Make a Donation

Consider making a donation to support SHAKSPER.

Subscribe to Our Feeds

Current Postings RSS

Announcements RSS

Home :: Archive :: 2007 :: July ::
Shakespeare Golden Ear Test
The Shakespeare Conference: SHK 18.0459  Monday, 9 July 2007

[1] 	From: 	Larry Weiss <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
	Date: 	Friday, 06 Jul 2007 15:14:59 -0400
	Subj: 	Re: SHK 18.0454 Shakespeare Golden Ear Test

[2] 	From: 	Ward Elliott  <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
	Date: 	Saturday, 07 Jul 2007 00:15:33 -0700
	Subj: 	RE: SHK 18.0454 Shakespeare Golden Ear Test

[3] 	From: 	Ward Elliott  <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
	Date: 	Saturday, 07 Jul 2007 00:15:33 -0700
	Subj: 	RE: SHK 18.0454 Shakespeare Golden Ear Test

[4] 	From: 	Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
	Date: 	Saturday, 07 Jul 2007 23:22:26 -0700
	Subj: 	RE: SHK 18.0454 Shakespeare Golden Ear Test

[5] 	From: 	Jim Carroll <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
	Date: 	Sunday, 08 Jul 2007 12:37:28 -0400
	Subj: 	Re: Shakespeare Golden Ear Test


[1]-----------------------------------------------------------------
From: 		Larry Weiss <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date: 		Friday, 06 Jul 2007 15:14:59 -0400
Subject: 18.0454 Shakespeare Golden Ear Test
Comment: 	Re: SHK 18.0454 Shakespeare Golden Ear Test

Ward Elliott's "golden ear" experiment might be worth trying, but I 
caution extreme care in selecting the subjects and the control plays. 
The question is, as Ward frames it, whether "educated intuition" is as 
good as computerized number crunching, the key word being "educated." 
Frankly, I am dubious about the validity of "Pilot studies with 
undergraduates from The Claremont Colleges [which] indicate that 
individual amateurs can distinguish Shakespeare from Non-Shakespeare 
with up to 79% accuracy." In particular, I would like to know more about 
the non-Shakespearean controls.

Several years ago, when Charles Hamilton's book suggesting that The 
Second Maiden's Tragedy" (probably by Middleton) was the lost Cardenio 
had just been published, a troupe in New York put on a laboratory 
production of that play.  There was a discussion period after the 
performance in which the discussion leader asked the audience if the 
play "sounded like Shakespeare."  Almost all responded that it did.  It 
seemed to me that the audience-who appeared none too sophisticated but 
were interested enough to go out of their way to attend such an off beat 
production in a loft-were reacting to nothing more than iambic 
pentameter, or maybe just verse.  Since it is probable that most 
audiences never attend plays written in verse except Shakespearean 
plays, any play in verse is likely to "sound like" Shakespeare.

[2]-------------------------------------------------------------
From: 		Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date: 		Friday, 06 Jul 2007 17:50:12 -0700
Subject: 18.0454 Shakespeare Golden Ear Test
Comment: 	RE: SHK 18.0454 Shakespeare Golden Ear Test

Here's our progress report on the first eight hours of the Golden Ear 
test.  We've gotten 18 takes and found one Golden Ear, two possible 
Bronze Ears, and some good offline feedback.  Our thanks to all who have 
given it a try.

But we also have a problem.  Many of the "takes" are actually retakes 
and should not be counted.  The test gives you your score and the 
correct answers before asking you for contact information, but you can't 
get the correct answers again after you've given us contact information 
- so some people are retaking the test to recheck the correct answers. 
Others just want another shot at it, and who are we to grudge them that? 
  As one student put it about a different web exercise: "Now how am I 
going to get any work done?  This is WAAAAAYYY better than solitaire." 
And the program can't be changed any time soon because the student who 
did it has graduated.

Nevertheless, we are still looking for real Golden Ears, to say nothing 
of real Silver and Bronze Ears, and it's the first try that's the proper 
test to count, not the second or third.  There is also a fix: to ask 
people to use "SHAKSPER" as their group name on the first try and "test" 
(both without the quotes) on all others.  If you give us contact info, 
do that on the first try also, or e-mail me, 
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 , if the 
program is not letting you give the right message, so we'll associate 
your contact info with the proper test.

Again, thanks to everyone who has given the test a try.  We encourage 
others to do likewise, as often as they wish, but to sign in under 
"SHAKSPER" for the first try, "test" for all others.


 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
  will soon cease to work; please use

 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
  instead.

Ward Elliott
Claremont McKenna College

[3]-------------------------------------------------------------
From: 		Ward Elliott  <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date: 		Saturday, 07 Jul 2007 00:15:33 -0700
Subject: 18.0454 Shakespeare Golden Ear Test
Comment: 	RE: SHK 18.0454 Shakespeare Golden Ear Test

Golden Ear Progress report end of first day:  we have what looks like 14 
takers, 20 takes, 3 bronze ears, one golden ear found, who gave 86% 
correct answers. Wow!  Nobody got less than 60% of the questions right. 
  Keep up the good work, SHAKSPERians. Our thanks to those who have 
taken the test and our continued encouragement to those who have not.
Ward Elliott
Claremont McKenna College

[4]-------------------------------------------------------------
From: 		Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date: 		Saturday, 07 Jul 2007 23:22:26 -0700
Subject: 18.0454 Shakespeare Golden Ear Test
Comment: 	RE: SHK 18.0454 Shakespeare Golden Ear Test

Golden Ear progress report toward end of second day, after removing the 
three most obvious second tries:

To date: 31 takers, 31 takes. Identified: one golden, three bronze. 
Anonymous: one golden, one silver, five bronzes.  14 takers considered 
themselves pros, 12 were critics, 5 were writers, 10 in the arts, 4 in 
natural or social sciences.  Average score, all: 18.84.  Average score 
of pros: 19; amateurs: 18.7

Observations so far:

1. all these averages are much better than chance, probably better than 
any other group we have tested, and better than any computer we know of 
in identifying very short passages. Nice going, SHAKSPERians, keep it up.

2. Amateurs are surprising close to pros in accuracy with no correction 
yet for pros' expected higher rates of recognition of the texts we used. 
  Nice going, amateurs.
3.  2/3 of the highest-rated takers, bronze or better, are anonymous, 
hinting that these takers are less impressed with their own performance 
than we are.  They should be taking a bow, but they are not. It's OK; we 
would rather have high participation and low identification than low 
participation and high identification.  We'll be sorry not to have the 
extra talent reachable for the second Golden Ear test, but the anonymous 
takes are still useful for aggregation on this test. If you got a bronze 
or better and change your mind, remember your score and e-mail it to me.

4. The second-try problem seems to have sorted itself out.  The 
remaining takes all look like first tries.

5. 2/3 of takers are not identifying themselves as coming from SHAKSPER. 
  It's not a big problem so far, since they're all coming from SHAKSPER 
these days, but it could make future comparisons with other groups more 
difficult, a pity, because it looks like SHAKSPERians are taking the 
test more seriously, and doing better on it, than other groups we have 
tested.

Our thanks, again, to all who have taken the test, and our encouragement 
of the others to go and do likewise.

Ward Elliott

[5]-------------------------------------------------------------
From: 		Jim Carroll <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date: 		Sunday, 08 Jul 2007 12:37:28 -0400
Subject: 	Re: Shakespeare Golden Ear Test

Short passages taken out of context are no more useful for "intuitive" 
judgments concerning Shakespeare than they are for tests that involve 
counting by computer. Notice that I did not say "computer judgments" 
because computers don't judge anything. No one would (well, maybe some 
would try) use a 150 word passage in a computerized statistical study 
because it is simply far too short to be meaningful. Likewise this 
online test tells you nothing. I doubt anyone would try to ask anyone to 
identify painters by cutting out two inch squares of their paintings. 
This test in fact reminds me of the old Oxfordian trick of taking 
_single lines_ from various authors and running them together, then 
saying "See, you can't tell the difference." In this case, the 
Shakespearean passages have been chosen to minimize what is distinctive 
about Shakespeare: his interesting diction and his constant stretching 
for metaphorical expression. As a result, I was able to identify all the 
non-Shakespearean works except the one by Spenser as non-Shakespearean, 
and identified several Shakespearean passages as non-Shakespearean. For 
the record, excluding a couple of passages I recognized (FE, Shall I 
Die?) I judged the H6 passage to be Shakespearean, the ALC passage to be 
Shakespearean and the E3 passage to be non-Shakespearean.

As I said, computers don't judge. The numbers they provide are only as 
useful as the humans judging them, and as I have pointed out many times 
here before, the McKenna clinics studies are seriously flawed. For 
example see: http://www.shaksper.net/archives/2003/1257.html

Jim Carroll

________________________________________________________________________
AOL now offers free email to everyone.  Find out more about what's free 
from AOL at AOL.com.

Subject:
Re: Shakespeare Golden Ear Test
From:

 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 
Date:
Sunday, 08 Jul 2007 12:37:28 -0400
To:

 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 


 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 


Short passages taken out of context are no more useful for "intuitive"
judgements concerning Shakespeare than they are for tests that involve
counting by computer. Notice that I did not say "computer judgements"
because computers don't judge anything. No one would (well, maybe some
would try) use a 150 word passage in a computerized statistical study
because it is simply far too short to be meaningful. Likewise this online
test tells you nothing. I doubt anyone would try to ask anyone to identify
painters by cutting out two inch squares of their paintings. This test in
fact reminds me of the old Oxfordian trick of taking _single lines_ from
various authors and running them together, then saying "See, you can't tell
the difference." In this case, the Shakespearean passages have been chosen
to minimize what is distinctive about Shakespeare: his interesting
diction and his constant stretching for metaphorical expression. As a
result, I was able to identify all the non-Shakespearean works except
the one by Spenser as non-Shakespearean, and identifed several 
Shakespearean
passages as non-Shakespearean. For the record, excluding a couple of
passages I recognized (FE, Shall I Die?) I judged the H6 passage to be
Shakespearean, the ALC passage to be Shakespearean and the E3 passage
to be non-Shakespearean.

As I said, computers don't judge. The numbers they provide are only as
useful as the humans judging them, and as I have pointed out many times
here before, the McKenna clinics studies are seriously flawed. For
example see: http://www.shaksper.net/archives/2003/1257.html

Jim Carroll

_______________________________________________________________
S H A K S P E R: The Global Shakespeare Discussion List
Hardy M. Cook, 
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 
The S H A K S P E R Web Site <http://www.shaksper.net>

DISCLAIMER: Although SHAKSPER is a moderated discussion list, the 
opinions expressed on it are the sole property of the poster, and the 
editor assumes no responsibility for them.
 

©2011 Hardy Cook. All rights reserved.