Twenty years ago, the American Jewish Committee commissioned the Roper Organization, a reputable polling firm, to gauge public knowledge about the Holocaust. One finding in particular grabbed headlines.
- Gideon Levy / Read the Fine Print
- Yehuda Ben Meir / No to Apartheid
- Israelis Say No Vote to Arabs if W Bank Annexed
Twenty-two percent of the respondents — thirty-four if you counted those who answered ‘don’t know’ — were open to the idea that the Holocaust might not have happened. The question itself, to an untrained eye perhaps, seems straightforward enough: ‘Does it seem possible or does it seem impossible to you that the Nazi extermination of the Jews never happened?’ But the result was so shocking that, after the initial round of predictable handwringing passed, questions emerged on the methodology of the poll.
The question, when read, seems clear, but when Roper went back two years later and asked a similar number of people the same question with a slightly new wording - ‘Does it seem possible to you that the Nazi extermination of the Jews never happened, or do you feel certain that it happened?’ - the result was more reassuringly lopsided. This time, only one percent said it was possible and only eight percent answered ‘don’t know,’ leaving a full 91 per cent certain that the Holocaust had indeed taken place.
Last week a news article in Haaretz - like Mark Twain’s proverbial lie - traveled halfway round the world. Splashed across the top of its front page in the Hebrew print edition was the provocative headline ‘A majority of Israelis support apartheid policies in Israel'.
The English online version, doomed to be permanently hyperlinked in every anti-Israel blog for at least the next five years, ran under the equally risible headline 'Survey: Most Israeli Jews would support apartheid regime in Israel,’ [this headline has now been corrected to "Most Israeli Jews wouldn't give Palestinians vote if West Bank was annexed"]. Gideon Levy’s poll analysis encapsulates the intention and effect of the poll succinctly: ‘Nice to make your acquaintance, we're racist and pro-apartheid. The poll proved what we always knew.’
The poll had been conducted by Dialog and commissioned by the Yisraela Goldblum Fund. The New Israel Fund, beleaguered target of a recent right-wing witch-hunt, has distanced itself from the survey, with which it was erroneously credited with commissioning. Understandably so.
The survey and its presentation in Haaretz have all the hallmarks of a serious failure of oversight. There were four kinds of problems: misleading translations from Hebrew to English, sloppily-worded questions producing meaningless ‘findings’, dubious aggregations of the data that produced misrepresentations of opinion, and agenda-driven editorializing.
The adjoining graphic on the English edition, unlike the one in Hebrew, inexplicably leaves out the following findings which feature prominently in the Hebrew version: A majority of Israeli Jews report that they would not be bothered having an Arab neighbor in their building; a majority of Israeli Jews report that they would not be bothered if their children studied in a class with an Arab pupil; and a majority of Israeli Jews reject the proposition that Arabs not be allowed to vote in parliamentary elections. In all three cases there were worryingly large minorities who openly chose the more racist option.
Even in the original Hebrew, it's not clear what respondents are being asked to favor or oppose, and they are not given options that might clarify their choices. This is especially true on the annexation questions, which could be read as referring to three very different ideas — annexing the whole West Bank, annexing the settlement blocs, or annexing all the settlements without the Palestinian population centers.
On the question of separation of Israelis and Palestinians on roads in the West Bank, respondents were actually offered three choices: ‘it’s a good thing,’ ‘it’s a bad thing but there’s no alternative,’ and ‘it’s a bad thing and should stop.’ Haaretz in English reports the second option erroneously as ‘it is necessary’ and Levy aggregates, both in the text and in the accompanying graphic, the 24% who say it is good and the 50% who it is bad but there’s no alternative into ‘74% support separation’ (only 17% said it should stop). The natural aggregation would be that 67% say ‘it is a bad thing’, a figure that completely contradicts the conclusion of the text and graphics; perhaps Levy thought few readers would ever see the raw data, which was not linked to the articles.
It gets worse. One question asks if respondents support partially ‘transferring’ Arab-Israelis to the Palestinian Authority. This question introduces a very politically-loaded term from one context into a seemingly different one. In the 1980s there were right-wing fringe parties which advocated ‘transfer’ of Arabs to Jordan or other Arab countries in order to magically obtain a Jewish majority in the West Bank — a kind of ethnic cleansing that never garnered any real electoral support and for which there is negligible support in the public discourse. In the last decade, there has been a very different proposal in some right-wing circles to draw the international border between Israel and the future Palestinian state to include certain Arab population centers west of the Green Line in Palestine, and not as currently in Israel. What is the reader to make of 47% who say they support it or the 40% who say they oppose it? To which idea is the poll referring, and why is the provocative word ‘transfer’ used if the question actually refers to border adjustments?
Responses to the question as to whether de facto apartheid exists in Israel (not whether it should be implemented) are segmented into ‘in most ways,’ ‘in some ways,’ ‘there is no apartheid,’ and ‘don’t know.’ The question, as the data from the original poll show, was confusingly asked in the context of the remarks of a fictitious ‘American author’ ("the interviewees were asked whether "a famous American author [who] is boycotting Israel, claiming it practices apartheid" should be boycotted or invited to Israel") and not, as the Haaretz headline presents it, as a standalone assessment of the situation in Israel. Again, Levy aggregates the two affirmative responses to yield 58 percent who believe there is apartheid in Israel, but the question does not make clear whether it is referring to just the State of Israel, the West Bank or both. As Levy himself points out, it’s not entirely clear that the Israeli public at large has any real understanding of what apartheid was, and in all likelihood is using it as a term of opprobrium for institutionalized racism of any kind.
Far from expressing support for apartheid, a majority of the Israeli public is sounding the alarm about something they are clearly very critical, if a bit ignorant, of — an ignorance further demonstrated by the confused responses to an unreported question testing their historical knowledge of South Africa.
In contrast to that, the question that is the basis of the screaming headlines does not even mention the word apartheid, but rather refers to voting rights in the event of an annexation which in the previous question confusingly refers only to settlements. It’s not easy to discern what kind of annexation is being proposed in the question, and anyway the majority clearly opposes it in any form.
The Israeli public is in need of self-examination in terms of its attitudes on racism and of the corrosive effects of 45 years of ruling another people. A well-constructed public opinion survey could do much to advance a mature discussion of the problem. This survey, and Gideon Levy’s poor reading of it, gave us neither.
Shany Mor is Senior Research Associate at the Britain Israel Communications and Research Centre (BICOM). He is writing a doctorate at Oxford University and was previously a Director for Foreign Policy at the Israeli National Security Council in Jerusalem.