Bartlett starts out on decent ground, describing exponential growth and reminding us how easy it is to misunderstand. And he's right that if the Earth's population grew to one person per square meter of land (which would happen in 2780 if we maintained a 1.3% annual population growth rate), it would be a bad thing. But then his argument goes downhill.
Bartlett notes that the world's population growth rate has fallen from 1.7 in 1986 to 1.3 in 2000. This is not just two isolated data points -- it's part of a larger historical trend, in which the world's population growth rate peaked in the 1970s and has been falling ever since. The most industrialized countries are increasingly facing sub-replacement fertility rates (replacement fertility is 2.1 children per woman, which will maintain the population size if there is no migration). There would have to be a significant change in the world's current trajectory for us to approach Bartlett's one-person-per-square-meter future. Demographers generally predict that the world's population will stabilize between 9 and 10 billion by 2050-2100 (that's .07 people per square kilometer). So runaway population growth is probably not on the table, and Bartlett's invocation of a simple exponential growth rate misses the mark.
That being said, it's quite possible that the growth rate still isn't falling fast enough, and that 9-10 billion people will be too many. Bartlett attempts to offer evidence that we're already at or near the overpopulation threshold, but his evidence fails. His first argument is that the US imports 60% of its oil, 15% of its gas, and 20% of its food. But import levels alone (particularly for globally traded resources like oil and food, somewhat less so for more localized ones like water and electricity) tell us nothing about sustainability, except under the absurd premise that the people living in any arbitrarily defined segment of land ought to be able to supply all of their needs from that same segment of land. By this logic my apartment is inherently unsustainable, since I have to import 100% of my food from outside of the building. The reason the US imports so much oil is that oil is unevenly distributed -- there's lots of it in Canada, Venezuela, and Saudi Arabia, but not so much in the US anymore. It would be quite odd to say that Barack Obama should buy a Prius because the US has little domestic oil, whereas it's fine for Hugo Chavez to drive an SUV because his country has lots of oil within its borders. It's unfortunate that Bartlett chose such a silly way to try to make his point, since limits to oil supply are a significant issue.
Bartlett also asserts that if any amount of climate change is anthropogenic, that proves the world is overpopulated. This too does not follow, despite the fact that it gestures toward a serious issue. In passing, Bartlett adds the clause "living as we do" to his statement about overpopulation. But that's a critical issue. Overuse of the Earth's resources and sinks is a function of both population and the per capita use implied by our lifestyle. Considering the enormous variation in per capita carbon emissions, it's rather silly to elide the lifestyle side and place the blame squarely on the population side -- especially when population growth rates are falling but carbon emission per capita is rising.
Let's grant, though, that population growth rates aren't falling fast enough. What can we do? According to Bartlett, "bad things." He lists a series of "good things" that unfortunately also increase population: "large families, medicine, public health, sanitation, peace, law and order, and accident prevention." Then there are the bad things that would inhibit population growth: "small families, violence, stopping immigration, and pollution."
Bartlett asserts that we're going to have to pick from the "bad" list, or nature will pick for us. However, he isn't bold and politically incorrect enough to tell us which bad thing he personally advocates (indeed, his whole article reads as an awareness-raising exercise, despite casting doubt on the effectiveness of education as a solution). The only specific "solution" he mentions is his contention that AIDS should be seen as a case of nature choosing for us. It strikes me as a highly dubious proposition that population growth itself caused the AIDS epidemic, much less that AIDS can be understood as some sort of inevitable correction by nature for our overpopulation. Certainly crowding can, ceteris paribus, promote disease epidemics. But there is much more at play than simple population density in determining the course of epidemics. And if we look at the global scale distribution of population and HIV infection rates, it certainly doesn't look like AIDS is preferentially attacking the densest parts of the world -- or even the densest parts of Africa.
In any event, Bartlett is simply wrong to say that only "bad" things will slow population growth. In fact, the most effective means to slowing population growth is a very good thing: increasing women's rights. Put simply, when women have the ability to make their own choices about pregnancy, and have access to opportunities in life beyond "barefoot and pregnant in the kitchen," they will on average choose to have far fewer children than they do under more severe forms of patriarchy. Equitable economic improvements for people are another good thing that can reduce population growth rates, since people who are economically secure tend to focus their energies on a smaller family (directly contrary to Malthus's prediction that if you let the poor earn a better wage they would just blow it all on cranking out babies). It certainly makes you more politically incorrect to warn that fixing overpopulation will require "bad" things, and there is a history of overpopulation fears being used to justify imposing bad things (like forced sterilization and Hardin-esque "lifeboat ethics") on the poor and people of color. But luckily we can deal with population through good things.
The essay ends with an absurd rhetorical question: "can you think of any problem on any scale, from microscopic to global, whose long-term solution is in any demonstrable way aided, assisted, or advanced by having larger populations?" This is easy -- any problem that would benefit from more brainpower. Think of the great technological innovations and amazing novels that we'd have with 148 trillion people (that one-person-per-square-meter prediction) working on it. Further, anyone with an unusual fetish could find a big community of like-minded individuals in that world, rather than feeling like an isolated freak. Bartlett speaks as if it's absurd to imagine increasing the population of Colgate -- but any alum of a big school like ASU could rattle off a long list of cool things that were available to them that Colgate doesn't have a big enough student body to support. The point here is not that I necessarily support either of those scenarios, but just that saying growth is always entirely bad is a silly oversimplification.