the real proportion of Qassams explained by Palestinian retaliation to killings by Israelis was (.08 * 4,874)/3,645 = 10.70%, not the 9.83% they arrived at.

For comparison, (.576 * 1,062)/4,874 = 12.55% of Palestinian fatalities were attributable to Israeli retaliation over the killings of Israelis. An additional 8.23% might independently be attributable to the Israeli response to Qassam launches, for a total Israeli retaliatory part of 20.87%. The Palestinian retaliatory effect for lives lost, in terms of Israeli lives, is a highly similar 21.11% of total Israeli lives lost.

The reason the proportions whose lives are lost to retaliation are so similar despite the fact that the Israeli response in terms of deaths inflicted is almost thirteen times larger per Israeli fatality is that more Palestinians suffer fatalities, meaning that there are more events to retaliate over. This ratio is comparable when using the number of days in which events occur, as the ratio to deaths is similar for both sides.

Because Israeli retaliation to Qassams in terms of deaths inflicted matches up with the number of excess Qassam launches following Palestinian fatalities, and because Israel is more retaliatory no matter how you slice the numbers on retaliation to their side dying, they are ultimately vastly more retaliatory than Palestine.¹⁰

---

Will We Go Blind?

The aggregate violent dynamics of the Israel-Palestine conflict are not an example of “an eye for an eye”.¹¹ Roughly one-fifth of the fatalities attributable to either side can be ascribed to retaliation for earlier fatalities and only around a ninth of the rockets launched by the Palestinians can be attributed to these fatalities. Given that killings and bombings are among the most severe things a group can do to another group, it’s unlikely that these estimates are even that far from the ground truth of the matter, all things considered. Adding in the Palestinians’ incendiary kite and balloon attacks, encouragement of murder between Arab and Jewish Israeli citizens, etc. would likely not make a major difference, and it might actually reduce the importance of retaliatory violence.

To the extent that there’s a risk of the whole world going blind from tit-for-tat violence, short of nuclear war, it’ll take a while, and it’s hard to imagine it happening with the notable technological and organizational asymmetries of the sides involved in the Israel-Palestine conflict.

It’s debatable whether any long-term conflict really ends up being dominated by tit-for-tat.

When I described the Nakam, I noted that they faced a great deal of trouble in their quest for revenge, and they eventually failed to get what they were looking for. That’s probably for the better, and it’s probably typical. It’s hard to look around in the real world and seriously contend that revenge plots generally succeed rather than the more likely possibility that they generally fail; that plotters become frustrated and quit, or that would-be plotters simply stop caring about whatever slight they felt they experienced, and they move on. That this even happened with a committed terror cell whose members had shown—and many of whom would also go on to show—that they were competent should really come as no surprise.

So what of Israel and Palestine? If not the cycle of violence, what model?

That isn’t for me to say, but there are still plenty of things that can be said.

For example, if Hamas laid down their arms, they would be fine. There’s no chance at all that Israel would ever embark on a genocide against Palestinians, and if violence against Israelis ended, then Israel would also have no issue peacefully coexisting with the Palestinians as they wish to—and mostly do—with other Arab states. The risk of “ethnic cleansing” in the sense of a population transfer is actually much greater for the inhabitants of Gaza and the West Bank if they are violent.

Or that Israel is not and has not been engaged in the initiation of violence. Much of the discussion on whether or not Israel is responsible for the violence it was subjected to on October 7ᵗʰ, 2023 seems to be framed within a world where the cycle of violence exists as a self-perpetuating dynamic initiated by Israel. But that is not the world we live in on either count. Israel follows rules of engagement: “soldiers are allowed to fire freely at rocket- and mortar-launching cells immediately before, during or after a launch, and with permission from a senior officer, they can also fire at Palestinians trying to lay bombs within half a kilo-meter of the border fence. Other than that, however, no offensive operations are permitted.” The Israeli military follows these rules, and violations are both exceptional, punished, and responsible for few lost lives relative to the conflict as a whole. More importantly, Israel follows these rules to its detriment, and groups like Hamas and Hezbollah, on the other hand, do not have a rule book; they attack violently and indiscriminately, and practically everything they do is a war crime. Placing the responsibility for being attacked on Israel’s shoulders is a peculiar form of victim-blaming.

If we want to be specific about how groups like Hamas and Hezbollah are constantly involved in war crimes, just look at the Geneva Conventions and, specifically, the concept of distinction. This matters for Hamas in a twofold manner:

1. They do not distinguish combatants from civilians. In fact, they go further than this, and they intentionally disguise combatants as civilians (e.g., in the case of suicide bombers), whilst also locating military assets among civilians in an effort to reduce the odds of being attacked. In a practical sense, Hamas engages in the attempted maximization of civilian casualties on the side they represent.

2. Their attacks are indiscriminate. Qassams are fired towards civilian centers; incendiary devices are used to burn farms; civilians are slaughtered in droves when they have the opportunity. This is, to some extent, due to Israeli policy. Namely, the policy outlined in those rules of engagement above: Israel destroys any rocketry that it can identify, forcing terrorists to attack in an opportunistic fashion rather than with direction towards military targets only. But given their rhetoric, actions, and history, it’s clear that groups like Hamas also do not mind attacking civilians at all. In their recent English-language address, they even suggested that Israelis were all combatants.

There is far more to the case that practically everything Palestinian and Lebanese terrorists do amounts to a war crime. But, for clarity, if we were to compare Hamas and Hezbollah, the latter obviously does relatively fewer war crimes. As a matter of necessity, Hamas can’t practice distinction or they will be eliminated. But that would be good for the world, as they are a terror group. Not only that, but they are a terror group that clearly harms Palestinians, both directly, and through the fact that they avoid distinction.

This specific war crime has major importance when it comes to monitoring events in the Israel-Palestine conflict. The point of distinction is the separation of civilian and military resources, so as to minimize the collateral risk to civilians and the danger of confusing them with military targets. Hamas uses civilians as shields to survive, and they also use civilian deaths for their highly-effective public relations. Accordingly, they do bring in enormous amounts of money from outraged foreign supporters.

Since Israel is a party to the Geneva Conventions and it abides by the law, its responses are targeted at perpetrators, while the Palestinian response to Israel’s existence is often simply firing missiles into civilian areas to kill random people. The fact that they’re bad at killing people in this way doesn’t absolve their attempts; no one would say “no harm, no foul” to a crazed gunman firing randomly into a crowd who just so happened to only hit the people wearing body armor. This is part of why pictures of rubble in Gaza shouldn’t move the needle on anyone’s views about who’s in the right in the conflict. Blowing things up is not, in itself, a war crime, and the moral valence should not be that it is purely bad, because what Israel does is clearly not indiscriminate. This applies very broadly, because Hamas uses a wide variety of civilian targets for combat purposes, including such oddities as using ambulances to transport soldiers.

Some of the victim-blaming Israel receives has had a vaguely academic background. For instance, numerous groups have claimed that Jews settling Israel is itself a form of violence. But buying land¹² is nonviolent, instances of people ‘moving in’ to someone’s house are mostly fictitious, and non-military interpersonal violence within Israel, even in settlement areas, is lopsided. While the Arab homicide clearance rate is low and has worsened under Ben Gvir, we still sporadically obtain information about perpetrators, and the difference is always lop-sided, with Arabs killing each other more often and committing more interreligious homicides to boot. In 2023, the Arab murder rate has apparently been some twelve-times the Jewish one.

The claim that “settler colonialism” is violent ends up being the claim that people who move into an area through peaceful means rather than by killing its inhabitants are doing some sort of violence by merely immigrating, working the land, and—in many cases—operating under well-defined property law instead of previously-extant informal, often tribal rules. Opposition to settler colonialism is a concept that was more or less invented from whole cloth to justify its adherents’ preferred positions in any conflict. Everything the outgroup does can be colonialism, one merely needs to recontextualize it.¹³

Settler colonialism is, in any case, a poor excuse for violence. In Hamas’ recent English-language address, they made a remark that is reminiscent of just how absurd the idea is. They claimed that their violence was “just a matter of time” due to Jewish settlement. This is effectively a claim that they are somehow possessed by settler colonialism; that it takes control of their bodies and compels them to commit acts of evil. But how? But why? There are no real answers, because the truth is that they chose violence.¹⁴

A related line of thinking is that Israel must be violently resisted because it has imposed apartheid on the Arabs. But this is simply a lie. Gaza is so far from a Bantustan that it’s hard to take proponents of this idea seriously. The story doesn’t make sense because there is only one clearly-defined group who it could apply to and, when they’re identified, the motivation for apartheid no longer makes sense. That group is non-citizen Arabs who either cannot or will not prove their ancestry in the region.

Israel’s Arab citizens comprise around a fifth of its total population and they have the same rights as its Jews. They’re not pushed into camps or “open air prisons” that are, notably, larger and less densely populated than Manhattan. If these people were able to provide proof of their family’s history in the region, they would be able to obtain citizenship as an Israeli. I’m sure some people can do that and don’t want to, but if their goal is their own personal freedom and welfare, they’re making a mistake. I’m also sure others might have ancestry in the region, but they can’t prove it for whatever reason. For the small number of them whose ancestors were among the nearly 700,000 who left Israel after the War of Independence, their situation is now certainly worse than the descendants of the 900,000 Jews who, at the same time, fled or were expelled from the Arab countries and Iran. But living in a tragic state does not compel violence, nor is it sufficient to create terrorists.

---

What Is Israel to Do?

Since Israel is the powerful party in the region, many think they are the only party who can make a difference. And Israel does make a difference. Their strategies change and improve.

For example, when it comes to preventing suicide bombings, Israel has learned that the preemptive arrest of suicide bombers prevents their attacks, but if they are preemptively killed, a would-be suicide bomber’s martyrdom inspires recruitment. Insights like that are especially useful for Israel, because Israel values its citizen’s lives more than perhaps any other nation.

The man pictured at the center of this photograph is Gilad Shalit. He’s seen saluting prime minister Benjamin Netanyahu at Tel Nof air force base, shortly after being released from Hamas captivity. Shalit was the first such prisoner to be released alive in 26 years. In that time, many other prisoners were released without their lives. The price Israel paid for his release was immense: Hamas agreed to give him back in exchange for 1,027 of their own prisoners.

The price Israel is willing to pay to protect its own is further exemplified by considering the death toll inflicted by the prisoners Gilad was exchanged for: those 1,027 men were responsible for 569 Israeli deaths. The very fact that Israel was willing to let them go in order to rescue just one person speaks to the intent of Israel’s operations in Gaza, the West Bank, and further abroad. The Israeli strategy is simply focused on preserving Israeli lives.

Through learning that it is better to capture than kill potential suicide bombers, Israel has likely increased the number of available prisoners with which they can make this sort of deal. But this is a conundrum, because the mere fact of being open to negotiations incentivizes kidnappings. Since most negotiation attempts fail, this means lives lost with little to show for it, and it also renders most arrests examples of mere incapacitation. Furthermore, when negotiations do succeed and Israel gets back a person or a corpse, released prisoners may recidivate, resulting in more Israeli lives lost even though Israel managed to get back a citizen.

This is a bad position for Israel, but it clearly shows that the country will use long-range strikes, heavy artillery, a wide variety of preemptive measures, dense intelligence networks, drones, and whatever other resources it has at its disposal to prevent as many Israeli deaths as possible. This comes with a secondary conclusion: if terrorists stray from strict distinction, Israel will not relent. If Israel can save a single Israeli life, that is far more valuable to them than umpteen terrorists and umpteen more Palestinian civilians those terrorists have on their side, have kidnapped, have convinced to stay put, or who have otherwise been caught in a position where they’re standing alongside Israel’s legitimate targets.

If Israel wants to be perceived as more humane towards Palestinians, then so long as they continue fighting, it can only do so by caring less about its own citizens. That will not happen, and that is why the future of this conflict—which is not a cycle of violence—can only be decided by Palestinians. For this to end, they must give up fighting. Israel will not lose any fight with the Palestinians, nor will it lose any bouts with the rest of the Muslim world. They’ve also demonstrated an unwillingness to perpetuate potential cycles of violence, to instigate conflicts, and to strike against non-military targets. Since Israel will not end the Palestinians, the Arab world will not take the Palestinians, and Israel cannot afford to make all of them citizens, peace is a process that Palestinians must initiate. Until then, it won’t exist.

1. This last difference is important to understand, because it reveals Islam’s early obsession with lines of descent. The Muslims chose Abraham over Moses because Abraham was the father of Ishmael and Isaac, making him the origin of the northern Arab ancestors of Muhammad’s tribe, the Quraysh and his clan, the Hashim. This legitimized the Muslim claim to an original monotheistic origin in Arabia rather than Babylon.

2. Some claim that the 44ᵗʰ clause (in versions with sub-clauses) of the Constitution of Medina provided such an obligation, but this is not consistent with how it was interpreted before then, nor does it fit with the nature of the battle, as Jews shared no obligations to participate in Muslim religious conflict, and it only follows if we treat the Constitution of Medina less as a pronouncement, and more as a formal agreement for all parties who were affected by it, which is not likely.

In the Ibn Ishaq version, the relevant clause is the 45ᵗʰ (in versions without sub-clauses), and it reads “They (referring to Jews) will aid each other against whosoever is at war with the people of this treaty.” The interpretation of parts of the document to mean that this would call the Nadir, Qurayzah, and Qaynuqa’ into Muhammad’s support when it came to any incursion into Yathrib is dubious, and may be possible to wring out of Wellhausen’s interpretation, but not out of reality, because, as Goto mentioned, the “Yahud Bani so-and-so” could not be a correct interpretation of the document due to the presence of “the name of a fourth tribe, which had to be much smaller than the three”—Banu Tha’laba. Because of the documented, separate negotiations with other Jewish groups, it’s hard to treat the claim that violation of the Constitution of Medina is explanatory seriously.

Because I do not wish to expand on this history and the various interpretations of what did or didn’t happen at any greater length, see footnote 4 and Lassner’s Jews, Christians, and the Abode of Islam. Lassner’s and Lecker’s in particular are works that I rely on heavily for the entirety of this section.

3. Like Julian Wellhausen.

4. As suggested by Michael Lecker in The “Constitution of Medina”: Muhammad’s First Legal Document, “I argue that the majority of the Jews of Medina, including the main Jewish tribes Nadir, Qurayza and Qaynuqa who are not mention in the Kitab, were not party to it." This does not mean that the main Jewish tribes had no agreements whatsoever with Muhammad; on the contrary, after the hijra, the Nadir, Qurayza and Qaynuqa concluded separate treaties with him. But these were basic non-belligerency treaties mainly including security arrangements, unlike the far-reaching commitments undertaken by the Jews who did participate in the Kitab. The rather limited Jewish participation in the Kitab means that it is far less significant with regard to the Prophet’s relationship with the Jews of Medina than has previously been thought.”

5. This may seem unintuitive, but the p-value here corresponds to the probability of the empirical residuals being stationary and our time series being cointegrated being 0.081. More formally-speaking, the p-value here returns P(Observed | Residual Stationary/Time Series Cointegrated).

I recommend generating two time series with random walks where one depends on the other to gain an understanding of how this works. In that scenario, the p-value will shoot up. For a real-world comparison, directly cointegrated stock market items like mining companies and commodities rarely go higher than 0.50. To obtain, say, 0.95, would require something like two identical random walks that are time-shifted and have small and separate sources of error. A mean reversion Ornstein–Uhlenbeck process can be used to check this: make X(t) a Brownian motion, and Y(T) = -(Y(t) - X(t-1)) dt + s dW(t), where W(t) is another Brownian motion and s is some small number. This will result in a large ADF p-value.

On a completely unnecessary note, if you’re like me, you see Ornstein-Uhlenbeck processes and think “Huh, kind of like a Wiener process”. Well, follow the Girsanov theorem and you can turn any Ornstein-Uhlenbeck process into a Wiener process with the right change of measures, in a similar sort of way you can turn any continuous probability distribution uniform. The more basic version of ‘make any continuous distribution uniform’ is ‘suppose X is real-value and has CDF F, then F(X) is U(0,1)’. Cool, right?

6. To obtain a reliable FFT, and thus wavelet decomposition, you need the time series to be relatively dense, or else you’ll oversmooth, which is why FFT is robust to Kalman filtering but, at the same time, this exacerbates the issue that it may not see small-scale dynamics that aren’t represented. This is basically doing the opposite of importance sampling.

7. In this usage, the noise is exogenous and contains all factors like ‘will to commit violence’, mental illness, or ‘means to commit violence’, which are part of the “response”, which in standard terms means “the whole time series”.

8. Through 3,645 Qassam attacks, there were fifteen fatalities in this dataset.

9. Results were isolated to the first day after an attack to “obtain a conservative estimate and to avoid bias from different lag structures”. Extending this exercise to additional days could obviously lead to explaining more deaths than are available in the dataset, which is a very probable outcome even when it is untrue, because of the fact that periods of time containing fatalities and rocket launches overlap, the period to look beyond is 20 days, and some instances of retaliation probably did happen outside of the dataset. Doing this does not alter the qualitative result that Israel is relatively more retaliatory.

10. Around thirteen times in raw terms, but all reasonable adjustments suggest Israelis are multiple times as retaliatory. For instance, using the mean deaths per event, the ratio moves up to fourteen times.

It should be noted that these are estimates of what is, not what would be if, say, Hamas and Israel were equally powerful. Hamas is not pulling its punches. If they were as capable as Israel, Hamas has made it clear they would commit a genocide against every Jew in the country.

11. Unless something has changed.

12. Though I do not consider it relevant, it is also true that Jews were once the sole controllers of the land, and they are, indeed, the extant group with the most distant historically-distant claim to it. Some people care about this, others use rhetoric that suggests they should care about it but they somehow do not, and some have tried to maintain consistency by inventing conspiracy theories about the ethnic origins of European Jews in an attempt to claim that Jews’ ancient land ownership is irrelevant.

13. As a consequence, opposition to “settler colonialism”, support for “#landback”, and other beliefs in the same vein are often arbitrary provocations to genocide and ethnic cleansing. They are fully-general and can be used to suggest, for example, that many regions like Transylvania, Istanbul, Biafria, etc. need to start engaging in mass deportations, expropriations, land transfers, and, effectively, genocides. The peaceful version of these beliefs, whereby people buy land as or on the part of claimants considered rightful, is the only permutation of this that is not horrendously illiberal and likely to promote mass death.

The poverty of these beliefs is related to why the “cycle of violence” as a concept cannot be saved by reference to “hate” or “speech” as examples of violence, either.

14. I am aware that their other excuses often amount to veiled anti-Semitism.