Ganges Water Machine


"The Ganges River begins at the base of the Himalayas and flows over 1,500 miles to Bangladesh, where it discharges through a massive delta into the Bay of Bengal. It is one of the largest, most urbanized, most productive river basins in the world. The Ganges Canal System contains more than 4,000 miles of distribution canals and irrigates thousands of square miles with water from monsoons and Himalayan meltwater. Ganges Water Machine: Designing New India’s Ancient River (Applied Research + Design, $49.95), by Anthony Acciavatti, presents a landscape history of how the Ganges water system came to be. The book was recently awarded the John Brinckerhoff Jackson Prize by the Foundation for Landscape Studies, and was the subject of Acciavatti’s presentation last year at the River Cities: Historical and Contemporary symposium at Dumbarton Oaks. Acciavatti is a principal at Somatic Collaborative alongside Felipe Correa, where they are developing largescale urban and landscape infrastructure projects, primarily in the Ganges Basin and Latin America. Acciavatti teaches at Columbia University and is a founding editor of Manifest: A Journal on American Architecture and Urbanism. Recently I had a chance to speak with him about the book, and the conversation included topics such as representation as a tool to understand and operate in a politically charged fluvial landscape, the importance of specificity when conceiving landscape change, the role of soils in an urban river system, and the potential of a newfound emphasis on shape and form. This interview has been edited and condensed. Brian

Davis: This book is beautifully drawn and useful—comprehensive and specific at the same time. Could you tell me a little bit about the basic intent of the book? I know that you worked on it for a very long time; how did it start? 

Anthony Acciavatti: I began writing the proposal in 2004 and got a Fulbright grant to do the work in 2005. The intent was always to create what I called a “dynamic atlas” that would be a set of drawings that would not just kind of map out the space of this area, the kind of static object — settlement versus agriculture versus open lands, roads…that sort of stuff—but map out how this space changes over time in two specific ways. One kind, historically over time—over the last 200 years—and another in terms of cyclical time, with the monsoon. I knew this area was one of the world’s most densely populated river basins, that it was agriculturally productive, and had not been mapped comprehensively in almost 40 years at that time. It is very hard to talk about an area that is so spatially diverse and undergoes these radical changes of state every year. The project was to create a set of maps by walking the land, traveling the river by boat, and to travel much of the area by car as well. The intent was to make an object, a dynamic atlas, that people could look at and use to make their own propositions about how to deal with issues such as settlement and monsoon agriculture. 

Davis: A great deal of the book was really a fantastic history of just how this machine, as you call it, came into being. Maybe we could talk a little bit about how you chose the period of significance? You referred earlier to about a 200-year period. How do those timescales relate to the idea of a river landscape as a “water machine”? 

Acciavatti: I think the idea references a paper from Science magazine in 1975 by the head of population studies at Harvard, Roger Revelle, and an engineering professor from the Indian Institute of Technology, V. Lakshminarayana, on thinking of the Ganges as a larger machine. In this case they were talking about how the monsoon and all the infrastructure, as well as the groundwater table and groundwater banks, act as a giant or colossal machine. I saw that concept as going back to the creation of the Ganges Canal, where the engineer working on that project in the 19th century refers to it as a machine. When it opened in 1854, it was five times the length of all the canals in Egypt and Lombardy, Italy, combined. This becomes a model for India but also for the United States and Australia. In fact, because of the demands of this landscape, India develops its first school of engineering—before the United Kingdom. This history is similar to what Richard White was describing in the late 1990s in The Organic Machine, his book on the Columbia River. The Ganges Basin is one of the most engineered spaces on the planet. Yet usually when people talk about this part of India, they speak about it as being atemporal, traditional, and outside of modernity, when in fact you could argue that this is one of the most modern places on the planet. 

Davis: One of the things that’s striking about this 200-year historical frame is it’s really, or at least approximately, when the British come on the scene. Yet throughout the book a lot of the either traditional practices, or the ways that Indians are adapting some of these ideas for their own purposes, seem to be a part of a story that you’re telling. 

Acciavatti: Often when people think about the history of technology, the concept of succession is very important—newer technology replaces older. But in this context it is really about layering these things. I think of those things as layers and how they intersect with one another. It gives you a very different constellation, almost galaxy, of canals and tube wells within settlement. The layering of new technology together with other modes and forms of knowledge had all kinds of effects on human and ecosystem health. For instance, the water table was changed, and malaria outbreaks became even more frequent. The system imposed a new spatial order, a kind of urban agrarian grid. The engineers were aware of this, because they looked at flanking the canal with shops, and how bridges would cut across it and also allow for ghats [areas in holy cities along the Ganges where stairs are created to reach the water]. It was never thought of as monofunctional, even though I think today a lot of times the irrigation department in India thinks of it that way.

Davis: I’m wondering, were these previous efforts also done with this dynamic idea you focus on at the forefront? If not, where did this interest in dynamism, or emphasis on dynamism, come from in your work?

Acciavatti: In the maps that had been done before, they’re your kind of typical survey maps where they show villages and settlements and major roadways, and they show surface water bodies. They don’t really show the dynamism of this area. One of the reasons that I became very interested in this is I grew up near the Mississippi. I lived for a time in New Orleans, and I had always been very interested in the relationships between cities and rivers.

Davis: Dynamism and change have always been important in landscape, but in the last 15 years it has come to the forefront. A lot of that was coming out from folks [such as Kristi Cheramie; Jane Wolff; Bradley Cantrell, ASLA; or Elizabeth Mossop, ASLA] who were living or working in Louisiana, but others are interested, too. I am working on a project in SЛo Paulo and was talking with a designer down there, Alexandre Delijaicov. He was trying to describe urban rivers and settlements in that city and said that rather than thinking of a river as a channel with two parallel lines and a floodplain, he found it a lot more useful to think about this idea of big river and little river. It is a simple but significant shift. Suddenly the flood is not this aberrant condition, but a fundamental aspect that is ephemeral. I think that is something that your drawings are really driving at and making operative. In them, it becomes palpable how a designer could start to work with that idea. Is that the intent?

Acciavatti: I was always very much interested in the relationship between monsoon and river channel in a context like India, where the channel goes from being only four and five meters wide to being two and three kilometers wide, and then shrinks back down to those four or five meters over the course of a year. Even more interesting than just water is the soil composition. Soil is how these landscapes hold water or move water along. It is fascinating to see that in the context of northern India, where it goes from being arid and parched land to being wet, gelatinous silt every year. Those soils act as sponges, primarily because of the almost 1.6 billion tons of alluvium that are shed from the Himalayas every year. [Through] the drawings I tried to formulate a way of using a single solar cycle to map out all of those changes from soil structures to watershed. 

Davis: You mentioned that you started the work before high-resolution satellite photography was readily available. That required you to spend a tremendous amount of time on foot collecting your own data. I think that is something that often gets lost nowadays, where a lot of design work is done using data that someone else has produced. Your work here was the opposite of that, it appears. How did you go about this, trying to organize this fieldwork?

Acciavatti: I think the question of data is always an interesting one. A lot of times we talk about it as being apolitical, and about it being neutral or objective. In fact, just by the very nature of having to collect it, data is a highly constructed object or representation that has to be interpreted. For me, in some ways I was lucky that there was very little spatial data in these areas. I really had to just set out. All that I had with me was a digital camera and a handheld GPS unit. Roadways and river edges are incredibly important in a context like this just in terms of transportation and movement of goods. Looking at roadways, then I think trying to basically say, “Okay, what can I actually find from being on the ground?” Of course I can detect roads. I can detect settlement. I would then map and photograph how these soils and these landscapes changed over the course of a year, trying to index those changes. Even just in terms of color, they go from beige to chartreuse green, then to a deep, almost forest green, then to brown and beige again every year. Also, just because of the scale of this territory, I relied on using this very old technique of the transect that was first developed by Alexander von Humboldt. Its lineage starts there, but it has also been a powerful tool for Patrick Geddes, Ian McHarg, and nowadays, the New Urbanists. I saw the transect as a method to go into greater detail about the landscape and to vividly figure the changes that take place here every year.

Davis: Those transects are the most striking and richest drawings. You used those more than the maps themselves to give an incredible account of all the instruments that make up the landscape, be it the tube wells, specific crop species, soils, or local farmers. To make these drawings, how were you deciding what went into the transect and what did not? Was the work of Anuradha Mathur and Dilip da Cunha an influence for you?

Acciavatti: I really appreciate their work, but would say my work is very different. I am quite concerned with issues of private and public ownership of land and infrastructure. I explore issues of ownership in terms of surface water and groundwater, as well as the ways in which private property shapes the vast mosaic of agriculture and settlement across the Ganges Basin. For me, this mode of drawing is more tied to the tradition of, say, Patrick Geddes and von Humboldt, as well as the Limbourg brothers. Geddes’s shorelineto- ridgeline transect showed how changes in topography and elevation were related to different tools and [practices]. While I don’t share Geddes’s technological and environmental determinism, I do think it is important to draw how infrastructures are used to mediate climate and weather, as well as how people use them to augment and change their property, and inadvertently that of their neighbors, too.

Davis: How do you imagine these drawings making a difference in practice in general? Maybe not these drawings specifically, but this way of working and thinking? And are you working with local partners there to put some of these ideas into practice? Acciavatti: Yes, at Somatic Collaborative, Felipe Correa and I have been working on a set of proposals that we hope will be developed into pilot projects either in Uttar Pradesh or maybe a little bit to the east in Bihar. But more generally, one of the practical uses, I hope, and one of the reasons I do not include the set of proposals in the book is because I really wanted this to be an object that people could use as a resource to make arguments and to make their own propositions. Basically hallucinate upon alternative futures for this area.

Davis: Jumping back to a representational question, there was something I was interested in that was a beautiful

part of the drawings. Using the thalweg —the lowest line in a river channel— you do these beautiful studies of river morphology. I rarely see that in landscape architecture drawings. I am wondering, how did you use that concept to understand specific morphological characteristics of the river?

Acciavatti: When I was an undergraduate, I was in a studio looking at the Charles River. In that studio we were focused on the point where the river meets the bay, and the tidal fluctuation that happens there. I became very interested in trying to record and understand how those sediments and debris move over time. In my drawing I use the thalweg to index the curvature of the river, and that makes it easier to compare and contrast changes in its geometry. It also indexes the movement of soils and edge conditions, which are related to groundwater levels and sedimentation patterns. In a lot of these areas where the river expands and contracts significantly, these patterns are foundational, not just for agriculture but also for sand mining. One of the questions this brought up is, how do you move away from New Orleans-style barrages or levees that line the river and instead allow for the dynamism to take place here within this area? How can we define points that would work wellin terms of developing a port and linking infrastructure? Those are some of the projects we are working on right now. You have to do a drawing in a very particular way to explore a set of issues like this. Those thalweg drawings help me to think differently about what I usually would take to be a pretty straightforward interpretation of how a river expands and contracts.

Davis: One of the things that I love about this drawing is that you’re studying the shape of things. Understanding that those different shapes have meaning and possibilities is a really important way forward. There is a tendency right now to look toward ecology when talking about landscape issues. That’s really interesting and important, but ecology focuses on the relations between things. It tends to deemphasize morphology, shape, and form. But in fact, form is fundamental—contemporary discourses about social, cultural, and environmental issues are impoverished without it. And it is something designers excel at.

Acciavatti: You touched on a really important issue. You’re right that ecologists do not have a great handle on it. What ways work better for doing certain things and undertaking certain processes and giving primacy to some and not to others? I think this is really important, and it requires judgment