The Sad Shepherd first appears in Jonson's 1640-1 folio Works (F2(3)) , the only authoritative text of the play. It collates as follows:

2 ^o : ² R-V ⁴ [$2 signed (- ² R1)], paged ² 115-122 [123-132] 133-155, misnumbering 151 as 143, 154 as 146.

CONTENTS: R1: title page; R1v: blank; R2: Persons of the Play; R2v-V4: text of the play; V4v: blank.

The most significant textual study of the folio’s printing, that by Percy Simpson, has condemned The Sad Shepherd in particular for sloppiness (H&S, 9.96) . As this essay will show, however, the basis of much of his condemnation is misguided, and F2(3)’s text of The Sad Shepherd is printed overall with considerable skill.

Printing

Thomas Walkley published the 1640-1 folio, but legal troubles over his rights to the works meant that printing was interrupted when the volume was nearly complete (see the Print Edition, 1.clxxiii). Walkley, by his account, bought the rights to all of the works in the third volume from Sir Kenelm Digby, Jonson's literary executor, for £40, and there is little doubt that the manuscript from which the play was printed derives legitimately from Jonson’s papers.   But Walkley neglected to enter this material in the Stationers’ Register, which was to cost him a legal battle with John Benson and Andrew Crooke, who secured rights to the works in a series of Stationers' Register entries in 1639 and 1640 (see Gants and Lockwood, ‘The Printing and Publishing of Ben Jonson’s Works’ in vol. 1). Before the legal dispute became problematic, Walkley’s printer, John Dawson Jr, finished printing most of the volume, but some time before 20 January 1641 the printed sheets (probably comprising almost all of the edition) were legally attached by John Benson, with the help of John Parker, another London stationer.   Walkley was eventually able to effect the return of the attached sheets, but there was significant interruption to printing.  

Using the evidence of title-page dates – each major work in F2(3) has its own title page –Percy Simpson assumed that the interruption to printing occurred just before work began on Discoveries and The Sad Shepherd, the only two works dated 1641 (H&S, 9.96) . (They were following the tentative conclusions of Greg, 1931b). This interpretation is partially supported by the petition put in by Walkley on 20 January 1641 for the return of the confiscated, finished sheets. The date of this petition suggests that anything printed in 1641 must have been started after the interruption to printing, and probably after the attached sheets had been returned. However, bibliographical analysis suggests that much of The Sad Shepherd and nearly all of Discoveries, despite their 1641 dates, were printed before any major interruption, putting into question Simpson’s claim that those texts were hastily printed following the delay.

Headline evidence especially supports this revised chronology, showing that most of Discoveries was printed in the same skeletons as A Tale of a Tub, a work dated 1640. In F2(3), The Sad Shepherd is invariably bound after The Magnetic Lady and A Tale of a Tub (although very rarely The Sad Shepherd is missing from the volume altogether).   The grouping, which also corresponds to signature assignment (see the full collation for F2(3) in Peter Happé’s essay in this database), suggests that the plays were printed together. But they were not, as the following headline evidence suggests:

Table 1: Headline Distribution in The Magnetic Lady, A Tale of a Tub, The Sad Shepherd, and Discoveries (1640-41)

* Except the four works listed here, the entire folio was printed using just two skeletons.

Skeleton I

Headline A [rules consistent, running title changes]

Mag. Lady: B1, B3, C1, C3, D1, D2, E1, E2, F2, F4, G3, G4, H3, H4
Tub: I3v, I4, K1v, K2v, L3v, L4v, M3v, M4v, N4v, N3v   , O3v, O4v, P2v, P4v, Q1v
Disc.: M1v, M2v, N1v, N2v, O1v, P1v, P2v, Q1v, Q2v

Headline B, state 1 [rules consistent, running title changes]

Mag. Lady: A3v   , B2v, B4v, C2v, C4v, D3v, D4v, E3v, E4v, F1v, F3v, G1v, G2v, H1v, H2v
Tub: [I1v blank], I2, K3, K4, L1, L2, M1, M2, N1, N2, O1, O2, P1, P3, Q2
Disc.: M3, M4, N3, N4  

Headline B, state 2 [different rules, same running title]

Disc.: O4, P3, P4, Q3, Q4

Skeleton II

Headline C [rules consistent, running title changes]

Mag. Lady: [A1v blank], B1v, B3v, C1v, C3v, D1v, D2v, E1v, E2v, F2v, F4v, G3v, G4v, H3v, H4v
Tub: I2v, I4v, K3v, K4v, L1v, L2v, M1v, M2v, N1v, N2v, O1v   , O2v, P1v, P3v, [Q2v blank]
Disc.: M3v, M4v, N3v, N4v, O2v, O3v, O4v, P3v, P4v, Q3v, Q4v

Headline D, state 1 [rules consistent, running title changes]

Mag. Lady: A4, B2, B4, C2

Headline D, state 2 [with bottom rule flipped over]

Mag. Lady: C4, D3, D4, E3, E4, F1, F3, G1, G2, H2

Headline D, state 3 [with bottom rule flipped back over]  

Mag. Lady: H1
Tub: I1, I3

Headline E, state 1  

Tub: K1, K2

Headline E, state 2 [bottom rule flipped over, top rule changed, running title the same]  

Tub: L3, L4, M3, M4, N3, N4, O4

Headline E, state 3 [bottom rule changed]

Tub: O3, P2, P4

Headline E, state 4 [top rule changed, running title changed]

Tub: Q1
Disc.: M1, M2  

Headline E, state 5 [bottom rule changed to the one used for headline D, running title added]

Disc.: N1, N2, O1, O2, O3, P2  

Headline E, state 6 [top rule changed]

Disc.: P1, Q1, Q2

Skeleton III

Headline F

Disc.: R1, R2, R3, R4

Headline G, state 1

Disc.: R1v, R2v, R3v

Headline G, state 2 [top rule changed]

Disc.: R4v

Skeleton IV

Headline H [rules consistent, running title changes]

Sad Shep.: R2, R3, S3

Headline J [rules consistent, running title changes]

Sad Shep.: R2v, R3v, S2v

Skeleton V

Headline K

Sad Shep.: R4

Headline L

Sad Shep.: [R1v is blank]

Skeleton VI

Headline M

Sad Shep.: R4v, S1v, T2v, T3v, T4v, V1v, V3v

Headline N

Sad Shep.: [R1 has no headlines], S4, T1, T2, T3, V2, V4

Skeleton VII

Headline O

Sad Shep.: S3v, S4v, T1v, V2v, [V4v is blank]

Headline P

Sad Shep.: S1, S2, T4, V1, V3

That Magnetic Lady, Tale of a Tub, and Discoveries use the same skeletons and that these skeletons do not appear elsewhere in the folio suggests that the works were printed consecutively, and the frequent changes to the headlines that partner headlines A and C, in skeletons I & II, allow us to determine the printing order here. All four of the headlines used in the initial quires of Discoveries (A, B, C, and E) were also used in the final quires of A Tale of a Tub. (A, B, and C appear in addition in The Magnetic Lady.) The number of headline dislocations across the variant states of B and E establishes that A Tale of a Tub was printed just before Discoveries. The changes to headline E suggest that the beginning of Discoveries was printed just after the end of Tub, as state 4 of the headline appears on only three pages, one at the end of Tub and the other two at the beginning of Discoveries. Headline B provides similar evidence. Halfway through the printing of Discoveries, headline B takes on different rules, because the top rule had broken and become badly damaged. This suggests the likelihood that headline B, state 1, is earlier than headline B, state 2, and it is the earlier state alone that appears in Tub. So most of Discoveries was printed after The Magnetic Lady and Tale of a Tub, using skeletons I & II, and probably before any interruption caused by Walkley's legal dispute. Only quire R of Discoveries seems to have been printed separately, when a new skeleton, III, began to be used. It is very likely, then, that the bulk of Discoveries was printed before the legal dispute, with the change of skeletons for quire R suggesting that only those pages were printed afterwards. Dawson would have no justification for preserving the skeleton formes for an indefinite period and for works he might not even be able to continue printing, nor is there any other apparent reason for the skeletons to have been altered, so it seems that the most likely explanation for the change of skeletons is the legal interruption.   Discoveries is not bound immediately after The Magnetic Lady and A Tale of a Tub in any of the copies consulted – that position is occupied by The Sad Shepherd – which may explain why this headline evidence and printing order have not been noticed before.

The headline evidence also points to a major disruption in the printing of The Sad Shepherd, the other work with a 1641 title page. The play stands out by being printed with a separate set of skeletons, and this in a volume fairly consistent in its overall use of just four skeletons for the majority of printing. But the different skeletons do not necessarily suggest that the play was printed at a different time or place. The Sad Shepherd was certainly printed in the same printing house, as the distinctive bottom rule of headline A occurs again in headline M (showing that the first part of Discoveries and the final part of The Sad Shepherd were not printed at the same time).   Headline evidence also shows that roughly one-quarter of the way through the printing of the play, in the middle of quire S, work was interrupted when two new skeletons, VI & VII, replaced IV & V. Similar patterns are associated with the points at which the sets of skeletons I & II and IV & V stopped being employed, so work on The Sad Shepherd probably also began before the legal interruption to printing. That both Discoveries and The Sad Shepherd suffered major interruptions suggests, too, that the initial printing of both works may have been concurrent.  

In addition, headline evidence suggests that the printers held off setting the title page of The Sad Shepherd, R1 (and its conjugate page R4v), because the later skeleton VI is used to perfect that sheet, whereas skeletons IV & V had been used for the other early sheets, including one of those in quire S. So R1:R4v (containing the title page) was printed after most of quire R and some of quire S had been finished, whereas the normal working pattern in the volume (as suggested by the chronology of the minor alterations to headlines) was to finish one quire before beginning work on another. It is likely that Dawson left the title page of The Sad Shepherd blank, possibly with an awareness of Walkley's impending legal troubles. By the time the dispute had settled, the calendar year had changed, and when Dawson resumed printing he gave the title page the 1641 date. (Although the 1641-dated title page of Discoveries (M1) fits the normal headline sequence, it is possible that late in 1640, when beginning to print Discoveries, Dawson realized that the work would not be completed until the following year.)

Other evidence also points to an interruption coming at quire R of Discoveries and quire S of The Sad Shepherd. The normal composing stick width for the volume is 124mm. This applies to Discoveries (excluding the additional rule for marginalia) up to the end of quire Q. For quire R the width jumps to 128mm.   In addition, the page begins to include 51 lines instead of the usual 50 from R1v, and the catchword for Q4v does not match the first word of R1. In The Sad Shepherd, the change to page numeration also begins in quire S (to correct the overall total for that section -- see ‘General Textual Essay’), perhaps again reflecting a disruption to the flow of printing. S2v:S3 were the first pages of the play to be printed with the corrected page numbers (136 and 137 instead of 126 and 127), and the correct numbers continue to the end of the play. These types of evidence -- stick-width and line-count changes and catchword and page number shifts -- on their own would be inconclusive,   but, coupled with the skeletons evidence, they do enhance the likelihood that the major disruption to printing occurred here. It seems probable, then, that signatures R1:R4v, S1v:S4, S1:S4v, S2:S3v, and T-V of The Sad Shepherd and quire R of Discoveries were the only pages to be printed after the 1641 legal dispute, as every other page in the volume reflects a smooth and continuous use of skeletons and stick width.

H&S’s conclusions about the supposedly late and hurried printing of Discoveries and The Sad Shepherd led them to suggest that: ‘in both these works there are clear signs of hasty printing’ (9.96) , and this by a printer whose ‘press-work was abominable’ for the volume as a whole (9.106). In the end, H&S’s conclusions that all of The Sad Shepherd and Discoveries is ‘ill-printed’ because of haste after the legal dispute do not stand in the light of fuller bibliographic analysis, so there is less reason to be particularly dubious about the printer’s handling of those texts. It is true that roughly three-quarters of the sheets of The Sad Shepherd were printed after the dispute, a dispute that probably added significantly to the production costs of the volume, but it doesn't necessarily follow that an interruption to printing led to greater haste and sloppiness in the printing of subsequent sheets.

Seemingly in support of H&S’s accusations of sloppiness, The Sad Shepherd underwent remarkably little stop-press correction (see Appendix 1). Only three points across two pages were corrected in the entire play, the lowest frequency of correction by far of any work in F2(3). Furthermore, all surviving stop-press corrections emerge on pages R4 and S3v, which interestingly are sheets printed around the time of the interruption. (This further change in working patterns gives additional support to the claim that the interruption occurred here.) It seems possible that Dawson did not bother to proof-read the post-dispute sheets as thoroughly as he had done earlier sheets in the volume.

However, although we will never know what has been lost by this apparent lack of proof correction, it seems on the whole that the typesetters were remarkably accurate, leaving very few uncorrected errors or corruptions. There are a few obvious errors (see Appendix 2), but most of them are concentrated on one sheet, S2:3 inner and outer. This sheet, again, corresponds to the point at which printing was probably interrupted, so it seems likely that the setting and printing was rushed here, as H&S suppose. But uncorrected errors are infrequent elsewhere, so like Discoveries, the interruption to printing induced sloppiness only in a very localized context.   Furthermore, it is difficult to determine whether even these minor errors derive from corruption in the manuscript or corruption in the printing. (Notice, for instance, that there is no obvious printer’s error, such as an upturned letter, left uncorrected in the play.) And the relative absence of uncorrected errors and stop-press correction may in fact suggest that Dawson had more time to print the final sheets, and thus took more care in correcting them. In either case, The Sad Shepherd, despite being printed anomalously in so many other ways, cannot stand out as being less authoritative or any more subject to error than the folio as a whole. Indeed, it’s very rare to find a play text printed so well in the period.  

In the incomplete Sad Shepherd we have an important witness to Jonson's working methods for constructing plays, as the manuscript seems not to have been revised for performance or print publication. But the raw state of this work makes it difficult to apportion errors in the text to a printer, when in many cases the errors might have been present in Jonson's manuscript. The available evidence nonetheless attests to high standards of printing, with remarkably few uncorrected errors, suggesting that any limitations in the play (and these are numerous – especially in the metre) probably stem not from an inadequate printer, but from the unfortunate fact that Jonson died, and left the play unfinished.

Karolin's Song

A special word must be given to Karolin's song (1.5.65-80), because other print and manuscript witnesses survive which might be taken as authoritative. In 1640, John Benson, Walkley’s competitor to the rights for Jonson’s works, printed Q. Horatius Flaccus, His Art of Poetry. English by Ben Jonson. With Other Works of the Author, Never Printed Before (Benson12mo.). The final work in this volume, called simply ‘A Sonnet’, is Karolin’s song, with several significant variants. Benson prints ‘and’ for ‘as’ (line 70), ‘and’ for ‘or’ (74), ‘band’ for ‘brand’ (77), and ‘Will’ for ‘May’ (78). None of these readings are improvements, but they seem to attest to an independent manuscript source. Peter Beal’s work on the manuscript record identifies 14 MS exemplars, and shows that the song was probably in circulation before F2 was printed in 1640-1. Some of these manuscript copies differ slightly from the folio’s text. A common difference, found in BL Add.MS.25707, Bod. MS.Eng.poet.c.50, and Bod. MS.Mus.b.1, is the alteration of ‘Except’ in line 77 to ‘Unless’, but no other alternative readings appear widely across the manuscripts. (Some of the differences, particularly those found in Bod. MS.Eng.poet.e.97, are nonsensical.) For the purposes of this edition, the song as printed in F2 has been taken as authoritative, as there is little reason to suspect that Jonson's manuscript is corrupt here. And even though F2 has a misreading, ‘heart’ for ‘heat’ (line 70), it corresponds on the whole with earlier manuscripts. It seems likely that the song as printed (barring the single error) was intended to be performed in The Sad Shepherd, although an earlier version of the song, circulated in manuscript, may have been written for another context.

Modernization and Dialect

The play has been modernized in its spellings, stage directions, and speech headings in keeping with the Guidelines for the Cambridge Jonson. The Sad Shepherd differs from other works by Jonson in its use of dialect, however, which is notoriously inconsistent, or ‘amateurish’ in this play (H&S, 10.363). The approach of this edition has been to retain the dialect spellings for characters who usually speak in it – Maudlin, Lorel, Douce, and Scathlock – and modernize the occasional slips into dialect of the others. (The reverse is not true, however. Although all of the above-named characters slip out of dialect at points in the play, there has been no attempt to change Jonson’s ‘standard’ English to dialect.) Earine is an interesting case, as in her one scene she speaks largely in dialect (2.2), perhaps an oversight on Jonson’s part, but also perhaps in keeping with her character’s attempt to mock Lorel (see the Introduction to the play). Scarlet too is unusual, as he differs from his brother, Scathlock, in speaking without apparent dialect. Since both Earine and Scarlet speak few lines, it's difficult to tell either way, but this edition has retained their respective use of dialect or standard speech.

There are many examples of words with dialect spellings that would no longer be differentiated to our ears. In particular, the ‘o’ of words like ‘to’ and ‘do’ is rendered ‘u’ in F2 (‘tu’ and ‘du’) for Maudlin’s dialect (e.g. at 2.3.2, 21, 23, 25). These words are now generally pronounced with a long ‘u’ sound, in any case, so preserving the original spelling would have little purpose. Jonson’s distinction between ‘o’ and ‘u’ here reflects a lost pronunciation. (He pronounced these words with a short ‘o’, like that in ‘mother’ and ‘dozen’, as he explains in chapter 3 of his Grammar.) So, too, it is not always easy to distinguish when dialect is intended. Some words, like ‘merkats’ (2.2.17), ‘sate’ for ‘sat’ (2.3.45), and ‘thrid’ (2.3.46), seem dialectical, but are spelled similarly elsewhere in Jonson, including his autograph manuscripts. Although the most substantive of such modernizations have been signalled in the collation, readers wishing to give Jonson’s dialect serious consideration should consult the old-spelling edition, which preserves the dialect much better than a modernized text can.

Other Editions

Editions of the play, from the 1692 folio (F3) to the present day, have been consulted in detail (see Appendix 3) in the preparation of this edition. Of these, the most important are the earliest. F3 corrects many minor errors, and some of its readings, noted in the collation, have been adopted. Waldron, Whalley, and Gifford, although perhaps having an excessive desire to emend sense or metre, contribute a very sharp awareness of staging, and between them they add most of the necessary stage directions. W. W. Greg’s old-spelling edition stands out among those of the twentieth century, especially for its incisive commentary. And even H&S’s misguided notions about the authority of the copy-text did not greatly harm their textual work. Although they are somewhat free in emending and supplying ‘missing’ words, they signal any changes in their textual apparatus.

We have been in the enviable position of producing the first modern-spelling critical edition of The Sad Shepherd. It is hoped that our efforts will advance the critical assessment of a play whose unfinished state has been lamented by generations of readers and theatre-goers.

Appendix 1

Collation of stop-press variants in The Sad Shepherd F2(3)

COPIES COLLATED

1. Bodleian Library, Oxford University, Don.d.66 [H&S 'B2']

2. Bodleian Library, Oxford University, Gibson 520

3. British Library, 79.1.4 [H&S 'M3']

4. British Library, C.39.k.9 [H&S 'M1']

5. British Library, fol. 1482.d.15 [H&S 'M5']

6. British Library, C.28.m.12 [H&S 'M2']

7. Cambridge University Library, Brett-Smith.a.7

8. Cambridge University Library, Keynes.D.6.23 [missing R3v]

9. Cambridge University Library, Syn.4.61.20 [missing 2E3-2E3v and S3-S3v]

10. Cambridge University Library, Syn.4.64.14 [H&S 'S1']

11. Cambridge University Library, Syn.4.64.15 [H&S 'S2']

12. Cambridge University Library, Syn.4.64.12 [2E3-2E3v and S3-S3v only]

13. Christ's College, Cambridge, Rouse 8.11

14. Fitzwilliam Museum, Cambridge, STC 14754

15. Gonville and Caius College, Cambridge, L.34.3

16. King's College, Cambridge, Keynes C.5.14

17. King's College, Cambridge, C.10.6

18. Newnham College, Cambridge, Young 205b

19. Trinity College, Cambridge, Grylls 32.140

20. Trinity College, Cambridge, Grylls 32.180

21. Trinity College, Cambridge, VI.12.11

22. Trinity College, Cambridge, Capell.F.8

23. Folger Shakespeare Library, STC 14754, copy 3 [UMI microfilms of STC, reel 671]

24. David Gant's personal copy

25. Leeds University, Brotherton Library, Spec. Coll. Strong Room Engl. fol. 1640 JON

26. Leeds University, Brotherton Library, Spec. Coll. Brotherton Coll. Lt. q. JON

27. Martin Butler copy

COLLATION

R1:R4 (inner)		STATE 1	STATE 2
R4	121
20		Robin-hood?	Robin-hood
21		Wood.	Wood?
S2:S3 (outer)		STATE 1	STATE 2
S3v	138
32		su[upturned ‘r’]e	sure

DISTRIBUTION OF VARIANTS

R1:R4 (inner)

STATE 1: 10, 23, 27

STATE 2: all others

S2:S3 (outer)

STATE 1: 4, 5, 6, 8, 19, 21

STATE 2: all others

Appendix 2

Uncorrected errors in The Sad Shepherd F2(3)

R2: ‘Larine’ for ‘Earine’ (Persons, 19)

R4v: ‘Mor.’ for ‘Mar.’ (1.2.20sh)

S2: ‘Alhen’ for ‘Alken’ (1.4.79)

S2v: ‘streames’ for ‘streame’ (1.5.21)

S2v: ‘Dorks’ for ‘Docks’ (1.5.34)

S3: ‘heart’ for ‘heat’ (1.5.70)

S3: ‘world’ for ‘wold’ (1.5.103)

S3: ‘Kar.’ for ‘Lio.’ (1.5.108sh)

S3: ‘Lookes’ for ‘Looke’ (1.5.95)

S3v: ‘so so’ for ‘so’ (1.6.5)

T1: ‘distate’ for ‘distaste’ (2.1.15)  

T2: ‘thy felfe’ for ‘thy selfe’ (2.2.45)

Appendix 3

Editions Consulted

F3	Works of Ben Jonson (1692)
Benson12mo.	Q. Horatius Flaccus, His Art of Poetry. English by Ben Jonson. With Other Works of the Author, Never Printed Before (1640)
1716	Works of Ben Jonson, 6 vols. (1716)
Wh	Works of Ben Jonson, ed. Peter Whalley, 7 vols. (1756)
Waldron	The Sad Shepherd, ed. F. G. Waldron (1783)
	The Sad Shepherd, ed. F. G. Waldron (1783); with Waldron's further MS notes on the play, mostly from Gifford, BL C.45.c.4
	The Sad Shepherd, ed. F. G. Waldron (1783); with Waldron's MS corrections and additions, BL 643.g.15
G	Works of Ben Jonson, ed. William Gifford, 9 vols. (1816)
Cornwall	Works of Ben Jonson, ed. Barry Cornwall (1838)
Gifford (1853)	Works of Ben Jonson, ed. William Gifford (1853)
Gifford/C	Works of Ben Jonson, ed. William Gifford, with Francis Cunningham, 9 vols. (1875)
Greg	The Sad Shepherd, ed. W. W. Greg (1905)
Schelling	Ben Jonson: Complete Plays, ed. F. E. Schelling, 2 vols. (1910)
Schelling2	Typical Elizabethan Plays, ed. F. E. Schelling (1926)
Potts	The Sad Shepherd (Cambridge Plain Texts), ed. L. J. Potts (Cambridge, 1929)
Porter	The Sad Shepherd . . . now completed by Alan Porter, with a foreword, ed. Henry Noble MacCracken and Richard Albert Edward Brooks (New York, 1944); with modernisation of the text of H&S
Baskerville	Elizabethan and Stuart Plays, ed. Charles Read Baskerville (New York, 1934)
H&S	Ben Jonson, ed. C. H. Herford and Percy and Evelyn Simpon, 11 vols. (Oxford, 1925-52)
Adams	Ben Jonson: Plays and Masques, ed. Robert M. Adams (New York, 1979)