Scientific disciplines must link up to serve development
A greater commitment to multidisciplinary research, and to local problem solving, is essential to achieving future development goals.
Last week, UN secretary-general Ban Ki-moon announced the members of the panel that has been given the challenge of defining the global development agenda after the Millennium Development Goals (MDGs) expire in 2015. 
The panel's first meeting, scheduled for this September, will mark the start of a process that will run parallel to — but separately from — the process of defining goals, targets and indicators for the Sustainable Development Goals (SDGs) that are likely to succeed the MDGs.
The process of creating a new set of development goals is expected to culminate in a proposal to the 68th General Assembly of the UN in September next year. 
How well the two processes will be coordinated remains to be seen. But in both cases, targets and recommendations are likely to be shaped around the three dimensions of sustainable development: economic, social and environmental.
Translating this triad into meaningful indicators of progress must involve both a solid evidence base and a focus on the needs of poor communities. It will also require scientists to work effectively across disciplines, and journalists to reflect this in their coverage.
No field has all the answers
Holistic approaches to tackling the challenges facing the livelihood of millions of people around the world can be interdisciplinary, multidisciplinary, cross-disciplinary or trans-disciplinary.
The principle behind each of these concepts is that no single discipline or branch of science holds all the answers when it comes to complex social and environmental problems.
The value and need for this approach was reflected in the rationale of proposals at Rio+20 for SDGs, and is something that few would disagree with.
A good example of the need for a transdisciplinary approach is a recent proposal, from Samantha Watson, of the London School of Medicine and Tropical Hygiene, and colleagues, for integrating knowledge in the physical, life and social sciences to improve the capacity of health systems to cope with, and respond to, natural hazards. 
The authors argue that such an approach would result in technological tools already used for preparedness and response, such as remote sensing, having a greater impact through better coordination and integration into planning.
There is much to learn from experience. Medical tools such as vaccines can do little good without health systems to deliver them to those in need. New technologies like solar panels can have little benefit without understanding social expectations. And new crop varieties can help agricultural productivity, but not without understanding the practices of smallholder farmers.
Barriers to cooperation
Growing acknowledgement by scientists, funders and policymakers of the need for a holistic approach to problem solving, and of the limitations of single disciplines working on their own, is one difference between the target setting of ten or 20 years ago and the thinking of today's scientific community
But putting this new awareness into wider practice is another matter.
There are the practical considerations of finding the resources to put multidisciplinary teams together, to get them to understand each discipline's scientific language, and to navigate the contributions and limitations that each brings to the table.
Then there is the (understandable) preference of disciplines, individuals and institutions to work with long-established methods that have proven reliable, rather than venture into uncharted territory.
The challenges are more pronounced when the goal is to bring together natural and social sciences. As recently as last June, in the run-up to Rio+20, social scientists appealed for a bigger and more visible role in research tackling global environmental changes. The obstacles are as basic as defining what counts as acceptable evidence, to agreeing on joint research methods.
The question now is how the scientific community and policymakers can make concrete progress on integrating economic, social and environmental sciences in the interests of development.
Global and local imperatives
Understanding the assumptions, methods and knowledge that different fields of science can contribute is essential to bringing them together effectively. So is clarity on how each can contribute towards joint goals.
Beyond this, it is important to ensure that scientific research involving several disciplines is linked to, or even rooted in, local problems. The SDG framework could set the stage by agreeing on a mechanism for integrating scientific disciplines in a way that channels research activities towards practical objectives and tangible impact.
Such a mechanism will have greater impact if not restricted to goals and indicators. Holistic approaches should also include incentives for linking up on the ground — for instance, making multidisciplinary teams working on practical questions a requirement for funding.
Practical questions are not necessarily known before research or development agendas are set. Take polio, where the impact of attitudes towards vaccination emerged relatively recently and remains an obstacle to achieving eradication. 
This example shows why researchers in the developing world must be in a position to shape research agendas at both grassroots and global levels, with knowledge of needs and socio-cultural contexts.
Journalists can play a role too, helping to raise the awareness of project managers and policymakers by highlighting how different fields of science can bear on development issues.
Research around the economic, social and environmental dimensions of development will continue regardless of global goals. But global goals are an opportunity to maximise its impact.
Post-2015 strategies need to be explicit about how to bring together different fields of science in a way that serves evidence-based policy not only at the global level, but also locally. That is where the impact will — and should — be most acutely felt.