Entries tagged with “Science”.


So, climate change and conflict is back in the media, seemingly with the strength of science behind it.  I’ve been a rather direct, harsh critic of some work on this connection before, at least in part because I am deeply concerned that work on this subject (which remains preliminary) might disproportionately influence policy decisions in unproductive or even problematic directions (i.e. by contributing to the unnecessary militarization of development aid and humanitarian assistance).  So, when CNN, the Guardian, and other media outlets jumped on a new paper in Science (sorry, paywalled) last week, and one of the authors was responsible for the paper I critiqued so harshly before, I felt compelled to read it – especially after seeing Keith Kloor’s great post on the issue. After reading it, I feel compelled to comment on it.

My response is lengthy, so for those on a time budget, I offer some takeaway points. The main post, with details, follows.

Takeaway points

  • The Hsaing, et al paper in Science makes claims that are much more nuanced than what is represented either in the press releases from Princeton and Berkeley, or in many of the media stories (especially the big outlets) about them.
    • The actual findings of the paper simply reiterate long-held understandings of the connection between climate change and conflict
    • These findings are, in summary:
      • The climate affects many arenas, including food supplies, markets, and employment. The climate affects each of these in different ways in different places.
      • Climate-related changes in one or more of those arenas could (but do not always) affect rates of conflict
      • Even when climate-related changes to these arenas do provoke conflict, the provocation can occur in any number of locally-specific ways
      • Therefore, all we can really say is that climate change might affect rates of conflict in different ways in different places in the future
    • We already knew all of this
      • The authors’ claims (as stated in this press release from Princeton) that this study was necessary to establish a causal relationship between changing climate conditions and conflict is based on a straw man of “people” who have been skeptical of “an individual study here or there.”
      • Much of the literature, and those working on this issue, have long accepted the idea of a complex link between changing climate/weather conditions and conflict. The real question is that of how climate variability and change contribute to rates of conflict.
      • The paper does not answer this question
  • The quantification of increased risk of conflict in the paper is problematic, as the authors appear to assume a constant relationship, year-to-year or season-to-season, between climate conditions and their influence on various drivers of conflict.
    • This assumption has long been discarded in studies of food security and famine
    • This assumption likely introduces significant margins of error to the findings of this paper regarding increased risk of conflict associated with climate change
  • The paper does not address the real research frontier in the study of conflict and climate change because it does not further our understanding of how climate variability and change result in increased risk of conflict
    • To the author’s credit, the paper does not purport to explain how observed climate variability and change are translated into conflict
    • The paper merely summarizes existing literature exploring this issue
    • The findings of the paper do not present an opportunity to adjust policy, programs, or diplomacy to avoid future conflicts, as they do not identify specific issues that should be addressed by such efforts.
    • To some extent, this makes the critique under #2 above irrelevant – the “risk of conflict” figures were never actionable anyway
  • Media coverage of this paper amounts to much ado about nothing new

 

Main Post

The Hsaing, et al paper bears little resemblance to the media stories written about it. It makes very measured, fairly contained claims about climate change and conflict that, if represented accurately in the media, probably would not have made for interesting stories. That said, the article deserves critical attention on its own terms so we can understand what, if any, new information is here.

First, I want to start with the good in this paper. This is a substantially more careful paper than the one I critiqued before, both with regard to its attention to existing work on the subject and to the claims it makes about the connections between climate change and conflict. The authors deserve credit for noting the long history of qualitative work on conflict and the environment, a literature often ignored by those conducting large, more quantitative studies. They also should be commended for their caution in identifying causal relationships, instead of basic correlations.

In my opinion, this much more measured approach to thinking about climate change and conflict has resulted in more nuanced claims. First, as the authors note:

“Social conflicts at all scales and levels of organization appear susceptible to climatic influence, and multiple dimensions of the climate system are capable of influencing these various outcomes.”

But later in the paper, the authors temper this point:

“However, it is not true that all types of climatic events influence all forms of human conflict or that climatic conditions are the sole determinant of human conflict. The influence of climate is detectable across contexts, but we strongly emphasize that it is only one of many factors that contribute to conflict.”

And in the end, the big summary (my emphasis):

“The above evidence makes a prima facie case that future anthropogenic climate change could worsen conflict outcomes across the globe in comparison to a future with no climatic changes, given the large expected increase in global surface temperatures and the likely increase in variability of precipitation across many regions over coming decades”

Every bit of this is fine with me. Indeed, had the reporting on this paper been as nuanced as the claims it actually makes…there probably wouldn’t have been any reporting on the paper. The hook “the climate affects a lot of things, and some of those things could affect rates of conflict, so climate change might affect rates of conflict in different ways in different places in the future” isn’t exactly exciting.

And this is where I have to critique the article. My critique has two sides, one intellectual and one from a policy perspective. They are closely linked and blend into one another, and so I present them both below.

Intellectually, I fundamentally question the contribution of this paper. In a nutshell, there is almost nothing new here. Yes, there appear to be some new quantifications of the risk of conflict under different climate situations, and I will return to those in a minute. But overall, the claims made in this paper are exactly the claims that have been made by many others, in many other venues, for a while. For example, the Office of Conflict Management and Mitigation at USAID put out a report back in 2009 (yes, four years ago) that reviewed the existing literature on the subject and came to more or less the same conclusions as this “new” article.  So I was a little bothered by the Princeton press release for this paper in which quoted lead author Solomon Hsaing several times, because I think his justification for the paper is based on a straw man:

“We think that by collecting all the research together now, we’re pretty clearly establishing that there is a causal relationship between the climate and human conflict,” Hsiang said. “People have been skeptical up to now of an individual study here or there. But considering the body of work together, we can now show that these patterns are extremely general. It’s more of the rule than the exception.

I’d love to know who the “people” are who think there is no relationship between climate conditions and human conflict. Critiques of the study of this connection (at least credible critiques) have not so much argued that there is no connection, but that the connections are very complex and not well-captured in large-scale studies using quantitative tools.  So, when Hsaing goes on to say:

“Whether there is a relationship between climate and conflict is not the question anymore. We now want to understand what’s causing it,” Hsiang said. “Once we understand what causes this correlation we can think about designing effective policies or institutions to manage or interrupt the link between climate and conflict.”

…he’s really making a rather grand claim for an article that just tells us what we already knew – that there is a connection between climate conditions and human conflict. And he is burying the real lede here…that the contribution we need, now, is to understand how these causal relationships come to be. This argument for “where we should go next” is also a bit grand, seeing as everyone from academics to USAID’s Office of Conflict Management and Mitigation have been conducting detailed, qualitative studies of these relationships for some time now because we already knew a) that there were relationships between climate and conflict and b) we needed to establish what caused those relationships.

Second, I feel this article suffers from a critical methodological flaw, in that the authors never address the variable coupling of climate outcomes and changes in even those drivers of conflict identified in the literature. For example, it is not at all uncommon to have market shifts take place seasonally, in a manner that can be either coupled or uncoupled with shifts in climate: that is, sometimes a bad rainy season damages local harvests and drives market prices for food up, while other times it could be a great rainy season and a very productive harvest, but factors on regional or global markets could still generate price spikes that end up limiting people’s access to food. In both situations, the people in question would experience a food stress, one closely linked to climate variability, and the other experience a food stress uncoupled from climate. This is why, as I argued back during the Horn of Africa Famine, drought does not equal famine. Famines are far more highly correlated to market conditions than climate conditions. Sometimes climate events like a failed rainy season can trigger a famine by pushing markets and other factors over key thresholds. However, we’ve also had famines in times of normal or even favorable climatic conditions for agriculture.

Simply put, the authors appear to assume a constant relationship between a conflict driver like access to food and the local/regional/global climate. To be fair, this seems to be a pretty prevalent assumption in the literature.  But to the point, this is a bad bet. As best I can tell, the authors have not managed to address the intermittent coupling of conflict drivers like access to food and markets with climatic conditions in their analysis. This, to me, casts significant doubt on their findings that risk of inter-group conflict will rise 14% at one standard deviation of temperature rise – in short, this is far too precise a claim for a study with such large margins for error built into its design.  My suspicion here is that the margin of error introduced by this problem is probably larger than their analytical findings, rendering them somewhere between weak and meaningless. And this, to be honest, was the only really original contribution in the paper.

Third (as I begin to pivot from intellectual to policy critique), while the authors claim to have focused on causal relationships (a claim I think should be tempered by my methodological concerns above), they cannot explain those relationships. I’ve made this point before: in the social sciences, causality is not explanation. Even if we accept that the authors have indeed established causal relationships between climate variability and change and the risk of conflict/rates of conflict, they do not know exactly how these changes in climate actually create these outcomes. This is clear in the section of the paper titled “Plausible Mechanisms”, in which the authors conduct a review of the existing literature (much of which is qualitative) to lay out a set of potential pathways by which their observed relationships might be explained. But nothing in this study allows the authors to choose between any of these explanations…which means that all the authors have really accomplished here is to establish, by different means, exactly what the qualitative literature has known for a long time. To repeat:

  1. The climate affects many arenas, including food supplies, markets, and employment. The climate affects each of these in different ways in different places.
  2. Climate-related changes in one or more of those arenas could (but do not always) affect rates of conflict
  3. Even when climate-related changes to these arenas do provoke conflict, the provocation can occur in any number of locally-specific ways
  4. Therefore, all we can really say is that climate change might affect rates of conflict in different ways in different places in the future

We already knew all of this.

At this point, allow me to pivot fully to my fourth critique, which comes from a policy perspective. People tend to see me as an academic, and forget that I served as the first climate change coordinator for the Bureau for Democracy, Conflict, and Humanitarian Assistance (DCHA) at USAID. I was Nancy Lindborg’s first climate advisor – indeed, it was in this role that I found myself first dealing with issues of conflict and climate change, as I was responsible both for briefing my Bureau’s leadership on these issues and guiding the programming of the Bureau’s dedicated climate change budget (some of which I directed into more research on this topic). In short, I do know something about policymaking and the policy environment. And what I know is this: this paper gives us nothing actionable to address. Even if I accept the finding of 14% greater risk of intergroup conflict at one standard deviation of temperature increase, what am I supposed to do about it? Without an explanation for how this temperature rise produces this greater risk, I have no means of targeting programs, diplomacy, or other resources to address the things that create this greater risk. In short, this paper tells me what I already knew (that climate variability and change can contribute to conflict risk) without giving me anything concrete I can work on. If I were still briefing Nancy, my summary of this paper would be:

  1. There is nothing new in this paper. Its key findings are those of CMM’s (four-year-old) report, and are already well-established in the literature
  2. The paper does not provide any new information about how climate change and variability might contribute to increased conflict risk, and therefore presents nothing new that might serve to guide future policy, programs, or diplomacy
  3. I have methodological concerns with the paper that lead me to believe that the rates of increased risk of conflict reported in this paper are likely stated with too much confidence. These rates of heightened risk should not be cited until put under significant scrutiny by the academic and policy community*.

In summary, the supportable parts of this paper are nothing new – it is a reasonable summary of the issues with establishing a connection between climate change and conflict, and a decent (if truncated) review of the existing literature on the subject (I’d suggest that a real review article of this subject would have to go wider and look at the conflict and environment literature more broadly). But it doesn’t say anything new that really bears up to scrutiny, and even if the “risk of conflict” figures are correct, the paper provides no information that might guide policy, programs, or diplomacy in a manner that could avoid such conflicts. For that information, we have to return to the qualitative research community, which has long espoused the same general findings as those in this paper.

The press releases from Princeton and Berkeley, and the more hyped of the media coverage we’ve seen around this paper (likely driven by those press releases) is much ado about nothing new.

 

 

*In my third point I am indeed taking issue with the peer review process that brought this paper to publication. I believe that Science wanted this paper for the same reason Nature wanted the last one: headlines. Let’s see how the findings here stand up to serious scrutiny.

Eric Cantor’s recent call to shift funding from the social sciences to the hard sciences (“Funds currently spent by the government on social science — including on politics of all things — would be better spent helping find cures to diseases”) reflects a profound misunderstanding of the complementary role these two epistemological arenas play.  John Sides has covered a range of reasons why the social sciences should not be seen as superfluous to needs, all centering on the fact that social phenomena are central to human well-being and happiness.  As he notes:

My problem with this laser focus on the hard sciences and on medicine is that it pretends that people’s quality of life simply depends on physical phenomena—how fast computers are or how much their knee hurts and so on.   That’s simply not true.  Much of people’s happiness—indeed, including whether they have access to computers or can endure a physical malady—depends on social phenomena.

Even more compelling is Mark Slouka’s 2009 article in Harpers, which offers one of the clearest defenses of the humanities I have ever read: simply put, without the humanities it is very difficult to be a functional citizen in a democracy (but in their absence it is very easy to produce a docile population of workers).

Let me take Slouka’s argument past what really read like something of an either/or tradeoff between the humanities and what he called “mathandscience” and toward a point of complementarity here: simply put, science is a way of seeing the world that enables particular understandings of that world. Science has facilitated spectacular changes in the way we live, from household technologies to medical advances.  But science is but one way of seeing the world, one that does not tell us what we should do, or what else we should do.  Those questions are the province of ethics, justice, and empathy. Science is poorly equipped to address any of these.

This is why science and technology require the social sciences and humanities. They help us separate what is possible in the world from what should be done in the world. Remember, history is littered with examples of highly rational, scientific projects that killed huge numbers of people in the name of a greater good or a logical goal (anyone remember the Soviet collectivization of agriculture under Stalin? How about the far less brutal, but still problematic ujamaa collectivization in Tanzania?). Without the arts, humanities, and social sciences, we are left with a tool (science) and no guidance about how to use it.  Further, the growing field of science and technology studies shows that the capacities of particular technologies, in and of themselves, tell us little about who will adopt them and why. Trevor Birkenholtz’s work in India, for example, demonstrates that farmers continue to use tubewells, even though they know that this practice contributes to groundwater depletion, because the use of tubewells is closely bound up in one’s identity as a good and prosperous farmer.  Without such insights, how can we work with farmers in this region to identify locally-appropriate alternative water-supply technologies?

Cantor, and those like him, live in an odd world where technologies and commodities are social goods unto themselves with universal and obvious value. Existing social scientific work already demonstrates this to be untrue. Defunding such work will not make his beliefs more true, it will just make it harder to make the world a better place with the scientific tools we have and will develop in the future.

One of the things I am (not so) fond of saying is that when it comes to climate, I am not really worried about what I do know – it’s the things that I don’t know, and cannot predict, that worry me the most. The climate displays many characteristics of a nonlinear complex system, which means that we cannot assume that any changes in this system will come in a steady manner – even a fast but steady manner. Instead, the geologic record suggests that this system changes in a linear manner (i.e. slowly warms up, with related shifts in sea level, precipitation, wind patterns and ocean circulation) up to a certain point before changing state – that is, shifting all of these patterns rather dramatically into a new state that conveys the extra energy in the atmosphere through the Earth system in a different manner. These state changes are frightening to me because they are highly unpredictable (we are not sure where the thresholds for these changes are) and, at their worst, they could introduce biophysical changes like increased temperature and rates of evaporation and decreased rainfall with such speed (i.e. in a decade or two, as opposed to over centuries) that the rate of change outpaces the capacity of biomes to adapt, and the constituent species in those biomes to evolve. This is not some random concern about biodiversity – people seem to forget that agricultural systems are ecosystems; radically simplified ecosystems, to be sure, but still ecosystems. They are actually terribly unstable ecosystems because they are so simple (they have little resilience to change, as there are so few components that shifting any one of them can introduce huge changes to the whole system), and so the sort of nonlinear changes I am describing have particular salience for our food supply. I am not a doomsday scenario kind of guy – I like to think of myself as a hopelessly realistic optimist – but I admit that this sort of thing worries me a lot.

So, to put this another way: we are running like hell down a long hallway toward an open door into a darkened room. We can’t see what’s in the room, and it is coming up fast. Most normal people would probably slow down and enter the room cautiously so as to avoid a nasty collision with something in the dark. When it comes to climate change, though, our current behavior is akin to running right into that room at full speed and hoping with all our might that there is nothing in the way.

This is a really, really stupid way of addressing the challenge of climate change.

The good news on this front is that we are starting to see the emergence of a literature on the early warning for these tipping points. I had a post on this recently, and now the July issue of Nature Climate Change has a review article by Timothy Lenton on early warning of tipping points. It is a really excellent piece – it lays out what we are currently doing, shows the limitations of what we can do, points to significant challenges both in the science and in the policy realm, and tries to chart a path forward. I think Lenton comes in a bit science-heavy in this piece, though. While he raises the issues of false alarms and missed alarms, he spends nearly all his time looking at methods for reducing the occurrence of these events. This is all well and good, but false and missed alarms are inevitable when trying to predict the behavior of complex systems. Yes, we need more and better science, but we also need to be thinking about how we address the loss of policymaker confidence in the wake of false alarms or missed alarms.

To get to this point, I think we need to be looking to arenas where people have a lot of experience communicating levels of risk and the importance of addressing that risk – the insurance industry. Most readers of this blog will have some form of insurance – be it health insurance, life insurance, car insurance, etc. I have all three. Every month, I pay a premium for a product I sincerely hope I never have to use. I’d rather hang on to that money (with a family the size of mine, it gets steep), but the cost of a catastrophic event in any of these areas would be so high that I gladly continue to pay. We need to encourage the insurance industry (they are already working on this issue, as they stand to lose a hell of a lot of money unless they can get their actuarial tables adjusted) to start communicating their sense of the likely future costs of climate change, and the costs associated with potential state changes – and do so in the same way that they sell us insurance policies. Why do we have scientists working on the marketing of our ideas? We are not trained for this, and most of my colleagues lack the salesman’s charisma that the climate change issue so desperately needs.

It’s time for a serious conversation about how science and the for-profit risk management world can start working together to better translate likely future climate impacts into likely future costs that everyone can understand. Science simply does not carry the weight we need in policy circles – the good data and rigorous analysis that are central to scientific legitimacy are, in the policy realm, simply seen as means to achieving a particular viewpoint, not an ever-improving approximation of how the world works. Until the climate science (and social science) community grasps this, I fear we will continue to talk past far too many people – and if we allow this to happen, we become part of the problem.



It’s been a while since I focused on the environment side of the whole “global change” thing that this blog is supposed to be covering . . . at least directly.  Pretty much everything we do in development is connected to the environment – indeed, of late I have been referring to climate change as development’s Kevin Bacon while at work: I can get you from climate change to a development challenge, or vice versa, in three steps or less.  But I have not been writing much on the subject directly.

However, thanks to Garry over at Resilience Science, I’ve just read a really interesting article in Science (and a nice counterpoint to the recent bin Laden ambulance chasing in that journal) by Steve Carpenter and a bunch of others on Early Warnings of Regime Shifts in ecosystems.  For years, I have been teaching my students about the challenges of global environmental change, and trying to impress upon them that the part of these changes I find most worrying are the parts that are hardest to predict – the thresholds when particular biophysical systems might make sudden, discontinuous transitions to new states.  What has worried me, and I think much of the global change community, the most is the fact that we are not sure where these thresholds are, nor are we sure what it looks like when we approach one.  Thus, there is a pervasive concern within the community that we won’t know we’ve crossed a threshold or done something irreversible.

Carpenter and his co-authors, however, tested the hypothesis that “catastrophic ecological regime shifts may be announced in advance by statistical early-warning signals such as slowing return rates from perturbation and rising variance” by artifically inducing a regime shift in an aquatic food web (Carpenter is a limnologist – he does lakes, as it were) while monitoring a nearby similar lake as a control.  Their finding: they could see statistical warnings of an impending regime shift for more than a year before it occurred, validating their chosen early warning indicators (chosen from previously constructed understandings of the food web in question, and a bit much to synthesize here).

That there might be early warning indicators, or that the variables chosen by Carpenter, et al served as useful early warning indicators for regime change in this particular system are not terribly surprising.  What is interesting, though, is that the authors were able to demonstrate in a real-world (experimental) context (as opposed to desk theorization) that the early warning signals of regime shift are in fact detectable and measurable.  Granted, this is for a small, bounded food web – but the demonstration is important in a much wider way.  If we can find early warning indicators for regime shift in a small food web, there is no reason why we cannot find indicators for other complex systems – we can find a lot more early warning indicators of the discontinuous changes we fear, and in enough time to possibly address those changes before they occur.

But one big caveat here: this study did not reveal the actual mechanisms of regime shift.  As the authors note:

The precise mechanism of the nonlinear transitions is not known for our experiment; it could be one of the processes proposed in the literature, or something else. These early warning signals are expected to occur for a wide class of nonlinear transitions (7). Even though the mechanism is not known, manipulation of an apex predator triggered a nonlinear food web transition that was signaled by early warning indicators more than a year before the food web transition was complete. Thus the early warning indicators appear to be useful even in cases where the form of the potential regime shift is not known.

It seems to me that there is a serious risk of conflating correlation and causation here – that the authors got a bit lucky in this experiment, but that in other systems without an adequate understanding of the mechanisms of change, false correlations could cause us to lose the signal of regime shift in the noise of inappropriate data points.  I’m not sure how, or if, they intend to address this . . . but I think they will have to, if we are to usefully apply this to our food-producing ecosystems in a manner that allows us to think about sustainable development and food security in a meaningful way . . .

Inquiry is dead when the flagship journal Science starts ambulance-chasing . . . but hey, its Osama bin Laden week in all media, so I guess it should be of no surprise that they are running a story on three-ish year old efforts to get a sense of where bin Laden might be hiding.  To their credit, the folks at UCLA are hardly crowing – it was a student project, and Thomas Gillespie, the faculty leader of the project openly noted “It’s not my thing to do this type of [terrorism] stuff,” and made it clear that he had no intention of shifting from his biogeographic interests:

“Right now, I’m working on the dry forests of Hawaii where 45% of the trees are on the endangered species list,” says Gillespie. “I’m far more interested in getting trees off the endangered species list.”

I’m waiting for the gentlemen over at floatingsheep.org to weigh in on this particular project – they are much more qualified to comment on the substance of the study.  However, I applaud Gillespie for refusing to get caught up in the hype.  Sadly, I’m sure some of my disciplinary colleagues will want to trumpet this as an example of how useful geography is, and why it should get more attention.  Because, you know, we’ve just recently shaken off the colonial origins of our discipline, where we proved our usefulness by mapping local populations and resources to facilitate their control, and lord knows we wouldn’t want to put that sort of thing behind us. As one of my colleagues in grad school once pointed out (tongue-in-cheek) after listening to some of our colleagues complain about how some engineering and science departments had much larger budgets, “if we were willing to help kill people, we wouldn’t have this problem.”

And people wonder why I get itchy about the militarization of aid and development.