Palestrina - The Tallis Scholars, Peter Phillips

The Tallis Scholars sing Palestrina
The Tallis Scholars, Peter Phillips
Gimell CDGIM 204

Palestrina is the composer The Tallis Scholars have sung and recorded most frequently. There need be no surprise in this: the quality of his music merits all the fame it has been accorded over the centuries, making him probably the most talked-about writer in the history of Western classical music (Beethoven, Mozart and Wagner are possible rivals in this, but with them the process hasn't been going on for so long).

But there is more to Palestrina than his ability to write a masterpiece every time he sat down to compose: his style makes demands of its performers which no other composer quite made. We have come to realize that if a group can sing Palestrina well it can sing any choral music well, for in his music there is no hiding-place. The sonorities are so clear, the logic of the writing so compelling, that one sound out of place is immediately detectable; and a blemish is more serious in music which depends on sheer sound for its impact than in more pictorial or rhetorical compositional styles. With Lassus and Byrd, for example, interpretation through the words alone will go a long way to producing a convincing performance; Palestrina requires his performers to think more carefully about the sound itself. The nearest comparison is with the passage-work in Mozart's piano music, which is equally so clearly and logically conceived that a stray note can acquire a disproportionate influence. Just as a pianist must rehearse scales and arpeggios to play Mozart well, so a vocal ensemble must work on blend and tuning to sing Palestrina well. There is no better or more rewarding way of learning how to sing Renaissance polyphony.

These two discs contain some of the best of our view of Palestrina, from the very first record we made commercially (in 1980), to one of the most recent (entitled ‘Lamenta' and released in 1998). In general we have concentrated on his Mass settings - there are three more in our catalogue as well as another version of Papae Marcelli - not least because they make such effective concert music. Palestrina had an unusual ability to write positive, outward-going, major-key music which, over the length of a Mass-setting, is a great strength: penitential writing tends to be more effective in shorter bursts. Three of the four settings in this collection rely on bright sonorities; the fourth - Sicut lilium - is more subdued and sensuous, as the words of the motet require. The Lamentations, on the other hand, show a completely different side of Palestrina's art. By dividing the text into short movements (and leaving out the Hebrew letters which are traditionally set as part of the whole) he was able to deliver a series of plangent statements worlds away from the mood of Papae Marcelli and Assumpta est Maria.

The Missa Assumpta est Maria, based on his own motet of that name which in turn is based on a short phrase of chant, is a classic example of sonorous Palestrina, its excitement achieved in large part by doubling the sopranos and tenors. The brighter sound this high-scored six-voice (SSATTB) choir produces is then emphasized in the style of the writing, which is more chordal than usual. The parody motet Assumpta est Maria shows the way in its opening bars: the three upper voices are grouped against the three lower ones in easily audible antiphony. This late-Renaissance method is then transferred to the Mass, most obviously in the Gloria and Credo but also in the first Kyrie, where greater elaboration was more customary. The sheer verve of this music has ensured that, along with Papae Marcelli, Assumpta est Maria has remained the most performed of all Palestrina's 107 Mass-settings.

Sicut lilium by contrast is an early work, based on a motet which was published in the composer's First Book of Motets in 1569. By contrast with the trumpet-like writing of Assumpta est Maria, the musical style here is more for string quartet, or viol consort: lucid, the working of the counterpoint at times very elaborate indeed. This can be heard at the beginning of the motet, where the point is treated to two complete expositions, from where it is transferred to the first Kyrie. Eventually all the movements except the Gloria open with imitative counterpoint. But underlying all this finely crafted polyphony is the perfumed atmosphere which the words of the motet (from the Song of Songs) originally inspired in Palestrina, and which, despite the long sentences of the Gloria and Creed, he managed to transfer to every corner of the Mass itself.

The second disc opens with a six-voice (SSATTB) set of Lamentations, originally the third lesson on the Saturday of Holy Week. Palestrina obviously felt drawn to the Book of Lamentations: he left nearly sixty individual verse-settings from it. Standardly each of these would begin with the initial Hebrew letter preserved in the Bible, but in this set they are missing. However the customary opening words are retained, as are the deeply emotive final ones: ‘Jerusalem, Jerusalem, turn to the Lord your God'. It seems that no matter how many times Palestrina set these words he found something new to say through them.

The Missa Brevis (for four voices SATB), although relatively ‘short' and straightforward in musical style, shows Palestrina's idiom at its most accessible. It was a success from the start, being first published in 1570 and reprinted repeatedly up to 1620, since when there have been countless modern editions. Despite this, no one has been able to say what gave the music its starting-point. The most likely candidate is chant, rather than a polyphonic model as the other Masses in this set have; but if it was chant, the melodies are not applied very consistently. In fact the title ‘Brevis' may indicate that there was no pre-existing material - rare in Palestrina's Masses - this title filling in for the missing name of the original. The music is anyway not that short, being conceived in a fluent idiom which seems to glide effortlessly through the longer texts, until it culminates in the second Agnus. Here Palestrina abandoned any pretence at being straightforward and wrote a canonic movement which is a model of the form. The canon (a tune sung first by one voice and then by another some beats apart) is between the two soprano parts (a new one having been added to the ensemble just for this movement), making it as audible as anyone could wish. As with the best canons it seems as though it can never end.

The Missa Papae Marcelli also culminates in a canonic second Agnus, but here both the build-up and the canon itself are very different. This is the Mass which was said to have ‘saved church music' by proving to the cardinals at the Council of Trent that words set to music could be audible even in polyphony. Dedicated to Pope Marcellus II, who had reigned for three weeks in 1555, it was probably written in 1556 and therefore dates from the years when that argument was raging. However it is hard to believe in this story entirely. Although the style of the music is syllabic at times, especially in the Gloria and Credo, it is not the case that the words are consistently more audible than in other settings of the period, something anyway made harder by using six voices (SATTBB) rather than four. Indeed the canonic second Agnus (now for seven voices SSAATBB) is one of the most complex movements of the High Renaissance, its mathematical density unequalled elsewhere in Palestrina's output. What one can say, however, is that the setting as a whole is an unparalleled masterpiece, of the kind one returns to again and again over the years. The Tallis Scholars gave their first performance of it in March 1977 and their 75th in June 2004. I hope there will be countless more.

Peter Phillips 2004

Thomas Tallis - The Tallis Scholars, Peter Phillips

The Tallis Scholars sing Thomas Tallis
The Tallis Scholars, Peter Phillips
Gimell CDGIM 203

A specially-priced selection of previously-issued recordings. This recording of Tallis's Spem in alium was featured in "Soul Music" on BBC Radio 4.

The supposed birth of Thomas Tallis in 1505 - the date is largely conjectural - gives us the last opportunity to celebrate him for many years. By 2035 - the 450th anniversary of his death in 1585 - one guesses the scene may be rather different. So I feel encouraged to feature our eponymous composer's work in the concerts we shall give during the 2004/5 season, and to release an anthology of the music we have recorded. It is perhaps worth recalling that The Tallis Scholars launched their career in London with four all-Tallis concerts in 1977/8; and made their English Anthems recording, much of which is included here, in 1985, alongside anniversary concerts in the Wigmore Hall, the Queen Elizabeth Hall and the Utrecht Early Music Festival.

My view of Tallis's genius has only deepened with time. Not only was he the arch-survivor but also, unlike those who trim and so build their monuments on shifting sands, he had the ability to create masterpieces in whatever style was the currency of the day. This should not be underestimated, because those styles changed out of recognition during his eighty-or-so years. First it was the traditional Catholic style of Henry VIII's reign; then it was the most severely chordal Protestant style of Edward VI's reign; then it was back to Latin and Catholic writing again under Mary, though this time in a more mature idiom than in Henry's reign - Tallis was by now turning fifty; then it was the compromise style for Elizabeth whom he served for twenty-six years and who left him sufficiently alone for him to produce some of his greatest music.

It was not considered desirable on these two discs to present Tallis's music according to any chronological sequence, but the four styles outlined above can be followed clearly enough. Disc 1 starts with the exception to every rule - indeed so outstanding is Spem in alium that it still seems impossible that one mind without a computer could have managed it. To write for forty voices which do not repeat themselves in consecutive motion and not to lose control of the whole colossal edifice, is to set a challenge which even the Art of Fugue scarcely rivals. The actual compositional style of it is slightly blurred between those characteristics implied by stages three and four above - sometimes imitative between (some of) the parts, sometimes setting the text syllabically, never dealing in the unrestrained melismas of much of his purest Catholic music - and so it is not fully established whether Tallis wrote it for Mary or Elizabeth (both of whom celebrated their fortieth birthdays whilst on the throne) or for some more abstract reason, perhaps to do with the Biblical number 40. But for us in our modern terms, as for Tallis himself, Spem remains the ultimate technical challenge - supremely difficult to bring off, supremely rewarding when one comes near.

Sancte Deus is a classic example of Tallis's first style, illustrating what I mean above by ‘unrestrained melismas'. A melisma is a melodic line which only uses one syllable, like the ‘A' of Amen, allowing the composer's imagination to fly free of text-setting. This essentially abstract way of thinking was admired by the pre-Reformation Catholics, and needless to say was particularly objected to by the Protestants. The Salvator mundi settings (the second much less famous than the first) were Elizabethan and so more compact; but Gaude gloriosa is one of the most elaborate Catholic compositions of the entire period. Unlike Spem it is colossal in length rather than height, using the nine exclamations of ‘Gaude' in the text to work up a construction which is essentially architectural. The music flows from one scoring to another to yet another, never using more than six voices at any given moment, but with such an exquisite control of melody and sense of overall direction, that the final pages feel as if the listener has just completed the journey of a lifetime. It comes as no surprise that Gaude gloriosa was influential - William Mundy based his Vox patris caelestis on it - and would have been more so if the Catholic style hadn't been so soon overturned by Elizabeth's accession.

The seven-voice Miserere nostril is both a demonstration of technical skill, and, in its music of the spheres way, possessed of an unearthly beauty. It is a canon six in two with a free tenor, which is to say there are two canonic melodies, one sung by the two top parts which is easily audible, while the other is shared between four of the other voices. This second canon has its four contributors starting at the same time, but going off at different speeds (the first countertenor has the model melody which the second countertenor sings in double augmentation, the second bass sings this melody inverted and augmented and the first bass has it inverted and in triple augmentation). Both this piece and Loquebantur variis linguis are scored for SSAATBB and are probably Elizabethan.

All the remaining pieces on the first disc were written in Tallis's second period - for the Protestants of Edward VI's reign. Nothing could be further removed from the glories of Spem or Gaude gloriosa. Gone are the melismas, the Latin texts, the interweaving of the lines in polyphony. The accent was now on simplicity and comprehension - hence the English texts and the chordal style, which was designed to make the words audible. One may think one knows what Tallis must have thought of this clipping of his wings, but at least he was not a man to sulk. His craftsmanship enabled him to adapt swiftly to the new realities and in a matter of a year or two he wrote some of the best-known and best-loved Anglican music there has ever been. Not all these tiny masterpieces are as famous as If ye love me, but they all bear repetition, as Vaughan Williams thought when he based his Fantasy on a Theme of Thomas Tallis on the third Tune for Archbishop Parker (part of track 15).

Disc 2 is made up entirely of Latin-texted music from the first and last periods listed above - from the reigns of Henry VIII and Elizabeth I - apprentice works in the style of Tallis's immediate predecessors alongside some of his maturest thoughts in old age. Tracks 1 to 5 are from Elizabeth's reign, tracks 6 to 8 from Henry's.

Nowhere is Tallis's mature style more perfectly on display than in his two sets of Lamentations, which in modern times have been the yardstick by which every set of Lamentations of the period is judged, whether English or from the Continent. As was standard, Tallis divided his settings between the Hebrew letters which preface each verse of the text in the Bible, and the lament itself. Although Tallis's Lamentations I and II are stylistically identical they are in different modes, which strongly suggests they were not intended to be sung together. The next three pieces are beautiful examples of Tallis's late style, with O sacrum convivium perhaps the most renowned. In each case he took the prevailing Flemish technique of imitative entries between the voices, built up a full sonority as each voice joined in, and then moved on to the next phrase of words and the next set of entries. It is a transparent idiom in which the words are set more or less syllabically - thus fulfilling the Protestant need for clarity - yet the music is allowed to expand and breathe.

The last three pieces show Tallis learning his trade. The obvious influence here is John Taverner, though in the background one can hear the music of Robert Fayrfax, John Browne and William Cornysh. Salve intemerata is Tallis's longest single-movement piece. Though it lacks the contrasts and sheer verve of Gaude gloriosa, it shows the same instinctive grasp of musical architecture, whose bricks are melodic lines of memorable fluency and grace. Of all the Tudor composers, one comes away from a big Tallis piece humming the tunes. By keeping strictly to the Phrygian mode throughout this colossus (except once in the ‘Amen'), Tallis gave himself no harmonic place to hide, and so was obliged to develop his melodic sense. The same is true of the very early Magnificat (which does creak a bit here and there - Tallis never revisited such banal moments as the downward scale at ‘et sanctum nomen eius') and Ave, Dei patris filia, which was probably written in the 1530s. The obvious influence on this piece is Robert Fayrfax, whose own setting of the text formed a model for Tallis. Indeed, so closely did Tallis follow Fayrfax that David Skinner, who recently discovered enough new source material to make the piece reconstructable, was able to decide where to add missing parts by copying the layout of Fayrfax's piece. Even if Tallis was no more than thirty when he wrote Ave, Dei patris filia, it includes some of his finest phrases, for example ‘semper virgo Maria' just before the ‘Amen'. To go from this to the stark directness of If ye love me, to the soaring phrases of Gaude gloriosa, to the compact, refined world of the Lamentations and O sacrum convivium, to the simply incredible sonorities of Spem in alium is to travel as far as one man can ever have taken his listeners. Living a long time does not in itself explain how Tallis did it.

Peter Phillips 2004

Josquin - The Tallis Scholars, Peter Phillips

The Tallis Scholars sing Josquin
The Tallis Scholars, Peter Phillips
Gimell CDGIM 206

‘In his unprecedented stature and his undisputed pre-eminence in the eyes of his contemporaries and posterity, Josquin has never failed to remind recent historians of Beethoven, who was similarly regarded 300 years later, and who retains a similar quasi-legendary aura' (Richard Taruskin).

If one were looking for a superstar among Renaissance composers - and identifying such people does no harm to the subject as a whole - Josquin is unquestionably the front runner. He was a star in his lifetime, travelling more widely, being paid better and having more desirable employment than anyone else; and he has become a star again more recently. It is true that in the centuries in between it has been Palestrina and Tallis who were performed more consistently, where Josquin was forgotten, but this was on account of their simple music, which choirs of any ability could sing. Josquin didn't write any simple music. All his music is complex, intellectually and vocally, posing problems which have only recently been found to represent a supreme challenge. As with Beethoven it is now recognized that facing up to Josquin's message can bring unparalleled rewards; and, more than anywhere else, it is now clear it was in his Mass settings that he expressed that message at its most fluent.

There are perhaps fifteen Mass settings by Josquin, all of which are essentially scored for four voices. This scoring in itself distinguishes them from much of his other writing, since elsewhere he could delight in fuller sonorities. Praeter rerum seriem (track 12 on the first disc of this collection) proves the point. Here is a motet for six voices whose opening bars have the sonority of an entire string section in a Romantic orchestra (this passage always makes me think of the funeral march in Mahler's first symphony). Such richness is rarely to be found in the Mass music, where instead Josquin concentrated on sparse detail, intense dialogue between the voices, the working and reworking of tiny melodic ideas: more string quartet intimacy than string section grandeur. For this, four wide-ranging voices were the ideal medium, able to be used in duets and trios as much as all together, without forcing too great a change in the impact of the sound. The one proviso was that the ranges should be wide, so that the musical meetings could take place, as it were, anywhere: high, low, at the unison or octave at will. These ranges are part of the challenge for modern groups - no conservatoire will train its singers to have the lightness of touch Josquin demands over nearly two octaves - but the challenge has been found to be worth shaping up to, and a modern solution to it has increasingly been found both in Europe and America.

The process of searching out such a method was significantly advanced when the two Masses included on the first disc of this collection won the Gramophone Record of the Year Award in 1987 and were widely played. It helped also that they were two of the finest of all Josquin's compositions. The Missa Pange lingua may have been his last Mass setting, even his swan-song, and it was not published until after his death, in 1539. Certainly it shows some of the characteristics associated with ‘late period' compositions, not least by Beethoven: the relaxing of purely mathematical ways of writing in favour of freer ideas, often in the style of a fantasy. The Missa Pange lingua has been called ‘a fantasy on a plainsong' (by Gustave Reese), in which Josquin broke the Pange lingua melody up into smaller phrases, motifs and rhythmic units, which the voices explore in just the kind of polyphonic dialogue I mentioned above. Nowhere else in the repertoire is this endlessly supple style managed so perfectly. After spending almost the entire setting being pulled to pieces and examined from different perspectives, the chant melody - which was originally sung as a hymn for the feast of Corpus Christi - can finally be heard complete, for the first time, in the soprano part of the third Agnus Dei: a true culmination of the whole work.

The Missa La sol fa re mi was published in 1502, making it a relatively early work. Here Josquin seems to have been fascinated by what could be achieved with the most restricted kind of mathematical framework, though that was not unusual for him at that time as will be seen below. Virtually the whole Mass is derived from the single five-note phrase which the medieval notes la, sol, fa, re and mi yield in the modern scale: A, G, F, D and E. This motif may be heard in different note-lengths and occasionally in different pitches in one or other of the parts, though it is mostly to be found in the tenor. In all there are perhaps two hundred repetitions of this melodic idea, culminating this time in the first and third of the Agnus Dei settings where its note-lengths get shorter and shorter, intensifying the mood of other-worldly experience which the music has gradually built up.

Of the two motets which make up the remainder of the first disc, the Ave Maria comes closer to the sparse musical style of the Masses. This simple but enormously influential setting of one of the central texts of the Catholic faith is scored for four voices, but unlike the Masses does not adopt an argumentative musical idiom. Instead, much of the writing is for duets and trios, made up of spacious musical phrases and uncomplicated sequences. Here the culmination of the piece is not an intensifying of anything musical, but an exceptionally long-breathed final passage, made up of some of the least hurried chords imaginable (to the words ‘O Mater Dei, memento mei. Amen' - ‘O Mother of God, remember me. Amen'). By contrast Praeter rerum seriem is a six-voice Christmas motet, of arresting sonorities and intricate musical detail. It is based around the melody of a devotional song. For much of the piece the polyphony is presented antiphonally between the three upper voices, when the song is in the first soprano, and the three lower voices, when it is in the tenor. The second half of the motet is less dependent on this melody than the first, becoming a more consistently six-part texture, eventually breaking into triple-time where the text makes reference to the mystery of the Trinity.

The two L'homme armé Masses of Josquin, which make up the second disc of this set, at first sight seem to be worlds apart: one might guess that Super voces musicales was a medieval composition and Sexti toni a mature Renaissance one. In fact the manuscript evidence suggests that they were roughly coeval; and they were published together by Petrucci in 1502 (in the same collection as the Missa La sol fa re mi). In choosing to paraphrase the popular L'homme armé melody, Josquin was contributing to a tradition which was already several decades old, which would continue for many more, and which would finally yield thirty-one settings by composers across the whole of Europe.

The title Super voces musicales indicates that the melody is quoted in turn on every note of the hexachord, almost always in the tenor part. The complications inherent in this are fascinating to follow. The ascent starts on C in the Kyrie, proceeds to D in the Gloria, to E in the Credo, F in the Sanctus (given again, complete, in both ‘Osannas'), G in the first Agnus Dei (incomplete) and A in the third (by which time it has at last become too high for the ‘tenors' to sing and has been transferred to the top part). The only sections to be completely free of it are ‘Pleni sunt caeli' in the Sanctus, the Benedictus and the second Agnus Dei, of which the two latter are mensuration canons for two and three voices respectively. The second Agnus Dei is made particularly complicated in that the top part is given the canon in triple time against the different duples of the two parts beneath it. The second halves of the Gloria and Credo (beginning at ‘Qui tollis' and ‘Et incarnatus est') are based on the melody in strict retrograde, with the Credo containing one more statement of the melody, the right way round, from ‘Confiteor' in a syncopated rhythm. This kind of mathematical framework for a four-voice Mass is what calls medieval compositional practice to mind; such things were much rarer in the sixteenth century, and are not found in the same way in Josquin's Sexti toni setting.

His Mass Sexti toni (‘in the sixth mode') is so called because he has transposed the melody to make its final note F (as opposed to the more normal G), giving it a major-key tonality. Much of this music has a more relaxed air than the other Masses in this set, though en route Josquin can be heard trying out new speeds, new rhythms and new scorings for the tune, now complete, now with a few notes used as the basis for an ostinato pattern or a canon. However the wide overall range of the four voice-parts brings to the writing the kind of sonority which is associated with Palestrina and the High Renaissance, rather than the more cramped textures of Dufay and Ockeghem, and the general impression is one of a broader sweep. The only exception is the final Agnus Dei which not only adds two new voice-parts, making six in total, but adopts a compositional method which certainly harks back to the ‘medieval' world of Super voces musicales. Taking an idea he used in that setting - of quoting the L'homme armé melody forwards and backwards in consecutive statements - Josquin here quotes it forwards and backwards at the same time. These statements form the lowest two parts of a six-voice texture, above which the upper voices revolve in two paired canons at the unison. This creates a sound-world all of its own - which since this recording was originally released in 1989 has been much discussed - reminding listeners not only of Josquin and the nascent sixteenth century but also of the methods of such modern minimalist composers as Philip Glass. It is a definition of super-stars that they are not only profoundly of their time, but through that profundity acquire a relevance for all time. In the last Agnus of his Missa Sexti toni Josquin proves that maxim gloriously.

Peter Phillips 2006

William Byrd - The Tallis Scholars, Peter Phillips

The Tallis Scholars sing William Byrd
The Tallis Scholars, Peter Phillips
Gimell CDGIM 208

England has never produced a greater composer than William Byrd. For range of expression he towered above his contemporaries, with only Tallis, already middle-aged when Byrd was born, in the same category of achievement. It stands to reason therefore that for a group such as The Tallis Scholars, dedicated to exploring Renaissance polyphony, Byrd's music has been central. We have made two discs entirely devoted to him; included his Tribue, Domine on our Live in Oxford collection; included his Lullaby on our Christmas Carols and Motets anthology; and made a BBC TV programme exploring his life and work, recently released as a DVD-Video. His music has featured in over a third of our 1,500 concerts. This double CD has been put together from the best of our recordings.

Anyone listening to this survey of Byrd's church music will be struck by the fundamental difference in outlook between his Protestant and his Catholic writing. The former has English words and a style which, at least in theory, was simple enough to ensure that those words could be heard; the latter has Latin words coupled to an elaborate compositional method which referred back to the kind of music the Reformation had hoped to put a stop to.

Although it is true that this division between simple and complex was not as firmly maintained by Byrd's time as it has been in Tallis's (Queen Elizabeth was less doctrinaire than her predecessors), it still explains why the two repertoires do not resemble each other. The message of the Latin-texted Masses, for example, seems to be directed inwards, towards contemplation through melody, suitable for a family circle; which contrasts with the manner of the Anglican Great Service, where the lines push outwards through studied declamation of the texts, suitable for performance before a big crowd.

The first disc in this set is devoted entirely to Byrd's Catholic output, beginning with perhaps the most significant, and the most daring, pieces he ever wrote. Nothing is more essentially Catholic than settings of the Mass, a point which would not have been lost on Elizabeth I's secret police, dedicated as they were to tracking down and harassing believers in the old religion. No one had set these texts in England since Queen Mary's reign some decades before (and would not do so again for three hundred years), so Byrd's trilogy stands isolated in time. But the really daring part of the story is that he published this music, admittedly in small volumes without title-pages, but with his name clearly given. Having taken such risks it is not surprising to find that the music itself is deeply expressive. The four-part Mass is perhaps the most personal as well as the earliest of the set, almost certainly written in 1592. It retains some techniques from the distant past, such as blurring the boundaries between the tenor and alto parts, yet there are moments of intensity - like the ‘dona nobis pacem' - which Byrd never surpassed in all his later music. The three-part Mass seems to have come next, using just alto, tenor and bass voices, an unusually restricted ensemble even for someone who was writing for small recusant choirs. Clearly Byrd relished the technical challenge involved, while fulfilling the scheme of three-, four- and five-part settings. The five-part was written last, probably by 1595. It seems the most mature of the three, both in the confidence with which Byrd handled his texts and in remaining within ever more focused boundaries: the Kyrie and Gloria are shorter than in the four-part. But this concision means that when Byrd does repeat a word, as at ‘Agnus Dei', the impact is unforgettable. Of the two motets which conclude the first disc Ave verum corpus belongs to the world of the Masses, having been published in the 1605 Gradualia but probably written in the 1590s. Like the Masses it was written for recusant choirs to sing, and similarly aims for deep expression of a fundamental text while creating a rapt, almost mystical atmosphere. Infelix ego, by contrast, is an earlier work which deliberately explores older techniques. Here Byrd is reaching back to the votive antiphon method of Taverner and Tallis, which he was too young to practise when it was standard. The long melodies, the duets and trios, and above all the use of the high treble voice, show him trying his hand at a great but lost art.

The second disc begins with three Catholic motets and an Anglican anthem which were included on the BBC television programme made in 2002 about Byrd's life (which also includes a filmed version of the four-part Mass different from that included on disc 1). All three of the motets were published in Byrd's 1589 Cantiones Sacrae, which indicates that they were written before he left the Queen's service to live with his recusant community, and its few singers, in the countryside. These motets are written on a much grander scale than those on disc 1, probably for the much larger choir of the Chapel Royal, where Elizabeth allowed Latin to be used. Vigilate is an essay in madrigalian word-painting, one of the most exhilarating there is. Tristitia et anxietas and Ne irascaris, Domine are elaborately penitential pieces, both reckoned to be expressing a deep longing for the country to return to Catholicism but in sufficiently veiled terms for the court to be able to swallow it. There are passages in both these pieces which have become almost iconic in recent years, of which ‘Jerusalem desolata est' at the end of Ne irascaris, Domine is perhaps the best known.

With Prevent us, O Lord we abruptly enter the world of Byrd's Anglican music. Here is music which evidently has a functional purpose: to tie like-minded people together in unaffected, down-to-earth expressions of piety. So that the words may be heard, much of the writing is chordal. The mood is no longer mystical, nor nostalgic for a lost past. This analysis also holds for O Lord, make thy servant Elizabeth, ‘A prayer for the Queen' as the manuscript source puts it. Affectionate and beautifully crafted, it lacks the one-to-one intimacy of the Masses.

This can also be said of the remaining pieces on disc 2. Even though they cultivate a more elaborated musical style, one has the feeling that O God, the proud are risen, Sing joyfully unto God and The Great Service are public gestures, built on the same principles of straightforward, non-tactile worship as the chordal prayers. In this company O God, the proud are risen is relatively little known, setting verses from Psalm 86: on stylistic grounds it may date from the late 1580s. Sing joyfully unto God, Byrd's most popular anthem both in his lifetime and today, is a work of full maturity from the 1590s. However there is little even in this substantial work to prepare us for The Great Service, Byrd's largest single composition.

This most impressive of all service settings thrills the listener as much by its rich scoring as by its sheer length or craftsmanship. Perhaps the first four of the six movements show Byrd coming to terms with his broad canvas, with his chosen ten voices (with no fewer than four altos) divided into two five-part choirs. But by the Magnificat the style is splendidly set, each section typically beginning with ‘Protestant' chordal writing before shaking free into ever more intricate imitation. Every verse of these six movements is set with its own scoring and its own character: one which always sticks in my memory begins at ‘as he promised to our forefather Abraham' in the Magnificat, wonderfully scored for AAATB.

Many people have said that Byrd did not really engage with these Anglican texts; yet he engaged with them enough for his English-language settings always to have found favour with the general public. No doubt he sympathized deeply with the Catholic texts at a personal level - he was a fervent believer - but such a reaction does not guarantee effective communication. Our ears are different from those of his original listeners. But one thing is certain: in these Anglican masterpieces he joined Tallis, Sheppard and eventually Weelkes, Gibbons and Tomkins in fashioning a new repertoire which has proved more durable and been more loved than anything else from the entire period.

Peter Phillips 2007

Requiem - The Tallis Scholars, Peter Phillips

Requiem
The Tallis Scholars, Peter Phillips
Gimell CDGIM 205

Settings of the Requiem Mass are among the most frequent requests for concert music. This may seem unlikely, given the subject matter, but in fact it is just that subject matter which makes them so compelling. There is a drama inherent in the text which never fails to move audiences, having, in the first place, brought out the best in the composer. It is not a modern kind of drama such as we are used to seeing in the cinema or on television, but rather of the opposite: of the light which is shining on the deceased (whose body would have been present in the original performances), of the immediacy of heaven, of the peace which death will bring. Put in words this may sound a bit far-fetched, but from the split second that the opening ‘Requiem aeternam' chant is heard, every listener is inevitably transported. It is a classic instance of the power of music over every other art-form to communicate without reserve.

This drama gradually moves through different stages as the music proceeds. The essential mood is the one of the opening - long-held chords inviting the contemplation of eternal rest. This is the Requiem's alternative to the atmosphere of desperation, noise and betrayal which underpins so many television thrillers. It returns at regular intervals - in the Gradual, the Sanctus and Agnus Dei, through the promise of perpetual light in the Communion - but is interrupted in the Offertory and the Responsory by the thought of what will happen if Christ does not deliver the departed soul from the pains of hell. In every setting the ‘essential mood' becomes unbearably intensified in these passages, though the musical style may not change very much. One recoils from the ‘poenis inferni' (the ‘pains of hell'), the ‘ore leonis' (the ‘lion's mouth'), the ‘dies irae, calamitatis et miseriae' (the ‘day of wrath, calamity and woe'). Indeed the Responsory represents a mini-drama within the whole, piling agony on agony as the pace of the music quickens by alternating brief chant passages with abbreviated polyphony. But in both the Offertory and the Responsory calm is restored by the idea of light: ‘let Saint Michael bring them forth into Thy holy light' in the Offertory, and ‘lux perpetua' (‘light perpetual') in the Responsory.

The main reason why Victoria's six-voice Requiem is one of the greatest masterpieces of the entire renaissance period is that this mood perfectly summed up the composer's view of life and death. There was no better text for a committed Catholic priest to set. In doing so Victoria created a sound-world which, although it was not original, gained a dimension not imagined before. In fact the Requiem (or Missa pro defunctis) had long been a favourite text of Iberian composers, from the late fifteenth-century setting of Pedro de Escobar onwards. This continued through the sixteenth century with, in particular, two versions by Morales, through to the High Renaissance and the setting by Guerrero amongst others. By the early years of the seventeenth century the school of composers based in Evora, Portugal, were making their own contribution, which seems to me in various ways to take up the possibilities inherent in the Victoria. Having recorded the Victoria, it seemed logical for us to continue with the best of the Portuguese settings, which eventually led us to make discs of the Cardoso Requiem and the six-voice version by Duarte Lôbo. There remain quite a few more, in particular the six-voice setting by Felipe de Magalhães and the eight-voice of Lôbo. This set represents the first time the fruits of this mini-project have been made available together.

Victoria wrote his Requiem for the funeral in 1603 of the Dowager Empress Maria, daughter of Charles V, wife of Maximilian II, mother of two emperors and sister of Philip II of Spain. For some years Victoria had been her chaplain. The music was published in 1605 in a print which contained nothing other than the movements associated with the funeral service, though some of these were extra to the normal sequence. In particular Victoria included the four-voice Taedet animam meam and the six-voice Versa est in luctum, though no one is entirely sure when these would have been sung. We omit the Taedet here not least because its style is very different from that of the Requiem proper, but include the Versa est in luctum as a postlude.

It is the six-voice texture (SSATTB) of the Victoria, used in long sustained chords which hide a plainchant melody in one of the soprano parts, that sets the scene for the Evora compositions. Both the Cardoso and the Lôbo apparently begin as carbon copies of the Victoria, the music expanding from the plainchant ‘requiem aeternam' as it were from a single point with infinite spaciousness. In fact the musical language of these Portuguese writers is not entirely derivative. The Cardoso in particular stakes out its own harmonic territory in that first phrase, making towards an augmented chord which suggests a date of composition well after Victoria's time. It was this chord which so struck listeners when the original disc of this music was released by Gimell in 1990, establishing Cardoso (c.1566-1650) for the first time as a major figure of the period. This Requiem was published in 1625, but it is not known for whose obsequies it was written. The Requiem proper follows Victoria in being scored for six voices (SSAATB in this case); the concluding Responsory Libera me however reduces to four (SATB).

Duarte Lôbo (c.1565-1646) was an exact contemporary of Cardoso, who must therefore have been a close colleague in both Evora and Lisbon. His penchant was for full sonority, as shown in both of his Requiem settings: the eight-voice (not recorded here) technically follows the trendy double choir baroque format of the time, but in fact proceeds for much of its length in eight-part counterpoint. His six-voice Requiem is apparently a later work - published in 1639 as opposed to the 1621 of the eight-voice - but simply continues the dense and sonorous idiom of his earlier years. His six-voice choir is SAATTB, which emphasizes the denseness, though he continued to put the chant in the single soprano part, instead of the more usual tenor, as both Victoria and Cardoso had done before him. The final Responsory, to different words, again sees a reduction to four voices.

To fill out the Iberian picture of these two discs, we have included four motets each by Cardoso and the Spanish composer Alonso Lobo (1555-1617, not related to Duarte). Two of the Cardoso motets, Non mortui and Sitivit anima mea, originally appeared in the same publication as his Requiem (1625) and therefore, since that book only contained Mass-settings, were obviously viewed as an integral part of the funeral rite. The other two Cardoso motets have more general texts, though Nos autem gloriari would not be out of place at the conclusion of a funeral or memorial service. Similarly two of the Alonso Lobo items are associated with the Requiem, most obviously his magnificent Versa est in luctum which may be compared with that of Victoria. Lobo's setting does not appear to have belonged to a Missa pro defunctis proper, though his Credo quod redemptor is also associated with the service of Matins for the Dead. Vivo ego is a more general text within the period of Lent. Lobo's Ave Maria is a masterpiece of a rather different kind, being based on a complex canon 8 in 4 at the upper fifth. Although the eight voices are divided into two choirs and the bottom part of each choir sings the same music, the other three voices are rearranged: the top part of the choir which begins becomes the third part in the choir which responds, the second part of choir I becomes the top part of choir II and the third part of choir I becomes the second part of choir II. Yet despite the mathematical intricacies the resulting music seems artlessly serene, as befits the text. It culminates in the most perfect ‘Amen', where the beauty inherent in these canons is particularly telling.

Peter Phillips 2005

Tudor Church Music Vol. 2 - The Tallis Scholars, Peter Phillips

Tudor Church Music - Volume 2
The Tallis Scholars, Peter Phillips
Gimell CDGIM 210

English sacred music of the sixteenth century has formed the backbone of everything The Tallis Scholars have done, right from the beginning. The original reason why we chose to have ten core singers in the group was because the English composers of this period had developed a choir with five basic voice-ranges, instead of the more normal four, and I had decided that the dazzling sonorities inherent in their writing benefitted from having two singers on each part. This dazzle is emphasized if the music is sung at a high pitch, which led us increasingly to perform with four sopranos - two high and two lower - above two countertenors, two tenors and two basses. Thus was born the classic Tallis Scholars line-up, which later could be adapted to sing much mainland European renaissance music as well. Such a line-up sounds almost inevitable now, but it wasn't in 1973.

This second volume of The Tallis Scholars sing Tudor Church Music brings the story of English sonority to the middle years of the sixteenth century, when the gap between it and what was being written on the continent of Europe was just as wide as in the time of Browne, Cornysh and Taverner who are featured in Volume One (CDGIM 209). Instead of the densely packed texures of Gombert and Willaert there is the gothic spaciousness of Sheppard. Instead of the relentlessly argued imitative counterpoint of Clemens and Manchicourt there is the fluency and sheer beauty of melody of Tallis and White. Where Palestrina strove to perfect his sound world, Thomas Tallis set about experimenting with every imaginable combination of voices, from the forty in Spem in alium, to the miniature If ye love me, and from the low choir of In ieiunio et fletu (Track 13) to the high choir of O salutaris hostia and O nata lux (Tracks 14 and 15). This story of the way English sonority developed is amplified and brought to a conclusion in The Tallis Scholars sing Thomas Tallis (CDGIM 203, which does not contain the three tracks issued here) and The Tallis Scholars sing William Byrd (CDGIM 208).

John Sheppard (c1515-December 1558) ranks as one of the most original, if not wide-ranging, composers to come from Britain. Media vita is his masterpiece, unrivalled for its breadth of phrase and expressive power, summing up everything about Sheppard and the creative world which surrounded him: the high treble part leading to breath-taking sonorities, the doubled altos, the old-fashioned reliance on a chant cantus firmus, the persistent, quirky use of a certain kind of dissonance. This is also the formula which underlies Reges Tharsis and Verbum caro, though worked out more concisely than in Media vita. But take away the treble part and you have Christe Redemptor omnium and In manus tuas I; take away the treble and mean parts and you have In manus tuas II. Early in his career Sheppard must have found a basic method which pleased him, for he rarely adapted it. The only exception here is to be found in Sacris solemniis, where his search for sonority took him into new territory. In this masterpiece he divided the altos, means and trebles to create, with the tenors and basses, an eight-part texture. Apart from his own Libera nos settings I know of no other example of this scoring. Certainly there were plenty of contemporary examples of eight-part writing from Europe - from Crecquillon's Pater peccavi to Lassus's double-choir music - but they only point up how different the English were. Lassus, cosmopolitan as he was, would have scarcely been able to believe his ears had he ever heard Media vita or Sacris solemniis.

Sheppard's Western Wind Mass is the third and last of the three settings based on this melody (the Taverner and Tye are included on The Tallis Scholars sing Tudor Church Music - Volume One, CDGIM 209). This is the shortest of them, in number of bars nearly half the length of the Taverner, involving twenty-four repetitions of the melody which are to be found in every part except the mean. This relative brevity can be explained by Sheppard's musical language, so different from that just discussed. Although there are passages which pay homage to the melismatic, rhythmically complex style of the early sixteenth century, this much more syllabic style must come from near the end of the composer's life when he, like everyone else, was influenced consciously or unconsciously by the new Protestant ideals of textual clarity. Nor is the brevity without creative impact: each movement has a drive through it which does not characterize either Tye's or Taverner's setting.

Robert White (c1538-November 1574) was arguably the leading figure in that lost generation of English composers, including Robert Parsons and William Mundy, which came to maturity between Tallis and Byrd in the third quarter of the sixteenth century. White as it were formed a school within a school, whose musical instincts were to look back to the Catholic style of Tallis's youth (a style he had missed) while putting that style to the service of Elizabeth I's Protestant Church. The result is a fascinating hybrid: the lines unwind slowly, much of the old sonority is still there, the cadences in particular can sound deliciously archaic. Yet the expressive power is more modern, more direct. There is no more thrilling example of that power than in the ‘Amen' of Exaudiat te Dominus.

The six-voice Magnificat is the most archaic-sounding of the pieces included here, making use of Taverner's choir of treble, mean, two countertenors, tenor and bass, and even dividing into a triple gimell (divided trebles, means and basses) at ‘Esurientes'. It is monumental and impressive rather than supple, in the way Regina caeli and Exaudiat te Dominus are. Another side of White's treble writing can be heard in the two settings of the Compline hymn Christe qui lux es, where the part-writing is so perfectly crafted round the chant - especially in the last verse of the fourth setting - that White seems to defy gravity.

But perhaps the most modern pieces - by which I mean most obviously caught between old and new - are the Lamentations (two sets joined together). There is in fact little pure polyphony in this music, but rather parallel movements between the parts, organized in blocks. It makes for a mesmerizing effect, and I am not alone in thinking so. The scribe who wrote the music out for the first time, added in Latin at the end: ‘Not even the words of the gloomy prophet sound so sad as the sad music of my composer.'

Peter Phillips 2008

Tudor Church Music Vol.1 - The Tallis Scholars, Peter Phillips

Tudor Church Music - Volume 1
The Tallis Scholars, Peter Phillips
Gimell CDGIM 209

English sacred music of the sixteenth century has formed the backbone of everything The Tallis Scholars have done, right from the beginning. The original reason why we chose to have ten core singers in the group was because the English composers of this period had developed a choir with five basic voice-ranges, instead of the more normal four, and I had decided that the dazzling sonorities inherent in their writing benefited from having two singers on each part. This dazzle is emphasized if the music is sung at a high pitch, which led us increasingly to perform with four sopranos - two high and two lower - above two countertenors, two tenors and two basses. Thus was born the classic Tallis Scholars line-up, which later could be adapted to sing much mainland European renaissance music as well. Such a line-up sounds almost inevitable now, but it wasn't in 1973.

This volume contain some of the most substantial music written anywhere in Europe in the early years of the renaissance; and it shows just how different English music was from what the Franco-Flemish composers had developed on the Continent. Instead of an inward-looking, contemplative mood, there is extrovert, brilliant writing. Instead of melodic lines which tend to move by step, chant-like, there are leaps and jumps so virtuosic that they rival the coloratura writing of solo vocal music from centuries later. Where Josquin has a reliably developing harmonic background, Browne sets down massive chords which change slowly, allowing the upper voices to festoon them with ornaments. Looked at historically the English composers of this period can almost be classified as ‘late medieval' for their archaic compositional practices and quirky scorings, yet the sound world they created is one of the most identifiable and compelling of all time. Once the listener is hooked on these sonorities nothing else ever seems quite so good. The second volume of The Tallis Scholars sing Tudor Church Music (CDGIM 210) shows how this characteristically English sound developed in the high renaissance, a picture which is amplified and brought to a conclusion in The Tallis Scholars sing Thomas Tallis (CDGIM 203) and The Tallis Scholars sing William Byrd (CDGIM 208).

There is no better place to start in illustrating pure sonority than with the music of John Browne (d.1505). The sheer length of these five antiphons shows that something unusual is happening; and then one notices the scorings, each different from the last. It seems that once Browne felt he had squeezed every last nuance from a sonority, it was time to try a new one. The largest in every sense is the eight-voice O Maria salvatoris (TrMAATTBB). This is backed up by three six-voice compositions, all for different groupings: Stabat iuxta (TTTTBB), O regina mundi clara (ATTTBarB), and the Stabat mater (TrMAATB). The most ‘normal' is the five-voice Salve regina (TrMATB). All the pieces come from the Eton Choirbook, in which Browne was given pre-eminence: O Maria salvatoris, the only eight-voice composition in the volume, opens it.

The Salve regina and the Stabat mater are the pieces which for years have maintained Browne's reputation as a composer. They are both highly expressive, though for many commentators the Stabat mater is the supreme masterpiece of the period, contrasting dramatic writing with contemplative passages in an emotional world of contrasts thought to have surfaced first with Monteverdi. This drama famously breaks through the surface at the word ‘Crucifige', which Browne hammers into place before turning inwards again with the phrases which follow: ‘O quam gravis illa poena' (‘O how bitter was your anguish'). For me, however, the piece which sums Browne up most perfectly is the Stabat iuxta. Its scoring (TTTTBB) has probably militated against frequent performances, but it is just that scoring which makes such an impact. With six voices operating within a compass of less than two octaves the opportunities for dense, almost cluster chords are unrivalled. The use of low thirds in chordal spacing is not encouraged by text-books of correct polyphonic procedure, but Browne simply couldn't avoid them with this scoring, and they are thrilling. Density of sonority leads to other delights, like false relations and other dissonances, which characterize much of the piece and culminate in the final bars.

William Cornysh's four-voice Gaude virgo mater Christi (AATB), also in the Eton Choirbook, is a brief example of the genius of a very different composer, a kind of alter ego to John Browne. Where Browne was inward and at times mystical in mood, Cornysh (d.1523) was typically extrovert and clear-textured. In Gaude virgo he sends his four low voices rushing around with characteristic verve, towards the end employing strict imitation between them at the resonant words ‘Mortis in periculo'. Later in this volume there are two more pieces by Cornysh: the ineffable Salve regina, one of his grandest and at times most theatrical conceptions, and the more modest and purely beautiful Ave Maria, mater Dei, a concert favourite of ours. The miracle of the Salve regina is how Cornysh never lets the momentum sag over such a broad canvas, constantly alternating powerful trios between the lower voices with more luminous sections involving the higher ones. All this culminates in a rhapsodic passage in praise of the Virgin, building unforgettably to the final word ‘Salve'. Never was Cornysh's dramatic sense more tellingly on display.

The remainder of the music in this volume is made up of two of the three Western Wind Masses - those by John Taverner (c1490-18 October 1545) and Christopher Tye (c1505-before 15 March 1573). The third one - by Sheppard - is included on The Tallis Scholars sing Tudor Church Music - Volume Two (CDGIM 210). Together these three Western Wind Masses make a unique set. Not only was there no other example in English renaissance music of a linked series of Mass compositions by different composers, there was also no precedent in English composition for a Mass based on a secular tune. If Taverner's was the first of this set to be written, it was breaking important new ground, probably deliberately emulating the justly famed continental Mass-settings on such popular songs as L'homme armé and Mille regretz. Tye and Sheppard then took up the challenge of turning Taverner's initiative into a Continental-style series. From the listener's point of view the linking is both obvious and satisfying since the three settings use the same unusual four-part scoring of treble, mean, tenor and bass; and the Western Wind melody, beautifully tuneful in itself, is almost always audible within the polyphony.

It is always assumed that Taverner's setting was the earliest, and that the other two were written to complement it. Certainly it is the most inventive, using the melody thirty-six times in all and treating it as the scaffolding for a set of variations. These variations were written without any free passages in between the quotations of the melody, nine variations in each of the four movements. Twenty-one of the statements are in the top part (where they are at their most audible), ten in the tenor and five in the bass. The melody is never used as a traditional cantus firmus. The end result is that the Taverner version is the most varied and the most sectional of the three, the beginning of a quotation almost always bringing with it a new scoring and new figuration. It is for this reason that tribute is paid to Taverner's resourcefulness in inventing these figurations and their counterpoints - he came up with well over thirty of them, each one adding something to one's appreciation of the melody itself.

The one thing Taverner did not do was put the melody in the second voice down - the mean. It was surely to fill out the scheme, and to pay a kind of homage, that Tye put the melody in the mean only. Keeping it in one part made it easier for him to elide the quotations, as happens at the very beginning where three statements run into each other. This in turn means that Tye's setting is more homogeneous than Taverner's, though also more virtuosic, for example at ‘Benedictus'. In this, and in some truly bizarre harmonies elsewhere, one has the sense that Tye, like his contemporaries White and Sheppard, was searching for new expressive options. They did not look to Europe to find them.

Peter Phillips 2008

Flemish Masters - The Tallis Scholars, Peter Phillips

The Tallis Scholars sing Flemish Masters
The Tallis Scholars, Peter Phillips
Gimell CDGIM 211

Just as the Flemish school of composers was central to every aspect of Renaissance music, so these composers have been central to the work of The Tallis Scholars. At regular intervals we have dedicated whole albums - as well as whole concerts - to their compositions, and particularly to their Mass-settings. Here one finds the essence of their art, the showcase in which they displayed their most sophisticated techniques.

This selection contains five of the greatest examples of Flemish Mass-composition. They were written in the hundred years between about 1475 (the Ockeghem) and 1575 (the Lassus), in which time the scope of such music had expanded enormously. The time-line is instructive. The earliest - the Ockeghem - is for four voices. Next came the Isaac, for six voices, followed by the Rore, for seven, and the Brumel, for twelve. The sequence is rounded off by the Lassus for eight voices, disposed in the late-Renaissance manner of double choir. Interestingly the greatest of all the Flemish writers - Josquin des Prez (whose Masses we have recorded elsewhere, intending eventually to record them all) - kept to four voices in all his Mass-settings. For him the Mass was a vehicle for the most subtle chamber music; for the composers represented here it was more an opportunity for grandeur, which meant increasingly polychoral sonorities.

Other techniques link these settings, proving that they were the products of a school of training which passed its techniques from one generation to the next. For example they are all based on pre-existing material: three on polyphonic models (all of Flemish provenance) and two on chant. This means that each composer had to find a way of quoting his model while surrounding it with his own inventions. Typically for Flemish writers these inventions tended to be motivic and imitative, the melodic lines pieced together like a jigsaw puzzle of interlocking fragments. This could lead to extreme contrapuntal ingenuity - much beloved of the Flemish more generally - in such devices as canon. Only by the time of Lassus did these national techniques begin to become genuinely international and proto-Baroque, harmony taking over from the more linear thought of the earlier writers.

The Missa De Apostolis of Heinrich Isaac (c.1450-1517) is unusual in being an alternatim setting. This means that by constantly alternating between chant and polyphony many of the polyphonic sections are contained within a single phrase. Isaac met the challenge of brevity by intensifying his part-writing and using sonority to spectacular effect: a six-voice texture was a rarity around 1500, and only the English at that period could dream up quite such thickly dissonant cadential formulas. This Mass was based on a selection of Gregorian chants taken from the repertoire of the Feast of the Apostles, and must have been written for court use in Vienna, where Isaac was employed and where it was customary not to include the Credo as part of the Ordinary.

By contrast the Missa Au travail suis by Johannes Ockeghem (c.1425-1497) was based on a secular model - a chanson either by Ockeghem himself or the unknown ‘Barbingant'. In fact Ockeghem only quoted the tenor from this three-part model (at that time the tenor was often held to be the ‘tune'), doing so infrequently but obviously: the first ten notes of it may be clearly heard at the beginning of each movement as a head-motif. After that references are hard to identify and the music settles into a pattern of gentle, undemonstrative chamber-music, the second Agnus Dei summing up the stillness of Ockeghem's conception. The reason why this music sounds earlier than anything else on these discs is that it lacks High Renaissance sonority. As was customary in the fifteenth century, Ockeghem kept his four voices within a narrow compass (except in the extreme case of the second Agnus) and regularly scored down to duets and trios.

From the other end of the historical spectrum comes the Missa Osculetur me of Orlandus Lassus (1532-1594). Here everything is sonority: brilliant sounds created by using eight voices in the ‘Venetian' double-choir arrangement, replete with high soprano parts and wide overall tessituras. In this, as in much else, he had effectively ceased to be a Flemish composer, his long years in Italy having taught him a more avant-garde style.

In fact Lassus's three double-choir Masses - the other two are the Missa Vinum bonum and the Missa Bell' Amfitrit' altera - represented an important step in the general advance towards Baroque music. Andrea Gabrieli visited Lassus in Munich in the 1560s after Lassus himself had made repeated visits to Venice. Clearly he was influential in the development of the new idiom, and there is a certain irony in the fact that it was a Flemish composer, whose training was in techniques antipathetic to everything Baroque, who led the way. Nonetheless one admires how he instinctively avoided the tedium which later would beset so much polychoral writing and its too-obvious effects. Here his alternating phrases are quite long, so that simple imitation between the parts is possible; and he was careful to make a strong difference in sonority between passages for one choir and those for both choirs together. The result is an early, but genuinely virtuosic, exercise in an idiom which would last for at least another two hundred years.

The Missa Praeter rerum seriem by Cipriano de Rore (c.1515-1565) was probably written only a decade or two before the Lassus - Rore was an older contemporary of Lassus but died considerably younger; and if the exact date of the Lassus is not known, the Rore must come from the period (1547-58) when he was employed by Duke Ercole II d'Este of Ferrara, whose name he invokes throughout the composition. Yet despite this temporal proximity to Lassus, and the fact that Rore spent an unhappy year (1563) as maestro di cappella at St Mark's in Venice, this Mass is a trenchant example of Flemish writing. It is based on one of Josquin's greatest motets, already in six voices, to which Rore had the skill to add an extra soprano part. He then turned the first alto part into a long-note cantus firmus, perpetually quoting the words ‘Hercules secundus dux Ferrarie quartus vivit et vivet' to the devotional song melody used by Josquin. Much of the power of this wonderful music derives from the Josquin motet, but close inspection shows how resourceful Rore was in his use of it. New counterpoints abound, and the extra soprano part adds a brilliance to the sound which is patently of the High Renaissance.

Much of what I have talked about here is summed up in the astonishing Missa Et ecce terrae motus by Antoine Brumel (c.1460-c.1520). Clearly this is a composition of an earlier period than the Rore and Lassus, yet it is the most sonorous of all. Since twelve-part writing was extremely rare in the period before 1520 Brumel was effectively making up the idiom as he went along. That he was successful is shown by how famous both he and this music became in the sixteenth century: the only surviving source was copied in Munich under the direct supervision of Lassus, who himself sang tenor II in an early performance; and only Josquin received more laments from his fellow composers on his death.

This idiom is quintessentially Flemish, blown up to vast proportions. Many of the twelve-voice sections are underpinned by a canon, quoted in long notes between the first bass and the first two tenor parts, using the first seven notes of the Easter plainsong antiphon at Lauds, Et ecce terrae motus. The purpose of this scaffolding was to give Brumel huge harmonic pillars, which move slowly. At first hearing this measured harmonic turn-over may seem disappointing, but soon the logic of it becomes apparent: Brumel needed to move slowly in order to work out all the motifs he has invented. In doing so he effectively abandoned polyphony in the sense of independent yet interrelated melodic lines, and resorted to sequences and figurations of the kind Josquin was exploring with no more than four voices.

Yet despite its fame, Brumel's masterpiece remained an isolated achievement, an unforgettable one-off. It is a tribute to his genius, and the vibrancy of the Flemish cultural scene, that such a glorious experiment could ever have been commissioned, written and performed.

Peter Phillips 2009

Renaissance Giants - The Tallis Scholars, Peter Phillips

Renaissance Giants
The Tallis Scholars, Peter Phillips
Gimell CDGIM 207

The High Renaissance is well known for its cultural giants. Leonardo da Vinci, Raphael, and above all Michelangelo epitomize a period when the human spirit seemed to grow and gain in confidence. There is no more visible proof of this than Michelangelo's seventeen-foot-tall statue of David, more than twice the height of any major piece of sculpture before it. And Italy was not the only country which suddenly seemed to be populated by more-than-life-size men of genius, Shakespeare and Cervantes among them. This collection is designed to illustrate the musical side of this astonishing period in European history, from England via Flanders and Spain to Rome.

Music and musicians can bestride their world in one of two ways. They can either write a masterpiece which is in itself colossal; or they can write so many great works that they slowly change the countryside around them. Thomas Tallis did both. By writing music of the highest quality in every style of the period for nearly sixty years he influenced everybody who followed him. And he also wrote Spem in alium, perhaps not his most influential composition - who could follow it? - but unquestionably the largest single work of the period. Conceived for forty independent voice-parts arranged in eight five-part choirs, Spem in alium, like Michelangelo's David, seems to break through what lesser men had come to accept as normal in music and enter a new world. The effect of forty parts coming together in properly argued polyphony is quite staggering. How did Tallis do it, without modern aids like computers or even sufficiently large pieces of manuscript paper on which to line up all the voices? There is much about Spem in alium which is not known - like why he thought of such a thing in the first place - but the greatest imponderable is how any mind could invent so much detail. Nothing comes close to rivalling it.

The Western Wind Mass by John Taverner is big in a quite different way from Spem in alium. Throughout the Renaissance period there was a vogue for taking popular tunes of the day and dressing them up, a technique which jazz musicians were to emulate in later times. Taverner chose the beautiful melody known as ‘Westron Wynde', a love song which encourages the wind and the rain to do their worst so long as the singer and his beloved can be together. In choosing such a profane model for a Mass-setting Taverner was not in fact doing anything very uncommon; what was unusual was quoting the tune thirty-six times. I do not know any set of variations before Purcell's ‘grounds' to go so far; and, like Purcell, Taverner has the imagination to make them all interesting. He is quite deliberate about it: nine statements in each of the four movements, each one taking on new ideas and counter-themes. No one can miss the tune itself, since it is quoted twenty-one times in the top part, where it is always perfectly audible. It also comes ten times in the tenor part and five in the bass, though it is never sung by the altos. Taverner also helps the listener by making each variation self-contained and running them straight into each other without extra material. This makes the music quite sectional, alternating solo and full passages, but with the melody always present. Taverner was sufficiently a giant of his time to start a brief tradition of Western Wind settings: there are examples by his younger contemporaries Christopher Tye and John Sheppard.

If ever there was a giant among Renaissance composers, it was Josquin. It has become something of a modern cliché to compare his standing with that of Beethoven: a composer who could take any form of music and transform it. Part of his great influence came from the fact that he travelled throughout Europe, the first super-star among composers, a fact which was acknowledged at the time not least because he expected to be paid more than anyone else. He was especially present in Rome, from where he undoubtedly shaped the course of Italian polyphony in general, and Palestrina's style in particular. Josquin's Missa Pange lingua is perhaps his best-known work and possibly his last Mass-setting. In general outline it is not dissimilar from Taverner's Western Wind Mass: a set of variations on a well-known tune, but in this case the tune is a chant melody from the liturgy for the feast of Corpus Christi; and Josquin almost never quotes it straight. Indeed it is so hidden in the polyphonic texture that one may think of the whole composition as a fantasy on a plainsong, rather than a set of variations. The clearest statement is at the beginning of the third Agnus Dei, where the melody finally emerges in recognizable form in the soprano part.

Giovanni Pierluigi da Palestrina was by far and away the most celebrated Italian composer of the High Renaissance and, like Josquin, a legend in his lifetime. He was also, unlike any other Renaissance composer of any nationality, celebrated from his own time to ours without interruption. I wonder if any other composer in the history of music, apart from Wagner, has had so much written about him. And apart from all this he was working in and around the Vatican at the same time as Michelangelo in his later years, whom he must have known. As a result there is no other composer who is so closely identified with the culture of the Italian Renaissance period. His Missa Brevis was probably written for the Sistine Chapel choir to sing, which would mean that its first performances would have taken place surrounded by Michelangelo's newly painted frescoes. Why the piece is called ‘Brevis' is something of a mystery, since the music is not especially short and all the usual movements are present. It may be because it is scored for only four voices, though this was quite commonplace. The final, glorious Agnus Dei increases the number of voices to five by introducing a second soprano part, in canon with the first. Unlike his teacher Tallis, William Byrd didn't write any single gigantic work by which he may be remembered. Indeed he is best remembered for his many small-scale pieces which, despite their size, revolutionized English composition. Like several of the other giants in this collection, Byrd turned his hand to every form of music available to him, transforming them as he went: music for keyboard, lute, viol consort, voices with viol consort, sacred vocal writing for both the Catholic and Protestant churches, and madrigals. To compare him with Shakespeare has some force since their lives overlapped, they both worked in London at the same time, and they both had the same characteristic intelligence of mind which penetrated to the heart of the words they were involved with. Byrd's Mass for four voices is one of the three Masses he wrote in the 1590s and published, without title pages, in defiance of the Protestant ban on Catholic artefacts. This is not gigantic music in any sense, but subtle, intimate writing which in recent times has achieved greater renown than many much weightier musical edifices of the period.

The Spanish sixteenth century had its own great men and women to celebrate, from St Ignatius Loyola and St Teresa of Avila, to El Greco, Morales and Victoria. It will be seen from these names that the accent there in the High Renaissance was on the Catholic Church and spiritual life. Tomás Luis de Victoria was no exception to this. After being ordained priest in Rome in 1575 he spent the years from 1587 until his death employed at the court in Madrid, initially acting as chaplain to the Dowager Empress Maria, for whose funeral he wrote this Requiem in 1603. It comes as no surprise to discover that Victoria only wrote sacred music, and not very much of it by some standards, but what he did write is of such intensity that for many people his larger works, and especially the six-voice Requiem, are without rival amongst High Renaissance masterpieces. The slow, inevitable unfolding of this music, movement by movement, in complete serenity surely has a message for all time.

Peter Phillips 2006

Bruckner: Symphony No. 7 - Georg Tintner

Anton Bruckner: Symphony No. 7

Royal Scottish National Orchestra, Georg Tintner

Naxos 8.554269

.flac+.cue+.log

Classics Today Rating: 10/10

Georg Tintner's generally excellent survey of the complete Bruckner symphonies for Naxos reaches a peak with this exceptional performance of the Seventh. What makes Tintner such a wonderful Bruckner conductor is his unerring ability to set a naturally flowing tempo for each of Bruckner's thematic complexes. The result, in the first movement, is richly contrasted, but never eccentric or disruptive of Bruckner's musical architecture. Tintner's flexibility gives the opening a genuinely powerful sense of momentum that makes the following slow movement sound all the more inevitable. Too often, this symphony sounds like two slow movements followed by two quick ones, despite the "Allegro moderato" marking on the very first page. Tintner also disciplines the RSNO's brilliant brass section so that they produce a glowing sonority that never sounds forced or harsh. Add to these several pluses a splendidly rich recording, and the result is as fine a Seventh as you're likely to hear - and a steal at budget price.

David Hurwitz

Bruckner: Symphonies Nos. 8 & 0 - Georg Tintner

Anton Bruckner: Symphonies Nos. 8 & 0
National Symphony Orchestra of Ireland, Georg Tintner
Naxos 8.554215-16

.flac+.cue+.log

I didn't find any review "with a pedigree", but found many of the reviews at the amazon page linked, both informed and competently written.

Mahler: Symphony No.6 - Chicago Symphony Orchestra, Haitink

Gustav Mahler
Symphony No. 6
Chicago Symphony Orchestra
Bernard Haitink
CSO Resound CSOR901807

.flac+.cue+.log

FANFARE

Appropriately enough, this recording arrived on July 7, Mahler’s birthday. My most recent encounter with Bernard Haitink’s Mahler Sixth wasn’t a particularly positive experience (26:4); this was a live recording from France in 2001 (issued in 2002), and though I found much to admire, the performance lacked the kind of urgency heard in the then recent performances conducted by Michael Tilson Thomas and Benjamin Zander; more damaging in this listener’s estimation was Haitink’s uncharacteristic omission of the exposition repeat in the first movement. Every Mahlerite should be thankful to the CSO, therefore, for its decision to preserve Haitink’s most recent Mahler performances, first with the Third (31:2), and now with this outstanding Sixth.

Haitink’s opening tempo in the Allegro is measured, very similar to Zander and Rattle. Both the >ma non troppo and aber markig notations are the clue here: while a faster tempo is certainly appropriate to the energico indication, a case can be made for a more moderate and precise march (“sturdy” is Michael Steinberg’s translation of markig and that is appropriate here); think of it more as determinedly implacable, rather than brutally insistent. It should come as no surprise that the orchestra is superb, with excellent detail (thanks in part, no doubt, to the sound production). The “Alma” second theme is a ray of sunshine illuminating dark places, expressive of happiness rather than impetuosity - the difference isn’t so much one of tempo as of content.

The development section is expansive and very atmospheric, injecting a much darker sense of futility; the “music from far away” is therefore that much more effective, achieving a sense of time in suspension. The cowbells are evocative without being obtrusive, and the strings are ethereal in their sensitive interplay with the celesta; the exposed solos for horn and clarinet add more contrast to the airy strings. The return to the march theme returns us to the hurly-burly of daily life, not at a frenetic pace, but simply as more of the same. The same kind of darkness heard in the development infuses the recapitulation; the coda swaggers to its triumphant conclusion as a firm refutation to the negativity of the march.

Haitink’s first movement, at just shy of 25 minutes in duration, is slower than the timings noted at Mahler’s own performances, which lasted between 22 and 23 minutes. Gergiev’s is a frenzied and heavy march, just shy of 22 minutes. I maintain, though, that Haitink’s is the more effective performance, the extra time supplying the kind of contrast that heightens Mahler’s thematic layout; Gergiev’s performance is without nuance and becomes monolithic.

The Scherzo is next, and with performances as good as this one, the logic of the original order of the middle movements becomes practically irrefutable (for a thoughtful examination of the middle-movement sequence debacle, see La Grange’s comprehensive analysis in Volume 4 of his Mahler). This parody of the first movement is heavier in its emphasis but otherwise continues the sense of implacability. Mahler’s most rigorously composed symphony is given rigorous control, and Mahler’s brilliant orchestration is exposed by the sound production - it is detailed and has depth both in an acoustic sense and in low-end reproduction. Haitink employs modern seating for the strings, but there is little loss of clarity. One notable demonstration of this clarity is the dance-like transition after the Trio, with its dry-bones xylophone. There is a truely pathetic quality to the Trios: they may evoke childlike images, but they seem too fragile for this environment, so that the resumption of the main theme is that much more sinister - Haitink really underlines the heaviness with groaning, heaving cellos, basses, and tuba.

As Steinberg says, the Andante is indeed “balm” after the Scherzo’s “music of disintegration and suppressed violence.” Haitink’s performance fits that characterization: initially calm but (Mahler being Mahler) with strong undertones that eventually erupt into powerful (but not violent) emotion, eventually to reach the heights, accompanied by the cowbells. There is more exquisite playing from the orchestra, the very poignant oboe of Eugene Izotov being the standout.

Haitink’s finale is painted on a very broad canvas: at 34:00, his is a couple of minutes longer than Tilson Thomas and Zander (and Mahler). How that time is utilized is the key. The exposition is dark, almost primordial, with life crawling, scrabbling for any handhold. The striving in the music for some semblance of triumph is battered down by very convincing climaxes: these make the hammer blows less of an idiosyncratic effect and more like exclamation marks - Haitink’s treatment of dynamics is another strength of this performance. In the expansive development, the gloom gives way to reminiscence: pointillistic effects of celesta and harp are so clear; there is plenty of vitality in the upwardly striving music, and then the first hammer blow falls. The sound is deep and dull, almost felt rather than heard (Zander’s is a much more singular sound).

Haitink’s pacing is very effective and there is ample dramatic tension in the warring themes at the heart of the development. The second hammer blow is less conspicuous, in keeping with Mahler’s wish that each successive blow is diminished in sound, though this one is integrated into the percussion battery almost too effectively. The decaying themes in the recapitulation are marked by a very heavy harp and mournful brass, like birds of prey on a corpse. The reminiscent music (“from far away”) is incredibly poignant here, the cowbells barely audible. One last attempt at life: the section, all brass chorales and desperation is almost frenzied in this performance, with the leering march as counterpoint, until all striving ceases with the coda. There is no third hammer blow, but Mahler’s revision is realized through the thunderous timpani. The “epitaph” for brass is practically motionless, hovering near death’s door, until the final, decisive stroke, more deadly than any hammer.

Coming so soon after Gergiev’s disappointing recording (31:6), this Haitink performance makes for a salutary corrective. Where Gergiev is all momentum and pulse, Haitink applies the lessons of a lifetime devoted to conducting Mahler and produces a performance of depth, nuance, and insight. Though Haitink’s expansive conception rivals that of the Third in duration, there is never a sense of time weighing heavily - all is drama and contrast. For those (like me) who often think of Haitink as the “restrained, sober” Mahlerian, this Sixth should be instructive.

The sound on this release is all one could ask for: spacious but offering exemplary inner-voice detail; ample low-end reproduction but not of the boomy or boxy variety. Mahler’s extraordinary orchestration, particularly that of the finale, where various instrumental sonorities - tremolando strings, celesta, tuba - play against each other, is given every shade of color and individuality.

The Sixth has had an outstanding track record on SACD, beginning with the recording by Tilson Thomas and now including Zander on Telarc, Eschenbach in Philadelphia on Ondine, Fischer on Channel Classics, and Jansons on Amsterdam’s RCO Live label (I haven’t heard the SACD of Abbado’s excellent recording with the BPO on DG). If price is a concern, Fischer’s and Gergiev’s are single-disc recordings, though neither is a bargain; Zander’s is an outstanding deal at three-discs-for-the-price-of-one (I can’t recommend the Gergiev at any rate). I would hope, however, that considerations of price alone won’t deter anyone from obtaining this outstanding performance of one of Mahler’s most dramatic and complex symphonies.

FANFARE: Christopher Abbot

JS Bach: Cantatas BWV 140, 147 - The Monteverdi Choir, Gardiner

Johann Sebastian Bach
Cantatas BWV 140, 147
Holton, Chance, Johnson, Varcoe, The Monteverdi Choir,
The English Baroque Soloists, John Eliot Gardiner
Archiv 431 809-2

This is the third disc in John Eliot Gardiner's Bach Cantata series for Archiv. The works he has chosen contain what are probably the two bestknown pieces in the entire canon, the chorales "Zion hOrt die Wachter singen" ("Zion hears the watchmen sing") from No. 140, and "Jesu bleibet meine Freude", popularly called "Jesu, joy of man's desiring" from No. 147. These masterly cantatas date from different periods in Bach's life. No. 147, in its earliest form, belongs to Bach's Weimar period; but for a Leipzig performance in 1723 he added recitatives, the famous chorale which concludes the first and second parts of the work, and perhaps the bass aria, too. No. 140 is a purely Leipzig piece dating from 1731 when it was sung on the rarely occurring Twenty Seventh Sunday after Trinity.

I have had mixed reactions to Gardiner's previous two cantata discs in this interesting series and I feel somewhat ambivalent about this one, too. Readers will not need to be enlightened concerning performance standards for these are, as usual, commendably high. There is no shortage, either, of affecting gestures in Gardiner's direction which is both warmly communicative and plentifully endowed with insight to the music. No, my problem here lies more with the strength of the interpretations which strike me as lightweight. The opening chorus of No. 140, for example, is a very dramatic affair yet this aspect of Bach's art takes second place to refinement of articulation and gracefulness of phrase. I would not for a moment be without these sterling qualities but the imagery of this great movement is vivid and declamatory and Bach's intention was surely to arouse the passions of his audience. The opening chorus of No. 147 is another radiant piece, this one dominated by a glittering high trumpet. The soloist, Crispian SteelePerkins, plays it almost impeccably yet, once again, I felt the effect was decorative rather than assertive.

The solo singing is by and large excellent. I very much enjoyed the unpretentious vocal quality of Ruth Holton though would like to have felt she had a little bit more in reserve. Michael Chance and Stephen Varcoe are both on characteristically strong form but my greatest pleasure derived from the eloquent declamation of recitatives by Anthony Rolfe Johnson. Choir and instrumentalists are as responsive to Gardiner's direction as we have come to expect of them but even so I am left with the feeling that the great sense of occasion generated by this music has only been realized in part. The Archiv documentation is up to its usual high standard.

N.A., Gramophone Magazine 1992