Exhuming the Mastodon


“Let us tenderly and kindly cherish, therefore, the means of knowledge. Let us dare to read, think, speak, and write.”

—John Adams

As a cultural historian, and one involved in rethinking graduate education, the notion of pathways is resonant in obvious ways. We are heirs to a tradition of valuing archives that are arranged synchronically and chronologically (classes, curricula (L. to run), and credentials) to effect a set of knowledge outputs and practices—the educated individual, critically forged and capable. That person extends the means and ends. So, John Adams, thanks.

But what happens when those means clot or forestall the impulse to dare and act in language—when the pathways become sclerotic and unnecessarily difficult? I’m thinking, for the moment, of the dissertation as we’ve inherited it from the nineteenth century. It takes the form of a thesis, but really a book, chaptered, indexed, bound. It must be “defended,” in the form of an oral meeting that theoretically works as an opportunity to counter and call bullshit on written material that can cloak error or ambiguity in its formal, officializing guise of print. The defense completes the delivery of new knowledge, by the newly “minted” scholar.

We might view it as a kind of curtain lifting, not unlike the iconic Charles Wilson Peale, in his self-portrait as gatekeeper to the objects of knowledge: “The Artist in His Museum,” 1822.


Since 1822, the museum of scholarly production has advanced through a few more chambers, but the performative and architecture are basically the same. Of late, we then take the text product, make it a codex via arbitrary formatting, and then contract with Proquest to digitize it, make it available on the Internet (not open-access, but close), and then usually provide it to the degree-granting institution’s library to archive. Many humanities students have begun to choose to forego publication at the moment of credentialing, for fear that they might be precluding their pathway not into “knowledge” but into the publication systems that market knowledge—academic presses embedded themselves in a shrinking trade in knowledge commodities.

But that access issue is almost the least of the problems with the PATHWAY of doctoral credentialing. It’s the form itself. That culminating experience is the place where the “running” in curriculum hits obstacles, stalls, crashes, burns, evaporates. Perhaps the digital offers ways to dredge the riverbed and make that knowledge system much more fertile.

I’d like to see dissertations that continue the curriculum—that are, as the MLA and AHA are making preliminary steps toward advocating for, process projects. They would arise out of a richer mix of inputs than an advisor and several other co-advisors to include communities of intra- and inter-institutional faculty and students. They would break down the wall between institutional knowledge and its publics by inviting widespread access to the project as a work in process. Graduate faculties would be configured to critique and follow real-time progress rather than dangerously episodic check-ins. The archive too would not be spatially remote, giving the student little excuse to get “lost.” Indeed, the line between reading and curating would be forever blurred. And indeed the metaphor of “defense” becomes unnecessary, since that need to complement the discrete bounded knowledge-output, the one we must “suspect” of flaws, has always and already been produced through an engagement with multiple voices and assessments.

So rather than Peale in his museum, we’d have the dissertation as collaborative dig, pulling forth, over time.  As in:


Also Charles Wilson Peale, this is an image of “The Exhumation of the Mastodon, 1805-08.”  Note the temporality Peale foregrounds, the wheel in motion, the dating over a three year period—this is a rendering of process. And it’s a process of manufacturing knowledge collaboratively, over time. It is a lesson from the past about how not to bury things.

Traveling the Landscape of the Book


Here is what I want to ask: Books provide us paths through the world. Can the world provide us paths through books? Or, more appropriately, what can the world itself tell us about how we should sprint beyond the book?

So let me digress with a comment on reading and books as sensory experiences. Books are read; text is visual. Nearly every assumption built into the imaginary of books depends on reading and sight. Too often we often fail to appreciate the breadth and depth of books in terms of their sensory evocation, much less how we might experience what is within. Of course, books themselves are tactile. Old books, in particular, have a certain smell—for the historian, opening an old book is akin to the experience of that new car smell. Ahhh, yes, the mustiness of an old library. I fondly recall the reddish hue of the archives that adhered to my white gloves. More typically, we think of the senses in terms of the sensory experiences evoked by a book, a petit madeleine, chocolate, or the smell of baked bread in a Bret Easton Ellis novel. Do these evocations go only one way, from the book to the imagination to the senses? Can we reverse that path, bring the physical experience—of the senses, of the material, to the book? Wouldn’t that enhance our experience? I am thinking presently of how a sound historian has used the digital humanities to evoke the auditory sensibilities of early 20th century New York City. Our senses might offer entirely new paths into and through literature, allowing us to move beyond the book, envisioning a multisensory experience.

Likewise, reading itself is not just a literal act of moving eyes over text and processing that text, but it has become a metaphor for the production of knowledge itself. We do more than read text. We also “read” landscape, images, and environment. And yet, this imagining still elevates reading above not only the senses but also above the material world with all its depth and expressions. Of course, books have never been isolated from the world, but discussions of the book usually imagine them as knowledge systems all but closed from anything outside the human imagination. I would argue that imagination is shaped by social and historical experience. Rather than imagine books as blazing paths through our minds, perhaps we should look to social and historical experiences—to the materiality of everyday experience—to find ways of imagining paths through books themselves.

Consider how the landscape can be exposed, confronted, and expressed to create a path through a book, one where the materiality of space helps us find the logic of a book, or perhaps the materiality of experience—perhaps the work of an aged craft iron worker, whose voice and talents reveal narrative. What about hyper-textual approaches to the book, where links structure our reading—connections to the material, the ephemeral, the momentary?

I want a world of non-textual paths, generated by the materiality of the world, that structures our paths through individual books, libraries of books, or literatures. I don’t want to abandon the narrative, the story, the text, the argument in favor of the archival. Rather, I want a connectedness between book and materiality of experience that transforms not only our reading of the world but also our reading of the book.

As we sprint beyond the book, let’s not race toward the book as an individuated form (and I’m not advocating abandoning authorship) without connection to other books or to the materiality of experience. Rather, lets build something that is interlaced with the world, with the materiality of experience, including especially a richer sensory experience. Let’s create books that are meta-analytical and meta-experiential.

Creating Multiple Adaptive Paths Through the Book


A traditional book encourages the reader to take a direct path from beginning to end. Pages are arranged in a fixed order and numbered. But there are many cases where a book is not read in the order of its pages. Imagine Mary, who consults her textbook to understand a particular physics principle. She looks up the name of the principle in the index, and then turns directly to that page. After reading the description, Mary realizes that she doesn’t understand. She flips the pages to earlier in the textbook where she remembers a key related concept was first introduced.

Instructional texts are not the only contexts where you might want to navigate non-linearly. James is reading a crime novel. He reads a few pages, and then, as he always does, flips to the last chapter to see how the story ends. He finishes reading the book, remembers a part that he particularly liked, and then flips back to re-read it.

Digital technologies have opened up new possibilities for facilitating the way we navigate through texts. If Mary were reading a digital book, a search for the concept she does not understand might return a variety of relevant information: where the concept is first explained, what she needs to know to understand the new concept, and where that concept is later used in the text. The book could recommend, based on her knowledge, which content she should view first. Using hyperlinks, it is now possible to easily jump between different parts of a book, and using adaptive recommendations, a system can indicate which parts of a book are most relevant to a particular reader.

If James were reading a digital book, the possibilities of new technology suggest a more interactive and more personalized reading experience. The author could indicate multiple ways a book could be read to suit different preferences. For James, the book could be automatically reordered to present the final chapter first. Based on James’ reading behavior, the book could automatically infer which parts James liked the best, and link back to those parts at the end of the book.

To facilitate multiple paths through a book, there are several considerations related to technology and user experience design: semantic indexing, designing for non-linear navigation, making intelligent recommendations, and adaptive reconfigurations.

Semantic indexing. At a minimum, the content of the book needs to be indexed (either through natural language processing technologies or crowdsourcing) so that semantically meaningful links between different parts of the book can be made.

Designing for non-linear navigation. With non-linear navigation comes the need to design the book’s interface to support the user in taking multiple views of the text. Side-by-side split screen views should be facilitated so students can make direct comparisons between content. Reading history should be saved so the reader does not lose the page they were interested in, and can retrace their steps through the book if necessary.

Making intelligent recommendations. As the number of navigation paths increase, the reader may need recommendations for which path to view next. The quality of these recommendations depends on how effectively the book can construct a reader profile, interpret reading history, and understand how its contents can meet the reader’s needs.

Adaptive reconfigurations. For an engaging reading experience, a book could adaptively reconfigure its contents based on reader reactions and preferences. Using different navigation paths, writers could author multiple reading experiences within a single book, tailored toward different profiles.

One final consideration in this discussion is ensuring that these adaptive technologies support how readers perceive their own needs. In general, users want to maintain control when interacting with technologies. For this reason, recommendations may be better received than adaptive reconfigurations. Readers will want to be able to understand how the book is being reconfigured and potentially select their own path. As adaptive technologies become more sophisticated, the goal should be to enable the reader to make more informed choices about how and what they read. 

Setting the Demons Loose


Many of the interventions offered to book culture or to what you could call the reading-writing economy are currently coming from start-ups, entities described by one entrepreneur-cum-academic as organizations formed to search for a business model. As such, they may fail to find that business model even though they succeed at finding outcomes. One that I worked with closely, Small Demons, found that fate. What we did find, while not a business model, is a tacit cultural map, one formed by the culturally resonant details set jewel-like within books, one which, when illuminated by a kind of UV light, glows so as to allow one to navigate through the storyverse—our term at Small Demons for the universe that exists parallel to the “In Real Life” one in which we live. A Borgesian world, then, a planet-like library with paths that may be traversed to allow a richer life for us humans.

The company created a taxonomy of keywords grouped as persons (fictional and/or real), places (fictional and/or real), and things (encompassing songs, movies, other books, events, sports, drugs, foodstuffs, cars, and so forth) and managed to use entity extraction software  to highlight those words in books, collect useful information about them, and link them to one another. One might then travel from Nick Hornby’s High Fidelity to Haruki Murakami’s Kafka on the Shore via Prince’s “Little Red Corvette.” Unlike the typical recommendation engines explored by C. Max Magee, these paths are not designed to lead from one recommended cultural artifact to the next but merely to offer an alternative mode of browsing. However, much like those services, it does offer signal amidst the noise, a heat map that offers clues to those artifacts, much like how surveying the restaurants in a urban plaza allows a prospective diner to gauge the vibe of each restaurant, see how the diners are dressed, the music playing, check out the decor.

In this respect, what Small Demons envisioned is books not just referring to one another but to entire cultural tapestries, situating these narratives within and around all other narratives, actual and imagined. From a commercial standpoint books transcend their ghetto, without abandoning their edges, they become permeable—which is in fact what they’ve always been. As such, the books become more truly themselves. As Rick Joyce, Chief Marketing Officer at Perseus, a consumer books company, likes to remark: “There are lots of books about shoes, but no shoes about books.” Books, by their very nature, contain worlds.

Now, how might Small Demons live on, not as a business but as a vision? During our existence what became clear was that there was an intense appetite amongst some (though by no means all) of the people who visited the site to actively participate, not just marvel (or frown). All the data we generated was generated in-house via automated entity extraction and a small group of editors tweaking the data. Users wanted to add data, both stuff that the computers missed and stuff the computers couldn’t ascertain. We could tell you that Dewar’s appeared in a book, but not who drank it, and what the role of the whiskey-drinking was in the plot. Was the protagonist drowning his sorrows? Was it spiked? Did she order Glenmorangie and was told nope, all we’ve got is Dewar’s? And so forth. As Erin Walker wrote, books are props in people’s lives, and so are the details within books, and people want to share those details, just as they like to share the books that contain them.

So if we are going to create tools to foster and support that impulse, the key thing will be to build into the system from the beginning the ability for users to add, amend, clarify, correct, and connect details they themselves see. We were not unaware of this need, we just didn’t move quickly enough to respond to it, and ran out of resources before we could deploy those tools.

Further to this principle, this data—from both an output and input standpoint—should live on the entire web, not just within the site or app. In other words, a read-write API. Again, this was something we were aware of, as there was a real appetite from web media companies large and small to integrate our data into their user experience, interest from libraries, interest from geo-location apps, interest from e-commerce retailers, from textbook publishers. But we ran out of time, in part because we didn’t prioritize it early enough. From a revenue-generating standpoint, this appetite for the API is clearly a major opportunity, if not the major opportunity, and would apply both to a for-profit or nonprofit entity.

That said, if it were a nonprofit it would be particularly wise to be aware of the larger context of linked open data. In other words, it should play well with others. Just like books do.

Following the Path from Book to Book


Some rights reserved by Walt Stoneburner

The question I get asked most often by strangers when they find out what I do: “What should I read next?”

The question is asked eagerly, and yet we are supposed to have solved this problem by now through the power of algorithms that ingest reader habits and learn reader behaviors and deliver book recommendations precisely calibrated to sate reader hungers.

Are these algorithms giving me the kind of life-changing book recommendation that I have received from other readers from time to time?

Is technology helping readers find better paths from book to book, with fewer false starts and pitfalls and more transformative and transporting experiences along the way?

The best book recommendation engine is the knowledgeable clerk at a well-stocked, well-curated independent bookstore. To this recommender you verbally input the last few books you read and liked, and she outputs a title, physically handing you the book which you can buy and read alongside a cup of coffee in the café next door.

This recommendation engine has been replicated in the online space via the very low-tech Biblioracle, an occasional feature of magazine themorningnews.org. In this feature, author John Warner, the son of an independent bookstore owner, gives bespoke recommendations to online commenters. They input the last five titles they read and enjoyed, he spits out a recommendation. To this eye, his recommendations are quite good.


Like the real-world experience it replicates, however, it is not scalable.

The question that I get asked so fervently from time to time—“What should I read next?”—is surprisingly fraught. Books represent a large investment for readers in money and especially time and emotional energy. Acquiring a book and investing the time to read 25 or 50 or 100 pages only to cast it aside is a souring experience, maybe enough to sour certain readers on reading entirely.

The stakes are high.

Part of Amazon’s business model hinges on the notion that it can mine your behavior to suggest products—for our purpose, books—that you will like and want to read.

In the real world space, this function is served by the “featured” front table in the bookstore, or by the books face-out on the shelves.

But these efforts are laden with commercial conflicts that seem bound to get in the way of providing a useful recommendation.

Publishers and bookstores engage in “cooperative advertising” by which publishers pay bookstores to secure prime shelf space and placement on front tables.

Amazon engages in similar practices, with promotion in its online bookstore often contingent on payments from publishers. Whether or not these considerations come into play with regard to Amazon’s book recommendations, they are opaque to the reader, and a temptation to push books or categories based on outside factors is undoubtedly strong.


Amazon’s recommendations are also curious in that they are, by default, based on what readers have bought and not necessarily what they have read and loved.

What should a recommendation engine strive to do?

  • Be transparent
  • Ignore retail considerations
  • Base recommendations on a reader’s reading habits
  • Seek clues to what factors might make reader enjoy a book that they wouldn’t otherwise pick up

Neither a human nor an algorithm can meet these requirements perfectly, but a human is better suited to grasp the intangibles in play.

So what can algorithms strive to do?

Cataloging sites like Goodreads and LibraryThing seem best placed. The sites give the reader control over which books they catalog and therefore which books are the basis for the recommendations. The sites also do not have an explicitly retail function (though Goodreads is now owned by Amazon), hopefully lessening the possibility of conflicts of interest.


But the human element shouldn’t be dismissed as unworkable in the digital era:

Book communities may hold the most promise. Like-minded readers can offer recommendations that have the human touch, while crowd-sourcing makes the process scalable.

These idea may have to suffice until technology allows us each our own personal Biblioracle.