Here's a bit of advice that people who aren't English majors can often stand to hear: in an English paper, don't talk about the “real world.” Talk about writing.
What do I mean by that? For starters, don't assume literature is a transparent window that shows us the real world — it's not something we can reliably look through. Often it's more like a painting than a window, and instead of looking through it we should learn to look at it. In other words, you have to pay close attention to the language, at how your author makes his or her points.
Why do I keep harping on language? Because it's all we've got. Literature is nothing but words — well, okay, some forms might have music or pictures, but usually we're dealing solely with texts. Historians try to talk about the real world; lit-folk content themselves with talking about the way that world gets represented in language.
Now, this doesn't mean you can't be interested in the real world behind the text. Some of the best papers are sociological, looking at the way the economy works in this book, or how gender relations are treated in another. Just remember that you don't have any direct access to that real world, only representations of it. Never lose sight of that fact.
Let's say you're writing about Chaucer's Wife of Bath's Prologue, and want to discuss the treatment of women. Fair enough — it's a good topic. The problem is that you can't write with confidence about the treatment of women in medieval England just by reading Chaucer. What makes you think he's telling the truth? — or that his experience is in any way typical of medieval English experience? It can certainly help to know a little something about that subject to understand Chaucer, and if you already know something about the subject Chaucer's poem might illuminate it. But if you want to know how real women were treated by real men in the real England of the real Middle Ages, you're going to have to do a lot of outside reading. One source just ain't enough.
(If you've actually done research on the real world, of course, you can bring it to bear on your paper. If you know something about the history of slavery, that might make your paper on Uncle Tom's Cabin more persuasive, and if you've read up on seventeenth-century Anglican theology, you might see things in Donne's poems other people have missed. Having some historical knowledge about the actual treatment of women in fourteenth-century England can make your discussion of Chaucer much better. Be careful, though, not to be reductive: it's easy to take commonplaces about the past and exaggerate them into falsehoods. If you assume “Women had no rights before 1900,” or “All Christians despise Muslims,” or “Neoclassical poets uncritically admired ancient Rome,” you're bound to say stupid things about the works you're reading.)
The fact is, you can't assume any single author gives us a trustworthy account of the way things were. You can, however, talk about how Chaucer portrays women in his works, and although that might not be interesting to a historian, it's exactly what you should be looking for in an English class. In other words, though you can't reliably talk about women in medieval England, you can talk about women in Chaucer's texts.
And it's not only grand generalizations about all of society that can give you trouble: even comments about a single author can misfire. Reading As I Lay Dying doesn't authorize you to describe the real-life, flesh-and-blood William Faulkner: it authorizes you to talk about As I Lay Dying. Be careful not to make generalizations about an author you can't back up.
It's especially difficult with nonfiction. When the poet Robert Burns says his love is like a red, red rose, we don't assume all women in late eighteenth-century Scotland were really like red, red roses; but when James Baldwin tells us about the way children treated him in a Swiss village in 1953, we might easily assume that it's an accurate description of Swiss villages in the '50s. You still have to be very careful. This isn't necessarily because nonfiction authors lie (although some of them do), but because an author's perspective may be limited, because some facts may have been changed for effect, and for a dozen other reasons. Remember that writers are not on oath.
Perhaps it sounds a little abstract, but it has real consequences in how you approach your thesis.
Let's see some examples. The bad ones all presume to talk about the actual world; the good ones shift the focus to the works of literature.
[Still under construction — sorry.]
|Honor was a very important concern in Homeric Greece, and people ???.||Homer depicts a world in which action is motivated by concerns of honor, and he is always careful to ???.|
|Augustine lived a||Augustine's metaphors always stress ???|
|Samuel Johnson had no respect for women, as we see in his many misogynist comments in Boswell's Life of Johnson.||In The Life of Johnson, James Boswell often goes eout of his way to emphasize Johnson's misogyny, drawing attention to his negative comments on women while passing over his more evenhanded statements.|
The moral of the story: never go beyond your sources. You may think you have to say grand-sounding things about timeless questions on human existence, but you can't. If you're writing a paper about Virginia Woolf's Room of One's Own, you can't say things about early twentieth-century feminism generally — you can talk only about Virginia Woolf's Room of One's Own. The good news is that A Room of One's Own is rich enough to warrant a lot of attention; you're not likely to run out of things to say about it.