Fingerprinting

[Peter]

As an example of my above post, while playing through a piece by Liszt I considered the pedaling technique used as compared with Chopin. There are a lot of similarities, I think between the composers, but the technique is much different. The overall effect that each was trying to achieve is different. These kinds of things are not expressly integrated in the scores, but through the scores we can get an idea of what messages are being conveyed. However, the actual performance will bring these nuances out that simply cannot be notated via standard notation, i.e., a pause here, a brief accelerando, all created without notational indication to establish the personality of the composer within the work.

But these are interpretive qualities which will vary from performer to performer and there are those who are capable of distinguishing one pianist from another based on what you are saying - this doesn't however alter the fact that a highly trained musician can distinguish a composer merely from the score without the aid of performance.

[Sorrano]

I think you miss my point, Peter. We are taught how to perform certain composers; do you play the music of Beethoven in a similar fashion as Schubert, even though they are contemporaries? Or Weber, or Kuhlau? This is just an aspect of discerning a composer by method of performance, obviously this does not include the fingerprints already present in the score, but it helps us to determine with more precision of what is in the score by how we approach it. The playing of staccato, for example, will differ from Beethoven to Mozart, but the total appearance in the scores may exactly the same. I look at this, not so much as a direct connection to the fingerprints that do exist, but as an evidence that there are, indeed, fingerprints within the music, performed or simply scored.

[Peter]

I think you miss my point, Peter. We are taught how to perform certain composers; do you play the music of Beethoven in a similar fashion as Schubert, even though they are contemporaries? Or Weber, or Kuhlau? This is just an aspect of discerning a composer by method of performance, obviously this does not include the fingerprints already present in the score, but it helps us to determine with more precision of what is in the score by how we approach it. The playing of staccato, for example, will differ from Beethoven to Mozart, but the total appearance in the scores may exactly the same. I look at this, not so much as a direct connection to the fingerprints that do exist, but as an evidence that there are, indeed, fingerprints within the music, performed or simply scored.

No I see your point, but it is about interpretation which is not what Euan is referring to. Yes staccato can be performed in many ways and sometimes the staccato dot in Beethoven (and later composers) doesn't even mean staccato, but an accent as in the demisemiquaver passage in the 1st movt of Op.110, but if a performer misinterprets that it doesn't alter the fact that a musicologist should be able to recognise the fingerprint of Beethoven.

[Sorrano]

Interpretation practices can go a long way in identifying specific details about the music. Dynamics, for example play a different role in Beethoven than they do in Mozart. To compare a Beethoven Symphony with a Mozart Symphony (score only) we do not see that difference that we hear when the scores are performed. But the sounds that we do hear that are not necessarily differentiated by the scores indicate with immediacy whether we are listening to Beethoven or Mozart.

I might add, too, that you have an inherent problem in the score of editorial interpretations that never met approval by the composer.

[Peter]

Interpretation practices can go a long way in identifying specific details about the music. Dynamics, for example play a different role in Beethoven than they do in Mozart. To compare a Beethoven Symphony with a Mozart Symphony (score only) we do not see that difference that we hear when the scores are performed. But the sounds that we do hear that are not necessarily differentiated by the scores indicate with immediacy whether we are listening to Beethoven or Mozart.

I might add, too, that you have an inherent problem in the score of editorial interpretations that never met approval by the composer.

I'm not quite sure about your point Sorrano - are you saying that we can only identify the composer fingerprint when we have a performance? I'd say that was true for the average music lover who may not be able to hear a score mentally, but not for a musicologist who should be able to identify a composer from score alone. I disagree with your remarks about dynamics, there is a very clear difference in a Beethoven score to a Mozart score where often the dynamics are actually editorial. This of course becomes even more apparent with late Beethoven where he is meticulous about his markings - where in Mozart would you find 2 pages that contain over 50 instructions to the performer as you do in Op.110? Just by seeing that alone we would know it wasn't Mozart without the need for a performance.

[Euan Mackinnon]

Sorrano

I have been following the discussion you have been having with Peter on fingerprinting and interpretation with great interest. However, as someone who has minimal technical knowledge of music, I am finding some difficult in seeing the distinction you are drawing.

Would an analogy help? We stain biological specimens in order to bring out a certain characteristic more clearly. This doesn’t mean that the said characteristic is only present when the staining has taken place; it is there all the time and intrinsic to the specimen concerned.

By analogy, are you trying to say that you agree with Peter (and my own instinct) that the fingerprint is entirely embedded in the score (and therefore does not need a performance) but that this or that performance can heighten (or diminish) our perception of it?

Euan

[Euan Mackinnon]

I am suggesting that we have reached the point where there are four elements left to consider - (i) Intervals, (ii) Harmony/Instrumentation, (iii) Melody/Note progression, and (iv) Rhythm.

Of course there are some dissenting opinions notably Dynamics and Register which Peter in particular feels should be included in the model and Performance which Sorrano and Philip have raised.

For my part, I instinctively believe that the four elements above – all seen in the score - contain all that we need for a potential model of fingerprinting. In short, they are necessary and sufficient.

Before trying to build our fingerprinting model (or, more accurately, to lay out some guidelines for it), I have two major questions to raise.

The first question concerns the Harmony/Instrumentation element (I will raise the second question in a subsequent post).

(Note: for simplicity I will stick to orchestral music but I assume the question is more general than that.)

I have read that Mozart (and, presumably, other composers – Handel and the Messiah, for example? Brahms? see Roehre, 25th April, Brahms and Beethoven thread) composed by initially writing out the ‘top’ and ‘bottom’ lines (presumably melody + bass). Then – and this is the important point that leads to my question – the rest (the orchestration?) followed more or less automatically so much so that it could often be ‘sub-contracted out’ to ‘professional’ orchestrators.

First, is this true?

And, second, if it is true, can we limit the Harmony/Instrumentation part of our fingerprinting model to just the ‘top’ and ‘bottom’ lines – typically 1st violins + cellos/basses - thereby greatly simplifying that element in the model?

I should add that my strong instinct is that one cannot exclude inner parts from the model. Indeed, I would go further and guess that a significant slice of the fingerprinting model will be concerned with this but I would welcome the response of those far, far better qualified than I am to address such matters.

Euan

[Peter]

I think you're talking about the initial sketching process Euan which I'm not sure is really relevant to our question - composers worked in different ways yes, but we're trying to establish a fingerprint from the final product? Baroque composers would often simply write out a bass line with numbers beneath known as figured bass which would indicate the intended harmony. Brahms would write out an orchestral piece initially for two pianos on four staves and then to orchestrate. When it comes to orchestration itself, no you can't limit it in the way you suggest - a composer such as Schumann for example was criticised for his very thick sound and it is interesting to compare his original versions, not only with his own revisions but with Mahler's retouching of the orchestration to see just how different the same material can be made to sound in different hands.

[Sorrano]

I think that methodology has a lot to do with the fingerprinting in the final analysis. Understanding of a composer's compositional technique helps understanding of what is in the music itself.

[Sorrano]

Sorrano



By analogy, are you trying to say that you agree with Peter (and my own instinct) that the fingerprint is entirely embedded in the score (and therefore does not need a performance) but that this or that performance can heighten (or diminish) our perception of it?

Euan

That is a good way to put that.

[Euan Mackinnon]

Peter, Sorrano

Thank you for your recent responses.

I am away on business for the next week but will post my remaining comments on musical fingerprinting when I return

Euan

[Euan Mackinnon]

Before proceeding to suggest guidelines for a model of musical fingerprinting I want to ask the second of my two questions.

How would you define Rhythm (in music)? In short, what fundamentally is rhythm?

It would seem self-evident that Rhythm is essential in any such model, but is it? Fingerprints are captured by patterns and surely rhythm is above all else a pattern and, as such, must be in the model.

I began to have doubts when I searched for definitions. I quote extracts from what appeared the most detailed of these below (I apologise for the resulting length of the post).

None of these extracts seems to me to pin down exactly what rhythm is. And all three state that ideas about rhythm have changed recently.

I would welcome your views and, even better, your definitions (or references to definitions).

Thank you.

Euan


From http://www.dolmetsch.com/defsr2.htm

[A]ny aspect of music having to do with time, which, since music must exist in time, means that all music is rhythmic. At its simplest, rhythm may be thought of as the disposition of strong (or accented) and weak (or unaccented) beats in a piece of music. This has led some analysts to see 'metrical organization' and 'rhythm' as being identical. If periodic stress defines musical groups, then a primary definition of 'meter' includes the opposition of weak and strong. This opposition is shown most clearly in the contrast of arsis and thesis across the barline, a model which can be expanded, as it was by Kirnberger, Hauptmann, and Weber in the 18th- and 19th-centuries, to include musical form as a hierarchical extension of the metrical impulse

[F]or contemporary analysts, musical rhythm and form receive a more sophisticated treatment. Most recent literature ties higher-level rhythm and meter to pitch structure, using methods founded on the historical association between certain harmonic and melodic progressions and their corresponding metric positions or stress. Musical rhythm is seen as a product of goal-directed melodic and harmonic activity. Schenkerian notions of prolongation and interruption inform their analyses, and are used to locate tonic accents and "structural downbeats" at important junctures in pitch structure. These voice-leading events assume formal significance when they are correlated with surface activity to direct the rhythmic flow of a work

From http://en.wikipedia.org/wiki/Music_theory#Rhythm

Rhythm is the arrangement of sounds and silences in time. Meter animates time in regular pulse groupings, called measures or bars. The time signature or meter signature specifies how many beats are in a measure, and which value of written note is counted and felt as a single beat. Through increased stress and attack (and subtle variations in duration), particular tones may be accented. There are conventions in most musical traditions for a regular and hierarchical accentuation of beats to reinforce the meter. Syncopated rhythms are rhythms that accent unexpected parts of the beat. Playing simultaneous rhythms in more than one time signature is called polymeter. […]

In recent years, rhythm and meter have become an important area of research among music scholars.

From http://www.britannica.com/EBchecked/topic/501914/rhythm

[R]hythm, in music, the placement of sounds in time. In its most general sense rhythm […] is an ordered alternation of contrasting elements. […]

Attempts to define rhythm in music have produced much disagreement, partly because rhythm has often been identified with one or more of its constituent, but not wholly separate, elements, such as accent, metre, and tempo. As in the closely related subjects of verse and metre, opinions differ widely, at least among poets and linguists, on the nature and movement of rhythm. Theories requiring “periodicity” as the sine qua non of rhythm are opposed by theories that include in it even nonrecurrent configurations of movement, as in prose or plainchant. […]

Unlike a painting or a piece of sculpture, which are compositions in space, a musical work is a composition dependent upon time. Rhythm is music’s pattern in time. Whatever other elements a given piece of music may have (e.g., patterns in pitch or timbre), rhythm is the one indispensable element of all music. Rhythm can exist without melody, as in the drumbeats of primitive music, but melody cannot exist without rhythm. In music that has both harmony and melody, the rhythmic structure cannot be separated from them. […]

The article then goes on to discuss Beat, Tempo and Rubato before turning to Time

The mind apparently seeks some organizing principle in the perception of music, and if a grouping of sounds is not objectively present it imposes one of its own. Experiments show that the mind instinctively groups regular and identical sounds into twos and threes, stressing every second or third beat, and thus creates from an otherwise monotonous series a succession of strong and weak beats.

There follows a series of sections that go deeper especially with respect to historic periods. Included in these sections are the following passages:

[…] Thus far, music’s structure in time has been examined separately from its structure in tone, but no such separation is really possible. Melody and rhythm are intimately connected.

[…]

In music employing harmony, the rhythmic structure is inseparable from harmonic considerations. The time pattern controlling the change of harmonies is called harmonic rhythm. In 17th- and 18th-century music, harmony tends to limit rhythmic subtleties and flexibility of the melodic elements (as well as determining the basic type of melody) in regard to stress accents

[…]

In European music the great variety of styles derives its relation to melody from different concepts of rhythm. They include the strict rhythmic modes of the 13th century; the free oratorical speech-rhythms of the Renaissance; the almost stressless flow of Renaissance polyphony; the strong body rhythms of the Baroque; the freedom of the late Romantics; and the primitivistic rhythms of the 20th century with composite and ever changing time signatures.

Thus, study of musical history shows a varying attitude toward rhythm, sometimes closer to strict rule, sometimes to “freedom,” as the temper of the times and the relative influence of poetry, dance, and decree.

[Peter]

I would simply define rhythm as an organised pattern of sounds including dynamics, accents and rubato.

[Euan Mackinnon]

I have an apology to make to anyone who is still actively following this thread.

By now I had hoped to have posted the final messages summarising my suggested model for musical fingerprinting. This dealt with what I feel are the only two essential elements in such a model: Harmony/Instrumentation and Intervals.

However a proper explanation required Images of various kinds (diagrams, graphs, facsimiles, etc) inserted in appropriate places in the text.

To my surprise, and after some frustrating days trying various things, I have found that the subset of HTML used on this forum allows a very limited and restricted range of commands for inserting Images – two to be precise and even these do not have all the usual options.

These restrictions (on such things as Sizing, Placement, and so on) meant I could not use Images as I wished and this effectively rendered the messages I had in mind impossible. The restrictions were confirmed by someone on the forum who is far more knowledgeable about such things than I am.

In retrospect I should have checked this before I started the thread and so I repeat my apology. If I initiate a new thread in the future I shall be more careful.

It just remains for me to thank those who contributed to the thread, especially Peter who hung in there until the end.

Euan

[Sorrano]

The overall scope of this thread has been pretty daunting to time restricted contributors. Perhaps it would be well to break down the various subsets and work with those one at a time; as for myself I have been thinking a lot along the lines of rhythmic similarities among certain composers. Bruckner, for example employed the 2 + 3 rhythm fairly frequently, as well as repeated dotted eighth followed by a sixteenth during critical moments of his works.

There is much that can be discussed per each criteria, and with a lot of patience we can delve through this.