Jan van Eyck Academie 2009 Yearbook

This book alphabetically composes an unfiltered list of all the written words produced at the Jan van Eyck Academie in 2009, and counts every recurrence. The texts, in English and Dutch, which the list takes apart, range from the projective abstractions of theoretical research to the instrumental specificity of administrative emails. The language of the academy, in all of its institutional registers, is thus enumerated as a totality.
A necessary surplus to this written word count, as image and speech, finds its place on the book's jacket and cover.


notes

vocabulary of the academy, index of the academie
is the set of all w such that w is an element of V (the vocabulary) and w has property P
{w | w ∈ V & P(w)}
(set theory articulation of procedure)

rules for when an image appears

how a machine reads
how to show a human intervention
how to show human input in fact comprises this machine?
when the machine gets to an image, it doesn’t know what to do with it ...

formulate and execute rules for formal iteration
for instance: when and how the gray background emerges and varies.
also: when images are introduced.
these rules will conflict in some cases.
the final book is a performance of execution.
certain cases will be illegible or overlapping.

include a block:
regular expression experiments. for example: all instances of “a [blank] man”; or 3 or more words in a row beginning with the same letter; or “x and other y’s”

incorporate blank spaces ...

include linguistic data: number of characters, words, etc

alphabet order everything

index w keywords
concordance of these words

the walls have ears

execute aleatory procedure for all formal elements including typeface
a flickering grey background
the lightness of index versus dirty heavy transcripts
every page there is a 1 in 5 chance a transcription line will appear (how will it be chosen)

reference of dark screen deleuze

images
fit into index?
or lay them out on their own page
see all together
perhaps need to ask someone to produce images
—what would be the protocol for this?
there are so few images in comparison to text
—ask for more?
fit according to a different grid

word list
what is the rule for reducing redundancy?
—turn over to human?
—keep singular; keep noun or verb?; don’t add pure stem if it doesn’t occur in text
what is the rule for english/dutch?
first: eliminate redundancy; then: eliminate stopwords; then: decide on further insignificance; finally: make keywords

and then: does human perform some action from keyword list? generate a new utterance?
the conclusion is a new utterance/collective utterance

transcription:
transcribe
turn over to someone else to underline

how are keywords identified?
way to produce a hyper set of keywords a subjective territorialized set from the institutional set
highlight keywords?
make staff perform a reading, have them identify the keywords
what are the keywords, the control words of this discourse?

determine list by eliminating stop words and comparing to another corpus
determine first set of keywords by having researchers underline statements
and by some automated term extraction query
then have researchers review this list (2nd level) and underline the words, a la raymond williams:
the words in which some historical change or action is being worked out.
the words which do the signifying work of the academie.

if the index is the map of jve’s symbolic order, how does the real intrude, rupture?

something to interrupt the alphabetical order
another type of page

red type and black type
red type

library notices

what has not been said
what has been said but not captured
white space
put in relation to an outside, a second text. if there is no word from text1 which matches text2, there is a blank space.

incorporate blanks

erase index words which are not key

speaking against the index

24 letters, 12 months, ? researchers

what program might relatively quickly produce a concordance?

an experiment in information science:
deliver texts to a set of readers, each of whom is asked to index the texts.

want to make the index speak
to select phrases

how the corpus is processed

map this system
image / text index

One hundred thirty seven thousand and sixty-one