I’d say it seems pretty fundamental that Religion is a kind of meme. One measure of success for memes is their ability to spread (virulence). If that is your only measure, you’re likely to have some ethically terrible things going on. It seems (to me) like an obvious constraint that the meme must spread without causing any obvious harm in its wake (except to related memes that may be in competition).
What useful purposes does religion serve?
I consider [create an in-group] to be a pretty central (usually unstated) optimization point (not exactly a purpose) in most religions. This has some well-studied psychological effects on said group that can benefit all its members socially, psychologically, and (further down the causal line) medically. Although it tends to lead to some unfortunate side-effects for the out-group.
How would you design a “rational religion”, if such an entity is possible?
Problem is that I’ve seen more than one source define “religion” as something like “systematic belief in the supernatural”. I’m not convinced that such belief can be “rational” (optimal for making more effective decisions). Perhaps as a stepping stone—use a religious-type belief system as an infection vector, then strip the supernatural elements away bit by bit; but that seems awfully deceitful to me.
This seems like a promising direction. So let’s say that by religion I mean a useful meme. The meme consists of a doctrine. That is, a collection of statements regarding human belief or action. A person infected with the meme lives in accordance with the doctrine. The doctrine is designed to cause a useful effect, apart from its own flourishing. The effect is caused by changes in behavior of the people infected by the meme. The effect need not be explicitly stated within the doctrine. This a broad definition of religion, as it doesn’t require that the doctrine contain any statements about the supernatural. But at this point I want to keep things open. As some posts have suggested, there’s a lot of argument about what does or doesn’t constitute a religion.
In order to be successful, a useful meme must be sufficiently:
Transmissible, meaning that it must be pragmatically possible to expose many people to the meme.
Infectious, meaning that a significant number of people exposed to the meme must become infected.
Contagious, meaning that it must spread of its own accord, until it has infected a threshold number of carriers. The threshold may differ from meme to meme, and is whatever threshold is needed to enable the successful creation of the desired useful effect.
Viable, meaning that it must be possible for a person to survive and prosper when infected with the meme.
Robust, meaning that once a given person is infected, it must resist eradication within that individual.
Hardy, meaning that it must be able survive, and possibly flourish, in a variety of intellectual environments, and there must be no commonly found environment in which it cannot survive.
Resistant. Meaning that beyond a certain degree of spread, there cannot be an obvious method by which a (presumably non-infected) person or group can eradicate it.
Stable, meaning that it must not change its nature significantly over time. Changes that continue to allow it to flourish and to cause the desired effect are permissible.
This list is not meant to be definitive or exhaustive. And I don’t claim to be using the best terminology.
For example, some religious doctrines contain the idea that if you cease to believe in any part of the doctrine, you will suffer in Hell upon death. This idea might enhance the robustness of the religion by discouraging the development of disbelief. Others contain the idea that it’s your duty, or that you’re rewarded in some fashion, for converting non-believers. This idea might enhance the contagiousness of the religion by encouraging those who are already infected by the meme to work to infect others.
Using this framework, perhaps the original post might be improved a bit. Putting some of the questions asked in the original post into the new framework, we get:
What useful purposes does religion serve? That is, what kinds of “useful purposes” can be designed into such memes?
Are any of these purposes non-supernaturalistic in nature? That is, can a completely non-supernatural religion flourish and create a useful effect?
What is success for a religion and what elements of a religion tend to cause it to become successful? That is, what elements of currently existing religious doctrines have helped them to become successful?
How would you design a “rational religion”, if such an entity is possible? That is, is it possible to design a religion that encourages “less wrong” cognition?
What are the relationships between aspects of a religion, and outcomes involving that religion? For example, Catholic doctrine includes elements that discourage birth control. Lack of birth control encourages higher birthrates among Catholics. This encourages there to be a larger number of Catholics in the next generation than would otherwise be the case. Thus the toolkit now contains one element: the prohibition of birth control as a means of increasing the contagiousness of a religion.
How do aspects of religion cause them to evolve differently over time? For example, Catholicism contains a permanent, authoritarian hierarchy of individuals who are dependent on the church for their survival and satisfaction. Without the Church, a Cardinal with no other job skills might starve or suffer disrespect from the community. Thus he is incentivized to increase the authority of the Catholic church in order to help secure his own survival. Over time, Catholic doctrine adopted the idea that the Pope is infallible. It might be possible to draw a line between the two phenomena and say that in a doctrine which supports a social structure that includes a permanent, authoritarian hierarchy of individuals who are dependent on believers for their survival and satisfaction, the doctrine will tend to evolve towards the accumulation of power for the hierarchy so as to ensure and increase their survival and satisfaction. Possibly another element for the toolkit.
What other such heuristics exist? Would a large enough collection of such heuristics aid in the analysis and design of religious movements?
I’m doing a little reading on both of them now. Big question: how to make them successful as social movements. I wonder if their elements can be modeled in a fashion similar to that which I did above. And if so, if there’s anything that such an application can tell us about how to improve their chances for success.
I don’t think it’s even helpful to model them in that way.
Your list for example doesn’t contain a point for “community”. It doesn’t consider the need for charismatic leaders.
There not much reason to focus on issues like contraception because a good new ideology can spread much faster if the community can recruit well and give new members a lot of value.
This sort of problem is susceptible to the red queen’s race—if you solve it, that just means that lots of people will copy you, and you will find yourself in competition with a new crop of viral religions. You don’t make much long-term progress by being innovative, so perhaps your best bet is to copy some moderately successful cults and hope that power won’t corrupt you. The Hare Krishnas are a good template—just avoid the racketeering, mail fraud, murder, and child abuse, and you should do fine.
I’d say it seems pretty fundamental that Religion is a kind of meme. One measure of success for memes is their ability to spread (virulence). If that is your only measure, you’re likely to have some ethically terrible things going on. It seems (to me) like an obvious constraint that the meme must spread without causing any obvious harm in its wake (except to related memes that may be in competition).
I consider [create an in-group] to be a pretty central (usually unstated) optimization point (not exactly a purpose) in most religions. This has some well-studied psychological effects on said group that can benefit all its members socially, psychologically, and (further down the causal line) medically. Although it tends to lead to some unfortunate side-effects for the out-group.
Problem is that I’ve seen more than one source define “religion” as something like “systematic belief in the supernatural”. I’m not convinced that such belief can be “rational” (optimal for making more effective decisions). Perhaps as a stepping stone—use a religious-type belief system as an infection vector, then strip the supernatural elements away bit by bit; but that seems awfully deceitful to me.
This seems like a promising direction. So let’s say that by religion I mean a useful meme. The meme consists of a doctrine. That is, a collection of statements regarding human belief or action. A person infected with the meme lives in accordance with the doctrine. The doctrine is designed to cause a useful effect, apart from its own flourishing. The effect is caused by changes in behavior of the people infected by the meme. The effect need not be explicitly stated within the doctrine. This a broad definition of religion, as it doesn’t require that the doctrine contain any statements about the supernatural. But at this point I want to keep things open. As some posts have suggested, there’s a lot of argument about what does or doesn’t constitute a religion.
In order to be successful, a useful meme must be sufficiently:
Transmissible, meaning that it must be pragmatically possible to expose many people to the meme.
Infectious, meaning that a significant number of people exposed to the meme must become infected.
Contagious, meaning that it must spread of its own accord, until it has infected a threshold number of carriers. The threshold may differ from meme to meme, and is whatever threshold is needed to enable the successful creation of the desired useful effect.
Viable, meaning that it must be possible for a person to survive and prosper when infected with the meme.
Robust, meaning that once a given person is infected, it must resist eradication within that individual.
Hardy, meaning that it must be able survive, and possibly flourish, in a variety of intellectual environments, and there must be no commonly found environment in which it cannot survive.
Resistant. Meaning that beyond a certain degree of spread, there cannot be an obvious method by which a (presumably non-infected) person or group can eradicate it.
Stable, meaning that it must not change its nature significantly over time. Changes that continue to allow it to flourish and to cause the desired effect are permissible.
This list is not meant to be definitive or exhaustive. And I don’t claim to be using the best terminology.
For example, some religious doctrines contain the idea that if you cease to believe in any part of the doctrine, you will suffer in Hell upon death. This idea might enhance the robustness of the religion by discouraging the development of disbelief. Others contain the idea that it’s your duty, or that you’re rewarded in some fashion, for converting non-believers. This idea might enhance the contagiousness of the religion by encouraging those who are already infected by the meme to work to infect others.
Using this framework, perhaps the original post might be improved a bit. Putting some of the questions asked in the original post into the new framework, we get:
What useful purposes does religion serve? That is, what kinds of “useful purposes” can be designed into such memes?
Are any of these purposes non-supernaturalistic in nature? That is, can a completely non-supernatural religion flourish and create a useful effect?
What is success for a religion and what elements of a religion tend to cause it to become successful? That is, what elements of currently existing religious doctrines have helped them to become successful?
How would you design a “rational religion”, if such an entity is possible? That is, is it possible to design a religion that encourages “less wrong” cognition?
What are the relationships between aspects of a religion, and outcomes involving that religion? For example, Catholic doctrine includes elements that discourage birth control. Lack of birth control encourages higher birthrates among Catholics. This encourages there to be a larger number of Catholics in the next generation than would otherwise be the case. Thus the toolkit now contains one element: the prohibition of birth control as a means of increasing the contagiousness of a religion.
How do aspects of religion cause them to evolve differently over time? For example, Catholicism contains a permanent, authoritarian hierarchy of individuals who are dependent on the church for their survival and satisfaction. Without the Church, a Cardinal with no other job skills might starve or suffer disrespect from the community. Thus he is incentivized to increase the authority of the Catholic church in order to help secure his own survival. Over time, Catholic doctrine adopted the idea that the Pope is infallible. It might be possible to draw a line between the two phenomena and say that in a doctrine which supports a social structure that includes a permanent, authoritarian hierarchy of individuals who are dependent on believers for their survival and satisfaction, the doctrine will tend to evolve towards the accumulation of power for the hierarchy so as to ensure and increase their survival and satisfaction. Possibly another element for the toolkit.
What other such heuristics exist? Would a large enough collection of such heuristics aid in the analysis and design of religious movements?
I think Effective Altruism fits the eight criteria you gave.
I don’t think rationality!CFAR currently has all eight at the moment but I think there a good chance that it will get them in the future.
In both cases calling them religions is likely not helpful.
I’m doing a little reading on both of them now. Big question: how to make them successful as social movements. I wonder if their elements can be modeled in a fashion similar to that which I did above. And if so, if there’s anything that such an application can tell us about how to improve their chances for success.
I don’t think it’s even helpful to model them in that way. Your list for example doesn’t contain a point for “community”. It doesn’t consider the need for charismatic leaders.
There not much reason to focus on issues like contraception because a good new ideology can spread much faster if the community can recruit well and give new members a lot of value.
This sort of problem is susceptible to the red queen’s race—if you solve it, that just means that lots of people will copy you, and you will find yourself in competition with a new crop of viral religions. You don’t make much long-term progress by being innovative, so perhaps your best bet is to copy some moderately successful cults and hope that power won’t corrupt you. The Hare Krishnas are a good template—just avoid the racketeering, mail fraud, murder, and child abuse, and you should do fine.