Many studies have concluded that the current global economy can transition from fossil fuels to be powered entirely by renewable energy. While supporting such transition, we critique analysis purporting to conclusively demonstrate feasibility. Deep uncertainties remain about whether renewables can maintain, let alone grow, the range and scale of energy services presently provided by fossil fuels. The more optimistic renewable energy studies rely upon assumptions that may be theoretically or technically plausible, but which remain highly uncertain when real-world practicalities are accounted for. This places investigation of energy-society futures squarely in the domain of post-normal science, implying the need for greater ‘knowledge humility’ when framing and interpreting the findings from quantitative modelling exercises conducted to investigate energy futures. Greater appreciation for the limits of what we can know via such techniques reveals ‘energy descent’ as a plausible post-carbon scenario. Given the fundamental dependence of all economic activity on availability of energy in appropriate forms at sufficient rates, profound changes to dominant modes of production and consumption may be required, a view marginalised when more techno-optimistic futures are assumed. Viewing this situation through the lens of ‘post-normal times’ opens avenues for response that can better support societies in navigating viable futures.


Transcending fossil fuels by initiating a swift decarbonisation of the global economy is one of the defining challenges of the 21st century. The most prominent factor necessitating this shift is climate change (and its related impacts), driven primarily by greenhouse gas (GHG) emissions for which fossil fuel combustion is the leading source (IPCC, 2018). Alongside this driver, the geological inevitability of fossil fuel depletion, with its potential to disrupt economies due to the increasing costs of maintaining anticipated energy supply rates, is relevant on a similar timeframe (Mohr, Wang, Ellem, Ward, & Giurco, 2015Wang, Feng, Tang, Bentley, & Höök, 2017).

In light of the transition imperative’s urgency and the high-stakes implications for sociopolitical stability, the extent to which alternative energy sources can reprise the physical economic roles of incumbent primary sources demands close and thorough investigation (Moriarty & Honnery, 2016, 2019). Will alternatives – specifically renewables and/or nuclear energy – be able to replace, in an economically and energetically affordable (let alone equitable) way, the fossil energy sources of today’s complex and globalised industrial civilization? Might a transition to post-carbon energy systems (‘a post-carbon transition’) imply fundamental discontinuities or step changes beyond present cultural, social and political-economic arrangements, rather than incremental techno-economic adjustments along a relatively smooth trajectory?

Today, more than thirty years after the IPCC was established, fossil energy sources still make up 84 % of global commercial primary energy supply, and global emissions continue to rise (BP, 2019), suggesting that transcending fossil fuels may be harder and more problematic than some optimistic studies suggest (see, e.g., Jacobson & Delucchi, 20112017a). Assessing the theoretical performance of systems comprising alternative energy technologies in the abstract, via quantitative modelling exercises that consider historically unprecedented developments unfolding decades into the future, cannot hope to address the full spectrum of questions relevant to establishing the practically realisable potential for such systems (Lenzen et al., 2016). Given the rate and scale of economic change required to minimise climate risks (i.e. net-zero emissions by 2050 or sooner), in navigating the terrain ahead we should expect that knowledge systems and practices established to deal with past and even current change processes will at best provide partial guidance, and at worst be misguiding.

In light of this, we believe that there is much insight to be gained by locating the investigation of energy-society futures squarely within the domain of post-normal science (Friedrichs, 2011). That is, we are dealing with situations that accord fully with Funtowicz & Ravetz’s (1993) original characterization of ‘post-normality’ in terms of uncertain facts, disputed values, high stakes, and urgent decisions. At the same time, the arguments we present in this paper are not limited to such framing. Even if energy-society futures are considered from a viewpoint of normal science and policy formulation, the case we make has direct relevance.

This paper assumes that transitioning to 100% renewable energy supply is an urgent and appropriate goal for humankind. However, informed by and consistent with the cautions from the post-normal perspective (Funtowicz & Ravetz, 1993Ravetz, 1998Saltelli & Funtowicz, 2014Saltelli, 2019), we present a critical and somewhat sobering assessment of the potential for quantitative analytical approaches to provide conclusive answers about whether renewable energy conversions can meet in full the demands for work, heat transfer, lighting and data manipulation made by today’s globalised and growth-orientated world economy (Alexander & Floyd, 2018). Reflecting an implicit positivist orientation, the findings of model-based energy and sustainability transition studies are frequently presented as if they relate directly to a ‘real world’ that a model is purported to represent, rather than relating to a ‘model world’ (McDowall & Geels et al., 2017Ramirez, Ravetz, Sharpe, & Varley, 2019). Even where correspondence between findings and ‘model world’ rather than ‘real world’ is strictly observed by study authors, such correspondence is often neglected in third party interpretation and reporting. This gives rise to what we see as a dangerously misleading optimism. While the post-normal framing directly challenges the grounds for such optimism, we emphasise once more that the case we set out is relevant even under the presumption of normalcy.

In this paper we show that energy transition modelling exercises are necessarily based on myriad complex and often controversial assumptions that necessitate the interpretation of their findings strictly in relation to the model as an abstract representation of a real world as understood by the modeler. Any conclusions drawn from such studies should be presented and applied with due acknowledgement of the deep uncertainties and limitations inherent therein. When the range of uncertainties and controversies is given due weight, we argue that a position of ‘knowledge humility’ is called for when assessing and developing scenarios and policies for a post-carbon transition (Amara, 1975Fazey et al., 2018Jasanoff, 2018McDowall & Geels et al., 2017Ramirez et al., 2019Sardar, 2010Sovacool & Brown, 2015). A disposition of knowledge humility entails reflexivity with respect to the epistemological foundations and commitments that inform transition-oriented decision making and action. Here the response to the dilemma of uncertainty and ignorance is not to deny it or seek to eliminate it, but to learn to live with it through reflexive governance (Voß et al., 2006).

Our review of the evidence and arguments suggests it is highly plausible that the transition to post-carbon energy sources and technologies implies reducing demand for energy services, per-capita and perhaps overall, below the levels of energy services enabled by existing fossil fuel-dominated global energy supply. While this statement is subject to the same circumspection that we argue should apply to contrary findings, we contend that such futures, which are presently marginalized (Laugs & Moll, 2017), should be elevated from a peripheral concern to one that actively shapes the ways in which actors engage in energy transition praxis. The case for this is sufficiently plausible that in the energy-intensive developed regions of the world, a post-carbon transition should include policy making and planning for what can be called ‘energy descent’ (see Odum & Odum, 2008Holmgren, 2012); or, to use the terminology previously introduced to Futures by Friedrichs (2011), ‘peak energy’ (i.e. futures characterised by significantly reduced energy supply). This would mean planning for and managing major supply reductions in coming decades, not just ‘greening’ existing supply (Moriarty & Honnery, 20082012a). This has profound implications for the basic social and political-economic formations that underpin our current modes of production and consumption.

The uncertainty attending energy-society futures, and the knowledge humility it demands, supports the case for adopting an anticipatory stance that is open to energy descent. Energy is a critical factor in economic production, but appreciation for the significance of this is weak within orthodox economics (Keen, Ayres, & Standish, 2019). Reduced overall availability of energy services implies economic degrowth, or downsizing of economies in terms of physical production (Sakai, Brockway, Barrett, & Taylor, 2018). This view diverges from mainstream green growth aspirations that involve ‘decoupling’ GDP from physical production, and physical production from energy and other resource use, enabled by technological efficiency gains and greater emphasis on services (Hatfield-Dodds et al., 2015). Evidence continues to mount that decoupling is incapable of meeting green growth expectations (Bithas & Kalimeris, 2018Hickel & Kallis, 2019Parrique et al., 2019). Furthermore, empirical studies cast doubt on the intuitively appealing idea that orienting economies towards services and ICT-mediated activity will reduce their energy intensity (Fix, 2019Palmer, 2017aParrique et al., 2019). On the other hand, the case for reduced economic growth allowing much more rapid decarbonisation is strongly supported (Foran, 2011Victor, 2012Le Quéré et al., 2019).

Before beginning our assessment of renewable energy’s physical and economic prospects, a brief note on nuclear power is required to delimit the scope of the present analysis. We appreciate that nuclear energy will play an important role in global energy systems for many decades ahead. Whether its share of total final energy supply increases modestly (Froggatt & Schneider, 2015), or perhaps even declines as old plants are decommissioned faster than new plants are brought on-line, nuclear energy’s persistence contributes only marginally to the question of energy descent plausibility and does not fundamentally alter our conclusions. We justify this on the pragmatic ground that, regardless of their relative techno-economic and environmental merits, considered globally, renewable energy sources seem to have achieved a large advantage over nuclear energy in terms of social and political support.

Given the extent to which renewable energy dominates visions of post-carbon futures, and the associated weight of research effort that it receives, it seems reasonable to focus attention on this prospectively dominant share of total supply. Nonetheless, the broad arguments that we make about the relationship between quantitative modelling and knowledge pertaining to plausible futures apply also to visions of alternative futures in which the relative contributions of nuclear and renewable sources are reversed. On this basis, the primary focus in this paper is the question of whether existing energy service expectations in the developed industrial economies can be satisfied primarily – and as close as possible to ‘entirely’ – via renewable sources.

Ed. note: You can find the link to the rest of the paper here. The link was updated 16/09/20 as the previous link was behind a paywall for many readers.