Welcome to Hyperion Records, an independent British classical label devoted to presenting high-quality recordings of music of all styles and from all periods from the twelfth century to the twenty-first.

Hyperion offers both CDs, and downloads in a number of formats. The site is also available in several languages.

Please use the dropdown buttons to set your preferred options, or use the checkbox to accept the defaults.

George Frideric Handel (1685-1759)


First reconstructable version (Cannons), 1720
Dunedin Consort, John Butt (conductor)
Studio Master FLAC & ALAC downloads available
Studio Master:
Studio Master:
Download only
Recording details: July 2011
Greyfriars Church, Edinburgh, Scotland
Produced by Philip Hobbs
Engineered by Philip Hobbs & Robert Cammidge
Release date: May 2012
Total duration: 99 minutes 23 seconds

Cover artwork: Queen Esther (1878) by Edwin Long (1829-1891)
National Gallery of Victoria, Melbourne / Bridgeman Art Library, London

Named the 11th Greatest Choir by Gramophone, the Dunedin Consort has established a reputation as the finest single-part period performance choir currently performing. The Dunedin Consort's highly anticipated new recording of Esther is the third recording in its hugely successful Handel series. The Consort has set the bar high for this Handel performance with a Gramophone Award in 2007 for Messiah and a BBC Radio 3 ‘Building a Library' First Choice accolade for Acis and Galatea. For Esther, director John Butt has reunited his award-winning team of soloists, Susan Hamilton, Nicholas Mulroy, Matthew Brook and Thomas Hobbs, plus well-known guest soloists Robin Blaze and James Gilchrist.

A 24-bit 192 kHz studio master for this album is available from the Linn Records website.


'So much in Butt's carefully prepared performance feels exactly right' (Gramophone)

'A two-voices-to-a-part chorus achieves more immediacy than a larger choir, coupled with a stylish and delightfully intimate band' (BBC Music Magazine)» More

'The Dunedin Consort perform as a tight and capable band and Butt once again has proved himself the master of dramatic pacing. Thoroughly recommended' (Early Music Review)» More

'One of the most attractive versions on disc' (International Record Review)» More
Handel’s Esther has long enjoyed the cachet for being the first English Oratorio, but its origins during Handel’s short period composing for James Brydges (who became Duke of Chandos in 1719) are obscure and the exact identity of its earliest version uncertain. What has become increasingly clear is the fact that ‘The Oratorium’ (as it was initially called) went through two versions between 1718 and 1720. What survives of the earliest version (much of which was clearly discarded during the revision) corresponds almost exactly to the vocal and instrumental forces required for Acis & Galatea of the same year, while the revised score corresponds to the expansion in the group of musicians that Chandos employed at his sumptuous establishment of Cannons, in Edgware. The libretto, like that for Acis, seems to have originated in the close circle of poets associated with Handel and Brydges during these years. Alexander Pope may have been involved, as he probably was with Acis, but most scholars agree that the bulk of the libretto was the work of John Arbuthnot.

It is clear that Esther was a topical subject during Handel’s Cannons years since Thomas Brereton had recently published his translation (1715) of Racine’s celebrated three-act play of 1689. Several turns of phrase from this appear in Handel’s libretto (e.g. ‘O Banks of Jordan’s stream … when shall we behold your Charms again?’, ‘Both Root and Branch they seek to spoil our Race!’), but it seems likely that the author also consulted Racine’s original, having adopted the French version of the name of the Persian King Ahasuerus, ‘Assuerus’. However, as Brereton notes in the opening dedication for his translation, there were many who were of the ‘Opinion that Religion and Polite Literature are incompatible’. In other words, these believed that religion should be kept apart from anything that could remotely be called ‘entertainment’. The Puritan strain clearly persisted in English culture, to the extent that Brereton believed those ‘who are over-run with Superstition, or Religious to Melancholy and Enthusiasm’ outnumbered those who are ‘decently’ devout by twenty to one. Given that a dramatic presentation of a biblical story would doubtless have been publically unfeasible in 1718, it is not surprising that Handel’s first attempt at a musical setting should take place within the context of a private establishment.

What was the point of producing sacred dramas set to music if there was likely to be so much opposition in the public domain? Brereton clearly belonged with those (a minority?) who believed that art, correctly used, actually enhances religion, ‘Tragedy wou’d, next to Preaching, be of all Ways the most conducive to Morality’. Moreover, he suggests that such productions provide a useful antidote to ‘atheism’ and attract those who might otherwise despise religion, quoting from George Herbert’s Church Porch, ‘A verse may find Him who a Sermon flies, And turn Delight into a Sacrifice.’ He also notes that the use of a Chorus, after the Greek manner will ‘to such as are especially inclin’d to Musick … have all the good Effects of the Modern Opera, without any of its Absurdities.’ Handel, of course, would have had considerable experience of the Italian genre of Oratorio, having written two examples of his own in Rome. But it is likely that the topicality of sacred drama in England provided him with an excellent opportunity to develop the genre in new directions, capitalizing on his already considerable operatic experience. What the English context in particular afforded was the ready-made establishment of the chorus in the various cathedrals, collegiate and private institutions, with the concomitant genre of the anthem. The chorus could have a greater emphasis than had been possible in the Italianate works, providing its own moral outlook on the drama and almost becoming a character in its own right. Here Handel could also draw on his native German choral tradition, adapting a large number of movements directly from his recently composed Brockes Passion (English audiences would fortunately not have been aware that the evil Haman’s last aria, lamenting his fate, was originally sung by Jesus). In all, the challenge was to integrate what was essentially actual church music into a music drama that benefitted directly from operatic practice.

The libretto for Handel’s Esther presents some of the most striking elements of the Esther story (which constitutes the book of Esther, and which is also the subject of a Greek version in the Apocrypha). However, it is very sketchy in places and it may well be that the revisions between 1718 and 1720 resulted in some unevenness of detail. The main characters are Assuerus, the Persian King (sometimes identified with Xerxes); Haman, his evil henchman; Mordecai, leader of the Jewish community in Persia and guardian of his orphaned cousin, Esther. Esther has recently been chosen as queen following an extensive search for an attractive young woman to replace the disobedient Queen Vashti. The libretto opens with three contrasting scenes: first, the order from the king’s chief minister, Haman, to massacre all the Jews (here it would be useful to know, from the full story, that the Jewish leader, Mordecai, had previously refused to bow before him), and the eagerness of his soldiers ‘to execute the blow’; secondly, the Jews, unaware of the decree, celebrating their seeming felicity now that Esther has become queen; thirdly, the sudden change of mood in the Jewish community when the news of their forthcoming massacre arrives. The central section of the story (Act 2 in Racine/Brereton and probably in Handel’s 1720 version) begins with Mordecai informing Esther herself of the news and imploring her to go and intercede with the king. But anyone who enters the king’s presence unbidden is to be put to death, regardless of status. Here Handel’s librettist, following Racine, uses more of the Greek version of Esther, which describes her deadly fear and fainting when she finally approaches the king. However, the king extends his sceptre to his queen, thus allowing her access without harm, and he expresses his deepest love towards her. Their relationship resumed, Esther invites Assuerus, together with Haman, to a feast. The final act concerns the arrival of the king and his offering Esther anything she desires. At this point it would be helpful to know that the king was not yet aware of Esther’s Jewish identity, but in Handel’s version it seems to be assumed, and the king seems ill-informed about Haman’s evil decree. In reminding Assuerus that Mordecai earlier brought to light a plot to assassinate him, Esther convinces him of Haman’s guilt. The latter is condemned to death and the Jewish community celebrates its freedom.

As is rather the case with Acis & Galatea, the events in this story are somewhat spare and the few changes of state occur very quickly. The opening act has something of the stasis of the first half of Acis, in which there are basically three tableaux in succession, Haman’s evil decree, the joy of the Jews at their supposed freedom, and their lamentation at the forthcoming massacre. Here the music (most of it from the 1718 stage of the composition) comes closest to resembling the charmed world of Acis, with Haman’s ‘Pluck root and branch from out the land’ almost recalling the comic evil of Polyphemus. One of the most attractive scenes of all concerns the Jewish celebration of freedom, very much recalling the pastoral felicity of Acis: the charming pizzicato aria ‘Tune your harps’ and then the actual appearance of a harp in the childish expression of devotion ‘Praise the Lord with cheerful noise’. This sense of a state of new-found peace – rather than a sequence of events – is enhanced by the repetition of the chorus ‘Shall we of servitude complain’. A similar technique marks the complete change of state for the third scene, with the chorus of lamentation ‘Ye sons of Israel, mourn’ repeated on either side of the centrepiece, the priest’s aria, ‘O Jordan, Jordan, sacred tide’. Here the interlocking violin lines allude both to the ripples of the river and the sadness of the community’s mood.

The central act shows a strong turn towards the solemn but highly intimate mood of the German Passion tradition (the majority of the movements here having their origins in Handel’s Brockes setting). Here the highly charged encounters between Esther and Mordecai, and then between Esther and Assuerus, can benefit from the type of recitative that Handel had developed in his operatic career. The borrowed music, almost miraculously, delineates the three characters, Mordecai’s solemn piety in ‘Dread not, righteous Queen’ and Esther’s frailty in ‘Tears assist me’ and in her duet with Assuerus (originally designed for Jesus and his mother). But it is the king himself who receives the greatest depth of character: a desperately concerned lover in the duet ‘Who calls my parting soul from death?’ and then – in perhaps the emotional high spot of the entire oratorio – he sings his greatest expression of love in ‘O beauteous Queen’. As John Roberts has recently shown, this is an aria on which Handel expended the greatest compositional effort, adapting the melodic material from three different sources. This is also marked by the first significant change of instrumentation since the harp back in the second scene, the addition of two bassoons, which provide a glowing sonorous support for the tenor register. ‘How can I stay, when love invites?’ follows almost immediately, now showing the inescapable joy of Assuerus’s love.

If Esther is already marked by its introduction of unexpected instrumentations in each of the first two acts, the final one introduces a pair of horns in the opening scene and a trumpet in the final chorus. Suddenly we seem to be in the world of Israel in Egypt or Saul, with their vast choral landscapes. The chorus ‘He comes to end our woes’ is perhaps the liveliest music in the work, marking the Jews’ anticipation of God’s intervention to end the tyranny of Haman. The final chorus, perhaps Handel’s longest, is a vast rondo interspersed with duets of praise, first between the priest and trumpet, then between Esther and Mordecai, and finally between two basses announcing the felling of cedars of Lebanon to build the new temple. These two choral blocks enclose the final stage of the drama, musically expressed through Haman’s recitative and aria of sorrow and supplication, and Esther’s aria of rage at his impertinence. In all, we almost hear the genre of English Oratorio emerge in the course of the 1720 Esther itself. It begins like a sort of ‘pastoral entertainment’ along the lines of Acis & Galatea, it becomes more operatic and adopts some of the pious intensity of the German Passion, and emerges as a choral oratorio on the largest scale. But this development is somehow central to the success of the music, giving the entire work – containing so many symmetries or complementary blocks – a succinct trajectory that is unique in Handel’s output. If the chorus ultimately emerges as the largest character, it is also there from the start, providing a dimension that is indispensable to the success of the drama. If Handel remained nervous of unleashing this extraordinary genre on the London stage, it was Bernard Gates, music director of the Chapel Royal, who took the plunge and organised a staged production of what was essentially the 1720 Esther, for a private audience in central London in 1732. In the wake of this success, and the likelihood of a pirate performance of the work, Handel correctly gauged that a revised, unstaged, version of Esther would become the beginning of a new venture in public music making.

John Butt © 2012

Sources and Performing Edition
This recording presents the earliest recoverable performing version of Handel’s Esther, responding to recent findings and hypotheses about the origins of the work. The new performing edition (published by the Early Music Company Ltd) differs from the scholarly edition in the Hallische Händel-Ausgabe (HHA; ed. Howard Serwer) in that the latter is based entirely on the evidence of the existing sources, while this performing edition reassesses the relationship between them based on inferences about the missing performing sources for the 1720 version. I have examined reproductions of all the relevant sources in detail: Handel’s autograph (GB Lbl R.M.20.e.7), the Malmesbury (Malmesbury Collection; deposited in Hampshire Public Records Office, Winchester) and Aylesford (GB Mp MS 130 Hd 4, vol. 93) manuscripts and Gates’ libretto of 1732. I am most grateful to Hampshire Public Records Office and the Earl of Malmesbury for access to the original of the Malmesbury MS.

The primary catalyst for reconstructing afresh the earliest recoverable performing version is John H Roberts’s seminal article of 2010, ‘The Composition of Handel’s Esther, 1718–1720’ in the Händel Jahrbuch, 353–90. Roberts reiterates what has been obvious for some time: that many details about the origin of Esther are unclear, contradictory or simply unknown. According to the wordbook produced for Bernard Gates’s performance in 1732, the oratorio was composed for the Duke of Chandos in 1720. While Handel’s autograph lacks the final page, which would normally carry the date, the earliest copy, now owned by the Earl of Malmesbury, states that Handel composed it in London in 1718. Thus there is a discrepancy in dates and also the issue of whether it was performed in the Duke of Chandos’s estate (Cannons) in Edgware or at his house in London (or indeed, somewhere else entirely). As the research of Graydon Beeks has shown (‘Handel and the Music for the Earl of Carnarvon’, in Bach, Handel, Scarlatti: Tercentenary Essays, ed. Peter Williams (Cambridge, 1985), pp. 1–20), the first part of the work in many respects conforms directly to the forces that Handel had while working at Cannons in 1717–18 (most noticeably lacking violas and including a second tenor line), thus matching much of the Cannons church music and Acis & Galatea in many respects. However, Esther as transmitted in Handel’s autograph and all the early copies contains an increasing number of movements with larger forces, including viola, extra solo singers, two bassoons, two horns and trumpet. Beeks notes that these larger forces very closely reflect the membership of the Cannons Concert by the middle of 1720.

Roberts’s research shows that one piece in particular could not have been composed at the earlier date. While the basic material of ‘O beauteous Queen’ comes from the Brockes Passion (c1716), Roberts observes that it also contains borrowings from operas by Keiser and Lotti, and from a serenata by Heinichen, all skilfully intertwined. The latter, and most likely the opera of Lotti too, Handel could not have heard before late 1719, so at last we have firm evidence that at least some of Esther as it survives today (particularly after the first few scenes), could not have been written in 1718. Given that Handel’s autograph contains none of the material he must have discarded around 1720, only the latter version is in any way reconstructable. Building on Serwer’s study of the sources, Roberts notes that the two most main copies of the first version, the Malmesbury and the Aylesford scores, are copied largely from the autograph and thus do not communicate the changes Handel would have made for performance. Unfortunately, Handel’s performance score (often referred to, in the case of his dramatic works, as the ‘conducting score’) is missing, probably lost in the chaotic distribution of all the contents of the Cannons estate.

It is the likely contents of such a score that Roberts most usefully proposes and that this recording and the new performing edition aim to represent. First, he suggests that the autograph’s division into six scenes was replaced by a division into three acts or parts, of three scenes each, as is suggested in the Gates wordbook of 1732 (Handel himself adopted the three-act format in his own later versions of the work). The division into acts mitigates the hitherto awkward tonal and dramatic progression from the chorus ‘Virtue, truth, and innocence’ to the arioso and chorus ‘Jehovah crown’d/He comes, He comes’, given that these now straddle acts 2 and 3 (Roberts remarks that this incongruity was first noticed by Chrysander). But, I might add, given that the three acts are shorter than in Handel’s later oratorios, it may well be that Esther, as the very first experiment in the English oratorio genre, was designed as a shorter work, to be performed in one go without substantial intervals between acts (thus making it about the same length as Acis & Galatea, if performed in its 1720 form without an interval).

The first major change to the content of existing editions of Esther comes with the aria, ‘Praise the Lord’, clearly an insert in the autograph (and coming from the later stage in the composition, given its inclusion of a viola part). Two recitatives are added to the end of this insert, the second, ‘O God, who from the suckling’s mouth’ being the one normally sung after ‘Praise the Lord’. But, as Roberts notes, both the tonality and the sense of the text point to this being the recitative designed to lead into ‘Praise the Lord’ rather than follow it. Moreover, given that the B section of ‘Praise the Lord’ reuses that latter half of the original aria coming next in the autograph ‘Sing songs of praise’, and much of the rest of its text appears in the final chorus (undoubtedly from the latter period of the composition), ‘Sing songs of praise’ is rendered redundant. In other words, ‘Praise the Lord’ should be preceded by its recitative and it almost certainly replaces ‘Sing songs of praise’. The recomposition of the recitative (with the text about the ‘suckling’s mouth’) also suggests that Handel decided to use a boy to sing the aria (the Gates text assigns it to the ‘Israelite Boy’). Moreover, the autograph contains relatively clear indications of an added flute part, presumably designed to add more sound to the harp’s melodic line. However, Serwer, as followed by Roberts, was wrong to state that the flute part is also indicated in the Malmesbury manuscript. I would also suggest that, given that the roll of musicians at Cannons in 1720 contains three or possibly four boy trebles and no sopranos, it is possible that Esther was performed without female singers. Certainly, the doubling of the vocal line of Esther’s last, rage, aria with oboe is unusual, and may reflect that this role was originally sung with a smaller voice than operatic practice might suggest.

The other important change suggested by Roberts is the insertion of a viola part into ‘Who calls my parting soul from death?’, found in a score now in the Moravian Music Foundation in Bethlehem, Pennsylvania. Although this is a late source, it may well reflect some of the readings in the performing score: much of the viola part comes directly out of Handel’s Brockes Passion which would have been unknown to any in England but Handel in 1720; secondly, since Handel had reworked the two violin parts in places to accommodate the lack of a viola in 1718, he seems to have altered the replaced viola part at some of these places. Most of Roberts’s other suggestions relate to the names given to some of the characters: the ‘Israelite Boy’ has already been noted, and the Gates wordbook also suggests that the solo alto part belongs to the ‘Priest of the Israelites’. This role had clearly become a major one by 1720 and Roberts suggests that it may have been designed for Thomas Bell, a celebrated countertenor at that time, who entered service at Cannons that year.

This recording is designed as a sequel to Acis & Galatea (presented in its 1718 Cannons version), reflecting the expansion of the Cannons Concert and Handel’s evident expansion of his conception of ‘The Oratorium’. The expanded band (as examined and presented by Beeks) allows for three violins per part, viola and two bassoons. The single trumpet clearly reflects the 1720 scoring although there is no sign of any horns. The vocal complement (ten singers, including the boys, being on the list of employees), seems to allow for roughly two singers per part in the choruses (nearly all singers thus taking a solo at some point, including the major roles), although it seems a little short in terms of the double tenor line. The Gates layout of scenes and acts (admittedly not so clearly evident in Handel’s autograph) provides a nicely symmetrical structure of nine, roughly equal, scenes which offer opportunities for planning a performance that attempts both a consistent musical integration of each situation and an overall dramatic trajectory.

John Butt © 2012

Waiting for content to load...
Waiting for content to load...