Taking Memetics Seriously

I am currently reading Darwinizing Culture, edited by Robert Aunger. I just finished David Hull’s contribution titled ‘Taking Memetics Seriously’ and I’m blown away by it. It’s the best writing on memetics that I’ve read for a long while. Hull manages to remain crystal clear on muddy waters of theoretical memetics with his bold arguments supported by very appropriate examples. His optimism is infectious, so much so that I sat down and typed some of the key points of the chapter in the hope of spreading his message to memeticists.

Geneticists know much more about the complexities of genetics than of social groups. Conversely, anthropologists and sociobiologists tend to be well-versed in the details of social groups. To them genetics looks pretty simple. Contrary to what we were all taught at high school, genes are nothing like beads on a string. So both memes and genes are likely to have comparable complex structures. (45)

Complaints about the lack of conceptual clarity in memetics arise in part because of an unreal view of how clear and uncomplicated certain familiar terms in science actually were or are. For example, look at the term ‘gene’ itself. Was it all that clear when it was first introduced in 1909 by W. L. Johannsen? [He goes on to argue that the changing definitions of the term ‘gene’ were never absolute but always operational in different contexts such as Mendelian genetics, molecular biology or evolutionary biology.] In general, critics of memetics assume standarts so high for scientific knowledge that few, if any, areas of science can possibly meet them.

However, memeticists are not totally off the hook. (…) Just as Mendelian geneticists and molecular biologists have provided operational criteria for applying their gene concepts, so must advocates of memetics. These operational criteria will not be ‘definitions’ as philosophers use this term. At best, they will be highly context-dependent rules of thumb. Even so, if memetics is to be taken seriously, such criteria must be provided, and they cannot be provided from the seat of a comfortable chair. They can emerge only as one sets about doing memetics. (46–48)

[As a reply to the objection phrased as:  “Until I get really clear about what a meme is, how can I conduct any empirical investigations on memes?”] In this respect, memeticists are in the same position as any scientist working in a new area. (…) The solution to this ineluctable circle is obvious if not very intuitively pleasing: you work on all fronts simultaneously. Crude empirical investigations lead you to develop your theoretical perspective more clearly and extensively, and as it improves, you are in a better position to run more sophisticated empirical investigations, and so on. (48–49)

The primary message of this chapter, then is that memeticists cannot begin to understand what the science of memetics is until they generate some general beliefs about conceptual change and try to test them. These tests are likely to look fairly paltry, but in the early stages of a science, attempts at testing always look fairly paltry. (49)

Quick and easy metaphors and popular science are likely to lead to the ‘debasement of memetics’. (49)

Below are some important remarks against two of the most common generalizations about the nature of memetic evolution.

As strange as it may seem, the tendency of thinking in terms of genes and organisms pervades the literature on memetic evolution and gives rise to numerous misunderstandings. For example, one commonly hears that conceptual evolution is so much faster than gene-based biological evolution. Certainly, memes can be transmitted much more rapidly than the genes of such organisms as whales, people, and sequoia trees. However, even from the organismal perspective, viruses and bacteria reproduce themselves much more rapidly than the vast majority of memes. (55)

Nearly everyone who discusses memetic transmission claims that it can be both vertical and horizontal. If parents teach their offspring something, that is vertical. Any memetic transmission that differs from that genealogical direction is horizontal. The preceding claims follow, however, only from the perspective of organisms and their genes, but this is not the appropriate perspective for memetics. (57)

And this, in my view, is the heart of it all:

Memetics does not involve analogical reasoning at all. Instead, a general account of selection is being developed that applies equally to a variety of different sorts of differential replication. Instead of genetics forming the fundamental analog to which all other selection processes must be compared, all examples of selection processes are treated on a par. (45)

I don’t like to quote at this length on this blog – this post is an exception in that these words accurately express what I have been thinking on these issues for a long time.