The Meaning of "Massacre"

The debate between Benny Morris and Martin Kramer over Israel's wartime conduct enters its second round.

An aerial view of Lydda in the 1930s. From the G. Eric and Edith Matson Photograph Collection, Library of Congress.
An aerial view of Lydda in the 1930s. From the G. Eric and Edith Matson Photograph Collection, Library of Congress.
July 28 2014
About the author

Martin Kramer is a historian at Tel Aviv University and the Walter P. Stern fellow at the Washington Institute for Near East Policy. He served as founding president at Shalem College in Jerusalem.

Benny Morris:

I have to admit that, prior to reading his essay, “What Happened at Lydda,” I had never read anything by Martin Kramer. But I had heard that he was a serious Middle East scholar, albeit of subjects far removed from the 1948 war. His essay, however, is imbued with clear political purpose—“Israel is defined by much of liberal opinion as an ‘occupier,'” Kramer writes at one point in an essay that ostensibly deals with July 1948—and thus smacks more of propaganda than of history (even though the minutiae of his criticism of Ari Shavit’s manipulation of texts and facts regarding one minor episode in the war—what happened at a mosque in Lydda on July 12, 1948—are illuminating, if not so much about the war as about Shavit).

In my response in Mosaic to Kramer’s essay, I argued that “disproportion” speaks “massacre.” Kramer has now replied to my argument in a manner disingenuous if not forthrightly mendacious. Yes, in contemporary warfare between advanced technological societies and Third World societies—the U.S. versus Iraq, for example, or Israel versus Hamas—the application of air power and sophisticated artillery by a Western power can lead to completely disproportionate losses on the part of ill-armed Arab ground forces, and these do not necessarily speak of massacre. But in the Israeli-Arab war of 1948, two or more relatively primitive armies came to grips. When, in a specific battlefield, one side was more powerful than the other, a “disproportion” in losses might arise. That happened, for example, in the successive battles between the Haganah/IDF and Jordan’s Arab Legion at Latrun in May-June 1948, where many more Israelis died than Jordanians due to the Legion’s efficient use of its mortars and 75-mm artillery batteries and to Israeli paucity in or misuse of heavy weaponry. But when the disproportion is 250:0 or 250:2, as occurred, according to contemporary IDF documents, between the IDF and the Lydda townspeople, some of them armed, on July 12 of the same year, then “battle” is surely not the name of the game; “massacre” is more like it.

To Kramer, this was a “battle with two sides.” And now, to mislead his readers, he says in his reply that there was indeed a “battle”—between the Yiftah-brigade soldiers and the two or three Jordanian armored cars that had penetrated Lydda. But that is not at issue. Sure, there was an Israeli-Jordanian battle (or, more accurately, a skirmish, in which there were Israeli casualties) around noon on July 12. But the question is whether what transpired afterward, between the townspeople, some of whom sniped at the Israelis, and the Yiftah troops—an action that ended in 250 dead townspeople–was a battle. Given the vanishingly small number of Israeli losses, “battle” is a tendentious misnomer, Kramer’s sophistry and verbal acrobatics notwithstanding.

In his reply, Kramer dredges up new oral testimony about what happened at the small mosque. (I bow to Kramer’s expertise in Arabic as to how the name of the mosque should be transliterated.) But this testimony still fails to prove that anyone from within the mosque threw a grenade at the Israeli troops outside, triggering the IDF rocketing of those inside. Anyway, the event at the mosque was merely one (small) part of the Lydda massacre that afternoon (“small” insofar as it accounted, reportedly, for only 30 to 70 of the 250 Arab dead).

Kramer may be right in saying, as he now does, that “250” as recorded in the IDF documents was a rough estimate. I doubt that the IDF soldiers actually counted the bodies as they gathered and buried them, or that they recorded the process. But even if the ratio was 200:0 or 200:2, it would still point to a massacre. Kramer, incidentally, writes of 250 “casualties” when the document actually says “some 250 dead . . . and many wounded.” Historians—indeed, English-speakers—should know the difference between “dead” and “casualties.”

Kramer says that the figure of 250 and its corollary of “disproportion” are my sole proof that a massacre or massacres occurred in Lydda on the afternoon of July 12, 1948. As he puts it: “On this slim reed rests Morris’ claim” of massacre. But that is simply not true. In my books, and in my response to Kramer’s essay, I also quoted from contemporary documents—Kramer, for some reason, avoids documents like the plague, preferring interviews by others conducted decades after the event (was he trained as a historian or as an anthropologist?)—showing that Yiftah HQ ordered its troops to shoot “anyone seen on the streets.” Lastly, Mula Cohen, Yiftah’s commanding officer, when recalling these events in Sefer Hapalmah, vol. II (1956), p. 885, wrote: “The brutality of the war here reached its peak. The conquest of the city . . . awakened instincts of revenge [yitzrei-nakam] that sought an outlet.” When you couple a desire for revenge, and shooting anyone seen in the streets, with 250 dead townspeople, what do you get? 

As for the run on the previous day (July 11) by Moshe Dayan’s 89th battalion through the peripheries of Lydda and Ramleh, all the documents agree that dozens of people—maybe 100-150—were hit, men, women, and children. True, the column suffered nine dead and some wounded. But the shooting was hardly a targeted killing of militiamen, and couldn’t have been one given the nature of the run (I’m sure even Kramer will agree with this). Granted, it wasn’t a deliberate massacre—but without doubt it was a mass killing that included civilians. Did these comprise many of the casualties? Most of the casualties? I don’t know. No one does. 

Perhaps part of the problem stems from the meaning of the word “massacre.” Of course, all would agree that if you line up 100 civilians or unarmed POWs against a wall and shoot them, you have a massacre. But what occurred in Lydda was more complicated. A firefight with two Jordanian armored cars and sniping by armed townspeople provoked mass killing by a small IDF contingent that felt vulnerable and panicky: 300 to 400 men in the center of a town that they thought had surrendered (it hadn’t) and that contained tens of thousands of locals and refugees. And the Arabs were the ones who had started the war.

But whatever the extenuating circumstances, had IDF troops acted in such a manner today, given current legal and moral norms, they would most likely have been put on trial—by Israel. One can argue that one shouldn’t “judge” soldiers’ behavior in the past by today’s standards. Agreed. But this doesn’t change the fact that they committed a massacre.


I shouldn’t really waste time on this, but Kramer’s assurances to the contrary notwithstanding, Natan Alterman’s poem Al Zot was indeed probably about Lydda and not, as he says, about Dawayima (a village near Beit Jibrin that was the site of an IDF massacre in October 1948). Why Kramer is so “certain” about this—as he is so certain about almost everything he writes—I don’t know. The fact that Dan Laor, Alterman’s latest biographer, believes it is neither here nor there; Laor offers no proof either way. Certainly, there is no proof ; we don’t have anything like, for instance, a diary entry by Alterman saying explicitly that he has composed a poem about Dawayima. But what he describes in Al Zot conforms to what happened in Dayan’s raid in Lydda—and the poem does speak explicitly of an incident in a town or city (ir) and not a village (kfar). In any case, there is no doubt in my mind that Alterman published his poem in the daily Davar, on November 19, in protest against a series of atrocities committed by IDF during October and early November, of which Dawayima was just one. Those interested in the whole series can find detailed descriptions, insofar as the accessible documents allow, in my 2004 book, The Birth of the Palestinian Refugee Problem Revisited. (Of course, Kramer might dismiss these descriptions of mine as anti-Zionist propaganda as they are based on IDF, Israeli government, and Western diplomatic documentation from 1948 rather than on oral testimony by the participants given decades after the event.)

Lastly, Kramer continues to elide—in effect, deny—that following the massacre, the IDF expelled Lydda’s (and neighboring Ramleh’s) inhabitants. He says in his reply that this wasn’t the subject of his essay, and that the debate about the expulsion is a “well-worn subject.” (I suppose World War II is also a “well-worn subject,” but I believe historians still write about it.) He denies that he has whitewashed or elided. But that’s precisely what he did, clearly and, to my mind, for political reasons. In his reply he continues to ignore the plain, simple, explicit import of the July 12-13 IDF documentation on the subject. The Lydda (and Ramleh) townspeople were expelled, on orders from on high, and the officers expelling them knew that they were carrying out an expulsion and, at its end, knew that they had carried out an expulsion. This may not have sat well with the conscience of Shmarya Gutman, the Israeli military governor of Lydda; indeed, in his famous November 1948 article on Lydda, Gutman compares what happened to the townspeople with the exile of the Jews by the Romans 2,000 years earlier. But that’s what happened, whatever justifications or stories he may have concocted afterward.

Gutman—and he is the sole source for this story—writes, as paraphrased by Kramer: “The next day, the Israeli military governor reached an agreement with local notables that the civilian population would depart from Lydda and move eastward. Israeli soldiers, acting under orders, also encouraged their departure. Within a few hours, a stream of refugees made its way to the east, emptying the city.” So: there was an Israeli-Arab “agreement” for the Arab exodus and there was some Israeli “encouragement.” Now really. I’m sorry, but what can I do? The documents speak clearly, explicitly of geyrush: expulsion.

In my 2004 book, I give due credence to other factors in the exodus of the inhabitants of Lydda and Ramleh: they wanted their menfolk released, they didn’t look forward to life under Jewish rule, and so forth. But none of this detracts from the fact that the event as a whole was an expulsion. (Kramer’s fellow expulsion-deniers, Alon Kadish, Avraham Sela, and Arnon Golan, the last of whom is actually a good, serious historian, conclude in their Hebrew book on the conquest of Lydda that there was a “partial expulsion,” since some 500 of Lydda’s inhabitants remained in place. Again: now really.)  

Kramer’s goal, throughout, appears to be to create or enhance a white-as-snow image of Israel. Like me, he is outraged by today’s widespread, untrue, and ill-willed misrepresentation of Israel, in the media and on college campuses, as a monstrous state. Well should he be outraged. But unlike Kramer, in countering this image I am unwilling to distort and misrepresent the past.

Benny Morris is a professor of history at Ben-Gurion University and the author of, among other books, 1948: A History of the First Arab–Israeli War (Yale, 2008).




Martin Kramer: 

First, a word of thanks to Benny Morris. His work wasn’t the subject of my essay, but he accepted an invitation from the editors to wade in as a respondent. Were it not for him, there wouldn’t have been any debate in Mosaic at all. Ari Shavit, the author of My Promised Land, whose account of Lydda was the subject of my essay, remains silent. So do those who boosted Shavit’s book while shedding belated tears of contrition over Lydda. They haven’t so much as tweeted the existence of the essay or the exchange that followed. What are they waiting for?

Here once again, as in his original response, Morris would distract us from his dubious claim (elaborated by Shavit) that Israeli troops committed a “massacre” in Lydda by reverting to the subsequent expulsion of the town’s Arabs. He even calls me an “expulsion denier.” So I will state my view more plainly for his benefit. On July 12, after the aborted Palestinian uprising in Lydda, an order came down from on high—just how high is debated—to expel the Arab inhabitants. But there were many thousands of Arabs on the road out of Lydda who didn’t wait for an order to leave, and never heard one. As I wrote, “the word ‘expulsion’ cannot suffice to describe everything that happened in Lydda” (emphasis added). Whether that makes me guilty of “expulsion denial” (which, like “Nakba denial,” draws an abhorrent analogy to Holocaust denial), I leave to readers to decide.

Now back to the Lydda “massacre.” Morris thinks the aim of my essay was to “create or enhance a white-as-snow image of Israel.” In a summary of my essay that appeared elsewhere on July 1, I wrote this: “My motive hasn’t been to protect Israel’s honor against the charge of massacre. There are some well-documented instances from 1948. It’s just that Lydda isn’t one of them.” Morris now speculates that I might dismiss those other instances “as anti-Zionist propaganda as they are based on IDF/Israeli government/Western diplomatic documentation from 1948 rather than on oral testimony.”


Untrue: documentation dispels doubt. And its absence increases doubt. The Lydda “massacre,” according to Morris the “largest” of 1948, is dubious precisely because of the total absence of any IDF, Israeli government, or Western diplomatic documentation—Morris’s own catalogue of what it would take to prove it. So he must cling tenaciously to the one shred he has, from a document not kept secret but published almost sixty years ago: the disproportion, in the Yiftah brigade’s report of the battle, between 250 enemy killed versus only two Israelis. According to Morris, in a battle between “relatively primitive armies” in which neither side has a clear advantage in firepower, that “speaks massacre.”

But the battle in Lydda (after the retreat of the Arab Legion) wasn’t between “relatively primitive armies.” It was between the Israeli army, outnumbered but equipped with antitank PIATs and heavy machine guns, versus many more locals armed with “sten guns, tommyguns, rifles (French, Italian, and German), and some dating back to the Ottoman period.” Morris again: “If, in a specific battlefield, one side was more powerful than the other, a ‘disproportion’ in losses might arise.” Isn’t that a precise description of the Lydda battle? What is the story of the small mosque, if not one of superior firepower (a PIAT) exacting a high toll (“dozens,” writes Morris) with one shot? The sides to an exchange of fire don’t have to be equal for their (mis)match to be a battle, and that battle doesn’t become a “massacre” simply because the outcome is lopsided.

But just how “disproportionate” was the killing in Lydda? That still depends largely on whether the figure of 250 deserves to be taken seriously. Even Morris now recognizes that it wasn’t the result of a count. “I doubt that the IDF soldiers actually counted the bodies as they gathered and buried them,” he concedes. (Indeed, they may have imposed that task on the remaining Arabs: one veteran of the battle said that “we didn’t gather the bodies. It wasn’t our business.”) So even Morris has no idea how the number was reached.

As I mentioned in my essay, the historian Alon Kadish was the first to discount the number, and deeply (“it is doubtful that the number of Arabs killed on July 12 reached 250 or even half that number”). And as I noted in my earlier reply to Morris, the historian Itamar Radai has warned that in the 1948 war, 250 was probably a “symbolic” number, simply meaning “a lot”—like the estimate for Deir Yassin, where 254 turned out to be 100 (or like Teddy Katz’s discredited claim about a “massacre” at Tantura, which he also put at 250). In his latest reply, even Morris seems willing to bargain to make a point: so what if it was 200? (Perhaps if we go another round, he’ll come down some more.)


I empathize with Morris: he needs to find estimates of Arabs killed and expelled to do his work, and the main (sometimes only) source is Israeli battle reports. But from the Bible to Vietnam, enemy body counts are the most inflatable figures in history. I wasn’t shocked when I read Yehoshafat Harkabi’s matter-of-fact statement that Israeli commanders often exaggerated enemy losses in 1948. I don’t imagine the exaggeration was ever on the scale of Samson killing a thousand with a jawbone, but it would have conveyed the same message: a courageous handful of men, with few means, defeated the many.

While the “disproportion” may speak massacre to Morris today, it seems safe to assume that Mula Cohen, the Yiftah commander who signed the report of the battle, didn’t imagine he was confessing to a “massacre.” And I doubt that the editors who included it in the heroic Sefer Hapalmah in the 1950s regarded it as the smoking gun of a war crime. So just what did such “disproportion” speak at that time? A handful of courageous Palmahniks, with few means, defeated the many—and conquered Lydda.

Why does that last phrase need to be emphasized? After his earlier, guns-ablaze race past Lydda and Ramleh, Moshe Dayan met with Mula Cohen, who would remember the encounter this way:

Dayan said: “I’m going to Tel Aviv to Ben-Gurion, to inform him that I conquered Lydda.” And that’s the whole big story of “I conquered Lydda.” In the meantime, they didn’t conquer Lydda. We advanced slowly, we added a company of the 1st battalion to the 3rd battalion. Night falls, and Dayan’s battalion is gone.


So who conquered Lydda? Moshe Dayan’s 89th battalion, made up of former Lehi fighters and village boys, in its 47-minute blitzkrieg? Or Cohen’s (and Yigal Allon’s) Yiftah brigade, led by kibbutzniks, who occupied the town, repelled the Arab Legion’s incursion, put down an incipient uprising, and drove the Legion out of Lydda’s police station? Dayan’s report claimed he killed 100-150 of the enemy (“very exaggerated,” wrote the campaign’s historian); Cohen’s report claimed (counter-claimed?) 250. Are we supposed to take these numbers literally? Or did commanders err on the side of glory? It’s a valid question, for which there won’t ever be a clear answer. What’s certain is that for the persons who reported the figure of 250 enemy killed on July 12, 1948 and published it in Sefer Hapalmah a few years later, the number didn’t “speak massacre.” It shouted that the Palmah, and no one else, conquered Lydda.

As supporting evidence, Morris adduces Cohen’s statement that the conquest of Lydda put the soldiers in a vengeful mood. This, says Morris, primed them for massacre. (His quotation isn’t from a contemporary document but from a later recollection by Cohen, proving once again that Morris never relies on the flawed memory of soldiers—except when he does.) There are two problems here. First, Morris has torn the quote from its context. Cohen was explaining why the soldiers of the Yiftah brigade descended into looting. Stealing and theft, not killing, were the “outlet” that “relieved the tension.” (The full page in Sefer Hapalmah is here.)

Second, there is testimony by squad commander Hanan Sever, given even earlier than Cohen’s, to a very different mood among the conquering soldiers after they took the town:

In all the coffee houses, soldiers sat sipping fine coffee from demitasses. The tension dropped and slacked off entirely. The battle ended in carelessness, as each man turned to amuse himself. We entirely forgot the fact that we were conquerors who numbered only 300, whereas the conquered numbered in the thousands, many of whom were Arab Legion soldiers still with their weapons; in only a few minutes, the conquest could turn into a defeat. Still, for some reason, a deep sense of confidence settled in our hearts, and with the carelessness of youth, we pushed aside the last bit of caution. 

(An evocative photograph of Yiftah soldiers in Lydda, by Boris Carmi, perfectly illustrates this passage.) On the morning of the “massacre,” Sever said his own soldiers patrolled streets “in apathetic and calm relaxation,” and he had to urge them to stay alert. If Morris thinks he can establish the “massacre” by portraying Yiftah’s soldiers as prowling Lydda for revenge, he’ll have to work harder—and with oral testimony.


Morris finally hits the nail on the head with this sudden observation: “Perhaps part of the problem stems from the meaning of the word ‘massacre.’” Morris should know, since he’s confused the meaning by expanding it. When he first wrote about Lydda, in the first edition of his book on the refugees (p. 206), he didn’t use “massacre” to describe the events there, reserving it for cases such as Deir Yassin and Al-Dawayima. But in the revised edition (p. 428), he inserted the word in a sentence on Lydda without any new evidence or explanation. Lydda thereafter featured prominently on his list of 24 wartime massacres. Not only that, but Morris presumed that all 250 enemy supposedly killed in Lydda were massacred (see his book on Glubb Pasha, p. 177).

Still, despite his condemnation of other historians for denying the “massacre,” Morris himself now allows that the issue isn’t obviously self-evident. In this second response to me, he admits that the Lydda case is “more complicated” than his archetype of massacre allows. In his earlier response, he even created an entirely new sub-category for Lydda, describing it as “a massacre, albeit a provoked one”—the provocation presumably consisting of enemy bullets and grenades flying at soldiers. In fact, nowhere does Morris provide any rigorous definition of “massacre,” how it differs from “mass killing” (of which he accuses Moshe Dayan), or how it relates to “atrocities” (which he applies to Tantura). Instead, he substitutes repetition for definition—reiterating the same claim again and again, as if this established it as a fact.

I’m not a historian of 1948, but I’m a historian practicing in a much larger and more established field of study, and I know best practices when I see them. I just don’t see them in the narrative of a “massacre” at Lydda. Indeed, were it not for the bogus claim of “massacre” at Tantura, the Lydda accusation would constitute the most blatant excess of Israel’s “new historians.” As the 1948 veterans disappear, such claims have grown ever more extravagant: first, the creeping reclassification of complex battles as “massacres,” then the spread of the notion that Israel’s leaders “covered up for the officers who did the massacres,” and finally the florid elaboration of freshly discovered “massacres” in popular works ranging from “imaginative reenactments” to theatrical plays.

For the last 30 years, new myths (in the guise of “new history”) have replaced old ones (the much-derided “old history”). This process has now peaked in a single decadent sentence, written by Ari Shavit and indebted to Benny Morris: “Zionism commits a massacre in the city of Lydda.” That this misplaced and overwrought confession has gone unchallenged by American Jewish thought leaders is proof that they aren’t competent to reconstruct an accurate narrative of Israel. Perhaps a few younger readers of this exchange will be inspired to attempt the task.


Martin Kramer is president of Shalem College in Jerusalem and past director of the Moshe Dayan Center for Middle Eastern and African Studies at Tel Aviv University.

More about: Ari Shavit, Israel, Lydda, Martin Kramer, Massacre, My Promised Land, Zionism