The Double Chorus in Bach's St. Matthew Passion


© 2003-2005 Yale Institute of Sacred Music


The one thing everybody knows about the St. Matthew Passion is that it is a double-chorus composition, and this scoring is widely considered its most characteristic musical feature. The idea that the work balances two matched ensembles against each other - a feature typically regarded as "symmetry"-has itself been a theme of almost every discussion. The view of the St. Matthew Passion as a symmetrical double-chorus work has arisen largely from the experience of modern performances that use distinct soloists and two large choirs. Seen from a distance, the two ensembles, each visually dominated by a large number of singers, do indeed look equal.

But in light of what we know about the forces Bach used in his own performances, we need to ask ourselves what it really means to call this work a double-chorus composition. It turns out that it owes a great deal to ordinary single-chorus technique, and thus is closely related to typical passion repertory. This aspect of the work is much harder to recognize in most modern performances.

The best way to understand the work's construction is to consider a famous movement from Bach's St. John Passion, the aria "Mein teurer Heiland." It combines a solo bass singing an aria with the ripieno group, which overlays a hymn harmonized in four parts. There are two points here. First, this scoring gives the ripieno group something slightly independent to do: they sing a four-part chorale essentially on their own. (Actually only the bass is entirely on his own - the concertists double the other three lines.) Second, this movements requires the presence of the ripieno ensemble because there are two vocal bass lines in this piece, one for the aria and one in the chorale. Bach's scoring creates the slightly contradictory case of an essential ripieno group. This is an exceptional use of vocal forces that Bach specified because the text of this movement is a dialogue, realized as a musical conversation between the bass soloist and the ripieno ensemble.

It appears that this very aspect of the St. John Passion was an inspiration for Bach, and for Christian Friedrich Henrici, the librettist of the St. Matthew Passion, in creating that work. Each of its most important commentary movements is cast as an allegorical dialogue between the Daughter of Zion and the Believers. The aria with chorale from the St. John Passion just mentioned is, in fact, a dialogue between those same two allegorical characters; this is specified in the source from which Bach and his librettist borrowed the text. That movement evidently served as a model for the many dialogues in the new piece, and I am convinced that they represent Bach's starting point for his design of the work overall.

Bach handled the dialogue texts in the St. Matthew Passion by providing the work with two complete SATB vocal ensembles, expanding on the principle he used in the aria from the St. John Passion. In setting many of the St. Matthew Passion texts, Bach has one side of the dialogue sung by a soloist from one group (Chorus 1) and the other side sung by the entire four-part ensemble of the other group (Chorus 2). In a few pieces (including the opening and closing numbers) Bach uses all four voices of Chorus 1 and answers them with the four voices of Chorus 2.

This organization is intimately connected with the conception of vocal forces discussed earlier. In the original parts for this work, which also survive, we find four principal parts (Soprano, Alto, Tenor Evangelist, and Bass Jesus, all in Chorus 1) containing essentially all the Gospel narrative in addition to arias and chorales. These are, in essence, the concertists' parts, just as in the St. John Passion - in fact they are disposed exactly as in that work. There are also four additional vocal parts (Chorus 2) whose singers double all the chorales and many of the choruses, most of which proceed with the voices in each range in unison. This, too, is just like the St. John Passion - these are effectively ripienists' parts.

This is what a performance along the lines of Bach's, with its particular vocal forces, lets you hear: that Chorus 2 is fundamentally a ripieno group, acting most of the time in support of Chorus 1, which is the concertists' group. You can hear this by noticing, as I mentioned, that essentially all the Gospel narrative is in Chorus 1. You can also note that in almost every dialogue movement Chorus 1 takes the lead and Chorus 2 follows: typically, Chorus 1 makes the opening textual and musical statement, whereas Chorus 2 merely provides commentary.

A good example is the accompanied recitative and aria "O Schmerz" and "Ich will bei meinen Jesum wachen" in Part 1 of the Passion. In the recitative the tenor of Chorus 1 sings the poetic text; he is answered by a beautiful but subordinate four-part chorale in Chorus 2. In the following aria, the tenor of Chorus 1 sings the solo aria line, answered by occasional interjections from Chorus 2, which presents important but less prominent material as a group. In a performance with Bach-sized forces one can hear the secondary role of Chorus 2 compared to Chorus 1, and understand this inequality as the difference between a concertists' group and a ripieno ensemble. This is much more difficult to perceive in a performance with larger forces; in this recitative and aria, for example, to hear Chorus 2 as subordinate when its members greatly outnumber those in Chorus 1 is difficult.

Chorus 2 is subordinate throughout the work in other ways. For example, Chorus 1 sings almost all the Gospel narrative; only a few passages are in Chorus 2. When Bach uses only one ensemble for a Gospel chorus, that group is most often Chorus 1; Chorus 2 gets only two such pieces. In several of the longest and most important Gospel choruses the two vocal groups begin antiphonally, but after a few measures they combine, and the texture collapses into four-part writing. The effect, with Chorus 2 doubling Chorus 1, amounts to traditional ripieno practice.

There are a handful of true antiphonal Gospel choruses but they are few and extremely short - just a few measures each. Overall, there is very little symmetrical double-chorus writing in the whole work. From this point of view, one could say that the St. Matthew Passion isn't a double-chorus piece at all, at least not a symmetrical one, because real double-chorus scoring plays such a small role, and because the two choruses' roles are so lopsided. In a performance with large forces, the sheer volume of the combined voices and instruments tends to over-emphasize the few pieces that do use the two vocal ensembles equally.

But there is another dimension as well. In this work Bach has given his ripieno singers - Chorus 2, in my analysis - more to do than we would expect from vocalists of this type. In fact they have a lot more responsibility. Probably inspired by the dialogue texts and the subordinate but somewhat independent role Chorus 2 plays in their musical settings, Bach found additional things for his "liberated" ripieno group to do. You can see this immediately in the original vocal parts for Chorus 2, which include not just choruses and chorales but - amazingly - solo arias as well.

Dividing the usual concertists' duties among two sets of vocalists, Bach assigned solo arias to his secondary singers, the ones in Chorus 2. Some of the arias in the St. Matthew Passion are performed by the singers and instrumentalists of Chorus 1, as you would expect from their concertist status, but other arias come from Chorus 2. From this point of view, the four singers in Chorus 2 are not really ripienists but four additional concertists, just like those in Chorus 1. In their solo arias, the voices of Chorus 2 move out of their subordinate role into the spotlight usually reserved for principal singers.

Many modern performances obscure this essential feature of the St. Matthew Passion because they use only one soloist to sing all the arias in a given vocal range, backed sometimes by the instruments of Chorus 1 and other times by those of Chorus 2. This is in contrast to a performance with the labor divided as Bach partitioned it, in which arias are presented from each side. The modern disposition gives the impression that the double-chorus aspects of the work, such as they are, reside in the choral ensembles and in the instrumental groups. In fact, the division of forces starts with the principal singers.

A performance with two sets of singers also lets us hear that Chorus 2's status is secondary because Bach's assignment of arias is hardly equal. Once again Chorus 1 takes priority. Each of the voices of Chorus 1 other than the tenor (who sings the Evangelist's words) has more arias than its Chorus 2 counterpart. Overall, in fact, Chorus 1 has twice as many arias, and its arias also make much greater vocal demands. For example, Tenor 2 sings "Geduld" with basso continuo only, and its companion recitative "Mein Jesus schweigt" with its transparent accompaniment. Tenor 1 has to contend with a solo oboe, strings, and with the second-chorus forces both in the recitative "O Schmerz" and in the aria "Ich will bei meinen Jesu wachen." The lone aria for Soprano 2, "Blute nur," hardly compares in difficulty to "Ich will dir mein Herze schenken" and "Aus Liebe will mein Heiland sterben" required of Soprano 1. In a typical modern performance, with the arias in each range all sung by the same person, this distinction disappears.

As Bach revised the St. Matthew Passion over the years, he made changes that strengthened the work's double-chorus features. Perhaps the most important was the provision of two distinct basso continuo groups, one for each chorus. Originally, the two ensembles had been served by one continuo group, just as a typical single-chorus passion would have been. The new scoring made each of the choirs musically more complete and potentially independent, and strengthened the impression-or the illusion-that this is a double-chorus work. In this regard, a typical performance by balanced forces takes the work a step further in a direction mapped out by Bach himself.

But it still represents something modern. We inevitably do this in performing old works, creating new versions or even (in some sense) making new compositions of them. The St. Matthew Passion presents an especially striking example, because even the way we staff the vocal lines in a performance has so many effects: on the nature of the chorus and on the role and duties of the soloists; on the prominence or even the very presence of a dramatic element; and on the work's fundamental disposition as a double-chorus composition. The meaning of a musical work clearly does not reside in the score alone, but depends greatly on how it is performed. This is worth remembering, if only now and then.


The material in this essay is drawn from Daniel R. Melamed, Hearing Bach's Passions (New York: Oxford University Press, 2005), which includes suggestions for further reading and listening. The recording of the St. Matthew Passion conducted by Paul McCreesh (Archiv 474 200-2) deploys forces almost exactly as documented in Bach's performing materials and discussed here.

Daniel R. Melamed is Professor of Music at the Indiana University Jacobs School of Music. He is the author of »Hearing Bach's Passions« and »J. S. Bach and the German Motet«, and co-author with Michael Marissen of »An Introduction to Bach Studies«. He serves as associate editor of the Journal of Musicology.