Friday, May 28, 2010

Not all contemporary art is rubbish

I’m thrilled to see my friend, photographic and video artist Lindsay Seers, being given some respect in Ben Lewis’s excellent piece for Prospect on why modern art is in a decadent phase. Like Ben, I think Lindsay is doing serious and interesting stuff, and I say that not just because (perhaps even despite the fact that?) I’ve been involved in some of it. I wrote a piece for Lindsay’s book Human Camera (Article Press, 2007), which I’m now inspired to put up on my web site.

Monday, May 24, 2010

Creation myths

Artificial life? Don’t ask me guv, I was too busy last week building sandcastles in Lyme Regis. However, now making up for lost time… I have a Muse on Nature’s news site (the pre-edited text of which is below – they always remove the historical quotes), and a piece on the Prospect blog. The Venter work may, if it survives the editor’s shears, also be briefly discussed on an episode of Radio 4’s Moments of Genius that I’ve also just recorded with Patricia Fara, due to be broadcast this Sunday (30th May).

Claims of ‘synthetic life’ have been made throughout history. And each time, they are best regarded as mirroring what we think life is.

The recent ‘chemical synthesis of a living organism’ by Craig Venter and his colleagues at the J. Craig Venter Institute [1] sits within in a very long tradition. Claims of this sort have been made throughout history. That’s not to cast aspersions on the new results: while one can challenge the notion that this new bacterium, whose genome is closely modelled on that of Mycoplasma mycoides, stands apart from Darwinian evolution, the work is nonetheless an unprecedented triumph of biotechnological ingenuity. But when set in historical context, the work reflects our changing conception of what life is and how it might be made. What has been done here is arguably not so much a ‘synthesis of life’ as a (semi-)synthetic recreation of what we currently deem life to be. And as with previous efforts, it should leave us questioning the adequacy of that view.

To see that the new results reiterate a perennial theme, consider the headline of the Boston Herald in 1899: ‘Creation of Life. Lower Animals Produced by Chemical Means.’ The article described how the German biologist Jacques Loeb had caused an unfertilized sea-urchin egg to divide by treating it with salts. It was a kind of artificial parthenogenesis, and needless to say, very far from a chemical synthesis of life from scratch.

But Loeb himself was then talking in earnest about ‘the artificial production of living matter’, and he was not alone in blending his discovery with speculations about the de novo creation of life. In 1912 the physiologist Edward Albert Schäfer alluded to Loeb’s results in his presidential address to the British Association, under the rubric ‘the possibility of the synthesis of living matter.’ Schäfer was optimistic: ‘The [cell] nucleus – which may be said indeed to represent the quintessence of cell-life – possesses a chemical constitution of no very great complexity; so that we may even hope some day to see the material which composes it prepared synthetically.’

Such claims are commonly seen to imply that artificial human life is next on the agenda. It was a sign of the times that the New York Times credulously reported in 1910 that ‘Prof. Herrera, a Mexican scientist, has succeeded in forming a human embryo by chemical combination.’ It is surely no coincidence that many media reports have compared Venter to Frankenstein, or that the British Observer newspaper mistakenly suggested he has ‘succeeded in ‘creating’ human life for the first time’.
What is life?

Beliefs about the feasibility of making artificial organisms have been governed by the prevailing view of what life is. While the universe was seen as an intrinsically fecund matrix, permitting bees and vermin to emerge from rotten flesh by spontaneous generation, it seemed natural to imagine that sentient beings might body forth from insensate matter. The mechanical models of biology developed in the seventeenth century by René Descartes and others fostered the notion that a ‘spark of life’ – after the discovery of electricity, literally that – might animate a suitably arranged assembly of organic parts. The blossoming of chemistry and evolutionary theory spurred a conviction that it was all about getting the recipe right, so that nature’s diverse grandeur sprung from primordial colloidal jelly, called protoplasm, which Thomas Henry Huxley regarded as the ‘physical basis of life’.

Yet each apparent leap forward in this endeavour more or less coincided with a realization that the problem is not so simple. Protoplasm appeared as organic chemists were beginning on the one hand to erode the concept of vitalism and on the other to appreciate the full and baffling complexity of organic matter. The claims of Loeb and Schäfer came just before tools for visualizing the sub-cellular world, such as X-ray crystallography and the electron microscope, began to show life’s microstructure in all its complication. As H. G. Wells, his son George, and Julian Huxley explained in The Science of Life (1929-30), ‘To be impatient with the biochemists because they are not producing artificial microbes is to reveal no small ignorance of the problems involved.’

The next big splash in ‘making life’ came in 1953 when Harold Urey and Stanley Miller announced their celebrated ‘prebiotic soup’ experiment, conjuring amino acids from simple inorganic raw materials [3]. This too was obviously a very far cry from a synthesis of life, but some press reports were little troubled by the distinction: the result was regarded as a new genesis in principle if not in practice. ‘If their apparatus had been as big as the ocean, and if it had worked for a million years, instead of one week’, said Time, ‘it might have created something like the first living molecule.’ Yet that same year saw the discovery of life’s informational basis – the source of much of the ‘organization’ of organic matter that had so puzzled earlier generations – in the work of Crick and Watson. Now life was not so much about molecules at all, but about cracking, and perhaps then rewriting, the code.

Burning the book

Which brings us to Venter et al. Now that the field of genomics has fostered the belief that in sequencing genomes we are reading a ‘book of life’, whose algorithmic instructions need only be rejigged to produce new organisms, it’s easy to see why the creation of a wholly synthetic genome and its ‘booting up’ in a unicellular host should be popularly deemed a synthesis of life itself. Here the membranes, the cytoplasm, everything in fact except the genes, are mere peripherals to the hard drive of life. (The shift to a new realm of metaphor tells its own story.)

But what this latest work really implies is that it is time to lay aside the very concepts of an ‘artificial organism’ and a ‘synthesis of life’. Life is not a thing one makes, nor is it even a process that arises or is set in motion. It is a property we may choose to bestow, more or less colloquially, on certain organizations of matter. ‘Life’ in biology, rather like ‘force’ in physics, is a term carried over from a time when scientists thought quite differently, where it served as a makeshift bridge over the inexplicable.

More important than such semantics, the achievement by Venter et al. is a timely reminder that anything laying claim to the function we might call life resides not in a string of genes but in the interactions between them. Efforts to make de novo organisms of any complexity – for example, ones that can manufacture new pharmaceuticals and biofuels under demanding environmental constraints – seem likely to highlight how sketchily we understanding how those interactions operate and, most importantly, what their generic principles are. The euphoria engendered by rapid whole-genome sequencing techniques is already giving way to humility (even humiliation) about the difficulty of squaring genotype with phenotype. Yet again, our ideas of where the real business of life resides are shifting again: away from a linear ‘code’ and towards something altogether more abstract, emergent and entangled. In this regard at least, the latest ‘synthesis of life’ does indeed seem likely to repeat the historical template.

 1. D. G. Gibson et al. Science doi:10.1126/science.1190719 (2010)
2. E. A. Schafer, Nature 90, 7-19 (1912)
3. S. Miller, Science 117, 528 (1953)

Tuesday, May 11, 2010

Debunking is hard to do

In his excellent article on ‘denialism’ in this month’s New Humanist, Keith Kahn-Harris mentions that one of the problems debunkers face is that they have to engage in ‘a minute and careful examination of the sources… [which is] a time-consuming task that requires considerable skill and fortitude.’ This was precisely what I found myself up against when I reviewed Christopher Booker’s climate-change-denial tract The Real Global Global Warming Disaster for the Observer. I examined in detail just a very few of the claims Booker made (that is, ones that we not transparently false or misleading), and in each case found considerable distortion. I put the results of that trawling on this blog, but even then there was too much information for me to find the time to get it into an easily digested and streamlined shape. The real problem is that the denialists seem to have endless time on their hands. Happily, Booker’s book doesn’t seem to have had a huge impact, but less happily that is perhaps because there is now just so much climate denialism around, thanks largely to the silliness at UEA.

This issue of New Humanist is as full of good stuff as ever, but I particularly liked A. C. Grayling’s skewering of Terry Eagleton’s book On Evil: ‘Eagleton has been too long among the theorists to risk a straightforward statement… as we are dealing with Eagleton here, note that this is of course not a mish-mash of inconsistencies, as it appears to be; this is subtlety and nuance. It is, you might say, nuance-sense.’ For one reason or another, I have recently found myself having to read various texts issuing from the cultural-studies stable, and I can regretfully say that I know just what he means.

Sunday, May 09, 2010

Private Passions

I was the guest today on Radio 3’s Private Passions, where I get to choose half an hour of music and talk about it with Michael Berkeley. It can be heard here for the next seven days, I believe, but after that it vanishes into the BBC’s vaults. As ever with radio interviews, only afterwards do I realise what eloquent things I could have said in place of ‘um, you know…’. But I enjoyed it.

Wednesday, May 05, 2010

What a shoddy piece of work is man

It seems kind of cheap to win the ‘most commented’ slot on Nature News simply by writing an article about science and religion. You just know that will happen; there is nothing like it for provoking readers to offer their tuppence’ worth, and in particular for drawing reams of comment from the fundamentalist fringe. My latest Muse (pre-edited version below) is no exception. I am, however, entertained by the thoughtful remark of Bjørn Brembs, who says:

“As usual, your article is very reasoned, thoughtful and balanced. Reading some of the comments here, however, I fear you are making a common mistake, so accurately described by PZ Myers: "Where scientists are often handicapped is that they don't recognize the depth of the denial on the other side, and that their opponents really are happily butting their heads against the rock hard foundation of the science. We tend to assume the creationists can't really be that stupid, and figure they must have some legitimate complaint about some aspect of evolution with which we can sympathize. They don't. They really are that nuts."
Does it make sense to to try and reason thoughtfully with someone who prefers "magic man did it" over "I don't know" as an answer to scientific questions? Couldn't it be that this peculiar and revealing preference alone constitutes evidence enough that this person may not be amenable to reason at all?”

Bjørn is probably right in most cases, but I should say that I’d be a sad fool indeed if I wrote pieces like this under any belief that they would convert creationists. No, I do it because I think the issues are interesting, namely: how well has evolution done in designing our genome? (Not very.) To what extent does evolution optimize anything at all? (Not much.) And how come we work pretty well despite all this mess? (That’s the really big question.)

Our genome won't win any design awards and doesn't speak well of the intelligence of its 'designer'.

Helena: They do say that man was created by God.
Domin: So much the worse for them.

This exchange in Karel Capek’s 1921 play R.U.R., which coined the word ‘robot’, is abundantly vindicated by our burgeoning understanding of human biology. Harry Domin, director general of the robot-making company R.U.R., jeers that ‘God had no idea about modern technology’, implying that the design of human-like bodies is now something we can do better ourselves.

Like most tales of making artificial people, R.U.R. contains a Faustian moral about hubris. But whether or not we could do better, it’s true that the human body is hardly a masterpiece of intelligent planning. Most famously, the eye’s retina is wired back to front so that the wiring has to pass back through the screen of light receptors, imposing a blind spot.

Now John Avise, an evolutionary geneticist at the University of California at Irvine, has catalogued the array of clumsy flaws and inefficiencies at the fundamental level of the genome. His paper , published in the Proceedings of the National Academy of Sciences USA [1], throws down the gauntlet to advocates of intelligent design, the pseudo-scientific face of religious creationism. What Intelligent Designer, Avise asks, would make such a botch?

Occasional botches are, meanwhile, precisely what we would expect from Darwinian evolution, which is blind to the big picture but merely tinkers short-sightedly to wring incremental adaptive advantage from the materials at hand. Just as in technology (and for analogous reasons), this produces ‘lock-in’ effects in which strategies that are sub-optimal from a global perspective persist because it is impractical to go back and improve them.

Intelligent design (ID) does not have to deny that evolution occurs, but it invokes an interventionist God who steps in to guide the process, constructing biological devices allegedly too ‘irreducibly complex’ to have been assembled by blind random mutation and natural selection, such as (ironically) the eye or the flagellar motor of bacteria [2].

As Avise points out, ID is problematic in purely theological terms. Were I inclined to believe in an omnipotent God, I should be far more impressed by one who had intuited that a world in which natural selection operates autonomously will lead to beings that function as well as humans (for all our flaws) than by one who was constantly having to step in and make adjustments. I’m not alone in that: Robert Boyle felt that it demeaned God to suppose he needed constantly to intervene in nature: ‘all things’, he said, ‘proceed, according to the artificer’s first design, and… do not require the peculiar interposing of the artificer, or any intelligent agent employed by him [3].

But ID must also confront the issue of theodicy: the evident fact that our world is imperfect. Human free will allegedly absolves God of responsibility for our ‘evil acts’ – but what about the innocent deaths caused by disease, natural disasters and so forth? Infelicities in the course of nature were already sufficiently evident in the eighteenth century for philosopher David Hume to imply that God might be considered a ‘stupid mechanic’. And in the early twentieth century, the physician Archibald Garrod pointed out how many human ailments are the result not of God’s wrath or the malice of demons but of ‘inborn errors’ in our biochemistry [4,5]

Many of these ‘errors’ can now be pinpointed to genetic mutations: at a recent count, there are around 75,000 disease-linked mutations [6]. But the ‘unintelligent design’ of our genomes, Avise says, goes well beyond such flaws, which might otherwise be dismissed as glitches in a mostly excellent contrivance.

The ubiquity of introns – sequences that must be expensively excised from transcribed genes before translation to proteins – seems to be a potentially harmful encumbrance. And numerous regulatory mechanisms are needed to patch up problems in gene activity, for example by silencing or destroying imperfectly transcribed mRNA (the templates for protein synthesis). Regulatory breakdowns may cause disease.

Why design a genome so poorly that it needs all this surveillance? Why are there so many wasteful repetitions of genes and gene-fragments, all of which have to be redundantly replicated in cell division? And why are we plagued by chromosome-hopping ‘mobile elements’ in our DNA that seem only to pose health risks?

These design flaws, Avise says, ‘extend the age-old theodicy challenge, traditionally motivated by obvious imperfections at the levels of human morphology and behavior, into the innermost molecular sanctum of our physical being.’

Avise wisely avers that this catalogue of errors should deter attempts to use religion to explain the minutiae of the natural world, and return it to its proper sphere as (one) source of counsel about how to live.

But his paper is equally valuable in demolishing the current secular tendency to reify and idealize nature through the notion that evolution is a non-teleological means of producing ‘perfect’ design. The Panglossian view that nature is refined by natural selection to some ‘optimal’ state exerts a dangerous tug in the field of biomimetics. But we should be surprised that some enzymes seem indeed to exhibit the maximum theoretical catalytic efficiency [7], rather than to imagine that this is nature’s default state. On the whole there are too many (dynamic) variables in evolutionary biology for ‘optimal’ to be a meaningful concept.

However – although heaven forbid that this should seem to let ID off the hook – it is worth pointing out that some of the genomic inefficiencies Avise lists are still imperfectly understood. We might be wise to hold back from writing them off as ‘flaws’, lest we make the same mistake evident in the labelling as ‘junk DNA’ genomic material that seems increasingly to play a biological role. There seems little prospect that the genome will ever emerge as a paragon of good engineering, but we shouldn’t too quickly derogate that which we do not yet understand.

1. Avise, J. C. Proc. Natl Acad. Sci. USA doi:10.1073/pnas.0914609107.
2. Behe, M. J. Darwin’s Black Box: The Biochemical Challenge to Evolution (Free Press, New York, 1996).
3. Boyle, R. ‘Free inquiry’, in The Works of the Honourable Robert Boyle Vol. 5, ed. T. Birch, p.163 (Georg Olms, Hildesheim, 1965-6).
4. Garrod, A. Inborn Errors of Metabolism (Oxford University Press, London, 1909).
5. Garrod, A. The Inborn Factors of Inherited Disease (Clarendon Press, Oxford, 1931).
6. Stenson, P. D. et al., Hum. Mutat. 21, 577581 (2003).
7. Albery, W. J. & Knowles, J. R. Biochemistry 15, 5631-5640 (1976).