A REVIEW
OF VACLAV SMIL'S SHOULD WE EAT MEAT?
In
my late twenties I went vegetarian for a year. Some of my friends
were strict non-meat-eaters, and I wanted to try it out. Plus
I was flying a lot for work and found that the airplane meals
made with tomatoes and beans just tasted better than the shoe-leather
beef. In the end, though, I couldn’t keep it going, and
I eventually returned to my carnivorous ways.
Years
later, I came to realize that it was a luxury for me to spurn
meat. In most places, as people earn more money, they want to
eat more meat. Brazil’s per-capita consumption has gone
up fourfold since 1950. China’s nearly doubled in the
1990s. Mexico, Indonesia, South Korea, and Japan have also seen
big increases.
More
countries are sure to follow, and that’s a good thing.
Meat is a great source of high-quality proteins that help children
fully develop mentally and physically. In fact part of our foundation’s
health strategy involves getting more meat, dairy, and eggs
into the diets of children in Africa.
But
there’s also a problem. Raising animals can take a big
toll on the environment. You have to feed the animal far more
calories than you extract when you eat it. It’s especially
problematic as we convert large swaths of land from crops that
feed people to crops that feed cows and pigs. Plus clearing
forests to make more farmland contributes to climate change,
as do the greenhouse gases produced by all those animals.
The
richer the world gets, the more meat it eats; the more meat
it eats, the bigger the threat to the planet. How do we square
this circle?
I can’t
think of anyone better equipped to present a clear-eyed analysis
of this subject than Vaclav Smil. I have written several times
before about how much I admire Smil’s work. When he tackles
a subject, he doesn’t look at just one piece of it. He
examines every angle. Even if I don’t agree with all of
his conclusions, I always learn a lot from reading him.
That
is certainly true of his book Should We Eat Meat?. He starts
by trying to define meat (it’s surprisingly slippery --
do you count kangaroos? crickets?), then explores its role in
human evolution, various countries’ annual consumption
(the United States leads the way with roughly 117 kilograms
of carcass weight per person), and the health and environmental
risks. He also touches on ethical questions about raising animals
for slaughter and covers some simple ways to eliminate the needless
cruelty involved.
As
usual, Smil offers up some intriguing statistics along the way.
A quarter of all ice-free land in the world is used for grazing
livestock. Every year, the average meat-eating American ingests
more than enough blood to fill a soda can. And Americans eat
a lot of pepperoni.
Another
thing Smil loves to do is question the conventional wisdom.
For example, you may have read that raising meat for food requires
a lot of water. This has been in the news lately because of
the drought in California. Estimates vary, but the consensus
is that between watering the animals, cleaning up after them,
and growing crops to feed them (far and away the biggest use),
it takes several thousand liters of water to produce one kilogram
of boneless beef.
But
Smil shows you how the picture is more complicated. It turns
out that not all water is created equal. Nearly 90 percent of
the water needed for livestock production is what’s called
green water, used to grow grass and such. In most places, all
but a tiny fraction of green water comes from rain, and because
most green water eventually evaporates back into the atmosphere,
it’s not really consumed.
As
Smil writes, “the same water molecules that were a part
of producing Midwestern corn to feed pigs in Iowa may help to
grow, just a few hours later, soybeans in Illinois . . . or,
a week later, grass grazed by beef cattle in Wales.” One
study that excluded green water found that it takes just 44
liters -- not thousands -- to produce a kilo of beef. This is
the kind of thing Smil excels at: using facts and analysis to
examine widely held beliefs.
Returning
to the question at hand -- how can we make enough meat without
destroying the planet? -- one solution would be to ask the biggest
carnivores (Americans and others) to cut back, by as much as
half. Although it might be possible to get people in richer
countries to eat less or shift toward less-intensive meats like
chicken, I don’t think it’s realistic to expect
large numbers of people to make drastic reductions. Evolution
turned us into omnivores.
But
there are reasons to be optimistic. For one thing, the world’s
appetite for meat may eventually level off. Consumption has
plateaued and even declined a bit in many rich countries, including
France, Germany and the United States. I also believe that innovation
will improve our ability to produce meat. Cheaper energy and
better crop varieties will drive up agricultural productivity,
especially in Africa, so we won’t have to choose as often
between feeding animals and feeding people.
I’m
also hopeful about the future of meat substitutes. I have invested
in some companies working on this and am impressed with the
results so far. Smil is skeptical that it will have a big impact
-- and it is true that today the best products are sold mostly
in fancy grocery stores -- but I think it has potential.
With
a little moderation and more innovation, I do believe the world
can meet its need for meat.
Publication
courtesy of Bill Gates
Blog