Emergence of this new form of observation from precedents in modern science

The wholeness is a system of centers which, working together, create the gestalt of a given part of space.

Now, the task of distinguishing some of the elementary centers which occur, and setting them apart from other fragments of space which are less centered, happens without difficulty in simple cases.

In seeing an apple as an entity, or the core of the apple, or the apple pits, something we do every day without thinking, again we calculate and notice that this center is more salient, more coherent, than many other nearby overlapping portions of space.

This is practically a definition of categorization. Using “apple” — a basic level category — as an example.

Or, for example, if I look at a black line on a white ground, the black line may be identified as a center. It is distinct, bounded, symmetrical, contrasting in color; its appearance as a center comes about because of these features which cause its differentiation. The objective character of this circle as an entity is well known as a fact of cognitive psychology.

By about 1930, there was already a well-documented literature on the problem of defining greater and lesser strength in centers (then referred to in simple cases as “goodness of figure”). By about 1940, this goodness of figure had been pinned down as being dependent on characteristics like convexity, differentiation, boundaries, and so on, “precursors of the fifteen properties I have identified. This goodness of figure was a weak form of the quality which I call “life”.

There it is! A confirmation that what I had suspected — that gestalt psychology and the fifteen properties are connected — pretty much directly stated by Alexander himself.

It is possible that one day computer programs designed for cognition might be able to pick out these centers and rank order them by their degree of life, but in this case we are moving to a level of complexity where human cognition is a more reliable measuring instrument than any presently available mathematical theory.

First time I want to dive into machine learning to see if today’s technology can be trained on recognizing centers and life in pictures?!

We do still have the number of centers, or density of centers, as something that can help to settle the argument. But this is an analytical evaluation based on a particular method of counting. It can be used to bolster a doubtful judgement. But the judgement itself as a pure judgement, the measurement itself, needs a direct method.

So far, the studies I have cited have still been viewed mainly as studies within the realm of social science, showing that psychological considerations should have a bearing on the evaluation of architecture. Their validity has been viewed as a validity within psychology, not within physics. The method which I propose, on the other hand, though continuous with these studies, takes a very important further step. I suggest that these apparently “psychological” methods give objective insight about the objective state of a system in the world and should be considered as measurements of this state — and hence as part of physics.

What I am proposing is that these observations of an observer’s inner state are not merely a reflection of a person’s attitude or psychology, but can actually can be used to measure something real about the external world itself.

#book/The Nature of Order/1 The Phenomenon of Life/9 Beyond Descartes: A new form of scientific observation#