What is classical?

Tags: #<Tag:0x00007f4d55466fa0>

I’ve managed 9 months with MusicBrainz before addressing this thorny issue…
Is this classical or not: https://musicbrainz.org/release/cbb883b5-dd08-4953-84b1-bfe7ed60e1ca ?

Should I care? Why should MusicBrainz?
The current entry is completed as if it were not a classical album. Undoubtedly it is cross-over. The Hilliard Ensemble are a classical outfit, but Jan Garbarek is jazz. The consequence of this treatment is that the original source of the music is sometimes lost and rather inconsistent. Most (but not all) of the composers are anonymous/traditional and detailed information is given in the CD booklet. Some of these have been given by associated “works”, but not all - e.g, track 1-4 is attributed to “16thC Scotland” in the CD booklet and track 1 to “Anonymous (Peru)”, but no mention on the MB listing. Track 6 lists a “composer” of “St Albans of Great Dunmow”. Clearly no such person existed (they are just two places 40 miles apart - the booklet lists the source as “St Albans / Great Dunmow” without further elaboration). Track 2-8 has the source as “Iroquois & Padleirmiut fragments” - the MB entry includes this in the recording title, similarly track 2-7 “16th Century Russia”.
So the treatment is inconsistent, but I do not blame the editor as it is not at all clear how to enter a release such as this when following the non-classical style. Ideally the “source” of the work would be clear, whether it is composed or traditional/anonymous and, if the latter, the place of origin (if known).
If the entry had been made in the “classical” style, then the “source” would be in the Artist entry, with an “as-credited” entry to reflect the release notes - e.g. “[traditional]” credited as “16th century Scotland”. This would also obviate the need for introducing rather artificial “works” and “composers”. It seems to me that the classical style would follow the intention of the release better. However the artificial classical/non-classical style divide seems to have led the editor the other way, resulting in a bit of a mess.
I’d appreciate some comments before messing with it!

1 Like

These are good questions.
There is no great answer that I’ve seen.
The better so far is "Is it in the classical tradition?"
The Penguin Test Score (still in development) is not universally recognised as being definitive or even faintly indicative. :grinning:
If you want to read more old discussion here is one place.

But a new start at the mountain seems fitting.

Wow, that is one long thread, with quite a lot of different points of view. My approach to these issues (as indeed most issues) is pragmatic.
The issue I faced with the release I quoted is that there is no good way of including “credited as” information for composers when a non-classical style is used (i.e. composer not in track artist). There is no composer type for artist-recording relationship, so the only place to set the “credited as” data is in the artist-work relationship. However, that relationship will be common to any release containing the work and different releases might credit the composer differently. I also question whether it is appropriate to create a work entity purely to enable the entry of a composer relationship, especially where the “work” is an interpretation of a piece of traditional music.
So it seems to me that the only good way of documenting this release is to use the classical style, which arguably fits with @reosarevok 's inclusion of

as being classical.
In that thread it was also said that

but little evidence was given for that.

Arguably, we could use the track artist for both the main performing artist and the composer (if different). This style could be used for both classical and non-classical recordings. The role of each should be clear from the various relationships. Both can then have a “credited as” entry which reflects what is on the CD booklet or whatever. I admit I haven’t fully investigated the implications of this, but it seems to me that it would not invalidate existing data, just allow greater flexibility. Comments on this proposal?

Amazon wasn’t letting me play every song, and I only got 30 second samples of the ones that did…

It doesn’t song like typical classical music. It seemed like something Yanni or John Tesh would do.
How are those two artists classified?

No25 in the Classical charts 1999 http://www.officialcharts.com/charts/classical-compilation-albums-chart/19991010/148/
But my point is that “classical” is a purely arbitrary distinction and that the style should enable the release to be catalogued properly. My suggestion to put both (main) performing artist and composer as track artist may sound like heresy, but what is wrong with it?

I happen to own this CD, and I agree it is difficult to classify. I’d say that some of the tracks belong to the ‘classical’ performance tradition in the sense that they are interpretations of existing composed pieces, while others are more like one-off ‘performances’, perhaps seeded from some existing piece or fragment but transforming it into something very different. John Potter’s liner notes are worth quoting here:

“[…] Mnemosyne really contains two sorts of music. The most straightforward are those pieces where we sing existing music (conventionally notated) and the saxophone improvises around us. We may reorder the music a bit but we know more or less what we’re going to do (we never know what the saxophone is going to do …). […] A lot of the newer repertoire on Mnemosyne consists of very small amounts of material with minimal notation. These are rarely complete pieces and often just a few scraps, recovered from old book bindings or buried for centuries under desert sands. We may decide on an outline form and share out the material, then we all improvise and none of us knows what will happen next. […] We did it for each other, in the absence of an audience, and these are complete, one-off performances, which will never sound the same again.”

I don’t know if that helps at all …

Hmm… yes and no. I must confess to having the CD also. What I am trying to achieve is to provide the credits as written at the end of the liner notes. For each track, there is the title, the composer/source, and then (sometimes) “Hilliard Ensemble, Jan Garbarek”. My assumption is that when the last of those three is not included, that is because it is

but otherwise there is a fair amount of improvisation from The Hilliard Ensemble as well as Garbarek.

My inclination is to include them all as track artists, as written in the notes, but with “[traditional]” or “[unknown]” as appropriate with (for example) “16th century Scotland” in the “credited as”.

Since there’s been no objections, I’ll make the changes as indicated and see what happens…

1 Like

I would say that if the “composers” did not merit credit on the back cover, they are not track artists (and by default the track artist should be the album artist).

I suggest adding origin information using a composed in relationship or, for something like “Iroquois & Padleirmiut fragments”, extra title information.

…but many classical compilation albums do not list all the composers on the back cover, while the guidelines say to enter the composer as track artist.

That could be useful in some circs