The Syrian crisis and the multi-year drought that preceded it have become emblematic of contemporary discussions about the possible security implications of climate change. Jan Selby and colleagues argue in a recent study that there is insufficient evidence to support a significant link between climate change, drought and violent conflict in Syria. Adrien Detges (adelphi) takes a close look at this study and provides an outlook on the points and critiques raised by it.
The Syrian crisis and the multi-year drought that preceded it have become emblematic of contemporary discussions about the possible security implications of climate change. Researchers have argued that extreme drought conditions in the Eastern Mediterranean at the end of the last decade have contributed to the emergence of the Syrian civil war. The argument, in short, goes as follows: anthropogenic forcing contributed to a record drought in Syria, which in turn led to the deterioration of rural livelihoods and accelerated migration to urban centres. In combination with other socio-economic and political pressures, this led to increased anti-government grievances, which resulted in protests, violent repression and eventually civil war.
A recent study by Jan Selby and colleagues (hereafter Selby et al.) in Political Geography takes issue with this argument. The authors claim that there is insufficient evidence to support a significant link between climate change, drought and violent conflict in Syria, especially when considering an indirect link via increased rural-to-urban migration. Behind this statement lie concerns over methodological weaknesses and premature conclusions in contemporary research, which would encourage sensationalist claims at the expense of scientific rigour and credibility.
The critique does a good job in pointing at the gaps in our current knowledge about the security implications of climate change in Syria, but seems bent on undermining the work done so far rather than suggesting further research to address the lacuna. Ultimately, the critique is not balanced enough, possibly misleading and risks encouraging undue scepticism vis-à-vis climate-conflict arguments.
Syria is not Darfur
Selby et al. start out with a list of quotes by high profile figures that serve to highlight what they describe as a tendency in contemporary policy discourse to overestimate the impact of climate change on fragility. In all fairness, Selby et al. do right by pointing at statements that go ahead of what available evidence suggests, but their critique is not differentiated enough. Matching a comment on Darfur from 2007 with a statement on Syria from 2015 tends to downplay recent scientific achievements and progress in the way experts and non-experts discuss possible connections between climate change and fragility.
Nearly the same can be said about their literature review, which emphasises the weaknesses of a handful of selected papers, but remains silent about other work. Casual readers of their article might get the impression that academic and non-academic discussions about climate change and fragility are far less nuanced and conscious of their own limitations than they actually are (in fact, recent reviews of the state of the art by Idean Salehyan or Halvard Buhaug are quite open about it). Interestingly, comments made in the main body of the article are often more commensurate, but they refer to more detailed aspects and do not seem to have much weight in the overall argument.
These problems are compounded by the fact that Selby et al. do not offer much constructive advice past generic calls for ‘greater caution and rigour’ in climate-conflict analyses.
How important is important?
The main issue with the paper by Selby et al. however, is its attempt to discuss the importance or significance of climate change as a factor of fragility on some sort of causal scale. To begin with, their argument is confusing. Selby et al. claim that previous analyses of the Syrian crisis tend to exaggerate the significance of climate change and drought as factors in the 2011 uprising, but it remains unclear what they mean by that. Since they do not quantify significance – arguably an impossible task, given problems with data and attribution – we cannot know by how much previous research is departing from what would be a ‘more adequate’ level of significance (see also Peter Gleick’s response to Selby et al.). Readers are left with a list of mistakes made in prior work, but without any indication on how to assess their gravity or relevance, thus again leaving them prone to misinterpretation and exaggeration.
More importantly, tussles over what constitute ‘more’ or ‘less significant’ causal factors can be highly misleading from a policy perspective, as they mask relevant interactions between different conflict drivers and mechanisms. Discussions like these suggest that conflict prevention needs to choose between addressing ‘climatic’ or ‘non-climatic’ issues, when in fact there is room and need for working on both conjointly. Whether or not climate change can be considered an ‘important’ factor of fragility does not really matter when there are opportunities to effectively address climate change in conflict prevention programmes and vice versa.
Caution where caution is due
That said, the article by Selby et al. does offer some valuable insights into the current limitations of empirical research on climate change and conflict in Syria. As have others, Selby et al. highlight the inherent difficulties in attributing particular climatic events like the drought in Syria to anthropogenic forcing. They advocate a probabilistic rather than deterministic rhetoric when talking about possible connections between climate change, extreme weather events and conflict risk. This point is also stressed by Cullen Hendrix in a recent commentary on the Syrian case.
Moreover, Selby et al. warn against squarely attributing particular social effects and grievances to drought, when the record of events says otherwise. For instance, they criticize an earlier study by Collin Kelley and colleagues for misattributing sharp declines in agricultural output in Syria to exceptionally dry conditions in the winters of 2006/2007 and 2007/2008, when in fact some of the steepest declines over the period 2003-2008 were recorded in 2003 and 2004. This is not to say that drought conditions could not have had an effect in addition to other processes that preceded them, but it relativizes their importance in explaining migratory dynamics and violence in Syria.
Closely related to this, Selby et al. stress that an overemphasis on drought in the Syrian case might mask other factors, such as unemployed youths, state repression, nepotism, mismanagement of natural resources or the cancellation of important subsidies. What is missing from this discussion, however, is how far these factors might have increased farmers’ vulnerability to drought and hence the salience of the drought in regard to their attitudes towards the regime of Bashar al-Assad. Building upon Francesca de Châtel’s analysis of the Syrian uprising, one could argue indeed that the drought helped to reveal the government’s poor commitment to climate resilience as part of a more general reluctance to ensure a sustainable agricultural policy.
As this discussion shows, there clearly is a need to better understand the possible mechanisms connecting drought and fragility in Syria to identify their central social and environmental components and to understand how they interact. As part of this reflection, it is worthwhile considering that droughts - as meteorological, but also hydrological and socio-economic phenomena - elude simple classification as purely exogenous shocks. The discussion of their causes and possible consequences for conflict and fragility needs to take this into account.