Reply to Haldane 153a well-agreed hypothesis of the origin of proto-life and its development into
bigger self-reproducing molecules which then carry the information to build
other structures around them, such as the coating that a virus has, and how
there might evolve bigger and more complex structures, namely living cells.
Still, the plausibility, in the light of recent knowledge, of how in sketchy
outline it mighthave happened is enough for me to prefer the naturalistic
hypothesis to a supernaturalist one. A philosopher who antecedentlyfinds
supernaturalism plausible can reasonably jump the other way. I am not expect-
ing agreement with John Haldane. Sympathetic understanding of one another’s
position is what I here aim for. Still, we should keep one eye on the scientific
literature: a more detailed and acceptable account of how life could have
evolved might well be in the offing.
Though Haldane has given a good and sympathetic account of my philo-
sophical methodology, I think that he may possibly have misled the reader
in his talk of the physicalist as a reductionist. Of course I do not believe that
talk of tomatoes, say, can be translatedinto talk of electrons, protons, and
other entities postulated in physical theory. For one thing ‘tomato’ is learned
partly ostensively. For another thing the molecules in a tomato are immensely
numerous and their arrangement immensely complicated so that we could
never give a complete description. Moreover the arrangements in one tomato
would not be the same as those in another. In a certain sense ‘tomato’ is a
more abstract word than ‘hydrogen atom’ is, for example. In calling a thing
a tomato we abstract from very many constitutional differences. Also words
of ordinary language can be very contextual, linked to anthropic interests.
Thus ‘tree’ is not a word of botanical classification, and if we were small
enough a dandelion might count as a tree. I can concede all this without pre-
judice to my conviction that a tree is just a very complex physical mechanism.
My physicalism is an ontologicalone, not a translationalone.
I can even talk, in a weak sense, of levels of organization. Consider an
old-fashioned radio receiver. One can look at it and see thermionic valves,
capacitors, inductors, resistors, a transformer and a loudspeaker, all connected
together in determinate ways by wires and at one end to an aerial wire. Now
consider the components themselves. The thermionic valve (I take the simple
case of a triode valve) has an evacuated glass tube which contains a wire, the
cathode, heated by an electric current so that electrons are given off and are
attracted to a bit of metal called the anode, which is positively charged. In
between the anode and the cathode is another bit of metal in the form of
a grid, into which is fed a varying charge, the signal, and which causes
amplified variations in the flow of electrons from cathode to anode. In fact all
the components can be explained in physical terms in this sort of way. Now
consider the radio receiver itself. It can do things that a mere jumble of
components cannot do. The components have to be connected together in