Friday, October 23, 2009

Night of the Living Benthamites

A scientist friend has drawn my attention to an article on the moral uncertainties posed by climate change. In the article, “Anthropogenic climate change: Scientific uncertainties and moral dilemmas” (in Physica D 237;2008: 2132–2138), Rafaela Hillerbrand and Michael Ghil say they know “a way to correctly incorporate all the relevant uncertainties into the decision making process” (2133). Their suggested method is Expected Utility Theory, an offshoot of game theory, decision theory, and economics.

Hillerbrand and Ghil are “utilitarians,” meaning that for them “the morally correct action is one that maximizes overall human welfare” (2134). Expected Utility Theory is the formal, logical expression of that philosophy. Here is Hillerbrand and Ghil’s quasi-mathematical definition of it (brace yerself!):

[single-click image for larger view, then go back a page on browser to return here]


I call this
quasi-mathematical because, for all its Greek symbols and function talk, the idea that people have “preferences” that can be assigned specific numbers is pure whimsy. One-dimensional preferences, as a few seconds of introspection will confirm, don’t even exist. Your desire for a cheeseburger cannot, even approximately, be assigned a unique Ui in the domain of real numbers R: it varies with time of day, how much you have been conditioned by marketing, your desire for healthier food, your desire to indulge in unhealthy food, what you remember of Fast Food Nation or The Omnivore’s Dilemma, a hundred other aspects of self. It is not a quantity at all. So nobody actually computes numbers to plug into equations like this: even economists and utilitarian philosophers aren’t that crazy. All this jargon and notation are gestural, not operable. They are there to lend an air of scientific authority to a philosophical creed first advanced a couple of centuries ago by Jeremy Bentham (1748–1832) and John Stuart Mill (1806–1873). The notational fancy dress adds nothing to the basic belief, which many have challenged and which strikes me as essentially crude. “The needs of the many outweigh the needs of the one,” as Spock intones annoyingly in The Search for Spock (1984).

But one needn’t get into the shortcomings of Utilitarianism to find problems with Hillerbrand and Ghil’s approach. A big one is their assumption that if we spend money to mitigate climate change, we will be making people worse off in the here-and-now for the sake of people yet unborn. “[M]itigation of major changes in future climate,” they say, is a “predominantly altruistic goal”; “Investing in the mitigation of climate-change effects means foregoing other investments” in human welfare (2134). This assumption is basic to Hillerbrand and Ghil’s whole argument, because if fighting climate change does
not require us to rob a present Peter to pay a future Paul then there is no trade-off, no dilemma, and nothing to write papers about. In particular, Expected Utility Theory (EUT) is irrelevant even if valid.

But the assumption is questionable, even by orthodox economic measures. The Intergovernmental Panel on Climate Change (IPCC) found in 2007 that by 2030, “macro-economic costs for multi-gas mitigation, consistent with emissions trajectories towards stabilization between 445 and 710 ppm CO2-eq, are estimated at between a 3% decrease of global GDP [Gross Domestic Product] and a small increase” (2007 IPCC report on Impacts, Adaptation, and Vulnerability,
Summary for Policymakers). “Small increase” means that the world economy might well be
strengthened by mitigation—a negative cost—as market distortions are rectified, waste is cut, smart technologies are deployed, and positive side-effects kick in. For example, “near-term health co-benefits from reduced air pollution as a result of actions to reduce [greenhouse] emissions can be substantial and may offset a substantial fraction of mitigation costs” (ibid.)

To support their claim that climate-change mitigation is inherently dilemmic, Hillerbrand and Ghil cite figures (in a section titled, ironically, “Jumping to Conclusions”) showing that it might cost $450 billion per year, about 1% of global GDP, to mitigate climate change sufficiently to avoid “major hazards,” whereas providing 80% of rural Africa with water and sanitation by 2015 would cost “only US$1.3 billion per annum” (2134). This is their only example of a mitigation “trade-off.”

But as their own use of the word “only” flags, this argument is weird. $1.3 billion is 0.3%—one third of one percent!—of $450 billion. The US spends $1.3 billion on the Iraq War every
two days. [1] If the world were able and willing to pony up $450 billion for what Hillerbrand and Ghil consider the “predominantly altruistic” goal of climate-change mitigation, wouldn’t it probably be able and willing to come up with a few altruistic pence for African health as well? Wouldn’t it have made more rhetorical sense for Hillerbrand and Ghil to pit an expensive here-and-now opportunity against the possible costs of climate mitigation?

But that is the least of this example’s problems. A bigger one is that Africans happen to be more endangered by climate change than almost anybody else—especially in the water department. To quote the IPCC’s 2007 report again,
Africa is one of the most vulnerable continents to climate change and climate variability . . . Climate change will aggravate the water stress currently faced by some countries, while some countries that currently do not experience water stress will become at risk of water stress . . . (Ch. 7 of the Impacts, Adaptation, and Vulnerability report).
And water would only be part of it: “Agricultural production and food security (including access to food) in many African countries and regions are likely to be severely compromised by climate change and climate variability . . .” (ibid.)

So let’s get this straight: Failure to mitigate climate change will screw Africa, big-time. It will undermine, perhaps entirely destroy, any benefits purchased by the $1.3 billion per year that Hillerbrand and Ghil cast fallaciously as an
alternative investment. Climate-change mitigation is a precondition for meaningful, long-term aid to Africa, not an alternative to such aid. The whole idea that mitigation would rob the Africans is thus cockamamie, backwards-ass. In fact, one of the nastier tensions in the mitigation debate these days is that the up-front costs of mitigation (usually assumed to be positive) must be borne by strong economies like those of the US, Europe, Japan, and China, while the benefits are going to be greatest for Africa and the other poorer, “southern” regions of the world that are threatened most by climate change to begin with. Hillerbrand and Ghil fantasize that mitigation would steal well-being from the Africans: in reality, the G-20 are quietly worried that mitigation is a giveaway to the Africans.

Another fundamentally wrong, wrong, wrong thing about the either/or notion of climate mitigation that justifies Hillerbrand and Ghil’s whole exercise is the idea that the harms which climate-change mitigation seeks to prevent are
distant harms, that mitigation is about “the well-being of future generations” (2133). That is only partly true: yes, what we do today will affect human life for many centuries to come, but the years 2050 (which many of us will live to see) and 2100 (which many of our grandchildren will live to see) are the most commonly referenced points in climate and mitigation scenarios. And let’s not forget Africa, of which the IPCC says: “Changes in a variety of ecosystems are already being detected, particularly in southern African ecosystems, at a faster rate than anticipated . . . ” (ibid., emphasis added). So this is about the future starting now, not the future starting after we are all long dead.

Fortunately, the things we need to do to prevent climate change from becoming disastrous are much the same things we need to do to save ourselves from several other catastrophes: ocean acidification, resource exhaustion, soil death, more. There is therefore no moral puzzle at all about mitigating climate change. Urgent altruism aligns with long-term altruism; self-interest aligns with intergenerational interest; climate mitigation aligns with numerous other benefits; long-term sanity may even, for a miracle, align with economics.

But let’s grant for the sake of the argument Hillerbrand and Ghil’s assumption that mitigation is dilemmic, and see how they deal.

I note, first, that on their first page they subtly
minimize the climate threat, stating that climate change will last “for decades to come.” Decades? Try centuries, based on the long residence time of CO2 in the atmosphere. The Intergovernmental Panel on Climate Change stated in 2007 that “anthropogenic climate change will persist for many centuries.”

On the second page (p. 2133), they set up a classic
straw man: “If there is a moral obligation to preserve the climate in its present state, where does it stem from?”

But everyone who knows the science understands that we cannot “preserve the climate in its present state.” That ship has already sailed. The climate has already changed. It is changing all the time and will continue to change. If we were to zero out our greenhouse emissions tomorrow, simply stop emitting altogether, anthropogenic climate change would continue for centuries, albeit at a greatly reduced rate and to a less extreme endpoint. The question is not whether “there is a moral obligation to preserve the climate in its present state”: it is whether there is a moral obligation to do our best to prevent climate from changing uncontrollably and irrevocably to a
much different state.

Hillerbrand and Ghil end their article with the rather nonspecific and uncontroversial suggestion that we give scientific prognoses a major role in decisions about climate mitigation while acknowledging that such prognoses, “taken on their own, give no sufficient reasons for acting or not acting, this way or the other” (2137). Well, no duh!

It is hard to see how we could apply Expected Utility Theory (EUT) to questions of mitigation even if we wanted to:

First, because it depends on a bloodless pseudo-math that will never move millions or billions of people to change their lives. Perhaps Hillerbrand and Ghil assume, autocratically, that all important change will be imposed from the top down by an expertocracy: they seem to, but don’t say this explicitly.

Second, EUT won’t work on climate for the reason Hillerbrand and Ghil acknowledge on p. 2135: “assigning actual likelihood values to expected impacts on human welfare [in EUT] is often difficult or even impossible with the current state of knowledge” (or, I would add, any plausible future state of knowledge). In other words, they would have us adopt a quantitative method without having access to the required quantities. This is better known as “guessing.”

Third, because of EUT’s admitted anthropocentrism (“valuing the environment solely as a basic resource for humanity, as done in the present paper”—p. 2135). Hillerbrand and Ghil exclude any idea that the nonhuman (or at least non-“sentient”) world is sacred and that we have any obligation to steward it. If humans can be fine without giant redwoods, then the redwoods can go. In Hillerbrand and Ghil’s vocabulary, only “environmentalists” will bemoan this outcome very much. Jesus, have these people never gone camping? The idea of valuing the non-human is deep-rooted in hundreds of millions of people, if not billions. How can it “promote overall human welfare,” Hillerbrand and Ghil’s ultimate value, to discard a value cherished deeply by a large portion of humans?

No: it won’t do. There is some hope in moral common sense, philosophically raggedy-ass though it may be: there is no hope in the robotic gameboarding of Expected Utility Theory. Common fear of total disaster, common care for the fate of our descendants, common devotion to the non-human as well as to the human—love, in fact, love for land, for life, for people, for non-people, for children: this is the only hope, unless one assumes (as I think Hillerbrand and Ghil do) that paltering, marginal actions are really all that’s going to be necessary . . .

I see it now. Full circle. The issue-minimizing and straw-manning that Hillerbrand and Ghil deploy early in their piece lower the bar enough so that their pet ethical calculus, the Expected Utility Theory, can hop over it. If the question were only whether we are obliged to make a feint at attempting the impossible (i.e., “preserve climate in its present state”), and that over a scale of mere “decades,” then a morally contorted pseudo-quantitative system operating on admittedly insufficient data would suffice. Because accomplishing next to nothing would really be OK. If you’re selling plastic sporks, you want people to think they’re going to be eating pudding, not steak.

Utilitarianism is a spork.

----------------------------------

[1] As of 2007, the American Friends Service Committee calculated that the US was spending $720 million per day on the Iraq War (
reported in the Washington Post). That’s $1.3 billion every 1.8 days.

----------------------------------

I will e-mail a PDF of the Hillerbrand and Ghil article to anyone who writes me at lnpgilman [ a t ] wildblue [ d o t ] net to request a copy.