January 23, 2004

Conversations with John Smart & Milan Cirkovic

Both John Smart and Milan Cirkovic provided me with assistance for my latest column, "Space: Not the Final Frontier." I corresponded with them via e-mail and I'd like to share those conversations.

Smart is a developmental systems theorist who studies science and technological culture with an emphasis on accelerating change, computational autonomy and the technological singularity. He is the founder and chairman of the Institute for Accelerating Change, a nonprofit community that seeks to help individuals better understand and manage continuous accelerating change.

Cirkovic is a cosmologist based in Serbia. After graduating from the Department of Physics at the University of Belgrade in June 1994, Cirkovic enrolled in the Astronomy graduate program, Department of Physics and Astronomy at the SUNY at Stony Brook, where he received his M.Sc. degree. He is currently working toward his PhD.

Both Smart and Cirkovic are transhumanists and both take great interest in seeking to understand the role, if any, that intelligence plays in the future of the Universe. Consequently, they share an interest in the anthropic principle, the singularity, post-singularity and extra terrestrial intelligence, the Fermi Paradox, and cosmological eschatology (the study of the life-cycle of the Universe, particularly its death). Smart, like a number of cosmologists these days (including Lee Smolin), study the Universe as it were a biological organism -- as something that came to be as a result of selectional processes. It's for this reason that Smart considers himself a cosmological embryologist. Similarly, Cirkovic believes that the Universe has a life-cycle, and that from here on in, this life-cycle cannot be predicted and understood without taking the potential role of intelligence into account; intelligence should be considered an integral part of the Universe.

Here are my conversations with them:

Conversation with John Smart

G. Dvorsky: Suppose we don't venture out into space. What about the "all the eggs in one basket" problem? What will post-singularity intelligences do about existential risks such as gamma ray bursts and collisions with foreign objects? Unless I'm missing a vital concept about MEST (the compression of matter, energy, space and time), the physical locality of post-singularity intelligences will still be here on Earth.

J. Smart: I think you are right about the need for redundancy of local planetary intelligence. In conversation, [Vernor] Vinge has himself suggested that post-singularity intelligences would build local secondary systems no matter how stable or immune Earth's A.I. grid seemed, just in case there were something we'd ignored in all our theorizing. But you don't need too many of them to achieve statistically near-inevitable immortality of the infocomplex, however. The problem might easily be solved with just a few off-Earth repositories of local knowledge (L5 points or the moon seem sufficient). Such systems could easily spring forth to maintain lineage immortality, as needed, if Earth were somehow extincted.

As for gammma ray bursts from nearby supernovas, I suspect technology can fashion far more powerful local shields than any type of radiation sent from far away. Inner space always seems more formidable than outer space in that regard. I'm not a physicist but I'd expect even just a bit of lead shielding around your repositories would presently protect them from even the nearest blasts, wouldn't it? Would you need repositories in geosynch orbit around large planets like Jupiter, so that some would always have planetary shielding, no matter the direction from which came the blast? I don't know, but I don't think so, and we are not even considering the femtotechnology shields that post-singularity intelligences will have access to.

Certainly we can also expect near Earth surveillance and redirection capacity for all foreign objects (meteors, comets, etc.) shortly after the tech singularity arrives. But I'd expect that we'll discover that all of the planet killers are already gone (due to the self-protective nature of solar development), and all the really big ones (Shoemaker Levy, etc.) probably only get sucked in by the really big gravitational attractors in our solar system (like Jupiter).

Again, as you and I know, space appears to be an informational desert, what I call a "rear view mirror" on the trajectory of universal intelligence. There is so much space within even our own solar system that there seems to be no realistic possibility that we'll send intelligence even to the edge of it. No matter, we can simulate it to amazing levels already, with even the primitive eyes and brains we've developed locally.

Conversation with Milan Cirkovic

G. Dvorsky: Are you familiar with the work of John Smart and his transcension scenario for post-singularity intelligence (migration of intelligence into vastly smaller compressions of matter, energy, space, and time)? If so, what do you think of Smart and his work? Do you see his hypothesis as a credible reconciliation for the Fermi Paradox?

M. Cirkovic: I'm afraid I have to admit that I have only a very superficial acquaintance with Smart's ideas (mostly from the transhumanism web sites), and I'm eager to read his upcoming book. However, from what I have learnt so far, I gather that it maybe a form of what I call partial solutions of Fermi's Paradox, for the simple reason: it is not sufficiently non-exclusive (or generic), as far as we know today. Among thousands (if not more) sites in the Galaxy where intelligent communities may arise, at least some of them will be unable/unwilling to go the Smart way (pun intended!). This is not something derogatory -- after all, 99% of the proposed solutions of FP suffer from the same disadvantage (and it is in most cases far worse than for Smart).

G. Dvorsky: When discussing the Fermi Paradox, why do we assume that Type III intelligences embark on interstellar colonization?

M. Cirkovic: There are at least two plausible answers. The first is that the entire definition of Kardashev's classification pertains to the spatial size, and since it is reasonable to assume that any civilization originates in a single (or a number of) discrete place(s), here it is. However, I guess that you aim at something more subtle, i.e. that the resources corresponding to the Galactic-level civilization could be in fact found in more compact environment, which would then obviate the material need for colonization. I'm skeptical, frankly, about this option, with one proviso below, mainly because we find most of the material resources of our universe extremely tenuously distributed, meaning that you need to control a lot of spatial volume in order to exploit its resources for creating value-structures. The proviso (and I gather that Smart counts at something similar) is that we don't have quantum theory of gravity which will judge once for all whether so called wormhole solutions to the quantum field theory are legitimate. If they are, several important ramifications occur, notably (a) the possibility of gathering resources from other places in our universe via non-local transport, and (b) the possibility of gathering resources from other universes through the wormhole throats. Again, although I'm an old SF fan, I don't see any meaningful way of saying anything sensible about these options prior to sufficiently well-developed quantum gravity, which is not something I expect very soon (i.e. within the current decade).

G. Dvorsky: Let's assume for a moment that we live in a non-typical Galaxy and that other Galaxies tend to be colonized rather early in their cosmological cycle. Would we be able to see evidence of this from our vantage point? Could we detect hyperscale engineering of some sort?

M. Cirkovic: Well, this is really a BIG question. I do tend to think that we would indeed be able to observe it, in such a exceptional case. However, one should keep in mind that even the closest big spiral galaxy is seen as it was 2 million years ago, as you notice below -- which is exactly the timescale invoked by Fermi and skeptics for the rise of Type III civilizations in the Milky Way. Thus, in galaxies farther away colonization and re-ordering of resources could be well under way without our noticing it so far. But then again, I don't see any sufficiently good reason for arguing for our exceptional position. Rather, I would bet (and that's something Robert Bradbury and I have extensively discussed and probably will write a short paper on it) that advanced civilizations will, in the process of becoming Type III (or equivalent), make itself rather invisible by moving into the outer regions of the Galaxy, mainly for the reason of heat (believe it or not!), i.e. impossibility to get their computational resources in contact with the temperature of the CMB. On the other hand, since you have read precious little which I wrote on the subject, you probably gather that I'm inclined toward a "catastrophic" viewpoint which argues that all civilizations in the Galaxy (and, per analogiam, in other galaxies) will turn to be of approximately the same age (i.e. ours), so I don't really expect such situations to occur. Of course, even a difference of million years or so may make a large difference concerning the ladder of ascent toward Type III, so the point is still very much debatable... (This is also under assumption that laws of classical computing will apply to advanced civilizations; but, see "Quantum Information Processing: The Case of Vanishing Interaction Energy" for an alternative option. If you are in contact with John Smart, you might wish to point this paper to him -- I guess he'll like its conclusion -- that under some special conditions we might one day create a "superconduction of information" analogue...)

G. Dvorsky: Assuming that we could detect hyperscale engineering, are we constrained by our perspective from our Hubble bubble (we are seeing these Galaxies as they existed millions upon millions of years ago)?

M. Cirkovic: Oh, of course, if we are going back billions of years, we certainly become limited, but mostly by the resolution of our observations. One should also keep in mind that we as of recently have numerical evidence that terrestrial planets (at least in our galaxy) cannot be older than about 9 Gyr. (see e.g. Lineweaver et al. paper in the New Year issue of Science)

I forwarded the previous conversation to Smart, who responded:

J. Smart: Thank you kindly for forwarding your thread. I've followed some of Milan's work from afar, and find him fascinating and very informative. I've added a blurb on his vanishing interaction energy paper to SingularityWatch. This is yet another promising example of MEST efficiency, the surprising emergent efficiencies of the microdomain. The empirically observed and still theoretically inadequately explained phenomenon of superconductivity is another, as you observed.

Cirkovic writes: Among thousands (if not more) sites in the Galaxy where intelligent communities may arise, at least some of them will be unable/unwilling to go Smart way (pun intended!).

This is an understandable response to the developmental singularity hypothesis. Few believe it would be fundamentally generic, as constraining as, for example the process of biological development once you are beginning with tuned DNA (or in this case, the tuned parametric DNA underlying this universe's physical-computational architecture). But it is. Developing organisms can either fail or create mature seed-propagating parents of very defined characteristics. You don't see multiple, non-generic solutions, ever.

Any attempt by any civilization to not transcend would leave it in computationally crippled slowspace. It would have to ignore constant accelerating innovations, innovations that just drop out of our playing with the physics of the microcosm, innovations that are far more intelligence discovered than intelligence created. I think not creating quantum computational devices, for example, would be like not discovering and using language, or mathematics, or electricity. Such systems seem as developmentally convergent and statistically determined as all the lower level features of the universe.

Your comments about wormholes are intriguing. If they were to be used in this universe that would violate every previous example of hierarchical substrate emergence in universal evolutionary development, which has lead to increasingly local interaction (even as we encode models that have increasingly nonlocal insights). Vastly greater MEST compression of compututation is the overwhelming developmental trend. Whether such nonlocality is possible or not, I would not expect it to influence the evolutionary development of spacetime within this universe. We can quantum entangle photons and instantaneously determine their states miles away. But beyond research into the historical processes that created us, I don't think we'll be doing much of that in coming years, within this universe. It takes finite time to traverse the space and becomes computationally uninteresting vs. more localized structure.

Inner space, not outer space, seems to be our developmental destiny, as I see it.

This is an ongoing conversation. Hopefully I'll have more soon, including Milan's response to John's last message.

1 comment:

Anonymous said...

Dear Mr. Smart,

Would the following analogy help explain the absence of supernovas from extremely massive stars.

Take a small baloon blow it up until it is destroyed by pressure from within. There will be a pop/supernova, now take a very large balloon and do the same, no pop occurs because the baloon has too much mass/gravity to explode.

Clyde L Harris
USA, Texas