Wikipedia talk:No original research/Primary v. secondary sources discussion/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Rubenstein's Axe

To refresh your memory let me point out these three edits YOU made:

• Edit1

Entirely removed primary sources as a possible source, except in a few narrow cases.

This edit says that ORIGINAl research drawing on primary sources is not allowed - which was, is and remains true. Some pointed out that this sentence was poorly phrased, and I rewrote it. This is because some people here act in good faith by assuming I act in good faith. When someone points out an error in what I wrote, or suggests an improvement, I act on it or support it. I even did this in regard to some your edits.

Wrong. Check your edit history *I* rewrote it. I am the one who removed "Original" in the above phrase. Wjhonson 02:04, 25 August 2006 (UTC)

In any event, you continue to misrepresent what I did. I didn´t "completely" remove primary sources as possible when the two exceptins remains - look of the word "completely." Now, you define the two exceptions as "narrow" but you are wrong. They are not at all narrow and they allow for many uses of primary sources. Slim Virgin asked you to come up with examples that would not be allowed according to the rule. You provided the example of the Anglo Saxon Chronicals and I have explained to you in detail that the so-called "narrow" exceptions you are obsessed with definitely permit the use of that source. So what is the point of this example from the edit history? It does not support your case at all. Slrubenstein | Talk 19:26, 24 August 2006 (UTC)

Irrelevant. The exceptions are to the use of primary sources ONLY. Get it? Not to the use of primary sources. But rather to the use of primary sources ONLY. I've said that now at least four times. And the way the wording was after your edits, my example could NOT be used, as it provides an evaluative statement, which you disallow from a primary source. Evaluative statements from EDITORS are completely disallowed in general so to try to say that is what you meant is disengenous as that area is already covered, and has nothing whatsoever to do with whether a source is primary or secondary or any other ary. Wjhonson 02:04, 25 August 2006 (UTC)

• Edit2

Edit 2 erroneous restatement that "drawing on primary sources is generally discouraged". Reworded following paragraph to completely remove "primary" from "research that consists of collecting and...."

There is no need to include the word "primary" in this sentence when the next sentence (or paragraph) explains very precisely when primary sources are permitted. Slrubenstein | Talk 19:26, 24 August 2006 (UTC)

• Edit3

Edit 3 "Again denigrated primary in favor of secondary with "in favor of" Again re-emphacized that contributors "drawing" on primary sources "must be exceptionally careful". Both erroneous statements seeking to deny the use of primary sources except in a few narrow cases. Again removed "primary" from "collecting and organizing.

Wjhonson 18:12, 24 August 2006 (UTC)

As others have said, it is a fact that Wikipedia encourages the use of secondary sources over primary ones. But anyone - like you - who then complains that this amounts to a prohibition of primary sources is just trolling. I have said it over and over again: primary sources are allowed under certain conditions, and these conditions are essential to maintaining the NOR standard, and these conditions are sufficient to allow or justify loads and loads of uses of primary source. They are not narrow, just precise. Slrubenstein | Talk 19:26, 24 August 2006 (UTC)

It is not a fact, it's your opinion. I have shown in the three edits where you removed the word "primary" even after you complained that you didn't change the wording you just clarified it. Several editors have pointed out to you, that what you did is not clarification at all. You are the one trolling, and you are the one seeking to change long-established policy. Hopefully by now that should be clear. Whether or not you agree, it should be at least clear that your changes are not consensus. Wjhonson 19:30, 24 August 2006 (UTC)

Excuse me, I never said I didn´t change the wording. I said I didn´t change the policy. Of course I changed the wording, that is one of the ways I clarified things. And if you look at the first paragraph of the preceeding section, I stated up-front that I believe that our policy distinguishes between the appropriate use of primary and secondary sources and encourages the latter over the former. I have never hid this. I have been very up-front about this. So the edit changes you provide accomplish nothing but illustrate what I have been saying all along. What is at issue here is not what I did, as I have never contested that. The question is why, and I have provided lengthy explanations. If you still think that the policy as I reworded it prohibits the use of the AngloSaxon Chronicle, as you seem to think, then you need to learn to read more carefully as I have explained repeatedly that this is not the case. Slrubenstein | Talk 20:02, 24 August 2006 (UTC)

And you are obstinate in the face of several editors disagreeing with your very changes, for the very reasons I pointed out. Instead of being obstinate which will get you exactly nowhere, why not try to figure out exactly what our issue is, and respond to *it* instead of some issue that we are not bringing up? The issue being that the wording as it *was* is very clear that Apple Pie is a narrow case where *solely* primary sources may be used. The policy said nothing about favoring primary *over* secondary, nor vice versa, but stated both in clear and simple language side-by-side. The ideal articles use *both* primary and secondary sources, on an equal footing. That has been long-standing policy which you continue to try to change by calling it "clarification", against consensus. Wjhonson 20:10, 24 August 2006 (UTC)

Editing using primary sources is one of the most typical ways people insert original research into articles. It's fine to quote a primary source when one is being completely non-analytical, but as soon as one starts evaluating the source material one must avoid using primary sources to back up that evaluation. Slrubenstein's edit made that clearer. Jayjg ^(talk) 20:46, 24 August 2006 (UTC)

Map Of Misreading

JA: Making a new section for collecting concrete examples of worst-case sinarios later. Remember, You Asked For It … Jon Awbrey 22:48, 23 August 2006 (UTC)

JA: Well thanks, FM, for saving me a trip to the archives.

Example 1. <Title Pending>

SL [SLR?, SV?] is right, Wikipedia has historically not viewed primary and secondary sources equally. Over the last year myself and other have repeatedly told others that secondary sources were preferred to primary sources, many, many times. FeloniousMonk 22:57, 23 August 2006 (UTC)

JA: Will discuss later, time for dinner now. Jon Awbrey 23:23, 23 August 2006 (UTC)

Time to go for me as well - O^O 23:51, 23 August 2006 (UTC)

JA: Having been the victim of some rather vicious Stalker Puppet Attacks in the past — thanks, I'm much better now — I am rather timorous about mentioning any articles that I still care about, but wot the hecque …

JA: Here's a list of articles where it was necessary to cite primary sources, more reliable secondary sources, or multiple sources of both kinds, by way of correcting common hearsay, easily recognizable misrepresentations, misquotations, and popular misconceptions, many of which were not sourced at all and many of which evidently devolved from popular writings and tertiary sources. Now, it could be argued — and I'm sure if it could argued anywhere it will be argued here — that I was seeking to "advance a position", say, the position that accuracy is preferable to baloney, so I think that we will eventually have to sew up that loophole, but like I said, one can'o'worms at a time.

Example 2. Tabula Rasa

In Western philosophy, traces of the idea that came to be called the tabula rasa appear as early as the writings of Aristotle: What the mind thinks must be in it in the same sense as letters are on a tablet (grammateion) which bears no (methen) actual writing (grammenon); this is just what happens in the case of the mind. (Aristotle, On the Soul, 3.4.430a1). Aristotle writes of the unscribed tablet in what is probably the first textbook of psychology in the Western canon, his treatise Περι Ψυχης (De Anima or On the Soul). However, besides some arguments by the Stoics and Peripatetics, the Aristotelian notion of the mind as a blank state went much unnoticed for nearly 1800 years. But the human intellect, which is the lowest in the order of intellects and the most removed from the perfection of the Divine intellect, is in potency with regard to things intelligible, and is at first "like a clean tablet on which nothing is written", as the Philosopher [Aristotle] says. (Aquinas, Summa Theologica 1.79.2).

JA: Let's start with an easy case, one where nothing bad has happened — yet — and yet I still lose sleep over it, on account of the very real and continuing possibility that it might. I don't want to get into diffs and all that unless it can't be helped, so let me just say how I remember things from the events of several months ago.

JA: There had been a running discussion about who invented the tabula rasa idea that took place on one or more of the articles that linked to that concept — maybe it was Scientific Method or Empiricism or both. I had never really thought about it before. Most folks associate the idea with John Locke, but I seemed to recall that Latin was not his main medium, so I suspected that it must have been a legacy from Medieval times, at least. Finally tiring of all the Blue Sky discussions, I get off my duff and rummage through some old books, eventually tracking the idea back through Aquinas to Aristotle, which data I duly type into the article. Is there something sinfully original about what I did or not? Sensible folk would probably say not.

JA: But wait, what if somebody objects that I'm using direct quotations of primary sources to "advance a position", all without having their insidious primacy canonized and sanitized to bits by some duly infallible Western Canonizer? Do I have to dig up some third party source as a notary public? Or wait around with bated breath for some secondary sourcerer to confirm the literary link to which my own sore eyes bare witness? I know, you can't imagine how that would happen. Sadly, I don't have to imagine. Jon Awbrey 17:25, 24 August 2006 (UTC)

If I followed this, the position that you think ought to be in the article is that Aristotle first invented the tabula rasa idea, and you found a quotes from both Aristotle and Aquinas saying just that. You did not synthesize, analyze, or deduce from primary sources that Aristotle invented the tabula rasa idea, the primary sources make that direct statement. The quotes are acceptable. --Gerry Ashton 17:47, 24 August 2006 (UTC)

JA: Thanks, I will slate you as a "sensible person". Still, I have run into cases where people would delete similar sorts of citations on charges that I personally felt were rather trumped up. But I said that it was an easy case, and I am just trying to establish a baseline for future discussion. Jon Awbrey 18:04, 24 August 2006 (UTC)

Example 3. Ockham's Razor

William Ockham (c. 1295–1349) … is remembered as an influential nominalist, but his popular fame as a great logician rests chiefly on the maxim known as Ockham's razor: Entia non sunt multiplicanda praeter necessitatem. No doubt this represents correctly the general tendency of his philosophy, but it has not so far been found in any of his writings. His nearest pronouncement seems to be Numquam ponenda est pluralitas sine necessitate, which occurs in his theological work on the Sentences of Peter Lombard (Quaestiones et decisiones in quattuor libros Sententiarum Petri Lombardi (ed. Lugd., 1495), i, dist. 27, qu. 2, K). In his Summa Totius Logicae, i. 12, Ockham cites the principle of economy, Frustra fit per plura quod potest fieri per pauciora. (Kneale and Kneale, 1962, p. 243).

The provision of the citation of a secondary source makes me wonder how anyone would think this violates NOR. Slrubenstein|Talk 19:31, 24 August 2006 (UTC)

JA: The context is this. Ockham has lately acquired something of a cult following in certain quarters, and folks who have evidently read even less of him than yours truly have taken to attributing magical powers to that humble razor that he drew on a day from a stone and struck on yet another rock to draw forth the mainstreams of modern science in one swell foop. So my yeoman labor of dredging up this more refined secondary source, which so compactly epitomizes the facts behind the epic genesis — well, you'd think it was some kind of iconoclastic blasphemy the way they rent their garments, and shredded not a few of my e-pistles over it. I mean, who are Kneale & Kneale to whittle down the razor whitticisms broadcast every hour by the Ockham Gospel Podcast? Jon Awbrey 04:00, 25 August 2006 (UTC)

Your language is obtuse and I have no idea what you are trying to say. If Kneale and Kneale are appropriate verifiable sources no one can challenge their inclusion. If someone challenges that they are biased, that is an NPOV issue but still would not warrent deleting the referenxce; the solution would be to identify clearly their perspective and add alternative perspectives from other appropriate secondry sources. Again, I see no problem here, certainly no problem relevant to this discussion on primary and secondary sources. Slrubenstein | Talk 15:13, 25 August 2006 (UTC)

JA: Thank you for confirming my judgment. Unfortunately, you were not there to assist when this inclusion was challenged, and you may not be available the next time a similar case arises, so I need to have a clearly written policy which says that it's okay to add credible information to an article, even if it discomfits somebody who cannot produce credible sources for their additions. What you are not appreciating is the fact that people can and did "challenge the inclusion", quite strenuously. I eventually won out, but y'know Nothing Is Binding here, and it was only because I had WP policy statements that were just barely clear and firm enough to back me up at the time. The frozen version of WP:NOR and the parallel changes in the others that were added in recent cycles are just not that clear or firm anymore. Jon Awbrey 06:42, 26 August 2006 (UTC)

It all depends on what the sources cited in this article say. If Kneale and Kneale said in their The Development of Logic that, after a thorough search of Ockham's writings the closest we found was the principle of economy, then the article should say that (which the present article does not do). It can then, as embellishment, cite Ockham's primary texts to illustrate the point. The essence of the argument that Ockham's razor in the form Entia non sunt multiplicanda praeter necessitatem is not found in his writings can only be based on a thorough reading of all of Ockham's published and manuscript writings, which calls on a special brand of expertise. In that regard, I'm uncertain if even a general history of The Development of Logic is an adequate secondary source. I'd like to see a citation of something by an expert on Fourteenth-century thought or preferably on Ockham. --SteveMcCluskey 14:15, 4 September 2006 (UTC)

JA: The purpose of citing these few particular examples is to illuminate a general consideration of some signficance for the overall quality of WP articles. In the case at hand, the aptness of the quotation from Kneale & Kneale was discussed at length by the editors on board, and it was decided to keep it. At least it was still there the last time I looked. It served to increment the quality of the information in the article in some measure and very compactly summed up what was known to the writers at the time of writing. Because I did check the information with several other sources on Ockham, I know that it reflects the general state of what was known to be in and out of his writings at the time. We all understand that some old incunabulum can turn up in an archive, an attic, or a cask of amontillado at any time. But in the meantime the best way of saying what Kneale & Kneale said in 1962 is simply to quote (Kneale & Kneale, 1962). Jon Awbrey 16:00, 4 September 2006 (UTC)

Example 7. Philosophy of Mathematics

JA: Later, Jon Awbrey 02:10, 24 August 2006 (UTC)

Discussion of Cases

"Verifiable" does not mean that someone has verfied the source or provided a citation, only that one could. My sense is that in most cases here there are apparent violations of NOR, but the remedy would not be to delete the primary sources but rather to find the secondary sources that make the interpretive or explanatory or evaluative claim. Doing so would improve the article by verfying the claims and by adding sources that are resources for readers. This is and should be the general response to most claims of violating NOR - to look for the appropriate secondary source. I just do not see this as a problem. I see Wikipedia as a constant work in progress, and I see this as a part of that endless process of improvement. Slrubenstein | Talk 16:43, 24 August 2006 (UTC)

JA: I understand the difference between "verified" and "verifiable", but I can't follow the rest of what you say. I guess my Virgil here — if you catch my drift — is to ask the question: "What would we expect WikiParadiso editors to do?" Wouldn't we expect them to consult primary sources and to collate multiple secondary sources into a critical compendium of the $\sum$ of human knowledge, like encyclopedia writers have proverbially claimed to do for as long anybody can remember? Would you shell out good money, or time, or energy for it if it didn't do that? Maybe, but not for long. Jon Awbrey 19:38, 24 August 2006 (UTC)

I challenge your basic premise; people shell out money for encyclopedias produced by acknowledged experts. Wikipedia is free and anyone can edit. It is for this reason, that we have a very heterogeneous group of editors and no editorial board supervising them, that we have these policies. So: I would start by accepting that there are many different kinds of editors.

1. Some editors come here with a POV agenda they wish to push. One way that they do that is by stringing along quotes or raw statistics and then providing their own analysis of it. This really does happen, and it is wrong. One purpose of this policy is to provide other editors with established principles to help police against such abuses.
2. Then there are other editors who know a lot, from both primary and secondary sources. Another purpose of this policy is to assist such editors in turning their knowledge into a good encyclopedia article. It encourages them to be very careful about primary sources, which often do not speak for themselves and thus invite intepretation. It reminds such editors that the editors themselves shouldn´t provide the interpretation nor present the primary sources in such a way as to privilege the editors´interpretations - which is not at all the same thing from prohibiting them from using primary sources. It also encourages these editors to find secondary sources to represent the interpretations, explanations, and so on that are so important. I think this is a good idea and I would be surprised if you did not think it is a good idea too.
3. Finally (well, maybe someone else can come up with other kinds of editors), there are editors who have heard things but who do not have the sources, but want to add to the article. This leads to an article that might violate NOR or Verifiability. In this case, two things can happen.

1. First, someone who knows our policies really well but who does not know the topic really well will leave the article as is and add tags asking for sources, or will try to edit the article so as not to change the content but make it more NPOV.
2. Second, someone who DOES know the topic really well comes by. When s/he does, one of two things happen.

1. On the one hand, s/he might say:this is BS, I know this stuff and it is just wrong. If there are no citations, s/he has every right to delete what s/he considers false information and if someone comes back and reverts the deletion and restores the possibly false information, the knowledgable editor can say "I think you are violating NOR but can you provide me with sources, if you have appropriate sources I will back down."
2. On the other hand s/he might say "Yes, this is right this is good - but where are the sources?" and then being a knowledgable editor s/he will start putting in the sources.

Well, this is what my Virgil tells me. The result ought to be an encyclopedia filled with articles that draw on primary and secondary sources - even though many of our editors are untrained in research and have widely varying kinds of knowledges. It is our policies, plus a large heterogeneous group of editors, plus time, that will lead to a FREE encyclopedia that will be as good or better than expensive ones people shell out lots of money for (money which pays for, among other things, PhDs and graduate students and an editorial board policing their research) Slrubenstein | Talk 19:51, 24 August 2006 (UTC)

JA: Gosh, it may take me till dawn to read all that, but a qwik-scan makes me think that you might have misread my premiss. I am just saying, wouldn't we all expect a self-respecting article writer to check primary sources? — isn't that just the sort of thing that we'd consider the normal and normative conduct of experts and competent amateurs alike? I just think so. Back later, Jon Awbrey 20:12, 24 August 2006 (UTC)

Jon, I really appreciate your willingness to dialogue on this and start clearing up issues. I think I understood your point - at least trust me, I took it on good faith. Take the time to read over what I wrote carefully. I do not think you and I clash on values. I do think that I am trying very carefully to be realistic about the process of Wikipedia - how things really work here. You have an ideal about what to expect from a self-respecting article writer. My response is that our policies are not written for the ideal article writer. They are written for a community of heterogeneous contributors who are generally unregulated. Many may well be self-respecting, but might have very different ideas of how and why one would have self-respect; many self-respecting authors may find it hard to believe that other editors are self-respecting. Maybe some editors have no self respect. I do not know. I just know that there are lots and lots of wikipedia editors with different views and interests and backgrounds and there is no ideal self-respecting editor who can or should serve as a basis for our policies. I DO NOT ask the question how an ideal editor would act. I ask the question how have REAL editors acted in the past, and the answer is, in many many many different ways over a whole range of good and bad. I also ask, what is the collective process by which articles are edited, because no article is written or edited by one or even a small group of editors at one time. It is a long long process. I think our policies make most sense when viewed in terms of this process.Slrubenstein | Talk 20:19, 24 August 2006 (UTC)

Jon Awbrey wrote "wouldn't we all expect a self-respecting article writer to check primary sources?" My answer: it depends on the field. I can't imagine writing an article about a novel without first reading the novel. On the other hand, I've never met an engineer or physicist who actually read the original Maxwell's equations. I tried once, and it took me about 10 seconds why the original form is never presented in any modern textbooks. --Gerry Ashton 21:28, 24 August 2006 (UTC)

JA: Fashions in education tend to go in cycles like most other fashions. When I was going through the first couple of wringers — speaking of cycles, do people still know what a wringer was? — there was a call to "Read the Masters", and so my sophomore QM class took its main text from Dirac's Principles of Quantum Mechanics (1st ed. 1930, 4th ed. rev. 1967), supplemented by the Feynman excursus and the radically reformed vulgate from Berkeley. We all hang out in different bars, I guess, but I have to report that I have not met many physicists or even EE types who haven't read Maxwell and tinkered a bit with all his silver hammers. Go figure (as Leibniz hath commanded us). Jon Awbrey 16:24, 4 September 2006 (UTC)

Well-studied primary sources

Given that a number of objections to the changes seem to have been prompted by the issue of primary sources in historical contexts, could we perhaps simply get away with noting that primary sources whose accuracy has been commented on by secondary sources can be used in correspondence to that evaluation? In other words, that a primary source can be used directly if it's a known quantity in historiographical terms? This might eliminate the concerns about using sources like Guicciardini and the Anglo-Saxon Chronicle, for example. Kirill Lokshin 23:13, 23 August 2006 (UTC)

It strikes me that perhaps what is bothering you and others is not my edit at all (which I think is less strict than what you propose here) but rather another element of the policy which was NOT an issue in my edit: that all sources must be verifiable. Perhaps you think that this means that the source must be in any public library, therefore excluding all rare manuscripts? This is something that perhaps we should discuss, but for what it is worth my reading of the policy is that it does not exclude such manuscripts. My belief is that while we should strive to use sources that are widely accessible, in some cases this just may not be possible with some primary sources. But I do not see this as meaning the primary source should not be cited. I DO however see it as a reason to encourage secondary sources. In other words, my ideal would be in an article to cite the primary source even if there are only 7 folios in the world, and then cite secondary sources that reproduce or refer to these primary sources but that are indeed available in most libraries. The Yeshu article is a perfect example of this, check it out. Slrubenstein | Talk 23:39, 23 August 2006 (UTC)

If a primary source is reproduced exactly and published, isn't that still a primary source? john k 23:46, 23 August 2006 (UTC)

I meant a secondary source that reproduces in part e.g. through extensive quotations. But John, do you think that this alters the policy, or requires us to change it?Slrubenstein | Talk 23:51, 23 August 2006 (UTC)

I'm not sure. At what point does something become not published? For instance, I am currently doing research at the French Foreign Ministry archives in Paris, looking at microfilms of the political correspondence with Austria (mostly). These documents have been put on microfilm, and are available to anyone who wants to go into the archives and look at them. How different is this from a manuscript that is available in only a couple of libraries in the world? It seems to me that the stuff I look at is definitely not published. I don't feel as though I could include quotations from it in wikipedia (assuming I wanted to, which I don't, particularly), even if they're making wholly uncontroversial, narrative points. Am I being too conservative? Where do we draw the line, exactly? john k 13:57, 24 August 2006 (UTC)

I think you raise an important set of questions that merit thoughtful and deliberate discussion. For the moment, my feeling is this: microfilm that is at only one library or archive should not be used as a primary source here because of its relative inaccessibility. If libraries around the world can easily get copies of the microfilm, then it should be allowed. Also, if someone (like you) has published a book or article making use of the microfilm and re-presenting the information contained in the information (in what is in efect a secondary source) then it can be used. These are my provisional thoughts and I would like to hear what others think. Slrubenstein | Talk 16:26, 24 August 2006 (UTC)

Archive ?

JA: Did someone forget to complete the archive op? Jon Awbrey 03:58, 24 August 2006 (UTC)

What's the archive op? SlimVirgin ^(talk) 05:28, 24 August 2006 (UTC)

Operation. I think JA is referring to someone deleting from this page, but not posting to an archive. Wjhonson 05:43, 24 August 2006 (UTC)

Awbrey´s right, sort of. It is not that I forgot to complete the op, but rathe that I did not double-check. I am working from an internet connection that is excruciatingly slow. I saved the archived material but what happened is that it didn´t save (sometimes, when I click save, I get a preview instead and the material is not saved). Anyway, I think I have now effectively fixed the problem. Thanks for catching it, Jon, Slrubenstein | Talk 16:39, 24 August 2006 (UTC)

Primary sources — my thoughts

(NB:This gets quite long, but for the clarity of future readers, please make any comments at the end)
Ok, I have a habit of finding a heated debate and wading in. After skimming the above disuccusions, I doubt that any consensus has been reached. I appologise if it has, and this ends up stiring up an old dead debate.

Firstly, it is stated that WP:NOR exists to support WP:VERIFY, rather than the other way around ("Another effect of this policy is that as original research will not be supported by reputable sources, it cannot be included." - from WP:VERIFY). In formal writing, it is good practice to cite your sources, because it allows your reasoning to be checked and further information to be found, hence WP:VERIFY. WP:NOT is a logical extension of WP:VERIFY, that if a a fact cannot be referenced, it should not be present. Failure to verify is far more easily established than original research, and indeed a conclusion of original research is an extension of failure to verify. Stating something as being original research implies that not only is not not sourced, but it is a conclusion reached by the editor. It cannot be reached without the verdict of failure to verify (as a properly verified statement is by definition not original research).

Thus it follows on that, in my opinion, in the case of sourced statements, WP:RS is, despite being a guideline, more important than WP:NOT. Without determining is a source is reliable or not, it cannot be shown to original research. WP:RS is a guideline not because it is not important, but because of the inherant difficulties in defining a good, reliable source. Consider the statements within WP:RS about the validity of primary and secondary sources to be the lead, not the follow.

The key facts within WP:NOR in my opinion are (using => as 'implies'):

• You cannot introduce entirely new information within an article that is without any sources. Most people thankfully understand that.
• You cannot do any form of analysis on a reference - however this should be noted as being a more specialised example of the next point.
• That "if A=>B", and "if B=>C" cannot be used to conclude that "A=>C". Whilst this is an obvious and logical step, it is one we must leave to either the references or the reader. This step is by far one of the hardest ones to convince people of, and hence it should be detailed as fully as possible.

There is often some confusion over what the primary source for a subject is. For example, when referencing a song, I have seen people attempting to use YouTube as a reference. However this is not the correct reference - the recording of the song itself that is depicted in the sample is the reference. If a politician makes a speach, it is the speach that is the reference, not the politician (although the speach is a work by the politician). The original painting is the reference for the painting, not a photograph of it. However the photograph of the painting is a reference for the photographic work of the photographer.

This confusion continues into the realm of secondary sources. here, however, it gets more difficult as we have to consider the purpose that a reference is being used for. For example, let us return to that photograph of a painting. it certainly does class as a secondary source for the existance of the painting - for the photographer has documented it. However it may not be used as a secondary source for the evaluation of the painting, as the photograph makes no attempt to do this. It is interesting to note that when an expert evaluates something, they become a secondary source on something, but also a primary source for their evaluation.

Finally, come come onto the crux of the matter. In my opinion, based on WP:VERIFY, WP:RS and WP:NOT (the intent of, rather):

• Primary sources may be used for referencing:
  1. That a statement was made or something does exist within the reference. For example, that...
     • ...a book contains a character
     • ...a recipe contains an ingredient
     • ...within a speech, a politican said the quote "Education, Education, Education"
  2. That something exists within a work that can be determined by observation not analysis and is entirely non-contentious. This must be something that can be determined through casual observation. For example, that...
     • ...A painting contains the colour blue (this is not definate existance, as colour is to an exent subjective)
     • ...A note was hit within a song (this requires identification of the note, but in clear cases there is no debate over this)
     • ...A person within a photograph has red hair
     • ...Within a song, the chorus is sung significantly louder than the verses
     • ...A politican said "Education" many times within a speech
     • ...A politican said "Education" more than "Healthcare", when it is casually observable that only a single refernce to "healthcare" was made in a speach about school reforms.
  3. With respect to statements by people, that they hold an opinion
• Primary sources may not be used for referencing:
  1. That something exists within a work that can be determined by observation not analysis, but there is any reasonable doubt over the observation. For example, that...
     • ...A note was hit within a song, but it is unclear which of two notes it is
     • ...A person in a photograph has black hair, but they could just as easily have dark brown or purple hair
     • ...A shape within an abstract painting represents a house (by definition in this case, the shape is abstract)
     • ...A person within a video is running slowly rather than jogging
  2. That a non-trival count of something exists within a reference. By non-trivial, I refer to both the ease of counting and the number to count. For example...
     • ...A politican said "Education" exactly 12 times within a speech. Given that a speach normally features more than just a single word, it is hard to casually count the use of a specific word (due to the separation between instances), making this non-trivial.
     • ...Noting that a painting has 11 hot-air ballons depicted is reasonable, however if the number was significantly higher it would not be (as counting would then become non-trivial)
  3. That any element within a work is, on a non-trivial level, bigger, quieter, higher, than any other element within the work. Such a non-trvial statement implies a prolonged analysis quite different from allowable casual observation. For example, that...
     • ...The highest note reached within a recording of a song was E6. This implies a comparison between all the notes within a song, which is clearly a major undertaking
     • ...Within a painting, the hot-air ballon 5cm from the left hand edge and 37cm down from the top is the smallest. If a short description cannot be used to locate something, there is a strong possibility that it was a non-trivial venture to find it and it is not casually observable
     • ...A speaker said "Education" more than "Healthcare" witin a speech about education and healthcare reforms (hence featuring both words regularly)
  4. That a narative is predominantly of a certain type. For example, that...
     • ...A character in a book is 'surly' or 'brash', but no third-person description of the character exists describing them as this.

...I could go on, but not right now, I'm out of steam. I'm going to post some of the above over on WP:RS, as it may be of some use to them. The key thing here is trivial uses are allowed. This is important, as it is often the case that commentators behind secondary sources will choose not to mention trivial points, as they often believe them to be obvious. Without allowing for trivial observations, common sense facts become hard to cite at all. LinaMishima 19:07, 24 August 2006 (UTC)

JA: You are misgrokking something very basic here — that all three, er, principles are basically one — they are the Not Making Stuff Up Principle. Jon Awbrey 22:48, 24 August 2006 (UTC)

How, perchance, is the above missgroking? Because I'd certainly agree with that. I know from experience, however, that without a clear definition, people find things both hard to follow, and hard to enforce. What I've written only differs from that statement in terms of progression of principles, and this is more because of the nature of the above discussions and my belief that WP:V is the best starting point (as it is the easiest to achieve and hardest to argue with). LinaMishima 01:42, 25 August 2006 (UTC)

JA: I only had time to give it a quick once over, but there seemed to be several places where you were trying to order the policies by some kind of logical dependency, precedence, or priority, for instance here:

Firstly, it is stated that WP:NOR exists to support WP:VERIFY, rather than the other way around ("Another effect of this policy is that as original research will not be supported by reputable sources, it cannot be included." - from WP:VERIFY).

JA: Given the intricate entanglements of the three main policies I think that any attempt at a hierarchical ordering is ultimately futile. Of course you can chase the tail of mutual recussedness around, but it just keeps going round and round. Jon Awbrey 03:18, 25 August 2006 (UTC)

As I believe I said, much of the ordering was inspired by the arguments by others about the ordering, featured above, and attempts to use their orderings as arguments with respect to content changes. In this respect a definative ordering and opinions thereof are useful. I do get your point, however, and recommend further commenting on the primary sources discussion which follows, which has also been a matter of contenion here - and, if I recall correctly, the cause for the ordering debates. LinaMishima 03:32, 25 August 2006 (UTC)

JA: DGMW, I'm not discouraging analysis of the logical and pragmatic relationship among the Big 3, indeed, I continue to pursue it. It's just that the excursion so far does not lead me to believe that we'll be qwite so qwikili getting to the $\bot$ of the ⊥-less lake.