The Till-Kill Argument

Yesterday my animal rights class discussed the "Till-Kill Argument"--an argument due to animal scientist Steven Davis (made famous when Michael Pollan appealed to it in this NYT article).  The argument is that if we want to do the least harm to animals, we're better off with a "veg & beef" diet than a vegan diet.  Here's his math:

120 million ha is the total amount of cropland in the US.  That's about 19% of US land (another 25% is animal pasture). Suppose (just suppose) that all that land were used to grow an all-plant human diet.  There would still be many animals killed, because of "till kill."  That's like road kill, except the animals--voles, rabbits, mice, ground birds and the like--are killed by farm machinery.  Davis estimates a death toll of 15 per ha.

Now suppose half that land is used for plant farming, and the other half for pasturing beef cattle.  The more tilling required for a crop, the more animals killed.  Cattle farmers plant low-till forage (Davis says), so there are fewer passes over the land.  That means, he says, a death toll of just 7.5 per ha.

Of course, you've got to kill the cattle too. He assumes 60 million ha worth of plant food is equivalent to 74 million cattle worth of beef--double the number of beef cattle now killed yearly in the US.

Bottom line:  fewer animals killed in the "veg & beef" scenario.  QED.

Note:  he's not saying this would be the case no matter what animal species is eaten.  If chickens rather than cattle are "grown," and we double the number of chickens currently eaten in the US every year, that would be 16 billion chickens.  The death toll of the vegan scenario is way lower than that.  This is a beef-specific argument.  If we want to minimize killing, "veg and beef" is the way to go...says Davis


What should be said about this argument?  Here's how a debate might go:

Pro-vegan:  The killing in the vegan scenario is all accidental.  It's foreseen, but not intended.  In the veg & beef scenario, the killing is intentional, so worse.  And don't think that's a merely scholastic distinction.  We all think people who deliberately kill squirrels in their backyards are doing something worse than people who accidentally run over them in their cars.

Pro-omnivore: Even if that's so (and it's debatable), there are trade-offs.  If one intentional killing would prevent a billion accidental killings, we'd all think it was a "necessary evil."  74 million intentional cattle-killings are worth it to prevent a half billion accidental till-kills.

Pro-vegan:  These numbers are implausible.  Plant farming till-kill may now be lower than 15 per ha, since new genetically engineered crops are low-till, compared to traditional crops.  Plus, the argument doesn't take into account other ways that livestock crowd out wildlife, because they compete with them for resources, destroy river banks, interfere with migration (because of fencing), and so on.  There's more death on the beef side than Davis allows.

Pro-omnivore:  The point is that vegans just assume that plant farming doesn't kill animals. But it does--a lot!  At the very least, it's an open question what kind of agriculture kills the fewest animals.


What next?  Are there better, stronger arguments available on one side or the other?  My class will continue discussing the article on Monday, so it would be great to have some input from the smart people who comment at this blog!


Adam said...

Here are a couple of responses to Davis's argument:
http://www.springerlink.com/index/v0726k81713341m1.pdf (Gaverick Matheny)
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1253172 (Andy Lamey)

One point is that the data just aren't that good. Notably, you'd need to use a lot less land for growing only vegetable crops, but Davis assumes that we just convert all of our agricultural land to cropland.

The philosophical questions are interesting, though, even if accurate data were to render them irrelevant. (There are some philosophical objections raised in those articles.)

Jean Kazez said...

Thanks a lot for that. I will follow up those links.

As far as "less land" goes.... I've been trying to figure that out. We can start with Davis's 120 million ha figure. Clearly, that will feed some % of the US population. It doesn't matter what that % is.

So suppose we use only half of that for veg. And we make up the difference by also "growing" beef. How much pasture will it take? He assumes 60 million ha will do. But maybe it will take more land, so there will be more till-kill.

A lot of the standard data on this stuff (in Singer, for example) is about the land it takes to grow grain for animals, not the land involved in pasturing cattle (as I recall). So I have to dig a little deeper to figure this out...

Maybe the links will help.

Jean Kazez said...

Oh good, I found the pdfs--



Faust said...

My two cents ammounts to:

1. The data is very important given that the argument is making empirical claims and it sounds like you're getting that covered


2. The two philosophical points of interest would seem to largely revolve around the question of intentional vs. non-intentional killing, and then utilitarian questions about the kinds of suffering generated by the different forms of killing.

Example: if we really could develop "humane" slaughterhouses (hotly debated obviously) then you might have a condition where the incidently killed animals are killed under conditions where there is more suffering.

Aeolus said...

I see Adam has beaten me to the punch, but, yes, it's important to tell your class of the responses by Matheny and Lamey. And be sure to have your students check out this "Animal Visuals" page:
(While you're there, click on the "Slaughter" and "Harvest" headings.)

So it seems that Davis's figures don't stand up. But what if they did? The issue of intentional versus accidental/foreseen killing is still worth pondering.

s. wallerstein said...

I haven't had time to read the links, but it seems that the calculations above don't take into account that the killing produced by planting crops will diminish with each crop, since wild animals will either be killed (and not reproduce) or seek a more "friendly" environment, while the killing of cows will be constant with time.

s. wallerstein said...

Jean: I'm Amos, but I've had problems with blogger and had to change my Google account number.

Jean Kazez said...

Faust, Yes, it seems like intentionalness is not decisive. If you have to choose between humane intentional killing and inhumane accidental killing, it's not obvious the former is preferable. Death by farm machinery--it's probably not a good way to go.

Aeolus--Matheny seems to slaughter Davis. I'm reading the other thing to, and will follow the link. I like/need good visuals! By the way, I was tempted to brag about knowing the "A. Taylor" at the end of Davis's article:-)

S-Amos, If there were gobs of mice and rabbits, they'd be eating the crops. I think the second article gets into what the number really is.

Aeolus said...

Here's something else: Davis claims to be arguing on the basis of a "Least Harm Principle" that he attributes to Tom Regan. On page 388 of his article, Davis refers to, without quoting directly, an e-mail he received from Regan, and then goes on to cite this passage from The Case for Animal Rights, which he says Regan himself calls the "minimize harm principle":
"Whenever we find ourselves in a situation where all the options at hand will produce some harm to those who are innocent, we must choose that option that will result in the least total sum of harm."

This passage can be found on page 302 of The Case for Animal Rights. But here Regan does not accept the minimize-harm principle: he REJECTS it because it is a consequentialist principle that aggregates harm done to different individuals -- and thus clashes with the ascription of intrinsic value to subjects-of-a-life. What Davis's argument is really aimed at would seem to be the implications of Regan's Miniride Principle.

Jean Kazez said...

Yes, and in fact there's no reason to assume the Miniride Principle will kick in here. (1) The Miniride Principle covers situations where all possible victims stand to lose the same thing. In the situation we're looking at, that's not necessarily so. The animals have different types of lives and different lifespans. (2) Regan introduces the Miniride Principle at a point where he's discussing special situations in which rights are overridden. There's no reason to assume this is one of those special situations.

Adam said...

Another objection to the data, which I don't think Matheny or Lamey raise, is that some (half, if I recall) of Davis's data is based on sugarcane. However, there's no reason that a vegetarian diet should include any more sugarcane than a meat-eater's diet. Nor is there any reason to believe that sugarcane is representative of vegetarian crops. (It's been a while since I've read the articles, but I think Lamey provides some reasons why it isn't.) Thus, I don't see why data for sugarcane is at all useful.

Deepak Shetty said...

If one intentional killing would prevent a billion accidental killings, we'd all think it was a "necessary evil."
not true in principle. We wouldn't kill a healthy human and harvest his organs to save 5 people. It wouldn't be accepted as a necessary evil so a pure number matching doesn't work.