I have long made a point of trying to avoid references of the form ‘X cited in Y’ (see Eleven Thoughts on Reading and Citing), at least without first checking source X. Having spent quite a bit of time recently checking multiple references, mainly made by Canguilhem, with a lot left to do, I thought I’d try to distil some of the reasons for this.

1. X claims that Y says Z, but on checking, you find that they say something different – either a minor slip, something more significant, all the way to a downright deception
2. What appears to be a quotation is actually a paraphrase.
3. What appears to be a paraphrase is actually a quotation.
4. The quotation, when read in context, is more complicated or nuanced.
5. The reference to the work is incorrect (title, subtitle, publisher, place, year, edition, etc.)
6. The page reference is incorrect.
7. Some combination of these reasons.

This work of reading and referencing checking is not always easy. For example, Canguilhem cites works in French, German, English and Latin, and these are often difficult to find. Canguilhem used obscure editions, some not even the most current at the time, especially in his writings during World War Two when books were hard to access. Sometimes more recent editions don’t have the same pagination. In some of the translations the translators have provided a reference to an English edition, which means that there are two sources to check. Canguilhem sometimes provides his own translation of a text initially in another language, and might indicate the work, but not the page – that can take some tracking down.

So if Canguilhem provides a reference, with a paraphrase rather than quotation, in which the title given is not of a book by that author, the publisher is incorrect, and he doesn’t provide a page number (and his English translator simply copies his reference)… but I still find the quotation, do I win a prize?

Whenever I take down notes with references to other sources I write ‘[check]’ in the text or note. The square brackets, or sometime highlighted text, make it easy to locate these at a later point. Some of these I can fix with my own books; many more require a bit of tracking down. I tend to store these up and do a consolidated round of checking – starting with Warwick library, then the British Library and if neither works turning to Worldcat to work out where I might be able to locate the source. Some early sources are available online, but many are not. This long list gets progressively whittled down, and added to as I do more note-taking and read other sources. I find it helpful to consolidate these things into a list which I try to work on periodically through a project – rather than leave everything to the end – although there are always some intractable references that end up on a final ‘to do’ list. When I’m in London or, less often, Paris, I can work through a lot of these; references to check at home, the office or Warwick library can be fitted in around other tasks – they are quite good things to fill small bits of time like a cancelled appointment.

The more dishonest route, of course, is citing X and dropping the reference to Y, making it appear you’ve read something you have not. More generally it’s revealing that the same reference mistakes often appear in multiple places by different authors – a product of reference copying rather than tracking down the source. If you want to be scrupulous, I think you should check the source. I’m certainly not suggesting that my work is without referencing errors, but they are the result of mistakes rather than inaction.