Wikipedia talk:No original research/Primary v. secondary sources discussion/Archive 4

Toward a consensus

The way the page reads now:

"Original research that creates primary sources is not allowed. Research that draws predominantly on primary sources is generally discouraged, in favor of research based on secondary sources. However, where an article (1) makes only uncontentious descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and (2) makes no analytic, synthetic, interpretive, or evaluative claims, a Wikipedia article may be based entirely or primarily on primary sources (examples would include apple pie or current events). These are relatively rare exceptions and contributors drawing predominately or solely from primary sources should be exceptionally careful to comply with both conditions.
"Research that consists of collecting and organizing information from primary and secondary sources is, of course, strongly encouraged. This is not "original research"; it is "source-based research", and it is fundamental to writing an encyclopedia."

Note the use of the word "predominatly" in the second sentence. The following sentence then addresses when a person can use solely primary sources (no secondary sources needed) in the case of apple pie or current events. And then states those are rare.

Finally it follows up in the next paragraph by reminding over-zealous editors that "source-based research" is not "original research".

That articles should be ideally a mix of both primary and secondary sources, has been the policy for quite a long time. The above wording is satisfactory to me. Wjhonson 02:16, 25 August 2006 (UTC)

SlimVirgin you asked me "where I'm coming from". Read the above wording, and then my comment that "the above wording is satisfactory to me". This is where I'm coming from. This wording has been policy for a rather long time now. I challenge you in fact to find how far back this wording of "primary and secondary" extends. And I have yet to see any proposal from you, only apparent confusion about what the argument is actually about. Wjhonson 06:41, 26 August 2006 (UTC)

This sounds good to me, with the possible exception of highly technical topics that may require specialist knowledge at the same level expected of the typical reader. Deco 03:42, 25 August 2006 (UTC)

I was thinking about how much I agreed with Deco, and an unintended consequence occured to me. If I were to write "Eric Barbour wrote in the August 1998 issue of Spectrum that in a test of four solid state amplifiers and two vacuum tube amplifiers, a triode amplifier had the best second-harmonic distortion" that would probably be acceptable, because Spectrum in general and this article in particular would probably qualify as a secondary source. But if I'm wrong, and the article is actually a primary source, then I can't include the information because Barbour didn't actually write that the triode was best, he wrote that the triode had -52 dB of distortion, and the next lowest number in his table was -48 dB. I used my specialist knowledge, that smaller numbers are better when measuring distortion, to interpret his table.

As I understand the current policy, any reasonable adult must be able to compare a statement in WP to the primary source from which it came, and see that they are equivalent, without specialist knowledge. For secondary sources, the general spirit of the WP:NOR and WP:V policy also require that statements be verifiable, but the exact knowledge requirement to do the verification is not specified. One could argue that the standard should be that anyone with enough knowledge to understand most of the article should be able to verify the statement. --Gerry Ashton 04:25, 25 August 2006 (UTC)

JA: I'm having trouble distinguishing what's being asserted as proto-quasi-consensus from what is merely its embedding excess of excipient explanatory expletives in this section. Could folks please consider blockquoting their proposed proposals? Gratia in futuro, Jon Awbrey 15:01, 25 August 2006 (UTC)

I am not sure I agree with Gerry. For one thing, secondary sources are usually more freely available than primary sources. As far as this article is concerned, the real issue in NOR is "in what way are sources verifiable" but rather how to use any source properly so that one is doing appropriate source-based research rather than original research. "In what way are sources verifiable" which seems to be Gerry´s main concern is a reasonable question - but it should be asked and discussed not here but on Wikipedia:Verifiability talk page. Slrubenstein | Talk 17:21, 25 August 2006 (UTC)

You seem to be confused about what OR is. Quoting a primary source is never OR. Not... ever. So how are these two issues related? OR is creating your *own* new facts *yourself*. Such as your *own* laboratory experiment. Quoting the results of someone else's experiment, previously published, is not OR. So can you explain this more clearly? Wjhonson 17:27, 25 August 2006 (UTC)

It's perfectly possible to be doing OR and to have everything sourced. Do you understand that? SlimVirgin ^(talk) 02:28, 26 August 2006 (UTC)

I can't speak for Wjhonson, but I certainly understand that. - O^O 04:12, 26 August 2006 (UTC)

Suppose, for a moment, we define OR as anything prohibited by the NOR policy. Under that policy, in an article entirely based on primary sources, a paraphrase that is not easily verifiable by any reasonable adult without specialist knowledge is OR. So quoting a primary or secondary source is never OR, but paraphrasing it is unless the paraphrase is very easy to understand. If instead the article is predominantly based on secondary sources, then the "no specialist knowledge" clause does not apply. Somewhat more sophisticated paraphrases might be allowed in the article based predominantly on secondary sources. As for Slrubenstain's contention that this should be discussed in Verifiability, the words that create the issue, "no specialist knowledge" are in this policy so this is the right page to discuss it. --Gerry Ashton 18:42, 25 August 2006 (UTC)

Gerry, I implore you to stop mixing up NOR with Verifiability. They are two separate policies. You keep bringing up verifiability. If an edit is unverifiable, it is bad because it violates the verifiability policy, not because it violates the NOR policy. If you want to debate the verifiability policy, do so on its own page. As to NOR, which is distinct from verifiability, Wjhonson is confused and simply does not understand out policy. I implore him to actually read it and learn it and comply with it. Wjhonson writes that "Quoting a primary source is never OR." This is flat out false. Sometimes it is OR, and some times it is not OR. He writes, "OR is creating your *own* new facts *yourself*." This is only one example of OR. OR also applies to using facts including those taken from primary sources to express a new (i.e. the editor´s own) interpretation, explanation, anlysis, synthesis, or argument. The issue is HOW one uses primary sources, and if one uses primary sources in this way, one is violating the policy. Slrubenstein | Talk 18:49, 25 August 2006 (UTC)

Slrubenstein wrote Sometimes it (quoting a primary source) is OR, and some times it is not OR. - I'm having trouble thinking of an example where quoting a primary source could be OR. I am clear, of course, on how sumarizing a primary source, or using a primary source to further some analysis could be OR, but not making the quote itself. Slrubenstein, could you provide an example of how quoting a primary source could be OR? - O^O 18:56, 25 August 2006 (UTC)

The meaning of a text depends on its context. The question "do you know what time it is?" may seem very straightforward, but in fact it means one thing if a stranger approaches you in the street and asks you, and another thing if your mom asks you if you are a teen-ager coming home at 2:00 am. My point is that the context in which primary sources are being used is important to assess whether the use violates NOR or not. Slrubenstein | Talk 19:27, 25 August 2006 (UTC)

I understand that, but I would point out that the OR doesn't come from making the citation (Jack said "do you know what time it is"), but in doing further analysis from there (Jack said "do you know what time it is" so he must have been mad at his teenage son. In other words, it is what is DONE with the citation that could lead to OR, not the citation itself.- O^O 04:12, 26 August 2006 (UTC)

Suppose I had put a paraphrase into an article that Barbour found that of the amplifiers he tested, a triode had the lowest 2nd order distortion. Along comes editor Q and removes it because the article is mostly based on primary sources, and Q thought my paraphrase of Barbour involved so much analysis that it was original research. Q is a reasonable adult, so, according to NOR he's right, and the paraphrase is out. Later, the article has been rewritten to rely mostly on secondary sources. I put the Barbour paraphrase back in, and Q objects on the talk page. I reply the paraphrase stays, because the "reasonable adult" clause does not apply, so only people who can understand Barbour's article are qualified to judge whether my version is a paraphrase or OR. Ten editors with specialist knowledge come along and all agree it's just a paraphrase, so it stays in the article. --Gerry Ashton 19:23, 25 August 2006 (UTC)

What Slrubenstain wants is a policy that allows him to remove certain facts he does not like, like this chart that he currently opposes being added to the capitalism article. Ultramarine 17:49, 25 August 2006 (UTC)

Both Smith and Marx observed and discussed a revolution in the economic system and emphazied the importance of studying economic development. This graph illustrated this revolution. World GDP/capita changed very little for most of human history before the industrial revolution and the introudction of capitalism. (Note the empty areas mean no data, not very low levels. There are data for the years 1, 1000, 1500, 1600, 1700, 1820, 1900, and 2003.)

That is BS and a you are a liar. I explicitly stated that this chart must be in Wikipedia. Slrubenstein | Talk 18:49, 25 August 2006 (UTC)

You have opposed adding the chart to the Capitalism article. Ultramarine 02:46, 26 August 2006 (UTC)

Slimvirgin, do not edit my talk page comments.Ultramarine 02:46, 26 August 2006 (UTC)

Thank you Ultramarine for bringing to our attention the specific issue that brought Slrubenstein here. I believe we can at least all (except rubenstein) agree that grinding axe's is not what this page is for. The policy as I quoted it, is long-standing, and should not be changed simply because one editor wants to then use it to beat another editor. Now that we understand that, to rubenstein, this is about one specific issue, I think we can all approach the topic with that perspective. And I think rubenstein's extreme reaction to the charge is enough evidence.Wjhonson 02:08, 26 August 2006 (UTC)

SLRubenstein helped to write this policy and has been involved in editing it since 2004, whereas your first edit to this talk page was just a month ago. It's perfectly standard for an editor who is very familiar with a policy to want to clarify it if he sees that editors are misunderstanding a particular point. That isn't "grinding axes": it's the way the language on policy pages is improved. It seems to me that this situation has been caused, not by SLR's edit, but by your failure to understand the policy. I'm sorry to speak like that, but you seem to be writing long screeds of posts to cover one very, very simple point. SlimVirgin ^(talk) 02:40, 26 August 2006 (UTC)

The point may appear simple, but we have failed to reach consensus. - O^O 04:12, 26 August 2006 (UTC)

I would like to make clear that SlimVirgin removed comments from this page that another user had posted. That may not be apparent to those who do not view the history. Wjhonson 05:23, 26 August 2006 (UTC)

Whether Rubenstein "helped write this policy" as you assert with no proof, is irrelevant. The policy has certainly evolved since 2004 and the fact that Rubenstein did not show up until now to contest what has been there for quite some time, is what is relevant. The language change was not an "improvement" and the more you two or three harp on that, the further you get from any consensus. Is this the way to create consensus? To repeat yourself ad infinitum? I don't understand that logic. Wjhonson 05:25, 26 August 2006 (UTC)

I don't need to prove it to you. I'm informing you that he helped to write the policy and has helped it to evolve too. He didn't "show up to contest" anything, but simply clarified the writing.

I asked you a question above but you appear not to have answered, so I'll repeat it. Material can be OR even if it's entirely sourced. Do you understand that about the policy? SlimVirgin ^(talk) 05:44, 26 August 2006 (UTC)

What I understand is that quoting a primary source is never OR. Until that point is clear its not going to be productive to go any further. Wjhonson 06:11, 26 August 2006 (UTC)

It's a simple matter to answer that one question, and it would help me to see where you're coming from. SlimVirgin ^(talk) 06:15, 26 August 2006 (UTC)

SV, from what I've seen here, no reasonable argument is going to convince those who object, they're simply immune to reason that supports any view but their own. The appeals to "finding consensus" fall flat... there's simply no consensus that the original policy's formulation is even flawed, much less that any one particular change is required or right. Nor is there likely to be as long as the basic, long-accepted fundamentals of what constitutes original research are wilfully ignored. All the yammering here is not going to change such a fundamental policy so significantly; many long-term contributors would not stand for it. Any consensus arrived at through a faulty understanding of basic foundation issues will be null. FeloniousMonk 06:32, 26 August 2006 (UTC)

After a day

I had no time to join the discussion yesterday, but I've tried to read and understand everything that was posted since my last edit. Doing my best to interpret all that was written, here is my summary of what I think the disputes at hand are. I post this in hope that I'm correct in identifying where we disagree, and that by clarifying this, we are closer to resolution.

First obsevation: There are two very distinct sets of edits that got caught up in the series of reverts and article protection. The first disagreement is over "expert editors", when they can cite themselves, and whether they cab make "vanity" postings. The second disagreement is over the use in general of primary and secondary sources. If there is another disagreement caught up in the protection that I didn't note, please describe it immediately below this paragraph.

I am not a party to the first disagreement over "expert editors". I hope that disagreement can be reconciled so the article can be unprotected swiftly, but I don't have any key insights there.

With regards to the second disagreement over primary and secondary sources, I believe that there is quite a bit of people misinterpreting what each other has to say. I see three separate issues being discussed.

1: Prohibition of primary sources: There is a lot of argument above about whether primary sources should be (or would be) prohibited under any of the proposed edits. Oddly though, I don't see anybody claiming that they want primary sources prohibited. Can all parties simply agree that prohibition of primary sources is not on the table? Once we agree to that, I think we can save a lot of typing. Of course, if anybody is championing the idea of prohibiting primary sources, they should clearly speak up.

2: Secondary sources are preferred to primary sources: I think on this issue there is genuine disagreement. Several editors have said straight-out that they prefer secondary sources over primary sources, and that the NOR policy has historically backed this preference. Several other editors (including myself) have said that primary and secondary sources are equally welcome in articles, and they NOR policy has never given preference to secondary sources over primary sources. This is, I think, the stickiest issue here. Until we agree on this issue, I fear there is little hope of agreeing on language.

3: Restriction on exclusive use of primary sources: There is obviously debate around the wordsmithing of the "makes only uncontentious" and "make no analytic" paragraphs. This is the language that I think Slrubenstein set out to clarify, but without agreement on issue #2 above, I think we are all talking past each other in our attempt to find consensus.

So, without me actually making any proposal here as to how the policy should be worded, could the parties to this debate point out if they think I have identified the issues at hand? - O^O 18:08, 25 August 2006 (UTC)

This is constructive, thank you. I agree completely with your first observation, and believe we should treat the two issues seperately. I would add that some editors here are also raising issues about verifiability. I believe those questions belong on the verifiability talk page. As to your three questions, I think we can all dismiss (1) as I do not believe anyone here has ever called for a flat out prohibition of primary sources. You are right about (2) and I have stated my own position and my reason for it very clearly. As to (3) I tried to clarify my position in my response to Wjhonson above (18:49, 25 August 2006), and Jon Awbery below (19:02, 25 August 2006). Slrubenstein | Talk 19:06, 25 August 2006 (UTC)

I would most respectfully ask some of the editors on this page to bear in mind that you may simply not have understood the policy, and yet rather than read it carefully and gain editing experience, you're asking volunteers to spend a lot of time trying to address your individual queries. There comes a point where that is no longer fair or reasonable.

The policy did not and does not prohibit the use of primary sources. It discourages their use in favor of reliable secondary sources. The reason it discourages them is that the use of primary sources often involves original research. There is nothing inherent about (published) primary sources that should lead to OR. But in practise it does, for the simple reason that many, if not most, of our editors are not good editors. They have no or little academic training. They do not know how to use primary sources properly; or perhaps they know but don't care. They therefore use primary sources to make analytic or synthetic claims that violate the OR policy.

The straightforward use of primary-source material is not prohibited by this policy. If a journalist witnesses a car accident, and writes a story giving his eyewitness account, the story is a primary source. That does not mean that, if he writes: "I saw five people die," we are not allowed to write: "Journalist X saw five people die." Taking up large amounts of time arguing that the policy prohibits this is frustrating and completely wrong-headed.

Secondary sources are preferred because most people don't know how to use primary sources, as WP:RS makes clear. That's it. SlimVirgin ^(talk) 19:10, 25 August 2006 (UTC)

JA: With all due respect, SV, I understood the policy quite well, as it was written at the time when I first read it. My understanding of that policy is precisely what causes me so strenuously to object to the most recent attempts to subvert it. Jon Awbrey 19:20, 25 August 2006 (UTC)

Then you have not understood it. SlimVirgin ^(talk) 19:30, 25 August 2006 (UTC)

SlimVirgin, respectfully, I feel this is inappropriate. You cannot definitively say what the policy was, you can only say what you interpreted the policy to be. Your understanding is not automatically superior to everyone elses. - O^O 20:09, 25 August 2006 (UTC)

Why do suppose that experienced editors who helped to write this policy might not understand it better than editors who just flit in from time to time to cause a problem on the talk page? SlimVirgin ^(talk) 02:43, 26 August 2006 (UTC)

If you are referring to me in your "flit in from time to time to cause a problem" comment then I am insulted by your comment and request you apologize. I think it would be best if you took a few days away from editing this page, what people have been writing here may be easier to understand with some time for you to cool off. - O^O

Home On The Range

JA: Here are what I think are some generally held common sense notions:

1. There must be no discouragement of research that draws on primary bibliographic sources.
2. There is nothing inherently "neutral", "objective", or "unbiased" about being a secondary source, in other words, that accrues to a source simply by dint of its being secondary. Neutrality is a distinct matter that has to be decided on its own merits.
3. There is nothing inherently "reliable" or "reputable" about being a secondary source. Reliability and reputation are separate matters that have to be decided on their own merits.
4. If you think that editors of anything anywhere can avoid the sweat of making "evaluations" of neutrality, reliability, and reputation, then you have omitted to read the fine print in the job description carefully enough.

JA: Jon Awbrey 18:45, 25 August 2006 (UTC)

JA: My own last, best hope for WP-kind got lost in an edit conflict, so I'll have to go dig it up, but here is the penultimate version:

Original research that creates primary sources is not allowed. Original research that draws predominantly on primary sources or single secondary sources is generally discouraged, in favor of research based on independent and multiple sources.

JA: I think that this states what is just plain common sense in the Real World of journalism and scholarship. Jon Awbrey 18:50, 25 August 2006 (UTC)

JA: Here is the more complete and correct version, which was discused briefly above — but I'm hopeful that all our heads are much clearer today.

Primary sources present information or data, such as archeological artifacts; film, video or photographs (but see below); historical documents such as a diary, census, transcript of a public hearing, trial, or interview; tabulated results of surveys or questionnaires, records of laboratory assays or observations; records of field observations. Secondary sources present a generalization, analysis, synthesis, interpretation, or evaluation of information or data from other sources. In discussing the works of an individual writer, for instance, in compiling a bibliography of writings by and about an author, primary sources are works actually written by that author, while secondary sources are works written about that author's life and work by others. The relative value of primary and secondary sources with respect to verfiability depends on the primary topic of the article. For example, when discussing an author's work, what the author actually wrote in that work is a primary source that is definitive as to its content, and it is treated as the most reliable source, over and above what any secondary source may say about it. In contrast, when discussing primary source reports of real events, common sense tells us that a comparatively disinterested secondary source, especially one that collates and verifies many different primary accounts, may be capable of more neutrality in its reporting.

JA: Jon Awbrey 19:02, 25 August 2006 (UTC)

Let us bracket common sense (or rather, Jon´s beliefs about common sense) and stick to the issue at hand. Here is what I think is Wikipedia policy.

1. Wikipedia must discourage, indeed prohibit, research that draws on primary bibliographic sources when a contributor does so in order to express or forward the contributor´s own interpretation, explanation, analysis, or synthesis of a topic.
2. Secondary sources by their very nature represent a particular point of view, and we should encourage research based on secondary sources because it is easy to be very clear about the point of view (e.g. bias, subjectivity) of the source and thus comply with our NPOV policy.
3. Reliability is not an issue in this particular policy (NOR), indeed, it is not an issue at Wikipedia. What is an issue is Verifiability, and we have a whole separate policy concerning verifiability - discussion of that policy belongs on that policy´s talk page.

Slrubenstein | Talk 19:02, 25 August 2006 (UTC)

JA: With regard to Point 1, to wit:

Wikipedia must discourage, indeed prohibit, research that draws on primary bibliographic sources when a contributor does so in order to express or forward the contributor´s own interpretation, explanation, analysis, or synthesis of a topic.

JA: Here you are confounding two separate matters. Wikipedia must discourage, indeed prohibit, all research that a contributor uses in order to express or forward the contributor´s own interpretation, explanation, analysis, or synthesis of a topic. Whether a contributor is doing that or not is a question that is wholly independent in principle from what order of research they are using. Jon Awbrey 19:12, 25 August 2006 (UTC)

Jon, you wrote again: " ... when discussing an author's work, what the author actually wrote in that work is a primary source that is definitive as to its content, and it is treated as the most reliable source ..."

This is sometimes false, as a number of people have pointed out. We can't keep on pointing it out. We are volunteers, and therefore the time we can spend on any one person's concerns is limited. Suffice to say, that cannot be added to the policy page. If I don't comment on it again, it's not because I agree with you. Please assume that I disagree and will revert if you try to add it to the policy. SlimVirgin ^(talk) 19:16, 25 August 2006 (UTC)

JA: It is not false in the ordinary sense of the words that I actually used. The work as published is definitive of its contents. That is why we cite sources in the first place. There were a number of misquotations and bizarre interpretations placed on what I wrote, but I am simply not responsible for those. Jon Awbrey 19:30, 25 August 2006 (UTC)

Jon, I suggest you take a break from this page for a couple of days. You're writing more than anyone will read, and you haven't fully understood the policy. When the page is unprotected, SLR can tweak what he wrote to clarify things if necessary. More than that is not required, and we shouldn't be wasting time with this kind of wikilawyering. SlimVirgin ^(talk) 19:32, 25 August 2006 (UTC)

Judging from the way people are responding to Jon's statement that "... when discussing an author's work, what the author actually wrote in that work is a primary source that is definitive as to its content, and it is treated as the most reliable source ..." I think others must be interpreting it more broadly than I do. Looking in John Toland's book Infamy (page 3) I read "On Saturday morning, December 6, 1941, one of the translators...." If I came across a secondary source that quoted Toland, but the date were December 4 instead of 6, by default I would believe the primary source over the secondary source. I think that is all Jon is saying. -Gerry Ashton 19:53, 25 August 2006 (UTC)

JA: Exactamundo. There are a few consequences of that. It's why we have things called "authorized editions" and "critical editions", and "variora", but I would not dream of suggesting that we should get the first pass reader of WikiPolicy O Lost in a' that. Jon Awbrey 20:14, 25 August 2006 (UTC)

Why would we need a policy that stated such a basic, obvious point? How many thousands upon thousands of other basic, obvious points does Jon want to add to this policy, so that at every stage in the editing process editors are told do this, no, do that, according to Jon's understanding of what an algorithm for common sense and editorial judgment might look like? Jon, please stop posting to this talk page for a few days. You're trying to wikilawyer us to death. SlimVirgin ^(talk) 02:49, 26 August 2006 (UTC)

JA: The Precipitating Parents of the US almost made the mistake of thinking that the principles in the Bill of Rights were too basic and obvious to bother writing down.

JA: I used to think that the things I wrote were basic and obvious. But when I came to WP I found out that they were neither obvious nor taken for granted, especially out there in Trenchtown. When I first read the Big 3 WP policies, that was not news to me. Except for 1 or 2 peculiar things that are causing most of the current fiasco, WP did not invent the lion's share of those norms. They simply acronymized what some of us got grooved into our brains from about the 6th grade on.

JA: The NOR policy as I knew it was working just fine. It constituted one of the last bulwarks against people who are too lazy to read that 2nd book, and who use every excuse in that 1st book not to do the work of writing quality articles. I am not the one who starting messing with it. Jon Awbrey 06:06, 26 August 2006 (UTC)