In Oxford Composer Companions: J. S. Bach, John Butt calls the B-minor Mass "one of the 'key works' in the movement promoting historically informed performance." At a minimum "historically informed performance" (HIP) means using "period instruments." Beyond that, it means that the performers are informed about the music's historical background, particularly performance practice. Additionally, the performers should be informed by the period instruments they play. They need to take them not as "primitive" forms of modern instruments but respectfully, on their own terms, so as to discover their peculiar strengths.
Of these three components of HIP, understanding and mastery of the instruments were the last to be widely achieved. Of course there were some outstanding performers on period instruments in previous generations, but it has taken time to develop a critical mass of professional "early" musicians who compare favorably with the best of their "modern" colleagues. When Nicholas Harnoncourt and Gustav Leonhardt recorded all the Bach cantatas a generation ago, they did not have performers who were consistently capable of excellent "period" playing and singing. That some of their performances were brilliant was due to their compelling conception of how the music should "go," and to some outstanding performers (e.g., Franz Briiggen on flute and recorder). That it is inconsistent is due to supporting casts without sufficient experience to execute fully their conductors' conceptions. Today their heirs—John Eliot Gardiner, Phillipe Herreweghe, Ton Koopman, and Masaaki Suzuki (to mention but a few)—work with instrumental ensembles, choirs, and soloists who consistently measure up to very high professional standards.
Nicholas Harnoncourt's 1968 recording was the first HIP recording of the B-minor Mass. It issued a challenge to a Romantic tradition of performance represented by Otto Klemperer's recording of a year earlier. That tradition favored large choir and symphony orchestra. It was imbued with the spirit of nineteenth-century Romanticism and had little concern for matters of historically appropriate style. In 1967, performances like Klemperer's had about a century of tradition behind them, which puts about a century between Bach's death (1750) and the beginning of the tradition. What happened in that gap? Next to nothing as far as the performance of Bach's music is concerned. He died shortly after completing the B-tninor Mass, never having heard it performed in its entirety, and in the century following his death, the performance of his music nearly vanished.
We know of only one performance of any part of the B-minor Mass during the half-century that followed Bach's death. The "Credo," replete with an instrumental introduction by his son, Carl Philipp Emmanuel, was performed in Hamburg in 1786. A few more partial performances took place during the first half of the nineteenth century, but the first known performance of the entire Mass did not occur until 1859 in Leipzig. In 1861 it was performed in Frankfurt. After that, performances spread and became quite numerous. The first performance in England took place in 1876, conducted by the husband of the famous "Swedish Nightingale," Jenny Lind, who trained the choir, sang the soprano solos, and also sang in the choir. The American premiere was by the Bethlehem Bach Choir in 1900.
During the century between Bach’s death and the revival of his works, major changes had taken place in musical style and performance. So when his music began to be performed again, it was according to nineteenth-century taste, with little regard for stylistically appropriate tempos, dynamics, phrasing, and articulation. Baroque violins, oboes, trumpets, etc., were seen as "primitive" so no thought was given to resurrecting them. By the early twentieth century there were a few voices crying in the wilderness. Wanda Landowska was a formidable advocate for the harpsichord. Albert Schweitzer spoke against the performance of Bach with nineteenth-century choral and orchestral forces and for clearing away nineteenth-century accretions from editions of his music. But throughout the first half of the twentieth century the typical performances of Bach's music continued to be in the Romantic tradition.
Harnoncourt's recording started a trend in which HIP took the lead. I don't know whether the lead is numerical, but its lead in influence is unquestionable. Due to the influence of "early" performers, "modern" performers are changing their ways of performing Bach. A few years ago, the violinist James Buswell played and spoke at Calvin College. He talked about experimenting with a Baroque violin. Though much inspired by the instrument, he concluded that he could not do full justice to both Baroque and modern violins. Not wanting to abandon repertory that required the modern instrument, he put away the Baroque instrument, but not what he had learned from it. He plays Bach wonderfully on his modern violin, incorporating what he learned from the Baroque violin. Another example is the pianist Andras Schiff who, as John Butt has said, "has gone right to the forefront of Bach performance because he has... integrated what he has learned [from harpsichordists] into a pianistic sound." A similar thing is happening with Bach's choral music. Compare earlier and later performances of the B-minor Mass by great "modern" conductors like Robert Shaw and Helmuth Rilling. The later ones clearly show the influence of HIP even when aspects of the Romantic tradition still remain (as they do especially in Shaw).
The most obvious difference between HIP and the Romantic tradition of performing Bach's choral works is in size. The Romantic tradition favors massive forces while HIP insists on small ensembles (3-4 singers per part in a choir and a comparably sized orchestra). Now, except for some performances by school ensembles and choral societies where educational or social considerations are important, "modern" performers usually reduce their forces considerably, often to HIP size. For example, Rilling, in his latest recording of the Mass uses practically the same number of vocalists and instrumentalists as Gardiner and Briiggen. There is widespread agreement that smaller ensembles are necessary for making one of the greatest glories of Bach's music audible—its intricate, contrapuntal textures.
In 1982 Joshua Rifkin recorded the B-minor Mass using one singer on a part—in the choruses! His performance is based on evidence that suggests that one singer on a part was the norm for Bach's choral music. Though his theory is plausible, it is hardly unassailable. Not surprisingly, it has not been widely accepted. Nonetheless, Andrew Parrott and Conrad Junghanel have each recorded performances that follow Rifkin's lead (though with some use of two on a part). They are worth listening to for their clarity of texture and generally excellent performances.
There is one aspect of the "one-on-a-part" theory that has gained wider acceptance. Both Shaw and Gardiner use one voice on a part in sections of choruses where the voices are accompanied by continuo only or only sparsely accompanied. There are good historical reasons for this practice. I find it highly convincing musically.
Smaller forces allow for another obvious difference— livelier tempos. Simply because they are less weighty, smaller forces are quicker and more agile. Of course that doesn't necessarily mean they should go faster. But quicker tempos are generally more effective for Baroque music for two reasons. First, the movement of much Baroque music is based on dance. Therefore it needs to be "felt" in relation to bodily movement, even in slow movements. Second, quicker tempos allow for more incisive articulation of the rhetorical figures that are essential for its expression. Articulation in general is central to Baroque expressivity, and it is in this area that the most important difference between HIP and the Romantic tradition resides. This is also the area where the gap is hardest to close. Putting subtly nuanced articulation of rhetorical figures at the center of a "modern" performer's expressive arsenal is not easily accomplished because it requires unlearning some deeply ingrained attitudes and techniques and learning new ones. It helps, at least for the techniques, to have the right tools—namely the period instruments. All other things being equal, it stands to reason, and to my ears has been amply demonstrated, that the tools designed specifically for the task—namely period instruments—work the best.
Having said that, I hasten to add that I have no desire to see "modern" musicians leave the performance of the B-minor Mass to their "early" colleagues. Our battered and broken world needs as many opportunities as possible to hear "the greatest musical artwork of all times and peoples" (as it was advertised by its first publisher). This music that William Buckley said is "inexplicable without a belief in God" can challenge those who do not yet say "credo," and it can wonderfully fortify the faith of those who do.
Calvin Stapert is the author of My Only Comfort: Death, Deliverance, and Discipleship in the Music of Bach.