Botany online 1996-2004. No further update, only historical document of botanical science!

Walter Arndt Lecture:

The Autonomy of Biology

Ernst Mayr

Science as we know it is the product of developments in the 17th and 18th centuries. What was conceived as science is indicated by the names Galileo, Kepler, Newton, Descartes, and Leibniz. It was the glorious rise of the physical sciences advancing from triumph to triumph. It was physics what people meant when they talked about science, biology was never included. The science of biology was a creation of the l9th century. Of course, there were branches of medicine such as anatomy, physiology, and embryology that we now include in biology but they were concerned with healing and were not considered to be the part of a separate science of biology. At the same time there was also a flourishing pursuit of natural history under the label of Natural Theology. A genuine biology remained dormant until the 19th century.

It took more than two hundred years and the occurrence of three different sets of events before a genuine science of biology originated. One can assign these events to three different sets: (A) the refutation of certain erroneous principles, (B) the demonstration that certain basic principles of physics cannot be applied to biology, and (C) the realization of the uniqueness of certain basic principles of biology that are not found in the inanimate world. Let us now review these three sets of events.

A: The refutation of certain erroneous basic assumptions.

Under this heading I deal with the support for certain basic ontological principles which later were shown to be erroneous. Biology could not be recognized as a science of the same rank as physics as long as most biologists accepted certain basic explanatory principles not supported by the laws of the physical sciences and eventually found to be invalid. The two major principles here involved are vitalism and a belief in cosmic teleology. As soon as it had been demonstrated that these two principles are invalid and, more broadly, that none of the phenomena of the living world are in conflict with the natural laws of the physicalists, there was no longer any reason for not recognizing biology as a legitimate autonomous science, equivalent to physics.


The nature of life, the property of being living, has always been a puzzle for the philosophers. Descartes tried to solve it by simply ignoring it. An organism is really nothing but a machine, he said. And other philosophers, particularly such with a background in mathematics, logic, physics, and chemistry, tended to follow him and operated as if there was no difference between living and inanimate matter. But this did not satisfy most naturalists. They were convinced that in a living organism certain forces are active which do not exist in inanimate nature. They concluded that just as the motion of planets, suns, and stars, is controlled by an occult, invisible force called by Newton gravitation, analogously the movements and other manifestations of life in organisms is controlled by an invisible force, Lebenskraft or vis vitalis. Those who believed in such a force were called vitalists.

Vitalism was popular from the early 17th century to the early 20th century. It was a natural reaction to the crass mechanism of Descartes. Henri Bergson and Hans Driesch were prominent vitalists in the early 20th century. The end of vitalism came when it could no longer find any supporters. Two causes were largely responsible for this, first the failure of literally thousands of failed experiments conducted to prove the existence of a Lebenskraft, and secondly, the realization that the new biology, with the methods of genetics and molecular biology, was able to solve all the problems for which one traditionally had invoked the Lebenskraft. In other words the proposal of a Lebenskraft had simply become unnecessary.

It would be ahistorical to ridicule vitalists. When one reads the writings of one of the leading vitalists like Driesch one is forced to agree with him that many of the basic problems of biology simply cannot be solved by a philosophy as that of Descartes, in which the organism is simply considered a machine. The developmental biologists in particular asked some very challenging questions. For exarnple, how can a machine regenerate lost parts, as many kinds of organisms are able to do? How can a machine replicate itself ? How can two machines fuse into a single one like the fusion of two gametes when producing a zygote ?

The logic of the critique of the vitalists was impeccable. But all their efforts to find a scientific answer to all the so-called vitalistic phenomena were failures. Generations of vitalists labored in vain to find a scientific explanation for the Lebenskraft until it finally became quite clear that such a force simply does not exist.


Teleology is the second invalid principle which had to be eliminated from biology before it qualified as a science equivalent to physics. Teleology deals with the explanation of natural processes which automatically seem to lead to a definite end or goal. To explain the development of the fertilized egg to the adult of a given species Aristotle invoked a fourth cause, the causa finalis. Eventually one invoked this cause for all phenomena in the cosmos which led to an end or goal. Kant in his Critique of Judgment had tried to explain the biological world in terms of Newtonian natural laws but had been completely unsuccessful in this endeavor. Frustrated he ascribed all Zweckmässigkeit as being due to teleology. This was, of course, no solution. A widely-supported school of evolutionists, for instance the so-called orthogenesists, invoked teleology to explain all progressive evolutionary phenomena. They believed that in living nature there is an intrinsic strive toward perfection. Here belongs also Lamarck's theory of evolution and orthogenesis had many followers prior to the evolutionary synthesis. Alas, no evidence for the existence of such a teleological principle could ever be found and the discoveries of genetics and paleontology eventually totally discredited cosmic teleology. The distinguished philosopher Quine told me once that he considered it as Darwin's greatest achievement to have refuted Aristotle's fourth cause by showing that the attainment of a definite goal in evolution could be explained by natural selection. There are numerous seemingly goal-directed processes in nature, particularly in biology, which one now no longer explains by occult teleological forces but by scientifically supported chemical-physical factors. I distinguish four groups of such factors which formerly had been referred to as teleological (Mayr 1992, 1998). The fifth group of factors, attributed to an intrinsic potential of the cosrnos to cause goal direction, could never be substantiated. Such cosmic teleology is now considered not to exist.

What is biology ?

When we investigate this we find that biology actually consists of two rather different fields, functional biology and historical biology. Functional biology deals with the physiology of all activities of living organisms, particularly with all cellular processes, including those of the genome. These functional processes can be explained by chemistry and physics.

The other branch of biology is historical biology. A knowledge of history is not needed for the explanation of a purely functional process. It is however all-important for the explanation of all aspects of the living world that involve the dimension of historical time, in other words, as we now know, all aspects dealing with evolution.

The two fields of biology also differ in the nature of the most frequently asked questions. To be sure in both fields one asks what? questions, in order to get the facts needed for further analysis. The most frequently asked question in functional biology, however, is how?, while in evolutionary biology it is why? This difference is not complete because in evolutionary biology one also asks occasionally how? questions, for instance how do species multiply ? However, as we will see, evolutionary biology has developed its own methodology, that of historical narratives, to obtain its answers, particularly in cases where experiments are inappropriate.

In order to truly appreciate the autonomy of biology one must know the remarkable difference between these two branches of biology. Indeed some of the most decisive differences between the physical sciences and biology are true for only one of these branches, for evolutionary biology.

The emergence of modern biology.

The two hundred year period, from ca. 1730 to 1930, witnessed a radical change in the conceptual framework of biology. It included the period from 1828 to 1866 during which both branches of modern biology, functional and evolutionary biology, were established, which resulted in the eventual refutation of vitalism and cosmic teleology. Yet biology was still largely ignored by the philosophers of science from Carnap, Hempel, Nagel, Popper, to Kuhn. The biologists, even though they now rejected vitalism and cosmic teleology, were unhappy with a purely mechanistic philosophy of biology. But all endeavors to escape from this dilemma, such as for example the writings of Jonas, Portmann, Uexküll, and several others, invariably invoked some non-mechanical forces. It was not until almost the middle of the 20th century until it became evident that a solution could not be found by a philosopher who did not have a background in biology. At the same time it became equally clear that, however the solution would be, it had to be completely compatible with the natural laws. No solution was acceptable that would invoke any occult forces. But how could such a solution be found ?

It turned out that in order to develop an autonomous science of biology one had to do two further things. First, undertake a critical analysis of the conceptual framework of the physical sciences. This revealed that some of the basic principles of the physical sciences are simply not applicable to biology. They had to be eliminated and replaced by principles pertinent to biology. And secondly, it was necessary to investigate whether biology is based among others on certain additional principles that are inapplicable to inanimate matter. This required a restructuring of the conceptual world of science that was far more fundamental than anyone had imagined at that time. lt became apparent that the publication in 1859 of Darwin's Origin of Species was really the beginning of an intellectual revolution that ultimately resulted in the establishment of the autonomy of biology.

B. Physicalist ideas not applicable to biology.

Darwin's ideas were particularly important in the discovery that a number of basic concepts of the physical sciences, which up to the middle of the 19th century were also widely held by most biologists, are not applicable to biology. I will now discuss four of these basic physicalist concepts for which it had to be demonstrated that they are not applicable to biology, before it was realized how different biology is from the physical sciences.

1. Essentialism.

From the Pythagoreans and Plato on the traditional concept of the diversity of the world was that it consisted of a limited number of sharply delimited and unchanging eide or essences. This viewpoint was called typology or essentialism. The seemingly endless variety of phenomena, it was said, actually consisted of a limited number of natural kinds (essences or types) each forming a class. The members of each class were thought to be identical, constant, and sharply separated from the members of any other essence. Therefore, variation was nonessential and accidental. The essentialists illustrated this concept by the example of the triangle. All triangles have the same fundamental characteristics and are sharply delimited against quadrangles or any other geometric figure. An intermediate between a triangle and a quadrangle is inconceivable.

Typological thinking, therefore, is unable to accommodate variation and has given rise to a misleading conception of human races. Caucasians, Africans, Asians, or Inuits are types for a typologist that conspicuously differ from other human ethnic groups and are sharply separated from them. This mode of thinking leads to racism. Darwin completely rejected typological thinking and used instead an entirely different concept, now called population thinking (see below).

2. Determinism.

One of the consequences of the acceptance of deterministic Newtonian laws was that it left no room for variation or chance events. The famous French mathematician and physicist Laplade boasted that a complete knowledge of the current world and all its processes would enable him to predict the future to infinity. Even the physicists soon discovered the occurrence of enough randomness and contingencies to refute the validity of Laplace's boast. The refutation of strict determinism and of the possibility of absolute prediction freed the way for the study of variation and of chance phenomena, so important in biology.

3. Reductionism.

Most physicalists were reductionists. They claimed that the problem of the explanation of a system was resolved in principle as soon as the system had been reduced to its smallest components. As soon as one had completed the inventory of these components and had determined thc function of each one of them, they claimed it would be an easy task to explain also everything observed at the higher levels of organization. This claim was vigorously objected to by biologists already more than one hundred years ago. Darwin's friend T. H. Huxley asked the reductionists why the reduction of water (H2O) to hydrogen gas and oxygen gas did not explain the nature of water ? What the reductionists confused was reduction and analysis. Of course we learn a great deal about any complex system by analyzing it. Indeed analysis is a most important and heuristic method in all branches of science, including biology. But first of all, in many cases of analysis, one does not have to go anywhere near the smallest parts (electrons, protons, etc...). Furthermore, and this is what the reductionists usually overlooked, in order to understand a system one needs to know not only the properties of its components but also the nature of the interactions among these components. And it is precisely these interactions that are so important in living systems. Their study is the objective of the holists (see below).

4. Laws.

All theories in the physical sciences are based on natural laws. Normally they have no exceptions which led Popper to the claim that any exception to a theory amounted to its falsification. Indeed, on the whole, properly articulated laws are the basis of the theory structure of the physical sciences. But is this also true for biology ? This has been seriously questioned by a number of philosophers (Smart 1963, Beatty 1995) and only few biological theories are based on laws. They are usually rather based on concepts, as we shall presently see.

The inapplicability of these four principles that are so basic in the physical sciences, has contributed a great deal to the insight that biology is not the same as physics. To get rid of these erroneous ideas was the first but perhaps hardest step in developing a sound philosophy of biology.

C. Autonomous characteristics of biology.

The last step in the development of the autonomy of biology was the discovery of a number of biology-specific concepts or principles. One is the concept of evolution. To be sure even before Darwin geologists knew about changes on the Earth's surface and cosmologists were aware of the probability of changes in the universe, particularly in the solar system. However, the world was seen as something quite constant, something that had not changed since the day of Creation. This view totally changed after the middle of the 19th century when science became aware of the comprehensiveness of the evolution ofthe living world.

The adoption of the concept of the biopopulation is responsible for what now seems probably the most fundamental difference between the inanimate and the living world. The inanimate world consists of Platonian classes, essences, types, with the members of each class being identical, and with the seeming variation being "accidental" and therefore irrelevant. In a biopopulation, by contrast, every individual is unique, while the statistical mean value of a population is an abstraction. No two of the six billion humans are the same. Populations as a whole do not differ by their essences but only by mean statistical values. The properties of populations change from generation to generation in a gradual manner. To think of the living world as a set of forever variable populations grading into each other from generation to generation, results in a concept of the world that is totally different from that of a typologist. The Newtonian framework of unalterable laws predisposes a physicist to be a typologist, seemingly almost as if by necessity. Darwin introduced population thinking into biology rather casually, and it took a long time before it was being realized that this is an entirely different concept from the typological thinking traditional in the physical sciences (Mayr 1959).

Population thinking and populations are not laws but concepts. It is one of the most fundamental differences between biology and the so-called exact sciences that in biology theories are usually based on concepts while in the physical sciences they are based on natural laws. Examples of concepts that became important bases of theories in various branches of biology are territory, female choice, sexual selection, resource, and geographic isolation. And even though, through appropriate rewording, some of these concepts can be phrased as laws, they are something entirely different from the Newtonian natural laws.

All biological processes differ in one respect fundamentally from all processes in the inanimate world; they are subject to dual causation. In contrast to purely physical processes these biological ones are controlled not only by natural laws but also by genctic programs. This duality fully provides a clear demarcation between inanimate and living processes.

The dual causality, however, which is perhaps the most important characteristic of biology by which it differs indubitably from the physical sciences, is a property of both areas of biology. When I speak of dual causality I am of course not referring to Descartes' distinction of body and soul but rather to the remarkable fact that all living processes obey two causalities. One of them are the natural laws which, together with chance, control completely everything that happens in the world of the exact sciences. The other causality consists of the genetic programs which characterize the living world so uniquely. There is not a single phenomenon or a single process in the living world which is not controlled by a genetic program contained in the genome. There is not a single activity of any organism that is not controlled by such a program. There is nothing comparable in the inanimate world. Dual causation, however, is not the only unique property of biology to support the thesis of the autonomy of biology. Indeed it is reinforced by some six or seven additional concepts. I will now discuss some of these.

The most novel and most important concept introduced by Darwin was perhaps that of natural selection. Natural selection is a process that is both so simple and so convincing, that it is almost a puzzle why after 1858 it took almost 80 years before it was universally adopted by evolutionists. To be sure, the process has been somewhat modified in the course of years. It is rather a shock for some biologists to learn that natural selection, taken strictly, is not a selection process at all, but rather a process of elimination. It is the least well adapted individuals that are eliminated in every generation, and those that are better adapted have a greater chance to survive. Also, in recent years, there has been a great deal of argument, what was more important, variation or selection. For me, there is no argument. The production of variation and true selection are for me inseparable parts of a single process. At the first step variation is produced by mutation and recombination, and at the second step the variants are sorted by selection. Of course, during sexual selection real selection takes place. Natural selection is the driving force of organic evolution and represents a process quite unknown in inanimate nature. This process enabled Darwin to explain the "design" so important in the arguments of the natural theologians. The fact that all organisms are seemingly so perfectly adapted to each other and to their environment was attributed by the natural theologians to God's perfect design. Darwin however showed that it could be equally well, indeed even better, explained by natural selection. Before and even after the publication of the Origin, it was widely postulated that God had given Nature the capacity to move toward perfection. Indeed many natural processes seem to move toward a final goal such as the fertilized egg toward the adult stage. Aristotle had referred to this as the "final cause", later philosophers had called it teleology.

Evolutionary biology is a historical science. It is very different from the exact sciences in its conceptual framework and methodology. It deals, to a large extent, with unique phenomena, such as the extinction of the dinosaurs, the origin of man, the origin of evolutionary novelties, the explanation of evolutionary trends and rates, and the explanation of organic diversity. Evolutionary biology tries to find the answer to why questions. Experiments are usually inappropriate for obtaining answers to evolutionary questions. We cannot experiment about the extinction of the dinosaurs or the origin of mankind. There is, however, a remarkably heuristic method available, that of historical narratives. Just as in much of theory formation you start with a conjecture which you thoroughly test for its validity, so in evolutionary biology you construct a scenario, a historical narrative, which you test for its explanatory value. Let us take the case of the extinction of the dinosaurs. An early conjecture was that they had been the victims of a devastating viral or bacterial epidemic. For various reasons this narrative was not very credible. A second scenario suggested that a drastic climatic event had led to the mass extinction. Geologists however could find no evidence for a drastic, climatic change. Eventually the physicist Alvarez suggested that the Earth had been hit by an asteroid which had kicked up such a large amount of dust that for a short time it had made life on Earth very precarious. The dinosaurs became extinct but a few probably nocturnal and small mammals had been lucky to survive. The impact crater of the postulated asteroid was eventually found near the Yucatan peninsula in Mexico and all other recent findings since have confirmed the impact theory to such an extent that it is now quite generally accepted.

The methodology of historical narratives is clearly a methodology of historical science. Indeed evolutionary biology as a science, is in many respects more similar to the Geisteswissenschaften than to the exact sciences. When drawing the borderline between the exact sciences and the Geisteswissenschaften it would go right through biology and attach functional biology to the exact sciences while including evolutionary biology with the Geisteswissenschaften. This, incidentally, shows the weakness of the old classification of the sciences. That classification was made by philosophers familiar with the physical sciences and the humanities but completely ignorant of the existence of biology.


The natural laws usually effect a rather deterministic outcome in the physical sciences. Neither natural nor sexual selection guarantee such determinism. Indeed the outcome of an evolutionary process is usually the result of an interaction of numerous incidental factors. Chance is also rampant in the production of variation. It governs both crossing-over and the movement of chromosomes in the reduction division. It was curiously this chance aspect of natural selection for which this theory was most often criticized. Some of Darwin's contemporaries, for instance the geologist Sedgwick, declared that invoking chance was unscientific. Actually, it is precisely the chanciness of variation that is so characteristic of Darwinian evolution. Even today there is much argument concerning the role of chance in the evolutionary process. Selection, of course, always has the last word.

Holistic thinking.

Reductionism was the declared philosophy of the physicalists. Reduce everything to the smallest parts, determine the properties of these parts, and you have explained the whole system. However, in a biological system there are so many interactions among the parts, for instance among genes of the genotype, that a complete knowledge of the properties of the smallest parts by necessity gives only a partial explanation. Nothing is as characteristic of biological processes as interactions at all levels, among genes of the genotype, between genes and tissues, between cells and other components of the organism, between the organism and its inanimate environment, and between different organisms. It is precisely this interaction of parts that gives nature as a whole, or the ecosystem, or the social group, or the organs of a single organisms, its most pronounced characteristics. To repeat what I said before, rejecting the philosophy of reductionism is not an attack on analysis. No complex system can be understood except through careful analysis. However the interactions of the components must be considered as much as the properties of the isolated components. And this is what the reductionists had neglected.


We are now ready to summarize the results of our study of the autonomy of biology. Yes, we have concluded, that biology is a science, as much as chemistry and physics are sciences. However biology is in many respects a very different science from the so-called exact sciences. Perhaps the most pronounced difference is that biology, in part, is a historical science. In this part of biology, evolutionary biology, the method of historical narratives is the most heuristic approach. Furthermore, biology has as its subject matter living organisms which in several ways differ quite fundamentally from inanimate objects. They have, in particular, two characteristics for which there is no equivalent in the world of the physicist. One is that every process and every activity is controlled by a double set of causations; the natural laws and the genetic programs. The natural laws are the same that control the physical sciences. There is no room in biology for anything, as for instance vitalism, that would be in conflict with the natural laws. But every living organism and its parts is also controlled by a second source of causation, the genetic programs. The absence or presence of genetic programs denotes the sharp border between the inanimate and the living world. Typological (essentialistic) thinking is misleading when applied to organisms. What must be used instead is population thinking which realizes that in a biological population every individual is unique and differs from all others. The statistical mean value of a population is merely an abstraction. Dual causality as well as the uniqueness of every individual of a biopopulation characterize the world of living beings and are therefore characteristic for biology.

This explains why all endeavors prior to the last fifty years or so to construct a philosophy of biology within the conceptual framework of the physical sciences were such a failure. Biology, we now realize, is indeed largely an autonomous science and a philosophy of biology must be based on the peculiar characteristics of the living world, recognizing at the same time that this is not in conflict with a strictly physico-chemical explanation at the cellular-molecular level.


[l] E. Mayr: The idea of teleology. J. Hist. Ideas 53, 117 (1992). -
[2] E. Mayr: The multiple meanings of "Teleological". Hist. Phil. Life Sci. 20, 35 (1998).
[3] J. J. C. Smart: Philosophy and Scientific Realism. Routledge & Kegan Paul. London 1963.
[4] J. Beatty: The evolutionary contingency thesis. In G. Wolters, J. Lennox (Hrsg.): Concepts, Theories and Rationality in the Biological Sciences. University of Pittsburgh Press. Pittsburgh 1995.

adapted for the internet: Peter v. Sengbusch -