GM-free Scotland

COMMENT

The article below was first posted on this website in June 2006. It concerns a small-scale study of the cultivation and marketing of GM corn.

As you will read, GMfreescotland.net was unimpressed by the paper, especially its liberal use of vague statements about claimed 'findings' unsupported by data. We were not alone in our reservations about the quality of the science in the paper. Professor Joe Cummins wrote to the Editor of the British Food Journal which had published paper to voice his concerns. The Editor countered his criticisms by stating “A common misconception is that science and research are about facts” (Institute of Science in Society Press Release 30.08.07). Well, we can certainly agree that, in this paper, the science and research don't seem to be about very many facts.

Over a year later, a fresh scandal has erupted over the presence of signs above the two types of corn on sale. These were not described in the paper but photographed by a journalist at the time. They suggested an attempt to influence customer decisions about which corn to buy and bias the results in favour of the GM choice. When you have read the article below, read SLAPP ON FREE SPEECH – News November 2007.

QUALITY OR WORMY
(First posted June 2006. The final three paragraphs in the original have been deleted as the references given there are now out of date)

Until the more pressing question of GM safety is resolved, purely agronomic or retail considerations are not of prime importance.

However, one scientific paper on these subjects was brought to our attention by GM Watch sceptics because it has received the British Food Journal's “Outstanding Paper Award for Excellence”.

The paper compares “Agronomic and consumer considerations for Bt and conventional sweet-corn” in Canada.

Out of interest as to why this paper merited such accolade in a British journal, we took a peak.

The study was a “farm-to-fork” trial in which two varieties of Bt and two varieties of conventional sweet-corn were grown side-by-side and then offered for sale side-by-side in the farm's shop. The aim was to assess economic differences in producing GM and conventional crops, and to identify factors which influenced consumer purchasing decisions.

The introduction to the paper sparked a waffle-alert. It seemed to include superfluous information, irrelevant to the rest of the paper. For instance, Bacillus thuringienses is a “gram-positive” bacterium which produces an insecticide “in the form of a crystal”.

The abstract sparked a numeracy alert. Here, it is explained to the reader what a dozen means, and that the GM maize outsold the conventional maize by a “margin” which was then given as a ratio.

These first impressions were re-enforced in the course of reading the paper.

The results section is peppered with opinions based on vague observations unsupported by actual measurements (a.k.a. hearsay). For example:

picking and packing were found to be “less time consuming” for the Bt crop (a second less, or a minute, an hour, a day a week, a month ...?).

professional scouts regularly checking the crops for pests “noted” that pesticide sprays killed non-target insects (how many insects, of what kind?), and, “noted” that pest damage to conventional corn was 10-20 per cent compared to 1-2 per cent for Bt corn (what did they count? plants, leaves, bite-marks ... anything at all?).

somebody in the farm shop (the cleaner?) threw away ten cobs out of every 100 of the conventional crop daily while “almost no” Bt corn (the odd kernel or stalk perhaps?) was thrown out.

In the discussion, the use of unspecified quantities continued. An entire paragraph revolves around the revelation that a “few” customers (how many? 2, 5, 10, 20 ... or 1 plus some imagination?) in the farm shop were observed (by a fly on the wall presumably, or maybe it was that cleaner again?) to fill their bags with the regular corn, pause to read the signs above the bin (more on signs anon) and refill their bags with Bt corn.

The suspicion that this was a pro-GM PR initiative dressed up as science is strengthened on reading the results for 'Economics' and 'Sales' considerations.

Economics?

(If you have a phobia for maths, skip to the last two sentences in the paragraph.)

The total input costs for the three staggered plantings of all the sweet-corn varieties tested were:

Planting 1 – Bt crop $7,000.55; Conventional crop $7,290.80

Planting 2 - Bt crop $7,000.55; Conventional crop $6,829.54

Planting 3 – Bt crop $7,000.55; Conventional crop $6,626.97

(Our table, not theirs)

In other words, the Bt crop cost the same each time, while the conventional crop cost more than the GM one for the first planting and less than the GM one for the other two (this happened because the conventional crop had varying insecticide applications while the Bt crop had none). This is reported in the paper as “The first planting of conventional had total costs of $7,290.59, |close to $300 greater than the Bt. The second and third plantings of conventional corn, at $6,829.54 and $6,626.97 respectively, cost slightly less than the Bt”. Problem? “Close to” and “slightly less” are not scientific measurements. Since the actual figures are given, we can do some simple arithmetic to find out what they mean: “close to” means $290.21 more; “slightly less” means anything between $171.05 and $373.62 less. Now you know. “Slightly less” can mean a difference of $83.41 more than “close to”, or, a difference of $119.16 less than “close to”. Or, to describe it in more usual comparative terms, the conventional crop cost 4% more than the Bt one for the first planting (i.e. “close to” means 4% more), and 4% and 5% less than the Bt one for the second and third plantings respectively (i.e. “slightly less means 4-5% less). Or, put another way, when the sums show an advantage to the GM maize, the figures are rounded up to the nearest whole hundred they are “close to”. While, when the sums show an advantage to the conventional maize, they become “slightly” less, even if the amount is in fact greater than the one rounded up to prove the worth of the GM crop.

Sales?

The reader is told “Bt sweet-corn sales were consistently higher than regular sweet-corn sales”. The data, however, displayed in a graph, suggests that sales of Bt corn were lower on 2 occasions and no different on 4 occasions out of the 31 days on which sales were recorded. Perhaps this should have read 'The Bt sweet-corn sales were consistently higher except for 19% of the time when they were not'.

If you have trouble believing that any scientist would deliberately try to bias the outcome of his experiment, here's what happened to the consumers whose “purchasing decisions” were being measured.

Before and during the test period, potential customers were invited (by hand-delivered letter and media advertising) to attend a public meeting. Inside the farm shop, a perceived need for “open communications” warranted that they were regaled with posters, bookmarks, pamphlets and signs. In case a customer should miss all these, three press releases and a press conference ensured coverage in TV, radio and print media. And, in case anyone was housebound but dying to know all about what was going on, there was an interactive Web page on a Web site containing “a weekly update on the research (including online video footage), and any press releases and relevant news items.”

Interestingly, after all that hype, it is noted in the Discussion that “Many customers at the ... farm market know very little about GE foods”. Also, the results indicated that 32% of customers failed to get the message and failed to want Bt corn, and that sales of Bt corn were only 10% above a random selection level of 50:50 Bt:conventional when both types of sweet-corn were displayed side-by-side, and clearly labeled, for sale.

This brings us to what started all this in the first place: the bit about being clearly labeled.

According to the paper, the two types of corn were presented in separate wooden bins labeled with either “genetically engineered Bt sweet-corn” or “Regular sweet-corn”, along with information on the number of chemical sprays used. A leading Canadian journalist for the Toronto Star, Stuart Laidlaw, who visited the farm several times, reported and photographed the actual signs placed above the bins. The one above the Bt corn reads:

“Here's What Went into Producing Quality Sweet Corn”, followed by a list of fertilizers.

The fact that it was Bt corn was shown on a separate sign.

The one above the conventional corn reads:

Would You Eat Wormy Sweet Corn?

Regular Sweet Corn:
Insecticides: Carbofuran – sprayed 3x

or

Bt foliar Spray – sprayed 4x
Fungicide: Bravo – Sprayed Once
Herbicide & fertilizer: 1 Application of each

Note. According to the paper the corn sprayed with Bt foliar was so disastrous is was not offered for sale. Either the paper was telling fibs and the wormy corn actually was offered for sale, or, the sign was telling fibs to give the impression that a lot more insecticide had been used than was actually the case.

Ah, this is fun, but we've got to stop somewhere.

OUR FURTHER COMMENTS

How did this paper come to be published in a mainstream scientific journal?

The British Food Journal describes itself thus: “After over 100 years of publication, the British Food Journal continues to be highly respected world-wide for its broad and unique interdisciplinary coverage of food-related research. The journal has a strong commitment to publishing the latest food research from around the globe, all of which is reviewed and adjudicated by an international editorial board of leading experts. ...”

Were the “leading experts” asleep, or did they fail to understand what had actually been done (we had to re-read the paper several times) and didn't like to admit it? Or, were they colluding because agreeing to nice things said about GM would help fill their pay packets and get them a pat on the back from similarly-minded colleagues?

How did the paper come to get an Outstanding Paper Award for Excellence?

Can no one on the British Food Journal's international editorial board recognise waffle, careless English, and hearsay dressed up as data, or apply simple arithmetic and percentages?

If this is a quality paper, just think what the wormy ones are like.

SOURCES

British Food Journal, 2003, 105 pp.700-13

www.emeraldinsight.com/0007-070X.htmwww.emerald-library.com/Insight//ViewContentServlet?Filename=Published/www.gmwatch.org/print-archive2.asp?arcid=6359