“Sorry, but I have to ask…. is it /ˈɛksɪʌnt/?”, the BAC’s box office manager asks me as we don hard hats to enter the beautiful grand hall in its crumbling post-fire glory.
“I’m /hərˈmaɪ.əni Granger.”, Emma Watson announces to a cinema full of surprised Irish children, whose brief 12 years have not brought them in contact with Greek mythology and who have, until this moment, been moderately certain that Harry Potter was acquainted with a girl called Her-my-own.
Pronunciation is just one of the key ways in which we understand and form ideas about each other when trying to communicate in the same language. Anyone who has forayed into conversation with a native speaker when first learning a new one will be intimately familiar with the laughter and requests for repetition that greet our attempts to bend our tongues around less familiar words, and the confession that, even had we pronounced it correctly, our accent would have given us away immediately. As actors are regularly taught, a perfectly recognisable word can be rendered incomprehensible through incorrect pronunciation, intonation or “because you’re constantly mumbling” (sorry mum!), and a badly attempted accent can throw an audience off instantly.
Despite our multifarious backgrounds as English speakers, it is generally accepted that there is a pronunciation standard for this veritable stew of a language – begged, borrowed, stolen and stitched together. That standard is, to this day, RP or Received Pronunciation, and English dictionaries and online guides all give ‘default’ British English pronunciation in RP (in the US it’s American Standard). The notion of RP is a relatively recent one, yet it pervades media, popular culture and indeed, the halls of acting schools up and down the country.
When Dr. Johnson published A Dictionary of the English Language in 1755, it included very little guide to pronunciation – perhaps through lack of etymological knowledge, but almost certainly because it was very difficult to reach a consensus on recommended forms at that time. When Daniel Jones’ English Pronouncing Dictionary came along in 1916, however, RP was included as the accent “most usually heard in everyday speech in the families of Southern English persons whose menfolk have been educated at the great public boarding schools”, and the BBC had been broadcasting in it for almost 2 years. The latter would be instrumental in the ‘standardising’ of English to RP, and many linguists came to prefer referring to it as ‘BBC English’, feeling that RP, being the keep of such boarding schools and of Oxbridge, implied somewhat of an exclusive superiority. Not much has changed there, then…
Accents, though, are not fixed and unchanging. They evolve, as language does, and the ‘lend’ of early 20th century RP (just imagine the Queen saying it) is as to ‘land’ as ‘inebriated’ is to ‘wasted’. The above is a lovely example of this evolution, as the narrator is speaking modern RP, while the broadcasts are in what we would now regard as the comically plummy tones of early RP.
The question of why RP is still considered accent ‘neutral’ for performance is an interesting one, though. A much wider range of accents are now on stage than was the case around 50 years ago, but every single acting graduate is expected to have a usable grasp of RP, even if it bears no relation to where they grew up or what they naturally sound like. It is, in point of fact, estimated that less than 2% of the UK population are native RP speakers, making it very unlikely that many of us even know one. Though linguists agree that no one accent is superior to another, the myth that RP is the clearest persists at the same time as the idea that it is the sound of education, of a certain social class and certain historical periods. The latter is the most common error; RP being a relatively recent arrival in the world of English accents, Henry VIII certainly did not speak it and do not let any costume drama tell you otherwise.
With the BBC moving on to a wider range of broadcasting voices, Shakespeare productions remain one of the last bastions of the accent, and although there have been myriad interpretations and modern versions of his plays, anyone staging a ‘straight’ Shakespeare is often likely to default to RP – historical accuracy be damned.
But the reason that sections of Shakespeare don’t rhyme, or seem funny where there seems a yawning gap for a joke is because they weren’t originally written or performed by anyone who spoke RP. David Crystal (who is a DUDE. If you want to know more/anything about English, go for one of his books) and his son Ben have done several versions of Original Pronunciation, or the closest approximation we have of the way Shakespeare and his acting company would actually have spoken.
Based on this realisation, The Globe did stage an OP performance of Romeo & Juliet in 2004, but it’s still not something you’re going to commonly come across. Which is such a pity, as a] it sounds lovely (a little bit West Country) and b] SO MUCH MORE OF THE LANGUAGE MAKES SENSE.
Take, for example, Sonnet 116, which comes in at 5:51 in the video:
Love alters not with his brief hours and weeks,
But bears it out even to the edge of doom.
If this be error and upon me proved,
I never writ, nor no man ever loved.
‘Proved’ and ‘loved’ DO rhyme in OP. As You Like It has a whole pun about ‘hours’ and ‘rot’ – whores and rut in OP. ‘Loins’ has the double meaning of ‘lines’ in the opening speech of Romeo & Juliet. And it’s a faster mode of speech, with the Globe’s production clocking in at 10 minutes less in OP, so we could all be getting home a bit earlier!
But using varied accents on stage is not so straightforward a process. We have almost subconscious, early learned prejudices about certain voices, and ideas about what they connote. A character’s accent, particularly in a historical piece, is sometimes chosen to tell us something about them – perhaps their class status or profession. It’s a shorthand that we can simultaneously revile and understand, and even those who do not believe that RP codes any kind of superiority will be hard-pressed to deny that an inner-city London or Birmingham or Glasgow accent doesn’t immediately conjure up associations for listeners that you must either choose to go with or to consciously subvert. British people often, for reasons baffling and frankly offensive to me, hear my Irish accent and presume they have free licence to begin talking about the IRA…
However, when it comes to performing accents, affectation is not restricted to RP. There is no such thing as a completely correct way of speaking for everyone with a particular accent (even RP); so any time you hear an actor taking on any accent other than their own on stage, they are performing an approximation of it, and probably an extreme one. There are intonations that are correct for that accent, but no hard and fast absolutes. Next time you’re down the pub, try going around the table and all saying ‘bath’, ‘laugh’ and ‘goat’. If you don’t all end up modulating your tone to sound the same, you might be able to gather yourselves into groups that sound similar, but you will all say the words slightly differently, even if you’re from roughly the same region. It’s because our accents are a whole medley of the places we’ve lived and the sounds of people we’ve been with.
I, as mentioned, am Irish, grew up around a lot of Americans, and have lived in London for 6 years. Much like Aidan Gillen, I’m not even sure I can do my own accent anymore (unless I’m with another Irish person). Quite often, people think I’m from the West Country; the mixture of my slightly clipped London/RP and the occasional words that escape in full-blown Irish rhythm (potato, hilariously enough, being consistently one of them) are confusing to the ear.
I’m also a mimic; I absorb and reproduce the intonation of whoever I’m speaking to. Sometimes people are surprised that I have an English accent after just 6 years, yet I’ve had it since the plane touched down. If you try the pub exercise, you’ll notice that as you repeat the words to each other, you will all start to sound more and more similar. Not quite the same, unless you have as strong an inclination to mimic as I do, but you will adjust your natural mode of speech. This is yet another social mechanism, one which we actually use to bond with people – particularly those we’ve just met – as we want to sound familiar, to build empathy and understanding. Being familiar and recognisable, or at least trying to be, is part of what makes us human.
So now I have a (seasonally appropriate) confession to make. When I was about 7, my friend’s brother thought it would be funny to show us Gremlins. Even as an adult, Scream is about my coping level when it comes to horror films, so when we adjust for age, Gremlins was the most terrifying thing I had ever beheld. Fast forward a few years, and for NO KNOWN REASON, I owned a Furby. I was mostly fascinated by it, partly determined to take it apart, and only a little bit convinced it would come for me in the night.
Furby was the first toy of my generation that came close to mimicking artificial intelligence. The fact that I now carry something far smarter around in my pocket every day should probably alarm me more than it does, but for those who missed this particular 90s fad; the Furby marketing schtick was that the toy would learn from you and eventually speak more English, mimicking what you were saying to it. Even a moment’s pause for thought would have you realise that this kind of technology would not have been on sale for the price of a child’s toy in 1998, but the myth pervaded for quite some time afterwards, and even led the NSA to ban Furbies from their offices.
The truth was some (equally remarkable) clever programming, which had the Furby increase the ratio of English to made-up Furbish it spoke over time. Its microphone could only detect a single tone being directed at it, and had no ability to record speech. Furby never copied its owners, or it would have sounded less fucking creepy. Its inability to do this led it to fall firmly into the uncanny valley, an unease that that we experience when something is almost, but not quite human. This is why we naturally mimic each other and strive to sound familiar, because that which sounds alien immediately unsettles us. For example, nothing will chill your soul like having poetry read to you by this robot. It is quite a basic programme and can’t, for example, cope with the homonyms of ‘lives’ and ‘lives’ – it doesn’t understand that context alters the meaning, something a human brain can do instantly. Two artificial programmes trying to communicate with each other when they are not programmed to do so goes something like the above video. Neither can respond creatively outside their pre-sets (the Furby is only responding to the fact that there is a voice, not what it’s saying) and it’s instantly apparent that neither is sentient.
So what exactly is it that tells us someone is human? Accent and intonation are large factors. In 2013, some TIME reporters came across a telemarketer that seemed human, but quickly went on to fail the Turing Test when asked some probing questions. The recording had been done in individual, fluid sentences, meaning that ‘Samantha’ intonated words in a way that, for the most part, made sense. Her voice varied across sentences and went up at the end of questions. While she didn’t have a particularly location-specific accent, there was something about her voice that was clearly unique, and that of an individual. All plus points. Fatally, though, while she could respond to a range of expected questions (including “are you a robot?”), she couldn’t react to the tone and mood of the person at the other end of the line and remained light and breezy, no matter how suspicious and interrogative the reporter became. There were also long pauses after certain unexpected phrases, while the software scanned the spoken sentence for key words to respond to.
So as far as we’ve come, we still haven’t created anything that can convince us it’s a human being, because we spend most of our early lives learning to do so. With each iOS update, Siri gets funnier and more responsive, but that’s all down to its human creators. AI may never learn to speak as naturally as we do, with pauses, reactions and adjustments for whoever we’re talking to, but as we move towards a lifestyle more and more integrated with companions like Siri, we may find that these increasingly human sounding voices mean that we will arrive at some kind of standard robo-RP in every household. Or perhaps the American Standard speaking cylons of Battlestar Galactica, individual but bland enough to be indistinguishable from the general populace, are our linguistic future. What will this mean for regional accents that are already in danger due to stigmatisation and population movement? As much as accents can carry inherent prejudice, they are an even more important source of identity and heritage. Just as broadcasting fuelled the rise of RP, technology may well accelerate the dilution of these regional voices. Who knows, maybe the stage will be where they make their last stand, and in another 200 years we’ll see a Mr.Burns-esque patchwork reproduction of Coronation Street’s Mancunian tones.
But don’t worry, I hear the National’s 2027 Waiting for Robot is still on course. Though the machines are definitely coming for our jobs.
Siri, what is the meaning of life?
I can’t answer that now, but give me some time to write a very long play in which nothing happens.
This is the second in a series of columns that will explore the art of science and the science of art, and (hopefully) break down the idea that they are discrete and unrelated. First Thursday of every month, same bear time, same bear channel!