Joi Ito's Web

Joi Ito's conversation with the living web.

Susan Crawford quotes an essay by John W. Patterson called "Thermodynamics and Evolution", part of a volume of scientific responses to creationism. She ties it neatly to Internet governance at the end.

Susan Crawford
Here is Patterson's conclusion:

"In reality, ... the 'uphill' processes associated with life not only are compatible with entropy and the second law, but actually depend on them for the energy fluxes off of which they feed. Numerous other kinds of backward processes in simpler, nonliving systems also proceed in this way, and do so in complete accord with the second law."

This all ties to internet governance. A sufficiently open net will tend towards order, not chaos -- and will do so on its own, with no external pilot.

Just reading the conclusion, you might think she's making a techno-utopian quantum leap, but the idea of open systems allowing evolution and order and seeming to defy entropy is any interesting one. Order can emerge in a system with increasing chaos around it if the system is open. I don't think being merely open guarantees that it will tend towards order. On the other hand, closed systems will tend to become disordered and the best way to maintain order in such a system is to move very slowly...


Well, Stephen Hawking's conclusion within that framework is:

"My only fear is this. The terror that stalks my mind is that we have arrived on the scene because of evolution. Because of naturalistic selection, and natural selection assumes natural rejection, which means we have arrived here because of our aggression. And my hope is that somehow we can keep from eating each other up for another 100 years. At that point science would have devised a scheme to take all of us into different planets of the universe and no one atrocity would destroy all of us at the same time."

Ravi Zacharias, a Christian apologist, commented on Hawking’s lecture, "Hawking was unavoidably caught on the horns of a dilemma. On the one hand, if there is no God he could feel the hold of determinism from which evolutionary theory could not escape—out of flux, nothing but flux. What followed from that deduction was even more troubling. For on the other hand, if evolution held true, he could not further ignore the aggression and violence through which man has evolved. Therefore, Hawking offered mankind’s only hope—that the savior of technology would come riding on the wings of science to rescue us from the clasping teeth of determinism."

Not so sure if physics and laws of the universe can be applied to open systems and Internet governance. Sidenote, Crawford should broaden her reading to people like Einstein, Freeman Dyson, and John Polkinghorne.

Albert Einstein, while not believing in a personal god, concluded through his years of research that the universe could not have been created by chance or from chaos:

"I want to know how God created this world, I am not interested in this or that phenomenon, in the spectrum of this or that element. I want to know His thoughts, the rest are details."

Freeman Dyson, father of Esther Dyson , is "one of the outstanding physicists of our time" and has "written extensively on the meaning of science and its relation to other disciplines, especially religion and ethics." While I don't know the details of his faith, he stands on the side of God's involvement in creation process.

There is another world-class physicist, John Polkinghorne, who became an Anglican priest, who is a strong Creationist. South Korea's leading materials scientist, Younggil Kim, who invented alloys Motorola uses in their chips, is also the country's most outspoken Creationist.

The idea that our universe was created by some random event and life came from a chaotic mass of gases are unthinkable for these people. Even in numerous experiments attempting to see if some complex structures could be form from a ball of gas only results in simple sugars.

Gotta stop. It's getting long. Sorry.

The evolution of open systems, and the emergence of form/order out of chaos is a major theme in the work of the Nobel prize winning chemist Ilya Prigogine, and the entire, as of now not fertile as expected, field of chaos theory stemming from Rene Thom's book "Structural Stability and Morphogenesis". Karl Popper extends the notions of the value of open systems in generating order in his "Open Society". All these works underline how broad and fundamental these notions are.

The 2nd Law is a bad argument against evolution, since processes that existsed before "the fall" such as breathing, chewing, etc. require it. We have to go back to the "no new information added by mutation" argument that Bernard hints at.

Hayek's writings on economics are analagous, a revelation I had while reading this post.

Well, surely it depends on some sort of organizer emerging? Something that will pull some part of the chaotic soup up into a more ordered state. This is what happens when a potter turns clay into vessels. The availability of vessels makes mankind more efficient and so leads to the development of extra vessel-makers. These in turn increase the amount of order in the world and the effect becomes self-perpetuating. But somehow there has to be a controller or a builder. Pot-making (or for that matter IP number allocation) won't happen without someone actually pushing forward and doing it.

In 'capitalism' the idea is that order emerges through concentration of capital. (For example, one family builds up enough spare resources from hunting, farming and trading activities to be able to put the time in to figure out where to get the clay and make the pots). In communism, order emerges through workers' revolutions. So workers come together at the local soviet and somehow decide to pool their spare resources towards developing this new technology. (All broadly speaking, of course.)

The point, though, is that from an internal perspective, there still has to be decisionmaking and somebody or something in charge, even though the whole system may look 'emergent' or 'self-governing' from an external perspective.

It's all a matter of the subjective or objective perspective. Objectively, there must be some sort of emergent way for creating order. Subjectively, it just feels like there's a boss, possibly an oppressive one, controlling things.

Antoin. I don't think that's true in all cases. I think there are numerous biological systems where we thought there was a leader or a "pacer" cell in heart cells or in slime mold and we find emergent order without any single point of control or leadership. It's the way the units are programmed that allow the order to emerge.

Well, then the order does emerge from somewhere. It's the genome imposing its order on the soup, in the same way the potter imposes order on the clay.

It's a matter of how you look at it. The important lesson is that you can't just wait for emergence to happen; there has to be a degree of management.

This is fuzzy thinking based on mistaken notions of thermodynamics. The old view is that closed systems tend towards entropy, unless you have some sort of negentropic mechanism like Maxwell's Demon. But today, there are well known examples of closed systems that naturally tend to negentropy and order. A classical example: seal a can full of multicolored M&Ms, the more you shake, the more randomly each color is distributed throughout the can; entropy is increased. The modern negentropic example: shake a can of mixed nuts, the largest ones will float to the top; entropy is decreased. This effect is known to physicists as "the Brazil Nut Effect." I've seen it demonstrated with smaller particles, Tang and instant coffee granules, shake it and they separate into layers.

Openness (ie interaction)is not the only condition for the emergence of order. There also is a need for sub-level or meta-level rules that determine the behavior of individual elements whose interaction produces the emergent order.

The parallel with Internet Governance is a very interesting one. The Internet and the web have been able to evolve as an emergent system through the aggregation and interoperation of a multitude of independent actors sharing the same interaction codes (TCP/IP, HTML, and other protocols) much like living organisms have their genetic code. TCP/IP and HTML are somehow the DNA of the present infrastructure.

People mostly understand the way genetics work as a linear blueprint that pilots a sort of factory plant. This is a wrong approach : living entities are dynamic systems. The body shapes produced through morphogenesis (baby to boy to man) are fundamentally a set of attractors for a dynamic biological system. And this set of attractors is inherent to the corresponding DNA but only visible when you let the process unfold.

Protocols for the network play the same role : a very limited set of rules produce a resulting pattern and architecture. It is hard from TCP/IP and HTML to guess in advance that the topology of the web will be the one described by Barabasi (see his papers at University of Notre-Dame : But it is a natural attractor.

Internet governance is not about doing nothing and establishing no rules hoping order will appear by magic. It is about setting the simplest set of underlying common rules that allow interaction between actors to produce viable and/or desirable patterns. It was done at the technical level (how the web works) in a very elegant way. it should now be done on other levels to address how the web is used.

This is why it is so different from the traditional mechanisms of Government. Government is fundamentally about checks and balances, a linear, mechanistic view of systems well adapted to the industrial era. Our complex and networked world needs order-producing mechanisms more akin to biology.

Those interested in Internet Governance should follow the activities of the Working Group on internet Governance (see WGIG at : set up surprisingly in the framework of the World Summit on the Information Society.

Sorry this is too long.

Shaking a can of particles affected by gravity proves NOTHING but that gravity tends to pull the finer/denser particles into lower strata SEEMING to defy entropy.

Shake that same can in space without the effects gravity and you prove NOTHING. Heavier/denser particles cannot squeeze their way to the bottom because there is no bottom (gravity CAUSES a bottom to exist).

Random processes (the supposed unintelligent creators of all things) are like the shaking can in space, except one serious flaw in logic whether shaking the can with gravity or without: ALL the particles in the can, PLUS the can, are/ were made by INTELLIGENT beings, the "experiment(s)" is/are conducted by INTELLIGENT (we think) beings, and analyses and interpretations of the experiments are performed by INTELLIGENT beings.

With all this intelligence floating around how can ANYONE ascribe the creation of ANYTHING to unintelligent random processes! RANDOMNESS IS CHAOS!

ORDER cannot come from CHAOS without INTELLIGENCE! It is the other way around. That's the whole premise of the Second Law. From ORDER to CHAOS, unless INTELLIGENCE intervenes.

Taking the dishwasher analogy from Susan's post, what do we have to say about the "heat" spewed out of the internet?

Obviously we are witnessing the evolution of the internet towards higher order through this chaos of millions of actors using it for various purposes (and sub-level rules, and a general lack of control of the potential uses...)

but what systems outside of the internet are becoming more chaotic because of the increasing order within the internet, along the lines of the dishwasher analogy?

or is the entropy captured in increasingly weird sub-systems within the internet?

Should we really be discussing the increasing order in certain parts of the internet alongside the increasing disorder in other parts?

David’s mention of Prigogine, Thom, and Popper is quite apt. Note that Alan Turing focused on related research about morphogenesis and emergence for the last few years of his life. Ironically, the Stalinist censorship of Prigogine’s early research foreshadowed Popper’s arguments. Popper’s self-criticisms as a devout Christian seems particularly appropriate for this debate about evolution vs. creationism.

The discussions of “closed” versus “open” here seem a bit muddied on either side. Susan might have benefited by discussing an open system that adapts to provide dynamic closure (“operational closure” in the literature). Think of "boundaries" which are flexible. However there are other necessary conditions for emerging systems, other factors which lead to metastability. For instance, Fisher-Kolmogorov-Petrovskii provided math to describe the propagation of a species in an active, “rich of energy” (food) medium of unlimited space. As that work gets applied in research (e.g. Prigogine and dissipative structures) the notion of “species” sometimes refers to chemical phenomena as well as biological. It has also been shown to obtain for sociological and legal phenomena (see Niklas Luhmann, Gunther Teubner, Terry Winograd, et al.)

The rhetoric in that recent Wired issue barely scratches the surface of a relatively large body work for which there exist long debates in many circles – debates which the lingering Aristotelians among us will not likely concede anytime soon. Prigogine earned his Nobel Prize on the quantitative side, but arguably the more rigorous and profound work was performed on the qualitative side by Humberto Maturana and Francisco Varela in “Autopoiesis and Cognition: The Realization of the Living”. Stafford Beers’ forward in their text is highly recommended reading, particularly for anyone who wants to extend systems theory to social interactions online. Warning: the subtext of Maturana and Varela demonstrates how cognition and living almost directly refute most notions of “information”, with dire consequences for political structures; beware the implications for those of fixed worldviews.

FWIW, the application of dissipative structures and autopoietic systems theory to the Internet, globalization, and transnationals has become my research focus over the past several years. For recent applications of related theory to the Internet, see a paper on the use of non-equilibrium systems for measuring network-based attacks globally:


That may help address some questions here about "order" and "disorder" regarding the Internet.

All of you are talking out of your rear ends. Negentropic? The nuts moving to the surface lowers the center of gravity and allowing a more energetically favorable system.

Seriously, Stop talking like you are Newton or Hawkings.

Wow, it is amazing how unintelligent educated people can be. Or is it "willful ignorance"?


THAT is DISORDER without intelligence. I disengaged my mind and just hit the keys as if I were a monkey banging on a keyboard.

Now IMAGINE what DNA would be like with a monkey at the helm! Or better yet, let a Cosmic Fart (the Big Bang) run everything!

WAIT, The letters I randomly typed were created by man to express concepts, and Love, and Friendship and so on. SO was the keyboard I typed on!

Here is what the universe would look like if random processes REALLY made everything:

Did you read what was typed? Oh Wait, Letters do not exist in a chaotic universe! Keyboards do not exist! Even Monkeys do not exist! And, last of all, HUMANS do not exist to argue over what was printed or theorized or created!

i see, Comment from STFU is allowed, but an honest comment from a Believer isn't. Hmmm

1 TrackBacks

Listed below are links to blogs that reference this entry: Thermodynamics and Evolution.

TrackBack URL for this entry:

Joi makes one of those posts that makes me wish I had more time to investigate – this one was inspired by a Susan Crawford post, and Susan was inspired by a Seth Schoen post about evolution and the second... Read More