= 27d567bc2d48d9f40e9ccc20ea370b15)
DanArmak comments on Fiction Considered Harmful - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (80)
The consumption of fiction is a way to produce a pleasurable experience. We act in a certain way to induce a desirable state of mind and body. How is fiction different from eating tasty food, listening to moving music, going on a nature trip with pretty views, or having sex? These other things also use time and energy that could be "spent on fact".
What's wrong with "escapism", and why do you call it "shallow"?
If you argue that we should spend more resources (time, attention, money, etc) on improving this world rather than "escaping" into a fictional one. Then why would you not also argue we shouldn't "escape" using music, sex, or ice cream? What makes sex or music part of this world but (the experience of) fiction not? Both engage the senses. (Fiction doesn't yet engage taste or touch, but when technology allows it to do so, it will and it will be better fiction for it.)
If you argue that we should make this world more like the fictional ones so we can enjoy it directly, it seems neither possible nor desirable. A big attraction of fiction is that everyone can choose and consume different works and genres at different times and places, while we all live in the same world. Another is that people enjoy e.g. tragedies they wouldn't want to happen in real life. (As well as most comedies, for that matter.)
Fiction is used to trigger strong emotional responses, but we don't want the real world to contain things that would trigger them, because things in the real world aren't chosen by people - they're the precise thing we need escape from.
I think abramdemski doesn't construe this as a either/or-dichotomy but as a continuous dimention from max-productive to wireheading. And he seems to argue that fiction is more on the wireheading side:
You ask for other aspects like listening to music and sex:
And I think you can place these also somewhere on the space. For music in particular there is not a single place but a range from jolly music song and laughter in a community to passive listening to random music during a commute. Hm, I guess the same range can be shown for sex.
Then, what places fiction further along the axis than these other things? What makes it less "productive"?
How do you define "productive" in the first place - productive towards what goals? What is special about those goals (whatever they are) that is different from consuming fiction as a goal, or a means towards the goal of experiencing the kinds of thoughts and emotions fiction produces? Sex, music, ice cream and mountain climbing are also "merely" means towards brain states.
As I inticated in this reply, I think the wireheading claim is separate from the productivity question. Fiction being productive (as a means of relaxing or stretching the imagination or such) is something that could save it from the wireheading complaint, because then it would not have to be justified as an end-in-itself.
The way I see it, wireheading is a complaint about things which produce counterfeit value. If the complaint is valid, wirehead-like things are things which fool the brain into thinking there is value where there is none. In the case of a literal wirehead, this is by direct stimulation of the value-detecting brain-bits. In the case of fiction, it's by a kind of superstimulus for social interaction and other human values. No actual social interaction takes place, but you feel as if you are getting to know people (usually important, sexy people) and getting caught up in an important chain of events.
The question is: is this counterfeit value, or real value?
Thanks for the explanation. This lets me get back to my original point: why focus on fiction? What makes fiction more like wireheading or more 'counterfeit' than the great majority of the things we do that aren't immediately necessary for survival? Compare the following:
All these activities also contribute to various forms of social bonding and of mental health, but that's just as true for fiction.
These are intended as a few random examples out of many others. Other than working to earn money, maintaining social relations, and a few necessary maintenance activities like shopping, almost all our actions are intended to create pleasurable experiences internal to our brains. What makes fiction different from any other such activity, and the value it creates more 'counterfeit' or wireheading-like?
I feel confused, and am likely missing some bad assumption. For the purpose of working through the assumptions, I'll keep arguing the anti-fiction side...
The part of me that feels like doing away with fiction could be a good idea also would be OK with doing away with many of those other things you mentioned. Eating ice cream is bad as a matter of fact (this doesn't seem to require much argument). It's just a superstimulus for "good food", and furthermore, negatively impacts health. Noticing this (consciously noticing it on a repeated basis) can in fact move preferences away from ice cream and toward healthier food, to the point where ice cream doesn't even feel tempting except socially.
(My actual motivational state is not like this, but rather flips back and forth between finding ice cream appealing and not. I have not decided to adjust my emotional state entirely toward the reality, largely because this change in motivational state would have some negative social consequences.)
Trips to beautiful natural sites do seem kind of silly to me. Looking at nice scenery is nice, but on the order of nice things, it seems like something I'm willing to pay significantly less for than what most people are. That's neither here nor there for the debate, though. The part of me that is interested in doing away with fiction says that at least this experience is fact-oriented. There is something valuable about going and seeing real scenery -- historical sites of importance, and things like that -- which is not there when the scenery is entirely simulated. The part of me concerned with wireheading says that this is enough to distinguish between the kind of pleasure produced by visiting real places vs simulating pleasant scenery.
The difference between real and simulated scenery in this respect can easily be blurred. A natural landscape is very different from a landscape specifically optimized by human hands to be pleasant. The part of me concerned with wireheading starts to be concerned about the second. (My actual motivations don't, though -- if things have been arranged in what feels like good taste to me, I enjoy it. Highly optimized landscapes such as malls and theme parks rarely feel like they're in "good taste" however.)
Casual sex isn't desirable to me. The part of me which is concerned with wireheading-like things says that this is because it's not connected to a wider web of meaning. This might be my actual reason. (I prefer a prolonged relationship -- "just sex" sounds like a painful thing emotionally.)
Music is good. The part of me concerned with wireheading says it isn't -- it's just an empty superstimulus.
Overall, I'd say the conclusion of this chain of thought is that to count things as actually-good rather than merely seemingly-good I'd like them to be connected to a wider web of meaning, rather than isolated. "fake" really means "shallow" (surface-level, lacking deeper connections). Taking things out of devil's-advocate mode, this does not seem entirely damning to fiction. It suggests that fiction can in fact be valuable, to the degree that its meaning is interconnected with other things.
It also bears noticing that this argument applies rather heavily to me and my preferences, not necessarily to other people.
So fiction was just an example of a more general proposition: enjoyment is bad. Sensual pleasure of any sort is bad. These things are a snare and a delusion.
What are they a distraction from, that should be pursued instead?
There's a big difference between saying wireheading and superstimulus are bad and saying enjoyment is bad. The way I'm framing it, that's roughly like the difference between saying that counterfeit money is bad and saying money is bad.
In the view that you're devils-advocating, fiction is fake, admiring nature is silly, casual sex is meaningless, and music is empty. If these are counterfeits, what are they counterfeits of?
And what's the thing about ice-cream?
I'd better climb out of the devil's advocate position before I dig myself too deep a hole.
gjm's reply is perfect in terms of describing the position being outlined.
I really do want to make a distinction between pleasurable things and terminally-valuable things, though. At least I think I do.
The way you're reacting makes me think that you don't -- that you find it puzzling that I want to differentiate between superstimulus and actually good things at all, regardless of questions about fiction and such.
I think it would be unfortunate if future civilizations decided maximum wireheading was the greatest ethical good.
I think it would also be unfortunate (but less so) if future civilizations decided that finely crafted full sensory experiences, akin to movies, were the ultimate good.
I furthermore think it would be unfortunate (but significantly less so) if future civilizations decided that finely crafted interactive experiences, akin to 1-player games with only non-sentient NPCs, were the ultimate good.
(With significantly more uncertainty, I think it would be much worse if all of the movies or interactive experiences were identical. The image of billions or more identical clones (human-optimal in whatever sense) watching identical recordings of a single extremely well-crafted thousand-year movie does not appeal very much to me. I'm not sure it's more preferable than a single human experiencing this best-of-all-possible-movies. Similarly, but less so, for interactive experiences.)
The ideal case seems much more like a massively multiplayer one, despite the fact that players will tend to clash with one another and it's much harder to optimize properly (will have to be worse in other respects as a result).
Applying the intuitions from these rather distant scenarios to more everyday matters, the enjoyment from ice cream does fall rather far toward the beginning of the spectrum I've just outlined. It seems rather like a small dose of wireheading (except when enjoyed socially).
(I find it quite amusing that I'm getting push-back on the ice cream thing.)
Fiction is counterfeit learning or counterfeit human relationships. Admiring nature is a side-effect of preferences that evolved to help us find good places to live or stay. Casual sex is a counterfeit of not-so-casual sex, which helps to make families (in at least two ways). Music is counterfeit pattern-spotting. Ice cream is basically sugar and fat, neither of which is very good for your health when consumed in large quantities.
Something along those lines, anyway.
(Full disclosure: I read fiction, admire nature, have not-so-casual sex because I'm married, spend an appreciable fraction of my life on music, and make my own ice cream.)
That looks to me like a very uncharitable reading of (or extrapolation from) what abramdemski has said. I take it to be, rather: enjoyment is (to abramdemski, at least) less valuable than we are apt to think it and enjoyment of things that harm us is a snare and a delusion; the existence of superstimuli (and especially the fact that superstimuli can be engineered by others who don't necessarily have our best interests in view) makes it more dangerous.
The ice cream example aside, I think it would be wrong to say fiction is something that harms us even as we enjoy it, except in the sense of opportunity costs, which is what abramdemski seems to be arguing. Fiction can use superstimuli to manipulate people, but so can lots of other things.
What is it about ice-cream? I had one as recently as a month ago, and, well, what?
If you did away with all those things and everything like them, what would be left? It feels like so little would be left you should be able to give a pretty complete list.
A popular piece of fiction that many people enjoy creates bonds and shared experiences and ideas. It's connected to a lot of things many people think and do, and it helps give meaning to their lives. I feel I could replace "fiction" with "culture" here and the argument would be much the same.
What is it you want human activities to be connected to? Fiction is very well connected to other human activities.
You seem to be saying sex in a prolonged relationship is a good thing. That's sex that builds on and reinforces the relationship. But shared experience of fiction can also build on and reinforce a relationship. People watch movies together, they talk about books they've read, they share their opinions and bond over shared opinions. What's the difference between sex and fiction as relationship tools? What's special about a relationship in the first place that makes it "less like wireheading" and "more connected" (to what)?
Yes, my actual position on this is much closer to "fiction is bad if it's not a social activity" rather than "fiction is bad".
This does not work as an argument against the extremist position, however. Continuing the devil's-advocate line of thought, I say: if fiction is just good as a social activity because I have friends who like fiction, isn't that just me being the elephant tied with a chain to a non-optimal social situation?
I am not saying you're wrong -- in fact I think you are right. What concerns me is that I think we should be striving for something better, not justifying the status quo. That's why I think this is a useful exercise. I'm very skeptical that our current behavior here would just happen to be anywhere near the best we can do. In fact I think fiction is very often more like the ice cream. Our motives for binge-watching an entire series or such are more often self-defeating than good, in any plausible interpretation of the word "good".
As for the final question -- what makes fiction feel more like wireheading than a relationship -- my answer is that there's a real person as opposed to the projected image of a non-real person. The difference is somewhat analogous to the difference between visiting your bank's website and seeing a large sum of money in the account, and visiting a fake banking website whose sole purpose is to simulate the experience of seeing a large sum of money in your account. The actual relationship with an actual person is good in that it not only creates a sequence of feelings and impressions in the brains of both people, but furthermore creates a richly interconnected dance between the two people which is lacking in fiction.
That's begging the question: what's better and why?
Many good things are best consumed in moderation. Very few things have no upper limit on 'more is better'. The very name binge-watching labels it as an injurious behavior akin to binge drinking. That doesn't say anything much about fiction generally. (And I think the same applies to ice cream.)
That seems like a fully general argument against any solitary activity, and even some activities that are done together (like watching movies) that aren't about complex interpersonal interaction. (Well, it's not an argument, it's a value statement.) You're free of course to have such values in your own life, but why do you recommend them to others? Plenty of people, like me, enjoy some time apart from others. And there is no social activity which produces the experience of consuming fiction, which I value.
"Fiction considered harmful" sounds like it should mean more than "I, the poster, enjoy / prefer other things to fiction". There are good arguments that we wouldn't want everyone to wirehead. But I don't see a good argument why we wouldn't want everyone to consume some fiction, as indeed most people do.
It's not begging that particular question... the implicit assumption is that the current state of affairs is far from optimal, not that my particular definition of optimal is correct. In fact part of the point is to explore what values we might hold (and still hold after reflection on those values) that would value fiction. I feel this is a valuable exercise largely because when I do reflect on it, arguments to the effect that fiction is something I reflectively value are rather difficult to make. If I were to discover that I don't in fact value fiction on reflection, that would be good news: easy life improvement by no longer acting as if I value fiction.
True. The point I was trying to make is that when I talk to people about this, they tend to give rather high-minded justifications of the value of fiction (usually as a means to other ends, not as an end in itself). While these high-minded justifications may in fact be correct, they seem very different from the motivation which actually causes people to consume fiction. The result of this difference is that the kind of fiction which is readily available on the market is more often "potato chip fiction" as opposed to "baked potato fiction": still food, but awfully greasy.
This point may not be that relevant to the overall set of questions.
I feel like this remark ignores the part before the text you quote ("there's an actual person") which is very much not a fully general argument, but rather an argument against solitary activities which are misleading superstimulus telling system 1 it's achieving things it's not.
There's also a big difference between creative activities (spending solitary time writing a book, say) and consumptive activities. It's certainly possible to spend alone time without the activity being "isolated" in the sense that I mean.
It's also possible for someone to be entirely creative and not engage in fiction at all while still being "isolated" in the sense I mean. When I imagine a version of me or someone else toiling away at something they love with plans to totally burn it without showing anyone before they die, I feel like something is lost in this sense; not interconnected with the web of life.
Perhaps that's just a bad intuition I have about my values, and/or perhaps it's not a thing many other people value. I don't know.
Beyond that, I think the difference I'm pointing to is that fiction feels disproportionately like a good thing (because it's designed to). If there is any distinction to be made between what feels immediately valuable and what I'd find valuable on reflection, fiction will tend to optimize for the first. (This is also true of other types of information I might consume, but fiction has particularly large freedom to optimize these differences.)
I think people (not just me) generally develop a kind of "memetic immune system" defense against the types of superstimulus which are present in abundance in their culture. We have a much higher bar for humor on television than in real life, because funny things happen with much greater frequency on television. No matter how much more attractive the people on television are, we adjust our expectations (somewhat, at least) and are able to find the people we meet in person relatively attractive. And so on. This generally gives us the ability to not waste too much time on the superstimulus. (It also means that they may not have similar defenses against superstimulus available in other places; an example is people over-doing recreational drugs when they visit places where more things are legal. The native population of those places is not prone to the same excess.)
In the absence of a suitable definition of (human) value I (or abramdemski) can only appeal to you intuition of this. You apparently agree that there is a range between the extremes and that the example fall in between. But you seem to refuse to compare these. That could imply that it a) measuring it to sufficient precision is impossible/impractical or b) might not be possible to find a continuous range where all activities can be compared (at least in principle) and instead a more abstract like a lattice has to be used. I could agree to that.
No, I don't. I'm asking for your definition of the extremes and the measure of what falls where on the range, because I don't share whatever intuition you're using.
ETA: Wireheading is a single thing with arguments for and against it. Part of your (or OP's) argument seems to be "wireheading is bad so let's avoid things that are sufficiently like it". In that case I'm asking, what makes fiction more like wireheading than those other things?
Actually, I think it's not a continuum from "productive" to "wireheading". I think there's a continuum from "immediate end-in-itself" to "delayed gratification" to "doing what's best for the future societies" (let's call this the fun vs productive continuum), and then there's a continuum in what you consider to be an end-in-itself which ranges from wireheading (max-happiness utilitarianism; good/bad is exclusively about good/bad mental state) to preference utilitarianism (what's good is defined in terms of minds, but involves stuff outside of minds) to something like "the divine aesthetic" (maximize some fully external-to-minds notion of beauty in the universe, so things like dead planets with beautiful clouds count as positive, even if they are never observed by a conscious entity).
This second spectrum is the one I'm pointing at when I say fiction is further toward wireheading.