This page is the fourth of a series containing the article
Putting the Systems back into Systems Biology by
Athel Cornish-Bowden, published in Perspectives in Biology and Medicine (2006) 49, 475–489.
A PDF file is also available.
We must therefore not be discouraged by the difficulty of interpreting life by the ordinary laws of physics. For that is just what is to be expected from the knowledge we have gained of the structure of living matter. We must also be prepared to find a new type of physical law prevailing in it. Or are we to term it a non-physical, not to say a super-physical, law?
Pauling, L. (1987) Schrödinger’s contribution to chemistry and biology. In Schrödinger: centenary celebration of a polymath (ed. C. W. Kilminster), 225–233. Cambridge: Cambridge University Press
Schrödinger, E. (1944) What is life? Cambridge: Cambridge University Press
Schrödinger’s book What is Life? (Schrödinger, 1944) was enormously inuential in persuading physical
such as Max Delbrck that biology had something to attract their interest and attention. Others, such as Linus Pauling
and Max Perutz, were far more skeptical about its value, at least when they came to look back on it long afterwards.
Pauling (1987), for example, gave the highest praise to Schrödinger for supplying the theoretical basis for all of
chemistry, but answered the question of whether he had contributed anything to our understanding of life by saying that
it is my opinion that he did not make any contribution whatever.
Buchner, E. (1897) Alkoholische Gährung ohne Hefezellen, Ber. Dt. Chem. Ges. 30, 117–124
Anyone coming to Schrödinger’s book for the first time more than half a century after it was written is liable to find it a mixture of the obvious, the familiar and the bizarre. His main points illustrate this: that biological systems obey the laws of physics has been accepted by almost everyone since vitalism was overthrown by Eduard Buchner (1897) at the end of the 19th century; regarding genes as aperiodic crystals seems an unhelpful way of looking at something that is now understood in great detail, and understanding of inheritance was not much illuminated by this description even in 1944; that organisms feed on negative entropy seems an unnecessarily poetic way of saying that organisms are open systems that maintain themselves far from equilibrium by exporting their entropy production to the environment.
However, in 2006 we cannot put ourselves into the minds of Schrödinger’s audience in the Dublin of 1944; in any case, he has another point that may well be more important than any of those that I have so far mentioned, though it has sometimes been shrugged off by biochemists as an embarrassing excursion into vitalism. As well as arguing that biological systems obey the laws of physics, the part that anyone but a vitalist would now regard as too obvious to be worth saying, Schrödinger also suggested that new laws of physics, beyond those needed for physics itself, might be needed to explain biology. Seeing no difference between this and vitalism is failing to distinguish between necessary and sufficient conditions: it would certainly be vitalism to say that adherence to the laws of physics is not necessary for a biological system; it is much less clear that the known laws of physics are all that we need for understanding biology, or that admitting this amounts to vitalism.
Rosen, J. (2004) Autobiographical reminiscences, http://www.rosen-enterprises.com/RobertRosen/RRosenautobio.html
Quite early in my professional life, a colleague said to me in exasperation,
The trouble with
you, Rosen, is that
you keep trying to answer questions nobody wants to ask. This is doubtless true. But I have no option in this; and
in any event, the questions themselves are real, and will not go away by virtue of not being addressed. This attitude,
I know, has estranged me from many of my colleagues in the scientic enterprise, and has put me far from today’s
Joslyn C. (1993) Review of Life Itself, Int. J. Gen. Systems 21, 394–402
Although Schrödinger asks what life is, he does not try to answer his question in any depth. This task was taken up
by Robert Rosen, who defined a living organism by saying that
a material system is an organism if and only if it is
closed to efficient causation; this, unfortunately, will seem obscure enough to readers well versed in Aristotelian
categories of causation, and utterly unintelligible to everyone else. In his illuminating review of Rosen’s book
Itself, Cliff Joslyn (1993) explains that it means that no production rule in an organism is given from outside;
must be generated from within the organism, as anything else will result in an infinite regress. In biochemical terms a
production rule can be equated with an enzyme that determines what reactions a given metabolite can undergo, so
Rosen can be understood as saying that all of the catalysts necessary for an organism to live need to be produced by
the organism itself.
Woese, C. R. (2004) A new biology for a new century, Microb. Molec. Biol. Rev. 68, 173–186
The essential difference between an organism and a machine can be explained without resorting to mathematics or philosophy. Although usually ignored in analogies between machines and organisms, there is a gulf between the two that is so wide and so far from being bridged by any existing or currently conceivable machine that it is not absurd to regard it as unbridgeable. This point was taken up recently by Woese (2004), but it has has very little impact on biology in general, which continues to develop as if it were a branch of engineering. Both organisms and machines consist of components with finite lifetimes that need to be repaired or replaced when they wear out. The lifetime of a modern throw-away machine may be the same as that of its weakest component, but for more classical machines, and for all machines in principle, if not in practice, the lifetime of the machine is longer, often very much longer, than the lifetimes of some of its components. Although some very advanced modern machines (such as the computer on which I am typing this essay) may include some internal checking to alert their operators to faulty parts that need to be replaced, the actual maintenance requires external intervention, and in any case no machine keeps track of the state of repair of all of its parts.
Yet that is what a living organism does. Not only does it make itself (something no existing machine can do), but it also monitors the working state of all of its parts, and replaces those that need replacing, all of this being done from within. As we know, in good conditions a human being typically lives for about 70 years (even without the intervention of modern medicine, which is, of course, external), while containing necessary components with lifetimes in the range of minutes. This is a discrepancy of well over six orders of magnitude in the lifetime of the organism compared with that of its parts. Proteins vary considerably in their lability, and in many cases they are degraded as a result of specific catalyzed processes and not just by being worn out, but, even if this is ignored, none survive completely unchanged for decades. Some proteins, such as the crystallins of the vertebrate eye, are never replaced, and remain in use for 90 years or more, but this does not mean that they remain in perfect condition and suffer no damage. On the contrary, cataract is just the most obvious indication that crystallins do not survive unchanged for decades. The difference for crystallins is that the organism can survive without repairing the damage, whereas the overwhelmingly more usual case is that damage needs to be repaired at a rate essentially the same as the rate at which it occurs.
Understanding how this maintenance is achieved is a huge problem, and even if we restrict attention to the purely
chemical part of what it means to be alive, that is to say to metabolism, it is still a huge problem. The chemical
reactions that constitute metabolism require enzymes to catalyze them, and these enzymes survive for periods that are
several orders of magnitude shorter than the period in which the metabolism continues to function normally. They
therefore need to be replaced. (Rosen referred to
repair rather than replacement, but that was an unfortunate
choice of term, especially now that we know of many examples of genuine repair of nucleic acids, and a few examples of
repair of proteins.) The enzymes themselves must therefore be regarded as metabolites, i.e. products of metabolism, and
other enzymes are needed to catalyze the replacement process. However, these other enzymes also have finite lifetimes,
and also need to be replaced, in processes catalyzed by yet other enzymes, which also need to be replaced, and so on
for ever unless there is a way to close the circle. We therefore need a way of conceiving that the organization of
metabolism is circular, so that at no point do we need to rely on any external help.
For almost all modern organisms a small amount of external help does exist, in the sense that apart from strict chemotrophs we are all parasites, as we need some of the products left by other organisms in order to survive. However, this dependence on other organisms solves only a tiny part of the problem, even for the most thoroughly parasitic of organisms, and it cannot even have solved a tiny part for the first organisms, which needed to survive in a world with no others to parasitize. Even the first organisms, of course, required some inorganic nutrients, just as all modern organisms do, so of course no organism is closed to material causation (and Rosen did not suggest that they were). This statement of the problem is probably clearer and easier to understand than the proposal of circular organization as a solution to it.
4Aristotle’s word αιτια is usually translated as cause, but it is sometimes simply transliterated as aitia, to avoid confusion with the present-day idea of a cause, which corresponds more with the efficient cause than with the other three. This word is the origin of the English word etiology.
Having presented the problem without reference to Aristotle’s four causes4, we can now return to it
to see how
machines and organisms differ in terms of them. As Joslyn explains in his his review of Life Itself, we
different categories of causation according to the answers we give to
why? questions. If we ask why a car engine
produces water and carbon dioxide, an answer in terms of fuel and oxygen provides the material cause of the water and
carbon dioxide. However, that is not the only possible correct answer: they are also caused by mixing and sparking the
starting materials in the carburetor, and this is now an efficient cause. If we ask why the engine mixes fuel and oxygen
and then ignites them, then answering that it is to provide power to drive the car forward appeals to a final cause,
which for a machine, and indeed for everything else in classical Aristotelean philosophy, is always something outside
the machine itself. The final cause remains essential for discussing engineering, but it has largely been banished from
the modern scientist’s view of the natural world, which has no room for an external designer with definite
The fourth category is the formal cause, which concerns the essential nature of the process; it plays little role in
If we ask similar questions about metabolism, for example why glucose 6-phosphate is produced by most organisms in glycolysis (catalyzed by the enzyme hexokinase mentioned earlier), then initially the argument runs in parallel to that for machines. If the answer is that glucose 6-phosphate is produced because glucose and ATP are available then that is the material cause; if we answer that it is produced as a result of the action of the enzyme hexokinase, then that is the efficient cause. So far so good, but when we ask where the hexokinase comes from there is no final cause: it certainly does not come from outside the organism, so it must be produced from within. We can trace back material and efficient causes for that, but we never reach a final cause. This is what it means to say that organisms are closed to efficient causation. In admitting it to be true, however, we are stepping outside everything that we know about machines, and everything we can derive from classical philosophy.