I am using a somewhat obscure mathematical technique in one of my papers, and there is only one reference text (so far as I am aware) that collects all the important examples of the technique that I need. However, I am leery of pointing my readers to this book without a warning, because the reference book has numerous problems.

The book was clearly compiled from a number of earlier sources, including the original research literature. These were put together without much care. Notation for the various functions involved changes from one table to the next. (The same object may be referred to an upper case U, lower case u, or a cursive letter.) Even more seriously, there are many misprints, some of which are more obvious than others.

I feel like when I reference this text, I should put some caveat lector—something like, "Details about the transformation and all the important examples that will be needed in this paper may be found in [1]; however, the reader should be advised that [1] contains multiple typographical errors." But that sounds somehow unprofessional to me.

Is that kind of statement unprofessional sounding? If so, how can I convey the information that this is a poor quality source—it just unfortunately happens to be the only comprehensive source in existence.

  • 5
    Why not cite the earlier sources? – Morgan Rodgers Dec 15 '18 at 6:51
  • As per the other comment, the solution is evident : go to the originals... – Solar Mike Dec 15 '18 at 8:02
  • 2
    If the errors are generally typographical and the publisher has posted an errata page, you could also point there. – Buffy Dec 15 '18 at 12:16
  • 8
    You might phrase it with a kind of balancing statement like this: "despite numerous typographical errors, [1] contains valuable details and examples...". if relevant, you could also use "notation inconsistencies" instead of "typographical errors". – Erwan Dec 15 '18 at 12:58
  • If you you use that book to point to specific facts/examples I wouldn't care. If you mention it in an introduction then I would had to ask SE myself. – Alchimista Dec 17 '18 at 11:02

Citing the original sources (as suggested in Comments by @MorganRodgers and @SolarMike) would generally be preferable even if the flawed book were perfect. By doing so, you credit the discoverers directly and give a sense of the timeline of various advances in the field.

However, I think it would be a mistake for you to ignore the flawed book entirely. It may be of some use to your readers in spite of its errors. Also, you wouldn't want to appear to be ignorant of the existence of that book.

As the occasion arises to point to the work of others, credit the original sources. However, if the original sources may be difficult to obtain, you could also mention the later book (along with a brief comment in passing about any mistakes on the specific topic under discussion). You might also mention the later book if it has any value added: translations from other languages, explanatory figures, discussions of connections with other results, historical perspectives, and so on).

If the later book is in your bibliography and you mention it a few times, you have at least given it due credit and shown you are aware of its existence. If you gently point out a couple of specific relevant errors, your alert readers will get the idea that the book has some flaws.

It is always best to point out errors in gentle, non-judgmental, matter-or-fact language. In spite of your best efforts and those of referees and editors, your own paper will likely have some mistakes. You will see some of them immediately upon publication, even though they stealthily avoided earlier detection.

  • 2
    -1: Crediting original sources is not the question here. Citing the earlier work would be irrelevant to readers who just want reliable results, if the book in question was itself reliable. – Sylvain Ribault Dec 15 '18 at 17:44

Just saying that [1] contains errors is not terribly helpful: it would be better to be more specific. For example your article could have an appendix where you would list the errors that you found, ideally with some justification. (Such as: "inconsistent with equation X or reference Y".)

On the other hand, it would be unprofessional for the authors of [1] to take offense of justified criticism, or for you to mislead your readers about [1] out of fear of being offensive.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.