The most recent and refined version of the evolutionary worldview that was first presented in Evolutionís Arrow can be found in the 34 page document The Evolutionary Manifesto which is here
Chapter 2. The Causes of Progress
What do I have to do to show that evolution is progressive? If I am to prove that evolution continually improves living processes in some general direction, what must I show about the evolutionary process?
One way to attack these questions is to examine a process that is clearly progressive to discover the type of mechanism that drives progress. Then we can see if the process of evolution also contains such a mechanism. If it does, we may be able to use an understanding of the mechanism to see where it is taking the evolution of life.
One of the areas that shows obvious progressive change is human technology. Technologies that serve key areas of human need such as communication, transport, and lighting have shown long trends of improvement in performance. Completely new and improved technologies have repeatedly emerged in each of these areas. Electric lighting is more efficient and convenient than gas lighting, which in turn is cleaner and more effective that lamps that burn mineral oil or animal fats. Aircraft are more efficient at some transport tasks than railways, which in turn have proved more useful than vehicles hauled by animals. And each century has seen new ways for humans to die in war.
As well as through the discovery of completely new technologies, technology also progresses incrementally. Each successful technology usually shows long sequences of improvement in performance as it is developed. The motorcar of the 1990's is faster, more reliable, and relatively cheaper than cars of the 1920's.
The case that technological change can be progressive is overwhelming. Even strong opponents of the progressive view of evolution generally accept that technological change can be progressive. But what is it that causes progressive technological change? What features of the processes that cause technological change are responsible for progressive improvement? And does the evolutionary process share these features?
Two features seem to be essential if a process is to produce progressive change. First, it must contain a mechanism that searches for improvements and reproduces any that are discovered. In the case of technological change, it is our mental processes that search for improvements. Individuals and groups are continually looking for new possibilities, trying to invent new technologies or to improve existing ones. Whether any particular discovery is reproduced depends on the extent to which it attracts sufficient interest in the economic or political marketplace. Of all the innovations thrown up by inventive minds, the market selects those that will be reproduced. Technology improves by being better able to satisfy the needs of individuals and groups who have economic or other power in these markets.
Second, a process that searches for innovation will produce progressive change only if there is potential for improvement. This potential must be on-going and not be exhausted immediately by the mechanism that searches for innovation. If the potential improvements can be discovered and implemented right away, they will not drive a sequence of progressive change. Obviously, motorcars would not have been improved throughout this century if there had been no potential for on-going improvements, or if all possible advances and supporting technologies had been discovered soon after cars were first developed.
Progress will occur if the potential improvements are discovered in a series of steps, with each step improving on previous steps, but leaving potential for further improvement that will drive further discovery. This will be the case where improvements necessarily build on previous steps, and are able to be made only after the previous steps have been taken. An example is the development of electronic monitoring and regulatory systems in car engines that had to await the emergence of computer technology, that in turn was not possible before improvements in semiconductors.
The pattern of change will also have this structure where better technologies are more complex, or where their development requires comprehensive and detailed research. The simpler technologies will generally be discovered and implemented earlier than more complex ones. For example, the first motorcars and aeroplanes were far less complex than those we see today, and improved versions that are even more complex are likely in the future.
We can therefore expect progressive change wherever it can be shown that there is a potential for improvement that is on-going in this sense. Provided the mechanism that searches for improvements is smart enough to take advantage of the potential for improvement, and not so smart that it discovers them all at once, progressive change will be produced.
With the benefit of hindsight we can see that technology has progressed where there has been potential for on-going improvements, although the potentials were often not obvious beforehand. It is not so easy to show exactly where there will be technological progress in the future. To do so we would need to identify where there are on-going potentials for technological improvement. The difficulties in doing so are notorious. In the middle of the 19th century, engineers could prove beyond doubt that flight by heavier-than-air machines is impossible. Nevertheless, assessments of where there is potential for improvement are made as a matter of course in the planning of research strategies. And where the potential exists, progress can be reasonably predicted.
Other processes will also produce progressive change if they have these two key featuresa search mechanism that discovers and reproduces improvements, and potential for on-going improvement. For example, the processes that develop skills in children often have both these features. Children search for new ways of doing things that will improve their skills, often with the help of parents and teachers. Changes are reproduced when they rewarded by success. And it is obvious to all of us who have previously mastered new skills that there is often an ascending scale of potential improvements that will drive progressive change. As a result, children progressively improve their skills as they grow and learn.
This is particularly clear in learning a musical skill such as piano playing, where each improvement necessarily builds on previous ones. There is a sequence of potential improvements in playing ability, with each improvement building on the skills gained earlier. We therefore can predict confidently that if a child has the will and ability to learn (i.e. an adequate search mechanism) progressive improvement will follow.
Unless a process contains both of these key features in full, change will not be progressive. For example, there is obvious potential for on-going improvement in the ability of human society to satisfy the needs and to develop the potential of its members. Technological progress has opened up the potential to improve the living standards of all people on this planet. But this potential for social improvement has not been fully exploited this century. It has not driven on-going social progress. Instead, at the end of the 20th century over 800 million people are chronically undernourished, and each year nearly 13 million under fives die as a direct or indirect result of hunger and malnutrition. In industrialised countries, many people spend much of their lives in unsatisfying, boring and meaningless work, while large numbers choose drugs over reality.
Human society has failed to produce on-going social progress in these areas because it does not include a mechanism that selects and reproduces social improvements when they are discovered or proposed. If society stumbles on improvements either by accident or by conscious effort, there is no process that locks in the improvements and perpetuates them. The economic and political markets that select and reproduce technological advances do not do the same for social improvements.
As we shall consider in detail later in this book, it is possible to organise human society so that social improvements are selected and reproduced, and so that social progress is as inevitable, unstoppable and natural as technological progress. And we will see that evolutionary success for humanity will ultimately depend on organising ourselves in this way.
This discussion has now got us to the point where we know what must be done to show that evolution is progressive. First we must show that the evolutionary process contains mechanisms with the ability to search for improvements in living processes and to reproduce any that are discovered. Second we must show that there are potentials for improvement in living processes that are on-going.
It seems certain that any process that successfully evolves living processes must meet the first condition. The central feature of an evolutionary process is that it searches for useful changes in living things, and reproduces improvements from generation to generation. If a process is capable of producing evolutionary change in living processes, it must contain a mechanism that searches for improvements and reproduces any that are discovered.
The most familiar evolutionary mechanism, natural selection acting on genetic variation, clearly meets this condition. Animals produce young that may vary genetically from their parents because of mutations or different combinations of genes. If the variants prove to be better adapted because of their different genes, they are likely to have relatively more offspring, with the new genes eventually taking over the population. In short, this process searches for improvements by throwing up genetic variants, and then perpetuates any improvements that are discovered.
The fact that the genetic mechanism searches for improvements largely by blind trial-and-error does not prevent it from producing progressive evolution. If there are potentials for on-going improvement and if it can stumble on them, it will produce progressive change. It does not matter that the genetic mechanism has no foresight or intention to make progressive discoveries.
Although nearly all living organisms known to man evolve through the genetic mechanism, it is not the only mechanism that produces evolution. There are other evolutionary mechanisms that search for better adaptations in living processes, and reproduce the improvements from generation to generation. A number of the processes which adapt humans as individuals provide obvious examples. As individuals, we continually use our mental processes to search for better ways of adapting to our circumstances, better ways to stay healthy, to prolong our lives, and to satisfy our needs. Often assisted by others, we develop and test possible adaptations mentally, select those that we think will work, try them out in practice, and adopt those that are best for us. When a successful improvement is discovered, it is likely to be adopted by others. If so, it will be reproduced throughout the population. It will be preserved across the generations until it to is replaced by something better. The result is an evolving culture of adaptive knowledge.
Change produced in this way is as much evolution as is genetic change produced by natural selection. When humans discovered how to build aircraft, and passed the ability from generation to generation, the result was evolutionary change. It was as much evolutionary change as the discovery by natural selection of wings in the first birds, and the passing of the relevant genes from generation to generation. The mechanisms that discovered and reproduced each of these types of change were different, but the end result was the same: evolutionary adaptation.
In humans, the processes that adapt individuals are not the only evolutionary mechanisms. Other mechanisms adapt human societies and organisations. The processes that adapt the cells within our bodies also fall into two classes. One class adapts individual cells, and the other adapts the organisations of cells that form tissues, organs, and the body as a whole. And in ant colonies, some processes adapt individual ants, and supra-individual processes adapt groups of ants and the colony as a whole.
In the case of the mechanisms that adapt individuals, possible improvements are developed, tried out, and selected at the level of the individual human, cell or ant. But in the case of organisational adaptation, it is the collective activity of many individual humans, cells or ants that search for and reproduce adaptations. Through these processes the organisation can solve adaptive problems collectively that individuals cannot. Just as our brain solves problems that are not understood by any cell in the brain, human markets and governments can solve adaptive problems that are not fully understood by any individual.
We will deal in detail later with the evolution of these various evolutionary mechanisms. We will look at what has driven the evolution of new mechanisms and how each mechanism has evolved and got smarter at evolving. Then we will consider how the mechanisms that currently evolve human individuals and human societies will evolve and improve in the future.
However, what is clear already is that the processes that produce evolutionary change each include a search mechanism. But this by itself is not enough to prove that evolution is progressive. We must also show that there are potentials for improvement in living things that are on-going and that can be exploited by the search mechanisms.
Are there on-going potentials for improvement in living things? Will an evolutionary search mechanism that is smart enough discover a sequence of changes, each better than the one before, and each an improvement in purely evolutionary terms?
Once it is accepted that evolution includes adaptive change in human culture and society, we can immediately point to an area where potentials for on-going improvement exist and have driven progressive evolutionary change: human technology. As we have seen, technological change is often progressive, producing sequences of improvements, each step better at satisfying human needs than those before. And it can be shown that there is further on-going potential for improvement in many areas of technology that will drive further progressive change.
We can go a step further by noting the general similarity between human technology and the bodily features of organisms that adapt them to their external environment. We often use our technology to adapt to our external environment. But most organisms achieve this through specialised features of their bodies such as legs, wings, fins, eyes, gills and lungs. The technology that enables them to move, see and communicate is part of their bodies.
Because of this general similarity, we can expect that the potentials for on-going improvement that have driven technological progress will also exist for the bodily technology that adapts organisms to their environment. And our experience with technological progress points to where these on-going potentials will be most pronounced. They will be most obvious in the evolution of complex adaptations such as the eye where a number of components need to cooperate together to make the adaptation effective. Evolution is unlikely to be able to perfect these complex adaptations in one step. Instead we would expect that the adaptation would be improved in a series of steps as new components are added and developed, and as components are adjusted to changes in other components.
These sequences can be expected to be most noticeable where the improvements are chasing better adaptation to environmental conditions (living or non-living) that are themselves changing. The most obvious examples are arms races between predators and their prey: the predator evolves greater speed, the prey counters by increasing its evasive ability, the predator responds with improvements in its ability to anticipate the movements of the prey, and so on.
The fossil record, where it is good enough, confirms the expected progressive evolution of complex adaptations. A particularly clear case is the evolution of warm-bloodedness in the ancestors of mammals. An ability to continually maintain the warm body temperatures that are best for movement and other activity has significant advantages. Cold-blooded reptiles are unable to do this, so they need to wait for warm air temperatures and the sun if they are to get to a temperature that is best for active movement. Early warm-blooded mammals could be out feeding all night while cold-blooded reptiles waited for the sun.
Complex changes to the blood system, physiology, and metabolism were necessary for efficient warm-bloodedness. Amongst the key changes needed were the evolution of external hair to retain heat, and the development of a four-chambered heart to improve blood circulation. The history of the evolution of the early mammals reconstructed from the fossil record shows how these various components were improved individually and as a combination over a long period. This progressively enhanced the ability of mammals to maintain a warm body temperature no matter what the external temperature.
So we have good reason to believe that much of the technology embodied in the adaptations of organisms will show the same potential for on-going improvement as many areas of our external technology. And the search mechanism of natural selection acting on genetic variation will progressively discover these potentials, producing progressive evolution in particular adaptations until the potential for improvement is finally exhausted.
Although the potential for progress in external technology and in the complex adaptations of organisms is similar, the actual rate of progress and its directness is very different. And these differences go a long way toward explaining why progress in genetic evolution is not as widely recognised as technological progress. Although both processes are fundamentally progressive, technological progress is a lot easier for us to recognise because it is more direct and it is faster, occurring during our lives.
The reason for these differences is that the genetic search mechanism is far inferior to the mechanisms that develop our technology. Genetic evolution is not very smart at exploiting the potentials for on-going improvement. Unlike us when we try to improve technology, it has no capacity to plan ahead, to use foresight, to visualise possible improvements, or imagine alternatives and try them out in its head before making them. Instead, without any idea of what might work and where it might lead, the genetic search mechanism blindly changes adaptations when new organisms are produced, and sees how the changes work in practice. There is nothing to stop the genetic search moving in the opposite direction to where the greatest improvements lie. A genetic change will be favoured by natural selection if it is better than what has come before, even though it might move away from greater potential improvement. And a change that is a step towards greater potential improvements will be reproduced only if it is itself an improvement.
Genetic evolution operates a bit like trying to improve a 1930's motorcar by making random changes to it in the dark, without any knowledge of how the car works or how it might be improved, and using parts you dont understand. The only feedback you would get about how you are going is to be told when a particular change is an improvement. We know from the history of cars since the 1930's that there are many on-going sequences of potential improvements that you could discover. However, if you set out to discover improvements using only blind trial-and-error, it would take you a very long time to make much progress and when you did, it would be very indirect.
Even where your search mechanism eventually exploited some of the potentials for improvement, the history of the changes that got you there would not look very progressive. There would be periods in which the search got stuck and no improvements were made, periods in which only minor enhancements were discovered and much better improvements were missed, and periods in which the changes that were made moved in the opposite direction to the potential for greatest progress. If you tried to decide whether such a process was fundamentally progressive by looking only at the history of changes, it would be very easy to get it wrong. It would be easy to get lost in the trees, and never see the wood.
This is the way that genetic evolution proceeds, even where the existence of potentials for on-going improvement makes it fundamentally progressive. The evolution of birds provides an example. It is evident that with the emergence and rise of land insects, large flying organisms that fed on the insects could be successful. Eventually this potential role was filled by birds. Despite the potential for success, the genetic mechanism was unable to evolve a group of insects to take up this role, probably because the insects were locked into having a hard external skeleton that is not as efficient for larger organisms as an internal skeleton. Although the insects were already there on land, it was not until many millions of years later that evolution eventually filled the role, and then only by an extraordinarily circuitous path. The ancestors of the birds came not from the land but from the water as amphibians.
The genetic evolutionary mechanism was completely unable to get itself unstuck by developing an insect with an internal skeleton. If it had done so, the role could have been filled quickly and directly by large flying insects. But in the absence of an asteroid or other catastrophe to undo enough of the evolution that had got it stuck, the genetic mechanism could not directly and immediately exploit the potential role.
So the opponents of a progressive view of evolution have found it very easy to go to the history of life on earth and point to absences of progress over long periods. But they have failed to see that extensive periods without progress are an inevitable consequence of a blind trial-and-error genetic search mechanism. The survival of some groups of animals without significant improvement for millions or even billions of years is not conclusive evidence against progressive evolution. There may be potential improvements in those groups that genetic evolution has not yet been able to discover. Life may have to await the evolution of smarter evolutionary search mechanisms before evolution can explore some of these potentials.
Whether or not evolution is progressive, there will be species that do not progress for very long periods of time. To prove the case against evolutionary progress, it is not enough to point to organisms like bacteria that have flourished unchanged on this planet for more than 3,500 million years. Anti-progressionists must also show that there are no potentials for on-going improvement that the evolutionary search mechanisms have failed to exploit. They must show that there can be no alternatives to current bacteria which, if eventually produced by evolutionary search mechanisms, would prove far more successful than current bacteria in strictly evolutionary terms. They must show that, for what they do, bacteria are perfect, and there can be nothing better. And they must show that this will remain true whatever happens in the future evolution of life in the universe.
Gould and other anti-progressionists have not even attempted to meet this critical challenge. They have not explored the possibility that there are potentials for on-going improvement that evolutionary search mechanisms have failed to discover. They have not considered whether future evolutionary developments are likely to overtake bacteria in their current form.
A major task of this book is to take up this challenge. I will show that there are general potentials for on-going improvement that will drive future progressive evolution. And I will be showing that bacteria (and humans) that dont explore these potentials for progressive improvement will not participate successfully in the future evolution of life in the universe. They will progress or perish. We will progress or perish.
So far in this Chapter we have seen that there are likely to be potentials for on-going improvement in the technology embodied in the complex adaptations of organisms. Wherever the use of genetic trial-and-error is able to exploit these potentials, progressive evolution of the adaptations can be expected. But will this evolution produce a general advance in living processes? Will life as a whole progress in evolutionary terms? Or, as argued by Gould and his supporters, will this evolution produce only better adaptation to local conditions? Will progressive change occur only while there is room for improvement in adaptation to the specific and localised environmental problems met by each population of organisms?
We would expect to find general progress if all organisms had similar potentials for on-going improvement, no matter what the nature of their local environmental conditions. The general potentials would drive general advance. In what circumstances might we expect to find general potentials of this kind? Potentials would be general if a particular adaptive problem is encountered by all organisms. We would expect general progress in the development of the adaptive technology to deal with the problem. General potentials for improvement would also exist if there were complex adaptations that would improve all organisms, no matter what their local environment.
Where these general potentials exist, problems and opportunities faced locally by populations would also be problems and opportunities faced generally by all other populations of organisms. So if a population were to exploit some of these general potentials for improvement, it would not only improve its adaptation to its local environment, but also would improve in a general sense. It would improve in all environments. By achieving better adaptation to its local conditions, the population would participate in universal advance or progress.
But do these general opportunities and problems exist? It is certainly the case that some adaptive challenges and opportunities are very general, affecting many species. The need for some form of vision is widespread, as are the benefits of warm-bloodedness. And when we examine the history of life on earth, we find general advances in particular adaptations across many species where common adaptive problems or opportunities are found.
For example, eyes have evolved in many different organisms, as have the various features that were needed for warm-bloodedness in mammals. There are also many examples of general improvements that evolved in one or a small number of closely related species and then spread widely because they brought the species general evolutionary success. The general nature of these improvements enabled the original species to adaptively radiate into many other environments, often ousting species that had not developed the general improvements. The fossil record shows many spectacular adaptive radiations triggered by the discovery of complex general improvements. For example, major radiations followed the emergence of the first primitive fishes in the Devonian, the evolution of the first primitive reptiles in the Permian, and the rise of the first dinosaur-like reptiles in the Jurassic.
Of course, even where improvements provided general advantages in this way, many species were not replaced immediately by those with the improvements. In the case of some living fossils they have still not yet been replaced. A species would not be ousted where it had valuable and specialised adaptations to local conditions that were not outweighed by the general improvements of the more progressive invaders.
This is another example of where the limitations of the genetic evolutionary mechanism mean that genetic evolution does not progress as quickly or as directly as technological evolution. When a general technological improvement is discovered, such as a lighter but stronger alloy, it can be readily combined with any other existing technology to produce improved solutions to particular problems. Technological evolution can easily combine the best with the best.
The genetic mechanism is unable to do this. It cannot directly combine adaptations discovered in different species. When a relatively general advance such as a better eye is discovered, there is no process that enables it to be immediately adopted by all those species that have inferior eyes. Again we find that although genetic evolution is as fundamentally progressive as technological evolution due to the existence of similar potentials for on-going improvement, it is far inferior at exploiting these potentials.
It is clear that there are adaptive challenges and opportunities that are not purely local. Some are general in the sense that they affect many species. But are there potentials for on-going improvement that affect all living processes, that are universal, and that will therefore drive the progressive evolution of life as a whole?
As I indicated in the first Chapter, I will be arguing that evolution progresses towards increasing cooperation amongst living things. To succeed with this argument I have to demonstrate that increased cooperation provides a general potential for on-going evolutionary improvement. To demonstrate this, I will have to establish a number of things. First, that increased cooperation amongst living processes has the potential to provide greater evolutionary success. Second, that this principle is general and the potential for greater evolutionary success applies to all living processes. Third, that the potential for improved evolutionary success is on-going, because evolution is unable to immediately exhaust the potential benefits of increased cooperation.
I will show that evolution exploits the benefits of cooperation amongst living entities through the formation of complex organisations of those entities. The organisations are structured so that cooperation is supported within the organisation. On this planet, evolution has produced cooperative organisations of molecular processes to form cells, cooperative organisations of cells to form multicellular organisms such as insects, frogs and ourselves, and cooperative organisations of humans to form human societies.
However, the formation of these structured organisations enables cooperation to be organised only between the living entities within the organisations. Evolution will not be able to exploit immediately the benefits of cooperation between the organisations. These benefits will not be exploited until the organisations themselves are organised into organisations, producing cooperative organisations of organisations. For example, the organisation of molecular processes into cells produced cooperation between molecular processes. But the benefits of cooperation between cells were not exploited until cells were organised into multicellular organisms. And cooperation between multicellular organisms was not exploited until the evolution of societies of organisms.
Continued repetitions of this process forms cooperative organisations of larger and larger scale, each containing the smaller-scale organisations that have evolved previously. As a result, human social systems include humans which include cells which include molecular processes. This evolution of organisations of larger and larger scale extends the scale over which living processes are organised cooperatively, but leaves unexhausted the potential for cooperation between organisations of the largest scale. The potential for further beneficial cooperation will not be finally exhausted until all living processes are permanently organised into a single entity that is of the largest possible scale. The potential for increases in the scale of cooperation in this universe will end only when the entire universe is subsumed in a single, unified cooperative organisation of living processes. It will end only when the matter, energy and living processes of the universe are managed into a super organism on the scale of the universe.
This evolutionary sequence has all the features of a process that is fundamentally progressive. The potential for almost indefinite expansion in the scale of cooperative organisation and the advantages this will bring provides a potential for evolutionary improvement that is on-going. And each step in the sequence of improvements builds on and goes beyond the improvements made in the previous steps.
The next five chapters are devoted to the development of these arguments in detail. We begin in Chapter 3 by exploring how cooperation can benefit living processes in evolutionary terms. We will see that by cooperating together, living processes improve their ability to meet whatever evolutionary challenges they face. Increased cooperation has a general potential to provide greater evolutionary success. And the wider the scale over which the cooperation is organised, the more successful the cooperators can be.
. For a readable, comprehensive and insightful account of the progressive evolution of communication and related technologies see Levinson, P. (1997) A Natural History and Future of the Information Revolution. Routledge: New York.
. See, for example, Chapter 15 of Gould, S. J. (1996) Full House: the Spread of Excellence from Plato to Darwin. New York: Harmony Books.
. Loftas, T. Ed. (1995) Dimensions of need: an atlas of food and agriculture. Rome, Italy: Food and Agriculture Organization of the United Nations.
. For a detailed argument in favour of this position see Durham, W. H. (1991) Coevolution: Genes, Culture and Human Diversity. Stanford, CA: Stanford University Press.
. Dawkins, R. (1997) Human chauvinism. Evolution 51: 1015-1020.
. Dawkins, R., and J. R. Krebs. (1979) Arms races between and within species. Proceedings of the Royal Society, London. B Biol. Sci. 205: 489-511.
. For example, see Carter, G. S. (1967) Structure and Habit in Vertebrate Evolution. London: Sidgwick and Jackson.
. For example, see Ayala, F. J. (1997) Ascent by natural selection. Science 275: 495-6; and Gould: Full House: the Spread of Excellence from Plato to Darwin. op. cit.
. See Chapter 7 of Stebbins, G. L. (1971) Processes of Organic Evolution. New Jersey: Prentice-Hall.