Person A is an Olympic-level athlete. He can perform amazing physical feats. The limits of his ability can be scored against some sort of metric (lap time, distance jumped, etc.), and since he’s working to improve on them, his own personal limits are known to him.
Person B is of average physical fitness.
Person C has a moderate chronic illness. He struggles to perform basic physical feats, but can function independently with some difficulty.
If all three of these people were secretly transplanted into an environment with lower oxygen levels and began to experience mild hypoxia, it seems that Persons A and C would both be more sensitive to this change than Person B. Person A would notice it because he would no longer be able to perform outstanding physical feats to the level he’s accustomed to. Person C would notice it because he’d struggle to carry out basic activities.
[Edit for clarity: I’m not saying that Person B would never notice this, but that he would be less sensitive to it, because his performance is higher-variance and subject to less of a “state change”, and doesn’t have a fine, frequently-scrutinised boundary between what he can and can’t do.]
Alternatively:
Person D is a voracious infovore with high reading comprehension. She’s used to grappling with precise language.
Person E is an average-level reader.
Person F has some sort of reading-related disability.
It seems that Persons D and F will be more sensitive to badly-punctuated writing than Person E. For example, Person D might be able to parse a sentence in two or three plausible ways, while Person F might not be able to parse the sentence at all.
Both of these cases involve both ends of an ability distribution being more sensitive to degradation of the environment than central cases. Are there better examples? Is this a phenomenon we actually see in the real world? If so, does it have a name?
I think the basic difference is that people B and E just don’t care and are less likely to notice these things because they’re not interested in them.
Replace your person B with a person B’ who is also of average fitness but recently started a new fitness regime and has been busy quantifying himself. He will notice his mild hypoxia as soon as A.
A and C are much closer to the limits of their respective physical bodies than B. A—due to his high motivation (he is a maximizer, while B is a satisficer, if A didn’t have a motivation to maximize his results, he could do the same (or better) as B), C—due to his limited physical abilities. Therefore, they have lower tolerance for the worse environment. In other words, if we plot physical abilities on x axis and the expected/desired result (i.e.result that they have a motivation to achieve) on y axis, we would probably obtain a convex function whose graph is below the line y=x (which corresponds to pushing the physical limits of the body).
I’m really not seeing either of your examples, unfortunately. What’s stopping average fitness person from noticing their times aren’t as good in the same way olympian-level person would notice that their times aren’t as good? What’s stopping him from noticing his morning jog or whatever is tougher?
Why wouldn’t average reader have more difficulty parsing badly punctuated writing as well? Why wouldn’t they be able to parse it in different plausible ways too?
I’m just not seeing it.
Edit: To go into more detail:
Person A, B and C are secretly transplanted into a low oxygen universe.
Person A noticed their 200m backstroke time is consistently 4 seconds slower than usual.
Person B notices they have to take walking breaks more often than usual on their morning jog, and took 5 minutes extra to complete their usual route.
Person C has more difficulty doing basic tasks.
Person D, E and F are each given 10 badly punctuated sentences to read.
Person D finds they can parse 4 of them in different plausible ways.
Person E finds they can parse 2 of them in different plausible ways, and can’t parse 2 of them at all.
Person F can’t parse 4 of them at all.
I see these examples as being just as plausible if not more than yours.
I imagine person B as someone who doesn’t do formal exercise at all. If their capacity for exercise goes down, they might think vaguely that they’re a little sick or getting older, but they’ve never been concerned with the details of how much they can do, and it’s going to a good-sized change for them to notice it.
However, I’m not sure this pattern extends to reading—some smart people who read easily seem to have the metaphorical proof-reading/copy-editing gene.
Well then actual fitness level is basically irrelevant, and the ability for someone to notice the effects of the environment change is based mostly upon the factor of whether or not they do any excercise.
In the first example, the average-fitness person probably has a lot more variance and a lot less visibility on his physical performance than the Olympian. The Olympian presumably also has a selection of meta-skills surrounding his chosen discipline, and is capable of judging when he’s off his game or when he falls short of his own standards.
In the second example, well...
I don’t know if you’ve ever gotten into the timeless identi-discussion with someone who is literate, but refuses to learn how to adequately punctuate their sentences, because they don’t see the difference. A lot of people decry poor standards of reading comprehension, and my pet hypothesis is that many readers don’t actually parse and evaluate sentences, but just let their eyes suck in the words and have a good guess at what it’s supposed to mean.
I’ve never explicitly stated that hypothesis, but now I have, it seems like a case of attribute substitution: i.e. “Parsing this sentence is hard, so I’ll round it off to the nearest sentiment and assume that’s what it’s saying”.
If you’re interested in some actual research on that hypothesis, try Ferreira for a starting point. Any of the papers on her page with the phrase “good enough” in the title will be relevant.
A lot of people decry poor standards of reading comprehension, and my pet hypothesis is that many readers don’t actually parse and evaluate sentences, but just let their eyes suck in the words and have a good guess at what it’s supposed to mean.
If you read a legal contract it’s important to understand every word. In most cases it isn’t.
If you focus too much on details you can also lose context.
Years ago when I was a forum moderator I was reading forum posts in a way where I could decently reidentify a banned member that reregistered. On LW I now don’t seem to have that ability anymore to the same extend. My focus is elsewhere.
You trade different ways of reading against each other.
refuses to learn how to adequately punctuate their sentences, because they don’t see the difference
If someone wanted to check their map versus their territory on this one, wanted to determine if the problem was with their reading ability, the language itself, or the writing’s author. What advice would you give them?
(I see what you did there, if you did what I think you did. If you didn’t do it, please disregard this.)
I can see two cases you’re plausibly asking about here:
The case of not being able to make sense of a sentence and wanting to know whether it’s you failing to read something that is well-formed, the author failing to write something that is well-formed, or the language being able to support this content in a well-formed manner.
The case of reading a lot, and apparently understanding it, but not being sure whether you’re actually doing the necessary work to understand it, because the mechanisms of that work are inscrutable.
Which of these are you asking? Or are you asking something else?
I’m asking about the case of not being able to make sense of a sentence and wanting to know whether it’s you failing to read something that is well-formed, the author failing to write something that is well-formed, or the language being able to support this content in a well-formed manner.
To add context, I frequently encounter writing in legalese that I can’t break down into an if/else flow chart and the writing should be able to be broke down into a simple logical chain. If the author is using precise language to keep things as brief as possible and I’m unable to properly parse it, it means I have found a new blind spot. My old assumption was that the authors were constrained by word count and the language therefor settled with insufficient explination, or were oblivious to the ambiguous wording, which fits Hanlon’s Razor better.
Legal documentation has its own conventions which I am not remotely qualified to advise upon. In the event of uncertainty in legal documentation, it seems prudent to consult someone with a background in law.
(Law is an area I’m both quite interested in and very ignorant about, but I’ve yet to find a good on-ramp for learning about how to practically interact with legal institutions and processes. It seems like legal texts should be extremely unambiguous, because they are produced by institutions who are presumably aware of the problem of linguistic ambiguity, and otherwise how are legal documents supposed to do their job? Then I remember that the same argument can be made of technical documentation, and people who product that for a living can do a spectacularly poor job of it. Other fields presumably also have people who are bad at their jobs.)
To be clear, I’m using the broadest definition of legalese, in this case: design guides, building codes. Technical material recognized by the authority having jurisdiction that is not intended to be ambiguous, it is just complex. Stuff where the advice is consult your engineer instead of consult your lawyer.
I find that often this sort of writing—technical-ish, e.g. trying to describe a flowchart or a boolean circuit in casual text, as you see in law or documentation—has various sorts of ambiguities (e.g. issues with associativity and quantifiers) that would be obvious if you tried to transcribe it into code.
If I understand the vocabulary of a text but the syntax is unclear, I’ll assume it’s badly written. If it’s composed of pathologically malformed sentences, I’ll assume my English skills are better than the author’s. If the vocabulary is unusual, or it’s a subject I’m unfamiliar with, I’ll give more weight to my own ignorance being the problem.
It seems like legal texts should be extremely unambiguous, because they are produced by institutions who are presumably aware of the problem of linguistic ambiguity, and otherwise how are legal documents supposed to do their job?
That assumes that the linguistic ambiguity is bad for enough stakeholders.
A lobbyist who writes an amendment for a law doesn’t always want that the politicians who vote on the amendment understand what it does.
An ambiguous version might also be a compromise that allows both sides of a conflict to avoid losing.
A not quite nit-picking critique of this phenomenon is that it’s treating a complex cluster of abilities as a unitary one.
In some of the (non-Olympic!) distance races I’ve run, it’s seemed to me that I just couldn’t move my legs any faster than they were going. In others, I’ve felt great except for a side stitch that made me feel like I’d vomit if I pushed myself harder. And in still others, I couldn’t pull in enough air to make my muscles do what I wanted. In the latter case, I’d definitely notice the lower oxygen levels but in the former cases, maybe I wouldn’t.
So dial down my oxygen and ask to do a road race? Maybe I’ll notice, maybe I won’t. But ask me to do a decathlon, and some medley swimming, and a biathlon? I bet I’ll notice the low oxygen on at least some of those subtasks, whichever of them that require just the wrong mix of athletic abilities.
For the reading one, I can believe this if I’m doing some light pleasure reading and just trying to push plot into my brain as fast as possible. But if I’m reading math research papers, getting the words and symbols into my head is not the rate-limiting step. If there are some typos in the prose, or even in the results or proofs, it doesn’t make much of a difference. There might be some second-order effects—when I try to fill in details and an equation doesn’t balance, I can be less certain that the error is mine—but these are minor.
So maybe sharpen your claim down to unitary(-ish) abilities?
No idea. Factor analysis is the standard tool to see that some instrument (fancy work for ability) is not unitary. It’s worth learning about anyways, if it’s not in your toolbox.
It is already in my toolbox, but I’m not sure how it helps figure out if this phenomenon is present in the real world. It’s still not obvious to me that, if the phenomenon does exist, it would survive when reduced to a unitary ability. I can think of a couple of mechanisms by which it may be more prevalent in a multivariable scenario.
This may be a case of regression to the mean, with the thing which parameters regress being conscious and not caring about these particular parameters.
Person A is an Olympic-level athlete. He can perform amazing physical feats. The limits of his ability can be scored against some sort of metric (lap time, distance jumped, etc.), and since he’s working to improve on them, his own personal limits are known to him.
Person B is of average physical fitness.
Person C has a moderate chronic illness. He struggles to perform basic physical feats, but can function independently with some difficulty.
If all three of these people were secretly transplanted into an environment with lower oxygen levels and began to experience mild hypoxia, it seems that Persons A and C would both be more sensitive to this change than Person B. Person A would notice it because he would no longer be able to perform outstanding physical feats to the level he’s accustomed to. Person C would notice it because he’d struggle to carry out basic activities.
[Edit for clarity: I’m not saying that Person B would never notice this, but that he would be less sensitive to it, because his performance is higher-variance and subject to less of a “state change”, and doesn’t have a fine, frequently-scrutinised boundary between what he can and can’t do.]
Alternatively:
Person D is a voracious infovore with high reading comprehension. She’s used to grappling with precise language.
Person E is an average-level reader.
Person F has some sort of reading-related disability.
It seems that Persons D and F will be more sensitive to badly-punctuated writing than Person E. For example, Person D might be able to parse a sentence in two or three plausible ways, while Person F might not be able to parse the sentence at all.
Both of these cases involve both ends of an ability distribution being more sensitive to degradation of the environment than central cases. Are there better examples? Is this a phenomenon we actually see in the real world? If so, does it have a name?
I think the basic difference is that people B and E just don’t care and are less likely to notice these things because they’re not interested in them.
Replace your person B with a person B’ who is also of average fitness but recently started a new fitness regime and has been busy quantifying himself. He will notice his mild hypoxia as soon as A.
A and C are much closer to the limits of their respective physical bodies than B. A—due to his high motivation (he is a maximizer, while B is a satisficer, if A didn’t have a motivation to maximize his results, he could do the same (or better) as B), C—due to his limited physical abilities. Therefore, they have lower tolerance for the worse environment. In other words, if we plot physical abilities on x axis and the expected/desired result (i.e.result that they have a motivation to achieve) on y axis, we would probably obtain a convex function whose graph is below the line y=x (which corresponds to pushing the physical limits of the body).
I’m really not seeing either of your examples, unfortunately. What’s stopping average fitness person from noticing their times aren’t as good in the same way olympian-level person would notice that their times aren’t as good? What’s stopping him from noticing his morning jog or whatever is tougher?
Why wouldn’t average reader have more difficulty parsing badly punctuated writing as well? Why wouldn’t they be able to parse it in different plausible ways too?
I’m just not seeing it.
Edit: To go into more detail:
Person A, B and C are secretly transplanted into a low oxygen universe.
Person A noticed their 200m backstroke time is consistently 4 seconds slower than usual.
Person B notices they have to take walking breaks more often than usual on their morning jog, and took 5 minutes extra to complete their usual route.
Person C has more difficulty doing basic tasks.
Person D, E and F are each given 10 badly punctuated sentences to read.
Person D finds they can parse 4 of them in different plausible ways.
Person E finds they can parse 2 of them in different plausible ways, and can’t parse 2 of them at all.
Person F can’t parse 4 of them at all.
I see these examples as being just as plausible if not more than yours.
I imagine person B as someone who doesn’t do formal exercise at all. If their capacity for exercise goes down, they might think vaguely that they’re a little sick or getting older, but they’ve never been concerned with the details of how much they can do, and it’s going to a good-sized change for them to notice it.
However, I’m not sure this pattern extends to reading—some smart people who read easily seem to have the metaphorical proof-reading/copy-editing gene.
Well then actual fitness level is basically irrelevant, and the ability for someone to notice the effects of the environment change is based mostly upon the factor of whether or not they do any excercise.
In the first example, the average-fitness person probably has a lot more variance and a lot less visibility on his physical performance than the Olympian. The Olympian presumably also has a selection of meta-skills surrounding his chosen discipline, and is capable of judging when he’s off his game or when he falls short of his own standards.
In the second example, well...
I don’t know if you’ve ever gotten into the timeless identi-discussion with someone who is literate, but refuses to learn how to adequately punctuate their sentences, because they don’t see the difference. A lot of people decry poor standards of reading comprehension, and my pet hypothesis is that many readers don’t actually parse and evaluate sentences, but just let their eyes suck in the words and have a good guess at what it’s supposed to mean.
I’ve never explicitly stated that hypothesis, but now I have, it seems like a case of attribute substitution: i.e. “Parsing this sentence is hard, so I’ll round it off to the nearest sentiment and assume that’s what it’s saying”.
If you’re interested in some actual research on that hypothesis, try Ferreira for a starting point. Any of the papers on her page with the phrase “good enough” in the title will be relevant.
Thanks.
If you read a legal contract it’s important to understand every word. In most cases it isn’t. If you focus too much on details you can also lose context.
Years ago when I was a forum moderator I was reading forum posts in a way where I could decently reidentify a banned member that reregistered. On LW I now don’t seem to have that ability anymore to the same extend. My focus is elsewhere.
You trade different ways of reading against each other.
If someone wanted to check their map versus their territory on this one, wanted to determine if the problem was with their reading ability, the language itself, or the writing’s author. What advice would you give them?
(I see what you did there, if you did what I think you did. If you didn’t do it, please disregard this.)
I can see two cases you’re plausibly asking about here:
The case of not being able to make sense of a sentence and wanting to know whether it’s you failing to read something that is well-formed, the author failing to write something that is well-formed, or the language being able to support this content in a well-formed manner.
The case of reading a lot, and apparently understanding it, but not being sure whether you’re actually doing the necessary work to understand it, because the mechanisms of that work are inscrutable.
Which of these are you asking? Or are you asking something else?
I’m asking about the case of not being able to make sense of a sentence and wanting to know whether it’s you failing to read something that is well-formed, the author failing to write something that is well-formed, or the language being able to support this content in a well-formed manner.
To add context, I frequently encounter writing in legalese that I can’t break down into an if/else flow chart and the writing should be able to be broke down into a simple logical chain. If the author is using precise language to keep things as brief as possible and I’m unable to properly parse it, it means I have found a new blind spot. My old assumption was that the authors were constrained by word count and the language therefor settled with insufficient explination, or were oblivious to the ambiguous wording, which fits Hanlon’s Razor better.
Legal documentation has its own conventions which I am not remotely qualified to advise upon. In the event of uncertainty in legal documentation, it seems prudent to consult someone with a background in law.
(Law is an area I’m both quite interested in and very ignorant about, but I’ve yet to find a good on-ramp for learning about how to practically interact with legal institutions and processes. It seems like legal texts should be extremely unambiguous, because they are produced by institutions who are presumably aware of the problem of linguistic ambiguity, and otherwise how are legal documents supposed to do their job? Then I remember that the same argument can be made of technical documentation, and people who product that for a living can do a spectacularly poor job of it. Other fields presumably also have people who are bad at their jobs.)
To be clear, I’m using the broadest definition of legalese, in this case: design guides, building codes. Technical material recognized by the authority having jurisdiction that is not intended to be ambiguous, it is just complex. Stuff where the advice is consult your engineer instead of consult your lawyer.
I find that often this sort of writing—technical-ish, e.g. trying to describe a flowchart or a boolean circuit in casual text, as you see in law or documentation—has various sorts of ambiguities (e.g. issues with associativity and quantifiers) that would be obvious if you tried to transcribe it into code.
OK. Gotcha.
If I understand the vocabulary of a text but the syntax is unclear, I’ll assume it’s badly written. If it’s composed of pathologically malformed sentences, I’ll assume my English skills are better than the author’s. If the vocabulary is unusual, or it’s a subject I’m unfamiliar with, I’ll give more weight to my own ignorance being the problem.
That assumes that the linguistic ambiguity is bad for enough stakeholders.
A lobbyist who writes an amendment for a law doesn’t always want that the politicians who vote on the amendment understand what it does.
An ambiguous version might also be a compromise that allows both sides of a conflict to avoid losing.
A not quite nit-picking critique of this phenomenon is that it’s treating a complex cluster of abilities as a unitary one.
In some of the (non-Olympic!) distance races I’ve run, it’s seemed to me that I just couldn’t move my legs any faster than they were going. In others, I’ve felt great except for a side stitch that made me feel like I’d vomit if I pushed myself harder. And in still others, I couldn’t pull in enough air to make my muscles do what I wanted. In the latter case, I’d definitely notice the lower oxygen levels but in the former cases, maybe I wouldn’t.
So dial down my oxygen and ask to do a road race? Maybe I’ll notice, maybe I won’t. But ask me to do a decathlon, and some medley swimming, and a biathlon? I bet I’ll notice the low oxygen on at least some of those subtasks, whichever of them that require just the wrong mix of athletic abilities.
For the reading one, I can believe this if I’m doing some light pleasure reading and just trying to push plot into my brain as fast as possible. But if I’m reading math research papers, getting the words and symbols into my head is not the rate-limiting step. If there are some typos in the prose, or even in the results or proofs, it doesn’t make much of a difference. There might be some second-order effects—when I try to fill in details and an equation doesn’t balance, I can be less certain that the error is mine—but these are minor.
So maybe sharpen your claim down to unitary(-ish) abilities?
Do you have any suggestions for such unitary(ish) abilities?
No idea. Factor analysis is the standard tool to see that some instrument (fancy work for ability) is not unitary. It’s worth learning about anyways, if it’s not in your toolbox.
It is already in my toolbox, but I’m not sure how it helps figure out if this phenomenon is present in the real world. It’s still not obvious to me that, if the phenomenon does exist, it would survive when reduced to a unitary ability. I can think of a couple of mechanisms by which it may be more prevalent in a multivariable scenario.
In medical literature, it’s not uncommon to read about “U-shaped curves.”
Anyway, I am reminded of this:
http://www.youtube.com/watch?v=Y3lQSxNdr3c&t=0m24s
Although I suppose it’s pretty similar to the “Goldilocks Principle.”
This may be a case of regression to the mean, with the thing which parameters regress being conscious and not caring about these particular parameters.
The actual key difference in the second example is that people D and F are using their system II, whereas person E is relying on his system I.