3.1 What I mean by a system

I take a system to be a recognizable set of parts that dissipates energy through participation of those parts in mutual collective (organized) behavior. Parts can be systems (subsystems). The technosphere is a system of systems. Other examples of systems include leaves, clocks, humans, rivers, cities, power grids, and the internet, some of which happen to be parts of the technosphere. Systems as described here are “dynamical” systems, i.e, systems that do something.

This organization-based definition also requires that a system persist through many internal “time cycles”, so that the parent system, as well as each part (if it happens to be a sub-system) has time to express its collective properties. Thus a dog expresses characteristic canine behavior over times that are long compared to the time between its heartbeats, and the time taken by a smartphone to send or receive a text-message is long compared to the typical transistor switching times of its logic gates.

To identify a specific system it must be recognizable and describable. The length of description appropriate to a system with many parts depends on what level of detail is of interest. Different levels of description are resolved by “coarse-graining” the system. For example, at the upper, or least resolved, level the technosphere can be defined as the global, interlinked collection of humans and technological artifacts that metabolizes fossil fuels and other sources of energy to power communication, industrial, governmental, military, financial and the other interconnected bureaucracies and institutions on whose function modern civilization and society depend.  Similarly, a specific computer might be described with low resolution as an electronic device that stores and manipulates digital information.

These broad descriptions can of course always be broken down to finer levels of coarse-graining, generally with shorter time constants, so, for example, the technosphere includes components such as cities, factories, high schools, people, cars, and computers, on down to picture frames, books, and buttons, and finally, at micro-scale, the technosphere contains hyper-abundant components like nano-fibers and human red blood cells. In a similar way the computer might be parsed into its central processing unit, random-access memory, power supply, visual display, and so on, all the way down to its chips, wires, and transistors.

A system as defined here: Has parts, does something, has a brief description, and persists through many internal time cycles

For present purposes what is important is that the overall behavior of a system can be described briefly. A brief description at the coarsest level is possible because the system is organized, i.e., the parts do not act independently, but in coordination with one another. With many system parts acting in concert, it is unnecessary to include all their individual behaviors in a top-level description, which can therefore be short.

Thus, even if one lacks knowledge of the specifics about what an entity does, but where organization and use of energy are manifest, it can still be classified as a system. This is because dissipation of energy by collective activity of its parts is the most basic action expressed by a dynamic system. Examples might include a strange organism wiggling under a rock, or a piece of unfamiliar machinery humming away by the roadside. These two coarse-grained observations would be sufficient to classify with some confidence the organism and machine as systems.

Looked at the other way around, the requirement that a system have a short top-level description implies limitations on the behaviors available to its parts. These are limitations intrinsic to the state of being organized. So dynamical systems “automatically” exhibit built-in structural tendencies, which can be expressed in terms of rules or constraints. This observation is key to the present study of the technosphere.

Systems also come with constraints or requirements that are imposed by the laws of physics. For example the law of conservation of energy implies that the rate of energy inflow minus the rate of outflow equals the rate at which energy accumulates in the system. Useful formal rules of organization, together with certain rules arising from basic physics, will be treated later in the this chapter.

It is not hard to see that the descriptions given above for the technosphere and the computer are each incomplete. They are incomplete because, for example, the descriptions do not explain all the terms they contain, such as “function” and “modern civilization” in the case of the technosphere, or “electronic device” and “manipulates” in the case of the computer. Incompleteness of description is a consequence of coarse-graining, and is a virtue in the sense that it suppresses reference to vastly numerous but irrelevant small-scale details of system behavior.

More than a virtue, incompleteness is a necessity. Humans could not survive in the world if their existence required access to complete information about their surroundings. To use a computer, no knowledge of anything inside the box is required, and, similarly, human interactions with other systems that constitute their environment, such as banks and bridges, are based on, and usually require, very little knowledge about the details of how those systems work. Insensitivity to the corpus of information potentially available in their environment is generally true of all dynamical systems, not just of humans, and is reflected in a fundamental organizational rule (rule of reciprocity) to be discussed in a subsequent post.

System descriptions are also subjective. I could have couched my description of the technosphere and the computer in different terms, but chose the particular ones given, such as “function” and “electronic device”, because it seemed to me that most readers would have enough knowledge and experience to recognize these choices and understand their use in the description of the two systems. The coarse-grained description of a system is subjective, and necessarily so, because it depends on a shared understanding between interested parties of the terms of description, which depends on the knowledge that each person has.

Subjectivity may seem to invalidate the goal of a scientific, non-anthropocentric approach to analyzing the technosphere, but acceptance of a shared-knowledge background is an intrinsic feature of any kind of scientific investigation, and the present study is not an exception. In the end, to sustain a scientific approach, one’s conclusions must only be consistent with the observable world and open to modification or elimination as the result of such a comparison. At the end of this essay, I will confront the concept of the technosphere with a series of critical questions, which I will then endeavor to answer. In the meantime feedback from interested readers would be helpful.

Certain other aspects of system definition might merit further discussion as well. For example, defining a system through its top-level (coarse-grained) description tends to leave some ambiguity around the edges. For example, if the Earth’s atmosphere is defined as its circulating envelope of gases, do the submerged bubbles of air in a surf zone count as atmospheric parts? If highway transportation systems, including cars, are part of the technosphere, is the disabled automobile on the side of the highway also a technospheric part?

One might also wonder whether once it enters a system, the material source for the system’s energy (e.g., fossil fuel in the tank or food in the stomach) becomes a component of the system that consumed it? Similarly, are a system’s waste products a part of that system before they are expelled to the external environment?

Finally, my definition of the technosphere includes humans among its parts. But humans, being animals, are almost universally considered to be parts of the biosphere. Do the biosphere and the technosphere share parts? Or are humans today (as I will argue) more logically viewed as intrinsic parts of the technosphere than parts of the biosphere? I will try to sort out some of these questions later in this essay.

Further reading
Accessible discussion of coarse-graining and many other topics in complex systems science, in Murray Gell-Mann’s The Quark and the Jaguar, Freeman, 1994.

Discussion of the technosphere as Earth’s newest sphere, P. K. Haff, Technology as a geological phenomenon: implications for human well-being, in A Stratigraphical Basis for the Anthropocene, 2014, Waters, C.N., Zalasiewicz, J.A., Williams, M., Ellis, M.A., Snelling, A.M. (eds). Geological Society, London, Special Publications, Volume 395, pages 301-309.

Persistent citation for this post: P. K. Haff, 3.1 What I mean by a system, in Being Human in the Anthropocene blog, 2018. https://perma.cc/C8LE-UGAW.

Next up: 3.2 Behavior versus origin of the technosphere. Emphasizing that at the system level my focus is on the technosphere’s dynamical limitations and imperatives, not its origin or “evolution”.

2 thoughts on “3.1 What I mean by a system

  1. When reading this entry, I was intrigued about your use of the term ‘subjectivity’. I think this is the point of entry for adding an aspect that I feel is somewhat marginalized, but clearly implicit to your argument. Every system is embodied information, and this is because every system is ‘subjective’: That means, in processing information about its environment, it focuses on information that is relevant for realizing the functions necessary for maintaining its operations. In my own work, I treat this as an essential element of the theory, since we can ask what is the most general function of systems: Well, that’s mobilizing and dissipating energy. If this is most general function, then energy and information are closely aligned concepts, which is very clear when you consider origin of life theories that build on thermodynamics. I think that this also holds for the entire ‘tree of life’, as envisaged in general theories about evolution as proposed by Eric Chaisson. The important consequence is that if you want to understand a system such as the technosphere, there are two positions of observers: The internal and the external. A system can be observed ‘from within’ and ‘from without’. If we humans are part of the technosphere, we are in a very difficult epistemic position, because we need to disentangle the two perspectives. This ties up with your previous post discussing agency. Our own subjectivity may be shaped by implicitly ‘seeing’ the technosphere as mediated by the functions that the technosphere also imposes on us. This is a form of ‘subjectivity’ that may be turned in ‘objective’ analysis if we achieve an external position: That is what you are trying to do. I apologize for always ending up with tricky philosophical questions!

    1. Thanks for the comment. Yes I think that by adopting an external perspective we go as far as we can in analyzing the technosphere objectively, certainly further than if we take what you call an inside view. And there is nothing contradictory in an nonanthropocentric view of anthropocentric behavior. Still, external analysis is not subjectivity-free, for the reasons given in the post. Defining what the system is that one is talking about, is, one could say, half the problem of analyzing the system, because it requires buy-in from whomever one is talking to. But if you try to define a sufficiently complex system, you soon find that you have to start making subjective choices (or maybe this is a circular argument–a definition of complexity, that a complex system is one with a high and irreducible level of subjectivity–but I am not going down that road), even though, once the definition has been made, an objective analysis can be brought to bear.

      And yes agreed that energy dissipation and information/entropy are closely connected, as orderly (low entropy) systems are generally much more efficient at “using” (dissipating) energy than less organized structures. I’ll try to touch on this when/if I get around to discussing transport processes in the technosphere.

      [6-8-18 Re the above remark on “circular argument”, see See Dominique Chu’s discussion of the problem of defining a complex system: “Complexity: against systems”, Theory in Biosciences, 2011, volume 130, pages 229–245].

Leave a Reply