What Evolves?
What can we develop and what evolves beyond our reach?
The difference between Development and Evolution
Hardin [The Tragedy of the Commons, 1968] (the TOTC) created a parable that says a great deal about our relation to the ecology. The ecology is literally our concept of what we live in. Our concepts—learned by long experience in getting what we want—are less than the complex endowment from which we take. Hardin turned that into an example of sheepherders who exploited in common a field of grass. They could only try to get more of what they wanted (more sheep) but they all got what no one wanted: An exhausted, over-grazed field.
We cannot answer where the field came from—although the idea of an abandoned commons is tied to industrial development in England—but the social system of herders illustrates very well the difference between individual development and a social form that evolves jointly, unintended, accountable to no individual, and wanted by no one. Regarding our governing State, our motto E Pluribus Unum summarizes the whole idea.
The TOTC is about the difference between a level of behavior and the formal level of that behavior. This pairing of focal levels (that we identify our behavior with) and a formal level (the shared “field” and rules of behavior) is the recursive pairing of all “complex” systems.
Biology is a classic model of complexity. We got the terms “evolve “ and develop” strictly from biology. And that is the problem with any further conceptual application to our eco-logy. The genomic form (a species) evolves to add or subtract new forms that interact in the ecology. Each form is represented (and reproduced) at the focal level of interaction by its instances (the organisms). The organisms develop, mature and die while the form persists.
A species (genome) evolves as a new form of organism development. That is the biological case but NOT the general case for evolution of and in the ecology.
Biology also has a crossover of evolutionary development (evo-devo) that is about how the realization of the organisms from the genome occurs. But the set of biological and other objects interact so that another formal level evolves that we conceive of as the ecology. The subject here is how we manage that ecology. The TOTC warns that we persistently focus on our selfish wants and ignore what formally evolves. We use social organization (governance) to amplify projects of “development” and that only strengthens the formal effects that evolve beyond our intention and accountability. We add the State as a level of social form to the ecology beyond our biological evolution and development. How do we cope with complexity as we socially extend and intensify our development by exploiting the evolved form of the ecology? That is the subject of Limits to Mobility (LTM) with the emphasis on the development of mobility by the State, with technology, that extends our exploitation of urban space and becomes a TOTC.
1980’s: The decade of Complex Systems
The TOTC of 1968 came with the recognition that our global ecology was being altered beyond the intention and accountability of individuals. Can we socially advance to managing, regulating and preserving the formal level of our shared existence with other genomes?
My interests started a few years later and specifically with our urban ecology, the part closest to us and the form of most of our interactions. The first draft of LTM was started in 1975. I still had a long way to develop and in the 1980’s I was working as a systems engineer with what we called complex systems, that may have been artificial but also ecologies as emphasized in a book by a friend of mine in Troy [Atsushi Akera, Calculating a Natural World: Scientists, Engineers, and Computers During the Rise of U.S. Cold War, MIT Press, 2006].
Coping with these ecologies was a subject of our Mitre Nonlinear Systems group. In the wider DC area was the Washington Evolutionary Systems Society (WESS) that brought together national and international workers in complexity. Later I also attended conferences of the New England Complex Systems Institute (NECSI). By that plus long dialog with Stan Salthe I developed some appreciation of complex systems. They are also evolutionary and nonequilibrium/nonlinear. These are all aspects that take us away from simpler concepts of the ecology that we learn by a long history of local actions in a persistent-equilibrium form. As long as the ecology, at any level above our actions persists despite our actions we are more sure of what to do. There is less risk and that is a condition for the persistence of any system through persistence of a form of learning by the parts. When form changes (evolves) the contained level of learning is obsolete.
Previous posts have been about the collision of governance with complexity. Can we get beyond the myopic focus of development by “higher” levels of social organization or does that just enlarge our selfish wants into the Drill Baby? Current results are not hopeful. But LTM was concerned with the documented history of “urban removal” projects—especially highway—that were clear examples of the unaccountable destruction of the urban form. LTM is all about the evolution of the transit-oriented development (TOD) form into auto-oriented sprawl (AOS). That is a true formal change that altered the development of every instance of city. It is not “progress”. The history of all the projects and developments shows how little their ecology is and how unaccountable they are for the “unintended” formal change. We are deluded by the TOTC we call “progress” that is actually more risk and inequity.
We are in the TOTC of our past indifference. We look for easy political solutions to problems that seem to be piling up and polarize according to our willingness to listen to a demagogue/autocrat who offers we He Wants as easy solutions. We should go back to just getting what we want.
A system (an ecology) is a network and so a form that contains interacting parts. And right there is a paradox of how the “whole thing” has a simple description (e.g., the “highway network” or an “urban place”) and then we need many bits to describe everything in that object. Complexity has inseparable levels of scale.
Readers of previous posts will understand that this paradox is resolved by the scale hierarchy model of complexity. There is more about that in Limits to Mobility (LTM) Part I but I owe the concepts to my tutelage with Stan Salthe {e.g., Stanley N. Salthe, Evolving Hierarchical Systems, Columbia University Press, 1985]. I found Salthe’s ideas about 1987 when I was working on a research project about scales of weather information. That was about winds affecting flight trajectories and so I stumbled into the vexed issues of turbulence. Complex, evolutionary systems have also been characterized as “generalized turbulence” [G. Nicolis and I. Prigogine, Fluctuations in Nonequilibrium Systems, Proc. Nat. Acad. of Science, 68(9), Sept. 1971: 2102-2107].
Then about 1989 I encountered the scale hierarchy in another way concerning the management scales of the airspace system [more in my memoir AERA 3 and the NAS Plan]. When humans logically distinguish scales and try to manage them, governance evolves as social aggregations. We get the State. The ecology (what we live in) has levels of complexity that evolve with interaction of the contained objects (called emergence). But we are liable to the fallacy that our social groupings are just a bigger “Us”.
We develop in a life cycle in the ecology. We are as complex as what can be constructed from our genomic form and our social networks are yet more complex. The ecology that includes our social networks is then more general and complex than biology. Even as levels of governance evolve as our social organization they are challenged by the ecological levels of complexity whether looking into their constituent parts or out to what we all live in. This is manageable only by mapping level to level as the relative simplicity of formal information rather than the contained information. The Constitution already had that figured out.
The point of distinguishing development from evolution, and the levels that evolve with developing parts, is to separate what we are or want from the ecological form that contains what we are and want. We need stability of evolved forms to operate in yet we want to exploit and destabilize them. Then we are faced by a paradox: How can we preserve the ecology that evolves? This is a practical question when we mistake progress for exploiting and destabilizing the ecology, even our immediate urban ecology.
To Evolve, or To Develop?
In the 1980’s system engineering concepts also developed. Did they “evolve”? We will consider that. I was pretty much in the center of witness to projects and research dealing with more complex systems. One development was the concept of system “architecture”. While the term had been around [e.g., Herbert Simon, The Architecture of Complexity, 1962] it was surprisingly absent from system engineering doctrine until the 1980’s. We could speak of a “system level specification” but without the formalisms attached later to architectures [e.g., the Department of Defense Architecture Framework, DoDAF, that was actively in use ca. 2000, and with legislative background from 1996].
The architecture concept was recognition that a bunch of parts needs a formal level to describe what is contained plus the rules of interaction (interface protocols). That had long been implicit in any project that was more than making or modifying an object (craft that becomes design-engineering). But the Internet and its participatory use and development fomented the formal architecture concept, starting then with “Information Technology” (IT). That superseded the previous system engineering doctrine based on building weapons. The SAGE system [developed ca. 1953-58] was the forerunner of all computer-mediated communication and decision systems (ecologies with IT) and so the apparent motivator of the architecture concept and any of the systems-engineering work we were doing by the 1980’s.
What disturbed me at the time was the colloquial shift of how complex systems were discussed. We no longer “developed” a system, which is all engineering does, but perhaps “evolved” them? No. This issue is treated in my paper Organizational Evolution, Life-Cycle Program Design: Essential issues in systems engineering and acquisition of complex systems [2006]. Evolve is not a transitive verb as an efficient action of some agent. What evolves emerges, unintentionally, from interaction.
This semiotic distinction between develop/evolve is not trivial but affects how systems are to be managed. The air traffic control (ATC) system we were developing under the National Airspace System (NAS) Plan was cancelled as a project and that blamed on too much of a Big Bang project. The implication was that we were developing too much in tasks that were too tightly coupled under an evolving technology. I argue that we were not. We were analyzing and integrating many developments (which is what Mitre was chartered for, by Congress). Congress just could not manage the complexity of some sizable architecture that spanned many (and risky) project chunks. And muffing that responsibility was not going to help applying technology. Congress ended up with piecemeal projects driven by the delays in modernizing the NAS. And so on to failure to keep up with technology and so perpetual complaints about “outdated” systems. That was the complaint for which the NAS Plan was devised and it is still the complaint now. Something wrong here?
The main target of LTM critique is how our urban ecology has been managed. The State at best claims a responsibility for “infrastructure” but the perpetual “decay” or obsolescence of that is just evidence of how that simplification-of-complexity strategy has failed. The State lacks an “architecture”. I have argued that the Constitution is just that. But we are now in a casestudy of how confused the State is about that architecture. Is the ecology not a thing for Drill Baby to autocratically develop?
Nature Seeks Stability
Having just read Frank Close’s excellent book Destroyer of Worlds [2025] I am reminded of the principle in physics that “nature seeks stability” by, among many other processes, radioactive decay. “Decay” is a pejorative term, but why? Is it not better to shed some nuclear particles to be more stable and persist longer? Is that an evolution or a development? What about the process inside stars that built up a ladder of “higher” elements, including the ones that make our ecology, even the “naturally radioactive” ones?
We need another scale hierarchy, a logical one, to address these issues. That was nicely provided by Stan Salthe who stated it as:
physics > chemistry > biology > ecology
Actually, I added “ecology” as an improper containment hierarchy. The statement is that physics contains all possibilities specified by chemistry and so on. We go from particles/fields to materials to organisms to networks of materials and organisms. Each level of specification truly evolves. At each level there is realization of only a vanishing fraction of all possibilities allowed it from a previous level. That is what limits prediction as we go down the hierarchy from physical law to trying to predict behavior as in sociology or economics. That whole topic is better addressed in my Prediction Needs Regulation that emphasizes formal invariance.
We have strived to understand the evolution of levels beyond physics. That is the subject of self-organizing systems (SOS). That eco-logy (our logical concept) has been suspended between diffusion back to the un-organized mix of physical parts and interaction within a formal field up to the further levels of organization. The key principle is that since the further levels specify the possibilities of physical law, the particular focal states that develop (the space-time location and interaction of every part) are permutations whose probability shrinks to nothing. That is the basic inconsistency with Boltzmann’s maximum-entropy (most probable formal distribution) model of diffusion. Self-organizing evolution cannot be from just “random” parts and interactions. However, all possible permutations have to be consistent with or “fit” the form. There is a bootstrapping recursion between part-development and form-evolution.
Consider the case of cities. Does their form and development correspond to genomic-organism biology? There are metaphors that result from that analogy. This issue has lately been clouded by findings of “universal scaling laws” for cities [e.g., A stochastic theory of urban metabolism, August 11, 2025]. These suggest that cities have a consistent form (e.g., a homogeneous distribution of size and energy dissipation, or streets and population) as do organisms. That is simply the argument for an evolved urban form that contains specific instances of development, not the genomic model of reproduction. The consistent form is an argument for the self-organization of cities that projects cannot gainsay anymore than improving on biological evolution (the Island of Dr. Moreau parable). We do not plan cities. They develop according to a form of part-icipatory interaction. Participation is just what State projects override.
The concept of “city” (the urban ecology) propagates because there is a form for our interactions. That includes spatial agglomeration seen as the density and access form of places. We are the carriers of the idea and from that cities emerge. But how that happens and whether the form evolves depends on on changes in our behavior, especially from mobility and bandwidth technology. That is why I spent time in Urban Rail and Troy on the question of whether such technologies are endogenous (of development) or exogenous (an evolving form) following Romer’s economic analysis [Paul M. Romer, Endogenous Technological Change, 1990].
Are we defenseless against technologies of consumption that do not fit the ecology? That is the political question for the State and so a test of whether the State is competent to change rather than preserve our form, our “way of life”. There is a system engineering view on that question addressed on the post on the ConOps: Engineers applying technology are not competent to substitute for the risk accountability of operators of the technology.
Is there any correspondence to evolution in urban forms? Yes. LTM is a study of the formal change from TOD to AOS. The change involves the difference in how transport mode and location interact to define access. The automobile did not fit the TOD form. But there was powerful agency by the State in ignoring rail and urban form but promoting projects for developing the auto mode. That leads to the important question of whether any of our infrastructure evolves with the urban ecology or is rather a development that has to take accountability for any instability of the evolved urban form.
The Paradox of Evolution and Stability
The idea that “nature seeks stability” can be traced from physics. The most basic version is thermodynamic equilibration (maximum entropy) that is completely inconsistent with evolution of self-organizing systems (SOS). The complexity of SOS requires the recursion between an evolved but persistent form and the contained parts that interact but fit the form. The biological example makes it clear that the form must persist relative to the ephemeral parts and behaviors. That is an existential condition beyond any “moral” rule of our behavior. But our social rules are part of the form and have to persist over our interactions.
What we see and describe as “the ecology” is both unspecific and persistent. We can share a concept (the literal ecology) only of what persists over a group of us. The “think globally” injunction is about a persistent ecology and implicitly the behavior (act locally) that fits that form. The simplest formal rule is to obey limits of consumption, the point of Limits to Growth [1972]. The form will then maintain itself but we strive to exceed the limits by mis-behavior and projects. Evolution implies change and evolution toward a specificity that, being nonequilibrium, is thermodynamically unstable at least. Ill-fitting behavior can topple the form and limits are the thresholds for that.
We speak of the qualities of ecological persistence as sustainability, adaptability and resilience (SAR). We are confused about what these qualities are and mis-apply them to our exploitation of the ecology (mobility must have SAR!). But an evolving ecology implies a deeper paradox of formal persistence with instability and change. What can our logical-social response to that be? It should at least be risk management by the State. But Drill Baby is hardly up to these “complex” issues.
We have the evident evolution of more complex forms from Darwin and Wallace in the 19th century, although traceable back at least to Lucretius in the first century BCE. Things change formally and not just in the configurations of a given set of parts. The spluttering indignity about genomic evolution at least persists in anti-evolutionary doctrine and so the belief that all parts have stayed (should stay!) formally the same even as the instances develop and die. This has political effect, including the fallacy that the ecology must be immutable and so so we can exploit it as much as our god-given ability allows. We have dominion as the “highest” biological form because we are the image of the form of all creation. We are seeing where that leads, right to Drill Baby and looking backward to the way things should have stayed.
I found “things change” to be the excuse in urban traffic modeling for indifference to the formal change of the transport projects. That is why it is important to travel the long road to understanding SOS and complexity. We started in the 1960’s by recognizing the instability of the ecology and then further advanced on understanding its dynamics by the 1980’s. Meanwhile a lot of ecological exploitation had occurred, and it is not decreasing.
That brings us to the more crucial paradox: Our concern for “the ecology” is of a persistent form, a form that can be preserved and maintained. At a formal level of description, that can only be statistical about what is possible. And yet our concern is for the evolved level of specificity that reaches an instability as if a cone is standing on a point over the thermodynamic pit of primordial soup: Or, urban form over diffusive sprawl. How can we manage that?
The Quality of Formal Persistence and Risk
Our ecology—our concept of what we can exist in—is going to be human-centric. The previous post was about how that tempts us to expand what we want as individuals into social projects and so the TOTC.
The human-centricity also leads to arguments of why preserving the ecology is somehow in our best interests, perhaps a moral imperative. I think that altruism toward other parts of the ecology is less useful than the existential argument that we are an inseparable part of the ecology. It is a form “we” (in our social organization of governance) must fit into and have a responsibility to maintain despite its instability while recognizing evolutionary change that is not managed by us.
There is also something to be said for our evolutionary learning. What genome persists that trashes its own house? But we get into the Red Queen race between what we learned in the past and the extension and intensification, by technology, of our ability to exploit the ecology. We are shrinking the ecology to the home we live in with all mod cons while oblivious to the far-flung supply chain that wrecks other places.
The conclusion of LTM is that the function of the State can/should be defined as risk management with equity. That is a prescription I find consistent with our political form, the Constitution. From current events and working many years around the Federal gummint I know how much that form has been trashed. How did we get to this abysmal pass?
The TOTC shows that formal change cannot be evaluated as an intentional change. That becomes the unaccountability of projects that expand in scale to change urban-ecological form, as discussed in the last post on benefit-cost analysis. We cannot compare the TOD and the AOS but we can ask the State for accountability of the projects that become the larger exploitation of a larger TOTC.
That also touches on the question of whether the technology is endogenous (of our intentional development) or exogenous as a threat to ecological form and so to be defended against. The failure to manage technology-application (e.g., the automobile) becomes the excuse for indifference to ecological risk management.
I will examine these questions further in the next post that will focus on the highway system. That is what the State took as its particular object of development and excuse for eminent domain projects (EDP). But is that not the most basic fallacy, of mistaking an evolved form for a manipulable object of development? And from that fallacy many failures of governance and our ecological TOTC follow, right up to now.