Note from Paul to John, October 3, 1999

John,

I focus mostly on that which is incomplete or that I think is incorrectly represented in the one page description of your thesis.

A) "Events are the result on interactions"..... actually perhaps better said as "some class of events are the results of interactions between complex autonomous systems". This leaves open the specification of classes of events (birth and death) events that are not solely the consequences of "middle world" interactions.

It is easy to fall back into thinking as though the world was single leveled.

The following is written to make each word used meaningful within the theory of stratified reasoning.

"We assume that "some type of inner core" of each individual autonomous system (i.e., an intentional complex system) interacts with an environment. This interaction is via an "epistemic gap" that is interpreted in the mind body case as the Cartesian gap, and in the quantum mechanical case as the Heisenberg uncertainty gap. These "cross gap" interactions between a core of an autonomous system and the system's environment results in the restructuring of the middle world. It is often that we can see artifacts left by events."

We cannot be as Wittgenstein and not say all that can be said. The reason why word-smithing is vital to conveying the motivating thought behind stratified reasoning is that a certain cognitive separation is necessary between notation and language regarding two separate classes of "unobservables"; the substructure and the ultrastructure. Often times this cognitive separation fails due to influence of the history of thought, up to the present. For example Jung's failure to make a distinction between two types of unconscious cause - and thus the failure of Jungian thought to see a distinction between memory and anticipation.

B) The notion that events are neither completely determined nor completely undetermined needs to be drawn out with specific reference to the Russian Applied Semiotic positions, on plausible reasoning and argument based on uncertain and changing data. There is a specific history, in the Russian literature, regarding the issue of uncompleted and changing information. Merely stating "events are neither completely determined nor completely undetermined" is a paradox that must be fully and yet compactly addressed with proper word smithing.

C) Events do not give rise to observations, unless the event is a perceptual event. These perceptual events are called "interpretations". Perceptual measurement is involved. Events are events but may leave artifacts. Observations are events of a special type that are relative to an observer and which produces an "interpretation" (see the notions in Peirce.)

Finally, a few comments.

John, the notion of a "signature space" is a recent invention of yours. This is a good invention. Calling the series of observation artifacts a space reminds us not to make a "Rosen" category error, where one mistakes a useful formalism for the Object Of Investigation (OOI).

Observations and representations (artifacts) are not often the same thing.

A state is from a state space, and the state space is the collection of all possible opportunities that the representation of the OOI could exist in. States are formalism not ontology and thus it is very proper to say, as you do, that states "come from" the representations of observations. Gestures are transitions between states and thus are formalism also. The gesture comes from a complex interior and thus is not "caused" by the state. It may be partially caused by the previous state.

The pairing of the state and gesture is the central engine for which we are attempting to demonstrate in the bead game - first with Bayesian inference and then with the voting procedure.

The gesture has it's own periodic table of memory substructure and expresses into the space of category policies as action in the world. The state, on the other hand is merely perceived, as it, the state, is an artifact produced from the external world, with respect to the middle world complex system.

Last two sentences in paragraph four are perfect.

The gesture is an internal observation of self, expressed as a state to the world. The world then "observes itself, (the not-self of the self of the OOI). So there are two kinds of observations. These are relative and thus there is considerable intellectual maturity needed to see and accept the differences.

Many mainstream people will not accept this relativism, and thus it is necessary to be careful, prepare a defense if stratified reasoning is attacked based of our use of a relativistic (self/not-self) distinction. Ultimately the attacks are not grounded but will be aggressively defended based on everything between patriotism and religion. Thus we need to have a demonstration of how the system works, rather than attempt to justify the distinction based on philosophical argument.

The last paragraph needs to move into an exact description of classical Bayesian inference networks.

-Paul