Tuesday, August 22, 2006
COMPLEXITY AND CONNECTIVITY:BAR-YAM AGAIN
Preface:
I've been pondering the relationship between complexity and connectivity ever since I was prompted by a post and an email from Dave at Thoughts Illustrated. Due to time constraints, I was not able to give the insights Dave offered in his email the proper attention they deserved then but I'm returning to the subject today.
The background here is a short paper on civilizational complexity, " Complexity Rising:From Human Beings to Human Civilization, A Complexity Profile" by Dr. Yaneer Bar-Yam that I forwarded to a number of thoughtful people, some of whom also were bloggers. My take on Bar-Yam's paper at the time had to do with " The Resilience of Civilizations"; Steve DeAngelis at ERMB responded with "Networked Civilization Revisited"; Curtis Gale Weeks at Phatic Communion put these ideas in the context of the theories of John Boyd with "Rule Sets and the Revised OODA" and Dan of tdaxp followed Curtis with "Comments on Verticalization and Progress". Dr. Von, who introduced me to the ideas of Dr. Bar-Yam in the first place, offered some verbal commentary in person but helpfully pointed to an older post of his, "Our Universe:Continual Emergence"as well.
The Relationship of Complexity and Connectivity in Civilization:
Civilizations are long enduring, complex adaptive social systems that remain distinctive from their neighbors. Differences between civilizatons are visible even to casual observers in the form of culture, language, religious belief, social customs and economic productivity. The origin or causation of civilizational differences have been hotly debated for dozens of centuries and at various times, climate, geography, divine favor, chance or superiority in terms of culture, genetics, martial prowess, political, economic or moral systems have all been offered up as explanations and all have met with fierce criticism.
However, my purpose here is not to weigh in on the merits of Baron de Montesquieu, Jared Diamond, Karl Marx, Victor Davis Hanson or Robert Wright but to point out, first, that the superficial manifestations of civilizations, that make them unlike others, all represent underlying patterns of complexity and of connectivity. Secondly, that while complexity and connectivity in a "superorganism" sized social system have a high degree of interaction and are often mutually reinforcing, they are not one and the same. On some levels, complexity can increase or sharply limit connectivity. Connectivity in turn, may simplify or complicate the workings of a system, increasing or decreasing " friction" as well as numerous other effects.
Take for example, late medieval to early modern central Europe. The Holy Roman Empire, at one point, had something on the order of 300 independent, overlapping, interdependent polities. That's more sovereigns than exist today on the entire planet, some ruling nothing bigger than a knightly estate, crammed into an area slightly larger than Germany.
Obviously, in the case of the Holy Roman Empire, the degree of political and social complexity were very high relative to the population of the time or to monarchical states like England . This complexity came at the expense of connectivity, particularly economic connectivity given the number of sovereign and semi-sovereign entities ( most with their hand out) that could and did interfere with trade in a myriad of irrational ways. Unsurprisingly, even after Napoleon consolidated this Germanic crazyquilt into a more manageable 35 state Confederation of the Rhine, the first serious exploration of German unification involved a proposal for a customs union, the Zollervein, that would have simplified, and rationalized trade, increasing the level of connectivity.
Dr. Barnett has made a similar argument regarding the connectivity effect of the Federal union of the United States in Blueprint for Action as well as elsewhere. If you look at the lowest panel of the Bar-Yam Historical Complexity diagram above, you will note the "hierarchy" line levelling off and falling while the "specialization/diversity" and "lateral connection" lines rising in parallel along a timeline. This is a nice visualization of changing the political or legal complexity of rule-sets ( reducing the number of competing rule-sets) intersecting with improvements in technology to permit higher levels of emergent connectivity.
Granted, the higher levels (and faster velocity of transaction) of connectivity are also, in themselves a form of complexity that will generate unanticipated spillover effects that will give rise to demands for regulation or control in the form of new rule-sets. As will the resistance of vested interests to emerging forms of connectivity or technology that endanger rentier arrangements, oligopolistic markets or the authority of corrupt tyrannies. As the relationship between complexity and connectivity is dynamic, these reactionary political responses to creative destruction are unavoidable.
Their success though, is far from inevitable. We have choices if we care to exercise them. ¶ 11:05 PM
Preface:
I've been pondering the relationship between complexity and connectivity ever since I was prompted by a post and an email from Dave at Thoughts Illustrated. Due to time constraints, I was not able to give the insights Dave offered in his email the proper attention they deserved then but I'm returning to the subject today.
The background here is a short paper on civilizational complexity, " Complexity Rising:From Human Beings to Human Civilization, A Complexity Profile" by Dr. Yaneer Bar-Yam that I forwarded to a number of thoughtful people, some of whom also were bloggers. My take on Bar-Yam's paper at the time had to do with " The Resilience of Civilizations"; Steve DeAngelis at ERMB responded with "Networked Civilization Revisited"; Curtis Gale Weeks at Phatic Communion put these ideas in the context of the theories of John Boyd with "Rule Sets and the Revised OODA" and Dan of tdaxp followed Curtis with "Comments on Verticalization and Progress". Dr. Von, who introduced me to the ideas of Dr. Bar-Yam in the first place, offered some verbal commentary in person but helpfully pointed to an older post of his, "Our Universe:Continual Emergence"as well.
The Relationship of Complexity and Connectivity in Civilization:
Civilizations are long enduring, complex adaptive social systems that remain distinctive from their neighbors. Differences between civilizatons are visible even to casual observers in the form of culture, language, religious belief, social customs and economic productivity. The origin or causation of civilizational differences have been hotly debated for dozens of centuries and at various times, climate, geography, divine favor, chance or superiority in terms of culture, genetics, martial prowess, political, economic or moral systems have all been offered up as explanations and all have met with fierce criticism.
However, my purpose here is not to weigh in on the merits of Baron de Montesquieu, Jared Diamond, Karl Marx, Victor Davis Hanson or Robert Wright but to point out, first, that the superficial manifestations of civilizations, that make them unlike others, all represent underlying patterns of complexity and of connectivity. Secondly, that while complexity and connectivity in a "superorganism" sized social system have a high degree of interaction and are often mutually reinforcing, they are not one and the same. On some levels, complexity can increase or sharply limit connectivity. Connectivity in turn, may simplify or complicate the workings of a system, increasing or decreasing " friction" as well as numerous other effects.
Take for example, late medieval to early modern central Europe. The Holy Roman Empire, at one point, had something on the order of 300 independent, overlapping, interdependent polities. That's more sovereigns than exist today on the entire planet, some ruling nothing bigger than a knightly estate, crammed into an area slightly larger than Germany.
Obviously, in the case of the Holy Roman Empire, the degree of political and social complexity were very high relative to the population of the time or to monarchical states like England . This complexity came at the expense of connectivity, particularly economic connectivity given the number of sovereign and semi-sovereign entities ( most with their hand out) that could and did interfere with trade in a myriad of irrational ways. Unsurprisingly, even after Napoleon consolidated this Germanic crazyquilt into a more manageable 35 state Confederation of the Rhine, the first serious exploration of German unification involved a proposal for a customs union, the Zollervein, that would have simplified, and rationalized trade, increasing the level of connectivity.
Dr. Barnett has made a similar argument regarding the connectivity effect of the Federal union of the United States in Blueprint for Action as well as elsewhere. If you look at the lowest panel of the Bar-Yam Historical Complexity diagram above, you will note the "hierarchy" line levelling off and falling while the "specialization/diversity" and "lateral connection" lines rising in parallel along a timeline. This is a nice visualization of changing the political or legal complexity of rule-sets ( reducing the number of competing rule-sets) intersecting with improvements in technology to permit higher levels of emergent connectivity.
Granted, the higher levels (and faster velocity of transaction) of connectivity are also, in themselves a form of complexity that will generate unanticipated spillover effects that will give rise to demands for regulation or control in the form of new rule-sets. As will the resistance of vested interests to emerging forms of connectivity or technology that endanger rentier arrangements, oligopolistic markets or the authority of corrupt tyrannies. As the relationship between complexity and connectivity is dynamic, these reactionary political responses to creative destruction are unavoidable.
Their success though, is far from inevitable. We have choices if we care to exercise them. ¶ 11:05 PM
Comments:
<< Home
I'm not sure I buy the portrayal of hunter-gather societies in the diagram, Mark. Consider the pre-contact Hawaiians, for example (I'd consider them hunter-gatherers practicing widespread horticulture so they fit either in the hunter-gatherer category or “early civilization”). Social organization was considerably more complex with distinct realms of primacy rather than a simple structure, the alii (nobility), kahuna (priests and skilled craftsmen), makaainana (common people), and kauwa. The kapu system bound everybody from chief to commoner and the medicine kahuna had real power within his area just as the canoe kahuna had within his.
If this sounds more like a network system than a simple hierarchy, it's because it is.
BTW I've known more than one Fortune 1000 company whose de facto organizational structure was the one listed under “hunter-gatherer”.
BTW I've known more than one Fortune 1000 company whose de facto organizational structure was the one listed under “hunter-gatherer”.
Hi Dave,
Thank you for the informed comment. i'm not well versed in Hawaiian history, knowing mostly the significance that like the Zulus in Southern Africa, the Hawaiians transitoned upward into a kingdom structure in the early modern era.
"BTW I've known more than one Fortune 1000 company whose de facto organizational structure was the one listed under “hunter-gatherer”. "
Says something about the personality of the CEO :O)
Thank you for the informed comment. i'm not well versed in Hawaiian history, knowing mostly the significance that like the Zulus in Southern Africa, the Hawaiians transitoned upward into a kingdom structure in the early modern era.
"BTW I've known more than one Fortune 1000 company whose de facto organizational structure was the one listed under “hunter-gatherer”. "
Says something about the personality of the CEO :O)
the superficial manifestations of civilizations, that make them unlike others, all represent underlying patterns of complexity and of connectivity. Secondly, that while complexity and connectivity in a "superorganism" sized social system have a high degree of interaction and are often mutually reinforcing, they are not one and the same. On some levels, complexity can increase or sharply limit connectivity. Connectivity in turn, may simplify or complicate the workings of a system, increasing or decreasing " friction" as well as numerous other effects.
I like the consideration -- well, the act of considering -- complexity and 'connectivity' interacting.
Lately I've been writing a lot on the ideas of social 'networks' and 'connectivity' as I explore these ideas. And no matter how I approach the topics, I keep coming to the same conclusion: that these are, after all, ideas about humans, societies, and cultures and may be fine as generalizations but less fine as explanations for how groups of humans interact. I'm more partial to complexity than connectivity, even if both are more often used to describe thoughts and thinking about societies than the actual, concrete societies they are meant to describe; complexity speaks to profusion, variation, etc., and really does get at the truth of the matter moreso than connectivity, since our concrete connections are often brief, ephemeral.
Civilizational or cultural 'connectivity' -- the things that 'connect' people well enough for us to call them a definable group -- is more often than not merely a description of how the members are alike, in: language, habits of thinking, methods of acting, etc. Lately I've been thinking of this in terms of the Periodic chart, and how two water molecules -- hydrogen and oxygen -- on opposite sides of the galaxy will act and react in more or less the same way if their local environments are very similar: they themselves are not 'connected' by more than reality and the Universe, but collections of them will still run downhill creating channels which we call rivers. Similarly, two humans within a definable civilization may go to a bank for a loan -- or to an oil baron -- because they think alike, but this does not mean they are 'connected' by their source of cash in any but the most generalized sense. (At the moment of shaking hands or taking money from the provider, they may each really connect with that person; but then each goes his own way and the real connection is ended.)
A civilization ends, then, when the members within it develop very different ways of relating -- not because the old 'network paths' are disrupted, although this may appear to be what has happened when in fact changes in the world produce different paradigms or ideologies and habits of thought and activity as people begin to observe that changed world.
Rule-set adoption is a personal activity rather than a social activity. If enough individuals within a group adopt a rule-set, even the rest will eventually adopt it -- even if grudgingly at first. Barnett's theories revolve around the notion that physically changing the world in particular ways will naturally lead to general adoption of specific rule-sets. Getting enough people to not only adopt those rule-sets but also enforce those rule-sets is the real problem, since the enforcers of opposing or varying rule-sets continue to make changes in the world that will make people question their original 'adoption'. I see Bar-Yam's final diagram to be a result of wide-spread adoption and enforcement of rule-sets rather than a top-down sort of enforcement: what has been called 'networking' and 'connectivity' will emerge rather than be implemented. That might be a fine line, since alteration of the world in specific ways may seem to be implementation; but it's the complexity of our contemporary world that limits a too-narrow focus on how the world can be changed to successfully create the desired results.
I like the consideration -- well, the act of considering -- complexity and 'connectivity' interacting.
Lately I've been writing a lot on the ideas of social 'networks' and 'connectivity' as I explore these ideas. And no matter how I approach the topics, I keep coming to the same conclusion: that these are, after all, ideas about humans, societies, and cultures and may be fine as generalizations but less fine as explanations for how groups of humans interact. I'm more partial to complexity than connectivity, even if both are more often used to describe thoughts and thinking about societies than the actual, concrete societies they are meant to describe; complexity speaks to profusion, variation, etc., and really does get at the truth of the matter moreso than connectivity, since our concrete connections are often brief, ephemeral.
Civilizational or cultural 'connectivity' -- the things that 'connect' people well enough for us to call them a definable group -- is more often than not merely a description of how the members are alike, in: language, habits of thinking, methods of acting, etc. Lately I've been thinking of this in terms of the Periodic chart, and how two water molecules -- hydrogen and oxygen -- on opposite sides of the galaxy will act and react in more or less the same way if their local environments are very similar: they themselves are not 'connected' by more than reality and the Universe, but collections of them will still run downhill creating channels which we call rivers. Similarly, two humans within a definable civilization may go to a bank for a loan -- or to an oil baron -- because they think alike, but this does not mean they are 'connected' by their source of cash in any but the most generalized sense. (At the moment of shaking hands or taking money from the provider, they may each really connect with that person; but then each goes his own way and the real connection is ended.)
A civilization ends, then, when the members within it develop very different ways of relating -- not because the old 'network paths' are disrupted, although this may appear to be what has happened when in fact changes in the world produce different paradigms or ideologies and habits of thought and activity as people begin to observe that changed world.
Rule-set adoption is a personal activity rather than a social activity. If enough individuals within a group adopt a rule-set, even the rest will eventually adopt it -- even if grudgingly at first. Barnett's theories revolve around the notion that physically changing the world in particular ways will naturally lead to general adoption of specific rule-sets. Getting enough people to not only adopt those rule-sets but also enforce those rule-sets is the real problem, since the enforcers of opposing or varying rule-sets continue to make changes in the world that will make people question their original 'adoption'. I see Bar-Yam's final diagram to be a result of wide-spread adoption and enforcement of rule-sets rather than a top-down sort of enforcement: what has been called 'networking' and 'connectivity' will emerge rather than be implemented. That might be a fine line, since alteration of the world in specific ways may seem to be implementation; but it's the complexity of our contemporary world that limits a too-narrow focus on how the world can be changed to successfully create the desired results.
# posted by 
: Thursday, 24 August, 2006
: Thursday, 24 August, 2006
One of the things that this analysis ignores is that individual nodes in a networked society, for example, can function at any level, down to hunter-gatherer. Dave notes, for example, that some Fortune 1000 companies act that way, and I would argue that most act according to the "early civilizations" model, at least in my experience.
The implication, of course, is that even in a highly networked society, individual nodes can deliberately choose to sever connectivity and to flout commonly-accepted rulesets. Indeed, it is easier for nodes in a networked society to isolate themselves than it is in a semi-hierarchichal society, because in a networked society you get to pick and choose your connectivity. Hizb'allah wants, for example, to be connected to the banks and the arms markets, but not to the Geneva conventions or any governing organizations (whether it be the government of Lebanon or the UN) except where the exchange is one way: benefit to Hizb'allah but not to the governing organization.
This actually begs the question of whether a networked social order can remain such, or must it reach a certain level of dispersion of authority, and then have forceful impositions of authority moving the order back to a less networked state.
The implication, of course, is that even in a highly networked society, individual nodes can deliberately choose to sever connectivity and to flout commonly-accepted rulesets. Indeed, it is easier for nodes in a networked society to isolate themselves than it is in a semi-hierarchichal society, because in a networked society you get to pick and choose your connectivity. Hizb'allah wants, for example, to be connected to the banks and the arms markets, but not to the Geneva conventions or any governing organizations (whether it be the government of Lebanon or the UN) except where the exchange is one way: benefit to Hizb'allah but not to the governing organization.
This actually begs the question of whether a networked social order can remain such, or must it reach a certain level of dispersion of authority, and then have forceful impositions of authority moving the order back to a less networked state.
Great commentary Jeff !
"The implication, of course, is that even in a highly networked society, individual nodes can deliberately choose to sever connectivity and to flout commonly-accepted rulesets."
Yes and no. I think that is contingent upon the nature of the network in question.
Some networks by design or inherent form, might impose unacceptable costs for so doing. For example, flouting the religious sensibilities of a tightly-knit, sectarian community will bring more immediate consequences than a powerful state flouting international law
"Indeed, it is easier for nodes in a networked society to isolate themselves than it is in a semi-hierarchichal society, because in a networked society you get to pick and choose your connectivity"
Again, in general I agree but I think you may be underestimating the costs. Going out to the wilds to live in a rude shack so you can be relatively untraceable in sending bombs to scientists and airline executives requires an unusual level of sacrifice. In terms of economic calculation, there is a bias toward remaining connected.
"The implication, of course, is that even in a highly networked society, individual nodes can deliberately choose to sever connectivity and to flout commonly-accepted rulesets."
Yes and no. I think that is contingent upon the nature of the network in question.
Some networks by design or inherent form, might impose unacceptable costs for so doing. For example, flouting the religious sensibilities of a tightly-knit, sectarian community will bring more immediate consequences than a powerful state flouting international law
"Indeed, it is easier for nodes in a networked society to isolate themselves than it is in a semi-hierarchichal society, because in a networked society you get to pick and choose your connectivity"
Again, in general I agree but I think you may be underestimating the costs. Going out to the wilds to live in a rude shack so you can be relatively untraceable in sending bombs to scientists and airline executives requires an unusual level of sacrifice. In terms of economic calculation, there is a bias toward remaining connected.
Jeff, following up on Mark's comments...
The implication, of course, is that even in a highly networked society, individual nodes can deliberately choose to sever connectivity and to flout commonly-accepted rulesets.
I think that in a 'highly networked society,' in which so many potential pathways to personal fulfillment exist, the severing of 'connectivity' may not be as dramatic as it would be in a hierarchical social structure. So...
in a networked society you get to pick and choose your connectivity
This may actually be part of the 'strength' of a 'highly networked society,' depending on the nature of that society. I imagine different levels and types of rule-sets, some of which may be quite inclusive of a variety of types of 'connection'; they will not break simply because alternate 'connections' are chosen.
Because I view rule-set adoption as being a very personal decision rather than an imposed condition -- and essentially, the formation of ideologies, habits of thought and acting -- the type of broadly adopted rule-sets within a resilient 'networked society' I envision will be broadly inclusive paradigms. For instance, the type of society America currently has allows one person to become a doctor and another a mechanic in the pursuit of personal financial and psycho-social stability; the 'American Dream' is a broadly inclusive paradigm held by many who might otherwise fragment into a plural monocultural polity. Of course, some paths for achieving that Dream are not allowed, and there is where we might find groups breaking off to pursue their own Dreams, and where friction with the rest of American society might occur: drug and prostitution rings, polygamists, etc. (Yet, I find the growing debate over polygamy and the legalization of some drug-use and even prositution, to be signs that a more inclusive paradigm may slowly be forming -- the "gay-marriage debate" is another area where friction can result in fragmentation or a greater inclusivity.)
This actually begs the question of whether a networked social order can remain such, or must it reach a certain level of dispersion of authority, and then have forceful impositions of authority moving the order back to a less networked state.
--This is an interesting question, and one I've had in mind for a long time. I think that a more interesting consideration will be if 'dispersion of authority' actually occurs in the envisaged 'failing networked social order.' It seems to me that a highly networked society may more likely fail to form than reach the kind of tipping point you suggest with your question -- i.e., smaller hierarchies may develop within a society, leading to plural monoculturalism in which the different hierarchies war with one another, preventing the development of a highly networked society; in which case, an imposition of order upon the disparate groups may force commingling of those groups and might actually help to lead to a highly networked social structure.
Probably, the level of disparity between monocultures within a polity will determine the necessary level of force required to break up the individual hierarchies. However, much of that force will not be standard martial or legal force that seeks to utterly destroy the monocultures -- they may be quite resilient, perhaps fighting to the last man -- but rather a 'consilience-focused' force working to establish inclusive paradigms which most of the parties can adopt.
Post a Comment
The implication, of course, is that even in a highly networked society, individual nodes can deliberately choose to sever connectivity and to flout commonly-accepted rulesets.
I think that in a 'highly networked society,' in which so many potential pathways to personal fulfillment exist, the severing of 'connectivity' may not be as dramatic as it would be in a hierarchical social structure. So...
in a networked society you get to pick and choose your connectivity
This may actually be part of the 'strength' of a 'highly networked society,' depending on the nature of that society. I imagine different levels and types of rule-sets, some of which may be quite inclusive of a variety of types of 'connection'; they will not break simply because alternate 'connections' are chosen.
Because I view rule-set adoption as being a very personal decision rather than an imposed condition -- and essentially, the formation of ideologies, habits of thought and acting -- the type of broadly adopted rule-sets within a resilient 'networked society' I envision will be broadly inclusive paradigms. For instance, the type of society America currently has allows one person to become a doctor and another a mechanic in the pursuit of personal financial and psycho-social stability; the 'American Dream' is a broadly inclusive paradigm held by many who might otherwise fragment into a plural monocultural polity. Of course, some paths for achieving that Dream are not allowed, and there is where we might find groups breaking off to pursue their own Dreams, and where friction with the rest of American society might occur: drug and prostitution rings, polygamists, etc. (Yet, I find the growing debate over polygamy and the legalization of some drug-use and even prositution, to be signs that a more inclusive paradigm may slowly be forming -- the "gay-marriage debate" is another area where friction can result in fragmentation or a greater inclusivity.)
This actually begs the question of whether a networked social order can remain such, or must it reach a certain level of dispersion of authority, and then have forceful impositions of authority moving the order back to a less networked state.
--This is an interesting question, and one I've had in mind for a long time. I think that a more interesting consideration will be if 'dispersion of authority' actually occurs in the envisaged 'failing networked social order.' It seems to me that a highly networked society may more likely fail to form than reach the kind of tipping point you suggest with your question -- i.e., smaller hierarchies may develop within a society, leading to plural monoculturalism in which the different hierarchies war with one another, preventing the development of a highly networked society; in which case, an imposition of order upon the disparate groups may force commingling of those groups and might actually help to lead to a highly networked social structure.
Probably, the level of disparity between monocultures within a polity will determine the necessary level of force required to break up the individual hierarchies. However, much of that force will not be standard martial or legal force that seeks to utterly destroy the monocultures -- they may be quite resilient, perhaps fighting to the last man -- but rather a 'consilience-focused' force working to establish inclusive paradigms which most of the parties can adopt.
# posted by : Friday, 25 August, 2006
: Friday, 25 August, 2006<< Home
