Wikipedia talk:No original research

The project page associated with this talk page is an official policy on Wikipedia. Policies have wide acceptance among editors and are considered a standard for all users to follow. Before editing this page, please make sure that your revision reflects consensus. When in doubt, discuss on the talk page. Always remember to keep cool when editing. Changes to this page do not immediately change policy anyway, so don't panic.

Citation This page was cited by Stvilia, B. et al. Information Quality Discussions in Wikipedia. University of Illinois U-C.

The spoken word version of this revision (diff) of this article is part of WikiProject Spoken Wikipedia, an attempt to produce recordings of Wikipedia articles being read aloud. If you would like to participate, visit the project page, where you can join the project and find out how to contribute.

Archives (Index)
1, 2, 3, 4, 5, 6, 7, 8, 9, 10 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 31, 32, 33, 34, 35, 36 This talk page is automatically archived by MiszaBot II. Any sections older than 7 days are automatically archived. Sections with fewer than two timestamps (no replies) are not archived.
About archives

We need some exceptions to the policy

I think this policy on Wikipedia is a bit annoying, and promotes deprivation of some information that could be vital. For example, I am under the impression that MP3 players are a probable replacement to radio (more to this topic can be read in Talk:digital audio player). However, my efforts to claim that ordeal on Wikipedia always get reverted (owing to this policy).

Besides, I think peoples' personal experience with concepts may be a basis for public awareness that is largely deprived as people could be missing out on experience that ancillary advantages of concepts may provide. Besides, people are so under some impressions that they forget the research is "too original" thus they get annoyed.

So, will the admins legislate some exceptions to this policy so Wikipedians can be less annoyed by it? Because I know I am. --Roadstaa (talk) 02:39, 2 July 2008 (UTC)

Admins don't legislate policy, the community does. You could create a proposal, but it's not too likely to succeed. WP:NOR is one of the core things that makes Wikipedia not "just a bunch of guys on the internet makin' stuff up." Somedumbyankee (talk) 02:57, 2 July 2008 (UTC)

But other websites can make information "sourceable" if the rumor becomes rampant elsewhere. --Roadstaa (talk) 03:14, 2 July 2008 (UTC)

Nominally we shouldn't be using just anyone as a source. Check out WP:SOURCES for criteria. Just because the rumor mill grinds it out doesn't mean it should be on here. Somedumbyankee (talk) 03:23, 2 July 2008 (UTC)

Further to my suggestion above - how feasible would it be to develop a policy along the lines of "this OR is more appropriate for (Wiki X/Website Y than for Wikipedia"? Some cases will always be marginal/personal experience/leading edge etc but it may be possible to place some work more appropriately than Wikipedia. Jackiespeel (talk) 14:02, 7 July 2008 (UTC)

I think we need exceptions to this policy. I did some research on a game to collect statistics about certain aspects of the game. However, I didn't realize until after I posted the information that no original research could be used. What I'm saynig is, I think that original research should be allowed if it has ___ people who agree, and can verify that said information is true. The underscores above indicate a number of people that would be needed to verify the information. --MaggotSoldier (talk) 17:42, 22 July 2008 (UTC)

It's understandable that you would want to get information you believe is correct and important out to the world, but Wikipedia has a very longstanding policy of not attempting to vet new research itself but relying on previously-established channels to do that, for a host of reasons. This means that if you have new ideas you want to present, you're probably better off presenting them in a different forum. Having the policy means there's going to be a lot of very valuable stuff that has to be turned down. Suggest finding a more open web site to present it. Best, --Shirahadasha (talk) 22:32, 22 July 2008 (UTC)

That is understandable. If I present my information on another forum, is that a credible source, or will everything I find information-wise not be accepted here, even if it is on another site, due to the fact that I did, indeed, gather it myself? --MaggotSoldier (talk) 02:04, 23 July 2008 (UTC)

If that "forum" is the National Academy of Sciences or the Communist Party's General Meeting, sure. If it's a forum on a website that anyone can post to or a personal website, that's not a reliable source. SDY (talk) 02:10, 23 July 2008 (UTC)

...and obviously there are many shades of grey in between. In the case of first-person research, there are usually better specialty sites where such details can be published, for access and discussion by people who share the same interests and experience (i.e. who can spot bullshit). Often on such sites there will be great subject-matter experts, who will be able to critique and expand on the issues you might raise. These sites may or may not rise to the level of peer review where they could be suitable for reference by WikiPedia articles.

If you think about it, it makes perfect sense that there should be lots of sources that are full of accurate information, but that nevertheless don't have the level of peer review and widespread notability expected for WikiPedia content. (And of course not all WikiPedia articles meet this standard, but that is the goal.) The only way for a resource like WikiPedia to maintain easy UNIVERSAL access is to have HIGH standards for source quality. For tighter groups, with shared specialty interest, it is possible to take more on faith – because the participants are already part of a circumscribed community, one that imposes its own quality standards through dialogue and the personal reputation of participants. (For example: If you post a bunch of inane bilge on a game site, the other regular contributors will flame you until you shut up. But that won't happen here. Instead, the standard for getting flamed on WikiPedia is posting stuff with no credible third party sources.) Hope this perspective is useful. Spinality (talk) 06:42, 23 July 2008 (UTC)

Application of definitions: synthesis?

Is applying a definition of a term an act of SYN?

In other words, for example, if the definition of discrimination is to treat people differently on a basis other than merit, and I can show source indicating a case where people are treated differently according to merit, is it SYN to therefore say that the case was one of discrimination, is that synthesis? Do I need an explicit source that actually associates the exact word with the topic? - Keith D. Tyler ¶ 00:54, 11 July 2008 (UTC)

It's not the spirit of SYN to prohibit trivial logical deductions, but in cases like the one you describe, you have to consider the connotations of your word choice. For example, a simple change of wording: Employees were discriminated according to merit is pretty bland in my consideration, but XYZ Corporation discriminates against its employees is blantantly displaying a context probably unintended by the reference. In a sense, whenever we choose to paraphrase a source instead of directly quote it (which is virtually all of the time) we are exercising some form of original research. But as long as our rewording gets across no more ideas than the references did, we're in the clear. As such, it's often a good idea to be careful around or avoid words that can easily be taken out of context, especially into a negative context. Someguy1221 (talk) 01:47, 11 July 2008 (UTC)

This is not only WP:SYN, but a classic example of it, and something I or any other administrator would immediately remove if it was done. "Discrimination" is a legal wrong, and any claim that a person committed a legal wrong has to be very carefully reliably sourced to sources who clearly and explicit say "discrimination." There are reams of law books and court opinions about what discrimination is and how the term applies in particular cases, and you can take a whole law school course on the subject and come out more confused than when you came in. You may think you know it when you see it, but Wikipedia can no more accept an editor's personal say-so on whether something is "discrimination" than it could accept a editor's say-so on whether something is kosher or obscene. Best, --Shirahadasha (talk) 05:52, 11 July 2008 (UTC)

Discrimination as a word far predates the modern implications, and it is still used today in its innocuous form (Oh, and it's not always a legal wrong by the cultural definition, anyway). The suggestion that the mere use of this word where the source did not is quite reaching of a statement; as I stated, avoidance of culturally charged words, or the negative connotations thereof, is sufficient to avoid SYN in this example. If Keith had been planning my second italicized example, you would be correct in removing it under this reasoning, but not in the case of the first, although it still may not be the best word choice. Someguy1221 (talk) 07:02, 11 July 2008 (UTC)

I don't agree with your definition at all, and I doubt many sources on discrimination would support it. Gobs and gobs and gobs of discrimination has (and does) occur within the confines of the law of the land. Nor does the definition of the term dictate a value judgment, so neither "legal" or "wrong" are relevant terms. And WP:NPOV prevents us from applying one-sided value judgments. Every form of discrimination has had its proponents who saw it as a good thing, so this notion that we must tread lightly because we apply or presume "wrong"ness (or "good"ness) to the concept is... well, POV.

I think focusing on applying presupposed value judgments to a term and then using these strawmen to dictate how we apply them in a neutral-POV encyclopedia is missing the point of what I'm trying to figure out. - Keith D. Tyler ¶ 16:40, 11 July 2008 (UTC)

Uh... wait, so you're saying that because a topic is currently considered to be negative, regardless of its true neutral meaning, it has to be carefully applied? That flies in the face of NPOV in my opinion... If the neutral definition of the term applies to a neutral description of the topic, what's the problem? Besides, this has nothing to do with SYN but with some sort of non-objective sensitivity issue. I don't see how using the passive voice in your example makes a difference, either ("employees of XYZ were discriminated against" versus "company XYZ discriminates against its employees" is the same thing, except one is in poor prose in order to pussy-foot around some company's PR image.) - Keith D. Tyler ¶ 16:32, 11 July 2008 (UTC)

Because good intentions, grammatical accuracy, and the neutral definition althogether are still insufficient from preventing people from having a bad reaction. Company XYZ discriminates against its employees is entirely accurate in this case, but invokes negative connotations unintended by the source. Like Shirahadasha did above, many people are going to look at that statement and assume the intended implication is that XYZ is an evil company that treats its employees unfairly. Where such deductions are possible, it is far easier to avoid certain words or skillfully construct sentences such that any negative deduction unintended by the source is impossible to arrive at; this is far easier than inviting edit wars and debates on the meaning of words every time someone misinterprets the sentence. Whether this is actually a SYN problem is as unclear as the meaning of the sentence that is written. Someguy1221 (talk) 20:41, 11 July 2008 (UTC)

Fine, though I disagree... so, then, is it valid to include this story on a page called Workplace discrimination... or categorize it in Category:Workplace discrimination? - Keith D. Tyler ¶ 21:38, 11 July 2008 (UTC)

I don't think that would be appropriate, given that merit-based discrimination is so incomparable to the other forms of discrimination listed at that article. In fact, Webster's considers the two connotations of discrimination so distinct it gives them seperate definitions. Thus, the theme of Workplace discrimination is utterly distinct from merit-based discrimination, which should not be listed there on the technicality that it is called by the same word. Someguy1221 (talk) 21:59, 11 July 2008 (UTC)

Comment: Using a phrase like "XYZ Corporation discriminates against its employees" is mentioning discrimination in the context of a workplace, and as Someguy1221 points out, such a use tends to conjure up an impression that XYZ Corporation is doing something illegal. One has to be careful about giving such an impression because it could be libel. Using the word "discriminate" in a different context, (such as saying "The food critic showed a discriminating palate") might not be a problem. Best, --Shirahadasha (talk) 07:31, 13 July 2008 (UTC)

I don't understand your logic. I thought Wikipedia was WP:NPOV, not dependent on some point of view, particularly one that has no basis. Like I said, many forms of discrimination are completely legal. Why should the content of WP be based on some people's false beliefs? That's not NPOV -- that's succumbing to one POV. Maybe workplace discrimination is a bad example, I don't know, the point of the question is being lost in reinterpretations and this notion that, well, some people will think something negative, so don't say it, even though that conclusion is presumptory and the characterization is perfectly accurate. So I have to come up with a synonym for "workplace discrimination" or "discrimination" itself that follows a subjective rule of thumb... which is forking, and leaves the original fork to be hopelessly defined by POV instead of by NPOV. Can you reconcile this dilemma for me? - Keith D. Tyler ¶ 15:51, 14 July 2008 (UTC)

We are required to abide by the law, whether we think it has a basis or not. This includes refraining from unsourced statements likely as a practical matter to convey an impression that negatively affects reputations, whether we think it should or not. Best, --Shirahadasha (talk) 16:06, 14 July 2008 (UTC)

That's a non-answer. I put it to you that the "law" says that we should not allow POV to influence the content of the encyclopedia, either by addition or subtraction. If there is a neutral, objective basis for a characterization or categorization, saying "it makes someone feel bad" is not a sufficient argument for avoiding it. NPOV is a pillar -- "making sure people feel" good is not. The question I asked, which relates to OR, isn't answered: Can I categorize duck under waterfowl or say that it is a waterfowl in the content, if I have a source that says it is a bird that spends its time in water, or do I actually have to have a source explicitly saying the words "A duck is a waterfowl" ? - Keith D. Tyler ¶ 23:10, 14 July 2008 (UTC)

This is actually very simple, and it has nothing to do with abiding by any rules, just common sense. You can even throw out all concerns about NPOV, OR, SYN, and RS...it's still very simple. Naturally, a goal of this project is to be reasonably comprehensible. So whenever we write something that may be reasonably misinterpreted, we should endeavor to make the actual meaning quite clear. It need not be a sensitive issue of possible negative perceptions; if you can bet that a good fraction of people will make the wrong conclusion from your words, you should reword them! That the alternative but still reasonable interpretation in this case is negative is a reason to be extra careful only because it is more likely to trigger edit wars and accusations of bad faith. So I'm looking at you here, and I see you have the choice between being correct and understood, or correct and misunderstood...I'm not asking you compromise the integrity of Wikipedia in consideration of others' feelings. Rather, I'm trying to advise you on how to add the same information with a minimum of fuss, because if you do what you were suggesting, then other editors will think you're violating this, that and the other thing. Someguy1221 (talk) 23:23, 14 July 2008 (UTC)

What would the wrong interpretation be? Ahh, perhaps you are saying there is "good" discrimination and "bad" discrimination, and perhaps a person might think I mean the "bad" one when I really mean the "good" one. The problem is... "good" and "bad" have no place here. Both are applications of POV. - Keith D. Tyler ¶ 05:13, 21 July 2008 (UTC)

"if the definition of discrimination is to treat people differently on a basis other than merit" Hold your horses. Who decides the what deeds are meritous or not? Recently in my country Henry Morgentaler was awarded the order of Canada. Those who awarded him said pointed to his unwavering commitment to the woman's right to abortion. Those who opposed him, pointed to the exact same actions of the doctor, and said he was responsible placing a "scourge[of abortion]" on Canada. Two very different views on the same deed.

Also what about treating people based on need, as opposed to merit? Is that discrimination? Surely everyone would agree that a hospital's treatment of some patients differently than others can't be called "discrimination".

But even if we do agree with the above definition, it is way too broad. Under it would fall international borders (discrimination based on national origin). Is that discrimination?

In short I'm making a case to stick to wikipedia's spirit of sourcing. If something is "obviously" an example of discrimination, then shouldn't that "obvious" observation have been made by reliable sources? Bless sins (talk) 00:24, 15 July 2008 (UTC)

[1]. - Keith D. Tyler ¶ 05:13, 21 July 2008 (UTC)

• Regarding the word "discrimination" itself. Why not just use the word the sources uses, instead of using "discrimination" (where the souce doesn't use it)? Why can't one be satisfied by simply saying what the source says? This makes life much simpler in my view.Bless sins (talk) 00:30, 15 July 2008 (UTC)

I think the word "discrimination" does have to be handlded with special care. We have to abide by the laws governing the Foundation,including the libel laws, and this means we sometimes have to take into account the rather strange ways the law sometimes looks at things. One of these oddities is we have to be careful using certain legal "magic words" that connote illegal conduct or similar issues, even though in ordinary English (as distinct from legalese) the very same words may have a different meaning. For better or for worse, "discriminate" is one of those special magic words that have a special meaning in legalese. The law basically compels us to go by the legalese meaning, even though the ordinary English meaning is different. The ordinary English meaning of "discriminate" might well be what you say it is and be no problem here, but the legalese meaning has to be the one we use here. The law sometimes leads to results that don't always seem logical. It's just the way it is. Best, --Shirahadasha (talk) 20:34, 20 July 2008 (UTC)

If third party sources show that X does Y, whether or not they use the word Y but a synonym, then Y is acceptable. Otherwise, every category will have to be split into all of its synonyms, based on what word the sources used to categorize the topic used. Madness, nonsense, and not the sort of exercise we should undertake. - Keith D. Tyler ¶ 05:08, 21 July 2008 (UTC)

I'll just point out the classic, long-standing example of synthesis given in the policy:

"If Jones's claim that he consulted the original sources is false, this would be contrary to the practice recommended in the Chicago Manual of Style, which requires citation of the source actually consulted. The Chicago Manual of Style does not call violating this rule "plagiarism". Instead, plagiarism is defined as using a source's information, ideas, words, or structure without citing them." -This entire paragraph is original research, because it expresses the editor's opinion that, given the Chicago Manual of Style's definition of plagiarism, Jones did not commit it.

As the example illustrates, WP:SYN has long prohibited the specific thing being done here -- an editor claiming a person has or has not done a wrong based on the editor's own application of a definition of that wrong to a set of facts. That's what the plagiarism example says can't be done, and that's what's happening in the example you gave here. Hope that helps. If you're requesting that this longstanding interpretation of the policy be changed, I wouldn't support such a change. The fact that the example you're giving is (like the plagiarism example) one where claims of legal wrong are involved, together with the fact that claims of legal wrong are fraught with pitfalls, makes me even less inclined to dare to tread, let alone rush in. Best, --Shirahadasha (talk) 06:16, 21 July 2008 (UTC)

You keep asserting that the definition of discrimination is necessarily tied to the notion of a legal wrong, but this is not supported by any source on the topic of discrimination. So I simply do not accept this basis for your argument, because it's unsubstantiated. Countless forms of discrimination both today and in the past are firmly within the law and many are even explicitly directed by law (black codes, affirmative action, ketuanan melayu, reservation in India, the list goes on). Go and read discrimination or many of the other articles in Wikipedia on discrimination topics to review. To assume that discrimination is "wrong" or "bad" is to endorse a particular POV. NPOV dictates that we assume a neutral point of view. You suggest that there is some sort of unwritten guideline of worrying about false assumptions that overrides NPOV but I don't see it.

Take the following source. [2] I can say based on this source that (as of the time that page was current) that "Jews are not permitted to enter Saudi Arabia." Yes? But I can't say "Saudi Arabia discriminates who can enter the country based on Jewishness" because the Saudi government does not use the word "discrimination" to describe their policy, even though the definition of religious discrimination is "treating a person or group differently because of what they do or do not believe"? - Keith D. Tyler ¶ 00:35, 22 July 2008 (UTC)

It would be preferable to have an independent reliable source to provide some context (and establish the noteworthiness of the fact), but your example is correct. Stick to the plain reporting of what the sources state. If you need to make an argument that the data in a source fits a particular interpretation or if you require multiple sources to establish a claim, it is very likely to be considered original research and inappropriate for Wikipedia. Vassyana (talk) 08:21, 22 July 2008 (UTC)

I can't give legal advice. Please contact a lawyer about the meaning of legal terms. Please contact your legislator if, after that advice, you think the law should be different from what it is. Best, --Shirahadasha (talk) 01:15, 22 July 2008 (UTC)

In terms of wikipedia content, "discrimination" is an emotionally-loaded word which many people react to immediately rather than perceiving it in its technically neutral way, so it would seem to many editors, myself included, to qualify under Wikipedia:Words to avoid. If there are other, more clearly neutral terms available, which do not have the emotionally-charged nature that word does, they should be used instead. In this case, I have every reason to believe that the point could be made just as clearly with alternate phrasing. John Carter (talk) 01:38, 22 July 2008 (UTC)

"Discrimination" is the neutral term. - Keith D. Tyler ¶ 08:02, 27 July 2008 (UTC)

This isn't a matter of law, nor is there a legal term involved, and I find your insistence on it without any basis, and you won't give me one. - Keith D. Tyler ¶ 07:03, 24 July 2008 (UTC)

published court decision

The WP:OR article defines trial transcripts as primary sources, and thus out of bounds. But what about a published decision? In my opinion, published decisions should be considered secondary sources (for the limited purposes of WP:OR), and potentially citable/quotable in WP. One relevant passage from WP:OR is: "Primary sources that have been published by a reliable source may be used in Wikipedia, but only with care..." Certainly a published court decision meets this qualification. Verklempt (talk) 06:30, 26 July 2008 (UTC)

I think you have a misconception as to what the policy is saying. While WP:OR defines trial transcripts as primary sources, that does NOT mean they are "out of bounds". Primary sources can be used. The key is to use them with care... and to not use them in ways that constitute Original Research. For example, a trial transcript might be appropriately used for a statement as to the opinion of a witness about some aspect of the case, but it should not be used to support a statement of fact about the case.

In the case of a published court decision... while such documents are indeed primary, they are definitely reliable published sources. If you stick to the source (ie limit the discussion to what is contained in the decision), and avoid drawing any conclusions from it... avoid interpreting it, analizing it or saying what it means... you should be fine.

All that said, I would have to see how it is being used in a specific article to know if you are using it appropriately or not. Blueboar (talk) 14:41, 26 July 2008 (UTC)

The classification of a source as primary versus secondary depends on context – on what elements from the source are being used to substantiate what points in the article. The majority of publications are primary sources for some new theses, syntheses, facts, findings, analyses, etc., because without offering some novelty, the only nonfiction works that would get published are textbooks, surveys, and encylopaedias. (If we forbade the use of all sources that contain ANY primary material, we would essentially limit ourselves to tertiary sources.)

In the case of a published court transcript, such a document would clearly be a primary source for details related to the subject matter of the case – it would be an error to cite deposition X as establishing fact Y. Fact-finding is left to judges and juries.

However, a properly published court transcript is a secondary source for some factual points – such as venue, parties to the case, dates of proceedings, and perhaps (depending on publication medium, transcription method, etc.) for reliable transcripts of verbatim testimony, establishing what statements were made by whom. Again, the transcript would not establish truth of falsehood of issues being addressed by the case, but rather about facts and events related to the court proceedings. Thus we could say "the transcript establishes that Smith said X" but not "the transcript shows X.") This at least is my reading of policy. Spinality (talk) 20:36, 26 July 2008 (UTC)

I believe an appellate decision, particularly a decision of a court like the United States Supreme Court, can be treated quite differently from a trial transcript and will often be usable as a secondary source. Appellate opinions meet all the necessary qualifications. Appellate judges are specially vetted for expertise and reliability. They check facts. Particularly for Supreme Court decisions, the very fact that they take and publish a case (The Supreme Court has a published opinion on only 1 or 2 percent of the cases appealed) is a reliable indication that the subject matter is notable and important in the legal world. Trial transcripts, on the other hand, are completely different. They often reflect mere allegations. The papers people file and the things people say at a trial aren't necessarily reliable, and there's no guarantee that anyone reliable has vetted them. Best, --Shirahadasha (talk) 07:58, 27 July 2008 (UTC)

RfC related to WP:SYNTH

I have been involved in a content dispute with an editor largely about WP:SYNTH for several months. There is (another?) RfC open here. There are several problems (notably COI), and we're clearly dealing with a WP:SPA, but I think that the fundamental one is that the editor truly doesn't "get" WP:NOR.

The background is this: The editor has self-published a book on his idea about a medical condition. Since the article he wrote specifically on his idea was deleted as non-notable, he's been steadily trying to introduce his idea into the Wikipedia article on a related, vaguely defined medical condition from the 19th century. Under threat of ban for obvious COI violation, he now confines himself to long complaints on the talk page.

His past work and talk page arguments primarily involve (carefully selected aspects of) primary sources. The rest is the addition of related conditions (like Chronic fatigue syndrome) that line up with his POV (and surely they're practically the same thing, because many of the symptoms are the same). He asserts that his work can't be "original research" because some of the studies are more than 100 years old. Thus he believes that he's only reporting someone else's original research, and (inevitably) drawing conclusions from those papers that just happen to support his novel, non-notable idea.

Now you know, and I know, that drawing novel conclusions from primary medical studies is practically the definition of WP:SYNTH. But this editor doesn't seem to know that. I am honestly wondering whether he's ever read this policy. Instead, he seems to think that I (and the other involved editors) are just invoking SYNTH as a form of anti-Truth™ wikilawyering.

I don't think that he needs a bunch of people to jump on him -- he's probably feeling pretty embattled already, and he's remarkably unfamiliar with basic WP conventions, like "{{Otheruses4}} belongs at the top of an article", so I think WP:BITE might apply -- but I do think it might be helpful if several editors who are particularly familiar with this policy would leave a polite note on the RfC's talk page to indicate that, indeed, the community consensus is that drawing your own novel conclusions from 100-year-old medical reports really is a violation of WP:SYNTH, and not just something that the other editors have invented as a means of obstructing him. WhatamIdoing (talk) 17:20, 26 July 2008 (UTC)

Primary vs. secondary

A phrase from Emanuel Ringelblum is literally quoted in History of the Jews in Poland as an alleged secondary source. I believe it's still a primary source if we don't quote the comment in Polonsky.Xx236 (talk) 12:59, 29 July 2008 (UTC)

The citation is correct... According to WP:SAYWHEREYOUGOTIT, if you took the quote from Polonsky you should say so. I would slightly reword the citation to make it clearer that Polonsky is quoting Ringelblum:

• "Was it inevitable that ... face of the greatest tragedy of all time?" Emanuel Ringelblum, as quoted in Antony Polonsky & Joanna B. Michlic, The Neighbors Respond: The Controversy over the Jedwabne Massacre in Poland. Princeton University Press, 2003

As for whether it is a primary or secondary source... that does not really matter. It is stated as being opinion and not as fact. There is no original research involved. In otherwords, even if the source is primary, it is being used appropriately. Blueboar (talk) 13:26, 29 July 2008 (UTC)