There are many historical case studies of past societies that show how they waxed and waned in the face of climate, technology, war, and disease. Classic examples include the collapse of the Mayan civilization due to land stress, warfare, and climate change and the end of the Roman Empire due to its overextension and financial mismanagement. Typically, these cases are used as simplistic analogies for modern predicaments. However, recent modeling of historical case studies that draw on new data about past environments can be combined with written sources and archaeology to create incredibly detailed simulations that provide a fresh approach to the past.
So far, these simulations are mainly used in order to clarify our interpretation of the history and to test the applicability of theory. That is a shame, because historians can run simulations that model food supply and biodiversity evolution from the past right up to the present to give us a unique evolutionary perspective on our current problems. Showing how the present has come into being but remains continually connected to the past allows us to create what Stewart Brand has called a “long now”—a perspective that allows us a more complex understanding of the long-term social, economic, and environmental challenges that all civilizations have faced.1
A new generation of models places decision makers, such as farmers and regulators, at the center of these simulations. An excellent example is the work done by archaeologist Tim Kohler and his team on the pre-Hispanic Central Mesa Verde region (currently, southwest Colorado).2 Careful modeling of the interactions between the local society and its environment has allowed the group to study in detail many of these interactions, including experimenting with different scenarios concerning resource exploitation and settlement location. In the case of settlement location, for example, the activities of families play out on a relatively detailed model of the actual landscape. The families have specific habits and techniques for exploiting the environment, and researchers can then choose to position families on that landscape in order to maximize their net caloric gain, as well as meet their requirements for protein, water, and fuel. In reality, of course, they are not expected to maximize all of these, but that hypothesis serves as a “null hypothesis” to better understand the relationship between humans and their landscapes. Comparing this null hypothesis to the actual distribution of the settlements in the landscape allows researchers to isolate the more complex dynamics underpinning the aberrations from the model results.
As with the best data-modeling comparisons, the exercise constrains existing ideas and speculations while opening up a new set of hypotheses to test. By adding to the model some of these additional hypotheses, and then testing again, a feedback cycle is set in motion that can ultimately lead to detailed insights into the dynamics involved. Learning about the degree of sensitivity of, for example, crop yields, not only gives more insight into how the historical socio-ecological system functioned, but helps to define the level of certainty in the model results. Eventually, the improved understanding of a particular, modeled case study feeds into the model’s usefulness as a heuristic for modern management.
Success in how these models simulate the future has to be measured in terms of how they anticipate the dynamics of the system, rather than in terms of concrete predictions. An example of how these models reveal the complex behavior of systems can be found by studying the theory that systems, such as watersheds or coastal plains, exist in a state of flux between alternate states. They shift from one to the other as external drivers push them over thresholds. This phenomenon explains why, for example, lakes can change rapidly from clear to turbid states in just a few years or decades.3 But for more complex processes, like land degradation, much longer timescales of observation may be needed. In a case study of southwest China, John A. Dearing showed how the relationship between land cover and soil erosion has existed in two alternate states.4 The transition between a stable, nondegraded state to a stable, but highly degraded state took several centuries to complete once a threshold in land cover had been passed.
To understand how these processes work, we need to overcome the modern tendency to support short-termism and to obsess about tangible events, while ignoring less tangible dynamics. Some argue that this position is the result of modern science’s privileging of reductionism, or the analytical process of breaking the world down into smaller and smaller pieces. Reductionism has driven technological progress over the last three centuries, but at the expense of understanding the interconnectedness of everything. When it comes to addressing many of the key world challenges, like climate change, macroeconomics, globalization, and resource depletion, there can be no substitute for a holistic and historical perspective. But a reductionist approach is so heavily ingrained and so widely spread in our culture and, especially, in science that changing our thinking will require a major effort. Only recently have scientists begun to explore explicitly the use of history to help elucidate complex system dynamics in the real world and to inform modern management of socio-ecological systems.5 Our worldview, language, and institutions all militate against such change.
By far the greatest challenge to overcoming global problems lies in shifting our mindset through education. This is where history simulations can play an important, pedagogical role. The current education system in the developed world is, overall, no longer adapted to the challenges of the twenty-first century, among which sustainability questions loom so large. We have to move toward an education focused on real-world problems where the aim is to balance simple cause-and-effect explanation with systems thinking. Understanding the history of systems, and the impact of human decision making over the long term, is a first step to transforming our mindsets.
