Monday, February 16, 2009

Arguing your opponent’s case

Back in 2003, the Cambridge palaeontologist Simon Conway Morris wrote a book called ‘Life’s Solution, Inevitable humans in a lonely universe’. In many ways, this work marked the culmination of the argument between Morris and the late Stephen Jay Gould over the Burgess Shale and the wider question of whether evolution is random and unpredictable or has some kind of inevitable pattern. Gould had used the Burgess Shale to argue that if you were somehow able to run the tape of life again, the result would look vastly different. Conway Morris had done most of the important research on the Burgess shale and felt peeved that Gould had misrepresented his findings in this way. The result was an ill tempered argument between the two, with Conway Morris pointing to the phenomenon of evolutionary convergence and the ability of evolution to navigate to certain solutions under selection pressure. It now looks like the weight of evidence lies with Conway Morris and that biology has become less Gouldian, although recent experiments by Richard Lenski with populations of bacteria showed the importance both of constraint and historical accident in the course of life’s history. The important point about Conway Morris’s arguments is that they are Neo-Darwinian; rather than introducing some alternate mechanism into the picture he is trying to demonstrate how the deceptively simple rules of evolution can produce the complexity and creativity of life, while being constrained to certain outcomes. This is in contrast to the meaningless empire of accident, so often invoked by materialists and metaphysical naturalists, but which doesn’t tie in at all well with the universe we actually observe.

The arguments of Conway Morris were absorbed and carried further by his colleagues in a series of essays published in ‘The Deep Structure of Biology’. Some of those who had contributed had begun to produce computerised maps of ‘biological adaptive space’. It is clear from looking at these that, out of the space of adaptive possibility, almost nothing works. In other words, the convergence of biological systems can be explained by environmental and physical constraints which act on all life. These restrict the boundless creativity of life to a series of ‘optimal solutions’. In the view of Conway Morris and others, evolution acts like a kind of ‘search engine’ to create more complex ecosystems in which different niches are enabled and filled by new species.

Perhaps a bit embarrassed by Conway Morris’s theological leanings, the contributors were keen to stress that their conclusions were firmly in the neo-Darwinian mould. Robert A Foley, one of Cambridge’s leading lights in evolutionary biology and the Director of the Leverhulme Centre for Human Evolutionary Studies, was highly critical of any notions of 'progress'', 'design' and 'purpose' in his contribution, yet he concluded:

'Rather the adaptive process which is driven by selection does have some law like properties that may well - under the right circumstances - lead to more purposive behaviour as a means of increasing or coping with complex adaptive integration and greater complexity and lead to contained directional trends. These characteristics can be said to give evolution a repetitive and, hence, to some extent. inevitable pattern....The final conclusion I would draw is that evolution on other planets - or a rerun of evolution on this one - will lead to many similarities because of the law-like nature of these processes...In a distribution of intelligences in the universe, or on a sample of one, we might speculate that conscious, purpose driven intelligence represents the mode'

On the anniversary of Darwin’s birth, Conway Morris launched into an article criticising the rather dismal atheistic spin which is put on evolution. I suspect he was trying to be provocative and re-instil a sense of wonder but he just attracted a torrent of abuse from the Guardian’s ‘Dawkinsia readership’.

Conway Morris’s rant was picked up by PZ Myers, who penned a scathing review of ‘Life’s solution’ some time ago. Myers's review was particularly dismal, being about 5% science and 95% statement of his personal philosophy. He clearly didn’t read the book as in his post he appears to be arguing on Conway Morris’s side most of the time.

‘I recommend an article in this week's Nature by Shubin, Tabin, and Carroll that argues for an important concept of deep homology. We do see similar structures, such as limbs in insects and invertebrates, that are not at all homologous on a morphological level, but when we examine their molecular genetics, we find similar substrates for both. This is the central idea of deep homology, that we have shared primitives, a set of regulatory networks, that see reuse over and over again in evolution. So while limbs arose independently in insects and vertebrates, when we look more deeply, we find that both use the distal-less developmental pathway. We see convergence because there are common functional demands that channel the solutions of selection, but there are also shared molecular constraints that limit the range of likely solutions.’

Which is exactly what Conway Morris has argued, that certain constraints, be they molecular or environmental , chart the course of evolution. Molecular convergence is one of the most fascinating features of evolution. A good example is provided by Carbonic anhydrases which are zinc enzymes necessary for dealing with carbon dioxide; these accelerate chemical reactions, assist with the construction of bones, allow the storage of carbon dioxide and are used by plants in photosynthesis. Carbonic anhydrases have evolved independently at least 4 times and we can therefore be pretty confident that if there are other organisms in the universe with are dealing with carbon dioxide, they are probably using the same zinc configuration. The appearance of deep regulatory networks seems to fit pretty nicely into the idea of a naturally emerging framework for life. What is not often appreciated is that most of the building blocks we need for more complex features - such as sentience, intelligence and nervous systems - evolved long before these complex organisations emerged. The molecular substrates to make these features all emerged in bacteria and microbes. So for example, the protein for the transparency of the eye's lens has been recruited from bacteria where it was used for a completely different purpose in a different context. Certain things which are optimal for simpler forms of life thereby contrive to provide a building yard for their more complex descendents.

The part of Conway Morris’s article which raised the most derision from the ‘new atheist contingent’ was where he said that:

‘Birds evolved at least twice, maybe four times.’

This was scoffed at by PZ but interestingly Conway Morris has been vigorously defended by Richard Dawkins, who said in a recent comment:

No, he does NOT mean bats, insects and pterosaurs! Of COURSE not. If he had meant that, he would not have said 'maybe' four times but 'at least four times'. He meant BIRDS, the creatures that we all call birds.

His whole book, Life's Solution is a hymn to convergent evolution. His thesis (and it is a very interesting and persuasive thesis, one that I largely agree with until we come to the religious nonsense at the end of the book) is that convergent evolution is far more prevalent than most people realise. In this bird passage, he almost certainly is advancing the thesis that the following statement is false: "Birds are a true clade in that all birds are descended from a single ancestor, and that ancestor would itself have been classified as a bird." There were several groups of feathered dinosaurs. Majority opinion says that only one of these groups has any descendants surviving today, and we call them birds. CM is advancing the interestingly heterodox thesis that some of today's birds are descended from one of those groups of feathered dinosaurs, while others of today's birds are descended from a different group of feathered dinosaurs. He is suggesting that the most recent common ancestor of all today's birds would not have been classified as a bird.

He could be right about that. Among all zoologists, I am probably, along with Conway Morris, the one most sympathetic to that kind of view (now that Arthur Cain is dead). We both love convergent evolution. But I am probably the least sympathetic to Conway Morris's next step, which is to drag God into the story. Convergent evolution, for me, is a wonderful testimony to the power of natural selection. Conway Morris at times seems to agree. But then at other times, he seems to think . . . well, let me put it this way. If you are the betting type, you'd be well advised to put some money on Simon Conway Morris as a future Templeton Prizewinner!

Another critic who unwittingly argues for Conway Morris’s case is Stephen Pinker on Jerry Coynes’s blog who argues:

My own take: 1. Though there’s much we don’t understand about the evolution of human intelligence, nothing about it is especially mysterious. A specific ability to do physics, abstract philosophy, higher math, and the other problems that vexed Wallace never evolved in the first place – they require millennia of accumulated knowledge in a culture, and decades of education and honing in an individual. A more generic ability entertain concepts of number, objects, living things, causality, and so on, and to combine them into lawful generalizations, is patently adaptive, as we see in the ways that all human cultures depend on acquired technological know-how for their survival, outsmarting the fixed defenses of local flora and fauna. While human-level intelligence is species-specific (as are many zoological traits, such as the elephant’s trunk), impressive levels of numerical cognition and cause-and-effect reasoning have evolved several times, including in corvids, cetaceans, cephalopods, and primates.

2. Nor is morality any mystery. Abstract, universal morality (e.g., a Kantian categorical imperative) never evolved in the first place, but took millennia of debate and cultural experience, and doesn’t characterize the vast majority of humanity. More rudimentary moral sentiments that may have evolved – sympathy, trust, retribution, gratitude, guilt – are stable strategies in cooperation games, and emerge in computer simulations.

3. No feature of consciousness has ever been discovered that does not depend 100% on neurophysiology. Stimulate the brain with chemicals or an electrical current, and the person’s experience changes; let a person’s experience vary, and you can measure the changes in chemistry or electrophysiology. When a brain is damaged, the person’s mental life is diminished accordingly, and when the brain’s activity ceases, the mind goes out of existence – Wallace’s séances notwithstanding, no one has found a way to communicate with the dead. The very existence of a subjective correlate of brain activity may not be understood (if it’s an intellectually coherent problem at all, which some would deny), but positing a “soul” simply renames the problem with no insight, and leaves the perfect correlation between consciousness and neurophysiology unexplained.

Or to summarise, human like intelligence and systems of morality emerge as convergent products of the evolutionary process; all of which was the thrust of Life’s Solution. Coyne argues that our big brained intelligence is a one off and that our colossal increase in brain size which began around about 6 million years ago is a unique evolutionary trajectory. This however, is deceptive. The dolphins and porpoises, experienced a vast increase in brain size before ours and have maintained it. Our brain only overtook theirs at some point around one and a half million years ago. What is remarkable is that they show all sorts of similarities in their cognitive landscape to us, including social play, communication, the ability to recognise themselves in mirror (the mirror test) and tool use; this despite an oceanic habitat rather than an arboreal one. The uniqueness of our intelligence is probably a question of degree rather than of kind. For example, a New Caledonian crow has a similar theory of mind to a chimpanzee despite a vastly different brain structure, research into sperm whales shows that diverse social groups can combine to produce a form of culture; this is remarkably similar to elephant societies which have similar practices, despite being in a very ecosystem. With 100 billion earth like planets in our galaxy alone, and with at least 500 billion galaxies in the universe, its fair to say that Robert Foley is probably right and that conscious-driven intelligence has other footholds elsewhere given the right environment.

Pinker is right with his third point, consciousness appears to be a product of the brain, but a particularly strange one. The ‘hard problem’ , which Pinker has recognised, is why the supposedly indifferent laws of physics and chemistry should contrive to produce a subjective experience from material processes?. Here it is worth recalling that Descartes himself said in the Sixth of his 'Meditations on First Philosophy'.

'I am not present in my body merely as a pilot is present in a ship.... I am most tightly bound to it, and as it were mixed up with it, so that I and it form a unit'.

Reductive explanations of the mind have all failed, the non-reductive physicalism which is in vogue amongst philosophers of mind seems to be the way forward, but as the above quote shows, it is close to what Descartes himself was really arguing; that humans are compounds of mind and body, and it is not natural or proper to them to be anything else.

The howler of the day perhaps comes from P Z Myers. In reaction to Conway Morris’s sense of wonderment - that we, as products of evolution, should be able to do science - he says:

We simply do not hesitate to point out a rational examination of the world of biology does reveal order and pattern! Science wouldn't work if the universe were purely chaotic.

I suppose we might phrase this as the ‘science anthropic principle’. We scientifically observe a universe which has a rational order and beauty to it because that is the only type of universe we could observe scientifically. The mindless Epicurean void the materialist wants to invoke, and which has returned in the guise of the multiverse, is not what we observe. Instead science was built on the expectation that the world is ordered by laws bestowed upon it from the outside by a benevolent creator, and that therefore, if we examine it with a certain scepticism towards both nature and our own intellectual abilities, we are capable of obtaining meaningful facts from it and fitting them into a coherent framework. You can reject the religion but continue with the categories of thinking.

Discuss this post at the Quodlibeta Forum

No comments: