4 A. M. Friday still feels like Thursday. Five hours remain until the contest. Bean dip
slowly dries around the rim of a jar, turning a darker, almost translucent brown. This
corner of the table, the one nearest the center of the room, is littered with the strange
and particular combination of plastic, paper, metal, glass, and organic debris that
typifies the remains of junk food. The room, a large but nondescript meeting room
with beige-painted cinder block walls, is bathed in fluorescent light, 60- cycle
radiation painting the few remaining occupants a lovely whitish green.
A few of them still hunch over keyboards, pecking feverishly, squeezing the last few
desperate instructions into their robots. Others sprawl on the floor around the test
course, watching carefully and hopefully as their fragile creations, their little Lego
and wire and motor golems, their tiny mind children, haltingly -- but autonomously
-- negotiate their way toward the goal. The expressions on their faces are variously
rapt, worried, and proud.
The scene is the early morning of the last day of Artificial Life III, a week-long
scientific hoe-down that took place last June in Santa Fe. The hardy hackers in the
cluttered room at the back of the building are entrants in a robot-building contest that
will be run as part of the "Artificial 4H Show" beginning at 9A. M. Their robot
creatures run the gamut from the eminently practical to the practically insane.
The insane ones, of course, are by far the more interesting. One, appropriately named
Rob Quixote, has only a single wheel, and therefore must steer by rotating an oversized
horizontal windmill-like contraption fastened to its head, effectively pushing against
the air to turn itself. Another moves by a sort of spastic lurching; throwing its entire
front section forward, it gains an awkward quarter inch, then gathers up its
hindquarters for another throw. This one is so inefficient that it requires twice the
usual number of batteries, and uses them up in a single run. Amazingly enough,
though, it successfully traverses the course, albeit slowly and with much ineffectual
thrashing.
"Artificial life," as a named discipline, appeared on the scientific scene relatively
recently. The first conference happened in the fall of 1987, and gave joyous birth to
this new field of scientific inquiry, or rather this new and rich confluence of many
different fields. Scientists who had been working in isolation suddenly discovered
others pursuing similar lines of investigation, and the meeting of minds was electric.
Artificial life is an attempt to create and study artificial systems -- that is, systems
created by humans -- that mimic processes or exhibit behaviors usually associated
only with living systems. Predictably, the primary medium that these systems are
created on (in?) is computers; this is a field that depends heavily on technology to get
its work done (they're doomed if electricity ever becomes unavailable). Also
predictably, a large proportion of its devotees are biologists, especially theoretical
biologists.
Why would biologists want to study artificial life? Don't they already have their hands
full trying to figure out the real thing? Well, for one thing, there are a lot of
experiments biologists would love to do that they simply can't: nature doesn't come
with convenient levers and knobs, and you can never roll back time and try something
over again. So if biologists can develop good models of biological phenomena, they can
implement them on computers and run clean and tidy experiments that are repeatable,
detailed, controlled, and manipulable down to the last detail. This is a far cry from the
messy, inexact, unrepeatable real world, and for some biologists would be tantamount
to scientific nirvana.
But there's another, larger reason for biologists to study artificial life. In the words of
Chris Langton, self-described "midwife" of artificial life (he organized the first
conferences and named the field), "Such systems can help us expand our understanding
of life as itcould be. By allowing us to view the life that has evolved here on Earth in
the larger context ofpossible life, we may begin to derive a trulygeneral theoretical
biology capable of making universal statements about life wherever it may be found
and whatever it may be made of."
I like it.
When I read this I was hooked. Visions of bizarre, unknowable alien intelligences and
strange, seething soups that cling and quiver and creep around filled my head. And here
are real scientists hanging around seriously discussing it! This is some serious fun!
And lots of different kinds of scientists are paying attention; biologists,
mathematicians, physicists, chemists, robotocists, and computerists are all well
represented at the conferences, with a sprinkling of philosophers, anthropologists,
economists, and others. The gee-whiz factor hooked me, but the interdisciplinary
thrust of artificial life reeled me in.
(In conversation people say "a -life." I've seen it written as Alife, A-life, alife, and
a-life. I wanted to use alife, but people tended to pronounce it like "get a life," so I'll
use a-life instead.)
Another appeal for me is the tacit approval of the "build it first, then study it"
approach in a-life. This method of building things and learning things (stumbling
around, really, butintelligent stumbling,directed stumbling) has always been my
particular forte. The premise is that we don't need to completely understand something
before we can build it or build a model of it, and that it's very often more instructive
to get a crude version up and working immediately than to try to refine the thing
completely before trying it out. By fumbling around and building things blindly, we
can often learn a lot by virtue of the happy accidents that inevitably occur. And it'stons
more fun that way.
There were far too many interesting things at the conference to describe them all here.
Instead I want to tell you about one particular talk that caused me to have a powerful
"Aha!" experience (and Ilive for "Aha!" experiences). If you know something about
evolution already, the following may not be news to you, but presumably most
computer programmers don't study biology.
The talk dealt with Lamarckian evolution. Lamarck was a contemporary of Darwin who
postulated that the things experienced by an organism during its lifetime could affect
the traits handed down to the next generation. As an example, a Lamarckian might
believe that proto-giraffes had to stretch their necks up to reach the leaves at the tops
of the trees, and because of all the stretching, their descendants were born with longer
necks. Unfortunately for Lamarck and his followers, this is rubbish.
It turns out that as far as biological evolution is concerned, Lamarckism is
nonexistent: there was no such thing at work in the development of life on Earth. So my
curiosity was piqued when I saw the title of this talk by David Ackley and Michael
Littman: "A Case for Distributed Lamarckian Evolution." What, were they crazy?
Talking Lamarck to all these modern scientists? (At the previous conference, Ackley
had one of the few really amusing presentations, so of course I would have gone no
matter what the topic, but this one looked particularly juicy.)
Ackley and Littman weren't trying to convince people that Lamarckian evolution had
anything to do with life on earth. What they did instead was compare the efficiencies of
the two types of evolution. (They created a simple evolution simulation, and then
compared Darwinian and Lamarckian evolution in their abilities to find a solution to a
particular problem.) Hey, this is after allartificial life, so if Lamarckian evolution
works better, we can use it, right?
What they found was that when Lamarckian evolution was allowed to enter the picture
-- when the things learned in one generation were at least partially passed on to the
next -- the system was much, much better at solving the given problem. It
consistently found better solutions faster in every single case they tried. This of
course makes some intuitive sense. Rather than waiting for genetic shuffling to find a
solution to the problem, the prior generation can point the current one in the right
direction. So Lamarckian evolution is pretty much a great thing, evolutionarily
speaking, because it gets you a lot further and it gets you there a lot faster. (Where it
is exactly that you're going is a question for the philosophers; for the moment, let's
just blithely assume that we reallydo want to getthere.) Their point was that as
simulation builders we should think about using Lamarckian inheritance in our
simulations, because it works so well. But this point reinforced something else that
had been rolling around in my head.
There's an evolutionary premise that I initially learned about through reading an
article by a robotocist named Hans Moravec in the first Artificial Life proceedings. I
learned more about it in Richard Dawkins's bookThe Blind Watchmaker and in a
fascinating book calledSeven Clues to the Origin of Life by a Glasgow chemist named
Graham Cairns-Smith. This particular concept is called "genetic takeover."
According to this idea, one substance can gradually replace another as the carrier of
genetic information. Cairns-Smith postulates that life began with replicating
inorganic crystals -- clays, as a matter of fact -- and that a genetic takeover
gradually occurred, with proteins and nucleic acids gaining in dominance until finally
the original materials were no longer needed. Dawkins and Moravec (and many others)
think that a genetic takeover is occurring now, with human culture taking over from
nucleic acids as the evolving entity, though they differ in their candidates for the new
"gene-equivalent."
Dawkins likes to speak about the "meme," a very useful term first coined in his
bookThe Selfish Gene . A meme is an idea, really, or a piece of information. It is
immaterial, and requires a material substrate of brains, books, computers, or other
media to exist. But given that substrate, the parallels with genes are very good. Just
like genes, memes replicate (we tell each other good ideas, or write them down for
others), memes mutate (we don't always get it right in the telling), memes mate
(ideas in combination often give birth to new ones), and memes compete for survival
("good" ideas stick around a long, long time, but "bad" ones die by not being passed on
to anyone: mindshare is their means of existence).
Moravec, on the other hand, seems to be more interested in the evolution of machines,
and speculates convincingly and entertainingly that our machines, our artifacts, will
eventually become the dominant evolving entities on Earth. Science fiction, or science
fact? I don't know -- there are compelling arguments both ways -- but in either case
it makes for very good reading.
In any case, they think that perhaps here on Earth biological evolution is thoroughly
obsolete, and almost despite myself I have to agree. Sure, it's still operating, but the
evolution of human bodies has been completely outstripped by the evolution of human
culture. Bodies evolve at an extremely slow pace, but culture evolves incredibly fast,
and humans are having such a profound impact on the Earth that biology simply can't
keep up. Look at the changes on Earth in the last millennium. Most of the species alive
a thousand years ago have remained physically about the same, yet there's no question
that the Earth has undergone a radical transformation, and primarily at the hands of
humans, as a by- product of their culture. (You might hesitate to call the rampant,
wanton destruction and boundless consumption of resources that Earth has suffered at
the hands of humans "evolution," but remember that the word "evolution" doesnot
necessarily imply improvement.) But why is it going so fast? How come humans do
this and other species don't?
One of the primary distinctions between human beings and their close animal relatives
is language. Humans can communicate with abstract symbols, and their
communications can be "fossilized" in time (that is, written down for later). Here
comes the "Aha!" we've all been waiting for: this ability allows humans to engage in a
form of Lamarckian evolution! The things we learn in our lifetimescan be passed on to
the next generation, though in a filtered sort of way. We can't change the way our
offspring are built, but wecan change their behavior (teenagers notwithstanding).
Other species do this to some extent, but humans are the unquestioned champs at
shaping their offspring.
As you can see, a-life -- just like life itself -- is rife with philosophical conundrums
and radical, thought-provoking concepts, and that's much of the reason I stay
interested. But probably the biggest reason of all that I like a-life is hard to express,
except by analogy: I get the same feeling peering through a glass screen into a
computer world full of digital critters that I do peering through the bars of a cage at
the zoo. The xenophile in me wants to see all the forms that life can take, and get to
know the minds of every other being. I want to puzzle out the motivations behind a
critter's behavior, and I love that shock of recognition I experience every time I look
into an animal's eyes -- even the ones that are so alien, like birds and reptiles and
fish. Again, it's this feeling that there are universal properties of life waiting to be
discovered, properties that apply not only to life as it has evolved on Earth but to
allpossible life, including the digital variety.
Are any of these a-life explorations really alive? That's an energetic and ongoing
debate among a- lifers, of course, and the answer ultimately depends on the definition
you pick for the word "life." Rather than arguing whether metabolism is more
necessary to life than reproduction, though, I like to duck the definition issue. I don't
really care too much whether wecall them alive, I want to see if people react to themas
if they're alive. I want to see that shock of recognition occur when people and digital
organisms collide. (What if "they" recognize "us"?!) It's sort of the Turing Test
approach for life: if it seems alive -- if people can't tell that it'snot alive -- then no
matter what we call it, people will treat it as if it's alive. That I'd like to see.
RECOMMENDED READING
DAVE JOHNSON's mother recently moved across the country, and sent him a total of
eight large cardboard boxes crammed with junk spanning his entire life that she didn't
want cluttering her garage any more. Among his old school stuff was a report card
from second grade that included a couple of N's, meaning "needs improvement." The N's
were in the categories of "Is Prompt" and "Works Steadily." Here's a quote from his
teacher, Mrs. Doris Short, that accompanied the report: "We've talked about being
prompt, but it's always 'I'll finish tomorrow.'" This is strong evidence for the claim
that personality is established early in life, and never changes. *
Dave welcomes feedback on his musings. He can be reached at JOHNSON.DK on
AppleLink, dkj@apple.com on the Internet, or 75300,715 on CompuServe.*