… there was a system, in which the mother was dependent on what the child had learned, as the stimulus for the next position [which] wasn’t well articulated until we got the cybernetics conferences going.
—Margaret Mead, on anthropological origins of second-order cybernetics1
You can now get an hour of entertainment, travel a mile in comfort, treat a headache, and feed your family more cheaply than ever before, thanks to innovations in using natural resources and human labor. As our energy sources shifted from timber to whale oil to coal to oil, the reward has been a vast array of consumer options and, for many, better lives. The march of progress continues: today people everywhere have access to more streaming videos, SUVs, ibuprofen, and pork than our forebears could have imagined. Yet the astonishing breadth of new products and services comes at a cost that consumers cannot easily see. Look about and you will see signs of a gathering crisis of resource depletion and pollution along with glimpses of its socio-economic reverberations.
Not every input can be swapped out. Water has no substitute, nor does clean air in our cities or an atmosphere free of excess greenhouse gases. Every now and then, resource crises and their attendant socio-economic dislocations come into our collective field of vision, only to be eclipsed by other news stories. In early 2008, the Wall Street Journal looked to alarming trends in rising prices across 22 global commodity markets, while observing that severe water and farmland shortages threatened economic and political stability in sub-Saharan Africa, India, and China. The pattern seemed to match the picture painted by those who had warned of limits to growth, from the eighteenth century’s Malthus to the 1970s Club of Rome reports by a team of MIT computer modelers along with ecologists and other observers over the decades.2 3
Shifts in thinking are visible. The 2008 Journal story noted that the economist and Nobel laureate Joseph Stiglitz had abandoned his formerly cornucopian world view for a more Malthusian one. Markets alone cannot solve resource and pollution problems, Stiglitz explained, when oil, water, and other key resources are underpriced relative to their environmental impact, especially given the extremely slim prospect that people would accept the drastic changes needed to price them correctly. But soon the specter of growth-limiting resource constraints and skepticism that markets would handle them again retreated from the news. It turned out that, for a moment, the world’s ecological footprint lightened: 2009 saw the largest ever drop in fossil-fuel greenhouse gas emissions, caused by the global financial collapse that first gathered speed in 2007. Attention turned to the firms, governments, and families who were struggling to survive the economic crisis. Today, greenhouse gas emissions are rising again while questions about how to spur economic growth continue and worries about climate change deepen.
I wonder, are we missing the connection? Could the forces that shaped the financial crisis be connected with the challenges we face in climate, pollution, human rights, and economic opportunity? Might a move away from reduction help us to understand our shared options?Social and economic systems figure into the answers, no matter what your philosophical leanings. But how, exactly, do we make sense of these systems?
Addressing the question, I argue, involves working against reduction. To help us get there, let’s explore the cybernetics underpinnings of the notion of limits to growth to map out how the very ideas that have fueled technological and economic growth could also enlighten our response to them. My hope is that we will find signs that there is a better way.
Thomas Malthus, whose name has become synonymous with the idea of limits to growth, posited that human populations could grow exponentially while agriculture—the backbone of the economy when he was writing in 1798—would grow only arithmetically. While the notion of self-reinforcing exponential growth has roots in ancient mathematics, formal approaches to logarithms had been developed in the seventeenth century. The Malthus equations, which underpin modern demography, explain how populations grow at accelerating rates without constraint unless checked by a rise in mortality (famine, disease, war) or by changes in human behavior (causing fertility rates to reach or fall below replacement). While the first-order differential equations produce smooth trajectories, Malthus understood that the effects of the limits would appear sporadically, though famines and plagues caused by crowding and poverty that would temporarily alleviate some pressures. Malthus wondered if we would ever be able to escape “a perpetual oscillation between happiness and misery.”4
You can apply the same framework to bacteria in petri dishes or the population of Easter Island (also called Rapa Nui). Malthus’ influential work shaped Charles Darwin’s thinking on how the struggle for survival gives rise to selection pressures, an idea that in turn shaped not only biology but also our thinking about business and countless social phenomena—the role of competition and the notion of fitness, for example.5 (With its eventual immoral application to social darwinsim and eugenics, as Joi Ito notes.)
Even today, when introducing dynamical systems, systems modelers often start with the population growth model proposed by Malthus. Its value lies not only in the intuitions that it maps into, making differential equations meaningful to the audience (and in the intuitions it challenges, when the exponential growth rate is graphed and the surprising hockey-stick trajectory emerges) but also in the thinking it sparks about strengths and limitations of the model, as people wonder what else to factor in. In such a use, a model is not just a computational machine to check the logical output of a set of assumptions but is also a uniquely valuable tool for dialog because it forces participants to surface their assumptions.
It also provides a third benefit: a wide-angle view of time. Using analysis or simulation, one can look far into potential futures using different assumptions.
The interplay between self-fueling processes of growth and the limiting forces of competition set the stage for economists and biologists who followed Malthus and Darwin. The same interplay found a parallel in control theory, an idea central to mechanical and electrical systems built on the physicist James Clerk Maxwell’s nineteenth century work on governing mechanisms for regulating steam engines. The twentieth century saw the flowering of electrical engineering, which began exploring how to use information from sensors to dynamically guide the state of a machine using feedback control, and cybernetics was born.
Soon, feedback thinking was shaping psychology, urban studies, ecology, management, philosophy, and more. Anthropologist Margaret Mead looked at family systems through the lens of feedback, using her experience to argue in 1967 that cybernetics should include itself in its mapping of systems, giving rise to second-order cybernetics.6
A decade before Mead’s speech, electrical engineer Jay Forrester founded the field of system dynamics, a feedback modeling approach that uses state variables to model interactions among people, nature, and technology. Like other threads of cybernetics it was made possible by the new power of computers to simulate the behavior of dynamically complex systems: after centuries of mechanization enabled economies to become more complex while jobs became more specialized, technology became a tool for moving beyond reduction in how we think.
Simplifying assumptions needed for analytical tractability were no longer necessary: the world could be cast as multicausal, simultaneous, and probabilistic, with forces for change and stability playing out at varying scales. Human behavior, with its panoply of explicit and implicit judgments, could be captured with unprecedented realism, taking insights from a growing body of experimental and field studies of human decision making. The study of bounded rationality had been pioneered by economists who drew on cybernetic thinking to build new theories of social and economic action that continue to influence the behavioral turn in economics, policy, and technology design today.
Instead of relying on theoretical simplifications of classical economics and closed analytical solutions, researchers and policy designers began to map out potential pathways of complex, interacting systems that captured the influence of multiple factors, without needing to rely on mental simulation alone. This approach, as Forrester explained to colleagues and students at MIT, was built on philosophy as much as technology; instead of looking at events and narrow slices of the world that assumed away everything else, computers allowed analysts to embrace a continuous and holistic perspective in which everything could connect in some way to everything else—something that philosophers, artists, spiritual leaders, and wise grandmothers have known for millennia could now be mapped digitally.
The systems view reminds us that in the real world, what we see as events are themselves the products of continuous processes that unfold over time, with no beginning and end, and that nothing is ever solely a cause or solely an effect. Events, cause, effect, side-effect, externality, main variable: all are artefacts of our thinking that we reify in ways that shape our perspectives in unseen ways. When you take a new medicine, your pancreas and kidney do not know which of its effects are intended and which are not, yet the language we use—“side effects”—tends to downplay the unintended consequences.
A famous example of the power of perception comes from a simple plot of exponential growth: depending on the scale you choose for your axes, what looks like a sharply rising hockey stick of growth can look like a flat line or gentle slope—which in turn shapes how you think about the situation and the options. It all depends on your perspective, which is in turn is shaped by the questions you are asking. The greatest shortcoming of the human race is our inability to understand the exponential function.7
As in many human endeavors, setting the focus is the first thing to get right when working with a model.8 Because it is not possible to represent everything, in each attempt to understand the world you need to set the boundary, level of analysis, and time horizon that will help to make sense of things. Usually, Forrester pointed out, we take too narrow a view in time and space.9 Short-sighted thinking set the stage for revenge effects: paradoxically, our actions often create the outcomes we most want to avoid.10 11 Drastic calorie cutting resets the dieter’s metabolism to a lower level, making weight gain forever more likely. Trying to boost profits, a company unleashes a price war that permanently erodes profitability. Football helmets prevent some acute injuries but pave the way for other, less observable harm that eventually causes irreversible encephalopathy and dementia. In seeking the upper hand that they envisage will enable peaceful relations with other countries, national leaders create an arms race that increases political risk.
By the 1970s, system dynamics was being used to illuminate some of the biggest challenges of all, connecting to the question Malthus had posed nearly two centuries before: would humanity encounter the same limits that constrained growth in other populations? Might the global problematique, as it was then called, represent the ultimate revenge effect?
Linking human action and nature through highly simplified equations, the early global models of the 1970s eschewed detail complexity that represented geographic and categorical variations in favor of more universal interactions. Simulations showed the results of failing to make wise choices about production and consumption: unchecked human activity could overwhelm ecosystems’ abilities to eventually recover and the world could cross a tipping point into unsustainability. Models of this type, system dynamicists held, were not predictions but were intended to show the result of certain policies and hypothesized processes. This point was not always taken by audiences.
It was difficult to find scenarios with good outcomes that did not rely on drastically changing consumption patterns. Put in economic terms, subsidies for resource-use coupled with the lack of price mechanisms for avoiding pollution meant that markets could not generate optimal, or even good, outcomes, particularly when the irreversibility and time-lags of ecosystem degradation were factored in.
The models used in early global studies are still available, and you can use them to explore potential paths.12 13 Two types of scenarios, you will find, avoid the most dire outcomes. The first set is built upon an extreme shift in the character of technology—casting it as far more beneficial, assuming away negative side effects. The second option is to limit consumption and growth, by choosing to live a less resource-intensive life before being forced to do so by worsening conditions. Today’s question is: what are our options now, when contending with an atmosphere that contains long-lived greenhouse gases, oceans infiltrated with micro-particles of plastic, e-waste dumps that contaminate the soil and groundwater poor communities rely on, and an ecosystem in which pesticides and antibiotics are measurable in every creature’s body?
The most plausible answer, it seems to me, is that we need to do both: to search for better technologies that move us towards the beneficial, and to design rich, joyful, and low-throughput lives for ourselves. To develop the case, let’s add to our roster of systems insights that may help move beyond reduction. We’ll link an exploration of tipping points, the notion that people and nature are separable, and exponential thinking to key ideas of feedback modeling that we have already touched upon, using the Limits to Growth debate to illustrate why they matter.
The study of complex feedback systems reveals that swings into undesirable states can be corrected by response mechanisms when the entire system is operating within a zone of stability. This understanding precedes computer modeling, of course: self-correction underpins the design of democracy, markets, and our other core social institutions. But history teaches us that some forays can take the system into an unrecoverable state, as the Easter Island/Rapa Nui example of an island civilization’s collapse reveals. Feedback can come too late to rescue you. Across every tradition, our songs and literature convey these truths. The twin stories of rebalance and collapse are the engines of comedy and tragedy.
As cybernetic thinking developed, examples of the dangers of crossing the threshold into unsustainability abounded. During the cold war era, runaway nuclear war represented one such unsustainable zone, eventually giving rise to stability-oriented policies of détente, containment, and test ban limitations. Rachel Carson’s Silent Spring fueled a wave of thinking about limits to growth due to contamination and pollution, raising the possibility of irreversible species extinctions. As the NRDC notes, she changed the world with the persuasive and meticulously-researched argument that some technology is so destructive to nature that it could not be absorbed by ecosystems without harm and therefore must be fettered.14 The study of climate change, launched by Svante Arrhenius in 1895, came to the fore of global modeling in recent decades and scientists outlined possible runaway physical positive feedback processes, such as the albedo effect, along with self-reinforcing social responses to physical changes, such as when warmer climates induce people to use even more fossil fuels for air-conditioning.15
How to avoid the collapse caused by overshooting the stable zone? We’ve already heard that economists are skeptical that prices for natural resources could be reset globally to reflect their true cost to the planet, but there are plenty of examples of countries imposing higher taxes or limiting carbon emissions via permits and auctions. Such reforms enable markets to move towards greater sustainability. Another solution is to create markets in avoided pollution (such as negawatts, which is energy that electric utilities do not have to generate because people are cutting back on power usage). Systems thinkers caution that such methods push back one set limits, buying us time, but if we do not manage other limits (water pollution, for example), becoming more efficient in one dimension may move the sum total of human activity towards a limiting constraint at an even faster clip. The ensuing collapse could be even worse.
In tandem with the work on global systems models, economists and ecologists developed a rich variety of ways to frame the global problematique.16 17 The 1970s saw arguments for steady-state over growth-driven economies, for example, and subsequent thinkers developed new ways to value nature and social wellbeing, posited data-driven thinking about ecological thresholds, and advanced the notion of resilience as a shared goal that integrates humans and the natural world (Panarchy, for example).18 Each of these perspectives rests on a key element of resisting reduction: things other than money matter.19 20
The debates we’ve just examined prompted what might be the most reductionist of questions: Are humans good or bad? The question strikes me as silly. Humans, like plankton and voles and lakes, are to be valued for themselves, as Native indigenous cultures, Thoreau and his compatriots, and countless religions advise. Yet how to make this idea relevant, to connect nature to the industrial economy that shapes our lives?
Measuring ecosystem services is one way to translate nature into monetary value, according to ecological economists, as a sort of stop-gap measure to illustrate the importance of nature.21 Yet nature’s intertransforming transactions do not involve money in any way, as Buckminister Fuller noted: “The grass doesn’t have to pay the clouds for the rain, the earth doesn’t have to pay the sun for radiation.”22 What do we lose when we try to monetize these interactions? Is the challenge one of measurement and computation, or something more than that?
Object-oriented philosophers caution against capturing nature this way, then go much further to advocate for learning to look at ourselves and the physical world in the same way, discarding the very notion of nature because it induces a sense that people are separable from everything else.23 The challenge is cognitive: massive phenomena like global warming or the internet are vast and complex hyper-objects that we cannot comprehend, so we treat them as abstractions when they are, in fact, as real as the spoon on your table.24 Meanwhile, age-old assumptions in science are dissolving as geologists reframe a discipline that for centuries has been seen as separate from humans: the study of the earth itself. Human activity is shaping the atmosphere, geology, hydrology, and the biosphere, they say, to the extent that we are now amid a new geologic era: The Anthropocene.25
In contrast, Julian Simon, a famous critic of neo-Malthusian thinking, argued that humans should not be seen as destroyers of nature but instead as the ultimate resource. A 1998 Wall Street Journal column explained: “Simon’s central point was that natural resources are not finite in any serious way; they are created by the intellect of man, an always renewable resource.”26 Against the backdrop of today’s global challenges, that looks like a strong claim. Yet it has its attractions. The view offered a more optimistic alternative to the deleterious pollution-generating, resource-depleting picture of humanity central to global models in which creativity, invention, and even the conscious choice to live more lightly were only represented in the aggregate effects of their impact on consumption. The models can’t tell us what we might create. In that sense, they look a lot like first-order cybernetics. Our responses lie outside the model. But global modelers felt that the key feedback loop of response lay outside of the model: if a Limits to Growth model succeeds, it does so by painting a picture that humanity responds to in ways that change the course of history—making the model itself eventually irrelevant.27
Today, Simon’s human-centered answer to Mathusianism has morphed into technological optimism. At its foundations is the notion of exponential thinking—which, in a deep irony, owes much to the very idea of self-reinforcing growth that gave rise to thinking about limits. Proponents of exponential thinking draw the same pathways seen in exponential population growth to argue that the future will be better in ways we cannot foresee. Yet often, the underlying mechanism—the technology equivalent of the population growth equations that would constitute a theory that can be debated and tested—are missing. Instead, we are told, the pattern speaks for itself.28
There’s another interesting parallel. If philosophers of the Anthropocene find limits in our ability to understand hyperobjects, advocates of the singularity find limits in our ability to understand the interacting, multiplying and presumably overwhelmingly positive effects of super-exponential technology development.29
Could it be that it’s the wonder of the future that we are failing to comprehend?
Alas, the data does not back up the story. A comprehensive 2016 World Bank study noted:
The effect of technology on global productivity, expansion of opportunity for the poor and middle class, and the spread of accountable governance has so far been less than expected. Digital technologies are spreading rapidly, but digital dividends – growth, jobs and services – have lagged behind.
A key conclusion: Technology, by itself, is insufficient to reduce poverty.
Perhaps the problem is simply that we’re not there yet, that around the corner is some future, post the hockey-stick transition, where technology will automatically unlock benefits for the poor. Such thinking is a common trap in future studies: that tomorrow will somehow be radically different from today, that the complexities and challenges we now face will be transformed. Might the event-based thinking that early cyberneticists took on in advocating for a continuous feedback view be to blame for such naïveté? Today and tomorrow are linked by myriad interactions. Any picture of a radically different future will need to explain how we get there, and acknowledge that things will keep changing.
With my skepticism in mind, it seems that we need more than just technology: we need ways to think, decide, and create together. In predictably bureaucratic language, the World Bank study identifies the antidotes to technology’s failure: reforms in the business environment, skills development, and good governance.
Can we cast these ideas in a creative, generative, and hopeful light?
Because it’s impossible to get things right the first time, and because the future is both already unfolding and at the same time always ahead on the horizon, we need to build the dialogs, institutions, interactions, and collaborations that will enable us to reform and redesign how we live, learning from our own experience and others as we go, missteps and failures included.
There is no “there” there when it comes to the future, so what we now need is a way to improve the journey.
What if we could redirect some of the effort that we currently devote to developing technology and business to new ways of working and living together? Cities, schools, neighborhoods, and companies could replace machines of blood and flesh with responsive, participatory, creative, and generative social organisms by harnessing the best of technology and rejecting its ills. To do this, we need a relentless, diligent, and clear-eyed ability to root out the unwanted effects of technology and markets and a commitment to nurturing the participation and creativity they unleash. Tech insiders are helping us to think more wisely: “Don’t simply focus on what would be ideal or critique the status quo,” danah boyd advises her fellow technologists, “Genuinely examine how what you’re seeking could also be corrupted and abused.”30 The critical requirement: empathy and self-reflection. Consumers, academics, and even journalists are entering the conversations too.
Today’s response to the anti-humanitarian aspects of mobile phone design that puts new pressures on the software and device providers—instead of blaming users for the anomie they experience—could presage a more socially responsive form of design where users shape technology.31 Interactive models could enable communities to collectively design the neighborhoods and parking lots and clinics for their own health and happiness in technology-enabled collectives designed by the visionary efforts pioneered by Rethink Health.32 We could move away from the separation of person and technology, humans and nature, ourselves and the world, all of which underlie our disembodied and siloed stance towards technology, markets, money, and expertise. We’d appreciate people in all their wonderful complexity.
Ito calls for a mindset change. This is the response I imagine. We’d teach our children and each other to understand the dangers the thought patterns built into the ways we look at the world and take action today. Recognizing that good intentions can unleash revenge effects, we’d learn to ask how things might go wrong, using every method at our disposal to imagine both the upside and the down side of our interventions. Grasping the lessons of tipping points, we’d value resilience. Understanding that people and nature are not separate, we would stretch our consciousness of everything around us and work against narrow categorization. In exploring both the wonder and danger in exponential growth, we’d value the principles, practices, and investment in human systems that would enable wise choices about how we interact with each other and the natural world. We’d move from a sense of separation to what Martin Buber called “I-and-Thou.” Wiener’s colleague Heinz von Forster used the same quote from Buber to lay out what he saw as the most important ethical principle for second-order cybernetics:
Contemplate the human with the human, and you will see the dynamic duality, the human essence, together: here is the giving and the receiving, here the aggressive and the defensive power, here the quality of searching and of responding, always both in one, mutually complementing in alternating action, demonstrating together what it is: human.33
It’s time to heed the advice of Weiner, von Forster, Mead, Forrester, and other founders of cybernetics, and put human connection back in the system.