J. E. Landau

Recently, I read a paper by the Russian semiotician Ivan Fomin (available here) which attempts to deal with some of the issues in the somewhat old-fashioned discipline of memetics. I think that this understanding of memetics might be an interesting way to understand sapience, the idea of what makes a creature intelligent and able to understand.

First, we should probably define the terms, since “memetics” isn’t exactly something that most people talk about. SCP readers may know that memetics is the study of transmissible ideas, particularly in the sense of treating them analogously to genes. In memetics, a meme is a unit of cultural information, and memes can mutate like genes and be spread between individuals in a population. This is the origination of the term “meme” as it’s used for internet memes, which are not exactly the same thing, but are transmissible in the same way. More important memes might be information on how to do something, or information on cultural practices (like religion).

It’s pretty easy to see how memes mutate. Religions change as they are transmitted between people and interact with one another, exchanging their memes with local customs. Memes encoding technological knowledge also mutate over time when someone discovers a new way to do something, like coming up with a better way to knap a stone knife.

Genes have a similar function, describing some biological information. Lots of genes create useful RNA or proteins that do things and accomplish the various tasks of a cell, and genes can mutate over generations, leading to new types of life.

The existence of “memes” as a concept that can be applied to things is not really debated, but the idea that meme theory is useful and distinct from other ways of understanding ideas and culture is controversial (or, perhaps more accurately, unpopular). Part of this can probably be attributed to the waning star of Richard Dawkins, who was the originator of the term meme, but other parts of it are probably more related to the lack of interesting results arising from meme theory.

Fomin’s work argues that memes and genes are both types of “sign”, which is a concept from semiotics, or the study of signs. This is a circular definition, which is perhaps appropriate given that semiotics is a highly introspective field and tends to be highly academic. It exists somewhere between philosophy, linguistics, and sociology.

Signs are things that stand for other things, which is also a fairly nebulous description, but the details of that aren’t important. Basically, Fomin’s argument is that genes “stand” for a specific biological thing- a trait, or part of a trait. On a molecular level, things are a bit more complicated, but this definition is fine. The gene isn’t actually the thing itself—you can have a strand of DNA that contains the genes for teeth and claws and neurons and whole organisms separate from the organism. It describes those things, like how a word or drawing can reference an object apart from that object itself.

Similarly, cultural ideas can describe objects without actually including them. You can know how to sew without being in the process of sewing, or be Catholic without literally being in mass right this moment, or know Swahili without speaking it right this second. The meme allows you to create these “cultural objects”.

Now that definitions have been established, my idea is this: sapience is the point where a creature’s existence goes from being genetic to memetic.

Imagine a bacterium. It has a purely genetic existence, where all of its traits are heritable. It has no ideas, it learns nothing from its environment except what is “learned” through the process of evolution. When bacteria develop antibiotic resistance, it is not because they have understood the concept and developed a mental model, it is because genes that code for antibiotic resistance have been developed in a population over generations (or through living genetic exchange).

Now, on a higher level of existence, we have animals, like cats and dogs. These animals live in a state that is determined more by genes than by memes. Animals have some limited memetic existence. You can teach a dog to do tricks or look for cocaine and it can learn how to open the fridge on its own if it’s motivated and clever, and these memetic components give it some new capabilities it wouldn’t otherwise have, but the vast majority of what it can do, what it thinks, and how it acts are determined by its genetics.

More intelligent animals have more and more memetic components. Birds and cetaceans talk to each other, spreading ideas and telling each other to watch out for dangers. Simians have been known to teach each other to make and use simple tools. But I don’t think that it would be accurate to say that these organisms have lives that are dominated by ideas.

Contrast human beings. Humans are dominated by ideas. The average human being lives in a city built based on memetically transmitted ideas, spread via architecture schools and civil engineering departments. Human beings work in jobs that are not based around their biological capabilities, but purely their use of tools. Even people working in manual labor positions, like carrying objects around or harvesting crops, are still carrying artificial objects to and from places determined by memetically coded informational rules, or harvesting crops that were planted intentionally in accordance with memetic ends.

This is not just a matter of technology. If we look back at the earliest behaviorally modern humans, these first people still made heavy use of tools and fire. They arranged themselves into groups, and adorned themselves with symbolic objects. They were dominated by ideas, not by physiology, when they exterminated the megafauna on every continent using weapons, or when they organized themselves into tribes and then split off from them based on their internal politics, or buried their dead. They constructed their own signs and symbols in ways that non-sapient animals do not.

Of course, that is not to say that we are at the end of this sapience spectrum. We are still limited by our genetics. People are born with life-altering illnesses or life-changing abilities through genetics, although our memetic inheritance often provides ways to enhance or overcome these biological differences. What would a purely memetic being look like?

One possible answer is that memetic beings are idea-based entities that exert their own will. Religions or corporations might fit this bill, although I don’t think that it would be accurate to describe them as exerting a “will”. It seems plausible that in the future we could see more advanced versions of religions or corporations, that can think and lord over human beings. I have no idea how this would work, but it could be a cool idea for science fiction. Perhaps laws and bureaucracy could be a mechanism for doing so, like a self-directing rule-based corporation with no leadership.

An alternative conception of a pure-meme lifeform might be an artificial intelligence. I don’t think that what we have today qualifies, but I’m also not sure what would. An artificially intelligent being would likely have “genes” in the form of the hardware environment in which it operates, the hard-coded rules it works under, and the training data it is given, all of which are fixed and immutable and generational. A hypothetical memetic AI would be something that can construct itself, alter itself at will, and which would have no constraints from its past. In other words, it would be a lot like the sort of artificial superintelligence the Yudkowsky crowd is afraid of.

Of course, this sapience scale doesn’t actually say anything about “intelligence”, but I’m not sure that “intelligence” is a meaningful concept that can be described on its own. Certainly, some people are more able to form logical conclusions, do math, etc. than others, and some clusters of those traits are probably correlated, but I’m not convinced that intelligence is a meaningful and rigorous concept on its own.

Certainly, you need some ability to connect things together mentally to have a life of the mind, but I think that with this scale you could have sapient creatures that are incredibly unintelligent by our standards. And, further, going up the scale wouldn’t necessarily make you smarter. Some sort of self-modifying self-assembling intelligence might be purely memetic… and unable to do anything useful or reason in any exciting way. Perhaps this is a flaw of the system.

One possible criticism that one might have is that this definition of sapience excludes groups of behaviorally and physiologically modern human beings who happen to use no technology. I think that there are no such groups, and that there cannot be any such groups. Even if you were to take a group of blank-slate human beings and drop them in an uninhabited place, they would quickly develop new ways of doing things and teach them to one another. They would form social bonds, they would gossip, and they would develop their own strange and novel rituals. All of these things are memetic in nature.

Another is that this excludes “singleton species”, where you have an intelligent and technological creature sui generis, since memes are a concept that only apply when one is discussing groups of creatures. I’m not sure how such a creature would arise, but it would certainly be a problem for this classification system.

I don’t think that this system is perfect, and I think that it would need a considerable amount of work to make it rigorous enough to use for, say, a Star Trek type galactic community and exploration program (and I’m guessing that in such a situation your method would probably be more like “what includes everyone in the Federation”). Even if it is not rigorous, I think that it is a novel approach.