sTo the Editor:
Some comments on Rudolf Klein’s article on The Limits to Growth [“Growth and Its Enemies,” June]. Obviously, the fact that predictions of this general cast have been made and discredited before constitutes only feeble evidence against the validity of this one, as even a casual familiarity with the history of science makes evident. Mr. Klein is quite right that what matters is the validity of the reasoning that leads to such conclusions. However, his attempt to analyze the methods used in The Limits to Growth reveals that he doesn’t understand the workings of such models well enough to perform any competent critique of their assumptions: the core of his review is, quite simply, nonsense. No model of this sort assumes that trends will remain constant in such a way that by the year 2000 everyone will be either an ecologist, in a hospital, a pot smoker, or belong to any other such patently non-self-sufficient category: one of the features of these models is precisely that their structure forestalls such silly predictions. The models do assume structural relationships that are invariant over time, and they do use trend-fitting, but from this Mr. Klein’s summary of the method does not follow. He clearly does not understand the significance of nonlinearities in feedback loops.
Further evidence of hasty, careless reading is Mr. Klein’s statement that The Limits to Growth ignores the proportionate growth of health and education expenditures in developed nations. Although the models for possible alternative policies are regrettably crude, it is wrong to suggest that the authors ignore the fact that there may be a price to pay for following their recommendations.
Basically, what needs to be done is to identify the real constraints and the real possibilities, not to counter oversimplifications with worse ones. . . .
Virgil E. Vickers
To the Editor:
Rudolf Klein compares the authors of The Limits to Growth with the soothsayers who consulted the entrails of a slaughtered ox. This is grossly unfair: the slaughtered ox was, presumably, accessible for inspection to the Roman public; but the authors of The Limits to Growth have yet to publish their computer programs.
University of Colorado
To the Editor:
It is very popular now to blame technology and technological advance for the problems of ecology and pollution. Some applied sciences such as physics and engineering are also under attack. The next logical step is to broaden the attack to include the very foundations of the physical sciences—mathematics and the scientific method. In “Growth and Its Enemies,” Rudolf Klein launches an attack on the tools of science—computers, systems analysis, etc.—using as his scapegoat The Limits to Growth, one of the most promising books of our time, which treats ecological problems quantitatively, employing the methods of science and modern technology.
Mr. Klein argues less with the results of the book than with the methods by which the results were obtained. For him the unforgivable sin committed by the authors of The Limits to Growth is that they use computers. . . . In his concluding paragraph he sums up his fears: “It is desperately important that complex problems should not be reduced to the simple symmetry required of systems analysis. This is, if anything, the final surrender to technology—to adjust our own vision of problems to the technical necessities of feeding them into a computer.”
As one who has been a systems analyst, I can assure Mr. Klein that systems analysis does not require that complex problems be reduced to simple symmetry. On the contrary, with the advent of modern computers, there is a tendency not to reduce complex problems to any simpler form, but to retain much more of the original complexity. It is true that the use of computers is a technical convenience, often indeed a necessity, but it permits a wider vision of the problems and is in no sense a surrender to technology. Instead, this technology has the potential to liberate us from ecological perils. . . .
Mr. Klein should not fear the computer, he should fear the uses to which it is sometimes put by those in power. He can best conquer his fears by offering constructive ideas in his own field of political science. It is when he ventures out of his field that he goes astray. . . .
Frank L. Paulsen
Rudolf Klein writes:
It is tempting to answer Virgil E. Vickers by saying that if I do not understand “nonlinearities in feedback loops,” then neither does he understand straightforward prose as distinct from jargon. However, I think that both his angry reaction and that of Frank L. Paulsen to my article point to a real difficulty in dealing with a book like The Limits to Growth. The book is addressing two quite different audiences: those who, like myself, are primarily interested in its social, political, and cultural assumptions and those who, like my critics here, are concerned chiefly with the technical processes of systems analysis and computer programming. The two audiences often do not speak the same language, with the result that there is a dialogue of the deaf.
So let me make one more attempt to spell out the theme of my article. First of all, I did realize that the projections used in The Limits to Growth were far more sophisticated than my (I had hoped faintly-funny but, as it seems, taken-too-seriously) examples about pot smoking, etc. But Mr. Vickers misses the real point here. I was not arguing that the methods were inadequate, since I am in no way competent to judge them. I was, however, questioning the way in which The Limits to Growth was applying its perhaps brilliant methods to draw what seem to me to be misleadingly firm conclusions about the future.
Indeed Mr. Vickers himself unerringly puts his finger on the basic flaw in the whole exercise. The models, he writes, “assume structural relationships that are invariant over time.” This seems to be a very big, very dangerous, and very dubious assumption to make; in my view, therefore, no amount of technical skill in computer programming can make the whole exercise other than an irrational form of predestinarianism. One example illustrates this point. A few years ago economists were arguing that the “structural relationships” of the British economy were such that excessive wage claims and inflation could only be prevented by increasing the unemployment rate. The unemployment rate has now gone up well beyond the recommended figure, but the rate of inflation has gone up faster still. It just so happens that the “structural relationships” have changed and thus made the model out of date. There could be no clearer indication of the perils of using historical patterns to construct models for predicting the future.
Again, Mr. Vickers asserts that the authors of The Limits to Growth do take into account “the fact that there may be a price to pay for following their recommendations.” If by that he means that the book contains the odd sentence or two recognizing that a zero growth rate may bring problems of its own, then he is right and I am wrong. If, however, one thinks that both sides of the equation—the debits of growth and the problems of non-growth—should be examined in equal depth and detail before even attempting to make any recommendations, then he is wrong and I was right. It is because The Limits to Growth is so one-sided in its approach—its intellectual naiveté disguised under a show of technical sophistication—that I consider it a rather frivolous book.
Lastly, let me assure Frank L. Paulsen that I am not anti-computer: one of my best friends, in my current researches, is a computer which faithfully churns out a great deal of data. My main concern is that technical wizardry should remain the servant of policy-making and not become a substitute.