I am Stanislaus J. Dundon, an agricultural ethicist with training in the history and philosophy of science. I have studied and taught history and sociology of agricultural science and have taught for the last nine years agricultural biotechnology ethics and public policy at U.C Davis. I have specialized in curriculum development in agriculture and the human values it impacts. I am a tenured member of the philosophy department at California State University, Sacramento where I teach  bioethics, business and computer ethics and contemporary moral issues. I have taught or studied at four land grant universities over a 20 year period. I am currently national Coordinator of the Soul of Agriculture project.[2]

                I volunteered to testify at this hearing because, as Soul of Agriculture Coordinator, I am concerned that a very promising technology, genetic engineering, in its current economic and institutional context, may cause environmental harm. This would impact most noticeably on family farmers and may damage the trust with which  the public regards farmers. But since this harm to farmers and the environment can be prevented mainly by the hard work of independent researchers in our great agricultural universities, I will try to make clear how the current work atmosphere in  applied genetics does not seem well suited to protecting the environment or farmers. One prominent agriculturalist at a nearby university described that atmosphere as "irrational exuberance."[3]


Known Environmental Risks

                Other speakers will deal with a broad range of environmental risks. My focus on farmers is intended to serve as a motivation to Land Grant leaders to be more concerned with their historical constituency: The farmers of America and though them, the consumers of the food they produce.

                Among the known risks to the environment are: 1.) the outcrossing of herbicide resistance to weeds producing "super-weeds" which will require more environmentally toxic countermeasures and higher costs to farmers. 2.) The development of insect resistance to the Bt pesticide due to plants which express Bt toxins constantly. Outcrossing is a completely  known risk because it has already occurred in  Canada where a weedy canola  showed up resistant to three herbicides, Roundup, Liberty and Pursuit, derived from the pollen of three open field tests being conducted in the area. Herbicide resistant crops have other risks, already in evidence, of lowered yields and higher costs as well as increased use of herbicides, such as glyphosate, whose toxicity profile is reported by the industry itself.[4] Monsanto reports residues in lettuce five months after application to fields planted to lettuce four months later, and in barley four months later. In California, glyphosate is the third most common cause of acute illnesses in farmworkers.[5] Adverse chronic effects are found in lab animals at less than the maximum dose in every category of chronic toxicity. Damage to beneficial insects is well documented as is glyphosate's mobility and persistence in soil and to a lesser extent in water. The Environmental Protection Agency's Ecological Effects Branch states: "In summary, this herbicide is extremely persistent under typical application conditions."[6] The massive increase in the use of glyphosate traceable to genetic endowment of agricultural crops with glyphosate resistance can be safely predicted to have effects on other arthropods, predatory mites and humus producing varieties. Earthworms and fish are impacted as well.[7]

                Bt corn and other Bt endowed plants bring Bt toxins into contact with beneficial soil organisms and  may have harmful impacts there. Organic farmers stand to lose Bt spray as a valued tool as insects inevitably develop resistance to it. Meanwhile conventional farmers may bear damage to their soils in return for only a few years of protection.


Unknown Risks

                My testimony focuses on unknown risks because I am convinced that the strongest and most long-lasting solution to obtaining biotechnological benefits while reducing its risks is strengthening the ability of the public scientific institutions to study biotechnology alongside the industrial innovators but independently of their corporate funding. It is not by regulating tightly the potential of "conflict of interest" research that the public will be protected, but by generously supporting publicly oriented basic research in biotechnology with a view to understanding its risks and limitations as well as its benefits. The reason for this need to deliberately and generously cultivate a balanced study of biotechnology is both the immense cost of this kind of research and the massive domination of the genetic engineering paradigm in our Land Grant Universities (LGUs) schools of agriculture. A significant proportion of applied genetics research is publicly funded, but with a goal of producing useable products and tools. This intensely practical orientation is not bad in agriculture. It is, after all, an applied science and there is no greater satisfaction for an applied scientist than to see his/her work succeed in the real world. And in today's world, the rewards, psychological, institutional promotion, professional prominence and even (perhaps regrettably)financial are derived from reaching applied science successes smoothly and quickly That industry should support the speed and smoothness in reaching success is very difficult to see as immoral. Moreover, fundamentally moral scientists  with strong professional stature are not going to choose to be involved in any project which they see as threatening to the public. They are generally "true believers" in the best sense of the word. They honestly see the attractive potential of the tools they have become masters of. But all of these factors make them less promising a group to turn to for sober evaluation of the hidden risks of those same tools. And that is true also because of the intensely subjective portion of any ethical estimation of risks and benefits. 


Subjectivity in Estimation of Unknown Risks

Mathematics exist for risk estimation only when some significant details are known about the mechanics or real world events which can produce the risk. The science of gene expression and the mechanics of gene insertion are such immature disciplines that they are loaded with uncertainties and huge regions of the simply "unknown."[8] We can estimate and get some empirical confirmation of the peak incidences of  drivers going the wrong way on urban freeways if we know what time the bars close. But that is because of our knowledge of human behavior and the impacts of intoxication. In more unfamiliar areas, some really persistent and informed imagination may be required to even anticipate risk. Some would argue that this kind of unfamiliarity removes our obligation. But the vastness of the impacts of the risks, such as a chronic toxicity in a national food supply, or the irreversibility of the release of some prolific and threatening organism into the environment require a more demanding standard.

It is true that human moral agents are not generally responsible for unknown risks. We would be paralyzed in inaction if we were bound by the absolutely unknown. The risks already noted above are completely known.. But a division of the unknown is needed to understand the precaution: “Don’t dive headfirst into murky waters.” Even some grave urgency, like escaping a fire or a lifeguard on a rescue

will not remove the obligation to jump in feet first. That is because the ignorance imposed on the diver by the murky water does not blot out the knowledge that a rock or tree stump could be hidden there, a submerged log could have floated there since the last time one swam there.

                Innovation would be stopped if we did not have some way of dealing with unknown risks  It is both easiest and traditional to deal with risk on the basis of foreseeability. Known risks are completely foreseeable,, as in the superweed above, but there is still a subjective element of estimating how severe an infestation of such weeds would be. Other risks can be divided into 1.) reasonably foreseeable, 2.)unreasonably foreseeable, and 3.) no known reason at all. In the analogy of diving in murky water, experienced persons who have swum and divided frequently in the area will not worry about rocks or stumps, but based on their knowledge of storms and logs which move below the surface, some would check every season, some after every storm and some every day. Some would always jump in feet first.

They are all implicitly assigning the risk to one of the three categories above. And there is an incurable subjectivity in how that assignment is made. There is a mathematics about how fast water must move to be able to move a rock of a given size and weight. But I wonder if it has been worked out for huge, neck breaking submerged logs which weigh almost nothing relative to the water and bump easily along the bottom. Knowledge of the usual direction of storms, of the location of forest, river or human activity sources of logs, would all help. But, given the risk, no experienced lifeguard on Great Lakes or ocean beaches is going to go head-first into storm-murky waters. The risk is too great, the benefits too small, and the knowledge base too uncertain, and risk-minimizing alternatives are readily at hand. On Lake Michigan we did not worry about rogue German submarines, but we still went in feet first on stormy days.   

                To exploit the analogy of diving in murky water further, the uncertainties  of gene insertion and expression are a perfect case of murky waters. In some instances the water has been checked for rocks and stumps by experience, but the water is still murky, the science is not mature even if the manipulation of the genome has been successful. With widely consumed GMO soy, corn and cotton seed oil, we dove in head-first and nothing happened, at least as far as acute toxicity is concerned. . Diving in feet first would be to have been much more cautious. We could have required enclosed testing of living organisms and much more careful testing of food stuffs before release to public consumption. Labeling of GMOs which have been on the market for years might help by providing a control population in a search for a cause of some novel chronic toxicity.


Subjectivity and Imagination

                University scientists intent on pursuing practical GMO products cannot imagine what the more cautious scientists are concerned about. And that is exactly the problem. Here imagination is a deliberate and subjectively motivated, pre-science exercise. Environmentalists, food safety specialists, and ethicists are guided by hunches based on history and multi-disciplinary factors. But that does not mean they are worried about German subs. Consider two scientists, both skilled geneticists, staring into the murky waters of genetic modification, one involved in traditional breeding, the other involved in genetic engineering. The former urges caution, the latter says: "What could you possibly be worried about? What we are doing is just a refinement on traditional breeding." The breeder knows that this is only true with respect to the desired goal, but not with respect to the technical means and unanticipated outcomes traceable to the means. . And these are where the unknown risks of GMOs are to be found. This is clearly the consensus of the FDA scientists who commented on the Office of Management and Budget  guided position in the FDA that GMOs are not of regulatory interest in virtue of the means (process) but only as to the product (the food itself).[9] So the breeder says: "I am not sure what exact risks I am worried about, except that the process is new,  and poorly understood. I can easily imagine, as one of the FDA scientists did, scenarios, at least five, where accidentally toxic outcomes could result from the process."[10]

The gene-splicer responds: "Well, I don't think we understand any better how traditional breeding works, certainly not in the details and such breeding brings in all sorts of genes we would just as well leave out." To which the breeder responds: "If you can imagine a scenario in traditional breeding where a serious risk to the public or the environment is reasonably suggested, then by all means we should take the same kind of precautions I am advocating." To which the gene-splicer will be able to respond. "One person's reasonably imagined scenario of risk is another person's German submarine."[11] What this dialogue should make clear is that risk estimates are subjective, in significant part, when the risk source is poorly understood. And this subjectivity is only more clearly seen when either or both sides use arguments rather than empirical data to either dismiss risk or suggest the reasonableness of its possibility.


Subjectivity of Estimates of Reasonable Foreseeability of Environmental Risk

                Reading the USDA/APHIS defense of its deregulation of Roundup-resistant canola, one sees what this subjectivity can do. Only a short time before the stacked resistance occurred, APHIS produced a document in which everything, including the behavior of honey bees, was optimistically interpreted to make the development of superweeds seem remote. This kind of optimism is like the assurance that large sunken logs do not move quickly along bodies of water. Why not check? It is reasonable, but time and money consuming.

                Even more significant openings for subjective optimism exists when the :foreseeable risk is based on a poorly understood and relatively infrequently experienced mechanism. Every geneticist knows what "pleiotrophic effects" means in their field even if it was rarely cited 10 years ago. It means that we know a gene may have many different and differing degrees of expression.[12] There is a possibility that some introduced  gene will be enabled to produce a protein in higher quantities that it did in its natural location or even have an augmenting impact on a protein natural to the plant which would be innocuous at low quantities but allergenic at high quantities. Optimists will say: "For pity sake, if you worry about everything, you will get nothing done." .Pessimists will say: "Don't dive head first."

                Another example of the pessimist/optimist subjectivity is in the impact of Bt and other genetically modified crops on soil quality. Benbrook, drawing on the work of D. Gleba et al., notes that up to 10% of the photosynthetically fixed carbon is exuded into the soil by plant roots. The interactions between these exudates and soil microflora and microfauna are poorly understood, but their importance for plant nutrition and overall soil and water quality are not doubted.  Optimists say "We have been altering soil ecology for years, especially with soil fumigants, so why hold bioengineered alterations to a higher standard?" Pessimists say: "The new crops and the bacterial, fungal or nematode changes  they may cause would be conceivably much harder to reverse than simply discontinuing a fumigant or tillage practice."[13] Notice once again that these are pre-scientific arguments. Given the risks involved it would be gross negligence  and desertion of the scientific ideal, and its ethical role in society to allow one argument to win without any gathering of data to support the winning hypothesis. Hunches are hypotheses calling for experimental design and testing. The "higher standard" objection is a complete distraction based on an appeal to "fairness" when fairness is not the issue. Environmental and consumer safety is the issue. But once again, this kind of pre-science dialogue reveals the subjective preference for clearing the way for  speedy application of new technologies. It is not using  science to  responsibly settle questions with research and testing. It is using argument to make such scientific work seem unnecessary. It is essentially a continuation of the non-scientific (or extra-scientific) way in which the FDA established its doctrines on biotechnological risk in 1992. 


Institutional Risk:

                One thing the Atlantic Monthly article, "Kept University" did not deal with in depth is a serious risk to the credibility of the scientific work done at our land-grant universities in this state if some serious environmental harm occurred as a result of the exuberant support of agricultural biotechnology.[14]

The root of this risk is in the degree to which biotechnological approaches to agricultural problem solving have penetrated traditional departments, or caused the reduction of more systems-oriented and sustainability oriented approaches. The history of land-grant universities has been peppered with instances of favoritism toward certain clienteles, crops or technologies, but there has never been so pervasive an enthusiasm as we find today with biotechnology.

                This is extremely troubling. Even in departments which have little to do with biotechnology directly we see academic professional societies allowing their journals or national offices to adopt public relations initiatives in almost identical terms and strategies to those advocated by the industry[15]  There is practically no place in the agricultural departments of a neighboring land-grant university where the enthusiasm for biotechnology is not present, even though many faculty oppose the enthusiasm and see it as unwarranted. This enthusiasm is not a wicked impulse, nor is it a quality a research administrator would want to find lacking. But for it to be so widespread that it seems to speak for the entire agricultural school or worse, for its entire scientific community, grave harm would come to university science credibility if some risk turned out to be real. If some harm to farmers, the environment, or God forbid, the public food supply occurred and it were traced to the complete lack of any significant research into the risks of biotechnology, the university would be seen as grossly negligent. It would be like claiming there is no possibility of  underwater logs moving to the diving spot. If the university seemed to routinely repeat industry advocacy of field releases, of diving in head-first, and claiming that good science supported this, even while it was clear that little risk-science was being done, would the public ever trust them again? And when the public and policy makers can no longer trust their institutions of scientific research, where will they turn? I do not hesitate to say that the university has no right to risk its credibility in this way, because we have no substitute for its tools, training and intellects. This is the most serious aspect of the Novartis/Berkeley relationship. 

                All these harms are almost entirely traceable to the tendency of those invested in an exciting and promising project to take an extremely optimistic view of the risks of their technology. After the tragedy they will say: "How did we know a log moved in to the diving place over the weekend?"


Dangerous Reticence

                During the extensive interviews which I did in preparation for this paper I did not hear of a single incident of administrative pressure or peer pressure to quiet critics of any given biotechnological projects. But I did have one active applied geneticist speak to me as if there were a definite pro and anti biotechnology polarization. I was, in his mind, anti-biotechnology. Criticism seems to be taken to mean implacable opposition, classifying one as an untrustworthy "activist." Faculty, however, freely admitted that they did not see much point in speaking out at meetings where overly rosy pictures of biotechnology were being painted One explained it to me this way. Working geneticists, whether in biotechnology or standard breeding, all know the science and its uncertainties equally well. Why bother to point them out? It informs no one and only marks one as an "activist." This can only hurt the objector and possibly his or her graduate students. This sounds like a kind of voluntary self-censorship. But it is a very dangerous kind of reticence, it seems to me, since it is an atmosphere which can easily allow significant risks to be papered over in the interests of harmony. For the public well being, more active discussion is needed



                The most natural thing to occur to one upon reading "The Kept University" article is to call for regulation and or vigilance against abuses. Certainly there should be full disclosure of income sources of scientific investigators. But, as indicated above, enthusiastic investigators with no monetary interest are just as likely to lean toward optimism. And they will be more convincing defenders of their optimism because they are completely honest true believers.. The danger comes not from bad people but from good people--enthusiastic and competent applied scientists who wish to see their work have an impact on the real world. Agricultural schools are applied-science institutions and they should be. Their service, as the article notes, to agriculture and even to industry, is lost if they work too intently on short term projects. The academic side needs to pay more attention to the basics of genetic engineering. But eventually applications must be pursued. And given the immaturity of the science involved, the enthusiasm must be tempered by those who think about submerged logs moving into the diving place. True believers are the most infectious and convincing speakers. But they need to be tempered by equally competent scientists with less optimistic leaning to new products and more leaning toward public issues.

                There is another way of reading the sexual metaphor of  "The Kept University"  If the faculty and administrators who must provide labs and support for investigators are seen as surrendering their virtue and consorting with corporations for financial support it may be because the have been tossed out of their formerly honest relationship by a faithless spouse. Or as Matthew reports: "Whoever puts away his wife causes her to commit adultery."[16] This is a bit harsh, but when public funding began to dry up, science faculty had no alternative but to look for funds elsewhere. Faculty are not given permission to do nothing, nor are they paid for complaining.!

                Faculty and department heads at a renowned land-grant university suggest the following. A concerted and respected research program on biological risk, emphasizing, but not limited to those encountered in bioengineering is needed. Coinciding with increased funding of biological farming systems research, such as proposed in AB 2663, there needs to be a generous endowment of  the landgrants with  the monies needed to look at risks. My interviewees grant that a person in a single new position in a department heavily invested in industry oriented technologies would have no viable professional life. A statewide institute which would draw upon scientists in and outside the University of California to do studies of the risks of agricultural biotechnology may be the only way to protect the environment and the university itself. 


[1] This paper has been published in Inquiry in Action ( #26&27 Spring/Summer, 2000), the publication of the Consortium for Sustainable Agriculture Research and Education

[2] This project's description and goals are viewable at www.soulofag.org.

[3]  Charles Benbrook puts it this way:

          This enormous shift in resources and the focus of agricultural science has occurred so quickly that there has not been adequate time for much reflection on the sustainability or value of the resulting technologies. The loss of  the benefits of research that has been abandoned in favor of molecular approaches has similarly been unexamined. In most quarters enthusiasm over the possible benefits of agricultural biotechnology has been infectious and unbounded. ("Who Controls and Who Will Benefit from Plant Genomics?", presented at the AAAS Annual Meeting, February 19, 2000, Washington D.C.)

Benbrook's paper is viewable at www.biotech-info.net.AAASgen.html

[4] Caroline Cox, "Glyphosate (Roundup)", Global Pesticide Campaigner, April, 1999, p. 13.

[5] Ibid. p. 15.

[6] Ibid.

[7] Ibid. p.16.

[8] It is interesting, from a history of science perspective, to see such terms as "junk genes" in the literature. It reminds one of dismissive references to colostrum in medical literature and practice in the 1930's and 40's or  the confident statements that infant formulas contained all the essential ingredients of mother's milk and no significant risks.

[9] These FDA scientists opinions as well as OMB's effort to guide the "product, not the process" doctrine, as well as "genetic engineering is just a refined advance of traditional breeding" doctrine, can be read at www.biointegrity.org

[10] Ibid.(same web-site) see the memo of Carl B. Johnson.

[11] FDA's Dr. Linda Kahl pointed out this same problem in the FDA doctrines. Is the goal to institute exhaustive testing of an unregulated industry with a century of safety (traditional breeding) or do you want an argument that allows you to avoid looking for risks in GMOs? See her memo at www.biointegrity.org

[12] It was to such pleiotrophic potential that FDA's Carl Johnson was referring in his criticism of the official FDA doctrines. See www.biointegrity.org 

[13] Benbrook, op. cit, p. 11-12. Quoted "dialogues" are my inventions but reflect those of Benbrook.

[14] Atlantic Monthly, March, 2000, pp 39-54.

[15] See my forthcoming "Public Values in Agricultural Biotechnology Communication" paper which reflects on the industry's public relations policy and the effort to bring university faculty into line with this policy. Two recent  examples of academic acceptance of this policy are the American Dietetics Association's flyer, supported by Monsanto funding, in its December 1999 issue of its Journal of the American Dietetics Association  and the "Genetically Modified Organisms (GMOs), A Backgrounder by IFT" put out by the Institute for Food Technologists. Both are leading professional societies for food and nutrition science faculty. Both publications play down risk emphatically.

[16] Matt: 5, 32.