Photo by jere-me/FlickrCC
Curator's note: Like pretty much every advocate of sustainable agriculture, including our own Bill and Nicolette Hahn Niman, I took immediate and sharp notice of James McWilliams's New York Times op-ed suggesting that pigs raised in the very epitome of evil, pig concentration camps, might have fewer pathogenic diseases consumers need to worry about. And got worried about the possible setback to the remarkable work of Iowa pig farmers, including the original leader of Niman Ranch pork, Paul Willis, that I witnessed first hand a few years ago and wrote a pretty rhapsodic Atlantic column about.
The op-ed raised several questions, some of which Marion Nestle asked right away, others that numerous blogs asked as well.
McWilliams is a friend, and someone I value highly as an original thinker who wants to challenge our most cherished beliefs. And, as he makes clear in the reply I asked him to give us, he shares those beliefs too! He'll also be an occasional contributor to this site, including a provocative piece--one that will rouse the very people aroused by his recent op-ed--that will appear soon. Here's his unscheduled debut. I'm glad to have it.
First things first: I strongly support sustainable agriculture. I've never been in contact with anyone from the pork industry. I'm not a hired bullhorn for mass-produced animal flesh (even if I am from Texas). I don't even eat meat. More important, I believe that alternative agricultural systems must always remain open to scrutiny. Just because a small, sustainable farm operates outside the realm of conventional agriculture does not mean it's flawless. Agriculture, by definition, is flawed. I thus believe in frequent and intense self-examination. It's healthy, and sometimes, when done properly, even feels good.
As anyone reading this knows, I've endured a wallop of criticism over my piece on free-range pork in last Friday's New York Times. The condemnation scans the spectrum of civility. A butcher in Iowa has offered to remove my testicles. Marion Nestle, as well as a host of other smart writers, have not. But they have legitimately questioned why I did not do two things in my Times piece: a) reveal that the study I quoted, headed by Wondwossen Gebreyes of Ohio State, was funded with a grant from the National Pork Board, and b) claim that the pigs in the study tested positive for trichinella, salmonella, and toxoplasma--when in fact they tested seropositive for these diseases. I'd like to briefly respond.
Would a handful of credentialed scientists, the board of a respected journal, and a host of outside peer reviewers engage in a grand conspiracy?The first criticism is a red herring. In the course of writing a book on the history of insecticides, I learned that it's very difficult for scientists to undertake large studies without industry funding. You think their universities are funding them? Not likely. Why do you think so many university scientists bail out of academia and go to the dark side of Monsanto and BASF? Money. Scientists, many of them reluctantly, depend on industry to carry out basic experimentation. Fortunately, this relationship is not inherently corrupt. Industries frequently end up supporting studies that do not present the results they desire. These results still get published (and the scientists, in turn, often lose their funding). Being fully aware of the realities of scientific funding, I chose not to cite the Pork Board, but the journal in which the Gebreyes study was published. Foodborne Pathogens and Disease is a prestigious peer reviewed journal. The scientists who undertook the study are well respected researchers in veterinary medicine working at top universities. These factors assured me that any undue pressure from the funding source would have been negated.
My critics, however, fingered the pork industry's involvement from the start as ipso facto evidence of the study's lack of credibility. Do they do the same when studies funded by the Organic Farming Research Foundation highlight the benefits of organic food? Think about it. Would a handful of credentialed scientists, the board of a respected journal, and a host of outside peer reviewers engage in a grand conspiracy to twist the results of a major study that only indirectly made CAFOs (concentrated animal feeding operations) look like a preferred option? On the flip side, I know full well that if I mentioned in my lede the fact that the study was funded by the Powerful Pork People, a large portion of my readership would roll their eyes, pour another cup of Fair Trade coffee, and dig the Thursday Style section out of the recycling bin. And this is the last thing I wanted, given that these readers are the food world's movers and shakers--the very people I believe should take heed of the study and think seriously about the unintended consequences of producing food in the gray zone between the wild and domesticated. (Okay, I'm not sold on Fair Trade coffee or the Style section either--but these are fights we can have later).
The second criticism--the seropositivity matter--is more serious. I'll admit I may have erred here. May. I know full well the difference between testing positive for a pathogen and testing positive for the antibodies of that pathogen. You can test positive for the antibodies of a disease and still not have it. It's very unlikely, but possible. Faced with the choice of spending a couple of paragraphs qualifying this distinction for lay readers, I looked to see how other reports of the study dealt with the matter. At this point I was thinking not like a scientist but a writer. I wanted to keep the piece flowing without getting bogged down in the distracting minutiae of seroprevalence. The Ohio State University research summary of the article put it this way:
A comparison of swine raised in antibiotic-free and conventional pork production settings revealed that pigs raised outdoors without antibiotics had higher rates of three food-borne pathogens than did pigs on conventional farms, which remain indoors and receive preventive doses of antimicrobial drugs.It decided not to make the distinction. The very scientific study I wrote about put it this way:
The results from this preliminary study suggest that all three pathogens were more commonly present in pigs that were reared in an ABF, outdoor, niche-market type of environment than the conventional, indoor-reared herds.The New Scientist, summarizing the results of the study, did make the antibody distinction, but then explained that antibodies were "tell-tale signs of infection."
Absorbing these analyses, I reminded myself of what I thought to be the ultimate point in the first place: whether tested for the disease directly or for antibodies of the disease, the pigs in the study were more likely to be infected than the conventional pigs. At this point, I chose to write that the pigs had these diseases. While this decision, made in the interest of readability, hardly makes me an apologist for CAFOs, I can fully understand why readers might have preferred a more technically accurate description--even if it does not alter the underlying point of my argument.
I hope my critics are willing to meet me halfway. Even granting my potential translation errors, does it make sense to kill the messenger before exploring the full implications of the message? I'm genuinely distressed that not a single high-profile food writer has said, "hey, I have problems with this McWilliams piece, but what if it's right?"
Sustainable agriculture already has enough preachers preaching to the choir. I--a supporter of sustainable agriculture--could be one of them. But every movement must challenge its own orthodoxies. Those of us willing to do so must be assured that we won't be burned at the stake for asking tough questions that will only make our shared quest for sustainable agriculture stronger.
Subscribe to Atlantic Food
Prof. McWilliams. First off, I'm happy that you took this step to respond to the critiques of your piece. More importantly I'm glad to read you clarifying how you envisioned the column (challenging orthodoxies). I have managed to get myself into a few heated exchanges over some of the responses to your column over at Honest Meat and on the Ethicurian and I made many of the same points you've just made concerning the extent to which the knee-jerk reactions and ad hominem attacks suggests that those of us in the sustainable ag. community are perhaps drinking the kool aid. I am particularly troubled by the extent to which responses to your post have both drawn lines in the sand using categories which are fuzzy at best and reactionary at worst (good science vs. bad science or the blind faith in particular scales of production - big is good. no! small is good). This perhaps is where my own academic experience influences my interpretation, but the power of stabilized categories to lead not only to exclusion (us vs. them, human vs. animal, etc.) but violence. It should be possible to offer a critique of CAFOs which doesn't depend upon creating simple binaries and straw men. Similarly it should be possible to support sustainable agriculture while also pointing out the potential for danger in how particular modes of production are operationalized and defined. If, as a movement we cannot do that, then I fear we have simply become a group of fundamentalist fanatics, which is exactly how we are seen by many conventional producers.
You should be willing to concede the following:
1) the largest variance in seropositivity for these toxins was regional.
2) treating animals with antibiotics is a good way to prevent them from acquiring resistance to said microbes; this is what the study measured.
3) the differences in seropositivity were marginal. the difference in trichinella seropositivity was statistically insignificant.
summarizing those claims, the study isn't very important, yet it probably is more widely known than every study published in this month's editions of Science, Nature, and Cell. why is that?
Professor McWilliams, in the interest of accuracy, also might want to acknowledge that industrial livestock production uses antibiotics to promote weight gain, not for the health of the animals. The use of massive quantities of antibiotics in animal feed--in quantities that far exceed the amounts used in human medicine--does endanger human health by promoting drug resistant bacteria. In contrast, the use of antibiotics in sustainable livestock production is limited to the health needs of the animals. This distortion in the op-ed piece is more serious even than the decision not to distinguish between infection and seropositivity.
Well said, Sasha. This particular food movement is looking more and more like religion, with true believers unwilling to look critically at the alternative farming methods. Why isn't it possible to recognize the shortcomings of conventional agriculture and also see it's benefits and/or see potential problems of the alternative? Joel, posting above, is a classic. Joel argues, with some justification, that McWilliams is selecting his results to make a case. But Joel then goes on to do the exact same thing: Aha! I found one result that makes my case so I'm going to close my eyes and ignore the rest! Gotcha! We win! Nooo. We lose if, as these preliminary findings suggest, there are potential problems with free-range-type swine production systems and we refuse to confront them for fear of what we might find.
ps. That there is regional variation in seropositivity says nothing of the differences across production models within regions.
Thank you for inviting this response and thanks to Professor McWilliams for writing it.
As someone who has made it a personal mission and vocation to help people to look beyond overly simplistic and often abused label claims (USDA Choice, "Free-Range," "Natural," Grass-Fed” to name a few), I was distressed by McWilliams' article for many reasons.
If you strip away some of the details, most of the debate has focused on your citation of a single study - focused on a very narrow subject - to draw the broad conclusion that pork from free-range pigs poses a health threat to consumers on par with salmonella-tainted pistachios or peanut paste.
I appreciate your admission that you may have erred on this point, at least from a technical perspective.
However, other aspects of the original Op-Ed piece personally put me off at least as much as the debate over the science. Frankly, I'm surprised no one else has commented on this so allow me to do so briefly here. You can then see more details in my post.
http://discoverbeef.blogspot.com/2009/04/dont-eat-that-pork-trichy-words-from.html
I feel that you impugned the integrity of those who are raising and promoting pork from free-range pigs, suggesting that they are intentionally misleading consumers in order to charge a premium for their product. I couldn't disagree more.
Your article also seems to suggest that those of us who seek to eat alternatives to CAFO-produced pork are doing so only because we "despise industrial agriculture." First, I would ask, what's wrong with that? We buy things all the time for emotional or reasons other than a product's functional benefit (in the case of pork, this would be great taste & texture).
Second, outdoor raised pork does taste different. "Terroir," which would include a pig's diet, has a direct influence on the flavor and texture of the meat. This doesn’t mean that all outdoor-raised pork tastes great. Some farmers are more talented than others. Plus, what tastes great to me may not taste great to you. But the bottom line is that many people are purchasing outdoor-raised pork because they prefer the taste and texture of it.
To conclude, I have two requests. First, let's definitely meet halfway and continue this debate. But let’s do so without offering up sound bite-ready, scary-sounding headlines that oversimplify it.
Second, I’d love it if you’d respond to these other points, too.
the study, by w. gelbrayes, has a number of flaws. it's the responsibility of the author (or the interlocutor in this case) to prove that his/her conlcusions stand up in the face of criticism. this is the nature of peer review.
here are the flaws, in detail:
1) the samples are unbalanced (as you pointed out). look along the vertical axes of table 2. the only samples in all three regions with statistically significant numbers are "conventional farming" samples.
they vary in salmonella seropositivity/exposure: 52% in wisconsin, 30% in ohio, 37% in north carolina.
those differences are far greater than the differences between abf and conventional farming measured in individual (~9% in the statistically significant regional groups [i.e. not the 5 pigs tested in ohio] and ~9% as a whole). why is that? not discussed.
2) measuring seropositivity in animals treated a priori with antibiotics is a proof of concept, isn't it? we already know that antibiotics work. we also understand that there are evolutionary-driven caveats with widespread antibiotic use, as hank discussed.
3) third, the trichinella seropositivity is insigificant. in fact, gebreyes knows this
no mention of statistical significance, because there isn't any. here's the winning line, however.
not to hate on gebreyes; he did say the following:
for the record, of course i'm biased. i have opinions, like everyone else. the difference is, i'm backing them up with the record. maybe you'd like to oblige me with the same.
I'm sorry, but I just cannot buy this.
To choose the title "Free-Range Trichinosis" for an op-ed piece in the New York Times - in part startling as well as satirizing - and then, cut to the chase in the first two paragraphs with alarming assertions and some hard numbers lifted from a study published in respected journal as proof of those assertions without bothering to fully elaborate on the context or conditions of the study or the scientific implications of the results is simply astounding. Now, after being confronted by readers who have read the study in question, Professor McWilliams writes “Faced with the choice of spending a couple of paragraphs qualifying this distinction for lay readers, I looked to see how other reports of the study dealt with the matter. At this point I was thinking not like a scientist but a writer. I wanted to keep the piece flowing without getting bogged down in the distracting minutiae of seroprevalence.” Is Professor McWilliams telling us that in a non-peer-refereed journal the flow is more important than the facts!!! Quite frankly this is unconscionable. Is this the way Professor McWilliams approaches his writing for academic journals…I suspect not.
My feeling on the matter could not be put more succinctly than it was in one of the posts on the Ethicurean about Professor McWilliams’ piece: “When he writes a piece to inform and influence public opinion on the serious matter of public health, he bears the responsibility of being no less rigorous in his research and no less careful in his choice of words.”
Joel: I'm sorry to use you as an example. Bad form on my part. But I don't want to back off from my general point: that we should be open to evidence, and not fall into the trap that McWilliams seems to have fallen.
I view McWilliams much as I do some of those that most vehemently disagree with him: as idealogues. They've staked out their positions and cherry-pick evidence to support those positions. We should discount their opinions accordingly, or at the very least view them with a healthy dose of skepticism. My main concern is that the most public supporters and dectractors of the local/organic/natural/etc. food movement too often seem to not want to look coldly at the evidence. See Prof. Nestle's original post on this topic, which questioned not just McWilliams but the science seemingly because she didn't like the result or the financiers of that science.
My reading of the original journal article is this: the alternative production systems are potentially more susceptible to some animal diseases than are conventional production systems, perhaps because they choose not to use profylactic doses of antibiotics. The work is preliminary, as the authors note, and raises some additional questions that could be the subject of further study. It does not in any way address a host of also relevant questions: among others, the causes of apparently different rates of serpositivity; the magnitude of the problem; the costs (and benefits) associated with profylactic use of antibiotics in conventional agriculture; and so on. But it is appropriate that this work be part of the public debate.
Professor McWilliams seems to appreciate the value of strawmen since he used them in his original article, and he used one to catch this blatant “red herring” in the choppy sea of responding words. If one questions the motivation behind a research study, one does not necessarily question the integrity of the study or the veracity of the results. While much academic research is sponsored by commercial entities - both directly and through veiled intermediaries such as “industry boards” - with the intent of confirming a specific position, one can only wonder what the affect of “unofficial” reporting of negative results to a sponsor has on the completion of a preliminary study and on further studies in that area. I am not sure that by raising a question about a study’s funding, one is suggesting a “grand conspiracy,” or a conspiracy at all, after all, we are not talking about “credentialed scientists” under the direct control of a company, and we are certainly not talking about the suppression of scientific results, as in the case say...of asbestos producers or big tobacco. No, this study was conducted in the light of day, and the results were reported in a peer-reviewed journal. One would have only hoped that the last sentence of the report’s abstract:
“This warrants a robust epidemiologic study to determine the role of various risk factors in the two production systems that may lead to persistence of bacterial (Salmonella) pathogens and reemergence of parasites (such as Trichinella) of historical significance.”
would be an indication to its readers of what can and cannot be drawn from it.