Read Everything
One of the stock pieces of advice that I give gradaute students is to "read everything." Of course it's impossible to read everything that has ever been written, but I do expect researchers to have read everything--literally everything--on subtopics within the field on which they are writing.
When I start a new project, I usually begin by collecting all the books and articles that have anything to do with that topic. I prefer to buy those books--especially those that are often cited in one of my own fields--rather than check them out from the library or through the ILL. I need to have all my intellectual tools at my disposal. That has always been my preference--even when I was a poor graduate student. Building a good professional library is an important investment.
I scan through them to explore the intertextuality--which sources get mentioned more frequently and how. I then collect more sources if I don't have them handy. Without this process, it wouldn't be possible to come up with viable research questions or to know what questions or concerns reviewers and readers might have.
Reading everything is especially important at the beginning of a researcher's career. One of the problem of novice researchers trying to read selectively is that they are not likely to have developed appropriate criteria for choosing what is important and what is not. The discernment in reading scholarly books and articles can come only from an extensive knowledge of the field, which is, after all, a synthesis of the diverse bodies of knowledge held by everyone who identifies herself or himself as a member of the field.
A novice researcher who doesn't try to read everything is like someone watching Star Wars Episode III without having watched the other five episodes. The main plot and some of the details might still be entertaining, but the person's appreciation of the film is necessarily limited by the lack of background knowledge and intertextual awareness.
Labels: grad school, reading, unsolicited advice