BACK

 
A DEFINITION OF ANALOGY

Tony Giovia
www.geometryofideas.com

 

Hi Richard,

>>>  Conscious response involves consideration of future alternatives, there is an evolutionary penalty to be paid... thinking time. The rapid growth of the human species could be explained by this advantage exceeding the break even level in terms of evolutionary advantage, once the thinking is quick enough and broad enough it works. How it got there may be a chance event. <<<

Killer point, very real and well put. It sets Thinking Time Penalties vs Pure Reflex ( ie, a variable vs a constant) as a criteria for evolutionary progress.

Using this to distinguish between Bot 1, Bot 2, Bot 3 etc. will take some work (there are a lot of angles here, and a lot of theory ), but the work will be very revealing. Of course, if I remember right, rates of change in C is one of your interests, so you had this in mind when you posted this and already have some of these angles on the board.

Are these C threads great, or what? <g>

Tony

 

Hi Richard,

>>> . I suggest that somewhere deep down, the reflex response decided it needed a means of getting the energy flows up and running in a potential conflict, amber alert if you like ...  A backwards analysis of the signs of danger prepares the organism for actual danger. <<<

I am trying to think of the mechanics of this scenario. You are right that it makes sense to assume C evolved from primitive reflexes, but it seems that to get to C would require a middle-state between instant response and deliberated response, because once reflex crossed the line into deliberation it would be C, not reflex.

Perhaps you suggest the solution when you say 'backward analysis' - maybe the middle-state is reflex plus 'event storage', ie, memory. Then the memories ( via physical laws) associated into C, with no cognitive builder required. The primordial soup strikes again.

This is an angle I haven't considered before, so these are just first thoughts.

Tony


Richard,

>>> The logic of analogy transcends various frameworks .... <<<

I'm not clear on what you are saying here. Are you suggesting that the 'logic of analogy' (ie, the relationships among the contexts which compose the analogy) is different from AND, NOT, COPY logic?

Analogy is itself just another context - composed of ideas like any other context. If Context A is composed of ideas X, Y, Z, and Context B is composed of L,M,Y,Z, then the physical underpinning of the analogy comparing A and B is their shared ideas Y and Z.

Tony


Richard,

>>> "But then no argument by analogy is intended to be mathematically certain. Analogical arguments are not to be classified as valid or invalid. Probability is all that is claimed of them ... We are therefore dealing with probable andness, probable notness so yes I would say we are outside the bounds of pure (as in certain) logic.<<<

OK, now I see what you are saying. I have Copi (the 3rd edition) and read that page, but I would not use the phrase 'transcending logic'  to describe analogy. If I may, I'd like to look at this process and show you my reasoning.

First let's look at what Copi means by probability. The dictionary definition of analogy (MW) is the 'inference that if two or more things agree in some respects they will probably agree in others'. This is where Copi's 'probability' comes in. Metaphor and analogy both use shared contexts as structural underpinnings. Metaphor makes no claims of identity beyond the shared contexts, while analogy does.

In other words, probability as Copi is using the term means (IMO) that some of the 'anded' ideas in an Analogy C are 'anded'  less directly than an actual simultaneously-included context. For example: if context S is shared by contexts A and B, then context S is included in the structure of Analogy C by definition. This context S is a definite connection between Contexts A and B.

But the nature of analogy is that other contexts not shared by A and B are also indirectly included in the analogy - for example, contexts T,U,V  'anded' to S in Context B but not in Context A (which therefore indirectly includes contexts T,U,V into Analogy C because they are already 'anded' to S ) - are the contexts which create the 'probable' aspect of analogy. ( BTW this type of probability does not suggest randomness to me.)

I myself would not characterize this process as 'transcending logic' - because logic is still holding the ideas within the analogy together.

Analogy identifies relationships, direct and indirect (via other ideas) but logic is still the organizing force.

Analogies are not mathematically certain comparisons because the compared contexts are structurally different - they are by definition not on opposite sides of an equals sign. The ideas of 'probable' andness and notness (as you define them) are difficult concepts for me to grasp; it doesn't seem that the andness itself is probable - it seems either 'and' is applied or it is not - but there seems to be a machinery at work that I can't get to at the moment (maybe it will come to me in the morning <g>).

THOUGHTS:

1) Assumptions are the reference points or base image patterns against which incoming data and "what if" scenarios are based.

2) Contexts are held together by "relationship bonds" which are dominant/submissive based. These power relationships serve the same function as chemical bonds, and shared contexts form the strongest bonds.

Tony

 

BACK