Alexander DeVolpi verse Amory Lovins: Part I

I continue to repost old essays. This two part essay reviews Physicist, Nuclear Weapons design expert, and Nuclear Arms Control expert Alex DeVolpi's assessment of Amory Lovins competence as an expert on nuclear proliferation and on nuclear powerreactor design.

Part I: DeVolpi on Expert Knowledge, Scientific Knowledge and Ethics
Amory Lovins has Chutzpah. Lovins, a Harvard & Oxford physics drop out, believed that he could argue with Alvin Weinberg about reactor technology. Not long ago he showed up at Argonne National Laboratory where he attempted to argue with Alexander DeVolpi about nuclear proliferation. This was a big mistake, first because DeVolpi ought to be regarded as one of the world's leading authority on nuclear proliferation, and further DeVolpi does not suffer fools gladly. And Lovins is nothing if not a fool, and only a fool would have argued with DeVolpi at Argonne about nuclear proliferation. For Amory Lovins to argue with DeVolpi on nuclear proliferation is more than a little like Alfred E. Newman attempting to argue with Albert Einstein on the theory of relativity.

Nothing illustrates the confusion of the Era of Confusion better than the fact that Amory Lovins is imagined to be an expert on energy by Al Gore and hundreds of of other political and business leaders in contemporary American Society. Lovins tends to overwhelm his auditors with a dense presentation of supposed facts and references, that cannot be easily or quickly or easily deconstructed. Thus am authorizing link may be in fact be to a statement that was published in an obscure South African humor magazine, and which is locked beyond a firewall, with the payment of a fee required for admission. Another Lovins trick is to reference a statement which he himself made 30 years ago, without any further support. Unless the auditor has a copy of the collected works of Amory Lovins, it is impossible to determine if the 30 year ago statement had just as weak evidence as the statement presented today.

Dr. DeVolpe can recognize shabby tricks for what they are, but most people, including Al Gore, doubt their own intelligence when confronted with a MacArthur Genius. To such people I can only recommend a reading of the story The Emperor's New Clothes together and Immanuel Kant's essay, What is Enlightenment. But such remedies may come with distress, because they invariably require a person to look at how he or she thinks and feels, and then to think about the answers to the questions those thoughts and feelings raise.

Dr. DeVolpe has written excellent and instructive Google Knols that should be read by anyone who wishes to think about let alone openly discuss nuclear issues like proliferation and nuclear safety. Beyond simply talking about proliferation issues, Dr. DeVolpi looks at a question that clearly strays into an area that might be described as practical epistomology. It particular how can we know if someone who claims expertise actually possess authority. Dr. DiVolpe shreds claims to authority on such matters, with a particular zest"

For progress in non-proliferation, we need be saved from the assumed or accorded authoritarianism of well-intentioned professors, especially from the East Coast, who have titles mistaken as credentials. Frank von Hippel of Princeton comes to mind. Notwithstanding good intentions, pleasant personality, teaching experience, and published papers — these do not constitute hands-on field or laboratory experience. Nor should one count time spent in Washington corridors, offices, and conference rooms.

I hold Frank partially responsible for the decade-long hiatus in reaching agreement with Korea on nuclear demilitarization, for decades of lack of progress in conversion of the Siberian plutonium reactors, for stalling growth of nuclear power in the United States, for misrepresenting the weaponizability of reactor-grade plutonium, and for sustaining radiophobia.

On the latter point, over two decades after the Chernobyl accident, Frank is yet to acknowledge in print that he was utterly wrong in projecting or implying a huge number of fatalities due to the accident. He and others cling to unvalidated beliefs regarding the effects of low levels of radiation . . .

DeVolpe scorns the authority of the under experienced.

were it not for the professors of the 1930s and 1940s who gained hands-on laboratory and field experience, we would not have succeeded in the timely development of nuclear weapons and nuclear reactors. With the demise of Hans Bethe and Pief Panofsky, a good example remaining is Dick Garwin (aside from some uncharacteristic overreaching he has done with regard to Chernobyl cancer projections).

DeVolpe offers us nothing less than a phenomenology of experties.

Expertise isn’t fungible; it can’t be bought or transferred; it’s accumulated from sometimes-tedious, but aggregate years of hands-on experience. Nor is anyone’s accumulated expertise unique or exclusive; some individuals have subsets of very relevant knowledge or experience that include skills and understanding of energy released in fission or fusion, or of policies and implications regarding nuclear weapons. These fields of knowledge overlap and supplement each other; there are no islands of expertise.

Incidentally, professional conferences and lectures are a common adjunct for keeping up to date, but they do not contribute directly to hands-on experience in the functioning of complex equipment. The same can be said about presentations, lectures, and facility visits; these are an integral aspect of technical development, but they are not substitutes for actual laboratory or field development, construction, experimentation, and analysis.

To a certain extent DeVolpi is an advocate of the tacit knowledge tradition without formally acknowledging Michael Polany's writings on the concept. In addition to focusing on the practical dimension of expert knowing, Dr. DeVolpi also focuses on ethical issues implicit in expert knowledge claims, authorized by a presumed claim to scientific authority. In particular Dr. De Volpi's, focuses on be called a phenomenology of scientific authority. That is he offers a description of the essential characteristics of a knowledge claim that is advanced with scientific authority. That description did not originate with Dr. DeVolpi, rather he found it in the U.S. Supreme Court's 1993 statement of standards for testimony regarding areas of science that required an explicit estimate of probabilistic error. That statement, presented in connection with the courts ruling on the Daubert v. Merrell Dow Pharmaceuticals case, set forth a four part qualifying standard:

peer review,
replicability,
documentation,
and stated rates of error

DeVolpi notes:

Judiciaries have retrospectively encountered deficiencies in ad hoc scientific/technical testimony and in forensic evidence that did not fully comply with a standardized methodology. Individuals have been wrongfully convicted of crimes; cancer and other illnesses have been incorrectly attributed; and epidemiological data has sometimes been misrepresented.

This is an ethical issue in DeVolpi's view. Making claims to scientific authority which do not conform to the Supreme Court's Daubert standard are not just epistomologically flawed, they are also ethically flawed, and such scientifically wrong statements have lead to moral wrongs.

Individuals have been wrongfully convicted of crimes; cancer and other illnesses have been incorrectly attributed; and epidemiological data has sometimes been misrepresented.

While DeVolpi focuses primarily on a case study of the over statement of anticipated public health consequences for the Chernobyl and TMI reactor incidents. He observes:

The scientific, technical and journalistic professions, though not alone, must share significant responsibility for premature and exaggerated predictions that have not materialized nor been rectified.

This professional ethical lapse have had serious public consequences:

Unsubstantiated characterizations contribute to public confusion, rather than clarification. Inordinate risk estimates have lead to the expenditure of tens of billions of dollars to protect against dangers whose existence is highly questionable.

(In Part II, I will address the substance of Dr. DeVolpi's "Bill of Particulars" against Amory Lovins.)

Alexander DeVolpi versus Amory Lovins: Part II

This section reviews DeVolpi's assessment of Amory Lovins' reputation as an expert on nuclear proliferation and nuclear power..

Part II: DeVolpi on Amory Lovins Expert Knowledge on Nuclear Proliferation and Nuclear Power

Alexander DeVolpi has offered a serious critique of Amory Lovins that I believe accurately raises questions about Mr. Lovins' authority. DeVolpi, points out that

Because Lovins renders no substantive academic or acquired nuclear credentials, the analyses he presents ought to be held to a strict standard of scientific credibility, such as that described by the Daubert U.S. Supreme Court decision. . . . This is in lieu of granting him interim benefit of doubt, a courtesy often extended to individuals who have an established scientific reputation . . . In other words, I would advise treating Lovins’ renderings on nuclear issues with healthy, but not dismissive skepticism. His presentation and publications should be judged by standard scientific criteria, no more, no less.

Next Dr. DeVolpi points to Lovins' scanty educational credentials and his lack of the sort of experience that would qualify him as an expert on nuclear matters.

Although Lovins seems to have completed some courses in experimental physics at Oxford University in England, he lacks any laboratory experience in nuclear physics or engineering. His vetted degree credentials are vague enough to induce caution, caveat emptor. Such a shortcoming has not prevented him from writing numerous articles, giving many briefings, and speaking frequently about nuclear technical policy. . . . Lovins has been a widely praised proponent of the so-called “soft-energy path,” as well has having been an habitual and readily available critic of nuclear energy.

. . . expertise alleged should not be considered credible simply because of personal experience, widely publicized image, or self-declared credibility — which can be crafted as concatenating substitutes for substantive technical analysis and publication. The individual being challenged should follow the same established guidelines for scientific analysis and peer-reviewed publication as the rest of us have during our professional careers.

Having offered this preamble to the question of Lovins' authority, DeVolpi proceeded to examine Lovins' method of presentation of "what appeared to be an informative but complex analysis . . .". DeVolpi thus offers a phenological approach to Lovins presentation by placing it into brackets, which examines what it appears to be at first in light of the accepted standards of scientific evidence which DeVolpi has suggested we apply to expert testimony. DeVolpi noted Lovins use of "extremely busy tables and graphs" which he found "difficult to sort through", and then suggested

his extrapolation from laboratory model to production product is unrealistic, being deficient in practical marketplace engineering. Faulty reasoning and extrapolation often reflect a lack of hands-on construction experience. Lovins did not put into evidence anything he actually built or was responsible for constructing, other than a viewgraph of a fancy banana greenhouse situated on his Aspen, Colorado, property.

The last comment is downright funny, because Lovins allegedly unheated Aspen greenhouse in which he grows bananas, is very much a part of the Lovins mystique. The greenhouse itself plays on role in establishing Lovins expertise, or the truthfulness of the case he presents.

DeVolpe noted that Lovins' presentation

more of an evangelical tirade against nuclear power, rather than a systematic case for balanced and alternative energy sources. A key indicator of stridency is the absence of explicit statistical characterization; Lovins presents almost everything in terms of absolutes, without conceding a range of uncertainty.

DeVolpi found an absence of a probabilistic perspective not only in Lovins' oral presentation but in his papers

Take a look at his papers and try to find any treatment or awareness of statistical uncertainty. What’s notably odd is that doubt/uncertainty is part of the natural order of things; to avoid recognizing it, especially in a paper about technical issues, is quite unnatural and unusual, and more indicative of proselytization for a cause.

Thus Amory Lovuns' presentations are

more of an evangelical tirade against nuclear power, rather than a systematic case for balanced and alternative energy sources.

DeVolpi subjects Lovins work to two related tests, those of "smell" and "ripeness". The smell test is designed to determine "legitimacy" or "authenticity". while the ripeness looks at "maturity", or "development".

DeVolpi notes that during a visit to Argonne National Laboratory 30 years ago Lovins called for a shutdown of all nuclear power plants. At that time almost no electricity in Illinois was produced by reactors, at the beginning of 2009 that figure had increased to between 70% and 80& of Illinois electricity being generated by reactors. DeVolpi observed

here’s a situation that has ripened enough for comparing actual outcomes with his original counsel. . . . If Lovins had his way 30 years ago, I would be paying . . . (for) coal-produced electric power. On the basis of cost, or feasibility, or environmental benefit, electric-power utilities and the state regulators would have been ill-advised if they adopted his anti-nuclear advice (at least in Illinois).

Nuclear power is not only commercially competitive, but extremely safe (no coal miners dying), no air pollution at all, no greenhouse gas emissions (such as carbon-dioxide). Nuclear-plant lifetime is being doubled from 30 to 60 years (which utilities, investors, and ratepayers appreciate). If Lovins had his way 30 years ago, considerably more particulates and gases would have been vented to the local and regional atmosphere from coal-fired plants (aside from the greenhouse gases emitted).

Moreover, if Lovins had his way, we would not have conserved the electricity-equivalent in domestic coal, imported and domestic oil, and domestic and imported natural-gas resources and reserves that we have for 30 years. A typical nuclear power plant each year avoids consumption of 3.4 million short tons of coal, or 65.8 billion cubic feet of natural gas, or 14 billion barrels of oil. (The United States has ample uranium resources.) So Lovins was wrong in implying that nuclear had no overriding societal or environmental benefits.

DeVolpi focuses on Lovins claims about nuclear costs:

Lovins displayed complex viewgraphs that, he purports, show that nuclear is the costliest of “low-or-no-carbon resources.” Yet, in the last 30 years, nuclear has displaced half the fossil-fuel combustion in Illinois while still being competitive. Inasmuch as nuclear-power plants emit no byproduct carbon-dioxide to the atmosphere, surely his claim that it is the costliest of low-carbon-emission sources fails the smell test.

Most of Lovins’ pricing and cost/benefit comparisons are based on “new delivered electricity” which frames the cost of U.S. domestic nuclear construction in the least favorable light.

He declares nuclear power an economic failure. Can someone explain that to my bank account which has benefited from compounding competitive electric power savings for the past 30 years? His rimy claim certainly fails the ripeness test.

DeVolpi then challenges Lovins' claims about nuclear reliability.

On the issue of electrical-grid reliability, Lovins asserts that there is no such thing as a “outage-free” source of electrical power. He must think that nuclear power runs by government fiat. Nuclear is a fixture on the grid because it is more economical to operate as base-load supply, while sources less reliable, intermittent, and more costly (such as wind, solar, and gas) provide supplementary power. During the past 30 years in Illinois, I don’t recall having the electricity supply and cost problems that California has had after it prohibited nuclear-power plants from being built within its borders. By the way, average U.S. nuclear capacity factor was about 92% in 2007. That’s excellent. Lovins pitiful effort to undermine the reliability of nuclear power egregiously fails the smell test.

DeVolpi examined Lovins account of nuclear power and finds that Lovins

chronically demonizes it on the grounds of the proliferation risk

opposes subsidies for nuclear power but favors them for renewables

Calls nuclear power a failure despite the reliable production of power at competitive rates

Opposes the construction of new nukes in the United States because of proliferation risks, even though new nukes are being built in other countries.

Argues that nuclear power is anti-democratic

called attention to Lovins 1980 statement in Foreign Policy that

the global nuclear power enterprise is rapidly disappearing

for nuclear power is ... the main driving force behind proliferation...

(nuclear power) retards oil displacement by the faster, cheaper and more attractive means which new developments in energy policy now make available to all countries...

DeVolpi next points to a 1980 article which the science journal Nature published. In this article Lovins, who was after all an undergraduate physics drop-out from Harvard (twice) and Oxford posed as an expert on plutonium weapons. Lovins concluded in the Nature article that

...It is therefore incorrect to state categorically that bombs made from reactor-grade or deliberately ‘denatured’ Pu are less effective, less powerful, or less reliable than those made from weapons-grade Pu.

DeVolpi was expert enough in nuclear weapons design to qualify as an expert witness on weapons design in the famous Progressive case and who had stood up to an Energy Department's attempt to intimidate him into silence about his "politically incorrect" views on "reactor grade plutonium" and nuclear proliferation gives short shift to Lovins "expert" information. De Volpi responded to Lovins

While Lovins convinced the editors and reviewers of Nature that a neophyte had figured out nuclear weaponry enough to become an publishable expert, his inverted conclusion is not supported by theoretical or anecdotal evidence . . .

Lovins further concluded

The foregoing argument also implies that power reactors are not an implausible but are rather potentially a peculiarly convenient type of large-scale military Pu production reactor....

De Volpi remarked

Coming from a neophyte who might never have seen the inside of any of those reactors, it reflects a hoary belief system that was as untenable then as now. Just show me a civilian power reactor that has been used to make military plutonium. This published proposition of his fails the ripeness test in 2009, just as it failed the smell test back in 1980.

Lovins added

In short, the somewhat greater technical difficulty of using power-reactor Pu for effective military bombs — assuming the reactor is actually operated at high fuel burn-up — may be more than counterbalanced by the greater political and economic ease of obtaining that Pu. It should not be lightly disdained in favour of purer material from dedicated facilities.

DeVolpi scoffed

This pitiful conclusion is the foundation of Lovins’ nuclear-proliferation belief system. It too, long ago, failed both the smell and ripeness tests. Incidentally, note the absence of measures of incertitude in this so-called technical paper.

DeVolpi asks

The Nuclear Illusion or the Nuclear Illusionist?

as he "fisks" Lovins 2008 paper “The Nuclear Illusion”. Lovins once again is allowed to go beyond a credible interpretation of evidence, ignoring the constraints of uncertainty.

Dr. DeVolpi in his Knols, exhibits outstanding and sophisticated critical skills. The DeVolpi Knols deserve far wider recognition than they have received to date. They are significant contributions to public discussion on many issues related to future use of nuclear energy in our society. In addition, the DeVolpi Knols are important examples of the sort of critical thinking that ought to be encouraged in the class room. DeVolpi's Knol on Amory Lovins ought to be read by any journalist, scholar, or politician who is considering using Lovins as a source.