Military Report on the Sinai Peninsula
Towards 'An English Fourth'
Seven Pillars of Wisdom, The Complete 1922 Text
'The Mint' and Later Writings About Service Life
Boats for the R.A.F. 1929-1935
reports and correspondence
The Forest Giant
Correspondence with Bernard and Charlotte Shaw
Correspondence with E. M. Forster and F.L. Lucas
More Correspondence with Writers
Correspondence with Edward and David Garnett
Correspondence with Henry Williamson
Translating the Bruce Rogers 'Odyssey'
Correspondence with the Political Elite 1922-1935
Check for programme updates on our News page
T. E. Lawrence, Seven Pillars of Wisdom, The Complete 1922 Text
A 'best text' edited by Jeremy Wilson from the manuscript and Lawrence's
corrected copy of the Oxford Times proof printing
Castle Hill Press 1997
Edition of 752 numbered sets
This edition is out of print
ABOUT THIS BOOK
This text of Seven Pillars of Wisdom, completed in 1922 but unpublished until 1997, is a third longer than the version issued after Lawrence's death. Despite its importance, it remained almost unseen for seventy-five years.
What follows is based on the prospectus issued by Castle Hill Press in 1997.
- You can also read a detailed history of the origins of the Castle Hill Press edition.
- Following further research in 2004, Jeremy Wilson wrote an essay titled 'Seven Pillars of Wisdom, Triumph and Tragedy'. You can read this on the T.E. Lawrence Studies website.
An amendment by Lawrence to Chapter 26 on a copy of the Oxford Times proof printing.
History of Seven Pillars texts
by Jeremy Wilson
In 1922 T. E. Lawrence finished work on the third draft of Seven Pillars of Wisdom. The first, written during 1919, had been incomplete when it was stolen from him at Reading station. The second had been a hurried re-write, dashed-off from memory. Using this as a basis, Lawrence worked for many months on a third version, which he corrected and polished. There were probably intermediate drafts of most chapters, because what finally emerged was a fair-copy manuscript. This, the earliest surviving complete text, is nearly 84,000 words longer than the version he later issued to subscribers.
Intending to produce his own private-press edition, Lawrence ordered colour collotypes of four of the projected illustrations from the lithographers Whittingham & Griggs. But then, while he was still working on the new manuscript, he began to drift into an unbalanced state of mind. In the belief that he could heighten his creative powers, he had deliberately subjected himself to near starvation and lack of sleep. He used to work at the book for many hours at a stretch. In the process, he had fallen victim to acute depression and self-doubt. He later wrote: 'I nearly went off my head this spring, heaving at that beastly book of mine.'
In this physically and mentally exhausted state, he became obsessed with the idea that the text was not good enough. His ambition was to write a classic of world literature - something that would stand alongside Moby Dick and The Brothers Karamazov.
But Seven Pillars had 'grown too long and shapeless'. By the time he reached the end of the narrative, the seven component 'books' he had planned (an echo of 'seven pillars') had grown to ten.
He decided that he must let his mind 'lie fallow for a time', after which he would attempt yet another revision. The medicine he prescribed for himself was enlistment in the ranks of the RAF under a false name.
Determined not to lose the new draft like its predecessor, he had Seven Pillars typeset by the Oxford Times newspaper, this being cheaper than making typewritten copies. Over a period of months, eight uncorrected proofs were printed as single pages on a proofing press. To make it impossible for anyone else to assemble the book, he sent the chapters for typesetting un-numbered and in random order, keeping back the most sensational material until the very end.
When the Oxford Times pages were delivered, he numbered and corrected a master set, making many slight amendments in the process. He repeated the corrections on four further sets of pages and had the five copies bound. They were then circulated for comment among friends, wartime colleagues, and literary critics. He warned that, while the text was largely accurate, the punctuation was 'the printer's own'.
It was this 1922 text which convinced readers that Lawrence had written a masterpiece. During the next few months Jonathan Cape, Constable, and other London publishers were keen to publish it in extenso. Failing that, Edward Garnett offered to make a popular abridgement. In January 1923 Lawrence came close to accepting Cape's proposals, but he withdrew when press attention led to his dismissal from the RAF.
In March 1923 Lawrence re-enlisted, this time as a private in the Tank Corps. His mental state, far from improving, was causing his friends serious concern. They sought ways to help him and, at the end of that year, persuaded him to undertake an abridged subscription edition of Seven Pillars. The private-press production would be a project to occupy his mind. In the interim, the bank held copyright of the 1922 text as security.
The 1922 text and the subscribers' abridgement compared
Lawrence later claimed that, with a few exceptions, his sole criterion for revision was literary. The whole truth was less simple. He found the task far more difficult than he had expected, because the experiences of the book had 'gone remote' from him. Worse still, his writing style had changed, reflecting his admiration for the contemporary writers whose work he had been reading avidly since the war.
His disappointment with Seven Pillars intensified but, in place of the wholesale re-writing that he came to feel was needed, he found he could do little more than abridge it. He made cuts almost everywhere, while generally tightening up the prose. Overall, the final subscribers' text which he issued in 1926 was nearly 84,000 words shorter than the 1922 original. It is this 1926 abridgement that was published in a trade edition after his death in 1935.
Among those who have read both the 1922 and 1926 texts, opinion about their relative merit has been equivocal. Robert Graves preferred the former: 'the greater looseness of the writing makes it easier to read. From a critical point of view no doubt the revised version is better. It is impossible that a man like Lawrence should spend four years on polishing the text without improving it, but the nervous rigour that the revised book gave me has seemingly dulled my critical judgement.'
E. M. Forster, who made suggestions to Lawrence during the final revision and therefore had reason to prefer it, compared the two texts in 1931 and concluded: 'I had to admit that the sentences in the revision were more concise and distinguished and showed a superior sense for the functions, and incidentally for the etymology, of the words employed in them. But the relation between the sentences seemed to me a little impaired: the correction, though logical, wasn't always easy.' Bernard Shaw, for his part, deplored the time Lawrence spent titivating a book that was already magnificent. He nevertheless took the opportunity of the revision to insist that Lawrence toned down the more obvious libels.
As long as the 1922 text remained unpublished, few other critics were in a position to judge. My own opinion, after spending a great deal of time with the two texts, is that the 1922 version is more readable - a conclusion implied in the comments made by both Graves and Forster. Setting aside the omissions in the 1926 abridgement, which are discussed below, I think that a reader who knows the 1926 text well will find the 1922 version familiar yet easier. The contrived 'literary' tone which can be intrusive in the 1926 revision is less often present. As a result, the text generally flows more naturally, as in Lawrence's letters.
The special value of the 1922 text
Seven Pillars is also important as autobiography and as history. In both these fields the 1922 text is, without question, superior to that of 1926. In the process of 'literary' abridgement, Lawrence cut out numerous personal reflections, some of which were important. Examples are the discussion of his motives for the secret ride to Damascus in the summer of 1917, his confession that the flogging at Deraa left him with a masochistic longing 'like a moth towards its flame', and his recollection of this event a few weeks later when he was present at Allenby's official entry into Jerusalem.
The historical record, likewise, often fell victim to abridgement. Scholars have been perplexed by a published narrative which, because of the cuts, does not always account for Lawrence's time or seem to square with independent records.
Worse still, the frustrations and abandoned plans of 1917-18 were largely suppressed in the 1926 text because, Lawrence said, some readers of the 1922 version commented that the dramatically 'flat' period before the final advance to Damascus seemed too long. In the 1926 revision, Lawrence cut the last four Books substantially.
While working on the authorised biography I compared the two texts of Seven Pillars with each other and with unpublished contemporary documents. I repeatedly found significant material in the 1922 version that was absent from the 1926 abridgement. Lawrence knew this, and ensured that the 1922 text would survive by presenting the manuscript to the Bodleian Library.
Why was the 1922 text not published long ago?
There was no question of publishing the 1922 text in Lawrence's lifetime. After he had issued the subscribers' abridgement in 1926 and Revolt in the Desert in 1927, he was determined that nothing more should appear.
When he died, Jonathan Cape, his English publisher, rapidly announced a reprint of Revolt in the Desert. A.W. Lawrence, as literary executor, was very keen to prevent that. He therefore allowed Cape to publish Seven Pillars.
The only version that was immediately available in a usable form was the 1926 subscribers' abridgement. In a miracle of rapid book-production, Cape's handsome trade edition was published on 29 July, only six weeks after Lawrence's death.
Nevertheless, the 1922 text was not entirely overlooked. In his Preface to the 1935 Cape and Doubleday editions, A.W. Lawrence wrote: "The remaining copies of the Oxford text printed in 1922 are still in existence, but it will not be made public for at least ten years, and then only in a limited edition."
Not all the reviewers were happy with this outcome. The one who put it most bluntly was the Arabian explorer H.St.J.B. Philby, who wrote in the Sunday Times: 'It is this [the 1922 Text] we want rather than the subscribers' edition which we are now given.' More diplomatically, E.M. Forster wrote: 'A later MS. forms the basis of the so-called "Oxford" edition of 1922. I have read through this edition twice. One always tends to overpraise a long book, because one has got through it; still the "Oxford" is in the judgment of several critics even superior to the version offered now, and it is good news that a reprint of it may eventually be made.'
In the event, the release of the subscribers' abridgement proved to be a very real obstacle to publication of the longer text. The reason is that Cape and Doubleday acquired exclusive rights to the subscribers' abridgement in 1935. Because there is substantial duplicatioin force, the 1922 Text could only be published with their agreement.
Cape and Doubleday went to the considerable expense of printing the subscribers' abridgement of Seven Pillars. Although it is an abridgement, 250,000 words is by any standards a long book. Luckily, their gamble paid off handsomely. Their version of Seven Pillars sold extremely well for many years. Then, in the 1960s, its value was further enhanced as a result of David Lean's film Lawrence of Arabia.
In that happy position, why should they agree to take on the huge expense of printing a 334,500-word text? If they did so, it could only be a direct commercial threat to the highly profitable investment they had already made. When A.W. Lawrence and others put forward the suggestion, they concluded unsurprisingly that such an edition was unjustified.
Their objection lost its force in 1986, at least in Britain, when copyright on the 1935 text expired. Prior to that, A.W. Lawrence and I had on several occasions discussed the idea of publishing the 1922 text. In the late 1980s, having completed the authorised biography, I began the editorial work.
In the last months of his life, however, A. W. Lawrence decided to postpone publication. A nagging concern was that a British soldier who had been involved in a punishable incident described in the book might still be alive and identifiable by other survivors.
Finally, by permission of the Seven Pillars of Wisdom Trust, the complete 1922 text was issued in 1997, in a limited edition of 752 copies. A one-volume Library Edition was published in 2003.
Since Lawrence's death, Seven Pillars has been accepted as a modern classic. The 1926 revision has sold well over a million copies and there are translations into many languages (no fewer than four French translations have been issued).
Release of the fuller 1922 text, almost unseen for seventy-five years, was the most important event in the history of T. E. Lawrence publishing since the issue of The Mint in 1955.
How the new edition was prepared
My aim in preparing the 1922 Text for publication was to issue it in its 'best' form. The initial typesetting was based on Lawrence's corrected Oxford Times proof. This was then checked against the Bodleian Library manuscript to correct the printer's many transcription errors and 'house style' punctuation changes. I also restored printer's omissions not corrected by Lawrence, two of which are of considerable interest. At the end of this process, the text published by Castle Hill Press was considerably more accurate than the text of the 1922 Oxford Times proofs.
However, I found that in addition to straight corrections, Lawrence had made some hundreds of amendments on his copy of the printed text. These were clearly intended as textual improvements, and I decided to retain them. That produced a version which incorporates his final revisions to the 1922 draft, and is superior both to the manuscript and to the surviving Oxford Times proofs. Finally, the book was lightly copy-edited to remove the grammatical and punctuation errors inevitable in a manuscript, and also to eliminate Lawrence's random variations in capitalisation and Arabic transliteration.
- Three volumes, numbered in Volume II. Spine height: 28.8cm. Trimmed page size 282 x 200mm.
- 650 sets (of 752), numbered 101-750.
Vols I and II, text
- Vol. I: Introduction and Books I-V, 456 pages.
- Vol. II: Books VI-X, editorial notes, 456 pages.
Typeset by Castle Hill Press in Caslon, Lawrence's preferred typeface. Printed by Cambridge University Press on 80g.s.m. Supreme Bookwove, a high-quality acid-free off-white book paper, and bound by The Fine Bindery in quarter cream Rohalbleinen cloth with grey Fabriano Tiziano boards. Top edges stained, coloured endpapers, head and tail bands. Issued in dust jackets.
- Vol. III: Illustrations
Part I (in colour): Forty-one Seven Pillars Portraits reproduced full-page - twenty-six are in full colour - as in Lawrence's 1926 subscribers' edition.
Part II: eighty-eight war photographs, mainly from the collection Lawrence formed and presented to the Imperial War Museum.
Originally printed in colour and duotone by T.G. Hostench SA of Barcelona on 115gsm matt art, and bound in cream Rohalbleinen cloth, with limp boards. This Spanish printing was found to be defective and relatively few copies were issued. Replacement copies were printed in England by The Burlington Press. Issued in dust jacket.
The three volumes were issued in a cloth-covered slip-case.
Of the 650 sets quarter-cloth sets, about 75 included an extra matching portfolio containing a proof set of the Seven Pillars portraits, printed on one side only of a heavy paper. These proofs were printed by the original Spanish printers of the book of illustrations, which was reprinted in England when the first printing was found to be defective. The same set of proofs (in a different-coloured portfolio) accompanied the full-leather 'specials'.
The three volumes and the portfolio of proof portraits were issued in a cloth-covered slip-case.
These extra-illustrated sets were issued to subscribers only.
80 sets (of 752), numbered 21-100, issued in an off-white cloth-bound slipcase
Three volumes, numbered in Volume II. Spine height: 28.8cm. Trimmed page size 282 x 200mm.
Vols I and I:
Vol. I: Introduction and Books I-V, 456 pages
Vol. II: Books VI-X, editorial notes, 456 pages
Typeset by Castle Hill Press in Caslon, Lawrence's preferred typeface. Printed by Cambridge University Press on 80g.s.m. Supreme Bookwove, a high-quality acid-free off-white book paper, and bound by The Fine Bindery in full Oxford-blue Harmatan goatskin. Hand-marbled endpapers by Ann Muir, using an unusual double-marbling process. Bound in full Oxford-blue goatskin, head and tail bands.
Vol. III: Illustrations and parallel text of the introductory book
- Illustrations, 152 pages:
Part I (in colour) The Seven Pillars Portraits
Part II a selection from Lawrence's collection of war photographs.
Printed by the Burlington Press, bound in quarter blue goatskin with off-white cloth sides, bound with:
- Parallel text of the introductory book:
An 88-page parallel text showing two versions of the Introductory Book of Seven Pillars. The left-hand pages reproduce Lawrence's revised text as sent to Bernard and Charlotte Shaw on 27 September 1924, and right-hand pages show the final text of the subscribers' abridgement. The parallel text displays exactly what changes were made as a result of advice from Bernard Shaw and others, and settles once and for all the argument about the possible scale of Shaw's alterations to Seven Pillars.
Portfolio of proof portraits:
The three volumes are accompanied by a cloth-bound portfolio of proofs of the Seven Pillars portraits, interleaved with Japanese paper.
Full-goatskin issue (left) with a full-goatskin copy of the 2003 edition (right)
Hand marbled endpaper by Ann Muir
We are sometimes aware of copies available privately or in the antiquarian book trade, so if you are looking for a copy it is worth contacting us.
20 sets (of 752), comprising six volumes including a complete parallel printing of the 1922 and 1926 texts in two volumes
The Castle Hill Press parallel text is surely the most fascinating of all published editions of Seven Pillars of Wisdom.
Throughout the design and production of the 1922 Seven Pillars, our decisions were strongly influenced by one idea: that the most important thing was the text. For this reason we took pains to reproduce accurately the wording of the manuscript, as modified by Lawrence's later amendments in the 'Oxford' proof. Also, reflecting Lawrence's own passion for good typography, we applied typesetting rules that most trade publishing houses and many private presses seem to have forgotten: no paragraph widows shorter than sixteen characters; no identical words at the beginnings or ends of successive lines; no hyphenation of proper nouns; no paragraphs beginning on the last line of a page, no fudged pairs of 'short pages', and so on.
The parallel text
In this spirit, we chose to accompany these twenty sets by something both remarkable and extremely interesting - two additional volumes containing complete texts of the 1922 and 1926 versions, typeset side-by-side in double column. Our aim in this setting was to align the beginning of each sentence that exists in both the texts. Readers can see at a glance exactly what was omitted, what was revised, what was moved, and so on.
Our edition of this 1000-page parallel text consisted of only 37 sets. Twenty sets, bound in quarter Harmatan goatskin, accompanied the twenty 'extra-special' copies of Seven Pillars, and two sets, not for sale, accompanied copies 'A' and 'B'.
Of the remaining fifteen sets, six went to UK copyright-deposit libraries, one to the copyright owner, two to the editor, and one was retained by the press.
These twenty sets of Seven Pillars of Wisdom, the Complete 1922 Text, comprise six volumes, all in the same format, contained in two solander boxes.
Box 1: The Caslon setting
The two text volumes printed by Cambridge University Press on 100 gsm Supreme Book Wove, and bound by The Fine Bindery in full Harmatan goatskin, to an inlay and onlay design by Glenn Bartley specially commissioned for the edition. All edges gilt; hand-sewn head and tail bands, leather joints and suede doublures.
The companion volume of illustrations is bound in full black goatskin.
An internal box contains an interleaved proof set of the Seven Pillars portraits, printed on one side of the paper only. This is one of 250 numbered proof sets from the first printing of the volume of illustrations.
Box 2: The Parallel Texts
The 1922 and 1926 texts: two volumes, printed on 80 gsm Supreme Book Wove. The typeface is Times New Roman which, being designed specifically for newspapers, is better suited than Caslon to two-column setting. Hand-bound by The Fine Bindery in quarter brown Harmatan goatskin, with brown cloth sides and hand-marbled endpapers by Ann Muir. Top edges gilt.
The eight chapters of the Introductory Book of Seven Pillars, in parallel 1924/1926 text, are quarter-bound in matching brown Harmatan goatskin with brown cloth sides.
Two sets of the edition, lettered 'A' and 'B', were reserved for the publisher.
Other editionsOne-volume Library Edition (2003)
|To ask us a question, click here >>|
About four times a year we send an email newsletter to subscribers
For prices see Online Shop
Some opinions of our work:
Considering the tastefulness of the physical design of the Castle Hill volumes - which undoubtedly would have pleased Lawrence, who was a devotee of William Morris's idea of 'the book beautiful' - and the spare tastefulness of their editing, and especially their making available important but otherwise hard-to-access texts, this is a project for which Lawrence scholars will indeed be grateful now and in years to come. [Professor Stephen E. Tabachnick, reviewing Castle Hill Press books in English Literature in Transition]
. . . I couldn't be more pleased. The attention to detail, and conception of this edition, are wonderful . . .
I cannot praise too highly the quality of the production, with exceptional clarity and beauty of print, the erudition of editing, and the excellent on-line service. Important correspondence in beautiful books - the perfect combination.
. . .Excellence in research and editing, and magnificently produced books in superb bindings. Last but not least, efficient and friendly service, with books posted in rock solid packaging.
. . . These books are a pleasure to own and read . . .. . . a quite invaluable job in publishing (very beautifully . . .) many of the writings of TEL which hitherto have been available only in manuscript form in museums, libraries or private collections, or in out-of-print books which are very hard to obtain.
An excellent set of publications that are beautifully edited and produced. A wonderful addition to my library and to any library.