Editor's note

Citation

(2006), "Editor's note", British Food Journal, Vol. 108 No. 8. https://doi.org/10.1108/bfj.2006.070108haa.001

Editor's note

A common misconception is that science and research are about facts, whereas in reality, research methodology concerns the unknown, hypotheses, probability, balancing and judging evidence or data. Thus, even in an objective research world, there is a need for interpretation and possibly an element of subjectivity.

Often with an incomplete picture, judgments are made, which maybe reported as facts by the media, leading to consternation in the minds of the public when new or contradictory findings later emerge.

One highly topical area, in which research and data collection needs to be undertaken, relates to genetically manipulated, or genetically engineered foods/organisms. It is also an area in which scientific opinion may not only be split but also polarised.

Another subject of debate in the research world relates to the accuracy and validity of published work. Several high profile cases, relating to concerns over published research, have recently made international news. However, at a lower level, advancement within disciplines requires vigorous debate and argument in order to progress. Most journals, even with peer review, will at some stage publish a paper which later goes on to be questioned. In the past this has been very rare with papers in the British Food Journal, although the validity of a paper involving GE corn, published in the BFJ in 2003, has now been challenged. The BFJ, in common with the approach used by other journals, has decided to publish a letter commenting on the GE corn paper. However, recognising there are two sides to every coin the BFJ has accorded the main author the right to reply. The philosophy behind the approach being that it is for the readership of the journal to make up their own mind. It is unlikely publication of these two letters will resolve the issue and interested readers are advised to monitor any continuing debate using the respective authors' web sites.

Dear Professor Griffith and Board

I am writing about an article: “Agronomic and consumer considerations for Bt and conventional sweet-corn”, D.A. Powell, K. Blaine, S. Morris and J. Wilson, British Food Journal, Volume 105 Number 10, 2003, pp. 700-713, which won the British Food Journal's Award for Excellence for Most Outstanding Paper in 2004.

A book by a leading Canadian journalist, Stuart Laidlaw (Secret Ingredients: The Brave New World of Industrial Farming 288 pages 1st ed. (15 April 2003) McClelland & Stewart; ISBN: 0771045956) raises serious questions about the method for collecting opinion data in the Powell et al. research.

Powell et al. reported consumers at a farm store showed a strong preference for GM sweet corn over non-GM corn. In the paper the choice appears straightforward – the bins were “fully labelled” either “genetically engineered Bt sweet corn” or “Regular sweet-corn”. The only other written information mentioned that might have influenced the preference of customers was lists of the chemicals used on each type of corn, and pamphlets “with background information on the project.”

However, according to Toronto Star reporter Stuart Laidlaw, when he visited the store on several occasions during the data collection period, the sign above the non-GM corn bin was headed, “Would You Eat Wormy Sweet Corn?” Above the Bt-corn bin the equivalent sign was headed: “Here's What Went into Producing Quality Sweet Corn”. Although Powell et al. describe in some detail the care taken to avoid biasing consumer choice during the research, there is no mention in their paper of the corn being labelled “wormy” or “quality”. Laidlaw includes a photograph of the “wormy” corn sign in his book. The photograph has been reproduced online: www.gmwatch.org/p1temp.asp?pid = 72&page = 1

The signs are not the only instance of methodological bias observed by Laidlaw but not reported by Powell et al. A number of fact sheets promoting genetic engineering were available at the farm store – some authored by industry lobby groups – but there was no balancing information from critics. Laidlaw also reports that on one occasion the lead researcher demonstrated to him his ability to influence a customer's responses in favour of Bt corn. The customer, who had bought non-Bt corn before Powell talked to him, told Laidlaw he would buy GM corn the next time he was at the store. Such interventions are of particular significance given that Powell et al. report that the store had a high number of repeat customers.

Powell et al. acknowledge that this was only a preliminary study and there were a number of limitations arising from consumer testing based solely on this one farm store. However, the cornerstone of science is full and honest reporting, and this experiment and its controls do not appear to have been reported either fully or honestly. The research should, therefore, be withdrawn, as should the journal's Award for Excellence.

Professor Joe Cummins Professor Emeritus University of Western Ontario, Wilkins Street, London, Ontario, Canada

May 30, 2006

Professor Chris Griffith

Would you eat wormy sweet corn? Or cabbage? Or broccoli?

That is what Ontario, Canada, producer Jeff Wilson often asks his customers. With 200 acres of fresh fruit and vegetables and a retail market on the farm, inquiring about his customers' preferences is not just good manners, it is good business.

Throughout the 1990s, Wilson's customers expressed a desire for reduced pesticides in the fresh produce purchased at his Birkbank Farms market. Wilson adopted an intensive integrated pest management program, but when cool, wet weather struck in 1997 – ideal for European corn borer – many of Wilson's customers who had previously said they could tolerate wormy corn by breaking off the damaged ends were no where to be found.

Wilson lost about $25,000 on sweet corn sales that year; an expensive lesson in people say one thing, but when it comes to grocery shopping, often do another.

So when I approached Jeff Wilson in 1999 about growing a genetically engineered Bt-sweet corn that in Florida field-trials had significantly reduced the need for pesticide sprays to control corn borer, he was enthusiastic.

I was eager to see what consumers would do when given a choice between genetically engineered and conventional whole produce – in this case sweet corn and potatoes – in a market setting instead of a survey or willingness to pay experiment which are both notoriously misleading.

As described in our paper (Powell et al., 2003), conventional (what we labeled as “regular” based on customer feedback) and genetically engineered Bt-sweet corn and potatoes were grown in similar eight acre plots, harvested, segregated and made available for sale at the Birkbank Farms Market.

Joe Cummins, and others on the internet, have accused me, and my co-investigators, of academic fraud and bias, because a sign sitting atop the bin of regular sweet corn asked, Would you eat wormy sweet corn?

That is a question Jeff Wilson cares about, with his pocketbook. It is also the language consumers use when talking about sweet corn, and what they are looking for when they peel back the silks of corn-on-the-cob.

But is it language intended to manipulate consumer purchasing patterns?

No.

The use of language, and its shared meaning, is always subjective. I have always based such work on the integrative risk analysis framework, first promoted by the US Presidential/Congressional Commission on Risk Assessment and Risk Management in 1997 (available at: www.riskworld.com/Nreports/nr7me001.htm) which argues that risk assessment, management and communication activities should be intertwined and reciprocal, rather than separate entities. And the best way to deal with value judgments in risk analysis is to openly declare potential sources of bias.

My bias is that science has a responsibility to lead, to explore the use of new technologies to enhance the safety and quality of the food supply while actively minimizing risks and respecting the concerns of affected consumers. For over a decade I have devoted my career to reducing the incidence of foodborne illness, and to the responsible use of new technologies to enhance the safety of the food supply.

Wilson and his staff at Birkbank Farms are committed to providing consumers with high quality food produced in the safest manner, as well as clear and accessible information regarding how that food is produced. Our shared goal is to understand consumer preference, not shape it.

The point-of-sale information in 2000 (and in subsequent years not described in Powell et al., 2003) at Birkbank Farms consisted of a large placard describing the options Wilson had to produce non-wormy corn, smaller handwritten signs describing the treatments received by corn available for sale on a specific day (which varied weekly throughout the course of the six-week consumer data collection period to reflect the different conditions under which different rows of corn were grown and variations in weather) and information pamphlets. This presentation can be viewed at: www.foodsafetynetwork.ca/images/sections/sweet-corn-model-farm.jpg

The large placard contained the following text:

“Delivering High Quality Sweet Corn

In order to provide you with the quality of sweet corn that you want we have three options

1. 1.

   Genetically engineered Bt-sweet corn:

   • •

     contains Bt protein in leaves and stalk; and

   • •

     requires fewer insecticides to prevent worm damage thus minimizing environmental impact.

2. 2.

   Bt-spary

3. 3.
   • •

     same Bt protein as in genetically engineered variety but sprayed on leaves; and

   • •

     protein exists naturally in environment and breaks down rapidly ...

4. 4.

   Conventional pesticides:

   • •

     used by most farmers to create worm free corn; and

   • •

     applied according to guidelines set by governments, but harm to beneficial insects observed”.

Because the work at Birkbank Farms was an overall risk analysis experiment in providing the public, and anyone else, with full and transparent information about how a particular commodity was produced, a press conference was held at Birrkbank Farms on 30 August 2000, to mark the beginning of the sweet corn harvest. The handwritten sign over the regular sweet corn asked, Would you eat wormy sweet corn, and then listed the treatments that corn had received to produce less-wormy sweet corn. The handwritten sign over the GE sweet corn (and we deliberately chose the label GE sweet corn because that is what it was – not just genetically modified, not a product of biotechnology or other terms that proponents of GE have suggested may be more palatable to the consuming public) said “Here's what went into producing quality sweet corn”, and listed no pesticides but herbicide and fertilizer. The handwritten signs were changed the following week.

A critic of GE may charge that simply asking the question, Would you eat wormy sweet corn, unduly influences consumer preference. A supporter of GE may charge that by labelling the corn genetically engineered unduly stigmatises the product and influences consumer preference (Powell, 2001).

I find such categorizations simplistic.

However, one journalist, among the dozens of other journalists, scientists, activists and hundreds of consumers who visited Birkbank Farms during the data collection period, and cited by Cummins, apparently interpreted the sign as evidence of manipulation.

We observed no evidence to support that charge, either through formal intercept interviews or anecdotal conversations; quite simply, no one else mentioned the wormy corn aspect of our signage (which was referred to in the description on the placard and, briefly, on the handwritten sign), although we admittedly did not specifically ask the question. What we did observe and respond to was heightened customer interest in methods of food production generally, and in response we developed and maintained a three kilometre self-guided walking tour on Birkbank Farms outlining the various tradeoffs and choices that face a commercial producer. Hundreds of people who wanted to know more about how their food was produced and the challenges involved took a stroll through the farm in 2000, and hundreds more in subsequent years.

In 2001, when we deliberately downplayed the research at the farm after the extensive media attention the previous year, sales of GE sweet corn outsold regular sweet corn 5:2. The presentation used that year is available at: www.foodsafetynetwork.ca/en/article-details.php?a = 4&c = 18&sc = 137&id = 889.

Cummins also alleges that the point-of-sale literature was promotional. The only literature that I am aware of present at point-of-sale was a brochure written by Katija Blaine and me, that contained information about benefits, risks and management strategies. Interested readers can make their own conclusions about the alleged persuasive nature of the brochures – one for Bt-sweet corn and one for Bt potatoes – by viewing them at: www.foodsafetynetwork.ca/en/article-details.php?a& = 3&c = 9&sc = 53&id = 886 and www.foodsafetynetwork.ca/en/article-details.php?a = 3&c = 9&sc = 53&id = 887,respectively.

The research at Birkbank Farms had strengths and weaknesses and both were related to the commercial nature of Wilson's operation. However, since May 2000 when we first wrote to Wilson's neighbours to inform them of our intent and hosted a public meeting for others to voice their concerns, we have been completely open about our intentions and results, and welcomed criticisms as a way to improve the project.

Powell, et al. (2003) explicitly acknowledged the limitations and applicability of the research by stating, “The labels on the produce bins may have influenced consumers to buy, just because they were there or perhaps because there was detailed information provided”, and concluded, “This research is a starting-point and describes the experience of one farmer on one farm during the 2000 growing season”.

Finally, to suggest that I possess some extraordinary persuasive skills, and that if I did, I would influence sweet corn purchases, one buyer at a time (with an intercept interviewee who was not included in the study) says more about the preconceived notion of my critics. What some allege is manipulation could more readily be described as conversation. Talking to people is good for Jeff Wilson's business and good for researchers.

Douglas PowellAssociate Professor, Department Diagnostic Medicine/Pathobiology, Kansas State University, Manhattan, KS, USA