I’ve decided that it is time for some new terms. I’ve been using “meme” as the cultural analog for the biological “gene”, which is more or less the use that Dawkins had in mind when he coined the term. But I’ve decided to scrap it. I also need a term for the cultural analog to the biological phenotype.
From “meme” to “coordinator”
The problem with “meme” as a term is that it has accumulated quite a bit of conceptual baggage. None of that baggage is compatible with my sense of how the cultural analogue to genes function, and much of it is harmful nonsense. Oh, as a popular term as in “internet meme” it’s fine and, in any event, it would be foolish of me to try to interfere with that. The trouble comes when you try to use more or less that meaning in a serious investigation into culture. No one has yet made it work.
I’ve expressed my views in a long series article, book chapters, posts and working papers going back almost two decades. Here’s my most recent thinking:
- Cultural Evolution, Memes, and the Trouble with Dan Dennett (working paper)
- Memes as Data: Targets, Couplers, and Designators (post)
- Cultural Evolution: Some Terminology (post)
If I jettison “meme”, however, what’s the alternative?
I don’t like coining new words so I’d like to use an existing word. I’ve decided up “coordinator,” which expresses nicely the function that these entities have to performer. The cultural analog to the gene coordinates the minds of people during interaction, whether face-to-face or through various media. This is not the place to explain that – you can find plenty of explanatory material in the above-linked documents, and others as well. I note only that the term sidesteps the notion of bits of “information” that get passed from person to person, brain to brain. On the contrary, it helps explain how we can, in effect, walk about in one another’s minds (cf. Peter Gärdenfors, The Geometry of Meaning, 2014 chapters four and five).
From “ideotype” to “phantasm”
What about the analogue to the biological phenotype, the individual organism in its environment? Ever since I first thought about cultural evolution as a process of “blind variation and selective retention” (in Donald Campbell’s formulation) I figured that the selecting was being done by groups of human beings and therefore that the environment of cultural evolutionary adaptedness had to be something like a “group mind.” In some way, what was being selected had to be public. I first articulated that position in an article I published in 1996, Culture as an Evolutionary Arena (Journal of Social and Biological Structures). But it took several years before I arrived at a satisfactory, albeit still provisional, formulation.
I did that in my book on music, Beethoven’s Anvil (2001), and there it rested upon quite a bit of preliminary work devoted to composing a physically plausible account of interacting brains constituting that group mind (you can download drafts of those chapters HERE). Given that foundation, here is what I said later in the book (pp. 192):
[The question of the cultural phenotype] is as difficult, obscure, and contentious as the nature of memes. Some theorists, Susan Blackmore among them, deny the need for a phenotypic entity at all. In contrast, David Hull has cast scientists themselves into the phenotype role in his treatment of science as an evolutionary process. Influenced by Hull, William Croft assigns the speaker and her grammar to the phenotypic slot in language evolution. Given the both the difficulty and the preliminary nature of this theorizing, these are reasonable choices.
In the case of musicking I take a more abstract view:
Performance-as-Phenotype: The phenotypic role in music’s evolution is played by performance level attractors.
Note that in identifying musicking’s phenotypes in this way I have not introduced a new entity into the theory; I have simply assigned an existing theoretical entity to a role in a different theory, that of cultural evolution. For we have already discussed performance level attractors in the previous chapter. A performance level attractor is simply a trajectory in a group’s collective neural state space that specifies a whole performance. As such, it is a different object from the groove stream attractors we discussed in Chapter 6, though a performance’s groove stream attractor would be a component of the overall performance attractor. Performance attractors are thus properties of brains-in-process, whether a single musicking brain or a group of brains coupled through musicking. They cannot be thought of as being inside brains in the way viruses are inside their hosts. They are self-organizing emergent phenomena arising when people make music.
The word “attractor” in the definition tells you that this conception is grounded in complex dynamics, which is more than I can explain here, but the chapter drafts I linked above should give you a good start.
The general idea is that cultural selection is operating on the set of states, a trajectory, a brain takes while it is perceiving or enacting some appropriate cultural phenomenon, such as a musical performance. In the case of a group making music, or even just listening to a performance, the trajectory unfolds in real time with the piece of music. The same with viewing a painting or a work of sculpture or even reading a book. If the trajectory gives pleasure, one recommends the work to others and seeks to re-experience it oneself. If, on the contrary, it creates anxiety, one avoids it.
But what do we call this thing? Last year I called it an ideotype. The “type” morpheme follows the “phenotype” word structure and “ideo” puts it in the head. And yet “ideo” also has connotations of idea, which is not quite right.
What to do?
After trying another term or two – including ethnotype – I decided to abandon the “type” notion altogether and go for something different: phantasm. The word is an existing one meaning “ghost or spirit”, “delusion or illusion.” While those aren’t quite right, upon further thought, they’re not bad either. Until vetted and solidified by the group, these things ARE will-o-the-wisps and, as neurodynamic objects, they’re pretty abstract and strange.
The word itself is not particularly common, which is, I feel an advantage. People may have heard or seen it, and may even know more or less what it means. But few will have any strong investment in its particular use. So my proposed use won’t be derailed by deeply ingrained habits of thought.
* * * * *
Here we have it:
Cultural evolution proceeds by blind variation among coordinators and selective retention of phantasms.
I like it, at least for now.
From the Boyd and Richerson school of gene-culture coevolution: There is not a special cultural analog of phenotype. It is just “phenotype”. Phenotype results from genetic, epigenetic, and cultural transmission and environmental influences, etc. This usage is compatible with the standard in biology. http://en.wikipedia.org/wiki/Phenotype
B&R just normally use the terms “cultural trait,” “cultural variant” or “cultural variation” to describe cultural inheritance. Also, on memes: http://www.des.ucdavis.edu/faculty/Richerson/CambMeme.PDF.
You have prompted some thinking.
Accepting that language, natural and artefactual (sensu Distin, 2010) functions to coordinate (sensu Benzon, 2013) I fear I cannot see attaching a new signified – that of a cultural
replicator – to the signifier coordinator proving to be a good trick (sensu Dennett) in the signifier’s replication.
My reading of David Hull is also slightly different. Surely he casts scientific institutions (sensu Di Maggio and Powell various) including Journals as part of the phenotype. Toulmin (1972)
said much the same in Human Understanding with his an evolutionary account of conceptual change.
Distin’s alternative is representations and the linguistically acquired capacity to meta-represent. She would, I know, accept signifiers (or presumably your designators) as essentially synonymous hence my (2012) ‘selfish signifier’. Rather than the dichotomy you pose:
One can think of memes as cultural analogs to biological viruses or one can them of them as analogues to genes. Dennett chooses the former while I choose the latter. (Benzon 2013 9)
I argue one can find both forms of transmission via what we might call institutionalization. So when you say:
My own view, the one I’ve been pursuing in these notes, is that these memes, the physical things playing a certain role in cultural evolution, have already been more or less identified by previous thinkers, but have not been recognized as playing a genetic role in cultural processes. (Ibid 58)
I would agree. Does cultural evolution have its unrecognised Gregor Mendel in a certain Ferdinand de Saussure?
References at http://shura.shu.ac.uk/4255/.
@Matthew Zefferman: I’m aware of B & R and gene culture co-evolution. I just don’t think it tells us very much about human culture, though it might be sufficient for animal culture. For example, I want to understand the cultural evolution of popular music in the United States during the 20th century, in particular, I’m interested in the interaction between African-America and European-America (which I’ve discussed in some detail HERE and more so HERE). I figure B & R have almost nothing to say about that. I rather doubt that gene flow explains, for example, why both jazz and rock and roll both spread like wild-fire among white performers but that hip-hop has yet to produce a large body of white performers, though there are some prominent white performers in the music. And I think that most cultural evolution is more like that than like lactose intolerance. As an account of cultural evolution gene-culture co-evolution has little to offer. It’s a non-starter.
@Ilfrym Price: “… I cannot see attaching a new signified – that of a cultural
replicator – to the signifier coordinator proving to be a good trick (sensu Dennett) in the signifier’s replication.”
“Replicator” is not my language; I don’t find it very helpful. Sure, it’s fine for printing presses and audio recordings and the like, but they’re late in the game and are secondary. As For Dennett, he has a big reputation, but he doesn’t understand culture very well.
“Does cultural evolution have its unrecognised Gregor Mendel in a certain Ferdinand de Saussure?”
Well, Kenneth Pike is the one who tried to generalize the emic/etic distinction to all of culture, which is what put me on to this line of thinking a few years ago, though I’ve never actually tried to work my way through Pike’s books on the matter. In my current terminology, coordinators are the emic properties of the physical embodiements of culture.
Phenotypes may be gene products, meme products – or the products of other kinds of copied information (for example, the information that is copied during individual learning). It is good to refer to the whole ensemble using the term “phenotype” – but it is also important to recognize that some phenotypes are coded for primarily by variations in DNA genes, while others are coded for primarily by variation in memes. The melanin content of skin is mostly coded for by DNA genes. The patterns on Scottish kilts are mostly coded for by memes. Hair colour is somewhere in the middle – influenced genes, memes and other factors.
The linked paper contains some infuriatingly obtuse content. For instance, what Daniel Dennett actually said that *Darwinism* was a universal acid. Replying that *memes* are not a universal acid is a straw man – and a pretty ridiculous one. Making content up like this is not a helpful way to criticize a rival scientific framework for cultural evolution. I suppose that this sort of thing is what you are reduced to if you lack coherent criticisms.
Dawkins categorized evolutionary entities into replicators and vehicles. Hull categorized evolutionary entities into replicators and interactors. According to Hull, the difference is that his “interactors” included replicators as a subset. Hull also says he was serious, while Dawkins ‘buried’ his vehicle concept. Both were early attempts to formalize the gene/phene distinction. These days, I think we can do a bit better using an approach based on information theory. Genes consist of heritable information and phenes are their products. This approach sidesteps many of the ink-wasting historical questions about copying fidelity that have preoccupied critics.
Cultural evolution is basically gene-culture coevolution with genes moving very slowly. We know that the genes of humans move very slowly compared to cultural change – because of humans having a long generation time. By contrast, microbes evolve at a speed comparable to cultural change – and memes and viruses have broadly similar generation times and evolve at similar rates. Academia has had a bizarre and unhealthy obsession with gene-culture coevolution (as opposed to cultural evolution) since the 1980s. It wasn’t just Boyd and Richerson – it was most of those seriously involved. However, you can’t coherently argue that this material is not relevant. Of course it is relevant.
“However, you can’t coherently argue that this material is not relevant. Of course it is relevant.”
Since it says little about cultural change it is marginal.