Skip to content


A veiled complaint about Elon Musk tweeting on how Neuralink creates the possibility of "conceptual telepathy."

I want to bring a curious set of expressions to some attention, from a Heinz von Foerster paper entitled Notes on an Epistemology for Living Things. At the end of that work, there are a set of more specific notes, which are intentionally formatted in the same way as propositions from the Tractatus.

The exposition on Proposition 8 interests me the most, but before we get to that, I'm going to try to distill some formalism that the paper uses into something simpler. The propositions leading up to the eighth articulate a few different ideas about the relationship between "an observer" and "an environment."

Here's my best effort at a brief summary of propositions 1-7:

  1. We [observers] experience the world as various stable or changing objects. The phrase "changing objects" is strange -- we see things become different, but simultaneously they retain some kind of stable identity.

  2. This paradox is a subtle indication that "changing" and "stable" are not properties of an objective world that exists outside of observers. There is no "change" or "stability" of objects or events themselves, but instead, of some kinds of representations.

  3. Observers behold distinctions (differences) in two domains: space and time.

  4. The notions of "an object" or "an event" follow from the fact that observers must embody some mechanism which computes whether or not two things (events/objects) are distinct -- each individual is doing something which determines the boundaries between things.

  5. It follows that objects and events cannot be considered to reside in "an objective reality" shared by all observers -- rather, objects and events are things that observers are individually doing -- specifically, they may be regarded as functions that are continuously computing whether some set of things are similar/distinct. Paraphrasing the paper: "the inter-relations between objects and events is a personal affair, whose constraints are anatomical or cultural."

  6. What we are calling "the computation of an object/event" is embodied by a physical (in our case, organic) mechanism which may itself be computed-upon or involved-with other mechanism -- like a higher-order function which maps between functions.

  7. If objects and events are functions embodied by organic matter, we can call "an organism" a kind of functor (still, embodied by something physical) which acts upon itself, so-as-to compute objects and events in such a way that it maintains its own integrity. In this sense, an organism (specifically here, an observer) is a fixed point in a recursion.

That's a lot of dense context, but I think it's important to set up the weight of these particular statements in the magical eighth proposition which are actually relevant to what I'm trying to say:

8. A formalism necessary and sufficient for a theory of communication must not contain primary symbols representing "communicabilia" (e.g., symbols, words, messages, etc.).

8.1 This is so, for if a "theory" of communication were to contain primary communicabilia, it would not be a theory, but a technology of communciation, taking communication for granted.

8.2 The nervous activity of one organism cannot be shared by another organism.

8.21 This suggests that indeed nothing is (can be) "communicated."


And in case it's not clear, the word "communicated" is quoted there because it stands for the misunderstanding that something is moved. In that way, "conceptual telepathy" makes the following mistakes:

  1. Of taking knowledge for something which is somehow distinct from the entire mechanism of "a relation (continuous computation) between a particular brain and the world as-it-is-beheld (another continuous computation)."

  2. If you have moved knowledge, you have necessarily, physically cloned the whole system of observer-plus-environment, or otherwise, have physically created something new which is complex enough to be an observer (and continues to embody that process).

How do you move a whirlpool from one river to another river? (Is it even the same whirlpool?)

Addendum: If my outline of the paper feels like solipsism, you are not alone -- but if you're interested, you should try to read the paper and maybe some of his other work, because the conclusion is actually the opposite! I can't do his whole position justice, but it's more or less the conclusion that what we call "objective reality" is a kind of consensus -- a fixed point in co-ordinating of all our actions together.