Technological Niche Construction

Technological Development & Niche Construction

Niche construction theory is the idea that in evolutionary biology organisms and their environments principally affect one another. Richard Lewontin expressed the notion of what later became known as ‘niche construction’ in his paper, Gene, Organism, and Environment (1983), as a pair of differential equations, which we will use as the starting point in our investigation of technological development. The equations are:

dOdt=fO, E, and dEdt=g(O, E)

Where O is organism, E is environment, t is time, and f and g are functions. In natural language: both the change in an organism and the environment over time are functions of themselves and one another. What this pair of differential equations expresses is the construction model of evolutionary biology, as opposed to the adaptation model, where either:

dEdt=0 or dEdt=g(E)

The difference is that in the construction model, both the organism and environment are functions of one another, which is to say that they are reciprocally bound into a single, coupled system. An example of niche construction is the activity of beavers, which, by the construction of dams, drastically increases the heterogeneity of the ecosystem by creating a large wetland. This ‘artificial’ wetland is then inherited by the offspring of the beavers and reinforces certain selection pressures that were selected by the beavers, so to speak. According to the adaptation model, on the other hand, the environment is seen as an independent and autonomous term that exerts selection pressures upon the organism from the outside in. An example of simple adaptation may be the long necks of giraffes, adapted for the tall trees of their environment. This second model has been the default methodological assumption of modern evolutionary biology since Darwin, though it must be noted that this is a convenient simplification of Darwin’s model, since Darwin often described the effects of his organisms on the environment (see Darwin, The Formation of Vegetable Mould Through the Action of Worms).

With human technological activity, however, it is not at all a question whether we have an impact on our environment, and whether the environment in return has an impact on our activity: We have a kinship with the beaver in our construction of dams, for instance. Looking at the construction model, we see that the change of the organism through time is a function both of itself and its environment, while the environment is likewise a function of the organism and itself. Since the environment is recognized as playing the role of the set of selection pressures upon the species, the coupling of the terms in these differentials means that the organism also plays some role in the selection of itself. As Lewontin puts it, “Organisms, then, both make and are made by their environment in the course of phylogenetic change, just as organisms are both the causes and consequences of their own ontogenetic development.”

In spite of the easily perceptible relationship between ourselves and our environment, however, we humans seem to take an opposite model when thinking about ourselves than we do for any other organism in evolutionary biology. The humanist model can be expressed as:

dEdt=gH, E and dHdt=hH

In other words, we as homo faber believe that we ourselves are the autonomous and independent force which shapes our environment to our liking, without considering that the environment in turn exerts a selection pressure on us. One need not look far for quotations about the mastery of man over nature, or how nature is designed for the use of man. Neither need we look far to witness rapacious actions against the environment with our own eyes: the paving over a rich patch of soil and wildflowers for a parking lot, the sudden clearing of a forest, the pollution of a river, fracking, & etc.

Whether we take the adaptation model or humanist model to consider our own position within nature, we end up in an absurdity which robs us of our freedom: according to the adaptation model, we are hopelessly determined by our environment, and we take for granted the environment which we inherit from our own predecessors without a clue that the state of that environment is within our power to modify; while according to the humanist model, we believe we are free to do as we wish, to take and to modify with impunity, with no forethought to any consequences that may circle back around to us from the environment which we take for granted as totally plastic and malleable under our hands. The first case puts us in that state of madness of the depressive, who cannot access their own power to act by virtue of believing they do not possess it, while the latter case puts us in that state of madness of the hubristic, who have no notion of their own limitations. Either case amounts to a loss of freedom, either when we believed we never had any to begin with or when we inevitably overreach our station, and what we may call natural justice promptly selects against us.

The construction model then seems to be the middle ground, the mutualistic expression of organism and environment which seems best able to capture our position within nature. The construction model amounts to the view that since we co-evolve with our environment, we actually have some sort of say in how our own phylogenetic history unfolds. The construction model expresses a proper relationship. But how is it that we affect our environment? It is surely by virtue of our technology above all else, but what exactly is technology doing? This is the question which can no longer be put aside by philosophy. We can no longer afford to let technology operate within the dark zone of otherwise clear thinking.

Yet, this is exactly where technology resides. Whether due to an ancient prejudice between those who philosophize and those who craft, or due to a perineal educational structure which cleaves the vocational from the theoretical, or due to some more physiological distinction between thinkers and doers, or whatever the case may be: we simply cannot afford to not understand technology as well as we can build it. We must cobble together some sort of picture of what technology is and what it is doing, how it relates to us and us to it, and we all know by now that it will not slow down to help us out. Luckily for us, we have access to many domains of knowledge which, in their specialized pursuits, have furnished us with a vast collection of new concepts which may be used as valuable principles or analogies to take into this most empty but fertile field of technological philosophy.

From evolutionary biology we begin to get an idea of what our ecological fate may be. From the theory of niche construction we get the idea that we may self-determine that fate in some way. I will not go on about how we may or may not undermine our own existence in the very (potentially misguided) pursuit of our own flourishing. That path is well-trodden by now, and by many who would do a better job at it than I (e.g. Odling-Smee, Niche Construction, ch. 10). However, those who have built these arguments often take technology as merely the means by which we construct, or deconstruct, our niche. This reduces technology down to something like a second-order phenomenon of our own biological evolution, and this shortcoming is manifest in all three models of human evolutionary biology. Under the adaptation model, we adapt to our environment by means of technology; under the humanist model, we control our own destiny by means of technology; under the construction model, we adapt to our environment by means of technology and the environment reacts to us by the effects of our technology upon it. None of this really offers much clarity about what technology is, however. Viewing technology as merely a means is exactly what pushes it into the dark zone of thought. The environment and ourselves are the two extreme terms in a bipolar relation between which technology sits, hidden in the middle.

As is the case with organism and environment, the relationship between us, our technology, and our environment turns out to be much more intertwined than a science would like it to be. However, might we lift the rapports expressed by the niche construction model out of biological evolution and apply them analogically to technological development? Is this a method by which we could throw some light on technology itself? After all, the fact that technological development is a higher order phenomenon of biological evolution gives us a clue that some of the same dynamics may apply, though we wish to keep in mind that technological development ought to be considered a distinct category of natural operations because technological beings are not living beings. Nonetheless, we may find value in conceiving of technology as an agent within an environment, as something which does not sit passively within a milieu, but which is a function as well as a force of its own within it. With that in mind, why don’t we apply the analogy to see how far it goes? If we take the construction model and swap organism O for technology T, then:

dTdt=f(T, E) and dEdt=g(T,E)

Which is to say, the development of technology over time is a function of the technology itself and of the environment, while the modification of the environment over time is likewise a function of technology and of itself—but what exactly is the environment here? Technology certainly exists within the natural environment, but it also develops in and is selected by our social and economic structures, which may operate as a more proximate environment from our technology’s perspective. In this sense, let us posit that democracy and capitalism act as an environment.

Since the environment is understood to select for certain traits, we may understand that certain kinds of technologies will succeed in our current environment while others will not, and still others will not even be recognized as possibilities. Far from being neutral or empty utility or means, technology has a very definite character, and is environmentally determined to evolve in particular directions. What are these directions? On the capitalistic side, technologies which are able to maximize profits, i.e. make indefinitely more money than they cost to produce, as well as technologies which are capable of reducing the cost of labor by automation will dominate the technological ecosystem. On the democratic side, technologies that are able to successfully influence public opinion as well as mass-producible objects that may be put into the hands of as many people as possible, thereby using public feedback to drive innovation, will dominate. Of course, both capitalism and democracy are thought to be valuable for their propensity to foster diversity. Even so, it is apparently the case that in spite of the initial technological diversity, over time technologies that successfully propagate tend to evolve towards automation, addiction, mass influence, and quality degradation. For the sake of this article, we will only look at the capitalistic mode of selection pressure.

When thinking about what maximizes profits we need to consider not only what consumers would be willing to pay for the product, but also what the producer is willing to pay to make the product. It is clear that automated manufacturing machines which do not need to eat, do not need to rent living space, do not need to entertain themselves, or support families are going to be much cheaper than human manufacturers. Therefore, we see that over time automated technologies will tend to proliferate unless some artificial guardrails are put in place to protect human workers competing for the same niche. For this is also a key factor to consider, that humans and machines may be induced to compete with one another within the same ecosystem. Hence, we see that regulation and deregulation of technologies come and go as the tides; this particular environment puts us at odds with our technologies.

On the side of how much a consumer is willing to pay for a product, there are endless ways one may be incentivized to spend more money on a product than was actually required to make it. This may manifest in supply and demand dynamics, advertising campaigns which seek to set up a tacked-on ‘social value,’ or may be driven by needs and dependencies. Each of these are profitable, and at times some more than others. However, I argue that the last of these, the model of needs and dependencies, is one of the best ways to guarantee profits: when something is truly a necessity, the value, whatever it is, will be met. Yet, there are only so many things that are truly considered to be necessary for human life, therefore, it becomes a matter of creating new needs. Indeed, the model of addiction turns out to be a very powerful way to guarantee profits, and even more so when the addictions are nonlethal. Hence, we may also expect many technologies within a capitalist ecosystem to often converge on an addictive strategy if they are to be long-lived. Like automation, this is only mediated if artificial guardrails are put in place to protect the consumer. Here again, the tug of war of regulation and deregulation symbolizes technology and humanity competing for advantage by modifying qualities of the environment, i.e. constructing the niche in contrary directions.

At the same time, however, it may be wondered how much the existence of this very environment was determined by the technology itself? If technology and environment co-evolve, is there a beginning to the process? Which of the terms takes priority? In some more recent pedagogical explanations of niche construction, the co-evolution of organism and environment is explained as a causal feedback loop. The feedback loop is something like: Environment has trait -> Organism adapts -> Adaptation changes environment -> Organism adapts… & etc. The feedback loop interpretation is convenient for understanding the basic concept, but it may actually lead to some problems.

If we find it problematic that our systems tend to select for automated and addictive technologies and we wish to alter the direction that our technologies evolve in, how can we intervene? We may posit that the environment is generally the first term in the beginning of a biological feedback loop, so we should therefore alter the environment of technology and it will follow in its course. But then, an organism which is successful in its niche construction is able to create a kind of stable equilibrium with how it alters the environment. The presence of these certain technologies may just outright resist or flip back environmental changes, meaning that the environmental traits conducive to modern technology would just resurface, if they were able to be changed at all. This ability of technology to actively structure its environment poses a problem to certain schools of thought which hope to simply alter our social and economic systems while utilizing the machinery of capitalism to support the new systems. The machinery is not empty utility, but plays an active role in shaping the social and economic structures to be as they are. For this reason, technologies that offload work and automate processes are sold as a way to liberate humans from labor, yet all that seems to happen is that we become busier, or our productivity increases while compensation stagnates. These technologies almost invariably reinforce or amplify the existing socio-economic systems instead of changing them in essence. If that is the case, then why do we not start by changing the character of our technologies, and then the environment will follow in its course? But then, the new character of the technology will be unfitted for the environment within which it already exists and again, the same technological strategies will eventually resurface, if they can be changed at all.

In this sense, the feedback loop seems to be somehow impenetrable. Once the cycle begins, the social systems, the economic structures, and the dominant traits of technology are simply locked in. It leads to the conclusion that all one can do is simply wait for the feedback loop to run its course. Assuming that this particular niche construction is problematic because it is not really a sustainable stable equilibrium, one might expect that over time the process would throw the whole ecology into instability and collapse it like any positive feedback loop does, for example, via climate change. Or, if one is more optimistic, one may think we will eventually be lead to a kind of technological transcendence at the ultimate end of the loop by landing in some sort of technological utopia where the economic and environmental problems are thought to be solved by means of the very technologies which posed the problems to begin with. One might even get the idea that the best way to get to the ultimate end of this feedback loop, whatever that may be, is to actively feed it, to accelerate it.

Herein lies the hidden thorn in the feedback interpretation: it quietly robs us again of our freedom to act; it becomes a kind of fate, or destiny. Look again, however, at the way Lewontin expresses the construction model. It is true that the differential pair could certainly be interpreted as causal feedback, but the fact that he uses dt implies a sort of continuity to the process over time. In fact, we see this pedagogical move of laying out discrete causal stages to describe actually continuous cycles often, e.g. the way in which we teach the Stirling cycle versus the continuous manner in which a Stirling engine actually functions. Additionally, there is no reason why the organism and environment, or technology and environment, must be thought to act in turns. The differentials seem to express a deeper relationship than that, a true co-evolution. We may instead think of the two variables as acting and being acted upon simultaneously between one another, making them a truly inseparable pair: much more like the relationship between voltage and current in a circuit rather than the relationship between, say, a thermostat and a room’s temperature.

What does this mean for technology? It seems to mean that we cannot simply change our technological design paradigms without changing our socio-economic structures; it also means we cannot simply change our socio-economic structures without changing our technological design paradigms. Both of these modifications must happen at once. Social and technological change are more radical than we often imagine them to be. We may find, for example, that the automobile is simply incompatible with some future socio-economic organization, but that if we truly want to manifest that reality, we must accept the fact. By the same turn, it will also be the case that new socio-economic organizations will go with technologies that we have never even dreamed of. However, as in evolutionary biology, technological niche construction is path-dependent, that is, the process of invention or the state of the environment are functions of their own previous states.

Yet, this may also be the point at which the analogy between evolutionary biology and technological development begins to crumble. Is there a different signification for biological and technological path-dependence? On the side of biology, we note that the physical world has broad and rigid constraints. We call these constraints laws of nature. These laws impose quite real limitations on organisms which mean that, despite the seemingly infinite creative capacity of evolution, organisms are not free to totally arbitrarily determine themselves like gods, but their evolution must occur according to the laws of nature as well as along lines laid down by their own prior phylogeny. Their freedom operates within natural constraints as well as within path-dependency. Certainly, our technologies inhabit the same physical world and are likewise subjects under the same rule of natural law. Nevertheless, when we consider the proximate environment of technological beings to be our social and economic systems, it seems that the ‘laws’ which inform technological development, i.e. invention, are much more changeable than the laws of nature which inform biological evolution. Path-dependency under these conditions becomes much more lax: true ruptures become a possibility. Such ruptures may be more radical than we expect, but also more possible. It sure beats waiting around for a feedback loop to snap.

Surely, the construction model is more helpful to us than the adaptation or humanist model, and within the construction model, a truly continuous, mutual interpretation of organism and environment, or technology and environment, may serve us better than a discrete feedback interpretation. We must choose our models carefully when they become so informative of our ethical orientations. It seems a poor choice to select a model which utterly undermines our freedom to act, such as the adaptation model, or more subtly, the feedback interpretation of niche construction when applied to technology. There is an equal but opposite folly in selecting a model where the agent is the autonomous force and the environment is mere background or utility, such as the humanist model, which amounts to rigid technological determinism when applied directly to technology. Between these extremes and outside the causal feedback model, we find a more difficult but more nuanced conception of freedom and of technology; one where technology is both a means and an end, both determined by but determining its environment, both a being in relation with but opposed to us; and one where freedom is neither null nor absolute, but something that operates within constraints and within path-dependency, as something that can act from within these complex co-evolving systems. By beginning from concepts borrowed from evolutionary biology, we may have just the beginning of a technological ethics, a compass just starting to point out a path. But this is all still just an analogy. How exactly is technological development different from biological evolution? What is an analogy anyway?