In this post, I’m going to complete my look at Mark Shepard’s book Restoration Agriculture: Real World Permaculture for Farmers, which I began in my previous post. My focus here is on Shepard’s analysis of the productivity of perennial polycultures – a subject dear to the heart of many a permaculturist.
In the chapter titled ‘Nutrition and Perennial Agriculture’ (pp.167-183), Shepard writes “The nutrition per acre under restoration agriculture outcompetes corn so much that it’s not even funny” (p.167). Let’s consider this in more detail.
Shepard is actually making three different arguments in this short sentence, two of which I find convincing and one of which I don’t. The first is that a mixed ‘restoration agriculture’ holding produces a more nutritionally complete and balanced diet than a corn monoculture. That’s not something I’ll dispute. Perhaps it’s not something anyone would dispute – I doubt even the most dogmatic proponent of agribusiness-as-usual would argue that a pure corn diet is a good idea. I guess a pure chestnut diet wouldn’t be so great either. But I think it’s true that the mainstream farming system is producing too much of a narrow range of crops which are not nutritionally optimal.
The second argument is that mainstream agriculture wastes a lot of its productivity in inefficient uses: primarily livestock fodder and biofuels. The claim that we need conventional arable annual farming in order to ‘feed the world’ indeed rings hollow when we use so much of its product to feed bio-digesters and livestock that service the demands of the wealthy, and Shepard makes that point convincingly. According to his figures, a corn monoculture can produce 13.9 million calories per acre, but the actual human nutrition derived from it (ie. direct plant food plus indirect food from corn-fed livestock) comes in at only 3.06 million calories per acre. That is a shocking discrepancy indeed, but what are its implications for the calorific productivity of an acre of corn? Zilch. Feed it to livestock, feed it to digesters, dump it in the sea, do any damn fool thing you like with it, but the productivity of a field of corn remains 13.9 million calories per acre. Don’t blame corn for what happens to it beyond the farm gate.
That last point is relevant to Shepard’s third argument, which is that an acre of restoration perennial polyculture outyields an acre of corn calorifically. Calories are important in debating different possible agricultural systems because it’s not an easy thing to get enough energy into the bodies of 7 billion humans, and proposals for agricultural systems that are unable to furnish the necessary calories are not an easy sell. So I’m glad that Shepard has bitten the bullet.
The productivity figure he uses for an annual corn monoculture in his comparison with a perennial polyculture is 3.06 million calories per acre (ie. that proportion of the US corn crop currently used directly for human food – see above). But the one he should be using is 13.9 million calories (the full calorific productivity of an acre of corn – incidentally, I’m mostly just using Shepard’s own reported figures here without corroborating them independently). Let’s now look at Shepard’s perennial productivity figures.
It’s worth pointing out that Shepard’s analysis is not based on actual results from a real live polyculture, but on data aggregated from various sources in the research literature. And it’s also worth pointing out that he’s chosen a maximally energy-productive combination of tree crops, including 86 chestnut trees, 208 hazels and 34 apple trees per acre. He claims a per acre productivity of 1000lbs for the chestnuts, which may not be unreasonable, although elsewhere in the book he states “Out of the thousands of Chinese chestnuts that have been planted at New Forest Farm in the past 15 years, only two of them are bearing” (p.81). This, and a few other considerations, provoke the thought that his reported productivity figures may be a touch on the high side, but let’s take him at his word when he says, in addition to the chestnuts, his system will produce about 2,900lbs of apples per acre, 400lbs of hazelnuts, 400 quarts of raspberries, 5,200lbs of redcurrants and 600lbs of grapes. By his calculations, that amounts to about 4.6 million calories per acre of fruit and nut crops.
There are also livestock grazing on perennial pasture alleys between the fruit and nut crops in Shepard’s system – a dairy cow, a beef steer, two pigs, two sheep and ten chickens, which he estimates will produce about 1,100lbs of meat and 2,100 gallons of milk per acre per year, providing another 1.1 million calories per year. Although he makes some good points about the complementarity of the different livestock species and the high returns from high stocking density, I’ve got to say that I find this an unfeasibly high return of meat and milk to expect to produce each year entirely from a 60x60m plot of perennial forage. But let’s go with it anyway – Shepard’s total system then produces by his estimation about 5.7 million calories per acre.
The conclusion I’d draw from this analysis is that if you choose the most calorifically productive perennial polyculture imaginable and then stretch its projected productivity to or beyond the limit of credulity, you can demonstrate that it’ll produce something like 40% the calories of a corresponding acre of annual corn. So the fact is, despite Shepard’s claims, a perennial polyculture under restoration agriculture is less calorifically productive than an annual monoculture of corn by a distance.
Does it matter? Yes and no, in my opinion. No, because of the following (admittedly simplistic) calculation. Taking the FAO figure of about 3.4×109 acres of arable land globally, and assuming the need for a daily calorific intake of 2200 calories for each of the 7 billion people on the planet, that would require a calorific productivity of about 1.7 million calories per acre – comfortably less than Shepard’s 5.7 million figure. Obviously some parts of the world could probably produce a lot more calories per acre of perennial crops than Shepard’s figure, others a lot less. Anyway, I’m reasonably happy to go along with his fundamental conclusion: yes, we can probably feed the world calorifically (and in every other way) with perennial polycultures. Especially if more people spend a bit more of their time working on them.
But the inferior calorific productivity of perennial polycultures does still matter, if only because there are lots of people in the alternative farming movement – including Mark Shepard – who persist in claiming that perennials outperform annuals on this measure when they manifestly do not. Why bring discredit on ourselves by making claims that are patently false when we don’t even need them to be true in order to justify what we’re doing? That matters.
Shepard draws the following conclusions about his proposed restoration agriculture system (p.180):
- It produces more than twice the human calories per acre as an acre of corn
- It is perennial and never needs to be planted again
- It prevents erosion [and] creates soil
- [It] can be managed with no fossil fuel inputs
I’m less sanguine. As I’ve just shown, in fact it produces far fewer calories per acre than corn, though not so few as to undermine its plausibility as an alternative to annual agricultures. Some of it will probably need to be planted again, though admittedly much less than an acre of corn. It will certainly do a better job of preventing erosion (though I do worry a bit about those livestock densities). Creating soil? Well, maybe. And yes it can be managed with no fossil fuel inputs, but then so can an acre of corn. Personally I wouldn’t fancy growing an acre of corn without motorised assistance, but nor would I fancy dealing with 1,400lbs of nuts and 8 or 9,000lbs of fruit.
I’m not sure how the human labour involved in the two cases stacks up. Shepard makes some interesting points about the possibilities, as yet unrealised, of mechanical harvesting in a multistorey perennial polyculture. I think he’s right that this may be possible, though I suspect not easy (surely there will be tradeoffs between the degree of polyculture ‘mimicry’ and ease of harvesting) and probably not especially efficient in terms of energy input/output ratios.
People do tend to wax lyrical about the work-free productivity of forest gardens, fruit forests, perennial polycultures or whatever you want to call them. I’m not yet convinced – I’d like to see some good figures. My sense is that there’s a lot of fiddly work involved in maintaining and harvesting these systems, including managing the successional dynamics of complex polycultures, which tends to go unaccounted in the enthusiasm of their proponents. I think our agrarian ancestors figured out correctly that, in most places at least, the best terms of the input/output equation are to be had from growing annual cereals, in the short run at least. Longer term, that approach has stacked up a host of problems for us, and there’s a lot to be said for moving towards perennial polycultures to remedy them. Increasing calorific productivity or saving ourselves work aren’t, however, among them. That’s why I prefaced this topic in my previous post with that neat quote from the Kansan farmer: “Let us not spend Nature’s accumulated fortune in riotous farming”. My contemporary take on that in the light of Shepard’s analysis would be: “Let us not pretend we can protect Nature’s accumulated fortune while continuing to farm riotously”. Or, “there’s no such thing as a free lunch”. If we’re going to embrace a perennial polycultural agriculture – and Shepard provides a lot of good reasons why we should – then we must also embrace working harder for less return.
Indeed, one of the attractive features about perennial polyculture is its affinity with smaller-scale more rural societies and a more peopled agricultural landscape. Bring it on, I say. But maybe grow a bit of squash along the way. Maybe make some weaker claims for the advantages of perennials and the evils of annuals. Mark, I’m with you most of the way, but I just don’t buy the simplicity of your mantra ‘perennials good, annuals bad’.