Ch 12: extract 6

More reasons we are not like machines ... and why do we find it so hard to say so?

7. Imprecise boundaries

That leads to the next issue. A machine has clear boundaries; a natural system does not. The machine model involves being able to identify viably distinct, stable things as parts, and a viably distinct, stable thing – the machine – as the product of their combination. Processes, by contrast, can overlap in a way that ‘things’ typically do not. Processes ‘have boundaries that are fuzzy or indeterminate’, write Dupré and Nicholson:

Processes are individuated not so much by where they are as by what they do … At no level in the biological hierarchy do we find entities with hard boundaries and a fixed repertoire of properties. Instead, both organisms and their parts are exquisitely regulated conglomerates of nested streams of matter and energy.^[1]

They liken organisms to Jupiter’s Red Spot, which, ‘by means of its continuous activity … demarcates itself from the flux of its surroundings’.^[2] Organisms are not tightly defined entities: neither even, as we saw, are some kinds of genes.^[3]The genome itself is not as defined as we imagine: ‘it is not even clear that there is any unique entity that qualifies as the genome of an organism.’^[4]

Whitehead pointed out the artificiality of drawing boundaries to organisms:

The truth is that the brain is continuous with the body, and the body is continuous with the rest of the natural world. Human experience is an act of self-origination including the whole of nature, limited to the perspective of a focal region.^[5]

We have seen that the environment is, in a concrete fashion, partially constituted by, and constitutive of, interdependent organisms, in which each is involved in the evolution of the other. ‘Some of these interactions’, according to Dupré, ‘are so intimate and so fundamental to the survival of the organism that it has been hotly debated whether the interacting entities are distinct at all, or should rather be understood as constituting a single life.’^[6] The philosopher of science Argyris Arnellos writes that ‘collaboration is one of the main characteristics of life and its evolution … it is very hard to imagine life (both at the intra- and the inter-cellular level) that is not collaborative.’^[7]

What he means by this can be illuminated by biochemist and philosopher Stephan Guttinger, who writes that ‘symbiotic life forms are the rule rather than the exception in the biological realm’:^[8]

the significance of symbiosis examples in general is the realisation that the entities we used to refer to as single living things … turn out to be less autonomous than we might have thought them to be … it is not just the termite that shapes the mound and that has the power to digest, reproduce and survive, but a system of interconnected entities that has to be treated as the centre and/or origin of these capacities or powers.^[9]

Too narrow a focus means that we conventionally think of organisms as sharply delineated individual entities, with linear causative relations. If, instead, we conceived them as systems, in which causation is both distributed and reciprocal, and these systems themselves nested in other, larger systems, and so on upwards and outwards, we would see a quite different picture unfold. We would then see, not just competition – that we would still, of course, see – but a combination of competition with co-operation. Such a relationship, in which division and union are fruitfully balanced, is what we mean by collaboration.

A superannuated myth of pop science has it that human beings evolved to be fundamentally selfish, striving to maximise their inclusive fitness by aggression and nepotism. As the anthropologist Christopher Hallpike reminds us, ‘studies of actual human behaviour, especially that of children show that, whatever evolutionary theory may say, while we must obviously act for our own preservation and well-being, we also have highly developed capacities for sympathy, co-operation, and group loyalty’.^[10]

Ultimately we have common ancestors. As I suggested, evolution becomes problematic when species are thought of as one sharply defined thing replacing another. However, ‘this is no problem at all for a process. No-one expects a thunderstorm or a battle to have precisely delineated boundaries.’^[11] Living organisms integrate their cellular communities by mingling their bodies and sharing their genes. In this way they can acquire learnt experiences from other organisms.^[12] The boundaries of species are, in reality, blurred. In the words of Daniel Drell, Programme Manager of the US Government’s Joint Genome Institute: ‘We can no longer comfortably say what is a species any more’.^[13]

8. Boot-strapping

In the case of a machine, the instructions for making the machine cannot themselves be the product of the very machine they are designed to make. Even in a computer, the software is separate from the hardware: the hardware has to be finished, before the software can be extrinsically inserted into it. The code for making the machine is not being simultaneously written by the machine in the very process of beginning to form itself as a computer. By contrast, as philosophers of science Paul Griffiths and Karola Stotz put it, with reference to the work of Susan Oyama,

the developmental information expressed in the organism is not present in the starting point of development, but is itself created by the process of development, through feedback from the current state of the organism to the states of the resources that will influence future development.^[14]

Even if computers can be brought to a point where they make apparently autonomous decisions, they can do so only because the capacity for them to do so depends on previous acts of human intelligence and human will. The ‘parts’ of an organism that are claimed to account for emerging phenomena lack adequate explanatory power, since the way in which those parts act when and where they do, and even their continued existence, cannot be understood without invoking the process they are supposed to explain, and, in something of an infinite regress, the processes to which, in turn, that process leads. A machine’s functioning is linear, whereas the process of living things is of entities that mutually co-arise.

Why has the machine model proved so attractive?

In the face of such overwhelming evidence of the inadequacy of the machine model to the study of living organisms, why, then, does this product of the mid-Victorian mindset persist?

One reason is its simplicity. We are familiar with machines, and because they are what we are used to making, taking apart and putting together, it is perhaps a natural assumption that other systems, ones that we didn’t make, will work in a similar way. This is particularly the source of the appeal in molecular biology, a discipline that necessarily concerns itself primarily with events that appear highly circumscribed in both space and time. Here chains of cause and effect tend to be abstracted from their wider context, and therefore seem to function in a broadly mechanical fashion. Molecular biology places ‘the entire explanatory burden on a relatively stable, epistemically tractable entity: a one-dimensional digital code that can be replicated, modified, and transplanted from one individual to another’.^[15]

Regular and reliable patterns of behaviour discoverable in an organism are unreflectively assumed to demonstrate mechanisms. But regularities do not always mean there is a mechanism; nor do they imply determinism. Random events can cause predictable outcomes. The rain that falls at a random moment in time in a random distribution on a landscape encounters potential stored in the landscape; that potential is actualised in the predictable flow of water down to a river mouth into the sea. Similarly, accidental deaths are by definition unplanned. Yet roughly the same number of accidental deaths, around 17,000, occurs in the UK every year. There is not an emergency team of bungee jumpers on standby to make sure the statistics remain steady. It cannot be overemphasised that regularities in nature do not imply mechanisms, nor do they imply determinism.

Christopher Hallpike makes the point that convergent patterns of social evolution do not imply determinism. ‘It is rather like the game of Monopoly’, he writes:

the players are all different and the throws of the dice produce a completely different game each time, yet the underlying constraints produce essentially the same result – a single player who owns everything and has driven all the others into bankruptcy. This is a good illustration that unique events, even randomness, and free will, are quite compatible with broadly predictable outcomes.

This is true without reference to the quantum nature of reality. But when one takes that into account, the case is strengthened. ‘Quantum physics showed that even the causal mode of description is only one very special way of tying together natural phenomena, and thus lawfulness in nature is not equivalent to mechanical causality’, wrote Bohr’s assistant Aage Petersen. ‘We have had to learn that the causal mode of description is a limiting case that encompasses only the simplest features of nature.’^[17] Physicist David Oliver puts it like this:

Nature is neither inevitably random nor completely lawful and predictable … Think of the universe as fabric woven on a loom. The warp threads are the laws of motion – rigid, and invariant, the weft, the emergent random strands that weave within the ordered warp. Together order and randomness form a creative whole … There exist physical events the randomness of which cannot be dispelled by the most determined efforts to expose ‘a deterministic cause’ … a stream of unpredictable information constantly enters the stream of definite information, a source of novelty, creativity, vitality…

Living things superficially exhibit some of the reliability and stability we associate with clockwork, though nothing about them in any way resembles clockwork, and they are, by many orders of magnitude, both much less stable, and at the same time much more reliable, than any machine.

I imagine some readers are thinking that, for all it may be a fiction, the machine model has persisted precisely because it has proved a very useful one. That is true enough. But that it can often prove useful does not mean it is true. In daily life, and even for most engineering purposes, we find it practical to assume that the earth is flat. If, however, we want not just to build a new garage, but to navigate the seas, the flat earth model is going to prove a serious handicap. In a conversation at the Niels Bohr Institute, David Bohm was asked if ‘science has shown us something that scientists do not want to see?’ He replied:

Well, they have become so used to their way of seeing it, that they don’t want to change, you see they feel uncomfortable about changing. And they feel there’s no reason to change, they say we’re doing so well now, why should we change. In one sense it looks as if we’re doing very well, but if you look at the broader view, it looks very dangerous.

The purpose of science is not utility: it can never be reduced to ‘how can I most effectively exploit the world?’ Schrödinger was right: science’s aim is nobler than that. It is to understand nothing less than who we are. So the truth matters, and matters greatly. Furthermore, in the context of evaluating science as a path to truth – the purpose of this section of the book – it is, of course, the central issue. And the machine model is found wanting. As von Bertalanffy put it, ‘even as a fiction the machine idea does not attain its goal, because … it proves to be inadequate in the face of a large and important section of biological data.’^[20]

It’s not even as though the fiction of the machine is harmless. It can be seen behind many of the more inhuman consequences of our technological society: our attitudes to both what human life is and to the living world at large. So, again: why does the model persist?

Probably not just because of familiarity or utility. The drive behind the machine model is, fascinatingly, much stronger and much less cold-blooded than an explanation on either ground would suggest. The sheer fanaticism of its advocates, and the animus directed at those who question the machine model, suggests a more impassioned investment, having little to do with science at its best.^[21] I am inclined to think that one element in the model’s popularity is that it encourages the sense that we can easily understand what life is and learn to control it – Faustian fantasies, in other words, of omniscience and omnipotence, that reductionists quite rightly dislike when they see them attributed to a God (I share their qualms).^[22]

Here the attentive reader will hear bells ringing from Part I. Breaking down an entity into its parts to see how it works and attempting to build it up again so as to manipulate and control it are familiar proclivities of the left hemisphere. So too is the phenomenon of denial, refusing even to countenance that a living organism might not be a machine when the evidence stares us in the face. It is reminiscent of the exclusively left hemisphere-dependent subject in Deglin and Kinsbourne’s experiments insisting that a porcupine is a monkey, because ‘it says so on the card’. Suggestive of left hemisphere dependency, too, is the facetious dismissal when its account of reality is challenged, something we saw in the cases of patients with right hemisphere dysfunction in Part I; and ultimately so is the anger which is ultimately provoked. The left hemisphere insists on its own vision, thereby interrupting the process whereby that vision needs to be referred back to the right hemisphere, so as to be taken up into the whole, as just one part – a valuable one, without doubt – of an understanding in context. Metaphorically speaking, it is as though many biologists now reside in the left hemisphere’s hall of mirrors, and not only cannot find the way out, but have stopped being aware there is a world outside to attain.

The left hemisphere’s serial, analytic approach is better equipped to deal with a system that is closed, static, linear and predictable – like a machine; not one that is open, constantly flowing, becoming and changing, and ultimately complex and indeterminate – like life. In the left hemisphere’s vision, things take priority over processes. It is good at understanding linear cause and effect, not so much reciprocal interaction, let alone a process of co-creation. It understands a whole as simply the assemblage of parts, and causation as from bottom up only, not from many directions at once within the whole. It is at home when it can follow procedures; less so when it comes to recognising new forms, or fields, at work. It prefers what is clearly defined, to what has imprecise boundaries. It doesn’t see Gestalten, of which life provides the pre-eminent examples.

This allegiance to the left hemisphere is, then, another reason why, however efficacious biology may have become in terms of manipulating the world, its claims to truth should be treated with care. But clearly a further question arises. If the hemisphere hypothesis, by positing the notion of left hemisphere capture, helps illuminate why biology has found itself in the strange predicament of claiming that black is white, why is it that physics, a science with the same pedigree of attachment to the machine model, should have escaped capture, as we saw in the last chapter that it had?

I think that the answer has much to do with the particular nature of Life. There are a number of aspects to this.

The reader will remember from Part I the difficulty the left hemisphere has with dealing with living things: its preference for inanimacy, its tendency to see parts of the body as lifeless ‘things’, its believing living beings (including the self) to be dead, its seeing people as machines or zombies, and its more general tendency to devitalisation. Then there is its tendency to halt the living flow of natural processes, substituting a succession of static representations, something I will say much more about in Part III. Encountering life, it needs to break it down into bits which can be manipulated and controlled; but life is ultimately recalcitrant to any such process. And, on top of all this, and most important, the left hemisphere can’t handle the concept of something with manifest intrinsic purpose (its own, inner, purposiveness), rather than something manifesting extrinsic purpose, engaged in instrumentalising or being instrumentalised by, its environment and other existing beings – the only kind of purpose the left hemisphere understands, and which is indeed its raison d’être.

In Part I, I quoted the philosopher Hans Jonas to the effect that the scientist assumes humans to be subject to causal laws from which the scientist himself is exempt. More than that, however, he saw that in this process inevitably ‘man-the-knower apprehends man-qua-lower-than-himself … since all scientific theory is of things lower than man the knower’:

It is on that condition that they can be subjected to ‘theory’, hence to control, hence to use. Then man-lower-than-man explained by the human sciences – man reified – can by the instructions of these sciences be controlled (even ‘engineered’) and thus used.

Philosopher of biology Stephen Talbott sent a draft book chapter to ‘a first-rank molecular biologist’; Talbott quotes from his response:

After describing my views as ‘very interesting, provocative, and necessary’, and before offering his support for much of what I had to say, he voiced this concern: ‘You very explicitly dispense with vitalism. Nevertheless, your piece is permeated by an atmosphere that says “There is something special about living things”.’

‘So I believe there is’, Talbott continues. ‘Animals and plants are a long way from rocks and clouds, and also from automobiles and computers. The need to point this out today is one of the startling aspects of the current scientific landscape.’^[24]

I agree with Talbott that, while there is no need for some ‘vital substance’ to be added for life to come about, there is obviously something special about living ‘things’. It borders on the absurd to deny it: and, as I have argued, organisms are just not machines. In a later chapter I will, however, argue that there is no absolutely hard and fast line to be drawn between the animate and what we call the inanimate. Not, of course, for the sort of reasons that usually lie behind such claims, those of the School of Nothing Buttery. Not, then, that the living is best seen by likening it to the inanimate; but, on the contrary, that the inanimate is best seen, as biophysicist Robert Rosen eloquently explained, as a limit case of the animate. That discussion will have to wait for now. What I want to draw attention to here is the curious phenomenon whereby scientists are made desperately uncomfortable by even the whisper of there being something special about living beings, or of their having intrinsic purposes of their own.

---

^[1] Dupré & Nicholson 2018 (27).

^[2] ibid (25).

^[3] Koonin 2016; and see p * above.

^[4] Dupré 2010.

^[5] Whitehead 1933 (290: emphasis added).

^[6] Dupré & Nicholson 2018 (20).

^[7] Arnellos 2018 (217).

^[8] Guttinger 2018 (309).

^[9] ibid (311).

^[10] Hallpike 2013. See more generally Tomasello 2009.

^[11] Dupré 2017a.

^[12] Lipton 2005 (14).

^[13] Quoted in Pennisi 2001.

^[14] Griffiths & Stotz 2018 (232), paraphrasing Oyama 2002.

^[15] Nicholson 2014. The paper provides a lucid and informative account of the history of the machine model in biology.

^[16] Hallpike 2013.

^[17] Petersen 1963.

^[18] Oliver 2019. A comment by philosopher of science Nancy Cartwright reflects her perception of a common structure to freedom in society and physics: ‘I imagine that natural objects are much like people in societies. Their behaviour is constrained by some specific laws and by a handful of general principles, but it is not determined in detail, even statistically. What happens on most occasions is dictated by no law at all’ Cartwright 1983 (49). For further discussion of determinism, in relation to matter and consciousness, see Chapter 25.

^[19] Bohm 1989.

^[20] von Bertalanffy 1933 (38).

^[21] In the introduction to J Scott Turner’s fascinating book The Tinkerer’s Apprentice, in which he dares to question the sufficiency of purely mechanical explanations, one senses his apprehension. Describing his book as not intentionally provocative, but nonetheless almost certain to irritate throughout, he writes: ‘The irritation culminates in the fourth stage, which is also the last chapter, where I bring the discussion back to what place design might have in a comprehensive theory of evolution. If you have not thrown the book against the wall already, perhaps this is the chapter that will make you do it’ (Turner 2007, 3). Is that the best he should be able to expect of his scientific colleagues? Would such a remark be expected in any other scientific context than the profoundly conservative life sciences?

^[22] One reason they may dislike such an attribution is that, that way, the ‘divine’ qualities are no longer available to be applied – in some cases, no doubt, unconsciously – to themselves. Some of it is quite explicit. The title of Mark Lynas’s popular book The God Species refers to ourselves.

^[23] Jonas 2001 (196).

^[24] Talbott 2010b (27).