LogoEkstig, B. (2004).The Evolution of Language and Science Studied by means of Biological Concepts.
Journal of Memetics - Evolutionary Models of Information Transmission, 8.

The Evolution of Language and Science Studied by means of Biological Concepts

Börje Ekstig
Department of Teacher Training
Uppsala University, Uppsala Sweden
1.  Introduction
2.  The Pattern
3.  Condensation
4.  Parallels in Cultural Evolution
5.  The Evolution of Language
6.  The Evolution of Science
7.  Summary


This study examines certain mechanisms underlying the evolution of language and science — including mathematics — using concepts developed in the field of biological evolution. Developmental processes are particularly emphasized. Analysis of developmental processes, processes such as human embryonic development, children’s verbal development, and adolescents’ scientific conceptual development reveals the unifying principle referred to as “condensation” — the successive shortening of developmental stages. The mechanism of condensation is coupled to the rate of evolutionary change.

The analysis examines the applicability of the concept of the meme. Regarding the evolution of language, we suggest a cooperative combination of genetic and memetic replication; while early on in the evolution of science only memetic replication is envisaged.

Key words: Evolution, development, memes, cultural evolution, language, science, mathematics.

1.  Introduction

This paper examines the mechanisms underlying the evolution of language and science, applying concepts developed in the field of biological evolution. Common to these various manifestations of evolutionary process is developmental growth, as observed in the embryo, child, and adolescent. The significance of the developmental process in its various guises forms the main subject of my study. Such a broad approach is highly multi-disciplinary, and in this introduction I will comment on some relevant fields of research.

In an early work, Stephen J. Gould (1977) reviews the widespread notion of a coupling between the developmental and evolutionary processes. More recently, our rapidly growing insight into genetics has lent this controversial field renewed respect. One such new field of research focuses on the developmental genetic machinery that underlies embryological phenotypes (Arthur 2002). It is recognized that evolutionary change occurs not by direct transformation of adult ancestors into adult descendents, but rather when developmental processes produce the features of each generation in an evolving lineage (Raff 2000). Wagner et al. (2000) have examined another theme, the role of development in the origin of evolutionary innovations. In this study I will build on the same principal insight of the central in role of development, but restrict my discussion to the lineage of humans.

Concerning cultural evolution, there is an overwhelmingly abundant literature on the relationship between cultural and biological evolution within which one can discern several different approaches. One such approach is socio-biology. As Edward O. Wilson has suggested, certain cultural norms may survive and reproduce better that other competing norms, causing culture to evolve along a track parallel to and usually much faster than genetic evolution. Wilson, however, contends that the connection between genes and culture is never completely broken and discusses this connection in terms of gene-culture co-evolution (Lumsden and Wilson 1981), a concept that supports the general notion of a close relationship between cultural and biological evolution.

Another approach that examines the connection between cultural and biological evolution is memetics—an approach likely familiar to readers of this journal. Memeticists attempt to explain cultural evolution without reference to any benefit for genes. The Darwinian principle of natural selection is expanded, and is regarded as working not only on genes but also on memes. Works by Dawkins (1976), Dennett (1995), and Blackmore (1999, 2000) introduce and develop the concept of memes. These authors attempt to grant memes an analogous role in cultural evolution to that played by genes in organic evolution. I will analyze this idea at length in this investigation.

In studying children’s psychological development, psychologists have attempted to apply concepts from evolutionary theory. Advocates of this influential field of research, called evolutionary psychology, emphasize that understanding the principles governing evolution is indispensable in attempting to understand human nature (Tooby and Cosmides 1992, p. 50). Such a general statement may stand as a guiding principle for the present investigation. However, if I have understood evolutionary psychologists correctly, they also seem to imply that all behavior ultimately comes back to genetic advantage. Such a view is questioned, for instance by Dennett and Blackmore, and I will demonstrate that genes are insufficient to explain many characteristics of cultural evolution.

Language is an important constituent of culture, and there is a tremendously rich literature dealing with the enigmatic question of the origin and evolution of language. There has recently been exciting debate as to whether or not language has evolved by means of a principle analogous to natural selection; Dennett provides a survey of this debate (1995, pp. 384-400). I suggest that any explication of the evolution of language, while incorporating elements from natural selection, should have recourse to additional mechanisms as well.

There is a widespread notion that the evolution of science has much in common with biological evolution. Thus Blute (2003), for instance, discusses a new field of research called evolutionary ecology. I will refer to such analogies, and in particular analyze their implications for children’s and adolescents’ learning of science.

The referenced literature testifies to the many notions of connections between cultural and biological evolution. These connections are elaborated in the analysis I will propose in this paper. It builds on my previously introduced model (Ekstig 1985, 1994), and here I will focus on the evolution of language and science. First I will recapitulate the main features of the model.

2.  The pattern

My study assigns the developmental processes of individual human beings a central role in the analysis of biological and cultural evolution. The role of development in biological evolution stems from the fact that genes carry and transmit instructions for self-replication and for development, not for evolution, and because organic evolution proceeds as a result of continuous modifications in the developmental program. The continuity of this process means that many vestiges of early evolutionary stages are stored in the genes and manifested as phenotypic developmental stages in the present-day embryo. Or, as Darwin long since observed, “the embryo comes to be left as a sort of picture, preserved by nature, of the former and less modified condition of the species” (1859, p. 338). Such common stages or traits, called parallels, are thoroughly examined by Gould (1977).

Likewise, the continuity of the transmission of cultural traits over generations means that stages from our cultural history are now discernible as developmental stages in the present-day child and adolescent. Such parallels in the cultural realm have long been observed and discussed, and are also surveyed in Gould’s review. In the field of mathematics and physics, Piaget and Garcia (1983) have reported a thorough study in which they find striking coincidences.

I have earlier (1994) analyzed the relationship between individual development and the evolutionary process, taking as a point of departure the pattern that appears when the developmental age at which each trait appears is compared to the period of time that has elapsed since it first arrived on the scene in evolutionary history. The essential features of the model appear in Figure 1. The pattern encompasses both the organic and mental aspects of human development, and indicates an intimate relationship between biological and cultural evolution.

Figure 1: diagram of ontogeny vs. phylogency in the lineage of humans
Figure 1. Diagram of ontogeny versus phylogeny in the lineage of humans. The ontogenetic age is measured from the moment of fertilization. The ordinate also gives the individual’s postnatal and fetal ages. Both scales are logarithmic. The diagram is a revised version of a diagram appearing in Ekstig (1985), and the reader is referred to that reference for comments. It should be mentioned that the uncertainty regarding the Phonetic organs point is very great. If the phylogenetic age is assumed to be one million years, which is not unrealistic, the point will be situated on the line. (Click on figure for larger version)

As is immediately obvious, the diagram includes very few of the traits found in a growing organism. The principle behind the selection of the traits forming the pattern, organic as well as cultural, is that these traits appear in an inflexible order in the individual’s developmental program as well as in the evolutionary process. I suggest that such traits be denoted cumulative traits.

Organic evolution proceeds through the successive addition of novel developmental traits, thus forming the evolutionary pathway of the species. These novel traits predominantly manifest themselves towards the end of individual development, i.e., just before sexual maturity, because, by and large, the progeny’s genetic constitution is not influenced by selection exerted on the parent after conception. These novel traits are called terminal additions, and since they manifest themselves before sexual maturation they tend to lengthen the development process and therefore also generation length. However, there is a counteracting process as well.

3.  Condensation

Lineages with shorter generation lengths increase their relative number in the gene pool. Hence there is a ubiquitous selection pressure favoring shorter generations, as analyzed in life-history theory. Stearns (1992) shows that for organisms that mature earlier, generation length is shorter and populations grow more rapidly. There is thus a trade-off between terminal additions and acceleration of maturation. Delayed maturation caused by terminal addition of a novel trait is selectively favorable only if the benefit of the novel trait is greater than the loss in fecundity due to the resulting prolonged generation length.

Shortening of generation length may be realized by eliminating unnecessary traits or by gradually shortening early developmental stages. Such gradual shortening of developmental stages is called condensation in the literature on evolution (the term “condensation” is also used in psychology, but with a different meaning).

Condensation and terminal addition act simultaneously and independently. Thus even if generation length increases—the most common trend—condensation may simultaneously have been at work. This implies that it is difficult to separate out and individually observe or measure the two processes. However, in the lineage of humans, condensation is indirectly revealed even by superficial analysis of the diagram in Figure 1. With time, all points in the diagram are displaced to the right, but due to the logarithmic form of the time scale, at unequal distances in the diagram. Therefore, linearity would not prevail unless the points were simultaneously displaced downwards at an appropriate rate. This reasoning rests on the assumption that the pattern will persist, because it is very improbable that it should appear just in our own time. I presume that the displacement of the traits towards earlier ages is accomplished by condensation, a conclusion that, in the case of biological evolution, is made plausible by the general selection pressure discussed above. The model also predicts the action of condensation in the cultural realm, and this subject will be explicitly discussed later in this paper.

One may conclude that the condensation of an evolutionarily old trait, appearing early in development, is less than that of a more recent trait appearing later in development. This seems reasonable, since one may assume that condensation of a specific trait is harder to accomplish the more condensed the trait has already become. Actually, mathematical analysis of the pattern revealed in Figure 1 (Ekstig 1985) indicates that the rate of condensation for a particular trait is inversely proportional to the age of the trait. For instance, the condensation of the development of the embryonic heart is 3% over 10 million years, while Newtonian science is condensed at a rate of 5% per decade. Thus, according to the present model, condensation of the manifestations of memes is about one million times faster than that of genes. This does not imply a claim to measure the rate of the evolutionary process; it only indicates the relative rates of change of biological and cultural evolution.

4.  Parallels in cultural evolution

This study will attempt to apply concepts that have emerged from a biological interpretation of the pattern in Figure 1 to the evolution of human culture. Therefore, my analysis of cultural evolution will bring the child’s mental development to the fore and examine how successive changes in the child’s developmental program are initiated by different modes of selection pressures and how they add up to evolutionary changes.

A conspicuous feature of the pattern in Figure 1 is that it includes the realm of cultural evolution, forming a continuous extension of the trend established by biological evolution. Indeed, just because organic change happens so slowly, compared to cultural change, as to be imperceptible over human time scales, it does not mean that biological evolution does not proceed simultaneously with cultural evolution. Cultural evolution comprises the cumulative development of many cultural manifestations involved in complicated interactions. Nevertheless, a few of the multifarious traits, namely cumulative traits associated with language and science, form the pattern. Other well-known cultural manifestations, such as mythologies, religions, and political ideologies, are not relevant to this discussion since they are not cumulative.

The selected traits exhibit parallels between their manifestation in the individual developmental process and in cultural history. This point is crucial to present model. Consider the ability to perform arithmetic calculations using whole numbers. This faculty emerged several thousand years ago and was later developed into the ability to handle the more sophisticated rational numbers. For an individual human it was and still is necessary first to be acquainted with whole numbers. As a matter of fact, it is generally impossible to omit a former stage of mathematical knowledge before acquiring a latter one, nor is it possible to study math issues in reverse order. This is characteristic of cumulative cultural traits. Therefore, the individual acquisition of these stages follows the same order in which the corresponding stages are observed to appear in the historic record. This cumulative characteristic is not related to psychology, education, or any other external factors, but is a logical necessity inherent in many mathematical concepts. It is only in cultural features possessing cumulative developmental stages, I repeat, that one can expect to find such parallels.

Not all stages of mathematical development, however, are strictly cumulative in the present sense. Thus negative numbers can be acquired somewhat earlier by modern youths than could be expected from their first appearance in history. This illustrates the fact that the points in the diagram are not arbitrary, but are selected according to the criterion of being cumulative.

During the course of cultural evolution new cumulative cultural manifestations are added to the individual’s developmental program analogously to the way terminal additions are added in biological evolution. Thus the time needed for a child’s mental growth has increased, and this can be broadly verified by referring to the evolutionary record. The infants of chimpanzees, to whom our ancestors are supposed to be related, reach adulthood faster than human children do. Relying on the continuity of the evolutionary process, this indicates that human childhood — viewed from a long-term perspective — has grown longer.

Simultaneously, according to our present approach, condensation has been acting on many cultural traits acquired before maturation, resulting in a continuous shortening of cumulative cultural stages. This hypothesis will be discussed with specific reference to the evolution of language and of science.

5.  The evolution of language

Noam Chomsky is reluctant to accept a Darwinian interpretation of the evolution of language, and instead suggests that an inherited universal grammar is the crucial, operative concept. Gould concurs, and further contends that “the universals of language are so different from anything else in nature…, that origin rather than as a simple advance in continuity from ancestral grunts and gestures, seems indicated” (Gould 1989, p. 14). Dennett (1995, pp. 384-400), however, vehemently rejects such non-Darwinian explanations. In considering language as an instinct, Steven Pinker tentatively approaches natural selection in suggesting that if language is like other instincts, it presumably evolved by natural selection (Pinker 1994, pp. 354–355).

Terence Deacon (1997) has articulated a more comprehensive Darwinian framework in his analysis of the evolution of language; I will take this as a point of departure for my discussion.

Deacon accentuates the role of development by stating that children are the vehicle by which a language gets reproduced (ibid., p. 109), and that “the structure of a language is under intense selection because in its reproduction from generation to generation, it must pass through a narrow bottleneck: children’s minds. … Language operations that can be learned quickly and easily by children will tend to get passed on to the next generation more effectively and more intact than those that are difficult to learn. So, languages should change through history in ways that tend to conform to children’s expectations; those that employ a more kid-friendly logic should come to outnumber and replace those that don’t” (ibid., p. 110).

Deacon indicates the existence of a mutual adaptation between language and humans, children in particular, in declaring that, “we should not be surprised to find complex human adaptations to language on the one hand, whose purpose is to ensure that language is successfully replicated and passed from host to host, and language adaptations to children on the other, whose purpose is to make languages particularly ‘infective’ as early as possible in human development” (ibid., p. 113). He thus arrives at a notion of language as an autonomous process: “The evolution of symbolic communication ... created a mode of extrabiological inheritance with a particularly powerful and complex character, and with a sort of autonomous life of its own” (ibid., p. 409).

If we assume that such an autonomous language has evolved as a result of some kind of selection process, we must make intelligible the very basis of such a selection principle. To this end we may take hold of Susan Blackmore’s suggestion: “I have assumed that people will both preferentially copy and preferentially mate with people with the best memes — in this case the best language” (Blackmore 1999, p. 104). Blackmore’s assumption thus implies two kinds of selection, one acting directly through imitation, a central principle in her works on memetics, and one through sexual selection.

I will now discuss these ideas in light of the present model, and attempt to distinguish between two particular evolutionary mechanisms. The first mechanism comprises the reproductive benefits induced by language, in other words, language as evolved by natural selection. The second comprises the mechanisms that might have caused language to evolve without reference to any influence on the genes of the host, the speaking human. As a consequence of these, I will introduce a third mechanism that has to do with the time children have available for learning.

The existence of the first mechanism is commonly accepted. It is associated with conventional natural selection and thus coupled to genetic changes, in particular those implying the growth of the human brain. I would like to be more specific about natural selection. Following Blackmore, I think that in the case of language, sexual selection is the most powerful mechanism; I will outline this view as follows.

By and large, an adult of high verbal ability will gain a high social ranking. Such people are supposed to enjoy more frequent access to mates and certainly also better access to food and other necessities of survival. In other words, they have greater reproductive success than the average citizen. But such high verbal ability is made possible by a genetically determined high brain capacity, and hence the selection on behalf of language also benefits genes for great brain capacity. In Deacon’s words: “The remarkable expansion of the brain that took place in human evolution ... was not the cause of symbolic language but a consequence of it” (Deacon 1997, p. 340).

In attempting to tackle the question of language evolution in terms of memes, Susan Blackmore has apparently come up with a similar line of reasoning. She suggests what she calls a meme-gene co-evolution that functions as follows. People are assumed to mate preferentially with those possessing the best language ability. “These people then pass on geneticallywhatever it was about their brains that made them good at copying these particularly successful sounds. In this way, brains gradually become better and better able to make just these sounds” (Blackmore 1999, p. 104, emphasis in the original).

Of course, other factors quite independent of language contribute to the selection pressure for a large brain, factors such as skill at hunting, foraging, tool making, and warfare. These are mainly coupled to natural selection and contribute to the co-evolution of language and the brain. Actually, one must assume that such abilities were the target of selection for increased brain capacity before humans crossed the symbolic threshold, to use Deacon’s terminology.

My second item, language evolution without reference to influence on genes, brings us back to Blackmore’s suggestion that language is acquired by imitation. The important point in this view is that one must assume that the best speakers have the greatest impact. In other words, imitation implies a process of selection: the best speakers are selected as models.

This process is most important in children’s learning. In their imitative activities, children are strongly influenced by peers with more developed language ability, in particular by older children. But of course, children predominantly learn from their parents and this circular coupling, parents-children-parents, opens up the possibility of an additional mechanism that I will call feedback.

Let us imagine a human being at the dawn of humankind who is endowed with a somewhat better than average talent for the primitive spoken language of the time. This person probably influenced his or her children such that they acquired language more efficiently than did other children. When these children became parents the process was repeated, but, as one may assume, at a somewhat higher level. Such positive feedback improves the language ability of those participating in it. Of course, one can also imagine a negative feedback process having a corresponding opposite effect. Thus, although not having any direct total effect on the evolutionary rate of change, such feedback affects the degree of variation in language ability that, in a second step, influences the rate of change as follows.

Every feature exhibits a certain variation, formed as a Gaussian clock curve, as measured with regard to its fitness in the actual environment. Natural selection works on this variation both by favoring the best-fit features and by disfavoring the less fit. In this way, the clock curve is successively displaced, which illustrates the continuity of evolutionary change. Obviously, the selection pressure is stronger the more a feature deviates from the mean value, and this has an important bearing on the functioning of the discussed feedback process in the evolution of language. Both the positive and the negative feedback processes contribute to broadening the distribution curve of language ability. However, since the selection pressure is strongest on the most deviating features, such broadening will strengthen the selection pressure. So when considering the evolution of language as a selection process at which people preferentially copy those with the best language, we may conclude that the feedback process will increase the pace of evolution towards higher language ability.

The suggested feedback process differs somewhat from what is characteristic of feedback circuits. In an electronic amplifier, say, a limited part of the output signal is fed back to the input of the same unit and causes a reinforced amplification. In the feedback that reinforces the evolution of language, the information is fed back to children in the next generation, and thus not to the same “unit.” However, to label the suggested mechanism as feedback is legitimate, I think, because of the otherwise great similarity in function.

Of course, someone could object that other animals’ ordinary reproductive processes also involve such feedback in the transition from parent to offspring. However, I think that the difference lies in the learning process human children undergo in childhood. This learning process enhances parental influence and it is that very process that may be seen as analogous to amplification in electronic amplifiers, whereas in animals the offspring’s behavior is determined mainly by its genetically unchangeable constitution.

Before leaving the discussion of the evolution of language I will suggest a third process that is a consequence of the mechanisms so far discussed. As we have seen, these mechanisms imply enhanced childhood learning and thus earlier language acquisition. This contributes to the evolution of language as follows. By acquiring language earlier in life the child will have more time to practice talking before adulthood, hence improving its verbal competence as an adult.

To sum up, there are two categories of mechanisms that I believe underlie the evolution of language. The first is the co-evolution of language and the brain, and this has an impact on genes. This mechanism works mainly through sexual selection. The second mechanism has no direct impact on genes, but works through imitation and an associated selection process. In addition to this selection process I have suggested a feedback process that either enhances or weakens parental influence, thus reinforcing the selection pressure and increasing of the rate of evolutionary change in favor of higher language ability. Finally, I have proposed a third mechanism according to which the enhanced and hence earlier acquisition of language lengthens the time children have available for learning. This mechanism would also contribute to increasing of the rate of evolutionary change.

The suggested processes are mutually reinforcing, resulting in progressively enhanced language sophistication. They also demonstrate the importance of children’s learning for the evolution of language, thus supporting the present model of condensation as an intrinsic principle of all evolution.

6.  The evolution of science

Let us now examine the most recent phase of cultural evolution, the evolution of science. I use the word science to refer to the natural sciences, including mathematics.

There is a widespread notion that science evolves in a way analogous to biological evolution. Thus Hull (1988) suggests that scientific ideas develop in lineages in an evolutionary process. A similar notion is expressed by Plotkin (1993), in his discussion of science as a product of a “Darwin machine.”

Holton and Brush (1985, p. 196), in their cogent account of the evolution of scientific concepts, analyze the analogies between the biological and scientific evolutionary processes in a most clarifying way. These authors compare the evolutionary mechanisms of species and those of science, finding four common points.

First, both processes presuppose continuity. This means that a species or a science can persist only if there is some stable means for handing on its structure from generation to generation. In science this continuity is identified as the operational and quantitative nature of concepts.

Second, there is the mechanism of mutation, the opportunities for individual variations. In science, these variations are assured by the boundless creativity of the human mind.

Third, there is the mechanism of multiplicity of effort. Science and species alike must rely on a large number of individual attempts from which ultimately come those few that turn out to be useful. The innumerable pages of scientific research documentation of past years testify to the wastefulness of this process.

Fourth, scientific theories are subject to a selection process not unlike Darwinian selection acting on mutant forms. Scientists create a multitude of competing theories and concepts; these are subject to tests in which the internal rules of science are applied, as are external conditions such as applicability, potential for further development, or contribution to social welfare.

It must be emphasized that the type of selection acting on concepts is not coupled to selection acting on human genes. In other words, there is no coupling between scientific ability and reproductive success, as we posited in the case of language. A scientist does not give birth to more children than the average citizen. I therefore conclude that the leash between genes and culture, to use Edward Wilson’s oft-cited metaphor, is broken as far as the scientific part of culture is concerned. Another consideration is that science has only been in existence for a few thousand years, too short a time to have had any influence on genes.

Though the similarities between scientific and biological evolution as discussed in the cited literature are descriptive analogies, that does not mean that they lack explanatory power. Selection in particular is commonly regarded as the decisive cause of scientific progress. In our own time such selection is intentional, performed mostly by scientists themselves. However, I doubt that such intentional selection was really operative in earlier epochs during which scientific thinking was tentatively forming its own character.

If one wants to posit a non-intentional selection process acting on the early evolution of science, one must pose the critical question: who would benefit from this selection? Since the benefit cannot be for the genes, the remaining possibility is that science has evolved to its own advantage, an idea in line with the notion of how memes work.

When humans passed the threshold of the symbolic, they obviously increased their ability to imagine abstract concepts. But in mathematics and science the employment of abstract thinking is still more accentuated, mathematics being totally abstract. The fact that we nevertheless put considerable effort into these activities shows, I believe, that such conceptual thinking must have increasingly attracted people. It must have been, as it still is, irresistibly attractive to the human mind to create the abstract conceptions that have led to the metaphysical notions of mythologies and religions, as well as to the logical and rational explanations of natural phenomena that have led to science. Deacon (1997, p. 421) reminds us in this context of the story of Archimedes running naked through the streets yelling “Eureka!” This myth illustrates the fascinating experience of recoding a familiar observation into abstract scientific concepts, and I think such experiences have had vital influence on the evolution of scientific thinking. Even more importantly, children’s minds must also have developed such a fascination in understanding abstract concepts.

We find another approach to selection in the reasoning Terence Deacon applies to the case of language. Regarding science, we may tentatively infer that scientific concepts are under intense selection pressure: in their reproduction from generation to generation, they must pass through a narrow bottleneck — children’s minds. Scientific concepts that can be learned quickly and easily by children and adolescents will tend to get passed on to the next generation more effectively and more intact than those that are more difficult to learn. For example, we may consider the invention of the position system that made arithmetic calculations much easier. This efficient although more abstract system quickly replaced the Roman way of denoting numbers and thus contributed to the evolution of mathematics.

Scientific concepts are, I believe, typical memes and as such must have proliferated because they are attractive to the human mind. This is also selectively advantageous for their own reproduction in their specific environment, the human brain. In this enterprise they have continuously been exposed to competition with other memes, most pronouncedly, with those of religion. It is thought provoking to interpret White’s (1896) classic account of the “warfare” — to use his own word — between science and religion as a struggle between memes.

The discussed process works by adapting concepts to children’s learning abilities and also, possibly, by adapting teaching content and practices in view of the requirements imposed by science itself. As a result, the learning of specific scientific concepts is being accomplished at successively earlier ages. In other words, our discussion supports the notion, as predicted in the present model, of the condensation of scientific concepts.

Let us discuss a second example. Euclidian geometry in its original form has been taught to children for about two thousand years. Nowadays in many countries such axiomatic geometry is replaced by abridged versions. This is because children can generally learn such a simplified geometry in less time, allowing room in the curricula for other parts of mathematics that have developed in the meantime and are considered as more important.

Earlier learning enables young adults to begin contributing to scientific knowledge from an earlier age, and hence contribute for a greater total time. These people will thus exert a greater than average impact on the growth of science and, presumably, on education.

The accumulation of concepts is equivalent to the terminal addition that we discussed in the case of biological evolution. By and large, this process causes a prolonged learning period in spite of the superimposed action of condensation.

It should be noted that the present model does not claim to cover the modern phase of scientific evolution, i.e., the last three hundred years or so. In our own time, however, a school system has been developed with the explicit purpose of, among other things, enhancing science learning. Moreover, recognition of the importance of science education has in recent years resulted in the development of the specific research disciplines of mathematics and science education, the aim of which is to improve instruction methods and thereby, of course, learning. Extensive educational projects with such a purpose have been developed, such as Project 2061 (AAAS 1989). No doubt, the school system has strongly enhanced concept learning in science, presumably also causing a rapid increase of condensation.

To sum up this discussion of the evolution of science, in addition to the general notion of intentional selection, for earlier epochs of science history I have suggested two co-operative mechanisms for the selection process. The first is a self-selection of scientific concepts for their own benefit in their own environment, the human brain. This can be considered a memetic explanation. The second is related to the way children and adolescents acquire scientific concepts, implying an adaptation of concepts to those most easily learned. This second mechanism leads to earlier acquisition, thus extending the individual’s active period as scientist, and hence contributing to the progress of evolution.

7.  Summary

Analysis of various manifestations of evolutionary processes in the human lineage indicates some common principles. One such principle is the crucial impact of development as manifested in embryonic development, in the verbal development of the child, and in the conceptual development of the adolescent mind. These developmental processes are characterized by a specific kind of trait, cumulative traits, which turns out to reveal a pattern that unites the evolutionary process in its biological, verbal, and scientific manifestations. This pattern is a consequence of condensation, a regular shortening of developmental stages. An explication of condensation remains to be made.

The present model shares central ideas with universal Darwinism in as much as it points out mechanisms common to both the biological and cultural realms of evolution. In so doing it also demonstrates the explanatory power of the notion of the meme, since symbols and concepts, as typical memes, are the cornerstones of culture. There is another, somewhat more speculative conclusion. If the observed trend toward the condensation of developmental stages, maintained since the emergence of early life, is not to be broken in our own time, the continued improvement of science and mathematics education is imperative.

I have also discussed several principles for the selection of information to be transmitted between generations in the various manifestations of the evolutionary process. Biological evolution employs natural selection of genetic information. As to the evolution of language, I suggest a process of selection acting cooperatively on both genes and memes. Finally, in the evolution of science, I conjecture that early in the evolution of science memetic selection was in charge, while in our own time we humans have taken over, and carry out an intentional selection. This sequence of principles of selection should be regarded in the light of the pattern visible in Figure 1, whereby one can perceive how the various principles of selection have contributed to the steadily increasing rate of evolutionary change.


American Association for the Advancement of Science (AAAS). 1989. Project 2061: Science for All Americans. Oxford: Oxford University Press.

Arthur, W. 2002. The Emerging Conceptual Framework of Evolutionary Developmental Biology. Nature 415:757–764.

Barkow, J. H. et al. 1992. The Adapted Mind. New York: Oxford University Press.

Blackmore, S. J. 1999. The Meme Machine. Oxford: Oxford University Press.

Blackmore, S. J. 2000. The Power of Memes. Scientific American 283(4):52–61.

Blute, M. 2003. The Evolutionary Ecology of Science. Journal of Memetics—Evolutionary Models of Information Transmission 7(1).  http://cfpm.org/jom-emit/2003/vol7/blute_m.html

Darwin, C. 1859. On the Origin of Species. London: Murray.

Dawkins, R. 1976. The Selfish Gene. New York: Oxford University Press.

Deacon, T. W. 1997. The Symbolic Species. New York: W. W. Norton & Company.

Dennett, D. C. 1995. Darwin’s Dangerous Idea. New York: Simon & Schuster.

Ekstig, B. 1985. The Regular Condensation of Development as a Mechanism of the Evolution of Man and Culture. Evolutionary Theory 7:195–204.

Ekstig, B. 1994. Condensation of Developmental Stages and Evolution. BioScience 44(3):158–164.

Gould, S. J. 1977. Ontogeny and Phylogeny. Cambridge, MA: Harvard University Press.

Gould, S. J. 1989. Tires to Sandals. Natural History April, 8–15

Holton, G. and S. Brush. 1985. Introduction to Concepts and Theories in Physical Science. Princeton, NJ: Princeton University Press.

Hull, D. 1988. A mechanism and its metaphysic: an evolutionary account of the social and conceptual development of science. Biology and Philosophy 3:125–155.

Lumsden, C. J. and E. O. Wilson. 1981. Genes, Mind and Culture. Cambridge, MA: Harvard University Press.

Piaget, J. and R. Garcia. 1983. Psychogenèse et Histoire des Sciences. Paris: Flammarion.

Pinker. S. 1994. The Language Instinct. London: Penguin Books.

Plotkin, H. C. 1993. Darwin Machines and the Nature of Knowledge. London: Penguin Books.

Raff, R. A. 2000. Evo-devo: The Evolution of a New Discipline. Nature Reviews Genetics 1:74–79.

Stearns, S. C. 1992. The Evolution of Life Histories. Oxford: Oxford University Press.

Tooby, J. and Cosmides, L. 1992. The Psychological Foundations of Culture. In The Adapted Mind, edited by Barkow, J. H. et al. New York: Oxford University Press.

Wagner, G. P., C-H Chiu and M. Laubichler. 2000. Developmental Evolution as a Mechanistic Science: The Inference from Developmental Mechanisms to Evolutionary Processes. Amer. Zool., 40:819–831.

White, A. D. [1896] 1960. A History of the Warfare of Science with Theology in Christendom. Reprint, New York: Dover Publications.

© JoM-EMIT 2003

Back to Issue 1 Volume 8