A conversation with Mark S. Miller

by David Krieger

Part one: Creole physics and the credit theory of identity[{“box_2d”: [73, 142, 945, 272], “label”: “text”, “caption”: “Mark S. Miller is one of the system architects of the Xanadu project, the electronic hypertext system conceived by Ted Nelson as the future of publishing. Miller is also author, with K. Eric Drexler, of the Agorics papers (published in The Ecology of Computation, B.A. Huberman, ed.; New York: Elsevier–North Holland, 1988) which first presented the idea of agoric open systems—computer operating systems in which system resources such as memory and processing cycles are traded by programs and processes on an internal open market. I spoke with Mark in Palo Alto, California in September 1992, and began by asking him about what he calls “creole physics”: physics’:

In several different areas, what we find is that we seem to have, in the case of language, a particular grammar wired into our brains as kind of our initial grammar. When I say wired into our brains, it’s not necessarily the case that you can look at the genotype and see the grammar directly encoded in there, but in the sense of Dawkins’ extended phenotype$^{1}$, that a particular grammar seems to be the phenotype of our genotype. The reason we have confidence in this, is that there is this process that there are at least three clear instances of, in which what linguists call a creole grammar arose. This process is one in which the first generation of children who grow up speaking creole are in a linguistic environment which does not constrain their grammar and to a large degree doesn’t teach them grammar.

You may want to go into the definitions of a pidgin and a creole.

When a bunch of people speaking a bunch of different languages are somehow thrust together by weird historical circumstances—especially when they’re trying to engage in trade with each other, which is a reason for different language speakers to constantly be in contact with each other—the language that they end up speaking is termed by linguists a pidgin language. The characteristics of a pidgin language are that the vocabulary is a compromise between the vocabularies of the different languages that fed into the pidgin, but the grammar is not. To a significant extent, to a good first approximation, a pidgin language has no grammar. It really doesn’t do very much positional encoding, and you simply communicate by trying to put all of the content into the vocabulary words. However, if you listen to a given pidgin speaker, you can tell what his original language was, because he’ll be using the grammatical constructs of his native language, he just will no longer be encoding meaning in that.

Right.

The children of pidgin speakers are in the closest approximation to sort of the “ideal fantasy linguistic experiment.”

The “state of grace.”

Yeah, the “state of grace fantasy linguistic experiment.” The sort of thing that you would clearly love to be able to do as a linguist would be to somehow get a group of human infants who’ve never heard a human language and put them together on an island tended by robots or deafmutes or somebody—specifically, deafmutes who didn’t know sign language, a completely non-linguistic environment—and let them just, in interacting with each other, spontaneously start speaking some kind of language, and just see what the structure of that is. Also, you’d want to do that experiment repeatedly with groups of children over and over again so you get some statistical distribution. Now, there’s all sorts of reasons why, logistically and ethically, and developmentally probably, this is not a possible experiment.

However, the children of the pidgin speakers are raised in what is probably the closest to this experiment you can ever get to, which is: They’re raised in an environment in which their parents don’t consider any grammatical construct to be correct or incorrect, so they’re free to use whatever grammatical constructs they use, and it doesn’t get corrected. Every time that there’s been a clear case of this historical circumstance, the creole language that results—the language that the children of the pidgin-speakers speak—has the same grammar. And the most recent case, which is Hawaiian pidgin and Hawaiian creole$^{2}$—they did this research while not only the original Hawaiian creole speakers were still alive but while the original Hawaiian pidgin speakers were still alive.

They were able to determine that the grammatical constructs of the Hawaiian creole speakers were not only the same

constructs as in other historical creole circumstances, but they were grammatical constructs that were found in none of the languages that were inputs to Hawaiian pidgin. It couldn’t have been transmitted culturally, so you have definite non-transmission of a grammar that arises spontaneously every time this thing happens.

In addition, once they had this hypothesis in hand with this one verification, they said, “Okay, if this hypothesis is true, then what we should find when studying children growing up learning to speak English, is that where English and creole agree, the children should basically get the English constructs immediately, and where English and creole disagree, the children should start out speaking the creole construct, and have to have it corrected into the English construct. And they found exactly that.

The one particular case that I remember from the article—although the article has a lot of examples of this—is “double negative meaning emphasized negative.” That’s a creole construct. From a logical first-principle point of view, it’s clearly more complicated than a “double negative meaning a negated negative,” i.e.—

A positive.

A positive, or at least the absence of a negative. So double negatives are something that children end up starting to use and they have to get it corrected into not using double negatives. As we all know in English-language culture at least it’s only partially corrected.

In any case, that’s one example of having this native inborn large structure that’s an approximation of the culturally-evolved structures that we’re used to, and it’s the thing that starts the boot-strapping process off, where we come at life with this biologically-constructed—

25

EXTROPY #10 Winter/Spring 1993

Set of pre-conceptions, actually.

Yeah, and then we get it corrected into a more culturally-evolved and hopefully a more sophisticated one.

Other examples are “naive physics.” With naive physics, what they’ve done is they’ve interviewed people, given them little tests, and said, “Intuitively, what do you think would happen if you took a ball that was going around a circular wall, and when it got to the end of the circular wall, which direction would it drop?” The correct Newtonian answer should be obvious —

Tangent, basically. It continues going in a straight line.

Right. But an amazing number of people will think that it either continues circular or will spiral outward. With “naive physics,” what you find is that, even in physics students who have gone through first- and second-year Newtonian physics, naive physics is fairly unperturbed. If you ask the questions in the right way, where you still get at what their intuitions are, as opposed to making them feel like they’re on a physics test and keying into their training, what you find is, their intuitions have been largely untouched by the training, and they still have the same intuitions as people who have not learned physics, and the intuitions actually resemble Aristotelian physics to a large degree. The significance is, the Greeks not being experimentalists, Aristotle cataloged not physics, but human psychology about physics, and I think it’s interesting to think of that as a theory.

Popper is very clear on the analogy between biological evolution and theories, and in fact over here we have a very nice mixed case of a biological evolution of a theory of physics, which was “good enough”. Biology stops, basically, when it gets to a “good enough” solution, especially when it’s an adequate-enough solution that any creature is able to carry forward from there using cultural evolution.

To an extent, biologically-derived theories are falsified through species or individual selection, in that, if your model of physics is “wrong enough” to expect the ball to do something completely — not even in line with “naive physics,” for instance to move out of the plane of its motion, something that wrong — animals with that model will presumably be selected out, if it’s wrong enough to make a survival difference.

This is a complete aside, but I think it’s interesting, and I also think it’s original: A lot of people look at naive physics, and think just on the kind of stuff we’ve already said. “It’s good enough, and there

are things about it that are wrong, but that’s okay; culturally, we can correct.” I think that it’s actually better adapted as a physical theory in many ways, than Newtonian physics is, to the world that our ancestors were in. Here’s specifically the thing that I mean by that: Everybody assumes that naive physics is primarily a physics of inanimate objects. However, I think that a lot of the ways in which it’s wrong are actually by virtue of being more adapted to being a physics of animals.

There’s a certain animism in expecting the ball to continue on a curved path as if it has an intention.

Exactly. There’s also a certain animism in thinking that “It’s been constrained, and now it’s free, so now it’s going to overcompensate.” Either one of those are intuitions that are good at trying to predict the behavior of intentional creatures.

Popper is very clear on the analogy between biological evolution and theories, and in fact over here we have a very nice mixed case of a biological evolution of a theory of physics, which was “good enough.”

I was going to say, you’ve “adopted the intentional stance” toward your particle or projectile.

Exactly. Most of what we needed to interact with and predict successfully were animate objects, many of whom were trying to evade us, so it would be appropriate to devote more of our resources to trying to predict them than to predict these nice cooperative, or at least not hostile, inanimate objects.

So to a certain extent, Aristotelian physics is still with us, and will continue to be with us until we start monkeying around with our genes.

By the way, there’s one very interesting example of this. This is the thing that really made me add the animism to this. The way I came by this animism thing is after reading a fascinating paper about some computer animation work.$^{3}$ An interesting peculiarity which has long been known to animators is that, when you draw an animation of a bouncing ball,

there are certain things that you can do that are violations of Newtonian physics — rather gross violations — that actually are known to make the animation sequence look more realistic. If you do it realistically, it actually looks less realistic than if you introduce these distortions. Now, what the distortions are, is that a ball, before it hits, elongates in the direction of motion, and then it hits and it flattens out. Also, there’s another similar thing — let’s just deal with that one, that’s sufficient.

In this computer animation work, it was work in animating animals. In which what they did, well, they actually used a Luxo lamp as a “virtual animal”, a Luxo lamp being this nice simple structure that’s easy to animate, that you already have the database for —

I’ve seen the cartoon; the one that I saw was called “Luxo, Jr.”

I’m not sure if “Luxo, Jr.” actually used this technique; I think instead it’s because the database from “Luxo, Jr.” was already available, that they decided to do this work with the Luxo lamp.

What they did is, they wanted to make the Luxo lamp move around the way you would imagine an animal would move around. They did that by setting up an energy minimization problem. They sort of said, “Okay, the struts of the Luxo lamp are a skeleton, the joints are joints, and we will just, in our calculation, come up with a muscular model for bending the joint.” Specifically, a model of the relationship of force and torque versus energy expenditure. Then what they did is they said, they basically had the following problem: the Luxo lamp is stationary on the ground over here, there is some kind of barrier between it and the goal, and it needs to get from where it is to the goal, and end up stationary at the goal, not just at the goal but stationary at the goal, and it needs to do that with a minimum expense of muscular energy. Fortunately, problems like this are sufficiently monotonic that you can solve them with straight hill-climbing.

What they ended up with was a motion in which the Luxo lamp first crouched down, then extended in a leap, then contracted back together, and moved its base forward. After it was over the barrier, it elongated itself —

In preparation for landing.

In preparation for landing, and then did a soft landing and came to a stationary position. So, they observed in their paper, that this has reproduced in a principled way, the trick that animators always knew about, and what I think they didn’t go on

EXTROPY #10 Winter/Spring 1993

26

to explicitly say, although they may have, but I’ll certainly go on to explicitly say, is that even at those very low levels of our visual apparatus, we have this very intentional physics, in terms of our expectations and perceptions of motion, and that’s why motion of the inanimate object which is distorted towards animate motion, actually looks more realistic.

So, back from the digression: Kahneman, Slovic, and Tversky$^{4}$, and also a book called Human Inference by Nisbett and Ross$^{5}$, catalog a whole bunch of ways in which human statistical inference is very badly flawed, but distorted in systematic ways, and there are many many problems that result from the distortion, but, to some extent, society — we are all escaping from those problems by virtue of the modern understanding of statistics.

Another topic I wanted to ask you about was your views on the question of identity, especially as it applies to cases like uploading and duplication.

It seems to me that most everybody comes at the identity question with the presumption that there is some hard notion of identity that’s real, and that all we need to do is find a definition that’s good enough that corresponds to the notion of identity that we have this compelling sense we must be able to find. I think that the whole sense that there must be some real notion of identity that’s preserved over time and over change of state, and change of information, et cetera, and it’s really a discrete sense of identity, that you can still use to reason about a world with uploading, with backup copies, and with, especially, I think the most challenging one, having a communication medium between people such that the bandwidth between brains is the same as the bandwidth within a brain, in which case, the discreteness of the individual — I think that’s probably the one most challenging to the notion of discrete individuals.

The reason why we have this compelling sense that there must be a definition of identity is what I’ll call ‘creole epistemology.’ I think that so far historically we haven’t lived with uploading and with backup copies and with high-bandwidth communication between brains —

So our intuition hasn’t been selected for accurate portrayal of that sort of world, and what we intuitively guess about those is probably not accurate and probably not the most evolutionarily-stable strategy of epistemology. So this set me on the following search. I think it’s also the case that trying to dispense with the notion of identity altogether is not the right way out.

Though it’s very popular.

There’s all sorts of abstractions that we use to understand complex systems, and denying any reality to some kind of notion of identity out of a reductionist sense that identities aren’t real, is to fall into the reductionist fallacy, to use a metaphor of Dean[Tribble]‘s, of ‘denying that the window on my Macintosh is real because it’s only built out of pixels.’ So what I’m looking for is some notion of identity which has the right kind of fluidity to it and still the right kind of sense of coherence and continuity over time and over change, such that, by using that notion of identity, it can be an aid to intuition about post-Singularity life, not a hindrance to it. What I came up with is the notion that we already have the right kind of intuition about the identity of civilizations, the identity of cultures, the identity of natural languages.

So what I’m looking for is some notion of identity which has the right kind of fluidity to it and still the right kind of sense of coherence and continuity over time and over change, such that, by using that notion of identity, it can be an aid to intuition about post-Singularity life, not a hindrance to it.

Let’s concentrate on civilizations. One of the key insights that led me down this path was something that I got from my pediatrician, Dr. Einhorn. When I was a kid, I was reading some stuff about the Greeks, and I was really fascinated by the Greeks, and I was home sick and my pediatrician was seeing me, and I got into a conversation with him about Greek civilization; and he also thought the Greek civilization was real neat, and then I mentioned that, ‘It’s really a shame the Greek civilization died.’ And he said, ‘They haven’t died; we’re it.’ And at first I just had no idea what he meant by that, and then he explained that so much of what our civilization is and where it comes from, is based on the ideas of the Greek civilization, and so many of the values that we have, evolved out of the values of

the Greek civilization, that, in a lot of ways, it really makes sense to think of the Greek civilization, for some descriptive purposes, as being who we are.

The thing that’s very nice about what people understand about the identity of civilizations is that people are already used to the fact that civilizations aren’t really discrete, they blend and flow into each other, there’s multiple influences from multiple paths, and drawing a boundary in space-time around a bunch of people and cultural practices and saying ‘This is Civilization X’ is clearly something that we do as observers, and that we have the choice of several different boundaries. We can’t argue that any one set of boundaries is correct. So, in that sense, no set of boundaries is objective, but neither is it the case that all boundaries are equally good; it’s not the case that ‘Anything goes,’ and ‘It’s all just subjective’ in the sense of ‘Whatever the hell, anything goes, it’s all just in the eye of the beholder anyway’ and ‘Why should my boundaries be any less good than your boundaries?’

Another thing about this is that, when we really think about whether or not we should say that ‘The Greek civilization isn’t dead, we’re it,’ whether there’s that kind of continuity, largely we’re not thinking so much in terms of ‘pattern identity,’ to use the phrase from Engines of Creation$^{6}$, because our civilization currently is very, very different from Greek civilization, and we’re also not using the notion that’s very common of ‘continuity of consciousness,’ because people generally don’t speak of civilizations as being conscious, and besides that, the Greek civilization was suspended in cryonic suspension until its literature was rediscovered.

The question is, what’s the criterion we use when we say, ‘We are the successor to the Greek in such a way that this vague descriptive notion of identity is preserved.’ That was resolved for me by hearing about Fukuyama’s book The End of History and the Last Man$^{7}$? So, leveraging off the description that I’ve heard of that book, I realized that the key issue is credit — credit, and respect, and admiration — Which previous state do we feel good about giving the credit for having been our younger selves?

I think this lets us get past what would otherwise be the terrible paradox of seeking long life in a post-Singularity world, which is: Should you seek to not grow? If you’re coming at things from a pattern identity point of view, then to the degree that you expand and learn and grow, you also die — your previous self has died. So, from the point of view of

27

EXTROPY #10 Winter/Spring 1993

someone who constructs his goal structure by starting with ‘I choose life, I want to live,’ the pattern identity notion would lead you to seek to constrain your future choices to prevent your growth into something that is so far beyond your current self that you’re no longer like your current self.

I think that the goal, if you reconstruct the goal of wanting to live in terms of this notion of continuity of identity through credit, a particular kind of credit that’s really neither of the ones that Fukuyama mentioned, that’s the one that I know I most crave in my intellectual pursuits, is the respect of people I admire. The greater my admiration for those people, the greater my satisfaction in getting any respect from them. They might not even respect me as a peer, but if they respect me at all, the higher my estimate of them, the more that respect for me makes a difference to me.

What I seek to become is a creature that is vastly, vastly greater than my current self, something that is magnificently greater than my current self; and I translate my desire to live into my desire to have that future magnificent self look back on my current puny self with some credit—

And recognize the debt that it owes you. Right. Yeah, ‘I’m thankful to that guy for having had the seeds to become me.’ Just like I look back on my four-year-old self and realize, ‘Oh, yeah, that guy, I can just kind of give him credit for having had what it took to grow into my current self.’ Let me make a meta-point, by the way. Both myself and the Xanadu project as a whole are very, very careful about credit and references which is in some sense why we’re in fact doing this hypertext system where you can make all these links to source material —

Yes, that’s been noted several times before about the whole XanAMIX crowd, and one of the reasons that attracted me to come up here and join in, is the meticulous pinpointing of giving credit where credit is due, and the willingness to spread around credit among sources. That ties in closely with what you said about credit identity — To the extent that we’re citing our formal selves, our more primitive selves, for the germs of what we become.

About these boundaries being between subjective and objective — Don Lavoie referred me to a book$^{8}$, the basic point of which is that philosophy has been stuck in this false dichotomy: that things are either objectively true, or they’re in an anything-goes state. In fact, most everything is in this third state.

Don comes at it from a hermeneutical

point of view, and you can think of these issues as very literary ones, where certain choices of descriptive boundaries ‘make for a better story’ about history than others. You can’t objectively say that one novel is better than another, but we all know that novels are not all equal; that novels really are better than each other, but there’s no objective way to determine in a disagreement which is which. In some sense it comes down to a matter of taste, but in some sense it doesn’t really, because there really is a content to what we’re arguing here.

A lot of matters of taste rely on being consistent with a set of cultural assumptions, in that, although there’s no direct reason for saying that one novel is better than another, there are cultural qualities, based on which, valuing what we consider to be the inferior novel would be inconsistent with certain other assumptions in our culture.

But I believe that even cross-culturally, and even, hypothetically, across species, you’ll find that there’s enough agreement about some of the extreme cases, especially the extremely bad cases, that there’s clearly something going on here that’s not just arbitrary and anything goes, but it’s also not objectifiable. I don’t come at this from a hermeneutical point of view, but from an evolutionary-epistemological point of view. That’s simply because I have not gotten into the hermeneutical literature yet. I believe, from what I understand of hermeneutics through Don Lavoie, that there’s actually a lot more agreement between the two philosophies than the practitioners of either are inclined to recognize or acknowledge, which is unfortunate.

From this evolutionary-epistemological point of view, all of our knowledge arises out of a process of variation and selection. You can think of the variation component of the evolution of knowledge as being the subjective component, and the selection component (where the selection is by external criteria — there’s also subjective selection, but we’ll leave that aside), the selection is by virtue of getting mugged by reality. It’s through the selection that we get the objective component, but therefore all of our actual knowledge embodies a mixture of the two where it’s impossible to separate the mixture. We can understand the process, but we can’t tell, for an individual piece of knowledge, how much of its content came from the variation process and how much came from the selection process.

*You said a little bit earlier about continuity of consciousness as a criterion for identity, and you said that Greek civilization was

‘cryonically suspended’ in the form of literature until it was revived. Does consciousness exist? Is consciousness necessary as an assumption, as a hypothesis, or can your personal epistemology get along fine without consciousness?*

I think that, similar to identity, there’s a lot of strong intuitions we have about consciousness that are a result of creole epistemology. As well as a result of things that culturally evolved between Freud and computers. There was a lot of evolution of the notion of what consciousness is between those two events, and essentially our folk psychology — which is a wonderful phrase of Dennett’s — is still largely a result of that pre-computer evolution.

What I seek to become is a creature that is vastly, vastly greater than my current self, something that is magnificently greater than my current self; and I desire to translate my desire to live into my desire to have that magnificent future self look back on my current puny self with some credit.

There’s many parts of what we mean by consciousness. The idea of there being a ‘real you’ at a point inside you that experiences things in a serial order and knows what it believes — understand, I have not read the book — but from what I understand, Dennett does a very good job of trashing that, and I think it well deserves to be trashed: homunculus theory?

That’s what Dennett calls it, the homunculus, the ‘little man in the control room’ — By the way, let me strongly recommend to everyone the segment of Woody Allen’s movie ‘Everything You Always Wanted To Know About Sex, But Were Afraid To Ask’ with Tony Randall in the control room of the guy who’s having sex, and Woody Allen as a sperm who’s agonizing about ‘What if he’s not even having sex? Maybe I’ll end up on the ceiling!’ And the thing that’s wonderful about that is it’s a very strange and wonderful mixture of homunculus theories, and society-of-mind theories; it’s kind of got a lot of

EXTROPY #10 Winter/Spring 1993

28

mixture of both. It’s actually very sophisticated. I really recommend going back and watching that after having read Society of Mind.$^{9}$

Okay.

So, our notion of consciousness that has much truth to it, is the notion of consciousness being reflective—that it really is quite a trick to be able to have been doing something without articulating to yourself what it was you were doing, and then to ‘hop back’ and say, ‘Oh, so I was doing that’ and then you’re articulating, you now have a thought you can think about. And this whole ability, not just to introspect, but one of the things we notice about articulate knowledge versus inarticulate knowledge is that inarticulate knowledge is learned gradually with training. Articulate knowledge, through, I’ll say, through the mechanism of consciousness—and what I mean there is purposefully vague—is able to learn many things on one hearing, because it’s manipulating a very different kind of learning process. A lot of the trick of the brain is the gluing together of those two very different kinds of learning, and, in particular, the way in which they each bootstrap the other.

Another aspect of consciousness that is valid is that part of the trick of consciousness is that your articulate construction of the world has as part of its articulate construction a symbol in there if you will—a designator, a node—that represents this creature itself. So this notion of consciousness is intimately tied in with the notion that we have of identity, which is: our creole notion of identity is the thing that’s there in that slot—

Experiencing the consciousness.

In the conscious articulation of how the world works and what’s going on there, there’s this idea structure of the identity of this personal self that’s a creole self—a creole identity of the self as well as a creole identity for consciousness—but that’s the thing that the rest of the conscious structure uses in thinking about its place in reality.

So I think that all of those are valid pieces and part of the pattern, but I think that not only is a lot of our other stuff about consciousness, like the homunculus stuff, wrong; but a lot of the stuff that’s right is much less important than people verbally think it is—that, in some sense, we’re much less conscious than we think we are, and the reason is that the thoughts that we’re aware of, are the ones that we’re aware of.

*Can you give an example of an instance in which people are less conscious than they

think they are?*

Okay. People generally think that they have a lot of introspective access to how they think. However, for relatively simple computational tasks—the kinds of things that people are conscious of what they do, of how they do it when they do it—are generally much simpler computational tasks than the things that people completely take for granted and do ‘without thinking.’ The fact that so many of those are of such incredibly greater sophistication than the things that we do ‘with thinking’ means that the fact that we think that it’s in our ‘thinking’ process that’s where the sophistication lies, is probably wrong.

One particular example that I really love is ‘Hansel and Gretel.’ In the book Distributed Artificial Intelligence, edited by Bond and Gasser$^{10}$, one of the papers has this diagramming notation which they’re exploring so they can diagram not only ‘Actor X believes A,’ but they can diagram what people believe about other people’s beliefs. ‘At this point in the story, this guy believed that that guy believed that this person was going to do that, whereas this person himself believed that this guy believed that he himself was going to do something else.’

Right.

So they’ve got a good diagramming notation for laying out these kinds of nested belief structures. They explain this by diagramming the beginning of the Hansel and Gretel story—the part of the story that is before it gets interesting—the part that you just read to a kid and assume that he gets it ‘without really thinking about it.’ It’s not considered to be a tough comprehension for very small children, but you end up with this massive structure to explain it. I think I’m good enough at evaluating notations and formalisms to look at it and say, the massive structure is not the result of a bad formalism. The formalism is good; the formalism does not create any combinatorial explosions here. The massive structure is the result of the fact that human beings are wired for this incredible ability to do nested modeling of belief structures—very nested modeling, three and four levels deep without straining.

It could also be considered empathic in the sense of putting yourself into someone else’s position and modeling their beliefs by adopting their viewpoint.

I think that one of the reasons that people always seem conscious to us, and no AI system has seemed conscious to us yet, is that when I form a model of you ‘without thinking about it,’ part of my model is my

model of your model of me, and my model of your model of my model of you. It’s certainly the case that no AI system has enough knowledge of our social conventions and of other human beings that, when we’re interacting with an AI system, the models that we construct that have explanatory power to us about how it’s going to behave, don’t include that kind of deep nesting because it doesn’t have enough ability to accurately understand my model of it, or its modeling of my model of it, to have influence in what my model of it is. There’s a lack of the kind of deep nesting that we normally do ‘without thinking about it’ with human beings, and I think it’s what that kind of deep nesting provokes, that is what we have attached the label consciousness to. That’s why, no matter how reflective a computer system is, it never seems conscious to us.

In Part Two of this interview, Mark and I discuss the five kinds of libertarianism, the Reverse Polish Moon Treaty, ‘nanarchy,’ and the role of gossip in a free society.

REFERENCES

$^{1}$Dawkins, Richard. The Extended Phenotype. Oxford: Freeman, 1982.

$^{2}$‘Creole Languages’ by Derek Bickerton, Scientific American, July 1983.

$^{3}$‘Spacetime Constraints’ by Andrew Witkin and Michael Kass, ACM SIGGRAPH, Vol. 22, No. 4; August, 1988.

$^{4}$Daniel Kahneman, Paul Slovic, and Amos Tversky, eds. Judgment Under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press, 1982.

$^{5}$Nisbett, Richard E., and Lee Ross. Human Inference: Strategies and Shortcomings of Social Judgment. Englewood Cliffs, N.J.: Prentice-Hall, 1980.

$^{6}$Drexler, K. Eric. Engines of Creation. Garden City, N.Y.: Anchor Press/Doubleday, 1986.

$^{7}$Fukuyama, Francis. The End of History and the Last Man. New York: Free Press, 1992.

$^{8}$Bernstein, Richard J. Beyond Objectivism and Relativism: Science, Hermeneutics, and Praxis. Philadelphia: University of Philadelphia Press, 1983.

$^{9}$Minsky, Marvin. The Society of Mind. New York: Simon and Schuster, 1986.

$^{10}$Alan H. Bond and Les Gasser, eds. Readings in Distributed Artificial Intelligence. San Mateo, Calif.: M. Kaufmann, 1988.

29

EXTROPY #10 Winter/Spring 1993

Now! Freely Available To Americans…

EUROPEAN SMART DRUGS!

You don’t have to wait for the FDA to “approve” of new memory and intelligence enhancing therapies. For your personal use, you can obtain therapies that can:

• Enhance mental function four times better than Hydergine!
• Increase mental energy, concentration, and alertness!
• Maximize the ability to memorize facts, figures, and scientific findings
• Increase learning, mental agility, and I.Q.
• Improve test scores leading to higher grades and work productivity!
• Prevent and treat degenerative brain diseases including Alzheimer’s Disease, Parkinson’s Disease, etc.

The Life Extension Foundation has compiled the sources, the doctors and the information to enable Americans to obtain life extension therapies from around the world. These drugs have been safely used in France, Germany and other countries for years, but until now, have been denied to Americans.

Europeans are using these drugs to improve their mental condition to unprecedented levels. Now Americans can gain access to advanced life extension therapies such as Piracetam, Centrophenoxine, human growth hormone, and a new memory-enhancing product that is four times more effective than Hydergine.

As a member of The Life Extension Foundation, you will learn about the scientific basis for using these life-enhancing therapies, their side effects, and how you can get them!

For just $50.00, you will receive Life Extension Report and Life Extension Update each month plus:

1. The Directory Of Life Extension Doctors. A nationwide directory of doctors who are knowledgeable about these advanced therapies and may be willing to prescribe them for you.
2. The Physician’s Guide To Life Extension Drugs. The first book ever published to provide American doctors with information about safe and effective “unapproved drugs”. This book is referenced to enable the lay person to understand and find therapies for specific purposes.
3. The Directory Of Innovative Medical Clinics. If you were told you had an incurable disease would you believe your doctor? A disease your doctor says is “untreatable” may already have a cure that the FDA has not yet “approved” of. There are scientists with impeccable credentials who are effectively treating so-called “terminal” victims of cancer, Alzheimer’s Disease, etc. You can now access these advanced research centers with The Directory Of Innovative Medical Clinics.
4. Discounts of 25% to 50% on your vitamin purchases. Members buy name brand nutrient supplements and advanced life extension formulas at super discount prices.
5. Discounts of 20% on all your prescription drug purchases including popular life extension drugs such as Hydergine and Eldepryl. THE MAIL-ORDER PHARMACY saves members hundreds of dollars a year on their prescription drug purchases.

The Life Extension Foundation is the only organization in the world that tells you how to obtain the most advanced life extension therapies in the world…long before they are “approved” by the FDA. You will be the first to find out about products that will enhance your life.

Mail to:

Life Extension Foundation

P.O. Box 229120

Hollywood, Florida 33022

To join, use the coupon or call: 1-800-841-5433

Enclosed is my $50.00 membership. Please enroll me in your life extension program which includes two newsletters each month and the three directories of life extension doctors, drugs, and clinics and the super discounts on my vitamin and prescription drug purchases.

Name

Address

City

Credit Card #

State

Zip

Exp Date

EXTROPY #10 Winter/Spring 1993

30