2.7.2 Living things
Throughout this course, we’ve been discussing animals and plants as living things, members of a distinct, biological kingdom, as if it was self-evident what life actually is, and that the distinction between living things (such as parrots, spiders and humans) and artefacts (such as bridges, spoons and computers) is clear cut. At first glance, the distinction does seem self-evident: surely we all believe that there is a sharp division between systems that are biological, or living, on the one hand, and non-biological, or inanimate, on the other. We would generally have no trouble in assigning an anteater to the first and a computer to the second.
But, as so often, such a clear and obvious separation may not stand rather closer scrutiny. Think about this difficult question.
What makes something ‘biological’? What makes it ‘living’ as distinguished from something that is ‘non-biological’ and ‘non-living’? This is a huge question, so don’t dwell on it for too long. Just jot down one or two possible answers.
Arriving at a hard and fast definition of ‘life’ as a basis for distinguishing living from non-living things has proved very difficult, if not impossible. The Oxford English Dictionary defines ‘biology’ as ‘the science of physical life’, but this hardly helps: it’s another piece of circularity. Some scientists would want to say that biological systems depend on complex interactions between certain types of organic molecules, notably DNA, amino acids and proteins. They might offer a rough definition of a living thing as one that shows evidence of all of the following at least once during its existence:
- metabolism: it maintains itself by taking in, storing and using energy and expelling waste
- response to stimuli
Unfortunately, as you can probably see, these definitions can easily be broken down by counterexamples. A cast-iron definition of ‘life’, and even of ‘biology’, has eluded science. Certainly, all known biology depends on certain complex organic molecules, but so do many other systems that we would not consider to be alive. Crystals grow; a laptop computer takes in energy, stores it, uses it for computations and radiates waste in the form of heat. Cars move under their own steam; and thermostats respond to changes in their environment. Viruses, which are only questionably alive, reproduce themselves.
According to The Oxford Companion to Philosophy:
Efforts to find some distinctive substance characterizing life have proven as futile as they have been heroic. The one thing which is clear is that any analysis of life must accept and appreciate that there will be many borderline instances, like viruses.
Many scientists believe that the definition-of-life question is a red herring. Life, they say, exists on a continuum. There are no simple alive/not-alive decisions to be made. A rock would certainly be low on any continuum of aliveness; a dog, a tree and a human would rank highly. Some systems would fall in a middle region of semi-aliveness – below bacteria, which almost everyone agrees are alive, and some way above rocks. Viruses would fall somewhere in the upper reaches of this middle ground. Below them would be complex systems that no one really considers to be alive but that display some behaviours consistent with living organisms – things such as the economy and cars.
The observation that all living things are built on an organic substrate is not particularly useful either. There are no grounds for asserting that biology must depend on organic molecules: some scientists call this belief carbon chauvinism. Indeed, one could make a persuasive case for the idea that the more sophisticated kinds of computer viruses also fit most definitions of life: they certainly move, respond, reproduce and share the metabolic properties of the computer, although there is nothing organic about them. But ordinary common sense would seem to insist that they cannot be alive.
These discussions are not irrelevant to the concept of artificial intelligence. If we accept that the distinction between ‘natural’ and ‘artificial’, ‘biological’ and ‘non-biological’ is not clear-cut, that natural things don’t necessarily have to be built on an organic substrate, and that intelligence, in the sense of successful, goal-directed, problem-solving behaviour, is widespread in the natural world, then a study of how intelligence (of this broader kind) comes about could have profound implications for the kinds of systems that computer scientists write.
We can now move on to explore some of these implications.