Transcript for J. Stephen Lansing: A Thousand Years in Bali
Good evening. I'm Stewart Brand from the Long Now Foundation. Well, a person that I've spent a lot of time paying attention to is Gregory Bateson and his wife Margaret Mead. I knew Gregory better. Back in the 1930s, when they had been infected by what was brand new, cybernetic theory, the idea of feedback, they went to Bali. (Stephen Lansing was just telling me). They invented visual anthropology. And did a book called Balinese Character, which is a classic. It is still studied by everybody in the business. That was a nice case of field anthropologists being set in a different direction by a very good body of theory, which in that case was cybernetics. Well the great, well maybe the grandchild of cybernetics is complex adaptive systems, complexity theory as it has been manifested at the Santa Fe Institute, where Stephen Lansing is a research professor. I've been on their board for fourteen years or something. It is where I first saw a version of this talk; not nearly as gorgeous as the one you'll see tonight. In Stephen and my case of taking a new body of theory, how complex systems work; emergent behaviors of the kinds of things you can do with agent-based modeling. And bringing that to bear on one of the most mysterious and wonderful bodies of behavior in Bali. And so you will see tonight, he is also bringing a new level of visual anthropology. Stephen Lansing. Thanks very much sir, that's very gracious. I need to apologize to all of you for turning up a week late. I zigged when I should have zagged awhile back on a ski hill in Wyoming. I'm grateful to be here now and actually I've been downloading and listening to some of the earlier talks in this series. It's really a very interesting series. I'm honored to have an opportunity to try and join this conversation. I'm going to take you in a different direction. And actually I'll wind up talking about the law now, rather than the physics of time in the future. I'm going to take you down the kinds of wandering paths that ecologists and anthropologists take. We'll wind up in the end, I hope, with something to say about time and about the law now. This is mostly about Bali, so off to Indonesia, which has the highest biodiversity on Earth in terrestrial ecosystems in the forests, also in the oceans and coral reefs. But things are not well in the tropics. Less than a third of the coral reefs are still healthy. Most of the Borneo in the Sumatran rain forests has been logged. Indonesia is once again been importing rice. We are losing some of the most precious treasures of the planet, the crown jewels. The question is why? Part of that has to do with politics, certainly. But I think also, I hope to convince you, that it has something to do with some weaknesses in our science and how we see the world. One of the things that began with the era of five year plans in the sixties, not just in Indonesia but all over the world, (wherever the World Bank got busy), was the implementation of hierarchical control systems. Anthropologists ask, "How did they manage before they had five year plans? Before 'you manage, we produce' system? The question is, don't complex ecosystems like rice paddies require firm hierarchical control? Well, maybe not. Maybe there are different lessons on how to organize the world. What I'll talk about tonight is partly about the sciences of complexities, or the Santa Fe Institute story, with reference to Balinese rice paddies. I want to talk a little bit also about Lisa Curran's work on the rainforests of Borneo, coral reefs and in the end, something about time. Begin with complexity. [In] 1995 the great biologist John Maynard-Smith wrote in the New York Review of Books that he has a general feeling of unease when contemplating complex systems dynamics. This is after a week in Santa Fe. Its devotees are practicing fact free science. Fact for them is at best the outcome of a computer simulation, rarely a fact about the world. These ideas that we talk about in Santa Fe are new and unfamiliar. One can ask that if these processes are really at work in the world, would we recognize them. I'm going to give you the quickest definition of what a complex adaptive system might be. This is from John Holland, one of the architects of this notion of complexity. A complex adaptive system is a network of interacting agents in which they are trying to maximize something. What he discovered is the aggregate behavior of the network that is formed by their interactions may be described without a detailed knowledge of individual agent behavior. So there can be emergent properties. That is a very nice and very general theoretical idea. But what I hope I can show you, the reason that I spent some time at Santa Fe, is that it turns out this can be useful in understanding something very real. Off to Bali. I went to Bali where the Balinese have been growing rice in terraces since at least the eleventh century. We know this from royal inscriptions at the time. They have also governed themselves in villages with a very ancient system of democracy. Now, democracies are quite rare in the world. As you know, the ones that the Italian Republic tried in the thirteen century lasted no more than a century in every case. But the Balinese have been, we believe, managing their rice paddies with groups called subaks, which are groups of men who share contiguous rice fields with a democratic system, since at least the twelfth century. It's an interesting kind of experiment in democratic governments. Let me tell you a little bit about rice paddies. They are artificial ponds in which the fertility of the water basically creates a kind of aquarium and the processes that go on in the water help to grow the rice. They provide the nutrients necessary to grow the rice, which is why we find terraces in Bali that, have been producing two crops of rice every year for centuries. It's a great trick and the control of water is the key to making that work. The reason it works is that volcanic rock is rich in mineral nutrients. [Let's see, I've got my little pointer here]. Those are just the phosphate content of rocks in Indonesian volcanoes and Bali is somewhere down there. Its actually not one of the highest but there is plenty of phosphate in the rock. As the monsoon rains fall on that fumacious rock, that very light rock, with these nutrients. They leech the mineral nutrients, which flows into the irrigation systems. That is delivered, finally, to the rice paddies. So it's sort of a continuous delivery system for phosphate potassium, the main nutrients required by the rice. Nitrogen is fixed by a zolo that grows in the water. You have a kind of perpetual motion machine. But the challenge for the Balinese to make this work is that the island of Bali is a steep volcanic island just south of the equator. As the rain falls on the slopes of the volcanoes it slices channels down the slopes of the mountains, creating these kinds of fissures. The Balinese need to get the water from the bottom of these ravines up to the top of the hill, some distance downstream, where they can grow rice. [This] means that they have to build tunnels to move the water down to the higher slopes We find inscriptions dated as early as the ninth century that refers to irrigation tunnel builders. They have been at it a long time so they have honeycombed the island with their tunnels and terraces by now. That's pretty clear: water can provide nutrients but it can also, less obviously, be used to control rice pests. Rice is vulnerable to a variety of things that like to eat it besides humans: Insects, which either eat the rice directly or carry bacterial and viral diseases, which can spread through the rice. Rats and so forth. There are lots of things that like to eat the rice. But by synchronizing harvests over a sufficiently large area it's possible to deprive the pests of their food and habitat. Here on this side you see that after the harvest they flood the fields and there is nothing for the pest to eat. But that will only work if all of the farmers plant at the same time. Generally they give themselves a five day window. The fields there, you can see, are all about the same stage, they are all at the same height. So there might be rice pests in the field now but later on, a few months from now after those fields are harvested and flooded, [there is] nothing for the pests to eat. The success of this system depends upon getting a large enough area fallow at the same time because the pests can move. If you get a kind of a patchwork system, if one field is fallowed but the other still has rice, the pests can just move around or be blown by the wind The success of this plan, of this method of controlling the pests, depends upon getting the right scale of coordination to deprive the pests of their habitat, which depends upon their dispersal characteristics. Ok, so that is the basic dynamics. The way the Balinese organize this is through this thing called a subak, A subak consists of all of the farmers who share water from a single source: a spring or an irrigation canal. In the picture, what I am trying to show you is a tunnel, excuse me, a temple where they will gather to make their decisions. There are the rice terraces which are managed by the farmers who maintain that water temple. Interestingly, in the meetings the farmers have within their water temples, they set aside the rules of caste. The reason that is important is that caste is very important otherwise in Balinese society. They are divided into a version of the Hindu caste system but they take it very seriously. So when you speak Balinese you have to first determine what is the relative caste of the person you are talking to. Then you use a different register, depending upon whether the person you are addressing is of higher or lower caste than you. A lot of this hierarchal structure is built into even their language. But when the farmers get together to hold a meeting to talk about the management of their rice terraces, all of that is set aside. You have to speak to your peers, to your fellow farmers, in middle high register Balinese or in the colloquial, informal language that you use within your own family. If you start trumping people with high language you can be fined (and it's very much frowned upon). That is part of what makes this democratic system work. So that's the basic subak. What is interesting is that these subaks then form a hierarchical structure which works from the bottom up. Here we have the water temple network. Subaks A and B; these are two subaks; there's the temple. But they both share water from this canal, which comes from that dam. They belong to this temple here. From time to time, once a year, they will send a delegation up to their neighbors up here to try and coordinate their irrigation schedules. Because the timing of the water released from this system is going to clearly impact how much water is available for their downstream neighbors. In this way they sort of build up from the bottom a hierarchal structure of relationships among the subaks using the water temples. The temples form a kind of network. At the summit of the network there is a crater lake. More or less near the real center of Bali there's a lake which the Balinese describe as a sacred mandala of waters. So they imagine that Bali is surrounded by a salt sea but in the center of the island at the highest point there is a lake that is full of life giving fresh water. In the lake there dwells a goddess and she has the power to bestow this blessing of water that makes things grow and also washes away pollution down the flanks of her volcano. Water actually belongs to the goddess and her temple is the supreme water temple of the island. She chooses priests by inhabiting the voice of a group of priest who"� [How do you say?]. When a priest dies, other priests will go into trance. They believe they are possessed by the voice of the goddess or some other deity that is connected to her retinu, who then speaks the name of the child to replace that priest. In this way there is continuity somewhat like the Tibetan idea. Anyway, these priests live up there, the village of Bater by the lake. They collect holy water from the points around the lake. The farmers come up to collect that holy water as a token of the blessings of the goddess of the lake. The Balinese call their religion Agama Tirtha, the religion of water. Each village temple controls the water that flows into nearby rice terraces. Regional water temples control the flow into larger areas. High on the volcano, Mt. Batur, there is an enormous crater lake. The Balinese believe that the lake is the home of the goddess of the waters, Dewi Danu, who makes the waters flow into the rivers and irrigation canals. The supreme water temple of Bali sits on the rim of the crater overlooking the lake. Once a year the priests give holy water from the lake to the farmers as a blessing from the goddess. First, priests ascend the crater to collect drops of water from the steam of the active volcano. Then the priests prepare the holy water for the farmers. A temple scribe writes letters on palm leaves, inviting farmers to the temple. Delegations from over 200 villages journey up the mountain to bring offerings to the goddess. At the supreme water temple, the Gamelan orchestra plays in honor of the goddess and the worshippers. Temple priests give a (soojung?), a bamboo container filled with holy water to each village delegate. The delegations then carry the holy water to their regional temples. Here the water from the goddess is mixed with more holy water adding the blessings of local gods and goddesses. Each delegate receives a soojung of this water to take back to his village. At the village temples local farmers receive a blessing for themselves and their fields. A few drops of holy water are sprinkled on every field. This ritual symbolizes the sharing of water that forms a coordinated system of irrigation in Bali. Working together in this way the Balinese have maintained the ecology of their rice terraces for over a thousand years. It's a rather idyllic system but there is a hidden problem because in each case that downstream subaks are at the mercy of their upstream neighbors for water. The guys upstream in any irrigation system have their hand on this, so they control the water. The question then is why do they release the water needed for their downstream neighbors? (Because the downstream neighbors can never give the water back.) Why doesn't this lead to a tragedy of the commons as it so often has in other irrigation systems?" We won't talk about the World Bank yet. What is different about Bali? What makes it work? John Miller at the Santa Fe Institute came up with this very simple formulation from game theory that suggested an answer to that question. Imagine we have two farmers, upstream and downstream, and they have a choice as to whether they want to synchronize their irrigation system or not, planting or not. If the upstream farmer is willing to give up some water, they both plant at the same time, then the downstream farmer is happy and he gets his water. If on the other hand, and the consequence then is they have a coordinated fallow period, they both harvest at the same time, [it] can help bring the pests under control because the pests don't move upstream to eat the upstream farmer's field. Suppose on the other hand that they don't synchronize. The upstream guy gets to keep more of the water but on the other hand the pests can come and attack his field. It looks as though pests can give leverage to downstream farmers if the farmers are thinking about pests in the context of this water allocation problem. We wondered if that very simple formulation had anything to do with the way the farmers actually think about this thing. We asked them. We asked 150 farmers in ten subaks which is worse, pests or water damage, water shortage? Here are the results. The upstream guys are worried abut pests and the downstream guys are worried about water. That seemed to be true in each of these ten subaks. There are the makings of a bargain. It turns out our economist was right. You can also look at this in the next scale; in six cases we had an upstream/downstream pair, upstream subak/downstream subak. In this agent is not an individual farmer but a whole subak. In that case the pattern is even clearer. The upstream subak is worried about pests and the downstream subaks, in these six cases, are worried about water. That looked as though, actually to my surprise that this actually worked out so well, they think about a lot of things but this seemed to be present in the mind of the farmers and helps to explain why the upstream guys are in fact willing to engage in these bargain with the downstream farmers. However, this two player game is unrealistic because in fact dozens and dozens of subaks affect one another. The irrigation systems are highly interdependent, there are lots of little subaks, dozens of weirs, and they are interconnected. The real question is how do they get it right not at the scale of two players but at the scale of, in the case that we studied, 172 subaks along two rivers. So I worked with a systems ecologist, Jim Cramer, who suggested that we build simulation model and see how this might work. At that point I had just come back from spending about a year studying one subak, so when he said we should scale up to 172 I was impressed. Anyways, here is a map. These are two rivers. The little squares, if you can see them, are meant to be the subaks. The little Christmas-y things are meant to be water temples. The stifled lines are the boundaries of catchments areas of the rivers. Bear with me, this is pretty clear. What happens in this simulation is at the beginning of the year, each of these 172 subaks, the village-size irrigation systems, chooses a cropping pattern. What are they going to plant? So it might be plant rice in January and August, for example. Then we simulate the flow of rain. The rain gets into the groundwater of the irrigation systems. The rice grows and the pests eat some of the rice. We simulate all of those processes and then at the end of the year we calculate the harvest yields for each of the subaks. Having done this, we then vary the scale at which they coordinate, from everybody does the same thing to everyone does something different to water temple scale. We can actually identify what kinds of clusters the subaks follow. Coordinated patterns based upon the traditional water temple system. To no one's real surprise, it turned out that the water temple networks optimize the trade-off between pests and water. They get it about right; get enough water to cluster of subaks and at the same time minimize the pest damage. That worked. In other words, one can see that theses water temples play a useful role in finding the appropriate scale of coordination to optimize those two opposing constraints. I gave a talk much like this, I think it was eight years ago at the Santa Fe Institute. At the end of my talk, a researcher named Walter Fontana asked me a question. He said "That's very interesting but it's really not surprising after all the farmers' have had centuries to get the scale right. Tell me did someone have to design this system? Was it organized? Did the Rajas have to work it out? Or could it have self-organized?" At that time I didn't know what he meant by that. He said, "Well, do you have to, [well],does someone have to impose this structure or could the interactions between the subaks lead to this kind of solution, to the formation of a network?". That seemed like an interesting question. The solution was, the solution we tried immediately was to just tweak our little model a little bit and turn it into what's called a hill climber, to see if the water temple networks might pop out on their own. The way that works is, start that same simulation but let the computer randomly choose cropping patterns all 172 subaks. So each of the little icons, upward triangle/downward triangle, they simply mean a random choice of cropping patterns. A downward triangle might mean plant rice in October and March, something like that. We are randomizing the conditions of water, which will effect the dynamics of water flow, rice growth and pest dynamics. We run the thing for a year and at the end of the year, each subak compares its harvest with its four closest neighbors. I kind of got bored writing little circles there but the little circles are supposed to suggest [that] these are the comparison groups. At the end of the year, I look around and see if any of my neighbors had a better rice harvest than I did. If so, I copy their copy pattern. If I did best, then I stick with my cropping pattern. That reinitializes the simulation and we run it again. Jim Cramer, my colleague, thought this thing would yield; that it would produce chaos that the model would keep flipping back and forth as they tried first one pattern to get the [model] or the other. We decided we would run it anyway. Here's what we saw. Here's year one of a simulation. This is the random choice of cropping patterns selected by the computer. The average harvest is pretty miserable. Its about five tons per hectare because there is lots of pest damage and lots of water shortages. Ten years later, synchronized patches, we you may or may not be able to see, have appeared. This group is doing one thing, that group is doing the same thing. The harvest yields have nearly doubled. They have walked their way to a solution and that solution looks very much like the water temple patterns. In fact, it is almost identical. As we repeated these simulations, we found that it is almost impossible not to grow a water temple system. We vary the pest constraints. The pests are more virulent then the fallow periods get larger. If the water is more of a problem then the patches get smaller. But no matter what the constraints are a network will form, given this kind of trade off, that will optimize the conditions. As this goes on, this also had another kind of interesting consequence. Here are the yields, rice yields, as time goes on in the simulation. That's the average, that's the highest. As time goes on you'll see that the average yield is going up until it comes very close to the highest yield, That flags the attention of any evolutionary biologist because we think that one of the reasons that people or animals disagree is because of envy; disparities in benefits. But if everybody is not only doing well but doing equally well, then there is no reason to be jealous. In fact, as we asked the farmers to compare their own harvests to the average in their subaks. You can't read that, that says that well, mine is the same and that says that it is worse. Even Balinese farmers like to badmouth their harvests. But they can't get away with it in their subaks because, in fact, everybody does equally well (and they do very well). That suggested a way in which this system could have persisted over time. You can imagine that, indeed from time to time, farmers decide to be renegades and they don't plan according to the schedule. It is possible to opt out; it is possible for subaks to become defunct, as we will see presently. But over time nature punishes you if drift away. Over time we can see a process in which, if not me then my grandchildren, are very likely to drift back into this pattern of organization because it is a very successful one. One more thing about this network idea; (It is interesting to see); the point is that it may be a very general phenomenon. We discovered in Bali that it may be quite general. Here is K; that is the number of, with apologies to Stuart Kauffman, the number of neighbors that we check in these simulations. In the model world, not Bali but in the model world, how many neighbors do I compare my harvest with before I decide what to do? That has an effect. If you look only at three neighbors, then as time goes on the number of subaks that change strategies, that flip back and forth, stabilizes with about 1/3 of them never finding the right optimal solution. They keep flipping back and forth. These are guys that never decide what to do. There may be a solution but they can't find it because they are not looking far enough out. You go to K = 4, then they find a solution faster. The network appears faster and fewer of them remain confused. As we look even further than thirteen, you get down to almost all of them confined to the solution because they can see it. If on the other hand you look not at your immediate neighbors but somewhere as in the system, for example, at the experimental farms, then you do get chaos. What happens is that everybody keeps flipping back and forth, identifying, and imitating the best farm in the entire landscape. But that loses the signal from the immediate vicinity; this is exploring your own immediate neighborhood and optimizing conditions there. That may be general. Anyway, it always succeeds and the search parameters are important. I hope that is clear. There is a complex adaptive systems explanation for water temples. Does that mean that they are just a kind of mathematical device? Well, I can't believe that [because] I'm an anthropologist. I want to talk about ritual a little bit. They devote a great deal of time and wealth and attention to their rituals that take place in the water temples. Remember that these temples are places where democratic decisions are made. Why do they do that, why is that important? It is characteristic of discourse of the Greeks, the Romans, and the Italians in the thirteenth century. Anybody who is interested in making a democratic assembly work winds up talking about the control of emotions. That becomes the central focus of Balinese discourse; how do you make communities, how do you cope with achieving consensual management? Balinese cosmology and I will have to do this briefly but I hope you will be interested, it posits a dualistic cosmos; the idea that the inner world of the self contains everything that also exists in the outer world. There are 102 components, by the way, of the inner world and the outer world. Given that presumption, there are two ways to think about it. One is the male way, purusa, which actually a high Balinese word for penis. Male power comes to the self, especially to males through their clan and it derives from the ancestors, [it is] something you inherit. It is what gives kings the power to rule. Pradana, in Balinese (they have modified a Sanskrit idea), it's the female principle. The female half of this cosmology and it is collective and transformative; it is about growth, it is about fertility. It is the pradana ritual cycle that exists in the water temple networks. Here are offerings in a water temple which relate simultaneously to the inner world and the outer world, depicting them in perfect order, as an ideal relating to both worlds at the same time. Water temple rituals are intended to tame the passions and to create order. Monthly meetings are here of subak heads, in which they must set aside lineage competition. We have these rights of equality; (He is about to pour some very nice Balinese wine) to the heads of the subaks who gather once a month to make decisions about the rice terraces. But what they talk about mostly is not ecology, they talk about politics. These [are] rituals of equality. Blind these elements of the inner world so that we can control the outer world. Here is another quick video clip. (You can read this). Here is the instrumental ritual, the sharing of water. These are offerings that grew in the fields of a subak. They are now putting them in the water that leads into that subak. It is returning thanks to the gods of that system. But we wondered if there is something more to it than the simple giving of thanks. Here are offerings at a water temple. Each family brings one of these offerings. That is what grew in their fields. These are moodras, aligning the inner world. They liken rice paddies to jewels. Jewels are an important concept in Balinese ideas. A jewel is something which symbolizes the mind. Rice terraces like these, once they are flooded with the moonlight shining on them, look like faceted jewels, as a subak had told me once. That order will fall apart in a week if all the farmers don't systematically manage the terraces to maintain that jewel-like precision. They say that the problem is really the same: the problem of creating a jewel-like rice terrace system. But it originates in itself rather than the mud. Okay, I will go quickly through this. I have gotten interested in this. (Quickly), there is a drive for universality and coherence in which they believe that the human microcosm is an intermediate between; it is a platonic view of how the world is put together, of how the universe is put together. It is reflected in the way in which they symbolize life. When you are born, they take the components of the afterbirth and align it in the household according to the directional symbolism which also relates to sounds and colors and letters. They align you with the macrocosm. At the end of your life, when you have died, they cover you with a shroud, which also depicts that alignment, although it shows the progression of changes that have occurred in your life. There is a sudhra. You can see there are lots of different components of his body. The Brahman priest has a simpler diagram because it is thought that he spent his life trying to simplify the inner world. That is reflected in the diagram that is drawn over his body. Alright, that is what I am going to tell you about Balinese ritual because we need to move along. In the 1970s everything changed, for excellent reasons. Indonesia at that time was also importing rice. Indonesia was poor; they needed to improve their agricultural productivity. The Asian Development Bank funded the Green Revolution. A new era of agricultural expansion, management in Bali as [it] did elsewhere in many parts of the world, certainly, in Indonesia. But it had unintended consequences. The basic ingredient was something they called a technology packet. Literally, packet technology in Balinese, which consisted of new rice varieties spread to grow rapidly and to take up chemical fertilizers effectively. (Plus, those fertilizers, plus organo-chloride pesticides). The farmers were urged to plant rice as often as possible and set aside the water temples. The planners of this system said that "it is perfectly fine to continue to have your lovely rituals in the water temples but don't think this is a practical management system". This system that was developed at the Rice Institute in the Philippines was moved to Indonesia. Within a few years there were unexpected problems, which you can probably anticipate. Miracle rice produced miracle pests. If indeed all of the farmers plant at different times it is like running our simulation model backwards, disrupting the pattern of coordination until you wind up with the patchy structure in which the pests can migrate from one field to another. It is just like the simulation model and that is precisely what happened. The first variety of rice that they tried this with, which was named IR8, proved vulnerable to an insect called the brown plant hopper. Two million tons [of rice] [were] lost [in] 77. The agronomists and geneticists bred up a new variety of rice which was resistant to the plant hoppers. But it proved to be vulnerable to rice tungro disease. That explosion began a few years later. They came up with a new variety, IR50, which was resistant to tungro disease not so good with the rice blast. Help me if you see where this is going. As time goes on, the farmers were required to put on heavier and heavier doses of pesticides until finally by the 80s they were flying the island spraying the fields. Meanwhile the extension agents are reporting back to headquarters chaos in the irrigation systems. We began to complain about this. The reason I got involved with the ecological modeling was really to convince the Asian Development Bank that the temples had a practical function. The director was not happy with this in 1984. Eight years later, they sent teams. We talked to them. In the end, the Asian Development Bank agreed with us that the substitution of the high technology and bureaucratic solution was counterproductive. In fact, they became converts to the water temple system. So that's a happy story. The planters have now dropped opposition to the water temple, in fact, they like to teach about them; they show our movie. However, the farmers are still urged to buy these technology packets of fertilizer and that is a problem because as you have already learned volcanic rock is rich with nutrients. Here is a test that we did adding phosphate fertilizer in six different fields, six subaks, from zero phosphate to the recommended dosage of 100 kg per hectare. [It] has no effect on yields. It is just wasted because the phosphate is already present in large concentrations. Once the government understood this, they allowed the water temples to regain control. But the Green Revolution still lingers. To this day, farmers add chemical fertilizer to this ancient self-sustaining system. For the last thirty years, the farmers have been borrowing money from the village cooperatives to buy fertilizer that they don't need, plying it to the fields. It washes out of the fields immediately; it flows back into the rivers and down to the sea. This little stream is flowing right out of those rice paddies up there. As it comes down, it is of course carrying all of the mineral nutrients from the volcanic soil, plus, all that fertilizer. All the fertilizer that wasn't needed by the farm; it is just washing down. By the time it gets here, to the sea, it is like a thin nutrient soup. The effect is that you grow simple organisms like algae. [Like] the algae that you see growing along the rocks there. That is what we find offshore, just blanketing the coral reefs. We only find it in places like this where you have that kind of agricultural drainage. The rest of the island, if there is no river carrying fertilizer, then the reefs are fine. But out there the reefs are nearly dead. Stephen Lansing and his colleagues are gathering samples from reefs around Bali, trying to understand the complex web we humans are part of. They hope to help Bali in its return to harmony. That's David Zuzuki. The problem is all of these nutrients flowing into the reefs and coral reef are adapted to low nutrient levels. That is not a good thing. There is Jim Kramer taking samples as one of the little streams emerges into the field. One more little clip; this is Dick Murphy, who may be here tonight. Talking about"� Tabular or table coral is accrapoor coral, some of them [are] the diameter of this boat. They are in perfect health. Whereas on the mainland, we also saw some pretty large corals but most of the large ones were dead. That tells us that in the past the conditions were such [that] the reef was healthy and those big corals lived long enough to really grow to a substantial size. But they're not alive now. So something has changed. Our conclusions are that the technology packets of fertilizer were mostly superfluous. I have a Balinese student who just got his Ph.D , sort of proving that by taking measurements along the river for over a year. The excess phosphate is not good for the terrestrial ecology and the excess nitrogen that grows the algae that destroys the coral. We can see that through looking at stable nitrogen isotope. The difference is that you can actually get the signature of the kind of nitrogen that is growing the algae; does it come from the sea or does it come from the technology packets? It really is a very foolish mistake. Why did we miss all of this? Why did we miss all this? Well, I think the answer is"�if we think about what are the elements of this complex adaptive system that I have been talking about? Its water temples, which look like a religious system to us; that is what we thought they were. The festivals that are held in those temples; here we see a priest from the Master Water Temple. He is carrying a little silver cup in which he has got holy water from the lake, from the goddess of the lake. But if he were to stumble upon this temple festival, (and you see a guy in white with his little silver cup sprinkling water on [the] offerings), you are not likely to be thinking about control mechanisms, right? Synchronized cropping, that is maybe a more obvious cue but it took something to pick it up I guess; the structure of the connections between the water temples through the holy water, which goes from the lakes, to the canals, to the fields. That really is the structure. Finally, there is this dimension of the control of the emotions. The cosmological order that allows the farmers to cope with the jealousies and disagreements that is common to everyone, including Balinese farmers. Not an obvious system. If you wanted to look at the connections between two subaks in the hierarchy you would need to be there on the right day when a few old men in white clothes go up and collect some cups of holy water to bring down to the fields of their own subak downstream. I gave this talk at the University of Michigan a few years ago. At that point I was ending it by saying, "Is this only in Bali"? The complexity dynamics really should be characteristic not only of Bali but they should be common, if we are right. The tradeoff between pests and water for example generates a network structure but if the farmers were growing roses and trying to control afas, you should still get some kind of dynamic. I posed that question and Lisa Curran, who is a tropical forest ecologist who spent"�well, she's younger than I am, not as long in Borneo anyway, said "We gotta talk". I'm going to tell you a little about Lisa's research on the forests of Borneo. (Let's see). Lisa has studied the dipterocarp; those are the tropical hardwoods of which there are 12 genera and about 470 species. She is the real expert; she can identify from the seeds, the fruits and the nuts what kind of dipterocarp it is. In the time that she had studied them, most of them in Borneo, have disappeared. Most of them are gone. It is a tragic story. Why did our forestry planning go so far wrong? More five year plans. Well, it is connected partly to, as Lisa found, to El Nino, the El Nino southern oscillation, the weather pattern which circulates across the Pacific, which you heard about. We knew that ENSO as it's called, the El Nino cycle, changes weather patterns, from the Galapagos. It sort of takes the weather from the Galapagos and Borneo cycling back in a several year cycle. What we didn't know, that Lisa helped clarify, is that these trees, the dipterocarps, use ENSO as a signal. What this signal does is to tell them to have a mast. A mast means: drop your fruit. It means 'reproduce' if you are a tree. Dropping the fruits and seeds, that happens for the dipterocarps only during the El Nino years. It is just a trigger signal I'm supposed to be on a new slide here. Here you have, if you can see that, these are the El Nino years and these are the fruits, patiently collected by Lisa and her team. The density of seeds dropped in the forest, so you can see it exactly tracks those El Nino years. Now you see why Lisa wanted to talk about the water temples. Many species, (because this is a tremendous pulse of food that arrives only in El Nino years): [including] the seed predators, which means lots of birds and mammals, the bearded pigs. The largest migrations of mammals outside of Africa are occurring out in, or used to occur, in Borneo in El Nino years when the bearded pigs would have their babies. These are great big animals, also the birds. All these creatures would have adapted. Interesting evolutionary dynamics, they will have adapted to the cycle their reproductive activity for the dipterocarp mast, which then creates a kind of master clock. It is a clock in the heart of Borneo. As far as we know, this does not exist in other tropical forests in Africa or the Amazon but then again, we haven't really looked very much. Lisa has begun to work in Borneo; we know so little about these things. The point I want to make here is that it is very much like the water temples. This is synchronized fallow controlling predators. The trees are doing pretty much what the farmers are doing, except in this case the trees are extending their fallow period over the whole island. That is a huge region; she has been working with NASA to see [just] how large is the synchronization of fruiting and masting in the dipterocarps. It is enormous as far as we can see in Borneo. That is what pulses food availability and keeps down the sea precia it makes it possible for the dipterocarps to reproduce because it provides so much food that the sea predators can't eat it all. But the plans for forest reserves, the five year plans, were based on the residency predators, rather than the pulses, the large numbers that occurred during the El Nino years. Lisa has done experiments showing that the sealings can suffer 99 or 100 percent mortality during the El Nino years. The rest of the time they would be fine and therefore the dipterocarps cannot reproduce. Here you have in 2002, from Lisa's Science [magazine] article, one of her Science articles, the logged areas. The yellow is the logged areas and the little green patches are what is left of the dipterocarps. There are not enough trees left for the mast. There has been a reversal and so now it triggers, not the regeneration of the forest but fires. Because the micro climate is gone and it is no longer wet, the trees have been cleared and not only will the forest burn but actually the peat swamp will also burn. Lisa is in Borneo now working on that problem. That turns out to be a lot of carbon emission. You lose the trees, the trees are no longer fixing carbon; [and] then you have forest fires in the El Nino dry years. Now also peat fires, so we are talking about .8 billon metric tons of carbon per year. The Kyoto global target for reduction was .5. So it is huge. Not to mention the loss of biodiversity. We are moving along here. Two examples then of complex adaptive systems, one in Borneo, one in Bali, different dynamics but in some way, maybe in some ways similar dynamics: bottom up control emergent networks that work and then were disrupted. I want to end with another way of thinking about this, not just the complexity perspective but a Balinese perspective. Maybe another way to think about the long now. The Balinese have very different ideas about time. Here is a simple little board which is a Balinese ukarutika calendar and it looks like, if you can see it clearly, which you can't. It is like a matrix. It is thirty columns and seven rows, well actually it's nine, one on the top [and one] on the bottom. [This is a] funny looking little thing, little wooden board. Let me explain what it is. Think about what our kind of calendar does. We have an Indo-European"�in the Indo-European languages we speak of time in a linear fashion. We think of time as a moving moment; the past is behind us, the future is in front of us. The present is this moment right now, in between the two; that is how we tell time. That's actually the arrow of time so to speak as someone who is embedded in our language because we make order of things but talking about tense. When did things happen? It's obligatory to speak in English or French or whatever in terms of past, present, future tense. That is how we talk about things. Balinese language, Indo"�excuse me, the Australasian languages don't have tenses. They create sense in a different way. They have a very interesting way calculating time. I think it is the most complex, at any rate the most complete way of thinking about time. It is in terms of multiple concurrent cycles. Time has dense. Let me just quickly walk you through what those calendars do; this will just be quick. First of all they have a lunar calendar. We kind of have a lunar calendar, from the Romans. But they are accurate about it. They count fifteen days of the waxing moon, fifteen days of the waning moon but then they subtract a day every sixty three days. Since the lunar cycle is twenty nine days, twelve hours and forty four minutes; that keeps their lunar calendar accurate. They are actually keeping track of lunar time in that way. They also borrowed from India the Ashaka calendar, which is a lunar solar calendar which keeps accurate track of the solar year by interspersing lunar months. These months are of different duration. You stick them in at an interval of two or three years so as to get it right and keep accurate track of the solar calendar. Ashaka calendar began in 79 A.D. and that is what they used. Finally, they have this uku. So what is the uku? It is ten concurrent weeks; I hope I can explain this clearly. We have the seven day week: Monday, Tuesday, Wednesday, Thursday and Friday. Well Balinese also have a three day week. So the three day week is (Pasa-batung-hajah). So today is Pasa, tomorrow is batung and it just recycles. It is concurrent with the Monday, Tuesday week. There is also a two day week. Today is maungay, tomorrow is paput. That is also cycling along. There is even a one day week. In the one day week there is only one day, luang, definitely today is luang. All of this is cycling and I haven't got room to put on my little Powerpoint the ten, nine and eight day weeks, all of these concurrent, all of these going constantly at the same time. So time becomes dense. Here is now an attempt, (this is what the Balinese are now compelled to do), to try to plot all that information on a western calendar. This is going to be the March calendar because they also have to keep track of our system. In one of these little boxes you are going to have the information to tell you what week, what lunar day is it. What day is it in the solar calendar and what day is it in all of these ten concurrent weeks. Most people can keep track of a lot of this. This just gives you some clues"�you can't read it anyway"�it is gunoontagalog, the fourth day of the four day week. All of that is going on. What do we make of that? It suggests that time is dense. It suggests that the way to think about it is in terms of patterns that emerge from interlocking cycles. That is how you schedule water temple rites, is by those multiple cycles, the water cycles. The duration of the calendar is 210 days. That turns out to be the most important cycle; the reason that little board has 210 days on it. 210 days is the old growth cycle of Balinese rice. The master calendar is the rice calendar cycle. Within it you keep track of many other things. For example, most of the time, nobody cares what day it is on the eight day week. But when your child is born, the day of its birth on the eight day week give you a clue as to who it is, who has been reincarnated into your family, from the mother's mother's side, to the father's father's side. Human cycles are there and the notion is that you couldn't reach the end of this. There are so many different cycles going on, they are interlocking in so many ways, all we as humans can do is keep track of the most important ones. So the three day week is the market week. The market happens in one village one day, the next village the next, the third village the third day and then comes back again. That has nice advantages. Instead of the poor old retailers (here) having to sit in their shops from nine to five everyday, they're there every three days. They triple their market size and there is less of just sitting and being. The reason it works is that we have inscriptions from a thousand years ago; the market cycle never changes. If this is the village and market is on day kajung, day one of the three day week, it will never change so it is entirely predictable. [There are] interesting ways to use time to create order and maybe save people some time. They also use the metaphor of music. Judith Becker, years ago, (a linguist, a great thing), 'Time and Tune in Java' showing how you think about music. Music is composed of multiple interlocking cycles. That is how gamelanÂ music is put together; something like fugues. Order appears if the cycles are integrated well and it doesn't if they are not well integrated. Music and time are clearly metaphors for one another. That little image from a Balinese traditional manuscript called Prakempra which tries to relate cycles of music and tone with letters, colors, gods and emotions (and all those things). It is one great package. I am at the end here because we began with John Holland's work on complex systems and his notion of what a complex adaptive system is. But his more recent book is called 'Hidden Order'. I want to suggest to you that there is an interesting hidden order here. It has taken me years as an anthropologist to begin to see that these funny little boards and things actually can hold very profound ideas. I want to suggest to you, seriously, if you are thinking about time, that this is a way to think about time that may be worth considering. Because this is the time that ecologists think about really, it is not the physics of time. It is the ecological view of time as having many cycles. If we don't think about those things, I think what I've shown you is a litany of horrors. We've made colossal mistakes. My colleagues, the people I work with in Indonesia, are very worried people. What I've seen in my lifetime, what Lisa has seen in hers (and these stories could be multiplied), there aren't very many people trying to study these systems. Lisa is now dividing her time between Borneo and the Amazon (because there is still a lot left of the Amazon). We are just beginning to twig to the most basics patterns, complexities giving us some clues, simple clues about what to look for. In the meantime, these systems are collapsing right before us. I am sorry to say people like us: we are the ones who are responsible for many of these changes. If we are going to continue to push forward, in what one geologist calls the "the anthropo-scene - this new geological epoch in which human beings control of the planet, [then] we need to get smart fast. That is my story. I need to acknowledge my many collaborators. This is many years of work and I think I'll just list them there rather than trying to read all their names. Thank you very much.