The founder of Wikipedia, Jimmy Wales, has got it slightly wrong. In the interminable debate over the value of Wikipedia (one, two, three, among others) he is interviewedby Business Week . Asked if Wikipedia should be cited by students and researchers he says

No, I don’t think people should cite it, and I don’t think people should cite Britannica, either — the error rate there isn’t very good. People shouldn’t be citing encyclopedias in the first place. Wikipedia and other encyclopedias should be solid enough to give good, solid background information to inform your studies for a deeper level. And really, it’s more reliable to read Wikipedia for background than to read random Web pages on the Internet.

Yes, encyclopedias shouldn’t be cited by researchers and wannabe researchers. But not because they’ve got errors in them — they shouldn’t be cited because they’re not primary sources.

If you don’t cite a primary source you can’t be sure you’re getting the original information. What you’re reading might be reinterpretted, filtered, edited, or spun. If you don’t cite a primary source you may have missed certain other points which influence your case. If you don’t cite a primary source you’re not looking at the evidence, you’re looking at what someone else says about the evidence.

Citing Wikipedia or Britannica is fine if what you’re saying isn’t that important in the grand scheme of things. But if you want to come across as authoritative — and if you’re a researcher or a trainee researcher then it’s your job to be authoritative — then you need to go back to the original text. It’s difficult and time-consuming, but without it we become a society build on hearsay.

Tags: primarysources, research, wikipedia