Wikipedia talk:Identifying reliable sources (history)/Archive 2

Archive 1

Archive 2

Define historian?

I think it will help if we add a quite rigorous definition of "historian". First draft: "A historian usually has a higher degree in history (occasionally a higher degree in another arts or social science subject). She or he publishes regularly in academic journals and/or academic presses." Thoughts? Itsmejudith (talk) 20:09, 11 November 2011 (UTC)

I think this is the kernel of the definition. My natural response was to gloss it, rather than to emend it. To gloss the obvious and the non obvious (Unsigned so anyone can edit) Fifelfoo (talk) 21:12, 11 November 2011 (UTC) :

• Higher degrees are usually Doctorates, Habilitations or Masters by research. They occasionally comprise other degrees, such as Masters by coursework degrees, but, this is usually associated with a vigorous or significant publishing practice by the historian concerned.
• Due to the concerted attack on academic disciplinarily in the last 30 years, Historians may have degrees in many cogently related fields. If in doubt, check the forums where their work is published and reviewed.
• Academic standards of regularity vary, and in some systems have changed considerably in the past ten years due to changes in academic publishing and the academic workplace. Similarly, professional historians are driven by similar work structures into a very similar publishing cycle. The time between books may be as little as one year, or as many as seven. Journal articles, chapters and conference papers appear more regularly, but delays in the publication cycle may mean that even a prolific historian has a number of years between a glut of papers becoming available. In general, history publications are published very slowly compared to science publications.
• The academic standard of a press, imprint or book series can be checked against the press's other publications, the libraries they are held by, and the journals they are reviewed in—most University Presses are safe. The academic standard of journals can be checked through their peer review standards, their calls for papers, and whether they are indexed in periodicals directories as peer reviewed.

Thanks. I like all this, but perhaps the wording can be reduced a bit by recourse to "usually", "we expect that". Also, I would like to propose the names of some writers as cases that have given rise to discussion in the past or that might prove limiting cases for us:

• Bat Ye’or. No academic formation in history. Published by Fairleigh Dickinson University Press which is not a member of the Association of American Academic Presses. I argued strongly against the "historian" label in her biography.
• Karen Armstrong. Publishes widely on religion. Writes on Islam but is not an Arabic speaker. Often on the BBC, has had an exhibition at the British Library.
• Andrew Wheatcroft. MA Cantab in history, which is a first degree. Published by Penguin. Secondary source work. Typical of a strain of UK writing, in which history is close to literary studies and cultural commentary.
• Frances Stonor Saunders. Journalist writing for a popular audience. All from secondary sources.
• Alison Weir. Popular populariser. Same "how different from the home life of our own dear Queen" territory as David Starkey.
• Marina Warner. Literary background. Work highly regarded by many academics.
• Martin Bernal. Original formation in Chinese studies. Long career teaching political science in a research uni. Black Athena series on Egypt-Greece cultural transmission exceedingly controversial, retains some academic support.
• Cheikh Anta Diop. Pioneering post-colonial writer on history and also researcher on scientific archaeological dating. Views have been widely dismissed but a re-appraisal might well show he was misrepresented. Has a university named after him.

To which of these should the label "historian" be applied? Itsmejudith (talk) 09:59, 12 November 2011 (UTC)

Putting me on the spot! These are rapid responses:

Not a historian, not reliable: Bat Ye'or: non-historian, I get a very strong FRINGE feeling almost immediately, I wouldn't rely on her works;

Not a historian, professionally reliable elsewhere: Karen Armstrong. I'd accept her as a Theologian, and her views when cited by historians, but I feel really hesitant about scoping her as a historian. In articles about the Axial age, if her theories are approved by other relevant disciplines (including history) I'd include them

Not a historian, popular work reliability (replace with historians asap, kind of unscholarly TERTIARIES actually): Andrew Wheatcroft; Frances Stonor Saunders; Alison Weir (when published as history, not historical novel)

Historian w/ use qualifications: Marina Warner, I'd double check reviews before using her as the core of a theory informed history article. Cheikh Anta Diop, aware that his views are considered not current (but needing to check this in depth via review articles), add him to "historiography" sections as appropriate even where he's out-dated if they were influential; where considered correct, write up. Martin Bernal, follow the majority of recent review articles, write up the historiographical controversy _where_ influential and relevant as observed by other scholars; where considered correct, write up. Fifelfoo (talk) 00:28, 14 November 2011 (UTC)

I am not sure I can accept limiting the definition of "historian" to those who have a degree. There are amateur historians who have very good reputations, and are considered highly reliable in their field. For example, Stan Fischler is best known for being a sports reporter for teams in the New York area, but he is also a well respected amateur historian when it comes to the history of the New York City subway (he has written several books on the topic). In fact, I would say he is probably more reliable on the narrow topic of the subway system than most academic historians. Ultimately, reputation is what makes a source reliable or not... and while an academic degree comes with an assumption of reputation, it is hardly the only way to earn a good reputation. And... some academic degree holders have very poor reputations. Blueboar (talk) 00:48, 14 November 2011 (UTC)

My question would be how we sound a reputation. For the academically supported areas of history it is straight forward: we have intros, review articles, and book reviews. For areas of history not academically supported like transport and logistic history, how do we determine the community of "High quality" experts, and their relationships of reliability? Is rail history supplied with amateur journals that set a consistent standard of historical quality? (I recently came across this on RS/N where a self-published militaria/military-history work is well regarded in militaria presses, but nothing has come forward from military history presses yet). Fifelfoo (talk) 00:52, 14 November 2011 (UTC)

• I cannot accept limiting "historian" to scholars with degrees in history, history is an integral part of many fields and one can be a historian of psychology for example and have a degree in psychology not history, or have a degree in philosophy and work profesionally with history (of ideas for example).·ʍaunus·snunɐw· 04:26, 14 November 2011 (UTC)
  • My impression (from working with them) is that history of philosophy academics, and history of medicine academics, regularly publish in modes that ensure peer review by other "Historians of particular fields of study," as the great and wise Australian Bureau of Statistics / Australian Research Council deems them. Correspondingly, practicing philosophers and medical practicioners who present histories of particular fields of study usually publish in similar locations, and thus get the shine of disciplinary oversight over their work. Just searching on a journal title (without knowing it) JHMAS shows all of the signs of a historical journal. I am assuming that a similar phenomena occurs with economic historians (I've only really read books, and a few articles in Past & Present, etc). And this same definition would allow the "flow through" of expertise for publishing but non-degreed historians in history of specific fields. Fifelfoo (talk) 04:38, 14 November 2011 (UTC)

My point goes beyond the example of historians of particular fields like "history of philosophy" and onto the fact that scholars of the humanities and social sciences often work as much in history as historians do, and publish about history in their disciplinary journals. History is always history of something - and people who are experts in that something are generally also reliable experts in that something's history whether or not they have a history degree or publish in history journals. (e.g. History of non-western peoples have generally been written by anthropologists - only recently has Ethnohistory come into the picture)·ʍaunus·snunɐw· 03:44, 21 November 2011 (UTC)

Many thanks, both, for thoughtful contributions. Do we have consensus to include a definition of "historian" and then also lay down guidelines for cases where reliable history is written by people who are at the edges of the definition or well outside it? Usually, there will be no problem at all with social scientists. I'll also add that some people are qualified in "Area studies" (European Studies, African Studies...). And Immanuel Wallerstein defines himself as a sociologist, although he also explains that that was one reason he found it hard to get his economic history published initially.

We are also going to have to warn about "quick historical overviews" in sources that are otherwise highly reliable. Some particularly bad examples have come up on WP:FTN recently. An article on a plant will cite a pharmacological research paper that starts with uncited claims about how and when the plant has been used medicinally. The pharmacologists are just claiming space for the relevance of their research, Errors in the introduction don't affect the validity of their research findings, but they are not qualified to comment on social use of plant extracts across time and space, and the paper introduction is not RS for such claims. Another quick overview I found to be in serious error was in relation to the Jesus myth hypothesis. A respected source on early Christianity claimed that Marx had adopted the mythical Jesus notion from Bruno Bauer, which if you know a bit about Marx is unlikely. An expert in the 1st century CE was repeating information gained from some unreliable source about 19th century thought. I didn't think it reflected particularly badly on the writer, but just shows that we should always exercise caution about statements that aren't part of the core of a text but only set the scene. Itsmejudith (talk) 19:39, 21 November 2011 (UTC)

Good idea but impractical?

While I appreciate the effort made here I have issue with the focus on historians as most histories are written by non historians (ie people who do NOT have a history or related degree--like historical anthology). For example, Pallasch, Thomas J. DDS; MS, and Michael J. Wahl, DDS (2000) "The Focal Infection Theory: Appraisal and Reappraisal", Journal of the California Dental Association gives a thumbnail sketch of Focal Infection Theory but Pallasch doesn't have a degree in history and Journal of the California Dental Association is not a history journal so does this make him unreliable regarding this piece of medical history? That is the impression this article gives and it is a very wrong one.

Furthermore this article ignores the complex relationship between history and anthropology. A historian is a good reference source for what is thought to have happened but an anthropologist would be the better source regarding why it is thought to have happened especially if there are few to no surviving written records.--BruceGrubb (talk) 03:34, 21 November 2011 (UTC)

• I agree very much. The problem here is that people who are experts in something are generally also experts in the history of that something. And as you say histories of non-western peoples are generally written by anthropologists who specialize in those peoples or areas, not by historians. I think the guideline would have to acknowledge that history as a disciplinary field does not have a scholarly monopoly on history writing as a practice.·ʍaunus·snunɐw· 03:53, 21 November 2011 (UTC)

• The article you cited isn't a history article, it is a kind of medical article whose genre, known to practitioners of medicine of various kinds, may be history. Consider this quote, "Modern restorative dentistry and endodontic therapy were essentially a development of American ingenuity," for example, that shows signs of idealistic nationalism (a method of conducting history that died out between the 1850s and 1930s), and which appears to be cited to that eminent paper on national mentalities, "17. Dussault G, Sheiham A, The theory of focal sepsis and dentistry in early twentieth century Britain. Soc Sci Med 16(15):1405-12, 1982." I'd suggest using MEDRS for the article you cite. JCDA is peer reviewed, and this appears to be a credible tertiary field summary in a peer reviewed medical journal. Regarding anthropologists and "what" versus "why," I'd suggest you read some historiography because your conceit is grounded in obvious ignorance. Fifelfoo (talk) 03:49, 21 November 2011 (UTC)

I think that accusation bounces right back at you. (Ok I agree that the difference between history written by historians and anthropoloogists isn't about description vs. explanation - but about areas of interest and approaches - but I do think your accusation of ignorance betrays your own ignorance of history written by e.g. anthropologists, political scientists, sociologists etc. - they generally do know how to use historical sources and methods)·ʍaunus·snunɐw· 03:52, 21 November 2011 (UTC)

I've read a number of very good histories written by anthropologists, political scientists and sociologists. Most of them clearly and instantly meet the definition of historian suggested here through the venue of publication, or the review of their works in scholarly journals. There is also a market for poorly written end of career pieces by academics of all disciplines in providing folksy quality material far below the standards of local history—and I've noted that these works rarely secure publication in the higher impact forums of journals devoted to research programmes in history, or anthropology. (I certainly wouldn't slang off anthropologists, political scientists, or sociologists as "what"ters.) Fifelfoo (talk) 04:04, 21 November 2011 (UTC)

Ok, I agree that the draft (which doesn't directly answer the question "Who's a historian") is sufficient for including also practicioners of history who have other academic backgrounds. I guess I was more reacting to the above discussion trying to delimit the question of being a historian to a question of identity rather than of practice. I think the right question to ask is not "who is a historian" but to "how do we know if this source is reliable as history" - i.e. focus on publication venues rather than on scholarly backgrounds.·ʍaunus·snunɐw· 04:13, 21 November 2011 (UTC)

The key reason why we have a difficulty and so have a reserve need to go back to the Author in reliability, is that historians commonly publish monographs or chapters in edited books. These are problematic for determining methodological competence to publish history because just because something is OUP doesn't mean that it is History. Non-historical sociologists publish with OUP. Non-historical political scientists publish with Cambridge. Non-historical anthropologists publish with Routledge. So this makes identifying books difficult. Showing that Jane, who has published in four history journals, and then brings out a book is competent for a historical book is easy. Showing that John's book chapter is safe to use for history, when his PhD was a historical anthropology is easy. Showing that Sue's book, which was reviewed in seven top history journals, is okay to use for history is easy. But there's no way to separate Yale from Harvard because they don't publish history books under an imprint. Fifelfoo (talk) 04:19, 21 November 2011 (UTC)

Here's a case we certainly have to identify as RS. Jeremy Sarkin, Germany's Genocide of the Herero: Kaiser Wilhelm II, His General, His Settlers, His Soldiers. "Jeremy Sarkin is Chairperson-Rapporteur of the United Nations Working Group on Enforced or Involuntary Disappearances, and is at present Distinguished Visiting Professor of Law at Hofstra University in Hempstead, New York. He is also an Attorney of the High Court of South Africa and of the State of New York. A graduate of the University of the Western Cape and of Harvard Law School he has been visiting professor at several US universities where he has taught Comparative Law, International Human Rights Law, International Criminal Law and Transitional Justice Southern Africa rights (South Africa, Botswana, Lesotho, Swaziland, Namibia and Zimbabwe)". It's a history of a genocide. Someone so well practised in relevant areas of law, who also has a long-standing interest in the relevant region is able to investigate this. With such a background he isn't going to put pen to paper unless he is sure he has the expertise to do it properly. Favourably reviewed in the Washington Monthly [1]. Itsmejudith (talk) 10:55, 24 November 2011 (UTC)

You had me at the obviously academic title and "Chairperson-Rapporteur". Visiting professoriates aren't always definitively indicative; teaching IHR ICL and Southern Africa rights is a very positive sign as it is highly connected to the publication. Washington Monthly isn't a scholarly journal, but it is a journal aimed at an intelligentsia. Why is this so obviously "right" for history? How can we encode this obviousness? In comparison, I recently checked Ervin Staub's Overcoming Evil: Genocide, Violent Conflict, and Terrorism, in relation to its capacity to critically evaluate contentious historical claims by other authors. Staub is a psychologist of genocide, and highly respected in this field, but he isn't a genocide demographer or historian of genocide. I concluded:

"Nobody has reviewed Staub's Overcoming Evil: Genocide, Violent Conflict, and Terrorism in a peer reviewed journal. Nobody has reviewed it in a journal concerned with genocide, or history. More: Genocide and Holocaust studies noted it in a bibliography but still didn't review it. Your assertion that Staub is "eminent" appears to be in question. PsycCRITIQUES (1554-0138) a non-peer reviewed journal (via Ulrich's periodicals directory) claims in a review (DOI:10.1037/a0025326), "highly personal book" "his research and applied work has focused on the psychology of good and evil" as opposed to demography or history "Staub writes in the first person" "Staub has certainly written for a wide audience, drawing examples from his own experiences in Rwanda, Congo, and Israel/Palestine" "Overcoming Evil makes a significant contribution to the corpus of psychology literature that highlights the powerful role of situations, rather than dispositions or personality, in creating “evil.” This work is suitable for both university students and lay readers." The review proceeds to praise Staub's contribution to psychology. Staub lacks the qualifications to evaluate history and demography. He would be greatly useful at Psychology of Genocide but not here. His capacity to make a judgement regarding Hollander's claims is null: he is appreciated as a psychologist, he is ignored by historians."

And believe myself to be correct in the conclusion that as an eminent psychologist, Staub is not placed to evaluate cumulative death tolls in "communist" states. Would you consider this an obviously correct evaluation of Staub's capacity as a historian; much like your evaluation of Sarkin was obvious to me from about the 14th word of your explanation? How do we encode Staub as "out" (if I am correct) and Sarkin as "in"? Or is this the Ignore All Rules limits of our capacity to evaluate history? Fifelfoo (talk) 11:18, 24 November 2011 (UTC)

Perhaps the only difference is that Sarkin is working with an archive that has only just been opened up whereas Staub is working with texts that have already been analysed many times before.

I think we might have consensus to make a statement up front "a history RS will usually have been written by a historian". And then somewhere start to spell out some exceptions. There can be an exception for anyone if the publisher, reviews etc. are good enough. The exceptions that have been urged on us here are:

1. Scholar in a neighbouring discipline (arts or social science). Possible solution: say that usually good, but scrutiny needed of the aims and purposes of the research, the quality of the publisher, reviews.
2. Specialist in a field writing the history of the field. Possible solution: say that usually good, but look to see if there is any critique of the methodology or findings and then only use with caution, also watch for bias if it is an official history.
3. Amateur. Possible solution: give exemplars of fields in which data collection is a major focus, and where amateurs often do some of that data collection (transport history, but not local history).

Itsmejudith (talk) 14:05, 24 November 2011 (UTC)

Comments after a quick look

I see some of these concerns have been discussed above, but anyway, as the results of one person's reading:-

I am confident the intentions are very good, but there does also seem to be something like a "closed shop" theme. I think trying to judge sources on the basis of author qualifications alone is bound to be open to abuse.

Defining historical articles as ones about the past is extremely broad and obviously not going to work in practice.

I can see that in order to be more nuanced, lots of complicated discussion is required. I fear this article will get longer and more complex. But very complex advice is also very open to abuse, by its very nature.

I was referred here from a discussion at WP:V about the old primary/secondary thing. I find that also concerning this point, the problems people have with using this distinction are not really being resolve by this article. Indeed, it uses the very simplistic approach of just basically saying primary is bad, whereas people constantly have quite reasonable problems even understanding the difference between primary and secondary sources, and even more reasonable questions about when it is acceptable to use primary ones. And it certainly is sometimes good to use primary sources.--Andrew Lancaster (talk) 12:49, 4 December 2011 (UTC)

Some Questions

Based upon the first sentence:

• 1. Historical articles on wikipedia should be the result of scholarly works.

What is considered a historical article on Wikipedia? Should, for example, "Malibeyli and Gushchular Massacre", "Iran Air Flight 655" or "1992 Los Angeles riots" be considered historical articles?

Question based on #3:

• 3. Relying upon medium or low quality commercial or popular works, or the results of reading newspapers, or journalistic opinion, is almost certain to constitute original research by synthesis if these sources are relied upon for weighting and article structure.

If an article is encountered that uses a source(s) written by a journalist or by individuals not recognized as historians, what actions are recommended? --Kansas Bear (talk) 06:30, 3 December 2011 (UTC)

Thank you for these questions which show where we need to clarify the guidance.

The three articles you mention are historical. They are about events from 1988 to 1992, and there has been time since then for historians to reflect on them and write articles and books.

The contrast is with an event still going on, or occurring only a few months ago. We have guidelines WP:RECENT that warn about the difficulties in those cases. Usually we use the mainstream media, e.g. AP, BBC, to write about current events.

What we should do when we find that an article is based on poor sources is look for better sources and gradually replace the poor ones.

So, what can we add to this guideline? 1) We can explain about the cut-off between current and historical, as described above. 2) We can explain that a current event will become historical. News media sources should be replaced by academic sources when they become available. 3) We might need to add to WP:IRS guidelines about how in practice you actually find good sources, and where you can go for help. Itsmejudith (talk) 08:01, 3 December 2011 (UTC)

I added a section to the guidelines, based on this discussion. Itsmejudith (talk) 16:58, 3 December 2011 (UTC)

Those were excellent questions, and the improvements to the guidelines are also excellent. Many events which were written as RECENT events may never receive serious scholarly attention; and, in those cases we'll probably need to figure out how to deal with "Old News." It is a matter of editorial judgement to determine how scholarly sources replace content written RECENTly. Often political scientists or sociologists may get at the topic first, with RECENTish analyses aimed at foreign policy or government practicioners. So quite often, by the time historians intervene, there are other scholars publishing in different ways who already ought to have been used for weight, analyses and colour! Fifelfoo (talk) 21:14, 3 December 2011 (UTC)

As a follow-up to my initial question, "What is considered a historical article on Wikipedia?", and your response(or part of it), "What we should do when we find that an article is based on poor sources is look for better sources and gradually replace the poor ones.".

If our "better sources"(ie. sources written by historians) change the term used to describe an event(the addition or non-useage of certain words "massacre", "civil war", "revolt", "riot" etc), by "poor sources", is a move request necessary(as per Wikipedia) and if consensus is against the term used by "better sources" what are our options? --Kansas Bear (talk) 21:45, 24 December 2011 (UTC)

Discussion elsewhere

Because this page has been linked to on WP:IRS I started a discussion there also. Is it really ready for the big time?[2] --Andrew Lancaster (talk) 11:13, 20 February 2012 (UTC)

I've been trying to push the beta so that it breaks by repeatedly mentioning the guidelines where apt on WP:RS/N. Sadly I haven't had any feedback nearly as critical as Andrew Lancaster's own critical feedback. Andrew's critical interest in this is of great importance. Fifelfoo (talk) 11:40, 20 February 2012 (UTC)

I think that as you are "trying to push the beta" then you should mark it as an essay so it is clear to inexperienced users that this page currently does not have the status of a guideline. I suggest in the you place a statement in the lead that after further development it will be formally submitted to the guideline ratification process. -- PBS (talk) 01:08, 24 February 2012 (UTC)

Thanks Philip, I hadn't marked it as an essay because it seemed to be below even that standard. Fifelfoo (talk) 01:21, 24 February 2012 (UTC)

Discussion on "Misuse" copied over from Wikipedia talk:Reliable sources/Noticeboard

Misuse of HISTRS

A frequent contributor to this noticeboard recently said at an article page, "Let me clarify this: you are ignoring the reliable sourcing standards for this article WP:MILMOS#SOURCES and WP:HISTRS..." and similar statements can be found by that same editor and (far less frequently) by other regular contributors in posts at this noticeboard, though most of the posts here do not so plainly claim HISTRS to be a "standard". HISTRS seems to be being quoted as if it were a policy or guideline when, in fact, it is currently no more than an essay and, indeed, an essay created by and mostly worked on by the editor who most commonly refers to it and also to certain criteria and sources cited at WP:MILMOS as if they were policies or guidelines. The WP:CONLIMITED policy makes it very clear that projects and, by extension, venues such as this noticeboard (and, indeed, dispute resolution venues such as WP:3O, WP:DRN, and WP:MEDCAB where I most frequently work) cannot establish policies and procedures which are binding on the Wikipedia community as a whole, at least not without going through the policy-making procedures set out in WP:POLICY. Treating and referring to HISTRS or criteria established by MILMOS as "standards" or as if they are somehow otherwise binding on all of Wikipedia is misleading, especially to newcomers (and who make up, I suspect, a disproportionate part of the folks who end up coming to this noticeboard) and others who are not sophisticated in the underpinnings of WP, and I object. If HISTRS is to be treated as a policy or guideline, it should be submitted to the Wikipedia community as such for evaluation (and I am not at all certain, I must frankly admit, that it will or should survive such scrutiny), but until it has been and passes that scrutiny, it should not be treated or referenced as if it already has. Regards, TransporterMan (TALK) 04:09, 22 June 2012 (UTC)

Newspapers, biographies and pulp tripe are clearly unacceptable for historical articles. If you can't read the existing policy, that exceptional claims (such as scholarly WEIGHT, or historiography) require exceptional sources (such as scholarly sources), then you ought to stop editing in fields covered by scholarly sources. Fifelfoo (talk) 06:05, 22 June 2012 (UTC)

I've worked on HISTRS too. The point is that codifying what makes a good source for articles in a particular subject aids decision making and sorts out arguments before they have time to develop into major rows requiring dispute resolution. If you think HISTRS is too restrictive, come and argue that on the talk page. Itsmejudith (talk) 06:26, 22 June 2012 (UTC)

I learned a lot just looking at HISTRS. The problem with it is that it calls for examination of literature that often does not exist, and if it does, is inaccessible. And, as noted above, it is neither policy nor a guideline. User:Fred Bauder Talk 12:41, 22 June 2012 (UTC)

"New York Times is shite as no NYT editor is a historian" is just nonsense. For most purposes The New York Times is an acceptable, even preferable source, especially for use by those experienced with its failings. User:Fred Bauder Talk 12:36, 22 June 2012 (UTC)

Which is crap. The NYT publishes on a daily schedule with no access to archival material, and its journalists lack the analytical techniques that historians possess. It is quite simply insufficient for history, it lacks any methodological claim to produce history. Fifelfoo (talk) 12:38, 22 June 2012 (UTC)

The New York Times is archival material, see footnote 1 at Cananea. User:Fred Bauder Talk 12:47, 22 June 2012 (UTC)

I absolutely agree that this is not the place to argue the merits, or lack thereof, of HISTRS, and being in the DR wiki-business I wholly agree that "codifying what makes a good source for articles in a particular subject aids decision making and sorts out arguments before they have time to develop into major rows requiring dispute resolution". My only point here is that it is inappropriate to use or refer to it as a standard before those merits have been established. Regards, TransporterMan (TALK) 13:34, 22 June 2012 (UTC)

Let's continue the discussion on the talk page of WP:HISTRS because I have several comments to make. Itsmejudith (talk) 14:14, 22 June 2012 (UTC)

This essay has very useful information and suggestions in it to the extent they are useable in the context of Wikipedia editing. As a continuation of the discussion may I point to the use of newspaper archives for the purpose of showing the nature of contemporary press coverage of a matter, Cananea strike. User:Fred Bauder Talk 14:44, 22 June 2012 (UTC)

(ec)So, to continue, comment 1, see what people say to this first. The NYT vs historian question is already addressed on the project page. For recent events, NYT and the media of similar standing are usually the best sources. Some time afterwards, there may be more in-depth, feature-length studies in serious magazines. Given a little bit longer, some scholars may take an interest in whatever the case is. After that the event "is history", and it is only historians writing about it. We gradually supplement the news sources with the accounts that have the benefit of hindsight. Eventually, the original NYT report is a primary source for the article. No-one is suggesting taking news accounts out if they are the only sources that exist. Occasionally there may be an event that seemed important at the time but is in the process of being forgotten. We have no real way of dealing with that, as far as I know. We assume that if something is notable it will stay notable. There aren't many such cases, anyway. Itsmejudith (talk) 14:56, 22 June 2012 (UTC)

A question it's always important to ask is: Is there an academic literature on this topic? And in the case of the Cananea strike of 1906, yes there is. At least there is this paper: C. L. Sonnichsen. "Colonel William C. Greene and the Strike at Cananea, Sonora, 1906". Arizona and the West Vol. 13, No. 4 (Winter, 1971), pp. 343-368. One of the first links that came up in Google, and then when I search in Google Scholar, I get 407 results. So there is a literature, and how can we not cite it. And here is a paper that eloquently explains why. Rodney D. Anderson. "Mexican Workers and the Politics of Revolution, 1906-1911". The Hispanic American Historical Review, Vol. 54, No. 1 (Feb., 1974), pp. 94-113, worth citing at length, though I have cut a little out:

Marjorie Clark's early study attributed the strife to socialist doctrines introduced into Mexico by a small group of intellectuals, primarily foreigners, and more recent scholarship has reaffirmed this interpretation.7 ... other scholars have maintained that the exiled Partido Liberal Mexicano (PLM) was the vehicle by which these doctrines were transplanted to Mexican soil, eventually enlisting the workers in a struggle against the Diaz regime and, as some have argued, against the capitalist system itself.8 Their arguments exhibit certain methodological weaknesses. Most are based on relatively few samples of Mexican workers' actions ... More significantly, they have virtually ignored what Guillermo Torres and his fellow industrial workers said about themselves ... Mexican industrial workers have been rendered "inarticulate" by historians who have written about them, aided and abetted by those of us who have assumed that they could not speak for themselves. Industrial workers in Mexico between 1906 and 1911 spoke often, and well, and if they were not much listened to then, the historian should do better now.9

In the light of those comments, can we really expect to write up a good article on the strike just from the documents of the time? No, we need to access and evaluate all the recent academic literature. Itsmejudith (talk) 15:16, 22 June 2012 (UTC)

That's all good, although I don't think I have access to that material. The contemporary press reports are, in addition, very interesting. It would be great if someone who has access to those sources would work on that article. User:Fred Bauder Talk 16:38, 22 June 2012 (UTC)

NYT is a good source for current news, and a good source for the view of the "popular press at the time" historically, it's a bad source for historical analysis, which should be the top tier of sources for any historical article. Also, just a sidenote for Transporterman, the "rulings" of editors at RSN aren't policy/law either, it's just our opinions. Granted, often, our opinions make a lot more sense than some of the authors that bring us questions, but what we say isn't binding on anyone. I don't think anyone that spends time at RSN has any agenda other than to try and help the articles that ask for our opinions to use only the best sources possible. -- Despayre  ^{tête-à-tête} 15:34, 22 June 2012 (UTC)

I think there is some misuse of RSN as authority too, as a weapon by tendentious editors, especially with respect to historical issues where there is a body of biased scholarly work extant, recent discussions at Talk:Occupation of the Baltic states being a good example. That said, I need to participate more at RSN rather than simply look on. I'm sure I would learn a lot about sources as well as offer opinions based on use of available sources. User:Fred Bauder Talk 16:38, 22 June 2012 (UTC)

Generally speaking I don't think anyone currently a regular contributor at RSN is so naive as to think that RSN isn't being used as a "weapon" in some disputes, and I know I'm aware of that when I respond, however, it doesn't often make a difference to the reading of the policies in place, and most of the time when that type of manoeuvre is attempted, what is really being discussed are issues of weight or notability, which I make clear, that I'm not commenting on. -- Despayre  ^{tête-à-tête} 16:55, 22 June 2012 (UTC)

One of the advantages with WP:HISTRS is that it sends editors attempting to only use limited or non-standard historiographies back to the scholarly literature, in particular through the concept of privileging literature reviews. For an example, see my response on WP:RS/N regarding the occupation of the Baltics bruhaha. Editors have been trying to argue that a scholar's views shouldn't be included, and have been suggesting that Western historiography is superior because of its origins, instead of due to its reputation with historians and its positive evaluation (comparatively speaking) in field reviews. The best response to an absence of scholarly sources in history articles, is to push scholarly sources; the best response to a use of FRINGE, unaccepted minority positions, or only one of a number of acceptable historiographical positions is to push scholarship even harder by pushing field reviews. This pushes editors away from their own innovative interpretations, and towards the preponderance of scholarly opinion. Fifelfoo (talk) 03:12, 25 June 2012 (UTC)

Newspapers, biographies and pulp tripe are clearly unacceptable for historical articles.

I think this comment is seriously wrong. The newspaper thing seems to have attracted the attention above, but the biographies claim disturbs me even more. The biography underway at our house is The Passage of Power: The Years of Lyndon Johnson by Robert A. Caro. It is 600 pages of text followed by 100 pages of footnotes. It is written by a prson who has won multiple awards for history and biography writing. It is the fourth of five volumes about LBJ, and I haven't yet seen a single good reason why it could never be a reliable source for all sorts of information about the subject. The idea that it is "clearly unacceptable" because it's "a biography" is just nonsense. The idea that it is unacceptable because the author earned is living by being a "journalist" rather than an "academic" is plain bigotry. The editors at that article agree with me about the source: they've repeatedly cited Caro's biography.

We should not be making these anti-policy claims. The actual policy is that "pulp tripe" is an acceptable source for some purposes, and that even gold-plated academic works are unacceptable for some purposes. I think there is some value on this page-in-progress, but these sweeping statements have got to stop. WhatamIdoing (talk) 18:27, 22 June 2012 (UTC)

That statement was an exaggeration. But we do need to distinguish between history and historical biography. Itsmejudith (talk) 18:40, 22 June 2012 (UTC)

How and why would you distinguish between them?

Consider a statement like "Johnson created the Warren Commission to investigate Kennedy's assassination." Would you demand different sources for that if it appears in the article about LBJ ("biography") vs the article about the assassination ("history")? What kinds of sources would you accept: school textbooks, newspapers, biographies of JFK or LBJ, the archived official order, or only formal scholarly sources written by paid, professional historians? WhatamIdoing (talk) 18:51, 22 June 2012 (UTC)

The bare statement hardly needs a source, so nearly any source will be good enough; however, I can imagine attempts to tell an "inside story" that might require very good sources indeed. User:Fred Bauder Talk 19:03, 22 June 2012 (UTC)

Is that the conclusion you think you would come to, if you were an inexperienced editor who had been pointed at this page, or if Fifelfoo told you that "Newspapers, biographies and pulp tripe are clearly unacceptable for historical articles"? WhatamIdoing (talk) 19:24, 22 June 2012 (UTC)

We always have to pay attention to what the core problematic of a source is. Tristram Hunt is a historian, and he wrote a well-received biography of Engels, and it contains verifiable facts about 19th century Manchester. It still isn't a history of 19th century Manchester, he wouldn't say it was. If he wanted to write such a book he could, but he hasn't, and we need to look elsewhere. Hope that makes sense. Itsmejudith (talk) 21:15, 22 June 2012 (UTC)

Biographies and specifically autobiographies has been discussed several times on the talk page of WP:OR. autobiographies have been explicitly excluded from inclusion as a primary source in WP:PSTS, because of military histories. The example I have used a couple of times is "Defeat into Victory" by William Slim ( here is one from 2008 with an example of usage and here) but there are dozens of other examples that can be used. This has led to the sentence in WP:PSTS: "A book by a military historian about the Second World War might be a secondary source about the war, but if it includes details of the author's own war experiences, it would be a primary source about those experiences." -- So I am not sure where the idea comes from the biographies automatically means inaccurate. -- PBS (talk) 12:23, 23 June 2012 (UTC)

I was gesturing with hyperbole at the fact that trade presses publish an awful lot of biographies written by journalists, non-specialists, or persons with specialities that mean that they have no historical standing. These vary from works like Caro (2012) The Passage of Power Knopf, above, in an imprint that makes me trust Caro's standing highly, through to mass produced pulp written by hacks on day wages. Seeing Caro was published by Knopf made me look for the reviews in historical journals (but its 2012, so they're not there yet). Biography by historians, which does exist but tends to be a minority in a discipline where political, social and cultural history dominates, is worthwhile. So is Caro's. They obviously the best sources for biographical encyclopaedia articles. They often make poor choices for political, social, cultural or military historical articles. I wouldn't trust Caro to evaluate the culpability of Johnson versus McNamara for the air war in Vietnam, but I'd trust Caro over Johnson's own participation in decision making. In contrast Robert Dallek (1998) Flawed Giant OUP looks even better given that Dallek holds a PhD and has published multiple monograph histories on the biograpies of US presidents. (Damnit, LBJ monographs seem to vary between high quality journalistic biographies and historical biographies conducted by specialists). But you do, I assume, get at what I'm talking about regarding coffee table biographies put out by book mills? Fifelfoo (talk) 03:28, 25 June 2012 (UTC)

I understand that you prefer the best possible sources.

Do you understand that "not the best possible source" does not actually mean "unreliable"?

Those coffee-table history books, or even children's history books, are reliable sources—for some, generally simpler, purposes. The majority of stuff we actually need to source in a typical history article isn't at the complex level of McNamara vs LBJ. Most of what we need to source is basic, indisputable facts at the level of "In fourteen ninety-two/Columbus sailed the ocean blue". We can, and IMO should, accept non-scholarly sources—sources that the typical reader can understand—for this sort of basic information. Ham-fisted assertions that only the best scholars will do really isn't on. A source is reliable if it meets our minimum standards, noy if it's the best possible source. WhatamIdoing (talk) 21:40, 25 June 2012 (UTC)

Not the best sources available can mean unreliable, unreliable for WEIGHT, unreliable for the preponderance of scholarly opinion, or reliable. Caro above is an example of a non-scholarly history, published in a highest quality non-fiction press. I'd consider it reliable for its subject as any scholarly work, including contributing towards a comparative evaluation of the WEIGHTs placed and structures used and opinions put. As itsmejudith points out, Caro's work is next to useless for America in the 1960s. In particular, allowing editors to coatrack together popular works means that they aren't reporting on the subject of the article at all if it is covered by scholarly history. There is a group of people who define what history is, and wikipedia's articles on historical topics are broadly and consistently ignoring the sources that produce weighting, structure and opinion. Fred's concern, which I share, about deviant historiographies produced by one scholarly community but rejected by the broad scholarly community is a case in point—the manner in which to resolve Fred's concern is by moving towards the scholarly meta-analysis. Not using the scholarly works means that the encyclopaedia is not infact discussing the topic, but conducting original research, and poor quality original research at that. Fifelfoo (talk) 22:57, 25 June 2012 (UTC)

No, that's not true. There is no rule that says "Thou must cite a scholarly source, or you aren't writing about the subject." You could use a bunch of non-scholarly sources to create a big mess, but you can do the same with scholarly sources—and that happens every day of the week.

You are misusing WP:COATRACK, by the way: that's when I declare that the subject of the article is LBJ but then put a lot of cruft into the page that isn't about LBJ. It's not when I cite a source that is primarily about LBJ to support some detail that is only tangentially related to LBJ.

Basically, I share your concerns about WEIGHT and neutrality. But I don't think that giving too much attention to a minority theory means that the source is unreliable, and I don't think that ignoring scholarly works automatically creates NPOV problems. To give a parallel example, I could write a perfectly NPOV article at Leukemia without citing a single scholarly source. Every sentence in that article now could easily be sourced to patient-oriented websites. The only thing that would change is the citations. I don't choose to, but it could be done. The sources would meet the minimum standards for reliability, and the article would comply with NPOV and NOR. I believe the same could be done for other basic subjects like Algebra or Library or Bread. You are saying that this isn't possible in history: that unique among all the subjects in the encyclopedia, an editor cannot write a decent article about history unless he cites scholarly sources, because a failure to cite a scholarly source is itself a WP:NOR violation, no matter what is written in the article.

That's not what the policies say. NOR, for example, demands only that the claim be made in some source. And then based on your novel idea of what the policies say, you go forth and make sweeping statements that you don't even believe in practice. Biographies are all useless, you say, but you'd accept Caro's biography. I think that you need to be far more careful about what you claim is in the actual policies. WhatamIdoing (talk) 17:35, 28 June 2012 (UTC)

Policy restriction?

As currently written, the "Nutshell" says, in brief, scholarly works should be used when available and, if not available, "the highest quality commercial or popular works should be used." If this essay were promoted to a policy or guideline, would that mean that sources which would ordinarily be considered to be reliable sources could not be used at all for historical articles because they were not of "the highest quality"? That is, does this restrict the general definition of reliable source by saying that some sources which are reliable sources for some purposes are not reliable sources at all for historical articles? I don't think that is the intent, but I'd like a clarification from those of you who have been working on it. On the other hand, I would suggest that if that is the intent, then the definition of what is and is not a historical article must be considered very carefully as this limitation changes the definition of reliable source for the vast majority of the articles in Wikipedia. The current definition, "Articles which deal with events in the past." can apply to articles about IBM or Leave It to Beaver as well as they can to more narrowly-historical articles such as Punic Wars. I do not want to give the idea that I am opposed to a help page setting out what the best practices are for sourcing serious historical articles and, hopefully, combating a trend towards lack of discernment between the quality of sources, but I have a great deal of concern about setting up new rules. Regards, TransporterMan (TALK) 14:01, 25 June 2012 (UTC)

The caveat is that using biased "scholarly" articles, for example in Soviet studies, in order to trump other sources is not acceptable. There should never be a problem with replacing popular sources with better scholarly sources; tendentiously changing the content of articles and replacing popular sources with nearly worthless scholarly sources is not. Most editors don't have much access to journal articles; we do the best we can, which, by the way, is acceptable for editing most articles, as that is the working definition of not available. User:Fred Bauder Talk 14:55, 25 June 2012 (UTC)

I read that to mean "the highest quality [available]", whatever that is, there is something that always fits that description. -- Despayre  ^{tête-à-tête} 15:14, 25 June 2012 (UTC)

Yes, you understand it correctly: this is an attempt to re-define "reliable" to mean "best possible" rather than "good enough for the purpose at hand". That's why it has no chance of being adopted in its current form. WhatamIdoing (talk) 21:43, 25 June 2012 (UTC)

This discussion is useful in helping us define 'historical articles. IBM is not a historical article but it ought to have a history section. All major companies have standard company histories, usually good sources that we should normally draw on. TV shows, no; we have a body of expertise in how to soure popular media articles. Itsmejudith (talk) 22:31, 25 June 2012 (UTC)

Popular sources are not good enough as they do not convey history. Scholars agree on what historical studies are, and they're not the bumper coffee table book of Nazi weapons, nor are they works which subsist outside the post-Rankean discipline and its review structures. History is defined, rather clearly, by the scholarly historiography. Fred's concern that the historiography may be incorrect is a specious one. We're not here to determine what is history, any more than we're here to determine if continents move or not. Editors who coatrack together a narrative and weight out of whole cloth are engaging in original historical research, and it is as FRINGE as citing the rejected historiographies that Fred fears. We're here to report what others believe happened. And when it comes to the past happenings of human society, there's only one group of writers who do that. Fifelfoo (talk) 22:36, 25 June 2012 (UTC)

Nope. Wikipedia does not have a "Historians' Point of View" any more than it has a "Scientists' Point of View". We present all the relevant and significant POVs, not just the POV of "only one group of writers".

But even if we did, that would not prevent us from citing pop sources. If (having considered the full range of sourceable material) you determine that the article needs to say A, B, C, and D to comply with NPOV, then, so long as the sources cited are reliable for those four points, it's okay if you cite four coffee-table books. Pop sources are plenty good enough for most of the basic information we need to present (like names, dates, and places). In fact, they may be better at "conveying history" than many an ill-written, jargon-laden scholarly source. WhatamIdoing (talk) 17:44, 28 June 2012 (UTC)

We seem to be a million miles apart. I have always argued that where there is a scholarly literature we must follow what that literature says. Otherwise, how do we know what is mainstream? Of course there are many articles which use non-scholarly sources, and where that poses little problem. Eventually, we should replace the non-scholarly with the scholarly. For example, I created the stub Stratford Langthorne Abbey, which another user then expanded. As the article developed, editors looked up the best sources and used them. There are plenty of non-scholarly sources around, for example tourist guide sites. They usually take all their information from other published work, and in the process they introduce bias and error. I corrected the article on Arundel to make it clear that the river's name is a back-formation from the town. I was reverted in good faith, because common sense and the guidebooks said it was the other way round. I held my ground because I had used good place-name sources and I knew that this was typical antiquarianism. From one poor source, one can work backwards towards their sources. Eventually you will come to the chronicles that are the only original primary sources. But that is to work back much too far, because the chronicles are difficult for non-specialists to interpret. Instead, we work back or sideways from the coffee table books until we find the standard histories of the particular topic/period. Virtually every topic has a written history that is definitive and accepted. We don't have to use only that source, but what we write has to be compatible with that source. Please come back and argue against this because it is extremely important. It's what writing history actually is. We demand that our physics articles be compatible with the scientific literature. Why not here? Itsmejudith (talk) 19:28, 28 June 2012 (UTC)

On the assumption that this reply is to me: "compatible with" is not the same as "cited to".

Let me give you an example: Pretend that I want to write, "In 1492, Columbus sailed the ocean blue" in an article about the European discovery of the Americas. This is an undisputed fact: nobody with the tiniest shred of credibility believes that Columbus sailed in a different year, or that Columbus didn't ever get on a boat, or anything except what I've written in this sentence. This is also an indisputably history-related fact: everyone agrees that it describes a significant event in the past. So (IMO) how to support this kind of sentence should be the main focus of any advice page related to history.

Fifelfoo is saying that I must cite a scholarly source for this sentence. Fifelfoo says that citing non-scholarly sources violates NOR because it means that I'm making up my own historiography. Fifelfoo has said that newspapers, biographies (e.g., a biography of Columbus), coffee-table history books, school textbooks, etc., are all clearly unreliable sources for any sort of historical statement. Fifelfoo says that "there's only one group of writers" who are permitted to tell us whether Columbus sailed in 1492, and that group is tightly limited to writers inside "the post-Rankean discipline and its review structures".

I say that citing a coffee table book, or a children's history book, is perfectly adequate. No matter what source you stick behind this sentence, the sentence will be presenting the mainstream view.

You've do have to get the article right. You don't have to WP:CITE scholarly sources to get the article right. WhatamIdoing (talk) 20:09, 28 June 2012 (UTC)

We agree that you could cite many sources for the 1492 date. In many such cases we have more than one reference, which is OK but beyond about 3 references for one point it starts to look weaker rather than stronger. What I would say is that there are some well respected and well reviewed histories of the Columbus exploration, of which at any one time, one is regarded as the standard work. Any university teacher will tell you that you must identify and refer to that standard work. You can use other sources as well, but you must as a basic read the standard work. Now, many of readers are not looking for that depth. Say the reader is a child. We should include good school textbooks in further reading, but we don't have to dumb down the article just to allow for child readers or for readers whose first language isnt English. I must say that I am starting to be shocked by anti-intellectual assumption, which I don't see as compatible with the goal of making an encyclopaedia. At the moment I am reading Jonathan Sumption's history of the Hundred Years War. Sumption isn't necessarily even "a historian"; he's a praticising lawyer. But you can see a quantity of research that leaves any coffee table hopeful standing. The archives consulted! The manuscripts ploughed through! In Middle English, Old French, Latin, Castilian, Italian, German, Flemish and German. These things ought to, and do, command respect, even if it is not your kind of history, which it isn't mine, really. I would beg you to consider the concepts of "scholarly community" and academic mainstream. If a reader has to visit a major library or pay £12.99 through Amazon to verify a reference, that is likely to be time/money well spent. Itsmejudith (talk) 21:18, 28 June 2012 (UTC)

(edit conflict) I agree with WhatamIdoing with perhaps with one clarification. Let's say Editor X adds an assertion that in the 1862 Battle of Antietam article that the 118th Maine Infantry (I'm making them up from scratch) was one of the Federal units that took Burnside's Bridge from the Confederates. That assertion is sourced with an inline citation to a 2008 New York Times article about a historical reenactment in which a representative from one of the reenactment units claims to be the "unit historian" and says, "Today we're portraying the 118th Maine, one of the Federal units who seized the bridge from the Georgia Confederates." Editor Y reverts and challenges the addition on the ground that Editor X failed to meet his WP:BURDEN because the source cited by X was not an academic source. Is Y correct? I say no, the NYT source is an acceptable source, subject to being corrected, replaced, or supplemented by a reliable academic source. If we say the contrary, there are many articles about historical matters which simply cannot exist or expand beyond stubs simply because editors do not have access to academic sources or because academic sources simply do not exist. Even when they do exist, there is no hurry; plain-old-reliable is good enough until someone cares enough to make it better and it should, in effect, shift the WP:BURDEN to the individual who chooses to do so. It is especially likely to be corrected in fairly short order if the assertion is wrong and the article is about an important piece of history. Citizendium failed largely because nothing could be done except to academic standards. Regards, TransporterMan (TALK) 21:29, 28 June 2012 (UTC)

Academic sources do not exist is a completely different case from editors do not have access to the academic sources. Newspaper sources are sometimes reliable but equally sometimes they are pay walled. Tough luckie. Itsmejudith (talk) 21:53, 28 June 2012 (UTC)

I took a year as randomly as I could've:,1581. There I find Edmund Campion , could be a lot worse, could be a lot better. I read a whole Pelican book about EC that isn't cited there. Are we going to be contented with the minimum? Or ar we going to spell out what good sourcing consists of? Itsmejudith (talk) 22:11, 28 June 2012 (UTC)

"Any university teacher will tell you that you must identify and refer to that standard work"—but the English Wikipedia is not a university and does not impose any such requirement on any article. I personally take that approach when approaching a new subject—but I do not confuse my personal preference for the "best possible" source with the "minimum actually required by English Wikipedia policies".

When we are defining "reliable" and "unreliable", we are talking about the line between a source that cannot be used to support a given statement and a source that is just barely at the absolute, rock-bottom minimum for what may be used. WP:RS and its friends do not spell out what best sourcing consists of, and they do not tell you how to go about researching a topic so that you can understand the subject. The primary purpose of these guidelines is actually to define the worst source that will still be considered (just barely) good enough for any given statement. That's a plain-old-reliable source, not a "practically perfect in every way" source or a "what you would pick if you had a team of professional researchers" source. The hypothetical newspaper article above is a plain-old-reliable source. It is 100% acceptable (so long as it is uncontradicted). We do not reject this source, and we do not remove (apparently) accurate information supported by it, even if we might optionally choose to "upgrade" the source to a classier one. WhatamIdoing (talk) 22:53, 28 June 2012 (UTC)

To put it more succinctly: This is about what is required or mandatory, not about what is good or wise or smart. WhatamIdoing (talk) 22:56, 28 June 2012 (UTC)

Still very much unconvinced. A page called Identifying reliable sources should do just that. Which sources are reliable for history? Books and articles by historians. I wonder if you are coming at this with more of a concern for happy editing than for happy reading? That's not meant to be a nasty comment, just that we often get bogged down in finding ways to resolve disputes, and sometimes it is useful to be recalled to the core purpose, which is making knowledge available in the form of an encyclopedia. An article on Christopher Columbus where one source is the New York Times might be somewhat surprising. An article on Christopher Columbus written up entirely from sources of that kind would make us a laughing stock. To repeat, our articles on natural science topics are based entirely on academic sources, and that is what makes them credible. Why should the same approach not apply to articles on social science topics? Where there is scholarship, the scholarship constitutes the knowledge base. To use anything else is a disservice to readers. The New York Times is a go-to source for news. For film articles, reviews in film magazines, fine. Itsmejudith (talk) 10:56, 29 June 2012 (UTC)

I fear that the topic of this thread may have been lost. The question is whether it will make academic sources mandatory. I again agree with WhatamIdoing, but I also agree with Itsmedjudith that a guide such as this is extremely useful so long as it is limited to developing best practices, not mandatory practices. I come at this from the dispute resolution point of view: If two editors come to 3O or DRN in a dispute about X adding an assertion based on a popular, but reliable, source and Y removing it because she claims that an academic source is required but not offering one, just removing the assertion, my opinion is going to be that, "while an academic reliable source is to be preferred, a popular reliable one is sufficient to allow the assertion to remain in Wikipedia, if Y feels that an academic source would be better, Y should let the material remain in Wikipedia, find such a source and add it to the article, but X has met his WP:BURDEN by giving an inline cite to the popular reliable source. X should also attempt to find an academic source, if it exists, but X is not required to do so." I'm not trying to figure out what is best, but what is the least. Regards, TransporterMan (TALK) 13:54, 29 June 2012 (UTC)

The thread may have been lost. Part of the problem is that we can't agree on what it means to identify reliable sources. Are we trying to identify Best Possible sources, or are we trying to identify Sources That You Can Rely On?

So here's a concrete example: Harper's Weekly (a major political news magazine) ran an issue on April 29, 1865 dedicated largely to the assassination of Abraham Lincoln. Is this source reliable for a claim that Lincoln was shot at the theater?

I will say that yes, the source is reliable for this claim. It is absolutely, without any doubt, a source I would recommend against using—but that it does meet the basic standard for being a reliable primary source. If you "rely on" this source, you will get that fact right.

I suspect that the authors of this page would say no. The point isn't to get the facts in the article right; the point is to show that someone with official credentials as a Real Historian has endorsed this fact, and that Wikipedians are sophisticated enough to produce gold-plated sources that prove the fact is not only right but also endorsed by the favored branch of academicians.

My solution to this difference is to rename the page to WP:Identifying the best possible sources (history). WhatamIdoing (talk) 19:07, 10 September 2012 (UTC)

Definition of History

I might have written this in the first place but now think if needs expanding. We need to say something about biography. Literary biographies are often written by literary scholars. Biographies of statespeople can be written by historians or literary scholars specialising in biography. Biographies of scientists can be written by scientists. We may need to warn against reliance on biographies for historical facts.

Then there is the question of historical fact in articles that are not mainly about the past. The History sections in articles about towns and cities, for example. Presumably we expect these guidelines to apply there too. Itsmejudith (talk) 07:28, 25 October 2012 (UTC)

I honestly don't believe that the community expects this proposal to apply anywhere. Most biographies are perfectly competent at reporting relevant historical facts. I'm sure, for example, that nearly every biography of every deceased US president so far has correctly recorded the dates of the subjects' deaths. Those dates are "historical facts" and there is absolutely no need to "warn against reliance on biographies" for relevant historical facts.

It might be worth talking about when a source is appropriate and when it is not. A biography is very likely to be an appropriate source for anything directly related to the subject. It is not likely to be a very appropriate source for anything unrelated to the subject, e.g., people or events mentioned in passing. The information might well be accurate (e.g., the date of JFK's assassination could be taken from a biography about JFK or LBJ or Jackie Kennedy, but not from a biography about a person wholly unconnected to that event, even though both will give the same dates), but passing mentions are not generally appropriate choices for your sources. WhatamIdoing (talk) 22:01, 12 November 2012 (UTC)

Review articles

Where do we place in the hierarchy review essays and review articles that review a group of similar texts together? I would say fairly high, but at the moment we only refer, explicitly, to "single item" reviews. And is the text sufficiently clear about the difference between consulting scholarly overviews in order to define the consensus (on the one hand), and drawing facts from articles and monographs (on the other)? There's potentially a lot of confusion because in the natural sciences people often treat articles as primary sources, in that they deal with "only one experiment". Yet in the social sciences articles - including reflective and argument articles - are usually regarded as secondary sources. Itsmejudith (talk) 16:13, 12 November 2012 (UTC)

I think it's current placement in the section on Weighting and Structure is fine (no. 2). Most monographs and edited volumes contain similar treatments of historiographical issues (often not a consensus - more an outline of the field over time) so it's ultimate placement is necessarily somewhat arbitrary. Facts, I think, are the easiest point to deal with and, in my opinion, you can generally set a somewhat lower bar for these in terms of appropriate secondary sources. Many "amateur" historians or otherwise deviant personalities are perfectly good at collecting facts. Where you need to be really rigorous with source selection is in the adoption of interpretations. Obviously, current journal articles for historians are not primary sources although what is a primary source is dependent on the object being studied. But there are fairly normative definitions of primary sources: e.g. a document created in the time under study. FiachraByrne (talk) 02:26, 13 November 2012 (UTC)

How to switch out "sources"

Per Fifelfoo's previous response, "Not a historian, popular work reliability (replace with historians asap, kind of unscholarly TERTIARIES actually): Andrew Wheatcroft; Frances Stonor Saunders; Alison Weir (when published as history, not historical novel).
Has anyone checked out the Wars of the Roses article? It contains at least 20 sources attributed to Alison Weir. With some of the sentences, in my opinion, of minor importance and not really needing a source, how should one go about replacing/removing said Weir sources? Carefully? With a comment on the talk page stating she is not a historian and should not be used in this type of article? Leave a link to this project page? --Defensor Ursa 03:08, 5 September 2012 (UTC)

Using a scholarly source which is generally recognized as a good reference in this area, which is easiest if you are reading it, as you read the book read our article at the same time. As you come to material in the source on particular pages, insert a reference to the page or pages of your source that covers the subject replacing use of the popular work. I haven't looked at the article, but do that in the form the article uses but use a form which includes page numbers in the reference. With respect to footnotes which actually need no reference, swap those out also; they do no harm. Please don't remove the information supported by the popular work without doing more, that wrecks the article and makes more work for someone later. User:Fred Bauder Talk 10:23, 5 September 2012 (UTC)

On a well-developed article, it can be a service to the readers to provide varying types of sources. That's why some of our articles about, say, cancer or heart disease include sources for basic facts that are distinctly non-academic and very patient-friendly.

Most of our readers don't actually look at the sources, so the main thing is to get the article's content right. If the only way to get the content right is to rely on esoteric academic sources, then you're stuck with that. But in most of our articles, it's not necessary, and using a variety of education levels among the sources makes not only our article, but also some of our sources accessible to our readers. Among the minority of readers who actually look at our sources, many of them are school-age kids or just starting at university, and they're looking for sources that might be useful in their introductory-level coursework. A decent "popular" book or a decent textbook is valuable to these readers, as are sources that are available free online. An academic journal that specializes in six-syllable words and unexplained references to previous academic publications is not generally going to be useful to our readers. We should use such sources whenever necessary, but we should not be rejecting reliable sources that are strong enough to support the actual claims they're behind just to put an academic sheen on the references section.

If a decent lay-oriented source is available, then I support using it to support some uncontested points in an article. WhatamIdoing (talk) 18:56, 10 September 2012 (UTC)

We have Further Reading for that. In disciplines that involve the analysis of complex language based sources popular works are often quite simply "wrong." Their authors lack interpretive capacity (as demonstrated by their publication mode, or with early publishers by their lack of an appropriate apprenticeship), their publications are produced for profit rather than to accurately represent the best efforts to understand the world; or, worse, in the case of works aimed at secondary students and undergraduates are produced with the intention of deliberately misrepresenting the world so as to better equip students to potentially correctly understand the world in future. This is when they're not fifty years out of date by default. "Simple facts" are rarely worth citing without an appropriate contextual frame in these fields. An annotated further reading section is almost always superior if the purpose of including a work is to recommend it to encyclopaedia readers whose capacity means they cannot directly access the state of the art. Fifelfoo (talk) 21:56, 10 September 2012 (UTC)

• Simple facts are frequently required by policy to have an inline citation even if you personally believe them "rarely worth citing".
• You derided the authors of popular books as being incompetent, money-grubbing, and dumbing down their works. To which I say: Surely You're Joking, Mr. Feynman. It's a popular book, it was published for-profit (in mass-market paperback, even), it simplifies and glosses over complex work—and it was written by a Nobel Prize winner. WhatamIdoing (talk) 01:03, 11 September 2012 (UTC)

And you propose using Surely You're Joking, Mr. Feynman for the scholarly consensus regarding work in that field? Perhaps I understated myself: "simple facts" are not worth including in the encyclopaedia because disciplines that analyse the contents of language objects do not produce simple facts, they produce complex contextualised findings. Introducing "simple facts" is unencyclopaedic because it is not what the knowledge system we are reliant upon for the production of information in this area actually produces. Fifelfoo (talk) 01:30, 11 September 2012 (UTC)

• You said, "popular works are often quite simply "wrong." Their authors lack interpretive capacity". Shall we ask the folks at WP:PHYSICS whether a Nobel-prize-winning physicist "lacks interpretive capacity"?
• "Simple facts" includes statements like dates. Consider Armistice with Germany, which "was an agreement that ended the fighting in the First World War. It went into effect at 11 am on 11 November 1918,". What's so "complex and contextualized" about that date? Or is this kind of information "not worth including in the encyclopaedia"? WhatamIdoing (talk) 21:54, 12 November 2012 (UTC)

I think for history where facts are contentious a rule should be established in terms of the proximity of the secondary sources to the relevant primary sources. If a secondary source has not actually consulted the archive or relies upon inadequate primary sources or "primary" sources that are actually 2nd or 3rd hand to the object of study then it is necessarily less reliable than a secondary source that has consulted the archive. On the other issue, i think that one could introduce a distinction between interpretations and simple facts. The selection of an interpretation(s) of historical phenomena in an article should rely upon a much more rigorous, scholarly and narrow selection of secondary sources than that which may be used for non-contentious facts. But the interpretation(s) adopted should guide and strongly inform the editor in the choice of which facts are relevant to an article. FiachraByrne (talk) 02:41, 13 November 2012 (UTC)

It is important to note the distinction between less reliable and unreliable. Yes, in general, a pop history is considered less reliable than an academic one (there are exceptions... at both ends) but, that does not mean we consider the pop history to be unreliable. If an academic history disagrees with a pop history, we should obviously give more weight to the academic source, but this does not necessarily mean we give the pop history no weight at all (see WP:UNDUE for more on this). That said... if the two sources agree, we would prefer you to use the more reliable of the two. It isn't required... but it is preferred. Blueboar (talk) 03:11, 13 November 2012 (UTC)

Historical articles

Here's the first real paragraph from this page:

Articles that deal with current events, or events occurring entirely in the previous one or two years are not regarded as historical articles, since they have not been studied by historians. When historians first begin to write about an event, then it should be regarded as a historical article. Sources that were previously satisfactory, such as reports in the mainstream press, should be replaced by sources from historical scholarship.

I don't believe that whoever wrote this actually understands what it means to have statements like this as our official advice. So let me translate that for you. The first sentence is basically fine:

Articles that deal with current events, or events occurring entirely in the previous one or two years are not regarded as historical articles, since they have not been studied by historians.

although a lot of people are going to wonder why "history" starts two years ago, putting things like Wedding of Prince William and Catherine Middleton and the 2011 Oscars in the realm of "history", and why any subject "studied by historians" is automatically "history", when it might look a lot like an article about science or music or medicine.

Then we go on:

When historians first begin to write about an event, then it should be regarded as a historical article.

Really? When historians first begin to write about an event? If you have an article filled with hundreds of sources (Barack Obama's got 340 at the moment), and you get just two historians writing their very first publications, then all of a sudden, the whole article should be considered "history"? Even the parts about what Obama did last week? Nobody who understands this sentence is going to buy that obviously overreaching claim.

Sources that were previously satisfactory, such as reports in the mainstream press, should be replaced by sources from historical scholarship.

Even if they're worse sources for the specific points being made? Even if the historian's POV is different from non-historian's POV, or the academic POV is different from the non-academic POV, and so replacing all the sources with historical scholarship would seriously violate NPOV? Even if the scholarly sources are inaccessible and impenetrable, and the material supported is really quite simple, like the date on which the event happened? Even if the couple of scholarly-historian sources that currently exist are inadequate to cover the subject? (Remember, we're supposed to do this as soon as the historians first begin to write about it, not when there is a well-developed body of scholarly literature.)

This kind of problem is typical of almost the entire page. There are good reasons to use historical scholarship, but this page is too sweeping and too easily misinterpreted in ways that will harm articles. You need to think this through more carefully, keeping in mind that anything written on an official guideline is going to be twisted and wikilawyered by POV pushers. Policy writing is hard; you have not succeeded here. WhatamIdoing (talk) 20:17, 9 May 2013 (UTC)

This is helpful. "Historical articles" is a problematic concept. WP:BLP applies to all articles, not just biographies. WP:MEDRS applies to medical claims whether made in articles about medical topics or as points in articles on other topics. I still think that we should point out that sources should be replaced, or at least supplemented, by historical sources, as they start to be published. I agree that POV pushers will attempt to wikilawyer guidelines, but better have some guidelines in place and then continually improve them than have only the general WP:IRS as first line of defence. Itsmejudith (talk) 22:47, 9 May 2013 (UTC)

Yes, I think you need to follow the BLP model: it's a historical claim if it's a historical claim, no matter what the rest of the page is about. Now you just have the impossible task of defining "history". WhatamIdoing (talk) 00:45, 10 May 2013 (UTC)

I don't get your point about "historian's POV", though. I accept that a journal article by a historian may be biased. And if it is an early paper on quite a recent event then there may not yet be other scholarship to balance it with. I do not see that there can be an "academic POV" against a "non-academic POV". There can be a view based on research and evidence versus a popular misconception. Perhaps you would like to give an example. Itsmejudith (talk) 22:53, 9 May 2013 (UTC)

Consider the scandal around The Man Who Would Be Queen as an example. The academic POV is that male-to-female transsexuals can usefully be divided into two groups on the basis of their sexual orientation. The non-academic POV (i.e., the one held by most transsexuals) is that there is only one type of MTF. The historian's POV is that the author and his children were abused shamefully by trans activists; the non-historian's POV is that the author deserved it for humiliating the trans community, and his kids deserved it for not denouncing him. Would you have an article that only presents one side of the story? (AFAICT, only one historian has ever written anything about this.) WhatamIdoing (talk) 00:43, 10 May 2013 (UTC)

Thank you for supplying an example. This underlines that we do indeed have to reflect on what is and what is not a historical article. The article you've directed us to is on a book, written by an academic for a popular audience. We already know how such articles should be structured. They should present the favourable, unfavourable and mixed reviews, which that article does. Although the book was published in 2003, few or no historians have written directly about it. Nor would we expect to turn to historians for a perspective on how trans sexuality is perceived in the 21st century because none of the debates are settled. By contrast there is a literature on trans in the 19th century - which might still be challenging for these guidelines as not all the writers in that scholarly debate would be historians. I don't think that you or anyone can distinguish between an academic and non-academic POV on trans sexuality. There is nothing remotely like a consensus In public opinion. Nor is there an academic consensus, within any one discipline or across disciplines. Psychologists are split. Sociologists are split, but differently. Historians have no view. I will think about wording that better expresses the intended scope of the guidelines. It would be useful if you had time to comment on the kinds of sources one would expect in a more clear-cut "history" article, like 1948 Arab-Israeli war. Itsmejudith (talk) 06:13, 10 May 2013 (UTC)

This claims to be a history of the scandal, and it is written by someone who is frequently described as being a historian (and specifically a historian of sex, gender, and society). Scandal==event, and it happened more than one or two years ago, so this page is claiming that who said what and whose feelings were hurt when is "history" and that only "historians" get to say anything about the events involved.

I think that the first thing that you want to do is to try to re-write ==Historical articles== to either define the limits as being material that is in an article that is clearly and primarily about history (so Battle of Britain, but not books or diseases with a bit of history tossed in at the end), or to change it to ==Historical claims== and try to explain what specific kinds of material should or shouldn't be covered (so Tuberculosis#History, which is another case that is worth reviewing. What sources do you think would be ideal? What sources would be acceptable, although not your first choice?) WhatamIdoing (talk) 22:23, 14 May 2013 (UTC)

It would be easiest to take a more traditional interpretation. That is, not till 30 years after the event is it suitable to treat a given topic as properly historical. 30 years is generally taken as the minimum as that's when most governmental material is made available in the archives. The other rider would be that the topic should have received substantial coverage by historians. I think it's important, by the way, to reactivate this essay. FiachraByrne (talk) 12:27, 13 December 2013 (UTC)

Moving this forward to be a content guideline + comments

Has anyone attempted to move this forwards to be a content guideline, as WP:MEDRS is? I also have two general comments:

1. I think "historian" should be wider than just PhD or advanced academic training in historiography. For instance, several prominent researchers of the Holocaust are sociologists - e.g. Barbara Engelking, Jan T. Gross, Nechama Tec - but most of their work is in history and they are commonly described as historians. The same is true of other fields - some anthropologists, economists writing about economic history, or strategic studies/war college alumni writing about military history. Instead of academic training in historiography, I would say "academic training in the humanities, with a professional focus on historical work".
2. I think that we should differentiate between contentious and non-contentious history. E.g. construction details of the German battleship Bismarck are probably not too contentious. But the 1948 Arab–Israeli War, Falklands War, or Collaboration in German-occupied Poland very much are. In the latter case, I think that works listed under "Historical scholarship may include:" should for the most part be excluded unless, perhaps, if they are widely cited. In particular in regards to memory politics or historical politics - some historians writing in popular media are willing to make dogmatic or even extreme claims. I would also reject conference papers - at least by less established historians (that currently are included in "The results of the scholarly process appear in numerous forms").

@Fifelfoo: I think making this essay policy will have a positive impact on contentious areas, thoughts?Icewhiz (talk) 11:57, 2 July 2018 (UTC)

• I haven't had my coffee so no comment on the above, but I've always thought this was an excellent essay. EEng 12:03, 2 July 2018 (UTC)

• Last time I dealt with who a historian was by writing essay, I think I incorporated "respected in discipline" elements into it. Another knowledge producer, who by publishing, has been accepted by historians with the PhD, DHist, Habil., etc. is a historian. Much as a knowledge producer who publishes regularly in peer reviewed journals in their field which are historical journals is. PhDs are a short hand to me. Fifelfoo (talk) 13:06, 3 July 2018 (UTC)

I believe this should move forward to policy, at least in part. I do not believe consensus is currently present. Fifelfoo (talk) 06:18, 3 July 2018 (UTC)

Lots of people with or without training in methods ("historiography") write history. However this discussion is how to reach a reliable source threshold status. In my opinion to reach that a publication has to be part of the historical discipline-- the "discipline" is the system by which historians discipline each other, by evaluating conference papers, theses and dissertations, unpublished articles and books, and publications. The vetting process is most severe when it comes to papers offered to scholarly journals, which are evaluated by multiple scholars who ask whether this is a mainstream contribution, or a fringe viewpoint or badly sourced, that ought to be rejected. Conference papers are vetted by the conference organizers, assigned scholars who criticize it at the conference, and also audience participation. If the paper appears in the published "proceedings" it passes the discipline test. Papers usually start out in less formal workshops, then get revised over and over by the author, before publication in a journal. If the paper has never been published it fails the Wikipedia RS test. I've been to a few hundred history conferences in the last 50 years, and can testify that the vetting process is generally quite serious and severe-- a wrong date, or even a misspelled name gets an author in trouble. Colleges and universities will not accept a thesis until it it is been approved by a faculty advisor, and also by a faculty committee. The school's reputation can be seriously hurt by accepting fringe c inadequate theses. Rjensen (talk) 14:09, 3 July 2018 (UTC)

The essay would need to cover archaeology, and probably sub-disciplines like art and economic history - the use of the former (which I know more about) by conventional historians can be very good but sometimes causes grinding of teeth among specialists. Johnbod (talk) 14:53, 3 July 2018 (UTC)

Interestingly - Holocaust studies (which encompasses quite a few sociologists and anthropologists) is a completely distinct topic from Jewish History (where scholars usually do not touch the Holocaust) - see Bridging the Divide: Holocaust versus Jewish History Research — Problems and Challenges - which is also quite distinct from Military history of WWII (and military history in general is a distinct sub-field). We should probably cover some specialized sub-fields. As for conferences, journals, and institutions - not all have the same reputation - Yale is one thing, Ewa Kurek formerly(?) of the Higher School of Skills - Kielce[3] is another (and would not be a RS). Icewhiz (talk) 15:30, 3 July 2018 (UTC)

it's not the educational institution that matters for RS, it's the established organized discipline based on an international network of scholars that comprise the discipline. The social science & humanities disciplines all operate much like the history discipline. We have famous cases of tenured profs at prestigious universities issuing fringe ideas OUTSIDE their own discipline. Rjensen (talk) 19:06, 3 July 2018 (UTC)

At this point in time, I do not believe that this is ready for promotion. To name only the problems within the first sentence of the nutshell ("Historical articles on Wikipedia should use scholarly works where possible")

1. It is still talking about whole articles. We have approximately 300,000 articles that contain ==History== sections. Are they meant to be exempt?
2. It takes an unrestricted, maximalist POV about scholarly sources, which means citing fewer sources that readers will have access to, even for non-controversial and uncontested facts. Don't get caught citing something like https://www.loc.gov/item/today-in-history/july-04/ for the date of the US Independence Day; that's "history", and a scholarly source is "possible", so this page says that you need to use the scholarly one.

I think that it might be helpful to spend a while talking about why we're here. What problems do we need to solve? I'm going to take "feeling envious that medical content has to get squeezed through the MEDRS straightjacket, and wishing to make it equally difficult to write about historical content" off the table; that's not real. So what are we trying to fix? Can anyone post links to a couple of disputes that would be easier to solve if this advice (which, if taken in moderation and applied to controversial claims, is IMO pretty good) were enforceable? WhatamIdoing (talk) 22:30, 3 July 2018 (UTC)

Thank's WhatamIdoing. I would like to see the encyclopaedia match the quality of sources used to the subject under consideration to the quality of the article in the area of history. Most particularly I'd like to see this applied to notability and weight of analytical / scholarly concepts in articles; and, for WEIGHTing for article organisation and narrative attention. I am also concerned about editor blindness to the subject of the article's fullness due to attending to lower quality literatures (class, race, gender). I'd like a document that provides guidance to help editors research and write to the standard appropriate for the article. Fifelfoo (talk) 04:32, 4 July 2018 (UTC)

So, to massively oversimplify, you're looking for:

1. When excellent sources exist, use them (or at least don't contradict them); when they don't, use the best you've got.
2. For a historical event, the "plain vanilla" historical POV should dominate over the sub-field POVs. So, for an article about the suffragette movement, the historical POV is most important, the gender studies POV is maybe almost as important, and other POVs (e.g., class, race, LGBTQ, economic, etc.) are of less importance, and this should be reflected in the article's structure (e.g., how much time it spends on each POV).

Does that sound sort of fair to you, albeit missing some nuances? Or am I completely misunderstanding you? WhatamIdoing (talk) 23:04, 5 July 2018 (UTC)

That's pretty much there, but with the second point it isn't about a "vanilla" position, as much as the position which appears in the scholarship as the chief consensus position. If the consensus amongst historians of suffragettes is that they use gender methods and emphasise gender as a central organising point. Other, lesser, consensuses should then be given lesser weight. Fifelfoo (talk) 04:08, 6 July 2018 (UTC)

• I wouldn't support promoting this to guideline status. The text on primary sources in "Reliable sources for purely illustrative purposes acting as a 'picture'" is confusing: "The primary source is not used to prove the fact, but to illustrate the proof of the fact ..." Sources don't prove facts or illustrate proofs. Primary sources are excellent sources to use in history articles so long as you're familiar with the secondary literature and don't go steaming in with something inappropriate. I agree that, for contentious issues, the weight given to a primary source should be guided by the secondary sources. I'm not sure what "[t]he fact is derived from a scholarly source" means. Also not sure what this means: "The primary source used is the one used from a scholarly source, or a very close analogue." Or "[t]he primary source is attributed".

I hope soon to add to Mary Winsor information about a study she made in 1914 of the way the British press treated the suffragettes. I'll be using the article she wrote as the source. There are secondary sources (although they don't go into much detail), but they aren't necessary. If this were a contentious point, it would be different. If, for example, someone makes a "study" concluding that AIDS isn't caused by the HIV virus, we wouldn't use that study as a source. In the bio, we might mention its existence, but we wouldn't let the bio become a platform for the views; we would be guided by secondary sources. But for material that isn't disputed, offensive, outdated, or wrong, a primary source can be not only fine but the best kind of source. SarahSV ^(talk) 05:55, 7 July 2018 (UTC)

this is exactly the kind of feedback needed for this essay. For this essay to be rewritten; to be able to propose to interested editors; that it be suggested for guideline consensus. That's exactly how far away from proposing a guideline I feel. And if people have more critiques I and hopefully other editors can develop them into a document worthy to propose. Fifelfoo (talk) 07:42, 7 July 2018 (UTC)

How about a "test drive"?

Knowing a little about science and technology, I would suggest that we check how the theory works in practice, before the essay be promoted to something more official. If I got it right, the author of the essay reckons that modern and post-modern historiography is better. So, let's take the article French Revolution and see how would be if we use mainly post-modern sources.

PS. I discovered this essay after a user posted a link in the discussion of article Souliotes. There, some users are trying to exclude sources of early 19th century, i.e. people who met personally the Souliotic community before the latter disappears, and prefere 21sth c. authors of the post-modern crop "history is what I write about the past".--Skylax30 (talk) 08:45, 5 July 2018 (UTC)

Well... Back in the 19th century, history was generally understood (in European and American academia) to be a field of non-fiction literature, rather than a scholarly endeavor that was attempting to ask and answer complex questions. It wasn't until the early 19th century that academia even thought about research as a serious thing; before then, the goal of a university was to promote truth, beauty, and goodness, and it took more than a century (and possibly longer within the humanities) for that shift to really happen. I would therefore be careful of how I used such old sources. WhatamIdoing (talk) 22:58, 5 July 2018 (UTC)

Re: “...before then, the goal of a university was to promote truth, goodness and beauty...”. Actually, no, before the early 19th century, the goal of Universities was primarily to train men for the priesthood. Ok... off topic, but I had to quibble. Blueboar (talk) 12:24, 6 July 2018 (UTC)

Here's a source for you. (I'm not sure that those two concepts are incompatible, though.) WhatamIdoing (talk) 19:46, 7 July 2018 (UTC)

Is the 21st c. university not promoting "truth, beauty, and goodness", as these are understood in the 21st c.? If WP is not attempting to be a kind of Ministry of Truth, she has to provide access to all truths. Older sources can be presented in separate sections, such as "19th century historiography". The claim "surpassed" has to be based on sources, not on axioms.--Skylax30 (talk) 06:26, 6 July 2018 (UTC)

No, we don't provide access to all truths, we provide access to notable and weighty consensuses in appropriate literatures. If the 19th century scholarship is weighty, and commented on, include it based on the secondary sources that comment on it. If it is post-Rankean and still respected in the appropriate scholarly community, then it is a high quality reliable source. If it is not respected, then we've got secondary sources that consider it weightily and notably as not-current, and we should consider including it in a historiography section—but include it primarily in the manner in which it is commented upon. It may even be useful as a supplementary PRIMARY: for instance if contemporaries cite it for fact while considering its analysis to be faulty, we could follow those contemporaries and cite it for more detail where contemporaries cite it for fact. Fifelfoo (talk) 06:31, 6 July 2018 (UTC)

thanks Fifelfoo--very well said. Historians form a community that spends a lot of effort trying to reach a consensus, and I would say 95% of the time the consensus is clear enough for Wikipedia. Rjensen (talk) 17:19, 6 July 2018 (UTC)

I am not sure that the historiographical consensus regarding the French Revolution is for post-modern historiographical methods. Just because a historiographical methodology is recent doesn't make it the accepted scholarly consensus. Fifelfoo (talk) 09:14, 6 July 2018 (UTC)

By George, you r right. Imagine all those neo- and paleo-marxists, liberals etc left alone without myths.--Skylax30 (talk) 09:43, 6 July 2018 (UTC)

@Skylax30: if you want to make political points, please find a more appropriate website. This isn't a forum. Doug Weller talk 12:07, 6 July 2018 (UTC)

A review in a historical journal must be favourable

It's probably stating the obvious, but just having a book or article reviewed in an academic journal is not enough: the review should actually be favorable.

I have recently read the review of The Picts, By Benjamin Hudson. Pp. xii, 266.ISBN: 9781118602027 in The Scottish Historical Review, Volume XCVII, 1: No. 244: April 2018, 119–127

This is a book by someone with a post at an American university writing in what appears to be his field - but the review is absolutely scathing and suggests that the book presents a risk to readers - it says: "This raises serious questions regarding the editorial procedures followed and concern must be expressed regarding the impact of the work on potential readers." The review lists a good number of errors, with page numbers - so one has to presume that the editorial committee of the Scottish Historical Review checked that the reviewer was justified in such a negative opinion. I think that such a review would disqualify the book from being a reliable source. Therefore the instructions in this essay should look for a favorable review in an academic journal as opposed to simply having been reviewed.
ThoughtIdRetired (talk) 22:49, 7 April 2018 (UTC)

Yes, one very negative review is bad but it perhaps can be outweighed by favorable reviews. I would phrase it -- if a review seriously attacks the book scholarship, that is an indicator it is not a RS. -- the advice is to check other reviews. In my experience the review editor will usually publish a highly negative review without sending the review out to edit board. Rjensen (talk) 01:08, 8 April 2018 (UTC).

To this I'd propose that a history written by a non-academic but generally reviewed favorably by academics is equivalently (and sometimes more) trustworthy. We all know people who love to talk ad-infinitum about an unbendable viewpoint, and even who obtain academic credits to make their windiness seem more erudite. I better trust a non-scholar who probes deeply and writes clearly. GeeBee60 (talk) 00:25, 16 July 2018 (UTC)

Acceptable use of primary sources

I think that the guidance presented here may be overly strict with regards to the use of primary sources. Although I think that secondary sources are usually preferable, high quality primary sources aren't necessarily worse than popular history accounts for non-disputed facts. The main issue is that some editors misuse them to push a certain POV by cherry-picking sources that emphasize certain aspects, perspectives, or controversies. This is where secondary sources are necessary to contextualize: is the primary source being cited as an example of a wider phenomenon or as an exception to a phenomenon? How strong is the secondary source involved? Some primary sources are repeatedly cited in secondary sources for facts without comment; this is usually a good indication that they are reliable.

Another thing that sometimes comes up is the use of sources where the author has a documented extremist political view. This was an issue, for instance, on the biography of Albert Kesselring (an editor argued that Kesselring's memoirs should not be cited; he confessed to ordering the extrajudicial execution of hundreds of people). It would be helpful to have better guidance on this issue, as it's not limited to primary sources. Catrìona (talk) 02:40, 25 November 2018 (UTC)

I think you're right about primary sources being used, without disputes, more than this page recommends. Autobiographies are generally accepted as reliable for whatever the author says. Old census reports are another primary source that nobody regularly disputes. "The population of Smallville was 2923 in the 1840 census" can be cited to the 1840 census. WhatamIdoing (talk) 05:26, 24 December 2018 (UTC)

No and no. An autobiography isn't independent of the subject and can be used only as a source on its own content -- and even then with care: "Smith wrote in his memoirs that his mother was cold and distant." And census data should be taken from a modern source, which might comment on accuracy issues in that locale, changes in political boundaries, and so on, not from contemporary primary sources. EEng 05:44, 24 December 2018 (UTC)

@EEng: While I am in agreement with you that primary sources should be used with care, that doesn't seem to be the current practice. See this featured article where a large portion of the citations are to primary sources, including more than a dozen to an explicitly fictionalized memoir. Catrìona (talk) 05:54, 24 December 2018 (UTC)

Well, I've never found that FA status is a guarantee of much beyond a peculiarly stiff style of writing, but AFAICS the Vrba memoir is used exactly as I describe above it should be -- as quoted, attributed material together with a discussion by other sources of its reliability. Not sure which other primary sources you're talking about specifically, but primary sources can generally be used to illustrate facts given in secondary sources, though only with extreme caution as fact sources themselves. EEng 06:08, 24 December 2018 (UTC)

Feel obliged to chip in regarding census reports. This is a good example of how lack of specialist knowledge in a subject can produce dangers. The situation that I have in mind affects Highland Potato Famine and related articles. The definitive academic study of the famine (Devine 1995) explains in some detail why census returns are misleading for anyone trying to assess how many people were affected (in short, the region always had a number of migrant workers who were away from home; these numbers increased and they stayed away for longer during the famine; the relevant censuses were taken at different times of year, so capturing different numbers of seasonal absences; etc., etc,). Without supporting guidance from a good quality secondary source, primary sources can seriously mislead.

I think primary sources have value in providing some expansion of the subject if there is little reason to believe that the writer had any reason to give a distorted account and they fit within the overall framework of information in secondary sources - for instance Westland Lysander gives the opinion from the memoirs of an AOP pilot (Munro) as to why the Lysander was unsuited to this role - general history of the aircraft shows that it was not usefully employed in this way (and Munro is used as a source by the writer of the official history of AOP squadrons).

Your first example illustrates perfectly why census data needs to come through a secondary source. I'm not quite sure what you're saying in your second example. EEng 16:50, 24 December 2018 (UTC)

Hence, careful, knowledgeable judgement is needed, producing edits that still fit within the framework of secondary sources.
ThoughtIdRetired (talk) 09:19, 24 December 2018 (UTC)

If one were comparing the 1839 and 1842 census numbers and drawing (or implying) a conclusion - it would be one thing (point taken). However, if one were to say that in 1700 the village has 50 people (or households, or what have you) recorded in the census and would contrast that with 10,000 in 1900 - that's a different matter - as the census figure does at least give us a handle of what size of locality it was. Icewhiz (talk) 11:39, 24 December 2018 (UTC)

Still no. If there's some significance to this there will be a secondary source reporting it. Otherwise you have no way of knowing that these numbers aren't subject to the kind of problems ThoughIdRetired described above, or any number of other problems. EEng 16:50, 24 December 2018 (UTC)

If we didn't accept the use of census data, cited to the census itself, then why did we let a bot add census data to tens of thousands of articles? Why do we have long tables of population data in some articles, cited to the relevant census database (e.g., Saint-Faust#Population)? Citing a census for its own contents seems to be very common and widely accepted. WhatamIdoing (talk) 17:01, 28 December 2018 (UTC)

But apparently we're not citing the census data "for its own contents"' it's being cited as "The Population"; citing for its own contents would be if we said, "The 18xx census lists X residents". Those automatically inserted tables should carry a severe disclaimer. EEng 00:43, 29 December 2018 (UTC)

A short reference simplifies

User:Icewhiz quotes in Stanisław Piasecki a short opinion from "The politics of literature: Poland 1945-1989" about a pre-war journal (the subject of the book is post-war literature) "the right-wing, pro-Nazi, anti-Semitic weekly paper Prosto z Mostu". It's a primitive simplification. I believe that a history of pre-war literature should be quoted to inform about a pre-war journal. Dr Małgorzata Domagalska describes the journal "Fear Or Fascination. Nazi Germany in „Prosto z mostu” (1935-1939) weekly" [4] So Icewhiz removes the fear.

The journal's attitude toward Nazism should be described in Prosto z mostu, but it isn't.

Stanisław Piasecki's attitude toward Nazism should be described in his biography. Now the radical opinion is presented only in the lead. The lead doesn't inform that Piasecki opposed the Nazis, that he was tortured and shot by the Nazis, which is quite important.

Xx236 (talk) 12:00, 21 January 2019 (UTC)

Piasecki was the editor in chief of the anti-semitic weekly Prosto z Mostu. Per another source - Holocaust Literature: A Handbook of Critical, Historical, and Literary Writings, Saul S. Friedman Greenwood Press he "Stanislaw Piasecki, who had been the editor of an anti-Semitic journal, Prosto z mostu (Straight from the shoulder), and an outspoken admirer of Hitler, was so shocked by the events following the German" - so hardly an exceptional claim (per Friedman - Piasecki himself (not that divorcing Piasecki from Prosto z mostu is easy - Piasecki's main activity was being editor-in-chief of this weekly) is presented as a Hitler admirer). That Piasecki was disenchanted of Hitler following the invasion of Poland in 1939 (in 1940 he wrote a column complained that the Nazis turned Poland into a "Jewish Paradise" - the occupants of Jewish ghetto being overprivileged and the Yellow badge being a "badge that protects"), does not change his pre-war activities (for which he is notorious). Icewhiz (talk) 12:08, 21 January 2019 (UTC)

There exists a difference between a biography and a page about a journal.

Why do you use a book about post-war literature as a source about pre-war journal? Do you appply the same methodology when you write about any other subject?

Does Friedman read Polish to understand Piasecki's texts?

You ignore my comments and my sources. You fight a war against me, against victims of Nazis crimes.

You dehumanize a victim of Nazism. It's pure Nazism.

Xx236 (talk) 12:47, 21 January 2019 (UTC)

Two sources - Carl Tighe published by University of Wales Press, and Saul S. Friedman published by Greenwood Press (nowadays ABC CLIO). Friedman's book is on Holocaust literature (which includes Piasecki's war-time and pre-war publishing). I don't know if Friedman knew Polish (wouldn't be surprising if he did) -he was an expert in the field per this 2013 obit in the Holocaust and Genocide Studies journal. Icewhiz (talk) 13:07, 21 January 2019 (UTC)

Piasecki isn't really the hero of my favourite story, but to be fair other sources indicate he grew disillusioned with Nazism by 1938 and declared it morally flawed.--MyMoloboaccount (talk) 20:24, 21 January 2019 (UTC)

As I have written - A short reference simplifies. No seriuos encyclopedia should be based om such references in books about war and post-war literature. Piasecki published "Prosto z mostu" before the war. Now everyone is wise.

A biography is a biography and a journal is a journal. If someone doesn't undetand the difference, he shouldn't edit such pages.

Xx236 (talk) 08:13, 22 January 2019 (UTC)

What are primary sources?

This page uses the term "primary sources" as if it is a well-understood term. But I can't find a definition of it anywhere.

• Is WP:PRIMARY an adequate definition of the concept for history?
• WP:USEPRIMARY describes what is a primary source for fields "other than history". So, where can we find a definition for history?

For the context of these queries, please see

• WP:Reliable sources/Noticeboard/Archive 266#Muhammad Bin Qasim page
• Talk:Muhammad bin Qasim#Tahdhib al-tahdhib, history of Tabari and Ibn Sa'd are secondary sources

-- Kautilya3 (talk) 11:18, 1 June 2019 (UTC)

I added some text to make clear the role for historians in WP:USEPRIMARY. At the introductory level it should work. Rjensen (talk) 11:59, 1 June 2019 (UTC)