On the linguistic and literary history of computing

On the linguistic and literary history of computing

A note on philology, passwords, and dictionary attacks

Diane Mehta, Hackable Book References and the Madwomen in Our Attics:

Dictionaries and books have turned against us. […] thinking about literature, and about literary passwords, as hackable because they existed in the public domain was something that made me uncomfortable. Public access to literature is a grand thing. But the secrets you carry around with you, after absorbing them from the books you love, is a deeply personal thing. The idea that literature was there to fill a database, that that the passwords I take from literature — passwords that mean something personal to me — made me think twice about what exactly it is I loved and how certain literary references feel more secretive simply because they mean so much.

This is the very essayistic type of essay that I used to enjoy both writing and reading somewhat more than I do now. Like any good example of the form, it gets something profoundly correct, in some ways more correct than is possible in the more strictly expository mode we mandate for each other as scholars.

So-called “dictionary attacks” are as old as computer password security systems themselves, which is to say as old as the time-sharing system as a historically specific form of modularization. Words, that is to say, have never served as secure “passwords,” just because they are words: that is, artifacts of the national-language standardization that is a legacy, the primary linguistic legacy, of our most recent modernity, the nineteenth-century technomodernity of empire and, later, decolonization; and that marks the birth of what within the same field of reference we call modern scholarship, as a philological undertaking, the standardizing compilation and arrangement of linguistic records under the direct sponsorship of the imperial and, later, the postcolonial state, which made them the basis of cultural modernization policies (in the many and various historical “language reforms” — really writing reforms in the first instance — of the nineteenth and twentieth centuries).

Which is why recommendations for mixed-case letters, numbers, and punctuation in “passwords” — the point of which is not to be “random” (they can’t be), nor to be non-memorizable (though that is an effect), but to retain some chance of remaining, as we used to say of our telephone numbers, “unlisted” — are older than you may think.

But that’s not my point here (and the issue is more complicated anyway).

My point is that the history of password security is not only a segment of the history of security techniques, beginning with currency watermarking and security printing, continuing with modern state militarization, imperial conquest and colonization, internal policing, border control, and secular education, and leading to the automated, and in many cases already consumerized biometric techniques of today (your new iPhone 5S’s Touch ID). It is also a segment of the linguistic and literary history of computing, a history that I think we in the literary humanities can cleanly task ourselves with both chronicling and analyzing. This is the project I’m pursuing in my current work, and it represents the kind of work I am encouraging and supporting as best I can. I think it offers literary humanists a project in their own domain of unqualified expertise, requiring no abjectly asymmetrical, likely to be fatally inexpert (at a level hardly to be ameliorated by shallow “interdisciplinary” “collaboration”), and in any case unreciprocated adoption of the methodologies and values of other disciplines. It even provides a structure — that is, a productively objectivizing framework or model — for the historical analysis of the melancholia that Mehta’s essay so skillfully, essayistically indexes, and which anyone ever sincerely drawn to “literature” (out of either a conservative or a progressive impulse!) must likely feel once in a while.