University of Virginia Library

Notes on the Text of Thierry and Theodoret Q1
by
Robert K. Turner, Jr [*]

In 1621 The Tragedy of Thierry King of France, and his Brother Theodoret was first printed by Nicholas Okes in an anonymous quarto edition for Thomas Walkley.[1] The authors' names, omitted from the Q1 title-page, were partially supplied by Q2 (1648), the first issue of which attributed the play to Fletcher and the second to Beaumont and Fletcher. Massinger's hand has also been detected in the work, and Cyrus Hoy,[2] the most recent investigator of the authorship of the Beaumont and Fletcher plays, divides Thierry and Theodoret between the three dramatists in the following way:

• Beaumont: III; V,i
• Fletcher: I,i; II,ii-iii; IV,i; V,ii
• Massinger: I,ii; II,i,iv; IV,ii.

From a purely literary point of view, the play is distinctly inferior to the great Beaumont and Fletcher collaborations, but Q1 is interesting to the textual critic because it apparently is a member of the rather mysterious cluster of King's Men properties, including Philaster, The Maid's Tragedy, A King and No King, and Othello, which appeared in the four-year period from 1619 through 1622. Thierry and Theodoret is the only one of these plays whose first edition was not succeeded within a few years by a second

edition containing a corrected or even a materially different text. Hence Q1 is of particular importance and is worthy of close scrutiny. In what follows I shall discuss some bibliographical features of the book which permit an insight into its printing and then attempt to make some inferences about the kind of copy from which it was set up. Let us first consider the following categories of bibliographical evidence:

Similarly, type which had been used in sheet C reappears in sheet D, but this time the source of the recurring types seems to have been only the outer forme of the old sheet. The last "e" of "Beleeue" (C4v, l. 14) is found again in "Were" (D1v, l. 3 — first "e"), the "m" of "my" (C2v, l. 29) in "comfort" (D1v, l. 37), the "o" of "owne" (C4v, l. 8) in "almost" (D1v, l. 12), and, in addition, other C(o) types reappear on D3v and D4. C(i) types do not show up before E1v: on that page in "of" (l. 16) the "f" seems to be the same type which previously appeared in "proofe" (C3v, l. 11) and in "liude" (l. 4) the "d" that of "aimde" (C3v, l. 21).

Evidence from recurring types must, I believe, be treated with reasonable caution: countless variations in the printing process (e.g., differences in inking, in pressure, in paper shrinkage) conspired to affect the impressions of the individual types and occasionally different types sustained nearly identical injuries, perhaps because of weaknesses inherent in the design of the letter or because of defects in the punches or the matrices.[4] Moreover, the location of a recurring type on a particular page does not mean that the forme which previously contained that type was distributed at the time the page on which the type reappears was set but by the time it was set. However, either by itself or, preferably, in conjunction with evidence from other sources, information furnished by the reappearance of types can tell a great deal about the method of printing a book.

In the case of Thierry, the evidence of the running-titles, type shortages, and type recurrences suggests that sheet C, as well as sheet D, was composed by formes. Had C been set seriatim, we would expect to find some type either from B(i) or B(o) on C2v since the former was distributed by the setting of B1v and the latter by the setting of B2. In addition, we would have to suppose that after running out of W's on C3v and using VV's in the composition of the rest of that page and all of C4, the workman distributed enough type to obtain the three W's which appear on C4v and then ran out of the letter again, for he apparently had no W's when he began to set D(o). It seems more reasonable to think that C(o) was first composed, B(i) being distributed after the setting of C2v but before the completion of C3 and B(o) being distributed after C3 but before the completion of C4v. Thus C(o), in skeleton I, could have been sent to the press either before or during the composition of C(i), which was imposed in skeleton II. From type reappearance, we know that C(o) was off the press and distributed at least by the time D1v was composed; however, we should recall that during the setting of D4v the supply of W's, which had been exhausted at C3v and which had remained so during the composition of C4, D1, D2v, and D3, was replenished. We may assume, then, that C(o), which contained fifteen W's, was distributed during the setting of D4v, thus freeing skeleton I for use in imposing D(o). But D(i) was also imposed in skeleton I; hence, by the time the composition of D(i) was completed D(o) was off the press

and skeleton I was again freed for use. It may be possible that type which had not previously been used in Thierry was also distributed during the setting of D(i); eighteen W's were used in the setting of C4v and D(i) but C(o) had contained only fifteen pieces of this type. Since there is no sign of the distribution of C(i) before sheet E and no further indication of type shortage in the remaining sheets of the book, it may be that the additional W's were secured from standing type which had been set for another job.

After the perfecting of sheet D, skeleton I was abandoned and skeleton II was used to impose both formes of all subsequent sheets. D(o) must have come from the press at about the time the composition of D(i) was completed because its skeleton was used to impose D(i). D(o) type begins to appear at E1; C(i) type, however, does not begin to show up before E1v. Hence, it seems probable that sheet E was also composed by formes, outer first, and that C(i) was kept standing until after the composition of E4v, at which time skeleton II was freed and placed around the type pages of E(o). Type reappearances show that D(i) was distributed before the setting of E3v was completed, thus freeing skeleton I, but E(o) was off the press in time for skeleton II to be used in preference to I for E(i). From this point on, the picture seems fairly clear. Composition so lagged behind presswork that the re-use of skeleton I (or the construction of a new skeleton) was never advantageous; in all cases the wrought-off forme seems to have come from the press before, or perhaps just as, a new forme was ready for imposition. There seems every reason to believe that composition continued by formes. In order to change to seriatim setting, Compositor B would have had to set three more type-pages in any given sheet to get the first forme of that sheet ready for the press than would have been required had he continued to set by formes. Moreover, had he been able to gain enough time with respect to the pace of his press to accomplish this additional work, the setting of only one more type-page would have made the second forme of the sheet ready for imposition. Under these circumstances, he would almost certainly have reintroduced his abandoned skeleton, for by doing so he could have avoided all of the inefficiencies of the one-skeleton process.[5]

From 1615 to 1626 Okes is known to have printed fifteen dramatic quartos, excluding eight triumphs (which are less than four sheets long).[6] With the sole exception of Lingua Q3 (1617), these were all multiple-skeleton

books; hence, we may take it as a general rule that for dramatic quartos two-skeleton procedure was standard in Okes's shop during this period. The reversion to one-skeleton printing in Thierry, therefore, may be regarded as distinctly unusual. Possibly an analogous situation is found in Philaster Q1, printed by Okes in 1620, where there is a temporary change from two- to one-skeleton procedure and where the general condition of the text justifies the inference that bad copy caused the composition to lag behind presswork.[6a] But, as I will attempt to show below, the manuscript from which Thierry was printed, while perhaps not ideal copy, was probably no worse than average; it certainly was far superior to the copy of Philaster Q1. Hence, we should look elsewhere for an explanation of the lag in composition, and I suggest, on grounds that must remain speculative until further investigation has been completed, that the difficulty lay with the compositor rather than the manuscript. The preferred spellings of Compositor B, the workman who set sheets C through K, compare reasonably closely with the preferences of the workman who set most of sheet G of The Maid's Tragedy Q1: the two agree in favoring blood, deare, eye, honor, and yeare to alternative spellings.[7] If my identification is correct, we have some reason to think that this workman was not a regular compositor but a person who had other duties in Okes's establishment. His part in The Maid's Tragedy was clearly that of a relief compositor: he set only seven pages of one sheet. His replacement of Compositor A in Thierry after a delay in the printing suggests that there too he stepped in to complete a job which one of the regular compositors for some reason could not finish. Perhaps, then, we should not be surprised to find that his work was rather more deliberate than that of an ordinary compositor, nor would it be unlikely that the material he set was proofread with more than ordinary care.[8]

Until all extant copies have been collated, we cannot say a great deal about the proofing of Thierry. However, the copy now part of the Thomas Pennant Barton Collection of the Boston Public Library has an extra leaf C1 inserted, the verso of which bears proof-reader's corrections (see plate).[9]

Like the proofsheet of Tichborne's A Triple Antidote, a book printed by Okes in 1609, this one shows a concern primarily for purely mechanical matters.[10] In line 10 the word "traine" is to be deleted, since it had been duplicated erroneously in the turnover (one wonders why the turnover was not deleted instead); in line 26 the compositor is evidently directed to replace or drive down a type whose shoulder was inking; in line 28 a letter (obliterated -- perhaps an "r") wrongly set between the "u" and the "ft" of "Mufter" is to be removed; and in line 37 a comma is to be inserted after the nominative of address "Martell." I think that the mark at line 37 which was cancelled and encircled was a period intended to replace the comma after "Vitry"; the proof-reader, realizing that the sentence was not concluded there but on C2, scratched it. Comparison of this page with a page in the corrected state shows that all indicated errors were repaired.

It is clear that the corrector was not meticulously concerned with the sense of what he was reading. "Mothers" in line 6 is a pretty obvious error for "mothes," and in line 10, the very line in which the first correction was made, the egregious omission of "her" before "vertuous" went unnoticed. There is no indication, of course, that the proofreader consulted copy. Although a collation might turn up other variants which could change our opinion, we are probably safe in thinking that even if Thierry was set by a relief compositor, it was proofed primarily for mechanical rather than literary errors.

Corruption in a text for which no control exists is often difficult to isolate because many of the sophistications of compositors and proofreaders must necessarily go undetected. However, in these cases an examination of the manifest errors can provide some insight into the nature of the manuscript which underlay the print. Of particular interest are errors which were apparently caused by misreading; if they abound we are probably justified in assuming illegible copy, although this assumption must be tempered by whatever knowledge we have of the compositor's performance in other books. The care with which the book under study was proofed is

also relevant since we know that readers, in addition to correcting mechanical errors, would also tamper with parts of the text which seemed to them defective. I have based what follows on an examination of the Folger Shakespeare Library's copy of Thierry Q1, which may contain one or more uncorrected sheets. A complete collation of the extant copies of Thierry would doubtless reveal some variants of which we are now unaware, but on the evidence of the Boston proof I doubt that they would affect the results of my analysis substantially.

Compositor A's one-sheet stint is too short to give us a fair sample of his work, yet in his eight pages twelve errors which can probably be attributed to him are detectable. Of these at least seven pretty clearly spring from the misreading of terminal flourishes or superscripts:

             

B1	13	Actions followes ] Actions follow
B2	25	Lyes ] Lye
B2v	16	watch ] watched or watcht
B2v	24	weeper ] weeper's
B3v	36	you ] you'r or you're
B4	10	you ] your
B4v	26	you ] your

One of the remaining errors (B3v 33 on ] of) is probably either a misreading or a foul case error; the others (at B1v 25, B2 14, B2v 4, and B2v 23) are mechanical errors — foul case, transposition, and turned letters.

In the rest of the book, all Compositor B's work, we can observe the same tendency toward minor misreading errors that we have seen in little in the first sheet. I count eighteen errors which can reasonably be placed in this category:

                           

C1v	6	mothers ] mothes
C2	14	neere ] neerer
C2v	2	doner ] owner
C3v	4	care and hidden acts ] rare and hidden arts[11]
D1v	37	mothers ] mothes
E1v	18	him, which ] him with,
F1	32	We are ] N'ere[12]
F3v	8	son's ] sum's
F4	23	I do ] I'd
G3	1-2	wishes/Outrunnes ] wishes/Outrunne
G3	13	Giue ] Giues
G4v	16	hard ] heard
H1	15	they'r ] thou'rt
H2	14	worthies ] worthes

       

I3v	11	good ] God
I4	12	preferre ] preuent
K2	24	prayers ] prayer
K3v	3	are ] haue

There are other errors which are probably attributable to misreading, foul case, or the careless substitution of one letter for another:                

D1	8	texde ] texte
D4v	3	man ] men
E2v	14	stags falls ] stags falle
F2	6	not ] nor
F2v	23	Forts ] Forte
F3	31	matcht ] marcht
G1	31	Bequeathe ] Bequeathd
I3	24	tones ] tongs

And others which most likely proceeded either from misreading or memorial failure:          

C4	5	your ] you
E1	7	you ] your
E4	28	worke ] worde
F3	21	to ] do
K3v	31	thine ] then

I cannot pretend that these lists are exhaustive; indeed, I have deliberately excluded errors of which misreading does not seem to have been a possible cause. Some of these are errors, primarily of omission, which seem to result from eyeskip or memorial failure; careless spelling; and such mechanical mistakes as transpositions of words and letters, turned letters, incorrect spacings, and obvious foul cases errors which result in nonsense words. In addition, the entire book contains a number of punctuation errors which indicate that the compositors misunderstood the sense of the material which they were setting. I have found some fifty errors of these kinds in B's stint, excluding mistakes in punctuation. Furthermore, other errors are certain to have been omitted which are concealed by the looseness of Elizabethan grammar and sentence structure or the sophistications of the compositors. That Compositor B made at least some textual changes in the light of his own judgement is suggested by a reading found at I4 32. The correct version, "Come there a band of em, i'le charge single," is rendered as "Come, there are a band of em, i'le charge single," a statement contradicted by the facts of the plot.

Although the handwriting created some minor difficulties, the copy was far more legible than some that Okes's compositors were asked to cope with. Nowhere do we find, as we do in Philaster Q1, such gibberish as "sight song" for "sigh'd, wept, sung" or "number without Probatum" for "nunne

without probation." Nor do we get such corruptions as "resterine" for "re-stem" or "Loue lines" for "loueliness" as we do in Othello Q1, a book which Miss Walker believes to have been printed from a reasonably legible manuscript.[13]

Thierry does, however, preserve certain characteristics which indicate that it is descended from a non-theatrical manuscript. I regard none of these features as conclusive in itself, but their cumulative effect is to point away from underlying prompt copy. Let us first look at the stage directions. Five are clearly permissive:

• C2v Enter Protaldie, with souldiers.
• F3v Enter Protaldy, a Lady, and Reuellers.
• H3v Enter Thierry, and Courtiers.
• K2 Enter Thierry, on a bed, with Doctors and attendants.
• K2v Enter Martell, Brunhalt, Deuitry, souldiers.

In addition, there are three directions, all at the beginnings of scenes, which use the formula "&c." apparently to indicate that the scene commences with the characters named but that others enter later in the scene:

• B4 Enter Theodoret, Martell. &c.
• C2 Enter Thierry, Brunhalt, Bawdber, Lecure. &c.
• E1 Enter T[h]ierry, Ordella, Brunhalt, Theodoret, Lecure, Bawdber. &c.

Such a notation suggests the author's rather than the book-keeper's hand, and it may be significant that a similar direction is found in Crane's transcript of Demetrius and Enanthe which is thought to have been copied from Fletcher's foul papers.[14] Stage directions do not furnish an altogether reliable means for distinguishing between theatrical and non-theatrical copy because the book-keepers often let authorial directions stand in prompt copies.[15] However, their presence combined with a complete absence of "professional" directions provides some ammunition for argument against prompt copy.

Minor confusion in the designation of characters also argues against prompt. A stage direction on D3v calls for the entry of two huntsmen, but the text and subsequent stage directions refer to them as keepers; the honest soldier is referred to variously in the text, stage directions, and speech prefixes as "Vitry" or "Devitry"; the astrologer who is impersonated by Lecure is known as both "Forte" and "Leforte." It is doubtful that all of these variations could have occurred in transmission, and they probably

would have been undesirable from the book-keeper's point of view. Hence, they provide another indication that the manuscript underlying Thierry was non-theatrical.

What evidence there is suggests that the Thierry manuscript consisted of the authors' papers, and the general freedom from confusion and serious misreading indicates that these were in the form of fair copy. Of particular interest is the frequent use of parentheses in Massinger's scenes chiefly to punctuate nominatives of address and non-restrictive modifiers. No parentheses whatever appear in the rest of the play except in one of Fletcher's scenes (III,iii) where they are used to mark several asides occurring within a longer speech (E2). Since parentheses abound in the early editions of Massinger's unaided works and are used profusely in the autograph manuscript of Believe as you List, we are justified in considering them as a distinctive Massinger characteristic.[16] Thus, we have reason to believe that the Massinger scenes of Thierry were set up either from fair copy made by the dramatist himself or from an unsophisticated scribal transcript of his papers, probably the former.

Similarly, the Fletcher and Beaumont scenes were probably printed from holograph manuscripts. Dr. Hoy points out that the linguistic divisions between the different shares are distinct: Fletcher's scenes, for example, contain twenty-one of the twenty-two ye's found in the play.[17] It is also interesting (and somewhat puzzling) that the contraction um is found only in the Fletcherian parts; the form occurs in certain of Fletcher's unaided works, yet in Philaster and A King and No King it also turns up in Beaumont's scenes. Therefore, um cannot be considered distinctly Fletcherian, and, as far as we know now, there is no reason to think that it is either scribal or compositorial. Nevertheless, Dr. Hoy suggests that the exclusive occurrence of the form in Fletcher's scenes of Thierry constitutes, for that play at least, a valid distinction between the work of the two dramatists,[18] and the fact that the distinction has been preserved in the print argues against the intervention of a scribe. Furthermore, the number of misreading errors is, I believe, about what we should expect if the compositors were working from author's fair copy; it is too high to reflect the very legible copy that would have been produced by a professional scribe and at the same time too low to reflect the illegibility of early working papers. A misassigned speech in Fletcher's share tends to support this view. The speech occurs at H1v 20 in the midst of dialogue between Ordella and Martell; it properly belongs to Martell but is prefixed Deui. It seems nearly impossible that this error could have been made by a compositor

or scribe, since the single incorrect prefix is found some ten quarto pages after Devitry's last speech and some ten before his next. The only reasonable way to account for the error is to regard it as an authorial slip, probably made during transcription.

If it is true that Beaumont's scenes as well as Fletcher's were printed from holograph, we must account for several serious errors which are found in them. The first of these is the incorrect reading in V,i (I4), which, as I have indicated above, could very easily be a compositorial sophistication. The second is two misassigned speeches in III,i (F3). Lecure enters to Brunhalt at F2v 31: in the short colloquy which follows, the two discuss the several evil plans which the Queen has set in motion. The interview concludes with the following speech by Brunhalt:

There is no way to interpret the action here but that Brunhalt and Lecure leave the stage talking to one another. However, immediately after, we find:

It is clear, as Dyce indicated, that the two speeches here assigned to Lecure must have been intended for the servant, who is brought on the stage solely for the purpose of delivering them.[19] The misassignments are undoubtedly transmission errors, probably resulting from memorial failure induced by the "Lecure" speech prefixes at the foot of F2v and the top of F3. The compositor may have been responsible, but it is not impossible that Beaumont himself made the mistake in transcribing his drafts into fair copy.

The last of these errors, a misplaced stage direction, is not so easy to explain away. At F3v 14 Protaldye and a Lady correctly enter to Bawdber and Devitry, but the stage direction — Enter Protaldy, a Lady, and Reuellers — also brings on revellers who are not required for some forty lines. The text is quite explicit on this point: at F4 23 Thierry, who has since entered, states, "Command the Reuellers in." At C2v 8 a direction that was evidently in the margin of the manuscript was placed by the compositor

one line late, but it is difficult to see how he could have been guilty of placing a direction forty lines early. Similarly, it is very unlikely that Beaumont, whose mind must have been on the working out of the comic discomfiture of Protaldye, would have introduced the revellers in the wrong place. I think, therefore, that this stage direction and possibly others in Beaumont's scenes were touched up by someone else, perhaps Fletcher. On the face of it, we should expect at least one of the collaborators to look over the whole manuscript to make minor revisions and to expunge glaring inconsistencies. In this connection it may be significant that in both Beaumont's and Fletcher's parts we find stage directions in similar phraseology, i.e., Exeunt all, but Brunhalt, Protal. Lecure, Bawdber. in II,iii (E3v), a Fletcher scene, and Exit all but Thierry. Brunhalt. in III,i (F2), a Beaumont scene. Massinger's corresponding form is Exeunt omnes, praeter Brun. Bawdber [etc.] (D1v).[20] Moreover, in the entire play there are only three stage directions in roman rather than italic type ("Reades" D1v 9, "aside" E1v 16, and "aside" E1v 26), a fact which suggests that in the copy they were written in the secretary rather than the Italian hand and therefore may have been added later than the original writing. The first of these occurs in a Massinger scene; the other two in a Fletcher scene.

To summarize, the evidence suggests that Thierry Q1 was set by a compositor whose speed was deliberate enough to cause him to adopt one-skeleton procedure, although there is nothing to suggest that his copy slowed him down very much. He may have been an apprentice or a relief compositor. The characteristics of the print indicate that he was setting from a composite fair copy written in the hands of the three collaborators which may have been reviewed and slightly revised by Fletcher. There is nothing to suggest that this manuscript had served as prompt or that it had been annotated for transcription into prompt. The print conveys the general impression of a clean job, marred by only a few accidents of transmission.

Notes