NETFUTURE
Technology and Human Responsibility
--------------------------------------------------------------------------
Issue #154 February 12, 2004
--------------------------------------------------------------------------
A Publication of The Nature Institute
Editor: Stephen L. Talbott (stevet@netfuture.org)
On the Web: http://www.netfuture.org/
You may redistribute this newsletter for noncommercial purposes.
Can we take responsibility for technology, or must we sleepwalk
in submission to its inevitabilities? NetFuture is a voice for
responsibility. It depends on the generosity of those who support
its goals. To make a contribution, click here.
CONTENTS
---------
Editor's Note
Quotes and Provocations
CyberTrackers: Bushmen and Information Technology
On Treating Hyperactive Children with Drugs
Do Plants Think?
DEPARTMENTS
About this newsletter
==========================================================================
EDITOR'S NOTE
Two brief notes:
I wish to thank all of you who responded so generously to the December
request for subscription donations. It is wonderful to be able to put
NetFuture out with the sense of warmth, gratitude, and heightened
responsibility that so naturally results when one receives such deeply
motivated support. While we fell a little short of our yearly fundraising
goal, we benefited from a surplus of energizing good will and even
sacrifice by many donors. Thanks you!
The main articles from In Context #10 (Fall, 2003) are now available
online. In particular, you'll find Craig Holdrege's study of the giraffe
(including a discussion of the not-very-salutary role of its long neck in
evolutionary explanation), and my own paper entitled "Qualities". These
and other articles are available at
http://natureinstitute.org/pub/ic/ic10. In Context
is a twice-yearly hardcopy publication of The Nature Institute.
SLT
Goto table of contents
==========================================================================
QUOTES AND PROVOCATIONS
CyberTrackers: Bushmen and Information Technology
-------------------------------------------------
"The art of tracking", writes Louis Liebenberg, "may have been the origin
of science". As a physicist who has spent many years tracking with the
Bushmen of the Kalahari Desert, Liebenberg speaks with some authority.
And there can in any case be little doubt about the remarkable
observational and interpretive skills of expert trackers -- skills that
would be the envy of many scientists (or, at least, of those relative few
who still occupy themselves with the appearances of the natural world
rather than with instrument readings and abstractions). A good tracker
can read a detailed public story written upon a littered forest floor
where the rest of us would see only a chaotic mass of dead leaves. The
bushmen of the Kalahari can identify an individual rhinoceros by examining
the pattern of cracks in its droppings -- cracks determined by the
intestine's distinctive convolutions. And they can make a good guess
about the psychological state of an elephant (are poachers harassing it?)
by examining the distribution of the droppings.
Not that the world's few remaining elder trackers are miracle-workers.
They readily acknowledge making mistakes. Mark Elbroch, who recently
spent several weeks in the Kalahari with Liebenberg and the native
trackers -- and who is himself one of the world's premier trackers --
tells me of the "tremendous humility" of the Bushmen. After exploring the
signs in a particular area, they may engage in heated debate among
themselves, refusing to venture an opinion to outsiders until they reach a
consensus. "Their curiosity is amazing", Elbroch says. "They are always
asking questions and trying to push their knowledge further".
Elbroch is concerned to do away with the miracle-worker idea. Certainly
there is special aptitude in some cases. But if you pay attention
to the environment long enough -- and especially if your life depends
on it -- you will learn the basic skills. In every domain of life,
whether it's a Picasso exploring the expressive potential of simple,
drawn lines, or a physician reading volumes about the condition of a
person in the appearance of his eyes, or a trainer getting to know the
ways of a particular animal, we find ourselves amazed by the results of
a disciplined, life-long attention to meaningful detail.
What such attention yields is a capacity to read a larger context, or
whole, through the qualitative expression of the part. The slightest
shift of intonation in the animal's growl may reveal to the trainer the
inner state of the animal, its circumstances, and the impending events.
Only a long habit of inattention to contextual relations makes such skill
seem miraculous.
In 1996 Liebenberg collaborated with a Cape Town computer scientist,
Lindsay Steventon, to develop the prototype of what is now known as the
CyberTracker system. Under this system, a Palm Pilot with a GPS unit and
icon-driven software can be put into the hands of an illiterate Bushman,
who then may enter up to several hundred "data points" in a single day --
tracks and signs, animal movements, predator attacks (successful and
unsuccessful), plant species distribution, and so on, depending upon what
is being studied. These observations are fed into a remote database where
the data is assembled into a larger ecological map. Many are predicting a
bonanza for wildlife conservation, habitat management, and ecological
research.
If encounters between one of earth's remaining hunter-gatherer societies
and modern technology are inevitable, this is the way you would like to
see it happen -- under the guidance of someone who has spent years
learning the native wisdom, who appreciates what the culture has to offer,
and who sees the technology as a way to encourage the practice of native
skills while bringing these into productive contact with a wider world
that will in any case increasingly make itself felt.
There are currently ten Bushmen employed to use CyberTracker in
conservation work. Many of these -- for example, a former road
construction worker -- would have found no other opportunity to pursue the
skills they love.
Liebenberg is currently setting up an evaluation program for certifying
master trackers in preparation for their employment. Elbroch tells me
that, in the case of one elder tracker whose sons considered his skills
useless in the modern world, the situation changed dramatically when the
father received his Master Tracker certification. The sons decided they
wanted to pick up these valuable skills for themselves.
If Liebenberg's spirit were to rule, the Bushmen might be enabled to
evolve in relation to the larger world, not simply through being
overwhelmed by it, but rather through exploration of the yet-unrealized,
more forward-looking potentials of their own culture. But we do need a
realistic awareness of the risks. The history of encounters between
cultures and technologies is replete with examples of social breakdown and
demoralization. My own unease on this score is hardly allayed when I hear
a Wired magazine commentator saying,
Liebenberg's work seamlessly connects the earth's oldest form of
knowledge to its most modern, sophisticated, and automatic counterpart.
It represents an extraordinary moment in technology transfer. Indeed,
Liebenberg has produced something akin to a Stone Age computer by
hacking into a bygone world.
What strikes me in this romantic, Gibsonian rhetoric is a preoccupation
with the "cool factor". But such a preoccupation is the one thing that
can virtually guarantee disruption rather than a "seamless" interaction
between cultures. Tracking, after all, is not "cool" for the few
remaining Bushmen who pursue the traditional ways; it is life and
survival. To inculcate our own gadget-worship in them would be to violate
almost everything worthwhile in their lives.
All this comes into better focus when we consider the scientific role
of CyberTracker. The Bushman's astonishing ability to understand the
interworkings of his environment through careful, integrative observation
is the polar opposite of the scientist's predilection for analyzing a
thing down to a set of decontextualized data points. The Bushman is
engaged in a thoroughly qualitative act of recognition that depends
on his having grasped, in a participative and imaginative sense, the
characteristic way of being of a thing. He knows the animal "from the
inside". The scientist, on the other hand, is bound to distrust both
the notion of sympathetic participation and the phrase, "way of being",
preferring to take the thing as an external given and immediately set
about the act of (preferably quantitative) analysis.
It is obvious enough that each of these poles could be enriched by the
other. Here we see the positive potential of Liebenberg's experiment.
The Bushmen stand to gain from a growing ability, not only to immerse
themselves harmoniously in their environment, but also to stand apart from
it in a mood of detachment and ever more sophisticated analysis.
Likewise, the scientist needs more than well-honed analytic skills. He
also needs the ability to participate in and thereby recognize the
wholeness, the integrity, the organic processes of becoming -- none of
them mechanistically describable -- that give us significances worth
analyzing.
We should recognize, however, that within the extremely one-sided
scientific and technological realms today, analysis always seems to trump
the recognition of wholes. Data wins out over the qualities of things.
So it is that the unity of the organism -- for example, of the animal
whose entire manner of being speaks to the Bushman through the smallest,
all-revealing signs -- dissolves for the scientist into a collection of
tissues, or genes, or survival strategies, or whatever.
If you doubt the difference between the scientist's and the Bushman's
styles of awareness, ask yourself whom you would rather have at your
side if you suddenly found yourself marooned in a wild and dangerous
place. An expert in genetics and evolutionary theory, or a Bushman?
And please note: the kind of insight serving your personal safety is
hardly unrelated to our long-term, collective safety upon planet earth.
Let us hope, against all odds, that an echo of the Bushman's qualitative
understanding will somehow be passed along with the computerized data
points to the researchers at the other end of the satellite link. And we
might hope further that this will counterbalance the single-minded
obsession with ever more massively assembled data, submitted to ever more
extensive computer manipulations.
There will truly be a renaissance within science when we can combine our
quantitative and analytic skills with a sense for the way the world
actually presents itself to us -- that is, with a receptive and
imaginative eye for the contextual unities and inner, meaningful
significances that give us more than a dead (if well-analyzed) collection
of parts. Some day, I'm convinced, we will learn to track our photons and
electrons, our genes and proteins, our Martian strata and cometary dust,
as much in the spirit of the Kalahari Bushman as in the spirit of the
computer analyst. Then, and only then, will we find the good and proper
place for computational technique within science.
(I try to point toward the essential qualitative dimensions of science in
the first of the articles listed below.)
Related Articles
----------------
"Qualities" in In Context #10:
http://natureinstitute.org/pub/ic/ic10/qualities.htm
"Hold a Blossom to the Light" in NF #141:
http://www.netfuture.org/2003/Jan2803_141.html
"Technology, Alienation, and Freedom" in NF #134:
http://www.netfuture.org/2002/Jul1802_134.html
On Treating Hyperactive Children with Drugs
-------------------------------------------
Here are some notes drawn from the discussion of children and drugs in
Beyond Therapy: Biotechnology and the Pursuit of Happiness. That
document, as I mentioned in the last issue, is a report of the President's
Council on Bioethics and is available at http://www.bioethics.gov. I have
interleaved comments of my own with the various remarks by the Council.
** The administration of psychotropic drugs to children tripled during the
1990s, in many cases approaching adult rates. Drugs are increasingly
given to children for the same wide-ranging reasons they are given to
adults. For example, antidepressants such as Zoloft and Prozac are
prescribed for depression, obsessive-compulsive disorder, tic disorders,
and anxiety disorders such as separation anxiety and school refusal. As
another example:
Research is actively under way exploring the use of mood stabilizers
(for example, Lithium) to treat children and adolescents for bipolar
disorder, oppositional defiant disorder, conduct disorder, episodic
explosiveness, and mood lability.
** Even in the adult case, we know precious little about the relation
between the psychotropic drugs we take, on the one hand, and, on the
other, the ultimate causes and deeper significance of our disorders, or
the needs they express. We are primarily treating symptoms. But with a
child the question arises: given a young human being with a lifetime of
adaptation and growth ahead of him; given the need for problems to work
against in achieving this growth; given that it is generally the parent or
teacher, not the child, who "suffers" the symptoms being treated; and
given the uncertain relation between symptom-removal and the underlying
needs of the child -- what clearly articulated reasons do we have, rooted
in the child's welfare, for attempting the chemical removal of these
particular symptoms? I'm sure that in at least some cases there are good
answers, but they don't seem to be widely advertised.
** The Council reports that, for whatever reasons, diagnosis of Attention
Deficit/Hyperactivity Disorder (ADHD) in the U.S. is on the rise, as are
estimates of the condition's prevalence. The conservative estimate ranges
from 3 to 7 percent of school-age children. Only slightly more permissive
criteria suggest 17 percent. Three to four million children are thought
to be taking Ritalin-like stimulants, and this number increasingly
includes very young children down to the age of two.
** Regarding the stimulants prescribed for ADHD, the Council writes:
For the worst cases, these drugs have proved a godsend, rescuing many a
child from failure in school, trouble with authorities, and general
shame and opprobrium. In the great majority of children diagnosed with
ADHD, stimulant drugs (frequently used in combination with non-medical
efforts to alter behavior) have apparently succeeded in enhancing focus
and attention, calming disruptive behavior, and improving performance
at school.
I do not doubt the cases where the drugs may seem a necessity. But
wherein lies this necessity? Is it intrinsic to the child? Or is it
instead found, for example, in a one-size-fits-all schooling system that
sets the child up for failure, and in a society that heaps shame and
opprobrium upon him? In how many cases is the pharmaceutical "godsend" a
patch for preserving a dysfunctional society?
From this perspective, drugs look like the perfect tools for avoiding
social reform. If you can erase the symptoms, turning children into
perfect citizens of the educational factory, why bother with the vexing
issues of reform? Far easier to administer a few pills for the removal of
disruptive symptoms than to deal with massive and deeply entrenched social
pathologies.
The same temptation -- to treat isolated effects in lieu of transforming
the underlying pattern -- beckons to us on every technological front. For
example, if you can genetically engineer a crop to thrive under
excessively saline conditions, why worry about those interwoven aspects of
farming and society that brought about the salinity in the first place?
** The Council is concerned above all with the "implications of inserting
the novel and precedent-setting use of drugs into child-rearing and
educational practices", as opposed to their use in strictly therapeutic
situations. This application of medicine "beyond therapy" allows parents
and teachers to "intervene directly in a child's neurochemistry when the
child behaves in a way that defies their standards of conduct". Moreover,
the coercion in this becomes nearly invisible:
Whereas the overt behavior of today's overbearing parents may elicit a
friendly reminder or a rebuke from grandparents or neighbors -- "Take
it easy on him; he's just a kid!" -- the use of drugs to attain similar
goals proceeds out of sight, immune to the correcting eyes of others.
** Regarding this use of drugs, I do not see a clear distinction between
therapy and realizing the broader potentials of human development.
The Council itself acknowledges that "the distinction between therapy
and enhancement ... is finally inadequate to the moral analysis".
Personally, I tend to think that the medicalization of domains not
previously subject to therapeutic technique is no more a problem than
the unhealthy medicalization of medicine itself. That is, the problem
doesn't lie so much in the widening of the sphere of therapy as in the
narrowing of this sphere so that it embraces little more than simplistic
notions of defect and remedy. Even a broken bone -- if treated with a
mere mechanical fix and ignored as an event that belongs to the afflicted
individual as an essential challenge along his path of development -- is
an occasion for medical abuse.
** An adult world pressing children to grow up as fast as possible easily
forgets that "childhood is generally marked by a spirit of
rambunctiousness that, especially in the case of young boys, often borders
on sheer unruliness and hyperactivity. Curbing the latter may too easily
stifle the former".
We tend to forget that temperaments selected over eons of evolution ...
are not obviously well-suited to sitting quietly in classrooms .... We
fail to consider that [children's] spiritedness might be part of a more
ambitious nature, their lack of attention part of an artistic
temperament, or their restlessness a fitting response of genuinely
eager students to uninteresting or poorly taught classes.
While the Council endorses the therapeutic use of behavior-modifying drugs
in difficult cases, it questions the casual reliance on drugs as a general
strategy for obtaining well-balanced children. It notes that "most
children whose behavior is restless and unruly could (and eventually do)
learn to behave better, through instruction and example, and by maturing
over time". Drugs short-circuit this learning process by acting directly
on the body, thereby separating achievement from the effort of achieving.
This raises the question whether we are looking for the mere outward,
behavioral result, or instead for the inner shaping of character that can
only be learned:
If the development of character depends on effort to choose and act
appropriately, often in the face of resisting desires and impulses,
then the more direct pharmacological approach bypasses a crucial
element. The beneficiaries of drug-induced good conduct may not really
be learning self-control; they may be learning to think it is not
necessary.
** Finally, in a profoundly important contextual survey, the Council
summarizes various social factors playing into the ADHD "epidemic" we see
today:
Anecdotes abound of schools and teachers pressuring parents to medicate
their children, often as a condition of continued enrollment; of
doctors pushed by hectic schedules and distorted insurance rules,
prescribing stimulants to children they have not fully examined; and of
parents seeking a quick way to calm their unruly child or pressuring
their doctors to give their son the same medication that is helping his
schoolmates. Powerful social pressures to compete, prominent in
schools and felt by parents and students alike, may play a role in
encouraging extra stimulant use. The Individuals with Disabilities
Education Act, without intending to do so, has created financial
incentives for schools -- and parallel incentives for parents -- to
push for an ADHD diagnosis and treatment. Insurance requirements that
tie reimbursement to diagnosis (rather than to need) also conspire to
push for more diagnosis and more drug treatment; so do insurance rules
that base doctors' fee schedules on the number of visits with patients
and provide greater compensation for short visits offering drug
treatment than for longer sessions exploring behavior-changing
approaches.
In a major (and worrisome) change from previous practice, drug
companies have taken to marketing drugs directly to parents, with spot
ads depicting miraculous transformations of anxious, lonely, or
troublesome children into cheerful, confident, honor-roll students.
The presence in virtually every community of children known to be
gaining advantages from stimulants creates a temptation for other
parents to offer similar advantages to their own children. In
addition, strong evidence suggests the growing illicit and self-
medicating use of Ritalin and similar stimulants by high school and
college students, taken (often by snorting and at higher doses) to
enhance focus and concentration before important exams or while writing
term papers.
Further, as the Council notes, a doctor's diagnosis of ADHD permits
college-bound students extra time in taking the all-important SAT exam.
"It will be interesting to discover whether more students now declare
themselves victims of ADHD, eligible not only for extra time on exams but
also for stimulant drugs that could improve their attention and
performance".
This last point underscores the truth that narrow remedial measures can be
self-defeating as a response to deep-rooted social issues. It is laudable
to make some allowance for individual circumstances -- for example,
by granting extra time for taking tests. But the realities of a test-
taking culture immediately overwhelm this positive gesture and may even
turn it into a negative. This is because the special allowance makes no
sense when you're administering tests the whole purpose of which is to
provide an objective and universal measure of individual performance that
can be compared quantitatively with the performance of others.
If you really want to assess the unique capabilities and potentials of an
individual student ... well, there's no escaping the intense and deeply
personal effort of the assessment. A single, standard, nationwide test
will not get you very far. And, short of a radical change in approach,
attempts to introduce an element of fairness here and there may lead only
to a muddle, inviting competition-driven students to take advantage of the
system.
Do Plants Think?
----------------
Time was when loose talk about how plants can think immediately pegged you
to the flaky fringe. (Remember Peter Tompkins' The Secret Life of
Plants?) How things have changed! Nowadays all you need to do is
mention computation and put "think" inside obscuring quotation marks and
-- whether you're speaking of plants or just about anything else --
suddenly the flaky fringe becomes the cutting edge.
So it is that we have this report in Nature (Jan. 21, 2004),
written by Philip Ball:
Plants appear to "think", according to US researchers, who say that
green plants engage in a form of problem-solving computation. David
Peak and co-workers at Utah State University in Logan say that plants
may regulate their uptake and loss of gases by "distributed
computation".
More specifically: plants open and close microscopic pores called stomata
in order to regulate their exchange of gases with the atmosphere.
Especially under the stressful conditions of drought, plants must (in
Peak's words) "solve a sophisticated formal problem: how to maximize
carbon dioxide uptake from the atmosphere while experiencing no more than
a fixed amount of evaporative water loss".
Peak and his colleagues discovered that stomatal openings and closings
occur in patches over the surface of a leaf, and the way these patches
shift around shows statistical similarities to the way elements of certain
cellular automata move around. (The cellular automata they have in mind
are, in effect, computer simulations of computers. More on cellular
automata below.)
The researchers believe their statistical correlations provide
quantitative support for the idea that "a plant solves its optimal gas
exchange problem through an emergent, distributed computation performed by
its leaves". All of which leads many to conclude, as one observer
exclaimed, "Cool -- plants can think!"
Well, okay. But then so can gases. Want a gas to calculate the result of
dividing X by 2? No problem. Just take a box with two equal
compartments, put X molecules of the gas in one compartment while leaving
the other one empty, and then open a small hole between the compartments.
Then wait while your nifty emergent distributed computer chugs away.
After a short time you'll find your answer given as the number of gas
molecules in either compartment. Through more elaborate divisions of the
box you could, in principle, obtain other, more complex calculations.
Cool -- gases can think! Nor is this a mere put-on. I have no doubt that
a mathematician could, at some abstract level, analyze the gas'
performance in terms of an emergent distributed computation in much the
same way as was done with the plant stomata.
But if you're suspecting there's something rather gaseous about all this
thinking, you're right. What's remarkable is not so much the amount of
thinking supposedly going on in plants as the amount of thinking not going
on in humans.
But first things first. It's not only plants and gases that seem to be
calculatingly thoughtful. The arctic tern migrates halfway around the
earth, from pole to pole, with the precision of an intercontinental
ballistic missile. The migrating monarch butterfly and the spawning
salmon perform similar feats. Come to think of it, no one really disputes
that every form of life we know is compacted of various forms of
thoughtfulness -- intelligence if you prefer. And even in the inanimate
world, the sun, moon, earth, and planets have an uncanny ability to
execute movements that almost perfectly duplicate, in one way or another,
the sophisticated calculations we have labored to put into our ephemeris
tables. Strange thing!
Or perhaps not so strange. In earlier eras the idea that the universe has
the character of thought was hardly remarkable. The cosmos is, after all
(as we now put it), lawful. What are these laws, even the most reduced,
most purely quantitative ones of our own science, if not a content of
thought? How could we even conceive of a world that was not, at least in
part, a world of concept? The universe is as fundamentally, as
primordially, as objectively, its conceivability -- its thought content
-- as it is anything else. And if, like a good Cartesian, you try to
imagine what that "anything else" might be -- how it can be wholly other
than thought -- I guarantee you a wild intellectual ride.
But surely distinctions are necessary. There is a difference between the
thought-imbued products of intelligent activity and the activity itself.
A world compacted of living thought is not the same as a world thinking.
(See "Intelligence and Its Artifacts" in NF #148.) If we have no reason
to imagine that the moon is busy thinking about and calculating the
ephemeris tables, neither do we have reason to imagine that the arctic
tern has conceptualized modern navigational theory. Nor are there grounds
for lumping artifacts with organisms, as if the butterfly makes its way
under the guidance of an ICBM-like computational guidance system. Or as
if plants are solving "sophisticated formal problems".
To leap to such conclusions is the sheerest anthropomorphism.
We solve formal problems because, thanks to our extraordinary powers
of abstraction, we have learned to pose those problems to ourselves and
then tackle them in a formal, computational way. The world in general,
whatever its thought-imbued character, does not happen to be engaged in
that sort of activity.
Allow me to carry the matter a little further. Philip Ball, the Nature
columnist, links the "plants can compute" idea to the view that the world
is almost nothing but a particular sort of computational algorithm:
the laws of physics might arise from units of matter, space and time
interacting with one another according to simple rules.
That's an amazing statement of an increasingly common sort. On such a
view the "simple rules" become, for purposes of explanation, almost
everything, with the world's substance disappearing into the anonymous
"units" of matter, space, and time. These units, which receive no
characterization, seem to serve little purpose beyond magically "giving
rise" to and embodying the rules. What the world's substance might be
according to its own nature is roundly ignored. Much of science today is
in flight from substance, preferring instead the neat, clean certainty of
simple computational abstractions.
This, of course, perfectly illustrates my thesis in "The Vanishing World-
Machine" (NF #151): the mechanistic thinker today not only takes all
things to be machines, but is rapidly substituting the machine's algorithm
(which is an abstraction originating in our own thinking) for the effort
to understand the embodied mechanism as such.
You may recall, from that earlier article, my description of the Game of
Life computer program, with its displayed grid of cells that turn "on" and
"off" according to simple rules relating to their own state and that of
surrounding cells. This is an example of a cellular automaton (CA). It
happens that, depending on the rules and the starting configuration,
particular aggregations of "on" cells move around your screen in
interesting ways. In fact, it's possible, in principle, to construct a
"universal computer" from these moving forms. Less ambitiously, CAs can
be designed so as to carry out certain basic computational tasks.
I spoke of constructing a "universal computer" (in quotes) because it's
obvious that you can't really do this. The shifting patches of light and
dark on the screen just don't have the kind of causal power required of an
actual computing machine. One patch of screen or blob of light is not
physically interacting with other patches or blobs so as to carry out the
computation. One cell does not (as is often said) "look" at the
surrounding cells to see their state. The interacting and "looking", such
as they are, take place in the silicon and circuits of the physical
machine executing the CA.
So here is another example of how scientists and engineers speak of a
"machine" when they mean "algorithm". What the computational CA's display
gives us is not a computing machine, but rather the visualization of a
computational algorithm. That is, it gives us the simulation of a
computer.
All this helps us to get clearer about the impropriety of claims for
computation or thinking in plants. Leaving aside the radically uncertain
meaning of "statistical correlations" between stomatal dynamics and CA
displays (what exactly is the significance of any such correlation?),
there remains the following crucial point:
Whereas the patterns of light on the computer screen do not themselves
carry out any computation and do not causally participate in the physical
execution of the CA program by the computer's hardware, the situation is
radically different for the leaf's stomata. These cells take their place
alongside the plant's other cells, and they are thoroughly integrated into
all the activities of the plant. They contribute structurally, helping to
determine the plant's shape and strength. They share in the genetically
related transactions of the organism. Their internal water is part and
parcel of the larger plant's water economy, and they participate similarly
in the plant's gas economy, nutrient economy, electrical economy, and the
vastly complex biochemical workings of the plant in general. Every aspect
of the plant's overall condition at any moment is likely to be reflected
one way or another in the individual stomatal cell.
The situation in the computer couldn't be more different. The real estate
of the screen display is not in the same organic way integrated with the
power supply or CPU or metal box of the computer as a whole. What we see
on the screen is a representation of an algorithmic abstraction. The
representation could just as well take on a completely different form --
for example, it could be given as columns of numbers -- and this would
make no essential difference to the operation of the algorithm. By
contrast, the plant stomata cannot be so arbitrarily meddled with. Their
form and functioning is in every detail bound up with the form and
functioning of the plant as a whole.
So, then, what is the message of David Peak and his colleagues to the rest
of us? That plants behave like computing devices? Nonsense. The only
way to show this would be to show it -- to demonstrate the same
sort of physical structure and goings-on in both cases. But, as I have
just suggested, it is a hopelessly long leap from the computer to the
plant.
If, on the other hand, the researchers are simply attempting to find,
at an extremely high level of abstraction, some sort of vague and
statistical similarity -- well, fine. But it doesn't tell us very much.
Of course there must be some similarity between stomatal dynamics and
a CA designed to perform an isolated task related to those dynamics --
just as there must be some similarity between a monarch butterfly and
a properly targeted ICBM. And if you abstract away all the differences
between the butterfly and the ICBM -- everything, say, except the starting
and ending points of their flight -- you could pronounce them in that
regard "the same". So what?
As for the leaf's stomata, the one thing you can be sure of is that the
pattern of their openings is not the result of a set of simple, CA-
like rules encoded as programs and executed by a computer-like machinery.
Show us the program and the machinery before you make any such claim. If,
on the other hand, this is not the claim, then we need to hear with much
more clarity what is being asserted.
The researchers conclude with the following remarkable sentences, which,
however, scarcely offer the needed clarification:
Our analyses are only a first step, of course, in connecting
computation and plants. If, like CA model systems, leaves really did
perform emergent, distributed computation, then the phenomenon of
stomatal patchiness would no longer be a puzzle: patchiness is a
companion of computation. Moreover, such computation would also be a
feat of surprising subtlety. In CA simulations, even though processing
occurs without a central controller, a central unit of some kind
initiates the computation and later reads out and interprets the
"result". In the case of leaves, stomata are simultaneously the
sensors of external information, the processing units that calculate
how gas exchange regulation should occur, and the mechanisms for
executing the regulation. Thus, in those plants that solve the dilemma
of optimal gas exchange, evolution may have found an elegantly
parsimonious computational technique in which input, output, and
processing are all accomplished by using the same hardware.
Here is an acknowledgment that plants are not at all like computers or
computer simulations. If they bear the imprint of thought -- and they do
-- it is in an organic manner very different from the way computers bear
this imprint. What, then, is the justification for the continued
reference to "hardware" and "computational technique" -- as if plants
were like computers? Sure, plants achieve a result one aspect of
which we can program into a cellular automaton, just as the monarch
butterfly achieves a result one aspect of which we can program into an
ICBM. But once we have seen this, it still remains to describe how a
butterfly actually functions like or unlike a missile. The same goes for
plants and computers.
Finally, how can Peak possibly say that, if his team's findings are sound,
then stomatal patchiness "would no longer be a puzzle"? In fact,
virtually all the causal goings-on explaining the pattern of stomatal
openings and closings would still remain to be elucidated -- goings-on
that have little in common with the causal activity driving computers in
general or CAs in particular.
A science that prefers algorithm to substance is a science that cuts
itself off from causal understanding.
(Thanks to Kevin Kelly for alerting me to this research. The paper by
David Peak et al. is entitled "Evidence for Complex, Collective Dynamics
and Emergent Distributed Computation in Plants", in Proceedings of the
National Academy of Sciences (PNAS), vol. 101, no. 4, January 27,
2004, pp. 918-22.)
Related articles
----------------
"The Vanishing World-Machine" in NF #151:
http://www.netfuture.org/2003/Oct3003_151.html
"Intelligence and Its Artifacts" in NF #148:
http://www.netfuture.org/2003/Aug0503_148.html
"The Lure of Complexity", parts 1 and 2, in In Context #6 and #7:
http://natureinstitute.org/pub/ic/ic6/complexity.htm
http://natureinstitute.org/pub/ic/ic7/complexity.htm
SLT
Goto table of contents
==========================================================================
ABOUT THIS NEWSLETTER
Copyright 2004 by The Nature Institute. You may redistribute this
newsletter for noncommercial purposes. You may also redistribute
individual articles in their entirety, provided the NetFuture url and this
paragraph are attached.
NetFuture is supported by freely given reader contributions, and could not
survive without them. For details and special offers, see
http://netfuture.org/support.html .
Current and past issues of NetFuture are available on the Web:
http://netfuture.org/
To subscribe or unsubscribe to NetFuture:
http://netfuture.org/subscribe.html.
This issue of NetFuture: http://www.netfuture.org/2004/Feb1204_154.html.
Steve Talbott :: NetFuture #154 :: February 12, 2004
Goto table of contents
Goto NetFuture main page