This is a question I often get from my students, especially since I have a reputation for being a strict grader (to them, this is an understatement; I am known to be judicious (think Lawful Evil)).
Here’s the short answer to the question: it depends on what you’re writing, who you’re writing for, and your current skill level. It is a complicated blend of these three factors, and it’s not easy to discern the contribution of each.
No, seriously, ask anybody and they won’t give you a straight answer to this question. I had to find my own sweet spot, but my sweet spot is either too much or too few according to reviewers. What other people deem “normal” may be excessive or not excessive enough.
Now, the long elaboration. Note that this should not be interpreted as an absolute guide; instead, think of it as one of many voices out there trying to advise you. It all comes back to you and (mostly) your professor’s judgement.
Two reasons. One, it shows you’ve done the reading. Sometimes, bibliographies or reference lists are just there to show off how much reading you’ve done. But more importantly, it shows you’re acquainted with the voices in your field. Two, it shows you’re responsible. You aren’t stealing from other academics.
To understand the complex interplay of the three variables I mentioned before, it’s easier to explain through my own experience.
During my undergraduate, there were some pointers I held on to, which mostly came from reading college tip websites and asking seniors:
- If it’s not your argument, cite it.
- If it has numbers in it, cite it.
- If it is about an event, cite it.
The reasoning for this is quite simple. For #1, if you don’t cite, you’re basically stealing. For #2, it allows readers to verify my dataset. In English, it means if someone wants to dispute my data, they will know where I got it from. For #3, perhaps the event is too obscure. I would want my readers, if they wanted to know more, to be able to consult the same sources I did.
You can quickly understand why then my papers would have at least five citations per page. I would cite online newspaper articles just to show that I have been reading a lot and I understood what I was writing about. This was okay and all, but it turned out to be quite tedious. I even started questioning myself: do these things really need to be cited? Like, do I really need to put a citation when I write “Indonesia gained independence in 1945…” if I’m writing a paper about nationalism for an Indonesian audience? Shouldn’t everyone know this by now?
So, I added one more pointer:
4. If it isn’t common knowledge, cite it.
This ties in nicely with point 3, as some events would not be common knowledge for readers outside Indonesia. I know when the Linggadjati Conference was held and what it was about, but a reader from the States may have no idea what I’m talking about. A footnote to another book about the event would be welcome.
However, this quickly becomes a bit hairy as you may ask “How do you know what common knowledge is?” Here, we need to refer to your audience and current skill level. Basically, a lot more things constitute “common knowledge” the longer you stay within a certain field and accumulate experience. This is also the reason why you may see some publications have few citations: the events they talk about are common knowledge to other scholars in their field.
Here’s how it works.
Let’s say you’re writing an article about the U.S. counterinsurgency efforts in Iraq and Syria, and your audience is largely like-minded scholars in the field. You start your paper with the background of the rise of ISIS, but you don’t really need citations at this point because it is likely that your audience (and you) already knows this. Citations would be needed if the event you’re describing is recent or obscure.
Then, you start to present your thesis statement. Before doing this, you point out previous research on ISIS and criticize them for lacking something. This is where citations are needed, because you’re acknowledging other people’s work. The citations are needed so readers can read the stuff you’re citing to make sure you’re mounting a credible argument against them.
After this, you show some statistics to prove a point. Unless you got the data by yourself, a citation is necessary; even more so if you “borrowed” visual representations of the data made by someone else.
I guess at this point, you have a fair idea of how many citations you need. However, if you’re still unsure, it’s better to err on the excess. That basically means it’s safer to have more, than be caught with less. But then again, also remember that too many citations obscure your own voice.