One of the most enjoyable things about studying as a post-graduate in a UK Geography department was the diversity of conversation topics I could get myself into in the corridors, over lunch, and after work in the pub. Investigating social, economic, cultural, atmospheric, geomorphological, and ecological patterns and processes (too name just a few) geography departments contain scholars with interests and skills that span the globe’s physical and social environments. This variety of backgrounds and worldviews can lead to widely differing perspectives on the current affairs of any particular day.

In many ways my PhD studies, funded by an interdisciplinary research studentship from the ESRC and NERC, allowed (demanded?) me to search out these differing perspectives and engage in these conversations. However, this diversity of perspectives isn’t appealing for faculty members focused narrowly on their own particular research specialism and the current paper they are writing about it. Maybe they just don’t have time. Or maybe there’s something deeper.

The distinction between the social sciences (human geography) and natural sciences (physical geography) has led to somewhat of a divide between these two ‘sides’ of Geography. As my former tutor and advisor Prof. David Demeritt highlights in the latest volume of the Transactions of the Institute of British Geographers, ‘human’ and ‘physical’ geographers have become so estranged that dedicated forums to initiate ‘conversations across the divide‘ of Geography now occur regularly at annual conferences. Demeritt’s article discusses how ‘Environmental Geography’ is often touted as having the integrative research potential to bridge the human-physical divide.

Environmental Geography (EG) explicitly sets out to examine human-environment interactions and is generally understood to be the intersection of Human and Physical in the Geography Venn diagram. Essentially, EG is the Geographical version of the Coupled Human and Natural Systems (CHANS) research program that has become prominent recently largely thanks to NSF funding. Whereas CHANS emphasises systemic concepts (thresholds, feedbacks, resilience etc.), EG emphasises concepts more at home in the geographical lexicon – scale, space and (seemingly most often absent from CHANS research) place. This is not to say that these concepts are exclusively used by either one or the other – whether you do ‘CHANS research’ or ‘Environmental Geography’ is also likely to be determined by where your research funding comes from, what department you work in, and the type or training you received in graduate school.

One of the main points Demeritt makes in his commentary is that this flat distinction between Human and Physical Geography is not as straight forward as it is often made out to be. Friedman’s world may be flat, but the Geography world isn’t. Demeritt attempts to illustrate this with a new diagramtic 3D representation of the overlap between the many sub-disciplines of Geography (most of which are also academic disciplines in their own right):

Thus, “Rather than thinking about geography just in terms of a horizontal divide between human and physical geography, we need to recognise the heterogeneity within those very broad divisions. …within those two broad divisions geography is stretched out along a vertical dimension. … Like the fabled double helix, these vertical strands twist round each other and the horizontal connections across the human-physical divide to open up new opportunities for productive engagement.” [p.5]

This potential doesn’t come without its challenges however. Demeritt uses EG to demonstrate such challenges, highlighting how research in this field is often ‘framed’. ‘Framing’ here refers to the perspective researchers take about how their subject (in this case interactions between humans and the natural environment) will be (should be) studied. Demeritt highlights three particular perspectives:

1. The Externality Perspective. This perspective might be best associated with the reductionist mode of scientific investigation, where a specific component of a human-environment system is considered in isolation from any other components. Research disregards or ignores other work in sub-disciplines, whether horizontally across the human-physical divide or vertically either side, and concentrates on understanding a specific phenomena or process.

2. The Integrated Perspective. We might think of this perspective as being loosely systematic. Rather than simply ignoring the connections with other processes and phenomena considered in other sub-disciplines, they are used as some form of ‘input’ to the component under particular consideration. This is probably the mode that most closely resembles how much CHANS research is currently done, and how most ‘interdisciplinary’ environmental research is currently done.

3. The Participatory Perspective. This third approach has become more prominent recently, associated with calls for more democratic forms of science-based decision-making and as issues expertise and risk have come to the fore in environmental issues. This mode demands scientists and researchers become more engaged with publics, stakeholders and decision-makers and is closely related to the perspective of ‘critical’ geography and proponents of ‘post-normal’ science.

Demeritt discusses the benefits and challenges of these approaches in more detail, as I have briefly touched on previously. Rather than go over them again, here I want to think a bit more about the situations in which each of these modes of research might be most useful. In turn, this will help us to think about where engagement with other disciplines and sub-disciplines will be most fruitful.

One situation in which the externality perspective would be most useful is when the spatial/temporal scope of the process/phenomena of interest makes engagement between (sub-)disciplines either useless or impossible. For example, reconciling economic or cultural processes with Quaternary research is likely to extraordinarily difficult (but see Wainwright 2008). A second would be when investigation is interested more in ‘puzzle solving’ than ‘problem-solving’. For example, with regards research on Northern Hardwood Forests the puzzler would ask questions like ‘what is the biological relationship between light availability and tree growth?’ whereas the problem-solver might ask ‘how should we manage our timber harvest to ensure sufficient light availability allows continued regeneration of younger trees in the forest understory?’.

The integrated approach has often been used in the situation when one ‘more predictable’ system is influenced by another ‘less predictable’ system. One system might be more predictable than another because more data are available for one than another, because less assumptions are invoked to ‘close’ one system for study than another, or simply because the systems are perceived to be more or less predictable. A prime example is the use of scenarios of global social end economic change to set the parameters of investigations of future climate change (although this example may actually have slowed problem-solving rather than sped it).

The participatory perspective will be useful when system uncertainties are primarily ethical or epistemological. Important questions here are ‘what are the ethical consequences of my study this phenomena?’ and ‘are sufficient theoretical tools available to study this problem?’. Further, in contrast to the externality mode, this approach will be useful when investigation is interested in ‘problem-solving’ rather than ‘puzzle solving’. For example, participatory research will be most useful when the research question is ‘how do we design a volcano monitoring system to efficiently and adequately alert local populations such that they can/will respond appropriately in the event of an eruption?’ rather than ‘what are the physical processes in the Earth’s interior that cause volcanoes to erupt when they do?’

Implicit in the choice of which question is asked in this final example is the framing of the issue at hand. Hopefully it is clear from my brief outline that it is a close relationship between research objectives and the framing or mode of the research. How these objectives and framings are arrived at is really at the root of Demeritt’s commentary. Given the choice, it will be easy for many researchers to take the easy option:

Engaging with other perspectives and approaches is not just demanding, but also risky too. … Progress in science has always come precisely from exposing ourselves to the possibility of getting it wrong or that things might not work out quite as planned’. [p.9]

Thinking clearly about the situations in which different modes of study are most useful might help save both embarrassment and time. Further, it also seems sensible to suggest that most thought should be done when researchers are considering engaging non-scientists in the participatory mode. If it is risky to expose ones self to fellow scientists, who understand the foibles of the research process and the difficulties of grappling with new ideas and data sets, it will be even more risky when the exposure is to non-scientists. Decision-makers, politicians, ‘lay persons’ and the general public at large are likely to be less acquainted with (but not ignorant of) how research proceeds (messily), how knowledge is generated (often a mixture of deductive proofs and inductive ideas), and the assumptions (and limitations) implicit in data collection and analysis. So when should academics feel most confident about parachuting in from the ivory tower?

First, it seems important for scientists to avoid telling people things they already ‘know’. Just because it hasn’t been written down in a scientific journal doesn’t mean it isn’t known (not that I want to get into discussion here about when something becomes ‘known’). We should try very hard to work out where help is needed to harness local knowledge, rather than ignoring it and assuming we know best (this of course harks back to the third wave). For example, while local farmers may know a lot about the history and consequences of land use/cover change in their local area, they may struggle to understand how land use/cover change will occur, or influence other processes, over larger spatial extents (e.g. landscape connectivity of species habitat or wildfire fuel loadings). In other situations, local knowledge may be entirely absent because a given phenomena is outside the perception/observation of the local community. In this case, it will be very difficult (or impossible) for them to contribute to knowledge formation even though the phenomena affects them. For example, the introduction of genetically modified crops will potentially have impacts on other nearby vegetation species due to hybridization, yet the processes at work are at a scale that is unobservable to lay persons (i.e genetic recombination at the molecular level versus farmland biodiversity at the landscape level).

The important point in all this however (as it occurs to me), seems to be that the ‘framing’ one researcher or scientist adopts will depend on their particular objectives. If those objectives are of the scientific puzzle-solving kind, and can be framed so that the solution can be found without leaving the comfy environment of a single sub-discipline, engagement will not happen (and neither should it). The risks it poses means that engagement will happen only if funding bodies demand it (as they increasingly are) or if the the research is really serious about solving a problem (as opposed to solving a puzzle or simply publishing scientific articles). As the human population grows within a finite environment the human-environment interface will only grow, likely demanding more and more engaged research. As I’ve highlighted before, a genuine science of sustainability is more likely to succeed if it adopts an engaged, participatory (post-normal) stance toward its subject.

Engaging researchers from other (sub-)disciplines or non-scientists will not always be the best option. But Geography and geographers are well placed to help develop theory and thinking to inform other scientists about how to frame environmental problems and establish exactly when engaging with experts (whether certified or not) from outside their field, or even from outside science itself, will be a fruitful endeavour. Geographers will only gain the authority on when and how interdisciplinary and participatory research should proceed once they’ve actually done some.

Demeritt, D. (2008) From externality to inputs and interference: framing environmental research in geography Transactions of the Institute of British Geographers 34(1) 3 – 11
Published Online: 11 Dec 2008
doi:10.1111/j.1475-5661.2008.00333.x