Sorting out Junk Science > Health Issues

By Gilbert Ross, M.D., Jack Raso, Sidney Shindell, David W. Cragin
Posted: Thursday, March 1, 2001

ARTICLES
Publication Date: March 1, 2001

Junk science is not a science centered on usable refuse. Neither it is garbology. Is it bad science? Can science be bad? Can it be junk? Should we trash "junk science"?

Whence "Junk Science"?

Reportedly, public use of the expression "junk science" began in a 1989 lecture by Michigan lawyer John Gleeson. But according to ACSH scientific advisor Donald G. Cochran, Ph.D., Professor Emeritus of the Virginia Polytechnic Institute and State University, the expression has been in use in his field—entomology, the scientific study of insects—at least since the early 1970s. In the 1998 Stanford Technology Law Review article "Trashing 'Junk Science,'" Gary Edmond, LL.B., and David Mercer, Ph.D., stated that use of the expression "junk science" has "become ubiquitous in discussions involving law and science."

In 1999, economist Edward S. Herman, Ph.D., Professor Emeritus of the University of Pennsylvania, surveyed 258 articles that included the expression "junk science" and had been published in mainstream newspapers within the previous three years. He found that in 62 percent of these pieces the expression referred to science-related arguments used by environmentalist and other critics of corporations, or by personal-injury lawyers engaged in suing corporations.

But the expression "junk science," however widespread its referent may be, has no generally established definition. Moreover, there is a borderland between junk science and science. Whatever characterizes junk science cannot be itemized definitively. In "Trashing 'Junk Science,'" Edmond and Mercer stated: ". . . [Historically] it is common for scientists to have strong commitments to their views during the early phases of new research. Evaluated in the context of their own time and place, the theories of Mendel, Kepler, Darwin, Newton, Galileo, and Copernicus relied upon leaps of faith and observations at the threshold of theoretical plausibility, together with the deliberate omission of counter arguments."

Science, Junk Science, and Pseudoscience

Peter William Huber popularized the expression "junk science" with Galileo's Revenge: Junk Science in the Courtroom, first published in 1991. In the index of its paperback edition, the expression "junk science" is followed by "See Science, junk." This suggests that junk science is a kind of science.

Dictionary definitions for the word "science" are disparate. But in biomedicine, public health, biology, and the physical sciences (e.g., chemistry and physics), "science" primarily denotes a continuous process whose basic purposes are to make phenomena recognizable and to predict outcomes, and whose fundamental activities comprise:

Science can be bad....[a]nd instances of science can be junk....[b]ut the activities the expression 'junk science' typically denotes are not instances of science.

observing and describing phenomena and developing general conclusions about them;
integrating new data with organized observations that have been confirmed;
formulating testable hypotheses based on the results of such integration;
testing such hypotheses under controlled, repeatable conditions;
observing the results of such testing, recording them unambiguously, and interpreting them clearly; and
actively seeking criticism from fellow participants in the endeavor called "science."

Science can be bad—i.e., immoral,^* disagreeable, injurious, severe (rigorous), and/or intense. And instances of science can be junk—i.e., shoddy, worthless, or only superficially appealing or useful. But the activities the expression "junk science" typically denotes are not instances of science.

If one uses the word "science" in the sense of "any activity that ostensibly requires study and method," such systems as anthroposophical medicine, astrologic medicine, classical Ayurveda, and Traditional Chinese acupuncture are reasonably describable as junk sciences—activities that seem scholarly and have appeal and/or superficial utility, but which lack efficiency, reliability, and/or soundness.

The "junk science" definition above, however, is too limited relative to usage of the expression. Many expressions consisting of more than one word have meanings very different from, or more limited than, their additive meanings. (Such expressions range from colloquialisms and everyday words—e.g., "cold turkey," "hot dog," "dead duck," "sitting duck," "eager beaver," "hog heaven," and "top banana"—to jargonistic expressions, such as "age of onset," "drug delivery," "expert testimony," 'incubation period," "managed care," "prospective study," and "relative risk.")

In Galileo's Revenge, Huber (notwithstanding the potentially misleading cross-reference mentioned above) describes junk science not as a variety, variation, or perversion of science, but as the opposite of science:

Junk science is the mirror image of real science, with much of the same form but none of the same substance. There is the astrono-mer, on the one hand, and the astrologist, on the other. The chemist is paired with the alchemist, the pharmacologist with the homeopathist. Take the serious sciences of allergy and immunology, brush away the detail and rigor, and you have the junk science of clinical ecology. The . . . physical therapist [is shadowed by] the chiropractor, the mathematician by the numerologist and the cabalist.

According to , "'Junk science' is faulty scientific data and analysis used to further a special agenda." If one accepts this definition, however, one must acknowledge that junk science is in no way science as science is conventionally defined. On the basis of the statement, "junk science" is reasonably definable mainly as "misinterpretation of scientific data, or faulty organized data from science, especially such misinterpretation or such faulty data that is used to promote an ideology and/or to increase finances." But junk science transcends faulty organized data from science and misinterpretation of scientific data.

In Voodoo Science: The Road from Foolishness to Fraud (Oxford University Press, 2000), Robert L. Park, Ph.D.—Professor of Physics at the University of Maryland and Director of the American Physical Society's Public Information Office, in Washington, DC—describes junk science as the crafting of "arguments deliberately intended to befuddle jurists or lawmakers with little or no scientific background." He describes pseudoscience as spiritual, superstitious, and other baseless beliefs and practices "dressed in the language and symbols of science."

Similarly, we define "pseudoscience" as "any activity, practice, system, methodology, or theory that simulates science, or that is described as science, but lacks a scientific basis." Pseudoscience encompasses:

...[J]unk science transcends faulty organized data from science and misinterpretation of scientific data.

endorsing a scientifically disproved theory and applying it in medicine or science,
rejecting standard scientific procedures,
citing myths as scientific evidence,
selecting and organizing data so that the result conforms to a particular bias, and
handling the scientific literature as if it comprised all statements by all persons who have ever been scientists.

In our opinion, using the expression "junk science" tends to disparage science, because it suggests, incorrectly, that junk science is a form of what most scientists refer to as science. We submit that the word "pseudoscience," because it more clearly denotes a kind of non-science, is a better descriptor for what "junk science" primarily signifies (i.e., any mode, instance, or product of pseudoscience). The two most characteristic sorts of junk science may be termed "ideologic pseudoscience" (junk science whose raison d'ętre is allegiance to a system of social, political, sociopolitical, socioeconomic, and/or religious tenets) and "vested-interest pseudoscience" (junk science inspired by self-interest).

....[T]he word 'pseudoscience'... is a better descriptor for what 'junk science' primarily signifies....

Junk Science Ploys in Public Health

Stated below are the gists of five common ploys of junk science in public health.

Noting that a chemical has not proved non-carcinogenic
It is impossible to determine that a particular chemical never causes cancer or never contributes to its development. Describing a chemical as a possible carcinogen on the basis of its not having proved otherwise tends to obscure health issues. In public opinion surveys in the U.S., cancer typically rates as the primary health concern. Thus, declaring that a chemical may be carcinogenic is an effective scare tactic.
Citing all possible adverse health effects of exposure to a chemical
Toxicologists typically identify only likely effects of exposure to any particular chemical. Devotees of junk science, however, tend to attribute numerous effects to such exposure, even effects extremely unlikely except perhaps at very extraordinary degrees of exposure to the chemical or in very unusual circumstances.
Describing particular insecticides as neurotoxic

In general, laypersons don't know that the relative safety of most modern insecticides rests on differences between human and insect nervous systems. Deltamethrin, for example, is 10,000 times more toxic to insects than it is to mammals. Deltamethrin is among several manmade compounds that are similar to natural, camphorlike plant substances called "pyrethrins." Human intake of such manmade compounds, called "pyrethroids," is so safe that in cases of large accidental pyrethroid intake, the low-toxicity solvent for the pyrethroid is often more dangerous than the pyrethroid itself.
Implying that an association, or correlation, is causal

In epidemiology, the words "association" and "correlation" refer to a relationship between two conditions or states such that they are likely to coexist. Such an association is not necessarily casual or statistically significant. Envi- ronmentalist groups often launch scare stories by pointing up an association that has not proved causal between a particular chemical and a particular health problem—and omitting that the association may not be causal or even noteworthy.
Implying that "environmental factors" and "environmental pollution" are synonymous
In epidemiology, an environmental factor is any nongenetic factor. Such factors include smoking, excessive exposure to sunlight, excessive alcohol intake, dietary factors, and viruses. According to health authorities, (a) these environmental factors are together responsible for about 80 percent of cancer cases, and (b) another environmental factor—environmental pollution—is responsible for less than two percent of cancer cases.

How Junk Science Gets Over

ACSH advisor Dr. Cochran states:

Junk science works best in areas that are complex. . . . with topics that are beyond the knowledge and experience of ordinary people. Frequently, practitioners of junk science base their claims on facts and figures that require specialized training and experience to interpret correctly. In these cases it is easy for such individuals or organized groups to claim that their position is correct and that those who oppose them are either wrong or have a different perspective on the data. To the ordinary citizen this sounds like two scientists or two groups of scientists having a disagreement about the facts, when in reality it is an overt attempt on the part of those practicing junk science to deceive the public.

A characteristic of the presentation of junk science is that it is usually . . . in the form of a press conference [notwithstanding lack of quality peer review]. . . . [But] for scientific data to be accepted by the scientific community, it must first be reviewed by peers. This is necessary to establish that the study is not flawed by some procedural error and that the interpretation of the results is consistent with what the data actually show. By holding a press conference to present information, this entire procedure is circumvented. This is a huge advantage because it is unlikely that there will be people in the audience with the expertise to refute the claims being made. Even if such experts are present, a press conference is not an appropriate arena in which to carry on a serious debate. The result is that the junk science gets wide coverage in the press. The scientific community is left with the task of trying to show how and why the press-conference "information" is incorrect, but with reduced hope of getting the wide coverage afforded by a press conference.

One of the areas in which junk science works best is issues related to safety. Whether it be the safety of food or pesticides or automobiles, it doesn't really matter. It is easy to make exorbitant claims that some product or practice is unsafe. The use of pesticides is a good example . . . . Before any pesticide reaches the commercial market it must be extensively tested to determine, among other things, any effects it may have on a variety of organisms, whether it causes cancer or birth defects, and how it can best be used in order to minimize any possible harmful effects on the environment. In the U.S., it must receive [approval] from the Environ-mental Protection Agency prior to distribution or sale. If a pesticide fails to meet EPA standards in these tests, it will not be registered. Yet there are individuals and organized groups who claim that the use of many approved pesticides will cause X number of extra deaths per million people even when used according to the manufacturer's directions. Such claims are difficult to refute, because no responsible scientist can say that a given chemical will never cause harm, even though the probability of its doing so is remote. The public may be left with the impression that the claim is correct because it cannot be categorically denied.

The best way to combat junk science is for responsible scientists to speak out whenever they find it in any field with which they are familiar. In this way, the public will at least be made aware that there is a question about the validity of the points at issue. A continuing effort is also needed to raise the level of understanding among the public about how science works. If this can be achieved, the appeal of junk science will be greatly reduced.

Whither Junk Science?

ACSH advisor Kathryn E. Kelly, Dr.P.H., a toxicologist, states:

In my line of work, junk science is most often seen when a regulatory agency, particularly one with oversight in scientific arenas, puts political objectives ahead of scientific integrity. It manifests itself when the agency establishes a political position and then selectively seeks data that support and further that position, even to the point of actively excluding any results which are inconsistent with that position and routinely ignoring the advice of its Science Advisory Board. This active rejection of basic principles of weight of evidence cannot survive in law, nor can it survive forever in science. Ultimately, ignoring scientific reality is not sustainable, and when the consequences of ignoring science become more evident, the politics will change.

Jack Raso, M.S., R.D., is ACSH's Director of Publications.

Former ACSH board chairman Sidney Shindell, M.D., LL.B., is Professor Emeritus, Department of Preventive Medicine, Medical College of Wisconsin.

Board-certified in toxicology, David W. Cragin, Ph.D., is a Councilor for the Philadelphia Association for Critical Thinking (www.phact.org), a science advisor for secondary school districts in the Philadelphia area, Risk Assessment Manager for ATOFINA Chemicals, Inc., and an Adjunct Associate Professor in Pharmacology/Toxicology at the University of the Sciences, in Philadelphia.

Gilbert L. Ross, M.D., is ACSH's medical director.

peer review: Evaluation of the work of at least one professional or technician by board members who are, or by a committee whose members are, similar in authority, in professional or technical caliber, and/or in occupation to the author(s) of the work.

double blind: 1. A procedure whose purpose is to prevent statistical errors in clinical experiments by preventing awareness among experimenters, clinical staffers, and subjects of the identity of the treatment any particular subject undergoes until completion of the study. 2. A double-blind study. 3. Any circumstance of trying to keep specific information from two different groups of persons so as to prevent that information's affecting a process (e.g., a measurement).

weight of evidence: Evidence (especially on an issue of fact) that, in terms of cogency and relevance, is better than evidence contrary to it.

^* for example, by gratuitously causing the deaths of study participants whose participation has been coerced

Signs of Junk Science

According to Steven P. Novella, M.D., Assistant Professor of Neurology at the Yale University School of Medicine, "junk science" and "pseudoscience" refer to endeavors that do not conform even marginally to scientific standards. Although science and junk science cannot be differentiated definitively, there are acts, inadequacies, aims, and attitudes that are contrary to modern science and are thus signs of junk science. Some of these signs are stated below.

disparagement or condemning of the clinical testing technique called the "double blind"
the passing off of non-data as data
affirmative citing of fatally flawed studies, and/or misinterpretation of results from scientific studies, to popularize a point of view and thus to pave the way for actualization of an economic, legislative, political, and/or social agenda
distortion of data, e.g., preferential and affirmative citing of experimental results irrespective of (a) the degree of accuracy of the results or (b) the degree of their consistency or inconsistency with relevant established results
absence or deficiency of peer review
presentation of correlations (associations) as causal without proof of such
presentation of anecdotes and testimonials as important evidence
approval of premises, or methods, that have been refuted scientifically or that the scientific community discredits
audacious claims, e.g., (a) unverified claims affirmative of alleged simple means of remedying complicated problems, and (b) assertions of boundless, or nearly boundless, applicability for a method or finding
supporting a claim by stating it has not been disproved
expressing a view on matter that is controversial and/or of considerable interest to a significant percentage of the public (a) through a mass medium, (b) in lay terms, and (c) in a manner likely to generate unfounded hope or fear among nonscientists
hostility toward scientific criticisms disaffirmative of particular beliefs, methods, and/or pursuits and a tendency to describe such criticisms (a) as defenses of the status quo, (b) as resulting from hostility toward innovations, or (c) as products of a full-blown conspiracy
rejecting the content of scientific criticisms disaffirmative of particular beliefs, methods, and/or pursuits (a) by condemning naturalism (the philosophical basis of science) or (b) by denying that science is a means of penetrating the criticized beliefs, methods, and/or pursuits
celebration of nonmembership in the scientific community (e.g., as a means of avoiding intellectual confinement)
creation of primary guiding principles on the basis of nothing more than a single unverified observation

—J.R.

FANSA's '10 Red Flags of Junk Science'

In 1995, the Food and Nutrition Science Alliance (FANSA)—a food-and-nutrition-centered federation of four professional societies representing 100,000 food scientists and healthcare practitioners—issued a list of ten signs of junk science:

recommendations that promise a quick fix
dire warnings of danger from a single product or regimen
claims that sound too good to be true
simplistic conclusions drawn from a complex study
recommendations based on a single study
dramatic statements that are refuted by reputable scientific organizations
lists of "good" and "bad" foods
recommendations made to help sell a product
recommendations based on studies published without peer review
recommendations from studies that ignore differences among individuals or groups

Source Notes:
Priorities Volume 12 Number 4 & Volume 13 Number 1