Bateson's two-stage differencing has been extended to indefinite recursive-differencing,
leading to limit cycles that are used in classification and pattern recognition.
In general terms, differences (or distinctions) embedded in signal-patterns of sensory
inputs are encoded. The patterns representing said encoding may also have
local distinctions (or differences). In order to interpret these latter distinctions,
further encoding-of-encoding is performed... and so on... so, operationally,
we have recursive encodings of distinctions that run indefinitely.
Luckily these recursive operations always converge on a closed cycle,
and the cycle as a whole is the interpreter of the input sensory pattern.
Best,
-- Joel Isaacson
PS technical detail in:
http://www.isss.org/2001meet/2001paper/4286330.pdf
From: Francis Heylighen <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
CC: [EMAIL PROTECTED] (Principia Cybernetica Discussion List), "Evolution, Complexity and Cognition group" <[EMAIL PROTECTED]>, Cybernetics Discussion Group <[EMAIL PROTECTED]>
Subject: [pcp-discuss:] Relational philosophy of information
Date: Mon, 19 Jul 2004 13:03:46 +0200
Seth:I was just thinking and came to somewhat of a conundrum. In the ever unsuccessful attempts to try to operationally define information, aside from Shannon and Weiner identification of it with entropy (I know its not exactly identical, but you know what I mean) the real problem occurs in trying to define it in terms of something else. What do we define information in terms of, matter? Energy? What does this mean?
What about Bateson's famous definition of information as "a difference that makes a difference"? The "difference" concepts refers to Shannon's "syntactical" view which defines information in terms of the possible number of states that a message could have (the more states, the more differences, the more potential information). The "making a difference" can be seen as referring to the "pragmatical" dimension of information: the message should not only be distinguishable, but relevant or meaningful, i.e. it should make a difference for the receiver, helping the receiver to make this decision rather than that one, and thus achieving a better or more desirable situation.
For example, if someone sends me the New York telephone book, but I don't know anybody in New York and am not planning to go there, this message contains a lot of information in the Shannon, syntactic sense, but none in the Bateson, pragmatic sense. I might as well have received several megabytes of random numbers and letters. On the other hand, if I was desperately trying to trace a person of whom I only know the name and the fact that she lives in New York, the message may be a godsend, and make a huge difference to my life.
Do we go the route of Fredkin and just insist information is the fundamental in which everything else is defined by?
The "difference that makes a difference" can also be interpreted in a more metaphysical, ontological sense as describing the fundamentally relational nature of reality: no phenomenon (difference) can exist on its own , it must somehow be related (covary) with some other phenomenon (another difference). This is actually the basis of my own philosophy and its "bootstrapping axiom", which says that distinctions (differences) are not given, but produce each other. It builds further on Leibniz's principle of the "the identity of the indistinguishables". See http://pespmc1.vub.ac.be/IDENINDI.html
In that sense, information (or rather relationality) is the fundamental in terms of which everything else is defined (including matter and energy). However, this is not the Shannon information which consists of independent "bits", but the Bateson one that consists of mutually dependent differences.
I'm not sure yet exactly how this ties in with a global brain, but you never know where inspiration will come from, you know? I just want to see what other people think?
The relation with the GB is of course that the GB is one huge network of relations along which information propagates, and as such merely a more complex organization emerging out of the simpler relational networks that have been existing all along... The intelligence of the GB consists in recognizing which differences make the more important differences, thus allowing it to filter out the meaning out of the sea of data.
--
Francis Heylighen Center "Leo Apostel" Free University of Brussels http://pespmc1.vub.ac.be/HEYL.html ======================================== Posting to [EMAIL PROTECTED] from Francis Heylighen <[EMAIL PROTECTED]>
