Skeptic: ‘a person who questions or doubts something (such as a claim or statement): a person who often questions or doubts things’ (Merriam-Webster Dictionary)
Over the past few weeks it’s slowly dawned on me that I’ve become more of a “foresight skeptic”. Although for some time I’ve thought the foresight term (e.g. “foresight practitioner”) is generally problematic, this is a new feeling of doubting some of the aspirations that inform fields like futures research, strategic foresight, and so on, and some of the practices that are used.
In this post I’d like to detail some of the reasons why my views have shifted, both as a marker of my current thinking (for later reflection) and, potentially, to stimulate responses from others.
Limited theorisation; tensions between theory and practice
As part of my doctoral research I’ve read papers questioning whether practices like scenario planning, visioning, and so on, have a sound theoretical basis. Slowly I’ve come to agree with many of these critiques. Critics (e.g. Shipley) expose important tensions between theory and practice.
There is no shortage of ‘tools’ and approaches (e.g. inductive or deductive scenario-building processes, backcasting processes, including some recipe-type guides), but I have often found these to be lacking in practice. For example, existing scenario guidelines proved to be inadequate when developing scenarios of the Copenhagen global climate summit (held in 2009) and its potential implications for future global climate regimes. Tools and practices also need theory; currently this aspect – in most cases – is largely determined by the practitioner’s background and/or improvisations. Scenario development models typically do not provide explicit supporting and underpinning theory for developing robust forward views. The specifics of scenario construction processes are, furthermore, generally vague.
Although efforts to address this problem have grown over the past five or so years (e.g. by linking to larger paradigms like constructivism, or to particular social theories, and complexity theory) they remain partial. Angela Wilkinson (Counsellor for Strategic Foresight, OECD), and colleagues further argue that there remains a “lack of well-grounded theory” to explain what scenario-based techniques work, when, in what circumstances, and why – and what doesn’t, when, and why.
At the same time I think this problem been amplified by the trend to use future techniques in inter-organisational settings. Examples are the shifts from product/service roadmaps to producing them at the sectoral level (rather than by individual firms) and the increasing use of scenario planning in community contexts or public policy contexts (rather than in a single organisation).
I also think the recent issue of On the Horizon that responds to Richard Slaughter’s book The Biggest Wake-up Call in History is a “wake up call” to our nascent professional field. It includes no unique theory and scant discussion of the methodologies used for speculating about possible and preferred futures. Contributors draw on theory and research from various other fields when making assertions and posing questions about the future. (I have a paper in this issue, so I’m also criticising my own work!). In some ways this is simply a reflection of the inter-disciplinary nature of futures inquiry; in others ways I feel that much of the issue fails to make a significant contribution to the debates it comments on. For example sustainability scientists and human geographers have deeper insights about population-environment dynamics and the complex relationship between humans and the natural environment.
I’ve started to call this perspective the “doughnut” view of futures inquiry – to provocatively suggest that there remains a big hole (i.e. in terms of theory) in the middle of current practice.
Lack of evaluative studies and critical scrutiny of methods
Over the past year I’ve looked into the lack of evaluative studies of prospective exercises, becoming aware of the fact that evaluation – when conducted – tends to be anecdotal. As other scholars note supporting empirical evidence is rarely cited by those advocating particular methodologies.
The reasons for this are complex. However, as I’ve probed a little deeper during my PhD research I’ve noticed that it contributes to gaps in knowledge that are problematic for practitioners and other users of these techniques. For example, it problematic if you’re looking at real-world effects as part of efforts to judge if particular techniques or approaches are more or less effective (e.g. in contexts like organisational decision-making, change processes, etc) and to assess their relative merit and worth.
The broader situation – which some call “methodological chaos” – may be worsening because we simply don’t know “what works”, stimulating ongoing trial-and-error style experimentation.
Coming to terms with complexity
On the one hand complexity, and the behaviour of dynamic systems, informs arguments for new approaches to planning, futures inquiry, and management. What some have termed the ‘complexity turn’ has had important implications for many domains. Some scenario practitioners argue scenario-based approaches are grounded in recognition of the inherent indeterminism of complex systems, thus necessitating exploration of multiple possibilities. E.g. Consistent with this Angela Wilkinson and colleagues argue we should “hold the future not as a belief, but as a fiction.”
On the other hand, as Tim Harford has written in Adapt, a more complex world, with more complex problems, also can be seen to militate against those seeking to develop foresight.
Similarly, when asked if forecasting techniques are improving, Nate Silver stated: “as-a-whole not by a lot, maybe a little bit at a time. Technology gradually improves, and scientific knowledge becomes more complete, but it’s still very incomplete… And in some things, and in human endeavour, society becomes more complex so you might have better methods, arguably, and more data, but you’re also are running against a moving target when the system itself becomes more complicated.”
Complexity science also suggests we should doubt some uses of scanning processes and early warning systems. Writing recently in the academic journal Foresight Averil Horton notes that complexity science has demonstrated that disruptive events (in complex systems) often don’t have earlier triggers, and he suggests people looking for “weak signals” may therefore be more likely to be caught unawares! ‘Foresight’ techniques may, instead, embrace “emergence” and the reality of unexpected disruptive events, but such dynamics make forward-looking assessments inherently problematic.
Cognitive barriers and related issues
Psychological sciences also provide a mixed picture for strategic foresight, futures research, etc. It points to the importance of mental models and need to address “cognitive inertia”. It suggests that imagining and considering scenarios can provide a useful basis motivating action by persuading folk of the plausibility of such futures, and this can also reduce overconfidence. However, it also suggests scenario-based exercises and related techniques can: exacerbate existing biases or create new ones; lead decision-makers to overestimate the likelihood of future events; and lead to maladaptive anticipatory actions. Drawing on a range of published studies Healey & Hodgkinson (2008, p.574) write: “contrary to popular wisdom, scenario analysis can narrow and confine as well as stretch thinking”.
Bradfield (2008) has done some important related research. He found that cognitive barriers can limit learning in scenario exercises, and individual cognitive and group behavioral factors are insufficiently considered. Bradfield observed – as I have too – that constructed scenarios tend to match the creators’ pre-existing belief systems and previous experiences, and surprising or extreme developments (so-called “out-of-the-box” thinking) get omitted. Cognitive science research indicates that the cognitive issues “are likely to be most prevalent when there is a multitude of inherently uncertain and complex situations and issues to consider as is generally the case in scenario work” (p.210). He adds:
“there are numerous cognitive factors at work in determining how people think, and therefore, stretching mental models and moving people beyond their business-as-usual thinking to create new insights about the changing world is a complex and difficult task… it should be recognized that overcoming the cognitive barriers discussed in this article requires a great deal of attention to the design of a scenario process, and the existing literature on scenarios provides little guidance of any substance in this area” (p.211)
I’ve grappled – often unsuccessfully – with the related ‘plausibility paradox’ (Healey & Hodgkinson, 2008). That is, scenarios must be considered plausible so that decision-makers meaningfully accept them and they, thereby, motivate change or inform action; however a highly plausible scenario is typically one that fits prior knowledge well and therefore is unlikely to challenge or, where necessary, transform existing mental models. This paradox is a key barrier to challenging status quo thinking.
Similarly, engaging more with uncertainty may not have the claimed benefits, such as improved decision-making capability. Potential pitfalls need to be considered, including perceived uncertainty leading to slower, rigid decision-making and heightened negative affectivity. The latter can trigger decision-avoidance and dysfunctional coping strategies (Healey & Hodgkinson, 2008).
The overall picture presented by such research is that practice is more complex than often presented.
There’s often more to the story than what’s told in “success stories”
Stories about Shell’s 1970s “oil shock” scenarios are a case in point. First, predictive success is celebrated when prediction is explicitly not the (stated) aim. Second, insider accounts show that Shell’s scenario planners often struggled to be heard and taken seriously – Pierre Wack himself remarked that scenarios that significantly challenged decision-maker’s mental models were “like water on a stone”. It took, in-part, some senior managers who saw the scenarios as well-aligned with their own agendas to drive change. They latched onto the scenarios that could help them to make the case for change.
Kees van der Heijden notes that the oil shocks helped to legitimise scenario planning at Shell but – and this is a big but – most manager’s continued to interpret the unfolding events as mere blips before a return to ‘business-as-usual’, instead of recognising change as the beginning of a new world.
These cases also seem to question that capacity of scenarios to be effective “cognitive devices” (van der Heijden, 1996) in surfacing underlying assumptions and enabling critical debate.
Lack of balance in writings of many practitioners and advocates
Many writings create the impression that practices like scenario planning are panaceas (e.g. for improved decision-making under uncertainty), rather than considering both potential benefits of the practices and the potential pitfalls (e.g. process risks, and the possible unintended consequences, etc).
Wider influences, underlying intellectual journey?
Since completing a Masters in Strategic Foresight program at Swinburne University I’ve immersed myself in interdisciplinary social sciences like Science and Technology Studies (STS). On the one hand research by those developing a Sociology of Expectations supports a focus on future expectations and guiding images the future (see my paper on nanotechnology). On the other hand, STS’s critique of determinisms of all kinds (e.g. technological determinism, social determinism) has strongly shaped my thinking.
STS scholars have persuasively demonstrated that technological developments – and their social consequences – are inherently unpredictable and, to a large extent, indeterminate. The desire to know which frontiers of advance will (or could) come to dominate, or lead to the most progress, is potentially fatally flawed. Consequently, anticipatory activities can in many ways seem pointless – although the stories that are told about scientific and technological futures clearly do have effects.
What is interesting from this perspective is the way that visions and expectations are strategically deployed. STS scholars focus on these dynamics, examining their ‘performative’ qualities.
Over the past five years I’ve also be heavily influenced by Professor Mike Hulme, who was an influential climate modeller and is now a sustainability scientist. Whilst he recognises the underpinning climate science, Hulme has critiqued the current emphasis on modelling the future. He argues that ‘climate reductionism’ (via climate modelling and related impact assessments) has become the dominant mode of analysing environmental change – both reflecting, and contributing to, overly pessimistic readings of social futures. Like STS scholars he is also wary of determinism – in this case the re-emergence of influential forms of environmental determinism. He also asks important STS-style questions about how climate models and climate modellers gain and then exercise social authority.
Hulme’s public doubting of the growing investments in climate modelling is thought provoking and important. Equally important has been his critiques of how climate change is framed and thought about, with Hulme questioning pessimistic readings of social futures. He has further argued that sustainability issues like climate change need to be understood both as ideas (which are situated in cultural contexts) and as physical phenomenon studied (and forecasted) through scientific practices.
Where to from here?
Over the past decade or so the increasing scholarly attention on futures inquiry has raised questions about these practices, leading to growing doubts (in my mind) about their use and effectiveness. At the same time, I note that threats and opportunities are increasingly “made real in the present” through model-based projections, scenario-building, and other modes of futures thinking.
Perhaps some of the issues raised by the critical scholars can be better addressed. Certainly, it’s been suggested by some management scientists that practitioners can increase their effectiveness by drawing on the latest psychological research (Healey & Hodgkinson, 2008).
Another possible way forward is to look anew at practice (e.g. scenario-building) through new theoretical lens. This may be a way to both critically examine and enrich these practices. An example paper I spotted recently examines and rethinks roadmapping practices through the lens of socio-technical transition theory and transition policy (see McDowall, 2012). Another recent paper also argues for more attention to social theory in backcasting exercises (Kiraly et al, 2013).
Practitioners often gravitate towards understanding theories of change and stability in an effort to ground their work (E.g. Sohail Inayatullah draws heavily on macro-history; Andy Hines and Hardin Tibbs draw on theories of cultural/social values change; Rafael Ramirez draws on social ecology theorists; and many others draw on a wide range of theories of technological change). There may be opportunities to go further, drawing on wider bodies of social theory produced by social scientists (e.g. in fields like sociology, political science, anthropology) and theories of socio-ecological systems (e.g. resilience theory). Prior to embracing integral theory, Richard Slaughter pointed to the importance of social theory such as those theorising the social construction of reality (Slaughter, 2002).
In Futures for the Third Millennium Richard Slaughter also noted that one of the criticisms of futures studies has been that its grounding in other fields has not been well ‘mapped’. Unlike Slaughter, I think this criticism still carries legitimate force. The recent focus on ‘methodological renewal’ has perhaps masked underlying issues regarding the theoretical basis of such methods. An example is the ‘framework foresight’ model recently published by the Houston program: it features a number of analytical steps but zero theory (some is alluded to, such as economic theory [e.g. Kondratieff Cycles, the business cycle], sociological theory [given the reference to ‘constants’], and stakeholder theory).
Or, perhaps, a largely indeterminate future means the hole in the “doughnut” ultimately cannot be filled. This thought refers to my lingering doubts about aspirations to achieve ‘foresight’ – be it organisational or social. Yes, we need action taken in reference to the future, but the dynamic complexity of reality may be too great for a sophisticated grasp of the not-yet. What I see, instead, are partial viewpoints shaped by belief systems. As Tim Harford persuasively argues in his book Adapt: Why Success Starts with Failure “the world we inhabit” may simply be “too complicated for anyone to analyse with much success” (p.8). The concept of organisational foresight is a problematic one if – as Harford argues – “it is impossible to know in advance what the correct strategy will be” (p.65).
Similarly, Jose Ramos has considered the implications of complexity theory for futures inquiry: “such an approach acknowledges that variables involved in understanding change are staggering enough as to make absolute certainty impossible. With human intention and action influencing ‘the future’, any kind of mastery over knowing change or creating change seems ridiculous.” (Source: ‘Action Research as Foresight Methodology’, Knowledge Base of Futures Studies, Foresight International).
Here, the “action foresight” work of Ramos appears useful. As I understand it, futures inquiry of this kind does not seek to discover “truths” (about the future); instead the core objective is to help to enable action in a particular local context. Without disregarding the issues raised above, participative futures inquiry like this may help to create change and to test ‘foresight’ methodologies. Ramos also notes that action research has moved toward conducting “systemic interventions”, and argues that well-grounded foresight may help to link larger patterns of change to action in local contexts.
Finally, if futures thinking and studies is, in essence, storytelling – as powerfully argued by Donald Michael – more interesting questions may focus on which stories get told and why. Given so many different possible futures and stories could be told, complex choices must be made.
I wonder if the above thoughts resonate with others and, if they do, how others are responding…