Algorithmic Rhetorics #rcdh14

This week in RCDH, we focused our reading and discussion around algorithms. As I mentioned last week, the topic felt kind of transitional for me—databases, archives, and metadata blend together fairly well (for me, at least), and they’re not topics that feel overwhelmingly technological for people. Whether or not we work on the back ends of those kinds of projects, the concepts themselves are not immediately intimidating, I don’t think.

That changes a bit with the shift to algorithms, which have a more machinic flavor. Whether that’s actually the case is something I was thinking about in class, and it’s persisting with me this morning. One of the things that I said last night is that, at heart, algorithms are simply procedures, and we spent a healthy chunk of our time trying to put that into practice.

Last week, we used a couple of pages from an old MLA job list, and brainstormed as complete a catalog of metadata as we could. Before class yesterday, I took that list and converted it into “variables,” such that Rank, for example, became a class with potential values like Assistant, Associate, Full, Open, Fixed Term, etc. I brought copies of that list of variables to class with me, and had the students write “programs” for evaluating job advertisements. We didn’t focus too heavily on programming languages or anything—I wanted to keep it as light as possible. The goal of the program was to generate a score S(n) for each of the sample ads I brought with me, and for the most part, the programs were simply a series of if-then-else steps that bumped up or down S based on a range of factors (geography, specialization, etc.). The students then swapped programs, and used them to score several ads.

Some of the obvious objections emerged in discussion—I didn’t provide any instructions about the value of S other than it should start at zero, and so we ended up with scores that ranged from single to triple digits. I didn’t give them copies of the ads ahead of time, so some of the variables they used weren’t especially relevant. I asked the students to execute the programs “as is,” which restricted what is normally a more recursive process. I probably should have been more careful about choosing information-rich ads and pruning down the variables to minimize wasted effort, but otherwise, it was an interesting exercise. It also turned us to some conversation about how the whole ecology of the process would need to change in order to implement this program for all to use, how search committees engage in their own procedures as they read materials, etc.

Maybe the most interesting thing for me about the exercise was the process of translating what is often a personal, value- and preference-oriented practice into algorithms. One of the pieces we read was Tarleton Gillespie’s The Relevance of Algorithms, an essay available online and also recently published in Media Technologies: Essays on Communication, Materiality, and Society, fresh off the press from MIT. Gillespie’s essay makes several points that really stuck out for me this week; among them is the idea that algorithms instantiate certain kinds of “knowledge logics,” such that it makes as much sense to study them sociologically as it does to think of them as technologies. He locates a “fundamental paradox” at the heart of algorithms:

Algorithmic objectivity is an important claim for a provider, particularly for algorithms that serve up vital and volatile information for public consumption. Articulating the algorithm as a distinctly technical intervention helps an information provider answer charges of bias, error, and manipulation. At the same time, as can be seen with Google’s PageRank, there is a sociopolitical value in highlighting the populism of the criteria the algorithm uses. To claim that an algorithm is a democratic proxy for the web-wide collective opinion of a particular website lends it authority. And there is commercial value in claiming that the algorithm returns “better” results than its competitors, which posits customer satisfaction over some notion of accuracy (van Couvering 2007). In examining the articulation of an algorithm, we should pay particular attention to how this tension between technically-assured neutrality and the social flavor of the assessment being made is managed — and, sometimes, where it breaks down (182).

It’s not simply that algorithms represent the operationalization or reification of what are social processes—there are also sociopolitical consequences to algorithms that make them our interlocutors and audiences, that place them among our available means of persuasion, and that make them rhetorical ecologies of their own. These are points that are obvious to some of you who might read this, but I don’t think rhetorical studies has quite yet gotten on board with the relationship between algorithms and rhetoric. I’m only just now articulating it for myself, to be fair.

One thing that occurred to me last night was an interesting parallel between Gillespie’s characterization and the definition of myths that Roland Barthes offers up in Mythologies. For Barthes, myths are a particular form of signification, an operation that takes the historical or the contingent and transforms it into Nature:

When it becomes form, the meaning leaves its contingency behind; it empties itself, it becomes impoverished, history evaporates, only the letter remains. There is here a paradoxical permutation in the reading operations, an abnormal regression from meaning to form, from the linguistic sign to the mythical signifier….The meaning will be for the form like an instantaneous reserve of history, a timed richness, which it is possible to call and dismiss in a sort of rapid alteration… (227)

I don’t know that I can convey to you how fascinating it’s been to read the first several pages of “Myth Today” in Barthes’ book, substituting “algorithm” for myth—in some places, the stretch is more casuistic than others, but by and large, it works. And this parallel might extend productively to Barthes’ thoughts about how to read myths: there are those of us who simply accept the results of algorithms as they are, those of us who understand their distortions, and then perhaps those who “focus on the {algorithm} as on an inextricable whole made of meaning {social flavor} and form {technological procedure},” producing an “ambiguous signification” that foregrounds Gillespie’s paradox.

There are strong parallels as well between a Barthesian “reader of algorithms” and the kind of computational literacy that Annette Vee advocates or the procedural literacy suggested by Lisa Gye:

procedural literacy entails learning, and thus being able to recognise, the procedures that enable algorithms and hence software to weave their magic but it is also a more fundamental literacy which takes into account of a range of human interactions. It allows us to model knowledge and to see the world as a system of interconnected parts.

It makes a great deal of sense to me to think of algorithms as myths in the Barthesian sense—and I wonder to what degree this parallel offers a more theoretically-inclined and/or humanities-friendly way of thinking about the issue of whether or not programming/coding belongs in our curricular core. In several ways, the arguments for the centrality of rhetoric are the same ones we might make for the importance of algorithms. I’m not sure that I’d completely recognized (or articulated) that resonance before now.

One last item. I want to write too about the other angle that our readings took last night, but I think I’ll save those for another post. I should note, though, that alongside several essays, we also read Stephen Ramsay’s Reading Machines, a book that makes much of McGann and Samuels’ notion of “deformance” or deformation:

In one sense, deformation is the only rational response to complexity. Nearly all deformative procedures (which include outline, paraphrase, translation, and even genre description) are intended to alleviate some difficulty…All textual entities allow for deformation, and given that interpretation occurs amid a textual field that is by nature complex, polysemic, and multi-referential, one might say that most entities require it. Seen in this light, deformation is simply a part of our permanent capacity for sense-making (48).

Are you ready to have your mind blown? According to Barthes, “The relation which unites the concept of the myth to its meaning is essentially a relation of deformation” (232).


(And yes, I think I have the topic/angle for the next essay I’m going to write.)

Metadata, Procedurality, and Works Slighted

Whenever I put together a course, I like to imagine that there’s some sort of narrative thread running through, whereby early topics and readings lead to the ones that follow. Sometimes that thread is brute chronology, but most often, it’s thematic, and I suspect that more often than not, the thread is one that only I can see, although I do try to suggest it at various points during the semester. In the case of RCDH, this has been a little tricky, not least because DH is still emergent, somewhat interdisciplinary, and my own field’s engagement with it is uneven. In my head, though, after we’d gotten an obligatory week of definitions out of the way, the first “unit” of the course was a trio of weeks gathered under the headings of database, archive, and metadata. (Here’s the schedule, if you haven’t seen it.)

We’re turning now to a week that didn’t necessarily fit that well as I was originally putting the course together, a week that combines Stephen Ramsay’s Reading Machines, some work on procedural literacies, and a few pieces/performances of algorithm. It’s an ambitious little week in its own way, but as we were working our way through a discussion of metadata last night, it got me to thinking about the transition between this week and next. Some of this I raised in class somewhat tentatively, but I wanted to write through it a bit today, partly for my own memory, and also because my guess is that there are others who have written about these ideas in more detail, whose work I may not have come across. So if any of this happens to resonate with other texts, please leave a note/suggestion in that regard.

If I have a hypothesis here, it’s that there’s an important connection between metadata and proceduracy (or procedural literacy) that I haven’t thought through as carefully as I want to. My own background/familiarity with the scholarship on metadata is a little spotty, so I worry that this is obvious to everyone other than me. I’ve read some of the classics, like Sorting Things Out, and I came to the topic through the Web2 discussions (like Everything is Miscellaneous), enough so that when I was in charge of the CCC Online Archive, we billed ourselves as an archive of journal metadata with a mix of approaches (mixing established classification schemes with emergent tagging, etc.). If I had to pin down a couple of dominant themes in the literature I’ve read, there was a focus in the mid-2000s on taxonomy vs folksonomy, and I think that’s an ongoing conversation. More recently, given mass digitization efforts, the quantified self movement, revelations of deep surveillance, and the proliferation of online archives, my sense is that there’s also been a turn towards the more basic question of accuracy, both in terms of getting it wrong and getting it too right. For example, last night we discussed Geoffrey Nunberg’s critique of Google’s Book Search, which chronicles some of its (many) egregious metadata failures and Jessica Reyman’s piece in College English on user data and intellectual property (PDF). Nunberg is a pretty straightforward critique of the consequences of getting metadata wrong, and Reyman (following Eli Pariser’s Filter Bubble) might be described as an exploration of the hazy line between data and metadata. Reyman closes by explaining that

The danger presented is that the contributions by everyday users will potentially be transformed into increasingly exclusive forms of proprietary data, available to the few for use on the many.

FB and others have become so adept at collecting and analyzing metadata that “privacy settings” are increasingly an empty gesture, a point that’s also illustrated charmingly by Kieran Healy’s “Using Metadata to Find Paul Revere,” another of last night’s readings. The connective tissue for me here is referentiality–the degree to which metadata presents us with an accurate representation of the data it is purportedly about. Referentiality is also one of the goals (among others) of various metadata standards, like the work that Cheryl Ball and the Kairos folk are doing.

One of the themes that emerged for me, though, during the discussion was the degree to which there’s a tension between (to borrow the subtitle of Sorting Things Out) classification and its consequences. That’s certainly a theme in Tarez Graban’s “From Location(s) to Locatability: Mapping Feminist Recovery and Archival Activity through Metadata,” (no link, sorry) which works from the assumption that metadata makes certain activities more visible and others less so, and that recovery work can proceed from questioning and complicating the categories that we often internalize about academic work.

I think, though, that I want to push that tension into the heart of metadata itself, something more like what Curtis Hisayasu gets at in one of his contributions to “Standards in the Making“:

What becomes absolutely transparent in these intersections is the actual work of historical knowledge-making which involves not simply “digging up” artifacts and placing them accordingly in the container of linear time, but making self-conscious decisions about how that artifact is to be organized alongside others to produce a narrative and an argument about the present.  “Tagging,” in instances such as these does not so much “describe” digital artifacts as much as it appropriates them and composes them in the name of more general logics of history and identity.

I don’t think that I’m after the claims, made variously throughout our readings last night, that metadata is necessarily rhetorical, ideological, and/or political, although I do think that it is all those things. In its nature as description, all of those qualities follow for me with respect to metadata. Maybe I’m pushing at something that is obvious or implicit in those adjectives, but I feel like there’s also space to think about procedurality in ways that might not be immediately apparent. Over the past few days, a couple of Twitter “events” helped this coalesce for me. First was Anil Dash’s essay about his “experiment” in retweeting, where he made the decision to tweet only women, and the second was Sara Ahmed’s discussion of gender and citation (in my head, I think of this as an essay on “Works Slighted”):

Sara Ahmed on the gender politics of citation

Sara Ahmed on the gender politics of citation

Ahmed’s not the first to point this out, nor are academic bibliographies the only venue where this (important) discussion is happening, nor is this solely a gendered concern, but her post(s) happened to coincide for me with this sense that metadata contains, at its heart, a ratio between description and procedurality. That is, there is no description degree zero, no purely descriptive metadata, and I think that sometimes we fall into the trap of imagining that there is. It’s not enough to simply say “it’s both,” though–I think that claim is easy enough to accept. The procedural literacy of metadata lies perhaps in figuring out that variable ratio and tuning it to the task at hand.

The exercise with which we began class last night was to take a couple of pages from an old MLA job list, and to develop a set of metadata categories based upon the entries, and one of things that crystallized for me was the degree to which this ratio varied from term to term. We also were able to lean on the ratio a bit, such that a seemingly descriptive phrase like “residential campus” could be read as a subtle (seductive) means of distinguishing an institution from others (vs. commuter campuses, online delivery, et al.). And on Sunday, I’d already taken an otherwise procedural term like “postmark” and used it as a descriptor for the impact of technology on the process. The deeper into the exercise we got, the more I think we were thinking both in terms of what metadata represent and how they functioned.

That same ratio lurks at the core of the bibliography, which is both descriptive (here are the works I have cited) and procedural (presented in a consensual format for their location), but Ahmed, Dash, and many others call attention to the procedural consequences of the bibliography. What we know now about network effects and filter bubbles should attune us to those consequences, even if we haven’t personally run afoul of the disciplinary fatalism that Ahmed describes in her followup to the Twitter conversation.

With respect to bibliographies, there’s an additional, material, fatalism–the reason we have traditionally constrained bibliographies to the works directly consulted for a given piece of writing is for reasons of space of the printed page. With the notable exception of the bibliographic essay, whose works cited sometimes also concerns itself with field coverage, there is an assumption that the bibliography must be primarily descriptive–these are the texts named directly here. But that list of citational attachments is, of course, implicitly preferential. As Ahmed puts it:

There is a ‘good will’ assumption that things have just fallen like that, the way a book might fall open at a page, and that it could just as easily fall another way, on another occasion. Of course the example of the book is instructive; a book will tend to fall open on pages that have been most read.

Maybe what I want to say is that there’s a similar assumption at play within metadata, or more precisely, within the way I’ve traditionally thought about metadata. As our work migrates slowly away from the printed page, it might open up opportunities for tuning the ratio away from description, for embracing the compositional implications of tools like bibliographies. Not that there aren’t important issues of preservation and persistence if we were to replace static bibliographies with links to a public Zotero folder, for instance, but I find myself thinking more and more about experimenting this way.

Geez. There’s plenty more to say, but this is long enough as it is. I don’t suspect I’ll have the time, energy, or inclination to do this every week. And please, if you’ve made it this far, and have suggestions in mind for other things I might stir into this mix, feel free to add them below…

Mining the JIL

This week, in Rhetoric, Composition, and Digital Humanities, we’re reading a series of essays about metadata, so that’s where my mind has been as of late. And one of the things that I’m asking my students to do each week is to imagine projects that they might do based upon the readings and resources for that week. So I spent a little time this afternoon messing around with the fabulous dataset that Jim Ridolfo has shared, the OCRed archive of MLA Job Information Lists.

I wanted to do something that had some kind of hypothesis, but also that I could do fairly quickly, without too much technological overhead. I settled for the question of how the job search process has changed over the past 10-15 years with respect to technology. When I was on the market for the first time, in 1997 (!!), I don’t recall whether the online version of the JIL had been introduced yet. But certainly the job seeker’s experience, even allowing for the online JIL, was predominantly paper-based.

I don’t think it’s particularly earth-shattering to suggest that this has changed. But can we find that change reflected in the JIL itself? I tried a couple of angles. First, I tracked all mentions of the word “postmark” (including postmarked). I was working mostly with basic pagecounts, and while I tried to be good about eliminating those few occasions where it appeared twice in one ad, I almost certainly missed some of them. I also combined the October and December JILs, so if an ad appeared in both with the word, I counted it twice, I’m afraid. This is a blogpost.

A line graph charting the use of the word 'postmark' in MLA JIL ads from 2000 to 2012

A line graph charting the use of the word ‘postmark’ in MLA JIL ads, 2000-2012

The only surprising thing about this graph to me is the little uptick in 2012, after references to postmarking had dropped into the single digits rather quickly from 2009-2011. Otherwise, though, this was about what I’d expect. As more institutions set up online HR forms and began to accept materials delivered electronically, the idea of indicating a separate postmark (as opposed to a simple deadline) has been on the decline.

The second approach I took was to look into how many job ads made reference to URLs, either in the form of departmental homepages or online (HR) submission forms. Given the steep climb from 1998 to 1999, you might forgive me for deciding not to count after a certain point. Instead, what I’ve done here is in terms of percentages. The JIL has front matter that stays pretty constant, so there are no URLs on those pages, making the maximum somewhere in the 90% range. I did a basic search count for “http,” then divided the # of page hits by the total number of pages in the JIL to arrive at these percentages. I also went backwards until I found a year (1994) with zero.


A line graph charting the percentage of JIL pages with at least one URL on them, 1994-2003.

This graph is a little misleading, though, going as it does from around one-third of the JIL pages in 1998 to nearly all of them in 1999. If I remember correctly, 1999 was the first year where participating departments were prompted to supply a homepage for their program and so, even if the ad copy doesn’t reference any URLs, the ad itself contains one. Were I to strip out those headings, the graph might be more gradual, and more accurate in its representation of the diffusion of the web into the JIL.

I know you’re curious. The five institutions that supplied URLs in 1995 were Clemson, Georgetown, Oregon State, Western Michigan, and an organization in Austria called the International Institute for Applied Systems Analysis.

Neither of these graphs necessarily lead to grand claims about the effect of the internet on the job search process. The second does suggest a point at which it was taken for granted that programs would have some rudimentary web presence. The first suggests that we’ve seen a gradual tailing-off of certain print- and postal-based assumptions about how materials circulate. And both suggest that if I wanted to claim anything more specific, I’d need to do much finer-grained work first.

That’s all.


The Limits of Facebacking

Last week, as part of their 10-year anniversary, Facebook released a tool that allows users to create (and later edit) movies based upon their FB usage. The “Look Back” videos offer “an experience that compiles your highlights since joining Facebook.” For a couple of days, my feed (and I suspect, most people’s) filled with “looks back” from a variety of friends, followed by the inevitable wave of parodies (Walter White, Darth Vader, et al.).

Like many of my friends, I went ahead and let FB sort through my photos and updates in an effort to set my “highlights” to music, but I didn’t end up sharing the results. This week in my DH course, we’re talking about archives, so I’ve been reading around somewhat alert to discussions of archiving, and I ended up thinking a bit about my “Look Back” and what it had to tell me about my relationship with FB as an archive of my life. I didn’t end up sharing my movie because I didn’t feel like it was particularly representative–while it did manage to hit on a couple of significant events (such as the fact that I bought a house), most of the updates and images included were pretty random. And so I’ve been thinking on why that was.

One obvious reason is that I don’t share as much of my life with FB as others do, and I say that without judgment. As FB itself notes, the content of the movie “depend[s] on how long you’ve been on Facebook and how much you’ve shared.” While I’ve been there for 6 or 7 of its 10 years, it would be an understatement to say that there are gaps in my self-presentation, the same gaps that show up in my movie. So on the one hand, I have no one to blame but myself–I can hardly be critical of a presentation when I’ve withheld the raw materials that it might draw on. And I don’t mean here to be critical of those who do share much more–I’m as fond as anyone of being able to keep up with my friends’ lives this way. But it was interesting to me just how much my Look Back ended up referencing my own attitudes toward sharing.

One of the texts we’re looking at this week is Johanna Drucker’s “The Book as Call and Conditional Texts.” It’s a compelling meditation on how we understand books, and how contemporary technologies might shift that understanding. Among other things, Drucker notes

The future book, which is already here, not just imagined but implemented in various ways in the configurable and mutable pages of the web, will be fluid, a conditional configuration based on a call to the vast repositories of knowledge, images, interpretation, and interactive platforms.

Of course, I read this article having already thought about the Look Back, and thus found echoes of it reverberating throughout. The “pages” of the Amazon website (or really, any e-commerce site) provide another fine example of this conditional configuration, a structured query whose contingent nature is disguised by “stable” terms like site and page. Of course the Look Back is a conditional text, but faced with mine, I was struck by how its conditions are really Facebook’s. I assume that its selection of images and updates is guided by the internal currency of shares and likes, FB’s way of calculating relative life-importance. And of course, the movies are shared on FB, themselves subject to the same economy that generated them.

I was especially taken with this passage from Drucker:

If ever the principles of a Heraclitan flux were embodied in the very ontology/phenomenology of an artifact, it is here, now, in the fleeting immediacies through which a document composes itself for our eyes only and for an instant’s disregard and then vanishes. Siblings and cousins and shades of resemblance may reassemble, so like the original we mistake them, momentarily, for that earlier temporary object brought into being, but then, with regret, relief, and other realization of the subtle but significant difference between the initial document and this “new” one, we realize the perils of our connection to “refresh rates.” No corpses remain.

I know that this is going to seem a little dark, but maybe the biggest absence in my Look Back was death itself. Among the life events that didn’t show up were the losses of my father and grandparents, and my surgery in 2011, perhaps the closest that I myself have come to death, at least to my knowledge. I have only really shared those things in very limited ways over social media, but collectively they represent a significant chunk of my life taken up with loss, depression, and grief. Even if I were better at sharing and dispelling those feelings, in many ways, my presence on FB was an intentional hedge against that time. If I couldn’t read or write academic prose (and there was a long time there where I couldn’t), at least I could toss up a link or a snarky observation about something or other. Nothing remains very long, and maybe for a while that’s what I needed.

Maybe what I saw in my Look Back, then, was a bearable lightness, or what passed for such at the time. If there’s something that Drucker’s discussion of conditional texts left me with, it was that, despite FB’s desire that I “become internally colonized by the form and formats of these reconfiguring texts,” something different happened. It produced just a little bit more distance between my own conditions and those that FB suggested on my behalf. That’s all.

Should I tweet this? Campus visit edition

Yesterday, there was a minor squall that swept quickly across my Twitterological system. One of the departments in my field that maintains its own, somewhat official Twitter account trumpeted the names and schools of the finalists for a senior search in their department. I do have a screen shot of the tweet, but figured that I’d have to redact so completely that there wasn’t a lot of point in sharing it here. But it read:

Delight! Our job search found exceptional candidates: [candidate1]- [school1], [candidate2]- [school2], [candidate3]- [school3]. Job talks coming up!

The post has since been removed, appropriately, but not before it was linked and critiqued by some folk with pretty substantial numbers in terms of followers. I don’t want to name, shame, or blame here; rather, my point is a broader one about social media and the search process, and why folks reacted so strongly and so negatively to what was in all likelihood a genuine expression of excitement and appreciation.

I think that many of us often assume that everyone in a department will know where to draw the lines when it comes to social media. And yet, the landscape changes fast enough that we don’t always have time to think about how they might interact with what may be tried-and-true procedures when it comes to things like job searches, which are not exactly everyday occurrences, particularly in the humanities. And the intricacies of the search process can be opaque, for anyone who hasn’t experienced it from every angle. There are parts of it that I know nothing about because I have neither a partner nor a family, for instance.

Whoever was responsible for the tweet, then, may not have understood some of those intricacies. In the case of a senior search, candidates aren’t always public about their applications, particularly within their home departments. It may not be a secret (although in some cases, it can be), but the fact is that an interview (and even a campus visit) is no guarantee of departure. Candidates don’t stop working or mentoring because they have a campus visit elsewhere, and they may wish to be relatively quiet about the fact that they are looking. The ethics of who to tell and when is a complicated one, and much more involved than “Are you leaving? YES NO (circle one)”

For junior searches, the issues are different, but no less important. Different schools operate according to different timetables, and broadcasting the names of finalists can affect their chances at other schools (“If that person’s a finalist at X, there’s no way they’ll come here. Next.”). It can affect their ability to negotiate other offers, and in some cases, it might even affect their strategy as they approach that campus visit.

It’s tricky. Social media accustom us to a certain level of sharing, to making a certain portion of our everyday lives public, but in the case of searches, campus visits, and the like, there’s a disjunct. A campus visit is a fairly “public” event on the campus being visited, but for the candidate, the campus visit is the final stage of what is supposed to be a confidential process. I’ve seen departments put the names of finalists on publicly-accessible calendars, announce them by name on their homepages, etc., for probably close to 15 years now. But that’s often an issue of one person making a mistake that’s easily correctable. With FB and T, every single person in a department has access to a more-or-less public audience.

At the campus visit stage, the membership of the search committee grows to include the entire department. All of a visitor’s interactions should be treated as “meeting with the (expanded) search committee.” Don’t post finalists’ names anywhere online. Err on the side of caution when it comes to tweeting or updating anything that happens in the context of the visit. Don’t tweet job talks. I’d even be hesitant to friend or follow a candidate until after the process was complete–even something as innocuous as that can be aggregated into privileged information if, say, several members of the same department happen to follow/friend the same person at the same time.

I say all this as someone who’s generally in favor of transparency when it comes to academia. But there are varied and complicated reasons for confidentiality during the application and interview process, and while I believe that hiring programs should be open about the stages of the process, candidate confidentiality isn’t ours to break.

Moretti, Franco. “Operationalizing: Or, the Function of Measurement in Literary Theory”

[It occurred to me that, after sharing my Reading Notes assignment, I should be doing more of this myself. I don't know that this will last, but in the interest of walking my talk, I'm going to try to contribute some entries to my class's shared bibliography. I'll post them here and at the course wiki, and I'll be covering essays that I wasn't able to include in the syllabus itself. My goal is to average an essay a week for the semester--I doubt that I'll do this evenly, but we'll see. I'll probably work ahead a little this weekend.]

Moretti, Franco. “Operationalizing: Or, the Function of Measurement in Literary Theory”

Moretti, Franco. “Operationalizing: Or, the Function of Measurement in Literary Theory.” New Left Review 84 (Nov/Dec 2013): 103-119. (paywall)

Abstract: Moretti suggests “operationalizing” as the bridge between the quantitative study of literary works and the theoretical concepts with which critics approach them. He demonstrates this by operationalizing Alex Woloch’s concept of “character-space,” testing it via word counts and network centrality maps. But Moretti also argues that this kind of operationalization must do more than simply provide additional detail or precision if it is to have value. In the case of character-space, he argues that “Two conflicting criteria for protagonism emerge from the two types of measurement…[and this] shows that the ‘protagonist’, far from being a fundamental reality of dramatic construction, is only a special instance of the more general category of ‘centrality’” (109, 112). Character-space as a concept “produced more than the refinement of already-existing knowledge: not the protagonist, improved, but an altogether new set of categories” (113). The second half of Moretti’s essay turns to Hegel’s concept of “tragic conflict,” testing it against a small corpus of Greek tragedies. Focusing on Antigone, Moretti suggests that operationalizing tragic conflict/collision reveals a mistake on Hegel’s part: “Face-to-face confrontations do occur in tragedy, and find a memorable expression in the rhetoric of stichomythia; but stichomythia does not convey the ‘ethically justified pathos’ that Hegel had in mind” (118). He closes by suggesting that computational critical methods may change our relationship to literary theory as it has begun to affect literary history.

Keywords: operationalizing, measurement, literary theory, digital humanities, literature, criticism, falsifiability, concepts, data, word counts, centrality, network mapping, character space, protagonism, tragic conflict, Hegel, Kuhn, Woloch


  • Alex Woloch, The One vs the Many: Minor Characters and the Space of the Protagonist in the Novel (2003).
  • Graham Alexander Sack, ‘Simulating Plot: Towards a Generative Model of Narrative Structure’, Papers from the AAAI Fall Symposium, 2011.
  • G. W. F. Hegel, Aesthetics (1975).


Operationalizing means building a bridge from concepts to measurement, and then to the world. In our case: from the concepts of literary theory, through some form of quantification, to literary texts (104).

Measurement does not lead from the world, via quantification, to the constructions of theories; if anything, it leads back from theories, through data, to the empirical world (106-107).

Computation has theoretical consequences— possibly, more far-reaching than any other field of literary study. The time has come to make them explicit (114).

Digital humanities may not yet have changed the territory of the literary historian, or the reading of individual texts; but operationalizing has certainly changed, and radicalized, our relationship to concepts: it has raised our expectations, by turning concepts into magic spells that can call into being a whole world of empirical data; and it has sharpened our skepticism, because, if the data revolt against their creator, then the concept is really in trouble. A theory-driven, data-rich research programme has become imaginable, bent on testing, and, when needed, falsifying the received knowledge of literary study. Of this enterprise, operationalizing will be the central ingredient (119).


Moretti raises 2 questions in a footnote on p. 114: “As will become clear, I assume that Hegel’s theory can be operationalized. This leaves open two further questions. First: and if it couldn’t? Would the theory lose all its value, and deserve to be forgotten? The second question is almost opposite in nature: if applied too loosely and widely, wouldn’t operationalizing lose the strict falsifying potential that had made it so valuable in the first place? In principle (though a full motivation will have to wait for another occasion), my answers would be, No to the first question, and Yes to the second.”

Another, more immediate question for me is the different relationship between concepts and practice that (may) exist in rhetorical studies. I don’t think that rhetorical figures are necessarily the same as literary concepts–they are empirically verifiable on the one hand, but their consequences seem less determined on the other. And yet, I can think of several writers who have at least dabbled in what Moretti describes as operationalizing, myself included.

Reading Notes

There’s a phrase that I got from Laurence Veysey, via Gerald Graff (although it appears in other places as well): “patterned isolation.” Veysey uses the phrase to explain the growth of the modern university and the way that disciplines grew without engaging each other, but I tend to apply it on a more “micro” scale. That is, there are many things we do as teachers and scholars in patterned isolation from our colleagues, tasks that call upon us to reinvent wheels over and over in isolation from one another. Fortunately, with the Internet and all, much of that is changing, as folks share syllabi, bibliographies, and the like online.

But I’m constantly on the lookout for ways to short circuit patterned isolation. For me, reading notes are one of those sites. I’m not a great note-taker and never have been–I’m too reliant on visual/spatial memory and marginalia. Disconnecting my own notes from the physical artifacts that I was processing didn’t make sense. Now, of course, I’m lucky sometimes if I remember having read a book, much less what I scrawled in its margins, so I wish that I’d been better about taking notes and I admire those people who have already internalized the lesson that it took me 20+ years to figure out. So one of the things that I like to do in my graduate courses is to aggregate the note-taking process. Rather than asking or expecting each student to take a full set of reading notes for the course, I rotate the function among them. There’s nothing to stop a student from taking more detailed notes on his or her own, but I want all my students, when they leave the class, to be able to take with them a full set of notes that they can refer back to later.

For me, the trick to this is making the notes relatively uniform–there’s a part of me that resists this, because different folks/strokes and all that, but I think it’s important to make the notes themselves scannable and consistent. Also, I think that the process needs to be sustainable–part of the challenge of reading notes is that they tend to shrink or expand based on available time, and they shouldn’t. The notes should be brief enough that one can execute them quickly (when time is short) but elaborate enough that they’re useful 5 or 10 years down the road when the text has left one’s memory. For me, this means keeping them to about a page, and doing them in a format that should take no more than about 15 minutes for an article or chapter. So here’s what I’ll be asking my students this semester to do for their reading notes:

  • Lastname, Firstname. Title.
  • Full MLA citation of article/chapter (something that can be copy/pasted into a bibliography)
  • Abstract (50-75 words, copy/pasted if the original already has an abstract)
  • Keywords/tags (important terminology, methodology, materials, theoretical underpinnings)
  • 2-3 “key cites” – whose thoughts/texts does this article rely upon
  • 2-3 “crucial quotes” – copy/paste or retype the 2-3 most important passages from the essay
  • 1-2 questions – either questions that are answered by the text, or questions it raises for further exploration

And that’s it. I’ve futzed around with different categories, but these are the ones that have stuck with me through multiple iterations of this assignment. The notes aren’t meant to be a substitute for reading, but they should provide the basic info about the article as well as some indication of how it links to others. And it’s meant to be quick.

This seems really obvious to me now but I can tell you that, when I was in graduate school, it wouldn’t have been. I can’t tell you how happy it would make me now to have taken notes like these on everything I’ve read since grad school. Especially for all those things that I’ve forgotten I’ve even read.

Penumbra & Digital Humanities

In a little less than 24 hours, my spring graduate seminar on Rhetoric, Composition, and Digital Humanities will begin (syllabus). I’m not typically the kind of professor who gives his students assignments to complete prior to the first course meeting–it always struck me as a little mean-spirited to bite into what little break we have between semesters. This year, however, I broke form, and suggested to my students that they read one or both of a couple of “fun” reads: Charles Soule’s graphic novel Strange Attractors and Robin Sloan’s Mr. Penumbra’s 24-Hour Bookstore. I may talk later about Soule’s book, which shares some features with Sloan’s, not the least of which is the apprenticeship that form the social core of each, but far more digital humanities folks that I’ve seen recommend Penumbra as the DH novel par excellence. As I finished re-reading it this evening, I thought about why exactly that recommendation is so prevalent.

Not that I disagree. On the surface, though, there’s really only a small part of the book that feels like a recognizably DH moment (whatever that is, I know). I’ll try to do all this without spoiling too much, but it’s the early scene in the book where data from Penumbra’s logbooks are mined and digitally visualized. Digital tools play various roles throughout the book beyond that scene, helping to propel Clay forward, but there’s also some irony in that a microscope and an audiobook on cassette tapes (arguably) play as large a technological role in the book as anything else. To say much else about the relative significance of various tech would be to drift into spoilerville, so I’ll trust you to take my word for this. (Or to argue with me obliquely, at least.)

In preparation for tomorrow’s night class, then, I wanted to be able to do more than simply send to page 94 to read the passage about metadata and visualization. At the risk of making this sound a little like a sophomore literature paper, I browsed some of the more popular definitions of DH, and thought about how Penumbra does or doesn’t reflect them. I do think that there’s a point to be made with respect to technology, and not simply novelty. There are a few places in the book where Clay reflects on the sophistication of technology (“Books used to be pretty high-tech, back in the day. Not anymore” (90), for example), and it’s not always the shiniest, most advanced piece of tech that ends up providing solutions. Sometimes, it’s a matter of thinking about old tech in new ways; mostly, it’s about not falling into the trap of assuming that there’s only a narrow field of use for technology (or that technology is of no use). So while Kirschenbaum writes that DH is “a field of study, research, teaching, and invention concerned with the intersection of computing and the disciplines of the humanities,” that definition has opened up onto broader inquiries into the relationship of media, materiality, and the humanistic disciplines in a way that the digital provokes, but may not completely capture. More recently, the collaboratively written Digital_Humanities volume from MIT describes DH as both transhistorical and transmedia.

The authors of that book also observe that

[Digital humanities] asks what it means to be a human being in the networked information age and to participate in fluid communities of practice, asking and answering research questions that cannot be reduced to a single genre, medium, discipline, or institution.

For me, this helps to open up what we might mean in calling Penumbra a DH novel–from the library of the Unbroken Spine to Kat’s efforts with Google Forever, there is a great deal of speculation in this book about what it means to be human, and what it means for our more traditional institutions (such as bookstores) to think in terms of networked information. And much of the book is driven by Clay’s ability to negotiate some fairly disparate communities of practice (even if these are sometimes plot conveniences). There’s a crucial moment for me, one that I think avoids any sort of spoilage: at one point late in the book, Edward Deckle tells Clay, “You seem very resourceful,” and this feels very resonant to me. Clay drives the plot forward, but he also functions very explicitly as a weak tie among many of the other characters (and disciplines, and institutions) in the book–the communities of practice ultimately coalesce around his efforts. So, he’s also resourceful in the sense that he’s able to fuel the collaboration necessary for the story.

[If I wanted to go full-on lit paper here, I'd probably talk about the etymology of penumbra, and the way that Clay bridges a number of tensions--property/piracy, human/machine, nerd/cool, natural/artificial, etc. Be relieved, because that's all I'll say on that.]

He’s also the one person (although Kat comes awfully close) who has access to the material resources necessary for all of this, and that’s a piece of it, too. Some fairly impressive resources are marshaled on his behalf in the book. I mean this in the economic sense, but also in the physical sense, because I was a little surprised, this time through, how often I thought about “making.” Perhaps it was having glanced back at Stephen Ramsay’s infamous claim that “Digital Humanities is about building things.” Most of the characters in the book are builders–even the minor ones like Ashley, who comes “alive” in their kitchen, or Grumble, when he holds the shrooms. Like Ramsay’s, Sloan’s is a capacious definition of building, I think.

I think I’ve probably said enough. I’m sure that there will be other novels that fall into this odd, little DH category — there are already pieces of some out there: my first introduction to text mining as a reading strategy came more than 20 years ago when I read Italo Calvino’s If on a winter’s night a traveler in college! But I think that Penumbra is more than a novel that happens to make use of metadata, visualization, and high-end computing as plot devices. In its accounting of network fluidity, collaboration, resourcefulness, and media consciousness, Sloan’s novel embodies for me a lot of the curiosity and energy that I associate with DH. Ultimately, that was probably the main reason I suggested it to my students.

We’ll see how it goes tomorrow.

Telescopic Text

So I’ve been slowly reading S by Abrams and Dorst, and slowly expanding my Twitter horizons with respect to bots, and today, I came across a really interesting app/tool that crossed the streams, so to speak.

It’s called Telescopic Text. Not unlike Tapestry, it’s an application that lets you write and store texts. Those texts, though, are like that word game where you create a ladder of words by adding a letter at a time (a, an, pan, plan, plane, planet, etc.). You start with a tweet-length sentence, highlight particular words, which then “unfold” as they’re clicked on. It’s like drilling down into a text to find more and more details.

The TT site itself starts with an example:

The tools for building one, and saving it, are at:
(registering for an account is free, which you’ll need to do if you want to save your efforts)

I ended up finding the site from a link to Tully Hansen’s “Writing,” which is located here (you’ll need to scroll down):

It reminds me too of Jon Udell’s classic screencast about the WIkipedia entry for the heavy metal umlaut:

I’m not entirely sure how I’ll be using this, but it’s been a lot of fun to play with this afternoon…

(x-posted from Facebook)


Apropos of positively nothing:

I was thinking just now about -aholic as our kludgy suffix for addiction, as in workaholic, chocoholic, danceaholic, Brookaholic (okay, that last one isn’t real). I figured that it must have come originally from alcoholic, but alcohol doesn’t match up with my spotty recollection of Greek and Latin. Briefly, I wondered (given the al-) if the word wasn’t originally Arabic.

Lo and behold, it is. It comes from the Arabic al-kuhul, or “the kohl,” which was ground ore used as mascara. Later it generalized to mean something like the pure substance of anything, including liquids, and it only acquired its modern sense in the 18th c or so.

Kohl comes from kahala, to paint or stain. Given the aversion to rhetoric as a merely decorative art that stretches back to antiquity, I begin now to understand certain rhetors’ obsession with beer.

from Facebook

Back to top
© 2014 Collin Gifford Brooke - A MeanThemes Theme