[Credit for horizontally transmitting these ideas to my brain goes mostly to Jennifer RM, except for the bits at the end about Bowling Alone and The Moral Economy. Apologies to Jennifer for further horizontally spreading.]
The concept of vertical and horizontal transmission felt like a big upgrade in my ability to think about cooperative/noncooperative behavior in practice. The basic idea is to distinguish between symbiotes that are passed on primarily along genetic lines, vs symbiotes which are passed on primarily between unrelated organisms. A symbiote which is vertically transmitted is very likely to be helpful, whereas a symbiote which is horizontally transmitted is very likely to be harmful. (Remember that in biology, “symbiote” means any kind of close relationship between different organisms; symbiosis which is useful to both organisms is mutualistic, while symbiosis which is useful to one but harmful to another is parasitic.) (This is discussed here on LW in Martin Sustrik’s Coordination Problems in Evolution.)
We can obviously generalize this quite a bit.
Infectious diseases tend to be more deadly the higher their transmission rate is. (Diseases with a low transmission rate need to keep their hosts relatively healthy in order to make contact with other potential hosts.)
Memes which spread vertically are more likely to be beneficial to humans than memes which spread horizontally (at least, beneficial to those human’s genes). Religions which are passed through family lines have an incentive to encourage big families, and include ideas which promote healthy, wealthy, sustainable living. Religions which spread primarily to unrelated people have a greater incentive to exploit those people, squeezing every last drop of proselytization out of them.
Long-term interactions between humans are more likely to be mutualistic, while short-term interactions are more likely to be predatory.
In general, cooperative behavior is more likely to arise in iterated games; moreso the more iterations there are, and the more probable continued iteration is.
Vertical transmission is just a highly iterated game between the genes of the host and the genes of the symbiote.
Horizontal Transmission Abounds
Wait, but… horizontal transmission appears to be the norm all over the place, including some of the things I hold most dear!
Religion and tradition tend to favor vertical transmission, while science, education, and reason favor horizontal transmission.
Free-market economies seem to favor a whole lot of single-shot interactions, rather than the time-tested iterated relationships which would be more common in earlier economies.
To this day, small-town culture favors more highly iterated relationships, whereas big-city culture favors low-iteration. (I’ve had a decent amount of experience with small-town culture, and a common sentiment is that you have to live somewhere for 20 years before people trust you and treat you as a full member of the community.)
Paradox One: A lot of good things seem to have a horizontal transfer structure. Some things which I tend to regard with more suspicion have a vertical flavor.
Horizontal Transmission Seems Wonderful
The ability to travel easily from community to community allows a person to find the work, cultural environment, and set of friends that’s right for them.
Similarly, the ability to work remotely can be a huge boon, by allowing separate selection of workplace and living environment.
The first thing I want to do when I hear that vertically-transmitted religion has beneficial memes is to try and get more of those memes for myself!
Similarly, I’ve read that many bacteria have the ability to pick up loose genetic material from their environment, and incorporate it into their own genes. (See horizontal gene transfer.) This can be beneficial if those genes are from organisms adapted to the local environment.
Paradox Two: In an environment where horizontal transfer is rare, opening things up for more horizontal transfer is usually pretty great. But an open environment gives rise to bad dynamics which incentivize closing down.
If you’re in a world where people only ever trade with highly iterated partners, there is probably a lot of low-hanging fruit to be had from trading with a large number of untrusted partners. You could arbitrage price differences, get goods from areas where they’re abundant to areas where they’re scarce, and generally make a big profit while legitimately helping a lot of people. All for the low price of opening up trade a little bit.
But this threatens the environment of trust and goodwill that you’re relying on. An environment with more free trade is one with more scammers, inferior goods, and outright thieves.
YouTube is great for learning things, but it’s also full of absolutely terrible demonstration videos which purport to teach you some skill, but instead offer absurd and underdeveloped techniques (these videos are often called “lifehacks” for some reason, if you’re unfamiliar with the phenomenon and want to search for it). The videos are being optimized for transmission rather than usefulness. Acquiring useful information requires prudent optimization against this.
Social Capital is, roughly, the amount of trust you have within a group. Bowling Alone is a book which researches America’s decline in social capital over the course of the 1900s. Trust in the goodwill of strangers took a dramatic dive over that time period, with corresponding negative consequences (EG, the decline in hitchhiking, the rise of helicopter parenting).
You might think this is due to the increasingly “horizontal” environment. More travel, more free-market capitalism, bigger cities, the decline of small towns; more horizontal spread of memes, by print, radio, television, and internet; more science and education.
And you might be right.
Paradox Three: Free-market societies have higher social capital. Citation: The Moral Economy, Samuel Bowles.
More generally: a lot of things are a lot better than naive horizontal/vertical thinking would suggest. I’ve already mentioned that a lot of the things I hold dear seem to have a pretty horizontal transmission model. I don’t think that’s just because I’ve been taken over by virulent memes.
By the way, my favorite explanation of the decline in social capital over the 1900s is this: there was, for some reason, a huge burst of club-making in the late 1800s, which continued into the early 1900s. These clubs were often very civically active, contributing to a common perception that everyone cooperates together to improve society. This culminated in an extremely high degree of social capital in “The Greatest Generation”—however, that generation was already starting to forget the club-making/club-attending culture which had fuelled the increase in social capital. Television ultimately killed or put the damper on the clubs, because most people wanted to catch their favorite shows in the evening rather than go out. Social capital gradually declined from then on.
(But, doubtless, there was more going on than just this, and I have no idea how big a factor club culture really plays.)
Why do so many good things have horizontal transmission structures?
How should we think about horizontal transmission, normatively? Specifically, “paradox two” is an argument that horizontal-transmission practices, while enticing, can “burn the commons” of collective goodwill by opening up things for predatory/parasitic dynamics. Yet the conclusion seems severe and counterintuitive.
Why do free-market societies have higher social capital? How can this be fit into a larger picture in which horizontal transmission structures / few-shot interactions incentivize less cooperative strategies?