In what I called A Theory of Reading 1.0, a series at The Thinking Grounds, I tried to articulate and justify what seemed to me to be the underlying assumptions and approach of literary interpretation in the English discipline. In its supplementary materials, I tried to tie up some loose ends, especially shoring up my explanation against authorial intent and reader-centred interpretations. I’m glad I did that work, because it led to my writing about Twitterary Theory, which I’m pretty happy with.
However, there are some outstanding problems with that version. I say “version” because I understood it as an initial attempt. One day I am going to attempt a 2.0, maybe a 3.0, as needed. I expect that I will need to update my understanding of interpretation as I learn more. Even as I wrote it, I knew that I was weak in the area of what Foucault calls the “author function.” I still do not feel equipped to deal with that problem. Since writing the series, a few more problems have come up, too.
First, what about singer-songwriters? My friend Jon Wong insists that listeners understand songs in relation to the real biography of singer-songwriters. For the most part I would say that my explanation of the problems of reading are universally binding; whether or not listeners attempt to understand songs in relation to their singers is utterly irrelevant to the dynamics of interpretation in the same way it is irrelevant with readers, books, and writers or viewers, paintings, and artists. I do want to bookmark Jon’s concern, though, because it will come up later.
Second, what about fan fiction? When I was discussing theoretical concerns with my friend Sunny Chan,1 Sunny waved away most of what I was talking about because she works on transformative reading, not the more strictly interpretative work that I do (or would do, if I were still in the discipline). In part I could lump fan fiction in with reception history, which looks at how people in various times and places have understood a particular work, but I think there is something more to fan fiction than just that: there’s good evidence that writing fanfiction helps young people understand themselves, their worldviews, and their society.2 I strongly dislike the idea of a headcanon—an interpretation of a text that knowingly operates without textual warrant and sometimes directly contradicts particular elements of a text—but it seems to operate in the same category: it is a practice that some people engage in, and not all of those people can be written off as ignorant of “proper” interpretative practice.
Third, what about J. K. Rowling’s tweets? My friend Natalie Morrill3 posed this question to me over Facebook:
j.k. rowling is tweeting about the hp world (i.e. that harry’s son got sorted into gryffindor today, which was not included in the books). i’m wondering whether i would consider this “canon.” then i thought: for sure christian would have thoughts.
I think canon is such a weird construct anyway. I mean in the sense you are referring, as in, “What are the bounds on this text?,” rather than, “Is this book part of the literary tradition?” The second sense is also a weird construct, but at least I know it’s history. I /assume/ the idea of canon you’re discussing comes from the canonicity of Scripture, via historical-critical/new critical literary analytical processes. Long story short, I’m not sure that “canon” is a sensible concept, really, as though a text is somehow perfectly bounded or pristine.
That said, whatever may or may not be canon, I usually don’t consider authorial pronouncements to be part of the text (nor should anyone); authors can and often do publicly interpret their own text, but such a thing is no more than an interpretation (from an admittedly unique relation with the text). So, musings on Dumbledore’s sexuality: author interpretation, though one that many readers had also come to independently. However… now I’m not sure. I haven’t seen these tweets, but … are they a kind of text? And if they are a text, and can count as part of “Harry Potter fiction,” broadly, then is there a reason that /oral/ texts could not? It’s an interesting problem. I think the solution to that problem is to deny the entire concept of canon altogether, and that whenever one embarks on an act of interpretation, one should mark out the text they are working with, including or excluding its paratexts and whatever texts it has a relationship with (sequels, prequels, etc.). There is no platonic Hamlet: there is only the version of Hamlet you are working on.
Yeah. It’s an interesting problem. I may need to revise quite a lot of my thinking on the nature of texts. Thanks for drawing this to my attention.
That’s very much in line with my thinking. At first I thought, This is like J.K. Rowling’s HP fanfiction – no more or less authoritative than anyone else’s. (E.g. If I’d written HP fanfiction wherein this kid got sorted into Slytherin, I don’t have to consider this “alternative universe” now.) But then the idea of Tweets as text (i.e. this is an authorized sequel to the series) started messing with that. I suppose there’s a possible middle ground. You’ve delved into this much more and have a better vocabulary for it than I have, I think.
I think my concern here is well-expressed in that conversation, but a reformulation might help: what are the bounds of a text? How does one determine or decide what does or does not count as part of the text one is studying? In fandoms, this is a question of asking what is or is not canon; while the practice of determining canon does draw on various forms of evidence and reasoning, in my experience this practice is inconsistent towards whether it is arguing, in an empirical or absolute sense, which texts are canonical and whether it is determining, for this community’s further conversation and practice, which texts count as canonical for them. Is canon about truth or about consensus? Let us assume we are not part of a fandom but interested in something more like academic interpretive practice. What is the analogous process for deciding what counts as part of a text?
One day I will write “A Theory of Reading 2.0” to address these problems. In the interim, let me outline what I’ve realized since writing 1.0 in light of these problems.
In “A Theory of Reading 1.0,” I began with a set of specific examples of interpretation which I hoped would, in combination, eliminate various interpretative fallacies—the Affective Fallacy, the Intentional Fallacy—and reveal a particular necessary and irrefutable truth about interpreting texts: that each text has its own field of possible meanings, in relation to a language, independent of both authors and readers. Interpreting a text’s meaning, therefore, involves making arguments about the priority, superiority, or at least legitimacy of particular meanings by looking at the elements of the text itself in relation to its language—a language specific to its time and milieu, loaded with allusions and connotations. You can look at the argument itself for more detail. To draw on the language of personal epistemology, I was insisting on an evaluativist approach: knowledge of the text is constructed by an interpreter and evaluated according to evidence (in this case, the words of the text itself and knowledge about its associated languages) and reason.
In retrospect, the sorts of evidence I was considering are maybe better considered as vexations rather than as absolutely limiting constraints. Understanding the author’s intent is of course impossible if you want absolute certainty; however, if I am allowing an evaluativist approach to understanding a text’s internal field or structure of meaning, then I might have to allow an evaluativist approach to understanding an author’s intent. The limitations I outlined—an author’s failure or imprecision in writing, the inherent ambiguity of language, the ways a language contains more than an author realizes, an author’s inconsistency in vision, an author’s own unexamined assumptions and biases—do not obliterate the possibility of assessing authorial intent so much as make such attempts imperfect. But attempts to understand the text’s own meaning are also imperfect. Of course I tried to address this, but I think I have a better way of articulating what is going on here.
There are possibly three kinds of interpretation: transitive, intransitive, and transformative.
Transitive interpretation attempts to understand something beyond the text object itself, something which the text communicates according to the conditions of its creation: the author’s state of mind, information which the text conveys, and so on. This is, as I said, a vexed process; no absolute understanding of that information is possible through the text (or out of the text, but that’s another story), yet the text is still evidence for better and worse arguments about the author, the text’s topic, and so on.
Intransitive interpretation attempts to understand the text object itself. The text does have an internal meaning-structure, as I described in A Theory of Reading 1.0, independent of particular authors and readers but dependent on particular languages; intransitive interpretation is the attempt to understand that meaning-structure. It follows the general rules I described before, but some elements of the author-function would still apply here.
Transformative interpretation attempts to move beyond the text object itself, like transitive interpretation, but it does not look at some originating communication. Instead, transformative interpretation concerns itself with the potential meanings which are unrealized in the text. These are not what I am calling “possible meanings”; in the text as it is, they are not good readings. However, by changing the text in some way—adding to it with fanfiction, making unwarranted interpretative rules as in headcanons, re-writing certain sections, even just changing its linguistic context—the interpreter not only creates their own text object but highlights something about the original text object, pointing out an absence, a thread of meaning that remained undeveloped, an assumption so deep as to be almost invisible.
So, to Jon I could say that transitive and intransitive interpretations are still distinct—the song’s own internal meaning remains independent of what it communicates about the singer—but transitive interpretation is still possible and valid. Moreover, intransitive interpretation might aid in transitive interpretation: beginning by understanding the song itself, without reference to the songwriter or singer, might help improve the ultimate aim of understanding the song with reference to the songwriter or singer.
To Sunny Chan I could say that I am dealing with intransitive interpretation, and that it is a different thing than transformative interpretation.
To Natalie Morrill I still do not quite know what I would say. This typology or taxonomy of interpretations does not address a fundamental question: what is a text? I’ve done a bit of research into this problem when discussing paratexts, but that might be for another day; in the meantime, though, I will point out that if adding to a text is transformative interpretation in some way, then even things like sequels and prequels could be seen (probably should be seen) as transformative interpretations of the original text. So, there is a text which is Harry Potter and the Philosopher’s Stone; there is another text, a transformative one, which is Harry Potter and the Chamber of Secrets added to The Philosopher’s Stone; there is another text which adds Harry Potter and the Prisoner of Azkaban, and so on. A transitive reading would need to keep track of this: while we have some extratextual indications that Rowling had plans for later texts when she wrote The Chamber of Secrets, we also have extratextual indications that these were limited plans which she later changed in some respects; what Rowling intended with The Philosopher’s Stone may not be identical to how Rowling interpreted it when she wrote The Deathly Hallows or tweeted about James Sirius Potter’s Sorting. An intransitive reading would also distinguish between these interpretations: The Chamber of Secrets includes The Philosopher’s Stone as allusion, but The Philosopher’s Stone does not include The Chamber of Secrets. However, you could do an intransitive reading of the entire series as a single object, which would allow you to refer to later books when interpreting earlier ones. Finally, as much as the series itself is a transformative reading in the succession of books, a person could do transformative readings of the entire series by writing fanfiction, creating personality typologies, and so on; those transformative readings could become text objects of their own, and the cycle goes on.
As is perhaps evident from those three responses, a person can try to make a clear distinction between these types of reading, but they tend to involve one another to some extent. What I hope is also clear is that all three types of reading place a significant degree of responsibility on the reader: given the vexations of the first two approaches, the non-absolute and always open relationship with something real (a person, a text’s internal meaning), an interpreter therefore has responsibility for their interpretation and whether it makes best use of the evidence and reasoning available to them; given the freedom of the third approach, an interpreter therefore has responsibility for what they have created as both a reader (in their interpretation of the text to which they respond) and an author (in their transformed version of the text).
While I do not necessarily mean to use the term responsibility in an ethical sense, I think we should point out that when interpretation becomes an ethical matter—you decide to act on an interpretation or you relate differently to someone based on an interpretation—then it matters who has responsibility for what. Authors have responsibility for the text they have created; interpreters have responsibility not just for creating the best possible interpretation of the text from the evidence available, but for selecting the most appropriate types of interpretation from those above.
While I have to do a lot more work before I could incorporate these into A Theory of Reading 2.0, which I imagine would begin with a similar from-the-ground up approach as the previous one, I did want to submit these insights to public scrutiny first. I would be thrilled to receive feedback, reasoned criticism, or suggestions.
1. Sunny is also a university instructor and professional if still fledgling academic, if that changes anything for you. We collaborated on the Weekly Wonders project and she has various projects of her own, too. One of them is called animorphs studies, which she has not updated and which you should prod her into updating. She has also written two excellent things for interfictions: “The Difference Between an Arm and a Wing” and “a Collection of things arranged in order.”
2. I have a list of sources somewhere supporting this point. Let me know if you want me to dig them up.
3. Natalie is a university instructor and an upcoming author of a book that will not be titled At the Top of the Wall, Alight. It is confusing. She has also written numerous other things (including an excellent poem in Carousel about a woman who realizes her husband is an octopus) but to my knowledge I cannot link to them because they are in print publications.