The Unprintable

(August 8-16, 2011)

Alan Sondheim

“A lur is a long natural blowing horn [...] The word lur is still very much alive in the Swedish language, indicating any funnel-shaped implement used for producing or receiving sound. [...]” Wikipedia.

So blowing might be a kind of scattering or a calling, might be a kind of signal. And I think of Lurid Numbers as a kind of calling, calling-forth both readerly and writerly textuality, but also an interpenetrating and entangled intermediary, the catatonic machine which might or might not open itself to the distending of symbols, graphemes. The machine reads itself; it doesn’t like to be bothered. It doesn’t like to read Lurid Numbers, because what is lurid, sleazy, is a bone in the works, can’t be read. And apparently the publishing machine of Blaze-Vox, Blaze-Lur, couldn’t read, broke down over, returned to catatonia (for what bureaucracy likes to be bothered), over what should have been a simple reproduction from code to paper – which did not, could not, occur? What do you say to this, to the insertion, beyond the readerly and writerly text, of the text which breaks the machine, which refuses purity?

Sandy Baldwin

Yes, I agree. You provoke me to think and respond. I see the lur as producing a signal, one that problematizes signaling behaviors. I see this in the phrase “very much alive in the Swedish language”: the word as “alive” is the problematization of signaling behaviors, where such behaviors as part of our modern understanding of communications that necessarily contain / are contained by systematization and closure. For modernity, to signal is to invoke an institutional culture that handles the call. (The lur problematizes any final response to your provocation as well, and instead leads me to a more “funnel-shaped” answer.) The blowing lur, the funnel-shaped implement (I read it as impediment, which already is the bone in the throat, the marrow in the grapheme), is a problematic signal. Lur also makes me hear a broken lure. Lure is a decoy, but also allurement and enticement. The lure is camouflaged and does not yet reveal the spoilage of the lurid. The breakage from lure to lur is the object escaping, the rotting out of the object. The kernel of matter that presupposes a machine – there is a machine threshing every “grain” of the real; the machine that makes everything itself, that ensures all is systematic and well-formed (machined) – is lost, leaving only an echo. In all this, the orphic song, the stirring of animal spirits, the original poem, the return from the dead. It is here that the where the machinic is not machinic.

You lead me to think that Lurid Numbers can’t be read or printed because the writing did not occur – which would require the iterability of closed world of the machine – it is not authored. I admit to be radically uncertain of the “author-function” of this text with my name on it. I recall that I did not intend for it to be unprintable or unreadable, at least not in any straightforward sense, though I do think I intended it to be undigestable. It remains in a lurid- or sleaze-world, so thin, flimsy, barely lit.

I wonder, in terms of the end of your provocation, is such a text still a text? I think of text as requiring reading and writing as advanced, modern processes; machinic processes, I suppose. I do not know if I can claim this for myself – not to have written a text – but I do wonder how far “text” can bear a lack of purity. I think more in terms such as “wryting” (which you coined), which suggests to me a kind of intensive, non-textual investment …

Alan

Wryting is of and in the midst of the body; Lurid Numbers skitters across the body, but drags, scrapes the escarpment of the body as well. “Text” is like “nice” – it can bear anything, it is not that smooth, but the family of usages containing it are smoothed out – in the case of Lurid Numbers, perhaps dilapidated as well. I’d think of you as an author for the simple reason – the text wouldn’t otherwise exist.

This might be the result of a crisis. To paraphrase L. Apostel, “The Justification of Set Theories” (in Logic, Methodology, and the Philosophy of Science, 1965), substituting texts for classes (the original is based on the comprehension axiom), “At the beginning of a textual procedure, when nothing whatever is known about sources or filters, complete freedom is given as to the construction of texts (1). When the textual procedure is already established, and when the results of earlier operations are stable, the construction proceeds without rearrangement: texts now are embedded in a formal and canonic history, which is taken for granted and is machine-decomposable; the roots are there, in a very classical sense (2). However, when a crisis is reached and the procedures hitherto followed must be altered, textuality returns to the original sources, accompanied by new procedures and filters (3). When the procedures operate without crisis, but when it is imperative that procedures at one level should be strongly distinguished from and yet completely determined by the procedures applied on the earlier level, we have (4).” Let’s think of (4) as conservation, and the third as permanent crisis, under the signs of capital and the fast-forward simulacra of signs. Then we’re in the skitters and jitters, aren’t we? Sufficient crisis, and the market breaks down, the machine breaks down, your self-identification as author breaks down. And that’s when things get interesting, when the machine reaches a state of indigestion.

What causes the crisis? On one hand the formal condition that upper and lower ASCII don’t exhaust the pixel-by-pixel gridwork of inscriptions; there’s always the glitsch, the fury beyond acceptable bandwidth. Most of our lives – in fact the phenomenology of our being-in-the-world – is conditioned by filtering-out, not only so the clean and proper body of the signifier may make itself felt against our skin, but also because anything else creates an ontological shift where the matrix of communication, community, disappears altogether. The crisis is of course the implosion. It’s that. It’s those. It’s when “a” replaces “the” down to the rootlessness. Lurid Numbers unroots numbers, signs, and symbols, replacing “a” itself with an unknowable little-object-a beyond the spectrum, smothering the spectrum. It rests, remains, sleeps there. So you do a pdf!

Sandy

Yes, I agree. It would be possible – maybe? I think? – to map the text as crossing thresholds or crisis points. Can we think of all texts in relation to their procedure and whether or not there is a crisis, an alteration? The quote from Apostel also situates crisis in a relation to origins, or pathology of the text, keeping in mind that these are inscriptive origins and symptoms, origins in the topography of transcriptions.

I was speaking to the poet and critic Chris Funkhouser about the problem of printing Lurid Numbers. He brought up the possibility of an unpublishable book. He saw this as the future of increasing information interlinking, where all possible sentences are linked into the net already; all possible texts already generated in some way; no “new” text possible without being plagiarism from the first. This is, of course, a version of Borges’ “Library of Babel,” and also the nightmare aspect of Ted Nelson’s Xanadu, where the density of hypertextual transclusion shuts down any new utterance. In practice, it is already being implemented with plagiarism detection systems such as Turnitin (my university purchased this software and wants it to be used on student papers). I found myself agreeing with Chris that this is a potential both for contemporary forms of writing, flarf being the example he used, and for the inseparability of the practice of writing from the apparatus of word processing software, in the largest sense. However, Lurid Numbers is unprintable rather than unpublishable. We could say that these are two forms of crisis. Unpublishability is a problem with the institution of authorship, and the confusion between economies of creativity and capital. Unprintability is a problem with the technical production of scriptive materialities, and a confusion between the conceptual potential of digital simulations and the institution that anchors these to a machinery of materialization, or more precisely, to a machine of printing out and handling. The confusion is already there in the simulation: MS Word and other software continue to simulate a page, while at the same time offering something else entirely. Lurid Numbers does have its own unpublishable pretexts, and unpublishability does relate to a materialization of the text – it is reasonable to see the legal institution of authorship in this way – but, whereas publishing involves a materiality of expression – creative expression capitalized through the author’s name – we could see the unprintability as a materiality of phenomena and the body. Is this the crisis and implosion (or explosion)? Of the book as object, as in my hand and before me, as technical product … The crisis of unprintability is worth studying on its own, and in relation to other crisis taking place in institutions of the book.

Finally, thinking of what you say of the cause of the crisis: you set the glit(s)ch as the “fury beyond acceptable bandwidth.” Perhaps “fury” is the correlate of crisis? A passion, a vehemence, beyond any law – and therefore in relation to the laws and institution of print and authorship – a fury that seeks the work, and – why not – seeks justice for the writer. Why justice? If unpublishability and unprintability involve the writer as name and capital (the writer’s head), then the fury of writing unpublishes and unwrites, leaving nothing but the body and the roiling of its interior. The work, the writing, the scriptive orgins – all these come from and go to this fury.

Alan

Just a couple of points here. No matter how much interlinking occurs, statistically, fantasy notwithstanding, all possible sentences etc. really constitute such an inaccessibly high number, that all the processing power of the universe would never complete the task, even given the 13+ billion years we’ve been around, even given incredibly high-speed processing. As you know, n-furcations quickly spin out of control. The second point relates to pain, and the inexpressibility of severe pain, and I want to end my contribution here, in relation to what might be foregone, the McGill Pain Questionnaire notwithstanding. So a few brief words left behind in lieu of an other text.

Inexpressibility occurs because of the difficulty of expressing interior states that might not have a clearcut symptomology (as thirst does, for example) – and also because severe pain derails speech and language and thought, as the internalized horizon of the flesh is muted or screams in abeyance. All of this touches on the pain of the signifier and its inexpressible relation to death – and all of this touches, as well, on the unprintable. The unprintable then is returned, as if by media mail, to the body that produced it. And from this moment, your text, as pdf, opens up and opens, as pain disperses, is dispersed, as the symbolic ultimately collapses, as is its wont…

What good is a writer if he can’t destroy literature? And us… what good are we if we don’t help as much as we can in that destruction? – Julio Cortazar

Geoffrey Gatza, fearless director of BlazeVox, that “publisher of weird little books,” took the final proofs of Lurid Numbers to his printer on July 27, 2011. Lurid Numbers is a collection of more or less “codeworked” text – much like i did the weird motor drive, my 2007 book with BlazeVox – written through simple computer scripts and word processings, and through my own impulse, inquiry, and idiocy. The next day he came back with some odd news in the form of an email from the publisher:

—— Forwarded Message
From: <no_reply@createspace.com>
Date: Tue, 26 Jul 2011 12:02:16 -0700 (PDT)
To: Geoffrey Gatza <editor@blazevox.org>
Subject: Files for Lurid Numbers, 978-1609640705 require your
attention

The interior and cover files for Lurid Numbers, 978-1609640705 have been reviewed.The cover file meets our submission requirements; it is not necessary for you to make any revisions to this file or upload it again.The interior file does not meet our submission requirements for the reason(s) listed below. Please make any necessary adjustments to your interior file and upload it again by logging in to createspace.com.The interior file contains pages with unreadable text or “jibberish” which we are unable to move forward with as it may appear as a file error in manufacturing. Please submit a revised interior file for further review.

Best regards,
The CreateSpace Team

As we like to say in academia, the email was “interesting,” that is, it could be read as linked to a number of other cultural domains and protocols. The relation of the “interior” to the “cover” repeats and takes part in the history of the “book,” where the cover is the limit of the work of writing; the cover is the enclosure or partition, the  event and inscription of multiple institutions: of authorship (if the work is under a pseudonym or in some way unsigned, the copyright page still must contain an author’s name, even if it is “anonymous”), commerce (the name of the publisher, legal descriptions of rights and regulations, and so on), and archiving (library of congress number, date of publication, etc.). Along with this, the fact that the interior of the book was somehow rotten or broken seemed both a judgment and a simple fact of this book. It was even better that this was expressed iconographically in the cover, which did meet “submission requirements.” I saw the cover as a submission of the contents to a single image. The cover shows a butchered and already old, slightly rotted fish. The image is photoshopped, neon and definitely lurid. Geoffrey directed me to this image, and I loved the combination of the repulsive and slimy, the mundane and organic, with the software transformation that keeps it real but artificial as well. It did indeed seem to submit and capture the interior. And then: “the interior file contains pages with unreadable text” seems to me an almost ontological statement, one that rubs against the proximity between the written work and the human. We may submit, we may submit a cover – ourselves – that meets requirements (of culture, of others), but our interiors are often quite different, unreadable. I also appreciated the misspelling of gibberish, suggesting a virality of the unreadable text into the printer’s email. Finally: “we are unable to move forward [...] as it may appear as a file error in manufacturing” suggested to me an event or force of the work beyond the interior file, a hidden explosion breaking the apparatus that machined it, and seeping or flooding past the cover.

In short, I was pleased to become more than just another job for the printer, to become a new process and something beyond the routine. At the same time, I was concerned, wondering what would happen with my interior file, as it were. I found out five days later, on August 1, 2011, when Geoffrey informed me in an email that

they cannot print this book and there is nothing I can do about it. [...] this is something completely new and I have to say I am perplexed by the mechanizations of modern times. The printers are not opposed to you or your work, this is a situation of a printing process that is highly automated and this registers exactly like a printers error to their machine. It is not a human that we must cajole into agreeing that this is art, which was my first take on this, as with the printer who cannot spell. This is a matter of a quality control camera that will reject books that look like this. I talked with a lot of people in the company and even had my lawyer call them to see if great weight would move the immovable. But no, their system will literally stop when it would try to produce your work.

A writing that stops the computer system, the very system designed to print out writing: what more could I ask for? What more frustrating thing, as well, so close to the print out of the book, that fetish object that makes authors out of writers? I was judged by the computer to have written something, i.e. it did not deny that there was an input that it could judge, but it evaluated my writing as unprintable, as a writing that can only remain in the space of the computer, within the possibilities of software. My interior file was bummed out but also filled or luridly lit up with a deep pleasure.

The act of writing is related to the absence of the work, but is invested in the Work as book. The madness of writing – this insane game – is the relation of writing; a relation established not between the writing and production of the book but through the books production, between the act of writing and the absence of the work. [...] To write is to produce the absence of the work (worklessness, unworking, [désoeuvrement]). Or again: writing is the absence of the work as it produces itself through the work, traversing it throughout. Writing as unworking (in the active sense of the word) is the insane game, the indeterminacy that lies between reason and unreason. – Maurice Blanchot

BTW, the book is here: http://www.blazevox.org/index.php/Shop/Poetry/lurid-numbers-by-sandy-baldwin-244/

Access image linked from University of Cambridge

Part 1 is here.

The printed institution of intellectual property holds that works cannot be reproduced “without prior written permission” (as the legalese runs). The printed work at hand is always documentary evidence of the printer’s permission for that work, whereas any additional permission – the permission of the subject to write and read in the face of the work – requires a chain of additional writings (prior written permission).

If chmod is tied to the body’s ontological topology in the network apparatus, it also renders this topology inseparable from crowds and communities. Consider digital rights management (DRM), perhaps the most intense site of debate around permissions. The debates around downloading, torrents, music sharing, and so on, are inseparable from the problem of controlling permission and its constraint to specific users and communities.

Keep in mind that on the net, domains of permission are separated into user, group, and world. Symbolic notation sets read, write, and execute permissions for each of these domains, so -777 is represented as –rwx/rwx/rwx. The first notation is left empty for a file or set to a “d” for permission on a directory. The next octet or notation sets permission for user, then group, and then world. A single string for topology of crowds. Take these as shifters: on the net the shifter can no longer be simply the familiar markers in language. Permission for user or group or world speaks those communities; speaks the community of one (user), a specific group, or anyone at all on the net. Group membership is complex; it can be temporary, overlapping, exclusionary. The chmod command can also set a “sticky bit” that allows or limits mass changing of modes. The sticky pit aggregates and speeds up operations. Stickiness involves retaining the read-only segment of a program in memory or “swap space,” so that users can create but not write files. The point is to prevent users from changing or deleting each others’ files. As a result, user permissions are collapsed into group and world permissions. The implications for digital writers are simple: where previously I saw myself as a creative writer, as modeled on the solitary artist producing from the depths of my psyche; in truth, I am shifted to be part of a more open and indeterminate group of writers who share constrained but communal permission. In this way, the voice and subject of digital poetics is never fixed but fluctuates between the plural and the singular through the setting of permissions.

Sandy Baldwin at E-Poetry in Barcelona, image by Chris Funkhouser

Each domain of permission demarcates the place for inhabiting and projecting onto the space of electronic writing. Once again: permission spaces is the netting of the subject. A site of “group” ownership is fundamentally different than “user” only, and so on, while “world” opens permission to all. Each case attempts to constrain the scope of the indicative (deictic) function of the shifter. DRM controls constrain permissions to certain users and groups, while sharing communities (torrents, etc.) open permission the world. The crux is less ownership than permission to access and the community (user, group, world) that is allowed this permission; or rather, ownership is within the domain of permission. Lawrence Lessig writes of the danger of the “read-only” internet. Perhaps unintentionally, he frames his argument with the terminology of permission. His call for a necessity and importance of a “read-write” internet is built on the space of permission described here. We are far from the pale remediations and idealizations of the writer and reader that still dominate discussions of digital writing and reading.

To write and to read text assumes at least a minimal narrative. Text is text because it is narrated. The structural narratology of Mieke Bal insists on this narrative premise in every utterance. Every <text> is readable because of the framing <I narrate <text>>. Even the blankest screen is an utterance. This minimal narrativity is tied to the deictic function of language. In the structural linguistics of Emile Benveniste, deictic utterances point to and invoke a world. Benveniste spoke of the signs used in the subject’s act of utterance as the “formal apparatus of enunciation.” The apparatus makes the subject present, an autobiographical apparatus allowing the subject to say and write “I.” Following Roman Jakobson, “shifters” are the linguistic deictics understood as speaking the subject: “I” or “me” or “Sandy” do not possess semantic value but syntactically speak the subject.

What are shifters on the net? Shifters enact the deictic function of language. Through shifters language “says” being. All digital writing is enunciated. What does it speak? What does digital writing utter? In part, it speaks permissive enframing and containment by the operating system. The indicative function of deixis references the operating system as the background world of the net. The system contains and holds language.

The psyche of the subject is circumscribed by the closure of the site. Permissive closure as shifter places and locates the subject’s enunciation. Nothing exits this closure. All that the subject is, is uttered here. The speaking subject is entirely a product of this apparatus. The shifter operates with a “punctual” reality concept (“wirklichkeitsbegriff”), in Hans Blumenberg’s sense. The “point” (or punctum) of the shifter holds the subject and system together.

Time is involved as well. The shifter fixes the time of the subject and creates a “pure present.” In digital writing, this is the real time of the screen, or the temporality of the “work.” This deictic time-space siting is at work in every surface, every web page, every electronic word, every font and pixel, and every space.

What kind of subject knows that they are permitted? A pervert, of course. The psychoanalytic terminology of “perversion” is specific here: I write and read and execute by assuming the desire of the other. The knowledge that allows digital writing and reading is the pervert’s knowledge. I only know the other’s desire because I act it out (I execute and perform) in my desire (in my reading and writing). A psychic model of digital poetics is found in the creativity of the pervert who wishes to recreate the world in the image of an other that can only be found precisely through this imaginary. What a pervert I am! I gaze at the screen or at the pixel or at the font, I imagine through the apparatus, and play until I am fulfilled. This is digital poetics.

How does chmod relate to the absent body?

It is too easy to emphasize the closure of the site. Permissions are openings. Setting permission to -777 or -775 allows access to write and alter files. A site can be taken over, owned, defaced, renamed. The chmod -setuid can allow trojan horse or other malware entry through “privilege escalation.”

To grasp the shifter as a sign and as part of a language is to inhabit a particular culture and a particular habitus. To see the site as closed and to take permission for granted is to punctually and permissively close the horizon of my culture, to say “I am a writer” and “I am a reader” with the confidence of a shared community and writing materials and techniques. In doing so, the sememe is narrowed to particular domains of knowledge. Or rather, to directories and files. Digital writing and writers today are caught in this narrow, constrained into file systems. The “emerging” field of digital writing is constituted through this closure of knowledge. We know what constitutes a work and a writer. Or rather, a file and a directory. What is a digital writer but a directory, a space of permission with the capability of siting files (works)?

Think here of Heidegger’s “enframing” technology but in terms of acts of permission rather than of the unfolding of being. The net is already a culture for us. It is thick with the other and our desire towards the other. It is lived and cultural. It is part of our world. Permission is at work here. The application and its features are permitted as objects of understanding. The “application” or technical object is a foreclosure of the shifter. Only in this way can we comfortably operate (write/read/execute). Protocol is definable because of this closing off. Protocol must not be understood as technical specifications. No, the reverse is true: every technical specification is the fictionalized residue of the body sieved and emitted through permissions. Protocol is a narrative of the body’s presentation. Permission is one of the protocological features that formalize actions, controls responsibility, and elaborate institutional personas. It is a concrete form of culture. The real but absent body is splayed across the files and directories of the permissive site.

To take permission for granted is to believe in the net’s existence. Could things be otherwise? Surely the opposite is the case? The net is fragile, built on the fly, barely or not at all existent, constantly happening and collapsing around us. (Think of the origin of the internet in Paul Baran’s desire for “survivable communication.” The net is the phantasm of this survival, always claimed in theory, sought in practice, lost in truth.)

Back with the shifter: we locate ourselves uncertainly in this projection. It is a partial source of the subject, a clot or coagulate without amounting to a body. The body is absent in every shifter. On the one hand, authority withdraws. The discourse of “protocol” following Galloway, or of “network culture” following Castells, or other cognate formations, formalizes the chmod command (and all similar commands), as if commands were at work as a performative ground of all writing online. Execution – the most significant but least graspable aspect of permission – is assumed everywhere. The net works. If permission must be given and set in practice, it is easier to assume the stability of the network in theory.

Listen to this: permission is prior to the deictic site. Or rather, permission opens the utterance through the possibility of narrative and quest. Deixis results from permission. The deictic display or pointing requires context. It invokes or carries semantics rather than containing a fixed semantic meaning. Enunciation always is other. The time of the screen is elsewhere, historical. A fundamental poetic point: permission creates mission. Narratives are stories unfolded of permission given. The materiality of media is emitted from permission to use the apparatus, as tools, as raw material. At the least, this means there is a voice caught up in the apparatus, a voice that must be “sourced.” Voice as material for enunciation but also as distant echo from outside the material. The apparatus allows speech but also speaks of allowance.

Is writing anything other than producing a work or a file? Is the digital writer anything other than a site or directory? The siting and existence of each, within the withdrawn authority of the net.

Every work is addressed to me. I court your permission. Do you give permission? There is no shifter here. There is only words on blank. There are never shifters, never any reference, never any world. All these formulas assume permission given and taken for granted. I can not know if I am permitted, I can only write. In the “absence of the work” (Blanchot) I write without guarantee, transitive and infinite, never knowing if I am permitted or not. The subject surges beyond the site of enunciation. Permission is absent, is everywhere, is uncertain, exorbitant and excessive.

Did you chmod -777? Do you permit me to, do you give permission? To write it to have permission, and this is true for every page and line and word and space. Permission is given for all writing on the net, from email to web pages. My writing is the unfolding and overflowing of your permission. I thank you, I celebrate you, I revel in you, but I also revile you, denigrate you, turn away. Why must I ask for permission to write? Writing only with your permission, I read you in every space and every word and every line and every page. This is the case even when you did not explicitly give your permission. Permission is withdrawn. I cannot be sure that I am permitted to write. I write in hope of your permission. I read your permission. I imagine that you give it and so I am able to write.

Why are discussions of digital writing not devoted to permission? Is this not the fundamental horizon of our writing? Digital writing, as digital and as writing, must be approached in this way. Or for that matter, why do we not discuss other aspects of our frantic, intense, overwhelming writing the net? We think reading is taking in of marks on a technologically enframed surface. Possibly we understand an author at the end of a circuit creating these surfaces and marks. The author is a function in the circuit, as is the text. We discuss links and Flash technique, generative and dynamic writing, form and narrative voice in virtual environments, and so on. Do not all these topics close down the netting of the subject in writing? Or rather, we take for granted the subject that enunciates and expresses on the net because this granting is necessary to our conceptual field of writing, held together as it is by instrumental topics, such as those in the list above. (I must say the only site I find discussing such topics is Alan Sondheim’s Internet Text, as ever the only philosophy of the net.)

Writers: is this the case? Do we not write because of compulsion, desire, passion? And also, we write through inertia, fatigue, anxiety? All these worldly orientations are missing from discussion of digital writing, but they are not missing from writing the net. Who does not feel the weight of fatigue in connection delays on the web, or deep anxiety at lags in email communications? These are inner orientations, part of one’s own disposition in relation to a body that inhabits the web intensely yet absently.

I will write of and with your permission.

What is chmod and what are permissions? The first 1971 implementation of UNIX included the chmod command. File permissions were a basic feature of UNIX and continue in all subsequent *NIX systems (POSIX, LINUX, etc.). Other file systems adopt related permission system. Web sites typically run on a UNIX-like system. They utilize htaccess and similar requirements to set permissions. Some file systems, such the Macintosh, refer to “privileges” rather than permissions. The semantics are similar, although “privilege” has a much more specific legal history as the designation of an individual’s entitlement granted by a government. By contrast, “permission” is traceable to individual intentional acts of granting a special access or right, an exception not covered by the generalized legal notion of universal rights. Is it any surprise that permission is also etymologically related to mission, to journey, to quest? Permission grants an opening to narrative. There is always a subject and a drama of permission.

Think of Robert Duncan describing the “opening of the field” as “a place of first permission.” Chmod opens and operates on a space of permission: the file system. A file system is built around methods for storing and organizing files, typically within directories. It starts from a directory as a file that contains the names of files within the directory, including itself. “The most important job of UNIX is to provide a file system,” write Ritchie and Thompson, as they described and created the operating system. They add: “A directory behaves like an ordinary file except that it cannot be written on by unprivileged programs, so the system controls the contents of directories.” Every space on the system is folded within itself, according to permissions. By default chmod is applied to a directory and only secondarily to files. Every directory and every file is a space of permission first, and only then a writeable or readable technical feature within the apparatus.

Chmod sets permissions to read, write, and execute directories and files within a directory. To write. To create a file, to edit it, to delete it. A file is written only if permission is given. Web pages are no different. Every file is subject to permission. To read. To show the contents of a file, to see the name. A file is read only if permission is given. To execute. To execute a file. To run a program. A file is executed only if permission is given. What if we approached digital writing in this way? What if we inquired into the permissions of each digital “writer” and “reading” and “text”?

Permission is an existential mode, a way of being for directories and files. It is often described through the “symbolic notation” of r/w/x. Permission to read (r) the file; permission to write (w) (or edit, create, re-name the file); permission to execute (x) the file. These characters name the permission given. Adding or creating files in a directory – permission to write – is adding names to a directory listing. Write (w) is permission to write names. The absence of a character or a dash (-) indicates a void, without permission. Symbolic notation writes (notates) the topology of permissions in the space of files and directories. It writes the shape of entities inhabit within that space. Is this writing not a minimal level of digital poetics? Instead of symbolic notation, chmod can also use octal notation to describe permissions. Octal notation uses base 8 numbers, typically in a string of three or four digits, allowing the precise state of permission to be expressed in a single number, such as 777. (A common joke is the octal notation for the symbolic setting -rw-rw-rw-“: 666 or “Permissions of the Beast.”)

What are we permitted? Permission is given to treat digital objects – files, directories – as textual objects to be written and read. Their qualities as object are textual because of this permission and do not preexist it. I cannot write or read a file unless I am permitted. The file becomes textual through permission that permits the objects to be “like a language.” Yet the levels of permission described by symbolic or octal notation are not to be understood as instrumental access to inscribe and to read, in the sense that handing a pen and paper to someone grants direct access to writing instruments. Technically, permissions give the right to use a “system call” on the file or directory covered by the permission. The system call instructs UNIX to make an edit or allow reading or execute a file. Not permission to write or read or execute, but permission to instruct the operating system to operate on files and directories. “I am a writer” or “I am a reader” means I am permitted to call on the system to write or read in my place. The system is the horizon of actions and meaning. Digital writers do not write but call on the system to do it for them.

The chmod and related chown command means files are assigned. The commands are acts of constitution. Every file is constituted as a file through permission and ownership. If the file is owned and its existence is formed through permission, does this not fit the conditions of intellectual property? If to create a file is to create intellectual property, is a file an act of expression? Think here of Cornelia Vismann’s fascinating description of the administrative logic of files. Files are always a problem of processing and recording as much as reading. Institutional power transcends or exceeds the files.

But what if permission were a struggle? What if we refuse it when it is given, or take what is not offered? To invent permission: what if this were the condition of digital poetics?

Part 2 is here.