III.
Modeling Cognition and Communication in Society
III.A.
Moving from classical to post-classical
1.
I have tried to suggest why °classical science and the mainstream linguistics
it sponsored° have not provided adequate foundations for a °science of text
and discourse°. The °classical scenario is resolutely monodisciplinary°
in dividing science into separate fields that pursue their several concerns in
isolation. The scenario has promoted assumptions like these: (a) science
maintains °disinterested objectivity in free-standing facts and timeless truths°;
(b) the worthwhile questions for research are those we can state and answer °within
a single discipline°; (c) those questions must be stated in terms of °testable
hypotheses and observable results°; (d) science has no °ecological
responsibility° to be concerned about the social impact of its work. Today,
these assumptions appear °ecologically
unsound° and unduly limit the scope of science.
2.
At this stage of human history, the vital issues are increasingly °geopolitical°,
bearing on the state of the earth and its global policies. In response, °post-classical
transdisciplinary science° should
be expressly committed to formulating and helping to implement a global °ecologist program for sustainable ways of life°. This task seems
all the more daunting in the °modern age of the great failure to connect°.
Modern societies and their institutions are now beset by interlocking °crises
in materials, knowledge, and communication° (II.133), which classical science
is poorly prepared to address, let alone resolve. Ironically, the ideology of °scientism° provides a general alibi for leaving things as they
are and expecting science to ‘change the world’ at some indefinite future
time (I.3).
3.
In this chapter, we shall explore how °post-classical transdisciplinary science°
might support a °science of text and
discourse in designing models of
cognition and communication in society°. Such models might doubly repay our
efforts by supporting both °scientific
progress toward renewed coverage, convergence, and consensus°, and °social progress toward an ecologist ambience of free access to
knowledge and to social equality
through discourse° (cf. I.1, 33
60; II.128). Conventional discourse practices may have been sustainable when
world and society were fairly simple and stable but not when they are swiftly
growing more complicated and unstable during °modernization,
a paradoxical process that diversifies
and specializes the knowledge a
society holds while imposing new modes of uniformity
and deskilling upon
the roles of participants° (VII.22). Like any °ecologist project°, more
progressive discourse practices will not emerge without a large-scale program to
describe and support them (III.97, 218; VII.174; VIII.118).
We have good reasons to suppose that the potential of discourse for accessing,
sharing, and developing knowledge and enhancing social equality is far greater
than most people know how to attain (I.33, 59; II.5; III.97, 239; V.89). And no
wonder, when the institutions of science and education in modern society have
devoted so little attention to exploring and supporting that potential. Here
again, we can recognize the °fundamental
contradiction in the official ideology of modern Western societies between
inclusive theory versus exclusive practice°. In theory, discourse is like a
‘free market’ welcoming everyone who is ‘competent’ to speak the
language; in practice, each individual is held solely responsible for developing
her or his own °communicative competence° and for taking the full consequences
of the outcome. The contradiction is nowhere more acute than in science and
education, where learners and trainees must take the specialized discourse as
they find it and improvise their own strategies for managing it; and where the
general public is excluded from discussions of technical policies and scientific
applications by the needlessly specialized jargon of the ‘experts’ (cf.
III.47; V.76-79; VII.49, 63, 219).
4.
Post-classical science should develop models to explore the systematic
interfaces among three types of parallel transactions: discoursal moves when making an utterance, cognitive moves when accessing and using knowledge, and social
moves when interacting with other living beings. Clearly, such a model
corresponds to a model of language °interfacing the linguistic
constraints with the cognitive and
social constraints from shared knowledge about world and society°.
Moreover, the same triad of aspects applies to our model of science too, which
has often been judged a ‘purely cognitive’ enterprise for seeking knowledge
apart from how it gets put into words or how scientists fit into society. The °classical°
triad of moves is to give a description of reality (discoursal), uphold realism
(cognitive), and refute other scientists (social).
5.
For a post-classical approach, knowledge is not a static array of facts or
truths built into the world and remaining the same for everybody, but the
dynamic operations of a modeling style. Each style evolves its own methods for °building
models of world and society° through the °dialectical
interaction between the data-driven
or bottom-up processing of experience
versus the top-down or theory-driven
application of prior knowledge° (II.96). In the ‘West’ or ‘Center’
(roughly, the area of Europe, the U.S., and Canada) the leading modeling styles
might be
characterized within the
basic contrasts
shown in
Fig. III.1. Realism situates
the essential domain of the
world in
phenomena,
i.e. in manifestations’ enabling ‘sensory experience’ (seeing, hearing,
touching, etc.), and proceeds by objective
observation; idealism
situates the essential domain of the world in consciousness, and proceeds by subjective
ideation. Also, mechanism holds the world to run on chains of causality, so that things happen out of predictable necessity, e.g.,
through physical forces; fortuity
holds the world to run on chance, so
that things happen out of unpredictable accident. At their extremes, these
contrasting ‘modeling styles’ cannot reconcile
‘subject’ with ‘object’ or ‘mind’ with ‘world’, but can only alienate
them by projecting the world to be either a fragmented array of isolated objects
(extreme realism) or else an insubstantial illusion (extreme idealism).
Reconciliation is vital for healthy, balanced cognition, but grows steadily
harder to attain in a °modernized world of swiftly changing ratios between
diversity and uniformity°.
6.
In the West, realism and mechanism have predominated and have been promoted
whenever a society or culture got °Westernized or modernized°. Their
commonsense versions are found in building-block models, where every object is made of identifiable
smaller pieces and is in turn a piece of a bigger object; and in billiard-ball
models, where one event triggers another the way a rolling ball strikes another
at an exact time and location and causes it to move along
a predictable path (Fig.
III.2). A
‘world-model’ consisting
of tangible objects
and connected events, where
one
thing makes another happen, can be termed classical reality. This model seems °data-friendly: you firmly
believe that what you see, hear, or touch
is ‘real’,
friendly° for enabling wide coverage of ongoing sensory impressions, and °user-friendly°
for supplying human awareness (‘the subject’) with a convergent means of orientation,
i.e., with a scheme for identifying objects and selecting actions or reactions.
And the model reassures you that you can gather knowledge about the world piece
by piece, and can predict one event after another. Ultimately, humans could
‘know’ the whole world and ‘control’ everything that happens in it —
the ultimate world-model with full °coverage, convergence, and consensus° (cf.
III.151).
7.
These sanguine prospects have encouraged strong versions of the classical
realism that collapses the distinction between the real
world versus human models of the world.
Knowledge gets °reified into a frozen
set of timelessly accredited, free-standing facts which experience ‘causes’ to be known and to become the
content of true statements° (cf. III.23); and anyone who sees or says something
else is simply mistaken and must be converted or silenced (III.10, 17, 23f;
VII.199). The mind is seemingly reduced to a mechanism for copying reality, the way a diligent camera records whatever appears
before its lens. But because sensory experiences always vary in fine detail,
simple copying could not sort them into convergent categories, and would thus
fragment the world into bits of data with no orientation for ‘making sense’
of them all or seeing how one thing is connected to another. Paradoxically,
striving too hard to collapse mind with world only alienates subject from object
(cf. III.5, 82).
8.
To organize experience and knowledge, the mind applies a set of basic
postulates of experience and knowledge. These ‘postulates’ cannot be
strictly verified or falsified: they are entailed
in all our accredited processes of
reasoning, discovery, or proof (cf. III.149). The fundamental postulates would
be:
(a)
identity: every object or event is
itself and not concurrently something else;
(b)
connectedness: every object or event
is connected to some others within patterns of part and whole, cause and effect,
and so on;
(c)
temporality and locality:
these connections pass through time and space;
(d)
substantiality: every object is
composed of a substance that has boundaries and properties, e.g. texture and
density;
(e)
observability: humans can register
objects and events with their senses and with suitable instruments;
(f)
dimensionality or measurability:
objects and events manifest properties that can be quantified in standardized
units, e.g., for velocity, mass, or force;
(g)
predictability: the foregoing
postulates make it feasible to specify which events should happen under stated
conditions.
These
postulates sustain a °dialectic
between any objective world
versus any subjective
activities for
constructing any model of
the world°
(Table III.1). The
dialectic follows from the straightforward
premise
that the world must be ‘knowable’ and experienceable’ if it is to be
‘known’ and ‘experienced’, and
vice- versa — things must have identity to be identified, observability
to be observed, and vice-versa. Otherwise,
humans couldn’t attain a °convergence among data and a consensus among
observers°.
9. These
activities are sustained by °cognitive,
discoursal, and social moves° in various patterns and combinations (cf.
III.4). If you repeatedly observe two
events in temporal and local proximity, for instance, you are prone to state a predictable connection such as ‘cause’ and ‘effect’, even
though, as David Hume once pointed out, the causality itself is not observable.
You can harden or weaken your assumption by making more observations under
changed conditions, measuring the variations, or testing your predictions. But
no amount of setbacks would make you state that the objects or events are in
principle not connected (cf. III.149). Instead, data pointing to that
conclusion will not be accredited
as worthy of investigation or
placed into the established categories, and will not be a topic among the
community of observers.
10. An
°objective outlook foregrounding the sensory data from perception and sensation°
implies that the objects and their properties supply their own data of identity,
substance, dimension, and so on. A °subjective outlook foregrounding the data
from cognition, and ideation° implies that the data are supplied by our moves
for identifying, measuring, and so on. But if, following our main premise in
III.8, these two supplies dialectically interact rather than compete, an extreme
emphasis on either supply leads to °radical cognitive moves° accompanied by °legitimizing
discoursal moves° asserting exclusive or absolute claims. °Radical realism and
empiricism° sternly reject any sources or types of knowledge that are not
directly connected to ‘physical facts’ or sensations, witness the standard
discoursal moves in °physicalism, positivism, verificationism, unified science°,
and the like [29-30]. °Radical idealism and rationalism° in turn insist that
no entities can even exist independently of the mind [31-32].
[29]
the only object of the abstract sciences or of demonstration are quantity and
number […] If we take in our hand any volume […] let us ask, Does it contain
any abstract reasoning concerning quantity and number? No. Does it contain any
experimental reasoning concerning matter of fact and existence? No. Commit it
then to the flames: for it can contain nothing but sophistry and illusion.
(David Hume)
[30]
of these languages, the physical, or that in which we speak about physical
things, in everyday life or physics, is of the greatest importance […]
physical language is the basic language of all science [and] a universal
language comprehending the contents of all other scientific languages. (Rudolf
Carnap)
[31]
the various sensations or ideas imprinted on the sense, however blended and
combined together (that is, whatever objects they compose) cannot exist
otherwise than in a mind perceiving them […] all those bodies which compose
the mighty frame of the world have not any substance without a mind, […] it
being perfectly unintelligible to attribute to any single part of them an
existence independent of spirit. (George Berkeley)
[32]
nothing reaches our mind from external objects through the organs of sense
beyond certain corporeal movements […] but even these movements, and the
figures which arise from them, are not conceived by us in the shape they assume
in the organs of sense […] all these things […] are presented to us by means
of ideas which come from no other source but our faculty of thinking (René
Descartes)
The agenda of claiming the sole power to define the ‘mighty frame of
the world’ fosters a °discourse economy° with peremptory generalizations and
exclusions: ‘the only object’ and ‘nothing but’ [29]; ‘universal
language’ and ‘all others’ [30]; ‘cannot exist otherwise’ and ‘all
those bodies’ [31]; ‘nothing’, ‘no other source’, and ‘all these
things’ [32]. Here again is our °contradiction between inclusive (define the
whole universe) versus exclusive (leave no space for the other definition)°.
Evidently, each side felt anxiously defensive and believed that its own validity
depended on flatly denying any validity to the other — on ‘burning its
books’ or reducing it to a straw-man, e.g., to the simple copying of
‘external objects’ in ‘the shape they assume in the organs of sense’.
And here again is the jump from °authority to authoritarian° (II.4, 129). The
disinterested search for knowledge mystified a power-hungry search for conflict
(cf. III.23f); sometimes it was the people and not their books who got
‘committed to the flames’.
11. The
dominance of °classical realism and mechanism° in the West has been vastly
reinforced by classical science, whose agenda is to be ‘reality-driven’ in a different mode from common sense: by
revealing a °fine-grained convergence among objects and events and attaining a
fine-grained consensus about their properties and relations° above and beyond
the °coarse-grained rich details and accidents of ordinary experience° (cf.
III.148; VIII.10). This agenda has enforced a °modeling style° that is both °data-driven
to make observations, discoveries, and so on, and theory-driven to identify and
define the significant properties and relations°. Building steeply upwards upon
the ways in which °ordinary cognition° also combines these two modes of being
‘driven’, science has developed highly specialized methods for deriving and
accrediting its knowledge, chiefly by exploiting realism and mechanism with
advanced technology — whence the importance of insulating the °scenario of
the laboratory experiment from everyday experience° (cf. III.38, 198). Since
ancient times, the °scientistic promise that science would eventually attain
coverage of all human knowledge° has
been sustained by the great discoveries in the °natural
sciences°, such as physics, chemistry, and astronomy, which, as defined in
dictionaries, ‘deal with matter and energy’ and ‘with objectively
measurable phenomena’. Success is gauged in practice by °testing and
verifying a prediction from theory on the basis of observations and measurements°
(III.1). It’s even better to correctly predict a phenomenon before it gets
observed, as when the planet Neptune was independently predicted by John Couch
Adams and Urban Jean Leverrier to explain the irregular orbit of Uranus and was
then discovered by Johann Gottfried Galle where they said it would be; or when
the elementary particle called ‘positron’ (the positively charged
counterpart of the electron) was predicted by Paul Dirac and later discovered by
Carl Anderson, who didn’t even know about the prediction.
12.
Thanks to its resounding
successes, ‘classical science’, as Ilya Prigogine and Isabelle Stengers
remark, ‘presents a vision of nature that would be universal, deterministic,
and objective inasmuch as it contains no reference to the observer, and complete
inasmuch as it attains a level of description that escapes the clutches of
time’. But in our own century, the same advancing techniques and technology
that underwrite scientific method have led to the irrefutable observation of
resolutely °non-classical phenomena°
unsettling even the main postulates in III.8. In the vanguard of the change, a new
physics has acknowledged that the description of the universe provided by °classical
science° is only a °well-behaved approximation°, and that the °classical reality at the base of
realism and mechanism° is just a °well-behaved
subdomain of the universe°. In ordinary experience, classical reality has only
one history (i.e., one time sequence of actual events); non-classical
reality has multiple histories (branching sequences of possible events). So the
fully deterministic and predictable
universe having a single history,
as assumed by classical science, is now reconceived to be a °probabilistic
quantum-mechanical universe having alternative
histories, each of which has only a particular
probability°. We cannot state what must
happen or must be true about reality
but only how likely things are to
happen or to be true. Still, as we move upward from the °sparse base of
elementary particles°, whose histories are quite unpredictable, to the °rich
world of human life°, large numbers of accidental
events interact and converge to
make most of the histories so improbable as to be safely ignored for practical
purposes — i.e., to be ‘summed over’,
in the terminology of physics. To say that one of the set possible events ‘actually
happens’ is to say that the others move their probabilities near to zero.
If so, the reality of both conventional science and common sense could, as
Murray Gell-Mann proposes, be more properly termed °quasi-classical°,
i.e., very nearly classical within a
universe that is fundamentally
quantum-mechanical.
13.
This outlook offers a
fresh context for the old question of how humans access knowledge and express it
in discourse. When we perform such basic moves as identifying, observing, or
measuring, we are not just discovering the
truth but manipulating the
probabilities and seeking a
convergence whereby a single version attains a very high probability while
all the others attain a very low one. When we use discourse to make statements
about ‘what really happened’ we are performing similar moves but with less
direct control, and seeking a convergence between our knowledge of the event
and its representation in discourse; the convergence can be at best a °sufficiently
fine approximation° to be accepted,
understood, or believed by other people. In a world so °richly imbued with
human knowledge and experience°, alternative statements are in principle always
possible; at most, we can try to make them less probable (III.142). And language
provides rich resources to reconcile alternatives and to indicate relative
degrees of certainty or belief.
14. The access to knowledge of non-classical phenomena can be readily demonstrated in the °new physics°. In the ‘EPRB paradox’ (called so after Albert Einstein, Boris Podolsky, Nathan Rosen, and David Bohm) (Fig. III.3), when a particle at rest decays into two photons, they go off in opposite directions. If a detector measures the ‘polarization’ (direction of spin to the right or to the left) of one photon, the polarization of the other is instantly known without measuring it — it must be the same. This result seems to disturb the °basic postulates of connectedness and locality°: how could
measuring
15.
It’s convenient and
popular to say that these °non-classical phenomena° at the base of the
physical universe just aren’t relevant to ordinary reality and our knowledge
about it. Why not simply rope off the °quantum world° as a theory disconnected
from the familiar practices in our °classical world° as if nothing else
existed? Two main answers are close at hand. For common sense, the practical
answer is that our urge to determine ‘the way the world is’ has seriously
restricted our inclination and abilities to understand the world in new ways and
to design a °new and sustainable ecologist program°. Our easy gestures of
grasping reality and truth encourage us to take things at face value and dismiss
other versions (I.58; III.97), as when °monoculturalism dismisses
multiculturalism° (VII.27-34). For science, the theoretical answer is that if
we are to attain a unified vision of nature we need theories extending from the
tiny and simple entities like quarks to large and complex ones like jaguars (to
echo the title of Gell-Mann’s landmark book). Our real world is rife with
non-classical phenomena, such as the patterns of weather or the eddies in a
rapidly flowing stream, whose indeterminacy and uncertainty are comparable to
those found in quantum mechanics (III.152). The domain of °chaos°,
where such phenomena are said to be situated, may be a projection of
indeterminacy from microscopic (sub-atomic) scales to macroscopic scales, as
Gell-Mann and others have suggested.
16.
We should now carefully
weigh the prospects for a post-classical
modeling style, which construes °classical reality and realism° to be not
the ‘real domain’ of all things but a highly successful model domain and modeling
style for attaining some modes of °coverage, convergence, and consensus°
but by no means all. Here, the physical world is ‘not a structure built out of
independently existing unanalyzable entities but rather a web of relationships
between elements whose meanings arise wholly from their relationships to the
whole’ (Stapp) (cf. III.51, 63, 81-84). For such a world, our °first
principle° might be:
What is matter and energy
on one level is information on one or
more other levels, and vice-versa; every entity, however objective or
subjective, is founded on the dialectical
interaction of a material substrate of
matter and energy with a data substrate
of information (Fig. III.4). Our
first principle
extends
from the basic construction of matter and the origin of the universe over to the
emergence and evolution of life-systems and onward to the organization of
cognition and communication in society.
17.
It is a ‘first principle’ not in the usual sense of an unquestioned
foundational ‘axiom’ in philosophy
(II.10) but in the sense that any theory, model, explanation, etc. —
any cognitive or communicative construct — which denies or severs the
dialectical interaction and admits only one aspect or the other, will eventually
lead to breakdowns in °coverage, convergence, and consensus°. Indeed, most of
the great unresolved controversies in the history of Western thought — about
‘body’ versus ‘mind’, ‘flesh’ versus ‘spirit’, ‘theory’
versus ‘practice’, ‘object’ versus ‘subject’, ‘event’ versus
‘observer’, ‘realism’ versus ‘idealism’, ‘physicalism’ versus
‘mentalism’ (and in linguistics, ‘form’ versus ‘meaning’, or
‘sound’ versus ‘idea’) — imply the thesis that the dualism of material
and data should be grasped as a dichotomy,
as a natural and universal line of demarcation. The ‘serious’ or ‘clear
thinker’ merely decides where to draw the line and sets about refuting,
converting, or silencing all the ‘fuzzy thinkers’ who don’t concur (cf.
III.7). So the history of ideas keeps sacrificing consensus under pressure from
disproportionate forced choices, oscillations, and rivalries among unbalanced
and incomplete theories.
18.
Moreover, °classical
realism and science° have ranked the dichotomy in a distinct hierarchy
wherein the data about a phenomenon are ultimately
fully constrained by its material substrate; what you can know about the
‘real world’ is °built into and totally predecided by its architecture° of
physical matter and energy, or of chemical elements and compounds, or of
biological cells and neurons, and so on. Within the °dialectical
outlook of post-classicism°, in contrast, the constraints are mutual and
fluctuating. The material has more fine-grained
constraints than any data access could attain or use and includes minute
irrelevant circumstances; certainty usually requires coarse-grained constraints. (You can readily perceive and know, say,
a ‘real plant’ but even an industrious botanist cannot know all the
fine-grained data built into its
material organization, e.g., the number of cells, molecules, or atoms it
contains.) In this sense, the material is ‘over-constrained’.
But, as the °new physics° shows, the material is also ‘under-constrained’ because without data it would be not just
meaningless but, at its sub-atomic base, a mere sea of possibilities (cf.
III.12ff). In the °quantum world° — where the ‘quantum’ itself is a
packet of energy plus information — even highly simple phenomena like the atom
do not show total certainty between material versus data. In their quest for
certainty, physicists kept analyzing matter in steadily more fine-grained detail
until they unexpectedly crossed the threshold where matter can be objectively observed to behave like information,
without any subjective intervention by humans.
19. For classical science, matter is just matter and not yet information until it gets observed, processed, interpreted, etc.; and ‘purely physical matter’ supplies the most certain knowledge and the soundest topic for ‘the language of all science’ (as in [30] in III.10). The chief lesson of the new physics, along with semiotics, cybernetics, organic chemistry, molecular biology and genetics, neuroscience, psychotherapy, biotechnology, artificial life, and much more, is that the dialectical interaction of matter plus energy versus information is fundamental to all modes of organization. So a post-classical outlook does not have to ask how information is ‘getting around so quick’ (III.13); every system in which particles, forces, and so on interact is a data field with a material base. When an observer enters, the result is not a sudden influx of data where only pure matter was before, but a dialectical interaction between two data fields (Fig.5) — a more ‘condensed’ one in the observed
system which contributes
20.
The broadest and richest
interaction between data fields occurs between the experienced world and a
person’s world-model (cf. III.5ff). This interaction is an exceptionally °dynamic
dialectic wherein the world provides experiences that constrain the
‘world-model’, while the world-model offers knowledge contexts to ‘make
sense’ of experiences°. In operational terms, the ‘world-model’ applies °top-down theory-driven control
by specifying constraints° on what
sorts of objects and events are likely to be encountered, whereas the
‘experienced world’ exerts °bottom-up
data-driven control by manifesting
constraints° in the exact details of the current occasion (Fig. III.6). We
must appreciate
the vital
importance of
this dialectic for every mode and event of experience, perception,
cognition, and communication, however abstract or concrete, including science
itself (III.156).
21.
The dialectic might look
like a ‘logical circularity’,
unable to explain which side produced the other and how, like the banal
‘chicken or the
egg’ problem.
But neither side by itself could originate from zero: neither as an
‘unknown’ world existing with no models of it yet, nor as an ‘empty’
model to which no world has yet been introduced. Instead, the dialectic would be
present at the very emergence of any world as a material-data array. From the
start, the °objectivity (e.g. identity) of the world implicates the moves of
subjectivity (e.g. identifying)° (III.8). Even in the earliest instant of the
universe, when all matter was condensed into a single point of supermass and
superenergy, data was present in extremely sparse modes, e.g., with all physical
forces having a unified identity (cf. III.46).
22. A
great challenge for °designing models of cognition and communication in society°
is to explain how human knowledge can have emerged and evolved up the °coverage,
convergence, and consensus° that it has, and how it might evolve into more
accessible and sharable modes (III.97, 186, 216, 218; IV.163;
VI.30). Since no world-model can ever be exhaustive
or totally exact, every model must be selective
and flexible; but its °approximation
or goodness of fit° with the ‘real world’ can always be enhanced
(cf. III.7). Human knowledge
is limited not just because humans
lack the diligence or their instruments lack the precision, but because the
set of knowledge about phenomena is indefinitely large, and because every
phenomenon is connected to a still wider or deeper context. In return, human
knowledge is also
unlimited precisely because its
current coverage can never be complete or final, and because new
connections can always be made among phenomena. Paradoxically, the necessity of constraints upon our knowledge guarantees our freedom
to shift those constraints (III.31; VI.40). We can vastly broaden our horizons
once we give up the idealized notions that
knowledge has absolute certainty as its endpoint, and that certainty is cumulative and transferable between part versus whole,
past versus present, or between one phenomenon versus another. As the new
physics has proven, any endpoint is ruled out by nature itself; and one mode of
certainty (e.g. position) may have to be ‘financed’ with another mode of
uncertainty (e.g. momentum) (cf. III.14). Also, a fairly certain whole (e.g. a
diamond) may have fairly uncertain parts (e.g. electrons); and understanding the
parts, however clearly, is no guarantee of understanding the whole and
vice-versa (cf. III.51). In some contexts (e.g., those described by Einstein’s
special theory of relativity), even such fundamental distinctions as matter (or
mass) versus energy, or time versus space, turn out to be interactions. The
prospect emerges that ‘all solutions to apparent contradictions lie in moving
away from the opposition and changing the nature of the question to embrace a
broader context’ (Humberto Maturana and Francisco Varela) (cf. III.46;
VIII.144).
23.
°Actualizing the human
potential for knowledge° more fully also requires us to transcend the °classical
primacy of content°, which organizes knowledge by °classifying things into
schemes° of what they are ‘about’, e.g., all objects into ‘animal,
vegetable, or mineral’ (cf. II.9). ‘Knowing something’ gets equated with
having a register of its ‘dimensions’, ‘qualities’, ‘properties’,
‘features’, and so on, which are waiting ‘out there’ to be discovered or
counted. So knowledge is directly built into the world and presented by it;
‘reality’ is consistently and reliably itself, and our only task is to go
gather in the content (cf. III.7, 18, 26, 86). Like classical reality, the
primacy of content seems °data-friendly and user-friendly°, but has serious
drawbacks (cf. III.6f). It encourages °realism in its reification of knowledge
as a modality of ‘facts’° we can refer back to a world that subsists
whether anyone knows about it or not. The set of ‘facts’ is in turn thought
to guarantee that every question has one ‘right answer’ (cf. VII.64ff). If
so, disagreement over content could only come from some distortion or error and
would be an occasion for confrontation and conflict rather than for negotiation
and integration (III.5ff, 10, 17, 24, 97; VII.199). This sinister prospect is
most imminent when whole world-models disagree, e.g., in conflicts between
narrow monoculturalism versus wide multiculturalism (VII.27 34).
24.
A °post-classical
approach to knowledge° might bracket
content by relating it not just to °reality or truth° but to its design
as a product of °modeling styles like realism°. This move differs
significantly from denying, contesting, refuting, falsifying, and so on, all of
which retain the supposition that there is ‘in fact’ some ‘true reality
out there’, and your opponent has simply ‘got it wrong’ through improper
perception, description, and so on, and needs to be forcefully ‘corrected’.
Instead, bracketing rejects the classical Darwinian struggle over the one
‘true reality’ (II.2; III.181) and explores the nature and evolution of
knowledge, allowing for the full range of alternative approaches to reality,
e.g., common sense alongside science as well as literature and poetry.
Bracketing seeks to °integrate the diverse alternatives° that classical
controversies °dichotomize and fragment°. Predictably, ‘bracketers’ may
have problems conducting a progressive discourse with ‘classicalizers’ who
insist on reducing all issues and models to contestable claims and who accuse
everybody with an integrative agenda of being ‘vague’, ‘indecisive’,
‘unscientific’, and so on, doubtless hoping to goad them into confrontation
after all. Their attitude is merely the academic version of the °right-wing
strategy of pursuing power through confrontation° (VII.37).
25.
The °post-classical
bracketing of content to highlight the design° also differs significantly from
the °formalist move of disengaging from content to grasp form as the static and
abstract shape and as an end in itself°,
e.g., a sentence as a °structure of words and phrases specified by a formal
grammar° (cf. II.115). Design, in
contrast, remains engaged with content while focusing on the dynamic
organization and evolution of a
domain as it undergoes drifts
(spontaneous changes, e.g., suddenly having a new idea) and transactions
(deliberate changes, e.g., stating an old idea in more useful terms). Form is
functionally reinterpreted here as the manifestation of a set of interconnected
design events that decide which means
will serve which ends, e.g., what
traits a species assumes in respect to its survival in the environment
(III.162).
26. Four design parameters seem essential (Fig. III.7):
26.1
Stability versus fluctuation:
how far and how fast a phenomenon retains versus alters its current state, e.g.,
its size or shape.
26.2
Familiarity versus novelty:
how far a phenomenon confirms versus disconfirms some known and expected state
or aspect.
26.3
Simplicity versus complexity:
how far the entities within a system and their mutual interactions are few and
uniform versus numerous and varied.
26.4
Determinacy versus indeterminacy:
whether decisions among competing alternatives are clear or unclear.
27.
These four parameters
can help to transcend the °classical notion in both common sense and science°
that content and knowledge have a °built-in or frozen organization° that
humans must accept. Built-in organization can only supply the standing constraints for a design potential to be further specified
by emergent constraints during
evolution. Rather than viewing stability, determinacy, and so on, as fixed
static qualities, we can see them as dynamic factors undergoing drifts and
transactions during the evolution in the design of content and knowledge. We can
then inquire how the ‘features’, ‘properties’, ‘structures’ and so
on, we encounter may have emerged and evolved — whether from short-range
fluctuating versus long-range stabilizing, or from episodic innovating versus
routine familiarizing, or from spontaneous complicating versus deliberate
simplifying, and so on. But we must take care lest we confuse these parameters
because they so often interact: a familiar phenomenon can seem simple and
determinate too as a matter of course, while a novel one can seem complex and
indeterminate. Distinguishing among these parameters allows us to pursue issues
like how familiarity can be deployed to help master complexity, e.g., during the
gradual transition from novice to expert (V.84).
28.
Contrary to what °common
sense and realism° would suppose, these design parameters apply to both °material
and data° in flexible correspondences. Simple material may be connected
with complex data (e.g., when a piece of rock is exhibited as a ‘found art
work’ or as a fossil bearing on paleo-biological theories), or complex
material with simple data (e.g., when a ‘pointillistic’ painting of
color-dots is seen as a human face). The same flexibility holds between a system
and its operations, again contrary to the belief of °classical scientists° who
hold that simple systems operate in simple ways; recent research in °complexity
theory, dynamical systems theory, and artificial life° reveals simple systems
displaying highly complex operations (cf. II.102; III.174). Similarly,
deterministic systems can manifest non-deterministic behavior; fortuity may not
be utter randomness but a °chaotic pattern that is precise yet unpredictable°.
29.
Control can now be more richly defined as a
hypothesis that fluctuation, novelty, complexity, and indeterminacy are
delimited by design constraints. Here, ‘control’ is not just an action
making some events happen and preventing others (the everyday sense), but an
action registering and adapting to constraints in the evolving design of events
or phenomena. Again, this design would not be a built-in or frozen measure of
the phenomena themselves, but an operational projection while experiencing and
acting upon phenomena, even if your action is merely to perceive them. To
control a domain, you apply the °standing constraints specified in your model
of it with the emergent constraints it is manifesting° and thereby attain a °convergent
framework for cognitive, discoursal, and social moves° like orientation,
explanation, and intervention.
30.
This °post-classical
sense of control actions° has a °coverage° ranging from the processes of
ordinary cognition and communication in society over to the most specialized
processes of scientific investigation. Here, science might be brought squarely
within its own scope of inquiry and reconciled with ordinary knowledge by
modeling the ways it builds up its specialized moves and models on the basis of
the general human moves for building models of knowledge (cf. III.11, 33, 68,
98, 174, 182, 188). In contrast, a °classical approach both in common sense and
in science° bypasses this inquiry by assuming that the parameters at the lower
end in the graphic in Fig. III.7 are the general
or normal conditions of reality, while
those at the upper end are special or abnormal
(but see III.42, 44). Ultimately, reality will prove stable, familiar, simple,
and determinate, and we may not find it so because we lack the proper knowledge
and experience, which (scientism tells us) science will eventually supply
(III.143, 151). To view the world this way is to °classicalize°
it, and the projects of science for doing so naturally appeal to humans, whose
ordinary processing also classicalizes and makes things easier to handle by
stabilizing, familiarizing, simplifying, and determining their ‘meaning’.
31.
The °basic postulates
of experience and knowledge° stated in III.8 do support °classical drifts° in
a sparse domain with general and
uniform constraints: there, identity, connectedness, temporality, locality, and
substantiality naturally enhance stability, familiarity, simplicity, and
determinacy; and our abilities to observe, measure, and predict are proof of
this enhancement. But in a rich
domain with specific and diversified constraints, this classical drift recedes
as identities, connections, and temporal and local conditions are proliferated,
and we begin to observe, measure, and predict the limits on our own abilities to
observe, measure, and predict, e.g., for the operations of discourse processing
(cf. III.238). It might seem paradoxical that increasing and specifying
constraints should make things less clear,
but only if we view constraints as °frozen markers built into or around the
phenomenon itself°. The paradox recedes if we view constraints as the
specifications and enrichments that offer steadily more freedom
to select among options and adapt to new contexts (cf. III.22).
32.
A post-classical
approach does not shun the classical by rushing to the other extreme, as some °post-structuralists
and post-modernists° seem to do by programmatically proclaiming absolute
instability and indeterminacy (cf. III.118-24). Instead, we assume the design
parameters to be undergoing continual evolution between stable and unstable, between determinate
and indeterminate, and so on. We can design a flexible model to reflect the
evolution rather than forcing it in one direction or the other to suit either
some staid classical notion about how the world is and what science ought to do,
or some post-structuralist defiance of that notion. We may thus avoid such
unproductive discrepancies as °mainstream linguistics trying to design stable,
determinate models of language by itself while officially discounting the
constraints of world and society that keep the system under control while it
functions°. That thrust toward ultimate certainty only undercut °coverage,
convergence, and consensus° because the constraints had to be unofficially and
sporadically reintroduced through the contexts the linguists were implicitly
supplying (cf. II.49).
33.
A °post-classical
science of text and discourse
seeks to control (describe,
explain, etc.) its domain by building models whose design parallels the design
of the domain° (cf. III.44). If the domain is complex and fluctuating, then a
complex and fluctuating model will be related to it in simpler and stabler ways.
Conversely, a simple and stable model would relate to such a domain in more
complex and fluctuating ways (cf. III.42). The common preference to make things
look simple and stable is °maladaptive° when it misrepresents them and hinders
us from controlling them (III.15, 82, 97).
34.
The transactions for
adjusting the relation between model and domain might be compared to ‘bookkeeping’.
The typical °classicalizing cognitive move is like ‘taking out a loan of control’ by projecting a domain to be simpler, stabler,
etc., than you can show°. Later, you can ‘cancel the loan’ by justifying
your projection (e.g., showing that the domain really is simple) or else
‘repay the loan’ by redesigning your model (e.g., admitting more
complexity). Failing to address such loans while they pile up is like ‘deficit financing’, running your operations on a model whose
relation to the domain becomes steadily more uncertain. Eventually, °disorientation
and alienation° set in when managing the domain events which don’t fit your
model consumes more resources than you saved by the loan. You end up like
ordinary people who pay dearly for their °ecologically unsound inclination to
finance an easy present at the expense of a difficult future° (cf. III.218;
VII.7; VIII.59); relying on easy, stereotyped responses to events in a changing
environment (VII.8), . which have a low adaptive value and become
‘maladaptive’ in the long range, but you may stick with them to avoid the
effort and disorientation of designing and applying more adaptive ones (cf.
III.97; VII.86, 108).
35.
To attain some
provisional control over a domain whose design is not yet well accounted for,
the gains of ‘bookkeeping’ can be raised and its risks lowered by distributing
the parameters within the model rather
than just trying to squeeze them all across the model (cf. III.40, 172). Some ways are graphically
suggested in Fig. III.8. You can distribute determinacy
by bringing some data into the foreground
and relegating others to the background
(in ‘Gestalt’ terms, by favoring ‘figure’ over ‘ground’), or by
treating some data as more coarse-grained
and other data as more fine-grained
Or, you can distribute fluctuation by
correlating some short-range
processes that seem less stable with some long-range
ones that seem more so. Or again, you can distribute complexity
by handling some data in a local
perspective of ‘micro-states’ and other data in a global
perspective of ‘macro-states’.
36. Distributing would be highly strategic if, as I have long surmised,
control is subject to conservational trade-offs
of design parameters. Making some things in your model more
determinate makes others more
indeterminate (Fig. III.9). The concrete
phenomena in
ordinary experience (e.g. a stone or a chair) appear fully °classical,
despite being more properly quasi-classical°
(in the sense of III.12). These concrete phenomena are coarse-grained
in being highly stabilized, and
their material base is fairly hard-coupled
to one data field about them (e.g. size and
density). Alternative
states of being or
alternative data sets have low
enough probabilities to be ignored. Stated in terms of quantum theory since
Richard Feynman, the °alternative histories
are summed over and the branchings in
evolution decohere°, such that only
one is selected to be made nearly
certain, while all others are made extremely improbable; and the selected one
incurs very slight interference from
the rest. When you enter a room and recognize some chairs, you don’t perceive
their fine-grained weight, height, width, etc., and you are not disoriented if
they vary in shape or if
you seethem turned at
different angles. You ‘make sense’ of the phenomena by determining which
relevant data should converge into °interobjectivity among your experiences
with the same object° (III.84) and leaving the rest indeterminate. If the
occasion required (e.g., you had to revarnish the chairs), you might determine
the size and shape more exactly. Still (like the botanist with the plant in
III.18), you would stop at a fairly coarse grain and never go down to the really
fine-grained data of, say, molecules.
37.
This °trade-off of
determinacy° would hold far more for abstract
phenomena; indeed, the larger the trade-off, the greater the abstraction. If you
are an Aristotelian philosopher seeking to define ‘the Good’, you won’t
list every ‘good’ object and event along with all its ordinary circumstances
— the total would be much too °rich and diversified to converge° into a
usable definition. Instead, you would take a set of actual or possible instances
and °rarefy out voluminous data° in hopes of determining just the relevant
data. So the determinacy of an abstraction gets traded off against massive
indeterminacy of concrete circumstance. Here, the branchings in evolution do not
decohere so strongly as for the concrete, and we may get interference among
alternative branches, e.g., when we try to evaluate a real person’s
‘character’ as ‘good’ or ‘bad’. Whether the person did
or did not do a particular action is a
quasi-classical question insofar as we have ways to tell whether the action was
selected or not; whether the action was good
or bad is a much less classical question, and whether the person had
‘good’ or ‘bad’ intentions (a key point in courts of law, VI.81) is even
less so. Evidently, ordinary experience and ordinary reasoning are not so
uniformly classical as common sense implies, but are more so in some domains and
less so in others.
38.
In science, our °trade-off
of determinacy° would hold in a more specialized way. The observations and
technical instruments in laboratory experiments are attuned such that specialized
modes of finer-grained data can be filtered out from the coarser-grained
specifics and irrelevant accidents of ordinary experience (III.11). If you run
an experiment to test the sparse ‘EPRB’ effect sketched in III.13, you
attend closely to paths of the photons but not, say, to location of the lab in
England or Switzerland, nor to the social status of the lab assistants. Your
determinacy is focused on defining the evolution of photons among alternative
histories, which physics tells you covers all photons and thus allows an
unlimited number of °convergent observations° (‘replications’), each one
differing from all the others in some coarse-grained but irrelevant details
(e.g., whether it was conducted by somebody who personally knew the late David
Bohm, who designed it). But if you do research on the rich domain of ‘human
intelligence’, the laws of physics can apply only through an elaborate °progress
of enrichment to supply the constraints specifying human control factors°,
e.g., the resources, such as attention, demanded by various tasks (cf. III.H).
And you will need to °diversify and integrate multiple tactics°, each of which
contributes its own selective constraints, e.g., some tests for social factors
and others for psychological ones; and besides the sparse data, e.g., about
‘microseconds’ of ‘reaction time’ on ‘closed-ended’ tasks, you will
need ‘rich’ and ‘noisy’ data, e.g., about ‘creativity’ on
‘open-ended’ tasks (cf. III.185). You might well find, as Howard Gardner has
eloquently argued, convincing evidence for ‘multiple intelligences’ rather
than the single unitary ‘intelligence’ supposedly measured by
‘standardized tests’ (cf. VII.109, 116).
39. In a similar trade-off on another design parameter, simplifying some things in your model would complicate others (Fig. III.10). The °commonsense building-block strategy° is to simplify a phenomenon merely by isolating it, which complicates its relation to its context. So a model with simple components requires complex interactions (Fig. III.11a), whereas a model with more complex components allows for simpler interactions (Fig. III.11b). To isolate °language by itself,
formalist mainstream
linguistics° postulated high
simplicity both for the components, such as the
phonemes and morphemes, and for their formal interactions in terms of
size and constituency° (II.D). Later, the complexity ‘squeezed out’ this
way exploded in syntax and semantics°, and
can only be integrated by a model of substantially higher complexity both for
components, such as °prosody and lexicogrammar, and for functional interactions
in terms of means and ends° (cf.
II.60f; III.61, 67). So huge a °loan of simplicity° can only be repaid through
a fundamental reorientation to °reintegrate language with world and
society°. Thereafter, how simple or complex our account should be is a °functional
data-driven issue rather than a formal theory-driven issue° (II.79). We may
discover a distribution wherein ‘peaks’ and ‘valleys’ of simplicity and
complexity function as ‘attractors’ for language data, e.g., wherein
specialized terms ‘attract’ ordinary terms toward specialized meanings, or
vice versa (cf. V.77).
40. All these trade-offs can apply not just to the internal organization of a model but also to the °goodness of fit° between the model and its domain. If you forcefully ‘squeeze’ complexity and indeterminacy out of your model by drastically simplifying and determining (Fig. III.12), they ‘flow
across’
into the goodness of fit connecting the model to its domain, thereby gaining
complexity and losing determinacy
(Fig. III.12). So, a drastically simple and deterministic model can connect to a
complex, non-deterministic domain only in complicated, vague ways. For example,
a simplistic model of language as a set of determinate labels, each referring to
just one object, can hardly connect up to real discourse, a medium where the
degrees of complexity and determinacy are continuously being regulated. This
regulation enables seemingly simple components like ‘the’ or ‘of’ to be
used in more complex and indeterminate functions than seemingly complicated
components like ‘ontogenesis’ or ‘superstrings’ (cf. IV.7).
41.
A highly adaptive
strategy for exploiting the trade-offs is to cross-factor the parameters. Raising
complexity can be offset by lowering
determinacy (Fig. III.13a), e.g., learning a new skill by relaxing the
thresholds of precision; or enhancing
determinacy can
be offset
by raising
simplicity (Fig. III.13b), e.g.,
describing a system of distinctive ‘sound-units’ (as phonemes) while
leaving aside
the intonation of discourse. Human processing has apparently
evolved just such cross-factoring strategies.
We manage the experiential complexity of ordinary experience by accepting
margins of indeterminacy, and we ‘finance’ some experiential determinacy in
selected domains through loans of simplicity. In this perspective, the notion
that determinacy is firmly built into material reality and must be held constant
while complexity rises has naturally created serious problems for °classical
realism and
the science and
education based
upon it° (cf. VII.85). Classical
scientists seek to manage the complexity of every entity by analyzing it into
simpler components, and expect that this isolation from context will enhance
rather than dissolve determinacy. Physicists were utterly unprepared for the °undecidable
quantum indeterminacy° they eventually encountered in the sparse sub-atomic
world (cf. III.12ff, 18). Similarly, °sparse formalist linguistics° sought a
complete description by analyzing a language into a set of fine-grained levels
with deterministic minimal units in phonology and morphology or with
deterministic rules in syntax and semantics and eventually stagnated in a
massive complexity of units, rules, and notations that stymied °coverage,
convergence, and consensus°. A °rich functionalist linguistics° does not
strive to combine completeness with determinacy but allows for the °correlations
between forms versus functions and between means versus ends° being more
determinate in some domains of the °lexicogrammar° and less so on others; and
for the description being finer-grained in some areas and coarser-grained in
others to adjust to the varying degrees of °delicacy° (cf. II.65, 78; IV.16).
42.
If we °bracket the
classical belief of both common sense and science° that a ‘normal’ or
‘good’ design is always a simple and determinate one (III.30), we can
reassess their values in terms of whether they are °supportive
or resistive for control° (Fig.
III.14):
42.1
Supportive stability (viz. a
long-term democracy) assists reliability that makes things dependable, whereas resistive
stability (viz. a stereotyped routine) leads to stagnation
that makes things turgid and immobile.
42.2
Supportive fluctuation (viz.
adrenaline regulating the diameter of blood vessels) offers flexibility
that strategically adapts to evolving conditions,whereas resistive
fluctuation (viz. an erratic heartbeat) leads
42.3
Supportive familiarity (viz. knowing
how to perform a difficult job well) assists orientation
that helps you get your bearings,
whereas resistive
familiarity (viz. getting bored with a dull, repetitive job) leads to
banality that reduces interest.
42.4
Supportive novelty (viz. breaking the
genetic code) assists creativity that brings a sensation of discovery and excitement,
whereas resistive novelty (viz. being
thrown into
a roomful
of bizarre strangers) projects volatility
that disorients sensible conduct
42.5
Supportive simplicity (viz. enduring
folk art) offers straightforwardness that captures the essentials and helps evolution
toward supportive complexity, whereas resistive simplicity (viz. a ‘simpleton’) leads to obtuseness
that makes a superficial, disorganized sampling and evolves either not at all or
toward resistive complexity.
42.6
Supportive complexity (viz. in a
scientific theory or art work) arises from successful integration
of many diverse factors, whereas resistive
complexity (viz. in an unmanageable calculation) arises from disintegration due to too many and too diverse factors.
42.7
Supportive determinacy (viz. precise
instructions) offers perspicuity that enables strategic decisions, whereas resistive
determinacy (viz. a blind prejudice) leads to bias
that ignores significant data about individual cases.
42.8
Supportive indeterminacy (viz. the
aesthetic experience of a modernist art work) offers openness
that enriches by integrating concurrent possibilities, whereas resistive indeterminacy (viz. totally hazy or contradictory
instructions) leads to vagueness that
founders among the possibilities.
43.
We can now formulate an
‘enriched’ definition of progress
as an evolution during which a system
shifts its parameters from resistive toward supportive; the opposite
evolution is regress. Obviously, such
progress need not occur from the mere passage of time, as common sense imagines;
nor need it follow from technological advances (III.52, 227; VII.7, 21). The °modernization
of societies° toward the end of the twentieth century reveals distinctively
regressive tendencies, where new technologies have not prevented the °disintegration
of control over a sustainable ecology°, as we shall see in Chs. VII and VIII.
44.
For human processing,
supportive drifts or transactions foster °reconciliation
between model and domain or between subject and object°, whereas resistive ones
foster °alienation° (cf. III.5, 7,
34). A ‘good’ reconciling design for a model is accordingly not just stable,
familiar, simple, and determinate, but rather parallels the design of the domain
while keeping its parameters on the supportive side (cf. III.33). By expecting
the ‘laws of nature’ to be simple and determinate with regular fluctuation
and little genuine novelty (III.30), °classical science encouraged scientific
theories with a deterministic design for a sparse domain wherein predictions can
be directly tested through a content-based ‘accuracy’ of facts and figures°
(cf. III.1, 7). Fluctuations within a system were believed to naturally seek
stability (at ‘equilibrium’), so that a disturbance (or ‘perturbation’)
is normally soon levelled out by °diminishing returns° (II.98). It remained
for °post-classical science to show that scientific theories can be valid with
a non-deterministic design for a rich domain wherein content-based predictions
are often not directly testable°; and that a °small disturbance can amplify
into a sweeping change through increasing returns°. These new principles have
demonstrated how °chaos can be unpredictably productive when its boundaries
provide unrestrained openness and unplannable creativity° — whence the
popular paradox ‘order out of chaos’ (Prigogine and Stengers). Today, new
perspectives on how life-systems might have attained their ‘order’ are
rapidly converging in numerous disciplines, from organic chemistry and molecular
biology over to demographics and economics (cf. II.102; III.163).
45.
Moreover, the °classical
view that parameters like determinacy are frozen attributes built into the
domain° rather than flexible design parameters in our models had imposed some
unproductive °stagnation (resistive stability) and bias (resistive determinacy)°
upon the range of theories to be constructed and upon the domains they should
address. Domains where fluctuation and indeterminacy seem rather high, such as
the meaning of language or discourse, were either avoided or pressed into models
that remained resolutely stable and deterministic, as in formal semantics
(II.48f). Such investigations may turn away from the main
phenomena toward the more deterministic and stable epiphenomena
that happen to accompany them, e.g., not meanings but the grammar of dictionary
definitions (III.170).
46.
Conversely, a °post-classical
outlook seeks to build models whose design is either stable or fluctuating,
simple or complex, and so on, in supportive modes that favor progress° in the
sense of III.43. We may succeed in °enriching our own modeling style° by
achieving more determinate knowledge about the non-determinacy of a domain, or
simpler knowledge about the complexity of a domain, or by discovering novelty in
a familiar domain, or stability within a longer-range pattern of fluctuation.
Rather than just ‘squeezing’ complexity and indeterminacy out of our model
by drastically simplifying and determining (as in Fig. III.12 in III.40), we
enlist them for °control and progress° by ‘re-evaluating’ them from °resistive
disintegration and vagueness over into supportive integration and openness°,
e.g., by °integrating phenomena that were hitherto considered unrelated or
diverse into a unifying account° (compare III.22, 153). This move would allow
the goodness of fit between the
model and its domain
to °raise straightforwardness (supportive
simplicity) and gain
perspicuity (supportive
determinacy)°, and the model
can then connect up to its domain in a simpler and clearer way (Fig.III.15).
The move can be seen most impressively in those Nobel-Prize-winning triumphs
of science that achieved massive °unifications°,
i.e., convergences between previously distinct theories. Some renowned
quasi-classical unifications united celestial with terrestrial motion by Isaac
Newton and electricity with magnetism by James Clerk Maxwell. A renowned
non-classical unification united the electromagnetic force with the ‘weak
force’ by Sheldon Glashow, Abdus Salam, and Steven Weinberg; and a feverish
race is now on to unite these two forces with the ‘strong force’ holding
together the nucleus of the atom and with gravity. In a fascinating development
of terms, a ‘force’ is now interpreted as an ‘interchange of messenger
particles’: ‘photons’ for the electromagnetic force, ‘gluons’ for the
strong force holding the nucleus of the atom together, ‘W+,
W-, and Z bosons’ for the
weak force regulating radioactive decay, and (presumed for consistency but not
yet observed) ‘gravitons’ for gravity (cf. III.94, 165; V.93). Superstring
theory is the leading candidate for the final ‘grand unification’ between
quantum theory (the three ‘quantum forces’, i.e., electromagnetic, weak, and
strong) with gravitational theory (force of gravity), whose mutual inconsistency
has been an agonizing problem for modern physics. I see some rapport between my
own post-classical work on modeling versus the ways that the °cognitive and
discoursal moves° of superstring theory strategically trade off complexity
against indeterminacy. By postulating more ‘dimensions’ inaccessible to °coarse-grained,
quasi-classical ordinary experience° besides the °classical three dimensions
of space and the Einsteinian fourth dimension of time°, and by postulating huge
ratios of energy, density, and velocity, the universe can be modeled in a
fine-grained state where entities (particles and forces) were only minimally
differentiated (III.21). In that state, matter and energy would be unimaginably
concentrated, and a modest repertory of indeterminately evanescent and unstable
simple particles could then spread out and eventually °stabilize and complicate°
into some of the basic elements of chemistry (III.49). Such is widely believed
to have occurred during the origin of the universe from the ‘big bang’.
47.
Such triumphant unifications display scientific theory °integrating a widened
range of phenomena° that differ or alternate starkly in their concrete
circumstances, and making them °converge into a network of fine-grained
connections° (III.8). As the experts settle down into a new °consensus°, and
the novelty of the unification wears off, the unified domain appears far simpler
and more determinate than before, as if all pertinent °loans have been repaid
or canceled°, opening a new action space for higher-level moves across
ever-vaster domains of data. In return, the theories whereby the unifications
were accomplished can seem forbiddingly complex to non-experts, who may
‘know’ that the domain is unified but certainly cannot demonstrate how or
why by physical experiments or mathematical calculations. The next step should
be to enlist discourse in extending access to the non-experts and supporting
them in making new connections for their understanding of nature (cf. III.3; V.C).
III.B. Prospects for evolution in the design of models and
domains
48. The general factors in the design of a model or domain, as they were sketched in section III.A, can now be °enriched to specify the evolution of design°. Our °first principle stating the dialectical interaction of a material substrate of matter and energy with a data substrate° can be specified for two types of coupling between the two substrates. Phenomena characterized by hard coupling have tight, sparse constraints between material and data (Fig. III.16a). Here, we can expect the more
°classical design of ‘lawlike’ connections from part to part (like building blocks) or from cause to effect (like billiard balls)° as mandated by the forces of physics (e.g. gravity) or the reactions of chemistry (e.g. bonding). Phenomena characterized by soft coupling have loose, rich constraints bet