This is a progress report on efforts to edit Hobbes’s Elements of Law for the Clarendon series of his works. The Elements was first printed in 1650 in two separate volumes, with the titles Humane Nature and De Corpore Politico, but these printings (Wing H2242 and H2219) were not approved by Hobbes, and lack textual authority. In 1650 it was well known that Hobbes planned to bring out a book on human nature – or de homine (on man) – and another on body – de corpore (which an inattentive reader could easily confuse with De Corpore Politico) – so it was tempting for enterprising booksellers to try to exploit the success of his De Cive (1642; 1647) by recycling under misleading names his manuscript Elements of Law of 1640. To find what Hobbes wrote in the Elements we need to look at the surviving manuscripts of it.

Currently, the standard edition of the Elements is the one edited (from manuscripts) by Ferdinand Tönnies (1855–1936). Tönnies was a German scholar who spent time in England in 1878, working on manuscripts of the Elements. He found five of these in the British Museum (or British Library, as we would say), and a sixth at Hardwick Hall in Derbyshire, one of the residences of the Cavendish Dukes of Devonshire. Tönnies returned to England in 1884, and signed a contract with an English publishing house for the publication of his edition of the book. However, the publisher withdrew from the contract two years later, and it was only in 1889 that the book appeared in print, from the press of Simpkin, Marshall, and Co., at London.¹ It was published again in 1928, this time at Cambridge, by Cambridge University Press, in the series Cambridge English Classics. In a brief note to this printing, dated from Kiel in November 1927, Tönnies stated that a “great part” of the 1889 edition had been “destroyed in a warehouse fire,” and asserted that “in the present impression the work is reprinted without change” except that the footnotes had been turned into endnotes. In fact this was incorrect, and the 1928 printing introduced new errors (as well as repeating old ones). For example, in 1:9:13 (part 1, chapter 9, section 13)² Hobbes says that we laugh at jests of which the wit consists of “some absurdity of another.” This is correctly printed in the 1889 text, but in the 1928 version it is replaced by the weaker and mistaken “some absurdity or another” (which is also the reading of J. A. C. Gaskin’s edition). In 1969 Frank Cass and Company brought out (at London and Totowa, N.J.) what was proclaimed to be a second edition of Tönnies’s book. It had a new introduction by M.M. Goldsmith, but otherwise was not new at all, since it was (perhaps fortunately) a facsimile of the original 1889 version.

Tönnies said that the manuscript which “holds the first place in point of authority” is Harley 4235 (in the British Library, and now conveniently available at <http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Harley_MS_4235 >). In this manuscript, Hobbes has signed the dedication, and made many additions and corrections in his own hand. Tönnies also emphasized the importance of the Hardwick manuscript, as it then was. It has since been transferred to another Cavendish residence, namely Chatsworth, where there are now three manuscripts of the Elements of Law – Chatsworth A2A, A2B, and A2C. Textual evidence makes it clear that Tönnies consulted A2B, which has many corrections and additions in Hobbes’s hand, and in which the entire dedication is in his writing. Tönnies suggested that “this is the copy kept by the author in his own possession” (1889, p. ix). Perhaps that is so, or maybe he kept it at the residence of one of his Cavendish patrons while Harley 4235 was at a different such residence (his main patrons around 1640 were his employer the Earl of Devonshire and the book’s dedicatee the Earl of Newcastle, much of whose library was later inherited by the Harleys). If Chatsworth A2B was his personal master-copy, we might expect that its readings would have been adopted in other manuscripts, such as Harley 4235. But in fact the texts of the two manuscripts differ in many (generally small) ways. Some of the other manuscripts agree in most of their readings with Harley 4235, but others follow Chatsworth A2B. In Tönnies’s day, these other manuscripts were four in number, namely Harley 4236, Harley 1325, Harley 6858 (which contains only the first thirteen chapters of the book), and Egerton 2005. Of these, the first three largely agree with Harley 4235, while the Egerton manuscript is close to Chatsworth A2B. Since Tönnies’s time, five more manuscripts have turned up. Two are now at Chatsworth, where they are named A2A and A2C. Another is in Derby Local Studies and Family History Library, and there is also one in the Thomas Fisher Rare Book Library at Toronto University. Yet another is in the Pepys Library at Magdalene College, Cambridge. The Toronto and Magdalene manuscripts, and the two at Chatsworth, are textually close to A2B, while the Derby text has readings that link it to Harley 4235.

The Pepys manuscript at Magdalene is notable for a passage of over 500 words that is missing from all the other manuscripts. In Chatsworth A2B at el1:14:2 there is an insertion mark in the body of the text, with the letter ‘C’ directly above it, and another insertion mark in the margin. So it seems that there once existed a document named ‘C’ with text for insertion into A2B. At just this point in the Pepys manuscript, the passage appears. It attacks the notion that people are by nature sociable, and compares humans with mastiff dogs. It is noteworthy that though this manuscript was not utilized by Hobbes scholars until the twenty-first century, it was listed in a printed catalogue apparently before any of the others. For it was recorded by Edward Bernard in his Catalogi Librorum Manuscriptorum Angliae et Hiberniae, Oxford 1697; this includes a catalogue of Samuel Pepys’s library, which lists “Mr. Hobbes’s Elements of Law Natural and Political” as item 6738. 23 (sig. 3B2b, p. 208). Montague Rhodes James, perhaps best known as a writer of fine ghost stories, also listed it, in his A Descriptive Catalogue of the Library of Samuel Pepys: Part Three – Medieval manuscripts, London, Sidgwick and Jackson, 1923; no. 2099, item 7 (p. 72).

Tönnies (1889, p. ix) contended that “the vulgar text” – that is to say, the readings in Molesworth’s edition and earlier printings, stretching back to 1650 – was close to the Hardwick version (Chatsworth A2B) and to Egerton 2005, while there was “much agreement” between the other manuscripts (the four relevant Harley items in the British Library) and his own edition. In fact, the manuscript that was closest to Humane Nature was the Pepys/ Magdalene one, while De Corpore Politico most nearly resembled the Derby text.

One finely produced manuscript was Harley 4236 (<http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Harley_MS_4236>), which has readings that are mostly the same as those in 4235, but sometimes includes emendations that are linked to Chatsworth A2B and the associated manuscripts. One of a number of examples is at 1:17:7. Harley 4235 says “An Arbitratour therefore, or he that is Iudge is trusted by the parties to any Controversie, to determine the same,” and this is also the reading of Harley 1325, the Derby manuscript, and the first edition of the printed book. However, in Chatsworth A2B (and the other manuscripts of its group) this passage begins “An Arbitratour therfore, or Iudge is he that is trusted.” In Harley 4236, the original reading was the same as in 4235, but a deletion and an insertion changed it in favor of that of Chatsworth A2B. On many other occasions, however, it retains the text of 4235 though Chatsworth A2B has different words.

Often, the text of Chatsworth A2B disagrees with that of that of Harley 4235. At 1:13:11 it disagrees with itself. There, at the end of the chapter, is a passage in Hobbes’s hand, saying that “Silence in him, who beleeueth that the same shall be taken for a signe of his consent, is a signe thereof indeed” since “the labour of speaking so much as to declare the same, is so little, as it is to be presumed [that he consented deleted] he would haue done it.” In the margin, Hobbes has written a slightly different version of the same thing, with “Silence, in them that think it will be soe taken” and then “soe little labour [is deleted] being requird to say No.” He has deleted neither passage.

The evidence suggests that Chatsworth A2B was the earlier version of the two main manuscripts, but that Hobbes later continued to revise both it and Harley 4235, which is in general the more polished of the two. He never definitively established the text of the book, at least in the surviving manuscripts. After signing the dedications (dated May 9 1640) he continued to make changes in both copies, but then a few months later left for France and embarked on new projects, including De Cive. It makes sense for a modern edition to include variants from all the surviving manuscripts and of the versions printed in Hobbes’s lifetime, but the best texts are usually to be found in Chatsworth A2B and Harley 4235. Moreover, Hobbes’s own additions and deletions in these manuscripts tell us much about the development of his thought.

Editing is a fine way to learn how easy it is to duplicate old mistakes. Errors which have persistently reappeared in printings of the Elements are often the work of nameless seventeenth-century compositors, and have sometimes proved remarkably tenacious. A good example occurs in 1:8:2, where Hobbes talks about harmony and the pleasures of hearing. Harley 4235 cites Galileo as showing that “two sounds differing a fift, delight the eare by an equality of strikinge, after two inequalityes” as “the higher note striketh the eare thrice while the other striketh but twice.” Talking about musical intervals, Galileo remarked that “The first and most pleasing consonance is, therefore, the octave since, for every pulse given to the tympanum by the lower string, the sharp string delivers two … The fifth is also a pleasing interval since for every two vibrations of the lower string the upper one gives three.”³ Clearly, Hobbes’s “fift” is Galileo’s “fifth.” What about Galileo’s “octave”? In Harley 4235 Hobbes says that “Sounds that differ in an Eight please by inequality, and equality alternate, that is to say the higher note striketh the eare twice, for one stroke of the other whereby they strike together euery second tyme, as is well proved by Gallileo.” Hobbes’s “an Eight” (spelled “an Eigth” in Chatsworth A2B) is an octave. In the Magdalene manuscript this became “any Eight” and in the first printed edition (Wing H2242) the reading was “any eighth.” Then, in the second edition (Wing H2243) a new reading with a spectacular future was introduced: “any height.” “Sounds that differ in any height” may not mean very much, but it has established itself as the correct reading for more than three and a half centuries. There is something sad about challenging that orthodoxy.