Sunday, February 01, 2009

Emergence 2: Weak vs. Strong Emergence

Modern study of emergence tends to separate "emergence" into two broad categories:  weak emergence and strong emergence.  This distinction is made by Mark Bedau, among other commentators.  The differences between the two species of emergence are significant.

Strong emergence:  This is the species of emergent discussed in the first post in this series--an emergence that is ontologically separate from its microstructure, not derivable from that microstructure, and capable of exerting downward causation on the functioning of that microstructure.

Weak emergence:  Weak emergence is the set of phenomena that is theoretically reducible to the known laws governing the microstructure, but where the calculations required to predict the resulting phenomena are so complex as to be effectively impossible.  Instead, with weak emergence, these calculations are carried out by means of simulation.

This study of weak emergence, also called more broadly the study of complexity, is seen in the (overlapping) modern disciplines of systems theory, neural networks theory, dynamical systems theory, agent based modeling, complex adaptive systems, etc.  One of my favorite authors, John Holland ("Hidden Order" and "Emergence") discusses this form of emergence.

The study of strong emergence has been less fruitful, so far, in science, perhaps because like weak emergence, it is not practicably reducible, but unlike weak emergence, it is also (so far) not capable of simulation.  Strong emergence--as is pointed out by many of its critics--is primarily either a subject of philosophical discourse or it is narrowly useful as a theory of human consciousness that really hasn't changed much since Roger Sperry's groundbreaking theory of consciousness in 1969.

For this blog, and the theory of Rhizome, are we interested in the effects of weak or strong emergence?  In my mind, the answer is clearly "both."  There is little doubt that human systems, evolving civilization, group dynamics, and economic dynamics exhibit weakly emergent traits.  The study of weak emergence, therefore, will likely give us insight into the operation of these systems, how to shape them, and the viability of alternative structures.  My theory of Rhizome communication, for example, and its information processing and economic coordination capability, can be compared to the capabilities of more hierarchal structures by way of simulation.  The functioning of these various structures are fundamentally reducible to known and understood interaction of their microstructure, but the resulting calculations are simply not feasible, and therefore the study of weak emergence may offer useful insights into their functioning.  I am also, however, interested in the potential strong emergence in human civilization.  Even if strong emergence exists nowhere but human consciousness, that alone is interesting enough to warrant further exploration--at a minimum, there is the question of whether a "weakly emergent" phenomena like human civilization and economics can be meaningfully separated from the strongly emergent consciousness present at each individual human component in the microstructure...

Intuitively, however, I think there is a good chance that strong emergence plays a much more significant role beyond human consciousness--in group dynamics, communication, cultural trends, and economic coordination.  This is where I think emergence has something fundamentally new and valuable to teach us--beyond what we can learn from studying complex human systems through the lens of weak emergence.  What exactly this influence of strong emergence is I do not know--one reason I find this so fascinating is that it is a possibility that is simply not being studied or considered.  The possibilities seem endless.  I do, however, have a theory, specifically pertaining to the influence of strong emergence on coordination and information processing.

My theory is that, just like human consciousness (which I presume to be strongly emergent) offers superior coordination and information processing capabilities when compared to non-strongly emergent coordination and information processing (such as computers), I think that strongly emergent phenomena in economic coordination, dynamic network structuring, and human communication may dramatically increase the efficiency of operation of human systems that foster strong emergence.  Such strong emergence may be present to varying degrees in our current economic and political systems, but my hunch is that structures designed to better foster strong emergence would leverage these effects to a much greater degree.  Specifically, as I hinted at in the first post in this series, decentralized but well connected networks (e.g. Rhizome) that mimic the connectivity of the human brain may, like the human brain, foster strong emergence.  Hierarchy, on the other hand, may actually act to dampen strong emergence--in fact, one of the main evolutionary features of hierarchy in human civilization may be that, by dampening strong emergence, system control is maintained at the top of the hierarchy, rather than ceded to some degree to a strongly emergent and ontologically distinct phenomenon.  These ideas are, of course, still very fuzzy, difficult to articulate, and poorly supported, but the potential here makes them worthy of further investigation in my opinion.

While strong emergence may not be subject to study by simulation, as weak emergence is, I think it may be replicable.  That is, if we accept that strong emergence is derivative of an underlying microstructure, then by replicating that microstructure, or at least the salient features of that microstructure, it may be possible to foster strongly emergent phenomena.  This has always been an influence in the theory of Rhizome--largely unstated until now because I haven't had the tools to explain why this was anything more than intuition and speculation on my part.  While I think Rhizome, as a theory, is valid even if there is nothing more than weak emergence in this world, the potential to leverage strong emergence may make the theory even more robust.

Readers may also find my litigation checklist of interest.


Jeff Vail said...

Readers may also want to see a comment to the first post in this series from Karl, who pointed out that my first take on emergence failed to distinguish between "strong" and "weak" emergence, and that the resulting lack of clarity made it look like I was minimizing the role of computer science/modeling/simulation in emergence studies. Hopefully this post has shown that, while those disciplines (and the Santa Fe institute) play important roles in the study of "weak emergence," they really don't address "strong emergence."

Karl also points out that the actual result of downward causation in social systems is very vague--what is it? Perhaps his question is itself an answer: I read his comment just moments after I put the finishing touches on this post (Part 2), and now it has helped me to integrate the two posts in this series so far, and to understand some confusion I was causing by my own approach to thsi problem.

That's synchronicity--it could be purely coincidence, but many (e.g. Jung) were convinced otherwise, though they had no framework to explain how synchronicity operates. I'll suggest that synchronicity may be the effect of strong emergence exerting downward causation on a system's microstructure. I think the potential for synchronicity to enhance the communication and coordination capabilities of an economic/political system should be clear, but it's something I plan to explore. I hadn't associated synchronicity with emergence before, and I had been struggling to answer the frequent question--"how would strong emergence actually help a Rhizome structure function?" Now I've at least connected the dots--thanks, Karl.

On a brief but (I think) entertaining aside, I've experience many very interesting cases of synchronicity in my life. One of my favorites: I had just finished reading Katherine Neville's "The Eight," a novel about a conspiracy/adventure involving the knights templar, the Cathars, etc., (and, of course, a veiled initiatory tale) when I found my travel plans suddenly and unexpectedly shot to hell: an injury prevented completing the Coast to Coast hike across England, and I found myself in a small town in the South of France (Vaour) that 24 hours earlier I had never heard of, nor had I even planned to visit France that year. Through a variety of "coincidence," I was sleeping in the farmhouse of my wife's relatives literally in the shadow of a ruined Templar fortress, next to a Cathar castle, reading Quantum Psychology by Robert Anton Wilson. Of course, that could all just be coincidence.

Jeff Vail said...

Another instance of synchronicity that might be of interest: I've recently finished reading Atul Gawande's "Checklist Manifesto," and I will be writing a series of posts in the near future about checklists, open-source OS, etc. One of Gawande's examples of the use of checklists to control complexity comes from the building industry--the complex set of checklists used to coordinate the function of the various trades in large-scale construction projects. Immediately after finishing Gawande's book (last week), I began reviewing some initial disclosures in a construction defect lawsuit that involve precisely this issue: Architect's construction change directives, field directives, extra work tickets, coordination slips, etc. between this building's owner, architect, engineers, prime contractor, subs and sub-sub contractors. While it's not the most glamorous job (and I'm fortunate that well less than 5% of my work time is spent in this kind of "document review"), I've found it invaluable as a concrete illustration of what Gawande discussed for a chapter in his book. In general, I've always found that my life provides interesting illustrations of the concepts about which I've just read, and I tend to read new books covering new topics just in time to find an application for the new knowledge. This is certainly a self-fulfilling prophecy to some extent (it's on my mind, so I look for examples of it in my life), but it also seems like a fertile ground for synchronicity in action...

ryan said...

great post, jeff.

"Hierarchy, on the other hand, may actually act to dampen strong emergence--in fact, one of the main evolutionary features of hierarchy in human civilization may be that, by dampening strong emergence, system control is maintained at the top of the hierarchy, rather than ceded to some degree to a strongly emergent and ontologically distinct phenomenon."

think of this in terms of how the media/US government/system of hierarchy in general act unconsciously (not in reflective awareness) to maintain the structure of hierarchy: we sometimes think that there are some hidden conspirators pulling the strings behind the media and gov - as if someone or group of elites "at the top" were making all the decisions... what if it's more a reflection of the dynamic interactions we all unconsciously make in order to support the system of hierarchy? each day we are (mostly) unaware of the nature and complexity of our social stigmergy which supports and feeds hierarchy.

example: one of the first and most common questions we ask one another upon meeting is "what do you do?" the unconscious, unspoken assumptions and reactions within this seemingly insignificant line of questioning might be crucial to maintaining the fabric of society. our typical expectation to the question, "what do you do for a living" is a response of some form of vocation which fits within the existing society - no matter how insane or maladaptive that social structure is. we socially accept killers (and reward them, give them benefits and education...) if they are employed as "soldiers" and yet reel almost endlessly against even the slightest violence within the social group - even if that violence is directed not against humans, but the property or possessions of those higher on the hierarchy. the recent removal of the youtube and facebook sites for the feb 6th planned day of global open source attacks on banks and governments is an example. hierarchy definitely stomps down on emergence... imagine how quickly society could shift if soldiers or bankers or capitalists or politicians of any flavor became widely rejected or even targeted by global open source insurgencies...

esme said...

"...economic coordination, dynamic network structuring, and human communication may dramatically increase the efficiency of operation of human systems that foster strong emergence..." - jeff, haven't you just described egalitarian, autonomous, hunter-gatherer cultures?

these cultures toyed with the parts but did not overly try to control them, their cosmological view of the world contributed to their synchronous understanding of its most minute details as applicable to their daily realities, and their own interpersonal interactions regulated their own part (via maintaining small, close-knit populations) in the total emergent system which would included mind-body-soul/consciusness inextricably, to their environment. Although modern man would like to confine hunter/gatherers to the "Stone Age," the reality is that they succefully attuned themselves and influenced the emergent properties of a supremely complex ecological system. Of course, we are only a timy part of the "parts" that govern the larger emergent properties of climate, earthquakes/volcanoes, and other species, that probaly led some to favor primitive agriculture which led to increased populations and, i would argue, LESS complex human interrelations (some not all of course) with all its present feedback manifestations.

I think there is a critical piece missing from this discussion and that is what emotions and diverse, often conflicting, present-day values play in emergence theories. Although, I think it was adressed by a commentor on the previous post.

If you have experienced that "fertile ground" of synchronicity, it is because of the pattern of human behavior generates similar thoughts, situations, and things or threads of, elsewhere at the same time you are experiencing them, no? Really, synchronicity is just the likelihood of probabilities, to which we have ascribed the concepts of luck, fate, chance, destiny, etc. It's just that since we are in them while we experience them (the epiphenomena referred to by a commentor), we've also ascribed the concept of the removed divine to it no?

anyhow, it's one thing to speak of the technical properties, but not at the price of the emotional/value properties since doing so has so manifested in our current modern dysfunction.

theputnamprogram said...

generally, I agree, particularly with your suggestion about the relation between culture and strong emergence. btw... the separation of strong emergence and weak emergence, due to Chalmers, is crap and not stable enough to build any argument onto it.
Well, all you say is nice, "honey for my ears"... but it does not help any further. Why? Because all these claims and suggestions remain kind of a mysticism if you won't give a mechanism.
Think about biology, around 1880 and since 1970. The difference is that 1880 they claimed an elan vital, while since the 1970ies (Tinbergen)biology looks for mechanisms... if you are interested, risk a look here: