Arrangement and Transcription

A commentor asked a good question about my last post:

In the sense you (both) have used the term here, what exactly is an arrangement?

To which I replied:

Thanks so much for this comment! I try not to use more technical terms than I need to, but sometimes I use one without even realizing it!

An 'arrangement' of a piece of music consists in altering it in some way so it can be played with, for example, different instruments. Given the day-to-day exigencies of practical musicians, this is very common. You go to a wedding and a harpist is playing the well-known "Wedding March". This piece is originally for orchestra--one of the movements from the suite of incidental music for A Midsummer Night's Dream by Mendelssohn. The harpist, therefore is playing a much-reduced arrangement of the original. Typically you retain the most recognizable features such as the melody, the rhythm and the basic harmony. 

Tognetti's arrangement of the Beethoven violin sonata is a different kind: instead of reducing and abbreviating the original, he is trying to create a new artwork by, well, I used the word 'unpacking' and that captures it fairly well. He makes explicit lines and harmonies that were implied in the original.

There is actually quite a lot to this question, so I am going to devote a post to it. Previously I put up a short post about transcribing Beethoven. I also got an interesting comment on that post (from the same commentor). I think I understand the interest so let me take up the question in a slightly more philosophical manner.

Musicians use two different terms: arrangement and transcription. Both refer to the practice of taking music written for one instrument or group of instruments and adapting it for a different instrument or group of instruments. The distinction between the two terms is that a transcription attempts to be an accurate reproduction of a piece of music in a different medium. A Scarlatti sonata for harpsichord may be transcribed for guitar and reproduce very closely the original. Ethnomusicologists may listen to music in an oral tradition and transcribe it into notation, trying to capture as accurately as possible exactly what they heard.

An arrangement on the other hand, implies some changes, perhaps significant ones, from the original. If you play one of the Bach cello suites on guitar, it is scarcely even a transcription because you don't have to change a single note. But if you take a movement from one of the orchestral suites, say the Air from the Suite No. 3 in D major, BWV 1068, and perform it on solo guitar, then radical note-ectomies will need to be performed! Here, have a listen to the original and an arrangement for guitar:

This is quite a good arrangement and it is well played, but I think you can hear that it is very different from the original. What we often hear are poor arrangements indifferently played and either done because of the needs of a particular engagement, such as a wedding, or to pander to the audience. For these reasons, arrangements are often looked down on in the classical world. This seems to be changing, however, as my post on Richard Tognetti indicates. You can certainly approach the whole question of arrangement from a creative position and not just from the needs of the day.

Some of this comes from the influence of pop music where each performer is expected to come up with an entirely new 'take' on every song from whatever anyone else has done with it. In the pop music world, when an artist records a song by someone else, this is known as a 'cover' version. It is, of course, a new arrangement, but where traditionally, as in the Bach example above, the attempt is made to capture the essence of the original by retaining things like tempo, harmony, note values, in a 'cover' version, the artist attempts to alter as much as possible to make the tune a vehicle for their own style. Here are a couple of examples. First, Jose Feliciano's instrumental version of "Yesterday": possibly the most-covered song of all time. The video does not seem to match the audio at all in this clip, so just listen:

And here is Paul McCartney's original:

Or to take some more extreme examples, here is the aria "Summertime" written by George Gershwin from the opera Porgy and Bess:

And here is an arrangement for solo piano by Keith Jarrett:

And finally, from the wild and wonderful 60s, Janis Joplin with Big Brother and the Holding Company:

If you have a philosophical disposition you might start squirming around right now because you want to say, but they are so different, which one is the song "Summertime"? Is "Summertime" a song (or aria), or is it a bunch of songs? Where is the actual song located? In Gershwin's mind alone when he was writing it down? In the notation on paper? Is the Janis Joplin arrangement as much the song as the original? Is it less so or more so? These questions of the identity or ontology of a piece of music are kind of fascinating. One of these days I am going to put up my long-postponed post (heh) on Plato and the Forms, where I hope to dig into these questions a bit more.

But for most musicians it really doesn't matter. Remember, musical thinking takes place in an entirely different part of the brain so we musicians tend to have intuitive or instinctual reactions to arrangements and transcriptions. Of course I don't mean intuitive or instinctual in the usual sense, I just mean "non-verbal musical reactions". A lot of the 'argument' that I present on this blog is, with the aid of YouTube clips, conducted in musical terms. I say, listen to this, listen to that, hear the difference and that is the musical equivalent of making points in a verbal argument. I have often heard a musician talk about music in a way that mixes verbal comment with examples played on a handy instrument. "Well, you could play the phrase this way

but it flows much better this way

Our musical reactions tend to be enigmatic unless you are following the musical argument. So you might argue for or against a particular arrangement by playing little excerpts and saying, "that doesn't really work, does it?" or "isn't that a nice effect?"

I've managed to get off the topic, but to return, it is simply the case that one has to judge the effectiveness of an arrangement or a transcription based on musical aesthetics. Some work better than others. It is always a good practice to check with the original version. You may or may not decide that the original version is the best.