Anything Said Is Said by an Observer

A short paper you can hold in one hand, and every line opens onto another one. I love the depth of the von Foerster paper and that it is short.

It is a talk, really. Heinz von Foerster standing up in front of a room of cyberneticians and opening with a theorem that is not his.

A theorem that is not his

He gives it to Humberto Maturana:

"Anything said is said by an observer."

Then he adds his own corollary, in his words "in a modesty":

"Anything said is said to an observer."

Two sentences and he has tied together the observer who describes, the language that connects two observers, and the society that two observers already make just by talking. He compares it to the chicken, the egg and the rooster. You need all three to have any of them.

I had just listened to Maturana and von Foerster together in a recorded conversation, and here is von Foerster baptizing a theorem with Maturana's name.

We do not see that we do not see

Then he turns to the blind spot. Every eye has one, where the optic nerve leaves the retina, and we never see a black hole there. We do not see that we have it. Von Foerster puts it in five words, "we do not see that we do not see", and calls that a deficiency of the second order. The only way through, he says, is a therapy of the second order.

And to make the point he reaches for Carlos Castaneda. Don Juan, the teacher in Castaneda's books, keeps trying to make him "see", to fill the blind spot with something new, and Castaneda keeps filing what he cannot explain as illusion.

Castaneda and Kuhn in the same few paragraphs of a cybernetics paper. That is a big part of why I love it.

Castaneda's books are widely thought now to be invented rather than fieldwork. Von Foerster's point does not need them to be true. It needs the shape of the story: a student who cannot see what he has no explanation for.

Kuhn, again

Then he turns to Thomas Kuhn.

I read The Structure of Scientific Revolutions a long time ago, at Bellarmine Prep or in college, I cannot place which. I was lucky at Bellarmine, introduced to advanced topics by brilliant teachers while I was still forming.

Von Foerster takes Kuhn's word and then turns it.

	Kuhn	von Foerster
a paradigm	a model for linking descriptions, specific to a culture or a language	the same, he borrows it as is
why it falls	it starts to fail, shows inconsistencies and contradictions	sometimes because it works flawlessly
the example	anomalies piling up	Copernicus, at the moment Ptolemy's system was at its most accurate
the shift he sees coming		from the properties of objects to the properties of the observer

A paradigm dropped for being perfect. He calls the old path flawless but sterile, and says some people can no longer pursue it "by their life".

He gives an example I did not expect either. Obscenity. If you show Mr. X a painting and he calls it obscene, you learn a lot about Mr. X and very little about the painting.

The question underneath

Then objectivity. The rule that the properties of the observer shall not enter the description. He asks how you could describe anything at all if you had no properties that let you describe. He does not let you escape into subjectivity either, because the negation of a nonsense proposition is another nonsense proposition.

So he asks a new question.

"What are the properties of an observer?"

And the catch is that we are the ones who have to answer it. We have to observe our own observing.

That is when he gives the line I quoted in Where I Learned to Look: the cybernetics of observed systems is first-order cybernetics, and second-order cybernetics is the cybernetics of observing systems. Gordon Pask had drawn the same two orders, he says. In the first the observer enters the system by stipulating the system's purpose. In the second the observer enters by stipulating their own.

He ends on responsibility. If we do not name our own purpose, somebody else will name it for us, and "I just obey orders" becomes the excuse. I did not expect a paper about observers to end there.

A decade earlier, the same title

Margaret Mead got there first, and she told it as a story on herself.

She went to the founding meeting of another new society, the one for general systems theory. Nobody in the room knew who she was. She suggested they use their own theory on the society they were founding: how big it should be, how it should grow, how it should connect to the rest of science. She was "slapped down without mercy". Apply the ideas a society is founded on to itself, of all the silly ideas.

So when the young American Society for Cybernetics asked her to speak, with her name known, "not disguised as a casual crank" this time, she made the suggestion again. What in thunder are we founding? Who should join, and who should not? How do we keep from getting steadily older until young people cannot talk to us? Should we plan for the society to end?

I heard the Atlanta story before I read it, either from her own voice or someone retelling it. I cannot say which anymore. I have been listening to a lot lately.

	Mead	von Foerster
title	Cybernetics of Cybernetics	Cybernetics of Cybernetics
given to	the new American Society for Cybernetics	the American Society for Cybernetics, a decade on
the turn	a society about self-correcting systems should correct itself	a science of observing should account for the observer
what she or he asks	what are we founding, and when should it die or transform	what are the properties of an observer
how it ends	"I commend them to you"	social responsibility, and a strong hand for the organizers

Mead points the tools at the society, and von Foerster, a decade later, at whoever is holding them.

On my own page

This October, outside in the sun, I wrote this in my notebook:

Observe the observer. Place the system itself onto the system.

And later on the same page, "the paradigm shifts with which time is measured". Mead's move and Kuhn's word, in my own handwriting. The page is Observe the Observer, posted as written.

I am keeping it all organized as best I can, on paper, here on the blog, in my head and on a few YouTube shelves. Some days the map gets ahead of the reading. These two papers are the reading, and short enough to carry around.

Where this sits

Before I opened either paper I wrote, in the reading list, that I would hold off on the words second-order cybernetics until I had read von Foerster and Mead. I have now, and the words stay.

The audio primers and printable sheets for both papers are in The Rest of the Map. Where these two sit among everyone else is The Cybernetics Family Tree, and the whole reading path is at /cybernetics/, with the talks on the watching path. Why the observer idea was already familiar when I got here is Where I Learned to Look. Maturana's own short paper, on time this time, is what A Moment Measured reads, and the eigen idea von Foerster named is where Eigen to Myself went the night after. All of it hangs from Complexity and systems.

Sources

Source	What it supports
Heinz von Foerster, "Cybernetics of Cybernetics" (1979), address to the American Society for Cybernetics; reprinted in Understanding Understanding (Springer, 2003)	Maturana's theorem and the corollary, the observer, language and society triad, the blind spot, Castaneda, Kuhn and Copernicus, objectivity, observed vs observing systems, Pask's two stipulations, responsibility
Margaret Mead, "Cybernetics of Cybernetics", in von Foerster, White, Peterson and Russell (eds), Purposive Systems (Spartan Books, 1968); excerpt at the ASC	the Atlanta story and her questions to the new society
Thomas Kuhn, The Structure of Scientific Revolutions (University of Chicago Press, 1962)	the paradigm and why it falls
Carlos Castaneda, The Teachings of Don Juan (1968) and the books after it; von Foerster cites Tales of Power (1974)	Don Juan teaching Castaneda to "see"
Truth and Trust: Maturana and Von Foerster, American Society for Cybernetics channel	the recorded conversation between the two