Table Of Contents

This Page

10/26/2011: A critic of Gaia by Kirchner

Notes from “The Gaia hypothesis: Fact, theory and wishful thinking” by J. Kirchner (2002)

[I apologize but the paper is so clearly written and right to the point that I have almost quoted the full paper]

“the singular term, ‘the Gaia hypothesis’, was being applied to many different propositions, ranging from ideas that most modern Earth scientists would consider self-evident, to notions that most would consider outlandish. The weak forms of the Gaia hypothesis hold that life collectively has a significant effect on Earth’s environment (‘Influential Gaia’), and that therefore the evolution of life and the evolution of its environment are intertwined, with each affecting the other (‘Coevolutionary Gaia’). I argued that abundant evidence supports these weak forms of Gaia, and that they are part of a venerable intellectual tradition (Spencer, 1844; Huxley, 1877; Hutchinson, 1954; Harvey, 1957; Holland, 1964; Sillen, 1966; Schneider and Londer, 1984). By contrast, the strongest forms of Gaia depart from this tradition, claiming that the biosphere can be modeled as a single giant organism (‘Geophysiological Gaia’) or that life optimizes the physical and chemical environment to best meet the biosphere’s needs (‘Optimizing Gaia’). I argued that the strong forms of Gaia may be useful as metaphors but are unfalsifiable, and therefore misleading, as hypotheses (Kirchner, 1989). Somewhere between the strongest and the weakest forms of Gaia is ‘Homeostatic Gaia’, which holds that atmosphere-biosphere interactions are dominated by negative feedback, and that this feedback helps to stabilize the global environment. I argued that if one defines it carefully enough, Homeostatic Gaia may be testable, but I also pointed out that there was abundant evidence that the biota can also profoundly destabilize the environment.”

“One part of Gaia that is clearly fact is the recognition that Earth’s organisms have a significant effect on the physical and chemical environment. Biogeochemists have devoted decades of painstaking work to tracing the details of these interactions. Many important chemical constituents of the atmosphere and oceans are either biogenic or biologically controlled, and many important fluxes at the Earth’s surface are biologically mediated (see Kirchner, 1989, and references therein). This was well understood among biogeochemists well before Gaia, although Gaia’s proponents have helped to educate a much wider audience about the pervasive influence of organisms on their environment.”

“Another well-established fact that is incorporated in Gaia is the notion that Earth’s organisms and their environment form a coupled system; the biota affect their physical and chemical environment, which in turn shapes their further evolution. In this way, Earth’s environment and life co-evolve through geologic time (Schneider and Londer, 1984). Over the long term, organisms do not simply adapt to a fixed abiotic environment; nor is the environment sculpted to conform to fixed biotic needs (as early versions of Gaia appeared to argue, e.g., Lovelock and Margulis, 1974b). A key theoretical element of Gaia is that, as with any complex coupled system, the atmosphere/biosphere system should be expected to exhibit ‘emergent’ behaviors, that is, ones that could not be predicted from its components alone, considered in isolation from one another. So understanding Earth history and global biogeochemistry requires systems thinking (in addition to lots of reductionist science as well). Gaia’s proponents have helped to promote a systems-analytic approach to the global environment, in parallel with a much larger and broader-based effort in the biogeochemical community as a whole.”

“Up to this point, nothing that I have said about Gaia would seem controversial to modern Earth scientists. [...] At the same time, Gaia’s proponents have consistently held that the Gaia hypothesis means something more than just the co-evolution of climate and life, and something more than just the idea that the Earth system can exhibit interesting system-level behaviors. Coupling between the biosphere and the physical environment can potentially give rise to either negative (stabilizing) feedback, or positive (destabilizing) feedback, and the consequences of this feedback can potentially be either beneficial or detrimental for any given group of organisms. But in the Gaia literature, mechanisms linking organisms to their environment are generally termed ‘Gaian’ only if they create negative feedbacks, and only if they are beneficial to the organisms involved, or to the biota as a whole (Lenton, 1998; Gillon, 2000). Positive feedbacks, or those that seem detrimental, are typically referred to as ‘non- Gaian’ or ‘anti-Gaian’ mechanisms. Thus Gaia’s proponents appear to view the Gaia hypothesis as combining elements of what I have termed ‘Homeostatic Gaia’ (i.e., biologically mediated feedbacks stabilize the global environment) and a qualified form of ‘Optimizing Gaia’ (i.e., biological modifications of the environment make it more suitable for life).”

“This is what makes Gaia interesting, but this is also what makes Gaia difficult. If Gaia meant only that organisms influence their environment, and that these interactions may give rise to interesting system-level behaviors, then Gaia would add little – apart from different language and different metaphors – to the general consensus of the biogeochemical community. By claiming that organisms stabilize the global environment and make it more suitable for life, Gaia’s proponents advance a much more ambitious argument, but one that is less clearly consistent with the available data, and one that sometimes may be difficult to test against data at all.”

After providing an interesting list of positive and negative feedbacks considered the most important to the climate, Kirchner concludes that

“This list of feedbacks is not comprehensive, but I think it is sufficient to cast considerable doubt on the notion that biologically mediated feedbacks are necessarily (or even typically) stabilizing.”

“Most of the work to date on biological climate feedbacks has focused on terrestrial ecosystems and soils; less is known about potential biological feedbacks in the oceans. One outgrowth of the Gaia hypothesis has been the suggestion that oceanic phytoplankton might serve as a planetary thermostat by producing dimethyl sulfide (DMS), a precursor for cloud condensation nuclei, in response to warming (Charlson et al., 1987). Contrary to this hypothesis, paleoclimate data now indicate that to the extent that there is such a marine biological thermostat, it is hooked up backwards, making the planet colder when it is cold and warmer when it is warm (Legrand et al., 1988; Kirchner, 1990; Legrand et al., 1991). It now appears that DMS production in the Southern Ocean may be controlled by atmospheric dust, which supplies iron, a limiting nutrient (Watson et al., 2000). The Antarctic ice core record is consistent with this view, showing greater deposition of atmospheric dust during glacial periods, along with higher levels of DMS proxy compounds, lower concentrations of CO2 and CH4, and lower temperatures (Figure 1). Watson and Liss (1998) conclude, ‘It therefore seems very likely that, both with respect to CO2 and DMS interactions, the marine biological changes which occurred across the last glacial-interglacial transition were both positive feedbacks’.”

About the “Optimizing Gaia”:

“it is one thing to say that we benefit from the environmental services that our ecosystem provides, and entirely another to say that our environment is in any sense tailored to our needs.”

“Organisms can strongly influence their environments. Organisms are also naturally selected to do well in their environments â which means doing well under the conditions that they, and their co-occurring species, have created. Thus it is likely that those particular organisms will be better off under those particular conditions than they would be without the environmental services that they, and their co-occurring species, are responsible for. In any such system, it will be true that ‘biotic effects strongly enhance the biogeophysical and biogeochemical conditions for life’ (Kleidon, this issue), precisely because the particular life in question is that which has been naturally selected to thrive under the conditions that those particular biotic effects promote.”

This is equivalent to the ‘anthropomorphic principle’. The constants of the Universe are such that if one of them has even slightly a different value, the Universe, as we know it, would not exist. Some then wonders how come the Universe is ‘built’ exactly so that it exists. Is this the signature of a super-power or God? Not necessarily because, due to natural selection of Universes, we could not be inside a Universe that cannot exist. Probably a better illustration comes from natural selection of species itself. A species that fits its environment is not due to a super-power that created it as such but is due to the natural elimination of species that could not survive the hardship of the environment.

“The Gaia literature further compounds this semantic confusion by making claims such as, ‘vegetation almost always influences climate for its own benefit’ (Lenton, 1998, emphasis added), rather than to its own benefit. What a difference a word makes! Saying that vegetation influences climate to its own benefit implies that vegetation alters its environment in ways that are beneficial. But saying that vegetation influences climate for its own benefit advances the much stronger claim that vegetation modifies its environment in order to reap the benefits that will result from doing so.”

Say differently, there may have been in the past and even present species that are altering their environment in ways that it won’t beneficial. So much so that the species had or may disappear. Species that are associated with a positive feedback are among them.

“The life forms that we observe today are descended from a very select subset of evolutionary lineages, namely those for which Earth’s conditions have been favorable. The other lineages, for whom Earth’s conditions are hostile, have either gone extinct or are found in refugia (such as anaerobic sediments) which protect them from the conditions that prevail elsewhere. As Holland (1984) has put it, ‘We live on an Earth that is the best of all possible worlds only for those who are well adapted to its current state’.”

About Gaia and natural selection:

“What Daisyworld actually demonstrates is that it is possible for such a system to arise by natural selection, but only given a very specific assumption embedded in the model. Because this assumption may not be true on the real Earth, Daisyworld only demonstrates a theoretical possibility rather than a guiding principle of the natural world.”

“Thus traits that confer a differential advantage (for carriers over non-carriers) will become more common via natural selection, but traits that confer a general benefit (to carriers and non-carriers alike) will undergo genetic drift; natural selection cannot have any effect on them.”

“Here, then, is the crux of the matter. The environment is that which is shared among organisms. To the extent that a trait improves the environment for life in general – and thus benefits its carriers and non-carriers alike – natural selection will not have any effect on it. Thus claims that ‘life-enhancing effects would be favored by natural selection’ (Kleidon, 2002) are not generally valid.”

“Thus environmentally beneficial traits and environmentally detrimental ones will both be favored by natural selection, as long as each confers a differential reproductive advantage to the individuals that carry it.”

“The Daisyworld model assumes that traits that benefit the environment also give an individual a reproductive advantage over its neighbors. Thus Daisyworld gives the impression that traits are favored by natural selection if they are environmentally beneficial, even though natural selection does not – indeed cannot – act on environmental benefits per se, precisely because they are shared between carriers and non-carriers alike. By assuming that individual reproductive success and environmental good deeds are linked, and thus that organisms will do well by doing good, Daisyworld to some extent assumes what it sets out to prove.”

“Traits that enable organisms to better consume resources or eliminate wastes will benefit the individual, and thus will be favored by natural selection, even though they also degrade the environment. Examples of such traits abound.”

“Natural selection is a mechanism, not a principle. It does not seek a goal; it just passes traits from one generation to the next, with the reproductively successful ones becoming more common over time. Thus there is no direction to evolution beyond the fact that whatever works (in a reproductive sense) works, and will be passed on to the next generation. Natural selection will favor both environment-enhancing and environment-degrading traits, as long as those traits confer a reproductive advantage, that is, as long as those who carry them have greater reproductive success than those who don’t.”

“if the connection between environmental good deeds and individual reproductive advantage is only coincidental, we should not expect Gaian traits to evolve any more frequently than anti-Gaian ones.”

It is still theoretically possible for evolution to systematically favor Gaian traits over anti-Gaian ones, but only if environments and their organisms are jointly subject to some form of natural selection, in competition with other environment/ organism assemblages. This would create a kind of metapopulation, and natural selection in metapopulations can produce evolutionary outcomes that would otherwise seem counterintuitive (e.g., Kirchner and Roy, 1999). However, a mechanism needs to be demonstrated by which such natural selection could occur in this context.”

I agree with this. If a species has a trait that gives it an advantage over other species, yet, alter negatively their environment, the species might disappear and may be replaced over time by a species that again a trait that gives it an advantage over species and that alter the environment in beneficial ways. Why does he say a mechanism needs to be found? It does not seem to difficult to come up with an illustrating mechanism, is not it?

“organisms that consume plants and respire their carbon could not multiply beyond the point at which they either run out of food or poison themselves with waste CO2. But these are far from the environment-enhancing feedbacks envisioned by the Gaia hypothesis.”

Is it?

I also agree with the following:

“One should remember that environment-enhancing feedbacks, if they occur, are intrinsically destabilizing (Kirchner, 1989). Organisms that make their environment more suitable for themselves will grow, and thus affect their environment still more, and thus grow still further. This is positive feedback, not negative feedback. Negative feedback arises when a growing population makes its environment less suitable for itself, and thus limits its growth. Environment-enhancing feedbacks are destabilizing; environment-degrading feedbacks are stabilizing. The Gaian notion of environment-enhancing negative feedbacks is, from the standpoint of control theory, a contradiction in terms.”

“Organisms are not merely passengers, riding passively on spaceship Earth. But to the extent that the biota are piloting the craft, they are flying blind, and the various life forms are probably wrestling over the controls.”

“anthropogenic emissions have pushed greenhouse gas concentrations far beyond the limits that they had previously remained within, for over 400,000 years. This indicates that the composition of the atmosphere is not tightly regulated, by either biotic or abiotic feedbacks, on human timescales. Yet both CO2 and methane are biologically active gases. And at least in the case of CO2, the anthropogenic fluxes are a tiny fraction of the gross fluxes entering and leaving the biosphere; only a small adjustment to those biological fluxes would have been needed to keep CO2 concentrations stable. There has been a modest increase in terrestrial photosynthetic uptake of CO2 (Ciais et al., 1995; Keeling et al., 1996b; Myneni et al., 1997), but not nearly enough to keep CO2 concentrations within their natural limits. This is an empirical rebuttal to Gaian notions of homeostasis and optimization, as it indicates that atmospheric CO2 is not tightly regulated at a biological set point.”

“In the human enterprise of science, our most daunting task is to see things as they are, rather than as we wish they were. Gaiaâs vision of Earth as a harmonious whole, engineered by and for the organisms that live on it, is a deeply evocative notion. It is emotionally very appealing to me. But I suspect that compared to the Gaian vision of global harmony, the actual Earth system â as it comes into clearer focus â will prove to be more complicated, more intriguing, and perhaps more challenging to our notions of the way things should be. Understanding the Earth system, in all of its fascinating complexity, is the most important scientific adventure of our time. We should get on with it, as free as possible from our preconceptions of the way the world ought to work.”

I globally like the paper. The only thing that needs to be further explored is the notion that natural selection could nonetheless favor Gaian-like traits. Kirchner seems to dismiss quickly this possibility and I do not understand why. He does, however, nicely debunk the Daisyworl model.