Walk Away! My Interview on "On the Media"

By Carl Zimmer | September 28, 2012 8:01 pm

My outburst last week about scientists trying to get reporters to sign a confidentiality agreement to see a paper on genetically modified food landed me on the radio. I spoke to Brooke Gladstone of “On the Media” for this week’s show. I’ve embedded the interview here.


Comments (8)

  1. Mary

    Thank you for coming out so loudly and clearly on this Carl. I really appreciate that you made noise and the other journalists who made the right call did so. Including Jonathan Amos of the BBC. Good job.

    And in this situation–and other cases of “reports” from activists that go directly to the press rather than through the usual science routes–the only thing we can do really is to shame that kind of behavior. There’s nothing else to be done in any kind of formal way. We can just hope that scientists who pull this crap think twice, and that stenographers of this sort reconsider the next time.

  2. gaddeswarup

    An article in The Guardian says (http://www.guardian.co.uk/environment/2012/sep/28/study-gm-maize-cancer)
    “In fact, there is one irony that a few scientists have pointed out but who have been drowned out in the furore. Séralini’s study was not so much about the dangers of GM technology, but the toxicity of the Roundup herbicide used on the crops.”

    CZ: The rats were fed different levels of genetically modified corn and Roundup, the herbicide to which the corn was modified to resist. I’m not sure where the Guardian gets the idea that the study was really about the Roundup alone and not GMOs. Ultimately, it tells us equally little about both.]

  3. MAUCH

    Continue in your tirade till this gets heard by more of the general public. More people have to understand that one botched research project does nor represent what science has to offer. Our culture could not exist without the advances that were born as a result of the majority of good science that can be found out there. The fate of the bad science like we see here is that it will be relegated to the scientific peer review junk heap.

  4. What?

    Response? http://www.gmwatch.org/latest-listing/51-2012/14226-response-to-monsantos-rebuttal-of-seralini-study-1

    [CZ: There are a few key points to consider.

    1. One is that the French researchers compared all the experimental male rats to the same 10 male controls, and all the experimental female rats to 10 female control rats. Typical studies will have many more controls. The GM Watch post mentions a study by Monsanto. Those researchers had data for 160 control rats to compare to 40 rats fed GMO corn at low or high levels. It’s also possible to use historical controls, which can include thousands of rats. The GM Watch post does not address this point.

    2. It is standard practice to show the statistical significance of results. There are a number of ways to determine that results are statistically significant. The most common is a p value, which essentially conveys the odds that the difference between two groups (tumors in GMO-fed rats versus control rats) is due to chance. The lower the better. Anything more than 5% is not considered statistically significant. The Seralini study describes some statistical methods in the methods section, but never presents any statistical analysis. Is the difference in the rate of tumors statistically significant? The paper doesn’t say. The GM Watch post does not address this point.

    3. There is no dose-dependent response in the results. That is, more GMO corn or Roundup does not lead to more tumors. In fact, in some treatments, the rats with the highest doses had fewer tumors than the controls. The GM Watch post does not address this point.

    4. The GM Watch post correctly notes that the Seralini study was not set up as a cancer test. They use this as a justification for some of the shortcomings of the experiment. If that’s the case, then people should not interpret the conclusions of this study as an ironclad indictment of GMOs. Also worth noting is that the paper itself is packed with pictures of GMO/Roundup-fed rats with tumors, but no pictures of the rats with tumors that ate a normal diet. This is hardly a representative sampling of the results. But these pictures were then used in the press materials released by the scientists–giving the impression that this study was, indeed, definitive proof that GMOs cause cancer.

    5. Finally, the GM Watch post says nothing about Seralini’s demand that journalists sign a non-disclosure agreement to see the paper in advance.]

  5. Precautionary Principle

    It appears that the esteemed Mr. Zimmer has not done his homework on this topic. There are reasons Seralini handled the press as he did. Scientists who don’t present the GMO party line are vilified. Go do your homework on what it means to be what is, essentially, a whistleblower on this industry. http://e360.yale.edu/feature/companies_put_restrictions_on_research_into_gm_crops/2273/

    [CZ: I’m well aware of the pressure that companies put on research they don’t like. However, I fail to see why that means that I should be prevented from reporting a scientific paper the way I always do, by talking to a range of experts to get a sense of its strengths and weaknesses.]

  6. Only dumb scientist would confirm this hypothesis by only testing few rats and only with one specie. It’s seems that no one learned with the test of aspartame in rats, thats shown us that we shared many biochemical pathways but not all. Rats tend to develop cancer with the same proporcional amount that it’s perfectly ok to humans.

    It’s “news” just scares the general public and that really is a problem concerning science progress. Journalist and public got to understand that science it’s a open book, that should be available to everybody. Both to learn and discuss

  7. Bill Roth

    I agree with what you said but there is a big component that I think has been missing from the discussion, and that is that the inherent power of rat studies to establish the safety of GM foods is limited by the feasibility of conducting large rat studies.

    I think that if the general public (and apparently a good number of scientists bloggers) were to listen to this coverage they would be left with the impression that it is possible to prove that GM food is safe with rat studies.

    It is not. The level of risk that most reasonable people consider “safe” is too small and it would require too many rats to establish that that small of an effect does NOT exist.

    Using this study’s goal as an example, if GM for were to increase the risk of getting cancer from 41 to 42% most people would think that was unacceptable. And really it would be because hundreds of millions of people eat GM food that would represent millions of people getting cancer over.

    Given a simple design with GM food/no GM Food, cancer/no cancer, the number of rats that is necessary to discover that small of a change is in the 10s of thousands of rats. It is not a feasible study.

    Rat studies of the size that are feasible to conduct (say a hundred rats in each group) can still only detect relatively large effects, e.g. increases of risk on the order of maybe 50%. Smaller but still meaningful effects would be invisible.

    This is not to say that there is any evidence that GM food causes cancer. But there are legitimate scientifically based concerns about GM food. Unfortunately, the anti-GM movement doesn’t seem to know what they are which seems to prevent intelligent discussion on this issue.

  8. Bill Roth

    I think this is also why they used the tumor-prone rats, too. If you, for instance, have 100 rats and 10 of your controls get cancer and 15 of your GM food rats get cancer then that isn’t significant. But if you expect 30 and see 45 that is significant, even though both are a 1.5X increase in cancer. If the tumor prone rats are still a good model for human you increase your statistical power by using cancer-prone rats. This is probably also why they use massive doses of whatever is being tested.

    This seems to be a bit of a Catch 22. If you design your experiment “properly” you’ll never find major cancer risks. If you design it to find the most cancer risk people complain it’s not done “properly”.

    Though obviously on the continuum of “proper” this particular experiment is pretty far off the beaten trail.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

The Loom

A blog about life, past and future. Written by DISCOVER contributing editor and columnist Carl Zimmer.

About Carl Zimmer

Carl Zimmer writes about science regularly for The New York Times and magazines such as DISCOVER, which also hosts his blog, The LoomHe is the author of 12 books, the most recent of which is Science Ink: Tattoos of the Science Obsessed.


See More

Collapse bottom bar