From: Lars Risan (lrisan@ulrik.uio.no)
Date: Thu Jun 24 1993 - 09:57:41 UTC
***********************************
I have been told that my posting two days ago was difficult to read on
IBM-PC. Hope this one works better.
***********************************
I am a student of anthropology at the University of Oslo working on a
cand.polit. thesis (somewhere between the American Master and Ph.d in
scope) on Artificial Life.
Anthropologists do fieldwork, that is, we spend some time in the culture
and among the people we want to study. My "culture" will be AL-theory and
related ideas, and my "people" will be researchers on an
AL/AI/system-science-lab. Such a fieldwork normally lasts one year or so.
This is meant as an informal inquiery to map places where research programs
on Artificial life and related ideas are taking place. I don't expect any
invitation to do my fieldwork at this stage, but if the reader of this is a
part of a lab/institute where there is several people doing "AL" (or
"emergent AI"), I would be grateful if you could E-mail me a few word about
your lab (what kind of projects that are going on in your lab, and how many
researchers you are).
If you want to know more about my project, the following is a brief outline
of my ideas so far.
Lars Risan
Centre for Technology and Culture
The Research Park, University of Oslo
Gaustadallaen 21
0371 Oslo - Norway
Tel (private): +47 22 46 40 90
email: lrisan@ulrik.uio.no
******************************************
Artificial Life - Computer Technology and Modernity.
******************************************
1. Introduction.
In this paper I will draw a brief outline of my project on Artificial Life
for my cand.polit. degree in anthropology. I am not going to use AL-theory
to study social phenomena, but to reflect about AL as a part of modern
society.
By situating AL in the history of ideas I will focus both on its
novelty and its roots in western scientific thinking; on its modernity and
its post-modernity. Two tendencies in contemporary society are of special
interest to me. Both of them are connected to the introduction of computer
technology and cybernetics/information theory. The first tendency is the
"Faustian spirit" (as Goethe+s Faust is interpreted by Marshall Berman
[1982]): "The desire for development", the modern drive to shape and
control our worlds. The flexibility of the computer opens up for new
possibilities to realise these wishes. The second tendency of interest to
me is the introduction of the concept of information and complexity in
science. By introducing these concepts in the study of nature, the "science
of complexity" (including AL) is changing our understanding of nature, and
thereby also our understanding of man's relation to nature. This may create
a new humility towards nature that was unknown to Doctor Faustus. How these
two tendencies meet, interact and create something new, is the main
emphasis of my project.
I am going to develop my argument further by first discussing the
"Faustian spirit", the concept of information and the concept of complexity
separately, and then see how these tendencies are emphasised differently in
different modern sub-cultures, from cyberpunks to the green movement. I
will not make any hasty conclusion as to where to place A-lifers in this
picture. Indeed there is probably no common "AL-ethic", but hopefully, at
some later stage in my project, I will be able to say something about some
possible artificial life world views.
2. Faust--and "the desire for development".
The myth about Doctor Faustus is as old as modernity itself, first to
appear in written form in 1587, telling the story about a marginal
character, an intellectual nonconformist, who liberates himself from
traditional social bonds, cultivating his individualism and searching for
novelties and experience. His companion is Mephistopheles, the Devil, and
his road leads him to hell. But he is still a hero, a man who excites us.
In Goethe's tragedy Faust ends up using his creativity and seemingly
endless power to construct new cities and design landscapes. Even
Mephistopheles is breathlessly left behind, and God grants Faust the
developer a place in Heaven. The morality is clear: What is valuable is not
only individual growth, creativity and liberation, but to spend these
resources in making progress for the whole of humanity.
This Faustian spirit is obviously present in the AL-project; "By
the
middle of the next century [man] will be able to create [life]." [Langton
1989:43] (It should be noted that this is not simply a descriptive
statement, but also a normative one--when said by someone who is trying to
do just that). Laing [1989] is speculating on how to colonise the moon with
self-reproducing factories, and Moravec [1989] is looking forward to when
superior robots--with better "hardware" than the obsolete human body--have
made the human race extinct.
There seems to be something with the computer that promotes this
wish to create and control one's world. Sherry Turkle [1984] examined this
desire among young hackers, and she found that they preferred the
relatively controllable complexity of the computer program, to the much
more uncontrollable complexity of human relations. In the virtual reality
of the computer they can develop their individuality without making any
commitments.
I am not saying that this extreme position is typical for A-lifers. This
becomes apparent when the above quotation of Langton is seen in its well
known context:
By the middle of this century, mankind had acquired the power to extinguish
life on Earth. By the middle of the next century, he will be able to create
it. Of the two, it is hard to say which places the larger burden of
responsibility on our shoulders. Not only the specific kinds of living
things that will exist, but the very course of evolution itself will come
more and more under our control. The future effects of changes we make now
are, in principle, unpredictable--we cannot foresee all of the possible
consequences of the kinds of manipulations we are now capable of inflicting
on the very fabric of inheritance, whether in natural or artificial
systems. Yet if we make changes, we are responsible for the consequences.
[Langton 1989:43]
There is something paradoxical in both making artificial life and
suggesting that it may have larger consequences than the atomic bomb and
stressing that these consequences are in principle unpredictable. But here
Langton is in line with Goethe. Goethe himself wanted to realise large,
modern construction-projects. He dreamt of building both a Suez-canal, a
Panama-canal and a canal connecting the Danube and the Rhine (which is now
being built!). Still he was very aware that the Developer's right hand was
the devil, that Faust was causing death, pain and devastation on his way.
It seems that being modern involves an acceptance of change and alienation,
an acceptance that being old isn't being wise, but being outdated.
3. William of Occham--and the concept of "complexity".
The rule of parsimony or Occham's razor states that if two explanations
with respect to everything else are equal, choose the simplest one. This
explanatory principle was first stated explicitly in different formulations
by the middle age philosopher William of Occham, for example as "plurality
must not be asserted without necessity" [1955:97]. The principle soon, at
least after Descartes, became one of the guiding stars of physics. Newton's
laws of motion and gravitation and Maxwell's unification of the electrical
and the magnetic forces, perhaps two of the greatest achievements of
classical physics, are both good examples of the use of Occham's razor.
They
are both a set of simple equations with great unifying power: A lot of
phenomena that prior to the formulation of these laws needed separate
explanations, could now be mapped onto them.
A lot of aspects or traits of the modern society as it developed from the
17th. and 18th. century, are correlated or somehow connected to this
principle.
The search for simplicity led us to discover those rare phenomena
of
nature that are linear because, as is well known, non-linearity generates
complexity. Occham's razor, then, was also made into a search for
linearity.
This correlates with the inclination, since Galileo Galilei, of asking how
instead of why, and thereby obtaining mechanical descriptions instead of
teleological explanations [Ian Stewart, 1989]. Science was then, in the
centuries to follow, increasingly legitimated and motivated by the Great
Wonder of Technology instead of the Great Wonder of Nature [Hoffmeyer,
1985]. This fitted well with the growing instrumentalistic attitude of the
growing modern, capitalistic societies.
The computer, then, is a consequence of Occham's razor, of unifying,
simple,
elegant, and linear equations. However, from one point of view, the
computer can also be regarded as the ultimate result of this development.
"Ultimate" in the sense of "last"--because it led to its own negation. Its
iterative power made science discover complexity and non-linearity; chaos
and computational irreducability--just to name a few.
There is of course a lot of continuity here. The new "science of
complexity" is understanding complex systems by reducing them to
formalisms. But still I think a growing understanding of complexity will
have a considerable impact on the instrumentalistic Faustian spirit that I
outlined above.
4. Descartes--and the concept of "information".
Rene Descartes is famous for his division of the world into "res cogitas",
the conscious, and "res extensa", the extended (material) world. "Res
cogitas" was something unique to man. Each human being had its individual
body and its individual consciousness. Even if Descartes thought that
machines might be made to reproduce language mechanically, real semantic
ability was something that required consciousness, something unique to man.
This view can be seen in Descartes understanding of the concept of
information. One only had ideas when something, via the hypothalamus,
influenced res cogitas. That influence was information [Kirkebyen 1993:6].
The rest of the processes--in the brain or elsewhere--were part of "res
extensa", that is, these processes were pure mechanics.
This fundamental division has been central to Western thought.
Semantics, meaning and information have been understood as something which
requires a human, interpreting subject, a consciousness. All else has, at
least to a major part of scientists and philosophers from the Enlightenment
to the end of the 19th. century, been conceived of as mechanics.
A major event in the 20th. century was when natural scientists and
mathematicians, like Warren McCulloch, Norbert Wiener et.al., started to
ask questions about information and communication. Information was taken
from the (elevated) sphere of human interaction and made into a natural
phenomena. However, as these scientists started to speak about information,
they also changed that concept. And the fundamental change was to make
information into an objectified, atomistic quantity, to be counted in bits
and bytes. They made a concept that, according to Shannon, was independent
of meaning. This means that the new (mainstream) understanding of
information didn't really break with Descartes' dualism--as the result was
that we got two understandings of information. The humanists retained their
subject-dependent Cartesian information (that isn't usually dedicated to
Descartes), and the realists defended their objectified
Shannon-information.
However, in the last decade or so it seems to me that the critique
of
Shannon's concept of information has changed. It isn't so much his
mathematical reduction of information that is criticised any more, as his
(and others') informal interpretation of these formalisms [Qvortrup 1993,
Nyrretranders 1991, Kristensen 1985, Kuyppers 1990]. Shannon wanted to make
a concept of information that was independent of meaning, and in his
description of his own formalisms he gives the impression that he succeeded
in doing so. His concept of information is however subject dependent, and
this dependency is hidden in a p (in the equation I = -ld(p) ) that stands
for the probability that the transmitted pattern of data will be priorly
known to the observer. Gregory Bateson, in trying to provide an alternative
to the objectified concept of information, defined information as "a
difference that makes a difference". This is actually a better informal
description of Shannon-information than Shannon himself did. If I know that
a message will be A, the A received makes no difference to me, and the
information content is nil. But if I have no idea of whether I will receive
A or B (just the knowledge that I will receive one of them), the difference
the message will make is at its maximum of one bit.
So, after 50 years of cybernetics and information theory,
Descartes'
dualism is under heavy criticism. There is an increasing understanding of
the subject-dependency of information, and information is seen as a natural
phenomenon. That is, meaning is about to be integrated into the concept of
nature, into "res extensia".
This change in our understanding of information is a part of a
broader
cultural drift in Western societies. I will give a brief outline of this in
the following section.
5. "The Bionic world view."
The distinction between nature and culture is a fundamental one in Western
culture. The former is, when seen in comparison with the latter, often
portrayed as chaotic, wild, irresponsible and often dangerous. While the
latter, in this context, represents the civilised, ordered, European
culture (with its satellites). The relationship between them is like a
bourgeois living room surrounded by wild bush (and wild bush-men). The
European has tried to maintain this picture by associating everything dirty
or disordered with the concept of nature. A variety of phenomena have, at
different times and places, been classified as "nature"; Negroes to the
white Anglo-Saxon Protestant, jazz music to the European intellectual of
the thirties, women and children--just to name a few.
Descartes' distinction mentioned above is an expression of this
on the philosophical level, a distinction which is institutionalised in the
academic world by the separation of humaniora and natural science. And,
contrary to the medieval thought systems that were both normative and
descriptive, both a theology and a natural philosophy, the modern system
has made a division of labour: Norms are given to the humaniora to puzzle
over, while the natural sciences describe the world. And what's more, the
validity of norms have been limited to "culture" while "nature" has been
treated more or less like a non-moral domain.
This is about to change. One important development in this
direction
is connected to the globalisation of the industrial society. There is
nowhere to hide the hole in the ozone layer; the nuclear cloud from
Chernobyl reached every part of Europe. "Nature" is knocking on our nice,
clean bourgeois door and telling us that it is about to be a part of us.
There is a parallel development in our moral attitudes: animals are given
rights, and there is a growing acceptance of the inherent value of species
and ecosystems. The idea that artificial life, machines, should be treated
like autonomous agents in a legal sense are--to the proper humanist--the
most
extreme expression of this development.
These new ideas are a part of what the historian David F. Channell
[1992] calls "The bionic world view." While the old clockwork-world view of
the Enlightenment was based on the metaphor "life is mechanics", the new
world view has turned the metaphor around and states that "mechanics is
life." Even if there always is a symmetry in metaphors, a "cross
fertilisation" as the poster to this conference states, the difference
between these two metaphors is profound. The former assumed that the
phenomenology of the living could be explained in principle by the language
of mechanics (later biochemistry), while the latter takes as its main
source the phenomena and language of the domain of the living, and then
apply this on machines. In the headlines of this conference there appear
words as such "history", "adaptive", "botanical growth" and "death"--words
that had no place in the science of the Enlightenment. The result of this
is that the organic world view (from alchemy, via vitalism and to holism)
and the clockwork approach are fused together. I will not go further into
the theoretical aspects of this tendency, it is both "unabashedly
mechanistic and reductionist" [Langton 1989:6] and toying with ideas like
".. the whole is more than the sum of the parts" [ibid. p. 41]. That is, it
is AL-theory--and you know that better than me. I would rather discuss what
ethos, what kind of pattern of attitude, motivations and especially ethics,
that may follow from, or promote these theories. I will discuss this in
relation to the two possible consequences of information technology that I
discussed above; the further development of the Faustian modernisation and
the introduction of the concepts of information and complexity.
The classical modernist society is only assigning instrumental
value to "nature". The latter, be it timber or slaves, is something to be
used and thrown away, something of no value except what it is worth to man.
The virtual reality dream of hackers and cyberpunks exhibits this attitude
taken to the extreme; one can enter into a whole new world that exists
entirely for one's own sake. No commitments, no responsibilities.
In strong opposition to this extremity is the endavour of
cybernetician and anthropologist Gregory Bateson. Hoping that by
understanding social processes through the use of cybernetics and system
theory, one could make the world more sane, he attended the early Macy
conferences on "control and communication in animals and machines" (to
quote Norbert Wiener a bit out of context). Disappointed that the world was
entering into another arms race, and that cybernetics did not bring
anything new but better missiles and commercial computers, he spent the
last 10 years of his life at the "New Age" centre Esalen in California.
Developing a pantheistic natural philosophy inspired by cybernetics, he
stressed the inherent value of natural systems, and tried to detach himself
from all kinds of instrumentalism, both physically through his retirement
to Esalen and through Zen Buddhism.
There is, however, no necessary contradiction between ascribing
inherent and instrumental value to something or someone. We do that all the
time. A good bargain is based on the mutual understanding that 1) both
parties are following their own interest and 2) both parties accept the
other one's legitimate right to follow his or her own interest. And,
according to Channell, in the coming era of the bionic world view our
relation to nature will be more like such a bargain than the old Faustian
instrumentalism--that had more the flavour of a rape.
As I see it, such a "bargain relation" is promoted by, firstly,
applying the relational concept of information on biological processes, of
realising that "nature" is a part of "culture", and secondly, by
understanding the complexity of natural systems that surround us, that is
by realising that "culture" is a part of "nature". Accepting the relational
concept of information means realising that e.g. manipulating genes is not
"physics", it is communication. This again implies that the organism
reading those genes is an interpreting subject and is thereby an autonomous
agent. But if we give messages, we should also be able to read the answers,
as all communication is a process of mutual co-ordination. Trying to
unravel these messages from nature implies that we have to situate
"culture" in "nature", we have to learn to see ourselves as part of a
hugely complex system. That is, there should be no genetic engineering
without a large dose of ecology.
If then, we make an axis with the instrumentalism of virtual
reality in one end and the detachment of Bateson on the other, where is AL
to be placed on this axis?
On the one hand there is the obvious Faustian instrumentalism of
artificial life. This is related to the fact that AL has to establish
legitimacy as a part of a larger socio-economic system which is sustained
by the following inter-dependencies:
- There is no technology without wealth, and
- no wealth without technology.
- There is no technology without (scientific) truth, (that's why nation
states invest in science) and
- no (scientific) truth without technology (because you cannot take part in
the scientific discourse without the technology to produce data).
The integration of truth, technology and economic factors is the
Faustian modernism institutionalised in science. The AL-project is a part
of this institution. The integration of basic science and applied science
may even be further developed in the AL-project where the domain to be
described is in itself technology, and where new discoveries sometimes also
are new inventions.
On the other hand there is the possibility for a new humility
promoted by the cautiousness that comes with the realisation of the
complexity of nature, and the unity (and thereby empathy and respect) that
may follow from the "cultivating of nature" [Hoffmeyer, 1985].
As I stated in the introduction, I am not going to place AL on the
"inherent value/instrumentalism" spectrum at present. Further inquiries
into the AL-project will bring answers, although I do not expect these
answers to be generalisations about some common "AL-ethics". It is
essential to map the variety of attitudes and thoughts of scientists and
philosophers who are shaping important features of our future relationship
to nature.
References.
Bateson, Gregory [1972], Steps to an Ecology of Mind, Ballantine Books, New
York, 1985.
Bateson, Gregory [1979], Mind and Nature. New York, Bantam Books 1988.
Bateson, Gregory [1987], Angels Fear. London, Rider Books 1988.
Berman, Marshall, All That Is Solid Melts Into Air, Simon and Schuster, New
York, 1982.
Channell, David F. The Vital Machine - A study of technology and organic
life. Oxford University Press, Oxford, 1991.
Hoffmeyer, Jesper, "Tvedelingen mellom kultur og natur" in Informations
samfundet, Thomas Suderqvist (ed.), Urhus, 1985.
Kirkeb?en, Geir, Descartes' tilnurming til persepsjon i lys av "cognitive
Science": Optikken (1632) vurdert ut fra D.Marss (1982) Vision. University
of Oslo, 1993.
Kristensen, Peer Hull, "Samfundets information og informationssamfundet" in
Informations samfundet, Thomas Suderqvist (ed.), rhus, 1985.
Kuyppers, Bernd-Olaf, Information and the Origin of Life, The MIT press,
Cambridge, 1990.
Langton, Chris, "Artificial Life" in Langton, Chris (ed.), Artificial Life,
Addison-Wesley, California, 1989.
Laing, Rickhard, "Artificial Organisms: History, Problems, Directions" in
Langton (ed.) Artificial Life, Addison-Wesley, California, 1989
Moravec, Hans, "Human Culture: A Genetic Takeover Underway" in Langton
(ed.) Artificial Life, Addison-Wesley, California, 1989
Ockham, William, Philosophical Writings, edited by Philotheus Boehner,
Montreal, 1955.
Qvotrup, Lars, "The controversy over the Concept of Information" in
Cybernetics & Human Knowing Vol.1, no. 4, 1993.
Shannon, Claude and Weaver, Warren, The mathemathical theory of
communication, The University of Illinois Press, Urbana, 1984.
Stewart, Ian, Does God Play Dice?, Penguin Books, London,1989.
Turkle, Sherry [1984], The Second Self, Teknisk Forlag A/S, Kobenhavn,
1987.
This archive was generated by hypermail 2.1.7 : Tue Oct 14 2003 - 21:44:17 UTC