Hickup over the Littany

One​ of the abiding mysteries in presenting music from the past is what the singers sounded like. There is no evidence for it, apart from written descriptions, all of which fall far short of telling us anything precise. What is one to make of this description of the singing in the Chapel Royal in 1515, written by the Venetian ambassador to Henry VIII’s court and included in Andrew Parrott’s The Pursuit of Musick? ‘More divine than human; they were not singing but jubilating [giubilavano].’ The exact meaning of ‘giubilavano’ has been long debated, to no avail. Or what does this résumé of national styles, written in 1517, tell us? ‘The French sing; the Spaniards weepe; the Italians, which dwell about the Coasts of Genoa, caper with their Voyces; the others bark; but the Germanes … doe howle like wolves.’

There is much less uncertainty when it comes to the sound of past instruments. Originals can be found in museums, and modern copies of them are common. A revolution in the playing of early music kicked off fifty years ago under the banner of ‘authenticity’ – the belief that the sound the composer would have heard should inform everything we can hope to achieve. This assumption heralded a dramatic change in listening habits, and if the battles it provoked are not now remembered, it is because the essence of what was discovered has passed into modern practice. Performing on copies of old instruments produced cleaner textures. Research indicated lighter and quicker tempi, and suddenly the colours inherent in the orchestration became apparent, like the colours concealed under centuries of varnish on Old Master paintings. Singing followed suit. Romantic slush became almost morally unacceptable, when it was realised that vibrato in singing, as in playing, had gone too far by the 1960s. The only difference was that the instrumentalists were building on solid foundations, and the singers on what was sometimes no more than guesswork.

Four ensembles founded in 1973 continue to work in the light of these discoveries: the Academy of Ancient Music, the English Concert, the Taverner Consort and Players, and the Tallis Scholars (which I started as a choir in Oxford while I was an undergraduate there). The first two concentrated on instruments only; the Tallis Scholars worked with voices only; and the Taverner Consort, led by Parrott, set out to combine the two. Parrott’s problem was always going to be dealing with voices in parallel with the instruments. No one could argue that the newly interesting old instruments weren’t what the composer would have heard. But with voices everyone could argue that what they heard wasn’t right, which of course meant not to their liking. In his first recordings, Parrott was pragmatic, using the instruments that were available at that time, and guessing how the singers should sound alongside them. The purists of the time accepted his stance.

In this company of pioneering ensembles, Parrott has proved something of a maverick. His group has given fewer concerts than the other three, made fewer recordings, and is now less active, yet its early discs were the most provocative. From the outset the Taverner Consort ranged over a wide repertory. In 1984 they released recordings of Machaut’s Messe de Nostre Dame (written in the mid-14th century) and the Monteverdi Vespers, and in 1985 Bach’s Mass in B minor. These were followed by a selection of Renaissance masters, especially Tallis (1989) and Josquin (1993), and an influential recording of Bach’s St John Passion (1991). For the Bach recordings, Parrott employed no more than two singers on each of the chorus parts and used the same singers for the arias. This was a response to traditional choral society performances, which tended to deploy a full orchestra, massed choirs and operatic soloists. By restricting the use of vibrato and radically reducing the number of performers, Parrott could achieve the much admired leaner textures.

At times he was simply contrary. His anthology of Music of the Sistine Chapel (1987), by no means the first or the last to cash in on the extraordinary popularity of Allegri’s Miserere, presents it shorn of its famous high C – a variant which was first heard in the 1800s and has since become emblematic of the work. This, and the daring of allowing his singers to embellish the lines of a Palestrina motet, gives some idea of how innovative (and uncommercial) Parrott was prepared to be. Tessa Bonner’s interpretation of the solo lines in the Miserere is fascinating and beautifully sung, but it would never outsell those with the top note.

By the mid-1990s Parrott had stepped back from his role at the cutting edge of performance practice, leaving him to concentrate on writing. His Composers’ Intentions?: Lost Traditions of Musical Performance (2015) is an anthology of essays in which he does more than anyone to pin down what is known about the sounds of the past. He has followed it up with The Pursuit of Musick, a collection of thousands of eye-witness accounts covering five hundred years of performance. It may be taken either as an erudite coffee-table book or a major research tool, saving countless hours of fractured looking on the internet; either way, it is full of diverting detail. In 1707, Thomas Brown wrote that some of the singing men at St Paul’s Cathedral and Westminster Abbey ‘dayly come wreeking hot out of a Bawdy-House into the Church; and others Stagger out of a Tavern to Afternoon-Prayers, and Hickup over a little of the Littany, and so back again’. There is more in this vein, and a lot more in solid reporting, though at times one wonders whether Parrott, after all, isn’t out to persuade us that standards in the past weren’t as high as they are now.

Behind all the inexhaustible detail lies the same question: what did the singers sound like? Some choral directors – I am one of them – have been prepared to conclude that the cake isn’t worth the candle, and one might as well go back to basics, look at the music carefully and create a sound which seems to suit what the composer may have wanted, as seen through his scores, and which pleases a modern public. Parrott was never one of these. The more the sound escapes him, the more he looks. In the preface to Composers’ Intentions? he says: ‘For the most part my researches have been driven by a performer’s simple aspiration: to understand as fully as possible how composers of the more distant past intended their written works to function in performance.’ This sounds straightforward enough, but expressed across the ambitions of the authenticity movement as a whole, it runs into the common accusation that the reinvention of the past was doing no more than anaesthetising and colonising it, to give modern people a comfortable sense of nostalgia and possession. And in fussing about minutiae of detail, which might have been misleadingly reported and interpreted, it could distract performers from giving themselves wholeheartedly to a convincing, living interpretation.

One of the most persistent critics of the authenticity drive was Pierre Boulez. ‘What is authenticity anyway?’ he wrote in 1988.

The more we tire ourselves out searching for it, the more it escapes us. Minds that have not experienced the era of the works that are being reconstituted cannot possibly know their reality and produce a real reconstruction. All research into authenticity is devoted to the reconstruction of the past such as we conceive it nowadays – more or less idyllic, coloured by a golden age that never existed. The more we look into this hallowed authenticity, the more we distance a work from our understanding, forcing it into a framework removed from our reality, which drives it into the land of pure utopia.

Much of what Parrott has written, and put into practice, is even-handed. But the original vision of the revolutionaries – to base every decision on what provably occurred in an actual performance in the past – lurks unhelpfully in the background. One example has been Parrott’s obsessive determination to prove that the falsetto voice didn’t exist until the 17th century. He first launched this campaign many years ago, but included an update in the reissue of Composers’ Intentions? The starting point for this theory was that the English choral tradition has relied on men singing the entire alto range in falsetto. This lazy assumption established itself in popular thinking, along with a number of theories about the contribution of boy choristers going back to the Reformation, from when the tradition of our cathedral choirs could be supposed to proceed unbroken. To correct this misunderstanding Parrott found it necessary to remove the possibility of falsetto singing from the picture altogether, since he could find no written evidence for it. Yet I have no doubt that falsetto (or ‘head voice’ as opposed to ‘chest voice’) was used when found convenient, as pop singers do today. When they reach a note that is uncomfortably high, they flip into head voice, just for a note or two. We all do it in our private crooning; and just as no one thinks to write a description of it now, nor did they then.

But to Parrott, for whom everything has to be itemised and substantiated, no evidence means it didn’t happen. This lack of flexibility was one of the central faults of the whole early music revolution, but it’s rare to find it when dealing with voices, since there isn’t much to be inflexible about. Parrott’s theory is an unusual one, and the practical effects of it can be heard in his recordings of a cappella English polyphony. In these, the countertenor parts are sung by high tenors, who can be over-dominant in the ensemble, forbidden to access their falsetto for the highest notes, as I believe their predecessors would have done. Maintaining such a tessitura over long spans of music is arduous, even for the bionic singers of today. However, this detail, small but significant, is nothing compared with the use of women instead of boys on the top parts, as Parrott’s recordings have it. If there is one mistaken belief still in currency today, it is that women can (or would want to) sound like boys. On the one hand, Parrott insists that avoiding falsetto is ‘authentic’; on the other, he has women singing parts that were undoubtedly sung by boys. To my ear, his polyphonic recordings have been strung up by theory.

But is the literal re-creation of concerts of the past so desirable? Boulez gives an interesting example, not from centuries ago but from his lifetime:

It is not clear that one would really be pleasing the composer … by re-establishing performance circumstances that could never have been entirely satisfactory … Stravinsky … asserted the unique documentary value of his own recordings and maintained that future interpreters should study them and be obliged to refer to them. Unfortunately, though, his precarious gifts as a performer, the circumstances and time pressures under which the recordings were made and the quality of the forces at his disposal do not let us regard this evidence as any sort of absolute model. In any case, can there be such a thing? Every interpretation conveys an essentially transitory truth.

This doubt can be extended back to what we know of the Sistine Chapel Choir in Palestrina’s time, when it was the premier choir in Christendom. What indeed would Palestrina say, given the standards of the choir that habitually sang his works at the time, if he could hear modern performances? His music, so perfectly formed, so gleaming, cries out for the kind of choral discipline which is rare today, but must then have been non-existent. ‘What is surprising, perhaps, is the number of papal singers throughout the 16th century who were thought by their contemporaries not to have been competent,’ Richard Sherr reports in Music and Musicians in Renaissance Rome and Other Courts (1999). ‘The adjectives used to describe them include “harsh” (aspra), “hoarse” (rauca), “dissonant” (disona: “untuned”) … and they are occasionally associated with the noun imbecillitas (“weakness”).’ Many of them were routinely ill, or absent, some were very old but couldn’t be sacked, and some had been admitted without taking an audition. ‘In short,’ Sherr concludes, ‘we may really not want to hear the music the Sistine Choir sang in the age of Palestrina in the way that they sang it.’

One way​ of measuring the way singing styles can change over a period of time is to listen to the earliest surviving recordings of singers, held in the National Sound Archive. The very first date to the late 19th century, but 1902 seems to have been a particularly busy year. It was then that Enrico Caruso made his first recordings (in a hotel room in Milan); the choir of St Andrew’s, Wells Street (then considered the best church choir in London) made the first gramophone record of choral music; and the only recording of a castrato was made. The rudimentary technology and lack of editing limit the experience somewhat, but there are modern ways of cleaning up old recordings, and there is no disguising how wide the gulf is between what was considered normal then and what now. Caruso is notable for having almost no vibrato, just very powerful high notes allied to an approximate sense of tuning in the lower ranges. The choir of St Andrew’s is even more distant from current practice, the men often favouring a dying, sentimental fall at the end of a phrase, and throbbing delivery in the approach, habits perhaps learned from operatic practice; the boys are fascinating in their pronunciation of the English texts, which radically affects the ensemble sound and makes it quite different from our own.

But at least these recordings are early examples of traditions that have continued. One sound from the past we will never hear again is that of the castrato. The recordings we have were made by Alessandro Moreschi in 1902 and 1904, when he was 45 and reportedly past his prime. Opinion has been divided over the usefulness of these performances, though I find his version of the Gounod/Bach Ave Maria one of the most astonishing pieces of solo singing I’ve ever heard. Much of the negative comment is misinformed. Cleaned up, the recordings show a strong, flexible voice, fully under control. Like Caruso, Moreschi disdained vibrato, though he had a way with swoops and portamenti (sliding from one note to another), which were developed to make the most of the acoustics in the Sistine Chapel, where he spent his career. No doubt these were welcome in his performances of the high soprano part in Allegri’s Miserere, a piece he made his own, though unfortunately never recorded. Unlike Caruso, his delivery is without cloying emotional charge: in the Gounod/Bach he ascends easily to a high B natural without any melodramatic effort. Most opera singers would make a meal of such a peak note these days. Such a ghostly, pure voice, coming to us from long ago, is beautiful. The film Farinelli (1994) attempted to re-create the sound of the greatest of all castrati, Carlo Broschi, known as Farinelli, by recording the voices of the Polish soprano Ewa Matas-Godlewska and the American countertenor Derek Lee Ragin separately, and then digitally merging them. It proved that a voice from the past can’t yet be resurrected.

Other examples of our muddled thinking about ‘authentic’ sound have to do with the relentless updating of trends, which never go back on themselves. Caruso would probably always have been a star, but his lack of vibrato, and the lack of emotional charge associated with it, might make him seem two-dimensional to a modern audience. The high period for vibrato came in the decades after the Second World War – the Maria Callas, Kirsten Flagstad days. The degree of vibrato favoured has fallen again since then, though it has new expressive qualities that modern listeners appreciate, however foreign they would have been to an artist of a hundred years ago. If trends can change that much in a century, how much more must they have changed in five hundred years? The key here is how much the pronunciation of English (and with it Latin) has shifted. The boys on the early recordings have clean, almost Germanic vowel sounds, with clipped over-accented consonants, probably developed by singing in over-reverberant acoustics. We know that vowels in spoken English changed dramatically in the period up to 1700, making the sound of earlier singing unrecoverable. And consider that before the 17th century performances of polyphony were customarily held in quite small chapels, not cathedrals, by people who might be more tuneful than you or me, but who sang very like us, their voices untrained by modern standards, not needing to project.

The success of the early music revolution has come from the fact that in creating an idealised view of the past it has appealed to contemporary audiences. I have no doubt that if we were really to re-create an evening with an 18th-century orchestra, or a service with a 16th-century choir, we would be horrified by the standard of performance, and disgusted by the smells. Inevitably the founding dogmas have been watered down. There was a time when the demand for period instruments extended as far as Brahms and Schumann. Now, to judge from the Proms this summer, Beethoven is fair game for traditional massed choirs and modern-instrument orchestras, though Mozart, interestingly, is approached with greater caution, and Bach even more so. To make a fuss about the instruments Brahms would have heard is nowadays considered quirky, though a modern symphony orchestra still wouldn’t dare perform Bach’s Brandenburg Concertos. The job is half done, and has found its niche. My regret is that the time and care that was required to master old instrumental techniques haven’t been extended to past vocal techniques, like high-pitch soprano singing or the exact methods of the old ‘countertenor’. Singing is still taken for granted, hardly thought to be in need of new study.

But the intermingling of old and new has meant that many lessons have been assimilated by modern players, to the extent that students now take them for granted – valveless horns, for example, or ‘straighter’ singing, even on the operatic stage. The festivals that advertise themselves with the moniker ‘early music’ are in reality just leaving out the compositions of the last two hundred years, while not worrying too much about ‘authenticity’ for the rest. The most diehard of these is the Utrecht Early Music Festival, which celebrated its fortieth anniversary last summer. In its early years I remember concerts starting at 10.30 p.m. and running long past midnight, when performers and audiences together would reassemble in the Polmanshuis, where, in smoke-filled rooms, the artists would be questioned by students sitting on the floor. Nowadays the organisers don’t dare hold concerts after 9 p.m., knowing that their audience will soon want to be in bed. I wonder whether the devoted people I see in the audiences today are the same students of forty years ago, whose children, missing the first wave of enthusiasm, have never showed much interest in joining in. These older audiences are loyal and numerous. But you are left with the impression that old music, when presented narrowly, is for old people.