For the past few months, I have been visiting with interest the various political battlegrounds of Wikipedia, even dipping in to a firefight for a little dustup of my own now and then. I have had a lot of thoughts, nothing very organized yet, but I felt the time had come to write some stuff down, before senility expunges it from my addling brain. There will be a lot of rambling here, so if anyone ever reads it, be forewarned.
I think the most revealing of the political battles is the article on Muhammad al-Durrah. The process of editing that article is held as a model, for better or for worse, of managing political disputes on Wikipedia. Poor Elonka took the bull by the horns, and got pretty badly mauled in the process. Whether she will emerge in the end as the savior of Wikipedia or its Ossama bin Laden has yet to be decided. But the question I ask is, is the article better or more neutral, for it?
I think that Muhammad al-Durrah is a fine article: it is clearly written, well-organized, comprehensive, copiously documented. It does not suffer from long apologetics and polemics, as do some other controversial articles. Grand Mufti of Jerusalem is so convoluted with arguments and counterarguments that it is almost impossible to read - you need a Rashi commentary to understand it. Muhammad al-Durrah does not suffer that way - he has gotten an article that puts forth the facts as clearly and as straightforwardly as can be.
Is the article neutral? Well, when I finished the article I came away with the clear opinion that the al-Durrah story was largely a fabrication; that in all probability, if he was shot, it was by the Palestinians, not by the Israelis; and that there is some likelihood that he was not shot at all, but that the whole episode was staged.
But I also felt that the whole article could have been rewritten, using exactly the same facts and exactly the same references, leaving the exact opposite impression: that allegations that the episode was staged were an insidious propaganda effort by the pro-Israelis.
POV, presentation and fact
Consider the following two statements on casualty counts in Deir Yassin:
Official relief agencies - the Red Cross and the UN Refugee Relief Agency - placed the casualty count at 107 to 120. But eyewitnesses, supported by official Israeli spokesmen, said more than 250 villagers were killed (footnote: Historian Uri Milstein claims the Israeli leadership inflated the casualty figures to discredit the IZL fighters, who were political opponents...)
or
Arab villagers who escaped the mayhem claimed they saw more than 250 slaughtered; Israeli officials, eager to discredit their political opponents, repeated the figure (footnote: Uri Milstein). But neutral relief agencies - the Red Cross and the UN Refugee Relief Agency - put the casualty count at 107 to 120.
Who do you believe? I think, in the first paragraph you believe 250 and in the second you believe 107. Which is simply to say that neutrality is not a function of the selection of facts; it is, first and foremost, a matter of presentation.
Is the Durrah article more neutral because of Elonka's heroic, if autocratic, arbitration? I don't think so. And I don't think the article could be more neutral. Because, in the end, I believe that the Durrah article and others like it are at the extreme outer edge of the neutrality policy of Wikipedia.
WP:NPOV is based on a number of premises, and those premises are challenged in articles of extreme political controversy. The first premise is that fact is distinguishable from opinion. "By 'fact' we mean 'a piece of information about which there is no serious dispute.'" But in hot topics, matters of fact get reduced to matters of opinion. That many people were killed at Deir Yassin is a matter of fact. How many were killed would normally also be a fact, but, because it is disputed, it is a matter of opinion. The issue rapidly becomes one of how much credence you put in the various reports of casualties. It is this issue of presentation that is the crux of the dispute, not the facts themselves.
Which brings me to the second premise of WP:NPOV that is challenged by political heat: the premise that it is possible to determine the appropriate weight given to an opinion, based on the currency of that opinion. "NPOV says that the article should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each." In intensely debated articles, determining the appropriate weight to be given to a particular view is a function of the point of view making the determination. To anyone convinced of the fundamental mendacity of Israeli spokesmen and partisans, any weight whatsoever given to their opinions is undue.
There is nothing new or original in my saying this. The whole concept of narrative has become central to modern historiography. "The basic idea of conceptual history is that all key social, political, and cultural concepts are both historical and, even when not always contested, at least potentially contestable," writes Matti Hyvärinen ( [1]). Certainly, in editing disputed articles on the Middle East, the possibility of coming up with a single text that accurately reflects the narratives of the opposing camps is at best unlikely and at worst impossible. And any success in resolving these conflicting narratives will be at the expense of the cogency of the article, as is the case in Grand Mufti of Jerusalem.
A modest proposal
What is to be done? I am the last person to suggest abandoning one of the pillars of Wikipedia. I do believe, however, that true neutrality can only be achieved by granting equal platforms to opposing camps to tell their story as they see it. This does not mean abandoning the principles of comprehensiveness, reliability of sources, and academic rigor. It means allowing opposing narratives free and equal voices.
So here is my modest proposal: parallel articles. Not as a desired final result, but as an editing tactic. The objective would be to end up with a single article, wherever possible, that would satisfy both sides.
The creation of the parallel articles would be governed by a number of strict rules:
- One side may not edit the article of the other side. Editors may comment on talk pages.
- Each side must include all the references included by the other side, and make every effort to present all sides of the dispute fairly and neutrally (according to the side's understanding). You must follow all of Wikipedia's guidelines for reliable sources, NPOV, and NOR.
- Disputes about reliability of sources will be resolved in the usual ways.
- The versions would be written within the article space: Deir Yassin/ProPal and Deir Yassin/ProIsr, for example.
This process should result in two articles containing substantially the same referenced materials, but different in emphasis, in organization, and in syntax. Now we begin the task of combining the two versions. Since the content of the two articles is the same, this might actually be possible. The objective would be to build the merged article reference by reference, section by section. Where differences are irreconcilable, the irreconcilable parallel sections would remain. But I believe that gradually the two versions could, in most cases, be subsumed into a single article which would meet the approval of both sides.
Ban vulture words
Just as there is a guideline against peacocks, there should be a guideline against vultures. Vulture words are words that are loaded with POV, but, in the subculture of Wikipedia political infighters, are allowed if they are used by reliable sources. The worst offending vulture word is "massacre". I participated in quite a long polemic on this word here. My main point was that, reliable sources or no, this word has only one agreed meaning: bad guys killing good guys. All the other attempts to refine the meaning of this word failed.
The only place in Wikipedia for words like "massacre", "blood libel", "rout" and so on, in the context of current political disputes, is in direct quotes. I don't care that, to participants in the dispute, the events in Deir Yassin are called a massacre. Since at least one party in the dispute (fringe though it may be) denies there was a massacre, the word should not be used.
There is a contest among pro-Israelis and pro-Palestinians to see who can get more articles into the Wikipedia with the word "massacre" in the title. This has resulted in a proliferation of articles which probably shouldn't exist at all. By unofficial count, the pro-Israelis are winning, with 12 articles, to the pro-Palestinians' 11. The pro-Palestinians have Israel and the apartheid analogy, which almost evens the score (I think that one apartheid is worth half a massacre). However, the pro-Israelis recently succeeded in getting blood libel and Deir Yassin into the title of one article, so they still have a decisive lead.
I have no objection to the use of the word massacre to describe historic events unrelated to current political conflicts. The Boston Massacre, the Saint Valentine's Day massacre and The Texas Chainsaw Massacre are all fine.
Politicization strategies
I found the working group's discussion of tag teams to be fascinating. Tag-teaming joins the repertoire of tactics (sock-puppetry, meat-puppetry, and so on) used to politicize the Wikipedia.
What I didn't see in the working group's final report was a discussion of politicization strategies, as opposed to tactics. Here are two that I think are worthy of study, and perhaps action:
Multiplication of articles
Proponents of a particular political position try to multiply the number of Wikipedia articles supporting their side. The Pro-Palestinians have, for example, written 153 articles on Arab villages depopulated as a result of the conflict. Almost all these articles are stubs, and are likely to remain so, as very little information is available about them. The pro-Israelis, not to be outdone, have 67 articles to their credit on Israeli settlements depopulated during the conflict.
I am not saying that each of these villages is not, theoretically, a legitimate topic for an article. As things stand, though, there is no question that almost all of the information in these articles could have been presented more succinctly and accessibly in other forms - for example in a table, showing the village size, population, date and circumstances of the depopulation. In those few cases where there is more information than this included in the article, a separate article would certainly be in order.
But I contend that the objective of these articles is not to present the information in the way that is most accessible, but rather in the way that maximizes the propaganda value.
Of course, it is best if you can make an article with some really inflammatory rhetoric in the title. The pro-Israelis get the prize for this with Blood Libel at Deir Yassin - an article about a truly marginal book by an author of dubious academic acceptance. The article gets the blood libel and the village into the article title in a way that is unassailable - it is the name of the book.
Undue weight
Here I am not referring to the weight of an opinion within an article (as used in WP:NPOV), but to the weight of the article itself. The al-Durrah article, for example, is 100 kilobytes long. The article on the 1948 war is 87 kilobytes. Was the al-Durrah episode really more significant than the 1948 war?
This disproportion is inherent in the nature of the Wikipedia. It is, in a sense, unfair to compare the two articles - the 1948 Arab-Israeli War is a summation of a much larger topic, while the article on al-Durrah is a comprehensive coverage of a recent dispute. Nonetheless, the amount of verbiage expended on topics that are hot but, in the long view, marginal, unquestionably has the effect of increasing their importance in the eyes of the reader - in other words, it is good propaganda.
In Groves Dictionary of Music and Musicians, Franz Schubert gets 56 pages; Zdenek Nejedly gets half a page. The size of an article in an encyclopedia is a reflection of the editors' estimate of its relative importance. Not so in the Wikipedia.
No solution?
The problems of article multiplication and of undue weight are inherent in the very idea of the Wikipedia. I do not see how they can be addressed. But I just wanted to raise them here. These strategies deserve serious study, and I don't doubt that someone will study them. Any doctoral candidates out there looking for a thesis topic?
For the past few months, I have been visiting with interest the various political battlegrounds of Wikipedia, even dipping in to a firefight for a little dustup of my own now and then. I have had a lot of thoughts, nothing very organized yet, but I felt the time had come to write some stuff down, before senility expunges it from my addling brain. There will be a lot of rambling here, so if anyone ever reads it, be forewarned.
I think the most revealing of the political battles is the article on Muhammad al-Durrah. The process of editing that article is held as a model, for better or for worse, of managing political disputes on Wikipedia. Poor Elonka took the bull by the horns, and got pretty badly mauled in the process. Whether she will emerge in the end as the savior of Wikipedia or its Ossama bin Laden has yet to be decided. But the question I ask is, is the article better or more neutral, for it?
I think that Muhammad al-Durrah is a fine article: it is clearly written, well-organized, comprehensive, copiously documented. It does not suffer from long apologetics and polemics, as do some other controversial articles. Grand Mufti of Jerusalem is so convoluted with arguments and counterarguments that it is almost impossible to read - you need a Rashi commentary to understand it. Muhammad al-Durrah does not suffer that way - he has gotten an article that puts forth the facts as clearly and as straightforwardly as can be.
Is the article neutral? Well, when I finished the article I came away with the clear opinion that the al-Durrah story was largely a fabrication; that in all probability, if he was shot, it was by the Palestinians, not by the Israelis; and that there is some likelihood that he was not shot at all, but that the whole episode was staged.
But I also felt that the whole article could have been rewritten, using exactly the same facts and exactly the same references, leaving the exact opposite impression: that allegations that the episode was staged were an insidious propaganda effort by the pro-Israelis.
POV, presentation and fact
Consider the following two statements on casualty counts in Deir Yassin:
Official relief agencies - the Red Cross and the UN Refugee Relief Agency - placed the casualty count at 107 to 120. But eyewitnesses, supported by official Israeli spokesmen, said more than 250 villagers were killed (footnote: Historian Uri Milstein claims the Israeli leadership inflated the casualty figures to discredit the IZL fighters, who were political opponents...)
or
Arab villagers who escaped the mayhem claimed they saw more than 250 slaughtered; Israeli officials, eager to discredit their political opponents, repeated the figure (footnote: Uri Milstein). But neutral relief agencies - the Red Cross and the UN Refugee Relief Agency - put the casualty count at 107 to 120.
Who do you believe? I think, in the first paragraph you believe 250 and in the second you believe 107. Which is simply to say that neutrality is not a function of the selection of facts; it is, first and foremost, a matter of presentation.
Is the Durrah article more neutral because of Elonka's heroic, if autocratic, arbitration? I don't think so. And I don't think the article could be more neutral. Because, in the end, I believe that the Durrah article and others like it are at the extreme outer edge of the neutrality policy of Wikipedia.
WP:NPOV is based on a number of premises, and those premises are challenged in articles of extreme political controversy. The first premise is that fact is distinguishable from opinion. "By 'fact' we mean 'a piece of information about which there is no serious dispute.'" But in hot topics, matters of fact get reduced to matters of opinion. That many people were killed at Deir Yassin is a matter of fact. How many were killed would normally also be a fact, but, because it is disputed, it is a matter of opinion. The issue rapidly becomes one of how much credence you put in the various reports of casualties. It is this issue of presentation that is the crux of the dispute, not the facts themselves.
Which brings me to the second premise of WP:NPOV that is challenged by political heat: the premise that it is possible to determine the appropriate weight given to an opinion, based on the currency of that opinion. "NPOV says that the article should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each." In intensely debated articles, determining the appropriate weight to be given to a particular view is a function of the point of view making the determination. To anyone convinced of the fundamental mendacity of Israeli spokesmen and partisans, any weight whatsoever given to their opinions is undue.
There is nothing new or original in my saying this. The whole concept of narrative has become central to modern historiography. "The basic idea of conceptual history is that all key social, political, and cultural concepts are both historical and, even when not always contested, at least potentially contestable," writes Matti Hyvärinen ( [1]). Certainly, in editing disputed articles on the Middle East, the possibility of coming up with a single text that accurately reflects the narratives of the opposing camps is at best unlikely and at worst impossible. And any success in resolving these conflicting narratives will be at the expense of the cogency of the article, as is the case in Grand Mufti of Jerusalem.
A modest proposal
What is to be done? I am the last person to suggest abandoning one of the pillars of Wikipedia. I do believe, however, that true neutrality can only be achieved by granting equal platforms to opposing camps to tell their story as they see it. This does not mean abandoning the principles of comprehensiveness, reliability of sources, and academic rigor. It means allowing opposing narratives free and equal voices.
So here is my modest proposal: parallel articles. Not as a desired final result, but as an editing tactic. The objective would be to end up with a single article, wherever possible, that would satisfy both sides.
The creation of the parallel articles would be governed by a number of strict rules:
- One side may not edit the article of the other side. Editors may comment on talk pages.
- Each side must include all the references included by the other side, and make every effort to present all sides of the dispute fairly and neutrally (according to the side's understanding). You must follow all of Wikipedia's guidelines for reliable sources, NPOV, and NOR.
- Disputes about reliability of sources will be resolved in the usual ways.
- The versions would be written within the article space: Deir Yassin/ProPal and Deir Yassin/ProIsr, for example.
This process should result in two articles containing substantially the same referenced materials, but different in emphasis, in organization, and in syntax. Now we begin the task of combining the two versions. Since the content of the two articles is the same, this might actually be possible. The objective would be to build the merged article reference by reference, section by section. Where differences are irreconcilable, the irreconcilable parallel sections would remain. But I believe that gradually the two versions could, in most cases, be subsumed into a single article which would meet the approval of both sides.
Ban vulture words
Just as there is a guideline against peacocks, there should be a guideline against vultures. Vulture words are words that are loaded with POV, but, in the subculture of Wikipedia political infighters, are allowed if they are used by reliable sources. The worst offending vulture word is "massacre". I participated in quite a long polemic on this word here. My main point was that, reliable sources or no, this word has only one agreed meaning: bad guys killing good guys. All the other attempts to refine the meaning of this word failed.
The only place in Wikipedia for words like "massacre", "blood libel", "rout" and so on, in the context of current political disputes, is in direct quotes. I don't care that, to participants in the dispute, the events in Deir Yassin are called a massacre. Since at least one party in the dispute (fringe though it may be) denies there was a massacre, the word should not be used.
There is a contest among pro-Israelis and pro-Palestinians to see who can get more articles into the Wikipedia with the word "massacre" in the title. This has resulted in a proliferation of articles which probably shouldn't exist at all. By unofficial count, the pro-Israelis are winning, with 12 articles, to the pro-Palestinians' 11. The pro-Palestinians have Israel and the apartheid analogy, which almost evens the score (I think that one apartheid is worth half a massacre). However, the pro-Israelis recently succeeded in getting blood libel and Deir Yassin into the title of one article, so they still have a decisive lead.
I have no objection to the use of the word massacre to describe historic events unrelated to current political conflicts. The Boston Massacre, the Saint Valentine's Day massacre and The Texas Chainsaw Massacre are all fine.
Politicization strategies
I found the working group's discussion of tag teams to be fascinating. Tag-teaming joins the repertoire of tactics (sock-puppetry, meat-puppetry, and so on) used to politicize the Wikipedia.
What I didn't see in the working group's final report was a discussion of politicization strategies, as opposed to tactics. Here are two that I think are worthy of study, and perhaps action:
Multiplication of articles
Proponents of a particular political position try to multiply the number of Wikipedia articles supporting their side. The Pro-Palestinians have, for example, written 153 articles on Arab villages depopulated as a result of the conflict. Almost all these articles are stubs, and are likely to remain so, as very little information is available about them. The pro-Israelis, not to be outdone, have 67 articles to their credit on Israeli settlements depopulated during the conflict.
I am not saying that each of these villages is not, theoretically, a legitimate topic for an article. As things stand, though, there is no question that almost all of the information in these articles could have been presented more succinctly and accessibly in other forms - for example in a table, showing the village size, population, date and circumstances of the depopulation. In those few cases where there is more information than this included in the article, a separate article would certainly be in order.
But I contend that the objective of these articles is not to present the information in the way that is most accessible, but rather in the way that maximizes the propaganda value.
Of course, it is best if you can make an article with some really inflammatory rhetoric in the title. The pro-Israelis get the prize for this with Blood Libel at Deir Yassin - an article about a truly marginal book by an author of dubious academic acceptance. The article gets the blood libel and the village into the article title in a way that is unassailable - it is the name of the book.
Undue weight
Here I am not referring to the weight of an opinion within an article (as used in WP:NPOV), but to the weight of the article itself. The al-Durrah article, for example, is 100 kilobytes long. The article on the 1948 war is 87 kilobytes. Was the al-Durrah episode really more significant than the 1948 war?
This disproportion is inherent in the nature of the Wikipedia. It is, in a sense, unfair to compare the two articles - the 1948 Arab-Israeli War is a summation of a much larger topic, while the article on al-Durrah is a comprehensive coverage of a recent dispute. Nonetheless, the amount of verbiage expended on topics that are hot but, in the long view, marginal, unquestionably has the effect of increasing their importance in the eyes of the reader - in other words, it is good propaganda.
In Groves Dictionary of Music and Musicians, Franz Schubert gets 56 pages; Zdenek Nejedly gets half a page. The size of an article in an encyclopedia is a reflection of the editors' estimate of its relative importance. Not so in the Wikipedia.
No solution?
The problems of article multiplication and of undue weight are inherent in the very idea of the Wikipedia. I do not see how they can be addressed. But I just wanted to raise them here. These strategies deserve serious study, and I don't doubt that someone will study them. Any doctoral candidates out there looking for a thesis topic?