Modelling meme units
The main criticism that can be raised against the memetic approach is that memes are difficult to define. What are the elements or units that make up a meme? Does a meme correspond to a complete symphony, or to a symphonic movement, a melody, a musical phrase, or even a single note?
In order to model meme structure, we may use some concepts from cognitive science. Perhaps the most popular unit used to represent knowledge in artificial intelligence is the production rule. It has the form "if condition, then action". In symbols:
If A, then B or A -> B
A represents a condition that is distinguished, B represents an action that is
executed or another condition that is activated.
The action leads in general to the activation of another condition. In fact a production rule can be analysed as a combination of even more primitive elements: two distinctions (which discriminate between presence and absence of the condition and the action respectively) and a connection (the "then" part, which makes the first distinction entail the second one) (Heylighen, 1991d; see also Heylighen, 1990). For example, a meme like "God is omnipotent" can be modelled as "if a phenomenon is God (distinction of God from non-God), then that phenomenon is omnipotent".
Production rules are connected when the output condition (action) of the one matches the input condition of the other. E.g. A -> B, B -> C. This makes it possible to construct complex cognitive systems on the basis of elementary rules. Even remembered melodies might be modelled in such a way, as concatenations of production rules of the type "if C (musical note distinguished), then E (note produced and subsequently distinguished)", "if E, then A", and so on.
A similar model applies to genes. A gene corresponds to a string of DNA codons, which respond to the presence
of certain activating proteins, or the absence of certain inhibiting proteins (condition) by manufacturing new proteins
(action). This may in turn activate further genes, depending on the present of specific chemicals in the cell, and so on. This leads to complex networks of "if... then" productions ( Kauffmann, 1992).
Variation of memetic units
It has been shown that production rules (or at least a simplified, binary representation of them, called "classifiers") can be used to build quite impressive computer simulations of cognitive evolution, using mutations, recombinations, and selection on the basis of "fitness" (Holland et al., 1986).
Distinctions can be represented as combinations (strings) of elementary yes-no
(1-0) observables. Mutation or recombination of
distinctions can then be modelled by either randomly changing certain binary digits in a string, or by concatenating the first part of one string (A) with the second part of another string (B), like in the following example:
A = 1001|001 mutation: A' = 1001|000
B = 0010|011 recombination ("crossing-over") of A and B:
A~B = 1000|011
Although these models do not as yet take into account distinct carriers, this looks like a very promising road to study memes formally and computationally.
Meme complexes
Even if we would model memes as connected sets of production rules, we still have the problem of how many production rules define a single meme. If we call a religion or a scientific theory a meme, it is clear that this will encompass a very large number of interconnected rules. In practice it will be impossible to enumerate all rules, or to define sharp boundaries between the rules that belong to the meme and those that do not. However, that should not detract us from using memetic mechanisms in analysing evolution.
Indeed, Darwinian models of genetic evolution have certainly proven their usefulness, even though it is in practice impossible to specify the exact DNA codons that determine the gene for, say, blue eyes or altruism towards siblings. As Dawkins (1976) notes, it is not necessary to be explicit about what are the constitutive elements of a gene, postulated to explain a particular characteristic or type of behavior. It is sufficient that we can distinguish the phenotypical effects of that gene from the effects of its rival genes (alleles). If we can determine the fitness resulting from these effects, taking into account the environment and the context of different, non-rival genes present in the genome, then we can make predictions about evolution.
The same applies to memes. If, for example, we observe that one meme (say Catholicism) induces its carriers to have more children than its competitors (say Calvinism and Anglicanism), and that the children tend to take over their memes from their parents, then, all other things being equal, we can predict that after sufficient time that meme will dominate in the population. Of course, in practice it is never the case that all other things are equal, but that is the predicament of all scientific modelling: we must always simplify, and ignore potentially important influences. The question is to do that as wisely as possible, and to maximally include relevant variables without making the model too complex.
References
- Dawkins R. (1976): The Selfish Gene, (Oxford University Press, New York).
- Heylighen F. (1992) : "
Selfish
Memes and the Evolution of Cooperation", Journal of Ideas , Vol.
2, #4, pp 77-84.
- Holland J.H., Holyoak K.J., Nisbett R.E. & Thagard P.R. (1986): Induction : processes of inference, learning and discovery, (MIT Press, Massachusetts).
- Kauffman S.A. (1992): Origins of Order: self-organization and selection in evolution, (Oxford University Press, Oxford).
