Time for a revision? Maureen O’Malley and Russell Powell on Major Transitions, part 2

One of the cool things about studying the so-called major transitions is that they are as interesting to philosophers of science as to biologists. So you really can’t help being exposed to the philosophy of science literature, and many (maybe most) biologists in the field cross the lines at least occasionally. I’ve been to both, and I’m here to tell you that philosophy conferences are more fun than biology conferences.

Last time, I briefly summarized the various forms of the major transitions framework and addressed one of O’Malley and Powell‘s criticisms, that the framework is progressivist. Now I’d like to look at their other two problems: lack of unity and missing events. By and large, I agree with these points, although there are some caveats I’d like to point out. Next time, I’ll consider their proposed solution, which I’m afraid I don’t find helpful.

Disunity is actually O’Malley and Powell’s first criticism, but it will be a bit more complicated than progressivism to address, and I was short on time on part 1. Essentially, they are arguing that the major transitions are not a natural kind, philosophese for groupings that belong together because of some fundamental commonality, as opposed to more arbitrary groupings whose members are only superficially similar. So what are the transitions? Here’s the list from the book:

Table 1.2 from Maynard Smith J, Szathmáry E (1995) The Major Transitions in Evolution. Oxford University Press, Oxford.

Table 1.2 from Maynard Smith J, Szathmáry E (1995) The Major Transitions in Evolution. Oxford University Press, Oxford.

Maynard Smith and Szathmáry recognized that their list of transitions include very different events, but

The justification for discussing such an apparently diverse series of changes in a single book is that they have features in common. The most important of these is that entities that were capable of independent replication before the transition can replicate only as part of a larger whole after it.

Additional features shared by at least some of the transitions include division of labor:

Every organism has many components — molecules, cells, segments, organs, and so on. In the evolutionary history of these components, an initially identical set of objects often becomes differentiated and functionally specialized. A division of labour can also occur between organisms within a population.

and “New ways of transmitting information,” such as the shift from RNA-based to DNA-based heredity and the emergence of human language.

As O’Malley and Powell point out, the transitions that best meet the “new ways of transmitting information” criterion — RNA to DNA and the emergence of language — meet the “replication only as part of a larger whole criterion” poorly if at all. In addition, the major transitions framework

…lumps historically contingent events with trends that are likely to occur across alternative plays of the ‘tape of life’…

The origin of eukaryotes, for example, is apparently a unique event, while multicellularity has evolved many times. The first of these criticisms is, I think, a real problem: the criterion of replicating only as part of a larger whole is coherent, and if Maynard Smith and Száthmary had stuck to it, their list would be as well. Division of labor often occurs in transitions that meet the first criterion; including it would result in fewer examples of some transitions, but still a coherent, list. Sticking to the first two criteria results in a set of transitions in which independently replicating entities — molecules, prokaryotic cells, eukaryotic cells, and multicellular organisms — combine to form new kinds of independently replicating entities — chromosomes, eukaryotic cells, multicellular organisms, and animal societies, respectively. Often these transitions were followed by division of labor among the formerly independent entities. We can argue about whether or not this is the way it actually happened, especially for the first of these transitions, but if it was, this smaller set of transitions makes up something like a natural kind.

But “new ways of transmitting information” is largely independent of the first two criteria, and it leads to the inclusion of transitions that don’t meet the other criteria. There is no doubt that the origin of the genetic code was a pivotal event, but ‘really important’ is not one of the criteria for major transitions (in this, the name ‘The Major Transitions’ is unfortunate, a point I’ll return to in part 3). Similarly, the origin of language is undoubtedly important, at least for humans, but we don’t imagine that there were independently replicating words that combined into collectively replicating paragraphs.

As for combining repeatable and contingent events, I don’t see this as a problem. First of all, we don’t know which events are, as O’Malley and Powell put it, “non-replicable, one-off events.” Language and eukaryotes are obvious candidates, but we don’t actually know that, if they hadn’t arisen in humans and Lokiarchaeota, something like them wouldn’t have arisen elsewhere in the tree of life. Even if some of the transitions truly are non-replicable, though, I can see nothing wrong with identifying a set of transitions that meet some criteria (say the first two) and asking what else they might have in common. Evolution is a combination of repeatable and contingent events.

The ‘evolutionary transitions in individuality’ framework developed by Rick Michod uses a different criterion and arguably leads to a more coherent list, as O’Malley and Powell acknowledge:

To avoid ontological and theoretical disunity, many evolutionary theorists discard the inconsistencies and contingent outlier events in MTE to home in on those that represent incontestable ETIs. Richard Michod, for example, sees transitions in individuality as crucial to understanding the history of evolution on this planet:

“The major landmarks in the divergence of life and the hierarchical organization of the living world are consequences of a series of ETIs: from genes to gene networks to the first cell, from prokaryote to eukaryote cells, from cells to multicellular organisms, from asexual to sexual populations, and from solitary to social organisms.” (Michod 2005:967)

On this view, major transitions are explained by selection favouring increases in hierarchical complexity despite fitness conflicts between levels of selection (Grosberg and Strathmann 2007).

Sorry for the quote-within-a-quote. Rick Michod was my Ph.D. advisor, so bear that in mind if you think my account may be biased. Whether or not the first two of these transitions (genes to gene networks to cells) occurred the way Michod imagines them is an empirical question; in his framework they result in new units of selection and thus new kinds of individuals. If we think that the transition from asexual to sexual populations similarly results in a new unit of selection, then the list is coherent and its members represent something like a natural kind. The question of whether or not the origin of sex fits this criterion is a complicated one that I may take up in a future post. Michod clearly thinks that it does, although

…there are several senses in which the evolution of sex is not an ETI in the same sense as, say, multicellularity.

So Michod’s ETI (evolutionary transitions in individuality) framework, with the possible exception of sex, resolves the lack of unity problem, as O’Malley and Powell acknowledge:

By focusing on ETIs, incongruous events can be removed from the MTE model.

Their one criticism of this modified framework is that it remains progressivist; I disagree, as I explained in part 1.

O’Malley and Powell’s final criticism of the major transitions framework, in all of its forms, is that it excludes events that meet its criteria. This is true, and it’s a real problem. The two kinds of events that they perceive as missing from both Michod’s and Maynard Smith and Szathmáry’s frameworks are

(1) missing events that do not fit the ETI criterion, but which are so transformative from a macroevolutionary perspective that they must be included, and (2) missing events that fit the ETI criterion, but which nevertheless have been excluded from MTE.

On (2), I completely agree. Events that result in a new unit of selection should be included in the major transitions framework, and some of them have been sorely neglected. As examples, O’Malley and Powell refer to the oxygenation of the Earth and the endosymbiotic origins of mitochondria and chloroplasts.

It’s true that Maynard Smith and Szathmáry focused most of their attention to aspects of the origin of eukaryotes other than the endosymbiosis that led to mitochondria. I wouldn’t go quite so far as to say they “did not include the acquisition of the mitochondrion as anything major,” since they devoted a section of the chapter on eukaryote origins (8.6) to it, but they did take a mitochondria-late view that foregrounds other sorts of changes. However, as O’Malley and Powell acknowledge, Michod’s ETI framework does treat the fusion of two prokaryotic cells as the central event in eukaryote origins.

Regarding plastid origins, O’Malley and Powell say that

The primary plastid acquisition is rarely mentioned in MTE, and it is never discussed as a major transition; this is equally true of subsequent revisions that assume an ETI focus.

This is mostly true, to my surprise. I have always thought of chloroplast origins and, for that matter, later endosymbiotic events, as part of the major transitions framework, but Maynard Smith and Szathmáry did not include it (see their Table 1.2 above). However, that plastid acquisition “is never discussed as a major transition” is clearly an exaggeration. Stu West and colleagues include it:

Table 1 from West et al. 2015.

Table 1 from West et al. 2015; highlighting added.

As do Patrick Keeling:

The many twists and turns of plastid evolution each represent major evolutionary transitions…

and Austin Booth and Ford Doolittle:

Paulinella and its symbionts [an example of a recent endosymbiosis] are mutually interdependent, and neither can replicate independently, indicating that the symbiosis is the result of a transition similar in many respects to eukaryogenesis.

In a section called “The canonical view of major evolutionary transitions,” Douglas Erwin treats chloroplast acquisition on equal footing with that of mitochondria:

For example, during the origin of eukaryotes, envelopment of a α-proteobacterium generated the symbiotic mitochondria, while endosymbiosis of a cyanobacteria produced a photosynthetic cell with a chloroplast.

But I agree that plastid origins are neglected in discussions of major transitions. I find this surprising because I think if you asked most biologists and philosophers interested in major transitions if plastid origins should be included, nearly all would say yes. Nevertheless, only a tiny minority of published papers on the topic include them.

On the other ‘missing event,’ the oxygenation of the Earth, I completely agree that this is not part of Maynard Smith and Szathmáry’s, Michod’s, Szathmáry’s revised, or any other version of the major transitions framework that I’m aware of. The question, which I’ll take up in part 3, is should it be?


Stable links:

Booth A, Doolittle WF (2015) Eukaryogenesis, how special really? Proceedings of the National Academy of Sciences, 112, 10278-10285.

Erwin DH (2015) A public goods approach to major evolutionary innovations. Geobiology, 13, 308–315.

Grosberg RK, Strathmann RR (2007) The evolution of multicellularity: a minor major transition? Annual Review of Ecology, Evolution, and Systematics, 38, 621–654.

Keeling PJ (2010) The endosymbiotic origin, diversification and fate of plastids. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 365, 729–748.

Keeling PJ, McCutcheon JP, Doolittle WF (2015) Symbiosis becoming permanent: Survival of the luckiest. Proceedings of the National Academy of Sciences, 112, 10101–10103.

Michod, R. E. (2011) Evolutionary transitions in individuality: multicellularity and sex. Pp. 167–197 in B. Calcott and K. Sterelny, eds. The Major Transitions in Evolution Revisited. MIT Press, Cambridge, MA.

O ’Malley MA, Powell R (2016) Major problems in evolutionary transitions: how a metabolic perspective can enrich our understanding of macroevolution. Biology & Philosophy, 31, 159–189.

Szathmáry E (2015) Toward major evolutionary transitions theory 2.0. Proceedings of the National Academy of Sciences, 10104–10111.

West SA, Fisher RM, Gardner A, Kiers ET (2015) Major evolutionary transitions in individuality. Proceedings of the National Academy of Sciences, 112, 10112–10119.

EDIT: added missing reference

EDIT: fixed two misspellings (2017-04-26)


  1. Matthew Herron says

    To be fair, most of the references I cite as including plastid acquisition within the major transitions framework probably came out after O’Malley and Powell’s paper was submitted. Times from submission to publication are often ridiculously slow for philosophy journals.

  2. Ray R. says

    I own a copy of Maynard Smith and Szathmáry’s “Major Transitions”, but I suspect I am a rarity in your readers. I like the notions of the hypercycle and error catastrophe threshold described in Chapter 1 (originally conceived by Manfred Eigen), and Maynard Smith and Szathmary do a nice job of describing them.
    In terms of listing of transitions though, I think that the transition from one type of replicator “nucleic acid” to a second type “thoughts” is a transition event in a category that dwarfs all the other events and is really a super transition. One side we have billions of years of evolution using one replicator, and on the other we have the start of evolution using a second replicator. The possible transition from one type of nucleic acid RNA to to another type DNA is minor by comparison (they differ by one oxygen on the ribose.). Of course, we don’t know what the original replicator was, and certainly getting to nucleic acids could have been a major transition by itself. But having biologists acknowledge that the arrival of language was a major transition at all was pretty significant, especially when one of those biologists is the legendary John Maynard Smith.
    Some scientists have noted the changes brought about by the rise of human civilization in terms of extreme numbers of extinctions, the massive changes to ecologies like deforestation and conversion to grass lands to farm lands, and have labelled a new age, similar to the labeling of ages like the Cenozoic or the Cambrian. When it is discussed, the term most commonly used is “Anthropocene” or “Anthrocene”. Personally, I prefer “Noozoic”, and I think it belongs in the Eon category.
    I’ll be interested in seeing what you think about oxygenation in part 3.

Leave a Reply