[Humanities computing as interdiscipline]

Rev. 22/10/99;
23/10; 24/10

A note on terminology:
humanities computing or humanistic informatics?

This is a brief attempt to consider what we might most usefully call the field whose major concern is the application of computing to the humanities. Discussion of the topic suggests two major alternatives: humanities computing or humanistic informatics. I begin with the assumption that names matter: they suggest, perhaps even shape the nature of what they name, and so care in choosing them is not misspent. I also assume that the poetics of naming is relevant: what they suggest by association of meaning, derivation or sound. We need to be concerned with how people tend to react to the names we choose, whether this is to respect, ignore or attempt to modify the reactions.

In paying such close attention to terminological poetics I of course limit my discussion necessarily to English. Since the situation is changing rapidly, among English speakers associations and reactions may vary across age as well as culture and location. The aim here, then, is to provoke discussion. Clearer thoughts on the same subject, especially those based on better research, are most welcome.

humanities computing

Of the two major alternatives I have identified, humanities computing seems the older, though it was preceded by computing and the humanities (as in the name of the professional journal) and computing in the humanities. It would be pleasing to discover that and preceded in, giving way to it as computing became increasingly assimilated into the disciplines concerned. A corpus study could presumably establish or disprove the historical sequence. In any case we know from the names adopted by centres in the field [X] that computing and… has all but disappeared.

The term humanities computing makes an adjective out of the noun but simultaneously puts it first, creating an interesting and useful blurring of the hierarchical relationship. So, of course, does computing in the humanities, though in a different way. Humanities computing, however, has the advantage of brevity, and brevity allows the entire term to enter the mind almost before any processing begins, whereas computing in the humanities is long enough to invite parsing, stop the intruder before it is entirely through the door, give time for its ejection. Anyone who answers the telephone by saying "centre for computing in the humanities" knows the effect this tends to have on many callers.

Because of the way computing is understood (about which more below), humanities computing also tends to be taken as oxymoronic; the effect is considerably enhanced by the abrupt juxtaposition of two nouns. We could even say that the juxtaposition constitutes a stark metaphor without the limiting copula, as famously in Ezra Pound's analysis of haiku, so that the violence with which the two apparently different, even contradictory things are yoked together is felt at maximum, the question asked most insistently without commitment to any particular answer. For those unfamiliar with the field, the mind is suddenly brought up short to wonder, "What does computing have to do with the humanities?" One might think that this is a problem, and in some ways it is, but the problem is a very good way to begin discussion with a curious person. Since we ourselves do not fully understand what computing is to the humanities and vice versa, we profit from being reminded that there is a problem here to consider. Furthermore, if I am right that the essential moment in our practice is in confrontation with the gulf between what we know somehow and what we can model with the computer, then the stark metaphor does precisely the job we need the name to do.

Etymologically, too, the term humanities computing is rich. Humanities points back to humanist, "A student of human affairs, or of human nature; formerly, sometimes, a secular writer (as distinguished from a divine)" (OED), which historically refers to Renaissance humanism. At root, however, is the word human or humane, an earlier spelling that has taken on the normative sense of the demeanour or action which befits a human being. This sense of an ideal suggests a focus on the fundamental, enduring concerns of humanity, whatever these may be -- much more a preoccupation or habit of mind than "values" (a word now coloured by the reactionary notion of returning to something lost).

Computer derives from the Latin verb puto, "to clean, tidy up; ponder, discuss; estimate, assess; regard as; think", through the medieval computus, "a set of tables for practically calculating astronomical occurrences and the movable dates of the calendar; a calendar" (OED). In historical usage, the words compute and computation (both from the 15th Century) have the sense of "reckoning by numbers", or sometimes just "reckoning"; computational (from the late 19th C) is exclusively numerical up to very recently. Computer enters the language in the mid 17th Century; it meant, "One who computes; a calculator, reckoner", then in the late 19th, when computational began, it comes to denote a machine for the purpose (cf. typewriter). Computer in our sense "cannot be dated to before 1945", when the term for mechanical calculating device met the ideas of Alan Turing in the ENIAC, acronym for "Electronic Numerical Integrator And Computer" (Twentieth Century Words, p. 267). Turing's ideas come out of the philosophy of mathematics; although numbers may be used to represent what happens in the Turing Machine as in the modern computer, the processes of both are basically logical not numerical. The etymological sense of computing as a means of putting things in order is thus not far off.

For obvious historical reasons, application of computing up to the 1980s was almost exclusively for numerical calculation in the sciences, and the expensive hardware was so often the focus for attention. Change began in the mid 1970s with the introduction of the personal computer (the term dates from 1976). Since then progressive sophistication in the interface between the hardware of logical operations and human intentions means that now, when we reflect on what we're doing with the machine as humanists, our focus can be much more easily on the root philosophical problem of what is by nature computable or can be stated in terms that are. But vocabulary lags behind and so interferes with thinking about and communicating what we do, as Searle complains of in Minds, Brains and Science. Thus this terminological note.

Contemporary evidence is mixed. The COBUILD Bank of English and the British National Corpus suggest that in American usage computing refers almost exclusively to numerical computation or hardware, in British usage, especially conversational, somewhat more widely to uses of the machine in general. Computational infrequently refers to applications beyond the sciences.

The story of humanities computing is further complicated by the connotations of unfamiliarity, foreignness and technical specialisation that cling to computing. These connotations not only repel the technophobe but also, more dangerously, attract the technophiliac, whose drive is toward exclusivity. If we focus on the technical aspects of what we humanists do with computers, then it may seem as if we are almost at the level of opening a musical greeting card (which contains a computer about as powerful as the first ENIAC). Indeed, much if not most of the intellectually important applications of computing in the humanities involve the most commonplace tools; the significant operations, as in textual encoding, take place in "wet-ware", i.e. the scholar's mind. So, when we use the term computing and even more computational for what we do, confusion is likely. The expectation that these terms refer to the unassimilated, obviously technical applications makes the term humanities computing seem old-fashioned, even quaint to some, as it appears to harken back to the days when the tools we actually use were new, raw, obviously computational.

My response to the situation is to use the words computing and computational whenever referring to operations partially or wholly in "wet-ware" -- a strategy not entirely dissimilar to Lenny Bruce's with respect to obscene words [X], in order to draw attention to a root problem.

As a colleague remarked in e-mail, "'Humanities computing' has two virtues: one is that everyone knows what the words mean, the other is that they don't. But at least neither word is strange...." (Norman Hinton, MedTextL 23/10/99).

humanistic informatics

What about humanistic informatics? It shares the same freight of meaning by derivation from human, humane, and some of the same oxymoronic force, but there are differences.

First, the juxtaposition is not so abrupt, since humanistic is marked as adjectival, so we avoid the metaphor and easily arrive at the qualified idea of a branch of informatics dealing with the humanities in some way or other.

Second is the common suffix of these two words, -ic(s), from which, I think, much of the force of the term comes. As the headnote to sense 2 in the OED entry for –ic points out, from 1600 onwards -ics "has been the accepted form with names of sciences, as acoustics, conics, dynamics, ethics, linguistics, metaphysics, optics, statics, or matters of practice, as aesthetics, athletics, economics, georgics, gymnastics, politics, tactics." In other words, an -ics is a specialised practice, with strong association to the sciences. Indeed, as Espen Aarseth says in From Humanities Computing to Humanistic Informatics: Creating a Field of Our Own [X], "'informatics' is roughly equivalent to 'computer science'".

In European usage scientific and its equivalents in the other languages denotes intellectually serious work, since as Susan Hockey notes, "the word scientia really means 'knowledge'" [X]. In North American usage and to some degree in the European, however, scientific often has a dangerously honorific function. As John Searle says in Minds, Brains and Science, "The situation with the word 'science' is even worse [than for 'mind']."

I would gladly do without this word if I could. 'Science' has become something of an honorific term, and all sorts of disciplines that are quite unlike physics and chemistry are eager to call themselves 'sciences'. A good rule of thumb to keep in mind is that anything that calls itself 'science' probably isn't - for example, Christian science, or military science, and possibly even cognitive science or social science. The word 'science' tends to suggest a lot of researchers in white coats waving test tubes and peering at instruments. To many minds it suggests an arcane infallibility. The rival picture I want to suggest is this: what we are all aiming at in intellectual disciplines is knowledge and understanding. There is only knowledge and understanding, whether we have it in mathematics, literary criticism, history, physics, or philosophy. Some disciplines are more systematic than others, and we might want to reserve the word 'science' for them. (p. 11)

The tendency, then, in humanistic informatics is to push our practice toward the more technical, more "scientific" end of things, as if it were a kind of computer science. Its divorce from the arts and humanities is further marked by the repetition of -ic/-ics, insistent if not offensive to the ear, at least in English. A stylistic blunder. The effect of -ics in a new term, such as informatics, is I would suppose much sharper than in one we have had in our mouths for longer, such as mathematics or linguistics.

In the OED, informatics is in a definition dating from 1967, "the discipline of science which investigates the structure and properties (not specific content) of scientific information, as well as the regularities of scientific information activity, its theory, history, methodology and organization." It is not a common word -- the Bank of English yields 1 hit in 35 million words -- so is likely to puzzle most people, again suggesting a specialised, arcane practice, a "science" in Searle's terms, even to those who do not know what it is. The associations are apt to be far too technical -- unless that is exactly what one wants.

Informatics refers, of course, to the word information. Etymologically, inform means "to put into form, to give form to", suggesting that to inform is to create -- another suggestion of order out of disorder. In modern usage, however, the word has sunk to a lowest or at least lower common denominator: information is the factual data you get from a reference book or a kiosk in a shopping centre; poets don't write poetry to convey information, most readers don't read poetry for it, though a scholar may be able to derive certain information from poetry. Information, more technically, is transmitted data wrapped in "information theory". Again, the overall force here is toward the technical.

Conclusion

This is certainly not a scientific note on terminology, rather an attempt to get at a certain set of partially conscious tendencies and associated feelings. The basic question it raises, however, is independent of its idiosyncracies, namely, what kind of a practice do we want? What approach to the other humanistic practices is best for it?


[Humanities computing as interdiscipline]