There is a paper about my research into Mendoza's work here: Resuscitating a Dead Rabbit

High-Entropy Essays were part of Cybernetic Serendipity at the ICA, London, 1968 along with
COMPUTERIZED HAIKU . They are frankly bogus physics essays. For the versions Mendoza published
in the Catalogue of Cybernetic Serendipity, ouputs from several runs of the program seem to have been
appended to make longer texts.

Hardly anyone remembers High-Entropy Essays now. This is a sad thing. They are the
forerunner of all the computer texts that have tried to pass as human-authored.
However, Mendoza's subterfuge was quickly discovered. This happened about 1962.

Professor Mendoza tells the story:

"You might also be interested to know the origin of
this work. Professor Flowers (no less) had a theory
that students never actually learned any real ideas;
all they learned was a vocabulary of okay words
which they strung together in arbitrary order,
relying on the fact that an examiner pressed for time
would not actually read what they had written but would
scan down the pages looking for those words...The end
point was when a colleague from another university secretly
sent me some first year examination papers a week or so
before the exam, and I wrote suitable vocabularies (without
cheating)and copied down what the computer emitted...
the script was slipped in among the genuine ones.
Unfortunately it was marked by a very conscientious man
who eventually stormed into the Director's office shouting
"Who the hell is this man, why did we ever admit him?"
So perhaps Professor Flowers' hypothesis was incorrect."

The original program is lost. Mendoza published some of the High-Entropy Essays.
He also published a flow chart of the program. This is what I used when I came to program my version.
My first version was made in 2003. But I have now reprogrammed (2020) it in a way that is more
faithful to the original. The code is here.

Wayne Clements. Feb 2020


HOME