Metronomes and microtime

[timothy42b]

I am one who sees a metronome as extremely useful, and many people here including Bernard have offered cogent arguments why they don't agree.  It occurs to me that we may be talking about two kinds of time.

I offer these definitions.  Macrotime is the seemingly simple ability to keep a steady beat, putting four beats into an average measure most of the time, at the same speed.  This is a skill beginners lack.  Some people never learn it.  It includes subdividing, accurately playing dotted rhythms, triplets, etc., but the focus is primarily on steady internal time.

Microtime is the subtle variation of performance around the macrotime beat, without degrading it.  This is not obvious, it is inferred more than recognized, but it is critical to good live performance.  Bernard noted in an earlier post that playing absolutely evenly gives a dead sound, and that top performers do not do so;  he suggests metronome practice can harm the ability to do microtime.  Listen to a big band, or any of the Latin or Cuban bands, live, and you will hear this but not hear it.  The performer may or may not be conscious of it;  the audience will definitely not be conscious of it, but it has a musical effect that registers. 

I assert, without proof but with some evidence, that a metronome is the most efficient method of internalizing macrotime in a beginner or a person with difficulties.  I think it must be active practice with it, passively listening to anything does no good. 

I don't see how a metronome can help at all with microtime, and it very well may hinder it.  I think the way you learn microtime is by doing a lot of listening to people who do it well, preferably live;  and by playing with other musicians.  One problem with recordings, especially popular music, is that so much of it is now electronically "fixed" in the digital studio.  Pop singers frequently have uncertain time and pitch, but after recording the notes are quantized so they all land on the beat, and even tweaked for pitch.  Listening to a lot of popular music of any genre is not going to help with microtime, though it is passive enough a process that it probably won't hurt either. 

It may seem like good macrotime is a given, but I play and sing regularly with people who are very bad at it.  It is more than common, in some types of performance it is almost the norm.  Because it annoys me so much, I've become a metronome fan, in moderation of course.

[bernhard]

I totally agree with you. I believe that there are many circumstances when a metronome is useful, and in at least two of them the metronome is mandatory:

1.   For students who cannot distinguish “rhythm” (in the sense of the relative time values of notes) from “pulse” (the steady beat that supports rhythm). In fact, I have never found an alternative to the metronome in these cases.

2.   For creating an aural representation of the music in one’s mind in real time. Again, I have not come across any better alternative for people with difficulties in this area besides the metronome.

Perhaps the most important warning against constant metronome use is that music cannot be satisfactorily brought to life with a totally even pulse. Although I agree with your ideas on microtime, this is something else altogether. Let me expand on it, because it came to me as a bit of a surprise and it gave new meaning to much discussed concepts such as “rubato”.

I happen to have a clavinova 970, which amongst other features came with an internal metronome and a set of 50 pre-recorded classical pieces. Meddling with this digital piano, I discovered that when playing the pre-recorded pieces, I could alter their tempo by changing the internal metronome speed settings. But there was something else as well. If I played (that is, the piano played) a prerecorded  piece with the metronome on, the speed setting did not remain constant. As an example, the prerecorded version of Debussy’s “Arabesque” which was supposed to be played around MM=134, showed a variation in the pulse that ranged from MM=34 to MM = 220! By way of contrast, the prerecorded version of Bach’s invention 1 ranged from MM=83 to MM=89, dropping to MM=44 on the last bar where the pianist used a rit.

There are several interesting (and to me unexpected) conclusions here: Romantic repertory cannot be played in an even pulse. Flexibility of note-value around a steady pulse (the usual definition of rubato – keep the left hand steady, but let the right hand dance around it – in Chopin’s famous direction) is not going to be enough. The pulse itself must be contracted and expanded in accordance with the music’s requirements, as evidenced by the digital piano metronome measurements at every point. Baroque repertory, on the other hand was played mostly evenly. One wonders if this is not just a consequence of our ears being used to it being played on a much more restrict range of pulse variation. I wondered what the Invention what sound like with a range as extreme as that used on the Arabesque. I did not have to search long. Wolfgang Rubsam’s recording of the inventions and Sinfonias for Naxos presented just such a rendition. It does take many listenings to stop getting irked by it, but by then the orthodox pulse-steady renditions start to sound dry and stale, and I for one wonder if Rubsam is not into something big here (Pletnev’s delightful interpretation of Scarlatti sonatas, which uses the same pulse elasticity also comes to mind).

So what I am trying to get at, is that this sort of pulse flexibility – which I believe is part and parcel of superior musicianship – rather then microtime, is what is really at risk when extensive use of the metronome is preconised.

At the same time, the random pulse of most beginners is if anything even worse than a completely even pulse.

So I would see three stages in students:

1.   Total randomness of pulse, caused by not even being aware the such a thing as pulse exist. At that stage metronome use is mandatory.

2.   A very steady pulse, even when the metronome has been discontinued. If this student is ever to provide musical renditions of a piece, he must go beyond that stage. At this stage metronome use must be discontinued.

3.   A fine judgement in regards to how to expand and contract pulse giving the most delightful musical renditions of a piece and truly bringing the music to life. The metronome has no place in such a situation. Now, it may happen that a total beginner, intuitively has this fine judgement. This where the teacher has to be very careful not to confuse this fine expansion and contraction of pulse with the chaotic pulse of the student in 1 above, and insist in “correcting” the student by insisting on a metronomic rendition of the piece. It is not uncommon for profoundly unmusical teachers to replace musicality with blind adherence to time values of notes in a score (forgetting that our musical notation although very good for pitch is very inadequate for rhythm). Or as Nikolaus Harnocourt once said, “let us not confuse faithfulness to the score with faithfulness to the music”

Best wishes,
Bernhard.

[whynot]

Good stuff, you guys.

[ted]

I am quite pleased somebody has actually said that at last, Bernhard. I've never thought that keeping one hand steady and varying the other is anything more than a subset of complete rhythmic freedom. Giving it the label "rubato", which has developed an ambiguous meaning anyway, doesn't help an advanced player much less a beginner. There are places for that "one-sided" bending of rhythm, of course; stride and ragtime are two obvious cases. The left usually remains solid while the right does perorations around it. It is actually pretty difficult to learn, judging by the seemingly small number of players who can do it well.

John Arpin, in his collected interpretations of Joplin, has shown us that even here, in an idiom often regarded as compusorily metronomic, a total bending of rhythm in both hands can produce musically valid and very exciting results. Granted, he has been subject to heavy criticism, but I for one experienced an opening of the gates when I first heard him. Of course, if you do it unthinkingly it sounds terrible, but so does insensitive variation of any element.

Rhythm permits infinite variation just like any other aspect of music. Sadly, I have come to the conclusion that a depressingly large number of otherwise top-flight executants have no intuitive grasp of it at all, mostly regarding it as a regularly arranged matrix of hooks on which the purportedly more important elements of melody and harmony can be hung. Exactly how this unfortunate state can be avoided during formative years lies with good teachers such as you. I haven't the faintest idea; I'm just glad that I dodged the trap myself.

[dougiedog]

  when I play with a metronome  I am slightly ahead of the beat. If I practice with it long enough will the metronome pull my brain in sync with it  or will it be a source of constant
aggravation?

[bernhard]

I am quite pleased somebody has actually said that at last, Bernhard. I've never thought that keeping one hand steady and varying the other is anything more than a subset of complete rhythmic freedom. Giving it the label "rubato", which has developed an ambiguous meaning anyway, doesn't help an advanced player much less a beginner. There are places for that "one-sided" bending of rhythm, of course; stride and ragtime are two obvious cases. The left usually remains solid while the right does perorations around it. It is actually pretty difficult to learn, judging by the seemingly small number of players who can do it well.

John Arpin, in his collected interpretations of Joplin, has shown us that even here, in an idiom often regarded as compusorily metronomic, a total bending of rhythm in both hands can produce musically valid and very exciting results. Granted, he has been subject to heavy criticism, but I for one experienced an opening of the gates when I first heard him. Of course, if you do it unthinkingly it sounds terrible, but so does insensitive variation of any element.

Rhythm permits infinite variation just like any other aspect of music. Sadly, I have come to the conclusion that a depressingly large number of otherwise top-flight executants have no intuitive grasp of it at all, mostly regarding it as a regularly arranged matrix of hooks on which the purportedly more important elements of melody and harmony can be hung. Exactly how this unfortunate state can be avoided during formative years lies with good teachers such as you. I haven't the faintest idea; I'm just glad that I dodged the trap myself.

Yes, Western music does tend to pay less attention to rhythm than to melody and harmony, and as I said, our notation system which is excellent for pitch is terribly poor for rhythm. Our rhythmic notation when taken at face value will necessarily result in regular, metronomic rendition. With live styles (such as the waltz or swing) this is understood and one does the necessary adjustments. With lost traditions (such as Bransles, Galliards and other dances/genres for which a live tradition has not survived) we are pretty much lost.

Year ago, it amazed me how some pianists/keyboardists of African/Latin origin could – without much musical literacy – play incredible renditions of their styles of music. Then I had an epiphany. They were not playing the piano. They were playing drums on the piano, that is, the piano for them was simply a drum with the added feature of having pitches (which some drums actually have). So the rhythmic element (which they had an intuition for, either from listening to it from birth, or because they were also drummers and percussionists) far outdid the melodic and harmonic elements. Stating it simply, they were doing the most trivial chord progressions, but they were tessellating them with the most complex rhythms. So when “exact” transcriptions of what they were playing became available, they were always disappointing because the rhythm flexibility was lost in notation. Trying to play it from the score always sounded dead an unmusical. The problem of course is that to bring such music to life in a like manner requires one to have (or try to develop) the same rhythmic intuition.

Best wishes,
Bernhard.

[JPRitchie]

If one takes a composition, implements it in MIDI strictly according to the score and then compares it with a pianist's version, there will be differences. In the case of Joplin rags, no tempo variations were marked and Joplin's tempo notes alluded to marches, as in "Slow March Time". Instrumental differences aside, which is to be preferred: the rendition that keeps the time in the score or that which does not? With all due respect to pianists' considerable abilities and popular successes, I've found Arpin's "Euphonic Sounds" to sound rushed in spots, as do parts of Rifkin's "Bethena". But, I might never have arrived at this conclusion had I not heard the corresponding MIDI, which revealed the composer's intentions. If I understand microtime correctly, it could be easily implemented in MIDI by modification of the start time and duration of
notes in a measure, while conserving their sum to the number of beats indicated in the time signature. So, with or without microtime doesn't arise solely from a technology limitation. If one then assumes at least a modest transcriptional ability of the composer, which is justified for Joplin, Liszt, and others by their demonstrated abilities to transcribe for piano train collisions, profane exclamations, and the flight of angels, how can one logically justify departures from the notated rhythms as artistic? Or that the composer was so careless with the score that it prevented a realization of the intended music? Composers can use phrasing to indicate microtime (or sometimes just legato) - so there is something to indicate it - should it be called for. Nancarrow is extreme, but used multiple time signatures. If you can play an existing  piece and put your name on it, I'm willing to accept it as your interpretation. And some analysis of what gives it its unique sound might be called for as a guide for others. But, compare a composer's approach: In the 1924 Warner Bros. edition of "Rhapsody in Blue" for 2 pianos and 4 hands, there is a forward that describes the rhythmic variations in the music. They adhere to an objective and rational device: in consecutive groups of 4 notes, every third one is accented. This sort of variation is designed and rational, so it's a high level construct. In contrast, microtime lacks an objective measure that distinguishes it from simply not keeping good time or whether it's rushed vs. inspired. I guess that the only thing I'm trying to say is that such things as microtime haven't led me to a better understanding of the composer's sonic version of the score, which I admit a desire to discover or create if it doesn't already exist.

Regards,
Jim Ritchie

[bernhard]

If one takes a composition, implements it in MIDI strictly according to the score and then compares it with a pianist's version, there will be differences. In the case of Joplin rags, no tempo variations were marked and Joplin's tempo notes alluded to marches, as in "Slow March Time". Instrumental differences aside, which is to be preferred: the rendition that keeps the time in the score or that which does not? With all due respect to pianists' considerable abilities and popular successes, I've found Arpin's "Euphonic Sounds" to sound rushed in spots, as do parts of Rifkin's "Bethena". But, I might never have arrived at this conclusion had I not heard the corresponding MIDI, which revealed the composer's intentions. If I understand microtime correctly, it could be easily implemented in MIDI by modification of the start time and duration of
notes in a measure, while conserving their sum to the number of beats indicated in the time signature. So, with or without microtime doesn't arise solely from a technology limitation. If one then assumes at least a modest transcriptional ability of the composer, which is justified for Joplin, Liszt, and others by their demonstrated abilities to transcribe for piano train collisions, profane exclamations, and the flight of angels, how can one logically justify departures from the notated rhythms as artistic? Or that the composer was so careless with the score that it prevented a realization of the intended music? Composers can use phrasing to indicate microtime (or sometimes just legato) - so there is something to indicate it - should it be called for. Nancarrow is extreme, but used multiple time signatures. If you can play an existing  piece and put your name on it, I'm willing to accept it as your interpretation. And some analysis of what gives it its unique sound might be called for as a guide for others. But, compare a composer's approach: In the 1924 Warner Bros. edition of "Rhapsody in Blue" for 2 pianos and 4 hands, there is a forward that describes the rhythmic variations in the music. They adhere to an objective and rational device: in consecutive groups of 4 notes, every third one is accented. This sort of variation is designed and rational, so it's a high level construct. In contrast, microtime lacks an objective measure that distinguishes it from simply not keeping good time or whether it's rushed vs. inspired. I guess that the only thing I'm trying to say is that such things as microtime haven't led me to a better understanding of the composer's sonic version of the score, which I admit a desire to discover or create if it doesn't already exist.

Regards,
Jim Ritchie

Composer intentions --> score --> midi ---> music according to composer's intentions.

I totally agree with your contentions in regards to midi. In fact, with disklavier the whole score may become obsolete.

However I do have a most fundamental disagreement over this part of the process:

Composer intentions --> score --> midi

You seem to be under the impression that the score actually contains the composer's intentions, and faithfulness to the score will guarantee faithfulness to the composer's intentions. This is not so. Here is an interesting quote from a supremely interesting - and necessary - book that I suggest you get aquainted with:

”We believe we possess a system of notation which will inform us about both the individual tone as well as the course of the musical piece. However, every musician should know that this notation is very inexact, that it does not precisely say what it does say: it does not tell us the length of tone, the pitch, nor the tempo, because the technical criteria for this kind of information cannot be conveyed by notation. The duration of a note can only be precisely described by a time unit; the pitch of a tone can only be represented in terms of vibration frequency; a constant tempo might be indicated by a metronome – if there were such a thing as a constant tempo.”

(Nikolaus Harnoncourt: “Baroque Music Today: Music as Speech” – Amadeus Press).

Maybe this should become the norm:

Composer ---> Disklavier ---> midi ----> perfect performance.

(Unfortunately, most composers these days cannot paly the disklavier with the necessary competence )

Best wishes,
Bernhard.

[JPRitchie]

Hi Bernhard,

   Thanks for the reply and the suggested reading material, which I will look at at least a little bit.  And, I will also look to see if Beethoven, Liszt, Joplin, and other competent performers ever remarked about how limited they found notation. Pardon my ignorance, but did Harnocourt ever compose something transcriptional - like the flight of a bumblebee? Although I really like some Baroque music (AMB's Mussette and Pachelbel's Magnificat and others), I've found authentication trails hard to find or verify, so haven't given it very serious attention.

    On the other hand, with more reliable authentic sources, here's what has continually happened when I've programmed compositions: they're different from any performance I've heard. Some are more different than others, but some are conspicuously and macroscopically different. In such cases, the score explicitly supports my interpretation. Case in point:"FigLeaf Rag"; it's marked at 100 qps, yet rarely played that way; likewise the B section of Maple Leaf Rag (marked staccato) and the climactic trill in Magnetic Rag ( marked "ff").  Yes, in these cases, I am under the impression that the score accurately and adequately, even if qualitatively, communicates the composer's intentions. Similarly, syncopation and note durations at faster tempi are both notationally precise. Moreover, in the case of Joplin, many of his works were cut by hand (reproducing pianos being unavailable until about 1910-1912) into player piano rolls. In such instances at least, the score certainly does have special significance as being a guide to the composer's intentions.

     My approach is:
Programmer+score --> MIDI --> Disklavier/Pianomation/Opus7/keyboard/synth --> composer's performance
The MIDI can alternatively be obtained by recording a manual performance. FWIW, I don't believe there is nor is it desirable that there be a "one and only" performance of any composition - scores are silent or qualitative on many aspects so, yes, the performance isn't completely pre-defined; instruments differ also. If there is some indication of microtime, one would certainly be remiss in not putting it in. My interest is to find the composer's performance with the aid of the score - whatever the inadequacies, authentic scores remain the guide the composer gave; it may take some work to understand it, but what else would one choose? And the work should be defended as any other performance. Another attraction of MIDI, aside from pedagogical applications, is its capability to create new kinds of performances. But, AFAIK, if not exclusively, most composers were initially capable performers.

    To get back to microtime, I can imagine circumstances in which it might be desirable and implied:  to achieve a specific effect - such as to emphasize the gondola's swaying in parts of Chopin's Barcarolle (Op. 60).

Best regards,
Jim Ritchie

[timothy42b]

I see that the distinction I was trying to make between basic time and microtime is nowhere near as clear as I'd thought, and maybe not so useful.   

I think of basic time as that which you count, and microtime as that which you can't count, but must feel. 

Rubato I include with basic time.  You count it, but you speed or slow the count. 

Gross examples of microtime, that might really fit on either side, would be swung eighth notes in jazz, or the early second beat in a Viennese waltz.  Better examples might be the need for the bass to slightly lead the beat in some eras of big band music, or for the soloist to lay slightly behind in a combo. 

Yes, MIDI can do it all, if you turn quantize off, but this seems hardly useful to me.  If you turn quantize off and play MIDI in manually from a keyboard, you will see all sorts of weird rhythmic figures.  Quintuple dotted notes, etc.  Turn quantize on, and they'll all line up as quarter notes or whatever you select.  But if I saw a quintuple dotted quarter in the music, I wouldn't have a clue how to count or play it.  Playing in a group, I might be able to feel it and play it correctly. 

This is one of the defects in the music I do for church.  When the organist isn't present, I record the service music in software and burn a CD.  I use a notation program and the notes are all perfectly in time.  They sound a little dead to me, and I would prefer live.  (I'm not at the performance level to just play it yet.  Maybe some day.)  The congregation doesn't complain, recorded is better than none, but it bothers me.  (There is one benefit.  Over time, I have trained them not to drag.  They sing at tempo or get left behind, none of this slowing down to follow them!) 

[JPRitchie]

Hello Tim,
     I only brought up MIDI to make sure I understood your microtime idea; but it could be used to include microtime in a programmed work.
     As long as you brought up church music, are you referring to the use of Disklavier's or other MIDI player instrument? Your reference to the notes lining up is suggestive. Such instruments offer a nice alternative to a recorded performance and, unlike Yamaha's Disklavier, some units can be retrofitted to an acoustic piano. The ones I'm familiar with also offer a record capability, so one can either record or program a performance for subsequent acoustic playback.
Regards,
Jim Ritchie

[bernhard]

Here is another relevant (I think) quote form Harnoncourt:

[…]
Let us take as an example […] Viennese dance music of the 19th century, a polka or waltz by Strauss. The composer tried to write down whatever notes were necessary, in his opinion, for the musicians who sat before him in the orchestra. After all, they knew quite well what a waltz or a polka sounded like and how such dances should be played. If this music were given to an orchestra which lacked this knowledge, which was unfamiliar with these dances, and the musicians were to play exactly in accordance with the notes, the music would sound totally different. It is not possible to write down such dance music precisely as it should be played. Often a note must be played earlier or later, or shorter or longer than it is written, etc. Thus we could play this music as precisely as possible, even with metronomical precision – and yet the result would have nothing to do with the work as originally intended by the composer.
[…]
(my highlighting)

And have a look here for more Harnoncourt quotes:

https://www.pianostreet.com/smf/index.php?PHPSESSID=4fa3efdbe079f94cdc6de62d8d92663e&topic=6879.msg68389#msg68389

Best wishes,
Bernhard.

[JPRitchie]

Hi Bernhard,

    The quotes you mention are specifically most pertinent to Baroque music, which has its own unique considerations and which Harnoncourt studied extensively.

Here's another quote from the book (p.29):

In the case of compositions written after 1800, using notation as the directions for performance (e.g. works by Berlioz, Richard Strauss and others), the primary consideration is to describe as precisely as possible how the written work is supposed to sound; only when these notes are precisely performed, only when all instructions are observed, does the music emerge.

The last independent clause seems especially pertinent - and Harnoncourt restricts it to post-1800 compositions. The history of music covers many centuries; some progress in standardization, etc. is apparent when carefully considered.

Moreover, there are instances where works were cut by hand into piano rolls and the resulting recordings taken as a legitimate representation of the work (Joplin's stuff on Biograph BLP 1006Q).

Regards,
Jim Ritchie

[faulty_damper]

Maybe this should become the norm:

Composer ---> Disklavier ---> midi ----> perfect performance.

(Unfortunately, most composers these days cannot paly the disklavier with the necessary competence )

I disagree that composers are the authority for the works they compose.  I think composers are actually the worst interpeters of their own music even if they had the technical ability to record it.  There is a big difference between an engineer/architect and those that use their designs.  Those that use it will find its most efficient and effective use, id est the performer.  The performer is the one who decides how to interpret a work.  The composer may fall short of ideas or get too caught up in trying to write something that may be out of context in relation to the whole, something that the performer will recognize as not being cogent enough to merit any consideration.


On the topic of rubato, more specifically the micro-robbing (and giving back) of time, the reason why it is such a powerful effect is because it creates very subtle contrast that is not readily percievable.  A broader approach to contrast is understanding perception because it all falls under how we percieve information and how our minds will extrapolate, interpret, and finally predict what will come next.

Our minds are exceptionally good at prediction.  In fact, our very livelyhood depends on how well we are able to predict to future with some degree of certainty.  Why do we know that when we let go of something that it will fall?  Or even more distantly predictable, why we go to college?  It is because we percieve information and interpret it in a manner that is rational and predict the outcome.

When we recieve one bit of information, it means nothing (assuming it is theoretically in a vacuum unrelated to anything else).  When we recieve another bit of information of similar type, our minds will grasp onto it and retroactively grasp onto the first bit of information.  Our minds then interpret both pieces of information, and predict what would come next.  Quick example:

1

This means little other than it is a number.
Second bit of information:

2

This means much more.  It is the number that usually follows 1.  We can now predict what the next bit of information could be and that could be...

3


This is how our minds predict the future.  But what if this information was steady, highly predictable?

4
5
6
7
8
9
10
11
...

We would soon lose interest because our minds have had enough information to interpret and predict what will be next.  It is simply numbers, the next greater than the previous by one.  Unfortunately, as interesting as seeing numbers appear in such a manner, we would lose interest very quickly and then ignore it.  This is what the metrenome does, it simple puts out a steady supply of reference beats that always falls in time.  If a metrenome were to be turned on, without using it in the usual manner, we would soon ignore it and forget it is even turned on, like the quiet drone of the air conditioner or the refrigerator.

But what if that steady supply of numbers were to change?

12
11
13
12
14
13
15
...

Then our minds will be much more engaged, perplexed by its sudden change.  What does it mean?  Why did it change the way it did?  If the pattern were to continue, we would soon see a pattern in the numbers.  Do you see it?  Or do you need more information to interpret?

Unfortunately, if this were to continue it would still not hold our interest.  Why?  Because we would have figured out what the numbers are doing and once it is understood, we lose interest because while the information is continuing, our minds are not interested in it.  It is like an ostinato, it may be interesting at first but if it were to continue we would lose interest because we would predict that it would just go on and on.  However, after we have completely ignored it, we can suddenly be aware of its presence by it stopping, just like when the air conditioneror the refrigerator stops humming.

This is how our minds are able to interpret information and predict what the next bit of information will be and occur.  As in music, it is not enough just to have contrasting pitches that fall in certain time.  The microtime effect, the subtle rubato that is not noticed has a profound effect in how we interpret and predict what will come next.  It adds to the mix of pitches and rhythm and it changes each with just enough contrast to make the information just a little bit more interesting.

My own short quotable on rubato:  The most effective rubato is the one you don't notice.


On a digression:  If our minds are unable to put related information together (synthesize it), there would be massive confusion.  As people, we don't like to be confused; it disrupts our sense of control.  But if the information were not very related, we may interpret it as separate information unrelated to the whole.  When we are in a concert and someone in the audience coughs, we would know that it is not part of the performance because the information is different from the more highly related information.  But for some reason, that noticeable out-of-place information is usually followed by even more out-of-place information.  Soon many audience members start coughing... spontaneous "chance music" (redundancy).

[leahcim]

I disagree that composers are the authority for the works they compose.

I think they have to be, by definition - who else is going to be "the authority" and the alternative, that there isn't one, wouldn't give you a place from which you could describe your own divergance or interpretation.

Simply put, if you're "interpreting" then I'd say you've already accepted some authority - it doesn't seem the same as buying a chair and using it to reach the top shelf or putting a piano in a room in the house the architect plans marked as "bedroom"

A broader approach to contrast is understanding perception because it all falls under how we percieve information and how our minds will extrapolate, interpret, and finally predict what will come next....<snipped>

The theory seems to fall down for anyone that listens to a recording of music more than once and has a memory.

If anything, the reverse will happen and, for example, a piano teacher that has taught a piece for over 35 years and guided the students they teach to play a particular piece, and listened to recordings they like over and over, would only be surprised / engaged / interested if the piece _didn't_ have the thing you're saying is required to keep their interest.

So why use it? Your theory suggests that if you play 100% mechanically the audience with a wide knowledge of classical music will get the effect of their attention being engaged etc. I'd say, that might be true, but perhaps not in the way you intended to get their attention.

Although the premise seems true much of what you're saying is experience, not prediction afaict. Gravity for example is one that gets pilots from time to time aiui, especially if the cues for level flight aren't there and something falls "up"

[JPRitchie]

Just a follow-up to my post here. Per Bernhard's suggestion, I read Harnoncourt's book. It's a rare kind, offering the first-hand insights of a contemporary professionally successful musician.

In general, Harnoncourt specifically and frequently distinguishes between music before and after the French Revolution (which incidentally many musicians supported)(1789-Assembly of Notables, Storming of the Bastille; 1792 - Louis XVI & Marie Antionette imprisoned and monarchy abolished; 1793- King and Queen guillotined; 1793-1794 - Regin of Terror; 1795-1799 - Rule of the Directory; 1799 - Napoleon becomes First Consul; 1804-Napoleon declared emperor; 1815 - Napoleon's final defeat at Waterloo). Prior to about 1800, Harnoncourt considers:

In general, music prior to about 1800 is notated according to the work-principle and thereafter as a direction for performance.

Although a single year is given, this is only a rough guide. Harnoncourt observes:

...there are several deviations: for example, as early as the 16 and 17th centuries, the tablatures (finger notations) for certain instruments are strictly directions for playing. ... This is an extreme example of notation as direction for performance."

Although not specifically not denoted as such by Harnoncourt, another exception, but in the opposite sense, to the year of division are the Vienesse waltzes of J. Strauss, Jr. (1825-1899), discussed above and rendered according to the work-principle. Richard Strauss on the other hand is cited by Harnoncourt specifically as using "performance notation" and lived 1864-1949. Mahler (1860-1911) is also given as another extreme example of using "performance notation". Rather than a specific time period, perhaps it might be better to distinguish among composers.

In this regard, those who worked with player pianos, such as Joplin and Nancarrow deserve special consideration because the technology permits highly precise scorings and vice versa. Moreover, Chopin's Revolutionary Etude carries a metronome mark of 160, while several of Joplin's works carry metronome marks, indicating a quite precise intention for performance.

Finally, in studying composers and whether their intentions can be adequately notated, one might look for evidence of their transcriptional ability. Onamonapia can serve as one benchmark.  Joplin's "Crush Collision March" seems an ample demonstration to me of his ability to notate a score precisely.

Regards,
Jim Ritchie
P.S. The example of number sequences is too simple. 3 does not necessarily follow logically from 1,2; 1,2,4 is equally valid by doubling the previous number. Any finite series of numbers can be assigned a not-identically-zero probability from random number generation. Sequences of musical notes sometime do follow common patterns - chord progressions proceeding to resolution for example. But, they can also be unexpected, even radical - and sometimes they are. Of course, neither random numbers nor musical notes are very interesting by themselves; one either seeks or imposes order to achieve something. And that... that can be a quite creative undertaking.
jpr