Still today, too many doctoral candidates start their work with an empty Excel workbook and not much else, and this ‘start from scratch’ recurs with every new generation of young researchers.
These widespread research practices are barriers to knowledge accumulation and slow down progress. For many published claims, the raw data, software used, and sometimes even basic assumptions such as system boundary choices (which processes and effects are included in a study and which are not) are not well documented, e.g., in life cycle assessment. Intermediate results are not reported, which means that often, other researchers cannot connect to existing knowledge. Consequently, a large body of repetitive research, incompatible datasets, and opaque modelling tools have accumulated over the last decades. Some of the data generated and tools developed cannot be improved or reused. Hence, the published works do not contribute to a cumulative body of scientific knowledge, and there is a danger that our work loses relevance and gradually turns into busy work. The prevalence of limited consultancy-type funding, the tendency to publish in small units and the recurrence of short project time frames mute incentives for cumulative data compilation and refinement as well as development of transparent methodology. If not overcome, the above-listed barriers to knowledge accumulation in the participating disciplines will curtail the impact of sustainable development research.
The lack of incentives for cumulative research in science for sustainable development has been bothering me for a few years now, and it is mainly our own responsibility to rethink our workflow and boost knowledge accumulation. While all actors involved, from funders to publishers, can facilitate knowledge accumulation, I believe that the main responsibility lies with us researchers and with the research associations and societies, in particular.
I put my thoughts and proposed solutions to the knowledge accumulation problem into a comment piece on the future development of sustainability science, newly published in Nature Sustainability (http://dx.doi.org/10.1038/s41893-019-0443-7). There, I list the required measures and infrastructure needs for cumulative sustainability science as well as the different responsibilities for fostering knowledge accumulation.
The solutions to the structural problems of sustainability science need to be community-driven, while securing stakeholder support is needed to bring about resilient structural change. Different processes need to be installed within science for sustainable development (Figure, inner circle) and several important exogenous determinants of a transition to more cumulative research need to be taken into account (Figure, outer circle).
We need a change in research culture to make cumulative science the new normal and embrace the new workflows resulting from the adoption of cumulative research strategies.
I hope you will find the time to have a look at the measures proposed in the comment and that you will join the process of improving sustainability science!