Organ majors in many colleges and universities today are being taught to use non-legato (articulate legato, "ordinary touch") and its associated techniques (early fingering, thumbs mostly off sharps, heels mostly off pedals, abandonment of glissando and substitution) in all the parts when performing early (pre-1800) organ music and to employ legato and its associated techniques (substitution, thumb glissando, liberal use of heels in pedaling) for all organ music written during or after the 19th century, unless otherwise specified by the composer [See blog, Touch, Parts I, III].
But with J.S. Bach there's a separation here; for purposes of settling on fingering it's probably safer not to call Bach's music early; in many cases we cannot apply only early fingering to his music because of its complexity; he also uses many more accidentals in his music, and, the more accidentals the music has, the more we have to use things like thumb under or thumb on sharps.
Thumbs under and on sharps cannot be avoided when the texture of the music is so thick.
The only thing we should not do in fingering his music in comparison with modern music or Romantic music is probably finger substitution and glissando.
There are exceptions however, even in Bach; fingering gets very individual because the hand layout for each person is a little different; the span of the palm is different for each performer, some can hardly reach an octave, in which case it's even more important to work out the hand division for the middle voices in advance of our practice.
With non-legato the organist is trying to get off the keys by inserting minute but definitely audible breaks between all the notes including the bass line which, if these breaks could be counted, would amount to approximately 1/8 the written value of the notes.
In actual practice some performers, in their desire to distance themselves as much as possible from the legato school in the performance of a Bach fugue, exaggerate this broken articulation even more where the audible breaks they insert between the moving notes are stretched to more like 1/4 the written value, which morphs the touch into marcato.
Bach organ fugues have even been recorded where every note in them from start to finish including the bass line is rendered staccato at half written value, sometimes even less than half, as fast as the organist can play it -- accelerating during the opening to a breakneck tempo that makes us wonder if we're listening to a rapid fire French Romantic organ toccata instead of a fugue -- complete with energetic head bobbing and smug ornamenting added to impress the senses.
This pathetic situation, if and when it occurs, carries the modern development of playing things broken to the worst possible extreme where the clarity of the moving lines upon which the entire effect of Bach's music depends is torn asunder and destroyed [See blog, Bach d minor, Part III].
Can a piece of organ music really be destroyed by the way its performed? ... yes indeed; it can be rendered insensible merely by drawing an unsuitable registration (choice of stops).
It's conceivable though, that in the course of our studies we may run across the score of a stand alone fugue written pretty much in 18th century common practice style by a modern composer where the music is very busy, there are no indications for the touch, and all the performer knows is that the music was composed in the 21st century (for possible examples, see Op. 6, 13, 22, 23 posted on this web site).
In this situation the question arises as to choosing a starting place for the touch ... i.e., whether to obey the rule and let the modern date of composition dictate the employment of legato in all the parts or to give the style of the piece primary consideration and employ broken touch from the beginning.
It can be argued both ways -- entrenched practice (legato) vs. stylistic authenticity (broken).
This should tell us something.
Still, at times we can feel like we're in a rope pull (photo) in which we're faced with deciding which "school" (legato or broken) we're going to have to assign to wear the roller skates.
Before we make up our mind as to which end of the rope to favor (or to favor either end at all) we need to make sure our feet will be planted on firm ground in this debate by having fallen back on what we already know that's very solid.
We know, for example, that no rule in organ playing is ever "absolute" and that there is no such word in the glossary of organ playing.
We also know that the entire effect of a fugue or any other highly polyphonic piece of organ music for that matter stands or falls on the clarity of its moving lines.
We also know that the more we connect the left hand line in polyphonic music the less the right hand line is evident, the more it's compromised; thus we have to sometimes break the left hand line by detaching so the right hand can be heard more clearly.
This too should tell us something.
We also know in the performing of repertoire that sometimes we have to "change the manuscript mentally" to get clear what's on the page [See blog, Listening For The Listener]; sometimes there are so many held notes that the other moving parts aren't really audible; they are just vaguely moving; we may have to pick things up in unusual places, not hold them quite as long, and this is true in so much organ music, not just Bach (in Reger, for example, we have to change hundreds of things in the way he wrote it, but we do it because our ear tells us that we can't hear clearly everything that's on the page, and it's wonderful writing).
In the Bach fugue we're not changing a lot of things, just not connecting everything.
We also know that a strict legato touch throughout was taught to organists in the performance of Bach and held sway for about a century and a half without serious complaint from academia or the general public; many world renown 19th and 20th century concert organists in fact built their performing reputations this way.
This also should tell us something.
And, we also know that stylistic fugues reminiscent of the Bach school are still being written in the modern era, pieces that are instructive, useful, and worthwhile for an organist to learn and perform; undoubtedly, due to the inevitable progress of the art, some of the best of these fugues are still waiting to be written.
And so, in terms of taking sides for a starting touch, the short answer for how to perform a modern fugue is that it's style is more important than a rule concerning its date of composition [See blog, Contemporary Practice].
Another question which arises is whether imperceiveably broken (clean legato) touch, which is middle of the road between legato and non-legato, a touch where the organist isn't trying to get off the keys, is good to employ with modern fugues.
The answer is yes, of course; a good argument may be made for using this midway "almost legato" touch for the fastest running notes in any organ music.
This tells us something, again.
Do we really have to declare ourselves for one side or the other in this rope pull then, when it comes to playing polyphonic music, early or modern? ... it depends upon our training and what we think is right, of course, but, it's important to remember to be very broad minded these days in exposing ourselves to many influences in our performing, composing, or improvising.
Bottom line: From the standpoint of clarity, to bring out all the best the music has to offer, what kind of touch(es) do we think to employ for performing a Bach fugue? Answer: ALL of them.
What kind of touch(es) do we think to employ then for performing a modern fugue? Same answer. ALL of them.
In a major Bach work we think to play some notes legato, some imperceiveably broken, some non-legato, some marcato, and some staccato -- 5 different things depending upon the duration of the notes, their rhythmic pattern, whether a note repeats, the way the instrument breathes in its own acoustical environment, and the tempo [See blog, Touch, Part I, II].
Above all -- it's about getting things clear.
A word to the wise: if we're not already steeped in and well satisfied with the results of any particular "school" of touch there's no real need to take sides in this rope pull between diametrically opposed schools when each piece we play, whether early or modern, must be approached stylistically -- neither is there any real need to be in denial of the fact that some of the finest stylistic recordings of a major Bach work on a period instrument have employed a mixture of every kind of touch there is [See menu bar, Videos, Bach Toccata & Fugue, for just one example of this].
When it comes to touch in organ playing, if we must insist on one school that reigns supreme -- it should be the school of clarity.