In a change to my published programme, I thought I’d engage with a couple of posts on nitrogen recently emerging from the Breakthrough Institute. In fact the issue is quite relevant to my last post, and to the next scheduled one. For more on the regenerative agriculture issue I’ve recently discussed, I’m following the debate over Andy McGuire’s recent blog post with interest. Meanwhile, for more on ecomodernism of the Breakthrough Institute variety, Aaron Vansintjan has just published this nice little critique. Doubtless we’ll take a spin around both these issues here at SFF again in the future.

Anyway, having directed some scepticism of late towards various aspects of the alternative farming movement that I consider myself to be a part of, perhaps it’s time I twisted the other way.  So here I want to take a critical look at the Breakthrough Institute’s line on the necessity of synthetic nitrogen in world agriculture, which is laid out in its agronomic aspects in this post by Dan Blaustein-Rejto and Linus Blomqvist (henceforth B&B), and in its historical aspects in this one by Marc Brazeau.

To begin, let me say that I’m not implacably opposed to the use of synthetic fertiliser in every situation, and I don’t think that a 100% organic agriculture globally is necessarily desirable or perhaps currently feasible. However, I think the narrative presented in the two BI posts is misleading. As is often the case, the sticking points lie not so much in what the posts say as in what they don’t say. I know Christmas is a long way off, but I’m going to lay this out in terms of the ghost of nitrogen past, the ghost of nitrogen present and the ghost of nitrogen future.

The ghost of nitrogen past

Marc Brazeau’s piece reminds us that, prior to the invention of the Haber-Bosch process for ammonia synthesis at the start of the 20th century, countries went to war to secure nitrogen for their farmers. He focuses on the international conflicts of the 19thcentury over the guano islands off South America, with their vast concentrations of richly nitrogenous seabird faeces.

It’s a nice piece in its own terms, but there’s a bigger historical story it omits. Brazeau broaches it, but doesn’t develop it, in this passage,

“The full lower 48 [US states, in the 1850s] was available for cultivation, and yet soil fertility was already a challenge. US agriculture is currently tasked with feeding 325 million citizens while exporting $150 billion worth of food. But in the 1850s, with just 25 million citizens to feed and hundreds of millions of acres of some of the most fertile soil in the world, on farms where manure-producing cattle, hogs, and poultry were well-integrated with crop production, US presidents were promising to get tough on guano prices and US business interests were verging on war in the Caribbean over fertilizer.”

For their part, B&B note that:

“During the 19th century, the populations of the United States and Europe were growing at an unprecedented pace — the U.S. population increased tenfold and Britain’s more than tripled…To raise farm productivity, these imperial powers started to import nitrogen-rich guano.”

What’s going on here? Well, the key surely lies in B&B’s phrase “these imperial powers” and in the spectacular US population increase, which wasn’t just a baby boom. In 1803, after defeat in Haiti, Napoleon gave up on his ambitions for an American empire and sold a fair old whack of that lower 48 to the US (another large tranche was subtracted from Mexico in 1848). The US spent much of the succeeding century progressively divesting the original inhabitants of their access to it and during that process, multitudes of European-origin settlers moved in – witting or unwitting foot soldiers of their government’s imperial ambitions. As historian Geoff Cunfer puts it, these pioneers “may have devoted most of their land, time, and energy to subsistence activities out of necessity” but they were “aggressively committed to…commercial cash-crop agriculture as fully and as soon as possible”1, because of their intimate connection to the global imperial nexus via their own government’s global ambitions.

Meanwhile in Europe, after Napoleon’s defeat Britain emerged as the dominant imperial and industrial power of the 19th century. With the abolition of its Corn Laws in 1846, cheap grain from North America (and, increasingly, other places with continental grasslands whose original inhabitants were also violently displaced in favour of export-oriented grain agriculture such as Australia and Central Asia) started flooding into industrialising Britain. The British agricultural workforce dwindled, and the British farmers who managed to survive the resulting agricultural crisis started favouring higher value, non-staple crops2.

All of which is to suggest that the search for cheap nitrogen in countries like Germany, the USA and Britain from the 19th century wasn’t just some inherent truth about the nature of farming and population increase, as the casual reader might surmise from the BI posts. Rather, it was the product of aggressively expansionist imperial-industrial ambitions, fuelled by fears among industrialising powers that lack of food autonomy made them vulnerable to enemies. If that point needs underscoring, perhaps Haber’s other main claim to chemical fame as the overseer of Germany’s successful chemical weapons programme during World War I might help to dramatize it.

Brazeau implicitly accepts this imperialist-expansionist aspect to the politics of agricultural nitrogen, but turns it into a world-historical truism:

“the Roman Empire was largely defined by imperial expansion, in search of fresh sources of nitrogen. They found it in the form of soil which had not yet been exhausted. The whole Mediterranean basin became tasked with feeding the city-state at the heart of the empire. All this is to say that this is not an industrial agriculture problem; clearly, it’s been a central obstacle of civilization for thousands of years. If the problem of nitrogen scarcity could be solved by cover crops and manure, it would have been solved long ago.”

But I think the direction of causality is wrong here, and so is the conclusion. Imperial expansionism sometimes involves a search for cheaper farm inputs, but the search for cheaper farm inputs is not usually the cause of imperial expansionism. And for a long time, in many parts of the world whose polities were not expanding aggressively, the problem of nitrogen scarcity was solved perfectly well by cover crops and manure.

The ghost of nitrogen present

But that was then and this is now. Whatever the rights and wrongs of the past, the fact is there are now 7.6 billion of us living on an ecologically fragile planet who somehow need to eat. The case set out by B&B in favour of synthetic fertiliser and against organic methods is, as they confess, the well-worn one that the lower average yields and higher average land-take of organic farming militates against it as a sustainable solution for contemporary food production.

Again, what strikes me about this argument is the things that aren’t said – four things in particular.

Thing #1. The idea that, as much as possible, we should aim to use less rather than more land for human crops surely commands wide agreement. So suppose you come to the issue afresh and take a look at global agricultural land use. You’d find that by far the largest proportion of the food that people eat is grown on arable land, which constitutes 29% of all agricultural land globally. You’d also find that about a third of this arable land was used to grow livestock fodder. You’d find that a small proportion of food comes from permanent crops, occupying 3% of all agricultural land. You’d find that the remaining 67% of farmland comprises permanent grassland, which produces a very small proportion of the food eaten globally in the form of meat – possibly no more than about 4%3. And you’d find that just over 1% of all this agricultural area was devoted to (formally) organic farming. If you did this, I think you’d probably conclude that the easiest way to reduce the global agricultural land take would be to reduce the amount of permanent pasture, followed by the amount of arable cropland devoted to livestock fodder, in view of the trophic inefficiencies involved. You might also wonder why B&B don’t mention this at all, and why they’re so exercised about the putative inefficiencies of the minuscule organic farming sector rather than the inefficiencies of the enormous livestock sector4.

Thing #2: Another idea that seems to command wide agreement is that it’s good to ‘reduce, reuse and recycle’ with nitrogen fertiliser, as with many other things. Fertiliser has major upstream (energy) costs and downstream (pollution) costs, so using as little as possible surely makes sense. In their post, B&B go through various options for improving crop fertilisation through such things as better management of cover crops, manure and food waste. They don’t give an overall figure for how much synthetic fertiliser could be saved, but totting up their numbers it looks to me like it might be as much as 80% – though maybe I’ve got that wrong. Even if it’s much less, that’s surely a good place to start for improving agricultural efficiency, rather than targeting organic farming. If the answer to the question ‘how much land should we use for agriculture?’ is ‘as little as possible’, the answer to the question ‘how much organic farming should there be?’ is surely ‘as much as possible’. We live in a world of awkward trade-offs.

Thing #3: labour is a missing variable in the BI posts, but it’s lurking in their shadows. B&B state that traditionally farmers reserved between 25-50% of their land for (not directly edible) N-fixing legumes. These figures seem to trace back to Vaclav Smil’s fascinating book Enriching the Earth5. Smil states therein that traditional Chinese agriculture never devoted more than 10% of cropland to green manures, while in parts of England the corresponding figure was 13% up to 1740 and 27% by 1836. In his definitive contemporary guide to organic farming Nicholas Lampkin argues for a minimum ley of 35%6. What accounts for this apparent historical decrease in the efficiency of organic fertilisation? Probably a number of things (including yield increase), but I suspect one of them is declining labour availability and increasing mechanisation. In contexts of low food insecurity, low labour availability and high mechanisation, it’s just easier for organic farmers to build fertility with long leys. But there are other options – as in labour-intensive Chinese or historical European agriculture, with their finer-combed local recycling of nutrients. Personally, I think more labour-intensive and local agricultures are the right way for agriculture to develop. I accept that other people may disagree. I don’t accept that current levels or trends in agricultural labour inputs should be assumed to be inherently the right ones.

Thing #4:  B&B write, “organic farms typically have 20% lower yields than conventional farms, requiring more land to produce a given amount of food. This means less land for wildlife habitats or other purposes”. But hold on – that’s only true if you assume that farms themselves aren’t wildlife habitats, that wildlife is indifferent to the habitats afforded by organic and conventional farms, that the possibilities for wildlife to move between habitats across farmland is unaffected by farming styles, that increased production or per hectare yields is always desirable, that ‘other purposes’ are more important than organic farming…and many other things besides. All of these points are at least debatable. I keep going back to this excellent brief critique of the so-called ‘land sparing’ argument by ecologist Joern Fischer, which to my mind effectively skewers the misplaced certainties of B&B’s one liner. As Fischer’s analysis suggests, while producing as much crop as possible from as small an area as possible using synthetic fertiliser certainly can be an appropriate goal in some situations, it’s an oversimplification to imply that the greater land-take of organic farming inherently limits its claims to environmental benefit7.

The ghost of nitrogen future

What would a future world that dispensed with synthetic fertiliser look like? Scarily profligate, according to B&B. They write: “Since synthetic fertilizer provides nearly 60% of current nitrogen for producing crops, eliminating it without making any other changes would require far more farmland to fix enough nitrogen to maintain production….The world would need to more than double the amount of cropland.”

The italicisation is B&B’s, not mine. Note its nervousness. Isn’t it a little bizarre to assume there would be an international drive so radical as to make global agriculture entirely organic but without making any other changesIn truth, ‘without making any other changes’ seems to be the leitmotif of the Breakthrough Institute’s entire programme, which amounts to the view that people in rich countries can carry on living as they do, people in poor countries will soon be able to live in the same way, and with a bit of high-tech magic it can all be achieved while lessening humanity’s overall environmental impact.

Well, it’s a view – a fanciful one in my opinion, and not one that I’d like to see manifested even if it were possible. But I’d note that it is just a view – one of many different visions about what a good life and a good future might entail. Trying to realise it is a choice that’s open to us. Other choices are also available. What I dislike about the BI posts is the way they implicitly lead the reader to conclude that a synthetic nitrogen future is inevitable and scientifically foreordained, rather than a choice we can make – one with consequences for better and worse, as with all choices.

The alternatives? Well, if we want to talk about inefficient agricultures, the vastly inefficient production of meat (disproportionately consumed by the world’s richer people) is an obvious place to start. I’m not a vegan and I think there’s a place for livestock on the farm and a place for permanent pasture in global landscapes – indeed, I’ve argued the case for it strongly in the past. But the scale of the global livestock industry doesn’t have to be taken as a given. As Fischer suggests, it isn’t incumbent upon humanity to meet every economic demand that arises. After all, the UN has a special rapporteur on the human right to food – it doesn’t have one on the human right to meat. Of course, it’s not fair that only the rich should get easy access to meat. There are various ways to proceed from that point: maintaining or increasing meat production levels is only one of them.

Smaller-scale, more labour-intensive agricultures geared to better nutrient cycling would be another alternative starting place. I won’t rehearse all the arguments here about depeasantisation, urbanisation and livelihoods, not to mention carbon and energy futures, but a large commercial farm that uses synthetic nitrogen and other relatively expensive inputs isn’t intrinsically better than a smallholding that doesn’t. I think it’s time we laid aside the expansionary and ultimately imperialist mindset that insists otherwise, and settled down a bit. If the US reined in some of that $150 billionsworth of food exports that Brazeau mentions (which it’s ‘tasked’ with only really through its own self-interested economic agenda), less input-intensive and more labour-intensive agricultural approaches may become a little more feasible again worldwide, and could bring many benefits. Moving towards less aggressively expansionist economic ideologies in general certainly seems worth pondering as a route for humanity’s future. You might take a different view – but it would be good if we could at least agree that we’re talking about different views, not the inescapable truths that the BI posts seem to suggest.

Just to crank a few numbers of my own around these issues, I looked at FAO data on current global production of barley, cassava, maize, millet, plantains, potatoes, rice, sorghum, soybeans, sugar, sweet potato, taro, wheat and yams (my calculations are here if anyone would like to probe or critique them). This list probably encompasses most of the world’s major energy-rich crops (oil crops excepted), but scarcely even begins to capture total agricultural productivity. Totting up the total calories produced from them and then dividing that figure by the total calories needed by a 7.6 billion strong humanity at 2250 kcal per day, I find there’s a 43% surfeit over human calorific need from those crops alone. If we then correct the production figure downwards by the 20% that B&B say is the typical organic yield penalty, include a generous 35% organic ley and make a few adjustments for existing organic production and livestock products from the ley, we find that organic production can probably meet around 90% of total human calorific needs just from those 14 crops at existing levels of land-take. That’s just a ballpark, back-of-envelope calculation, but it suggests to me that this ‘organic agriculture can’t feed the world’ trope is a bit overblown. I’m not too bothered about whether it can or not – but I think we’d be better off debating the subjective content of our visions rather than writing them in ways that seek to buttress their historical inevitability or objective truth.


 1. Cunfer, Geoff. 2005. On the Great Plains: Agriculture and Environment. College Station: Texas A&M University Press, p.99.

2. Thirsk, Joan. 1997. Alternative Agriculture: A History. Oxford UP.

3. A ballpark figure I’ve come up with from FAO data, based on all the cattle, sheep, goat and horse meat produced globally (so possibly an overestimate?)

4. Data in this paragraph from

5. Smil, Vaclav. 2001. Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation of World Food Production. MIT Press.

6. Lampkin, Nicholas. 1990. Organic Farming. Farming Press, p.150.

7. Actually, Blomqvist has written a longer piece on this specific issue here, which is quite interesting – but not to my mind ultimately convincing that the ‘land sparing’ concept is robust to the kind of criticisms levelled by Fischer.


Teaser photo credit: By Rasbak – Own work, CC BY-SA 3.0,