-Caveat Lector-

from:
http://www.socresonline.org.uk/socresonline/3/1/11.html
<A HREF="http://www.socresonline.org.uk/socresonline/3/1/11.html">Smith:
Social Structures and Chaos Theory
</A>
-----
Smith, R. D. (1998) 'Social Structures and Chaos Theory'
Sociological Research Online, vol. 3, no. 1,
<http://www.socresonline.org.uk/socresonline/3/1/11.html>

To cite articles published in Sociological Research Online, please
reference the above information and include paragraph numbers if
necessary

Received: 4/7/97      Accepted: 24/3/98      Published: 31/3/98



Abstract

Up to this point many of the social-scientific discussions of the impact
of Chaos theory have dealt with using chaos concepts to refine matters
of prediction and control. Chaos theory, however, has far more
fundamental consequences which must also be considered. The
identification of chaotic events arise as consequences of the attempts
to model systems mathematically. For social science this means we must
not only evaluate the mathematics but also the assumptions underlying
the systems themselves. This paper attempts to show that such social-
structural concepts as class, race, gender and ethnicity produce
analytic difficulties so serious that the concept of structuralism
itself must be reconceptualised to make it adequate to the demands of
Chaos theory. The most compelling mode of doing this is through the use
of Connectionism. The paper will also attempt to show this effectively
means the successful inclusion of Chaos theory into social sciences
represents both a new paradigm and a new epistemology and not just a
refinement to the existing structuralist models. Research using
structuralist assumptions may require reconciliation with the new
paradigm.

------------------------------------------------------------------------
Keywords:
Chaos Theory, Complexity, Connectionism, Consciousness, Memes, Network
Theory, Reductionism, Social Change, Soliton, Structuralism, Timelines,
Transduction

------------------------------------------------------------------------
Structures

1.1
One of the most enduring aspects of all social theories are those
conceptual entities known as structures or groups. In psychology
apparently repetitive behaviour patterns are assembled into traits of
personality. In sociology people are combined into aggregates which are
themselves described with terms such as citizen, resident of a
municipality, ethnic group, religion, political conviction, sexual
preference, race and so on.
1.2
Ways of assigning people to these aggregates vary. Initially people were
placed in categories which themselves had attained the status of
cultural validity. Castes, estates, tribes, sex are examples of these.
Assignment of individuals to these groups was taken as an a priori and
they were used to assemble definitive and clear-cut patterns of other
attributes.

1.3
The International Encyclopaedia of the Social Sciences defines a
'structure' in the following way:

The term 'structure' may be defined as a pattern, ie. an observable
uniformity, in terms of which action (or operations) takes place. ...
Structure refers to an aspect of empirical phenomena that can be
divorced from time. The patterns of action, qua patterns, do not exist
as concrete objects in the same sense that sticks and stones do. The
patterns of action in this sense are abstractions from concrete
empirical phenomena, and they 'exist' and are empirically verifiable in
the same sense that the squareness of a box exists and is empirically
verifiable. (Levy, 1968)

1.4
Despite the fact this definition is nearly 30 years old it is
sufficiently expressive of what most social scientists actually believe
to be deemed 'intuitively clear'. Most researchers talk of such
social-structural phenomena as 'class', 'ses', 'family', 'household',
'babyboomers', 'Xers', 'ethnic groups', 'minorities', 'women',
'homosexuals', 'races', 'castes' and so on without much further thought.


1.5
Structuralism has two problems. First is the practical matter of
determining what the structures are and how our perceptions are to be
classified. Second is the problem of 'change' - of how one structure
becomes another.

1.6
Structuralism has the problem of being a very fluid term. It could be
almost any pattern in this definition. The issue here was 'where do the
lines get drawn?' or what are the boundaries of the structure? The
history of social thought refers to this as the 'taxonomy' problem. It
is a debate which goes back at least to Plato and Aristotle.

1.7
Drawing the boundaries forces on us two decisions. One is 'what
categories shall we create?' and the other is 'how shall we understand
and interrelate these categories?'



------------------------------------------------------------------------



'I shall call this figure the duck-rabbit. It can be seen as a rabbit's
head or as a duck's. And I must distinguish between the 'continuous
seeing' of an aspect and the 'dawning' of an aspect. The picture might
have been shown me, and I have never seen anything but a rabbit in it.'
- Ludwig Wittgenstein (after Fuglesang, 1982).

Figure 1: Duck-Rabbit



------------------------------------------------------------------------

1.8
To return to the first definition given, structural entities or patterns
were also to 'be divorced from time' even as their very existence is
recognised as a temporary phase of a time-dependent process: evolution.
Little wonder then, that one of the most intractable theoretical debates
of the 'structural- functionalist' era was 'the problem of change'. How
one can hope to discuss 'change' in a set of conceptual entities which
have been specifically 'divorced from time' presents a considerable
intellectual challenge.[1]

1.9
A visit to any library will reveal shelves of books all published in the
post-war period in which systems theory was applied to social science.
Most notable in these titles are those that attempt to incorporate
change, or flux, into the array of boxes and arrows which represented
the formal paradigm of Systems Theory. Almost all these books start with
general flow charts and map out the various paths influences can and
must take in order to be able to influence the systems' 'functioning'.

1.10
The problem of change also has two features. The first is related to the
taxonomic boundaries. Boundaries change when, say, one ethnic group is
assimilated or absorbs another. This phenomenon is well known but the
formalism of social structuralism has never been able to deal with it.
People in ethnic groups can be seen to follow assimilation patterns
through time. Ethnic groups are not defined so much in terms of the
individuals who are in them but in the behaviour patterns the members
exhibit.

1.11
For example, the illustration in Figure 1 can 'change' from a bird to a
rabbit as an act of cognition divorced from any physical alteration to
the picture itself.

1.12
This is still bracketed, however, when we realise that while some people
may see a duck and others a rabbit, no-one is likely to see a cup of
coffee. Moreover, in all instances, what one is doing is seeing marks on
a page and being led to remember a bird or a bunny. Duckrabbit is in
this sense neither duck nor rabbit. It is a visual stimulus made of ink
and paper (or of pixels). The conundrum of Duckrabbit contains both
objective and subjective aspects. Objectively virtually all will agree
it is a picture. At the other end of the continuum it is probably as
futile to argue this pattern symbolises a 'duck' and not a 'rabbit' as
it is to dispute over opening eggs at the little end as opposed to the
big end.

1.13
Between these poles is a range of things the pattern might also be
symbolising. It is almost certainly not a cup of coffee, a tree, an
igloo or the eruption of Krakatoa. In short, we do not have the luxury
of using this or any other ambiguity of our symbol system as evidence
for the adoption of pure relativism. Some interpretations are simply
less believable than others. In this case we may be forced to resolve
our interpretation of the symbol into two, not one, defensible meanings.
Perhaps there are three such interpretations. But there are certainly
not an infinite number of them.

1.14
The fact we cannot have perfect congruence between our cognitive symbols
and the reality they purport to represent is not evidence of our
inability ever to know what that reality is. Those who make such
assertions are arguing not logically but ideologically.

1.15
Many ideas which seem clear at the level of abstract theory concepts
become less clear at the operational level. One can have a reasonably
good idea of what a bird is but when asked a companion question 'rate
the following birds regarding how good they are as examples of birds'
the matter becomes less clear. McNeill and Freiberger (1993: pp. 84 -
85) report that when respondents were asked to rate birds on a scale
from 1 (excellent) to 4 (poor) it was found that robins scored 1.1,
eagles 1.2, wrens 1.4, ostriches 3.3 and chickens 3.8. Bats scored 5.8.

1.16
Our categories themselves (such as 'birds') are not hermetically sealed.
Human understanding can take a creature which is known not to be a bird
at all and still classify it along a 'birdness' dimension. If a bat can
be a rather inadequate example of birdness then what kinds of potential
equivocations or misunderstandings can and doubtless do arise with
terminology which is not so formally clearcut as that of biological
taxonomy? What is the price in lost clarity and perhaps in wrong
analytic decisions when the existing ideas of social structure are
employed?

1.17
To take a common example of a social structure consider ethnicity[2] and
the change often associated with it: assimilation. Figure 2 shows the
classic signposts which are generally taken to represent assimilation
but even as we observe people making the transition we have some nagging
questions. For one thing, how do we conceptualise in terms of a
structural category those people who are 'in-between' or hybrids of one
group or set of traits and another? Should we assign a label to them
now? derive labels from logical categories? waiting until a certain
percentage are able to be so classified? wait for them to demand their
rights? something else?

1.18
In the vernacular of many societies these individuals are (usually
derisively) called apples, bananas, oreos, half-breeds and so on. Do
they form 'structural groups' or not? In a more direct example: what of
the people of Hong Kong? Are they British? Chinese? Something else? And
how does one decide?

1.19
The arbitrariness of assigning people to categories the did not seem
even to know existed led some theorists to embrace a 'negotiated order'
approach to all social reality. Given that the signal uniqueness of the
human species seems to be self-consciousness the work of Blumer (1969)
 and others sought to seek its patterns there.



------------------------------------------------------------------------

Subprocess or ConditionTypes or Stages of AssimilationChanging of
cultural patterns to those of host societyCultural or behavioural
assimilationLarge-scale entrance into cliques, clubs, and institutions
of host society, on primary group levelStructural assimilation
Large-scale intermarriageMarital assimilationDevelopment of sense of
people-hood based exclusively on host societyIdentificational
assimilationAbsence of prejudiceAttitude receptional assimilationAbsence
of discriminationBehaviour receptional assimilationAbsence of value and
power conflictCivic assimilation

Figure 2: Ethnic Assimilation Variables (source: Gordon, 1964)



------------------------------------------------------------------------

1.20
This trend became so disturbing that George Homans dedicated his 1964
presidential address to the American Sociological Association (ASA) to
the call for bringing individual human beings 'back in' to social
theories. In his address he observed that social structural theories did
not really seem to have any need for human beings (Homans, 1964).




Memes

2.1
As with Duckrabbit ethnic identity and assimilation is, at the level of
the individual person, some kind of interaction between sense data and
cognition. Thus ideas, too, may be thought of as being 'structured' (as
when they are assembled into a paradigm). In this guise an ethnic group
may be understood as a collection of people who both accept one-another
as members and define membership in terms of attitudes, attributes,
behaviours or understandings which comprise identifiable patterns.
2.2
Structuralism also leads us into another complication. Paradigms which
represent reality as a set of mutually exclusive categories have
difficulties other than the problems of change and categorisation. They
also lead into debates on 'reductionism'. The onset of sociology was
marked by the efforts of Emile Durkheim to insist that sociological laws
existed and could be studied independently from the study of the
psychological laws for individuals who made up society. Sociology did
not need to be reduced to psychology any more than biology had to be
explained in terms of physics. Social processes could be studied in
their own right. In the time since then some critics have observed
social theories were so abstract people were not really able to
influence the long-term outcomes that macro laws would predict.

2.3
Structuralism, in short, could not easily deal either with change or
with how or to what extent the actions of individuals could influence
the future state of their community, their culture, the world.

2.4
Ideas (or as Dawkins calls them, memes[3] (Lynch, 1998)) can flow
through time, being advocated by different individuals but retain highly
stable intellectual content. In this instance the membership of the
group changes but presumably the 'ideas' remain the same or, if they do
not, they change slowly enough so that the actual members of the ethnic
group do not experience the destabilising emotional consequences of
changes in understanding that are taking place too rapidly.[4]

2.5
Consider, for example, the matters of political affiliation, political
ideology and political campaign promises. In the 1996 US Presidential
election there was some discussion as to whether incumbent William J.
Clinton was a 'liberal' or a 'moderate conservative'. Comparisons of his
policies and principles with fellow-Democrat Mario Cuomo suggested to
many people that Clinton was not a liberal at all. The issue of
classification is this: both of these men were indisputably members of
the Democratic Party. This is a legal status and cannot easily be
rendered ambiguous.

2.6
We might also note that what passed for liberalism in the 19th century
was a devotion to laissez-faire economics, the restriction of domestic
government to the role of guarantor of contracts, and an aversion to
almost all forms of taxation. The late 20th century contents of liberal
ideology are rather different. The matter before us is to render this
apparent fluctuation or evolution in label and meaning (terminology and
membership) in a way consistent with a conceptual formalism. This
rendering will hopefully be relatively free of what is known generically
as 'hand waving'.

2.7
Within the current corpus of human intellectual life there is at least
one example which could prove to be of use. The conceptual paradigm of
quantum mechanics contains within its formalism a well articulated
approach to structure which allows easily for situations of this type to
be discussed.[5]

2.8
One could, for example, suggest that two memes (ideas) such as
individual liberty and collective responsibility might combine as in
Figure 3. In this way we can think of an idea as a soliton, a solitary
wave, moving through history.



------------------------------------------------------------------------



Solitons are described mathematically as Kortweg-de Vries (or KdV)
equations. They are of the form:

ut - 6uux - uxxx = 0

where: u (x,t) = exp{-k3t + ik(x-t)}; k  R, k  0

and: ux = u/x

(from Drazin and Johnson, 1989)

The figure shows the computed dynamics of a two-soliton interaction. It
is essential to realise this soliton is a wave travelling through a
medium or some other matrix capable of transmission. Water waves, for
example, pulse through the medium (oceans or lakes or rivers) but at
every stage of their travel they are made up of different collections of
water molecules. The water molecules which make up the wave at any given
time are largely distinct. As complex as these waves are they are not
infinitely complex (within a given dimensional space) unless the
dimensions themselves are continuous. Water, coming in molecules, is
quantised. The issue of infinitely variable waves must then rely to some
extent on whether space (or space-time) itself is amenable to being i
nfinitely subdivided or it is also quantised. (see for example: Misner
et al (1973). This consideration will appear later when the specific
case of the Lorenz attractor is examined.

Figure 3: Computer Rendering of a Two-Soliton Interaction



------------------------------------------------------------------------

2.9
Are these solitons 'structures'? Is the region of the two-soliton
interaction pattern also a structure? Those of us who do survey research
know how difficult it is to guess at what the underlying three waves are
if all we have is time-insensitive data collected from different people
with non-standard questions at irregular intervals. After gathering what
data we can we could employ such statistical techniques as Factor
Analysis, Cluster Analysis and Principle Components Analysis before
using LISREL to 'analyse' the structures we found. Even using
approximations to deal with temporal fluctuations by employing lagged
variables and autocorrelation analysis, long standing approaches in the
more mathematically sophisticated areas of macro-economics, is
increasingly being seen as having limitations which our computer
technology no longer requires we continue to accept (Rosser, 1991).

2.10
Viewing structural entities as waves forces us to revisit the question
of rigid taxonomic boundaries. In survey research it is common to regard
certain collections of individuals as 'cohorts'. If one extends this
then one possible visualisation for a soliton is that of a cohort (such
as 'Babyboomers' or 'Generation X-ers' who owe their designation to the
historical period in which they were born) moving through temporal space
and interacting with other cohorts. Structural entities can also, of
course, be based on attitudinal as well as demographic attributions so
that people who arguably qualify as boomers by virtue of their age may
have life experiences more akin to those of Generation X and thus occupy
some hybrid (or in the language of wave mechanics a 'resonance state').
[6]




Chaos

3.1
At the intuitive level, at any rate, chaos theory seems to provide a
means of escaping structuralism's two signal flaws: it is inherently
dynamic and time-sensitive and it permits a definition of social
structural entities in such a way that if real, living, unique human
beings vanished then the structures of society would also vanish.[7]
 This also provides, by means of its formal inclusion of human
cognition, an analytic paradigm which might allow us to see how
providing a set of democratic laws to a totalitarian culture does not
necessarily transform it into a democratic society, or how it is that
despite having a constitution written by American soldiers after World
War II, Japan still boasts a decidedly Asian culture. Things can change
and individual people can make a difference at the 'macro' level.
3.2
The fundamental observation of chaos theory is that small changes can
have large consequences. Butterflies might cause tornadoes, saving a
life in 1930 might mean that Earth's political future is that of being a
squalid dictatorship never to join the UFP and build the starship
Enterprise, and so on. Even as this staggering realisation started to
sink into the consciousness of philosophers and scientists other more
rustic souls were reminding us that 'for want of a nail a kingdom was
lost'. This was not new news.[8]

3.3
Historians have known for years that 'plausible' different or future
worlds (Hawthorn, 1991) are the stuff of worthwhile thought experiments.
This exercise forces us to distinguish between what our theories tell us
could not have happened and what our history tells us did not happen.

3.4
Pipes (1995) argues that the Bolshevik Revolution in Russia was not
inevitable but that given its success the emergence of Stalin was all
but assured. He also maintains that '[i]t is my considered judgement
that had it not been for the Russian Revolution, there would very likely
not have been National Socialism, probably no Second World War and no
decolonisation, and certainly no Cold War...' (Pipes, 1995: p. 1) He
also points out that had Fannie Kaplan's poor eyesight not spoiled her
aim she might very well have successfully assassinated Vladimir Lenin in
August of 1918. Without Lenin the Revolution would almost certainly have
failed.

3.5
When confronted with such 'if only' scenarios it is not hard to see why
many historians take the quest for a Grand Theory of society less than
seriously. But even if Fannie Kaplan had succeeded, what would our
present world look like? The truth is we do not know.

3.6
Skocpol (1984) reveals the difficulty in using the very limited tools of
social theory to deal analytically with historical events. What, then,
to do with an approach to structuralism or the 'structuring of society'
that muddies the distinction between the plausible possible and in
plausible impossible? How are we to differentiate the utterly impossible
from that which is merely formidably difficult? Answering these
questions will allow us to deal more directly with the question 'are all
worldviews equally arbitrary?'

3.7
The answer is equivocal. Pipes' earlier noted comments contain both
chaotic and non-chaotic elements. Fannie Kaplan's attempt to kill Lenin
and the consequences of her failure are presented in a manner consistent
with Chaos Theory: a very small event can have extreme ramifications.
Pipes also argues, though, that Kaplan's failure all but assured the
eventual emergence of Joseph Stalin and by extension National Socialism,
WWII and so on. Is he really trying to have it both ways? Did no other
opportunities emerge which, had they been seized, would also have
altered this outline of events?

3.8
Pipes' attitude is not atypical of many historians. There is a sense of
human history being played out on various levels or planes and mostly
these levels do not interact or interfere with one another. Sometimes,
however, they do. Sometimes a distant butterfly wingflap causes storms,
property damage, and stock market fluctuations half a world away.
Sometimes it doesn't.[9]

3.9
This being the case it becomes all the more essential to be clear about
the formal and the informal uses of the chaos concept in social science.
Chaos, like Relativity, Darwinism and Mechanism before it, has the
potential to be transformed into a metaphor and to have its terminology
misunderstood and misapplied. Indeed numerous new titles currently
available indicate this trend has already begun. (eg. Savage, 1988). The
best defence against this is a well-defined and accepted formalism

3.10
In order to have a formal notion of structure which is consistent with
the chaos theory paradigm we now need to return to a requirement so far
unelaborated. In its most precise rendering chaos can only arise when
the possibility of any given state repeating itself is potentially zero.
To take the illustration of a strange attractor such as the Lorenz
attractor (as in Figure 4) what is needed is a situation in which the
orbital pathway of a flow or flux can continue for an indefinitely long
period of time (for eternity) without ever passing through the same
point twice. If this condition is not met then the orbit is not in fact
chaotic but periodic even though highly convoluted. What this in turn
means is that the phase-space in which the flux is propagated should be
continuous and not quantised. A quantised space, however large, is
effectively finite and thus cannot provide for truly chaotic behaviour.
[10]



------------------------------------------------------------------------



For true chaos to be present the orbital pathway must be a line of
infinite length. If it passes through the same point twice then the
deterministic nature of chaos informs us the orbit is of fixed
periodicity and length. In order for true chaos to exist in nature,
then, it may be necessary for space-time to be continuous and not
quantised. If it is quantised then the number of possible positions is,
while incomprehensibly large, finite. Conservation of energy means that
the phase-space for such an orbit would likewise be limited.
Lattice-models in which both chaos and solitons arise, however, have
been proposed (Sayadi and Pouget, 1991).

Note that in this plot the lines clearly appear to intersect. They only
appear to do so because this a two-dimensional rendering of a
three-dimensional process. Chaos in a phase-space depends on the
dimensions of the phase-space. If the actual dimensions of the
connections (the lattice through which the signal or influence
propagates) is of evolving or growing dimensionality then apparently
finite restrictions may be ephemeral. See, however, for opposing views
on themes related to this, Dyson (1988) and Gell-Mann (1994).

Figure 4: The Lorenz Attractor



------------------------------------------------------------------------




Complexity

4.1
Even the briefest glance at the graph of the Lorenz attractor will
produce the sense that a pattern is clearly present. The fact that
'chaos' is involved in producing it does not remove the feeling that
something 'macro' is going on.
4.2
This is the position taken by complexity theorists (eg. Kauffman, 1995)
who insist (as Durkheim did a century ago) that large-scale stochastic
regularities can be empirically shown to 'exist' and can be analysed as
such.

4.3
Ever since social sciences first began to analyse groups of people as if
they comprised a single entity or 'structural component' a constant
objection has been raised: social structural entities to do not really
exist save as heuristics. People exist. Classes do not wage war on each
other. People do. This much is patently obvious. But we are also
confronted with what we deem to be irrefutable (or at least highly
persuasive) evidence to the contrary. Ethnic groups, socioeconomic
classes, demographic cohorts and the like are useful empirical
generalisations and, with the correct data analysis tools, can be used
to group individuals together in categories the people themselves may
not even suspect exist.

4.4
So while human consciousness is clearly necessary for many of the social
patterns we observe, the patterns may not be identified by the actual
people who participate directly in them, or they may be labelled
differently from the way relatively external observers classify them
(sometimes leading one group to accuse the other of possessing false
consciousness). This, though, does not argue in favour of relativism as
much as for the need of a formalism with the ability to be able to
translate (or to 'map' in the terminology of mathematics) one
perspective into another. Einstein's theory of relativity not only
stipulated that the way certain phenomena were perceived was
attributable to the frame of reference of the observer (where one sits
determines what one sees) but it also demanded that for such a
pronouncement to make any sense at all the underlying laws must be
invariant and must hold in the same way everywhere. Relativism thus
depends on absolutism.[11]

4.5
Obviously the world does not present itself to us as being as
complicated as it could possibly be. Despite the theoretical potential
for unbridled chaos we live in a condition sufficiently ordered to
permit us to exist and concoct theories. Solitons are complex and
largely non-chaotic entities within the life-span of their existence.
Does this fact mean that our world is made up of events which are
effectively sealed off from one-another's influences after all? Or is
this division of phenomena into such levels of abstraction as 'micro'
and 'macro' just another heuristic whose time has gone? If we accept
this latter position then our chaos- informed formalism must be able to
preserve or at least incorporate the apparent existence of these various
levels of analysis and our human notion of relative stability in these
things.

4.6
We must now go back to deal with Pipes' assurance that, given Lenin,
Stalin would prevail.

4.7
Just as the argument in the 19th century raged over whether laws of
history were reducible to laws of psychology, this debate now has been
transposed into the joust between chaos theorists (those who insist on
starting at the bottom and working up) and those complexity theorists
who claim this is both hopelessly complicated and in any event not
needed. Starting with the analysis of sandpiles, complexity theorists
discovered, after spending many arduous afternoons on Long Island
beaches, that when sand was piled up it eventually tended towards a
critical angle of 51� 52'.

4.8
The process whereby sand piles up effectively at random but sooner or
later a part of it achieves the needed critical angle is called
'self-organised criticality' (Bak et al, 1988).

4.9
If only sandpiles or Egyptian pyramids acted this way social scientists
might never have thought of structures. Many things, though, exhibit
very similar patterns in nature and in society. One area of extensive
study is that of diffusion of innovation. As early as 1971, Fisher and
Pry noticed that capital markets adopted financial products with a very
predictable form: the logistic equation. Ten years later, Marquette
(1981) applied a logistic model to political mobilisation and Cesare
Marchetti (1980) showed that the logistic equation not only predicted
very well the ability of a child to learn a language but also the
general cycles of discovery, invention and innovation. He concluded that
society acted like one huge 'learning system'. Societies learn to use
technology in patterns almost identical to those which are seen in
children learning language. Aggregates of human brains seem to adapt or
learn in ways parallel to the ways in which those aggregates of nuerons
known as human brains do.[12] Power laws can also be used to predict
earthquakes and asteroid strikes on planets including earth.



------------------------------------------------------------------------



Market penetration of four processes or machines in the United States.
The time constant is about 50 years for all four cases. Here the process
is one way there are no cycles involved. The assumption is that the rate
of these penetration processes is the primary clock of the economy. This
rate may be closely linked to human behaviour and may be the deep reason
for the stability of the cycles' length. (Marchetti, 1980)



Note that the power law here is based on a logistic relationship.

Marchetti's suspicion that a 'deep reason' exists for why these
relationships all conform to a power-law relationship is arguably a
manifestation of statistical self-similarity being detected by a
researcher. Self-similarity results from two or more apparently distinct
phenomena exhibiting commonalties. These commonalties apparently pique
human curiosity and cause us to ask "is this merely a coincidence or is
something else going on here?" Human psychology seems to be distrustful
of assigning too many coincidences to the reason of random chance.

Self-similarity is also evident in the lattice approach of connectionism
when people are in some ways networked together in patterns highly
reminiscent of those which connect neurons in the brain. This is by no
means either necessary or sufficient proof of the superiority of a
connectionist paradigm but it is indeed adequate to justify further
research and speculation in this area.

Figure 5: Power-Law Curves Exhibiting Market Penetration



------------------------------------------------------------------------

4.10
This property, if it really is there, is an example of
'self-similarity'. Self-similarity is a general concept which is
intended to convey the fact that a similar (or identical) pattern can be
detected at various levels of abstraction or observation. It is the
manifestation of self-similarity at various levels of chaotic systems
which formally differentiates chaos from randomness.

4.11
Truly random systems cannot reveal patterns but only coincidences which
humans mistake for patterns (Knuth, 1969; Tezuka, 1995). When real
patterns are revealed then chaos is said to be present. Chaos, the
apparent state of disorder or irregularity, is in fact a state in which
strict determinism reigns but the flow of the patterns is formidably
hard to predict owing to the primitive nature of contemporary
mathematical models and computer technology.

4.12
The techniques for identifying patterns within this investigative
paradigm is to speculate deductively about what the rules governing a
system might be (usually these rules are expressed as differential
equations) and then to carry out exhaustive computer simulations which
require the researcher to examine carefully the patterns which do emerge
and try to decide if they look like anything resembling (similar to)
situations of interest. Langton (1995) has applied this technique to his
study of artificial life. Danielson (1992) and others have turned their
attention to artificial morality.

4.13
Other more deductive approaches to large-scale pattern identification
can also be used. Recent hypothetical studies by Rinaldi et al (1994)
 have produced the following application of 'strange attractor' concepts
to the flux of political popularity.



------------------------------------------------------------------------



A Trajectory
P' � P''
close to the degenerate heteroclinic loop.

where:
x(t)= the public support (popularity) of politicians at time t.
y(t) = the hidden assets that the corrupt politicians hold at time t.
z(t) = the investigation effort at time t (ie. the sum of activities by
different institutions such as police, courts, and the press).
and:
A = A(x); A representing positive actions taken
A*(x(t)); A* is the level of positive expectations held by the public
B = B(x,y); B representing bribes and all other moneys earned illegally
by politicians
C = C(B,y); C representing the private consumption of the politician as
constrained by bribes and hidden assets.
D = D(y,z); D representing the discovery of the corruption. (Here: the
amount of money confiscated by the courts).
And this leads to the following model:

1.x' = (+ (A*(x(t)) - A(x)) - (- xD(y,x)
2.y' = xy + B(x,y) - C(B(x,y),y) - D(y,z)
3.z' = (D(y,z) - (z

Where ( represents the positive or negative reactions of the public, (
the persistence of investigations and ( represents the relative
importance of manpower expended to dollars recovered.

Figure 6: Strange Attractor dynamics for the popularity of a corrupt
politician (Rinaldi et al, 1994)



------------------------------------------------------------------------

4.14
Two points can be made here. Marchetti's work has demonstrated yet again
the formal self-similarity of learning systems to learning individuals.
Rinaldi's work shows additional similarities which we can assume would
be equally true of every individual in the population. Like all real
paradigm shifts, the epistemological consequences go beyond mere
redefinition of some terms.

4.15
What, then, are we to do with the attempts to analyse social change
using structuralist terminology? Clearly the success of those who
advocate the adoption of the chaos/complexity paradigm cannot expect to
be taken seriously if all they can offer is an analytic framework which
states something like 'what we call 'structure A' and what we call
'structure B' are simply different places on a complex pathway'. This is
obviously true and almost equally obviously useless as an analytic
statement. Taking a more specific variant such as: 'The collapse of the
USSR was a period of great turmoil and instability which saw the
reassertion of nation states and statelets, the boundaries of which were
much more closely aligned with traditional ethnic territories than the
Imperium they supplanted' will do little formally to aid our analytic
quest.

4.16
What is also true is that the formalism of traditional structuralist
paradigms cannot accommodate the examples above. By effectively
isolating at a formal level the micro from the macro forces,
structuralism all but makes the short-range and micro level all but
epiphenomenal. This confining further means that any attempt to
illustrate the effects of the micro on the macro are to be made
anecdotally or not at all. There remains the suspicion, however, that
the social world may not have an inevitable single final stage. There
may not be a social form which corresponds to the Natural world's Big
Crunch (or entropic death). There may be other options. The world - our
world - is revealed in the chaos/complexity paradigm as will and
representation within the constraints of natural law. What the
chaos/complexity paradigm adds is the possibility of human cognition
playing a role within these constraints.

4.17
We also need to resist the temptation to stray from the requirements of
a formalism. Influences must propagate through a medium such as a
lattice. Memes can, of course, lie dormant in books or carved on granite
blocks or in computer files. They lose any relevance at all if they are
lost or become, like Linear B, indecipherable. Memes, like genes, must
be passed by contact along the relevant dimensions. In the case of memes
it is required that they be passed from one living being to another and
that these beings be sentient.

4.18
It is therefore to the discussion of the matrix or medium within which
chaos arises that we must now turn our attention. Let me now propose
what I hold to be the underlying architecture of our epistemology and
how it can accommodate both our perception of chaos and of order
simultaneously. This architecture has already been dubbed in the
literature with the name 'Connectionism'.
--[cont]--
Aloha, He'Ping,
Om, Shalom, Salaam.
Em Hotep, Peace Be,
Omnia Bona Bonis,
All My Relations.
Adieu, Adios, Aloha.
Amen.
Roads End
Kris

DECLARATION & DISCLAIMER
==========
CTRL is a discussion and informational exchange list. Proselyzting propagandic
screeds are not allowed. Substance�not soapboxing!  These are sordid matters
and 'conspiracy theory', with its many half-truths, misdirections and outright
frauds is used politically  by different groups with major and minor effects
spread throughout the spectrum of time and thought. That being said, CTRL
gives no endorsement to the validity of posts, and always suggests to readers;
be wary of what you read. CTRL gives no credeence to Holocaust denial and
nazi's need not apply.

Let us please be civil and as always, Caveat Lector.
========================================================================
Archives Available at:
http://home.ease.lsoft.com/archives/CTRL.html

http:[EMAIL PROTECTED]/
========================================================================
To subscribe to Conspiracy Theory Research List[CTRL] send email:
SUBSCRIBE CTRL [to:] [EMAIL PROTECTED]

To UNsubscribe to Conspiracy Theory Research List[CTRL] send email:
SIGNOFF CTRL [to:] [EMAIL PROTECTED]

Om

Reply via email to