The Internet Unconscious

On the Subject of Electronic Literature

Sandy Baldwin

• Publisher: Bloomsbury Academic
• Year: 2015
• ISBN:
  9781628923407
• Edition: 1
• Topic: Social Sciences
• Subtopic: Media Studies
• Index: Social Sciences

Winner of the N. Katherine Hayles Award for Criticism of Electronic Literature from the Electronic Literature Organization There is electronic literature that consists of works, and the authors and communities and practices around such works. This is not a book about that electronic literature. It is not a book that charts histories or genres of this emerging field, not a book setting out methods of reading and understanding. The Internet Unconscious is a book on the poetics of net writing, or more precisely on the subject of writing the net. By 'writing the net', Sandy Baldwin proposes three ways of analysis: 1) an understanding of the net as a loosely linked collocation of inscriptions, of writing practices and materials ranging from fundamental TCP/IP protocols to CAPTCHA and Facebook; 2) as a discursive field that codifies and organizes these practices and materials into text (and into textual practices of reading, archiving, etc.), and into an aesthetic institution of 'electronic literature'; and 3) as a project engaged by a subject, a commitment of the writers' body to the work of the net. The Internet Unconscious describes the poetics of the net's "becoming-literary, " by employing concepts that are both technically-specific and poetically-charged, providing a coherent and persuasive theory. The incorporation and projection of sites and technical protocols produces an uncanny displacement of the writer's body onto diverse part objects, and in turn to an intense and real inhabitation of the net through writing. The fundamental poetic situation of net writing is the phenomenology of "as-if." Net writing involves construal of the world through the imaginary.

VIII

Plaintext

“Plaintext” is not a clearly defined term. As a starting point, consider the range of definitions in the National Institute of Standards and Technology Glossary of Key Information Security Terms. It starts with the context of cryptography, defining plaintext as “Data input to the Cipher or output from the Inverse Cipher,” and secondly as “Intelligible data that has meaning and can be understood without the application of decryption.” Thirdly, it also defines it simply as “information.”¹ Internet RFC 2828, the Internet Security Glossary by R. Shirley, defines plaintext as “Data that is input to and transformed by an encryption process, or that is output by a decryption process.”²

Note that plaintext is both the source and result and the throughput of the process. Plaintext as part of an encryption process may seem distant from plaintext as just vanilla unformatted text, but the fact is we cannot abstract “text” under conditions of today’s information systems from encryption processes. While the IP protocol and the link layer of the net are unsecure, drifting, and open, all higher layers of application and transportation—in short, all layers of human read-write-execution—are encrypted. Indeed, RFC 2828 self-referentially includes the Internet itself as one entry in the Internet Security Glossary. To use the net is to engage in encryption.

Cryptography, literally “hidden writing,” shadows the literary in that it theorizes the partial readability or unreadability of a text passing through transformations and handlings of a writing system. The cryptographic principles for “handling and filing” of plaintext, that is, for “the original, literal texts of messages sent in secret or confidential code,” are serious and strict. In his lectures training Signal Intelligence Service and National Security Agency cryptographers, William Friedman states:

There should be in existence only a definite, limited number of copies of the plain text of the message cryptographed in any secret code or cipher and these copies should be carefully controlled in distribution, handling, and filing. […] The plain text should never be filed with the cryptogram itself. The further the containers in which the plain-text copies are kept from those in which the cryptograms are kept the better. These containers should never be kept in the same safe. If only one safe is available, it should be used for the plain-text versions, keeping the cryptograms in a locked filling cabinet.

Also:

Plain text and its equivalent code or cipher text must never appear on the same sheet of paper for final copy or for filing purposes. Work sheets should be destroyed by burning.³

What interests me here is the maintenance and labor required to isolate the plaintext from its encryption, a distance that supports technical layers, cultural institutions, but also military logistics and rules of engagement. Such maintenance and labor produces the aesthetics of the text. The force that binds and separates plaintext and encryption requires burning should they brought together. The conflagration built into these texts supports and is supported by law and ideologies of identity and security. The uniqueness of the text is inseparable from this conflagration.

The destiny of plaintext is isolation because of uniqueness, destruction to prevent reading. Plaintext rests on an almost metaphysical topology of materiality and information. It construes an exterior to information systems that interrupts the text with a demand. What demand? Plaintext interrupts to announce a community of writers beyond the military logistics and rules of engagement that determine information systems. The literary is inextricable from this interruption.

Under today’s conditions of “Pretty Good Privacy”—where the NSA no doubt intercepts all our communications down to the merest post-it note, but probably cannot be bothered to read the critical mass of data beyond flagging key words such as “turban” and “peace” using data sniffing protocols such as Echelon—under such digital encryption standards, every byte of data—and this means every pixel and every screen of every computer everywhere, that is, every byte and its monitored output—figures in the open secrecy of plaintext and encryption. It is as if plaintext were the secret core of information. It is as if a total decryption would produce a textual superset of all possible knowledge.

Plaintext is the fantasia of the library and the cryptoanalysis of text is a model of the depositing and retrieval of knowledge in archives. Of course, the crucial thing is this “as if” which metaphorically repeats the notion of the archive without clarifying it. There is no plaintext but only systems of encryption and decryption, institutions of secrecy and revelation. Information is always totally secret and totally open.

At the same time, we all know that plaintext is also commonly used as the “lowest common denominator” file format in a system. If I am editing a text with someone else, perhaps using Microsoft Word, I may ask them to send it to me in “plaintext.” I expect an unformatted document, written in Wordpad or another text editor. Similarly, plaintext is the default format of email messages. It is a non-format for the simple transmission of a message, a technical format that defines “without format.” It is where the message is nothing but itself and is not the medium. It is amedial.

This is highly problematic. One implication of this concept of plaintext is inseparability from exemplary proximity to sharers, those to whom we need to get a message, and where format means surplus added to and possibly obscuring the message between friends. My point is that this notion of plaintext requires communities where text is read for the message, where the message is the medium. All of which implies the notion of a minimal communication circuit of nothing but sender and receiver. One thing that characterizes plaintext and ties it to the problem of the literal, and in turn to the problem of the literal qua literary, is this implication and invocation of friends, of a community of lovers of literature, and the implication of this community as a determination of its plain-ness, shifting focus from technicality to community. Even the plainest of text is addressed to another. This means plaintext is never plaintext.

March 11, 1968

President Lyndon B. Johnson approved the American Standard Code for Information Interchange, or ASCII, on March 11, 1968. Johnson was simply institutionalizing what was already the case: from its first implementation in 1963, ASCII was and is the standard for encoding data. All communication, transmission, and storage on computers or other telecommunications devices is ASCII-encoded. All characters must be ASCII-encoded or converted to ASCII and then encoded. Information is the transformation of ASCII encodings. We may think of the web as a zone of diverse multimedia, but its protocols are all coded through and with ASCII. Objects on the net are addressable because of a wrapping of ASCII and it is through ASCII-compliant protocols that communication is possible. It is the medium of our media.

Plaintext as unformatted text typically means ASCII-encoded text. This most basic text is encoded and encrypted, in order for it to appear as text. ASCII is the protocol name for the plaintexts we read. ASCII connotes plaintext and source code, and therefore the ethos of programmers and hackers, in contrast to proprietary formattings and layers used by applications such as Microsoft Word and its.doc filetypes.

At the same time, there is a politics of national characters in ASCII. The initial ASCII lookup table was ordered for English characters, with pounds sterling or kanji as complex exceptions. ASCII is the precedent for the recent Unicode universal character encoding, which assigns its first 128 characters to the ASCII character set, and debates continue over encoding of Korean or Ogham or Klingon characters. The story of escape codes and swap tables for alternate character sets is important but lengthy; my point is the contested and heterogeneous nature of the standard from the first. The problematic relation of universal encoding to local context is captured in the Japanese term “mojibake,” which names the incorrect mappings we see as rows of boxes and other glitch character when software attempts to render Japanese script. ASCII and now Unicode provide the framework for the net’s phenomenology, determining appearances and conditions of experience.

What does ASCII encode? Perhaps everything. Michel Foucault defined the lower limit of discourse with the example of the given-ness of “a handful of printer’s characters.”⁴ Friedrich Kittler’s critique broadened this media a priori beyond print technology. ASCII would seem to offer a similar handful of characters, but this is not the case. Instead, it is a heterogeneous field of transformations, a Zermelo formalized set, an immaterial and non-medial a priori. The definition seems simple enough: ASCII is a seven-bit character encoding specifying a relation between characters and bit patterns. Of course, ASCII is an eight bit code but uses seven digital bits to encode each character, from 0000000 to 1111111. I will return to the eighth bit below. ASCII maps a specific character to each value in the entire range, resulting in a total of 128 characters for the binary digits from 0 to 127. The first thirty-two digits (0 to 31) and the final digit (127) are non-printing characters.

ASCII encodings are transformations, mappings of f(a) to f(b), but they are also conditions of appearance. They will always be made phenomenal and material. The non-printing characters memorialize incorporated and optimized bodily skills. One way this is evident is in the need for escape codes or code breaks corresponding to keyboard SHIFT or SPACE keys, accommodating the manipulation and collation of material writing surfaces.

The difference between the DEL and BAK (or backspace) keys is complicated. In a computer where the writing surface is treated as an immaterial support for text, a backspace effectively deletes characters. In the physical medium the backspace pushes the typewriter carriage back a space to allow overtyping or revision. The physical surface is moved incrementally without necessarily affecting the character imprinted there. Does BACKSPACE mean delete or does it mean simply a physical move? Not a straightforward question but a matter of whether you deal with characters floating on an immaterial writing surface, or physical media with character markings.

Many of the non-printing characters are obsolete. Most of the non-printing characters are control codes used to alter physical mechanisms. The Carriage Return character originates with the typewriter, where the bar or button rotates and advances the cylinder or carriage holding the paper, and returns it to the left hand side of the page. In ASCII, the code commands printers to return the cursor to left hand side of the page. The locations, page size, printer type, and so on, are determined by particular output formats. Similarly, the non-printing character Form Feed causes the printer to advance and eject the current page and to begin printing on the next page.

All of these commands alter the physical mechanism that displays ASCII characters. At the same time, the commands themselves remain abstract and separate from particular printers, with the material particularities of manufacturer, parts, wear and tear, and so on.

The remaining ninety-six codes correspond to printable glyphs. These include punctuation marks, symbols, digits, and letters of the alphabet. 1000001 maps to the letter “A,” 1000010 to B, and so on. Lower-case and upper-case letters differ by a single bit, so lower-case “a” is 1100001.

The table printed in the original 1963 standard document reinforce the appearance of mapping bits to printed characters. The table’s courier typeface, familiar from terminal screens, seems to correspond to the bit patterns, since we are familiar with seeing both displayed in Courier. In truth, this is a matter of the “seeming” of representational practices. In practice, textual output was a secondary goal. The goal of the standard was not to define particular representations but to maximize possible encodings to a broad range of media.

ASCII is an immaterial grid. The 1963 standard did not represent encodings using digits but with a dense grid of eight columns with sixteen characters each, where each row corresponds to a four bit pattern (0000 to 1111) and each column to a three bit pattern (000 to 111). So, the upper left position in the grid was 0000 in the first four bits and 000 in the last four bits, and contained the null character. The bottom right position in the grid was 1111 in the first four bits and 111 in the last four bits, and contained the DEL character. The four columns in the middle of the grid contained the printable characters. As a result, the contents of a column or row could be transcoded into another column or row by simply changing a bit. This was quite intentional and the product of long debate, as some of the standard’s appendices make clear, so that symbols commonly used to precede numerical information (such as the #) occupy a separate column from the digits; and so the printable symbols cluster together visibly on the table, preserving the division between the visible and the invisible, the printable and the non-printable, at least in terms of the printed table. The 1967 revision to the ASCII standard added uppercase characters and some additional symbol characters, but the column and the dense centering of the graphic characters were largely preserved. Lowercase and uppercase only differed by a single bit.

As a result, ASCII is less an encoding than a transformation field dictated by the grid, by the printed table, where the position of the code is as important as its content. Codes are grid addresses and can be targeted as such.

The first code in the grid is a null character. The null character is not the number zero, which is code 48 or binary 0110000, nor is it a space character. It does not correspond to the appearance of a blank on the materiality of the screen. But it is also not the absence of inscription on the surface of appearances. Rather, ASCII code 1, digital 0000000, is a code for NOP or no operation. While the space or blank character may have effects in print, the null character may have effects in computer code. In C and other languages, the null character terminates strings. For example, “hello\n” tells the compiler to print hello and then to terminate the operation. The null character corresponds to no printed mark but does correspond to a medial difference. It is a condition of appearance but not a phenomenon in itself. In contrast to the inscription of characters on the material surface, ASCII exists in a dataspace of operations guaranteed by the negativity or syntactic gap of the null character. As Anthony Wilden and Brian Rotman note, true zero is crucial to placing and ordering of discrete digital elements. As the basic act in this space, as the first ASCII code, null mediates all other characters in the grid.

The eighth or parity bit is another digital supplement, a technical specification dictated by the alterity of ASCII. ASCII is often referred to mistakenly as an eight bit encoding, since microprocessor architectures at the time used an octet or byte, as many microprocessors continue to do today. In fact, seven bits are used for encoding, as already described. In a classic lack of foresight, the designers of the ASCII standard considered and rejected eight bit encoding as providing far more characters than the standard required or would ever require. In fact, eight bits means that in practice ASCII codes include an extra bit. The redundant bit travels with the code. It often functions as a parity bit for simple error checking. The odd or even value of the parity bit signals that transmission was successful or that the code is corrupted. The parity bit is a pure sign of digital existence, an ontological presence of ASCII in that other space of the net.

ASCII is less an encoding than a function. It does not describe the appearance of glyphs on screen. It does not prescribe type style, font, document structure, or markup. It does not specify how data will be recorded; it only prescribes encoding for interchange. The Foreword to the original 1963 standard states that ASCII is a “character set to be used for information interchange among information processing systems, communication systems, and associated equipment” and that the “means of implementing this standard in the principle media, such as perforated tape, punched cards, and magnetic tape” will be specified in later standard...

Table of contents

• Cover
• Halftitle Page
• Title Page
• Contents 
• Foreword
• Introduction
• I
• II
• III
• IV
• V
• VI
• VII
• VIII
• IX
• Notes
• Bibliography
• Index
• Imprint