Three adaptive landscapes

Three communities I've come across in the last few years have made me see language and order in a new way. Two I've read about - Peruvian potato farmers and Balinese rice-growers. The other - Mutawintji - I visited as a tourist on an Outback safari before the PhD started. I'll get all my caveats out of the way: no in-depth knowledge of any of these; it'll seem like a pretty functionalist argument; I know almost nothing about anthropology. Given that…


Mark, our Mutawintji tour guide, did a damn fine job of explaining the history of the place. It's now a national park, after local Australian Aborigines fought to take it back. He now manages the park, and is still slowly learning the stories that his forebears knew. There are few of them left to learn the stories now. Mutawintji itself was/is a nexus where many different people would come together. People marked the stages of their lives there; everyone would gather, eat and carry out tests and ceremonies. These stages were marked with hand stencils.

Stencils like this are a ubiquitous image; if I had thought about them at all before, some part of my brain had a mental picture of some people just messing about at the rock with dyes: primitive graffiti. I was shocked by how entirely wrong this subconscious assumption was. People would go back to their print -which started with just a hand - and have small lengths of forearm added, marking different stages of their life. The dye itself lasted for many years, but not forever - so the rock itself is a palimpsest of who knows how many generations. (Palimpsest: one of my favourite words - very glad to have an excuse to use it…)

Your first handprint was earned by demonstrating you'd learned enough about the landscape to feed a bunch of your peers. This was, of course, also a test of how effectively that knowledge had been imparted - though I got the sense it was a test in name only; no-one within that community would be likely to fail it. (Youthful rebellion in such a harsh environment was not tolerated - anyone breaking rules could be severely punished.)

The key to the landscape was the stories; as you grew up, you'd travel about and listen to others tell them. They would be about all the things you'd see, and the things that lived there. It was this knowledge you needed for that first handprint. Later, you might get more specific knowledge. For example, people took on animals, and it became their responsibility to make sure they were not over-exploited.

Mutawintji is famous for its rock carvings. Mark told us he was still trying to find out the stories behind them. Some older locals still remembered them, but still also held on to the intimacy of that knowledge, and were not passing it on as a job lot. Once, he'd been doing a tour with some old women from the area, and was explaining that a few markings on the rock must have been put there more recently, because they made no sense. "Yes they do - those relate to women's business," the ladies said. This was news to Mark, and he asked what they meant. "Those relate to women's business," they replied.

That's a fascinating point in itself - the intimacy of knowledge which, to my ears, immediately makes me think, 'waargh, just write it down! Why the secrecy?' But focus on the stories themselves: generations walked the landscape and re-formed them; their survival depended on them; knowledge of them meant the difference between eating and not eating. At the rock carvings, I had a lucid sense of how few people still knew them. If they were written down - and many of them are, of course - all you would be looking at was a fossil. The story Mark told was about a living, breathing, evolving thing: a creature whose sinews were made up of people, story and land, and regenerated itself in places like Mutawintji.

Imagine you're airdropped onto the same landscape 300 years ago, with a hundred others, and you have nothing but a book describing the local flora, fauna and geography, and the ability to read. Will you survive? Well - you'll do better than you would have without the book. But if it was to work, it would have to bootstrap a whole lot of other things - experiential knowledge, the minutiae of dealing with the landscape from moment to moment as a group. J.C. Scott calls this metis, Oakeshott practical knowledge. There isn't any reason why codified, written knowledge can't underpin metis - after all, you can learn to bake cakes very well from a book, and over time you'll pick up all the various nuances of cake-making through trial and error. I really just want to get across the contrast: Australian aboriginal communites used language entirely dynamically, deeply embedded in the socialisation process, and intimately connected to their environment. More than that - the stories were part of their nervous system and one of their genetic transmission mechanisms. Seventy dialects existed in the New South Wales region where Mutawintji is, an area more than three times the size of the UK. But as with the language itself, the boundaries would be dynamic - perhaps managed in special places like Mutawintji itself, where many groups met on common ground.

I'm trying to avoid any value judgements here, though its unavoidable. I also want to mimick something Mark conveyed to us implicitly; he clearly disapproved of the mystery-isation of his community. This lent his speech a matter-of-factness, giving a sense of what it took to survive in that place. One language choice of his shows this perfectly: the clever fellas among his community were just that: clever fellas, not shamen. That's not something that would sell well in the new age stores of Glastonbury. I've tried to reflect that in my recounting, but it appears to clash with a fact I've thus far omitted: the stories themselves are full of spirits. The same is true for my other two examples. It's possible to present an entirely functional description of them - indeed, to build a computer model. But as with Mutawintji, gods and spirits are woven into their structure. I'll come back to this.


Stephen Lansing worked with ecologist James Kremer to build a model of Balinese rice management. It is integrated through a water temple network, with 'holy water' flowing down from a crater lake. Lansing, in Priests and Programmers, "tells the story of well-intentioned but ultimately disastrous attempts by planners to reorganise farming systems on the island of Bali." This is learning through social pathology: "the very success of the temple networks in balancing water needs and sustaining good harvests made them nearly invisible" - until they were broken. At least a thousand years old, complex, always changing but only slowly, Balinese rice-growing combines irrigation engineering and water management. It works with a trade-off:

For a single farmer to try to reduce the pests on a field without coordinating with neighbours is useless because pests will simply migrate from field to field. But if all the fields in a large area are burned or flooded, pest populations can be sharply reduced… Both kinds of fallow periods - burnt fields or flooded - are effective techniques for reducing the population of rice pests, but both depend on synchronising the harvest and subsequent fallow period over many hectares… The temple networks sustain good harvests by finding planting schedules that provide enough water for everyone, but also permit pest control by synchronising fallow periods for each block of terraces.

When the Dutch began gaining colonial control of Bali in the 1840s, Lansing argues, "… it appears that once the temples had been pigeonholed as religious institutions, their practical functions became invisible." This blindspot carried right on through to the implementation of Green Revolution technologies in this century: replacing native rice strains with high-yielding hybrids. The seasons were also altered: traditional fallow periods and irrigation regimes were superseded by continuous, intensive cropping. The result: a very short-term productivity increase, then water shortages and "unprecedented outbreaks of rice pests and diseases".

So what exactly was it they managed to break? The water temples "define connections between productive groups and the components of the natural landscape that they seek to control." Each shrine or temple is associated with a specific part of the irrigated landscape. A local irrigation system begins with a spring or a weir in a river, which diverts part or all of the flow of water to an irrigation canal - "the congregation of the weir shrine or spring shrine consists of all the farmers who use the water originating from this source."

Holy water is a vital part of all rituals - and it must originate from a point upstream from the ritual. The first lot comes from steam from the volcano at the crater lake. Upstream water still 'belongs to the collective' and is yet to be divided. Holy water is collected from temples by delegations and mixed in for their own rituals, but this also only goes downstream, never the other way. So "the flow of holy water from temple to temple establishes hierarchical relations between the temples and links them to a common origin."

There are about 100 farmers who get their water from a common source - these collectives are known as subaks, and it's these Kremer modelled. Water temples exert a form of hierarchical control over irrigation in the subaks. Rituals are 'performative' - calling on people to do certain tasks within the system - but there's no simple, linear mapping between ritual and function. A complex cyclical calendar system is also used. Two quotes from Lansing sum this all up nicely:

Agricultural work is not merely a sequence of technical tasks; it is a meaningful series of interactions between social groups and the natural world… ritual as an integral component of the technology of farming.

Lansing's tale has a happy, if curious, ending. Already having in-depth knowledge of Bali, he had tried to impress upon the authorities that the rituals of the water temples were "'a system of ecological management with deep historical roots in Balinese culture." They didn't listen. Lansing then involved Kremer, and they modelled what they identified as the central dynamic. Lansing was already convinced that the water temples had a key role, but the model asked: 'could these systems of social coordination have measurable effects on rice production?' The answer was 'yes' - and this persuaded people in power. The water-temple based system is now supported / allowed to function. I say this is curious because the model itself is a fantastic simplification; it is telling that it took this 'quantitative' step, where Lansing's own anthropological knowledge failed to convince. I wonder whether anyone would have looked for that dynamic without Lansing's immersive knowledge of Bali - yet it seems to validate Krugman's maxim that 'we just don't see what we can't formalise.' More depressingly perhaps, policy-makers just don't see anything that hasn't been coated with a thick varnish of Quant.

From the Goddess of the Craker Lake down through all the lower-level temples and shrines, this appears to be a pretty unique management approach - though this may be because other parts of the world have yet to find their Lansing and Kremer. Balinese rice management is conservative - negative feedbacks are built in to maintain stability - but it's a solution to that classic economic problem: the management of scarce resources, in this case water. I'll leave it there for now and go on to the last example.

Andean potato cultivation

J.C. Scott uses the work of Jan Douwe van der Ploeg as a metis-imbued, adaptive counter-example to top-down 'scientific agriculture'. I'm only vaguely uneasy about calling it an historical example of successful open source production. Scott does a better job than I could of summarising van der Ploeg's work - here are some key chunks:

The typical farmer cultivates anywhere from twelve to fifteen distinct parcels as well as other plots on a rotating basis. Given the great variety of conditions on each plot (altitude, soil, history of cultivation, slope, orientation to wind and sun), each field is unique. The idea of a 'standard field' in this context is an empty abstraction. Some fields contain only one cultivar, others between two and ten, sometimes interplanted in the same row or with each in its own row.

Each cultivar is a well-placed bet in its niche. The variety of cultivars makes for local experimentation with new crosses and hybrids, each of which is tested and exchanged among farmers, and the many landraces of potatoes thus developed have unique characteristics that have become well-known. From the appearance of a new variety to its substantial use takes at least five or six years. Each season is the occasion for a new round of prudent bets, with last season's results in terms of yield, disease, prices, and response to changed plot conditions having been carefully weighed. These farms are market-oriented experiment stations with good yields, great adaptability, and reliability. Perhaps more important, they are not just producing crops; they are reproducing farmers and communities with plant breeding skills, flexible strategies, ecological knowledge, and considerable self-confidence and autonomy.

Compare this 'craft-based' potato production with the inherent logic of scientific agriculture. The process begins with the definition of an ideal plant type. 'Ideal' is defined mainly, but not only, in terms of yields. Professional plant breeders then begin synthesizing the strains that might combine to form a new genotype with the desired characteristics. Then, and only then, are the plant strains grown in experimental plots in order to determine the conditions under which the potential of the new genotype will be realised. The basic procedure is exactly the reverse of Andean craft production, where the cultivator begins with the plot, its soil, and its ecology and then selects or develops varieties that will likely thrive in this setting. [Scott p.301-2]

The farmers have their own taxonomy for discussing cultivars and plot conditions. Cultivars themselves are constantly exchanged - even given as presents at weddings. While the taxonomy appears simple - three binaries of hot/cold, high/low, hard/soft are most important - van der Ploeg argues they cannot be compared to a static description used in agronomy:

When one separates these concepts from the people who use them or from their context, they become 'inaccurate'. Of course this inaccurate character does not prevent farmers from establishing quite exactly the overall conditions of specific plots… The inaccurate nature of the concepts used even seems favourable for such an exact interpretation of a plot's condition and the ensuing dialogue. For interpretation and communication can only be active processes; concepts must be weighed against each other every time a specific plot is being considered. [van der Ploeg, Potatoes and Knowledge, p.212]

Again, all this is within a 'magico-religious interpretation of the world'. The soil itself is pacha mamma (Mother Earth), who is 'grateful for the respect paid earlier to her.'


So am I advocating religuous ritual as the foundation of 21st century distributed production? Wouldn't that be something? What gods and goddesses would British farmers be likely to adopt, I wonder? But no. In 'Perfect Order' - Lansing's follow-up to Priests and Programmers - he notes the history of Western science has been the continuation of -

- a centuries-old battle against superstition. From this perspective, magic is the antithesis of rational thought.

This battle for modernity is not merely a dry academic argument: agricultural specialists carry out these visions through a plethora of agencies. Lansing himself has taken many of them to see the Balinese system in practice, but they quickly returned to "the planners' real work, which always took place in hotels and government offices." [Perfect Order p.10]

It's in these places that the Lansing-Kremer model was able to sway opinion, of course. But while the model succeeded in capturing the functional dynamic at work, has it helped shine a light on the bind between that order and its cosmology? Are we stuck with the false dualism of functional vs magical? Lansing -

Both the water mountain and the desires of the farmers who sustain its artificial shape must be aligned to far more rigorous patterns than they would take if left to themselves. All this is a question of engineering, not magic. The magical idea is that the two problems are really one - that the forces needed to align the water mountain already exist in the inner world of the farmers. [Perfect Order p.124]

The Balinese contrast this order with the chaos of untended forest and coastline; Lansing notes their worldview has some similarities to Hegel's: 'they have objectified the Reason of their cosmological vision into innumerable symbolic forms, to the point that it pervades their experience of the world.' It also materially organises their world.

In Peru, van der Ploeg found that 'magic' there was intrinsic to managing the risks of constant adaptation - "'essential for achieving progress within the framework of the art de la localité" -

It is precisely 'traditional' magic, with its built-in explanation and fear of the natural world, which makes possible renewal and transformation. Magic delineates, that is, it reduces the space for experiments to socially acceptable proportions. [Potatoes and Knowledge p.216]

Lansing and van der Ploeg provide functional answers, in a way, but they're a far cry from a narrow functionalism. (It's also curious that few people seem to label any of the major religions as magic, despite their constant calls upon invisible beings to assist in day-to-day survival.) My interest certainly lies in the success of these three communities: they work, so on some level are functional. They're a small set in a vast space of possible cosmologies that can tie people and environment together. As discussed elsewhere on this blog, they occupy that space like a flock of birds. The modeller in me wants to know how they relate to underlying substrates - potato genes, energy in joules. What's striking is that they don't directly talk about these things, or even need to know about them. One might even describe them as virtual machines running on an underlying, but invisible operating system (if one were so inclined.)

I'm unsure what Hayek would have made of this. One the one hand, he believed our chance discovery of the price system was a gateway to the Great Society, and all previous forms of social organisation were untenable anachronisms. On the other, they share so many common features with his own conception of spontaneous order. As Lansing points out, the water temple system is "a model for an intrinsically dynamic social institution... if each group of local farmers acts in its own interests and responds to purely local conditions, all the groups benefit as a solution for the entire watershed emerges." [perfect order p.16] (There are places in the Bali's order where 'acting in your own self-interest' is socially condoned, and others where it is not - you would be ill-advised to start on a sole irrigation building project; self-interest and selfishness are not necessarily synonymous.) Price itself is a mediated symbol, attenuating a vast amount of situated knowledge, and it only works in context. The money system has its own performative and socialisation rituals.

But these three 'adaptive landscapes' seem, in different ways, quite vulnerable. The particularly tight coupling of language and landscape in the patchwork of Australian tribes, I'm guessing, was especially unable to survive a colonial onslaught. Peruvian potato farmers are among the poorest people in the world. [e.g. van der Ploeg, Labour Markets and Agricultural Production, p.151] One could make a case that the adaptive landscapes described here mean resilience and stability for those communities, yes. But the price system has enabled something of a very different order - to the extent that it now wraps the planet.

For my PhD, I want to know about the productivity of these 'adaptive landscapes'. But productive for who? (Or possibly whom.) In both Bali and Peru, 'high yield' cultivars have been attempted, but who was to benefit? And what say did they have in the matter? It's a spaghetti of democracy, ownership and exploitation. A slightly tangential example from the past month: President Garcia justifying denying indigenous Peruvians their land rights thus:

We have to understand when there are resources like oil, gas and timber, they don't belong only to the people who had the fortune to be born there.

How very convenient: a story as old as colonialism and enclosure, often accompanied by tales of the noble savage's shadow 'fading away' in the slow but inexorable sunrise of Progress. I suspect, to answer my own question, Hayek would have been as blind as most other western academics to any existing order; his Great Society was Progress incarnate, though he detested 'constructive rationalism'.

Mark's story, I said, was of a living, breathing thing. This is true for all three; it's also a strong strand of conservativism. Here's Burke:

In England we have not yet been completely embowelled of our natural entrails… We have not been drawn and trussed, in order that we may be filled, like stuffed birds in a museum, with chaff and rags and paltry blurred shreds of paper about the rights of man. We preserve the whole of our feelings still native and entire, unsophisticated by pedantry and infidelity. We have real hearts of flesh and blood beating in our bosoms.

He also glances off the idea of adaptation:

You see, Sir, that in this enlightened age I am bold enough to confess, that we are generally men of untaught feelings; that instead of casting away all our old prejudices, we cherish them to a very considerable degree, and, to take more shame to ourselves, we cherish them because they are prejudices; and the longer they have lasted, and the more generally they have prevailed, the more we cherish them. We are afraid to put men to live and trade each on his own private stock of reason; because we suspect that this stock in each man is small, and that the individuals would do better to avail themselves of the general bank and capital of nations and of ages. [Burke, Reflections, p.144-145]

'The more they have prevailed, the more we cherish them.' A nice memetic evolutionary algorithm. So should we celebrate ignorance, as Burke and Hayek do? That's certainly not the conclusion the Balinese have come to: they see themselves as continually, consciously seeking order. Indeed, it would appear it's possible for outsiders to achieve some understanding of these societies without venerating them. Burke appears to be attempting to achieve what Mark from Mutawintji most objected to: mystery-isation. Perhaps not coincidentally, this allowed him to conclude:

We fear God; we look up with awe to kings; with affection to parliaments; with duty to magistrates; with reverence to priests; and with respect to nobility. Why? Because when such ideas are brought before our minds, it is natural to be so affected; because all other feelings are false and spurious, and tend to corrupt our minds, to vitiate our primary morals, to render us unfit for rational liberty; and by teaching us a servile, licentious, and abandoned insolence, to be our low sport for a few holidays, to make us perfectly fit for, and justly deserving of, slavery, through the whole course of our lives.

So Edmund wants to keep his estate and seat in Parliament - it's only natural. Again, convenient. He celebrates an organic, earthy ignorance over the dead wood of 'paltry blurred shreds of paper'. But its a false dichotomy. There are many solutions to the problem of managing complex human societies, and we are capable of coming to some understanding of how they work. Solutions may be as diverse as human communities - indeed, perhaps, it may be a normal consequence of letting a bunch of people loose in a place, in the same way that ants organise themselves.

That's another cliche, of course. We're not ants. We cannot be airdropped into a valley and expect to insta-self-organise a local environmental management regime (though there is a management theory, holographic organisation, that proposes just this.) But these three cases make me optimistic about two things: the price system isn't all we've got. It's one tool, perhaps among many - one particularly suited to large-scale orders (by which I mean its arm reaches a long way, and we will need to keep some scale of international trade going.) Also, there's no necessary opposition of conscious design and blind adaptation: deliberation does not make us stuffed birds in a museum. That's not how humans work: we appear to have evolved a language system that couples us dynamically to our landscapes and each other. In one respect, the comparison to ants is sound, then: language may have co-evolved with self-organisation. The price system is a subset, an offshoot, of this evolved ability. We have to act deliberatively in order for these to function correctly.

The next fifty years are going to be fertile soil for uncovering new adaptive landscapes - both physical and social. It's rather a difficult thing to conceptualise, as its so normal to oppose conscious design to adaptation. Yet in all these examples I've covered here, people act with deliberation, constantly communicate - within given structures, taxonomies and cosmologies. A combination of informatics and enormous resource pressure may (if we're lucky) be an excellent growbag for all this. But it will require feeding with conscious thought, argument, experiment. Evolutionarily, the price system can be compared to the 'discovery' of the backbone. (A Dawkins example.) Hayek, I reckon, would be happy with this. It opened up a doorway into a new realm of organisation. We need more of these, and we need to seek them out.