What Good’s an Author?

When writing A Theory of Reading 1.0, I beat your heads a bit with the intentional fallacy. And I think it’s important to do so, as many people just getting into serious literary analysis still have an attachment to authorial intent and dismantling that attachment is an important part of getting people to treat a text as an object into itself which we can examine. Moreover, as I hope I demonstrated, it is simply true that authorial intent doesn’t determine what a text means. However, I may have given the impression that the author doesn’t matter at all to a text or to its meaning, and while that’s not the worst of errors, it is an error. Indeed, if I get around to A Theory of Reading 2.0, I’ll want to address this.

I think there’s rather a lot to be said about this, and rather a lot of wrangling to be done over the precise relationship between authors and texts, which I’m not going to attempt here. However, I want to note three different approaches to the author-text-meaning relationship, which I’ll call the Process Approach, the Function Approach, and the Political Approach.

The Process Approach

For all our huffing against authorial intent, literary critics tend to talk about authors a lot. First, there’s a formal requirement, where we use the author’s name as the agent creating the text: “In King Lear, Shakespeare depicts a disenchanted world haunted by the absence of ghosts, fairies, or gods,” or, “Joyce’s Ulysses, however, marks a departure from Homer’s Odyssey in that the world which the journey marks out is not geographically complete but rather biographically complete.” Intent is never quite claimed, of course, in the way that “when you said x you insulted me” differs from “you meant to insult me when you said x.” But authors are never far from the mind of even the most formalist critic.

What seems to be happening—and this is only based on observation of my former colleagues and of the theorists I’ve read—is that critics use their knowledge of an author, especially that author’s other writings, to frame their approach to the text. Someone who knows that Shakespeare was about to write Hamlet (and two notably different versions, at that) might read Julius Caesar differently than if they hadn’t read Hamlet.* Knowing that Shakespeare was about to make a breakthrough in the depiction of character psychology, the critic might be alert to moments in Julius Caesar which presage that breakthrough. In other words, knowing something about an author gives you clues for what you might look for in the text. It might help you overcome certain preconceptions you were bringing to the text. Knowledge of an author can pull out a certain pattern you hadn’t seen before.

But knowledge of the author isn’t going to give you evidence for your new reading. It will just put you in the frame of mind necessary to notice something that you hadn’t noticed before. From that point on, you’ll still have to do what I described before: make arguments about the text using the text’s own features as evidence. Thinking about the author is part of the intellectual process, and it might help to replicate that process in your output (journal article, conference talk, high school composition, whatever) to help your reader see the text in such a way that they’re amenable to the evidence…but the evidence must be there and the argument must work on its own.

The Function Approach

Of course, the author might well be a feature of the text, after a fashion.

I highly recommend that you read Jorge Luis Borges’s short story/essay “Pierre Menard, Author of the Quixote.” The narrator reviews the literary output of the recently-deceased Pierre Menard, most of whose work is unremarkable. Menard’s magnum opus is remarkable, however: Menard (re-)wrote several fragments of Don Quixote as though he had written them for the first time. The narrator goes on to explain how Menard’s Quixote is better than Cervantes’s, even though they use identical words in an identical order. The only difference between them is the author. And if the narrator is right, and Menard’s Quixote is better than, or at least distinct from, Cervantes’s, then this implies that the author, somehow, matters.

Now, an obvious explanation comes to mind. Menard was purportedly writing in a much different context than Cervantes, and in Menard’s time certain opinions present in Don Quixote are much more controversial or surprising than they would have been in the time period of the original. I remember my father speculated that the title of the second Lord of the Rings movie, The Two Towers, bore some relation to the then-recent attack on the Twin Towers of the World Trade Centre. I explained that the title was much older than that, but imagine if that title had only been chosen when the movie was produced. It would seem a lot different, wouldn’t it? We already discussed that the context of a text’s production matters because that is the context in which it means something, but it’s worth noting that the author is a way of indicating which context was the text’s.

There’s more, however, which I only want to indicate briefly. Foucault’s “What is an Author?” discusses the author function; put simply, the author function is the role the author plays in our analysis of the text. I admit to being unequal to the task of fully explaining the essay, since I’ve only just skimmed it today and it’s been a few years since last reading it, but the gist is this: the author of a text is not identical to the real person who wrote it, but is rather a sort of fictitious persona attached to the text (rather like a narrator, but more closely aligned to the writer). This author’s assorted characteristics are taken seriously by readers of the text. I advise that you go and read the essay; certainly I intend to, and I’ll come back once I have and explain it better, perhaps. But what I want to note is that certain ideas about the author might be a feature of the text itself. Note that I’m not sliding back into the intentional fallacy here, and for two reasons: 1) it is the perceived author or the received idea of the author, not the historical facts about the writer, that is a feature of the text, and 2) this author-function is only one feature of the text, and it might easily be outweighed by the other features of the text. For instance, consider Paradise Lost. Many people read Paradise Lost as a critique of Christianity, and it may be so, but if those same people go on to attribute that critique to Milton, they are mistaken. Milton, by all accounts (such as those of everything else he wrote), is about as devout a Christian as there possibly could be. Milton’s piety might be a feature of the text, but it doesn’t determine the meaning of Paradise Lost. Rather, it merely adds “despite Milton’s own intentions” to an explication of the text. As William Blake wrote, Milton “was a true Poet and of the Devil's party without knowing it.”**

But this creates a sort of messiness which makes me feel uncomfortable. It tends to make texts replicate beyond limit, because each time you attach a different author function to a text, you get a different text. Paradise Lost, read in ignorance of Milton, is a different text than Paradise Lost read with this perceived Milton attached. Similarly, Don Quixote is a different text when we pretend Menard wrote it than when we acknowledge Cervantes wrote it. This messiness, however, might be inevitable. Again, I’ll return to this question after I finish re-reading “What is an Author?” But I want to note that the author might be a feature of the text, but this isn’t the same thing as saying that the author’s intent determines the meaning of that text.

The Political Approach

I was talking about Laura Mulvey’s film criticism with another Master’s student (now a PhD student) a few years ago; Mulvey made substantial arguments about the male gaze in film, and how it determines the representation of women in those films, which became a very common mode of feminist analysis of and in popular culture, but this colleague expressed skepticism because many of the filmmakers were gay men. Gay men can, of course, still have gazes, and it will be male, but it won’t operate the way Mulvey describes. What’s notable is that both Mulvey and this colleague are concerned with who is doing the gazing. The author matters.

Many conversations concerning identity politics focus on who is saying something or wearing something. I mentioned this briefly in my run-down of different schools of criticism. Concerns about cultural appropriation abound. And there are also concerns that not enough women, people of colour, etc. have opportunities to make culture. There is further discussion about how people of different demographics create different kinds of culture; women tend to represent female characters differently than men do. All of these concerns focus necessarily on the author.

Now, there’s maybe a confusion of terms and concepts here. Concerns about gaze and voice, as they impact the text, can be chalked up to the author function, perhaps, depending on how that shakes out. Meanwhile concerns about women creators or queer creators can be treated separately from concerns about the meaning of the text; indeed, if we’re noticing that women portray female characters different than men do, we need to examine those portrayals as independent objects without reference to their creators first in order to show that it’s the portrayal that’s different and not our perception of it. It’s possible that a lot of this concern is part of the second step of an argument, where the first step is an explication of the text itself and the second is an explanation about why that text’s meaning matters to the culture as a whole. Only the first step is literary analysis as I’ve described it, but the second step is part of what makes literary analysis useful.

But I think there’s a lot more to be said about how the author’s particular demographics and positioning matters to the text as a speech utterance. For instance, look at this paean to the YouTube song cover, which is not trying to make a point about literary analysis but nonetheless shows how the singer of a song influences the meaning of that song. Singers, of course, appear in songs much more visibly than authors appear in books, but songs are texts, too, and can be read as such (though they require familiarity with a different set of conventions and technical matters), and so if the singer influences the meaning of a text, so too must the author to some extent. Probably the Political Approach is equivalent to the Function Approach, but certainly not all people concerned with the politics of authorship would think so.

So that’s three different ways of thinking about how authors matter to their text’s meanings without determining those meanings. I’m quite aware that my thinking in this area is still weak and needs more work, but that will have to come sometime in the future. If you have any concerns or contributions, let me know.

Index

---

*For more on the two versions of Hamlet, and how the second version is a watershed in not just Shakespeare’s own writing but English-language literature generally, take a look at Shapiro’s 1599: A Year in the Life of William Shakespeare. Honestly, if you have even a little bit of interest in Shakespeare, or literature, or English history, give it a read: it’s one of those academic-topic-for-popular-audience books, and it’s good at being such a book.

**I don’t actually buy the Paradise-Lost-as-critique-of-Christianity argument, but even if it’s the case, that doesn’t say anything about Milton.