13

I am writing an article for an audience that mostly consists of computer scientists. A specific part of the subject is motivated by Kirchhoff's laws. I intend to mention this fact and reference the rules. But that is easier said then done. Apparently, Kirchhoff predates referencing.

How to reference a source that is so old, it has no bibtex entry, but is also probably not well-known throughout the readership?

edit: I want to clarify, that I can also imagine to not reference it at all, but I find it difficult to draw a line here. Can I assume that every reader is familiar with a term? For instance, can I omit a citation to Damas/Hindley/Milner when introducing ML to modeling engineers? How about Newton's method for computer scientists?

Ooker
  • 8,266
  • 6
  • 54
  • 114
choeger
  • 307
  • 2
  • 6
  • 34
    What do you mean by predates referencing and has no bibtex entry (I can easily make one if I know the bibliographic data). Anyway, the German Wikipedia tells me this is the publication you are looking for and a quick glance seems to confirm it. – Wrzlprmft Apr 11 '17 at 19:18
  • 8
    Why would you cite this? Providing citations for standard undergraduate-level material will just make you look like an amateur, ill informed about academic norms. – Buzz Apr 11 '17 at 19:24
  • That is a valid concern. But I occasionally find it hard to define where the limit is. Especially when writing about subjects that are related to multiple fields. – choeger Apr 11 '17 at 19:27
  • 10
    @Wrzlprmft Even hand-crafting a BibTeX entry is unnecessary in this case, since you can download one from the journal's website... I too am a little confused as to in what sense this paper "predates referencing". I would just cite it like any other paper, whether or not I could find a canned BibTeX entry for it. – Pont Apr 11 '17 at 20:13
  • 4
    Re: Edit: Kirchhoff's laws are undergraduate (possibly high-school) material for anyone who's taken Physics, and Newton's method is covered in most Calc I courses. In both cases, you can expect an audience of college-educated computer scientists to know of (if not be familiar with) these laws. However, Damas/Hindley/Milner isn't typical undergraduate material for most modeling engineers, so that might need some explanation/citation. – apnorton Apr 11 '17 at 23:23
  • 10
    Not having a bibtex entry shouldn't prevent an article from being cited. Imagine if mathematicians were not allowed to mention Aristotle or Pythagoras just because bibtext didn't exist back then. – vsz Apr 12 '17 at 04:08
  • 1
    As your audience can't be assumed to know German, I suggest that while you may or may not choose to cite it, if you do, you also cite a more accessible text ("see any undergraduate electronics textbook, for example introduction to Electronics, by Smith and Jones") – Chris H Apr 12 '17 at 11:01

3 Answers3

33

For cases like this, unless you want to give a historical reference, there's usually no need to cite the primary source, you can just cite your favourite circuit theory book.

Anyway, if you really wish, you can certainly cite the original paper too, which can be found, e.g., here.

For what concerns the added question on where to draw the line, you can have a look at the following question, and the answers therein: How generous should I be with citations?

Community
  • 1
Massimo Ortolano
  • 55,428
  • 19
  • 166
  • 207
  • I might be in a minority, but if you are going to cite anything I think it is best to cite both the original source and "your favourite circuit theory book". For reasons see my answer below. But with Kirchhoff's laws it is probably not necessary to cite anything. –  Apr 13 '17 at 08:47
22

I am writing an article for an audience that mostly consists of computer scientists. A specific part of the subject is motivated by Kirchhoff's laws. I intend to mention this fact and reference the rules. But that is easier said then done. Apparently, Kirchhoff predates referencing.

Your statements make no sense. It's like saying that you want to use Pythagoras's theorem "but that is easier said than done. Apparently, Pythagoras predates the English language." It is trivial to cite any paper you wish: just write the necessarily BibTeX.

The fact that no references appear in Kirchhoff's paper is completely irrelevant to whether or not you should use references. You are writing today, not in the 1840s, and today's standards apply to you.

How to reference a source that is so old, it has no bibtex entry, but is also probably not well-known throughout the readership?

You write you own BibTeX entry! BibTeX is just a language, like LaTeX.

However, it isn't necessary to give citations for such basic material as Kirchhoff's laws. They're part of standard high school education, so you can assume that everybody knows them. And, even if somebody doesn't know them, the phrase "Kirchhoff's laws" is specific enough that they can Google it and get the answer.

David Richerby
  • 33,823
  • 6
  • 74
  • 143
  • "And, even if somebody doesn't know them, the phrase "Kirchhoff's laws" is specific enough that they can Google it and get the answer." That's what I've been wondering about throughout this entire thread. Also, I don't see how a precise reference to Kirchhoff's original paper would be of use to the intended audience. If the intent is to indicate how old the result is, one could simply say something like "Kirchhoff's laws (formulated in 1845)". Of course, now one has to spend some time making sure they weren't formulated (or communicated to others) several years prior to publication . . . – Dave L Renfro Apr 12 '17 at 20:12
  • 2
    @DaveLRenfro I'd certainly expect more of the audience to know Kirchhoff's laws than know enough German to read Kirchhoff's paper. – David Richerby Apr 12 '17 at 20:15
  • Yep, I completely overlooked that part, and I'm someone who struggles pretty much every few days with 1800s mathematical literature in French, German, and Italian (the 3 main languages that show up)! – Dave L Renfro Apr 12 '17 at 20:17
5

For well-known basic physics, it is not necessary to cite anything. But if you do include a citation, you should cite the original source (regardless of its age), and to be helpful to the reader, also cite one or two recent textbooks.

If you only cite a secondary source (“your favourite circuit theory book”), then you are misleading the readers, making it hard for them to verify what you are saying, and quite possibly propagating information that is incomplete, misleading, or inaccurate. Ole Bjørn Rekdal has written several excellent articles about these issues. Here are parts of two of the abstracts:

  • From Monuments to Academic Carelessness (2014):

    [Katherine Frost Bruner's] collection of advice to writing scholars has been widely quoted ... The most frequently quoted message in Bruner’s article deals with the importance of making sure that references in academic texts are complete and accurate. Exploring the citation history of this particular message reveals an ironic point: the great majority of those who have quoted Bruner’s words on reference accuracy have not done so accurately.

  • From Academic urban legends (2014):

    Many of the messages presented in respectable scientific publications are, in fact, based on various forms of rumors, [because] authors have lazily, sloppily, or fraudulently employed sources, and peer reviewers and editors have not discovered these weaknesses in the manuscripts during evaluation.

  • 5
    I fail to see the relevance of the quotation from Monuments to Academic Carelessness. In particular, if your point is "the importance of making sure that references in academic texts are complete and accurate" then you're contradicting your own advice by using a secondary source without citing Bruner's original. – David Richerby Apr 12 '17 at 11:00
  • Yes, the quotation from "Monuments to Academic Carelessness" is relevant because it is about the importance of references being complete and accurate. I don't think I am contradicting my own advice: Bruner's advice was for writers in scholarly journals, whereas I think it is clear that I was recommending Rekdal's articles, not passing on Bruner's advice. I just included the quotations to whet people's appetites. –  Apr 12 '17 at 11:22
  • 5
    “If you only cite a secondary source […], then you are misleading the readers, making it hard for them to verify what you are saying” But the implicit claim a reader would want/need to verify is not “these are the laws Kirchoff wrote down”, but “these are Kirchoff’s laws, as they are understood and used in the field today”. So an established modern reference source is much more appropriate for backing up the claim than Kirchoff’s original work would be. – PLL Apr 12 '17 at 14:51
  • 2
    I agree that it would be good to cite a modern textbook, as the original is probably hard to understand and a modern book would give the laws in the form they are used today. But citations serve other purposes as well, e.g. giving credit and enabling the reader to check the original source rather than rely on a source that refers back to another source, which refers back to another source, etc. So I would cite the original and the textbook. (What I wrote applies more to things that are less well known than Kirchhoff's laws.) –  Apr 12 '17 at 15:46
  • 2
    This seems to depend massively on the subject, whence the disagreement. In mathematics and theoretical CS (at least), the original source is often rough, buggy and unreadable, whereas later expositions are clearer, more useful and more correct. In experimental sciences, I can easily imagine the opposite dynamics (as suggested in "Academic urban legends"): the original source has done the experiment, whereas the secondary sources just repeat what they have read (unless they do replication studies). – darij grinberg Apr 13 '17 at 04:54
  • It might depend partly on the field or whether it is theoretical or experimental, but I don't think that fully explains the disagreement. For the reasons above, I would cite the original even if it was rough, buggy, and unreadable. If there were outright errors or it was truly unreadable, I would make a parenthetical comment about that and definitely cite a readable and correct later source as well. And, as Rekdal's articles show, secondary sources don't just repeat what they have read. They sometimes get it badly wrong. –  Apr 13 '17 at 06:55
  • It strikes me as abundantly silly to insist on citing "the original" in a scientific field at all. If you cite a paper that's over a hundred years old because it's the original, and there's a perfectly good modern citation available, then you are only doing a disservice to the readership and community. This begins to fall under "proof by reference to inaccessible literature". When you are quoting and analyzing something in the humanities, things are a bit different as the exact way it was originally said carries intrinsic value. There's no such value in math outside of history. – zibadawa timmy Jul 27 '17 at 16:25
  • @zibadawa timmy It's usually not inaccessible, and there is value, for the reasons I gave in my answer and my second comment. I was arguing for citing both the original and a recent publication, not just the original. Are you saying that is a disservice to the readers? –  Jul 27 '17 at 17:38