Dr David Barrett presents some fascinating insights on Elizabethan pronunciation.
I was always interested in the development of the English language and initially decided I would like to learn Old English in order to help me understand where our language originated. Old English is a fascinating language to learn, full of grammatical rules and interesting vocabulary. It is quite refreshing when reading in old English to meet a substantial amount of vocabulary which has been carried over into modern English. The use of phonetic symbols is not necessary to the study of Old English as the pronunciation is relatively straight forward and phonetic, like many other European languages of the time. Looking for a more practical application for language study, I started to read texts on Elizabethan pronunciation. It quickly became obvious to me that I had a huge gap in my knowledge as I had not looked closely at Middle English pronunciation. Many of the texts on Shakespearean pronunciation constantly refer back to the Middle English pronunciation and how it was affected by the Great Vowel Shift. The latter was a phenomenon which only affected English out of the European languages and caused marked changes in pronunciation, particularly in so far as the long vowels were concerned. We do not know what caused the Great Vowel Shift but its consequences were significant. In this chain shift, the long vowels were raised and the highest front and back long vowels became diphthongs (Middle English time and house were pure vowels, /ti:m/ and /hu:s/). The differing pronunciations of ‘ea’ words such as ‘meat’, ‘bear’ and ‘great’ today are a direct result of the GVS. In fact, ‘great’ is one of only a handful of words which preserve the original ‘ea’ pronunciation. A similar duality of pronunciation is found in some words ending in ‘ear’; compare, for example, ‘dear’ and ‘bear’. In Shakespeare’s day, both pronunciations were possible in many of these words.
It became apparent to me that I would have to get to grips with Middle English pronunciation before I could begin to look at Shakespeare. Like Old English, Middle English (ME) pronunciation is relatively straight forward and logical, and whilst reading Chaucer and other ME poets, I began trace a line of continuity between Old and Elizabethan English. Of course, English continued to develop through the ME period and didn’t change overnight, so the transition to Early Modern English (Elizabethan English) was seamless. My overriding interest was to look at Shakespeare’s language and see how the changes in pronunciation over the last four hundred years might have affected its rhythm, rhyme and other performance aspects, such as the use of puns (word-play on homophones for example). I was fortunate to discover that some brilliant minds had written about the development of language and pronunciation over this period and so there was some helpful literature to get started with. An understanding of Middle English pronunciation was essential to this study as sound changes are referred to in respect of changes to the ME pronunciation. A basic knowledge of phonetics is also necessary.
I was initially studying part-time for a masters degree in Script Writing for the Theatre at the University of South Wales in Cardiff. When it came to choosing a subject for my dissertation, I decided to write about Performing Shakespeare in the Original Pronunciation (OP). This accelerated my research and caused me to start thinking about the practical implications of using OP in performance. At the time, there had been very few productions in OP and very little had been written on the subject. I therefore focused on what was available and wrote about notable productions by John Barton in Cambridge in 1952, Helga Kökeritz in the USA in 1954, and David Crystal at the Globe Theatre in 2004.
One of my tutors suggested I enrol on the PhD course and, after consulting with the drama professor, I did just that (I completed the MA first). When I first looked for an MA scriptwriting course, I found that there were very few university departments which offered the option of writing for the theatre (rather than film or TV). This influenced my initial choice. The University of South Wales was a perfect choice to study for a PhD in Performing Shakespeare in the Original Pronunciation as I would be working in the drama department and there were plenty of students available to invite to my workshops. My work fell into two halves: the first part consisted of undertaking the scholarship necessary to have the expertise to speak with authority on the development of English pronunciation and to coach students in it; the second part, and the most exciting, was the practical research. This involved devising a means of teaching the pronunciation in a short time-frame to students who had little or no knowledge of language history or phonetics. This pronunciation was then used in rehearsal and workshop settings in order to discover how it might benefit actors and audiences today. At this time, I also ran some workshops in London for young professional actors who were interested in learning about OP. I was fortunate that at USW there was also a radio department and one of my workshops culminated in a recording for radio. Once enrolled onto the MPhil course, I was free to structure my own learning. I was very fortunate that Professor David Crystal was my external examiner for the MPhil transfer and he kindly recommended that I transfer to a PhD. David also helped me enormously with my first ever transcription of a play into OP, which was Twelfth Night.
In order to understand fully the mechanics of Shakespeare’s writing it was necessary to study not only the pronunciation of individual vowels and consonants, but also the effects of the pronunciation on the rhythm. Some words were stressed differently in Shakespeare’s day and some were spoken with more or fewer syllables. Therefore, it should be possible, to some extent, to reconstruct the original rhyme patterns of Shakespeare’s day as well as his metrical patterns. Owing to the effects of the Great Vowel Shift, many words had several different pronunciations and Shakespeare was able to pick the pronunciation that suited the rhyme, pun or metrical pattern. A modern example of pronunciation change may be seen in the word ‘garage’, the pronunciation of which is currently in a state of flux and differs from place to place and generation to generation. Some older people preserve a soft French ‘g’. giving the two syllables equal stress. Americans put the stress on the second syllable. Younger English people Anglicise the ‘g’ and only stress the first syllable, changing the vowel on the second syllable to an unstressed short ‘i’. Some aspects of speech, such as Elizabethan intonation could not be recorded and are lost for ever. During an intensive survey of Shakespeare’s texts (as well as poetry by other contemporary writers), I found I was able to reconstruct lost rhymes and restore metrical patterns, which I am convinced are a fairly accurate representation of the way the language would have sounded to Shakespeare’s audiences.
While I was in the process of reconstructing Shakespeare’s prose and verse and transcribing his texts, I became acquainted with the work of Dr Richard Flatter (R Flatter: Shakespeare’s Producing Hand). Flatter’s fascinating research looks at irregular metrical patterns and incomplete half-lines in Shakespeare’s plays and puts forward a convincing hypothesis for the possibility that Shakespeare was deliberately including hidden direction in his plays. Flatter makes the interesting point that, in his plays, Shakespeare was writing dramatic rather than poetical texts and so we should expect the lines to convey character, context and emotion. Flatter discusses, amongst other things, the fact that half-lines in asides should not be completed by a character who has not heard the aside. The same goes for characters who enter after a half line. He makes a convincing argument for simultaneous speech and proposes that irregular lines and metrical gaps are there for dramatic reasons. Flatter is critical of editors who, over the years, have attempted to regularise and smooth out the bumps in Shakespeare’s metre. It occurred to me (and Flatter also acknowledges this) that lines which appear to the modern reader to be irregular, may not have been so to Shakespeare but may be the product of pronunciation changes over the years. If one is to test Flatters theories, one has to do so wearing the hat of an OP practitioner.
The huge number of discoveries made during the course of my project cannot possibly be listed here, but a few examples will suffice. This passage, from A Midsummer Night’s Dream originally rhymed perfectly, every line rhyming with every other:
Flower of this purple dye,
Hit with Cupid’s archery,
Sink in apple of his eye.
When his love he doth espy,
Let her shine as gloriously
As the Venus of the sky.
When thou wakest, if she be by,
Beg of her for remedy.
The modern pronunciation of ‘wound’ below came from a regional variant. This and the older pronunciation below were both possible in Shakespeare’s day. This extract is another example where each couplet rhymes perfectly in OP:
But who is here? Lysander! on the ground!
Dead? or asleep? I see no blood, no wound.
Helena. O, I am out of breath in this fond chase!
The more my prayer, the lesser is my grace.
Happy is Hermia, wheresoe’er she lies;
For she hath blessed and attractive eyes.
How came her eyes so bright? Not with salt tears:
If so, my eyes are oftener wash’d than hers.
No, no, I am as ugly as a bear;
For beasts that meet me run away for fear:
Words beginning ‘qu’ would generally be pronounced ‘qw’ in words of Old English origin, such as ‘quick’, but might be pronounced as ‘k’ in Latin or French borrowings, such as ‘banquet’. This results in ‘quote’ and ‘coat’ being a homonym, which enabled Shakespeare to use this pun in The Two Gentlemen of Verona:
And how quote you my folly?
I quote it in your jerkin.
The first stage of my work was pure scholarship. I needed to get to grips with the whole idea of Elizabethan pronunciation and I needed to be able to formulate a transcription policy, which would justify all the sounds I had chosen to use. I found four texts enormously helpful in this regard. My main go-to text was EJ Dobson’s English Pronunciation 1500-1700. This falls into two volumes, the first being a survey of the contemporary sources, those writers of the period who described contemporary pronunciation in some detail (albeit without a standard phonetic alphabet), sometimes in justification of a standardized spelling system. These writers are known as the Orthoepists. The second volume of Dobson consists of a detailed discussion of the phonology of the two-hundred-year period. Also very useful was Helga Kökeritz’s Shakespeare’s Pronunciation. As well as including a survey of Shakespearean phonology, Kökeritz also includes a rhyming dictionary, a list of common syncopations (words containing elided syllables, which affect the metre) and evidence of Elizabethan pronunciation in Shakespearean puns. Cercignani’s Shakespeare’s Works and Elizabethan Pronunciation is useful for cross-referencing pronunciations and for an alternative perspective, although it is harder to use to look up individual pronunciations. Finally, Charles Barber’s Early Modern English gives a wider view of the language of the time but includes a really useful chapter on phonology. Understandably, these four works do not agree on all aspects of the pronunciation. This is not surprising when one considers the regional variations in pronunciation, differences between younger and older generations and the effects of a language which was in a constant state of flux. In a sense, this makes the subject more exciting as the scholar of OP is able to make informed choices of their own about which pronunciation to choose. The transcriber is able to include subtle differences of pronunciations between generations (or even stations in life) as David Crystal did in his transcription of Romeo and Juliet for the New Globe Theatre. The above sources provide some interesting spelling evidence, gathered by a study of the variant spellings and idiosyncrasies of individual writers (including Queen Elizabeth). The question I was often asked by students was, ‘How do you know what Shakespearean pronunciation sounded like? This is a point I clarified in the justification to my transcription policy:
We have the evidence of the orthoepists, those whose interest was in writing about the mechanics of their own pronunciation. This is problematic. There was no standard pronunciation in Shakespeare’s day and there is known to have been marked variation between the regions. Even today, regional dialects share a phonological system but there can be distinct differences in the detail of the pronunciation? Where there are marked differences in pronunciation, I chose to favour the writings of those orthoepists from the south of England, such as Charles Butler, whose pronunciation would have been near to that found in London, and those, such as Alexander Gil, who had lived for a considerable time in the capital before publishing.
But what about Shakespeare’s own pronunciation? At least in his younger years he would have been influenced by the Warwickshire accent, even if he felt obliged to conform to the London regional form of speech later in life (as well as to the hyper-correctness of the grammar school pedagogues). Evidence of Shakespeare’s pronunciation is embodied in the texts in the form of word-play, rhymes and rhythm, which may shed some light on his vowel sounds and syllable patterns. These, also, are open to interpretation, but when usage is compared over a range of texts and is viewed in conjunction with the orthoepists’ comments, certain patterns begin to emerge, such as frequent examples of the vacillation between different vowel sounds on the word ‘fear’. My project was concerned not solely with Shakespeare’s own pronunciation but with that of the London stage in his era.
Possibly least helpful to the transcriber, there is evidence embedded in the occasional spelling peculiarities of contemporary writers. Although the authority of a text does not necessarily derive solely from the author, the transcriber may have had a hand in it, I would argue that, in terms of what we can discover about pronunciation, it is not relevant whether the spelling derives from the author or a scribe. The authority it carries with it is of the age. A scribe may reveal through a spelling pronunciation the pronunciation of the day, however incorrect the actual spelling might be.
The authority of the transcription, then, derives from the combination of these sources as interpreted by the transcriber. Although not infallible, it may be as much as eighty percent accurate, in the view of David Crystal (Pronouncing Shakespeare, 2005). Crystal points out that we cannot be certain now whether Shakespeare heard something as a full rhyme or perhaps a half. As transcriber, I chose to place more weight on the evidence of Shakespearean texts than on the orthoepists or any vagaries of contemporary spelling.
My transcription policy guides the reader through the main features of the vowel and consonant systems of Shakespeare’s English, looking at unusual features of the language, such as syllable expansion and reduction, unfamiliar stress patterns and some variant pronunciations, often caused by lengthening, shortening, restressing or secondary stress. It is perfectly possible to rationalise the possibilities into a workable set of norms which may be applied universally to Shakespearean texts.
The overriding objective when transcribing is to keep the script actor-friendly. This immediately poses a problem: how does one accurately represent the pronunciation of the spoken word orthographically without using overcomplicated phonetic symbols? A precedent was set by David Crystal, who presents his text in a part broad (phonemic) transcript. This has several advantages over a full phonetic script. The actor is able to see at a glance where the common ground is with present day English (PDE) pronunciation and is able to focus on the differences. There seems little point in asking actors to interpret phonetics where the pronunciation is identical to today’s. In this type of transcript there is scope for variation in the actor’s accent, provided it is kept within the parameters of OP as defined by the phonetic symbols. Although the transcript might look strange to the actor at first sight, reading it will become fluent as soon as a small number of phonetic symbols is mastered, as well as a few rules of interpretation. These mainly relate to secondary stress and the pronunciation of words under weak stress.
The disadvantage of using a broad (phonemic) transcription in this way is that a common base accent is used as a point of departure. My project took the base accent to be RP, but productions in America have used transcriptions designed for the local accent. Any deviation from this base accent is notated phonetically and anything which conforms to the standard is left unaltered. It follows then that an American company would not be able to use a transcription prepared for a British production unless the actors were skilled enough to use British RP as their base accent, which may sometimes be the case. The only alternatives are to make a complete phonetic transcription, which would be less actor-friendly and more labour-intensive for the performers, or to train the cast to the level at which they are able to work from a regular script (perhaps with some footnotes), which is infinitely preferable, and achievable, given enough workshop and rehearsal time.
Once I had the scholarship under my belt and had written a workable and fully justified transcription policy, it was time to develop a methodology for teaching the pronunciation to actors. I decided the best plan was not to teach lines by rote but to teach actors where the pronunciation has come from and how the transcription policy works. They are then better able to apply the system themselves. In practice, I taught the long and short vowels in isolation, moving on to any peculiarities in the pronunciation of consonants then looking at syncopations and resolutions (the shortening and lengthening of words by syllable reduction and expansion). The effects of syncopation may be seen in the modern pronunciation of secretary. Does it contain three or four syllables? Syllable reduction was rife in Shakespeare’s day and, together with resolution (for example, giving the word resolution five syllables: re-so-lu-ti-on) may be the cause of many an irregular line when spoken in modern English. I employed flash cards and grouped the words from a given text into lexical sets of alike pronunciations. I presented the actors with the rhyming and pun evidence, which helped to justify my transcription policy, and they were happy to accept it as a possible set of pronunciations to use in OP.
Over the course of three years, I was privileged to work with students at the University of South Wales, students from drama schools in London and young professional actors. The enthusiasm and affection for OP amongst of all of these groups was palpable. As well as perfecting (as far as possible) my transcription methods and teaching practice, it was interesting to observe the way performing in OP affected their basic acting technique and enabled them to discover a new voice. Above all, almost without exception, the actors found using OP a great deal of fun.
Although I owe a great deal to the work of David Crystal, my transcription policy differs in the pronunciation of some of the vowels. The vowel sounds are open to interpretation and any variations simply reflect the sort of differences one hears in present day spoken English. This only goes to demonstrate that there is no single ‘correct’ interpretation of OP. Any variations of style are a positive feature; the sort of variety which one finds in a PDE production is quite acceptable, although more extreme variation is not. As Ben Crystal said, “[a]n accent from the outskirts of the village is acceptable but not one from the next village.”
During the course of my research I uncovered in old newspaper archives evidence of a hitherto unknown production of Macbeth in OP. This production took place in 1952, the same year as a well-known student production of Julius Caesar in OP, directed by John Barton at Cambridge University. The difference was that Macbeth, at the Mermaid Theatre in London, was a professional production, the first one known in modern times. The original Mermaid Theatre, before it moved to Blackfriars, was situated at the bottom of actor Bernard Miles’ garden in St John’s Wood and was a reconstructed Elizabethan Theatre. Surprisingly, I manage to source a programme from this production. The programme notes revealed that the voice coach on the production was AC Gimson, a phonetician from University College, London (and David Crystal’s phonetics teacher). Digging deeper still, I discovered that Gimson had done some voice recordings for the actors. If only those recordings were still available, they would reveal so much about the style of pronunciation adopted by the production. After a tricky and time-consuming search, I tracked these down to the British Library (although there was no mention of OP or the production in the archive descriptions). After a few hours of listening, I was convinced that these were the missing voice recordings by Gimson and his fellows at UCL. This was confirmed by David Crystal (who kindly came down from Anglesey) who even recognised his teacher’s voice. This was a bit of a scoop for me. Close examination of the style of pronunciation revealed a conservative, well-articulated style of speech which would probably have sounded old-fashioned to Shakespeare himself, but serves to demonstrate that there is a certain degree of tolerance afforded to the transcriber.
There seems to be a general assumption that Shakespeare’s lines must always be spoken with meticulous articulation, emphasizing every syllable. In respect of the original pronunciation, this could not be further from the truth. In Shakespeare’s day, amongst other things, aitches were dropped and -ing word endings reduced to -in. The pronunciation of the stage was far less formal than one would expect.
There is a myth that somewhere in the mountains or remote areas of the USA there lurks an isolated population which preserve the Elizabethan speech habits brought to the country by the founding fathers. Of course, this cannot be true. Even in an isolated community, language does not stand still but develops and changes over time. However, there are some remote communities which preserve some aspects of historical language, such as the people of Tangier Island in Chesapeake Bay. Some of the vowels spoken by that community (there is evidence on You Tube) are certainly more akin to Elizabethan than modern American English.
There have been, recently, a few notable performances which have faithfully reconstructed the speech of Shakespeare’s stage. David Crystal has done much to spread the popularity of OP over the last twenty years. The two productions of Troilus and Cressida and Romeo and Juliet at the New Globe Theatre, for which he prepared the transcriptions and coached the actors, brought OP to a wider audience. These, and other more recent productions involving both David and Ben Crystal (such as Ben’s Hamlet at Nevada University) have done much to sustain the momentum achieved by the New Globe productions. Paul Meier has done a great deal to popularise OP in America, particularly through his production of A Midsummer Night’s Dream at the University of Kansas.
My study of past OP performances has revealed some pointers which may be of benefit to directors considering performing in OP. The question of readability by the audience is important to anyone who is considering investing in a production of this type. In this respect, OP may be viewed as simply another accent of English, albeit one that is no longer in everyday use. The phonemes, syncopations and stress patterns are not wholly unfamiliar to present-day English speakers, and the consonant system of Elizabethan English is almost the same as today’s. Within the general sound map of OP there are various possibilities for interpretation which have been used in past productions and there is scope for actors to use their present-day regional accents. Audiences are surprisingly tolerant of linguistic variation and are able to accept a variety of pronunciations. The OP sound map intersects in many respects with that of Present Day English. Many of the phonemes are identical and most OP vowel sounds may be found in some regional context in the British Isles or around the world. Actors too are very adaptable, and their training in the use of different accents enables them to adopt the use of OP as they would a regional dialect.
The rate of delivery is significantly speeded up in OP performances. This might enable performances of plays to achieve a running time closer to that of the Elizabethan stage. The music had to be rescored for the Globe OP production of Romeo and Juliet as the underscored passages, which fitted the modern English performance perfectly, were too long for the OP dialogue. According to Alfred Hart (In Stolne and Surreptitious Copies: A Comparative Study of Shakespeare’s Bad Quartos ), Elizabethan actors would speak 2,300 lines in two hours, compared with around 1,700-1,800 by the RSC. This faster pace would significantly speed up the dramatic action.
It seems logical to trace the development of pronunciation forward a generation and take at look at how the application of OP might affect restoration drama, so there’s a future project. Dobson’s survey of English pronunciation stretches as far as 1700, so that seems an obvious starting point. Other writers also follow the sound changes until the very end of the Great Vowel Shift and onward to the present day. However, I remain fascinated by the way OP can inform Shakespearean productions, even those which do not attempt to restore the rhyme and metre fully. There is plenty more research to be done on the Shakespeare plays. What would be fantastic is for the students at our drama schools to be taught the benefits of using OP to reconstruct Shakespearean verse and bring new possibilities to performance. Everyone I have worked with has been enthused by OP and declared that it will bring a new dimension to their work.
David’s thesis and some other short papers on Original Pronunciation may be read or downloaded from:
David’s catalogue of his own play scripts and musicals may be found on his website:
EJ Dobson (1957), English Pronunciation 1500-1700, Oxford at The Clarendon Press.
H Kökeritz (1953), Shakespeare’s Pronunciation, Yale University Press
F Cercignani (1981), Shakespeare’s Works and Elizabethan Pronunciation, Oxford at The Clarendon Press.
C Barber (1976,1997), Early Modern English, Edinburgh University Press
T Nevalainen (2006), An Introduction to Early Modern English, Edinburgh University Press
D Crystal(2005), Pronouncing Shakespeare, Cambridge University Press
Richard Flatter (1948), Shakespeare’s Producing Hand William Heinemann, Melbourne, London, Toronto
D Weingust (2006), Acting from Shakespeare’s First Folio: Theory, Text and Performance, Routledge, Oxford
David Barrett has a PhD in Drama and English from the University of South Wales (his thesis was titled Performing Shakespeare in Original Pronunciation), a master’s degree (MA) in Scriptwriting for the Theatre and a master’s in music (MMus) from London University. David is currently an English and Drama teacher at Portsmouth High School. He has been writing and composing for the stage for over twenty-five years. David has written and directed plays for schools, theatre companies and London fringe theatre (Jermyn Street, Canal Cafe, Pacific Playhouse). His interest in the English language led him to research how actors might benefit by performing in original pronunciation.