Are RoundUp® and RoundUp Ready® corn safe?
21 September 2012
UPDATE: The Séralini "study" discussed below has finally been
by Nature magazine. The retraction was in response to overwhelming
pressure from the scientific community for exactly the reasons outlined
here. Basically, by choosing a specific type of rat, overfeeding it,
using an inadequate control group and hiding any information needed to
repeat the experiment, the original "scientists" rigged their "study" to
produce the results they wanted.|
RoundUp® is a broad-spectrum systemic herbicide that has been in
use since the 1970s and is the most widely used herbicide in the
U.S. Glyphosate, the active ingredient, is a chemical that blocks
the enzyme that facilitates photosynthesis in plants. Classified as
an herbicide, glyphosate is a plant poison, not a people poison. It
is a very popular plant poison probably because it is so cheap and
effective. Spray a little bit on your yard and in a few weeks
you'll have a giant patch of dead plants.
Despite it's widespread use, merely saying the word RoundUp can generate
a lot of fear and hatred. Saying the words genetically modified or
Monsanto really gets people going. I am neither going to bash
nor defend Monsanto. Instead I will concentrate on a recent
(19 September 2012) French study about RoundUp® and
RoundUp Ready® corn.
Here's a Reuters article about the study:
Here's the study itself:
Before going any further, let me explain my situation. My sole
source of income is from growing Kona coffee. That means I
need a commercial crop every year or else I will go out of
business. My family and I would prefer to avoid that.
Luckily, Kona coffee grows so well in Kona that it will grow even
without fertilizer and pesticides. The trees will be pathetic
little sickly things that barely produce any coffee but they'll
still grow. I call it farming through neglect. There's
nothing wrong with farming that way but I need my coffee trees robust
and happy so they will reliably produce a commercial crop.
As a farmer I feel it is part of my job to learn about and
understand pesticides and fertilizers. Knowledge is
power. I need a commercial crop but I also need it to be safe
because I drink my coffee too. Furthermore, my family and I
live on the farm so I would like to avoid poisoning my own backyard
or making anybody sick. I want to be as gentle to the land as
possible so future generations can live here too.
Long ago I made the decision to not use RoundUp. Not because I
deemed it dangerous but because I found other solutions that are
better for my needs. RoundUp is such a controversial brand
that there have been literally hundreds of studies about it.
Early studies covered the issues quite thoroughly. After more
than four decades of use, new studies often focus more on fear
mongering than fact.
Ignore the man behind the curtain
The recently released "study" mentioned above is highlighted on the
website SustainableFoodTrust.org. Visiting that website, the
very first thing I saw was the quote:
"Safe" levels of GM Maize and Roundup can cause tumours, multiple
organ damage, and premature death, finds first peer-reviewed
lifetime feeding trial."
My BS detector immediately started to twitch. My first thought
was "What changed?" If "safe" levels are so dangerous why is
it only now being seen after more than 40 years of constant use?
Farmers, landscapers and gardeners aren't all stupid, you'd think
someone would have noticed something wrong without scientists having
to tell us.
The Sustainable Food Trust website is very slick with lots of pretty
pictures, professionally produced videos and dramatic quotes that
zip across the screen. It looks to me like it's designed for
people with the attention span of a preschooler. I wanted
actual science, not a bunch of drama.
That brings up a real problem in our society. Hype and
propaganda are free but scientific papers usually cost an inordinate
amount of money. Anybody in a scientific line of work knows
that it is a real problem and can seriously hamper progress.
It is however quite convenient for bad studies because it is so
difficult to find the original source. This study, in a rare
exception, is available to the general public for free.
Let's Go Fishing
The first problem I noticed with this study is their control
group. The study took 200 rats and divided them into 20
groups. Two of the groups (one male and one female) received a
standard diet while all the other groups received diets with various
amounts of RoundUp and/or RoundUp Ready corn. See the problem?
Imagine that you're doing an experiment trying to prove that you can
catch more fish with cheese than with worms. You put ten poles in the water,
nine baited with different types of cheese and one baited with a
worm. The one pole with the worm is your control group.
The problem is that there are nine poles with cheese and only one
with a worm. A fish nibbling at a random pole is much more
likely to end up with cheese, thereby "proving" that cheese is
With careful statistical analysis it is still possible to get valid
results even with such skewed sample groups. At the same time
it is also a great way to go fishing for the results you want.
The question is, was there a good reason for using such a small
control group or were the scientists fishing for specific
results? After all, statistics never lie but liars use
statistics. If I were trying to overturn years worth of
previous data I think I would have designed an experiment that was
beyond reproach rather than right on the edge of viability. So
why did these guys use such a small control group? I don't
Back in the early 1900's, shortly after Darwin became well known,
the public was fascinated with animal intelligence. Around
that time there was a famous horse named Hans that could do some
pretty fancy arithmetic. For example, ask Hans "What is the
cubed root of 27?" and he would tap his hoof three times.
It was deceptively difficult to figure out what was going on.
As far as anybody could tell the horse wasn't being given signals of
any kind. The horse's owner didn't even have to be present,
anybody could ask a question and the horse would give the answer.
It was only after much study and far too much controversy that this
mystery was finally solved.
As it turns out Hans only gave the correct answer when the questioner
already knew the correct answer. If the questioner knew the
answer then the horse was correct 89% of the time but if the
questioner did not know the answer then the horse was correct only
6% of the time. Similarly, put the horse behind a curtain
where it could not see the questioner and all of a sudden the horse
was no longer a genius.
The horse, of course, couldn't really do math. What the horse
could do was pick up on involuntary body language in humans.
The horse would simply stomp its hoof until the crowd twitched,
gasped or started to smile, indicating that it had reached the
correct answer. This "Clever Hans effect", or Expectancy Bias,
is why it is important for scientific experiments to be
Here's another example. Imagine that you have two cups of
coffee, one of which is laced with poison. Knowing which cup
is which ahead of time will almost certainly affect your evaluation
of the taste. If I hand you the cups at random you won't know
which cup is which but you might still be able to guess based on my
reaction when you take a sip. That's why it is important for
researches as well as their test subjects to not know which group is
the control group. It's called a double-blind study.
I looked all through the RoundUp study but couldn't find the word
blind anywhere. I read "Section 2: Materials and methods"
thoroughly but could find no mention of making the various groups
anonymous. If the researches knew which group of rats were supposed
to have tumors and which were not then it is quite likely that their
evaluations were not totally unbiased. Once again either the
experiment was very poorly executed or the "scientists" were fishing
for specific results.
Trust Me, I'm a Scientist
Peer-review is a cornerstone of science. It is what allows us
to separate previously unknown phenomenon from experimental errors
and personal interpretations. Without proper peer review a
scientist might as well say "One time I saw Bigfoot flying a
UFO. I can't remember exactly where or when but trust me, I'm
Unfortunately this study includes very little actual data, making
peer-review nearly impossible. In the study they added RoundUp
to some of the food and water they fed to the rats. The problem is
that RoundUp is not a single product, it is an entire brand of
products. Each version of RoundUp has its own list of
ingredients, some harmless and others known to cause problems when
Glyphosate is the active ingredient in RoundUp but this study
ignores glyphosate, concentrating instead on a specific version of
RoundUp: GT Plus. Even though the study says it is
important to study "total chemical mixtures," it is left to the
reader to figure out what that mixture might be. The study
says they used "GT Plus" but the closest I could find is Roundup®
Grand Travaux Plus which is, I think, a French version. The
exact ingredients are not listed and some are a trade secret. In
other words, there is no way to know exactly what the researchers
fed to the rats.
Beyond "what" was fed to the rats, the study also fails to mention
"how much" was fed to the rats. This particular strain of rats
is quite prone to developing tumors. It is a well documented
phenomenon. In fact, if you want your lab rats to develop
tumors, all you have to do is give them an unrestricted diet.
For all we know that is exactly what this experiment did.
There is no mention of food quantities or weight gains anywhere in
It gets worse.
The conclusion starts with the following sentence:
This report describes the first life-long rodent (rat) feeding
study investigating possible toxic effects rising from an R-tolerant
GM maize (NK603) and a complete commercial formulation of
Despite the phrase "first long-term feeding study" being quoted
several times in the press, this statement is flat-out wrong.
Here is an
assessment of 24 long-term feeding studies
on exactly the same subject. Not a single one of them found
any safety problems linked to long-term consumption of GM
food. Either the authors are amazingly ignorant of their own
field or they willfully ignored prior work. I don't think
they're ignorant because in their own introduction they say:
long-term and multi-generational animal feeding trials have been
performed with some possibly providing evidence of safety, while
others conclude on the necessity of further investigations...
In other words, similar studies have indeed been done before but
none of them showed evidence of problems. The best the authors
could come up with was "necessity of further investigations."
I find it interesting that they so quickly discount all the prior
studies that "provide evidence of safety," making sure to use the
word possibly first. That's not very scientific.
Monsanto demands that their products are safe but I don't care
what Monsanto says. Anybody selling a product can't help but
be biased, even if they are well meaning. Monsanto has an
obvious conflict of interest when publishing information about
their own product.
Speaking of conflict of interest, my favorite line in the study is
the very last one:
The authors declare that there are no conflicts of interest.
I decided to ask Google what it knows about Séralini, the
primary author of this study. It turns out that
Séralini is a well known critic of Monsanto, RoundUp and GMO
foods. He has a vast array of prior work with very similar
conclusions to this one. He is open about being funded by
Greenpeace and he has a rather long list of books for sale through
Amazon. In other words, he makes a living bashing Monsanto,
RoundUp and GMO foods. The more he can get his name in the
press, the more money he makes. So not only does he have a
conflict of interest, he is willing to blatantly lie about it.
I find that sad and wish I could take away his Science license.
Like so many things in life, glyphosate and GMO foods are neither
all good nor all bad. As mentioned previously, glyphosate has
been in heavy use for more than 40 years now. It does have
some known issues that have been well documented. Overall, it
is far less problematic than most other pesticides. As time
goes on, the chances of discovering serious dangers caused by
glyphosate become less and less.
Genetically modified foods are a relatively new creation. In
one sense genetic modification is not much different that breeding animals
or grafting plants. On the other hand, genetics is a very
powerful tool, more like a chainsaw than a pocket knife, and if
we're not careful we could do some serious damage.
It's not like GM food is still experimental though, if you eat food
that you didn't grow yourself then chances are good that you are
already a consumer of genetically modified foods. Almost
anything cooked with vegetable oil, almost anything with sugar (corn
syrup and sugar cane), almost all corn products, soy beans, papayas,
zucchini, plums, animal feed, margarine, lipstick... are all common
I'm not saying that glyphosate and GM foods are safe. What I'm
saying is that too many false alarms will make us stop listening. We
need to be realistic, not panicky. When someone runs around yelling that the sky is
falling, especially for something that is already such a large part
of our economy, that person had better have some damned good
evidence. If not then they need to shut up because spreading
fear for no good reason is the same as shouting "Fire!" in a crowded
movie theater. It just ain't right.