Gotelli, N.J., and A.M. Ellison. 2004. A primer of ecological statistics. Sinauer, Sunderland, Massachusetts. xviii + 510 p. $34.95, ISBN: 0-87893-269-0.
[I reviewed A primer of ecological statistics in the March 2005 issue of Ecology. Click here to read the review. If access is prevented, email me for a pdf.]
Tuesday, August 22, 2006
Monday, August 21, 2006
Stauffer, D., and A. Aharony. 1994. Introduction to percolation theory. 2nd ed. CRC Press.
There is a classic book in environmental science with the title Consider a spherical cow (Harte 1988). Tongue in cheek, the author asks us from the outset to consider how far are we willing to depart from a perfect representation of the world in order to solve a scientific problem. Of course, we must first admit that simplifying abstractions are the basis of theoretical science, that in fact no scientific model will achieve the goal of isomorphism between dynamics in the model and dynamics in the world (van Fraassen 1980), at least not any model for a system sufficiently complex to be truly interesting. Instead, science proceeds by trial and error, in a dialectic of simplification and complexification, with the criteria for model acceptability determined by the problem at hand.
One of the simplifications that scientists have long had to “consider” is the fiction of homogeneous space. Fundamental models assume complete mixing in chemical reactions and disease transmission, and symmetry in the horizontal movement of water particles and foraging organisms. We assume that rates of spread for mutant genes, forest fires, or invasive species are the same in all directions. But, let us now instead consider a heterogeneous world. This we can do, for our experience is that wherever we look the world is heterogeneous, it is grainy. One of the great scientific insights of the twentieth century is just how thoroughgoing is the fractal geometry of nature (Mandelbrot 1982). Percolation theory is for scientists for whose problems the heterogeneity and graininess of nature matter.
Percolation theory is not a theory in the sense of the “special theory of relativity” or the “phlogiston theory of combustion”. Its purpose is not to explain a particular phenomenon. Rather it gives a set of tools to researchers interested in a variety of problems ranging from statistical mechanics to ecology. Imagine a grid in one, two, three or more dimensions. One can either focus on the squares (site percolation) or the lines between them (bond percolation). Now, randomly assign attributes to each square or line with a given probability. Since these attributes are assigned randomly sometimes squares (or lines) with the same property will be adjacent. Sometimes they will form clusters. Percolation theory is concerned with the statistical properties of these clusters. How many clusters are there of what size? What is the length of the perimeter around these clusters? At what point, known as the percolation threshold, does a major portion of the grid coalesce into one giant cluster? (This property, described here imprecisely, is defined for theoretical infinite networks as the point at which an infinite cluster first appears. Properties of finite networks can be related to the percolation threshold, too.) Initially it may seem that this level of abstraction and the simplicity of the properties that can be calculated would limit the range of applications. But, to the contrary, researchers in many fields have found ways to conceptualize discipline-specific questions in the general formulas of percolation theory, solving problems in physics, chemistry, biology, and social networks. The problems range from understanding transport of liquids in porous media (hence the name “percolation”) to the control of disease outbreaks to phase transitions in condensed matter physics.
Introduction to percolation theory is a widely recommended, much sought out-of-print introduction to the topic. As far as theoretical physics goes (theoretical anything for that matter) it assumes relatively little in terms of physics (none) and math (basic programming, differentiation and integration of functions of one variable, and elementary probability theory). It derives many of the basic known results and presents others that go beyond the level of mathematics for which the book was written. It introduces basic concepts of percolation, renormalization, and diffusion on porous medium. From this book, an interested scientist or student from any field in which stochastic processes are used to represent dynamics can obtain techniques to complement existing conceptual toolboxes, while the background obtained from reading this book will provide confidence to venture further into the technical literature. An ecologist by training, I found this book readable and informative with a rare style for the genre. As an introduction and reference for computational scientists in many fields, it is indispensable. Highly recommended.
Harte, J. 1988. Consider a spherical cow. Reprint edition. University Science Books.
Mandelbrot, B. 1982. The fractal geometry of nature. W.H. Freeman & Co.
van Fraassen, B. 1980. The scientific image. Oxford.
This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 License.
There is a classic book in environmental science with the title Consider a spherical cow (Harte 1988). Tongue in cheek, the author asks us from the outset to consider how far are we willing to depart from a perfect representation of the world in order to solve a scientific problem. Of course, we must first admit that simplifying abstractions are the basis of theoretical science, that in fact no scientific model will achieve the goal of isomorphism between dynamics in the model and dynamics in the world (van Fraassen 1980), at least not any model for a system sufficiently complex to be truly interesting. Instead, science proceeds by trial and error, in a dialectic of simplification and complexification, with the criteria for model acceptability determined by the problem at hand.
One of the simplifications that scientists have long had to “consider” is the fiction of homogeneous space. Fundamental models assume complete mixing in chemical reactions and disease transmission, and symmetry in the horizontal movement of water particles and foraging organisms. We assume that rates of spread for mutant genes, forest fires, or invasive species are the same in all directions. But, let us now instead consider a heterogeneous world. This we can do, for our experience is that wherever we look the world is heterogeneous, it is grainy. One of the great scientific insights of the twentieth century is just how thoroughgoing is the fractal geometry of nature (Mandelbrot 1982). Percolation theory is for scientists for whose problems the heterogeneity and graininess of nature matter.
Percolation theory is not a theory in the sense of the “special theory of relativity” or the “phlogiston theory of combustion”. Its purpose is not to explain a particular phenomenon. Rather it gives a set of tools to researchers interested in a variety of problems ranging from statistical mechanics to ecology. Imagine a grid in one, two, three or more dimensions. One can either focus on the squares (site percolation) or the lines between them (bond percolation). Now, randomly assign attributes to each square or line with a given probability. Since these attributes are assigned randomly sometimes squares (or lines) with the same property will be adjacent. Sometimes they will form clusters. Percolation theory is concerned with the statistical properties of these clusters. How many clusters are there of what size? What is the length of the perimeter around these clusters? At what point, known as the percolation threshold, does a major portion of the grid coalesce into one giant cluster? (This property, described here imprecisely, is defined for theoretical infinite networks as the point at which an infinite cluster first appears. Properties of finite networks can be related to the percolation threshold, too.) Initially it may seem that this level of abstraction and the simplicity of the properties that can be calculated would limit the range of applications. But, to the contrary, researchers in many fields have found ways to conceptualize discipline-specific questions in the general formulas of percolation theory, solving problems in physics, chemistry, biology, and social networks. The problems range from understanding transport of liquids in porous media (hence the name “percolation”) to the control of disease outbreaks to phase transitions in condensed matter physics.
Introduction to percolation theory is a widely recommended, much sought out-of-print introduction to the topic. As far as theoretical physics goes (theoretical anything for that matter) it assumes relatively little in terms of physics (none) and math (basic programming, differentiation and integration of functions of one variable, and elementary probability theory). It derives many of the basic known results and presents others that go beyond the level of mathematics for which the book was written. It introduces basic concepts of percolation, renormalization, and diffusion on porous medium. From this book, an interested scientist or student from any field in which stochastic processes are used to represent dynamics can obtain techniques to complement existing conceptual toolboxes, while the background obtained from reading this book will provide confidence to venture further into the technical literature. An ecologist by training, I found this book readable and informative with a rare style for the genre. As an introduction and reference for computational scientists in many fields, it is indispensable. Highly recommended.
Harte, J. 1988. Consider a spherical cow. Reprint edition. University Science Books.
Mandelbrot, B. 1982. The fractal geometry of nature. W.H. Freeman & Co.
van Fraassen, B. 1980. The scientific image. Oxford.
This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 License.
Kirsch, S. 2005. Proving grounds: Project Plowshare and the unrealized dream of nuclear earthmoving. Rutgers University Press.
In 1957 the Lawrence Radiation Laboratory in Livermore, California began an audacious experimental program to study potential peaceful applications of nuclear explosions. I reiterate, the intention was to carry out peaceful nuclear explosions. These nuclear scientists sought to exploit the energy of uncontrolled nuclear reactions in the service of society's greater good. From start to finish, the goal was constructive destruction. This sounds preposterous to modern sensibilities. How any nuclear explosions during the negotiation of the Limited Test Ban Treaty could be considered peaceful, even welcomed in some communities (the anti-NIMBY), is almost incomprehensible. This however, is the story of Proving Grounds.
Project Plowshare was the name of this experimental program, lasting more than fifteen years from 1957 to 1974 and costing the US government about 155 million $US in the currency of that time. The biggest projects contemplated by Plowshare scientists were to be demonstrations of nuclear excavation. The first was a harbor planned for an isolated corner of Alaska (Project Chariot), the second a sea level canal across the Isthmus of Panama. But neither of these projects was carried out. Indeed, in fifteen years the total earthmoving of Project Plowshare comprised a few minor explosions at the Nevada Test Site, basically leaving a few large holes (but we knew nuclear explosions could do that!), or, in one interesting case (Project Sulky), a pile of rubble. (You'll have to read the book to see why this is interesting.)
Individually, these episodes tell us a bit about the politics, economics, and social organization of science in the immediate post war era. What is more interesting about Plowshare, however, is not the idiosyncracies of the individual projects (themselves attesting to the idiosyncracy of the scientific process) but the program's eventual failure as a whole. Further, this failure was not primarily in the view of the physicists who participated in it, or in their peers at other institutions. Evidently, physicists by and large judged the models of Plowshare to be basically correct, so far as they went (the extent of nuclear fallout notwithstanding), and its applications sound. Rather, Plowshare failed because it could not command public assent, nor the assent of other non-physicist scientists, primarily ecologists and public health professionals. It is common knowledge among students of science and technology that (contrary to myth) science is not universal, that its communities skirt the borders of public and private knowledge. But here we have a science the outcome of which was wholly determined by non-experts. The bungled politics that makes up the bulk of this tale only accelerated the failure which in retrospect seems to have been Plowshare's destiny from the outset. (Surely at least, it could have been seen from the event of a college textbook, Constructive Uses of Nuclear Explosives, published prior to the existence of any real experimental program and certainly before any successful demonstrations of this new science.) Curious.
In the end, the end of Plowshare was politics. Kirsch writes that “Marcuse's fears of technocracy were quite legitimate, of course, and have in some ways been realized” (p. 208). But, I think there is no “of course” about it. It is not evident to me that Marcuse's fears were legitimate (or that they have been realized). The “of course” is to forestall any doubts we may have to the validity of this claim. But, the story of Plowshare, carefully researched and expertly told, tells us just the opposite. Technocracy is not the inevitable conclusion of technology, but just one of its possibilities. Dissent is not futile. But we, the dissenting, cannot slide into cynicism but must vigilantly call technocrats to account. Science also belongs to us.
This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 License.
In 1957 the Lawrence Radiation Laboratory in Livermore, California began an audacious experimental program to study potential peaceful applications of nuclear explosions. I reiterate, the intention was to carry out peaceful nuclear explosions. These nuclear scientists sought to exploit the energy of uncontrolled nuclear reactions in the service of society's greater good. From start to finish, the goal was constructive destruction. This sounds preposterous to modern sensibilities. How any nuclear explosions during the negotiation of the Limited Test Ban Treaty could be considered peaceful, even welcomed in some communities (the anti-NIMBY), is almost incomprehensible. This however, is the story of Proving Grounds.
Project Plowshare was the name of this experimental program, lasting more than fifteen years from 1957 to 1974 and costing the US government about 155 million $US in the currency of that time. The biggest projects contemplated by Plowshare scientists were to be demonstrations of nuclear excavation. The first was a harbor planned for an isolated corner of Alaska (Project Chariot), the second a sea level canal across the Isthmus of Panama. But neither of these projects was carried out. Indeed, in fifteen years the total earthmoving of Project Plowshare comprised a few minor explosions at the Nevada Test Site, basically leaving a few large holes (but we knew nuclear explosions could do that!), or, in one interesting case (Project Sulky), a pile of rubble. (You'll have to read the book to see why this is interesting.)
Individually, these episodes tell us a bit about the politics, economics, and social organization of science in the immediate post war era. What is more interesting about Plowshare, however, is not the idiosyncracies of the individual projects (themselves attesting to the idiosyncracy of the scientific process) but the program's eventual failure as a whole. Further, this failure was not primarily in the view of the physicists who participated in it, or in their peers at other institutions. Evidently, physicists by and large judged the models of Plowshare to be basically correct, so far as they went (the extent of nuclear fallout notwithstanding), and its applications sound. Rather, Plowshare failed because it could not command public assent, nor the assent of other non-physicist scientists, primarily ecologists and public health professionals. It is common knowledge among students of science and technology that (contrary to myth) science is not universal, that its communities skirt the borders of public and private knowledge. But here we have a science the outcome of which was wholly determined by non-experts. The bungled politics that makes up the bulk of this tale only accelerated the failure which in retrospect seems to have been Plowshare's destiny from the outset. (Surely at least, it could have been seen from the event of a college textbook, Constructive Uses of Nuclear Explosives, published prior to the existence of any real experimental program and certainly before any successful demonstrations of this new science.) Curious.
In the end, the end of Plowshare was politics. Kirsch writes that “Marcuse's fears of technocracy were quite legitimate, of course, and have in some ways been realized” (p. 208). But, I think there is no “of course” about it. It is not evident to me that Marcuse's fears were legitimate (or that they have been realized). The “of course” is to forestall any doubts we may have to the validity of this claim. But, the story of Plowshare, carefully researched and expertly told, tells us just the opposite. Technocracy is not the inevitable conclusion of technology, but just one of its possibilities. Dissent is not futile. But we, the dissenting, cannot slide into cynicism but must vigilantly call technocrats to account. Science also belongs to us.
This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 License.
Friday, August 18, 2006
Habermas, J. 2003. The Future of Human Nature. J. Habermas. Cambridge: Polity Press. 127 pp.
Recent developments in biotechnology – especially in cloning and human reproduction – have occasioned the rise of a bioethics cast in the idiom of “human dignity”. This new bioethics isn’t just about the moral gravity of technical knowledge, but also the politics of technology and social constraints on science. At its core, it is about what kind of beings we think we are, shocked as we are to learn that we have only about 30,000 genes and that a large number of them are shared with the genus Paramecium. Of course, the politics is about how much restraint should be exercised in various fields of research, and who should use what kind of muscle to leverage that restraint. As usual, there are progressives and futurists alongside more conservative pundits – writer and environmentalist Bill McKibben and doctor/philosopher Leon Kass come to mind – though there are many others who also would cease active research on controversial topics hoping for moral clarity, or at least that we will not seriously misstep, trudging as we are on ethically rocky soil.
The Future of Human Nature, by philosopher and social thinker Jürgen Habermas, is a recent and welcome contribution to this discourse. It is also the most compelling argument for the conservative position. In part, this is because Habermas takes a political angle on some of the metaphysically more windward questions. Compare with Kass’s Life, Liberty, and the Defense of Dignity. Whereas Kass’s position depends on the premise that political liberalism is antithetical to human dignity (indeed to many virtues worth preserving), and that human dignity cannot be understood apart from deeply metaphysical notions about free will and the like, much of Habermas’s argument actually relies on the suppositions of political liberalism. In three brief essays, Habermas addresses these questions:
Are there postmetaphysical answers to the question: What is the good life?
How might biotechnology influence the self-understanding of the human species?
- How should religious intuitions, beliefs and doctrine, and institutions address social policy in a secular society?
These are weighty questions. Habermas is to be commended for having addressed them, clearly and cogently, in just 127 pages.
Habermas’s first question is an old one, but not one much talked about by philosophers nowadays. Habermas observes that it is “precisely with regard to the questions that have the greatest relevance for us” that philosophers are the least useful. Seemingly, the best they can do is to tell us how we can’t answer a question, not how we can. Why is this? Is it truly the case that philosophy after Kant cannot guide self-understanding, because the philosophers themselves have lost faith in their ability to tell a coherent story? Habermas thinks not, at least not entirely, because language, even the language of analytic philosophy, is not a “kind of private property”; it is shared. For Habermas, the content of the answers we seek resides in the self-understanding of shared language-users. When the self-understanding of the language community is threatened, as Habermas perceives that it is by modern biological technologies, not only can philosophy intervene, it must.
Lurking behind this moral imperative is a thesis about the self-understanding of human beings as biologically indeterminate organisms. The second essay, comprising approximately the second third of the book, explores the political and moral consequences of cloning and other genetic technologies (e.g. preimplantation genetic diagnosis). Habermas argues seven points. First, Habermas believes that the most useful approach to a politics of biotechnology is to determine what it means to say that “the genetic foundations of our existence should not be disposed over” (p. 22). Therefore, (point number two) the comparison of reproduction technology with the abortion debate has been misleading. Third, the human genome is constitutive of our self-understanding as Homo sapiens, and therefore gene-manipulation dangerously interferes with the identity of the species. Then, fourth, a moral distinction should be made between “made” and “grown”. Thus, fifth, gene-manipulation will affect the self-understanding of the “genetically programmed person”. Sixth, the knowledge of a genetically programmed person that they are such may restrict the choice of one’s way of life and thereby disrupt the symmetrical relationship that obtains between free and equal human beings. (In particular, the parent generation will be empowered at the expense of the offspring.) Finally, certain research strategies exemplify the “dangers of liberal eugenics” and are to be strictly avoided.
Habermas’s defenses of (1), (2), and (4) cogently address some of the more important political considerations for a modern liberal society contemplating the gravity of biotechnology. Serious questions about privacy and legal standing sit at the juncture of made/grown and our previous political experience will not serve us well here. However, in Habermas’s view these considerations, which I think should be a part of the political discourse, are intertwined with (3), (5), and (6) which are (in my view) based on scientifically unjustifiable premises. I do not think, for instance, that our self-understanding as a species is much at all related to our (mostly) shared genome, which we share (mostly) with all other living organisms. Our common-sense self-understanding did not change an iota when we learned that the Human Genome Project had been 99% (or 100%) completed. I suspect it did not affect the self-understanding of the scientists involved in the project either, not even for the project leader whose own DNA was used in the project.
What has happened here is that the biological discourse has inherited concepts from a metaphysical one. These are ideas like “free will” and “determined” and “programmed”. But these concepts do not truly belong to biology; they are not a part of the workaday world of DNA extraction and sequencing. For the most part, they are not even a component of biological theorizing. However much we share Habermas’s premise that our shared self-understanding is based on our shared language, we also must admit that our scientific language and its concepts are not broadly shared (our society is by and large scientifically unschooled), that our scientific concepts and the language used to communicative them are continually changing, and that even if science was fixed and scientific knowledge was universal the majority of us would not take our existential cues from our scientific dogmas.
The final essay in this volume, “Faith and Knowledge”, is predicated upon the supposition that religion does in fact have something meaningful to say in a secular world. For Habermas, the 11 September 2001 terrorist attack in New York is sufficient evidence of this. But, when should society heed religion? When should religion defer? Habermas consistently cautions against univocal moral pronouncements, never affirming the complete moral authority of religion in either its supernatural or secular varieties. There are some who believe that science is a special language game particularly useful for technological applications, but that it is not pertinent to our deepest metaphysical and ethical intuitions. Habermas objects. Modern science, as alien and exotic as it is, does inform our common sense self-identity.
I differ. Truly, our world was changed by Copernicus and Darwin. But, this in no way justifies the presumptions of secularism. Modern political liberalism does not embody a single scientistic public; democracy is “many-voiced”. The upshot of this is that when it comes to bioethics—Habermas uses genetic engineering as an example—one need not assent to (possibly dubious) theological claims to find reason in the voice of religion. “So God created man in his own image, in the image of God created he him” (Gen 1:27). From this, Habermas finds insight for even the irreligious to appreciate: God the creator is not God the technician. “One need not believe in theological premises in order to understand what follows from this… [if] the place of God be taken by a peer… Would not the first human being to determine, at his own discretion, the natural essence of another human being at the same time destroy the equal freedoms that exist among persons of equal birth in order to ensure their difference?”
As should be clear from my comments, I think that Habermas has failed to identify the solution. I do not think that the genome is the essence of the man. Importantly, however, he has correctly identified the question. What does it mean to say that the genetic foundations of our existence should not be disposed over?
This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 License.
Subscribe to:
Posts (Atom)
