This is another archival repost, originally written for the old blog in 2005.
I’ve almost finished this article on the replication of memes and how it is and isn’t like the replication of genes (in terms of their evolution). It might not be completely finished, or even properly copyedited and structured yet, but I want to make sure it doesn’t get burried and forgoten in my real work:
Parents lend children their experience and a vicarious memory; children endow their parents with a vicarious immortality. — George Santayana, The Life of Reason, “Reason in Society,”
A common objection to the concept of memetics (What is memetics? Wikipedia has an introduction) is that ideas and other things that can be imitated have too low copying fidelity for a comparison to genetics to work. The objectors can be forgiven for not realising just how much the analogy works in terms of copying fidelity because they are comparing contemporary memes with contemporary memes–the obvious thing to do. But memes are young–they’ve been evolving for somewhere between a few hundred thousand years, and a couple of million years–while life on earth has developed its high fidelity copying machines over four thousand million years.
Today genes cooperate with each other to build complex survival and reproduction machines. The genes hide inside these machines in cells which protect the genes from the outside world of radiation and DNA degrading chemicals. They produce molecular machinery (enzymes) to make near perfect copies of themselves and fix any typographical errors or smudges that may make the DNA illegible. The information which is encoded in genes is digital, rather than analogue: a particular “letter” is either adenosine or not adenosine, not three parts adenosine and one part guanine. DNA uses positive-negative base pairing between just four bases, meaning that if the sequence on one strand is damaged, the correct sequence can be deduced from the complimentary strand. All this adds up to a highly stable gene with high fidelity and longevity.
But those are modern genes, four thousand million years in the making. Genes weren’t created with all their packaging and molecular machinery. Chemical replicators which may conform to a very loose definition of “gene”, and one (or many) of which may have led to modern genes, may not even have used digital information storage. The original replicators, precursors of life, may have floated free in the “primeval soup” with no protective membranes and little metabolism. They may have had no molecular machinery to assist in copying, let alone copyediting. But low-fidelity low-longevity replicators in the primeval soup managed to evolve and eventually give rise to high-fidelity replicators, possibly because in this early chemistry full of building blocks they had the advantage of high-fecundity and could therefore try out lots of different combinations, eventually happening upon higher fidelity and longevity.
Richard Dawkins, original proponent of the meme as a replicator analogous to a gene, addresses the objection of low fidelity in The Selfish Gene, the book in which memes were first described:
“I think a new kind of replicator has recently emerged on this very planet. It is staring us in the face. It is still in its infancy, still drifting clumsily about in its primeval soup, but already it is achieving evolutionary change at a rate which leaves the old gene panting far behind.”
Scientific ideas, fashion trends and music don’t evolve by mutations in a digital sequence, and sounds and sights are not stored in highly stable digital sequences in our memories. But we have invented lots of digital or quasi-digital methods of storing memes—not just books, sheet music, photographs or digital computer storage devices, but as Blackmore (1999) describes, the structure of language itself. All of the world’s languages consist of a few dozen discreet vowel and consonant sounds, usually chosen arbitrarily from what is actually a continuous scale of sounds. These sounds are put together to create short syllables making up discreet words separated by short but noticeable pauses. Most written languages have a few dozen letters representing the language’s sounds, and a few other punctuation characters and grammatical rules which break the text up into more discrete and digital-like chunks.
Humans have been inventing memes for increasing the fecundity, fidelity and longevity of memes, from the written word to the internet, such that so many new ideas are coming into existence that the bad ideas (not just silly ideas, but ideas that are bad at replicating, e.g. because they’re not interesting or useful), can be expected to be selected out very rapidly, just as a poor replicating molecule in the primeval soup would have been, or new mutations that cause miscarriage are. The good ideas, such as writing, cooking and farming, so far do have longevity. These ideas are not actually a single meme but a collection, and cooking has evolved though different combinations (campfires, argas, gas and electric ovens, charcoal barbeques, toasters and microwaves), but the same applies to all our biological organs. Cooking methods have evolved much faster than human eyes. But this is not indicative of just low fidelity and longevity, but of much much higher fecundity of memes. Memes are spread from person to person over time spans varying from minutes to years. With our modern methods of idea spreading a meme can be spread to millions of people in one go. People have generation times of around 20-40 years.
Memes are not exactly analogous to genes, and proponents of the idea of memetics do not claim that these two replicators behave identically, only that they are both subject to Darwinian natural selection. There may be ways in which memetics sometimes fails to explain cultural evolution, and memetics may end up being more useful as a thought experiment than a science, but opponents of memetics have not tripped it up with the fidelity or longevity arguments.
Further reading
- Bateson, Adrian, 2000. Life Without Genes. London: Harper Collins. (Amazon)
- Blackmore, Susan, 1999. The Meme Machine. Oxford University Press. (Amazon)
- Dawkins, Richard, 1989. The Selfish Gene 2nd ed. Oxford University Press. (Amazon)
- Dennett, Daniel, 1995. Darwin’s Dangerous Idea. New York: Simon & Schuster. (Amazon)