It seems to me that "permanent/temporary" is an instance of "general/specific".
As explained in another reply here, that occurred to me but I resist summarizing it so. Although it seems to be trivially true, it may turn out that if you collapse the categories in your mind you lose the benefits of hacking your mind when you do that.
Also, collapsing things to the most "general" observation may be a psychological turing tar pit, a place where "everything is possible but nothing of interest is easy." If you only think of the most general aphorism, it may require a lot of work to apply it to various situations. The "specific" rules (personal/impersonal, permanent/temporary, general/specific) may require three times the storage but be very easy and fast to apply.
I don't know, which is why I resist trying to editorialize. If I were a psychologist, I would take a conjecture like that and test it. Which is often the difference between pundits and scientists. A pundit wonders if such-and-such is the case and writes an essay. A scientist wonders if such-and-such is the case and then sets about trying to devise a means of testing the conjecture.
I don't know if being pessimistic abut my code would make it better. I do know that tending towards pessimism of the sort described by Dr. Seligman has made me unhappy.
Thanks for your reply. I agree with you that speaking abstracting "permanent/temporary" is an instance of "general/specific". To be precise: temporary is an instance of specific, over time. Categorizing an explanation as temporary means that it is contained within a specific (and short) period of time. (I also agree that concrete categories can often be helpful in practice even if they aren't strictly necessary in theory.) That's what I think you meant in your comment above.
But that isn't what I meant. My confusion is how to apply it - how to actually categorize concrete explanations in practice? The example again "gems are always a pain" seems to be generalizing over gems, and not over time. Therefore, it would be a better example of general rather than permanent. A better example might be "this gem/Windows/linux/PC never works! It always messes up, every single time I use it." My difficulty is that (it seems to me) that every example can be recast in either form, because different events in a person's life occur at different times - to generalize over events is to generalize over time; and to generalize over time is to generalize over different events.
It occurs to me that an explanation might be neutral along some of these dimensions - an explanation might not be explicitly permanent, nor explicitly temporary. Or one might have a default, and say "if it isn't explicitly temporary, then it's permanent". eg, "they're just jealous because I'm smart" doesn't mention time, so by default it is permanent. Or, there could be a continuum along a dimension, corresponding to the size of the set generalized over: only this one instance---similar events---absolutely everything (and similar for the permanence/temporary dimension).
It might seem that I'm picking on your example, as not an ideal one to illustrate permanent/temporary (which was its purpose). I am. But I make the same kind of error (assuming it is one). And it seems to me that categorizing something as more strongly along the general/specific dimension, or more strongly along the permanent/temporary dimension seems arbitrary, and one can do it just as one feels - there's no principle behind it. Perhaps it is still useful, even without a principle behind it, but I'm uncomfortable with this.
Summary: I don't know how to classify an explanation as "permanent/temporary" or "general/specific" in practice. [answered in replies]
Or maybe I just too much :-). But it was important to me to lay this out clearly. [I haven't yet read your other reply that you mention - now read]
"Gems" is general about gems, it suggests that all gems have this problem. And "always" is permanent about time. The statement also implies impersonal: It suggests this is true for everyone, as opposed to "I'm always flocking gems up on my projects."
I chose that to illustrate just one of the three axes, but to my ear it sounds like it is saying impersonal, general, permanent.
So I agree with what appears to be the general thrust of your argument which is that an explanation might be neutral on one axes, or imply something about one axis, or make statements about two or more axes.
I suppose this is why Dr. Seligman makes tests with many, many questions. You need to aggregate a lot of explanations from one person to get a picture of their underlying attitude toward positive and negative events.
I have been troubled by cognitive therapy techniques, because they are implicitly intended to be used in the way that Seligman does. But they purport to be truthful, when they are intended to be used in a biased way. If you use them in a non-biased way (for example, to undermine good feelings, not just undermine bad feeling), they'll make you feel bad.
That's what I like about the two levels of this definition of optimism: it comes right out and says that it is biased, as a way of containing bad things, and expanding good things. The categories at the second level (personal, specific, temporary etc) appear to be just the same as cognitive therapy techniques.
Thanks, discussing it has helped a lot. I think my confusion was in thinking the axes are exclusive, but of course an explanation can be general and permanent.
Can it be general and temporary? Everyone is driving like madmen today!
Or specific and permanent? Damn it, my car is always breaking down!
Yep. The specificity depends on the framing ("everyone is driving" generalizes over individual drivers, but is specific to the set of drivers - it excludes people in offices, at home, at sea, etc). But that's a different confusion :-) and maybe the explanation includes the frame (and thus fully defines the specificity with respect to it), and framing is an important tool for changing one's interpretation from pessimistic to optimistic.
As explained in another reply here, that occurred to me but I resist summarizing it so. Although it seems to be trivially true, it may turn out that if you collapse the categories in your mind you lose the benefits of hacking your mind when you do that.
Also, collapsing things to the most "general" observation may be a psychological turing tar pit, a place where "everything is possible but nothing of interest is easy." If you only think of the most general aphorism, it may require a lot of work to apply it to various situations. The "specific" rules (personal/impersonal, permanent/temporary, general/specific) may require three times the storage but be very easy and fast to apply.
I don't know, which is why I resist trying to editorialize. If I were a psychologist, I would take a conjecture like that and test it. Which is often the difference between pundits and scientists. A pundit wonders if such-and-such is the case and writes an essay. A scientist wonders if such-and-such is the case and then sets about trying to devise a means of testing the conjecture.
I don't know if being pessimistic abut my code would make it better. I do know that tending towards pessimism of the sort described by Dr. Seligman has made me unhappy.