Theory games: from monstrous puppetry to productive

Theory games: from
monstrous puppetry to
productive stupidity
Chris Bigum
Griffith Institute for Educational Research
Theory games
I would argue that academic work is
less a form of birthing than a monstrous
puppetry. We reanimate sometimes
dead sometimes living scholars in our
own work and make them co-perform
the theories and practices we are
intellectually invested in
Charlotte Frost
Familiar theory games
• Theory as spice, something to trade
on/with: have theory can travel
• Theory as salt: There is not enough
(too much) salt in your thesis, paper,
proposal, application.
• Theory as fad/fashion: low-carb
theory; theory snacks; deep-fried
theory …
• Theory/Practice (meat and three veg)
And then there are turns
(the menu may have changed)
• Material, nonhuman (Latour; Law; Mol;
Jasanoff)
• Practice (Schatzki; Kemmis)
• Posthuman (Braidotti)
• Non-representational (Thrift)
• Performative (Butler; Haraway; Barad)
Why?
Where, what and how is the human in
a world of prosthetics, printable body
parts, bots, and bio/nano-engineered
nature?
“the four horsemen of the
posthuman apocalypse:
nanotechnology, biotechnology,
information technology and
cognitive science.”
Braidotti
The GRIN technologies:
Geno; Robo; Info; Nano
Kelly
Any sufficiently advanced
technology is indistinguishable from
magic
Clarke
I am interested in machine
magic
Ever have a “wow” reaction to what a
machine has done?
We have had plenty of the other kind.
Fractals
Artificial life
Simple rules to determine if a new cell
appears or dies
Coming to terms with magic
Two options:
Fake it
We know what is going on. It’s just the
same old, familiar .... (McLuhan’s rearview mirror)
Oooh! And we can visualise it!
And if we can’t domesticate it we can
ban it.
Be productively stupid
Focussing on important questions
means being ignorant by choice
Be comfortable with being stupid
Willing to say: “I don’t know”
The new machine magic on the
block
Machine learning, big data, deep
learning...
A Primer: the Eureqa story
A perfect storm for old AI
algorithms
• Lots of data
• Internet of things
• Moore’s law
Meanwhile in the sciences of
the social
• Frantically adding the adjective digital
to everything
• Silly debates about what is big
• Re-doings of familiar enactments of
machines
• (Mandatory use of Powerpoint at conferences)
• Boosters/pragmatics
• Critics/doomsters
• Digitally homeless
• Artificials/algorithmics
How we think about machines in
general and the new machines
matters
• Predictive models
• Decision making
• Can include everything that can be
measured
• Algorithms that “learn” (code for
improve)
Becker
“Everything present in or connected to
a situation I want to understand should
be taken account of and made use of. If
it’s there, it’s doing something”
What are these new machines
doing?
•
•
•
•
Stats
Bayesian (generally)
A lot of brute force computing
The more data they are fed the better
they get. Think Google, xMOOCs,
Amazon etc.
Prospects for the sciences of the
social
A new & deeper ditch
Qual
Quant
Or
a little bridge building
In any old, current or yet to be
circumstance in which work is
delegated to a machine there is
always a re-distribution of
competencies between machine
and humans
Any consideration of the posthuman
that ignores negotiations between
humans and stuff, things, machines is
like physics without dark matter and
energy. (not even close... no cigar)