On Ant-Lions and Scholar-Programmers
May 07, 2011
Alongside the griffins and unicorns in medieval bestiaries frequently appears a fascinating but somewhat less notorious hybrid: the ant-lion, a creature with a magisterial feline head and an industrious, formican body. David Badke’s fascinating online anthology of bestiaries notes that fantasies of ant-lions likely originated with the Septuagint’s mistranslation of Job 4:11 which contains an “uncommon word for lion” that is commonly rendered, as in the King James Version, “old lion” (e.g. “The old lion perisheth for lack of prey”), but in the Greek translation as the “Μυρμηκολέων”—the ant-lion. Although medieval readers (somewhat disappointingly) read a reference to a large, aggressive, and possibly cannibalistic species of ant early Greek writers imagined an actual lion-ant cross-breed which, being the offspring of a carnivore and herbivore, could find no food to eat and so eventually “perisheth.” For my entry in this bestiary of alternative career paths in academia, I would like to turn to analogous hybrid: the scholar-programmer, a beast with the head of a scholar (and likely an advanced degree in a theoretical discipline to prove it) joined with the industrial and practical skills of a computer programmer.
The scholar-programmer is not, it should be noted, a software developer, but a scholar for whom algorithms are an important mode of inquiry and communication for work in his or her chosen field. This chapter will particularly examine the scholar-programmer whose interests lie mostly in the humanities. This focus is not chosen because scholar-programmers are only (or even primarily) found in such fields (indeed, scholar-programmers occur naturally in the sciences), but because it is in the humanities that the hybridity is arguably most striking. In this chapter I will examine four possible career paths for the scholar-programmer—the traditional faculty position in a humanities department, a faculty position in a school of library and information science, a research librarianship, and employment in a digital humanities center—along with the concomitant advantages and drawbacks of each.
First though, a word on the qualifications one should attain to become a scholar-programmer seems apt, as one goal of this collection is to plant, to borrow a word from the request that partly inspired it, “signposts” for graduate students in search of a career in academia. First, the humanities scholar-programmer should probably earn a Ph.D. in the humanities. Although one can clearly function as a scholar without a terminal (or any) degree in one’s field, a Ph.D. remains the most widely accepted certification of one’s qualifications to participate in the work of the academy. For some of the careers described below, a Master of Library or Information Science might also suffice and may in fact open more gates along certain roads, but for those who think of themselves as a humanities scholars first and programmers second, a Ph.D. demonstrates a clear and inarguable commitment to the field. Technical qualifications are somewhat more flexible. At most universities the computer science undergraduate major is less a course in programming and more a study in advanced mathematics and the theory of information one will encounter in computer science graduate work. This can often be surprisingly useful knowledge; but I do not think every scholar-programmer needs to have it. Instead, a scholar-programmer should be someone for whom algorithmic thinking, that is the ability to think about a problem in mental pseudo-code, comes naturally. Familiarity with a number of different programming languages is important and the ability to learn new languages quickly is essential. This will probably mean the scholar-programmer has taken at least a few programming classes, but some of the best scholar-programmers working today are largely self-taught. The best proof of one’s programming credentials, then, is being able to point to an impressive bit of functioning code written (mostly) by oneself on a major code respository such as Sourceforge, GitHub, or Google Code.
One might think that scholars with the above qualifications would be highly valued for the breadth of their skills, but in the current environment the humanities scholar-programmer is far more likely to find no completely suitable professional home and must often either temporarily favor one nature over the other to avoid starving for lack of gainful employment. Many humanities scholar-programmers will first attempt to find a tenure-track faculty position in a relevant humanities department. If it can be found, a faculty job is without a doubt the most widely understood and in many ways the most accommodating place for the scholar-programmer to conduct her research. The advantages of extremely flexible hours, summers off, and the permission (even the expectation) that one research in the field that led one to graduate school are undeniable perks. Certainly, burdensome committee work, teaching uninterested or even antagonistic students, and relatively low salaries and limited travel funds keep the ivory tower from reaching paradise, but these were factors most scholars knew and at least partially understood as they pursued their graduate degrees, and yet many finish undeterred and hundreds apply for most available openings—with the result that few actually find such a tenure track position.
Technical expertise, in most cases, gives the scholar-programmer little advantage in the competition for faculty jobs. Most humanities departments, like any professional organization, tend to select candidates with skills the incumbent employees can best appreciate. Work that is largely based on technical methodologies may seem unfamiliar and perhaps threatening to traditional humanities scholars. Surprisingly, even those few positions now sometimes advertised as requiring a specialty in “digital humanities” rarely require or strongly value the ability to program. Programming is a very difficult and time consuming activity that drains mental energy in a way most scholars only experience when in the midst of an important writing project, and the energy spent writing code takes away from that which could be spent publishing articles and books (the most valued currency to deposit into the humanities CV at present). Even if hired, the scholar-programmer in a traditional humanities department may find it challenging to communicate the value of her work to her colleagues.
For some, a faculty position in Library Science (or as it is now more often called, Information Science) might be an alternative. For probably obvious reasons, the value of algorithmic thinking became clear to those whose work centers around searching, sorting, comparing, and visualizing data rather more quickly than it has for those whose work mostly involves reading books and writing about them, and so technical sophistication is more likely to be understood and appreciated in such departments. On the other hand, writing a monograph on a traditional humanities topic (say, a post-colonial reading of Beowulf), even if one’s research is supported by clever algorithms, may not earn one tenure in some Information Schools, and a humanities Ph.D. is probably less valuable to a candidate than one in a field directly related to Information Science. Most of the other advantages and drawbacks of the humanities professorship still pertain, though, and the job market tends to be slightly better, so for scholar-programmers whose research fits the Information School paradigm, hopping disciplines might be a viable alternative.
Libraries themselves are also beginning to provide excellent opportunities for scholar-programmers. The practicality of providing resources to scholars who increasingly interact with their content and each other as much online as they do in physical space has forced library administrations to begin to develop the technical capacity of their work forces. Many now seek to hire librarians with technical skills to work on interesting projects, and are sometimes willing to forgo the usual requirement of a Masters degree in Library Science for demonstrable content and technical expertise. Further, many libraries offer tenure-track faculty status to these employees (often at higher salaries than humanities faculty of equal rank). Still the primary mission of most libraries is service to scholarship rather than scholarship itself, and so librarians, despite their faculty status, often are ultimately treated as service employees and usually have 12-month contracts (without summers off). Further, most digital librarians are responsible, not just for one content area, but for the holdings of the entire library. The work of the humanities scholar-programmer in such a position is therefore likely to drift further and further from that which inspired her to undertake the stress and poverty of graduate school in the humanities.
There is still another place, to (mis-)quote an 11th century bestiary entry on the ant-lion, not of those which I have told you. Over the past two decades there have emerged places known as Digital Humanities Centers that provide amenable ecosystems for such hybrids as the scholar-programmer. For those unfamiliar with the model, digital humanities centers are spaces, usually administratively connected to libraries or humanities colleges, which attempt to combine digital technology with humanities research. I will spend the remainder of this chapter discussing the place of the scholar-programmer at such centers. My two reasons for such disproportionate attention are a) digital humanities centers have come into existence far more recently than the other institutions described above and so the opportunities they offer may be less familiar to most readers and b) although far from perfect they seem to offer the most potential at present for satisfying the diverse passions of the scholar-programmer
Although there is now some variety in what such centers do, many were founded as service units tasked with providing advanced technical support to tenure-track faculty members in traditional humanities departments. The scholar-programmer employed by such a unit would likely find herself working on someone else’s research in a relationship that more closely resembled that of graduate student or lab assistant to her supervisor than a true scholarly collaboration. As funding became increasingly available for digital humanities projects (most notably with the founding of the Office of the Digital Humanities at the National Endowment for the Humanities), the mission of many centers began to shift from faculty service to internally generated, grant-funded research. For a few scholar-programmers, external funding meant it was possible to add one’s own research project to the work of the center.
Even so, new positions at digital humanities centers are still often classified as “staff” rather than “faculty.” It is tempting to think that this status does not matter, but it does, at least, suggest the value with which a university regards the employee to those outside the institution. An assistant professor has a title that clearly defines a stage in a career that is understood both throughout the university and can be transferred to other institutions—an assistant director of a digital humanities center far less so. Still, the enterprising scholar-programmer in this high profile, emerging field can often more than recover whatever professional status is sacrificed by the staff designation. As a result of the aforementioned increases in funding for digital humanities work, many scholar-programmers now have unprecedented opportunities to receive and oversee large, international research projects at a very early stage in their career. In the five years I have spent as a scholar-programmer at a major digital humanities center, I have been in small and productive meetings with the heads of many of the world’s major libraries, a few of the nation’s best performing artists, and consider top scholars from around the world my colleagues and friends. As Bethany Nowviske, editor of this volume, tweeted after such a meeting: “struck again by dues-paying crap I skipped in deciding against tenure-track jobs. How many junior faculty sit in on discussions like this?”
Of course, along with the glamour of high-powered meetings and world travel comes the administrative and managerial work that is also, in traditional departments, reserved for more senior scholars. Like department chairs and deans, successful scholar-programmers can quickly find themselves distantly separated from the actual work of humanities research. In large centers, even much of the actual coding is done by those hired on the soft money provided by grants. For many of the scholar-programmers who administer these grants, this “outsourcing” of the programming work feels somewhat like having a graduate student write one’s monograph (even if it is based on one’s own outline). My colleague Matthew Kirschenbaum observes, quite accurately, that, “Many programmers talk openly of the aesthetics of code, using terms like beautiful or artful in the same way that a grandmaster might describe a game of chess (another formal activity par excellence).” For the scholar-programmer the code is a place of discovery and scholarly expression no less than written natural language. Unfortunately, at large centers, it is likely that the one who conceived a grant project must oversee many others as well, and so is unlikely to spend much time working on the details of the code for the project she initiated. Thus, for many digital humanities programmers, (not all of whom, it should be noted, self-identify as scholar-programmers), the service model remains very much in effect; the programmer still works on a project the scholar cannot complete, only now because of lack of time rather than lack of knowledge.
There are, it should be said, very good reasons for an organizational model in which the director of a software project delegates much of the hands-on work to a team. In fact, it is generally the best model for the sorts of projectsthat intend to deliver reusable code that might participate easily in the academic open source ecosystem. These are the sorts of projects that are often funded, and rightly so. The really elegant piece of code written in the course of research by the scholar-programmer which, after having run once or twice, has completed its one and only purpose is not uncommon in more technical fields, but generally is not (and arguably should not be) funded by public money in the humanities. It must be understood, however, that large digital humanities software projects are more about programming and enabling humanitiesresearch than about research itself.
Yet many digital humanities centers, including my own, promote themselves as research units (usually to distinguish our work from the service function of, for instance, those groups that train faculty in the use of more common technologies such as blogs and PowerPoint). Clearly, digital humanities centers are places where scholars are exploring more advanced technologies, but to call tool building or interface design “research” seems to stretch the definition of the term well beyond the elasticity of its common interpretation by the rest of the academy. The reality is that while many digital humanities projects produce useful and innovative new software, few by themselves, actually advance knowledge in either the humanities or programming. We abdicate that work to the “content partner,” the scholar who oversees and contributes content but does not program. Yet most digital humanities centers have on their staff a few scholar-programmers who could, if afforded enough time, serve as both the scholarly and technical lead on important projects. If digital humanities centers are to become, as most aspire to be, places of research rather than service, they must abandon the anachronistic division between theory and application. Those who want to use new technology for their research should intimately understand it themselves.
Some might object that 1) there are not enough scholar-programmers who can do such work and 2) service and grant work help to justify the existence of an academic unit, especially a new unit like a digital humanities center, in lean economic times. Both objections could be addressed, however, if digital humanities centers became, like most other non-service units on campus, teaching units. As cultural heritage and human experience becomes increasingly digital, the relevance of non-technical humanities scholarship is in danger of diminishing to the point that tax payers, legislators, and donors may someday decide it is no longer worth supporting. Digital Humanities centers are uniquely positioned to begin training a new generation of students for this change, and if they undertake this mission now, few could fault them if they no longer build websites or tools for faculty. Such centers would have to be staffed by scholar-programmers whose jobs would so resemble the work of tenure-track faculty that fair administrators could not ethically withhold the status from them.
Until such a happy future, though, the ant-lion seeks its elusive prey. Which (if any) of the above career paths is best suited for individual scholar-programmers depends, unfortunately, on which part of their nature they are most willing to sacrifice. If one’s scholarly head must be fed by conducting humanities research, a tenure track position in a traditional humanities department is probably the only choice that will satisfy. This prey is as elusive as is rumored, though, and hunting it successfully likely means amputating (at least for a time) one’s programmer body in favor of prostheses made of articles and monographs. If, though, working on interesting programming projects in service of humanities scholarship seems sufficiently nourishing, a library job might suit. For those who are unwilling to completely cut off either part of their intellectual selves, and believe in the promise of what might soon be rather than what currently is, a digital humanities center will easily keep a scholar-programmer from starving and could, in time, provide the sort of omnivorous feast for which we hybrids hunger.
Badke’s site includes a copy of a 1923 article from Antiquaries Journal by George C. Druce that cites J. B. Pitra’s translation in Spicilegium Solesmense, vol. iii, p. 354, of
Physiologus veterum gnosticorum: ‘Eliphaz the Themanite says: “The ant-lion perisheth for lack of prey”. The Naturalist speaks thus about the ant-lion, that its father eats flesh and its mother herbs. If then they shall have produced the ant-lion they produce it possessed of two natures. It has the fore parts of a lion and the hind parts of an ant; so that it cannot eat flesh because of the nature of its mother (or herbs, because of the nature of its father), and therefore it perishes for lack of food.’(http://bestiary.ca/biblios/biblio45.htm)
 Scientific publications frequently cite algorithms used in research. Consider, for instance, the image post-processing algorithm “DUSTY”, used by astronomers to account for the effects of space dust, http://www.pa.uky.edu/~moshe/dusty/DUSTY_Cites.htm
From “Hello Worlds (why humanities students should learn to program)”, originally published in the Chronicle of Higher Education, and now reprinted on Kirschenbaum’s website at: http://mkirschenbaum.wordpress.com/2010/05/23/hello-worlds/