Open The Echo Chamber

Friend/colleague/journalist Keith Kloor has a very interesting piece about the problematic character of some research that recently purported to cast doubt on the health safety of GMOs.  The piece is an excellent effort to push back against crap science and crap evidence in the GMO debate, and stands as an interesting example of a general need to critically evaluate “scientific” claims about any number of issues from the harmful character of vaccines to those would deny anthropogenic climate change.  However, the piece, in making a strenuous argument for evidence, overreaches in its conclusions and too-quickly dismisses the ecological issues associated with GMOs – exactly the opposite outcome Keith was seeking.

In general, I agree with Keith that the screaming about the health impacts of GMOs has greatly outstripped the evidence, but I take issue with the idea that GMOs have been largely proven safe all around.  There remain significant questions about ecological impacts that have not attracted a solid scientific consensus (i.e the impact/cost/tradeoffs of gene flow between GMOs and the surrounding ecology, pest resistance, etc. –.  The evidence base in this area is pretty small, and ecological systems are very diverse and complex, so the levels of uncertainty here are fairly epic (Pamela Ronald, whose article Keith references approvingly, even notes that the ecological impacts of GMOs are still an open issue).  Honestly, Keith is too dismissive of this challenge:

Some of these folks are worried about new genes being introduced into plant and animal species. But humans have been selectively breeding plants and animals pretty much since we moved out of caves, manipulating their genes all the while. The process was just slower before biotechnology came along.

Modern genetic engineering does not have all that much to do with selective breeding.  We are doing things in GMOs that could not occur in nature, which is rather different then applying what amounts to a modified evolutionary force (human selection of seeds to plant the next season or animals to breed) in the selection of traits within certain crops.

Another big concern that has been widely reported is the “rapid growth of tenacious super weeds” that now defy Monsanto’s trademark Roundup herbicide. That has led farmers to spray their fields with an increasing amount of the chemical weed-killer. Additionally, some research suggests that other pests are evolving a resistance to GMO crops. But these problems are not unique to genetic engineering. The history of agriculture is one of a never-ending battle between humans and pests.

Again, this is a bit off to one side.  Yes, agriculture has always been about the radical simplification and management of ecosystems in a complex world, and therefore is indeed an endless battle for control between farmer and surrounding ecology.  What this elides, though, is the fact that some of these GMOs have made things harder for us in the long run.  It is hard to sell this as a good, or even neutral, outcome.

I raise this issue because of its implications for conversations about food security. People (some, but not all) working in food security tend to be a bit dismissive of the ecological concerns surrounding GMOs, loosely equating them with the human consumption concerns that have largely been disproven at this point.  The response to insistent questioning about ecological impacts is the argument that, on balance, GMOs have done more good than harm and/or that GMOs are necessary for global food security going forward.  The “more harm than good” argument seems to me based on evidence that is hardly complete, and often references a glowing, startlingly unproblematic vision of the Green Revolution.  The “we need GMOs” is a crisis narrative of food production that drives current and future claims about their value.  At the moment, much of the food security world is arguing that there is a dire food production crisis either coming or just arrived (embodied in the 2008 and 2011 food price spikes), and therefore we cannot wait for the burden of proof surrounding any issue associated with GMOs – we must act now!  This crisis narrative effectively reduces anyone who raises objections to an all-out push for more production to some sort of monster who would rather let innocents starve than engage with the messiness of the real world. Basically, precaution is conflated with timidity, making those who show concern over the ecological impacts of GMOs into the food security equivalents of Neville Chamberlain.

This would perhaps be true…if, in fact there were a global production crisis and we needed GMOs to feed everyone. This is a much harder thing to sell, however, when one realizes that the world produces roughly twice as much food as needed to feed everyone adequately each year.  There isn’t a global production crisis (though there are local production crises that arise from complex causes and need to be addressed in locally-specific ways). I’ve said it before, and I will say it again: globally, there is plenty of food. Now, and likely for quite some time. Period.

In the twitter conversations that emerged around Keith’s article, Robert Wilson, a mathematical ecologist, argued for “an adjusted precautionary principle where we consider the risks of inaction, as well as action.”  This strikes me as both reasonable and necessary when discussing food security policy and programs these days.  If we did this, though, I think we would find that the risks of inaction are, in fact, substantially smaller than the risks of action – this is not to say there are necessarily huge risks to action, just that when you are producing twice the calories needed already, the risk from doing nothing is, at least in the short term, pretty small.  There is time to test more crops, more widely, under more conditions, before we arrive at any production crunch.  There is no need to rush.

In short, we need to be sure that our efforts to push back against crap science (which is really what Keith was doing in his article) don’t overreach and inadvertently empower narratives and arguments that are not supported by the evidence – exactly the opposite outcome Keith sought.

As I mentioned a few posts ago, I am working through James Scott’s The Art of Not Being Governed (my endorsement is in the linked post).  In the course of my reading, I have been thinking about what Scott calls the State Accessible Product, which he sets in contrast to the Gross Domestic Product.  To Scott’s thinking, the states/kingdoms he is discussing in Southeast Asia were motivated not to maximize the value of all goods and services in their realms, as such maximization might include the production of goods that could not be transported/taxed/otherwise used to enrich the state.  Instead, it was in the state’s interest to maximize the production of things it could see, count and move – in other words, to push the growth of a State Accessible Product.  Two things resonate for me about this idea:

1) It scales. Just as states pushed for the production of SAP, so too the households I I discuss in Delivering Development tend to divide up livelihoods roles and activities in a manner that maximizes not individual well-being, but activities that only make sense when bundled with the activities of other members of the household – a sort of Household Accessible Product.  In an uncertain economy and environment, it makes no sense to focus one’s entire agricultural production on market sale, or to focus entirely on subsistence reproduction of the household.  Yet this is just what we see men (playing the former role) and women (playing the latter) doing in some of the households I examined in Ghana.  They do this for a lot of complex reasons, but certainly there is something to the idea that these roles force the members of the household into the production of a HAP that certainly does not maximize all possible production and income, but does a lot to reproduce social roles and social stability.

2) It explains why my argument that a lot of farmers on globalization’s shoreline strategically deglobalize was both surprising and, at least to some people, threatening: the opting in and out of global markets is exactly the sort of thing states fear, as it means that the production of these farmers goes in and out of legibility from year to year – making it hard to extract value from that production. In short, there is a GDP that is not coterminus with an SAP along most of globalization’s shoreline – and that non SAP production is critical to the well-being of those engaged in those activities.

It strikes me that a key question here is whether or not our focus on governance in development has led us to inadvertently emphasize activities, projects and programs that render greater and greater percentages of GDP as SAP – certainly, without access to the financial resources produced by the control of a SAP, states are in a weak position.  But if many of the activities that actually keep people alive on a day-to-day basis are non-SAP activities, what are we to do?  Are we to wipe out/make legible these activities so the state can profit from them?  If we do, are we going to enhance the vulnerability of the populations whose livelihoods we alter?  Is the enhancement of vulnerability an appropriate trade-off for the creation of a state-legible economy?  Can addressing vulnerability and building a strong state be made to rhyme at all?

In my previous post, I objected to the way in which Tyler Cowen’s recent NYT blog post pushed the dominant “crisis of production” narrative in discussions of food security.  In my opinion, the recurrence of this problematic claim in various popular outlets has a lot to do with people’s relatively surficial understanding of food security and the causes of hunger in the Global South.  For some reason, development seems to lend itself to dilettantism…at least in part, I suspect, because people assume that the global poor are so bad off that any new ideas would be an improvement on what is there.  Of course, there is also the subtle, durable assumption that poor people (especially of darker skin tones) somehow (re)produce their problems because they don’t think rationally/clearly/etc*.  Such arguments fall apart when they are tested with actual evidence, but most op-eds and policies have nothing to do with evidence…

This problem extends beyond how we talk about the poor themselves to how we think about the governments under which they live.  While governance (not the same as government, folks – please try to remember this) is really important to development outcomes, it is not everything…and government (as in the formal rules and structures of governance in a particular place) can be even less important, as many of the global poor live beyond the reach of the state.  So blaming the state and its policies for hunger can be a pretty tricky proposition. When Cowen, in an offhand way, wades into the role of the Malawian government as an illustration of how his presumed production shortfalls are exacerbated by problematic government policies, his lack of understanding of the African context becomes clear:

many African nations have unhelpful policies toward agriculture. Malawi, for instance, subjects corn to periodic export and import restrictions as well as to price controls, all of which thwart development of a well-functioning market. When market speculators save corn in anticipation of greater scarcity, they may be punished by law. These restrictions of market incentives exacerbate the basic supply problems.

First, Cowen cherry-picks Malawian government actions to make this point.  While price controls and import/export restrictions have been used, there is another side to Malawian intervention in the markets: the subsidization of inputs to boost overall farm productivity.  As a result, he ignores the near-perfect correlation between the years when the government intervenes in input markets – effectively, when the Malawian government subsidizes fertilizers – and the years when Malawi is a net food exporter to the extent that it can pay for the entire subsidy several times over (this correlation has proven very durable and very vexatious to some of the more theologically-inclined free marketeers out there).  In the case of Malawi, some market intervention, however distortionary, actually does work to ensure adequate food production within the country each year. Which gets to a much larger point: the Malawian government is doing this not out of ignorance or irrationality, but because it is being responsive to citizens whose short term needs are so dire that to take a long-term only view would result in mass morbidity, if not mortality, in the short run.

For example, in a priori assuming that Malawi’s decision to punish market speculators when they “save corn in anticipation of greater scarcity” (one person’s “saving” is another’s “hoarding”), Cowen fails to parse between the needs of an efficient market (a means of transmitting future price situations into current pricing decisions) with human needs (a means of obtaining adequate food such that members of the household do not die) – in most places I work, there is a large disjoint between the two.  It is this disjoint that the government of Malawi, and indeed many governments around the world must negotiate.  It is this disjoint, and its attendant reality, that is Cowen’s second major problem, as he doesn’t really understand it.  This reality has two parts:

1) Yes, in the long run markets can transmit information about pricing and preferences that can lead to more productive and useful decisions, but in contexts where people are living on a dollar a day, their margins for error are small and their ability to wait for markets to work things out is limited.

2) There is a presumption that the anticipatory price signal will result in actions to address the problem before the shortage actually hits.  However, the causes of shortage generally extend well beyond the management capacity of any single state.  In short, transmitting shortage signals into the present only serves to prolong the challenges that the Malawian poor are going to face, without producing any effective policy or market response because there is no government capacity to respond.  In short, why transmit the emergency into the present when you are going to need help to address it now or in the future?  This is why many African states punish hoarding…though they could be looked upon as comprising a de facto futures market, hoarders transmit not just information, but shortage into the market and onto very vulnerable populations earlier than would otherwise be the case, undermining safety and security sooner and to no good end.  Given the option of an efficient market populated by a lot of dead people and an inefficient, or even broken market populated by live people, most African states are going with the latter.  Until someone sorts out how to set up functioning markets near-instantly, builds enough financial resilience into African livelihoods to weather this sort of market behavior, or builds the financial and infrastructural capacity of African states to a point they can manage this short of shock without external assistance (or some combination of the three), states will continue to be forced into this sort of decision, and will make the same choice.  I am not convinced that the manipulation/corruption of markets Cowen describes is a cause of hunger as much as a symptom of a hugely problematic global political economy that no one small country can effectively manage.

In short, the situation in Malawi is very common in sub-Saharan Africa.  For most countries, the issues I raise above have been in play since independence.  The typical African country is dealing with a set of pressures that make straightforward economic decision-making nearly impossible – from state-building to market-building, these countries cannot just make economic decisions, they must make political-economic decisions that reflect the immediate reality around them.  Government is easy…until you actually have to govern.

*This is not to absolve all poor people of all responsibility for their situations. The global poor, like everyone else, are human – they are subject to emotions, biases, prejudices, etc. that sometimes do cause major problems for their well-being.  However, it has been my experience that this is not a dominant cause of the problems of poverty…mainly because if these problems were exacerbated more than they were helped by the efforts of the global poor, we’d have a lot fewer poor people because they would mostly have died. The global poor make fantastically difficult decisions about the allocation of scarce resources every day with a shocking degree of success…something we overlook at our peril.

While I appreciate the overall sentiment behind “World Hunger, the Problem Left Behind”, Tyler Cowen’s op-ed piece in the New York Times earlier this week, in trying to put forth an important message he reinforces really problematic understandings of hunger. Cowen, like so many others, continues to frame food security and hunger issues as a crisis of production and productivity.  Citing Michael Lipton, Cowen writes:

Rwanda and Ghana are gaining, but…most of the continent is not. Production and calorie intake per capita don’t seem to be higher today than they were in the early 1960s. It remains an issue how Africa’s growing population will be fed.

First, production and caloric intake per capita are not necessarily directly linked.  As I have observed elsewhere on this blog (and as Hans Herren and I argued at the New America Foundation/Slate event Feeding the World While the Earth Cooks), there is a massive loss problem in most food systems that more than accounts for most food shortages in the world today.  Considering the absolutely remarkable explosion of the African population since 1960, when it held about 300 million people and today, when it holds 870 million, holding the line on per-capita production means that people effectively tripled the agricultural output of the continent.  But when you lose 40% of that production between the farm gate and market, you are going to get a disjoint between production and caloric intake that has nothing at all to do with the skills of the farmers or their on-farm circumstances.  Are there ways to augment production and improve it?  Probably, but this is not the central problem at the moment, nor is it the low-hanging fruit.  Don’t reengineer the ecosystem, just fix the road!

It would be good to see some serious discussion of food insecurity that did not center on agricultural productivity.  It seems that the urge is to start from the hardest end of the process to fix, perhaps because it is in the fields that food goes from abstract to concrete. It is hard to argue that the biggest problem is loss in the supply chain, because that is harder for people to see.  We need more posters of rotting loads of food on the back of trucks, and less pictures of dusty fields with stunted crops.  Maybe that would start to shift the narrative?

Ben Leo at ONE.org (formerly of CGD) put forth an intriguing proposal recently on Huffington Post Impact: It’s Time to Ask the World’s Poor What They Really Want.  In short, Ben is trying to argue that the current top-down definition of development goals, no matter how well-intentioned, is unlikely to reflect the views of the people these development goals are meant to benefit.

Hear, hear.  I made a similar point in Delivering Development. Actually, that sort of was one of the main points of the book.  See also my articles here and here.

But I am concerned that Leo is representing this effort a little too idealistically.  Just because we decide to ask people what they want doesn’t mean that we will really find out what they want.  Getting to this sort of information has everything to do with asking the right questions in the right way – there is no silver bullet for participation that will ensure that everyone’s voices will be heard.  To that end, what worries me here is that Ben does not explain exactly how ONE plans to develop the standardized survey they will put out there, or how exactly they will administer this survey.  So, here are a few preliminary questions for Ben and the ONE team:

1)   Does a standardized survey make sense? Given the very different challenges that people face around the world, and the highly variable capacity of people to deal with those challenges, it seems to me that going standardized is going to result in one of two outcomes: either you ask focused questions that only partially capture the challenges facing most people, or you ask really general questions that basically capture the suite of challenges we see globally, but do so in a manner that is so vague as to be unactionable.  How will ONE thread this needle?

2)   Who is designing the survey? To my point above, what questions are asked determine who will answer, and therefore determines what you will learn.  While the information gleaned from this sort of survey is likely to be very interesting, it is not the same thing as an open participatory process – full participation includes defining the questions, not just the answers.  Indeed, I would suggest that ONE needs to ditch the term participatory here, as in the end I fear it will be misleading.

3)   How will you administer the survey? Going out with enumerators takes a lot of time and money, and is subject to “investigator bias” – that is, the simple problem that some enumerators will do their job in a different manner than others, thus getting you different kinds/qualities of answers to the same questions.  On the other hand, if you are reliant on mobile technology, how will you incentivize those rural populations with mobile handsets to participate?  If you can’t do this, you will end up with a highly unrepresentative sample, making the results far less useful.

This is not to dismiss the effort Ben is spearheading – indeed, it is fantastic to see a visible organization make this argument and take concrete steps to actually get the voices of the global poor into the agenda-setting exercises.  However, this is not a participatory process – it is, instead, an information-driven process (which is good) that is largely shaped by the folks at ONE in the name of the global poor.  If ONE wants this to be more than information-driven, it needs to think about how it is going to let a representative sample of the global poor define the questions as well as the answers.  That is no easy task.

In all sincerity, I am happy to talk this through with anyone who is interested – I do think it is a good idea in principle, but execution is everything if you want it to be more than a publicity stunt…

I am in the midst of (finally) reading James Scott’s The Art of Not Being Governed.  The book is a fascinating exercise in anarchist political geography – basically, examining how state power was limited by/shaped by various geographic factors in upland Southeast Asia.  As with many Scott books, it is a readably huge, sweeping view of a long timespan and a lot of diverse people, but thus far it hangs together well.

I was struck by a particular passage in chapter 3 that speaks rather directly to a lot of food security/agricultural development work being done in the world today, not least under Feed the Future.  The passage comes in a section where Scott is reviewing the historical efforts of various tenuous states to consolidate their control over the surrounding population, but contains an important message for those who are pushing the newest era of agricultural modernization (I call it part II, but there are compelling arguments for this being part III, or IV, depending on how you parse the history of development):

At about the same time as the Peloponnesian War, the early Chinese state was doing everything in its power to prevent the dispersal of population. Manuals of statecraft urged the king to prohibit subsistence activities in the mountains and wetlands “in order to increase the involvement of the people in the production of grain.” . . . The objective of this policy was, it seems, to starve the population into grain farming and subjecthood by separating them from the open commons.

There is no leap at all from this ancient Chinese case to what the inadvertent outcomes of much of our “efficiency-driven” food security and agricultural development work might do.  By compelling greater and greater market integration, and doing so through a focus on fewer and fewer crops, we are effectively closing the commons and prohibiting/constraining subsistence activities among the affected populations.  The result, in the best case, is improved agricultural outcomes and incomes that translate into improved well-being for all involved.  The worst case is a scenario where various marginal populations who have developed some expertise in managing the uncertainty of their particular contexts lose adaptive capacity, making them much more vulnerable to state violence and control.  Critically, these are not mutually exclusive scenarios – what works in one year or season to improve the quality of life for rural farmers might, in another year or season with different market and weather conditions, work to extend state control over marginal populations who already receive little for their status as citizens. I am sure that nobody who works on the donor side wants to be part of a campaign of state violence (at the worst) or part of a project that results in the further marginalization of poor, marginal populations (in the better to middle-case scenarios).  So, at least for Feed the Future, perhaps now would be a good time to call the Democracy, Human Rights and Governance folks (they’re just one floor up!) and have them take a look at this?

Not everyone wants a resilient population, folks.

The larger message here: incorporating people into agricultural markets is about much more than economic efficiency – there are much broader considerations, from state-level political economic issues to sub-household gender roles, that come into play when we radically rework existing agricultural systems and the livelihoods that go with them.  I seriously doubt we are doing enough to capture the wide suite of challenges that comes along with market engagement to ensure that our agricultural development programs are not enhancing vulnerability and stripping resilience from some of the most marginal people we are working with.  Output per hectare is not the only relevant metric. It’s probably not even the most relevant metric, given the fact we are not suffering from a production crisis at the global scale (despite what you hear from various outlets).

Incidentally, if you are at all interested in rural development, you should buy every book James Scott has ever written and get to reading.  Now.  Links below.

Weapons of the Weak is a classic – shame on you if you do rural development/community-based development and are not familiar with it. Or, shame on the people who taught you…

The Moral Economy of the Peasant is Scott’s most underrated text.

Domination and the Arts of Resistance is Scott’s first “meta” text – huge topic, giant sweep, really interesting.

Seeing Like A State shows up on everyone’s must read list…pretty much because you should read it.

I will be speaking about my book and research at the University of Florida on Friday as part of the Glen R. Anderson Visiting Lectureship.  Poster here:

Hope to see folks there!

So, a while back I decided to talk about how I negotiate peer review, semi-liveblogging my response to a revise and resubmit request from a pretty big development journal (see part 1, part 2 and part 3).  Well, I now have a response to my resubmission . . .

No.  To quote: “after much deliberation, the editors have reached a rather difficult decision. [The editors] feel that they cannot accept your revised paper.”

Yep, I have gone from revise and resubmit to outright reject.  This is . . . unusual, to be honest. More unusual, however, is the rationale for the rejection.  To quote from the decision:

What makes this difficult is that [the editors] recognize that you have in fact taken account of what the referees said, and have tried to accommodate their comments, but the editors feel that what has emerged from the revision process is not an appropriate paper for Development and Change.

Translation: you did what we asked, and addressed the referee comments, but in doing so you ended up with a paper that we think belongs at another journal.  Well, fair enough, this happens.  But why it is not appropriate is a little odd:

While they still believe that there is an interesting idea at the core of your paper, they don’t feel that the revisions have solved the initial problems, and they are not convinced that further rounds of revision would be any more successful. The intended contribution of the paper appears to be theoretical, but the paper hasn’t managed to work out that contribution in a way that will be accessible / comprehensible to our readers.

Soooo . . . I have an interesting paper, but the editors more or less think their readership can’t deal with the complexity of the argument.  [Note: I am disregarding the assessment that my revisons have not solved the initial problems, since they already have said that I took account of the referees’ issues – this is a contradiction I am just going to leave aside. That, and they did not show me any reviewer comments, so I have no idea what I did not resolve]  One of my colleagues has called this the oddest rejection he has ever seen.

Now, I want to be clear – the folks at the journal with whom I interacted throughout this process were very responsive and polite, and were kind even in their rejection (they were quite apologetic, actually).  I would submit to this journal again, though I admit to wondering exactly what aspect of my work might fit here, as I am confused by what they believe the capacity of their readers to be.

This, folks, is the nature of peer review – sometimes, you just have no idea what happened.  I am not privy to the internal conversations of the editorial board, and will not pretend to know exactly what happened here.  What makes this hard is that I did not receive any substantive comments on this second round of review, so I have no guidance at all on edits.  I am rereading the paper, adding a citation I had missed earlier, and making minor tweaks to the argument (the article I missed before actually strengthens the case for what I am doing in the manuscript).  I’ve sent it off to a trusted senior colleague to have a look, and to see where he thinks it might go next.  I will probably sound out the next editor in advance, just to make sure that s/he thinks the paper is appropriate before starting a long review process again . . .

Two years and counting, folks, since my initial submission.

Any editors out there interested?  Anybody?

I’ll be running my mouth about the book again at Chatham University on December 2nd.  Chatham has some very cool stuff going in sustainability and the environment (a new school!), including a new Eden Hall Campus in Richland Township, PA.  My talk will actually be out on that campus, and not in the Shadyside campus . . . directions are here.

The flyer (they’ve done a nice job on it):

Hope to see some of you there . . .

Chris Albon copied me on a retweet today from World Concern that said:

A beautiful sight: things growing in #Somalia. This is what’s possible in the #HornofAfrica. twitpic.com/7c8y24

For those not inclined to click the link, it went to this picture:

I have mixed feelings about this tweet and this picture.  On one hand, it expresses what I am sure is genuine relief from an organization that is concerned with the well-being of people living in the Horn of Africa.  On the other hand, the phrase “this is what is possible” suggests that this does not usually happen . . . except, of course, now we are in the Dayr, the October to December rainy season.  Though the Dayr is the shortest rainy season in this part of the world, wet fields and new growth do in fact usually happen right about now.  Further, the phrase “things growing in Somalia” suggests that nothing was growing before.  This was not the case – things have been growing, even in famine-struck parts of southern Somalia.  Not enough has been growing in some places, and this shortage has been compounded by all sorts of political challenges that have created a widespread problem.  Finally, there is a bit of tone to this – as if we are out of the woods in the Horn.  Well, maybe – but it will be months until a real harvest comes in, and much longer than that before accountable governance and functioning markets return, so we have a ways to go.  And given that this famine was not caused by drought (the drought exacerbated other underlying factors), the fact that we are having trouble addressing those underlying factors means the next drought (and there will be another one relatively soon) may create a very similar set of circumstances and challenges.

In summary, I believe in hope.  That is why I call myself an optimist.  But at the same time, we have to be careful about conflating hope with triumph . . . which is why I call myself a hopelessly realistic optimist.