[Journal of the Simplified Spelling Society, J23, 1998-1, pp3-7]
[See also Newsletter N1 'Androcles & the Lion' and Shaw Alphabet in Journal topics.]
The Shaw Alfabet: key for riters.

Sound-writing 1892-1972:
George Bernard Shaw and a modern alphabet.

Kingsley Read.

foto of Kingsley Read

We are grateful to Michael Twyman, Professor of Typography and Graphic Communication at the University of Reading, for permission to reprint this memoir which was written for an exhibition in the university library in 1972. It subsequently appeared along with the photograph of Kingsley Read taken in the late 1940s, in 1983, in the catalogue to the Shaw Alphabet archive in the University's Department of Typography and Graphic Communication. We also have to thank Read's daughter, Mrs Mavis Mottram, for her encouragement in reviving her father's work. Subheadings and endnotes are added. The SSS previously published items on the Shaw Alphabet in its Newsletter N1 (Bob Brown, April 1991, pp2-3) and in JSSS J18 1995/1 (Alice Coleman, pp25-30).

1. Origins and advantages of written speech.

Neither words nor alphabets have always been used in records. Cave men recorded hunting exploits pictorially. The earliest crude symbols to be written were unrelated to words; they were 'pictographs', simple standardised drawings, hundreds of which were needed to convey imprecisely a very limited range of ideas. With more precision, Chinese writing employed thousands of 'ideographs', which only experts could read and write.

Then, 3000 or more years ago, came the highly economical, easily applied, exactly meaningful, writing with 'alphabets'. Given readers who spoke the writer's language, a few graphic symbols (now called 'letters') could serve to represent the few basic sounds with which a whole language was spoken. Words became visible as well as audible. The Phoenician, Greek, Etruscan and Latin languages were adequately represented by as few as 22 to 25 letters.

Roman civilisation and the Roman church made Latin the international language of writers in Britain and throughout Europe for roughly 1500 years. Although by 1400 AD Chaucer and Wyclif were using a form of English, it was not the English we now speak. To the Latin alphabet a letter W had been added. Later, U and J became letters with sounds distinguished from those of V or I. But as Latin C, Q and X have sounds otherwise represented (by S or K or KS or GZ), only 23 of our 26 letters could serve us for sound-matching, even if used consistently in our spelling. As there are at least 40 significantly differing speech sounds employed in speaking English, we lack 17 single letters for single sounds. To write these 17 sounds by means of couplets, triplets or quads of letters (such as SH, THE, CH, WH, TCH, OWE, AWE, EIGH, OUGH) is ambiguous, unmethodical and wasteful. While we continue to use the Latin alphabet with only three added letters, spelling largely depends on memory, not on method. An alphabet of some 40 or more simpler characters would eliminate the wasteof labour and materials caused by our traditional spelling irregularities. Writing and printing would occupy far less space. It is this resulting economy, still not fully appreciated, that Bernard Shaw grasped and fostered. His aim was not conceived as educational but utilitarian.

2. Sweet's approach to a desirable modern alphabet.

The story told in this exhibition begins with an unusual kind of alphabet concerned with economies in writing, published in 1892 by Henry Sweet of Oxford, a great authority on phonetics, the science which analyses speech into its few significantly different sorts of sound. Sweet's analysis of spoken English into some 40 sorts of sound was not original. Isaac Pitman among others had used 40 sound-sorts matched by as many characters, both for an abbreviated shorthand and for longhand (romanic) sound-writing.

The most distinctive feature of Sweet's Current Shorthand was that his characters always kept their appointed place on the horizontal 'writing line'; wheras Pitman's and other fast shorthands, by joining ends to beginnings in any sequence of characters, makes words wander variously from a ruled or imagined writing line - a wandering much exaggerated where long words are fully spelled. For typewriting and type-set printing the aligned sequence of lettering is essential.

Sweet's lettering, then, conforms to the traditional three main kinds of characters: Shorts, which stand on the imagined writing line with their tops aligned on an 'upper parallel' (like orthodox letters a e m n o u); Talls, which (like b d f h k l) stand on the writing line but ascend well above the height of Shorts; and Deeps, which (like g p q y) are top-aligned with the Shorts on the upper parallel but descend well below the writing line. This is a neat and familiar manner of writing: Talls and Shorts keep an imaginary writing line well defined, while Deeps and Shorts equally preserve an imaginary upper parallel.

Less happily, Sweet employed two more categories of lettering: one so enlarged as to be both Tall and Deep (like a script letter f), the other of less height than the Short letters: neither the too large nor the too little letters serving to preserve either parallel's level at all. Furthermore, Sweet's own writing distorted the small letters in order to link them fore and aft with larger letters. He held the too common belief that for fast writing the writer may only lift the pen between words.

In using Short, Tall and Deep lettering, Sweet conformed to tradition. Quite apart from any use of abbreviated spelling, he gained speed by enlarging his alphabet to spell all single sounds with single letters. That is, he used no 'digraphic' sound-spelling such as TH, SH, IE, AY. Moreover, Sweet's characters are among the simplest graphic shapes known to geometry: they are mostly single penstrokes, without dottings, crossings or 'diacritical' markings such as dictionaries use to define a letter's pronunciation. Such markings would involve pen-lifting and hand movements additional to any required in advancing from one letter to the next. Sweet's alphabet served to spell, to write, (and could have served just possibly to type) with simpler, as well as fewer, letters than are used in orthodox English. It was in this respect that it provided a crude model worth refining as recommended by Shaw: not to serve still as shorthand, but as an all-purpose modern alphabet.

Dr Abraham Tauber's book, George Bernard Shaw on language (London, Peter Owen 1965, p30) states that Shaw first met Sweet as early as 1879. It is well known that Sweet became in some measure a prototype for Henry Higgins, society speech trainer, in Shaw's Pygmalion, written in 1912, the year of Sweet's death.

3. Shaw's appeal for a wholly new alphabet.

Shaw habitually drafted his own writings almost fully spelled in the 40-letter alphabet of Pitman shorthand. He may well have found this unsatisfactory for re-reading and revision. It could spell sounds unambiguously, having an adequate number of letters. But as its script was unaligned, it certainly could not serve also for typing and type-set print. Moreover, Shaw was very knowledgeable and interested in fine typography. At the age of 85, he appealed to "type designers or artist-calligraphers, or whatever they call themselves, to design an alphabet capable of representing the sounds of the following string of nonsense quite unequivocally without using two letters to represent one sound or making the same letter represent different sounds by diacritical marks." The nonsense test-piece was intended to cover all English sound-sorts and to discover designers who truly recognised them. He then went on to recommend Sweet's alphabet as a suitable point of departure for his designer (see pp26-27 of Shaw's preface to The Miraculous Birth of Language, by Professor Richard Albert Wilson, London, Dent, 1941).

This Preface, dated February 1941 but not published till the autumn, gives Shaw's most precise instructions, though his public campaign opened with a long and important letter to The Times of 14 April 1941. Only years later was the letter to The Times made known to me, but while I was myself experimenting with a sound-spelling alphabet, my attention was drawn to Shaw's appeal in the Preface.

How many others responded seriously to his appeal I was never able to discover, though I tried. Shaw dissuaded me from contact with or influence by others. But from acknowledgement postcards he had printed, it would seem that there was no lack of misdirected proposals and gratuitous advice; for there he stated concisely what he sought and what he repudiated. Especially notable is his dismissal of all "schemes spelling English phonetically with the old ABC".
He sought a wholly new alphabet - "to be used and taught concurrently with the old alphabet until one or the other proves the fitter to survive". He would not consider tampering with orthodox English spelling or its traditional alphabet: these were to be left undisturbed - and unimproved.

What - beyond courage - qualified Shaw to demand a new English alphabet? Though an Irishman to the last, he certainly possessed authority on the pronunciation of English. From 1926 to 1939 he served on the BBC's Spoken English Advisory Committee. When Robert Bridges, Poet Laureate and first Chairman of the Committee, died in 1930, Shaw succeeded him as Chairman for the next ten years. The Committee included several exponents of phonetic writing. Bridges himself had with the help of the calligrapher Edward Johnston, produced a large and graceful alphabet. Daniel Jones [1] and A Lloyd James [2], both expert in phonetics, later became professors. Sir Johnston Forbes-Robertson was, among other things, the best Hamlet of his day. Logan Pearsall Smith, with Robert Bridges, inaugurated the Society for Pure English.

By 1936, the Committee had grown to 24 members, of whom seven were senior academics. Other advisers included well known speakers such as Lady Cynthia Asquith, Kenneth Clark and Alistair Cooke. It is therefore not surprising that Shaw developed a keen interest in creating an alphabet fully allied to speech. His association, on this Committee, with phonetic experts must surely have helped him to crystallize his own ideas for a modern all-purpose alphabet.

4. Read's early attempts.

What were my own qualifications to further Shaw's intentions? It may be enough to say that in my teens I went with a scholarship to Birmingham School of Art and there learnt lettering and designing under the headship of Robert Catterson Smith, a one-time Kelmscott craftsman; and that between the wars I designed and commercially supplied large lettering in various fashions and materials. On the phonetic side I had taken a course of speech training, and had studied several phonetic alphabets, including those of Bridges and Sweet. If I was particularly qualified at all, it was in having some practical experience, both graphic and phonetic. When, around Christmas 1941, I read Shaw's Preface, I was 54, old enough to back keen interest with long perseverance.

After a month's preparation I submitted to Shaw
(a) a tentative alphabet of 47 letters
(b) reasons for choosing them, and
(c) their transcription of his test-piece of nonsense.

To these I added
(d) a sheet of variously styled lettering to show how the alphabet might be adapted in writing, printing or display, to scribble a note or engrave a monument, to print books or make neon signs.

His printed acknowledgment postcard, dated 27 January 1942, bears an exceedingly kindly, almost excited footnote. He showed my first crude attempt to others. To my repeated enquiries for advice from him or others helping him, he only replied that I was better left to my own devices. I am aware of two or three cases in which he subsequently commended to recognized authorities my grasp of his intentions.

At his desire, in 1943 I prepared a manual with examples, entitled Sound-writing: a method and an economy in spelling. Shaw found it "admirably clear", though he disliked some "graceless lettering". His belief that "for handwriting the words must be written without lifting the pen" is one I cannot share. Schools no longer require it. His own signature to this letter shows three harmless liftings of the pen in his name "Bernard".

This letter begins with advice to consult Mr I J (later Sir James) Pitman [3], of shorthand and publishing fame, whose experience of phonetic alphabets is unrivalled. Mr Pitman dissuaded me from immediate publication and encouraged me in further improvements of the manual's alphabet.

5. Developments before Shaw's death.

In the autumn of 1944 Shaw announced in The Author (quarterly journal) his intention to make a Will promoting a new alphabet. He had already in a letter dated 19 July 1944, told Pitman " ... so I wash my hands of the business, and leave the field open to you to do the job with a grant in aid from the Public Trustee... " It is certain that no abler and better situated co-ordinator could have been chosen to see the task through, even if Pitman's personal leanings were more educational, less specifically utilitarian, than Shaw's.

Three years later, in 1947, Mr Pitman and Dr Daniel Jones visited Shaw to urge upon him the aims of the Simplified Spelling Society. Their reception is related fully by Pitman in his introduction to Tauber's Shaw on Language. Their Society's commitment to using none but our accustomed 26 letters of the alphabet - and consequently to digraphic spelling of sounds - was anathema to Shaw: he was adamant against it.

The Will, finally signed on 12 June 1950, does not specifically exclude the use of familiar letters of the alphabet, but it was evident to the Trustee from Shaw's published writings that he had intended the use of a wholly new set of between 40 and 50 characters. If further evidence were needed, it exists in Shaw's private correspondence quoting my grasp of his intentions as a guide.

The Will was wilfully made in language more Shavian than legal in so far as its Clauses 35-38 dealt with the alphabet. Beginning with Sub-section 35(1), it calls in effect for some estimate of the world's man-hours wasted in writing and printing English with an alphabet of 26 instead of 40 or more letters; and a valuation in money of those wasted hours. This impossible task was entrusted to Mr P A D MacCarthy who, having investigated, could only report that no reliable data exists for any meaningful estimate. Sub-section 35(2), also in Mr MacCarthy's care, deals with transliteration of Androcles, which presented a few problems mentioned in his Appendix to Androcles.

Although Shaw's letter to The Times, his Preface to Wilson's book, and his private correspondence refer explicitly to an alphabet for printing from type as well as for script, the Will makes no definite provision either for or against using printers' type in Androcles. Clause 35(2) provided funds "to employ an artist-calligrapher to fair-copy the transliteration for reproduction by lithography, photography or any other method that may serve in the absence of printers' type". In brief, the Will permits, if necessary, a departure from normal letterpress printing. It was agreed that no such departure was necessary.

6. Subsequent developments.

Shaw died on 2 November 1950. It was not until royalties from My Fair Lady swelled the estate that his executor, the Public Trustee, could put into effect the Will's Clause 35 concerned with the alphabet. By then this Clause had been challenged and its validity had to be tested in the High Court. After a costly hearing, it was pronounced legally invalid.

An appeal being denied at first, Mr Pitman sought my help to implement Shaw's intentions without resort to his estate. One result worth mention was a leaflet showing the economy of letters and space made by my then proposed alphabet, compared with an orthodox type-setting. By taking the Lord's prayer as an example, the phonetic values of my lettering were evident without a key. Here I already used the alphabet which was destined to become a competition entry. However, largely by Pitman's exertions, the dispute was settled by allotting no more than £8,300 to execute Clause 35 relating to the alphabet.

7. Competition, discussion, and choice of alphabet.

Thereupon, the Trustee announced a world-wide competition to secure ideal designs for a Shaw Alphabet. Though this clearly reduced my own chance of formulating it, my previous work was not unknown to the Trustee who in January 1958 persuaded me to illustrate and discuss competition requirements on BBC's programme, Panorama.

Clause 6 of the Trustee's 'Advertisement M.4405.V' stated that "it is implicit in the Will and in Mr Shaw's writings" that the main object is "saving of labour ... a means of writing and printing the English language which will be more economical of the writer's time, of the paper and ink of the printer, and of transport and storage, yet convenience and ease in reading are of importance... Practical problems of typography will be taken into account". Clause 7 adds that "designs of shorthand codes for verbatim reporting and designs for reforming the existing alphabet by addition of analogous letters will be disqualified".

Competitors had a year in which to prepare their alphabetic entries. I saw no reason to amend my Lord's prayer alphabet, nor to submit alternative entries. The Advertisement offered inconclusive counsels on sound-sorts to be represented. I hardly believed it possible to arrive at a perfect alphabet without finally pooling the wisdom of competitor(s) and judges.

In view of Shaw's stipulated speech model, "that recorded of His Majesty our late King George V", I went to Broadcasting House to have a number of the King's recordings played over to me. His pronunciations varied according to context as with all other speakers. I also went to type-founders - the Monotype Corporation - and consulted printers, becoming convinced that Androcles ought to be type-set, not reproduced from a calligrapher's fair-copy as the Will permitted "in the absence of printers' type". I wrote to Mr Pitman on 18 November 1958 that fair-copying "is superfluous. Worse, the very absence of type provides a gratuitous argument for opponents ... The Will provides for propaganda costs. The fait accompli is our best, most widely intelligible propaganda". His reply agreed: he too had taken stock of the possibilities.

My competition alphabet was accompanied by examples, type designs, and detailed reasons for sounds and characters chosen. It proved to be one of 467 entries, many of them from abroad. None met exactly the ideals of the judges. However, I found myself among four competitors sharing the honour and the prize. Our four entries are best compared as scripts, though hardly as typography, in renderings of the Lord's Prayer reproduced in a trade journal, Print in Britain.

Mr P A D MacCarthy, from Leeds University's Department of Phonetics, was undertaking a transcription of Androcles in the new alphabet as soon as one could be adapted and approved. He was therefore asked by the Trustee "to collaborate with one or all of the four designers mentioned... (see the Foreword to Androcles) to produce the best possible alphabet... " Various revisions were considered till finally each designer's latest attempt was re-written by a disinterested calligrapher for comparison. The selectors chose mine as closest to their requirements, discussed with me a few possible alternatives, and nominated me for appointment as designer responsible to the Trustee and his adviser. My letter of appointment is dated 19 July 1960.

8. The Shaw Alphabet in print and typewriting.

A month later, on 18 August, I brought to London the finished Shaw Alphabet. It was fully discussed with Mr Pitman and with Mr J T Harrison (of Stephen Austin and Sons, Hertford, who produced type and printed Androcles) and it was adopted by the Trustee. I then proceeded to make the die-cutting drawings - 30 times print size - in three distinct styles required for stage directions, the names of speakers, and the dialogue.

Mr MacCarthy was by this time transliterating the play while on secondment to Lahore University, Pakistan, and a good deal of printers' proof revision fell to me. New and old versions of the play were printed on facing pages, matching exactly line for line, without either over-running the other. The task of securing tolerable typographic spacing was not easy. An edition of 40 000 paperback copies was issued commercially by Penguin Books Ltd. Their refinements of typography in the orthodox version inspired me to emulate it in the new alphabet. Our joint result was chosen as one of the National Book League's 'best printed books of 1962'.

Apart from this Penguin commercial edition, the Trustee distributed gratis to all Head Public Libraries of Britain, the Commonwealth, North and South America, and to all National Libraries of the world, a total of some 13,000 hard-back copies which should still be available. [4]

The Shaw Alphabet itself, and both editions of Androcles, were published on 20 November 1962, with a press conference and publicity on television.

No-one needs to know the new alphabet to see immediately that Androcles demonstrated a marked economy; for the lines of its orthodox text are exactly 50% wider than matching lines in the Shaw Alphabet. Normally, line-widths would not be shortened; but books in the new alphabet would occupy one-third fewer pages, using that much less type and ink; they would be lighter for handling, transport and shelving, and a good deal cheaper. Questioned in the press conference as to cost, Mr Harrison replied that his type-cutter and type-setter had used no unusual procedure or machine. Except for its novel letters, it was a perfectly normal type, normally printed.

It is also immediately clear that the new letters are consistent in their sound-writing. As to the economy in printing, rather less than half of it comes from single-letter representation of single sounds - ie from avoiding digraphs; more than half comes from simpler and narrower lettering.

Since that day, it cannot be said that alphabetic economy is technically 'impossible' - or even difficult. The fait accompli proves Shaw's point. A transliteration of part of Lincoln's Gettysburg address exhibits good typography in the Shaw Alphabet. An article on the new typography was commissioned by Indian Print and Paper, a Calcutta trade journal.

For my part I was determined to carry the accomplished evidence further - further than the Will specifically required. Throughout 1962 I had been preparing plans for a Shavian type-writer, and on propaganda grounds the Trustee accepted quotations obtained from Imperial Typewriters Ltd, Leicester. The special letters were cut for around £70 and thereafter a normal portable machine (44 keys, 88 characters) was available at the current catalogue price of £29. The Trustee provided Mr MacCarthy and myself with the first two such machines. The keyboard not only carried the Shaw Alphabet, numerals, punctuation marks and sundry signs: it retained 26 Roman capital letters for orthodox addressing of envelopes.

I used my Shavian typewriter to produce a quarterly journal called Shaw-script; for correspondents sought more reading practice than Androcles gave them. The original typescript was reduced and offset printed by Rank-Xerox Ltd, Birmingham.

The Shaw Alphabet.


9. Correspondence, evidence, and current developments.

We needed practical evidence that all sorts and conditions of persons, at home and abroad, can easily learn and write and spell with the Shaw Alphabet. Such evidence depended upon an organised correspondence invited by Sir James Pitman on page 16 of Androcles. By the time his invitation was published, he had become so fully engaged in other activities that he sent me an SOS. If correspondence was to be organised at all, I must do it.

I accepted the task with an entirely free hand, for it was possible that minor problems, unforeseeable by theory, might emerge from the alphabet's use by persons of all sorts, ages and dialects. A Guide to Shavian Spellings was prepared and I awaited results. Experience thus gained, being largely technical, is detailed elsewhere. Enough to say that Londoners, Scots, Americans, while raw beginners, regarded their personal speech as the 'proper' English, but were contentedly conforming in a matter of weeks to the printed spellings of Androcles and the journal Shaw-script; for a ready conformity saves thought and meets readers' expectations.

It was observed that unskilled or hasty scribblers wrote no less decipherably in the new alphabet, but that four of its characters tended to be malformed grotesquely.

After four years of handling correspondence it seemed clear to me that some graphic and phonetic changes in the alphabet would increase its already striking facilities. With this - possibly unique - practical experience to go on, it seemed a duty to implement it in a final alphabet, one differing even less from the now unalterable Shaw Alphabet than that had differed from Sweet's.

So, with help and encouragement from writers willing to test changes rigorously in circulated correspondence, I gradually evolved the 'Quickscript Alphabet'. Its manual, issued late in 1966, is in the British Museum Library, the Library of Congress and elsewhere, including Reading University Library (where the technicalities and history of these alphabets is documented).

Since early 1967 Quickscript has been used satisfactorily. Among those able to speak with equal experience of both Shaw-script and Quickscript are Professor Russell Graves of North Carolina University, who drafts his stage plays in Quickscript, and Mr E J Canty of Portsmouth, who was a fellow competitor in 1959. All who have experience of writing in both alphabets prefer Quickscript's facilities and its relative simplicity in sound-writing.

It is to be doubted whether the Sweet-Shaw-Read line of evolution can go much further. Its use is learnt with ease. It enables both script and print to be done with marked economies. If research establishes the greater efficiency of a modern alphabet in advance, another generation may see it "used and taught", as Shaw hoped, "concurrently with the old alphabet until one or the other proves the fitter to survive."


[1] President of the Simplified Spelling Society 1946-68.

[2] A Lloyd James wrote the Preface to the 5th edition of the Simplified Spelling Society's New Spelling (1940).

[3] President of the Simplified Spelling Society 1968-1972 and originator of the Initial Teaching Alphabet.

[4] A total of 265 remaining copies were passed to the SSS in 1991, and have since been distributed mainly to members.

Back to the top.