Baroque Tempo

[Vox Humana]

I doubt that many forum members have much appetite for scholarly articles, but this one might be of interest to those who like to know about how things were done historically, even if (like me) they do not necessarily feel bound to follow the information in performance.

I have long had a growing conviction that speeds during the Baroque era were often, if not usually, a great deal slower than very many performances we hear today, not only on the organ (the now-usual approach to Buxtehude seems especially wrong-headed to me), but also more generally, including in choral music. My convictions have been almost entirely empirical, so I was very pleased to find an article in an e-journal presenting hard evidence towards the same conclusion. It is Beverly Jerold's article Numbers and Tempo: 1630-1800 here: http://scholarship.claremont.edu/ppr/vol17/iss1/

I don't necessarily agree with every point Ms Jerold makes (I feel very unsure about the section "The mathematical possibilities of "half"), but overall the article seems to add up. I have often bemoaned how fast speeds can make Baroque music sound trite when it ought to sound noble, so I was rather tickled to see a quote towards the end of the article from Jean-Marie Leclair in 1753 making the exact same point translated with those very same words. It is nice to see Ms Jerold concluding that, in the end, it is the music that matters - but I have never yet come across a scholar who would disagree with that.

One thing I will say: I found it immensely difficult when reading this article to comprehend the data about speeds without involuntarily cross-referencing to the internal, mental metronome I have developed over the years, courtesy of the real thing. It really did highlight the fundamental difference between them and us that Jerold points out at the beginning of her article. One needs to strip out these preconceptions and stop cross-referencing to received habits. When a contemporary described Bach's playing as very fast, this needs to be understood by the standards of his time, not ours.

Incidentally, the May 2014 issue of the journal Early Music contains another article by Ms Jerold about notes inégales which is by far the most practical exposition of the rules that I have ever seen. It establishes definitively (so it seems) that notes inégales were not observed outside France (except by Muffat when deliberately writing in the French style). It also makes the interesting point that it was an aid to teaching students how to understand and feel rhythm (in an age without metronomes). Abstract here: http://em.oxfordjournals.org/content/42/2.toc

[Colin Pykett]

Tempo means time, and this has been one of my hobby horses for a good while. There are many questions and not many answers, so instead of pretending I know them, I'll just reel out a few of the questions.

 

In those days how was time measured? The long case (grandfather) clock had not been around all that long - Newton had only just explained properly how the pendulum worked. It depends on gravity, and that was not within the grasp of human knowledge until he came along.

 

And portable time - meaning how was time measured at the console for purposes such as checking beat rates when tuning (important for setting the imperfectly-tuned intervals in all temperaments, particularly an unfamiliar one) and checking the speed at which you are playing (the issue in the post above)? The metronome as we know it was not invented until the 19th century, though it was an acknowledged problem much earlier and attempts were being made to produce clock-like devices which musicians could use around the end of the 17th century. As far as I know, they were not very practical though. Apart from anything else, they didn't 'tick' very loudly, if at all.

 

Pocket watches? Yes, they were around and there is a portrait of Henry VIII wearing a huge one round his neck (therefore not really a 'pocket' watch). This is because they were extremely expensive and a symbol of great wealth. But they were not very reliable nor good timekeepers, and the sweep seconds hand which is required by a musician did not exist as far as I know. Harrison's exquisite chronometers did not appear until the mid-18th century, but even so they were scarcely the sort of thing your average working musician could have got hold of. In any case, we are speaking of the Baroque era here, not the Classical.

 

Wrist watches - duh, no, 19th century stuff.

 

So what did those 'ordinary' people do (bearing in mind that, apart from his mighty intellect, even Bach was 'ordinary' in the sense of his origins and means)? I don't know. Did they routinely have some sort of innate sense of time which we today have since lost perhaps? One of my favourite conjectures, which has generated both approval and ridicule elsewhere, is that they might have used a simple pocket pendulum such as piece of string with a bob (weight) on the end. (i.e. a builder's plumb line, which has been commonplace for millenia). A child could have observed that the bob swung periodically while her/his dad was doing a bit of masonry work. If s/he later became a musician maybe s/he might have hit on the idea of using one for musical purposes - the string could have had marks on it to correspond to the beat rates of a favourite temperament perhaps, or to indicate tempi much as the wand and sliding weight of a metronome does today. A musician could unwind the string to the required mark and off s/he would go.

 

Similar cheap plastic devices (pocket metronomes) paralleling the construction of the handyman's expanding steel ruler used the same principle. They were available from virtually all music shops until the advent of today's electronic metronomes. My first organ teacher used one at the console. They might still be available today - I haven't checked. Were they, by any chance, the final evolved form of the 'musician's plumb line' I've postulated above?

 

But if not, what else could they have done?

 

I could go on and on, and there are indeed yet other options, but as pointed out above, scholarly articles are not really what the forum wants so I won't give any references to my purple prose on the matter here.

 

CEP

[Guest Geoff McMahon]

I have long thought that the speeds we hear today cannot be historically correct. When we remember pieces which were written for particular performers who were unable to play them because they were too difficult to play and now are common place in the repertoire, it seems to me evident that the standard of performance has been going up over the centuries. What basis is there for playing fast? In the sleeve notes of Harnoncourt's Bach B-Minor Mass, he quotes one of Bach's sons as stating "Mein Vater spielte seine Werke recht lebhaft" (my father played his works very quickly). Quickly in relation to what? In that recording of the B-Minor Mass, Harnoncourt takes a pace which is so fast that the horn players simply can't play their notes and it sounds awful. Andrew Parrott has done some research on this and I believe he discovered a source which indicated how long a specific (Bach?) performance had taken, but I forget which and what. I might try to find out.

 

The interesting question, if we assume performing speeds have gone up over the years is why? I think there are two reasons. Firstly, simply because performers can play faster, just as our athletes can run (quite a bit) faster. In organ playing, I suspect some of the reason is the lack of acoustic in many American churches. Performers tend to play faster in a dead acoustic anyway, as the music seems to demand it, or, to put it the other way round, in a long and lively acoustic, a musician will tend to play more slowly, so that it doesn't all become an unclear wash of notes.

 

John

[Contrabombarde]

Would not improvements in performers' technique that allows them to play faster go hand in hand with technical improvements in organ design? Having played a few exceeding "clattery" Dutch organs of the seventeenth and eighteenth centuries with awkward pedalboards I felt the organs pretty much forced me to play no faster than a certain tempo. Fast forward a couple of centuries to electropneumatic action and concave radiating pedals and fast French toccatas become feasible.

 

I wonder what JSB would make of an instrument such as Liverpool Anglican Cathedral were he to be transported into the 21st century? I expect initial disbelief that the organ in the south case could begin to be controlled by a console in the north case, with an impossibly heavy touch - followed by disbelief that the touch is similar to a small portative organ!

[Vox Humana]

On 04/08/2014 at 08:00, John Pike Mander said:

The interesting question, if we assume performing speeds have gone up over the years is why? I think there are two reasons. Firstly, simply because performers can play faster, just as our athletes can run (quite a bit) faster. In organ playing, I suspect some of the reason is the lack of acoustic in many American churches. Performers tend to play faster in a dead acoustic anyway, as the music seems to demand it, or, to put it the other way round, in a long and lively acoustic, a musician will tend to play more slowly, so that it doesn't all become an unclear wash of notes.

I think a very important factor here from about 1950 onwards (I don't suppose it had much impact before then) is the increasing awareness of recorded performances to act as points of reference. I think it is also quite likely that, as the market gradually became awash with recordings of the more popular pieces, performers were driven to find new interpretations in a way that would not have arisen before the days of recordings. An easy way to make your performance more exciting than someone else's is to take it just that bit faster. Before recordings, assuming you were lucky enough to hear a good performance in the first place, all you could keep was a memory, which would very likely play tricks with you over time. I suspect, too, that it is recordings that have driven up technique by raising awareness of the benchmarks. Before recordings there were simply no means to make a performance permanent for future reference.

To pick up Colin's point about pendulums, Jerold's article details some French advice about these. Although pocket pendulums would be technically possible, how would you be able to tell what it was they were measuring and whether they were accurate? You could calibrate one so that it is X times as fast as a big pendulum, but, even if you made the latter according to the prescriptions in one of the books you would still not have an entirely reliable point of reference. Also, a pendulum does not have an audible click like a metronome so it is inherently less precise from the start. How do you tell the precise point at which the pendulum begins to swing the other way? The slower the swing, the more difficult it is to be sure. I feel this even with the click of a metronome: I find that maintaining 48 beats a minute consistently and evenly is less easy than maintaining 100. The other way of measuring time was with a heart beat, but this, too, is very imprecise.

[pcnd5584]

I have long thought that the speeds we hear today cannot be historically correct. When we remember pieces which were written for particular performers who were unable to play them because they were too difficult to play and now are common place in the repertoire, it seems to me evident that the standard of performance has been going up over the centuries. What basis is there for playing fast? In the sleeve notes of Harnoncourt's Bach B-Minor Mass, he quotes one of Bach's sons as stating "Mein Vater spielte seine Werke recht lebhaft" (my father played his works very quickly). Quickly in relation to what? In that recording of the B-Minor Mass, Harnoncourt takes a pace which is so fast that the horn players simply can't play their notes and it sounds awful. Andrew Parrott has done some research on this and I believe he discovered a source which indicated how long a specific (Bach?) performance had taken, but I forget which and what. I might try to find out.

 

The interesting question, if we assume performing speeds have gone up over the years is why? I think there are two reasons. Firstly, simply because performers can play faster, just as our athletes can run (quite a bit) faster. In organ playing, I suspect some of the reason is the lack of acoustic in many American churches. Performers tend to play faster in a dead acoustic anyway, as the music seems to demand it, or, to put it the other way round, in a long and lively acoustic, a musician will tend to play more slowly, so that it doesn't all become an unclear wash of notes.

 

John

 

Indeed - I would agree with this and the points made by Vox.

 

You raise the important point that we have no real reference point or bench-mark by which to judge these matters. However, I should be most interested in the outcome of your research into Andrew Parrot's conclusions.

 

Another factor which may have influenced performance speeds is the gradual increase in the pace of life. I suspect that if one were able to put Bach (or Mozart) in Oxford Circus, for example, they would both be entirely overwhelmed by the frenetic activity.

 

It also ties in with an earlier discussion regarding pupils' performance speeds and in particular, those from the Far East. I have had, on many occasions, to encourage pupils to slow down a piece (often considerably). This is particularly true of the music of Bach and Mozart. YouTube is partly the culprit, here. I suspect that many who post themselves playing on this media channel are doing it, partly in order to 'show off'. Unfortunately, it often gives the misleading impression that the piece in question has to be played at this speed. Unfortunately, as John has observed, this often results in mistakes and a sheer lack of musicality in the performance - simply because the pace is far too fast.

 

With regard to optimum speeds of examination pieces, I rarely resort to the use of a metronome. For one thing, the speeds given are often editorial (and when I have checked them, they almost always appear to me to be too fast). I prefer to allow the piece itself, the figuration and where relevant, historical context to give an indication of how fast the piece should be played. I see nothing musical, for example, in playing the first movement of a Mozart sonata as fast as possible, purely because it is largely notated in semiquavers. If the end result is unmusical, then the speed is wrong - particularly when it is evident that a student's technique does not allow them to play the piece accurately, fluently and securely.

[Colin Pykett]

Quickly in relation to what?

 

John's question hits the nail right on the head. Absolute, as opposed to relative, time is the issue which which Vox Humana started this topic and which I addressed in post #2. Issues such as notes inégales relate to relative time and are therefore of a different kind.

 

We should also remember that there is probably no single 'correct' answer in any of this. Let's assume that an organist in village A used a plumb line type of pendulum to set his tempi (or temperaments) as I suggested. This does not mean that his colleague in village B would have done so. If the two villages were more than a few miles apart they would seldom have met. Travel was so difficult and time consuming in those days that it is sometimes easy to forget this when we can just jump into a car - the only way to travel was on shanks's or a real pony, and then only if the weather was decent.

 

Travel difficulties were a major reason why pitch standards varied so much (Cammerton, Chorton and all that, and even these might have only specified a range of frequencies, not a single one). The same applies when we look at the plethora of temperaments used in the Baroque era - how many organ builders would even know what their colleagues more than a few miles away were doing? I can scarcely believe that many of them would have read books by Mersenne, Werckmeister, etc, and they probably hadn't even seen them.

 

CEP

[Tony Newnham]

Hi

 

My piano & organ teacher carried a portable metronome - a device like a tape measure, but with calibrations for musical tempi and acting as a pendulum. It's really not that difficult to obtain the relevant tempo from such a device. Not sure where she got it from - or if such things are still available (this would have been laye 1950's/early '60's).

 

Every Blessing

 

Tony

[pcnd5584]

Hi

 

My piano & organ teacher carried a portable metronome - a device like a tape measure, but with calibrations for musical tempi and acting as a pendulum. It's really not that difficult to obtain the relevant tempo from such a device. Not sure where she got it from - or if such things are still available (this would have been laye 1950's/early '60's).

 

Every Blessing

 

Tony

 

Although 'relevant tempo' is often either arbitrary or conjectural.

 

With regard to the Associated Board, often it is nothing other than a current editor's idea of how fast a particular work should be played.

 

There are also instances of composers giving metronome marks which were challenged by performers. In several cases, the composer was quite happy to accede to the view of the performer. One such case was Duprê and Cochereau; another was, I believe Howells, although I cannot recall the performer.

[Colin Pykett]

My piano & organ teacher carried a portable metronome - a device like a tape measure, but with calibrations for musical tempi and acting as a pendulum. It's really not that difficult to obtain the relevant tempo from such a device.

 

Exactly. The very thing I was describing above Tony, and as I said, my organ teacher used one as well. I think the salient point is that such a device does the job well enough, rather than being a highly precise scientific instrument in the sense a clock pendulum has to be if it is to keep good time over several days.

 

It would probably also be good enough for setting the initial beat rates of the tempered intervals in a chosen temperament, which could then be tweaked by ear to taste by an expert tuner when laying the bearings.

 

Although I completely take on board Vox's reservations which he made in #5, the fact that such devices were used in my (and Tony's) lifetime shows that they were not show stoppers.

 

Incidentally, calibrating a plumb line type of pendulum is not difficult and would not have been then (i.e. post-Newton). The length of a 'seconds' pendulum, one which takes one second to swing from one extremity to the other, is almost exactly one metre. Provided the bob is small compared to the pendulum length, and much more massive than the string, the time period does not depend on the weight of the bob for practical purposes (not many people know that!). The time period then varies as the square root of the pendulum length. (Good old Newton, and Galileo before him, were jolly clever to work all this out, weren't they).

 

Simple, as the meerkats would say, and it would have been simple then to those in the know I suspect.

 

CEP

[Malcolm Kemp]

I likewise agree that Baroque tempi these days seem to be far too fast. If you do the Bach Motets too fast they simply become a blurred jumble. John Mander singles out Harnoncourt as a culprit; I find nearly all Harnoncourt's tempi excessively fast, particularly so in Mozart Mass recordings. Ultimately we don't know for sure what Bach and his contemporaries did and we need to make educated guesses/judgments information available.

 

There is a similar problem, particular to organists, with the term legato. There are those who try to say that Bach played with the kind of absolute legato we would associate, perhaps with Franck or a Field Nocturne and there are those who try to play Bach with a totally detached staccato. I don't subscribe to either of these theories. I suspect that the nearest one could get to describing what is wanted is an "articulated legato" - neither a Romantic type legato nor staccato.

 

Surely ultimately, whether considering tempi or touch in playing Baroque music, the ear has to be the ultimate judge, taking into account the size and acoustics of the space it's being performed in and the number of people in the room. I have several DVDs of Barenboim giving both piano and orchestra conducting masterclasses; he constantly advises that the tempo is the LAST thing you need to decide upon after having taken everything else into consideration first.

 

Malcolm

[Colin Pykett]

Is there not another factor as well to take into account? Over a lifetime, would it have been the case that a given musician varied her/his tempi as well as other nuances of their playing? If so, might this have been partly unconscious and partly the result of experience and learning? Did Bach play a given organ piece in the same way when he was that rumbustious teenager at Arnstadt compared with the way he played it towards the end of his life in his sixties?

 

Mostly rhetorical questions, and I certainly do not have the answers even if others do (which is why I am asking the questions here). But I have observed similar phenomena affecting the way some modern players and conductors (by which I mean those whose lives have coincided with the ability to record their performances) subtly change their technique as they move towards a mature style.

 

CEP

[Malcolm Kemp]

Colin - Your comment about Bach gradually changing his playing as he aged/matured brought Dupré instantly to my mind. Some of his best known recordings - not necessarily his best technically - were made when he was getting quite old.

 

Malcolm

[Colin Pykett]

Speaking of slow tempi, though not a Baroque one in this case, how about this - Widor himself playing "the" Toccata in old age:

 

https://www.youtube.com/watch?v=J8vz1D_L_OE

 

It's well known of course and I expect most members will have heard it. However, the way he achieves the required subjective accent on the first of each of the recurring quaver-pairs comes through this old recording very well. Not all modern players come near this, and one reason might be that they take it too fast.

 

I also find the timing of his pedal part interesting. Compensating for the slowness of the action and the speech of the huge pipes perhaps?

 

CEP

[Malcolm Kemp]

Not only is ridiculously fast tempi a problem in Baroque music; some Anglican church music - and especially Psalms - is now performed significantly faster than it was fifty years ago. Perhaps the problem with the Baroque tempi is like so much with life these days; everyone's in such a hurry all the time. I'm sure I've read somewhere (perhaps in the writing of Peter Williams, I'm not sure) that Bach was very aware of the speed of the average human heartbeat. Surely that gives us some indication of what he was thinking?

 

Malcolm

[sbarber49]

Not only is ridiculously fast tempi a problem in Baroque music; some Anglican church music - and especially Psalms - is now performed significantly faster than it was fifty years ago. Perhaps the problem with the Baroque tempi is like so much with life these days; everyone's in such a hurry all the time. I'm sure I've read somewhere (perhaps in the writing of Peter Williams, I'm not sure) that Bach was very aware of the speed of the average human heartbeat. Surely that gives us some indication of what he was thinking?

 

Malcolm

Not to mention hymns!

[Vox Humana]

On 04/08/2014 at 11:17, Malcolm Kemp said:

There is a similar problem, particular to organists, with the term legato. There are those who try to say that Bach played with the kind of absolute legato we would associate, perhaps with Franck or a Field Nocturne and there are those who try to play Bach with a totally detached staccato. I don't subscribe to either of these theories. I suspect that the nearest one could get to describing what is wanted is an "articulated legato" - neither a Romantic type legato nor staccato.

Most organists - and not only them - don't bother to verify their information. (I wonder how many people reading these posts have read the article which sparked off this thread?) It proceeds almost like Chinese Whispers, people regurgitating the convenient bits they have been told that appear to support their preferred style of playing. The result is that all sorts of rubbish gets attributed to Bach and bandied around, e.g. "he used his heels." I suspect he did, but there's no proof. The only specific quote I have seen regarding his touch is a remark in Ernst Ludwig Gerber's Tonkünstler-Lexicon (1746) where he says "His [Schröter's] manner could not possibly please those who knew Bach's legato manner of playing, for he played everything staccato." But all this tells us is that Schröter played in a manner which agreed with the 1746 understanding of staccato (whatever that may have been) and that Bach's playing contrasted sufficiently to be described as legato. We can't assume that either term agrees with twentieth- or twenty-first-century understanding. What exactly it did mean would have to be teased out from other contemporary information. I am sure this must have been done, but I am not well enough read to take this any further. The current wisdom (unless I am totally misunderstanding it) is that early legato was a near-legato that admitted an almost undetectable sliver of silence between the notes. I have seen one or two early texts adduced to back up this near-legato, but I have never been quite convinced that this is their inescapable interpretation. It is certainly one possibility, but nothing I have yet seen precludes a clean, yet connected legato either because I haven't yet seen an unequivocal mention of that momentary silence. There may be one, of course. There are some clues in Bach's music that tend to support the legato hypothesis: e.g. a downward octave leap in the left hand where the first (upper) note is sustained over the second, thus precluding the staccato that many of us like for the sake of clarity. There are also some interesting observations in the commentary of the new B&H edition of (I think) the Passacaglia. In England, Francis Linley wrote, probably in the 1790s, that diapason movements should be played by "gliding the notes and chords into each other"- which doesn't really sound like "articulated legato" to me.

I agree that pace of life must be another factor likely to affect perception of speed. I mentioned it once before, but in baroque times the fastest thing in everyone's life (apart from bullets and arrows) was a horse. I was brought up in a relatively rural environment where the roads were relatively clear, farming was non-intensive (fields were full of Grey Partridges) and the pace of life was very much slower than it is today. Performance speeds seem to have increased in tandem. It would be interesting to see whether proper research would confirm this impression.

Jerold's article makes the point that pendulums and heart beats were all most people had as time-measurement tools. One copy of Thomas Tomkins's Musica Deo Sacra has a note (probably by his son Nathaniel) to the effect that a semibreve took the same time as two heart beats. I calculated this speed as minim = 72; Wulstan came up with minim = 76. It depends on the heart and the speed of a normal heart can be anything from 60 to 100 beats per minute! In any case, as Wulstan observed, whilst many of Tomkins's anthems work well at m = 76, many don't.

[Colin Pykett]

I will quite understand if members feel I'm becoming an utter nuisance on this topic, but it does coincide exactly with my interest in time, both as a professional physicist and an amateur musician. One of the beautiful aspects of our universe which unites both subjects and pulls them together in a most mysterious way is time.

 

But because of my amateur status as a musician, I am therefore grateful for what I have learnt from others who have posted here. I find Vox Humana's thoughtful posts, in particular, most revealing and helpful, and although he modestly tells us (often) that he doesn't know everything, he knows a darn sight more than I.

 

In #17 Vox said that "most organists - and not only them - don't bother to verify their information". That might be true, I don't know, but it's certainly the case that one can get a lot of help from what might seem to be dry academic texts at first sight. I'll quote from Fenner Douglass's wonderful book ('The Language of the Classical French Organ'):

 

"One antique organ, played with the grace and delicacy the music invites, can teach us more than a dozen treatises on registration, on ornaments, or even on fine points of organ building. The theoretical comments out of their proper contexts remain at best a dangerous asset".

 

Against that backdrop, Douglass then analyses in exquisite detail those pre-Revolutionary organs themselves and how they were played, including the way they might have been registered and nuances of performance including ornaments and staccato/legato.

 

Because of Bach's interest in French music, might he have absorbed some of this into his own performance style? He would probably have come across enough people who had worked on French organs and knew how they were played, of which Gottfried Silbermann's brother Andreas was but one. Andreas also trained Gottfried, and JSB knew him well.

 

I'll leave it there. Thank you for your patience if you've read this far.

 

CEP

[Vox Humana]

To my resounding shame, I have never yet read Fenner Douglass's book. I must make a point of it when I get back from holiday. It is not expensive.

Bach was certainly well versed in French taste from an early age since as a teenager he went frequently to hear the Duke of Celle's band, which was composed mostly of French musicians. It has been observed that his well-known ornament table is basically French. In 1796 J. F. Reichardt (1752-1814) mentions that Bach combined the French style of ornamentation with the individual style of fingering and performance that he had developed. Reichardt was recycling secondhand information at best, but perhaps it is accurate. I doubt one should give too much weight to the French element, however. Bach made a deliberate point of being an all-round musician and he was just as familiar with Italian music as French. In composition he took what was useful to him from all styles and I see no reason why this should not have been true of his playing as well. There is good reason to believe that, if there was any French element in his playing, it did not include notes inégales (see the last paragraph of my post #1) Nevertheless, registering An Wasserflüssen Babylon BWV 653 as a Tierce en taille and Dies sind die Heil'gen zehn Gebot' BWV 678 as a Cromhorne en taille seems to work for both pieces, although I would love to confirm this on an historical organ.

I wonder whether our modern printing, with its general emphasis on clarity, with neatly spaced and aligned notes, encourages speed. Most Bach autographs I have seen are clear enough, but there are usually a few elephant traps where the notation is cramped. I did once try to play the P&F in G BWV 541 from a facsimile of the very cramped autograph. I didn't get very far. I have also tried playing from the generously-sized facsimile of Neumeister's MS with more, but still less than complete success (partly due to the abundant errors). But it can be argued that the organists of old would be perfectly used to playing from such scores and therefore better able to deal with them. Also it may well be that memory played a more important part for them than it does for most of us, so I am not sure whether the manuscripts are a significant factor or not.