Life's solution

He starts with deepest chronological base, the formation of stars and planets. Even at this level, he stresses, there are constraints. Stars have sequential mechanisms, now fairly well defined. Following them, planets' structures and even orbits may follow almost predictable pathways. After the earliest emergence of life, rules of form, options of habitat and, ultimately, the way intellect occurs, may be broadly set and followed. Darwin understood this from the beginning - evolution builds on what's gone before. Even the most bizarre-looking sea or land life has resulted from a series of steps reaching into the past.

The body of the book portrays those steps, where identifiable in the past and as seen today. The steps, as Conway Morris rightly reminds us, are the results of adaptations through time - which he defines as "inherency". He bristles with indignation at the critics of the adaptationist programme who contend if you can't identify the "usefulness" of a trait, it's not an adaptation. Just because we are ignorant of a function doesn't mean there is none. Since evolution works on all parts of an organism, even if unequally in time or location, all evolutionary steps are adaptations. To Conway Morris, the frequent appearance of similar adaptations - "convergences" - in varying environments indicates that life operates under some general orders - it's not "rule by roulette". Among is many examples is the bizarre similarity in brain structure between human beings and mormyid fish. The latter is a creature living within an intense electrical environment. With high demand on its cognitive functions, the mormyid's brain uses about half the body's oxygen supply - three times that of the average human.

After fashioning his thematic structure with lavish amounts of material, Conway Morris nearly demolishes the edifice in explaining why he's constructed it. In a final, rambling chapter, he lays out the plans for the building. Most of it lashes out at things he deplores, including amazingly, "ultra-Darwinism", that catch-all phrase his target Gould used so ineptly. According to Conway Morris, something - not a deity, not "just six numbers", not anything even definable - but something is "out there" guiding everything from the construction of stars to evolving apes who can write sonnets. Much of this foundation is based on the thoughts of a few other writers, mostly philosophers. Here, Conway Morris exhibits a form of colour aberration - instead of using John Greene, he would better have chosen John Grey.
[stephen a. haines - Ottawa, Canada]

I can see that the book might be irritating to materialists (scientific or otherwise), but if its sometimes-controversial tone is overlooked, it has much to offer the general reader. When Conway Morris takes a position that is not orthodox, it is usually qualified with a question mark. I think the major positive contribution of the book is its many fascinating examples of convergence.

There is a remarkable relationship between the views of Stephen Jay Gould in "Wonderful Life," published in 1989, and those of Conway Morris in "Life's Solution," published in 2003. Conway Morris opposes Gould's idea of contingency. But the strange thing is that Gould, while claiming support for contingency from the Cambrian fauna, praised the work of Conway Morris on that fauna.

From the time of the Cambrian explosion of animal forms to the present there has been a marked reduction in the number of general forms. Gould would take this as evidence of the fragility of forms in the face of chance contingencies. But Conway Morris sees it as a consequence of convergence. The two men seemingly differ only in their conclusions from the evidence, but I think there is a deeper divide. To Gould nature is fundamentally probabilistic, but to Conway Morris it is deterministic. I agree, recalling that Einstein championed determinism in physics.

Gould used the idea of replaying the tape of evolution. He argued that contingencies would make the reappearance of man very unlikely. To Gould, a replay is only a thought experiment to help us understand. But Conway Morris asks what can be done in the laboratory? On pages 121-124 he describes experiments done by Lenski and Travisano with the bacterium Escherichia coli over a large number of generations. It was first separated into several populations. Then they were allowed to diversify, and were separated further. Finally all populations were switched from their customary and agreeable glucose diet to a maltose diet and allowed to try to adapt during 1000 generations. The degree and mode of their adaptation was partly due to convergence, in addition to starting points and chance, and the three could be separated statistically. Over the long term, convergence won.

Conway Morris questions the theory of the "RNA world," including the idea that the RNA was self-replicating. I think he overdoes his skepticism there. A Perspective by Leslie Orgel: "A simpler nucleic acid," in SCIENCE, 17 Nov 2000, discusses self-replication of the simpler nucleic acid TNA as well as RNA. It seems to me that the self-replicating property of RNA, TNA and similar nucleic acids assures the appearance of life by one route or another, and so discounts Conway Morris's notion that the conditions for life have to be "just right," as they are on Earth. He argues that those conditions are rare in the universe, and so account for our failure to see evidence of life elsewhere. My own view is different: My guess is that we don't find intelligent life elsewhere because when it reaches our stage of development it self-destructs. Maybe that creates a challenge: Can we be the first to acquire wisdom as well as technical skill?

Is evolutionary convergence merely a convergence of characters of two or more species when they adapt to similar ecological niches? Conway Morris would like to embed the concept in a more structured context. In reference to an interesting application he expresses it in terms of "morphological space." The particular application is to "skeleton space" as defined by Thomas and Reif. He seems to be saying that each of the conceivable morphologies in skeleton space is a fixed-point attractor. The attractor emerges as the laws of nature guide the unfolding dynamics of evolution.

Is this concept of fixed-point attractors in a character space too discrete? In "The Crucible of Creation" Conway Morris gives another example, from the work of D.M. Raup: the morphospace for the geometry of the shells secreted by the molluscs. Some regions of this morphospace are thickly populated. But other zones are more or less empty. In these, the solutions to the equations that govern the geometry can be used to visualize the hypothetical shapes, but they somehow look "wrong." Thus the general morphospace is continuous, but only particular points are realized in the real world determined by evolution.

Conway Morris makes a good case for the inevitability of humans, but the evidence is sometimes fragmentary. I think this is only the beginning. There may be as yet untapped evidence in our own present natures. In particular, I suspect that a physical understanding of the network dynamics of our nervous systems will lead to the conclusion that the brains which appeared in the Cambrian explosion would inevitably evolve to the present level, and perhaps beyond.

In Chapter 10 Conway Morris returns to the ubiquity of convergence. Convergence is found not only in directly observable phenotypic characters, but also at the molecular level. For instance, the protein rhodopsin for color vision is tuned to particular colors by substitutions at key sites, and different species adapting to the same color sometimes use identical substitutions. It can become uncertain whether molecular similarities and identities are due to convergence or common ancestry. Thus there is at the present level of knowledge a measure of uncertainty which could be exploited by creationists. But fortunately overall outlines of order are found in cladistic analysis based on molecular evidence. This reflects general human understanding as it looks out on the world with faith that order will be found.

Morris argues that evolution may have purpose, that life is not just a bleak working out of statistics. In his last chapter, he writes that "there has been a resurgence of interest in the connections that might serve to reunify the scientific world with the religious instinct." This connection of evolution to religion may make some readers uncomfortable. While Morris' writing style is generally lively, his digressions into the details of biology may leave behind non-scientist readers.

While many folks are familiar with a handful of examples of convergence (the camera eye and those marsupials in Australia come to mind), it is remarkable how pervasive the phenomenon is. In fact, although I still don't know what to make of it, Conway Morris convinced me that convergence is a fact about the world that deserves more attention than it has received.

But the book is much more than a mere compendium of examples. For Conway Morris uses the ubiquity of convergence as a counterweight to the almost orthodox view that the history of life is a governed by a large helping of luck and accident, and that, to paraphrase S.J. Gould, if we reran the tape of life's history, it would have turned out entirely differently. Convergence suggests that, whatever the role played by happenstance, natural selection has worked under narrow constraints built into the structure of reality.

Conway Morris concludes the book with some perhaps preliminary discussions about the possibility of religious and scientific understandings of the world peacefully co-existing. Here as elswhere, Conway Morris only hints at certain ideas rather than pursuing them exhaustively. As a result, some reviewers have written unfair and uncharitable things about the book. But I, for one, was left with much to ponder, and with the hope that Conway Morris will continue his provocative explorations.