Note: My title for this entry gives me this mental image of Charlie riding a dragon through a remote Carpathian valley, approaching a dark cave in which a-- well, I suppose it would have to be a cannon-- lies hidden. If I had the talent for it I'd draw a picture to go along with this entry. Unfortunately, the idea of riding a dragon wouldn't be canon, in the sense discussed here. Oh well. It would have been fun. Kind of like the late, lamented Snape-vampire theory; but that is a topic for another discussion….
I am not referring to the question of whether to define the "meaning" of a story in terms of the author's intention or of the reader's perception; that ground has been covered often enough-- perhaps too often-- before, by me and by others. Suffice it to say that I am assuming here that the reader shares my view that the authorial perspective is most sensible; anyone who disagrees probably isn't wasting his or her time reading my stuff anyway!
Rather, the issue I am discussing is this: For some questions about a story, a simple reference to the author's perspective is sufficient to provide a clear answer, either based on the text itself or on the author's explanations given in published interviews. But in other cases-- the ages of Bill and Charlie Weasley being a current hot topic among the various possible examples-- the evidence provided by the author may be confusing or even contradictory. What should we consider to be the real story, the "true canon," in those cases?
Various answers are possible, even among those who recognize the author's authority in defining the story. One Quiller has stated that "personally I consider the chat to be the true canon in this case" (of the older Weasley ages). Others would prefer to stick with the books when the chat disagrees; and still others might say that neither possibility can safely be considered canonical until JKR clearly addresses it (and does so with more careful thought than is possible in a one-hour rapid-fire chat session). (Of course, in the Bill-and-Charlie case, that's assuming that the evidence really is contradictory; Rather than simply diving into how a definition of "canon" should handle cases like this, however, I'd prefer to take a step back and consider some more general principles of what our recognized "canon" should accomplish for us. The ultimate idea lurking behind all of this will be, of course, that what we recognize as "canon" should enable us to have a good discussion of the books themselves; indeed, by looking at my principles of canon from a certain sideways angle, you'll probably be able to rephrase them into statements of what I consider a "good" discussion!
(Note: Ironically, as I've been typing this, I've been listening to my newly-purchased Canadian Brass CD of J.S. Bach's The Art of Fugue. As I was typing the above paragraph, they came to the end of Contrapunctus #14, which Bach left unfinished at his death. I was interested to discover that they chose to break off in the middle of the piece, where Bach had left it, even though music scholars are able to produce a plausible reconstruction of how Bach might have completed it had he lived. This is, in effect, a musical equivalent of the same question-- in this case, "what should we consider the 'real' Art of Fugue?" …Of course in the case of Bach, you'd want to avoid the term "canon" in that context, as it refers to a musical form closely related to the fugue. And no, I'm not going to write an entry titled "Johann Pachelbel and the Quest for the True Canon!" …But back to the topic at hand….)
So anyway, I would propose the following characteristics that I'd like to see in a definition of canon. Note, by the way, that these are intended to help us define the canon of a story. If we were talking about a different literary genre-- poetry being an obvious example, or for that matter postmodern literature-- we would no doubt use a different set of principles. But as My first proposed principle of "canon" in story discussions, then, is:
(1) Our standard of canon should reflect the realities of human knowledge.
That is: In ordinary life, we find that some things are so obvious you don't have to think about them; other things you have to think about, but still allow you to reach a conclusion that's definite enough for us to safely suppress doubts. Still other things in life are uncertain enough that we do have to keep an open mind, pending further evidence; and of these, there may be a range of probabilities, ranging from "quite likely but not certain" to "highly unlikely but still possible."
Similarly, since in discussing "canon" we're discussing what we know about the story, the same set of categories should apply here. Some things will be "obviously canonical"; others will be "clearly canonical but not obvious" (this distinction comes into play in JKR's differing answers to the H/H and D/H questions; more on that when I write an entry on the March 4 chat session). Other things may be "probably canonical," "realistically possibly canonical," or "unlikely but still possible in canon." A helpful definition of canon will recognize these degrees of our knowledge; it should allow for definite "yes" and "no" answers where appropriate, but shouldn't force us into "yes/no" answers on everything.
My second principle is related:
(2) Our standard of canon should be available for discussion.
I agree that it would be possible to define "canon" as "whatever JKR imagines, whether she has communicated it at all or not." Indeed, at times I have leaned toward that definition. However, since the various meanings of "canon" in the English language have generally implied a rule or standard for objective measurement, I now think it's best to define the HP canon in terms of what JKR has made available for us to know, so that we can indeed judge our interpretations objectively by it.
Having our canon "available for discussion" also accords well with the nature of language, and therefore of literature, as a communicative act. As I commented on Thus by this standard we may say that the HP canon expands every time JKR reveals new information about the story to the public. When it's an insider sworn to secrecy, like Steve Kloves or Alan Rickman or Robbie Coltrane, then I wouldn't consider it "canon." But when JKR is deliberately making information public, whether in a book or in a chat or interview, she is thereby continuing to tell the story that she started with the publication of PS/SS. It's all part of the "reading experience" that she has chosen to give us; therefore there is no harm in considering it "part of the story."
Also, I should note that "availability" doesn't mean that there's no judgment calls on our part involved; nor does the required exercise of the interpreter's judgment mean that canon is not "objective." It's "objective" in that it's something outside ourselves that we're observing based on evidence. However, the validity of our statements about canon will be only as good as the quality of our judgment of the evidence and the conclusions we can draw from it.
Next:
3. Our standard of "canon" should be in harmony with the author's intended presentation.
As I remarked at the beginning of this essay, this is well-trodden ground, and I'm not going to argue for it in detail here. However, one aspect of authorial interpretation that I wish to emphasize here is that, if the author is unlikely to have imagined something, then it shouldn't be considered part of canon. This sometimes comes up in the attempt to resolve contradictions by means of elaborate explanations. If the evidence suggests that the author made a mistake, let's just say that instead of trying to get too clever.
For example, it is unlikely that JKR has concerned herself with questions like why September 1 or October 31 have been on the same day of the week in two consecutive years, which isn't possible on a standard calendar. To suggest that wizards do like Tolkien's hobbits and put in one or two days a year that fall in between calendar weeks, or that the years in the story have only 364 days (or some other multiple of seven), would in my view be an exercise in ridiculousness unworthy of being described as "interpretation" at all.
My fourth and final principle is somewhat related to the "authorial" principle above, in that we want the "canon" to reflect the kind of story the author is trying to tell:
4. The "canon" should be internally consistent.
This standard of course is only relevant to genres where consistency is desired; I use it here because JKR is clearly intending to tell a consistent story. This standard can therefore also be applied to Tolkien, and indeed to most traditional storytellers. It may not always be applicable to poetry, to postmodern literature, or indeed to humorous parodies in which worrying about inconsistencies misses the whole point of it.
This can be illustrated by the wand-order mixup from the first printing ofGoblet of Fire. Even if the error had not been corrected, we would not wish to maintain the following three things all at the same time:
(1) "It is canonical that James was killed first";
To maintain a consistent canon, however, we'll have to allow for degrees of uncertainty in most cases of apparent mistakes. The wand-order one was pretty straightforward; after reading GoF I had little doubt that the "ideal story" would have Lily coming out of the wand first instead of James. (And the revised edition confirmed my expectations on that point.) But I must emphasize here that applications of this principle will rarely allow us certainty, but only a general sense of likelihood. Other cases are more difficult; in the Weasley-ages case, it is conceivable that we might have to say, "maybe JKR was mistaken in the chat; maybe she was mistaken in one of the books; we don't know."
I am aware that this principle can easily be abused. The "ship debates" provide obvious examples, as for example when people say that if JKR puts Ron and Hermione together romantically it will be a "contradiction," untrue to herself and to the story as a whole. But in those cases (ship-related and otherwise) I can only say: "It is not possible to have a reasonable discussion with an unreasonable person." It is for that reason that I prefer to avoid discussion forums whose participants' judgment I cannot trust to avoid this trap. But I continue to maintain that looking for the "ideal story" is a reasonable standard, and that it can be useful when applied by reasonable interpreters.
We are now ready to consider my proposed definition of canonicity:
Definition: The canonicity of a statement about a story
By this definition, then, it is (in my judgement) "probably canonical" that Harry is going to work toward becoming an Auror in the next year of his studies. Hermione's, Ron's, and Ginny's middle names only became "canonical" when JKR told us about them on March 4th (and, likewise, Snape having some form of vampirism had been a canonical possibility until that same chat, but is so no longer). None of the possible explanations of why Halloween was on Friday two years in a row is (realistically) canonical at all. And after the GoF release, it was (I think) "probably canonical" that Lily had come out of Voldemort's wand first.
Finally, let's consider the evidence concerning Bill's and Charlie's ages, and see to what degree we can consider the various possibilities "canonical." We observe that:
(1) In PS/SS, Bill and Charlie have already left Hogwarts, while Percy is in his fifth year;
The difficulty is that if Charlie were exactly two academic years ahead of Percy, he would have been still in school in Harry's first year-- unless he had left early; but given Fred's and George's reluctance to leave early, it seems extremely, extremely unlikely that Charlie had blazed that particular trail ahead of the twins. Furthermore, if Charlie had been at Hogwarts the year before Harry's came, the remarks about not having won "since Charlie left" would make little sense (and it would also contradict the general indications that Harry would have given them an improvement over the previous year's team).
At one point it had seemed natural to suspect that Gryffindor had won the Quidditch Cup in Charlie's seventh year, seven years before PoA. This would have mean that Ginny was not remembering Bill's first year, but only the fact of his being at school (Charlie would have been there also, of course). It would also have implied a large age gap of seven years between Charlie and Percy, contradicting the two years indicated in the World Book Day chat.
In attempting to reconcile all this difficult information, I recommend, not coming up with a far-fetched explanation that technically reconciles everything (like Charlie leaving school early-- even though nobody specifically said he didn't, it's still quite out of character and undermines the explanation for the twins' reluctance to do so), but rather establishing which of the points of evidence are most significant to the story, and giving the heaviest weight to those.
In particular, I would rank the significance of the relevant points-- the points that would "change the story" most if wrong-- more or less as follows:
(1) that Charlie was not in school in Harry's first year;
(Note that, since the exact numbers pretty clearly are not JKR's major focus in putting the story together, I give them the least weight as evidence, and allow them to be overridden most easily.)
If Charlie had finished his NEWT's in 1990 (leaving one year between his Quidditch tenure and Harry's), and if Bill had finished the year before, then Bill's first year would have been 1982-83. Ginny, having been born at some point in 1981, would almost certainly not have remembered Bill leaving for school in September, but might conceivably have had a dim memory (supplemented by family pictures, perhaps) of Bill coming home for Easter break and going back to school (Ginny could perhaps have been two by then). Gryffindor may not have won the Quidditch Cup in Charlie's seventh year; the "worst luck in the world" to which Oliver refers may have extended back into Charlie's time (although I don't know that we can be certain about this).
The biggest "contradiction" here is that it leaves a four-year gap between Charlie and Percy, at least in their school attendance years (their birthdates may have been only slightly over three years apart). But since JKR was groaning about "maths" in answering that question, I'd give the "two years" figure the least weight of any evidence here-- certainly less than the fact that everyone in PS/SS speaks as if Charlie had been gone at least a full year. Bill's "five years" (which by this storyline would actually have been six) may also be thought of as an approximation without harm to the story. (The "seven years" without a Quidditch cup does have some relevance to the story, as Oliver seems to have been waiting his whole school career for Gryffindor to win it; so I give that more significance than the chat answer.)
So I hereby suggest the following timeline, with my best assessment of the "degrees of canonicity" involved, based on my best understanding of how JKR would be most likely to resolve the apparent contradictions if she were to work through the issue:
Bill Weasley born: fall 1970 (tentative possibility)
This also puts the most tentative ages for the Weasley kids as follows (at the end of OotP):
The next-best possibility, I think, is that Charlie and Bill may each be a year younger than shown here. This would ease a few of the difficulties (Ginny's early memory of Bill, the "five years" since Bill had been at Hogwarts, and the "two years" between Charlie and Percy), but at the cost of forcing some awkwardness into PS/SS-- perhaps a reason why Charlie wouldn't have been on the team in his seventh year? But Oliver and Fred both seem to speak as if Charlie had been gone at least a year, and I consider that a greater difficulty than making two of the year-values approximations and giving Ginny a very early childhood memory.
On the other hand, given that there are three pieces of evidence suggesting a 1990 NEWT date for Charlie (JKR's "two years" in the chat, Bill's "five years," and Ginny's "ever since Bill came"), it may well be that JKR has generally imagined the timeline that way, and that she had miscalculated herself into imagining that Charlie had been gone a year when she was writing PS/SS.
But, given my definition of "canon," the question is: Which would she be most likely to revise? Given the emphasis on the poor team in 1990-91 and everyone's slightly distant memories of Charlie, I think it would be more difficult to revise that out than to change her assumptions on the age gaps; and for that reason, I think it makes best sense to think of that as the "most likely canon." But this remains a difficult judgment call.
(Note that, even among us in the authorial-interpretation camp, some might prefer to define "canon" as "whatever evidence suggests JKR thinks on the question at this time." If you use that definition, then yes, you would probably consider it "more likely canonical" that Charlie was only three academic years ahead of Percy. My reason for not defining canon that way is because the definition doesn't handle contradicitons well.)
In any case-- by either of these definitions, really-- I do not believe we can point to One True Canon on this issue-- only to a set of "canonical possibilities." Of the possibilities, I think Charlie finishing in 1990 is the best; but if someone else thought 1991 were more likely, I wouldn't argue the point. I would be more tempted to argue the case if someone did argue for it as a clear-cut case one way or the other; but even then, it might be just because we're working from different definitions of canon (in which case the difference might be only a miscommunication rather than an actual disagreement).
Most of the time, canon is a clear-cut matter; otherwise "canon" wouldn't be useful. But sometimes it's less clear; and if we expect it to be black-and-white every time, then I don't think we're being reasonable. This, I submit, is one of those times.
Suggested Characteristics of a "Canon" Definition
(2) "it is canonical that Priori Incantatem makes a wand regurgitate its spells in reverse order"; and
(3) "it is canonical that James came out of Voldemort's wand first."
* * *
The Definition, and Some Quick Examples
is the degree
to which it reflects available knowledge
of how the author would imagine the story
given sufficient opportunity to refine it.
Applying the Definition: Bill and Charlie's Ages
(2) also in PS/SS, McGonagall tells Oliver that "heaven knows, we need a better team than last year," having been "flattened in that last match by Slytherin," and Fred tells Harry (on hearing of his appointment to the team) that "we haven't won since Charlie left."
(3) in CoS, Ginny (in her first year) says she had looked forward to coming to Hogwarts "ever since Bill came";
(4) in PoA, Oliver says that Gryffindor haven't won the Cup for seven years;
(5) in GoF, Bill says he hadn't been back at Hogwarts in "five years"-- presumably since his NEWT's;
(6) in OotP, Fred and George stay in school for most of the year only because of how badly it would shock and upset Molly if they left early;
(7) in the World Book Day chat, JKR said that Bill, and Charlie were two years apart, and likewise Charlie and Percy.
(2) that it would have been out of character for Charlie to leave school early;
(3) that it had been several years before PoA since Gryffindor had won (apparently seven?);
(4) that there was not a large age gap between Charlie and Percy (although "two years" may be an approximation);
(5) that Charlie had apparently been gone at least a year before Harry came to Hogwarts;
(6) that Ginny could apparently remember Bill being at school, before Charlie was there;
(7) that Bill seemed to have finished school five years before (although this also may be an approximation).
Charlie Weasley born: summer 1972 (tentative possibility)
Percy Weasley born: fall 1975 (time of year tentative)
Fred & George Weasley born: April 1978 (certain)
Ron Weasley born: March 1980 (certain)
Ginny Weasley born: spring 1981 (time of year quite uncertain)
Bill begins at Hogwarts: September 1982 (tentative possibility)
Ginny's first memory of Bill going to school: post-Easter 1983? (very uncertain)
Charlie begins at Hogwarts: September 1983 (tentative possibility)
Charlie's last Quidditch Cup: spring 1986? (probable but uncertain)
Percy begins at Hogwarts: September 1987 (certain)
Bill finishes at Hogwarts: June 1989 (tentative possibility)
Fred and George begin at Hogwarts: September 1989 (certain)
Charlie finishes at Hogwarts: June 1990 (tentative possibility)
Ron begins at Hogwarts: September 1991 (certain)
…and from there on out we know things pretty well.
Bill 25
Charlie 23
Percy 20
Fred & George 18
Ron 16
Ginny 15