The instruction booklets (or manuals for short) consist of three different elements: text (which relates to the story or to Hyrule's history), images, and gameplay instructions. Let's analyze the relative canonicity of these three aspects of the manuals.
I define "text" to be the text which speaks about the history of Hyrule. Text includes the story at the beginning of the manual, any subsequent text in the manual that refers to the story, and descriptions which include references to the story (example: the ALttP manual's description of the Book of Mudora and the Smithy's Shop).
Now, a big question when analyzing the manual text is: who wrote it? In many other games, a wise sage, living within the same time period of the game, is the author. In the Zelda games, it appears that the text of the manuals was written by: somebody who lives in the same time period as the game (since text refers to game events as though they had happened in the recent past), who has access to Hylian scrolls, Hyrulian legends, and contemporary knowledge. It's not crucially important to know exactly who wrote the manual, as long as we recognize that the text is based on solid historical truths.
In the ALttP manual, the text gives the impression that whoever wrote it was analyzing the Hyrulian culture, but wasn't involved in it. The manual refers to "Hylian scrolls" and "legends" as its sources for history. For contemporary events, however, the author just describes the events without saying where he got the information. The author knows about contemporary events, but is detached from them.
When the author of the text in any manual speaks of contemporary events (i.e. events that happened just before or during the game) he must be reasonably accurate. It's kind of hard to get current events wrong, don't you think? BUT, when the author talks about the history of Hyrule (i.e. events that happened long before the game, before the author's lifetime), his account is more open to innaccuracies and confusion. This is because he might not understand the Hylian scrolls or legends, or because much of Hyrule's history is passed down as old wives' tales.
In relation to images and gameplay instructions, I believe that the text is the most authoritative. Text gives the facts in no uncertain terms. Real-world history is based on text, so I look to the text for the most solid historical truth.
Pictures are valuable because they help us to visualize the Zelda universe. I consider any picture, drawing, etc. in the manuals to be images. (This doesn't count screenshots, obviously, since they are from the game.)
To begin my dissertation on images, consider first what real historians do. Real historians have two things to base their histories on: artifacts uncovered by archaeologists, and written accounts. Sometimes, copies of written accounts contain illustrations, which represent what the author thought something might have been like. These illustrations are entirely fictional, but help one visualize the account described in the text. For example, many versions of the Bible contain illustrations, but the original texts were not accompanied by images.
Consider another real-world example. What did president George Washington really look like? There are several paintings of him in existence, but he looks slightly different in each one. He even looks a little different on the one-dollar bill. All the images we have of him give us an approximate mental picture of what Washington really looked like, but not an exact one.
Apply this to the manuals. We don't have any "real" artifacts to look at
here, only illustrations. We can argue two points of view:
(1) The author of the text is also the artist that drew the pictures. Thus, images should be taken at face value, and be considered as authoritative as text.
(2) Whoever compiled the manual (Nintendo in this case) simply used the text written by somebody else (a Hylian), and then added their own illustrations.
I think that the second point of view has more weight. Consider this: how many historical sources that you know of that include pictures? Whenever a contemporary historical source describes history, it is rarely or never accompanied by pictures. In Hyrule, much of the history is preserved by Hylian scrolls, legends, and the Book of Mudora. Real-world scrolls don't come with detailed color pictures. Pictures come from outside the source, from archaeological discoveries. Since, looking at the manuals alone, we have no archeaological discoveries, I conclude that the original author of Hyrule's history is not the one who drew the pictures.
I offer the following hypothesis to explain the origin of the images. Nintendo, as the agent who is presenting Hyrule's history to us in the form of a game, is the compiler of the manual. They got the text by either cutting and pasting in a contemporary Hyrulian account, or by looking at Hylian scrolls themselves and writing their own account. Because they have special insight into Hyrule, they drew all the images themselves. These images, while not drawn by a Hyrulian, are reasonably accurate because the compiler of the manual can see into Hyrule, an ability which no one else can authoritatively claim to do.
So, assuming that the images were not drawn by a Hyrulian, what does that do to the credibility of the images? First and foremost, while the images may be a very good approximation of what Hyrule is really like, they are open to bias. For example, in the days of the NES, Nintendo tended to base it's manual art on its in-game art. Since, within the game of LoZ, Link looks boxy and squished, Nintendo tried to match this art style by making Link look boxy and squished in the manual. This has the effect of making Link look younger. Many have pointed to Link's appearance in the LoZ manual as evidence that he's only ten or so, but this argument is invalid thanks to the fact that Nintendo was biased in the way that they presented their art. In fact, the text of the manual never gives Link's age: it calls him a "young lad," which could put him anywhere between the ages of 8 and 15.
I conclude, then, that images don't have the same weight that words do. Words are relatively exact and unambiguous, while images are only an approximation of the truth, open to the artist's interpretation and the way in which he wants to present it. Images do have canonical weight, but not as much as text does.
Last of all, manuals contain instructions on how to play the game. Gameplay instructions include such things as what button to press, contrived mechanical operations like Magic Meters or Heart Containers that have no bearing on the story, and any reference to "the game."
Basically, my view on gameplay instructions is that, since they have nothing to do with Hyrule's history, we will lose nothing by totally disregarding them. The only reason that the instructions are there is because Hyrule's history is being presented in the "game" format. Presented in any other format, these gameplay instructions would have no meaning for Hyrule's history. In my written storyline, for example, the fact that you press B to use the sword is irrelevant to the story.
For this reason I consider gameplay instructions to be non-canon. If you strip away all references to gameplay from the Zelda manuals, you are left with a solid core of historical truth. As historians, we are concerned with the history, not with the details on how the games are played.
Table of Contents: