For such a large-scale work, 'Spem in Alium' seems to have left remarkably little trace in history. We do not know when (or why) Tallis wrote it nor do we have any documentary record of its first performance. So over the years a tradition built up of large-scale performance, typefied by David Willcocks’ 1965 recording sung by King's College Choir and Cambridge University Musical Society. A sea change happened with the recording by the Clerkes of Oxenford under David Wulstan. Here the motet was sung at a higher pitch, with one person to a part in a performance notable for its transparency and beauteous clarity of line. Listening to this recording, I was struck again by the striking difference between it and most other recordings. Not everyone will like the sound-world that Wulstan conjures up, but one cannot help but admire the wonderful control of his high (very high) sopranos. And the recording transformed the way we think of the piece.

We have also been learning to alter our perceptions of why, and how the piece came to be written. It has long been assumed that the work was written in 1573 for the 40^th birthday of Queen Elizabeth I, hence the 40-part nature of the piece. Peter Philips has speculated, in an interview, that Tallis might not even have heard the work in full. But, though the documentary record is sparse, the work must have made some impression on Tallis's contemporaries at the Chapel Royal. For, in 1610, when preparations were underway for the coronation of Prince Henry (eldest son of King James I) as Prince of Wales, 'Spem in Alium' was dusted off in a version with new English words. After Prince Henry's death, the coronation went ahead with Prince Charles (future King Charles I) in his stead and 'Spem in Alium' was performed in its new guise as 'Sing and Glorify'.

And it is as 'Sing and Glorify' that the work has come down to us. The earliest manuscript, the Egerton Manuscript, uses the English words which are noticeably more cheerful than the Latin ones. The performance at Prince Charles's coronation is notable for other reasons as well. A contemporary diarist, in 1611, remembered how the piece had first been commissioned by Thomas Howard, 4^th Duke of Norfolk as an answer to Striggio's 40 part 'Ecce beatam lucem' and the work was first performed in the long gallery at Arundel House in the Strand.

If this memory is accurate, then we may presume that the private performance may well have involved the Chapel Royal and other members of the musical establishment. But given the contemporary preference for performing the work in echoing churches, the mind boggles at the idea of a performance in a private house, albeit a large-ish one like Arundel House.

On this recording, the Chapelle du Roi under Alistair Dixon give performances of both 'Spem in Alium' and its contrafactum 'Sing and Glorify'. They perform the work with one singer per part but recorded the works in the huge echoing spaces of All Hallow's Church, Gospel Oak, London.

In live performance the choirs enter singly, one by one. So the music is notable for the way the centre of the music moves across the choir leading to some thrilling acoustic effects and later on the climaxes can involve some brilliant antiphonal effects. This can be difficult to bring off on a recording. The recording by the Tallis Scholars under Peter Philips on Gimmell is noticeable for the brilliance with which the recording engineers have tackled this challenge. And the recording by Pro Cantione Antique under Mark Brown is also successful in this way.

Unfortunately, the recording by the Chapelle du Roi is less impressive in this respect. There are too many moments when the choir comes over as a generally aural wash with key personnel highlighted. The recording has an unfortunate tendency to spotlight individuals so that some of the sopranos, in particular, tend to stand out. The middle parts also blend into a single muddy texture rather than a combination of individual lines. The climaxes are certainly thrilling but recording in All Hallows, with the resultant huge backwash of sound, was a mistake I think. The recording sounds like a choir of 40 trying to sound like a bigger group.

What makes the recording special is the ability to hear the English contrafactum alongside the original Latin. In many places, of course, the English words disappear in the general aural wash. But singing in English does certainly give a different sound to the piece in some places.

This recording is quite an achievement. The pieces were recorded as part of the Chappelle du Roi's complete Tallis Edition and as such will be appearing in volumes 7 and 8. Surrounded by their sophisticated performances of Tallis's other works, I think that these recordings could be quite valuable. But here, issued as a short CD with just the two works, the performances have their shortcomings highlighted.

Quite how you want 'Spem in alium' to sound is a very personal matter and many people will already have a favourite recording. But this one is worth considering for its inclusion of 'Sing and Glorify'.

Robert Hugill