The history of the early printing of Sejanus cannot be divorced from its interpretation. This is partly because, as will be seen, the typeface and layout of the first edition, the quarto of 1605 (Q), themselves make a highly significant statement, becoming in that edition part of the play’s ‘meaning’; and partly because the text of the play as it was printed, whether in Q or the 1616 folio version (F1) , differs from the play that was acted in 1603-4. Jonson seems unequivocal about this in his epistle To the Readers in Q:

Lastly, I would inform you that this book, in all numbers, is not the same with that which was acted on the public stage, wherein a second pen had good share; in place of which I have rather chosen to put weaker (and no doubt less pleasing) of mine own, than to defraud so happy a genius of his right by my loathed usurpation.

The extent of revision, however, apart from the addition of marginal notes and prefatory material, may not be quite as substantial as ‘good share’ suggests. ‘In all numbers’ is derived from Lat. omnibus numeris, ‘all parts’, ‘all details’ (cf. Pliny, librum omnibus numeris absolutum’, Epistolae, 9.38; Cicero, ‘omnibus suis numeris et partibus’, De natura deorum, 2.37). Jonson used Pliny’s phrase ‘absolute in all numbers’ to mean something like ‘perfect in every respect’ in the title to Und. 84 (of Kenelm Digby), in Epigr. 79 (of Lady Rutland), and in his letter to Cecil of November 1605; the use in Mercury Vindicated, 157-8, is very similar, while in Discoveries Sir Nicholas Bacon likewise ‘filled up all numbers’ (657; cf. also Dedication to Epigrams, 16 and New Inn, Epilogue, 6). Heminges and Condell use the phrase similarly to attest the accuracy of their text in the epistle to the Shakespeare first folio (1623), where the plays are ‘absolute in their numbers, as he conceived them’. It is therefore likely to mean here that the text of Q is not the same ‘in all details’ as that performed, and this implication that the second pen’s part was not very large is supported by two contemporary witnesses. One is ‘Ev.B’, (possibly Everard Buckworth) whose poem about the 1603 performance of Sejanus refers to ‘the author’ and ‘his grave and learned toil’ in the singular. This, however, was probably written up to two years later, and intended for the emphatically single-author publication of Q. The entry in the Stationers’ Register for 2 Nov. 1604 is less compromised. Stationers’ Register entries usually record the titles of the manuscripts actually submitted, and the copy licensed in this case was ‘a booke called the tragedie of Seianus written by Beniamin Iohnson’, rather than ‘SEIANUS his Fall’, the title under which Q was published. This suggests that the text that was approved in 1604 may have been closer to the 1603 version passed by Tilney, than to the one finally printed. If so, the ascription of authorship to Jonson alone is significant, the more so since the licenser, Zacharias Pasfield, was an ‘acquaintance’ of his, and was requested by him to act as a spiritual adviser over his ‘scruple of conscience’ as a recusant in 1606 (Life Records, 32). Pasfield, who had previously licensed Every Man In, Cynthia’s Revels, and Poetaster, would probably have known something of the origins of the fourth Jonson manuscript he read. Friendship may also have disposed him to be uncensorious in his assessment of the play.

The firm statement about the second pen may, therefore, be defensive, an attempt to distance the printed play from one which when performed had caused offence, and which was being prepared for the press at a period close to that in which Jonson and Chapman were in prison for their part in Eastward Ho!, printed in the same year by the same printer. But whatever allowance is made for deviousness on Jonson’s part, he is explicit enough about the editing out of a second author’s share to leave no serious doubt that the text we have does differ to some extent from that which was performed. The ‘second Pen’, inevitably identified as Shakespeare’s in the eighteenth century – ‘Shakespeare himself assisted Ben Jonson in his Sejanus, as it was originally written’ (Edmond Malone, Plays and Poems of Shakespeare, 1821, 1.356 , and cf. Whalley, 3.130 ) – is now usually assumed to have been that of George Chapman, who seems to refer to a collaboration (though not necessarily by him) in which ‘other fords mix[ed] with his modest course’ (In Sejanum, 38). There is, however, no firm evidence for Chapman having written any of the 1603 version. Both the identity of the collaborator and the extent of his contribution are beyond recovery, and lost with them is the play that was acted at the Globe and at Whitehall. (For a plausible defence of Shakespeare as the second pen, set by the company to make the play more stageable, see Barton (1984) , 93-4. Gifford’s (3.8) suggestion of Fletcher is unconvincing, mainly because his career as a dramatist had not started by 1603.)

Thus no editor can hope to reconstruct a text that takes the reader or director closer to the original stage version of Sejanus. The choice instead is between a text based on the self-consciously literary version which Jonson chose to publish in 1605, with its unique and copious marginal notes and its distinctive typefaces, and the revised version of that text which Stansby printed in 1616, with the notes omitted, and with type, layout, punctuation and spelling standardised to accord with the style of the other plays in the collected Works. In a new edition of Jonson’s complete works it is inevitable that a similar degree of standardisation will again be imposed on Jonson’s text. An edition of Sejanus based closely on F1 would, therefore, involve two degrees of standardisation, two further moves away from the copy Jonson gave to Eld in 1605, and from the literary product he then conceived. Since, as will be seen, the revision of Sejanus for F1 was less painstaking than most subsequent editors have assumed, there are few arguments for preferring it as copy-text. This edition is therefore based on the 1605 quarto, reproducing the distinctive features of that edition as far as possible within the confines of standardisation and modernising. The most obvious departures from the appearance of Eld’s version are the move of the marginalia to the foot of the page, and the loss of upper case type used as a visual sign of Roman formal discourse.