I won't follow up too directly on the vibrato controversy (lots of interesting things in those posts), but try to go through my own thinking process on historical authenticity. John says some interesting things:
For a while now, the recordings of Bach which claim to be "Historically Accurate" have bothered me for some reason. I'm always left asking myself if it was worth it. To go through all that trouble to try to recreate a sound which no one can claim to be 100% accurate. In some cases the recordings to me come out calculated, and bland. There are some exceptions, but I won't go naming these recordings as that isn't the point of this exercise, but instead offer the following argument.
Is there a difference between "Historically Accurate" and "Historically Informed"? for example, can a choir of two hundred singers and an equally mammoth orchestra perform Handel's Messiah and call it "Historically Informed" considering they've taken the time to prepare the score with a sense of Baroque style, articulation, and nuance even though the performing forces are too large, and the instruments too "new"?
The question of vibrato, I think, is answered so well in Kenneth Woods' post:
"I can’t help but feel that in all music the “non-vibrato sempre” method is a weak-minded cop-out, an easy way to avoid thinking about whether, when, why and how to vibrate, a process which demands an awareness of harmony, instrumentation, color and taste. It stops the process of thinking, listening, responding and contemplating sound dead in its tracks."
How can we expect modern players to remove vibrato without removing the soul of their performance? I'm not saying we should be adding "Bel Canto" vibrato or rubato into the music of Bach, but we should not be afraid to let the instruments sing.
There is also evidence that early keyboard players did not use their thumbs! Why don't we ask our organists and harpsichordists to do the same? Because it would probably take away from the musical ability of the performer - which to me is like removing the left-hand vibrato motion of the string player.
I've been interested in baroque music and performance practice for a long time, beginning when I was an undergraduate, listening to recordings of music by Schütz, Bach and others, and I started a group called The Bach Ensemble in Seattle when I was 23 years old. This instrumental/vocal ensemble performed Bach cantatas once a month--everyone got the music in advance, I usually worked with smaller combinations (soloists, continuo, solo instruments) in advance, but the basic schedule was a choral rehearsal Saturday morning, overlap with the instrumentalists' rehearsal to run through the chorus(es), rehearse orchestra (and soloists), then on Sunday, do a dress rehearsal in the late afternoon, take a short break, then do the concert. We quickly added other baroque repertoire to the program, sometimes chamber works, solo vocal works with continuo and a few instruments, and occasionally orchestral or choral works. One season we did works by Schütz on every program.
This was a fabulous learning experience, as most of these works were not recorded, so I had to learn the music from the scores alone and make decisions about dynamics, articulations, tempi, bowings, etc. from the internal evidence in the music, plus whatever I could figure out from reading about baroque performance practice and listening to various recordings of baroque music.
We were, of course, using modern instruments (gambas and recorders for works such as Cantata 106), but I was beginning to think in terms of period instruments and this was when, for example, the big Bach cantata project with Leonhardt and Harnoncourt began. I'd done Cantata 4 (Christ lag in Todesbanden) with my church choir earlier and was looking forward to Harnoncourt's recording of the cantata to answer a performance question for myself: in one of the movements the bass line plays dotted eighth-sixteenth notes throughout, the soloists sing "Alleluia" with a series of triplets, followed by two eighths--so a question is: do you adapt the dotted rhythm to the triplets? Do the soloists adapt the duple rhythm of the two eighths to the triplet? I remember thinking, "In this recording I'll find out the right way to do this!" As I remember, when I finally heard the recording, the continuo adapted to the triplet, one of the soloists sang the eighths straight, and the other adapted to the triplet! So much for definitive answers!
However, the series had excellent liner notes, including details about problems in the original parts or score (this is always interesting: not just for baroque music--see Norman Del Mar's Orchestral Variations--Confusion and Error in the Orchestral Repertoire (unfortunately long out of print), Del Mar's other books on Conducting Beethoven, Conducting Brahms, etc., and the many articles in the Journal of the Conductor's Guild on errata. Harnoncourt, especially, wrote about articulation and whether slurs, dots or other markings in one part of a movement would translate to others.
During this time there was also a considerable debate in the Choral Journal about "authenticity," with statements by Rilling, for example, on how, even if one could hear a performance by Bach himself, given our ears (conditioned by listening to Beethoven, Wagner, and 20th century music), we wouldn't hear as an audience would have in the 18th century. This is true, but to me, not the point. Rilling, of course, has changed his performance practice considerably (even though he uses modern instruments)--I sang in a performance of the Mass in B Minor in 1972 with him (more about that one day--it was a transformative experience) and, for example, he added the violas to the violins in the Agnus Dei for a richer string sound (very romantic sound, lush, full vibrato). I just heard a performance of the Mass by him in Eugene last month and it's not even close to the same performance--a very different articulation in the opening Kyrie fugue, for example).
I'm not saying we shouldn't change our minds, of course! I hope I learn something between successive performances of the same piece! But in a very real sense, our ears change, too, as we adapt to hearing new and different ideas of performance style.
Of course, there are no definitive answers to some of these questions and absolutely there is an element of fashion here. Once something catches on (an articulation, "swells" on long notes, etc.), others pick it up and it becomes a part of "style," whether or not it is correct.
But just because you can't ever get to an historical "truth" isn't reason enough to fail to pursue it. The "truth" in a performance also has to include the psychological and emotional truth in the music--that's sometimes harder to get. In that sense, I agree totally with John that if you pursue authenticity that results in blandness of expression, you've lost what the music was about in the first place.
However, I don't think the pursuit of authenticity has to mean bland performances. And it shouldn't!
During the time I was conducting The Bach Ensemble, Stanley Ritchie moved to Seattle as first violinist of the Philadelphia String Quartet (in residence at the University of Washington). Stanley also had a significant background as a baroque violinist and was in a duo with harpsichordist Elisabeth Wright. (Stanley has now for some time been in charge of the baroque orchestral program at Indiana University). He was also concertmaster of the New York City Opera and assistant concertmaster at the Metropolitan Opera, so he's clearly an outstanding violinist in any style.
I worked with Stanley for a period of time to learn about baroque violin techniques and (at that time) how to adapt those techniques to the modern instrument and player. Here's where questions of period instruments really can begin to inform. For example, the baroque violinist didn't use a chin rest or shoulder rest. Without being able to hold the instrument between shoulder and chin, the left hand has to support the instrument more. That doesn't make vibrato impossible, but you can't vibrate all the time and with the same intensity that one can with a chin and shoulder rest. It also changes some fingerings, since the violinist has to "crawl" between positions part of the time.
Gut strings also make a difference in sound and how much one can dig into the string (at a certain level of pressure the string simply doesn't speak well). The bow itself, shorter and lighter at the tip, doesn't allow for as much pressure as one can make with a modern bow. That means that dynamics are created more by bow speed than pressure (Ken Woods makes the excellent point that in Elgar's time, playing into the string is a part of the style of the time).
These are just a few things that learning about period instruments tells us.
Briefly back to vibrato: in the opening movement of the Bach Johannespassion, the flutes and oboes play a series of suspensions. When I do it, I ask the winds to play senza vibrato, since that heightens the dissonance--and therefore, the expressivity of those passages. I think that "authenticity" doesn't have to mean bland--it's in how you approach it.
John mentioned earlier keyboard practices of playing without using thumbs. While this doesn't have to mean that a modern player can't use their thumbs, it does teach something about articulation and how notes would/could be grouped. I think the major point is to take this knowledge and use it to learn more about how the music was done and what it expressed (and how it expressed it).
The element of how audiences perceive what we do (through their own preconceptions) is also interesting. For example, for quite a while I've done a "middle-European" pronunciation of Latin in works where I think it's appropriate (Haydn, Beethoven, and Bruckner masses, for example). One does run the risk that the audience will focus more on the "weird" Latin than on the music. Any time we hear something that doesn't meet our expectations, it has the potential to take us out of the moment. This has often been mentioned with nudity in film--do you continue to be involved in the story and characters, or think, "Wow, that's so-and-so with her top off"?
We can't help those reactions and must take them into account in our choices. Are the gains in a German pronunciation of "Kyrie" or "Agnus" worth the potential losses as some audience members think about Latin pronunciation rather than the music? You've got to at least ask the question, I think. And it will be the more "educated" in the audience who will notice it! I remember doing the Rheinberger Cantus Missae with my PLU choir on tour one year and we were using Germanic pronunciation. I got almost no reaction to it (other than people liking the beautiful mass) until we did a performance in Tucson, where my predecessor at PLU, Maurice Skones, was the head of the choral program at the U of Arizona. Afterwards he asked me, "What do you think all the grad students will be talking about tomorrow after hearing your performance?" Of course, it was the pronunciation of the Latin!
This post has gotten too long and is definitely too rambling, but these are important and intriguing questions! So, enough already.