Arguments to treat references in scope

  • By declaring requests for references out of scope, we discourage some legitimate and - interesting history questions.

  • History is unusual for H:SE; a large part of history is the examination of sources.

Arguments to treat references as out of scope

  • References are not answers to questions. An answer to a question should be evaluated and judged based on the answer. If I ask "What are good references for X" and someone replies that "Sujarkama's Economics" is a good source, then everyone should be able to evaluate whether that answer is good or bad. I should not have to read Sujarkama's Economics and discover that it relies on the assumption that one of God's holy laws is that all currency should be in the form of comestibles. (Knowing that, I'll downvote the answer).

  • Eliminating references constrains questions to those that are accessible to all. Take as an example the original question about IKEA. I would argue that the edited question is superior because the resulting answer should include both a discussion of the history of IKEA design and reference sources where we can learn more. Good Questions get Good Answers. (I have to admit that I edited the question, so I may have some bias in determining that the resulting question is superior.)

Is there precedent? Have we decided this before?

I believe that we've discussed this before, but I didn't find a question that really encapsulated the issue that I believe @lohoris and I were discussing. (At this point let me eat some crow - I may have been party to a discussion on another SE site about requests for references which resulted in a strong consensus that they are out of scope. I have been working from the belief that we had collectively discussed and decided this. So if I could have the crow stir fried with a bit of curry?)


@BartArondson asked for a reference to the history of IKEA. The question as originally phrased explicitly requested references, but in the context of a question about history. Rather than discuss the dilemma in the comments, I asked that the question be moved to this forum, and I've flagged the set of comments as obsolete because I don't believe that they usefully respond to that question - particularly in light of my supper of crow.

Further examples

Other questions that relate to this and seem relevant

  • Where can I find Nuremburg videos - OP is asking for the location of/access to a single video. This is a discrete request, not a list, and the answer would not be ephemeral (barring link rot).

  • Feedback from a Newbie - @Chinastudy points out that our guidelines could be clearer.

  • But surely you must know that oatmeal cookies are the only true currency! (Though, Sujarkama notwithstanding, they do perform poorly in vending machines.)
    – Joe
    Commented Sep 17, 2014 at 23:35
  • At this point I would like for someone to make a proposal; we've discussed this sufficiently. Perhaps @rosalindRei could make a concrete suggestion for an amendment to our help center?
    – MCW Mod
    Commented Sep 26, 2014 at 0:16

8 Answers 8


I am in favor of requests for reference material with certain restrictions. When I say reference material, I mean source material (primary sources). So, asking

What source material for the fall of Jerusalem is there?

Would be on topic. The correct answer would be Tacitus and Josephus. There is only one right answer.

An example of an off-topic question would be

What is a good book about the fall or Rome.

There are many books on the subject and there is no right answer.

Reference questions can be quite beneficial. It is often very difficult to find the original source material for an event if one does not have much experience researching that particular time/area. If I wanted to find out what we know about ninth century England, that question would be much too broad for a single question here. However, asking what sources we have would be a very good strategy for me to start researching on my own. The correct answer would likely include the Anglo-Saxon Chronicle. When, researching the topic, I come up with more specific questions, those would be a good suit for History.SE.

  • 5
    I like that sort of nuanced approach. Also, we should strongly encourage compiling multiple references into a single answer, as community. In this case, I think the benefit of having good information on the site far and away eclipses any possible risk of "list of too many small answers".
    – DVK
    Commented Apr 20, 2014 at 23:29
  • Here is a fairly typical example of a "what is a good book" question producing good answers on Mathoverflow: mathoverflow.net/questions/44125/… Commented Sep 19, 2014 at 10:57

I'll try to contribute without adressing each interesting point in this discussion. Being new to the site I'm still trying to figure out some things. Sometimes questions strike me as too broad. Others as too narrow.

I don't have a problem with questions about primary sources. This is a fair exchange between people that aren't used to work with primary sources and people that are. But I would prefer that these kind of questions be narrow and very specific.

I do have a problem with questions of the kind of "what's the best book about x". But narrowly focused and maybe tagged as reading advice it could be workable. Anyway there's a way of putting this kind of question that would break free of the reference question constraint: "I've been reading x what do you think?".

As someone said sources are the marrow of history. Excluding questions about this it's too radical.

What I would like is a show of previous effort. I will not answer a question that can be checked on wikipedia.


Comments based on a question I asked How can I find the most complete list of the names of Kaytn Massacre victims?

One of my comments is a direct copy from there.

  1. SE gets a lot of attention and can be very helpful for tracking down sources, especially across international/language boundaries. I know of no better community to tap for these requests.
  2. Philosophically speaking, what we know about the past is subject to change based on new evidence. Not to sound too extreme, but every question about history is implicitly asking for an answer based on the best available information we have today. Many things are cut-and-dry, but non-source answers are also subject to change.
  3. Therefore, I think that specific source questions should be allowed. That is, questions that can have a definitive answer, if the information were to be available (regardless of whether or not it is).

Samuel Russell: 'These resources are unlikely to be available for consultation outside of "closed case" areas like Josephus'

Plainly wrong. For example he can conduct a simple search on the British national archive website to show him just how much is in the public domain, and moreover how recent such material can be.

I originally posted for references on the history of ethnopsychiatry which is an incredibly small field compared to the historiography of world war I or Stalinism which were the topics of examples given of posts which supposedly make specific, legitimate reference requests. Hence, anyone who deems themselves suitably qualified to answer my question (i.e. have read the relevant literature and which topics/sources have been covered or not) would necessarily converge on a small number of unattended sources. If one is familiar enough with the literature in a specialist area one would be able to identify the major debates within it and evidence already used to support arguments.

Broad requests are necessary also in preliminary research especially if like in my case research is being conducted independently of an institution - I've had to collate a reading list and identify major works independently so potentially could have blind spots. I could narrow my request to sources that would be available to me at a major library or online as I am not able to travel to seek archival material in different countries. But again, of course, anybody in a position to recommend I rifle through the patient records of a Transvaal psychiatric institution would also be able to answer my original question perfectly efficiently too.

  • 1
    Would you care to specify the closed domain for the research problem of English resistance to the change in English economics between 1776 and Peterloo? It is an equivalent time period to Josephus' core concerns. We know that it will be in the public domain: we cannot know without starting to produce the history what the appropriate source context is to select from the public domain sources. Josephus is closed by limited survival: for everything else critical and opinionated disciplinary selection needs to narrow context. Commented Sep 18, 2014 at 21:34
  • I'm sure @RosalindRei has something to say about this. Either way, if this is supposed to be a research board I don't see the problem with answers representing some degree of original research. Commented Sep 19, 2014 at 10:03
  • Votes are on this board to determine whether it's any good. Commented Sep 19, 2014 at 10:09
  • 2
    I do not see the reasoning behind shutting down an entire class of questions because there may be instances of it where it's not clear that anyone already knows the answer. Commented Sep 19, 2014 at 10:24
  • Further, if you happen upon a question where you could make a decent answer if it was more properly delineated, ask for the context of the asker, why they are interested, or suggest if there may be a more specific version of the question it would be helpful for them to have an answer to. Or type up your partial answer specifying how it is a partial answer. From there community wiki could collate these answers should the question produce several partial answers (if necessary). This is what cw is for, and it has worked wonders on Mathoverflow, I don't see why it wouldn't work here. Commented Sep 19, 2014 at 10:45

My opinion is that we should allow questions asking for references.

First, as a reference to Tyler Durden's answer, for which I do not understand downvotes, actually almost every question is asking about references. Did something happen? In history, we can be sure only if somebody has written (or stored in other way). But this can be lie, the source can be incomplete, or the asker was not smart enough to understand relationships between two events. So that's why people ask. We require references for such questions.

Second, there are questions for which sources exist. If someone asks "Who was Adolf Hitler?", we can close it. Such question does not require big querying and probably in most languages in the world there is at least a Wikipedia page for such a topic. This is an English speaking forum, so we can assume that anyone can find an answer in English.

Third, in some cases, it's not that obvious. No-one is able to read all the books that people have written since the beginning of the world. If somebody asks "Where can I find more info about Pontius Pilate?" this is not obvious question. We know that even many Christians consider him evil, but some think he was later a saint. Which sources show neutral point of view? Which are a sort of Christian/Jewish/Roman/etc. propaganda? I think this would be an interesting question.

Fourth, sometimes the answer would be too long and the asker knows it. Let's assume I am interested in "Age of Discoveries". I'd like to read a book about it, but I'm not sure which one is worth reading. I don't have particular question now - in fact I just saw "Pirates of the Caribbean" and liked how ships fought. But I saw some ghosts, magic, sea monsters etc. in which I don't believe. What would you suggest me? This is in my opinion a good question. I'm interested in all of the aspects, both everyday living (tools, clothes, were marriages contracted, etc.), particular topics (pirates, governors, conquistadors) but also in great events (conquest of Peru, Mexico, settlements in Virginia etc.). I know there are many sources but which one is the best?

Fifth, there can be questions about validity of the sources. For example "Marcus Brittanicus wrote that Nero was visiting Hispalis, while Romus Italicus said he was only thinking about it, and Propius Helveticus said three hundred years later that it was not Hispalis but Hortonium, and then in 18th century Gibbon said that Brittanicus had never existed, which one is right?"


I believe we shouldn't allow resource requests for primary sources.

  1. Most primary sources are archival in nature. These resources are unlikely to be available for consultation outside of "closed case" areas like Josephus.
  2. Non-archival primary source collections are books. These books display selectivity, translation, editorial control of meaning. They're works, in part, of art. As such listing one or many primary source books is a comparison of like-for-like and exactly to my mind like "What is a good book on Roman history."
  3. The answers are trivial, and tend to display the discipline of library science more than the discipline of history or historiography.

Alternate: we should allow questions such as "What are the seminal historical texts in XXX studies." as these are fixed by review articles and historiographical analyses of sub-disciplines. For example, "English Labour History" => Engels, Hammonds, Thompson.

  • 1
    I up voted this only for the "alternate:". I couldn't really say about 1 or 2, but I disagree with 3. Commented Sep 18, 2014 at 18:31

This question recently showed up in this question https://history.stackexchange.com/questions/15279/sources-for-colonial-ethnopsychiatry

Like I mentioned on there, I do think that this should be a perfectly legitimate question for a board on history. My argument was that from my understanding finding sources is a typical component of the day-to-day of a historian researcher, and that questions of this sort would likely for the most part produce perfectly useful and searchable answers to questions common in the field.

@Semaphore argued that questions like this don't have a definite answer (as mentioned as an argument above), which may indeed correct (though as @RosalindRei points out in the comments in the question above, her case should have a definite answer in the shape of a short list). It might still be worthwhile to remember that this requirement originates from the original stackexchange where sticking to the 1Q&1A format was an important point as there were a lot of different opinions floating around with about more or less anything (this programming language versus that, etc.). This may not be so on this board, especially when it comes to listing references.

While I am not a historian, I do feel strongly in favour of H:SE being a resource useful to actual researchers in the field. From my experience with Mathoverflow and from talking to historians, I believe our worlds to not be all that different in this respect. Thus I feel I must point out that I have benefited immensely from both reference requests and more open questions on that site.

In regards to this it may be interesting to note that Mathoverflow solves the problem of open questions (1Q possibly many A) using a big-list tag and making the question community wiki (meaning posts are more freely editable, and that contributors and the asker don't get reputation points). This is not unlikely to produce a number of useful resources here, as it has done there. Further, in opening up this venue H:SE will not immediately deter historians interested in using this site from asking what I believe to be perfectly reasonable questions for a historian to ask.

A couple examples of useful resources on Mathoverflow stemming from questions with more than one valid answer:





Since I myself have asked for references on Mathoverflow, where the policies are more in line with what would be useful for a researcher in the field (their policies initially developed independent of the original Stackexchange), and this produced (if several) good answers without having to resort to neither big-list nor community wiki, I'll share that question as well:


As for how to encourage good requests for references and discourage the lower quality questions, I suspect the nature of history as a subject means that this will not generate poor answers.

If I am to propose a change to current policy it may be to allow reference requests, but encourage the asker to be specific to the point where--and maybe present an argument for that--it is feasible a complete answer would consist of a fairly short (and probably slowly expanding) list of reference material. From there it's a question of common sense:

E.g., a request for introductory material on recent European history would be too general as a question (but could still generate an interesting resource via a big-list tag and community wiki, as per Mathoverflow), while a properly delineated question like the one by @RosalindRei in the top of this answer would be totally fine with no ifs or buts.

  • I think the problem with asking for source materials that have a specifically narrow focus is that it doesn't much benefit to anyone other than the OP. The problem with a general focus is obvious. The area in the middle is more appropriate - present the question you are looking for sourcing on and let the answers refer to sources. If you haven't formulated a thesis yet, a question about historiography is on topic, but will generate answers that are more broadly useful. The question in that case is how do I find source material - not what is the source material.
    – Comintern
    Commented Sep 24, 2014 at 4:24
  • It will of course benefit anyone who would be interested in sources within that particular focus. If a particular answer to a particular question is useful to someone other than the asker is for any particular user to decide, not a sweeping decision by the community moderators in general. People googling for that particular topic will also be able to see it. In this respect I'm not quite sure if you are being serious. Commented Sep 25, 2014 at 16:09
  • Care to suggest a change to the help center?
    – MCW Mod
    Commented Sep 26, 2014 at 0:17
  • Oh, I did not realize I had gotten a comment here. My suggestion would be to allow questions that ask for primary sources on a given topic and make them CW as necessary, e.g., if the question produces several partial answers. Commented Jan 23, 2015 at 12:21

All I can say, is that virtually every single question so far on the site either asks for references, has comments asking for references or has annoying people criticizing your answer by comments demanding references.

So, either you have to turn it into Wikipedia and start requiring references everywhere or you have to make it absolutely clear NO REFERENCES REQUIRED in bold letters at the top of every page. Otherwise, you are going to have the self-appointed reference police demanding reference every time somebody farts or makes a joke (do you have a reference for that joke?)

  • 14
    There's a difference between requiring references to back up an answer; and asking for reference material name as a question.
    – DVK
    Commented Apr 20, 2014 at 23:26

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .