Tuesday 30 October 2018

Memetic mass extinction

I've previously discussed the idea of a cultural-cambrian-explosion - i.e. a rapid explosion of diversity followed by competitive winnowing.

I've also talked about the role of memes in the Holocene extinction event. That there's a large-scale extinction event of DNA creatures is widely recognized. However this is also a good time to see memes going extinct - and that seems to be a less well recognized phenomenon.

Some modern memes are spectacularly successful. For example, the many memes that compose the english language are all doing pretty well. However typically, whenever a very successful creature arises, numerous rivals are forced to give up their niches. There must be many languages that are losing their speakers and going extinct. Religious memes also look as though they are probably dropping like flies. Hunter gatherer tribes have been dwindling for a long time. They often take many of their memes with them.

This modern memetic extinction is best seen as part of the Holocene extinction event. Rapid cultural evolution causes environmental disruption, many entities can't adapt fast enough and so go extinct.

Sunday 28 October 2018

Susan Blackmore - Artificial Intelligence - What is Our Role in the Future?

Susan speculates about machine intelligence and the future. Thanks to Adam Ford for the video.

I don't disagree with very much here. Susan seems a bit more pessimistic than me. Also, it seems to me that she exaggerates the extent to which machines are currently creating themselves. At the moment, I would emphasize that humans are still in the loop. Machines creating themselves will be a reality one day, but today the idea is a bit of a cartoon caricature. I also suspect that I spend less time than Susan thinking about whether the internet is conscious.

Monday 15 October 2018

For memes, minds are like islands

Islands are natural evolutionary experiments. The history of evolutionary theory prominently features researchers who visited islands. Most famous of these was Darwin himself. Darwin's trip to the remote Galapagos islands was famously documented in The Voyage of the Beagle. This took place before Darwin had really got interested in evolutionary theory - and it was clearly a catalytic event.

Islands feature restricted gene flow. They also feature rapid evolution due to the environment being different from the ancestral one, and due to small population sizes and genetc drift. The size of the island and how isolated it is are important factors to the resulting evolutionary dynamics.

It has subsequently become understood that islands are a special case of restricted flow of heritable information. That means that the great lakes are like watery islands for fish. It means that deserts are like islands for cacti. It means that that mountains are like islands for their plants and animals. It also means that many hosts are like islands for their parasites.

Dogs are like islands for their fleas. Zits are like islands to their bacteria - and minds are like islands for their memes. Evolutionists got the most mileage out of archipelagos far from the mainland. The memetic equivalent would be lost or isolated tribes. Sure enough, anthropologists have long been fascinated by these types of group. Remote monasteries another possible object of study - and of course groups of humans can themselves become isolated on real islands.

Islands are natural experiments in group selection. Theory has long suggested that if the number of migrants per generation between groups falls to around 1, then groups of sexual organisms can start to behave rather like individuals which develop their own, distinct traits. They start to develop persistent differences between one another - instead of simply being part of the same gene pool. One migrant per generation is not very many and so conventional wisdom has argued that group selection is not very important. However, one place where isolation levels can be high is on islands. If an island is sufficiently remote, it is not very challenging to get into the ballpark of 1 migrant per generation.

Reproduction within islands is subject to one set of selection pressures, while reproduction between islands is often subject to a quite different set of selection pressures. Strawberry plants illustrate how such selection can pull organisms in two different directions. Strawberries have runners for local, asexual reproduction and seeds for sexual reproduction over large distances. Strawberries must allocate resources depending on their local conditions. If they are cramped, and under seige from parasites, then seeds start looking like an attractive option. This type of resource allocation dilemma is a common one for organisms on islands. Memes too must allocate their resources between fighting off competition within their existing mind and spreading to new minds. As with strawberry plants, they are often pulled by selection in two conflicting directions.

Islands are widely considered to be incubators of evolutionary innovation. On the mainland experiments often don't get very far before reality catches up with them and calls them to account. On islands, experients can last for longer before their day of reckoning comes. Often, rats come to the island, and destroy its native ecosystem - and that is the end of that. However, once is a while something amazing happens, and the island evolution produces something valuable that would never have evolved without the isolation. In the USA, many small start-ups attempt to reproduce this innovation-on-islands effect. Most fail, but sometimes, something amazing happens.

The evolutionary dynamics of island ecosystems seems like fertile ground for cultural evolution and memetics. It is territory that has already been explored to some extent, but there is still much left to learn.

There's an interesting analogy between organisms colonising a new volcanic island and individual enculturation of human brains. In both cases, pioneer species come first and create the ecosystem for their successors.

One of my early essays was about evolution on islands. I wasn't thinking about cultural evolution much back then. It is interesting for me to look back on the essay, now I that know more about the relevance of island dynamics to memetic evolution.

Saturday 13 October 2018

Memes in U.S. Government Influence Campaigns

Thinktank CNA has a document out titled: "Exploring the Utility of Memes for U.S. Government Influence Campaigns". It's a 67 page document heavily loaded with pictures of internet memes and analysis of their potential influence.

It's one of a number of recent publications which attempts to reify the modern conception of an internet meme and retro-fit this onto the definition of the term "meme".

I think serious meme enthiusiasts need to reject this sort of thing as an unscientific perversion of the meme concept. IMO, it is OK to use of "meme" as an acceptable abbreviation for "popular internet meme". After all, popular internet memes are indeed still memes. However there's no good reason to redefine the term "meme".

I notice that even dictionaries are giving the nod to this new usage. For example, https://www.dictionary.com/browse/meme still lists the Dawkins-style definition first, saying:

a cultural item that is transmitted by repetition and replication in a manner analogous to the biological transmission of genes.

...but it goes on to say that "meme" can also refer to:

a cultural item in the form of an image, video, phrase, etc., that is spread via the Internet and often altered in a creative or humorous way.

The first usage is scientifically useful. The second definition, not so much.

It is worth noting that there was a similar movement to define "gene" in a frequency-dependent manner. In 1966 G. C. Williams defined a "gene" as follows:

In this book I use the term gene to mean 'that which segregates and recombines with appreciable frequency'

Going on to say:

In evolutionary theory, a gene could be defined as any hereditary information for which there is a favorable or unfavorable selection bias equal to several or many times the rate of endogenous change
It was pioneering to define a gene in terms of "any hereditary information" - but the idea of frequency-dependence wasn't broadly adopted. Few could stomach the idea that genehood depends on frequency in the way that Williams contended. It seems adaptationist to define genes under the influence of of genetic drift out of existence.

I like that "memes" are now very popular and widely-discussed - but now is not an appropriate time to forget about their roots.

Survivalism vs hedonism

Survivalism and hedonism are doctrines surrounding fitness maximization and pleasure maximization respectively.

Both schools vary along the individualism / collectivism axis. Survivalism ranges from individualists, who are out for themselves through those who profess to be against the extinction of our species. Similarly, hedonists vary from those out for their own pleasure, to those who think the total pleasure of all sentient beings is what is important.

Hedonists think they are more advanced than survivalists. They are further up in Maslow's hierarchy of needs. They view animal evolution as a transition away from basic survivalist goals towards more advanced optimization targets dictated by nervous systems. In turn, survivalists view hedonists as short-sighted folk prone to drug-taking, masturbation, casual sex, wireheading and other dubious activities oriented towards pleasure and often at the expense of basic survival. See the picture, for more examples.

For a while, I considered these teams' rivalry to be largely misplaced. Hedonists who neglected survival were short-sighted fools likely to end up with nothing. Only hedonists who recognised that hedonism depended on survival really needed to be taken seriously. However, such hedonists would be likely to behave a lot like survivalists until they were pretty certain that they were not going to encounter more advanced alien civilizations in the alien race. That might well be tens of millions of years in the future. If the two tribes behave in much the same way for billions of years, they don't seem too dissimilar.

I'm now a bit less confident of this assessment. Hedonists might think that alien races would also be hedonistic - and so might not be too concerned about the consequences of an alien invasion. I can also imagine a hedonist gambling on extinction vs making everyone three times as happy forever. A survivalist would be unlikely to take such a gamble - but possibly a hedonist might.

It also seems that in practice that the different optimization targets do result in different behavior. For example, they differ regarding treatment of non-human animals. Many hedonists are concerned with animal welfare causes: factory farming, wild animal suffering and so on. However, to survivalists these mostly look like bad causes: non-human animals are probably not going to help make sure that Spaceship Earth completes its journey. This kind of practical difference today does not bode terribly well for the thesis that the behavior of these tribes will converge in the future. Maybe the survivalists and the hedonists will remain at odds.

Friday 12 October 2018

Pretend to be part of your host

How do parasites avoid destruction at the hands of the host immune systems? Of course there are many ways, but one way is to pretend to be part of your host. Viruses are experts at this. They hide inside host cells, out of the way of the immune system. However the immune system has a window to cell contents the form of the major histocompatibility complex (MHC) proteins, which are detectable from outside the cell. Some viruses, for example cytomegaloviruses synthesize families of human-like proteins to fool the immune system into thinking that the host cell is still mostly human.

Is there an equivalent of this in cultural evolution? There's certainly a distinction between self and other and this is used by the memetic immune system when deciding which memes to reject. The components most clearly marked as beiong "self" are those that the host considers to be part of their own identity. So many memes compete for adoption by the identity of their host.

One way you can tell which memes are considered to be part of the identity of an individual is when they say "I am X". Some examples:

  • Religion: "I am a Christian"
  • Politics: "I am a Republican"
  • Jobs: "I am a nurse"
  • Sex: "I am transexual"
  • Ethnicity: "I am Jewish"
  • Disease: "I am disabled"
  • Age: "I am old"
One possible defense against unwanted memes masquerading as being part of you is to spring-clean your identity. If you reduce its footprint there will be fewer places for unwanted memes to hide. At any rate, regardless of whether you want to shrink your identity, it is worth looking closely at who you think you are, to make sure that it really is you.

Thursday 11 October 2018

Weaken host immune system by creating a diversion

I have a previous article about cultural immune deficiency disorders. I've also written about cultural opportunistic infections - the cultural version of the well knnown phenomena, opportunistic infections.

I've long recognised the possibility that infections could attack the host immune system to create a hole to penetrate host defenses - and that such a hole could then let in other parasites. Clearly these dynamics apply in cultural evolution as well - as is seen in cultural immune deficiency disorders.

This article is about a related set of strategies involving diverting host immune system resources elsewhere. Rather than directly attacking the immune system parasites may be able to create a diversion. One strategy would be to absorb host immune resources via deliberate immune stimulation using expendible, fake targets. Another strategy might be to liberate suppressed persistent infections already present within the host by making holes in the immune system for them.

The former strategy has some well-studied examples. Some parasites release what are known as superantigens - which act to stimulate the immune system, sapping its resources.

A couple of analogies illustrate the proposed effects. Invaders sometimes liberate prisoners from the territories they are invading. The reason is not always clear - but they may be creating a diversion which saps the resources of the defenders. They may also engage in arson. Causing chaos can help them to achieve their own goals. Another analogy involves the saying that "the enemy of my enemy is my friend". It is widely recognised that uniting against a common enemy can result in strength gains. Orwell wrote about this in the book "1984". Unrelated parasites may effectively gang up on their hosts - since they have a shared interest in a weakened host immune system.

I don't have much in the way of concrete evidence or examples in the cultural domain to offer in this post. However the possibility that these dynamics might also apply to cultural evolution seems intriguing.