The Shakespeare Conference: SHK 14.1152  Tuesday, 10 June 2003

From:           Jim Carroll <This email address is being protected from spambots. You need JavaScript enabled to view it.>
Date:           Monday, 9 Jun 2003 21:57:50 EDT
Subject: 14.1129 Re: King John, Titus, Peele
Comment:        Re: SHK 14.1129 Re: King John, Titus, Peele

First, let me make a correction to my post of June 6. I wrote "For
example, if 3 rejections of 51 tests pass a 20,000 word text, shouldn't
a 3,000 word text be allowed 6-7 rejections of those same 51 tests
((20,000/3000)x3)?" and of course I should have written "...shouldn't a
3,000 word text be allowed 20 rejections of those same 51 tests
((20,000/3000)x3)?" I set up the calculation, then forgot to carry out
the final multiplication by three, and since 20,000/3000 is 6.67, I
wrote "6-7 rejections" instead of the correct "20 rejections."

Ward Elliott <This email address is being protected from spambots. You need JavaScript enabled to view it.> wrote on Friday, 06 Jun
2003 15:21:56:

>WE, new: see WE, old, above.  Smaller samples have more variance than
>large, and wider ranges, often so wide that they no longer distinguish
>non-Shakespeare, and you can't validate them.  Hence, the shorter the
>sample or baseline block, the fewer tests you can validate properly and
>use.  It's not cooking the books to use tests validated for the sample
>size at issue, it's a commonsense methodological safeguard.

But your method of validation is the problem. You can't decide to
discard a test simply because it no longer matches enough of the smaller
blocks of text. If you work with 20,000 word blocks and find that 51
tests can reliably distinguish your "core" sample of what you believe to
be authentic Shakespeare, then when you divide the same texts into
smaller blocks, of course more blocks will fail, for precisely the
reason you give above. But you can't discard the tests which give too
many rejections of authentic Shakespeare, because some of the same
blocks of text are _passing_ the other tests _because_ they are too
short! The only logical procedure is to use the same tests for every
size block and either increase the range of values which cause a
rejection, increase the number of rejections that are allowed to a
particular block, or increase the number of blocks that can be rejected
before rejecting the whole as Shakespearean.

Here is what Foster said on the subject (CHUM vol. 32, 491-510, 1999;

"...Closer study of the Clinic's work, including tests mistabulated or
not reported at all, reveals that the early canonical poems also run
into trouble with the Clinic's attributional tests. In 1992-1994,
Elliott and Valenza circumvented this problem by investigating which of
the so called "Play-Validated Tests" could be included as
"Poem-Validated Tests" without having to report rejections for Venus and
Adonis, The Rape of Lucrece, and the Sonnets.  These poems mutually
survived eight of the Clinic's "Play-Validated Tests": Grade Level,
HCW's, Feminine Endings, Open Lines, Enclitic clingers, Proclitic
clingers, With-as-penultimate word, No x 1000/No + Not, BoB5, and BoB7
(in 1995b, 1996a, Appendix Four, 46 of the 54 "Play-Validated Tests" are
thus suppressed for the canonical poems). Next, "A Lover's Complaint"
and "A Funeral Elegy" were tested against the eight "Play-Validated"
tests that the early canonical poems were able to pass.  This procedure
resulted in four rejections for "A Lover's Complaint" and five for "A
Funeral Elegy". Elliott and Valenza neglect to mention that these two
late poems pass many of the original 54 tests for which Venus and
Adonis, The Rape of Lucrece, and the Sonnets receive "not-Shakespeare"
rejections. In attributional work this is called "stacking the deck".
When the Clinic's full battery of tests is rigorously applied to Venus,
Lucrece and the Sonnets, these canonical poems do no better than many
"Apocrypha" and "Claimant" texts that are rejected by the Clinic's badly
gerrymandered regime."

I found the correct url for your website in your "Smoking Guns..."
paper, and it is


Here you fail to answer Foster's points entirely, simply reiterating

"Giant, 20,000-word blocks average out much more variance than small,
3,000-word ones and permit a much wider range of valid tests. Big-block
play-validation and small-block poem-validation are two separate issues;
they must be treated separately, and they were by us."

failing to note again that it is the way you treat block-sizes that is
the fundamental problem. You can _not_ avoid rejections by reducing the
number of tests.

>examples Mr. Carroll cites are mostly Ford peculiarities, well validated
>for distinguishing FE-sized Ford blocks from FE-sized Shakespeare blocks
>and well adapted to the question we asked:  "Could Ford have written the
>Elegy?"  They worked well enough for this purpose to make the composite
>odds of Shakespeare authorship thousands (then, gazillions now) of times
>worse than those for Ford authorship.

You like to quote these numbers, but they are meaningless if the testing
procedure is invalid in the first place. Also, if there are possibly
thousands of tests that could be used, what are the odds that if you
apply only 51 of them, a certain configuration of results is obtained?
Astronomically high, but _some_ configuration has to be the result.

You can't just use tests that distinguish Ford from Shakespeare, unless
you have _all_ the tests that distinguish Ford from Shakespeare. In this
case you use even fewer than 51 tests, making the results that much less
reliable. And it's highly unlikely that all of the tests that you do use
for this case are reliable in the first place, and I've already
complained about the use of only one preposition, "of". Is it possible
that if you considered all prepositions, the picture might be that
prepositions don't tell you much?

Another dubious test is the "proclitic/enclitic" test. This is a
subjective determination of stress patterns. It would be rare for any
two persons to scan even something as short as a sonnet the same way,
and as Tarlinskaja's system is based upon identifying stresses,  the
test is impossible to apply objectively on texts the size of the
Shakespearean ones. In your paper "The Professor Doth Protest Too Much
Methinks" you answer Foster by saying (p430) "But this misses the point
of the test, which is not to count instances of normal stress, but
instances where a "clinging" monosyllable loses its normal stress for
metric reasons." I know that, but the point is you have to identify
where the stresses are to begin with, and that is a subjective
judgement. I can recommend the fine book "Meter in English", edited by
David Baker (University of Arkansas Press, 1996) for many examples of
the differences of opinion that can arise when lines of poetry are
scanned by different readers.

[long snip]

>Even the
>stoutest rejectors of Shakespeare have had to admit that LC has lines
>worthy of Shakespeare; even the stoutest defenders had to admit that it
>has lines not so worthy of Shakespeare.

Which means absolutely nothing, since you can say that for all of
Shakespeare's non-dramatic poems. But most of the objections are due to
changes in taste, not because of some lack of poetic technique. These
lines from Lucrece, for example:

"Oh hateful, vaporous, and foggy Night,
Since thou art guilty of my cureless crime,
Muster thy mists to meet the eastern light" (Luc. 771-3)

A little too much of the "It was a dark and stormy night" mode, plus too
much alliteration. The question is not "Is it "worthy" of Shakespeare"
(which requires us to decide what we would like Shakespeare's writing to
be), but "Is this writing consistent with his practice elsewhere"? V&A

"It shall be fickle, false, and full of fraud;
Bud, and be blasted, in a breathing while;" (V&A 1141-2)

Too much alliteration here as well, but the preponderance of soft "f"
and "s" sounds is typically Shakespearean. The Lucrece example has more
variety with its hard "g" and hard "c" sounds.

>We haven't surveyed practicing
>poets, but we have done a pilot study of a Claremont lit class for our
>Golden Ear test.  The class as a whole divided 6-6; the six Golden Ears
>in the class divided 3-3, close enough to make it interesting.  If our
>stylometric evidence is good, the nays seem closer to having it right.

What's a "Golden Ear"? I hope they've all been reading Shakespeare for
25 years or thereabouts.

Memorable lines are rare in Elizabethan/Jacobean poetry (actually, any
poetry of any time). Shakespeare has many, Donne has some, Jonson a few.
You've failed to answer the question: If Shakespeare didn't write LC,
_who did_? Who else could combine a pastoral setting with lines so
finely, so exquisitely wrought, such as these:

Of folded schedules had she many a one
Whe she perused, sighed, tore, and gave the flood;
Cracked many a ring of posied gold and bone,
Bidding them find their sepulchres in the mud;
Found yet moe letters sadly penned in blood,
With sleided silk feat and affectedly
Enswathed and sealed to curious secrecy. (LC, 43-49)

"Cracked many a ring of posied gold and bone". Get it, Ward?  Here
Shakespeare combines his typical verse style in an almost flawless mix
of soft and hard consonants, with his later highly enjambed style, and
startling imagery ("Cracked many a ring of posied gold and bone").  Try
Daniel's "Complaint of Rosamund", appended to his own sonnet sequence,
for comparison.

Jim Carroll

S H A K S P E R: The Global Shakespeare Discussion List
Hardy M. Cook, This email 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.

Subscribe to Our Feeds


Make a Donation

Consider making a donation to support SHAKSPER.