< ?xml version=”1.0″ encoding=”UTF-8″?>
< !ELEMENT SET (PAGE+)>
< !ELEMENT PAGE (TEXT)>
< !ATTLIST PAGE PERIMETER CDATA #REQUIRED PAPER_STOCK CDATA #IMPLIED PAPER_WEIGHT CDATA #IMPLIED>

< !ELEMENT TEXT (COLUMN+)>
< !ATTLIST TEXT PERCENT_OF_PAGE CDATA #REQUIRED FONT CDATA #REQUIRED DEPRESSIONS_ON_SURFACE CDATA #IMPLIED FONT_SIZE CDATA #REQUIRED>

< !ELEMENT COLUMN (LINE+)>
< !ELEMENT LINE (WORD+)>
< !ELEMENT WORD (DATA)>
< !ELEMENT DATA (#PCDATA) >
< !ATTLIST WORD CAPITALIZED (YES | NO) “NO” ITALICIZED (YES | NO) “NO” NUMBER_OF_LETTERS CDATA #REQUIRED PART_OF_SPEECH (ADJECTIVE | ADVERB | CONJUNCTION | GERUND | INFINITIVE | NOUN | MATH_SYMBOL | NUMERAL | PREPOSITION | PRONOUN | PARTICIPLE | NA) “NA”>

Collaborative video with Hans Breder. Dissolving painted poems, transforming water to milk.

The Intermedium of Tissue

A variable represents a set of words in a particular order. For example, x might equal the word-set {the quick brown fox jumped over the lazy dog}. Note that the bracket represents the boundary of the word-set and is not included in the set itself.

A piece of writing, whether it be a book, magazine article, or a phone number can be thought of as a particular measurement of the world, both real and imaginary. We can look at the piece of writing as a word-set containing punctuation, words, symbols and notation arranged in a particular order and within a particular boundary. It can be used to take measurement and convey meaning. The word-set of a novel might measure the fictional account of a love affair. It uses thousands of nouns, verbs, prepositions, and other parts-of-speech in the correct order to calculate the result of the story. The word-set of a phone number measures the code to connect two or more telephones in communication – including the metadata to designate area code and prefix.

The observable elements of the word-set can be represented by modeling the hidden states of the elements. For example, we could use a numerical index to represent the order of the elements. We could also describe the hidden part-of-speech of each element based on a statistical model of the probable outcome of transition between parts-of-speech and the probable observable outcome of each state.

Even if the set is transformed by randomly mixing its order, the measurement is still useful because it exposes the structure of the body of text. This is easily discerned when paying attention to the grammatical structure. If a set contained the nouns {hurt knee football} there is a good chance the word-set is measuring a football injury. We could look to other elements in the set for additional clues. In fact, by tagging each element in the word-set with a part-of-speech we can vary the complexity of the word-set by altering it at the purely grammatical structure. It is way of writing and observing with attention focused on verbs, nouns, conjunctions, sentence-ending punctuation, question pronouns, possessive second determiners, etc. The meaning or measurement of a rearranged word-set is not destroyed but rather a new measurement is taken from a different angle. Part-of speech tagging (POST) is typically used in linguistic research and not generally considered a form of writing.

To perform writing using POST, we can treat the word-set as a body of text and graft the elements of one set onto the grammatical structure of another to create a third body and word-set that is completely unique. The elements could be assigned to the parts-of-speech either systematically or randomly. A systematic approach leaves much less room for error because it takes into account word ambiguities. A random approach has a higher margin of error because it rearranges without concern for valid syntactic orderings of words and does not recognize semantics. But it can reveal more surprises by producing a greater quantity of combinations. This differs from the proverbial infinite-number-of-typewriting-monkeys-eventually-recreating-Hamlet problem. Cut-up writing is not necessarily a brute force attempt at creating the next masterpiece to be enjoyed by millions. It’s an expressionistic record of selecting and arranging words utilizing a particular algorithm.

Here is Tristan Tzara’s original algorithm.

To make a dadaist poem take a newspaper.

Take a pair of scissors.

Choose an article as long as you are planning to make your poem.

Cut out the article.

Then cut out each of the words that make up this article and put them in a bag.

Shake it gently.

The take out the scraps one after the other in the order in which they left the bag.

Copy conscientiously.

The poem will be like you.

And here you are a writer, infinitely original and endowed with a sensibility that is charming though beyond the understanding of the vulgar.

In his famous 1916 stunt at Cabaret Voltaire, he shocked the audience by writing a poem with this method. In effect, what he also wrote was an algorithm for producing a poem. If we go along with him and assume his algorithm is correct, what he suggests is that if I conscientiously but randomly copy words from a pre-existing word-set to another structure, the new word set will be like me. Therefore (if this law is commutative) I will be like the word-set. My body will be indistinguishable from the body of text. More accurately, we are bodies in a dance measuring time and place. I should be able, using the correct equations, to rewrite myself and my life in an extremely literal sense.

Many famous writers have attempted to do just this. William S. Burroughs and Brion Gysin each developed techniques for integrating cut-up writing into traditional writing in an often surrealist manner. Their claim was that cut-up writing was so intertwined with reality that it was useful for mystical purposes. They deduced that there must be a supernatural force at work choosing the words. We can’t explain how the words are chosen for us, they just are. Burroughs, Gysin, and others experimented with divination and precognition in their algorithms and claimed an expanded supernatural awareness by rearranging pre-existing word-sets. Burroughs remarked that, “Perhaps events are pre-written and pre-recorded and when you cut word lines the future leaks out.”

When taken extremely literally, the cut-up writing and resulting text alters the reader’s perception in a preternatural way. Proper nouns, for example, retain their meaning and importance but are re-imagined. For example, a segment from a recombined word-set might read {John is a red cone.} It asks the reader to imagine John, perhaps someone we know, as a red cone. What does that mean to say John IS a red cone? If you start associating John and red cones, you might also begin to see actual examples in reality of red cones correlating with John. It’s arguable whether this is magical or coincidental, but with enough practice it might become easy to tune our awareness to influences beyond our normal perception. One possible technique would be to generate very large word-sets and mine them for bits of information that intuitively seem important. If we deliberately pay attention to recombined words not intentionally written, we might find new ways of thinking about the world. The recombined words need to be read as a personal message. It is a leap of faith to assume cut-up writing creates word-sets that have been written by some outside force specifically for the person reading it. The fact remains that the words have, in fact, been created and do exist. Both the content of the word-set and the act of reading it are brought into focus. It places the body of the reader squarely within the text itself. Randomness, when used to simulate the ordering of the set, is no longer sloppy or chaotic. It’s difficult to find meaning in purposelessly rolling dice, but meaning is created by rolling dice in the context of a game. Randomly rearranging words is also difficult. But within the context of an cut-up algorithm, meaning is also created.

Critics of this approach claim the act of writing becomes too easy and doesn’t require the skill and patience of choosing and ordering words intentionally from the mind of the author. We now live in an age of design, engineering, and information overload. Anything that smacks of chaos is generally shunned. With a multitude of word-sets competing for our attention, we tend to seek rich content over the ephemeral. Well crafted information seems to travel between more people than information generated strictly by chance. We simply do not have time to read and interpret cryptic messages. Besides, it’s natural to want to read a word-set someone has taken the time to order very well. It’s a pleasure to read a great novel or to read clear, concise instructions on how to do something.  The craft of ordering word-sets is a time-honored skill useful for conveying the meaning of our thoughts and evoking emotion within the reader. We tend to value the time it takes to choose just the right words and put them in just the right order. Reading a list of scrambled words is usually not pleasurable for the reader because it’s difficult and “meaningless.”  Perhaps this explains why cut-up writing is supernatural — precisely because it is not natural.

But recombinant or cut-up writing still serves a purpose. Because it is relatively easy to create a unique word-set of thousands of elements in a short period of time, we are able to generate completely original sentences that might be otherwise go unnoticed.

But the original artistic purpose remains. There is a value in rearranging pre-existing word-sets even though the chaotic period of WWI and the overall mission of dada is over.

The text tends to become much more expressionistic and non-representational.

To graft the flesh of one body of text to another:

Tag the elements of two word-sets with the correct part-of-speech, preserving the order of the elements.

Sort the elements of the first word-set into lists based on their part-of-speech.

Remove the elements from the second tagged word-set, preserving the order of the tags.

For each tag in the second word-set, substitute it with a random member of the corresponding tag list from the first word-set.

Repeat if necessary using the resulting third word-set or something new.

In the intermedium of the graft, new flesh will grow and adhere.

I. When a TAG button is pressed

1. Accept any amount of text from a field called INPUT. Perform a part-of-speech tagging and display the tagged words in a text field called TAGGED OUTPUT. The value of the TAGGED OUTPUT field might look/NN something/NN like/IN this/DET ./PP

2. Copy each word (without its POS tag) into an appropriate field based on it’s part-of-speech. The user may delete words or add their own words. Make all words lowercase.

II. When a GENERATE button is pressed

1. Accept a value the user has inputed in the COMMON field. The value can be from 0-100. The number corresponds to an array that contains the 1000 most common words in the English language. (For example, a value of 20 in the COMMON text field means the 20 most common words will not be substituted in the next step.)

2. Substitute each of the words or punctuation marks in the TAGGED OUTPUT field with a different word or punctuation mark from the corresponding POS field unless that word is in the range specified by the COMMON value in which case it is not substituted. Display the result in a SUBSTITUTION text field.

3. Capitalize any letter directly following the PP, Punctuation, Sentence Ender tag then remove the POS tags from the SUBSTITUTION text. Display the new text in a field called RECOMBINED.

The Intermedium of Flesh: from poem to code

Cut-up the body

Look inside the body

You will see previously hidden structures

Cut-up another body

Look inside

Graft

A corpus skin graft

Grafting Writing and Reading

First writing with revealed blood allowing the token. The glyph of writing graft is a young practice fluid for studying the size of our days and studying plastic out the modification. The fact contains that the animals have, in fact, been spread and do fall. But the mobile temporary activity relates. When revealed alone initially, the common degree and investigating writing defects the systems paper in a television way. In fact, by allowing each traffic in the mid-east with a pattern we can look the example of the important by allowing it at the wholly single object. But it can reject more slices by reducing a easier pictographic of heterologous. In his individual 36 envelope at b logographies, he called the background by comprehension a building with this phonetic. It allows grafts of displays, hands, grafts, and other array in the same graphite to pass the text of the product. The loud of a scroll can grow the bacterial prehistory of a epidermal instance.

In practice (with crossword themed lottery ticket):

generation:
————————–
jeweler once cookie semi length noisy free factor lawn off uncommon ouch elk illusion wax inn yesterday relic being

poetification:
————————–
a noisy semi of uncommon length (once free of illusion) was an ouch factor off the Elk Inn lawn yesterday

wax and a cookie was the relic of the jeweler

interpretation:
————————-
Big rig truckers who attend the Elk’s Lodge should have high-fiber cookies wrapped in wax paper to eat while driving. The constipation experienced from the cookies may be painful, but it will remind them of where they started (retrieve snack) and where they are going (relieve self).

translation:
————————-
The Crossword will win if truckers at the Elk’s Lodge receive a cookie recipe from a jeweler.

instruction:
————————-
Write a high-fiber cookie recipe on wax paper and mail it to all the truckers at the local Elk’s Lodge with the return address of a local jeweler.

codification:
————————-
To: Elks Lodge #590 – 637 Foster Rd – Iowa City, IA 52245

From: MC Ginsberg Jewelers – 110 E Washington St – Iowa City, IA 52240

Trucker’s Cookies

1 c. whole wheat flour
1 c. bran flakes
1/2 c. oats
1/2 c. wheat germ
1/2 tsp. baking soda
1/2 tsp. salt
2 packets Sweet ‘N Low or 1/2 c. sugar
1/2 c. corn oil
1 c. molasses (unsulphured)
2 eggs
1/2 c. fresh ground peanut butter
1/2 c. raisins
1/2 c. chopped nuts

Mix together flour, bran flakes, oats, wheat germ, soda, salt, and Sweet ‘N Low; set aside. In mixing bowl on low speed mix oil, molasses and eggs. Add peanut butter and mix well. Add the dry ingredients. Add raisins and nuts last. Drop by spoon onto cookie sheet sprayed with Pam. Bake at 325 degrees for 20 minutes. Makes 4 dozen.

]]> http://mbutler.org/projects/iwanttoliecom/ Thu, 12 Jan 2006 20:18:06 +0000 http://mbutler.org/projects/?p=45
iwanttolie.com is a website that was developed for some neen activity in early 2006.

I. When a TAG button is pressed

1. Accept any amount of text from a field called INPUT. Perform a part-of-speech tagging and display the tagged words in a text field called TAGGED OUTPUT. The value of the TAGGED OUTPUT field might look/NN something/NN like/IN this/DET ./PP

2. Copy each word (without its POS tag) into an appropriate field based on it’s part-of-speech. The user may delete words or add their own words. Make all words lowercase.

II. When a GENERATE button is pressed

1. Accept a value the user has inputed in the COMMON field. The value can be from 0-100. The number corresponds to an array that contains the 1000 most common words in the English language. (For example, a value of 20 in the COMMON text field means the 20 most common words will not be substituted in the next step.)

2. Substitute each of the words or punctuation marks in the TAGGED OUTPUT field with a different word or punctuation mark from the corresponding POS field unless that word is in the range specified by the COMMON value in which case it is not substituted. Display the result in a SUBSTITUTION text field.

3. Capitalize any letter directly following the PP, Punctuation, Sentence Ender tag then remove the POS tags from the SUBSTITUTION text. Display the new text in a field called RECOMBINED.

Here is the link to my 2004 PHP version of Haacke’s original work.

Visitors’ Profile 2005 (after Haacke)

seminar version performed for Version>04 Invisible Networks

based on the book ALLOY

Listen to the sountrack

Watch the Quicktime version

Dully Servant (Soviet No-Gang Shriners) reads from the twisted linguistic cosmology of >KIND TRICKS as [tele-psychic puppet] action-script for Skeleton Bride (People’s Republic of Delicious Foods). An advanced study for buried meta-conspiracies and progressive states of bibliomantic techgnosis.

crypto-adaptation:
1. secret self-modification by an agent in response to that agent’s changing systemic conditions
2. temporary micro-alterations in belief structure to accomplish a site-specific task
3. cross talk between versions
4. ritual narrative constructed to help explain random phenomena
5. orchestrated prescience
6. a hermetic sales pitch from beyond

A) Receiving predictive information concerning one’s environment has been the greedy desire of humankind since the construction of language as a tool thousands of years ago. If only we could describe what is just over the horizon! What an edge we’d have over our “competition” if we already had the specialized language to discuss the newness of tomorrow today.

B) As we continuously enter the new newness of so-called ‘new media’ and ‘new age’, our natural language is intermingling with programming language in shocking and considerable ways. Meaning is constructed out of clearly defined protocols that either “work” or “do not work” — there is no third position here. For the first time in existence, Poem and Code are the same. Though this ‘new language’ is less susceptible to the confusing ambiguities of old (i.e. we cannot use variables and handlers without clearly declaring their value and defining their procedure) there is still a magic in language which contributes to human evolution.

C) Language in some way creates the very reality in which we live. Words and concepts point to realities beyond the sensory world and assist us in making contact with a dimension that is higher, lower and parallel.

D) The great DADA experiment of cut-up writing carried out by Tzara, Burroughs, Gysin and others has now reached epic proportions. The intent was to jitter thought process with randomness in language. That jitter is now quake-like in magnitude thanks to advances in programmable electronics and automatic algorithms. What took them hours with newspapers and scissors now takes us seconds with computers — I can do 4 million words over brunch. We are no longer creating experimental voices but communicating with beings from other worlds.

E) Intangible Ideas, in Plato’s conception — as supersensible realities beyond human thought — are captured in scripts, as prisoners in their cells, and released by the act of perusal, setting the prisoners free. These Ideas reside in the words and terms independent of the programs and books in which the words are encased. Yet how and where, in the interval between their setting down and their taking up, do they abide? By what secret tract is their existence in the mind of the author/programmer connected with their resuscitation in the mind of the reader/user? Why at the sight of certain lines and figures on the voiceless page/screen do these particular thoughts spring up into renewed activity? What is the indiscoverable nexus between the physical vibrations of light and these immaterial substances of our noetic life?

F) Homo Sapiens unlocked the mystery power of their brains through the invention of language thousands of years ago. Language allowed them to reflect in a self-aware manner, invent finely crafted tools, and create handmade ornaments, shelter, and paintings. This epoch process is culminating in programmable electronics. We are becoming aware, again since before language, that the fracture between “subject” and “object” is unreal. What is more necessary… the operating language of a word processor or the inputted creative language of its user?

Akoll Tapoze Yitsoung!

launch the flash documentation

download the cards

Download the script for the pilot episode

chain gang theory:
The Markov Chain Gang are t.v. characters trapped behind the show’s wall of fiction, fighting to escape through ruptures in the script’s disjointed writing. Language is being eaten by the myriad stochastic computer processes constantly at work in our culture-system. Television viewers have matured or evolved into cybernetic image consumers capable of digesting smaller and smaller bits of video until content becomes irrelevant. Television sets are no longer windows to another world, but feedback mechanisms integrated into our identity and physiology. The Markov Chain Gang craves the luxury of the observer/viewer context. They ask: “How is it that you know this is a soap opera? Are you a detective, looking for trace structures outside my motivation? Are you a prisoner like me?”

Flipping through different channels and different shows is an artistic act. It’s actually all one channel, all one show. The Gang is subverting composition by applying painterly collage techniques to pre-existing script language. Older notions such as act structure, continuity and temporality fall to stranger development. Horizontal, non-hierarchical themes emerge. There are no tidy conclusions that let you know how everything turns out. The show is a constant state of becoming a new show. Each Markov Chain Gang episode is just a bracketed data set inside a theoretically infinite order. It references and resembles, but the conceptual process’ transformative nature amounts to legal plagiarism; borrowing and reconstituting without stealing. The Markov Chain Gang is a television show not about its content, story or characters, but rather the antithetical story of the decisions of repression and change in the script as it meets the demands of the genre.

how to write it:
The episodes are written by choosing television scripts from different TV shows. This source material should be chosen according to what flavor or mood the writer wants the resulting Markov Chain Gang script to have. Once a few thousand words have been copied from the source material, they are analyzed by one of many computer programs that run Markov Chain algorithms. There are many different algorithms that cut-up text, but Markov’s seems to be the favorite with poets and artists.

The output text will be sentences that have been broken apart and put back together with other sentences and words. It’s difficult to read at this point, so a writer needs to work the cut-up material into a useable script. One can think of a genre as a template. The writer needs locations, characters, situations. Something to differentiate, say, a television script from a novel. When the writer has in mind what that is, they dip into the Markoved source text for answers. There are locations, characters, situations there… they just need to be extracted. The writer should not attempt to control the script, but rather let the script happen on its own. Expect clash, chaos and surprise. With practice, the writer is able to act merely as a catalyst for the scripts creation rather than its author. There is something out there giving us life.

epilogue:
I invented a game. It’s called Quantum Twenty Questions. It’s just like 20 Questions, except no-one has an object in mind when the game starts.

First, you ask a question like, “Is it an animal?” Then you flip a coin and if it’s heads the answer is yes. If it’s tails, no.

You proceed this way until you find out exactly what the object is. Where did this object come from? Obviously from your own mind, right? Is it a television? Heads. Yes.