gwern comments on To like, or not to like? - Less Wrong

2 Post author: PhilGoetz 14 November 2013 02:26AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (85)

You are viewing a single comment's thread. Show more comments above.

Comment author: PhilGoetz 14 November 2013 04:27:21PM *  2 points [-]

All this doesn't address my key observation, which is that the prior against one of the first one-hundred professional English writers turning out to be the best is simply incredible. Your list of other top-rated artists in other fields only reinforces this point. What are the odds that the best artist or practitioner in every field happened, by chance, to be one of the first 0.1% in that field? Either there is a strong bias to overrate the early practitioners, or the human race has been devolving rapidly for hundreds of years.

If the claims people made were along the lines of "X was the most influential in his field", we could expect this. But I often hear it stated as absolute ability.

Comment author: gwern 14 November 2013 04:51:38PM *  9 points [-]

All this doesn't address my key observation, which is that the prior against one of the first one-hundred professional English writers turning out to be the best is simply incredible.

I did address that, but obviously in a way you didn't understand. Let me try again: that is not an observation, that's a result of an unmotivated and unjustified model you postulated which leads to a result which you already had as a bottom-line. Your reference classes are post hoc cherrypicked to reach your desired conclusion, your data is minimal (see my point about your bizarre interpretation of the Paris interview criticisms and the comparison to other fields), and if your strategy was applied to any other field, give equally absurd results that noone before the 18th century should have been the greatest in anything because the human population has grown so much since then.

Comment author: PhilGoetz 14 November 2013 06:19:51PM *  -2 points [-]

if your strategy was applied to any other field, give equally absurd results that noone before the 18th century should have been the greatest in anything because the human population has grown so much since then.

If you apply my strategy to any other field, the numbers give the result that noone before the 18th century should have been the greatest in anything because the human population has grown so much since then, yes. When you do the numbers and they give you a definitive answer, you don't dismiss it as "absurd" because you don't like it (or because you have a bug up your ass about the person who ran the numbers).

My separation of classes chronologically is, like all good models, inspired by observation. In this case, the observation that a statistically-impossible number of the people considered "best in field" came very early in those fields, even in fields like literature where coming early is a disadvantage rather than an advantage as regards the quality or contemporary opinion of your work.

Why are you always especially rude to me?

Comment author: gwern 14 November 2013 06:34:39PM *  8 points [-]

When you do the numbers and they give you an answer, you don't dismiss it as "absurd".

One man's modus ponens is another man's modus tollens. You have amply explained why you reached the conclusion you did based on idiosyncratic personal preferences and constructed an unconvincing model to try to justify it; there are no 'numbers' here, there is only absurd reference class tennis. ('Stratford-on-Avon had 0.0001% of the medieval English population; the odds against the greatest English writer coming from Stratford-on-Avon is astronomically unlikely!') I am perfectly happy saying that the result refute the pseudo-premises - not that you gave a precise model in the first place: I will ask you again, what is the right amount of criticism for Shakespeare that would satisfy you that he really was the greatest writer ever?

My separation of classes chronologically is, like all good models, inspired by observation. In this case, the observation that a statistically-impossible number of the people considered "best in field" came very early in those fields, even in fields like literature where coming early is a disadvantage rather than an advantage as regards the quality or contemporary opinion of your work.

The damning point here is that you are willing to bite the bullet and say it applies to sciences as well, where we would, contrariwise, naturally expect the earlier a scientist to live, the easier it is to make incredible discoveries and pick up low-hanging fruit. Only an early scientists has a hope of discovering, say, gravity. Or an early mathematician something like calculus. You have to live as early as Parmenides if you want to discover something basic and extremely important like 'the moon is illuminated by the sun'.

Comment author: lmm 15 November 2013 01:20:25AM 3 points [-]

In the cases where we have objective measures (like memorization contests) we see records being broken all the time (which is as we'd expect). A lot of this can be attributed to improved general intelligence, but we'd expect that to be correlated with creative skill too. Are there any measurable world records from the Elizabethan era that still stand?

The damning point here is that you are willing to bite the bullet and say it applies to sciences as well, where we would, contrariwise, naturally expect the earlier a scientist to live, the easier it is to make incredible discoveries and pick up low-hanging fruit. Only an early scientists has a hope of discovering, say, gravity. Or an early mathematician something like calculus. You have to live as early as Parmenides if you want to discover something basic and extremely important like 'the moon is illuminated by the sun'.

You're measuring something here, but I don't think it's likeability. Newton may have been more historically important than Einstein, but no-one would prefer the former's theory of gravity to the latter's. If Shakespeare got pretty close to the perfect tragedy, but there was a slight refinement of the form from the 19th century that's better (if less significant), surely people would prefer to watch that, and count themselves fans of that author.

Comment author: gwern 17 November 2013 03:46:55AM 1 point [-]

Are there any measurable world records from the Elizabethan era that still stand?

I'm not sure what sort of world record you would have in mind, and given the parlous state of science at the time, what world records would you trust? If, for example, I exhibited a Chinaman from the Ming who lived for 231 years, which is surely a world record, you would rightly reject this by saying 'it is much more likely that this world record is inaccurate than he really did live to 231, given how notoriously bad records were at the time, the cultural value set on being the oldest man in the world, etc'.

If Shakespeare got pretty close to the perfect tragedy, but there was a slight refinement of the form from the 19th century that's better (if less significant), surely people would prefer to watch that, and count themselves fans of that author.

If Shakespeare helped define what the perfect tragedy was, and all later tragedies felt the 'anxiety of influence', this isn't so clearcut. See my other comment.

Comment author: Pfft 15 November 2013 05:32:07AM *  0 points [-]

the earlier a scientist [lives], the easier it is to make incredible discoveries and pick up low-hanging fruit.

You are ignoring the distinction PhilGoetz made in the grandparent comment:

If the claims people made were along the lines of "X was the most influential in his field", we could expect this. But I often hear it stated as absolute ability.

Comment author: gwern 17 November 2013 03:20:11AM 1 point [-]

Is that a real distinction? When Shakespeare is the 'most influential', then in some respects, he is setting what it means to be 'able'. He is setting our norms and expectations, laying down the language we think and write in. John Keats: "He has left nothing to say about nothing or any thing." Ralph Waldo Emerson: "His mind is the horizon beyond which at present we do not see."

When a writer is so influencing (should I say, 'distorting'?), is it really meaningful to draw a distinction between 'influential' and 'able'? Like Phil's implicit claim that every writer has an equal chance of being Shakespeare, this is not something I am willing to instantly grant without inspection.