As a part of our In-Depth Fellowship group this winter, we took time to consider how to answer this question in our consideration of longtermism. After the discussion, members of our fellowship wrote up our considerations into this forum post to bring our discussion to the wider community. While we are not overly confident in the contents, we hope this framework may contribute to the broader conversation of what longtermism may look like in everyday situations and encourage further thought and action.
But how can we go about productively thinking and discussing these ideas?
Our key intuition is that this level of everyday decision making should be grounded on very concrete examples. These examples should codify norms of conduct that we want society at large to adopt.
This is hardly a novel idea. Think about washing your hands.
It is a concrete action tied to a concrete stimulus—washing your hands is something you do after the toilet.
It has unambiguously good longtermist consequences in expectation—this helps society avoid disease at scale.
It is scrutable—people are more or less aware of the basic reason why washing your hands is such a good idea
We think those are good things that a good everyday longtermist norm ought to have. Washing your hands is, as we see it, a success story of an everyday longtermist rule.
In the rest of this post we will describe some longtermist norms in vignettes, through imagined examples. They are mostly things that are not that usual, but that we might want to happen more often.
Our main goal with this post is to suggest a standard of what everyday longtermism research might look like: introducing everyday norms through vignettes, and discussing the consequences of widespread adoption of the norm.
We are not confident that this is the best way of doing research in longtermism, but we think it is worth exploring. Do you have other ideas of norms to add that you practice or others should consider practicing? Let us know what you think in the comments!
Example 1:
Anna has as part of her job to decide whether to allocate funds to a research project within her company. She has already made her decision, but writes out her reasoning and shares it with colleagues. This helps her reexamine her assumptions and makes the process more scrutable.
Example 2:
Bob leads a team across time zones with much collaborative work. To help ensure that excellent practices of sleep, hydration, nutrition, and exercise are prioritized, he routinely includes one area of these on their quarterly team dashboard. This encourages the team to prioritize individual team members’ health, energy, and longevity in their time management and decisions in line with their ambitious goals.
Example 3
Casey is reviewing a CV for an applicant. The applicant does not meet the formal requirements for the job, but Casey wants to hire them anyway. Casey visualices the hundreds of people making a similar decision to theirs. They would be ok with hiring this specific applicant, because Casey trusts their instincts a lot. But Casey would not trust 100 people in a similar position to make the right choice; ignoring the recruiting guidelines might disadvantage individuals from underrepresented groups in an illegible way. Casey decides that the best course would be if everyone chose to just follow the procedure, so they choose to forgo their intuition to favor better decision making overall.
Example 4
Dylan is offered a job as a Machine Learning engineer to help the police with some automated camera monitoring. Before deciding whether to accept, he seeks out open criticism of that line of work, and tries to imagine what are some likely consequences of developing that kind of technology, both positive and negative. After doing a balance he realizes that while it will most likely be positive there is a plausible chance that it will enable abuse and illegitimate breaches of privacy, and rejects the job offer.
Example 5
Eve reads an interesting article. She wants to share it on social media. She could spend some effort paraphrasing the key ideas in the article, or just share the link. She has internalized that spending one minute summarizing key ideas might well be worth a lot of time saved to her friends who otherwise could use her summary to decide whether to read the whole article. Out of habit she summarizes the article as best as she can, making it clear who she genuinely thinks would benefit from reading the article.
Example 6
Frankie is a plant manager in an office with ~50 people. She hears an alert of a possible pandemic spreading through her country. While the authorities are not very concerned, she uses this as a chance to write an email to her colleagues, with a brainstorm of cheap ways to make the place more sanitary—eg distributing masks among workers, making gel available, hanging posters indicating how to wash your hands and reminding workers to stay home if they feel sick.
Everyday longtermism in practice
Owen Cotton-Barratt wrote a post introducing a key question for longtermism:
What does longtermism recommend doing in all sorts of everyday situations?
As a part of our In-Depth Fellowship group this winter, we took time to consider how to answer this question in our consideration of longtermism. After the discussion, members of our fellowship wrote up our considerations into this forum post to bring our discussion to the wider community. While we are not overly confident in the contents, we hope this framework may contribute to the broader conversation of what longtermism may look like in everyday situations and encourage further thought and action.
But how can we go about productively thinking and discussing these ideas?
Our key intuition is that this level of everyday decision making should be grounded on very concrete examples. These examples should codify norms of conduct that we want society at large to adopt.
This is hardly a novel idea. Think about washing your hands.
It is a concrete action tied to a concrete stimulus—washing your hands is something you do after the toilet.
It has unambiguously good longtermist consequences in expectation—this helps society avoid disease at scale.
It is scrutable—people are more or less aware of the basic reason why washing your hands is such a good idea
We think those are good things that a good everyday longtermist norm ought to have. Washing your hands is, as we see it, a success story of an everyday longtermist rule.
In the rest of this post we will describe some longtermist norms in vignettes, through imagined examples. They are mostly things that are not that usual, but that we might want to happen more often.
Our main goal with this post is to suggest a standard of what everyday longtermism research might look like: introducing everyday norms through vignettes, and discussing the consequences of widespread adoption of the norm.
We are not confident that this is the best way of doing research in longtermism, but we think it is worth exploring. Do you have other ideas of norms to add that you practice or others should consider practicing? Let us know what you think in the comments!
Example 1:
Anna has as part of her job to decide whether to allocate funds to a research project within her company. She has already made her decision, but writes out her reasoning and shares it with colleagues. This helps her reexamine her assumptions and makes the process more scrutable.
Example 2:
Bob leads a team across time zones with much collaborative work. To help ensure that excellent practices of sleep, hydration, nutrition, and exercise are prioritized, he routinely includes one area of these on their quarterly team dashboard. This encourages the team to prioritize individual team members’ health, energy, and longevity in their time management and decisions in line with their ambitious goals.
Example 3
Casey is reviewing a CV for an applicant. The applicant does not meet the formal requirements for the job, but Casey wants to hire them anyway. Casey visualices the hundreds of people making a similar decision to theirs. They would be ok with hiring this specific applicant, because Casey trusts their instincts a lot. But Casey would not trust 100 people in a similar position to make the right choice; ignoring the recruiting guidelines might disadvantage individuals from underrepresented groups in an illegible way. Casey decides that the best course would be if everyone chose to just follow the procedure, so they choose to forgo their intuition to favor better decision making overall.
Example 4
Dylan is offered a job as a Machine Learning engineer to help the police with some automated camera monitoring. Before deciding whether to accept, he seeks out open criticism of that line of work, and tries to imagine what are some likely consequences of developing that kind of technology, both positive and negative. After doing a balance he realizes that while it will most likely be positive there is a plausible chance that it will enable abuse and illegitimate breaches of privacy, and rejects the job offer.
Example 5
Eve reads an interesting article. She wants to share it on social media. She could spend some effort paraphrasing the key ideas in the article, or just share the link. She has internalized that spending one minute summarizing key ideas might well be worth a lot of time saved to her friends who otherwise could use her summary to decide whether to read the whole article. Out of habit she summarizes the article as best as she can, making it clear who she genuinely thinks would benefit from reading the article.
Example 6
Frankie is a plant manager in an office with ~50 people. She hears an alert of a possible pandemic spreading through her country. While the authorities are not very concerned, she uses this as a chance to write an email to her colleagues, with a brainstorm of cheap ways to make the place more sanitary—eg distributing masks among workers, making gel available, hanging posters indicating how to wash your hands and reminding workers to stay home if they feel sick.
--------------------------------------------------------------------
In conclusion:
We have suggested a concrete methodology to study everyday longtermism: exploring vignettes of behaviour we want to consider promoting.
We have given some examples of how to superficially apply this method.
This is hardly novel, but we hope this opens up the door for more discussion on this area.