I'm still hiding out from COVID-19
I want to write a draft for a textbook on Macromemetics. I take as a point of departure the work in memetics(1) (Micromemetics) that describe how humans like to imitate each other, and are good at understanding and recognizing when a certain meme has been enacted by another, and also which memes are appropriate to enact at any given time, either in response to a meme recently enacted by another or no. Humans do this in order to achieve memetic resonance, and the idea that humans (memetic agents) try to maximize memetic resonance is the First Law of Macromemetics.
I also want to touch on the three ways of representing a memetic system, which are the State Diagram, the Deployment Descriptor(s), and the Transition Matrix.
I figure a good place to start is introduce the laws I have so far and explain how they work. The Laws of Macromemetics depend on a concept of a model of a state made up of a system of agents, and memes which are eligible to be deployed, and which new state the whole system then transitions to. I hope to provide a simple linguistic example of how that might work.
Table 1. The Three Laws of Macromemetics
1. An agent deploys memes in order to achieve optimal resonance
2. Deployment of a meme causes transition to a new state
3. A mutation is a Modification, Addition, or Deletion of a State, an Agent, or a Meme (MADSAM)
Let's take a really simple example of a system in which multiple people could participate, lifted from the work of Dick & Jane.
fig. 1. Dick and Jane cover (Yiddish)
So we have three characters, Dick, Jane, and Spot. These are not agents in the system, but more like MIAOs (Memetic Iconic Anchoring Objects). The point is that as far as memetic agents are concerned, there are only certain speech events which are possible (2):
Table 2.1. Collection of all possible complete Dick & Jane speech events.
"See Jane run."
(These two work for all characters)
"See Jane laugh."
"See Jane pet Spot."
"Jane pets Spot."
(This works for Dick as well)
"See Spot bark."
(Only works for Spot.)
These speech events together are comprised of a small set of words, namely:
Table 2.2. List of All System "Words" (or memes)
"Dick," "Jane," "Spot," "laugh/laughs," "bark/barks," "pet/pets," and finally "See," and of course (.)"period."
If we propose a game where each playing (agent) gets to say one word, and another agent another, until somebody says "period," then we can call each of these words a meme, i.e., an enactable snippet of behavior which another agent (person) can recognize. Straightaway, we can envisage a graph that depicts all possible valid contexts and the allowable actions in each of those conects.
fig. 2. Memetic State Diagram of Dick & Jane Memeplex
Let's break down what fig. 2 represents. The "clouds" are memetic "states," which determine which memes are possible to deploy next (3). The "arrows" (4) represent the deployment of memes (5). We can fairly easily transition this state diagram to a set of deployment descriptors, and at this time I'd like to make some comments about notation. For one, capitalization. States' names are capitalized, and meme names are lowercase, and are followed by an exclamation mark, separated by commas on the arrows where they happen (MIAOs, by the way, are in ALL CAPS).
A couple of obvious issues with fig. 2. are inconsistent capitalization of the State Names and the meme! names, how every arrow does not have a label, and how states named "Jane," "Dick," and "Spot" appear twice, while they should be named something like Jane-2, etc. Some shorthands have been used to avoid visual complextity (6).
A More Rigorous Notation
The state diagram shows an overall picture, and it's accurate, but sometimes we want to rigorously describe all of the transitions in the system, so we can check that we haven't missed anything, design from the ground up, and so forth.
Table 3. Deployment Descriptors for Dick & Jane Memeplex
Start [JANE, DICK, SPOT] see!See, jane!Jane, dick!Dick, spot!Spot
See [JANE, DICK, SPOT] jane!C-Jane, dick!C-Dick, spot!C-Spot
C-Jane [JANE, SPOT] run!Run, laugh!Laugh, pet!Pet
C-Dick [DICK, SPOT] run!Run, laugh!Laugh, pet!Pet
C-Spot [SPOT, See] run!Run, bark!Bark,
Pet [JANE, DICK, SPOT] spot!Period
Jane [JANE, SPOT] runs!Runs, laughs!Laughs, pets!Pets
Dick [DICK, SPOT] runs!Runs, laughs!Laughs, pets!Pets
Pets [DICK, JANE, SPOT] spot!Period
Spot [SPOT] barks!Barks, runs!Runs (7)
Run [DICK, JANE, SPOT] period!
Laugh [DICK, JANE] period!
Period [ ] start!Start
Next time I think we'll take a look at what the transition matrix might look like for a system with these graphs and deployment descriptions.
Keeping It Real
Like any functioning human system, we need this to be self-correcting. We need action to be able to be taken by somebody when an invalid memetic deployment takes place. This makes the system self-correcting, and this brings us to the Laws of Immunomemetics.
Table 4. The Laws of Immunomemetics
1. Any stable memeplex contains an immunomemeplex
2. A system of rules or laws is equivalent to a collection of bullying behaviors
3. The function of an immunomeme is to deter mutations to a memeplex.
We covered some of the basics of how to describe a memetic system, or memeplex. Next time, we'll dive into what a Transition Matrix might look like. This is where we start to consider multiple agents in a real system (6), and also we'll want to start to look at immunomemetic concepts.
(1) Susan Blackmore, Daniel Dennet, et al.
(2) This is a highly-abridged version of Noam Chomsky's Universal Grammar, shall we say.
(3) And when relevant, by whom.
(4) or "edges" from classical graph theory. A memetic state diagram is not much more complicated than "a directed graph," for most intents and purposes.
(5) every "arrow" in the state diagram would have a name, like see! or jane!, or a list of names, indicating all the memes which may be deployed to enact the transition to the given next state.
(6) To see some good examples of rigorous State Diagrams, check out The Dining Philosophers essays.
(7) Note that I've only got one Spot state in the deployment descriptor list, but there are two State clouds in the transition diagram.