Reading the Norton Anthologies: Databases, Canons, and “Careers”

What type of canon do the Norton anthologies of literature construct? And how has that canon changed over time? These questions are somewhat unusual for humanists in that their answers could be framed not in a syllogism or thesis, but rather in the forms of the list and the table—and extensive ones at that.

We wanted, first, a way to see who goes in and who goes out of this canon that we so often teach from. In a sense, this project is as much about pedagogy as it is about literary criticism: We have no investment in the Nortons as being representative of “The Canon,” but rather see them as a medium through which undergraduate and graduate students of literature encounter major works and begin to formulate their ideas about the literary field. Following John Guillory, the Nortons seem to be one of the primary means by which the cultural capital of literature gets distributed and reinscribed within the university. How, then, has this medium—of texts and canons that inform courses, students, and scholars—changed along with literary criticism over the past half-century?

Our team built a database containing every work and excerpt featured in the Norton Anthologies that we have studied so far, with room to grow for those that remain. This allows us to easily see what we have been thinking of as the “careers” of both authors and individual works over time. How, for instance, have the works selected to represent Milton changed over time? When was Margaret Atwood first added to any Norton Anthology? Which poems represent Langston

Hughes in the Anthology of Poetry? Are they different from those that represent him in the anthologies of World, American, or African-American literature? What proportion of authors in the anthologies are women, and how has that changed over the last fifty years? Which authors have been cut from the anthologies? And which authors or works replaced them?

In order to answer these questions about the people and ideas admitted to these canons, we needed to restructure the data from the Nortons' tables of contents into a format that could be queried and would reveal the relationships among many different works and selections from works across a variety of different manifestations. The problems this poses from the perspective of data structure are easiest to think through with a major author like Shakespeare, who appears in every anthology relevant to his work. We need to know which works were selected to represent Shakespeare in each edition of every anthology in which his works appear. For example, which Shakespeare plays appeared in the first edition of the Norton Anthology of English Literature, and which in next eight? How do those selections compare to the ways in which Shakespeare is represented in the anthologies of Drama, Poetry, Western, and World literature, across each of their individual editions?

We achieved this by creating a structure based on a set of n-deep parent-child relationships and a number of many-to-many connections, using a web interface for parallel data entry and validation between several collaborators simultaneously. Using this structure and the Shakespeare example above, King Lear becomes a “child” of Shakespeare, and Lear's “Blow,

winds ... !” speech from Act III a child of King Lear. Because of this nesting, we can then measure not only

which anthologies any work of Shakespeare's appears in, but, of those, which contain King Lear in full, which only have the excerpt of Lear's speech, and which contain other parts of Lear. This allows us to be more precise about the ways in which we count authors' presence and absence across all of the anthologies that this project will eventually consider. This data entry interface was built with the Django, an open-source web application framework, and the code is publicly accessible on GitHub. The database will be demonstrated and described in more detail during the presentation.

Having produced new editions and types of anthologies semi-regularly for more than fifty years, W.W. Norton & Company has been in the business of binding literary canons longer than anyone else still publishing. Since M.H. Abrams edited the first Norton Anthology of English Literature in 1964, numerous editions and kinds of anthologies have followed: The Norton Anthology of American Literature, World Literature, Western Literature, Poetry, Drama, Theory and Criticism, Short Fiction, Literature by Women, African American Literature, Latino Literature, Jewish American Literature, etc.

Much can be learned about the ways in which the Nortons were designed from these titles alone. First, the largest anthologies are defined by both geography and a linear temporality influenced by conventions of periodization. The Norton Anthology of World Literature lays claim to it all, from Afghanistan to Zimbabwe, and from Gilgamesh to Orhan Pamuk. Western Literature claims a smaller (if vaguer) part of the world, and English and American literature focus on national literatures, including postcolonial and expatriate writers within the bounds of the nation-concept. While these geographic anthologies are ostensibly genre-agnostic, others are genre-specific (Poetry, Drama, Short Fiction). And the last type focuses on writing by and about writers of a specific gender (By Women), ethnicity (Latino), or religion (Jewish).

One of the premises we read as implicit in the Norton's design, then, is that some authors and works become significant enough to include only in specific contexts. Making it into the World Literature anthology seems to denote significance at a greater level than inclusion in the Western Literature anthology alone would. Likewise, seeing a writer anthologized in Short Fiction but not World Literature seems to imply a significance limited to that literary form. Canon formation has always relied on a logic of a ranking or tiering, and the pool of authors and texts against which a given work “competes” is greatest at the largest scale of population. We argue that the geographically bound Nortons can be read in such a way that they imply a hierarchy even among the canons the anthologies already connote.

Of course, all of the decisions we measure across these many tables of contents are underwritten by a human element. Many practical and historical factors that exist at a slant to the question of a work's “canon-icity” attend the production of an anthology that stretches to more than 6,000 pages, serves tens of thousands of instructors and students, and our analysis attempts to account for these factors. A quantitative approach is necessary but not sufficient to read the ways in which the Nortons have represented and continue to represent works that, taken together, lay claim to the status of a national, generic, or global literature. Two key examples of the incommensurability of the form to its implicit claims: Because of its length, the novel is poorly served by the anthology form. Some shorter novels and novellas do get anthologized. But, more often than not, writers who are primarily known as novelists are represented by a single short story, or an excerpt from a novel. The second overarching practicality is the influence of authorial estates and the cost of printing rights, especially for 20th and 21st century authors.

As a way of approaching these institutional and qualitative questions, we have begun a set of interviews with Martin Puchner, the current general editor of the World and Western anthologies, and will discuss some of his insights into the decision-making process in the presentation. Among these include the influence of instructor and student surveys on texts that get selected, the impact of rights costs on the texts that get chosen for a given author, and the place of editorial intervention in relation to these powerful practicalities.

Our database and the attendant institutional research on the Nortons as the product of both scholarly editing and the demands of the market allow us to see the trajectories not only of individual authors and works, but broader trends of inclusion and exclusion in the Norton's canon. By gathering data about both the works and the authors who wrote them, we reveal the ways in which the Norton has responded to the expansion of the literary canon, growing in size while simultaneously giving a greater share of its pages to authors and ideas that would not have been considered canonical 1962. In the process, we find authors whose literary reputation has waxed or waned (or both); those whose names have been a constant presence, but whose representative works have dramatically changed; those who were slated for canonization but never “made it;” and those who have arrived late but seem to be here to stay.

Like many of the Stanford Literary Lab's projects, “Reading Norton Anthologies” operates at several scales at once. We are interested both in individual texts and authors, as well as broader patterns of representation and contextualization within the confines of this object that occupies liminal spaces between statement and syllabus, and between the market and the canon.

Bibliography
Bausch, R., ed. (2015). The Norton Anthology of Short Fiction. Eighth. W.W. Norton & Company.

Baym, N., ed. (2012). The Norton Anthology of American Literature. Eighth. W.W. Norton & Company.

Bourdieu, P. (1993). The Field of Cultural Production. Columbia University Press.

Ferguson, M., Salter, M. J., and Stallworthy, J., eds. (2005) The Norton Anthology of Poetry. Fifth. W.W. Norton & Company.

Gates Jr, H. L. (2014) The Norton Anthology of African American Literature. Third. W.W. Norton & Company.

Gilbert, S., and Gubar, S., eds. (2007) The Norton Anthology of Literature by Women: The Tradition in English. W.W. Norton & Company.

Greenblatt, S., ed. (2012) The Norton Anthology of English Literature. Ninth. W.W. Norton & Company,

Guillory, J. (1995). Cultrural Capital: The Problem of Literary Canon Formation. University of Chicago Press,

Puchner, M., ed.(2014) The Norton Anthology of Western Literature. 9th ed. W.W. Norton & Company.

Puchner, M., ed. (2012). The Norton Antholog of World Literature. 3rd ed. W.W. Norton & Company.

Shakespeare, W. (2005). King Lear. Edited by George Hunter. Penguin Classics. Penguin.

Stavans, I., ed. (2011) The Norton Anthology of Latino Literature. First. W.W. Norton & Company.