(Note: This is a rolling canon, which is to say that it is not fixed. Although this canon does not claim to be comprehensive, it does aim to provide a foundation for anyone looking for the very best books of American literature. For more information about this canon, see this primer. When we recommend specific editions of works, they are hyperlinked by publisher in parentheses next to the title. When we have found no extant editions sufficient, we have created our own, which are hyperlinked on the name of the work itself.)
(Canon Conflict” is a series of posts that address issues arising from the process of creating the American Canon. These posts are extended versions of heated discussions held among the members of the Canonizing Board.)
A Paradox
The crest of the mountain Forever remains, Forever remains, Though rocks continually fall.
Amazing. Haunting. Canon-worthy.
Those are words I’d use to describe this and the other four works under the heading “Southern Paiute Poetry” in The Library of America’s anthology of Nineteenth Century American Poetry. The anthology includes forty pages of what it terms “American Indian Poetry,” and a lot of it is pretty great. Reading it constituted my first foray into Native American literature outside of assigned middle school English reading, and the high quality really struck me.
My hunt for a comprehensive anthology of Native American literature that contained both poetry and prose began, but I met a series of disappointing roadblocks, and, along the way, I learned a few valuable lessons, the major consequence of which is that none of what we consider “traditional Native American literature” is eligible for inclusion in our Canon.
What is “traditional Native American literature”?
Let’s start with a few disclaimers about the idea of “traditional Native American literature” itself. There is no such thing as the Native American “tradition” in the sense it’s usually meant. When Columbus landed on American soil, there were something like two thousand independent Native American tribes, of which roughly three hundred survive to this day. We cannot treat Native Americans as a homogenous group with a collective tradition. The idea of “Native American” as an identity category begins always already fraught with inconsistencies and colonial assumptions, which become endlessly reified by ongoing white discourse.
The various indigenous tribes did share some things in common certainly, but so did the English colonists and their Spanish colonial contemporaries who were simultaneously invading Mexico. The commonalities actually fall along the same lines. Although Spain and England had different monarchies, they shared comparable (but not identical) governmental structures. Although Spain and England had different national religions, they were both variants of Christianity (although many Protestants and Catholics at the time would have disagreed with this categorization). Although Spain and England used different languages, they both spent extensive time writing things down, an activity almost non-existent in “Native America.”
This last feature of these numerous indigenous tribes stands as the most important for our discussion of the American Canon because we consider the divide between oral literature and written literature a meaningful one. Although I do not disagree per se that oral literature is a form of literature, putting a work of oral literature on the page results in the creation of a new text, no matter how similar to its spoken precursor, and written texts are our exclusive concern for the American Canon. The boundaries of any canon are necessarily arbitrary; that an artwork must be written down to qualify for our Canon could prove its most important categorical distinction. But we must hold to that distinction.
When people speak of “traditional Native American literature,” they usually have in mind prose cosmologies (particularly creation myths) and ritual songs or chants. It’s unclear when works by Native Americans stop counting as “traditional,” but one could posit a reasonable dividing line as the moment between the transcription of formerly oral literature and the creation by Native American writers of texts solely on the page.
Traditional Native American “Poetry”
To start, there are no “traditional” Native American poems. The works I’m talking about are songs, chants, incantations, curative charms, and such. They look like poems on the page, but they were not intended as poems nor treated as poems when composed, sung, and heard.
The quatrain I cited above, the one about the mountain, is a powerful one. It has the effect of a Basho haiku or, as the 1918 boom in interest in Native American songs among the Modernists proved, a work of the Imagist school.
But that’s only how it looks on the page. When sung, these songs had a completely different purpose, impact, and flavor. What we see is a single image, pared of all extraneous matter and quick on the page like a lightning strike. But songs like these were not meant to occupy a few minutes of one’s time. They were sung over and over. Even a translation that accounts for this, reprinting the same lines many times, cannot impress upon the reader this disparity in length. There is a Navajo Night Chant that last upwards of three days. There is no way to demonstrate that on the page. This is just one of the dozen difficulties in representing on the page what these Native American songs were meant to do when sung and how the listener would have apprehended them.
Perhaps more importantly, to quote A. Grove Day, “the main purpose of poetry among the Indians [sic]…was to get hold of the sources of supernatural power, to trap the universal mystery on a net of magical words.” Most of the songs and chants were composed with this goal in mind; that is, the authors (anonymous in almost every case) selected words on the basis of their ability to manifest supernatural power. Native American songs and chants are not the same as Christian hymns, although they share spiritual subject matter. These songs and chants are charms in the literal sense, utilizing words not for personal expression of a spiritual experience or for praising their respective deities but as a means of invoking the power of spirits beyond the realm of human power via the words THEMSELVES. The creator(s) chose the words they believed would best serve as conduits for supernatural power. How can you judge a work in which words are chosen in this manner with something like an Elizabeth Bishop poem?
You can’t. They’re so different that they aren’t even in the same category of art.
Traditional Native American Prose
As I’ve said, one of the things that (almost) all Native American tribes had in common before colonization was that they didn’t write anything down, poetry or prose. When white people began taking down the prose of Native Americans, they were most interested in the stories that were part of their belief systems, particularly cosmologies. Today, most of the traditional Native American prose we have is religious in nature, not just cosmologies but other stories with spiritual significance, what are sometimes collected as “Native American mythology.”
But calling these “mythologies” is misleading because, unlike Greek and Roman mythologies, which are no longer part of contemporary belief systems in Greece and Italy, the cosmologies and “mythic” prose of Native Americans taken down from the 19th century onward still hold currency in many Native American tribes. Because there is no homogenous Native American “religion,” there is no homogeneous approach to traditional Native American spiritual writing. That said, viewing them as dead religious beliefs, as we view the Greco-Roman dramatis personae of gods and goddesses, rejects the possibility that they reflect valid spiritual truths.
Calling them fiction (and assessing them as we would assess any other work of fiction) makes a final, decisive ruling on their validity as a belief system. This would have them occupy similar taxonomic status as American tall tales, such as Paul Bunyan or Pecos Bill. But no one centered their relationship with nature, each other, and the universe under the assumption that some cowboy dude actually tamed a tornado. Reading an Iroquois creation story as a myth delegitimizes it as a potential religious text in the same way that reading Genesis as a myth would do so. One could consistently read all religious texts as fiction if one so chose, but that would be, in my eyes, both beside the point and pretty disrespectful. Even for someone who believes strongly in the exclusionary truth of my religion (and its sacred text), I don’t feel comfortable classifying the contemporary religious texts of other people as fiction.
Aside from the ethical problems that this would raise, I have found that religious texts, if read as literature, rarely make for good art, particularly ones written in prose. If we were to take, for example, the Book of Acts as fiction, we would find it unacceptably repetitive and dry, despite the fact that it contains many supernatural events. Of course, this is because Luke was not attempting to write fiction but to tell the story of events that truly happened. Some religious texts, particularly those written in poetry, do have literary merit, but, because they mostly aim to lead the reader to powerful truths about the nature of god and the human experience, considerations of style (particularly variation in theme, narrative, and sentence structure) are not a priority. I do not deny that there are portions of the Bible, which I believe firmly to be inerrant and directly inspired by the Holy Spirit, that are also good literature, but, tellingly, these are almost exclusively the poetry (Song of Solomon and many, but not all, of the Psalms are the most obvious examples). I am not making a once-and-for-all claim, but religious prose infrequently qualifies as canon-worthy literature if read using the same standards that we use to read secular short stories or novels. This isn’t true of literature written about spiritual matters, such as George Herbert, which is entirely different matter. I am not speaking here about texts that are literary responses to religion but texts that form part of that religion itself.
Personally, I think it is a very colonialist move to read Native American religious writings as fiction, especially if one is not willing to apply that same technique to any religious texts that one holds to be false. Approaching religious texts from a secular perspective is hard, maybe impossible, because it requires so much context to see their merits through the lens of the believer, which is necessary to fully appreciate their power. I don’t think it’s reasonable for someone outside of the religion to expect to read these texts on their own terms, even if that outsider could do so while simultaneously believing them erroneous. It’s just too hard of a mountain to scale, and the reward would at best have to be an individual one since the experience would be practically impossible to communicate to another outsider.
Ultimately, taking Native American spiritual prose as fiction is something I’m not willing to do, and fiction is the only prose that is eligible for inclusion in our American Canon. Therefore, no Native American religious writings (or spiritual texts from any other religion) will be included. I think this is the only respectful way to approach the writings of other religions, and I am strongly suspicious of the integrity of aesthetic evaluations of contemporary religious writings in any form, no matter how well-intentioned they may be.
Making a canon is an exercise in audacity. Who are we to rule over the entire history of American literature? Let me explain.
When I was in middle school, I listened toMatilda on audiobook. I don’t really remember why. What I do remember is the thrill I got when I reached the part of the book where Matilda’s local librarian recommends what ends up being a series of twenty books, which you can find in full here. I can still hear the person who did the audiobook reciting them in almost a lyric tone: The Old Man and the Sea by Ernest Hemingway, The Invisible Man by H.G. Wells…
As far as I know, this was my first interaction with a canon, and I was immediately hooked for life. As a fledging reader of “adult” literature, it was refreshing to have someone outside of my schoolteachers (whose taste I intensely distrusted) tell me which books I needed to read. This was before searching for “100 Best Novels” on the Internet was a possibility for me. Matilda’s reading list became my reading list, and, although I eventually moved on from it, it remains an important moment in my development as a reader.
Dahl wasn’t creating an actual canon when he made this list of twenty books of course. Matilda’s list contains only novels, and three of them are by Charles Dickens. It was just a list of books he felt like the librarian character might reasonably recommend to a precocious elementary schooler. But it’s also clear that Dahl considers these worthy texts, hand-picked by him for their high aesthetic quality. What mattered wasn’t that the list claimed to be comprehensive but that it gave me a starting point in the impossible project of figuring out which books I needed to read to become a well-read young adult. It was a watershed moment for me, and I can proudly say that I figured out that I didn’t need to read Mary Webb before it was too late.
Not everyone gets as excited over canons as I did then (and still do), but they are everywhere, and they are a necessity. As Harold Bloom says insightfully in his very problematic introduction to his equally problematic book The Western Canon, “We possess the Canon because we are mortal and also rather belated. There is only so much time, and time must have a stop, while there is more to read than there ever was before…we are in the pragmatic dilemma of excluding something else each time we read or reread extensively.”
There are thousands and thousands of books. No one could possibly begin to read even all the books written in one language. Without the help of well-read and dedicated literary scholars and layfolk, a reader would be paralyzed by possibilities. Although no one canon can be taken as definitive, cross-referencing them allows a reader to get a sense of what other people think is good and thus a foothold in the overwhelming world of existing literature.
If your goal is to only read good books, consulting as many canons as you possibly can before choosing your next one would be the smartest avenue to ensure you don’t waste a page of your time. If, like me, your goal is to have the fullest possible appreciation of the history of Western literature, you need canons to help you build the foundation for continued lifetime reading. If you hope to deeply appreciate the achievement of Edith Wharton, you need to have read Jane Austen; of Allen Ginsburg, you need to have read Walt Whitman; of Edward Taylor, you need to have read John Donne. In Will Self’s words, “To understand influence, or the relation of literary genres to one another you need to have absorbed sufficient quantities of text to make intuitive comparisons—while in order to lose yourself in a given work, you need to be able to get far enough into it in the first place.”
The practical nature of canons has huge implications, although they aren’t as huge now as they were before the advent of the digital age. When canons are created, whether implied or explicit ones, they mark some works as worthy of being saved and imply that the rest can safely be allowed to fade into obscurity. This has real life consequences, as many of these non-canonized books do, in fact, fall out of print and thus off of the reading lists of most. The act of recovering “lost texts” is a major component of contemporary literary studies, and I’m happy to say that the existence of things like Forgotten Classics and Google Books has made the process of “losing” texts much less likely. Amazon has forever changed the availability of books. Where you used to be restricted by what your local library and bookstores had to offer, you are now restricted only by what you can afford. I rarely find a book that I’d like to read that I *can’t* buy, although plenty of the more obscure ones are too expensive because they have been out of print for so long.
Still, the availability of books is one of the most important consequences of canonizing. Let me say clearly that everyone who talks about the quality of books contributes in their own microscopic way to the avalanche of assessment that ultimately determines whether this or that book will go into “the Canon” (or, more accurately, into one of many canons, including ours). If you’ve ever recommended a book to me, you have played a part in the creation of our American Canon, and our canon will doubtlessly influence as many as three people, who will in turn influence others.
Of course, most people don’t create explicit canons. However, implied canons are commonplace. Articles entitled “50 Best Books to Read on the Subway” or the syllabus for an Introduction to American Literature imply the existence of a canon, however small in scope. All literary anthologies are canons, but they necessarily favor shorter works, such as poems and short stories, and so do not include novels (or more than a couple plays at most). None of these claim to be true canons, but they are frequently where we turn to decide which books to read.
Part of the hesitation in making an explicit canon is that they tend not to age very well. The determination as to who counts as a major writer changes even from decade to decade, and many of the greatest American writers were not considered canonical until long after their deaths. And canonizers are truly guessing when it comes to literature written over the last 10-15 years or so. As such, we have found relatively few explicit canons, especially in print.
But the creation of these canons is important. Although no one canon will ever be comprehensive, the existence of multiple canons allows for cross-referencing, giving readers a better chance at figuring out which works are worth taking their time to track down. I don’t believe there can ever be too many explicit canons. In creating this one, we have frequently consulted the ones we could find in print (Ellen Moers, Harold Bloom), but the paucity of explicit canons has become even more apparent to us. It’s telling that Bloom’s canon, one of the twentieth century’s most influential instances of tastemaking, is frequently cited on book jackets to entice readers. Over the last week alone, I have run into two books that are advertised as having Bloom’s seal of approval. This is in part because Bloom is a highly respected literary critic, but a major contributing factor is that he is one of the few people to come out and say exactly which works are canonical and which aren’t. It provides a level of validity to his quotables that few critics have. And it means they we are relying way too heavily on his canon, which is a big problem since his canon is…well, a big problem.
I believe there ought to be a proliferation of canons. The more, the better. That gives potential readers, like middle school Zack Rearick, the best chance at getting to the best books. To that end, we have made this American Canon, knowing that it will be forever flawed, despite our sincere and tireless efforts.
I will add that I think our Canon is particularly useful because it is online rather than in print. The benefit of our canon being digital rather than in print is that it can be a rolling canon, with new works forever being added, rather than forever static on the page, as those in print must be. This gives us the ability to change it as our ideas about which writers are canonical changes. It will forever be in flux, but that’s the way the amorphous American Canon that exists in all of our minds works anyway.
A canon is a collection of artworks deemed to have high aesthetic value. Canons are usually implied rather than written down, although things like anthologies and syllabi create what are essentially canons. Most explicit canons are fixed objects; they are created once and then remain static. However, because this canon exists in the digital world, it can be continuously updated.
What does a canon do?
Both good and bad things.
There are thousands and thousands of books. No one could possibly begin to read even all the books from a single country. Canons help decide which works to keep and which to let fade into obscurity. Without the help of well-read and dedicated literary scholars and layfolk, a reader would be paralyzed by possibilities. Although no one canon can be taken as definitive, cross-referencing them allows a reader to get a sense of what other people think is good and thus a foothold in the overwhelming world of existing literature.
Canons are also weighty ideological tools, and there is a long history of them being used to reify the oppression of non-privileged groups. As a form, the canon inherently implies a degree of objective truth, claiming that the works in it are somehow empirically good and that their quality ought to be universally recognized. This can easily function as a means of normalizing the literature of whatever group is in power in a given society while dismissing works by writers who do not belong to that group. Although we have consciously worked against this tendency, anyone in America is inextricably bound in a social fabric that is both racist and sexist (just for starters), and it’s impossible for that not to play into the creation of a canon.
What is in our American Canon?
This is a literary canon, and it includes works of poetry, fiction, and drama by writers born or raised in America. All works that satisfy those criteria are eligible. We do not uphold the high art/low art binary.
What is not in our American Canon?
First off, anything that isn’t literature. Americans have succeeded in producing great writing in a variety of non-literary categories, such as essays, autobiographies, political treatises, literary criticism, and so on. These are not eligible for inclusion in our canon. Categories of art that are made of words but are not literature, such as song lyrics and screenplays, are also not included on the basis that they are different genres and that calling them “literature” implies a hierarchization of one art category over others.
Here are some other things that are not in our American Canon.
Children’s literature: As I’ve said, we do not uphold the high art/low art binary. Children’s literature absolutely is literature. However, it is a kind of literature whose audience and conventions are so different that it would require an entirely separate canon to do it justice. This is purely an issue of feasibility.
Graphic novels: Graphic novels are also literature, but none of the people on our Canonizing Board are well-read enough in the genre to have any ethos to include them. This is a limitation of the canonizers, not a value judgment of the graphic novel as a form.
Young Adult literature: The boundary between children’s literature and the works in our canon may be (relatively) straightforward, but the line between a Young Adult work and a work of “adult” literature has always been murky. We don’t regard any Young Adult works as fundamentally ineligible, but we do recognize that the genre has its own audience and conventions. There is no hard line, so it’s a case-by-case evaluation, but we do want to note that exclusion from this canon does not necessarily reflect a lack of quality in a Young Adult work. A Young Adult canon would look very different from the one we’re making, and, like a children’s literature canon, it would have a merit of its own.
Religious Texts: There is no safe ground in assessing the spiritual writings of any religion, and we will not be doing that. As a person with strong religious beliefs, I believe that judging works of these kinds, whether they are sacred texts or aspects of ritual/liturgy, is beside the point at best and disrespectful at worst.
What about popular literature?
Although it can be useful as a term in personal conversation, “popular literature” isn’t a tenable as a true category distinction.
A better word for what people typically mean when they talk about popular literature would be “genre literature.” Some examples of genre literature are: mysteries, romance novels, true crime, science fiction, and horror. The idea of genre literature as a discrete category comes from the assumption that it uses different conventions than “regular” literature. It perhaps allows, for example, for its writing to be more formulaic with the promise that it will make up for the lack of originality in plot structure in return for success on the terms of the individual genre. It’s okay for a mystery novel to use the same old narrative structure that all of them use if the mystery itself is clever and keeps you guessing.
We don’t necessarily believe that this is true of genre literature or that “genre literature” is even a real literary category, but we do recognize that some well-regarded works in the above genres rely on this kind of leeway for their success. In order for a work of genre literature to make it into our canon, however, it needs to succeed not just on the terms of its genre but on the terms we set for all works.
Why make a canon?
Implied canons are commonplace. Articles entitled “50 Best Books to Read on the Subway” or the syllabus for an Introduction to American Literature class all imply the construction of a canon. All literary anthologies are canons, but they necessarily favor shorter works, such as poems and short stories, and so they do not include novels (or more than a couple plays at most). None of these claim to be true canons.
Part of the hesitation in making a canon is that they tend not to age very well. The determination as to who counts as a major writer changes even from decade to decade, and many of the greatest American writers were not considered canonical until long after their deaths. And canonizers are truly guessing when it comes to literature written over the last 10-15 years. As such, we have found relatively few explicit canons, especially in print.
But the creation of these canons is important. Although no one canon will ever be comprehensive, the existence of multiple canons allows for cross-referencing, giving readers a better chance at figuring out which works are worth taking their time to track down and read. I don’t believe there can ever be too many explicit canons. In creating this one, we have frequently consulted the ones we could find in print (Ellen Moers, Harold Bloom), but the paucity of explicit canons has become even more apparent to us. Try Googling “works in the American canon,” and you’ll see what I mean. We believe there ought to be a proliferation of canons. The more, the better. To that end, we have made this one.
The benefit of our canon being digital is that it can be a rolling canon, with new works forever being added. This gives it the ability to change as our ideas about which writers are canonical changes. It will forever be in flux, but that’s the way the canons ought to work.
Any other notes?
A few, yes.
-We have made an exception to the “born and/or raised in America” rule for the first couple of generations of (white) American writers, especially those (such as Susanna Wright and Anne Bradstreet) who moved to colonial America in their 20s. It seemed unfair to exclude them when America was still in the process of becoming its own nation. This is an arbitrary distinction, and it arbitrarily ends with the birth of Phyllis Wheatley in 1753 under the assumption that if someone working against the disadvantages she had to face could write canonical poetry, then it was no longer necessary to make excuses for everyone else.
-Works are not eligible for canonization until a minimum of three years after their publication in order to give us an appropriate amount of time to reflect on them.
-When it comes to a work that consists of selections of an author, by which I mean something like a Selected Poems or a Selected Stories, not all selections are created equally. Whenever possible, we have indicated which selected text by which editor best represents the quality of the author. We have done the same when we felt that there is one published version of a text that is better than the other versions of it, such as Cristanne Miller’s version of Emily Dickinson’s poems.
Honestly, the pre-revolutionary era wasn’t a dynamite time for American literature. You will find only four books in our Canon written before the beginning of the Revolutionary War. Exclusively poetry as well, since the first American novel wasn’t written until 1789, and the first truly “American” play wasn’t written until 1787 (I won’t go into the details on that distinction). Pre-revolutionary American poetry pretty much all sounds the same.
The major exceptions to this rule in America are Anne Bradstreet and Edward Taylor, both seventeenth-century poets. These two poets have unique voices, and they don’t have that monotonous eighteenth-century feel that makes most poetry between Pope and the Romantics sound so paint-by-the-numbers. The major strength of eighteenth-century American literature rested not in its sincerity but its satire. I could make a selection of the best humorous American verse of the eighteenth-century, from broadsides to mock epics. Such a selection, which would include poems like Robert Bolling’s “Neante,” Jonathan Odell’s “Word to Congress,” and, of course, Joel’s Barlow’s “The Hasty Pudding,” would fit easily into our Canon. But anthologies aren’t eligible for inclusion. And no one writer composed enough high quality satirical poetry to make it in.
Anyway. The two other authors in our Canon from the pre-revolutionary period (Wright and Phyllis Wheatley) are quite characteristically eighteenth century, and readers can get a pretty comprehensive feel for what non-comic American poetry from the era sounded like by reading them alone. Aside from being the two highest quality American poets of their period, they also happened to have relatively small outputs, which wasn’t necessarily a factor in choosing to canonize them but which certainty doesn’t hurt because even THEIR poetry sounds mostly the same. It’s kind of a happy coincidence that using our Canon as a guide allows a reader to get the full eighteenth-century feel while only have to read 46 total poems.
Wright was probably the American best poet of the first half of the eighteenth century, but she’s a little known figure, in part because she wrote so few poems and in part because eighteenth-century poetry is as out of vogue now as an style of poetry can possibly be. She’s worth reading, but you can only find her poetry in an obscure compilation entitled Milcah Martha Moore’s Book, which also contains works by other lesser talents. And this thirty-poem selection doesn’t even include “To Eliza Norris—at Fairhill,” an unusually feminist epistolary poem from 1750 that is one of her best works. Taken together, these 31 poems form the entirety of Wright’s extant corpus. She’s worth reading, but it won’t be an easy task without access to a university library.
blood bank blues 