Navigating Bounded and Unbounded Spaces.
Unique to our age is the massive scale at which we are applying science and technology to the construction of our physical, social, and cultural reality. Our prowess in this construction game is so complete that it has altered the fundamental relationships between human beings, the rest of nature, and the artifacts that we create. As we marvel at the technological side we may overlook the human side: the fact that we are, for the first time in history, making it possible for all human beings on this planet to interact with each other, and that we have the power to shape the nature of that interaction.
Our next evolutionary step may have less to do with the construction of cognitive technologies of the computer-based kind, and more to do with the creation of patterns of consensual human interaction. These are the patterns, modes and moods of interaction that can enable or constrain the creative energy of the human spirit.
As we create useful patterns it will be important to articulate and propagate them across space and time, so that all of humanity, present and future, will be participants in the grand conversation. This is the point at which technology enters the picture. The Internet, the massive deployment of computer-based agents carrying out our wishes and similar technological constructions are mere forerunners of the technology-based socio-cultural infrastructure to come. The prerequisite technological evolution is well on its way; the technological side will take care of itself. It is time for us to pay more attention to evolution of the socio-cultural side.
Herb Simon (1962) speaks of the bounded-ness of human rationality in certain cognitive domains. These are precisely the domains in which computer-based cognitive technologies can play a prosthetic role; and we have seen spectacular advances in these areas. We have been focusing on this aspect of bounded-ness for four decades. Paradoxically, in terms of the evolution of the human socio-cultural enterprise it may be more important for us to articulate the boundedness of the formal world of machines, mechanisms, and a priori state-spaces within which computer-based devices operate. The work of Robert Rosen (1991) has shed some light on how we can formally establish these boundaries. Unfortunately, work in this area has been sparse, and we have not yet developed a conceptual framework in which to relate these two sets of bounded-ness. A failure to do so may place arbitrary constraints upon nature and humanity.
In this brief paper I offer a metaphoric look at modes of interaction within different domains. I will do so by describing different types of communicative or interactive protocols -- protocols that admit interactions between human beings, humans and artifacts, and merely artifacts.
Ideally, such discussion should be tangible and formal. I can only offer an abstract, metaphorical view of the situation by discussing maps, territories, and means of navigation. Consider two friends, one well versed in the intricacies of traveling by car -- let's call him the `driver'; the other with a fondness for exploring rivers while fishing -- let's call her the `angler'. Each person has learned to use maps to guide them in their respective domains. The driver is thoroughly familiar with all symbols to be found on a roadmap, knows all of the permissible ways in which the symbols may be combined to depict the routes to be traveled, and can also construct new maps to describe permissible paths for previously uncharted travels. Similarly the angler is competent in the use of maps depicting the topography of the terrain, the location of trails, camps, fishing holes and the various aspects of the outdoors to which one pays attention when one is fishing. We can expect the two map-making conventions to have a lot in common; however, their use may differ in significant ways.
Now imagine the two friends taking fishing trips together. After agreeing upon a particular destination the driver can employ roadmaps to guide him to the desired area; in fact he can add refinements to the map by describing uncharted roads taken on the way. Upon arrival, the angler takes over, and the trip continues based on the protocol of topographic mapping with symbols depicting aspects of interest for fishing. In the hands of the driver the roadmap defines a certain protocol for (inter)action; similarly in the hands of the angler the fishing-map serves the same function.
In this scenario we are looking at what may be called an analytic protocol. In more formal mathematical terms, the navigation is defined in terms of `direct products'. We agree upon an a priori set of symbols and a set of relationships permitted between/among that set of symbols. The repeated application of permissible operations upon the symbols yields a path through the state-space. The driver and angler each function in the state-spaces with which they are familiar, and have (complete) command of the permissible operations; in a sense their respective worlds seem to be closed under the operations with which they are familiar.
We should note that in the domain of the driver the map and the territory are in close correspondence; form, and function are easy to relate, as are the available courses of action and the consequences of such action. Such correspondence should not be surprising; after all, the roads and maps are both artifacts. The number of roads and the number of maps is in fact rather large, and the navigational problem-space can be very complicated; however, the situation is so well defined that we can express it in terms of effectively computable procedures and reduce it to matters of syntax. The legitimacy of doing so is demonstrated by the success of computer-based vehicle navigational systems. As technology advances we can increase the size of the statespaces. The ultimate size of the (a priori) statespaces can be so immense that for all practical purposes we can consider them to be unbounded.
In the domain of the angler the correspondences between map, territory, form, function, courses of action and results are more problematic. Nevertheless, for a competent angler navigation seems so easy as to be in the same category with that of the driver. However, the angler is (inter)acting in the domain of the living; this world is complex, not merely complicated. For now we retain the analytic protocol label for both types of navigation; keeping in mind that in the fishing domain the map is an artifact and the territory is a living system; consequently, the analytic protocol may be inadequate since the situation is unbounded in a way that is fundamentally different from that which is described in the previous paragraph. We shall examine this more fully in a moment.
Let us now turn to a more interesting case. Imagine that the friends have taken a number of trips together and have seen each other's map-using protocols. The driver now decides to take fishing trips. He is not yet an expert at fishing, must take the fishing-map at face value, and may not appreciate the full significance of all things on the map and all things on the river. Nevertheless, a number of driver-domain symbols and operations can be applied in the fishing domain, and (at least on the surface) the fishing symbols and operations have `meaning'. The driver is able to navigate and fish in a rudimentary fashion by mimicking the (inter)actions of the angler.
This is essentially a synthetic protocol. In more formal mathematical terms, we are appealing to `direct sums', as we construct a state-space based on surface features of the situation. Fully meaningful navigation is not yet possible since we are often mimicking (following) form without understanding function; however, we can now navigate over a wider terrain. Our state-space has been enlarged, but some states and transitions within that space are hidden from us or are not well defined. In effect we are fishing from the driver's perspective, or perhaps taking that which is common to both domains and applying it in the new domain being explored. It should be possible to describe a synthetic protocol formally enough to delegate the navigational task to computer-based agents; however, such navigation in the domain of the angler is problematic, and can lead to unintended side effects.
The problem arises because we are using a synthetic protocol in a complex domain. This would not be the case if the new domain was merely complicated. For example, the angler, if she was previously unfamiliar with the rules of travelling by car, could expect to navigate reasonably well, even after only brief exposure to the rules of the road. Actually, in the world of artifacts, where map and territory have a well-defined correspondence, the synthetic protocol can be used to combine domains that can be subsequently navigated analytically, and analytic descriptions of distinct domains can be combined to yield descriptions of the composite domain. This is a most useful property -- one of the bases of our technological prowess, and it gets us into trouble only when we attribute it to domains in which it does not apply.
The most interesting case involves the two friends taking repeated trips together. During such trips they become immersed in each other's world of driving and fishing. At this point we find the full interaction of two life histories and experiences as new consensual experiences are generated on each trip. During such trips the friends may find that they pay attention to different aspects of reality, in both the driving and fishing worlds. They may also find that in order to record descriptions of these experiences (on a map) they have to create new symbols and even new rules for combining symbols. The `meaning' of the symbols may become problematic in the sense that the experience(s) that gave rise to the symbols may (will) not be the same as the ones to be generated when the symbols come into play on subsequent occasions.
This new protocol is more difficult to classify. It has both analytic and synthetic aspects. In a sense it is integrative, but not in the sense of bringing together that which pre-existed in an integrable form. I would call this a systemic protocol; it has to do with consensual creation of the future. Note that once the new symbols and interaction rules are in place, once the new state-space has been created, we (and others) can rely on an analytic protocol for (reliable) navigation, or a synthetic protocol for (limited) exploration.
Let us return for a moment to the case of the angler on a fishing trip. We tentatively labeled this protocol analytic. In fact there is more at work here than can be accounted for by this protocol. The appropriate protocol is actually the systemic one, with the requirement of repeated (continuous) interaction between the angler and nature. In the absence of such interaction the angler would find that some of the features on the map would no longer apply as some aspects of form as well as behavior have changed. This is not change in the chaos-theoretic sense of selection between alternative admissible states; this is a new interplay of form and function. In Rosen's (1991) terms we are witnessing the continuously renewed interplay of material cause and efficient cause. This is the sense in which living systems are unbounded. This unboundedness is characterized from a different perspective by Maturana and Mpodozis (1992) as `the recursive interactions between living systems and medium as a process that necessarily flows in a continuous co-drifting that involves for each living system at every moment all the dimensions of its domain of existence while at the same time each living system operates as part of the medium of others'.
The world of living systems and the world of artifacts are each bounded (and unbounded) in distinct but complementary ways. It seems clear to me that we should employ systemic protocols to guide the interaction between human beings, as well as between human beings and nature. It also seems clear that when we deploy artifacts to mediate the interaction between human beings we may shift the mode of interaction to the analytic or synthetic. Finally, I maintain that interactions between artifacts are limited to analytic or synthetic protocols.
The manner in which we position the nearly unbounded a priori state-spaces of the world of artifacts in relation to the world of living systems has everything to do with the future unboundedness of nature and humanity. There is no need for either side to become a constraint upon the other as long as the design process continues in a systemic fashion. We are just beginning to develop the world-wide technology-based socio-cultural infrastructure that will at once enable and constrain our lives. It is my hope that these brief notes and insights will stimulate discussion on ways to shape that development.
Correspondence to: Bela Antal Banathy, International Systems Institute, Salinas, CA, USA.
International Systems Institute and Saybrook Graduate School, Salinas, California, USA
Maturana HR, Mpodozis YJ. 1992. Origen de Las Especis por Medio de la Deriva Natural. Publicacion ocasional no. 46/1992, Museo Nacional de Historia Natural, Santiago de Chile, Chile. (English translation published in 1999, cited in Bunnell P. 2000. Attributing Nature with Justification, in this issue).
Rosen R. 1991. Life Itself: A Comprehensive Inquiry into the Nature, Origin, and Fabrication of Life. Columbia University Press: New York.
Simon HA. 1962. The Architecture of Complexity. Proceedings of the American Philosophical Society 106.
|Printer friendly Cite/link Email Feedback|
|Author:||Banathy, Bela Antal|
|Publication:||Systems Research and Behavioral Science|
|Date:||Sep 1, 2000|
|Previous Article:||Attributing Nature with Justifications.|