There are the things we know that we know. Then there are the things we know that we don’t know. And finally, there are the things we don’t know that we don’t know. So goes the poetry of former US Secretary of Defence, Donald Rumsfeld. One might argue that he missed the possibility of tacit knowledge (the things we don’t know that we know) in the above quote. But otherwise, these four epistemological categories, taken together, comprise by logical extension the totality of what we may know, or not know, about the future. And then, of course, there are also our ‘intuitions’ or ‘habits of thinking’ about the future, which do not quite qualify as knowledge per se.

Some of these habits of thinking have to do with our perception of time. Take for instance the classic short story A Sound of Thunder by the world-famous science fiction writer Ray Bradbury. If you don’t already know it, it’s time you read it. The story revolves around two protagonists who travel back to the end of the Cretaceous period to hunt a Tyrannosaurus rex as a trophy. Having realized that even slight interferences might escalate into huge changes in their own present, they attempt to take all kinds of precautions to prevent this from happening. In the end, these precautions turn out to be in vain; one of the two protagonists accidently diverges from the preordained path, killing a butterfly in the process. Upon their return, they discover a change in the outcome of a vital presidential election, resulting in a win by a sinister totalitarian candidate. The story is famous for fleshing out the argument that the effect of slight changes may over time grow into major disruptions that can alter the course of events on a massive scale. It has sometimes been credited as the inspiration for the ‘butterfly effect’ – although the term was formally coined by the meteorologist Edward Lorenz, who discovered that tiny, butterfly-scale changes to the starting point of deterministic nonlinear systems may result in vastly disparate trajectories at later stages. Within the domain of evolutionary biology, this contention may find some theoretical comfort in Stephen Jay Gould’s classic contingency hypothesis, which states that life today would have looked very different if the patterns of extinction and survival had taken an alternative trajectory back in the Cambrian, or, say, at the end of the Cretaceous.

A CONtention

This ontological contention has an epistemological corollary in the contention that the further we move forwards in time, the harder it is to predict the outcome of things. This position may also be found in futurist writings, for example in Herman Kahn and Anthony Wiener’s classic 1967 milestone contribution The Year 2000: A Framework for Speculation on the Next Thirty-three Years. Again, this corollary may find some comfort in the models of chaos theory, as well as some apparent empirical support in the environmental sciences, where the long-term effects of pollution, invasive species and the like often appear to be out of reach for the scientists studying the particular phenomenon. The classic example of this is of course the apparent unpredictability of the toxic effects of the pesticide DDT on ecosystems, which were spelled out in Rachel Carson’s Silent Spring. In short, DDT was launched as an effective pesticide against malaria during the 1940s by authorities who believed that the concentration of this compound was too low to have any negative effect on wildlife and ecosystems. This assumption soon turned out to be wrong, as it was discovered that the concentration of DDT accumulates throughout the food chain, in effect poisoning the top avian predators. Hence the title Silent Spring, which alludes to a spring without the song of the birds killed by DDT. DDT was later banned in most industrialised countries, but has found continued use in areas of the world where environmental legislation is less regulated. Meanwhile, DDT has now spread to the entire planet, including, for example, populations of penguins in Antarctica.

Of course, the question remains as to whether these contentions about long-term unpredictability are actually credible on a more general level. It may seem easy to find counter-examples. The case of DDT mentioned above could certainly be interpreted as an example of the inherent unpredictability of nature, but it could also be written off as a rookie mistake made back in the 1940s and 1950s by the still emerging field of ecology. After all, taking account of the accumulation of toxic compounds throughout the food chain (or ‘food web’ as ecologists tend to call it nowadays) isn’t exactly rocket science. With the advanced understanding of the complexity of ecosystems that is now integral to modern conservation biology comes a theoretical perspective that is equipped with a series of ‘early warnings’, which at the very least tell us where to look when things go wrong. In evolutionary biology, Gould’s adversary Simon Conway Morris has argued that cases of parallel and convergent evolutions demonstrate that, although the appearance of a specific species may depend on circumstances, there is an underlying causal predictability in the kind of adaptive complexes that will emerge during evolution. We also know with strong astrophysical certainty that the Sun will expand its volume in about 6 billion years to the extent that it will swallow the Earth completely, ending all life on this planet. By contrast, it can be next to impossible to predict the outcome of a presidential election, even on the eve of voting. Thus, depending on our choice of examples and theoretical presuppositions, it would seem that the whole picture is rather more complex than might initially have been presumed. This in turn highlights the fact that the relationship between timescale and forecasting is an issue that must be taken into account by any observant futurist.

In fact, on closer scrutiny, it appears that the overall ‘problem’ of timescale and forecasting is actually an entanglement of at least four distinctive issues, each of which has its own semi-independent dynamic. The most obvious issue is that the various examples mentioned above are phenomena on incomparable levels. Natural physical systems may be complex and possess certain kinds of unpredictability, but they do not (we presume) possess the reflexive agency that gives human (and to some extent advanced animal) systems the ability to change their behaviour in radical new ways as a response to prior information. Furthermore, many systems of futurist interest are hybrid, in the sense that they comprise a mixture of human actors and their socio-technical practices, artefacts and infrastructures, and natural physical and biological systems. To navigate properly in the nexus between these disparate epistemic domains is a challenging task, full of the pitfalls that arise both from the common limitations of academic specialisation and from clashes between the incommensurable causal-explanatory frameworks that exist among these domains.

The second – and perhaps less obvious – issue is that there is no universal agreement regarding what exactly constitutes the necessary elements in a successful analysis of the future. Or – to put it another way – there is a notorious lack of coherence among futurists in terms of framing and methodology that goes right to the heart of how problems about the future are defined at all. The most obvious example of this are the tensions about the role of forecasting itself in futures studies, and the question of whether prediction or empowerment for change should be the main goal of this exercise. The practice of preferring or disregarding certain futurist approaches at the expense of others is in itself a powerful sorting mechanism that constitutes a possible bias in a futurist analysis.

In extension of this, the third issue directs our attention towards the rather troublesome fact that in a multidisciplinary community of experts with all sorts of backgrounds, there are some experts who are more equal than others. Despite periodic toasts for plurality, it is an unsurprising fact that future studies – perhaps with Rachel Carson as a notable exception – is dominated by thinkers from the Humanities or the Social Sciences side of the divide that CP Snow once named ‘The Two Cultures’ (the other being the technical and natural sciences). With this social context also comes the ever-present risk that members of the majority view may export their own prejudices and blind spots into the general state of future studies. However, should anyone succumb to this, they do so at their own peril. Ignoring material and natural conditions in future developments is something that may severely skew the conclusion of the analyses in question. Take for instance the prognoses of the late, world-famous demographer Hans Rosling, who predicted that the world human population would stagnate at a figure of approximately 11 billion people in the year 2100. This prognosis has been used by scholars in the Humanities and the Social Sciences, who have argued that the problem of population growth has been exaggerated and that it will play no major role in the fate of generations to come. The trouble with this view, however, is that Rosling’s calculations (which, among other things, assume increased gender equality and an increased level of literacy and education among women) presume a kind of geopolitical stability for the rest of this century that seems highly unlikely in the face of the challenges from anthropogenic climate change. Avid followers of future geopolitical dynamics will note that one of the major expected consequences of increased global temperatures will be a significant decrease in global food and freshwater supply – up to 35 percent in yields of global vegetable and legume production, according a PNAS review paper from 2018. Other expected corollaries include an overall degradation of ecosystems, global toxification, poorer child health, rising economic inequality, and an increased frequency of armed conflict, social unrest and terrorism. Clearly, the combination of these perils could easily put Rosling’s optimistic prognosis under pressure.

TIMESCALE

Finally, the fourth issue has to do with the choice of timescale. This choice, when appearing in any investigation, is not just a matter of ‘mere’ taste or preference – it also a source of additional bias in itself. By focusing our attention on future developments within a given timespan, we also turn our attention away from what may happen outside that given timespan.

The usual timeframe of futurist interest is typically between 10 and 30 years. When Kahn, in the early Reagan years, wrote in The Coming Boom, Economic, Political and Social with predictive success of a future based on progress and free-market capitalism, he did not envisage an early twenty-first century aftermath with the headaches of globalised jihadism, the Covid-19 pandemic, and the rise of anthropogenic climate change as a major existential threat to humanity. Likewise, it is a common geopolitical expectation that China will rise to the position of the world’s dominant superpower within the next 30 years. Sometimes this expectation is accompanied by the idea that India will emerge as China’s main geopolitical competitor. However, these prognoses, which are primarily based on pre-Covid-19 economic growth rates, fail to take into account that both China and – especially – India are expected in the latter half of this century to face serious challenges to their food and freshwater supply as a consequence of the developments mentioned above. If these challenges are not countered with due diligence, they may transform these countries’ population surplus from a geopolitical asset to a geopolitical liability, in turn endangering their long-term status as great international powers.

There is of course a wide array of unknowns in this analysis, but it does illustrate the perils of choosing a too narrow timeframe in futurist analyses. As Henry Kissinger once remarked, history has no ending.

AN EVEN DEEPER BIAS

Luckily, it does not seem impossible to counteract the problems associated with the four issues mentioned above through reflexive awareness and attention to problem framing. However, underlying all this is perhaps also a fifth and even deeper overall bias. This bias has to do with the notion of foresight itself and the origin of future studies as a strategic planning device, in military circles and elsewhere. There is, by logical extension, a common counterthesis to the tension within future studies concerning prediction vs. empowerment for change, and that is the rather simple proposition that neither of them is possible to the extent that we would like them to be. Contrary to optimistic approaches that seek to ‘plan’ or even ‘design’ the future, this proposition warrants a more pessimistic approach that pays closer attention to the dynamics behind systemic dysfunction and the dangers that can lead our plans astray. A coherent epistemology of futurism will probably need to utilise all these complementary approaches in order to reach a balanced understanding of our possibilities and limits in forecasting and influencing the lives of coming generations. We may learn much through studying megatrends, weak signals and the like, but it is equally important that we recognise the limitations of these analytical perspectives. In the future, the race will not be to the swift, nor the battle to the strong. Time and chance will happen to them all.

---