Rational feelings: a crucial disambiguation
Ever wonder something like, “I know it’s bad for me that I lost my job, but I actually feel happy about it… is that rational?”
What could a question like that mean? There is a divisive ambiguity here that really messes people up. A feeling as an experience is neither rational nor irrational. It’s like asking how ethical a shade of purple is. The point is that a feeling must be framed as a behavior or a statement to ask whether it is rational, and which one matters heaps and loads to the answer.
If you think of the happiness as a behavior, something that you’re doing, then the question is secretly asking about instrumental rationality: whether you’re applying your beliefs correctly to attain your values. In our opening example, the question becomes “Does feeling happy serve my values?”, or simply “Do I value feeling happy?”. If you’re almost anyone, the answer is probably “yes”.
If you think of the happiness as a statement or instruction that says “Your values are being served”, which can be true/false and justified/unjustified, then the question is really about epistemic rationality, and asks: “Am I justified to believe my values are being served?”. If “it’s bad for me” means “no”, then “no”.
Because of this ambiguity, although it can make sense to say “I’m happy” to indicate “my values are being served”, I propose that in the interest of epistemic hygiene it’s worth being more specific. Conflating feelings-as-behaviors with feelings-as-statements inflicts a great deal of pondering and confusion about whether feelings are rational (also precipitated by Hollywood), and to make matters worse, each of these similes has only limited validity:
1) A feeling is a behavior only insofar as you have control over it. This is something perhaps to strive for, but which certainly varies in feasibility. If someone carefully injects you with dopamine at a funeral, you might feel happy. That doesn’t mean you’ve made an instrumentally irrational choice. 2) A feeling is a statement only insofar as it has a given interpretation. In my opening example, the happiness might rather signify “There’s nothing you can do about this so you can relax and move on.” This might be clarified on reflection, and if the statement isn’t a rational one, you might consider retraining your feeling to offer more rational suggestions. But then, as with any statement, your epistemic rationality hinges on whether you believe it, not on how you feel. On the other hand, it is conceivable that even after introspection reveals the mechanism of a feeling, it still does not present itself as a statement. So sometimes the question “Is this feeling rational?” just isn’t applicable, but greater self control/awareness makes your feelings more often like behaviors/statements to be assessed as “rational.” For the ticklement of your visual cortex, the following table displays 15 scenario types (all of which can really happen), and what the scenario means for your instrumental/epistemic rationality (which is sometimes nothing: “—”). If you like, try thinking of an example scenario for the plausibility of each cell: Your instrumental/epistemic rationality with respect to a feeling:|
|
You haven’t interpreted the feeling as a statement |
You’ve interpreted the feeling as a statement | ||||
|
The statement is justified |
The statement is unjustified | |||||
|
You believe it |
You don’t believe it |
You believe it |
You don’t believe it | |||
|
You can’t (yet) control the feeling |
--/yes |
--/no |
--/no |
--/yes | ||
|
You can control the feeling |
You like (value) the feeling |
yes/-- |
yes/yes |
yes/no |
yes/no |
yes/yes |
|
You dislike (devalue) the feeling |
no/-- |
no/yes |
no/no |
no/no |
no/yes |
What’s the point? Understanding your feelings means you can put them to better use. For one thing, you don’t have to turn off a good feeling just because it makes a bad suggestion, as long as you can ignore the suggestion: if you lose your job, go ahead and be happy about it, just make sure you behave appropriately and keep looking for a new one. Likewise, you’re not obliged to feel a “smart” feeling that you don’t like, as long as you’re smart enough to remember what good advice it might have given you: if you’re worried about failing your exams, say “Thanks Worry, good idea, I’ll go study. Okay, I’m studying now, you’ve done your part, you can leave me alone for a while!” Don’t forget, in addition to communicating with the unconscious about epistemic issues, your feelings can be used for all sorts of other things, like energetic motivation, health benefits, and watching Avatar. And failing just one of these purposes doesn’t entirely preclude the others, as long as you can keep them effectively separate… feeling like blue people are real can be highly advisable at times.
Up-voted, with the following caveat...
Feelings are a communication of information. Ex: Feeling happy about losing your job may reflect 1) your knowledge about the stress the job actually held for you, 2) your desire for change (growth, creativity, etc.)...
Feelings can be probed with thoughts in order to ascertain their origin/meaning. To the extent that this is true, I prefer to think of feelings as a condensed form of communication with the ‘rational’ mind. ‘Here’s some information you really shouldn’t ignore—SLAM! [Insert feeling.]’
Yes, perhaps this wasn’t clear when I said “feelings can be used for all sorts of things besides epistemic suggestions”… I’m editing now to say “in addition to” instead of “besides”. Thank-you!
Yes, perhaps this wasn’t clear when I said “feelings can be used for all sorts of things besides epistemic suggestions”… I’m editing now to say “in addition to” instead of “besides”. Thank-you!
Important point. One nitpick:
Preferences over feelings can be context-sensitive; there are some circumstances under which many people would prefer not to be happy (or at least honestly claim to, whether or not they’d agree on reflection).
Yes, valuing happiness context-dependently is more general and to be distinguished from fearing its “irrationality”. Interestingly, until people reflect to distinguish these cognitive phenomena, they might not yet be operating separately in their brainware, in which case how they reflect can affect the development of their values and hence the conclusion. If one has been reflecting (consciously or unconsciously) with the bias that “rationality” is the only considerable meta-value of feelings, then one is more likely to end up in the second category.
On the other hand, it can be useful to encourage smart feelings in order to motivate you to achieve your values. If you want to do this, then the second and fifth cells in the last row can be “yes/yes”.
Agreed. I think delegating computational load to your unconscious by designing “good instincts” is a great life strategy. For the purposes of the table, I decided it makes most sense if you factor this into whether you “like the feeling”, so I’d place these in the middle row instead.
I think good feelings ought to be grounded in reality—caused by perceiving a good state, or good future prospects, according to our values. Whatever stimuli can cause the good feelings without actually causing the good state in the world, I take as an article of faith to be less authentic, secure, lasting, and complete (sure, you say this multi-vitamin has everything I’ll need; but what about the vitamins you haven’t discovered yet?). On the other hand, I do enjoy pointless diversions (fiction, music, games), so I’m sympathetic to your view.
You suggest a permissive attitude toward feelings that are at odds with what we believe about the world and/or what we believe we value, because once we intellectually disclaim “this feeling is not grounded in reality”, we’re inoculated against any unwelcome interference in our decisions. I’m not comfortable with that. I’m sure my feelings do bias my thoughts.
Since I don’t have any evidence that I can control my feelings significantly without also deceiving myself about the state of the world, I have to live with them. That means, whenever I find them at odds with my beliefs, and see no reason to change my mind, I have to believe that most of my feelings are those of a slightly more stupid person, who I have yet to convince with my higher mind.
The question of how often “You can control the feeling” is most interesting to me. Actions, decisions, distractions, and time can all palliate an unpleasant feeling. But can I do it just by deciding “I’m going to feel this way about it instead”?
This is a misunderstanding of an important function of positive feelings, which is to motivate positive action, in anticipation of a result. Many outcomes can only be achieved by having the feeling in advance, that you can only expect “grounding” for afterwards.
In particular, if “times are tough”, the worst thing you can possibly do is feel bad about it, since that is creating a positive feedback loop of negative expectation.
Feelings are intended to prepare current behavior in expectation of future outcomes; if you feel bad, your brain thinks you need to prepare for a negative outcome, and vice versa. The catch is that (at least at human-scale intelligence), these expectations have a significant self-fulfilling aspect.
I’m curious about how “learned optimism” or similar ideas fit into aspiring rationality. I’m somewhat nervous that learned optimism seems like ignoring reality – for example, say you lost your job due to poor performance, but you attribute it to some external factor, either due to your natural reaction, or through applying learned optimism techniques. My understanding is that the literature suggests that such a mindset will generally lead to better life outcomes, but how does it fit in with aspiring rationality?
That is, can we employ learned optimism techniques to improve our chances of winning, while still working to improve our rationality?
If you expand “rationality” to mean “improving our chances of winning”, you’ll notice that you just asked “can we improve our chances of winning while working to improve our improving our chances of winning?”
To which I would say, “I should certainly hope so.” ;-)
That having been said, I’ve in the last few months decided that the direct concept of learned optimism is bunk. Or more precisely, it’s based on an error in reasoning.
Specifically, reversed stupidity is not necessarily intelligence, and reversed learned helplessness is not necessarily healthy confidence. Also, just because confident people produce certain patterns in the output of their thinking, it does not necessarily mean that mimicking those outputs will produce the same internal states.
Yes, Seligman has some research on those points. However, his training methods call for essentially rote repetition of thought disputation until something “clicks” for you; AFAICT, there is no way to predict just how much disputation you will have to do in order to actually change your inner experience. In effect, there is a faith-based exhortation that sufficient repetition will bring about changes.
In contrast, I’ve found that by focusing on removing learned helplessness instead, the click to positivity is natural. I (and others) automatically begin seeing the positives in an area of action that was previously restricted. And instead of having to continually have the negative thoughts arise in order to be disputed, getting rid of the helplessness simply switches off the negative thoughts in that context, from that point forward.
I contend, therefore, that “learned optimism” is a nonentity, as optimism and confidence are the natural state. Helplessness can be learned and unlearned, and must be unlearned in order to experience actual optimism or confidence. Practicing optimism skills may eventually lead you to stumble on an unlearning of some bits of your helplessness, or not. But directing your attention to identifying and removing the specific bits of helplessness pays off in a much more predictable and permanent fashion.
(tl;dr: optimism good, learned optimism conceptually flawed, eliminate learned helplessness ftw.)
My “good future prospects” definitely covers your “anticipation of a result”.
Good point; I just was using a slightly different meaning of “anticipation” than the one implied by the context of, “grounded in reality—caused by perceiving … good future prospects”.
In that sense of “anticipation”, the mental activity is passive—perceiving that something good is already going to happen, by projecting from current circumstances.
In the sense I meant, the activity is active—you intentionally create a sense of anticipation for something that has no “grounding in reality” and is not “caused by perceiving”. (Indeed, in the ideal case, it is entirely ungrounded in current perceptual reality, so as to be based solely upon one’s intentions.)
I like your responses in this thread. I’m going to comment on the one thing that bothered me:
That’s appealing, but a little facile.
If times are tough, perhaps a negative enough feeling about them will encourage me to pursue drastic change or effort. Naturally, I should feel good in anticipation of my work beginning to pay off.
As soon as you’ve felt bad long enough to notice there’s a problem, the emotion ceases to provide you with any new information.
And the more complex or stressful the course of action required to change the circumstances, the more counterproductive a continuing negative emotion will likely be.
Essentially, our negative emotion systems were evolved to handle short-term emergencies (e.g. being stalked or chased by a predator) and seasonal-or-longer resource pressures like food shortages. These kinds of situations either require immediate but short term action, or else a passive strategy of energy conservation, resource hoarding, and risk-avoidance.
So, if the thing you’re feeling bad about isn’t an immediate emergency, or won’t be helped by passive inaction, a negative emotion will hurt more than it helps. (And most people’s goals, most of the time, aren’t helped by passive inaction!)
The “but maybe negative feelings will motivate me” idea is just plain wrong, in precisely the same way that “but death is a natural part of life” is wrong. It’s socially-cached sour grapes.
People think negative emotion makes sense because a person who holds power over you can get you to do things with it. But there’s still a high cost to your effectiveness in that case (narrowed focus, lower creativity, greater stress, etc.), and if you try to apply it to yourself, it has some rather nasty failure modes.
(Worse, sometimes, these negative motivations can be stealthy; i.e., you don’t really realize that the “carrot” you’re dangling for yourself is just a stick in disguise. Hence my saying that “what pushes you forward, holds you back”—i.e. anything you use to try to get yourself to do something, is a disguised “stick”. Positive motivation doesn’t feel like you’re trying to motivate yourself at all—you just are motivated.)
I absolutely agree, provided that you also roughly apprehend the problem’s relative importance to your happiness. Not all problems can be solved at once.
Plausible, but be careful—Caveman stories are ripe for mockery (see “Paleo” diet/fitness). But to put it in Caveman Bob terms—as I mentioned before, mounting unhappiness with a situation may be the only thing enabling a risky decision, like looking for food in a new area, or attacking a rival.
Negative feelings about my long-term situation do motivate me—not directly (like they would if they were in response to some immediately addressable threat), but in giving me a reason to plan/think about changing things. I readily admit that I wish to experience fewer negative feelings. So for me there are no sour grapes. I’m 100% behind reducing pointless discomfort. Maybe what you say is so for most people; I agree that such sentiments as “maybe we need death so we can appreciate life” are just as you say, suspiciously convenient and without sound basis.
Sure—but “plan/think about” changing things is not the same thing as actually taking positive action on a consistent basis. Otherwise, we’d all be skinny rich supermodels, just by worrying about our weight, money, or looks. ;-)
That’s my main point here. Negative emotions don’t provide much utility for the action part, even if they do promote thinking. Seth Roberts suggests that one function of negative emotions is a relative increase in thinking and observation, and it does indeed make sense. You just have to remember to switch to positive anticipation as soon as possible, by deciding what you want to change the situation to.
Notice that my hypothesis is actually a negative one—I’m saying that there’s no evolutionary basis for a long-term action motivation system based on negative emotions. Can you point to an evolutionary pressure that would create such a thing?
Neither of which is a course of action that requires sustained motivation in the face of opposition. Either your attack succeeds or fails, and the only reason you’re looking for food elsewhere is that there’s none here. IOW, nothing needing willpower to continue at.
Again, the thrust of my argument here is that negative emotions are not a viable anti-akrasia tool; they’re vastly more likely to be a source of akrasia than a cure for it, except in cases where you can distort the problem into the form of an urgent/emergency situation, or an ongoing resource shortage.
(Some people do one or both of these things all the time, but it’s not good for their physical or mental health, and there are many problems that can’t be abused into one of those shapes.)
Sometimes. Sometimes it’s more like “huh, now that I think about it, that doesn’t make a lick of sense”. Sometimes it’s “I detect that this emotion was arrived at by thus-and-so a process, which I can interrupt in this-and-such a place.” A bit more on this in the luminosity sequence (five posts of nine written).
Yes, I think it’s at least useful to make the attempt. The negative consequences if I turn out to be wrong seem insignificant—oh no, I tried to deceive myself about my ability to feel differently than I do!
As for making up stories about how my dumb emotional self arrived at an irrational belief, that sounds like fun, but I’m not sure how much I’d be confident that I had the right story. I’ve heard pjeby promote the idea (and claim that he can teach people to produce such stories by introspection). But I suppose if these stories work (get you to change your feelings) then there’s also little harm.
″ The negative consequences if I turn out to be wrong seem insignificant—oh no, I tried to deceive myself about my ability to feel differently than I do!”
Repression anyone? I think directly telling yourself, “I don’t feel that way, I feel this way!” can be extremely harmful, since you are ignoring important information in the original feeling. You are likely to express your original feelings in some less direct, more destructive, and of course less rational way if you do this. A stereotypical example is that of a man deciding that he should not feel angry that he did not get a promotion at work and then blowing up at his wife for not doing the dishes properly. Maybe there is nothing to actually be angry about, and screaming at his boss certainly wouldn’t accomplish anything, but ignoring the feeling as invalid is almost certain to end badly.
I think Alicorn is suggesting that if you attempt to understand why you have the feelings you do, and if these reasons don’t make sense, your feelings will likely change naturally without the need to artificially apply different ones.
I’m also not convinced that repression is real. I can’t imagine deceiving myself about how I feel in a moment (lying to others, yes, but what’s more obvious to me than how I feel?). I do believe people can self-deceive about the reason that they’re angry/sad/etc. and attribute that lingering emotion to innocent bystanders instead. Maybe that’s what you mean.
I think you’re underestimating human variation, and some people aren’t skillful at distinguishing their own emotions.
“In the moment” is an interesting point. The Kahneman video about the differences between experience and memory is relevant. I’d say the pattern of repression is feel some undesired emotion, dislike it (probably without naming is), distract oneself from feeling it and probably tighten the muscles which would be used to express it, and blur the memory in the process.
Also, how would you classify : “I’m not angry, it’s just that other people are such idiots.”?
Sounds like someone is both angry and a poor liar ;) I’d classify it as a defensive move by someone who was just accused of being an excessively angry person. I imagine the person is well aware of how angry they are.
Thanks for the video. I don’t agree that my calculation of “how much will I enjoy that?” which guides my decisions is based on the way I’d evaluate what I expect as if it had already happened and I were remembering it, but it’s a fascinating claim. I agree with the rest of Kahneman’s talk. I think he says that reflective happiness is determined by goals and wealth, but experiential happiness is mostly achieved by spending time with people you like (I’m not sure, because he appeared to mislabel one or the other, likely due to speaking error).
I concede that when people are motivated to hide their feelings, the way they do it can include some actual self-deception as well.
That’s plausible: “feeling is because of dumb reason X” → feeling retreats → “I must have been right.” I just don’t trust it entirely.
Actually, you point up a big issue in what I have to teach people: how not to make up stories, but rather to simply observe the sounds and images that flash through their heads in response to a mental inquiry. Invariably, these bits of information are not particularly logical or verbally sophsticated, unlike the information that comes from the “making up stories” bit of your brain.
Instead, the truthful information is usually childish, paradoxical, or downright ridiculous. A common reaction upon seeing these things is to go, “Are you kidding me? That’s what this is about? I’ve spent how many years doing/feeling this stupid thing because of that?”
(Ideally, that’s all there is to it—the feeling stops there. But sometimes, you realize that you still believe the stupid thing anyway, even though you know how stupid it is. That then requires a bit more work to construct alternatives, the way Eliezer eliminated his fear of the lurking serial killer.)
Anyway, as I tell people, if you’re not at least a little surprised, confused, or ashamed by what you discover when you do RMI, you’re almost certainly making it up yourself—using the apologist instead of the revolutionary, so to speak. Shut up and ask the question again, then watch, wait, and listen for the answer.