Although there are many recordings available, Bach's St John Passion is still far less popular than his St Matthew Passion. For this reason a new recording of the former usually attracts less attention than a new release of the latter. It could be different with this recording. It has two features which make it especially interesting. First of all, it presents a reconstruction of the very first version of this work. Secondly, it is performed with one voice per part, and in this respect it could be considered the counterpart of Paul McCreesh's St Matthew Passion.

There are no less than four different versions of the St John Passion. Nowadays the most performed and recorded one is the last, of 1739/49. The version of 1725, which is quite different from that one, has only been recorded a couple of times: the latest are by Philippe Herreweghe (Harmonia mundi, France) and Peter Neumann (MDG). There is a third version, which Bach started to prepare in 1729, but when he was told there would be no Passion performance that year, he left it unfinished. The St John Passion was first performed in 1724, but the score used in that performance has been lost. Fortunately, enough material has been left to try to reconstruct that very first version. And that is what the Dutch musicologist Dr Pieter Dirksen has done. It is this reconstruction which is recorded here. Its first performance took place in the Netherlands in 2001.

There are a number of differences in comparison with the versions which are usually performed and recorded. Some of the recitatives and arias are more simple, with less ornamentation. But the main difference is the instrumentation. In this reconstruction there are no transverse flutes. Dirksen believes that Bach originally didn't intend to use them and that they were only added at the last moment. In his first year as Kantor in Leipzig Bach didn't have flute players at his disposal and the cantatas from this period don't have them as well. They are dominated by the sound of the oboes, and when the St John Passion is performed without transverse flutes, the oboes get a more prominent role, which is in line with Bach's cantatas from the early years in Leipzig. Besides, the transverse flutes don't play such an important role in this passion in comparison with the St Matthew: they mainly play 'colla parte' with the oboes (opening chorus) or the violins (some of the 'turbae') or the choral tenor, an octave higher. The arias in which the transverse flutes are involved are both written in keys which make them difficult to play on the baroque flute. Bach usually composed his flute parts in D or G major, in contrast to 'Ich folge dir' (in B flat major) and 'Zerfließe mein Herze' (in f minor).

Dirksen believes the first of the two is a 'violin aria': Bach would hardly compose a major vocal work without an aria with violin solo. The character of the instrumental solo part (the articulation, for instance) is very 'violinistic'. The upper instrumental part of 'Zerfließe, mein Herze' was meant to be played by the oboe. The key is more playable on the oboe than on the transverse  flute, and the range of the part is much more limited than necessary for the transverse flute.

"These two reconstructed scorings result in an arrangement of the obbligato instruments for the arias which is logical and far more typical of Bach." If the aria 'Mein teurer Heiland' is left out - because it is accompanied by basso continuo only and has more the character of a chorale setting than an aria - the first and last of the remaining seven are scored for oboes, the second and sixth for a solo string instrument (violin and viola da gamba respectively) and the third and fifth for strings. The aria in the heart is 'Erwäge', which has the unusual scoring of two viole d'amore. This results in a symmetry which Bach loved so much and which can be discovered in other compositions like the St Matthew Passion and the Mass in B minor.

When I first heard this reconstruction I was convinced that this was how Bach had originally conceived his St John Passion. In particular the solo violin in the aria 'Ich folge dir gleichfalls' sounds so idiomatic that I wondered why I ever took the scoring for transverse flutes for granted. I also thought the arguments by Pieter Dirksen in support of his reconstruction are very plausible.

On the whole the present studio recording of this same reconstruction is a strong case for this very first version of the St John Passion.

One of the strengths of this performance is its inner coherence. The singers are all first class and completely at home in Bach's idiom. In a performance with one voice per part, in which the singers not only perform the solo parts, but the tutti as well, it is even more important that the voices blend well. And that is the case here. In the opening and closing choruses, the turbae and the chorales, one never gets the impression of a bunch of soloists who happen to sing together for one particular occasion.

The soloists all do well in their solo parts. I have not always been impressed by Gerd Türk, but here he does an excellent job in the role of Evangelist. He makes a difference between the more dramatic and the more pedestrian narrative moments. His diction and articulation are admirable.

Stephen MacLeod does not have a very strong voice and perhaps lacks a little authority in the role of Jesus, but - like Gerd Türk - I was pleasantly surprised by his performance in comparison to what I have heard from him on other occasions.

Caroline Stam has a beautiful voice, with a warm and pleasant timbre, which is very suitable to the arias she has to sing. I liked the dynamic differentiation in the B-part of 'Ich folge dir gleichfalls'.

Peter de Groot may not have the most impressive voice in comparison to some colleagues who hit the headlines. His emotional involvement in his arias is strong and makes his performances memorable. In 'Von den Stricken' he emphasises 'völlig' (zu heilen) in the B-part. Equally eloquent is the short pause before the closing exclamation 'vollbracht' in the aria 'Es ist vollbracht'.

Bas Ramselaar is perhaps the best of them all. The very tender and intimate aria 'Mein teurer Heiland' is outstanding and deeply moving. Most admirable is his ability to colour the words and phrases according to the meaning of the text. Impressive is also the dynamic differentiation within the melismas in 'Eilt, ihr angefochtnen Seelen'. By the way, I wondered why Stephen MacLeod is singing the arioso 'Betrachte, meine Seel', whereas all bass arias are sung by Bas Ramselaar.

Charles Daniels is one of those English-speaking singers who has a feel for German music, as he has demonstrated on several occasions (for instance in Paul McCreesh's recording of Schütz' Weihnachtshistorie). He sings the two - very different - arias and the arioso 'Mein Herz, indem die ganze Welt' admirably.

I have major problems with the tempi in the tenor arias, though. The first one, 'Ach, mein Sinn', is taken at a relatively slow speed. I can't figure out why. This aria follows on Peter's denial, and expresses anxiety, even desperation: "where shall I find comfort (...) In this world I can find no counsel". It seems to me this character is somewhat lost here.

The opposite happens in the second aria, 'Erwäge'. Pieter Dirksen sees this aria as the heart of the 'hidden symmetry' in the St John Passion, with a unique scoring of two viole d'amore, "and it is not by chance that this aria at the same time is by far the longest of them all". Not in this performance: at 6:40 it is the longest, but certainly not 'by far': 'Zerfließe, mein Herze' is only about 20 seconds shorter. Of all the recordings I know 'Erwäge' is here by far the fastest: most of them take at least 8 minutes.

I feel that, at this tempo, the aria doesn't get enough weight. To me a slower tempo would be more suitable to its meditative character and very strong symbolism.

I would imagine that Jos van Veldhoven has his reasons for these choices of tempo, but I can't find anything about that in the booklet.

My main problem with this recording is the lack of drama. This is partly caused by the rather slow tempo with which the part of the Evangelist is sung. It makes the telling of the story rather unnatural: it isn't as speechlike as it should be, and there is too little differentiation between stressed and unstressed syllables and words. But it also has to do with the 'turbae', some of which lack the sharp edges and the power which one would expect in a piece where 'the Jews' are portrayed as a mob.

The chorales are well sung, although some are rather too slow again, and sometimes there is a little too much legato. The opening chorus is sung with great expression. I only wished there had been more accents in the melismas on 'Herrscher'.

The contrast between the A- and B-section in the closing chorus 'Ruht wohl, ihr heiligen Gebeine' are well realised. The concluding chorale contains two elements: a prayer and a doxology. It is sung with great intensity and conviction.

The players do a great job: in particular the oboes are brilliant. In the liner notes, Pieter Dirksen writes: "without the doubling flutes, the oboes' chains of dissonant suspensions sound thinner, but also more searing". And that is certainly right: I haven't previously heard those dissonances done as poignantly as they are here. The use of the harpsichord for the accompaniment of the Evangelist is certainly historically justified, but I would imagine it takes some time to get used to it. I am less convinced about the prominent role of the theorbo in the basso continuo.

This recording comes with an opulent booklet, which contains three essays. The one I have already referred to is by Pieter Dirksen in which he explains the reconstruction he has made. In his contribution Jos van Veldhoven explains the reasoning behind the performance with one voice per part. There is also a third, written by Guus van den Hout, who writes about the way Christ's Passion has been depicted in art through the ages. The booklet also contains a number of pictures of art objects from the Museum Catharijneconvent in Utrecht, a museum for religious art.

The placement of the pages with those pictures isn't always very convenient: if you want to listen to the recording while following the lyrics in the booklet, from time to time you have to turn two or three pages to find the next lines.

I have certainly enjoyed listening to this new recording, and I shall return to it from time to time. It is a shame, though, that one of the main features of the St John Passion, its dramatic character, hasn't been fully realised.

Johan van Veen