This past Wednesday I participated in an Intelligence Squared US debate, "Don't Eat Anything With a Face." Dr. Neal Barnard and Gene Baur argued for the motion, while Joel Salatin and I argued against it. If you missed it, you can watch the video here. It was a great experience, from which I learned a lot. I will be publishing a series of reflections on the debate, the first installment of which is below.
It was an honor to share the stage with my partner Joel, a prominent and brilliant proponent of pasture-based farming and a leader in such farming himself, with John Donvan, an excellent moderator, and with our worthy opponents, Neal and Gene.
I was able to chat with both of our opponents outside the debate and found both of them friendly and personable. I have read most of Gene's book, Farm Sanctuary, which I will be reviewing soon as a guest post on LetThemEatMeat.Com, and had the opportunity to have a lengthy discussion with him later in the night after the debate. Gene and I seem to agree on a lot more than we disagree on when it comes to our relationship with animals, the appropriateness of widespread veganism being the major exception. I genuinely appreciate Neal's commitment to putting the food fork over the surgical knife, emphasizing the importance of nutrition in the prevention and reversal of disease, but he and I have strong disagreements about not only the proper interpretation of scientific research but even what the basic facts are. I will explore those disagreements in further installments of this series.
Here, I'd like to explore what I think Joel and I did well, what we could have improved, and why I think we lost the debate according to the audience vote.
As Rhys Southan pointed out, we were fighting an uphill battle to begin with. I would articulate the primary reasons as follows.
First, although the mainstream nutritional and medical establishments have not embraced veganism, many of the nutritional ideas that proponents can rally in support of it are simply extreme extrapolations of ideas firmly rooted in our nation's nutritional consciousness for the past sixty years. My strongest points about the nutritional value of animal products are unfamiliar and counter-intuitive in this context: when people think of vitamin A, they think of carrots, not butter; young people don't remember the era of cod liver oil and most older people have no idea why they took it as children or that it was the modern obsession with antibiotics, now realized to be wrongheaded, that led to its demise; people think of cholesterol as a cause of disease and not as a potential nutrient; my allusion to nutrients in bone probably confused people if they accomplished anything since I never explained how delicious a soup traditionally made from bone stock can be.
Perhaps most important of all, however, we were fighting an uphill battle on the ethical front. Compassion is rightly compelling. To argue that growing plants on a large scale kills animals or that intentionally increasing the proportion of wild animals and thus the numbers that will die even worse deaths than they die at the hands of humans can appeal to our logic, but the fact that this complicity is less direct means, rightly or wrongly, that it confronts our emotions in a far less powerful way. To argue that killing animals is an ecological necessity appeals to our logic but if it has any emotional impact it is probably more to induce a sense of ecological guilt than anything else.
The battle was uphill for other reasons as well, however, reasons more related to circumstance than the intrinsic dynamics of such a debate. Intelligence Squared declares the team that has the greatest net positive shift in the proportion of the audience that agrees with its position to be the winner. If team A shifts from 90 to 91% and team B shifts from 5 to 7%, team B wins. This is an excellent way of accounting for the baseline bias of the audience and makes the result based on the ability to persuade, especially to swing undecideds. But it also means that the team starting out with the smallest amount of audience agreement has much more room to improve, and therefore a better chance of winning. We started out with 51% of the vote, while the other side started out with 24%, leaving that side with much more room to improve.
One could make a rough analogy between this and the statistical concept of regression to the mean. If you test a completely ineffective drug against a placebo and look at its ability to, say, lower cholesterol, and the baseline cholesterol levels are higher in the treatment group than the placebo group, the likelihood they will either drop in the treatment group or rise in the placebo group is very high. This is why statisticians often recommend comparing the post-treatment values in each group rather than the change each group experienced (I've explained this concept in much more detail here.) The analogy is only a rough one, and persuasion is quite a different process than most pharmacological phenomena, but the high baseline agreement probably hampered us, especially if many of the undecideds were people who came because they were curious about veganism and looking to be convinced.
Given this uphill battle, I think Joel and I did a great job, but I have also learned a lot from this debate. One of my most crucial lessons has been that the skill set required for public speaking in a seminar or lecture format and the skill set required for debate and critical analysis in a written format do not add up to the skill set required for a live debate.
I had some sense of how precious time would be as I was preparing, but I appreciate this concept much more now. Numerous people told me, correctly, that my opening and closing remarks appeared too rehearsed and too read. Ordinarily, I never read anything when I speak, but until this debate I had never had strict time limits of such short durations -- seven minutes, two minutes -- so I focused my preparation on fitting my remarks into these limits and chose to write out my remarks in detail beforehand. In retrospect, I could have and should have done this in other ways that amounted to a more casual and conversational tone.
If I were to do this all over again, moreover, I would make my opening remarks very different. I chose to tell my personal story as a way of weaving key evidence-based arguments into an illustrative and relational narrative. Now that I realize how difficult it is to make a coherent scientific argument in a minute or a minute and a half, however, and how difficult it is in a debate with such a broad focus to make sure all the most important points get discussion time, I realize how valuable those uninterrupted seven minutes really are. I could have taken half of them to refute the opposition's strongest health-related points -- that observational evidence shows the benefits of vegetarianism and the harmfulness of meat, and that intervention studies such as Dean Ornish's show that meat-free diets reverse disease -- and taken the other half to present my scientific arguments about traditional disease-free populations all having animal products in their diets, animal products providing important nutrition, and strong individual variation in the ability to harness nutrients from plant foods.
The preciousness of time hit me in two other areas: making the strongest points first, and asserting more influence over the topics being discussed. Quite frequently, I would have two points to make and lead with the weakest one first so I could end with the strongest. I now realize you should always lead with the strongest points in a debate like this, because the time spent on making the weaker point may be just enough time for the other side to interrupt and either take a stab at the weaker point or simply change the subject. For example, I really wanted to discuss the confounders in observational studies looking at vegetarianism and meat-eating, and the fact that large studies attempting to adjust for these confounders suggest these effects are illusions, since this was one of Neal's strongest points, but I instead chose to address Neal's weaker point that not all smokers get cancer by pointing out there are meat-eating populations completely free of cancer. The time I envisioned having to address the epidemiological studies vanished before my eyes.
I also feel that I stayed far too faithful to the immediate points being made when I should have been more savvy in using opportunities to speak to move the discussion onto topics I considered most important, or interrupted other people at appropriate times to steer the conversation in a more desirable direction. The other three debaters were adept at this, and in some cases I think Neal and Gene went overboard by steering the conversation back to factory farming, but it is a skill that can be used very effectively in a debate like this and one I definitely need to develop and utilize better.
This again crippled my ability to fit in critical points about the observational studies or Dean Ornish's intervention trial. When John first asked me to respond to Neal on the basic science, I should have taken the point back to the basic science in his opening remarks, which would have allowed me to bring these issues up, but I instead responded directly to the points Neal had made immediately before that.
When Gene said that the only nutrient vegans need to worry about is B12, I should have interrupted him to ask, "Where is someone who is genetically unable to convert beta-carotene to vitamin A supposed to get vitamin A from?" Instead I let the conversation become dominated by B12 and let the point I made about vitamin A in my opening remarks completely fade from memory.
When I responded about the China Study, I should have taken the opportunity to talk about Campbell's animal experiments. Instead I responded directly to Gene's immediate comment about the China Project itself, but it was a weak point because you can't prove a negative, and you can't explain why the convoluted statistical arguments in the book are wrong without tediously parsing them. Had I pointed out that Campbell showed in his experimental research that animal and plant proteins act identically in their promotion of cancer, that high-protein diets protect against the initiation of cancer, and that in the most realistic animal experiments the only reason the low-protein animals don't get cancer is because they just get sick and die instead, this would have been a stronger point. Perhaps I could have concluded by quoting Campbell's 1972 paper that the changes experienced by his low-protein animals were "similar to the retardation of brain cell growth of young malnourished animals" for a nice sound bite.
There were a couple places where I should have chosen my words more carefully. During the discussion of B12, I said I wanted to bring the conversation "back down to earth," by which I meant that I wanted to move it from theoretical abstractions to the real, on-the-ground, practically relevant data, but it seemed to come across as an insult to the value of the discussion up to that point. It also seems that saying the efficacy of Neal's diet for weight loss was "mediocre" and "run-of-the-mill" may have influenced John's perception that it was a personal attack, even though I had genuinely lauded Neal for the good design and reporting of his studies and was making a point about the efficacy of the diet shown in his data, and trying to make a point that the positive effects are due mostly to weight loss, and perhaps in some cases to increases in the intakes of certain nutrients, and not to veganism per se. Perhaps if I had chosen a few words more adeptly I could have completed that point.
Overall, I think my strongest nutritional argument was for biochemical individuality. Anecdotally, some of the vegans thought this was my strongest point but that I never developed it very far. I think I could have made a much stronger nutritional case if I had steered the conversation back to this point more assertively and really hammered it home.
My second strongest argument would have been my analysis of the epidemiology of meat and vegetarianism and of intervention trials that supposedly support vegetarianism or veganism, such as Ornish's, but I missed making those entirely until I hastily rushed them into my closing statement.
I think Joel's strongest ethical and ecological point was that animals are necessary components of an ecologically sustainable agricultural system, and that each animal has an ecological niche it occupies, with most non-human animals becoming prey to another species as part of that niche. An ethic that violates the basic ecology can't offer a coherent and sustainable alternative. Joel and Gene had a very brief discussion about veganic farming, and I wish they had fleshed the issue out in more depth. I think Joel's point that plants have forms of communication and response was thought-provoking, but could have been delivered more effectively if he had stressed that communication is a continuum, without suggesting actual and absolute equality in the morality of killing something on that continuum regardless of what point it occupies.
Joel and I both made the point that producing plants leads to animal death. As Rhys pointed out, I made a quantitative claim that more animals die for the production of plants, and this is unclear. I don't think it hurt us because the other side didn't dispute it with numbers, but I should have just left it as a qualitative point: raising plants kills animals. Joel could have made his point more effectively if he had left the billions of bacteria out of it and focused on animals, especially small mammals that people have more sympathy for. I think both of us could have made these points more effectively if we had prepared a specific argument addressing whether and to what extent animals die during production of plants outside of industrial monocropping, and whether six to seven billion people can be fed a vegan diet without any industrial monocropping. When Gene said he didn't support industrial plant agriculture, I think Joel and I should have assertively pressed him to articulate how he would feed the world on his veganic farming model. Similarly, I wish that the other side had put less effort into going back to CAFOs and had instead expended that effort pressing Joel and me to articulate how our model could feed a large and densely populated world.
On a similar note, I think I could have benefited our side with appropriate and limited use of snappy one-liners interpreting the other side's statements. Joel did a great job with lines like, "So the thing is, don't wash your kale." When I asked Gene if his vision of a post-animal farming world involved the extinction of all domesticated farm animals and he alluded to American cows being replaced by buffalo, I should have interjected to clarify whether he meant American breeds of cows should all go extinct, because I think that point may have slipped through the cracks. If the one-liner brought the floor back to me, as it did for Joel when he made the kale comment, I could have asked him whether people on other continents should be vegans, and if so, what happens to the cow on a global scale. Extinction?
Of course, no post-mortem analysis of a lost debate or even a won debate will ever be free of critcism. A live, on-the-spot debate where you have to think on your feet at every moment precludes "perfect" performance.
This was my first ever debating experience and I am incredibly grateful to Intelligence Squared for the honor to participate in the debate and to learn these important lessons.
In future installments in this series, I will cross-post my review of Gene's book and explore some of my interpretive and factual disagreements with Neal in more depth.
Your thoughts? Please share them in the comments!