Contents
- Index
White Paper - Explanation Versus Intervention (Top)
Explanation Versus Intervention in Complex Systems
(Draft Paper)
Dr. Lawrence Boyd
August 16, 2003
Two Perspectives
There are two fundamentally different ways to study complex systems. The traditional approach is to view complex phenomena from the top down and construct components, relationships, and functions from surface phenomena like behavior. A more recent approach is to view complex systems from the bottom up and explain surface phenomena in terms of higher level self-organized patterns of low level entities - a process referred to as "emergence." A popular example is the study of the human brain. One approach looks at behavior and tries to figure out how the brain is built and functions to produce the behaviors. The other approach starts with neurons and low level entities and asks how they may have self-organized into higher levels of complexity. Another popular example is the study of ant colonies in which individual ants instinctively do what they do in response to their immediate environment, including the ants around them, yet the outcome is some higher level of organization necessary to the survival of the colony. (citations). As might be expected, at this early stage of development, the two approaches tend to be on separate journeys because they are so different. But, both approaches ultimately seem to run into dead-ends. Therefore, many scientists believe that the greatest advances will come from the interplay of the two approaches.
Political systems are complex systems and are therefore no less subject to these different perspectives than other complex systems. This is very much true of the study of "democratic political systems." Here to, it is possible that drawing on both will produce insights that we would not have with one alone. It is even possible that the bifocal view might provide the key to the seemingly intractable problems that we encounter in one or other of the perspectives alone. The traditional approach to democratic theory looks from the top-down at components of the system. The first premise of the system was that people should govern themselves while at the same time it was taken as a given they cannot. Consequently, it was simply assumed that the functions of decision-making had to be delegated to an entity that would represent the people.
To prevent the tyranny of government or the tyranny of the majority, the Constitution, as the blueprint of American democracy separated in law the function of the policy-making entity from the function of executing it and of enforcing it. As the system evolved and decision-making became more complex, other institutions emerged, including political parties and special interest groups, each having some independent influence over policy making and government actions. More recently, because of the explosive growth and power of corporations and their political power, special interest groups were divided into commercial interest groups and public interest groups.
The theory of democracy that emerged maintained that this collection of power centers gave people an indirect voice in matters of policy while the balance of power among the entities ensured stability and freedom from tyranny. Like all perspectives, this one contained within itself a definition of what is possible - in this case how much people could participate in decision-making. Reinforced by prejudices and vested self interests, the picture of reality that this perspective created became a self-fulfilling prophesy that endures even today, even while it does not seem to be working. Only a scattering of theorists in the West dared to argue that participation should be (discretely) extended, as in worker self-management, but they still accepted the basic principles of representative government and countervailing powers and foresaw no fundamental change. Others rested there hopes for better quality participation on the schools, even though it is well known that school systems tend to mirror their communities. Only the followers of Marx and Lenin viewed representative government as a stage in history that would wither away with the State. The fact that the state is still with us is taken by many as confirmation of the structural perspective and the theory of representative democracy and countervailing powers. Like the poor, the representative system will always be with us.
The founders of American democracy set up the system to avoid the abuse of power by monarchs or by aristocracies and they were determined to prevent the abuse of the government under the American representative system. Today many believe that in spite of these safeguards, power has become highly concentrated in the hands of a small elite that uses power to its own ends at the expense of the public. In the logic of the countervailing power system, the solution requires an independent source of power to redress the balance of power in the system. It must be independent of the system because vested interests in the status quo would prevent it from acting as a check and balance. The only independent source of power resides in the people. That is, the remedy for a power imbalance in an ailing democracy is more democracy. Unfortunately, the perspective itself appears to prohibit that solution. Supporters still argue that people are not qualified to handle the complex issues of today and that even if they were, it would not be technically possible for them to deliberate and vote on issues.
Thomas Kuhn's theory of normal and revolutionary science may be informative here. That is, maybe the structural perspective on democracy had exceeded its ability to explain what is happening in the American political system in response to changing conditions and like the Ptolemaic universe giving way to the Copernican universe the top down perspective of democracy should make room for to a new perspective. Perhaps, that new perspective lies in the bottom-up view of complex systems. Consider this scenario. Suppose five peers deliberated public issues of personal concern and each person discussed the personally most salient issue with at least one person in their extended circle of 150 friends and personal acquaintances, and each of these people deliberated their personally most salient issue with four or five close family and friends and each of those people connected with one or more people in their circle of 150 contacts, and suppose that each of these levels interacted with a higher level of even greater connectivity, then millions of people would have been involved in a process of identifying the most salient public issues and the positions on which there was consensus. All that would be necessary is and executive to carry out the will of the people and a judiciary to enforce the will of the people.
Obviously, connectivity and scalability of this magnitude is impossible and the whole thing is a pipe dream worthy of a modern-day Rousseau. Or is it really such a pipe dream? Certainly, it was before the Internet and high tech information tools. But, given the Internet and a new generation of information scientists and social network and connectivity geeks who know the technical wizardries and potentials, every piece of the scenario above is theoretically possible. These 21st century visionaries see democracy as a self-organizing succession of lower to higher levels of connectivity similar to other complex systems and call it "emergent democracy."
Perhaps the most exciting thing about this new perspective is that there are no obvious upper limits on citizen participation - at least in theory. In sharp contrast, as long as the democratic system is viewed as a representative system, it logically follows that direct participation of citizens is sharply restricted, As far as we know, there is nothing built into the emergence approach that makes direct democracy logically impossible or that an emergent democracy could not in fact replace representative democracy. But, emergent democracy is really still in its infancy and very little attention has been given real world limitations and how emergent democracy connects to the real world. Some would say it is fun, as far as it goes. But, it is not grounded in reality and certainly not ready for applications.
But, it seems that history, as it has a want to do, is proving this wrong. After much of the initial theoretical work on emergent democracy was done in the early 90's, something happened in American politics that will dramatically support the work on emergent democracy. This event also may have changed American politics forever. Seemingly out of nowhere, a bunch of Internet powered public advocacy groups emerged in response to the attempt to remove President Clinton from office and the cyber group movement grew in response to the impending war in Iraq, the 2000 Congressional Elections, rampant corporate corruption, media consolidation and violations of trust, the recall of Governor Davis, and the issue of re-districting in Texas. "Emergence" is precisely the right word to describe how these groups came about because no individual in these groups was conscious of being involved in some manifestation of little emergent democracies - which is exactly what they are. Consequently, emergent democracy as a bottom up process is not only a theory but a reality taking place in a rapidly growing number of real political entities.
Still, these "ad-hoc" groups do not of themselves make a system nor do they of themselves make for a coherent theory. For this we have to resurrect the top-down power balance approach. These groups taken collectively make up that independent, self-organizing force needed in American democracy to check the excessive and destructive influence of corporate, media, and government elites. In other words, collections of these little emergent democracies might be able to balance the centers of powers in the American political system and thereby save it from tumbling into self-destruction or worse. In this important sense, the collection of emergent democracies is the connection between the top-down model and the bottom-up model of American democracy. While the emergence approach frees the power approach of its built in constraints on citizen participation, the power approach grounds the emergence approach in the familiar reality of institutions and forces and outcomes. In this important sense the relationship between the perspective is a good example of theoretical synergy.
Besides providing the missing balancing element in the power model, there is much more that the emergence approach provides in our study of American democracy. It will tell us how the little emergent democracies like Moveon work and how they are different from traditional interest groups. It will tell us how leadership in those entities is different from traditional groups and it will tell us about the real world limitations on those entities and what their futures may bring. Also, of great importance to our study, the emergence approach will tell us how to evaluate those entities and how to measure performance and other properties relevant to democracy in transition.
This analysis suggests that studies of a democracy in transitions should use a framework that combines the strengths of both the power and emergence perspectives. That framework will retain the reality of representative government while acknowledging the need for greater participation. The exciting thing about the emergence approach side is that it leaves open the possibilities of progressively increasing participation, even to the point that much of public decision-making is done by the people. In this sense, our democracy is truly a "democracy in transition.
A few implications to make the point:
1. We should identify how and to what extent Moveon and other like groups in fact behave like the models of emergent democracies. Differences may suggest things that Moveon my try or they may suggest better models of emergence.
2. One of the above properties is the role of leadership. I know for a fact that Moveon leadership is very like the description of leadership in Ito's paper. For example, Ito says that the "role of the leader is not about determining the direction and controlling the followers, but about maintaining integrity, representing the will of the followers and influencing and communicating with peers and leaders above."
3. We should determine how issue deliberation and actions reach higher levels of connectivity or beyond Moveon. Is it a conscious act, as in alliances, coalitions, and joint efforts - or is it somehow self-organizing as emergence theory might imply?
4. What can we learn from emergence theory and Web technology that would help us measure the development of Moveon, in empirical terms such as in hits and links? What guidance can we get from theory alone?
5. What can we learn from the above about the potential futures of Moveon? What are the theoretical limits of connectivity? How is the future of emergent democracies tied to events in the real world? How is it informed by the power perspective?
We should keep in mind that this line of research is somewhat separate from our other empirical research efforts, which involve political outcomes predicted by the power balancing approach, such as the success of a candidate that was not created by the establishment or events like reversals of the media or government decisions to fire whistle blowers, or the stopping of legislation that is not in the public interest or the passing of legislation that is against the will of special interests.
The Significance of the Bottom-Up Perspective
The bottom-Up perspective provides an answer to a very important question - what is it about moveon-like groups that make them so unique and different from other public interest and special interest groups and anything in the past? The answer is that to an important extent Moveon groups are self-organizing little emergent democracies. One of our first lines of empirical inquiry should be to determine to what extent the various groups are bottom-up and draw out the implications. How much is the agenda defined by Moveon members and how much by leaders/ To the extent that leaders determine the agenda is this another form of representative democracy? Do members say, "Well my views are not represented in Congress or the Executive so I will join a group in which my views are represented?
Bottom-Up as a Framework for Understanding Versus Intervening: The Invisible Hand Versus Planned Change
Bottom-up emergence is a model for explaining complex systems, but as the role of leadership in emergent democracies implies, it is also a model for planned change. Human beings are not ants and all complex systems are not completely "natural" and self-organizing. Market forces operate from bottom up, but they assume complete, accurate, and accessible information - which is never the case. Also, efficient allocation of resources may not be the only social goal. Therefore, we intervene using the process as a base. In the case of the market, it is supply and demand. In emergent democracy, it is social nets, connectivity, and talk (deliberation).
Another interesting example of bottom-up organizing was a project in California involving an in-home support services program for the elderly and disabled. The current program was top-down, meaning that some program heads in Sacramento sent down instructions to social workers exactly how many minutes to give to washing dishes, bathing, meal preparation, and so forth. The role of supervisors was to enforce these directives. The problem was that this top-down process did not take into consideration the characteristics of each client, which the social workers knew best. Consequently, the workers would give their clients what they need and then back-in the forms assigning minutes for this and that. The result was that awards for clients with the same characteristics would often vary greatly and awards for clients with different characteristics would often be the same. A group from the University of California at Berkeley proposed a unique bottom-up solution called a worker decision support system (to distinguish it from management information systems). At a State Capitol hearing to get financial support for a demonstration, a high official said with tongue-in-cheek that he would put all of the social workers in a room and have each of them write down what he/she would give in hours to each client, discussion each case and why there were differences in the amounts among them selves, poll the workers again, and then take the averages.
The team from the University of California, returned "Thank you Sir, that's exactly what we will accomplish with a decentralized network of computers for social workers. First, they would construct an algorithm that would predict the hours that the social workers would give on average from survey data. Subsequently, social workers would enter the characteristics of a client and get a prediction based on the model - which they could use or not use. If not in agreement with the prediction, workers were are asked to explain why. When significant departures were encountered, supervisors knew that it was either because the worker was inexperienced and needed training or was simply incompetent or they had weighed important characteristics that were not in the algorithm - at which time those new characteristics would be added to the algorithm for future cases. After several months, when a worker used the predicted hours, he/she new that this number would be the number that over 98 percent of all of his/her co-workers would have given that client.
The point of this example is that there was a bottom-up process going on, but it was deliberate and man-made. In fact, it could not have happened without intervention. Nevertheless, there is a tension between the "natural" and the "intervention" models when it comes to emergent democracy. Moveon and like groups do not seem to fit comfortably into Ito's three-stage model of emergence. Are there ways in which Moveon fits - or is the Ito model the wrong social network model for the Moveon phenomenon? It seems like the Ito model might be applicable to issue identification, but not to action - and not to action groups. If the Ito model is not the right social net model, what is that better model?
The equity case above also points out the difference in leadership roles in bottom-up approaches. In a top-down system, the role of supervisors is to enforce the top-down "Thou shalt' rules. In a bottom-up system, the leader's role is to facilitate the process. We already know enough about moveon to know that a core principle for the leaders is not to tell members or other participants what to do, but to facilitate what they want to do. Through an Issue Forum and more informal processes, the leaders do get information on what is of most concern to the members and in what order. However, one of the objectives of Moveon is to detect and move quickly on urgent issues. In those cases (which in troubled times are many), the leaders tend to identify the issues, propose actions, and then seek members' responses. An interesting question we should ask is whether there are examples of cases where the leaders proposed, the membership response was cool or luke warm, and the leaders backed off? To what extent do Moveon leaders act in the more traditional role of persuasion rather than facilitation? In any case, they are often not performing a pure "facilitator" role and the group is not acting in some classic emergence sense. To the extent this might be the norm, are groups like Moveon playing a surrogate representative role? The "me-too" two-party system, with both parties as captive of big money, leaves the liberal citizen without representation. Is Moveon a way to feel represented? (Would a proportional representation, multiple party system do the same?)
Another set of questions involving leaders should be - if Moveon is playing a representational role, to what extent are the leaders accountable to the membership? To what extent is traditional "organizational democracy," including constitutions and elections important? Is it enough that members are free to leave and go to another group?
To Comment Click Here
(Top)