CEE New Millennium Colloquium
March 20-21, 2000
Wong Auditorium, Tang Center, MIT Building E51
Natural and Built Environment
DR. JARED L. COHON
President, Carnegie Mellon University
It's a wonderful honor for me to be back here, especially, speaking to this crowd, to my former faculty and my current faculty at Carnegie Mellon. I am, therefore, doubly nervous. I was Dave Marks' first Ph.D. student and he told me I was his best Ph.D. student until he got a second one. I thought was pretty cruel of him.
I will be as brief as I possibly can and still be provocative. That's my goal. It's really to set the stage here and get some discussion going.
The topic as you heard from Phil is "The Natural and Built Environment". I thought a lot about the title of this session and I wondered if they really meant "The Natural versus the Built Environment". I don't think so. Or, maybe we should just be talking about "The Environment" and why we make this distinction. It's a very important issue and indeed it goes to the most fundamental question of all: the role of humans in nature; our relationship to the natural world; and the impacts of the things we build on the natural world.
I'm going to talk about basically two topics. First, the built environment itself and a little bit about infrastructure and infrastructure planning. This is basically to prove that I've done something other than being an administrator over the last few years or at least an attempt to prove that. The other is sustainability, which is closely related to the first and provides an even broader context for our discussion. I'll close with some challenges for the future as I see them.
Let's start with the built environment. Just to review the obvious: the built environment -- the things we build, the facilities that we create and build and maintain -- are all intended to provide certain things, that are fundamental to people's lives, i.e., water, food, and shelter. Interestingly, some of the facilities we build, like flood control, are to protect the other facilities we built already. Others are built in support of economic activity, e.g., resource extraction products and transportation systems.
I want to take a moment to focus on cities and to talk a little about their history. Although I'm most interested in getting us to think about where we're headed as opposed to where we've been, we can learn from history. A colleague at Carnegie Mellon, an historian named Joel Tarr, has suggested four eras in the history of the US city. (This is applicable to the urban areas in the developed world generally; I'll say something about the developing world in a moment.) Up until 1880 it's fair to say our cities were pedestrian cities. They were designed and constructed to support pedestrians (and horse traffic). It was in the late 19th century and early 20th century that the city became, in Tarr's words, "piped, wired and tracked", as civil engineers built water networks, sewer networks, etc. With the rise in automobile use and over a several decade period, the cities stared to expand and expand very significantly. We saw the rise of the suburbs, and since 1960 the rise of edge cities and the movement towards the megalopolis.
Think for a minute about developing countries. I'm sure everyone in this audience has traveled to developing countries, and to the urban areas in those developing countries. Think about what you've seen in those cities, especially their scale. The point of all this is that the transformation of the earth is partly due to the development of huge cities, and we ain't seen nothing yet. E. Ayensu, et al., writing in Science in 1999 estimated that 40 to 50% of the earth has been irreversibly transformed. Now they weren't thinking just about cities, but much of the other land transformation, that they would claim has been irreversible, was pursued to create the products that feed the cities. So the city is not a bad surrogate for human impact on the natural world. And the question for us is, is this sustainable?
To expand on my point that we've ain't seen nothing yet, Ian Johnson, a Vice President of the World Bank, estimated that 1.5 billion people do not have access to clean water now. To us in this room, this is a number (or similar ones) that we've heard before. To me that is amazing. It's shocking, really, that we have not yet solved that problem.
Think also about the transportation systems in urban areas of the developing world. Someone once told me the average travel time from the airport to downtown Bangkok is 6 hours -- on good days. It's incredible and it's going to get worse. Here's one person's estimate of how it's going to change. Platt estimates that by 2025 60% of the world's population will live in urban areas (Platt, Rutherford H. (editor), The Ecological City: Preserving and Restoring Urban Biodiversity, February 1994). That is 5 billion people. That's more than twice what it is now. And, you can see from Platt's figures we'll be experiencing both population growth (from 5.1 billion people in 1990 to 8.5 billion in 1995) and the increasing urbanization of that population (from 43% in 1990 to 60% in 2025).
Let me get to the key point, though. Michael Huff, who doesn't sound like a friend of civil engineers, offered the two following quotes in his book, Cities and Natural Processes. "The underlying disciplines" -- that would be us "that have shaped the city have little to do with the natural sciences or ecological values." "The notions of humanity and nature have long been understood to be separate issues." To me that's the major challenge. Let me tell you about some work that was done a few years ago by a committee formed by the National Research Council to look at the problem of measuring and improving infrastructure performance. It was a highly multidisciplinary committee, a committee that I chaired and a committee that, in fact, produced such a framework. The committee had practicing engineers and public works in mind when it created its measurement framework
As part of its work, the committee visited three cities an older city in the east, a relatively new city in the west, and a progressive city in the Midwest. We spent two days in each city interviewing the public works director, local politicians, and getting to know people, trying to understand the kinds of challenges they faced and the way they went about assessing infrastructure performance. One Director of Public Works said, "Oh it's easy. I count complaints," which was his idea of measuring infrastructure performance. This same person, when asked what his major problem was for the future, said, "Our major problem is damage to road services from the trucks coming from and going to the port". Mind you, this is a city where the port was the major economic activity and the perspective of this Director of Public Works, charged with providing infrastructure to this city, was that that port was nothing but a headache. Part of the problem was the mix of jurisdictional responsibility: the port was operated by the state; he was a local city official. Nevertheless, that was his perspective. He was a trained civil engineer, but, I'm pleased to say, not a graduate of any university with which I've been affiliated. If that's the kind of narrow thinking that civil engineers are going to use in the management of infrastructure, we will become marginalized. We will lose the position of being the shapers of our cities in the future, of providing infrastructure, and we will deserve it.
That's what motivated the creation of this framework that I've referred to and that I'm going to spend just a couple of minutes on this. If nothing else, it will give you something to argue about. Figure 1 provides a framework for evaluating infrastructure systems. This is a big problem, and a very broad one with many different components. It's multidisciplinary, it's multi-jurisdictional and it's certainly multi-objective. First of all there are many different systems within infrastructure; there are water systems, transportation systems, and waste systems, and even other modes that we don't often think of as infrastructure. Well, maybe we do now but we didn't when we first built these systems. Here, I'm thinking of communications systems, which have become crucial elements of urban infrastructure. However, our committee only focused on waste, wastewater, water supply and transportation.
Infrastructure also exists in various spatial scales. As I said, it's a multi-jurisdictional problem. It goes from the individual or the household -- the customer, if you will -- all the way to the globe, at least to the extent that infrastructure systems have potential global impacts. We focused on county and city government, regional governments if they existed and states.
We suggested that to think productively about infrastructure systems the public works director (the civil engineer) should be thinking in terms of being embedded in a rather large, and in every case I've ever seen, very complicated political system. There's the Public Works Director surrounded by the city political organization, the mayor, city manager, whatever the organization may be, surrounded by regional organizations and the state and federal governments. That creates an environment for regulation and for decision making that the public works director must be sensitive to, i.e., he should not be thinking that the port is nothing but a headache because it tears up the director's roads.
We offered a process for developing an approach to assessing infrastructure performance (Figure 2). Before assessing performance, you should spend substantial time identifying the decision-making context and the framework for decision-making, identifying the system, agreeing on a vision, and identifying objectives and measures, all of which should be strongly influenced by local values. Only after completing all of these preliminaries should system performance be measured. I would submit -- and the Director of Public Works is a good example of this we spend too little time on these preliminaries. Civil Engineers spend too little time on these issues, not just about infrastructure, but in general about that which is the built environment.
We thought of infrastructure as a multi-phase process. You plan infrastructure, you implement it, you build it and then you evaluate it and hopefully you iterate and go back to the planning phase. Of course you don't build all infrastructure all at once. Within a given infrastructure system, different subsystems are likely to be at different phases. Furthermore, the measures one might use to assess performance are likely to change from one phase to another. The measures that you might use in a planning context might be different from an implementation process where you're much more interested in efficiency and containing costs whereas up here (in planning) you should be thinking about broad regional visions and what it is you're trying to achieve.
Our committee proposed a system of measures, not specific measures but classes of measures (Figure 3). We recognized that the traditional view of infrastructure systems tends to focus on what you might think of as inputs or dimensions of a system. So how many times has a transportation systems manager, when asked "how effective is your system?" might answer, "I've got 25,000 miles of road". That's not an indicator of system performance. It really says nothing about system performance. It misses the point about why you have a transportation system. Nevertheless, it's important to know things like how many miles of road you have, etc. These are system descriptions, but we also need the measures of effectiveness itself, like travel time and reliability and cost which represent a class of measures relevant to all infrastructure systems.
This is one approach to the very difficult and growing problem of infrastructure management in developed countries, where the major issue is rebuilding and maintaining existing systems and what city knows that better than Boston? Developing countries with their huge and rapidly growing cities face a different problem. There, the existing systems are poorly developed and are failing to provide necessary services. And it's going to get much worse before it gets better, I think.
Is what we're doing sustainable and are we moving in sustainable directions? That was a question I posed earlier but I didn't offer an answer, and I'm not going to offer an answer now. But I do want to talk about sustainability. It provides a broader context, the context within which infrastructure planning perhaps should occur. Sustainability is a difficult thing to define, but it's an issue that we need to struggle withand indeed many people are and I hope we will today in our discussion.
Sustainability is the topic that seems to come up whenever we talk about environment and especially if we talk about the built environment and the natural environment. It's especially motivated by the difficulty of non-point sources now that point sources in this country are largely controlled, cumulative impacts (a very difficult issue) and population growth. And in addition, of course, there is the increase in multinational and especially global problems of the sort that we heard Bob Correll speak so interestingly about this morning. Sustainability is intimately and unavoidably connected to this question of the role of humans in the natural environment.
The most widely used definition of sustainability still is that of the Brundtland Commission which wrote in 1987 that sustainable "development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs". This is a remarkably homocentric definition. But for understandable reasons, it is very general and difficult to implement in practice. Interestingly, our organization, ASCE, recently issued a definition that I find quite attractive for its implementability. The ASCE Task Committee wrote, in 1998, "sustainable water resource systems (but I think they would be willing to extend this to other systems) are those designed and managed to fully contribute to the objectives of society, now and in the future (and here's the key phrase for me), while maintaining their ecological, environmental and hydrological integrity." Now that's something we can work with, but does raise a very difficult question. What is "ecological, environmental and hydrological integrity?" We don't have a good answer for that. Actually this gets to that deep philosophical question of what's natural and what is our role in this world?
When I was the Dean of the Yale School of Forestry and Environmental Studies, I met a National Forest Supervisor in Montana who was really into sustainability and natural forest management, which led me to ask the obvious question: what to you consider to be a natural forest? His answer was the state of the forest before European settlement, which is not a bad answer. It's perfectly defensible. But why? Why is that a natural forest? Why not the state of the forest in 1940 or 1022, if he knew what it was? There's nothing magic about the European settlement of North America. We are here, after all. There are six billion of us; there will be 8 billion in 25 years and 10 billion, I guess, in 50 years. And we're here to stay, at least for probably many hundreds of thousands of years, or let's say tens of thousands anyhow. We have to reconcile this somehow. We have a role in nature. It's not the absence of humans that defines nature. We have to come to an understanding of how what we do, including our constructed environment, fits in with the natural environment. I don't have answers for you. I've got questions, some of them courtesy of Pete Loucks, Gene Stakhiv and Lynn Martin who recently published the paper "Sustainable Water Resources Management" in the recent (March/April, 2000) issue of Water Resources Planning and Management. Prompted by the definition that I just showed you from the ASCE Task Force, they posed the following questions. "What is or should be sustainable?" "Should everything be sustainable (or sustained which may be a better way of putting it), and if not, "what need not be?" "How do we account for future technological developments that may mitigate many of the adverse effects?" "What resources are vulnerable to irreversible decisions?" "Of all the possible strategies, which is the best?" That's actually my favorite question and a related question is "What are the goals and who should decide them?" I would submit that all of these are relevant to civil engineers and in fact we have or should have a role in answering all of them and I don't think we do to the extend we should.
Let me turn to challenges. There are many but I've chosen to focus on three categories, starting with integrated assessment. We've heard a lot about this already from Bob Corell. It's a very important challenge, on which we've made a lot of progress but with a long way to go. And I use the term broadly, to include global climate change, where integrated assessment is a very active research topic, as well as virtually any environmental system. And what it's trying to capture is the idea of both a systems view, a concept that needs no explanation to this audience, and the multi- and interdisciplinary nature of environmental problems. We're not just talking about the physical and biological connections among subsystems. We're also talking about the social, economic and political aspects of this as well. A related and very important challenge is uncertainty and dealing with very long planning horizons. How do we make decisions today that will have consequences for many, many decades, or even centuries or millennia to come? It's an extremely, uncertain decision-making environment. How do you make decisions in such a context?
The second major challenge is the need for more science and data. There are many areas of science that have to advance but the one that I want to focus on is ecosystem health. I mean both aquatic ecosystems and terrestrial ecosystems. We've made progress in this area recently, but we need to know much more if we're going to answer the questions that Loucks, et al., posed and that are imbedded in our definitions of sustainability. If we're going to try to set out to protect ecosystem health, and that will be the centerpiece of sustainability, we better know how to measure it.
And we need much more data. The same article that appeared in Science that I referred to before (by Ayensu, et al.) made a call for developing a global inventory of data on major ecosystems. A good call, indeed, though a very ambitious one. I've recently became involved in water and sewer issues in the Pittsburgh area, focused on the three rivers area. I was amazed at how spotty the data seems to be. I mean these are basic water quality parameters of the sort that we've talked about for more than 30 years. And this is a very important river, the Ohio River, and is tributaries. It is astounding to me how little data we have on that system. And I don't know why. I think it's because we disinvested in data collection, I assume, at the Federal level, and the State and the local region didn't pick it up either. I hadn't been involved in a local water quality problem for many years, but my guess is that Pittsburgh is not very different from many other regions.
My third challenge is decision-making frameworks. In the context of my infrastructure evaluation framework, I emphasized the importance of focusing on goals and who sets them and that goes directly to this point. It's related both to the way we organize ourselves as governments as well as the techniques that we use for making decisions and the degree of participation in those decision making processes. On the former, let me give you an example from some work I did for the UN in India over fifteen years ago. My job was to focus on integrating environmental impacts into river basin planning models, working with the Central Water Commission in India. One of the things that came up was the positive benefit of water development for human health. In a country where access to secure water supplies is such a big problem, it seemed like a no-brainer to me that that should be one of the components of our planning models. The answer I got was no. We can't do that. We're the Central Water Commission and water supplies are the responsibility of the Ministry of Health, not us. That really annoyed me. Mind you, we're talking about a stand pipe in a village which would cost a tiny, tiny fraction of the cost of the kinds of water schemes they were considering, while having an enormous potential impact on the health of the population of that region. They didn't deny that, but they felt totally constrained by the organization of their government.
Identification of criteria in our models, in our decision making efforts, is crucial and we don't always get it right (Figure 4). David Marks is going to want to kill me for presenting the following example. It is an early example of mathematical elegance that had absolutely no impact on real decision-making. This was one of the early products of the Rio Colorado Project, done at MIT in the early 1970's by many of the faculty here today. And I was a graduate research assistant in the project. We built a wonderful and beautiful, multi-objective linear programming model in one of the first major applications of systems analysis techniques to large scale river basin planning. And we realized that this problem was characterized by deep regional disagreements among the provinces that make up the Rio Colorado Basin. So we had a great idea. We developed two objectives: the usual net benefits objective and total deviation from an equal water allocation, as a measure of interprovincial equity. We used these two objectives to generate a stunning non-dominated set. Well, I didn't go but Dave took these results down to Argentina and he came back and he basically said, "Quick, publish the paper and forget this thing." The idea of an equal water allocation to some of the provincial decision-makers was a deep threat. They wanted nothing to do with it. Even the idea of talking about any kind of base allocation from which you would measure deviations was impossible. It was an elegant analysis and a great paper, but it didn't work. We didn't get the criteria right, and that is very, very important.
I think our work in Argentina was a good example for its time (Figure 5). Now, almost thirty years later we have new techniques, and we have learned from past mistakes. A good example of how to identify criteria was reported in a paper that appeared in 1992 in Water Resources Bulletin by Ridgley and Rijsberman about the Rhine River Delta. They worked with the Dutch Government in a highly participatory decision-making environment that was very successful, at least early on. I don't know what the final outcome was in terms of the implementation of projects, but it was very effective in getting the stakeholders involved. I was struck by Wayne Clough's comment this morning about his early encounter with people involved in the environmental movements in the San Francisco Bay and how he learned from that. Well, hopefully we will all learn from experiences like that and that we will realize what the Dutch Government has: you need to involve stakeholders. I don't mean inform stakeholders, which is still the way we tend to do it in this country.
Participatory decision-making doesn't mean "this is what I've decided and this is why you should accept it." It means involving people. But it's not easy to do. Just how hard it might be can be glimpsed from some of the work of David Orr of Oberlin College. Orr, who is very original and writes beautifully, coined the notion of "slow knowledge". What we do, everybody in this room, he would call "fast knowledge." In using this phrase, Orr had in mind the technological and professional nature of what we do; not just the pace at which we do it. In contrast, he uses the term "slow knowledge" to capture the knowledge of people who live in the ecosystem that's being affected by the infrastructure that you and I designed. It's the bamboo workers who put together the frame for the project that we saw in Bill Mitchell's talk. Orr observes that these indigenous people have developed a deep understanding of how their ecosystem works because they've depended on that ecosystem over many generations. This "slow knowledge" is very important knowledge. Sustainability, if we're ever going to achieve it, is going to rely in part on our ability to understand this knowledge, to integrate it into our decision-making and to integrate the people who have slow knowledge into our decision-making processes. We've never done that before. It may be among the hardest things we ever do, but I think we have to do it.
Chris Hendrickson, the Head of our Department of Civil and Environmental Engineering at Carnegie Mellon, offered three choices for where the Civil Engineering profession is headed in the future. Unfortunately Chris, who was here earlier, had to leave. Maybe that's fortunate so he won't hear me butcher this.
Civil Engineers as Master Technicians This would be surveying, as a dated and somewhat trivial example. We don't do surveying any more. It isn't even thought of as part of the Civil Engineering profession any more. In the future, we can expect other activities now thought of as part of our profession to become "commoditized" and performed by technicians. Or, if we are not careful, civil engineers will become the technicians.
Civil Engineer as Master Builder and Civil Engineer as Master Integrator are Chris' other two characterizations. We are and should continue to be builders, but without question, I think the Master Integrator is what we need to be. It's certainly is where MIT's Civil and Environmental Engineering Department is and wishes to remain and I think that's where we have to take our profession. To do otherwise, I think is to marginalize us and perhaps not to be sustainable if we're Master Builders. It depends on how one does it. There's more to be said about that. Maybe it will come out in discussion.
This is the vision statement from Version 6.6 of Civil and Environmental Engineering's Strategic Plan at MIT.
"Civil and environmental engineering research and education must cope with problems that require integration of science, technology, policy and management, often in situations rich in ambiguity and complexity. The important challenges lie in creating and restoring the built infrastructure in a manner compatible with the natural environment, and in understanding, and efficiently managing environmental processes that sustain life. Environmental responsibility and integrative thinking are requirements of the discipline. "
I think it's a wonderful vision statement. I like this very much. It captures, I think, all the key points that an MIT Civil and Environmental Engineering vision statement should. The emphasis is on synthesis, a great word that Rafael used this morning and has used before in his writings. It focuses on understanding complexity, emphasizes a systems view and on integrating all of the important pieces of decision-making.
Finally, I don't know if you saw this, but in ASCE News earlier this year, they published their top ten engineering achievements of the century, that is the 20th century, and it's something to be proud of. Civil Engineers have made today's society possible (Figure 6). Interestingly, it's airport design that was number one, not the interstate highway system as some would have thought or some of us might have said water supply in any event, they made it to the top ten. However, I wonder what the top ten list for the 21st century will look like. I hope it doesn't look like this. If it does, I don't think Civil Engineering will have achieved the transformation to a sustainable profession of being the Master Integrators that I think it needs to be.
Thank you very much.
Return to Speakers page
Return to the Colloquium main page.