Re: Time and change

That ties in with the earlier point Tim made about circularity (see below; my response to Tim's October 22, 2003 11:01 PM post). However, at that point in her post, Judith was not speaking directly about the nature of time, but rather about the general nature of systems as they relate to time. So, yes... her terminology is circular, but since she wasn't using it to define time, her comments work for me.

I don't think "process", "change", or "action" is fundamental. There might be some model in which you could represent any one of them symbolically as a fundamental parameter, but that would be a model in which you had no need for the concepts of time or space, or had somehow otherwise included them in the semiotic encoding of the model.

The models in which such concepts as process/change/action might serve as fundamentals, without invoking a concept of space&time, are the kinds of models I was grasping at in my early posts on this topic of time. I still have no clear idea of what these kinds of models might look like.

That's a masterstroke of of genius in more efficient semantic encoding of the formalized descriptions of the systems under study. In demolishing the prejudice that the referents of “one’s observables” must necessarily be “conventional reductionistic fragments”, he has liberated physical science from its limiting constraints as a tool for describing systems of any real complexity, which systems happen to be the most interesting ones. It obviates the sterility of context-independence by introducing context-dependent parameters into the fundamental descriptions from the get-go.

Here's an excerpt from my five (!) pages of notes on that single paragraph:

well be a useful methodology in constructing theoretical models that are better able to accommodate complexity. That certainly appears to be what RR was saying.

I've deliberately excluded the concept of "change" at this point, because Tim asked a very specific question about it. I'll come to that presently, but first I want to wrap up the discussion of "action" and "process", as they relate to Tim's musing about whether such concepts might be more fundamental than "time" and "space".

Here's a thought experiment:

1) I would suggest that defining a situation in which phenomena cannot be perceived and data cannot be gathered immediately precludes on logical grounds any possibility of "proving" any conclusion which would rely on the aforementioned phenomena/data. So I am not sure what this experiment says about "time". Indeed, I would argue that we cannot even say space exists within the box - only that the box itself occupies space. It is a matter of speculation (albeit, commonsensical) to say that space therefore exists within the box. From the situation given in which we have zero information about the interior it could be void of space. The only indirect indication that space exists inside is the precondition that the box contains a perfect vacuum, which presumes a presence of some space in which the vacuum could be claimed.

2) It seems to me that the phrasing of the question "Do time and space exist within the box?" assumes that time & space are both entities which have their own respective independent ontological existences - that they are "things" in the material world. I would argue that we have no meaningful way to discuss time in that manner. (see my post of 10/25/2003 regarding time and impredicativity)

I don't know how to define "action", or "process" without using the concept of time, which means that the way I use "time" conceptually is more fundamental than the way I use "action" or "process". My usage seems to be a reasonably close model of observed reality, so I would say that time is more fundamental than "action" or "process".

Now, as for "change", that's another matter entirely. Skipping ahead for a moment to Tim's October 22, 2003 11:01 PM post, he asked:

Short answer: Yes.

Longer answer: It depends on the context. For example, in a two-dimensional planar coordinate system, consider the straight line given by the equation y = x. How does y change with respect to x? Piece of cake, right?... take the derivative of y with respect to x:

If we draw the equation y = x on an x:horizontal/y:vertical two-axis graph, it's just the straight line that passes through the origin (x,y) = (0,0) at 45° to the x and y axes, and its slope = 1. Not very interesting, but really simple. Now, pick any two points anywhere on that line. You don't need to involve time at all to see that the y values of the two points represent a change of dy, and ditto for the change in x values, which we call dx. You can do the same thing with, say, the parabolic function, y = x²...except that in this case, dy/dx = 2x, which is not a constant slope. But it's the same difference, by which I mean that the changes represented by dy and dx are the same kind of changes. When you day "dy" or "dx", you are clearly speaking of "change", and it's ""change" without some (hidden) reference to time or temporality". (These are simple examples; they can be generalized to more complex systems.)

Ergo, the answer to Tim's question is "Yes".

I perhaps should have been more precise in asking my question: "Can one speak of "change" as it is observed in physical dynamical systems without some (hidden (i.e., tacit)) reference to time or temporality?

My question here is whether this example is relevant or not to the encoding of dynamical systems. Your example begins from a 2-D space and has no inherent dynamical or temporal quality. It is a static system. So it follows that the differential equation of such a system likewise has no explicit or tacit temporal reference.

Kampis argues that dynamical systems which are described formally by some function F, describing some kind of curve in a manifold, are abstractions which convert dynamics -> statics:

"It means that the objects of dynamical systems are time-global, and hence, time-less, unlike the time-local (and time-bound) observations. The existence of the trajectory as an invariant and independent, well-defined object means that the dynamics of the system is so effectively decoupled from real time that it becomes completely static. A curve is just a curve and nothing else; as such it has nothing to do with the concept of motion - with any motion or change whatsoever. So, instead of the dynamics universe where we live, one that embeds courses of change, we are left with a static and motionless conceptual universe. And, insofar as we consider this formal universe as the ultimate model, we arrive at the idea of a static and frozen-out reality; of a walled-in Universe where motion and no-motion are equivalent." [1991, p. 168]

There's an interesting aspect of that question in the phrase "(hidden) reference to time or temporality", with special emphasis on the word "hidden". Why should a temporal reference be "hidden"? >From my perspective, it can only be so if our thinking about the fundamental elements of the problem is not sufficiently precise to begin with. If we're going to be able to solve any given problem in a way that turns out to be useful, the theoretical model we use must be encoded in a way that enables us to interpret the results in the domain of applicability; that is, we have to be able to relate the solution to those aspects of the systems under study that motivated us to seek a solution in the first place. That is precisely what we mean by context-dependency.

The point is this: the question as to whether "time" & "space" are more or less fundamental than "process", "action", or "change" is domain-specific; that is, it's context-dependent. Any physical system can be reduced to, say, descriptions that are stated in terms of mass, length, & time. Physics does a helluva job of that. But so what? It's of no use to be a reductionistic smarty-pants right down to the quantum level if you lose all trace of the phenomenon you purposed yourself to study... for example, life itself.

I can boil it all down to what I believe is a quintessentially Rosennean perspective: questions about what are the "fundamental parameters" of any given problem must be determined by the contextual constraints of the problem itself. We cannot make a priori assumptions about the applicability of, say, the constraints of making time-based measurements in problems that have relativistic entailments, when such constraints might be utterly irrelevant to, for example, the error rate of DNA replication in undernourished humans living in a high-stress, polluted environment. The exclusion of such domain-specific (i.e., context-dependent) constraints can be prescribed by any number of principles -- epistemological, information-theoretic, scaling, general system-theoretic, phenomenological, etc.

That is an excellent point. I feel that ch. 4 of AS can be seen as an exposition of some of the various ways in which the context of the question notably alters the role and meaning of "time".

For my part, I'll stick with an inquiry into the nature of time itself, which I believe is more fundamental than the methodologies by which we measure it. That will be the subject of my next message.