OK, I don’t want to sound too harsh, and I’m really happy that you are writing and trying to do stuff etc. You can only be criticized because you wrote something, rather than nothing at all. This is laudable.
So. In the spirit of constructive criticism, I want to try to express why I can’t stand this post. Some of my criticism involves subtleties that might be hard to get across here, but we can continue discussing this on Skype etc.
The post claims to criticize a simplistic approach to understanding akrasia, but what it offers as an alternative is just another simplistic approach, which is to my eyes not much different and stuck in all the same harmful frames (“us vs them”, “solve”, “just aren’t remembering” etc.)
The naive intuition that “akrasia” points to some living, coherent thing with its own goals, which knows stuff “you” don’t and is sometimes actively undermining “you”, is actually a pretty good description of what is really happening. The OP is really trying to make it not be the case, but it seems driven more by wishful thinking than looking at reality. Maybe my intuition is wrong here but it tells me that you are doing some mental motion that puts you in a sorta fake manic optimism state to even be able to write a post like this. Some of that might be a good thing, but if you use too much it starts to sound very unconvincing to me.
The thing akrasia really points to is something like advanced model-based reinforcement learning on goals that you are not consciously unaware of, that is directly in control of your actions. (It is much more than the “conserve energy” theory mentioned in other comments, though “conserve energy” is indeed included as a significant goal.) The models used in it learn and predict the results of you following some plan or strategy used to reduce akrasia, including clever meta-plans like “decompose the problem and apply my best tools to solve each part”. This is precisely the reason why most “solutions” only last for a pretty short time, and your post offers nothing to change this state of affairs beyond “here’s another piece of meat to throw into the jaws of the monster, have fun with the additional 15 seconds that it earns you”.
I don’t know either you or OP, and am not in either of your heads. However, from this and other things OP wrote, I suspect that this model has worked for them. From what you’re saying in point 2, this model would not work for you and fails to accurately describe what’s going on in your head.
Lifelonglearner, I suspect you are making a broader claim than you can support, but that the narrower claim inside it is accurate. I agree with you that giving people new models may sometimes be net-negative. I agree with you that tabooing and/or breaking down akrasia is probably a useful tool to get at a symptom and figure out a way to deal with that symptom. I intend to give it a shot, and I’ll make sure to note what kind of results I get- if you like, I’d be pretty happy to put those notes somewhere you can get at them if you want datapoints to refine this.
I think akrasia is a useful model when in non-taboo form. When I read “instead of treating akrasia as an abstraction that unifies a class of self-imposed problems that share the property of acting as obstacles towards our goals, we treat it as a problem onto itself” my first thought was that description made it sounded like depression. Depression isn’t exactly self-imposed, but it can look an awful lot like it is. It’s worth noting that attacking depression as an entity of itself can be far more effective than trying to break it down into symptoms.
(I’m not suggesting that akrisia === depression! I am suggesting that there exist abstractions that unify a set of symptoms that are best attacked as the abstraction they are.)
I don’t know either you or OP, and am not in either of your heads. However, from this and other things OP wrote, I suspect that this model has worked for them.
I personally prefer the aesthetic of figuring out how stuff works as a priority. And one could argue that this also gives greater practical benefits in the long run… but sometimes the “long run” is pretty long indeed.
I personally prefer the aesthetic of figuring out how stuff works as a priority.
To me these are somewhat orthogonal issues. I like cute/pretty/elegant and I like practical/functional/effective, but I’m willing to get them from different things. Of course, if something manages to combine the two, that’s great.
I think there may be confusions about the things that we’re pointing to and it seems useful to try and clear those up first.
First off, by akrasia I’m pointing at situations where you look back and think, “Drat! In that instant, I could have done this other thing that would have better served my goals.”
(It may actually be the case that you were satisfying some other hidden part of yourself during that time. I’m not implying that instances of additional counterfactual productivity always indicate realistic places to improve, merely that they appear to be.)
Or something like that. Happy to try and zero in on a better working definition if you’d like to propose something different. I think the things we have in mind of examples of akrasia may not be the same.
Secondly, I want to stress that I don’t think that forcing / a brute force approach towards productivity is not optimal. Often, the sort of internal debugging of an Internal Double Crux and Focusing flavor seems like the right way to debug internal aversions as well as move forward.
The sort of “dropping your obligations” and sourcing internal motivation that Nate Soares writes about in his Replacing Guilt series is, I think, very important.
I also think, though, that this model sometimes isn’t at the right level of granularity to deal with smaller problems like forgetting to get things done or getting sucked into distraction-loops. For this, I think the above algorithm can be useful.
Models of akrasia that personify it or model it as a malicious agent don’t feel tractable or intuitive to me. Given the statement above, I want to once again stress that I don’t think this is a solution to all problems related to getting things done. I do think that the reductionist framing and considering the potential dangers of reification are important, though. Hope that tempers any impressions you have of optimism on my part.
I read the OP and I thought I understood really well what you were trying to say. I wrote some criticism.
You read the criticism and thought that I failed to understand you. You answered by pointing to this apparent misunderstanding and reiterating your view from the OP in a slightly different way.
I read your answer and I thought that your answer meant that you had understood around 0% of my original comment. I am now backtracking to point out how much my criticism was not about what you thought it was about.
I think there may be confusions about the things that we’re pointing to and it seems useful to try and clear those up first.
When not specified, I assume standard definitions like “a lack of self-control or the state of acting against one’s better judgment” (from Wikipedia), and your “want-want vs want” or “being unable to do what you ‘want’ to do” seems to capture it also. I don’t think that this was a problem in our communication. Your clarifications above are exactly what I’d have expected after reading just the OP.
Often, the sort of internal debugging of an Internal Double Crux and Focusing flavor seems like the right way to debug internal aversions as well as move forward.
I feel like you’re answering by pulling from your database of “smart things to say about akrasia”, instead of actually reading what I’m trying to say. This is frustrating. The database obviously includes CFAR stuff, like “Double Crux > willpower” and “reductionism + use tools to solve problems”, and Nate’s “Replacing Guilt”. But I know those sources too, I value them and I’m not arguing about these points at all.
Models of akrasia that personify it or model it as a malicious agent
This is some strawman that you are pulling from the naive approach, and not at all what I was saying.
“living, coherent thing” → this doesn’t mean “person”
“which knows stuff “you” don’t and is sometimes actively undermining “you”″ → this doesn’t mean “malicious”
I do think that the reductionist framing and considering the potential dangers of reification are important, though.
And once again, you are rebutting a strawman of what I said. “model-based reinforcement learning on goals that you are not consciously unaware of” is something that can be usefully treated in a reductionist way, and not a reification but the best short description of what’s most likely out there in reality that I could think of.
Uh, I’m sorry, I’m tired now, we can continue trying to understand each other’s points tomorrow. So far, I am losing hope
Sounds good, I’ll try to follow-up more later. You’re right that I wasn’t engaging with your points. There were a few things I thought I explained poorly in the original post and wanted to make sure those weren’t the points of disagreement.
This conversation didn’t move forward, which I think is unfortunate.
I think SquirrelInHell agrees with something like the view described on sinceriously.fyi, especially false faces. If you read that, you may see where the objection was coming from.
I don’t think such a model of Akrasia is productive. I don’t think such a model will help you overcome your inability to do that which you desire. Insomuch as such a model of Akrasia does not help you achieve your goals, it is not rational;
OK, I don’t want to sound too harsh, and I’m really happy that you are writing and trying to do stuff etc. You can only be criticized because you wrote something, rather than nothing at all. This is laudable.
So. In the spirit of constructive criticism, I want to try to express why I can’t stand this post. Some of my criticism involves subtleties that might be hard to get across here, but we can continue discussing this on Skype etc.
The post claims to criticize a simplistic approach to understanding akrasia, but what it offers as an alternative is just another simplistic approach, which is to my eyes not much different and stuck in all the same harmful frames (“us vs them”, “solve”, “just aren’t remembering” etc.)
The naive intuition that “akrasia” points to some living, coherent thing with its own goals, which knows stuff “you” don’t and is sometimes actively undermining “you”, is actually a pretty good description of what is really happening. The OP is really trying to make it not be the case, but it seems driven more by wishful thinking than looking at reality. Maybe my intuition is wrong here but it tells me that you are doing some mental motion that puts you in a sorta fake manic optimism state to even be able to write a post like this. Some of that might be a good thing, but if you use too much it starts to sound very unconvincing to me.
The thing akrasia really points to is something like advanced model-based reinforcement learning on goals that you are not consciously unaware of, that is directly in control of your actions. (It is much more than the “conserve energy” theory mentioned in other comments, though “conserve energy” is indeed included as a significant goal.) The models used in it learn and predict the results of you following some plan or strategy used to reduce akrasia, including clever meta-plans like “decompose the problem and apply my best tools to solve each part”. This is precisely the reason why most “solutions” only last for a pretty short time, and your post offers nothing to change this state of affairs beyond “here’s another piece of meat to throw into the jaws of the monster, have fun with the additional 15 seconds that it earns you”.
I don’t know either you or OP, and am not in either of your heads. However, from this and other things OP wrote, I suspect that this model has worked for them. From what you’re saying in point 2, this model would not work for you and fails to accurately describe what’s going on in your head.
Lifelonglearner, I suspect you are making a broader claim than you can support, but that the narrower claim inside it is accurate. I agree with you that giving people new models may sometimes be net-negative. I agree with you that tabooing and/or breaking down akrasia is probably a useful tool to get at a symptom and figure out a way to deal with that symptom. I intend to give it a shot, and I’ll make sure to note what kind of results I get- if you like, I’d be pretty happy to put those notes somewhere you can get at them if you want datapoints to refine this.
I think akrasia is a useful model when in non-taboo form. When I read “instead of treating akrasia as an abstraction that unifies a class of self-imposed problems that share the property of acting as obstacles towards our goals, we treat it as a problem onto itself” my first thought was that description made it sounded like depression. Depression isn’t exactly self-imposed, but it can look an awful lot like it is. It’s worth noting that attacking depression as an entity of itself can be far more effective than trying to break it down into symptoms.
(I’m not suggesting that akrisia === depression! I am suggesting that there exist abstractions that unify a set of symptoms that are best attacked as the abstraction they are.)
I’m sorry, but I’m not talking about what is useful in someone’s head. I’m referring to the best current ideas of how this really plays out in the brain. For some background, try https://sideways-view.com/2017/02/19/the-monkey-and-the-machine-a-dual-process-theory/
I’ll take “what is useful in someone’s head” over some harebrained ideas about how it really plays out any day.
Praxis is the criterion of truth :-P
Legit :-)
I personally prefer the aesthetic of figuring out how stuff works as a priority. And one could argue that this also gives greater practical benefits in the long run… but sometimes the “long run” is pretty long indeed.
To me these are somewhat orthogonal issues. I like cute/pretty/elegant and I like practical/functional/effective, but I’m willing to get them from different things. Of course, if something manages to combine the two, that’s great.
Hey, thanks for the thoughts!
I think there may be confusions about the things that we’re pointing to and it seems useful to try and clear those up first.
First off, by akrasia I’m pointing at situations where you look back and think, “Drat! In that instant, I could have done this other thing that would have better served my goals.”
(It may actually be the case that you were satisfying some other hidden part of yourself during that time. I’m not implying that instances of additional counterfactual productivity always indicate realistic places to improve, merely that they appear to be.)
Or something like that. Happy to try and zero in on a better working definition if you’d like to propose something different. I think the things we have in mind of examples of akrasia may not be the same.
Secondly, I want to stress that I don’t think that forcing / a brute force approach towards productivity is not optimal. Often, the sort of internal debugging of an Internal Double Crux and Focusing flavor seems like the right way to debug internal aversions as well as move forward.
The sort of “dropping your obligations” and sourcing internal motivation that Nate Soares writes about in his Replacing Guilt series is, I think, very important.
I also think, though, that this model sometimes isn’t at the right level of granularity to deal with smaller problems like forgetting to get things done or getting sucked into distraction-loops. For this, I think the above algorithm can be useful.
Models of akrasia that personify it or model it as a malicious agent don’t feel tractable or intuitive to me. Given the statement above, I want to once again stress that I don’t think this is a solution to all problems related to getting things done. I do think that the reductionist framing and considering the potential dangers of reification are important, though. Hope that tempers any impressions you have of optimism on my part.
So here’s what happened until now:
I read the OP and I thought I understood really well what you were trying to say. I wrote some criticism.
You read the criticism and thought that I failed to understand you. You answered by pointing to this apparent misunderstanding and reiterating your view from the OP in a slightly different way.
I read your answer and I thought that your answer meant that you had understood around 0% of my original comment. I am now backtracking to point out how much my criticism was not about what you thought it was about.
When not specified, I assume standard definitions like “a lack of self-control or the state of acting against one’s better judgment” (from Wikipedia), and your “want-want vs want” or “being unable to do what you ‘want’ to do” seems to capture it also. I don’t think that this was a problem in our communication. Your clarifications above are exactly what I’d have expected after reading just the OP.
I feel like you’re answering by pulling from your database of “smart things to say about akrasia”, instead of actually reading what I’m trying to say. This is frustrating. The database obviously includes CFAR stuff, like “Double Crux > willpower” and “reductionism + use tools to solve problems”, and Nate’s “Replacing Guilt”. But I know those sources too, I value them and I’m not arguing about these points at all.
This is some strawman that you are pulling from the naive approach, and not at all what I was saying.
“living, coherent thing” → this doesn’t mean “person”
“which knows stuff “you” don’t and is sometimes actively undermining “you”″ → this doesn’t mean “malicious”
And once again, you are rebutting a strawman of what I said. “model-based reinforcement learning on goals that you are not consciously unaware of” is something that can be usefully treated in a reductionist way, and not a reification but the best short description of what’s most likely out there in reality that I could think of.
Uh, I’m sorry, I’m tired now, we can continue trying to understand each other’s points tomorrow. So far, I am losing hope
Sounds good, I’ll try to follow-up more later. You’re right that I wasn’t engaging with your points. There were a few things I thought I explained poorly in the original post and wanted to make sure those weren’t the points of disagreement.
This conversation didn’t move forward, which I think is unfortunate.
I think SquirrelInHell agrees with something like the view described on sinceriously.fyi, especially false faces. If you read that, you may see where the objection was coming from.
I don’t think such a model of Akrasia is productive. I don’t think such a model will help you overcome your inability to do that which you desire. Insomuch as such a model of Akrasia does not help you achieve your goals, it is not rational;