The Emancipation of Note Releases

One of the most challenging things to notate for orchestral musicians without using additional verbal instruction is a long isolated note that doesn't decay in volume/presence over its length. Write a whole note with a forte and you get a forte with a decrescendo. For every orchestral musician, this is the absolute default. 

To put it more broadly, every note that is being produced in an orchestral context is being played like an attack with a release tail that is selectively longer or shorter but it always decays towards its end. The only way to prevent that from happening is to add another long note right after but as soon as you reach the last "long" note in a sequence of long notes it will decay. Of course, also employing hairpins will change that but there is no common symbol for "don't decrescendo".

Interestingly, our entire western notation system is conceptualized as very focussed on note attacks. It is very effective to notate when exactly and how a note should be attacked but it is incredibly cumbersome to notate what happens after that especially when you want it to run against the common performance practice and it is even almost impossible to notate any specific wishes for a release.

In orchestral context, this whole "how long should a note be" and "how should its release sound" is engrained in a common understanding of a performance practice that gets applied everywhere. You get decrescendos where you don't write any, you get shorter notes where you write them long etc. All this has become the default of interpretations in centuries of musical performance.

In case you are still wondering what I'm talking about, let me demonstrate a few things:

I'm pretty sure, everybody knows that one. In case you don't here's the recording for that specific excerpt. Give yourself some time and listen to it several times and compare what is played against what is notated (by the way, trumpets are transposing here so they sound a major second lower).

If we try to replicate exactly what the trumpets are playing, it looks something like this:

Particularly interesting is the half note in the second bar which gets shortened considerably (also in the trombones/tuba if you listen closely) and all the quarter notes we see with a tenuto mark don't even get anywhere close to their full note length in the interpretation. This extreme case is by the way a specific brass phenomenon which is not as extreme in other sections (or even other music) but the general interpretation of such a fanfaric segment is to considerably shorten the notes. The decrescendo on the long note is a universal phenomenon though.

Of course, this way of interpretation is absolutely justified and benefits the music greatly. If the players held all notes at full length the melodic gesture would become blurry as the reverb tail of the last note would ring into the next note. In a way this interpretation factors in the room reverb. The real issue with these things arise when you want to consciously go against that.

One of the most common issues  when recording or rehearsing with an orchestra is to get rid of "messy releases". A messy release happens when several instruments/sections hold a long note (for instance a chord) that is followed by a rest (or is the final chord of a piece) and players decide (analog to the example above) to shorten that long note but each to a slightly different actual ending point which will create a very unstructured and in worst case weirdly balanced end of that chord/sound.

Notational conventions have come up with very inelegant ways to counteract something like this and in his very comprehensive book "Textures and Timbres", Henry Brant dedicates an entire (1 page) chapter to this issue:

This whole page is a testimony of that release dilemma that we're dealing with. And you can see these things everywhere, for instance here in the published Star Wars Main theme right on Page 2:

When writing for orchestra, I usually notate long notes similarly. However, I have to say that in recent years I noticed a stronger discipline regarding note lengths with session players. Just last summer, I was orchestrating for a feature film where I had written several whole notes tied over to an eighth note in the score and the players at Synchron Stage Vienna actually played the additional eighth note length which needed us to tell them to shorten them all to the barline.

So you see, this whole releases issue seems to be a can of worms.

To make matters worse, this playing default is not universal and depending on the players and the genre of music, you might be getting radically different results.

The other extreme of that spectrum are Jazz players, particularly the ones that play in large Jazz ensembles like Big Bands. This whole genre puts incredibly more focus on note releases than in the "classical" world. The general understanding in this genre is, that a note release is a similarly important rhythmical event as its attack. As the rhythmical component in this genre is considerably more in focus than with orchestral music, there is a very different attitude towards note releases. In fact, in order to make some passages groove properly, the releases have to support that rhythmical idea.

Getting back to our Jurassic Park Trumpets, let me just quickly elaborate how notes are articulated in the different "worlds".

Brass players in general use their tongue to control the air stream through the lips into the instrument. The hardness of the attack of the note is controlled by either making a soft "duh" with the tongue or a hard "tuh". The end of the note is usually being executed by terminating the air stream from the lungs which automatically "fades out" the note more or less and gives a relatively soft release.

In the Jazz world, notes are way more often articulated with tongue movements like "dud" or "tuht" ending the air stream abruptly with the tongue which will create a noticeable end of the note. This playing style supports the more rhythmic gesture of the music (additionally to other stylistic differences in the playing). Also, you would way more regularly get long notes held at full volume to their exact end in this genre so the tied over eight note approach as mentioned above is not necessary in these cases.

Now with session musicians, you very often get players who work and play in both worlds and can similarly adjust to both styles. However, as you can imagine, with media music being very flexible in its genres and some players having experience in both worlds and others who don't (e.g. Woodwinds and Strings), some things will get lost in (notational) translation and interpretation. So even when taking a lot of care in the way you notate things, there are always chances that the players might misinterpret and need to be verbally guided towards what is wanted. 

Now while all these things are mainly relevant when working with real musicians, they also of course have a huge impact on music that is produced with samples. Knowing about the performance defaults of course will also increase the realism of your sample productions. Knowing that for instance you can not mock-up the Jurassic Park theme with just a legato patch but need different variants of short and long notes to get a relatively convincing result is tremendously helpful in executing realistic mockups.

Fortunately the strict separation between "classical" and "popular" music is slowly fading away also in the education of musicians and we very often get session players who feel similarly at home in both worlds. You might still get the defaults of the individual genres in the interpretation but your players will be much quicker in adapting when you say something like "Please hold all long notes without decrescendo to their exact end point".

It will be interesting to see how this develops and whether notation and playing defaults will develop into a more "universal" playing default or whether these two opposite worlds of interpretation will remain and we will struggle for years to come guessing whether our players will interpret what we write with one or the other hat on.

Coming back to my initial statement how it is incredibly cumbersome to notate release points of specific notes in our western notational system which really feels like it's not designed for it, you just need explore a few contemporary works for harp and the notational abominations that need to be used to communicate the dampening of specific notes or ranges clearly.

Here are a few examples taken from instructional literature:

All notes that have that circled cross attached to their stem in the first example are to be dampened at this right moment. The lower example shows two possible ways to dampen a specific range (also usable in glissandi). Besides these things, it is practically impossible to clearly notate "hard releases" on any instrument that is capable of sustaining notes without using a verbal instruction. So we're stuck with guesswork for now and ideally know the players that we're working with well enough to predict how they are going to interpret what we have written.