MASON: Thoughts as tools:
the meme in Daniel Dennett's work
BOYD: Should memes and viral-information be considered synonymous terms?
SPEEL: Memes are also interactors
VANEECHOUTTE: The replicator: a misnomer. Conceptual implications for genetics and memetics.
HEYLIGHEN: What makes a meme successful? Selection criteria for cultural evolution.
BEST: Memes and Genetic Opposition
Thu, Aug. 27, morning session: Evolutionary Mechanisms
SZAMADO: Basic Questions
In Memetics:Life-Cycle, Reproduction And Resources
CLEWLEY: Emergence without magic: the role of memetics in multi-scale models of evolution and behaviour.
EVERS: An explanation for general societal altruism consistent with Darwinian natural selection according to the memetic application of Hamilton's rule.
Thu, Aug. 27, afternoon session: Computers and Networks
MARSHALL: The Internet
BOLLEN et al.: Improving Memetic Evolution in Hypertext and the WWW.
HALES: Artificial Societies, Theory Building and Memetics
RIBEIRO & LUTKUS: Real product design assisted by abstract evolution
SAKURA: History of Sciences as a Division of Memetics? Implications from Comparative Comparative Studies on the Reception of Sociobiology.
Fri, Aug. 28, morning session: Social Sciences
DE JONG & SPEEL: Mimicry
behaviour in social and organisational life
GREINER: Memes and the creation of new patterns of movement in dance
KATZ: Dance and Evolution: a non-stop combination of biology and culture
MARSDEN: Operationalising Memetics - Suicide, the Werther Effect, and the work of David P. Phillips
PITT: Memes and monotheism
JACOBSON: Bridging the gap: Memetics as a methodological tool to close the ranks between social and traditional history
Fri, Aug. 28, afternoon session:
Heylighen, Vaneechoutte, Speel, Marsden & Boyd: Concluding Panel Discussion: What is Memetics?
From the scheduled symposium program, only 3 contributors did not make it: Liane Gabora, who had apologized because she was too busy preparing her long term visit to Belgium, Thomas Quinn and Koen Bruynseels. On the other hand, a guy whose name I don't remember (?Rosdeitcher?) presented an improvised, but entertaining talk in which he sketched his own "conversion" from being a follower of Ayn Rand's "objectivist" philosophy to becoming an advocate of the memetic/cybernetic paradigm.
From the other, scheduled talks, I particularly appreciated my co-chair Mario Vaneechoutte's speculations on the origin of life as a model for memetics, Michael Best's simulation of cultural vs. genetic evolution, Szabolcs Szamado's analysis of fundamental memetic replication processes, John Evers' application of memetics to explain altruism and Paul Marsden's review of research on "social contagion" as an existing body of empirical data that cries out for a memetic reinterpretation. The talk by my PCP collaborator Johan Bollen about our learning web algorithms also generated a very positive response, although I am of course not in an objective position to judge about its quality (;-). Practically all papers are by now be available on the web (see below). They will be published by the end of this year as part of the congress proceedings.
The symposium was concluded with a lively panel discussion, chaired by Gary Boyd, in which the absent panel member Gabora was replaced by Paul Marsden, and a short brain storming session with all remaining participants to generate a list of suggestions for us to advance the field of memetics. One of the concrete decisions was to steer the Journal of Memetics more in the direction of the system of commentary used by "Behavioral and Brain Sciences".
Michael L. Best
Our work to date has explored directly a number of these questions. Our principal method of inquiry has been computational methods of analysis and simulation.
We have studied the microevolutionary dynamics of memes within natural language discourse (Best 1997a; Best 1997b; Pocklington & Best 1997; Best 1998a). Since, "language is the best approximation of pure culture," (Gerard, Kluckhohn, & Rapoport 1956) an inquiry of discourse can illuminate issues fundamental to culture at large. Through techniques of information retrieval we have built computational tools which locate replicating units of information, track their movements through discourse elements, and measure their interactions. Thus, we have built a computational toolkit to study ecological microevolutionary phenomena within culture -- population memetics.
We have also, through a series of simulations, examined the relationship between the adaptive systems of organic evolution through genic selection, individual learning, and memetic evolution through cultural selection (Best 1998b). In the social animals these three adaptive systems interact in rich and ill-understood ways. To date our studies indicate that social learning converges on adaptive goals faster then individual learning within our simulation environment. Further, social learning is no real match when pitted directly against organic evolution.
Our point? Continued progress on the fundamental questions of memetics require theories with operational definitions along with supportive evidence acquired through direct analysis or simulation. Computational approaches are the only practical and promising method to acquire such evidence.
Memetics is the science of how ideas or units of information propagate through a population of human carriers. In analogy to genes these units of information are referred to as memes (Dawkins, 1976). Like their biological counterparts, memes are presumed to be shaped by a process of variation and selection (Heylighen, 1996) in which the most fit variations are more likely to be replicated and hence retained in the population while the less fit variations dwindle and eventually die out.
Two important processes can be distinguished in the propagation of memes. First the meme needs to be stored in the carrier's memory where it is processed and integrated into the already existing store of memories. Most established models of human memory are based on semantic and associative networks (Klimesch, 1994; Anderson, 1973). Since memes are stored in human memory this also seems to be the most appropriate model of meme structure. It is also compatible with the notion that memes undergo mutations and partially change form. Secondly, the meme has to be communicated (replicated) to other potential carriers, after being converted from memory into a form that renders it suitable for communication. This conversion is necessary because most means of communication like speech, books or films use a linear format while memes are most probably stored as semantic networks.
The WWW, on the other hand, sets itself apart from other media because it was explicitly designed to mimic the way information is stored and processed in human memory (Bush, 1945). Hypertext pages express ideas and concepts in the distributed structure of their links, similar to how semantic networks represent information. Consequently, the WWW's format of storage closely resembles that of memes in human memory. This makes the WWW more than just an efficient means of global communication; it is also an ideal medium for meme storage and representation.
Memes stored on the WWW could be said to undergo the same process of variation and selection as other memes do. Variation occurs when memes in the network connect to new pages and shift their links from one page to another. The WWW is deliberately designed to enable these kinds of changes. Selection upon this variation however only occurs in a rather indirect form, i.e. by retrieval frequency and selective linking (Heylighen, 1996). Even the least fit WWW memes can remain to be linked and retrieved simply because they are e.g. the only path to other more fit memes. As a consequence, memetic evolution on the WWW is slower and less efficient than it could be (considering the WWW's electronic and global nature).
We have implemented an automatic system that we believe can speed and improve memetic evolution on the WWW. The system introduces variation in the network by adding links that are logically presumed to be beneficial for network structure (transitive and symmetric closure). The system then applies selection by tracking user retrieval paths in the network and rewarding often-traversed connections. Users implicitly indicate which memes are most fit by simply retrieving them more than others. Dynamic network structure is in its turn fed back to the human browsers of the network by an automatic ordering of links from a certain page according to decreasing link strength. This ordering decreases the probability that connections to less fit memes will be selected because they are more likely to be positioned lower in the list than the more fit memes.
We believe the system tightens the feedback loop between variation and selection in the WWW and will speed the memetic evolution of ideas on the Internet. The system has also been shown to improve the mapping between network structure and users' cognitive maps (Bollen, 1996). This property might enable researchers in the field of memetics to more reliable measure meme structure as well as track the different mutations and variations of memes that might occur over time. As a consequence the system can also be applied as a research tool.
It is also argued that the Receiver-Uncertainty-Reduction definition of information developed by Claude Shannon can fairly easily be extended to provide a measure for memetic information the other more evolved forms of Viral information. This is important because the power of such concepts as signal/to/noise ratio and its correlate requisite variety, can then be quantitatively employed in memetics.
Noösphere - H. Consciencestr. 13 - 3000 Leuven - Belgium
The broad definition of a 'meme' [Dawkins, 1976] hardened the verification of the applicability of the meme-concept. Much attention is drawn by the influence of ideas on the physical behaviour of a person [Lynch, 1997], complicating a clear sight on ongoing processes. Application of memetics to situations of restricted communication can avoid obscuration by real-life complexity. Internet communication - which is mainly restricted to writing and reading texts - is a particularly suitable model, since it enables direct quantification of communication-events. We constructed a model for an evolution of ideas, based on the textual-iconographic communication on the internet [Bruynseels et al, 1998]. The model will be available on-line, and enables visitors to create a well-defined `meme'. The model functions as an `open' cellular automaton, with hierarchically ordenend functional units, on which genetic selection rules (both based on user-input and morphological characteristics of memes) are applied.
Here, we discuss different organisation structures in the modeling of the genetic selection rules. Both a 1D and 2D cellular automaton approach is discussed.
GENERAL STRUCTURE OF THE MODEL
Visitors at the web-site can construct a meme, consisting of a textstring (slogan), and an attached 'life-form' (composed from a set of feelers, bodies and tentacles; in different colors). The memes are stacked in 'orbits', these are containers (chromosomes) which can carry 6 memes. In the 1D model, the orbits are organised in a linear array, from 'low level' to 'high level'. In the 2D model, the orbits are organized hierarchically tree-alike (Fig.). Starting from the upper orbit, each orbit has two descendants. In both models, new memes fill up the lowest uncompletely-filled orbit.
Determination of the `fittest'. Authors of memes can evaluate other memes, by giving them a + or - sign. These evaluations are recorded to compute a score for each meme. Besides this, the server calculates which memes within a given orbit are morphologically most similar. The score of these memes is increased relative to the other memes. The relative contribution of voting behaviour and morphology is expressed as a factor () which can be adjusted during evolution. After each cycle (15 min), all memes will be redistributed over the orbits according to their scores.
1D model. The 3 most successful memes in each orbit n wil move to the orbit above (n-1); the other 3 memes will drop one orbit. (n+1). The highest orbit is at n=1.
2D model. In a given orbit 2n, the two `fittest' memes wil go to the orbit above (n), while the four unsuccessful memes are distributed over the two lower orbits (4n and 4n+1). Boundary conditions are obvious: no memes are distributed to lower unfilled orbits, and the `fittest' memes in the upper orbit remain at their place.
The 'reflux' or 1D model has the advantage that all memes will meet
each-other on their way to the top; giving a maximal 'collision' of ideas.
The system however is not flexible enough to enable the existence of long
lasting 'cooperations' of memes. The 2D 'hierarchical' model gives a stronger
validation of success/failure, since the way to the top level is shorter
for the same amount of memes. This might improve coöperation and competition.
Fig. Reorganisation of memes over the orbits in the 2D model.
The memes (numbers 1-6) in an orbit (green areas) are divided into 2 successful and 4 unsuccessful ones. The former go up one level, to the orbit above. The 4 unsuccessful memes are distributed at random over the 2 descendant orbits.
In this context, the nature of mathematical models in memetic theory is considered. In particular, it is argued that this class of models occupy a single position in a hierarchy of models of human society, mind, and the brain. A consequence for memetics is that it may not provide a comprehensive MR theory of so-called cultural evolutionary phenomena.
This analysis of the ontological status of memes is based on general philosophical arguments which are then applied to other apparently "emergent" structures as used in theories of evolution and self-organisation. However, the positive role of abstract concepts in an approach integrated with related disciplines is also recognised. The method of integration is explored, using ideas from modern theories of nonlinear spatially-distributed dynamical systems to indicate how it may be possible to rigorously connect levels of mathematical representation in a model hierarchy. The close relationship of these ideas to the metaphysical concept of a Meta-System Transition is discussed.
All of the preceding species have found ways to increase their survival chances by adapting purely physical aspects. In fact, they appear to have organised a masquerade to evade dangers posed by other organisms in their selective environment. If we define selective environment (Brandon, 1988; Speel, 1997) as all aspects in the ecology an individual, population, or species that weed out the weaker exemplars, we can say that mimicry is an interesting way to cover up one's weaknesses and deactivate the selective forces related with the perceptual ability of others that are a part of the selective environment.
This article reconciles intra-familial altruism with extra-familial altruism by applying Hamilton's rule to an altenative unit of natural selection: the meme. The article demonstrates that memetically encoded altruism can be exhibited in favor of any member of a freely communicating population (or, 'memetic family') and can survive natural selection according to Hamilton's rule. To apply Hamilton's Rule to the meme, the following demonstrations are made: first, that the meme is, in fact, a unit of natural selection; second, that memes replicate horizontally (i.e., accross family boundaries rather than down family trees); and third, that memes are capable of producing altruistic behavior. Finally, a modified version of Hamilton's Rule is provided which can account for the possibility of memetic 'conversion', a phenomenon which results from horizontal reproduction and allows for the transformation of a carrier organism from competitor to clone.
The topic of cultural origins is generally approached from an archeological perspective. For example, by dating artifacts such as tools, we learn approximately when humans acquired the ability to make and use these tools. Memetics offers a different approach. It focuses on the information structure undergoing this second form of rather than on the human host. The memeticist addresses the topic of cultural origins by asking how the selection, variation, and replication/transmission of memes could have established itself in the human brain.
The analogy between biological evolution and cultural evolution is, of course, imperfect. However, since we know more about the former than the latter, it can be helpful to use our knowledge of biological evolution as a scaffold against which to direct inquiry into the cultural analog. The origin-of-culture model proposed in this paper was inspired by a plausible and innovative model put forward by Stuart Kauffman to explain the origin of life.
The origin of life is something of a paradox: if living things come into existence when other living things give birth to them, how did the first living thing arise? That is, how did something complex enough to reproduce itself come to be? Reproduction is orchestrated by an intricate network of interactions between proteins and DNA; proteins are made by decoding DNA, and DNA requires the catalytic action of proteins to get decoded. So the question becomes: how could a system composed of complex, mutually-dependent parts come into existence? Kauffman proposed that life began with the emergence of a set of autocatalytic polymers, which has subsequently been shaped into life-as-we-know-it by natural selection. None of the polymers can catalyze its own replication, but there exists a catalytic pathway to the formation of each polymer in the set.
In this paper I show that the origin of culture presents an analogous paradox. The work of Karmiloff-Smith and others seems to indicate that we are endogenously driven to find progressively more abstract recodings of memes by recursively redescribing them, and Donald  argues convincingly that the ability to do this came about during the transition from episodic to mimetic culture with the arrival of Homo erectus about 1.5 million years ago. However Donald leaves us hanging as to what sort of functional reorganization could turn an episodic mind into a mimetic one. Thus the paradox of the origin of culture can be stated as follows: How does a mind come to assume a self-sustained stream of representational redescription that progressively shapes, and is shaped by, an internal, structured model of the world, or worldview? This paper proposes a tentative model of how this comes about through the formation of abstractions/category memes from episodic memes, which culminates in the emergence of an autocatalytic meme network.
The longer paper will expand on this model, illustrate it with data and make links to the sociological literature on persistence, in particular Giddens' theory of structuration.
Dennett (1996:155) explains that mental contents become conscious not by entering some special chamber in the brain but by winning the competition against others.
In dance, memes, as mental representations and the process of replication of these representations of ideas (or other imagined constructs), are phenotypically mediated by the dancing body. This paper points out that cultural dialogue in dance is often responsible for the creation of new patterns of movement in the dancing body, but not by a sociological effect of the external world on the inner world of the dancer, but by natural selection of the competition among mental contents, meme-effects in brains. It's the same process that is present in all nature: an algorithmic process (Dennett,1995,1996; Dawkins,1986,1995,1996) creating new designs. Japanese Butoh Dance and a reflection on the three phases of evolution (selection, variation and transmission), according to the pioneer of butoh dance Tatsumi Hijikata and his conception of the dead body, will be used to illustrate the presentation.
Artificial societies comprise societies "as they could be" not just societies as they are". It's remit is the "space of all possible societies" and as such it's concerns are with those common features and processes found in many social systems: co-operation, altruism, specialisation, hierarchy, group formation etc.
A brief overview of current work in artificial societies is given.
It is argued that artificial societies can be constructed to aid memetic theory building through experimentation. The ultimate aim is to produce robust theories and testable hypotheses which can be empirically tested in the real world.
Artificial society models are particularly applicable in memetics because the complex interaction of memes within real societies makes the deduction of testable hypotheses from existing meme theories difficult. Also the construction of formal models forces the researcher to express clearly their understanding of meme theory which aids unambiguous communication and debate.
Key assumptions on which the societies are based are often parameterized and varied, mapping a space of assumptions to a space of consequences. It is argued that such a process should be viewed as a combination of computer aided deduction combined with an inductive learning process based on a set of consequences (the "third way" described by Axelrod 1997 and "ceduction" described by Hales 1998).
A methodology is presented for such investigations based on the Popperian (Popper 1968) conception of theory refutation through the falsification of hypotheses and consequent theory development. The novel part of the methodology is that this process occurs within a deductive system (the artificial society) which due to it's complexity requires empirical investigations to formulate hypotheses which can then be tested against real societies.
Current experimental work using these techniques will be outlined. This work involves the construction of and experimentation with an artificial society in order to aid the building of meme theory around the processes of stereotyping and group formation.
Results obtained from previous more simplistic models (Axelrod 1995, Hales 1998) suggest that simple meme processes may be sufficient to account for simple forms of group formation, intra-group altruism and inter-group antagonism.
The new model attempts to capture the possible co-evolutionary relationship between stereotyping by individuals, group formation in the population, and co-operation and competition between groups (Heylighen 1995).
Which memes will most successfully pass all these stage can be modelled by a series of selection criteria. These criteria can be grouped in different families, which are defined by the main system responsible for the selection. Objective criteria denote selection by outside phenomena. They include distinctiveness, invariance and controllability. Subjective criteria represent selection by the subject who assimilates the message. They include utility, coherence, novelty, learnability and pleasure.
Intersubjective criteria depend on the interactions between different subjects. They include group utility, conformity, authority, formality, expressivity and publicity. Finally, the meme-centered criteria represent selection on the level of the meme itself. They include self-justification, intolerance and proselytism.
All other things being equal, a meme that scores better on one of these criteria is expected to become more numerous in the population than a meme that scores worse. This is a falsifiable hypothesis, which can in principle be tested through experiments or observations.
Dance is an outstanding event or series of events of variegated ordinariness. Dance cannot be studied apart from a body and its surroundings. But to what extent do surroundings and body guide its behaviour?
This paper aims to consider as evolutive the relation between a body and its surroundings in the sense that Charles Darwin introduced in "On the Origin of Species" (1859) and will take two theoretical instruments to discuss it: the Meme Thesis of Richard Dawkins and the Representational Thesis of Fred I. Dretske, chairman and professor in the Department of Philosophy of Stanford University.
This proposal asks for a calibration of the inflection of our thoughts. And this calibration is the stepping stone to sites of reversible destiny of established truths. Brains are structured in an extremely subtle and complex way of wiring and if we can assume dance as an experience, it must be configured that experience is taken as a representational fact about an informational function (Dretske, 1995).
If dance is information not encoded in DNA, it has to be brought to the body. If Culture is a space where occurs exploration and transformation of information through variation, selection and replication, as Dawkins points, and dance is information with patterns (Gabora, 1997), dance finds itself in a privileged place to the clarification of evolutionary processess and can be understood as an evolutionary process itself. As a non-stop combination of Biology and Culture, dance is a pool of dancing memes and their offspring.
This comprehension builds new statements about old dance questions. The commonly accepted evanescense and pervasiveness quality of dance abandon its auractic realm and falls directly into the sensuous representation of the sensory system of the human organism.
Two centuries ago Goethe's novel "The Sorrows of the Young Werther" was published and widely read throughout Europe. The story, in which the hero commits suicide, was widely held to provoke readers into committing suicide themselves. As a consequence of this "Werther Effect" the book was banned in several areas. For memetics, this event is of significance because it was perhaps one of the first tacit acceptances by those responsible for setting social policy that culture may act like a contagion. However, a century later the sociologist Emile Durkheim rejected the cultural contagion thesis in his seminal treatise Le Suicide, and it was not until 1974 when the Werther Effect became the focus of empirical testing.
Two years before Richard Dawkins coined the term meme, the sociologist David Phillips published an article in the American Sociological Review with evidence to support the claim that suicide could be shown to behave as a cultural contagion. He showed that exposure to suicide in media stories was a significant variable in accounting for US and UK suicide rates (Phillips 1974). Specifically, Phillips correlated media coverage of suicides positively suicide rates during the month following the reported suicides.
In a similar way he correlated aeroplane and car accidents, murder and violent crime rates to mass media reporting. (Phillips 1977, 1978, 1979, 1980, 1983, 1986). In doing so Phillips pioneered, and to a large extent pre-empted the operationalisation of memetics by attempting to map the spatio-temporal spread of behaviour through society. Whilst Phillips made no explicit reference to memetics, he argued that patterns of behaviour in society could be understood as cultural contagions (Phillips 1980) each with specific incubation periods, specific modes, and specific channels of infection. This paper will review Phillips' empirical work, and use his methodology for outlining a general method for conducting empirical memetics.
The operational level. With the World Wide Web, a system of connections (links) is maintained in a data space so that the space can be navigated. But the data space is supported by a network, and the connections of the network do not correspond directly to the connections of the data space. So, in a way that is transparent to users, a connection in the data space must be mapped onto a route in the network, making routing a key factor in network operation. Internet routing not only provides routes, it also provides them in a way that adapts to the state of the network . This adaptation is achieved by network management centres that collect information about the state of the network and then send out messages that change the routing accordingly.
This routing system can be interpreted as a memetic system in which the messages carrying the routing changes are memes and the network management centres act to mutate them to make them more fit for their purpose.
The service level. The use of agents is increasing on the Internet. Their purpose is often to provide, or to help provide, some service, as for example with a search engine. Now, when a number of agents are present in the network performing tasks for their dispatchers, it is possible for them to assist each other by, for example, sharing common tasks, ensuring that pre-requisite tasks are done first, avoiding interference, and so on. But the way that help is obtained by one agent from another can be benevolent or greedy, disruptive or co-operative, aggressive or disposed to negotiate. In fact, some norms are needed , both among the designers of multi-agent systems and, within the network, among the agents. If the normal exchange of ideas takes place among the designers, and some norm emerges, the problem then arises of how to represent and enforce this norm on the network.
The user level. The Internet supports the rapid and accurate world-wide transmission of the memes of its users. It has been argued that this supports the tendency towards a homogeneous world culture . It has also been argued that the speed of transmission makes it more difficult to distinguish between the more and less valuable memes . The widespread use of agents brings another aspect to the situation in that they can be used to control the memes to which their users are exposed, thereby supporting a situation in which narrow cultures are reinforced and society can fragment into numerous incompatible segments. These effects and their interactions will be discussed.
The latter is obviously a serious problem for any enterprise aiming at being scientific: the entities it studies had better be real in one way or another. Although the former, however, might seem irrelevant to "good scientists", who have only to worry about the truth/falsity of their work and not its social consequences, I suggest that it should be taken just as seriously. This is because I also think that we can show the humanist concerns to be misguided, just as we can explore objections to the "tough-minded" claim that "memes just don't exist". The way to do this is through a study of Daniel Dennett's work. While it is well-known that he advocates memetics, it is not so immediately evident that the concept of memes and the issues it raises are centrally tied to the rest of his work.
Moreover, his support of memes fits also with his more general project of dispelling mankind's illusions and yearning for "skyhooks", while at the same time avoiding the dangers of "greedy reductionism".
The first part of this paper explores the way Dennett has always considered thoughts as tools, both explicitly in his written remarks, and implicitly in his entire philosophical method and approach. This conception of thoughts is one that encourages us to take control of them, and use them to their full advantage: if we can now get a science (in the form of memetics) of the ideas that rule us then, according to Dennett, we will just have that much more power over ourselves and our environment.
The "universal acid" of Darwinism might via memetics eat through our cherished ideas of culture and of our very selves, but it will leave behind the chance for a genuine self-understanding and -control.
The second part deals with the ontological status of memes by considering Dennett's general ontological position. He has definitely tried to avoid such armchair metaphysical work, but has not on the other hand gone as far as Richard Rorty in rejecting altogether the question "what is real?". His rejection of language-of-thought theories, such as Jerry Fodor's, might seem to conflict with his claims of reality for memes. The tension between the two is easily dissolved, however, and in fact they turn out both to be entailed by Dennett's general position. For he sees himself not as a radical, but as an adherent of scientism, and orthodox scientism at that; his allegiance to science is what has led him to his "mild realism", which is particularly evident in his philosophy of mind. Just as patterns and thoughts (and haircuts etc.) are real without being uniquely instantiated and must be identified in the context of their environment, so too can memes be as real as electrons without being real in just the same way. The real question is not "how real are they?" but rather "how are they real?" or "in what way are they real?", and we can find the answer in Dennett's work on mind, patterns and intentionality.
I will show that the battle for ideas for survival is a very real one and that ideas evolve, mutate or are defeated.
There are striking parallels with biological evolution but also important differences. Both of which I will illustrate. I will start with a description of the ideological world that gave birth to Judaism, that is the Egypt of the fourteenth century BCE. (There is a debate about the chronology.) Freud argued fairly convincingly that Moses's monotheism was derived from the monotheism of the Pharaoh Akhenaton. After the defeat of monotheism in Egypt the priests had to flee. Moses escaped from Egypt taking with him some Hebrew slaves. His ideas (not without a struggle) were adopted by the Hebrews. The notion of God evolved from the only Sun God of Egypt into the God of a small tribe. The scriptures show just how narrow and how tribal. It evolved further to become the God of the Kingdom of David and Solomon.
After a string of military defeats the ideology evolved into one suitable for a rural community of Jews around Lake Galilee under Roman hegemony. It became one that awaited a son of David to lead the national liberation. This is the area that Jesus lived and preached. His ideas reflect those of that community. After his execution his ideas evolved from one suitable for the anti Roman poor Jews to one suitable for city dwelling gentiles scattered around the Mediterranean.
The Jewish laws about circumcision, and food became ignored by the new church, again not without a struggle. Anti semitism became part of the Christian Church. It had to evolve again from being that of a small persecuted minority to become fit to be the State religion for the Roman Empire. Once the might of the Roman Emperor was behind the new religion church organisation and doctrine changed, massive new churches were built and new bibles were made abundant. It was largely this ideology that survived till the reformation.
Some of these changes are slow and continuous, others are strikingly abrupt. The Bible is a long record of ideological and physical struggle. It is worth studying in spite of inevitable omissions, distortions and mistranslations.
Since the environment of the rich and powerful differs from that of the poor and weak, successful ideas may differ in the various classes. The material power of the rich to support or repress ideas must also be taken into account. Hence I believe that the Marxist approach is essentially correct.
This idea has been developed by taking concepts imported from the reverse engineering and it attempts to infer from the analysis of the external features of an object an auto-referential structure capable of codifying its dimensional and visual aspects.
The way proposed to obtain such results is to try to construe the forms as the outcome of the action of parameters coordinated by a hypothetical genetic code, considered as some sort of genotype, which according to its biological analogue will, in this context, control the phenotype - its visual expression - when implemented by a specific software.
The inspiration leading to this approach comes from experiments and mathematical modeling offered by the artificial life which will allow to incorporate into the product design properties such as morphodynamics, mutation and evolution, resulting from the possibility of information transmission and reproduction inherent to this process.
One of the consequences of this methodology is to foster an alternative way of teaching design, as well as a philosophical and differentiate approach, thus making this project concept to be a strategy which may be considered as an evolutionary design.
Darwinian evolutionary theory was firstly introduced into Japan during the late 19th century, but was accepted as a theory for social evolution, not for biological evolution, as it was in several other countries. We Japanese failed to create a biological Darwinian study program even after WW II. There was no Darwinism in Japan until 1970s. Then arrived Wilsonian-Dawkinsian sociobiology which broke the Japanese "seclusion" of evolutionary biology during early 1980s. We can observe the following two features in the process of the introduction of sociobiology: (1) ecologists in "branch" streams and younger scientists took initiative, and (2) the lack of controversy. The latter shows an impressive contrast against the situation in the English speaking countries. As the reasons for these features, I can point out (a) the absence of Darwinism itself; (b) the lack of opposition from the left, and (c) the absence of a popular science press.
Some preliminary surveys in the reception process among German speaking countries and Korean reveals structural and functional similarities and dissimilarities to the case in Japan. German speaking countries also experienced a long delay in the adoption of sociobiological thought. However, they experienced rather serious controversies of their own. Korea shows another pattern. Classical ethology, sociobiology and even evolutionary psychology were introduced simultaneously in early 1990s. But they started a unique popularization of these ideas, which has not yet occurred in Japan.
Based on these observations, I can hypothesize some patterns among meme pool dynamics. For a more powerful theory of "population memetics", it may be necessary to adopt more concepts from genetics, ecology, and evolution, e.g., island ecology theory and phylogenetic constraint. The fundamentals for analogy building are also required.
I argue (Speel, 1997) that we can and do judge memes by their merits without neccesary interference of the physical world. If we take the definitions of Hull and Dawkins seriously we cannot deny this conclusion.
If one wants to adopt the terminology of replicator, then the following considerations should be taken into account. The information processors referred to above can be defined as replicators: they replicate information. When replicator is used to intrinsically mean self-replication, then it is obvious that such a self-replicating system occurred thus far only once on Earth: the cell. In cells, both processors (metabolic and informational enzymes) and digitally encoded information (the chromosomal genes) form a closed semantic circle, that is an interdependence whereby processors can only be produced by translation of digital information and whereby this digital information can only be replicated by the processors for which it encodes. This enables us to define natural selection as a special case of selection, limited to biology (i.e. cellular replication) and whereby the survival of information is intrinsically linked to the survival of its processors. In memetic evolution, processors (humans, animals) continue to exist even in case that the information they replicated is unsuccessful, and on the other hand information can be successfully replicated even in case that certain processors die or stop the replication of a certain idea.
From these remarks, more useful definitions then can be put forward: living cells are the only self replicators on Earth. Enzymatic polymerases, resp. human minds, copiers, presses, etc. are all processors which can function as replicators of informational molecules, resp. informational symbols. Finally, informational molecules (polynucleotides) and informational cultural conventions (behaviours, spoken words, written, printed and electronically transmitted texts) are replicated elements of instantiated information, replicates.
In summary, the use of replicator for gene or meme is twice erroneous: they are not self-replicators and they even are not processor-replicators. Genes and memes are replicates or bits of transmissible information.
From this it follows that the comparison between memes and chromosomal genes impossibly can lead to formal analogy. Memetic evolution can better be compared with mobile genetics of plasmids. In memetics and plasmid biology, unlike in chromosomal genetics, a larger degree of recombinatorial freedom is possible, since even in most cases of nonsense recombinations, the reproductive success of the processors theirselves is not impaired, as is the case in chromosomal genetics.
Lecturer, Columbia University.
The categories of "replicator" and "interactor" have been shown to be abstract enough to be applied outside their original biological context without metaphor. However, this non-metaphorical (diagrammatic) usage requires a careful specification of both specific modes of replication as well as catalytic interaction. The concept of a "meme", linked as it is to behavioral patterns replicated by imitation, although of some use in the study of human societies (fashions and technological innovations do spread by imitation), is not adequate for the study of linguistic phenomena, where patterns are replicated by obligatory repetition (i.e. phonemes and other language patterns are institutional norms). The distinction between "meme" and "norm" will be discussed, and illustrated with examples from actual linguistic history (mostly from the history of English, and of several Romance languages). Questions of dialectal variation will be raised, with the invention of standard languages seen as a means to reduce the mutation rates in the population of replicators (and hence, of slowing down evolution) and viceversa, the creation of pidgins and creoles viewed as a reinjection of heterogeneity in the population, accelerating evolution and hence giving rise to new entities (e,g, Jamaican English, Haitian French). Then the question of "interactors" will be raised. In the micro-biological context interaction proceeds by catalysis, i.e. the triggering of large-energy results via low-energy expenditures. An abstract model of catalysis will be proposed (a catalyst defined as a dynamic element capable of switching a target system from one stable state to another), and several candidates in the world of linguistics explored. In particular, speech acts (warnings, commands, promises etc) will be shown to act as catalytic entities, switching human agents from one stable social state to another.