The Shakespeare Conference: SHK 24.0480  Monday, 14 October 2013


[1] From:        Gerald E. Downs <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         October 11, 2013 11:45:27 PM EDT

     Subject:     Re: Spanish Tragedy 


[2] From:        Gerald E. Downs <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         October 14, 2013 1:00:06 AM EDT

     Subject:     Re: Spanish Tragedy Additions 




From:        Gerald E. Downs <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         October 11, 2013 11:45:27 PM EDT

Subject:     Re: Spanish Tragedy


I haven’t read later sources but Lukas Erne’s Beyond the Spanish Tragedy (2001) and Philip Edwards’s 1959 edition provide enough commentary for a new look at the early transmission. If I were methodical this play would be low on my list because it isn’t labeled bad quarto. Yet I’m sure that someday many unsuspected texts will end up in the BQ category on comparison to acknowledged shorthand reports. By someday acknowledged I don’t mean tomorrow, but The Spanish Tragedy is a good test case for now.


There’s a strong presumption that the first 1592 text (printed by Abell Jeffes) was a bad quarto because the next printing (by Edward Allde), according to the title-page, was “newly corrected and amended of such grosse faults as passed in the first impression.” Gross faults are not good. While recalling that bad quarto doesn’t necessarily imply that bad = gross, the only way to discover reporting is by inference from anomalous text. Unfortunately, the “first impression” (like some I’ve made) didn’t last; no copies survive.


I follow Edwards “for convenience” in referring to Allde’s edition as 1592. (Though not the only edition that year, it’s not Q1, Q2, or even a Q, but an “octavo in fours.” The survival of books is an interesting topic, as here: the editions of 1592, 1594, 1599, and 1603 exist in one copy each; ten copies of 1633 survive. That isn’t because the earlier books were ‘more thumbed’; 99.9% of them have gone the way of 99.9% of small books and early editions were less valued, even by early collectors—to whom we owe what we have.)


Because 1592 seems to be less corrupt before 3.15 than after, Edwards makes a case that the early acts were printed from the author’s foul papers (the shrug-shoulder default) and that the latter part derived from the lost, gross, Jeffes edition. Erne questions the arguments for ‘memorial reconstruction’ adduced by Edwards, some of which I will revisit. To me the issue is not whether Jeffes provided copy for 1592, (following a common preference for printed copy) but whether all of its copy is reported. Erne observes that Edwards claims the 1592 copy-text beginning with 3.15 was a “version of the play which was acted . . . [and] reported or assembled by ‘unauthorized’ persons” (Edwards Introduction). Erne agrees that the first passage discussed is “textually very corrupt and in places difficult to make sense of” (Erne, 60):


            Enter Ghoast and Reuenge,


            Awake Erictha, Cerberus awake,

            Sollicite Pluto gentle Prosperpine,

            To combat Achinon and Ericus in hell.

            For neere by Stix and Phlegethon:

            Nor ferried Caron to the fierie lakes,

            Such fearfull sights, as poor Andrea see?

            Revenge awake.     (3.15ff)


Revenge probably had the better plan. These lines and the rest of the short scene have (potentially) interesting problems that may point to reporting. But Erne asserts that


“Most of the corruptions . . . are not those we would expect from a faulty memorial reconstruction. Why would a reporter mistakenly give the lines to the Ghost? add mistaken stage directions? transfer two words or forget a line? . . . If, however, the deficiencies mirror the quality of the copy from which Jeffes set up his edition, we would expect to find similar irregularities in the other passages which Edwards attributes to the lost quarto. This is not the case, however” (61).


Actually, the lines are those of Don Andrea’s ghost; the question is why ‘Ghost’ when the prefix elsewhere in 1592 is ‘Andrea’? The two characters are here given both entrance and exeunt when they were already on stage and when only one exits at the scene’s end. True, we can’t expect these anomalies to stem from controlled authorial work or a competent MR; but such mistakes are compatible with shorthand reporting because the dialogue imperfectly guides transcription and other agents aren’t left out.


Shorthand reporting is irregular for a number of reasons. At times reporter and actors are faultless, as in John of Bordeaux (our control text). Regular meter can be recovered easily, even if it is written as prose; faulty meter can be repaired or rewritten, or it may remain faulty. Much depends on the actors’ delivery and the care taken in the fixing. The evidence is in the deficiencies and not, as Erne implies, in seemingly good text.


Scholars have tried to make sense of the passage by positing a lost line—a tempting fix when meaning may be manipulated with some well-chosen words. However, it’s better to try to repair the text from its own substance. Besides, anyone can drop a line and even if that happened there is no conclusion to be drawn about reporting. I’ll take a stab at resolving the corruption as if it did occur during performance:


Nothing is more certain for a shorthand reporter ignorant of mythology than his screwing up the names; context will not help him during note-taking or in transcription. Compositors might compound error by misreading (as 1592 makes some bad Latin worse). Erictha is Erichto, Achinon is apparently Acheron, and Ericus is Erebus. Aesop’s Kid spelt better than that.


Amid the regular verse lines, ‘To combat Achinon and Ericus in hell. / For neere by Stix and Phlegethon:’ have twelve and eight syllables. Editors therefore prefer ‘in hell’ to end the second line. But ‘to combat Erebus in Hell’ seems OK (Why not?). The problem is, Acheron is a river (albeit personified); belonging with Styx and Phlegethon, it’s the one on which Caron ferried souls. I suppose the player misspoke lines that originally went (as Webb Pierce would say) something like this:


Solicit Pluto gentle Prosperpine,

            To combat Erebus in hell. For never

By Acheron and Styx and Phlegethon,

            Nor ferried Caron to the fierie lakes,

            Such fearfull sights as poor Andrea sees?


I suggest this reading only to show how easily theatrical reporting can account for corruption. Conclusive evidence is more likely to be found in the aggregate. For example, preceding 3.15 Hieronimo repeats a Latin proverb,


    Chi mi fa piu carezze che non suole,

    Traditio mi ha, o trader mi vuole. (3.14.168-9)


    Mi. Chi mi fa? Pui Correzza Che non sule

    Tradito viha otrade vule              (1592)


Pui sums it up. It may be that Kyd “could scarcely Latinize his neck verse” but he had to do better than this. The corruption looks like that in Bordeaux: lousy Latin but brilliant for the shorthand recording of a foreign language. Plain old transcription can’t account for it, yet editors hardly comment on a feature that often recurs in 1592. Edwards says of the 1594 reprint that


Two wrong attributions of speeches have been corrected [Why were they wrong in the first place?] . . . a half-hearted attack on the corrupt Latin has produced seven corrections. These corrections in the Latin cannot be from an authoritative source, since so much nonsense is allowed to stand (xl).


And yet “1592 is a good text.” No word about where the “nonsense” Latin came from, though nonsense trumps foul papers. My hypothesis is that both 1592 editions are reports, perhaps versions of the same stenographic transcription. Edwards observes that the “careful spelling of 1592 has its origin in Allde’s printing house.” That’s the treatment any shorthand report would get from various agents until many features of the report would be unrecognizable without comparison to a good text. But vestiges remain as evidence hard to explain by transcription alone.  As with other plays, a more complete case could easily be made for 1592.


As for the 1602 revisions: Edwards acknowledges that its text is “[rough] enough to make us suspect that [Pavier] got it surreptitiously—perhaps by transcript, but conceivably through the actors” (lxiv), but notes that “it is quite impossible that a ‘reporter’ in the audience should have known the old play so well that he would be able to spot that the brief Second Addition was an addition.”


If one posits “a reporter in the audience” it could be helpful to think of such reporting as possible, if only for the sake of argument. To an adept stenographer (Bordox), reporting the entire play would be work of a few hours (before transcription); he wouldn’t need to spot additions. Nevertheless, if the publisher thought it worthwhile to fit the added lines to his printed text (which he did, though they don’t really fit) it would be worth a reporter’s time. I believe theatrical reporting happened a lot. However, transmission won’t help to find the author of the Additions, as far as I can see. But without the transmission you may not get started.


Gerald E. Downs



From:        Gerald E. Downs <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         October 14, 2013 1:00:06 AM EDT

Subject:     Re: Spanish Tragedy Additions


Michael Eagan replied to my posting:


> I was stunned to read Gerald Downs’ long exposition on

> parallels and authorial attribution.


I haven’t stunned anyone since they took away my taser. As for the “long exposition,” I know the feeling; sometimes I look to see how many hundred pages are yet to go.


> Every one of his positive criteria apply to 1 Richard II,

> confirming its attribution to Shakespeare.


I’m not sure what Michael means by my “positive criteria.” I accept reminiscences of Shakespeare in both the 1602 Additions and in Woodstock and see an analogy between the two, where the Vickers method might force one to agree with M. Egan’s attribution or to deny Sir Brian’s; unless one agrees with me that parallels don’t always force attribution. I used the Q1 Hamlet example, wondering whether M. Egan would reopen the Woodstock can of worms himself. He is entitled, in my opinion; if Shakespearians accept the Additions attribution they should consider the other seriously.


> What clinches the matter, though not a single critic has ever

> addressed the evidence, is that verbal and phrasal parallels

> (to go no further at this point) exist between 1 Richard II and

> plays not attributed to Shakespeare until the 20th century.

> These include Edward III. The Two Noble Kinsmen, and yes

> even the Sir Thomas More fragment.


The Egan memory must be hereditary. Or maybe I’m no critic; but I offered my opinion back in the betting days. I accepted Woodstock / Shakespeare parallels as sufficiently demonstrated. But I also cited one of my rules-of-thumb: when a playtext exhibits parallels to numerous others, that’s no sign the play influenced the others. It’s the other way round; the text full of parallels is derivative.


As far as the question goes, it can’t matter when irrelevant someones attributed plays to Shakespeare; contemporary knowledge is what matters. For example, if players are behind Woodstock as it exists, they presumably knew their sources.


> These data completely destroy the argument that the play was

> written by a contemporary plagiarist. That leaves only one alternative.


I suppose Brian Vickers will agree. But I can’t. Woodstock doesn’t come with any credentials; it looks very much like a report. M. Egan denies that possibility outright and even suggests that it might be holograph Shakespeare. “Only one alternative” is one way to go; “back to square one” is another.


Larry Weiss observes:


> As my opinion in Egan v. Elliott . . . points out at length, verbal

> parallels are not particularly significant, especially in the face of

> astronomically divergent stylistic differences.


Parallels can often be very significant; so can stylistic differences. The recent Woodstock argument neglected to discuss just what the text is, transmission-wise. When a playtext is reported issues are not limited to “him or him”; a lot can happen between writing and reporting. Revision, addition, plagiarism, memorial reconstruction, or all of them, can happen and happen again (as I guess of Woodstock).


Michael Egan remarked:


> [I]n the Two Noble Kinsmen and Sir Thomas More the parallels

> are found exactly in those scenes previously identified by the style

> critics as Shakespeare. So the process is mutually confirming.


That would be so only if these texts had as many parallels (per line) with other plays. As I’ve noted elsewhere, Hand D’s text could derive from Shakespeare. If it does, that doesn’t mean Woodstock is itself by Shakespeare.


> I have never claimed that parallel phrases clinch the matter but

> without them there would be no case. What proves my thesis the

> quality of the writing.


Michael seemed to make that ‘clinching’ claim. Paradoxically, too many parallels weaken the case. Now he claims that “quality” proves the attribution, but quality is hard to prove. He was right the first time: with no measurable comparison there is no viable case.


> Brian Vickers has introduced a “four-word string” test for

> attribution, and run[s] the Works and other plays through a

> plagiarism program. I invite him to test 1 Richard II.


No doubt Woodstock strings “more than four” words at a time, since it borrows from Shakespeare works. The play should be tested, perhaps with more attention given to ‘run-of-the mill’ collocations that might show up in the lines Michael identifies as of 'Shakespearian quality without parallel.' If some “usual” collocations with Shakespeare appear, Egan's chosen lines (how many are there?) may not be dis-qualified. I suspect the lines aren't Shakespeare’s; for the same reason, I won’t deny they are from a previously unknown Shakespeare text: we cannot know.


Bill Lloyd made some fair observations, not all of which I question:


> But Jackson’s and Lake’s arguments that the play can be

> shown to be by Samuel Rowley seem to me to be very strong.

> Dr Egan disagrees—and there we rare.


You can say that again, though I only medium rare: Woodstock is quite corrupt and handled by a number of theatrical types. It is second only to Bordeaux in Werstine’s categorical list of manuscript alterations. The text should be studied with such evidence in mind before attempts at attribution.


> The parallels adduced for Shakespeare’s hand in the SpTr

> Adds have been vetted by negative check, database search,

> stylometrics etc.


The “negative check” is inadequate, though we learn Lyly was not the author. The stylometric studies were inconclusive (and not worth much if unknown writers were involved).


Gerald E. Downs



Subscribe to Our Feeds


Make a Gift to SHAKSPER

Consider making a gift to support SHAKSPER.