Why the axial line? (again)
Wow! Amazing messages.
I want to reply specifically to both Alan’s and Sheep’s latest but I need to think some more about them. In the mean time I have some basic questions which have been bugging me. Maybe some of the answers are in those last two posts but they didn't spring off the page at me. I will send this now in case anyone has some ideas to help me on my way.
It seems to me that some factors are logically inescapable if you are to have a field of study which relates space and society (or perhaps, configuration and people).
1) REPRESENTATION Saying that space syntax is ABOUT a set of analytical tools is a bit like saying that daylight analysis is ABOUT using photometers. Tools are a means to an end - what are you trying to find out about a place by applying spacesyntax analysis? You can’t find out about temperature gradients, for instance, because you are not using a thermometer. You measure only very specific features of the world; "where a human being can move to." Axial lines ONLY represent ‘lines of sight along which a human being can move.’ Boundary spaces ONLY represent ‘areas of floor across which a human being is free to move from any part to any part’. Convex spaces combine the constraints of both of these, representing ONLY ‘areas of floor across which a human being is free to move from any part to any part along a line of sight’. You might confine your work to statistical analysis of the representation (a map or graph), but the method of representation carries inherent meaning. Am I wrong in saying this?
2) HUMAN FACTORS For all the talk about abstract statistics, the only way you can be sure (as you stand in a real location) that you have something which can be represented (respectively) as an axial line, a boundary space or a convex space is by applying the above definitions. If we define them in other ('neutral') geometric terms we are leaving an opportunity for inconsistency where space is not simple or empty. If you can’t perform the above-mentioned movements or observations in any part of the location, I mean REALY go there and do it, you cannot accurately represent that space as a discrete entity in a spacesyntax map. Is that not so?
3) APPROXIMATION I appreciate the work done by Sheep and others on fractional (and angular) analysis which allow for degrees of change as an observer moves along a line-of-sight . I feel sure that this is important in reflecting how space is really used. But surely these methods are intended to make a closer representation of real world phenomena, not to smudge-out any existing inconsistencies of method. I am interested in Sheep’s comment about the robustness of the axial line. Doubtless this is the reason why it works better as systems get bigger and the particularities of places and persons become averaged out. Or perhaps my assumption is wrong?
4) PHYSICALITY Then there is the question of what is being correlated WITH the configurational statistics. Again, I appreciate that the items of interest in the world are often PATTERNS of phenomena which are expressed in an abstract way; ‘co-presence’; burglary rate’; ‘natural movement’; etc. So, can a variation in spatial configuration cause variations in ‘co-presence’ in a regular, predictable way? ONLY if it causes a greater likelihood of different individual people moving into the same space at the same time. Co-presence isn’t a thing, people are things. Until ‘things’ have changed, the purported cause has had no effect. As far as I am aware, the only ‘things’ which are the subject of space syntax are people, not other measurable changes in the world. Are there any other subjects?
5) CAUSE I note that the spacesyntax laboratory is always careful to use the word ‘correlation’ to denote the link between such phenomena and configurational statistics. I will dare to use the word ‘cause’ because if there is no causal relation then all we have are co-incidences, or the independent products of a third cause, in which case we are all wasting our time (and I (obviously) do not believe we are). I don’t know why I am telling you this, because you must teach it to every doctoral student you supervise. Is this not a basic fact of academic & scientific research?
6) SPECIFIC HUMAN MOVEMENTS The Social Logic of Space defines the relation between the physical constraints on spatial configuration and ‘where people go.’ There are fascinating discussions about this as an iterative process of rules-of-space and rules-of-movement developing in tandem and generating emergent properties of global form in settlements (probably global phenomena in society too)(?) But the only THING which is not ‘physically separated spaces’ is ‘people moving’. But even this is too abstract to measure. ‘People moving’ is a generalization about many different individual people performing specific movements at particular times and places. Is there something else to observe?
(I am still trying to work out Alan’s point about natural movement not being about movement).
7) MECHANISM If configuration is having a causal influence on patterns of peoples’ movement, then it is (at least sometimes) a casual factor in the particular decision of an individual person at a specific decision point (node) Of course it can’t cause a change in movement at a point where no options are available. The effect on any individual person may or may not be decisive, but it is logically necessary for such activity to occur IF there is a causal relation between space and society. This isn’t determinism which says that all subjects must react the same way - it is the pattern of probability for a category of people which is determined by the environment. Neither is it a precise mechanism, but is it not inescapably the TYPE of mechanism which MUST be involved?
[log in to unmask]