What is a song?

[Bell Green]

AABA, verse/chorus, verse/bridge/chorus, AAA, groove track, rap song, and on and on and on.

What is a song? How do you define a "song". Who wrote the first song? What is the evolution of the song.

I know, there's probably a thread somewhere about this, so if you know, then please tell me and I'll delete my post. I just thought it would be interesting to hear what you all thought a song was and your own personal definition, not a textbook definition.

[j$]

Self-expression, with music.

Probably a bit vaguer than you were hoping for!

[Jim of Seattle]

Well..........

There is no definition. It's what you decide. Some people say there is one, but they will never agree.

That said, I'm going to take a stab at it. People, chill in advance. I don't stand wholeheartedly behind this. It's a shot in the dark.

The primary melodic instrument should be the human voice.
It's usually relatively short. It's hard to justify something as a song that's over about 8 minutes or so.
It has to have a fairly simple structure. This ties into the length requirement. Something that's 15 minutes long probably has some significant musical development which make its scope larger than that of a song.
It has to consist of a single movement
There must be some purely musical aspect to it, as in, it can't simply be a recitation of text. If there is no accompaniment to the singer it has to be sung. If there is accompaniment, it can be spoken, but ideally should be spoken in rhythm.

The first songs, of course, are from the classical tradition, and were written long before what we have history of. Early music was divided between secular and sacred music. For a long time sacred music was the only thing being written down, so its all we have much record of. And it was not "songs" at all. I suppose secular madrigals from the Renaissance could be considered the first "songs", since they followed all the rules I listed above, although certainly Renaissance secular music followed in long traditions of songs prior to that period.

Perhaps the best way to define a song is to decide what something that isn't a song really is. First things that come to mind would be a "piece" or a "poem". These are two things mistakenly called "songs". The Blue Danube is a piece, cuz it isn't sung. Carmina Burana is a piece, but could be argued that it consists of a bunch of songs, since many of the individual movements would qualify. I've recently written a single "song" that's over half an hour long, but otherwise would qualify. So it's really a piece.

There's my shot at it.

[blue]

song Audio pronunciation of "song" ( P ) Pronunciation Key (sông, sng)
n.


1. A brief composition written or adapted for singing.
2. The act or art of singing: broke into song.

2. A distinctive or characteristic sound made by an animal, such as a bird or an insect.
3.
1. Poetry; verse.
2. A lyric poem or ballad.

[jute gyte]

It's usually relatively short. It's hard to justify something as a song that's over about 8 minutes or so.
It has to have a fairly simple structure. This ties into the length requirement. Something that's 15 minutes long probably has some significant musical development which make its scope larger than that of a song.

in an exception to these mostly agreeable rule, there are groups like khanate whose songs are sometimes over fifteen minutes simply because they play at such extremely slow tempos. (i'm chilling in advance and all, but it seems worthy of notice.)

[HeuristicsInc]

khanate reminds me of old swans with a different vocalist.
-bill

[Bell Green]

interesting, very interesting. keep it coming. I dug this up with Google - makes quite interesting reading
http://en.wikipedia.org/wiki/Song

But what do I think? Well, there are three basic building blocks to music for me, and those are rhythm, melody(pitch) and tempo. But what came first - someone banging a drum or someone blowing a digeridoo? Don't know, have to do more research. But song refers to something vocal, melodic and I guess in the beginning they were songs of praise. But praising what?

Uh, sorry. I'm dozing. Have to pick this up tomorrow.

[jute gyte]

khanate reminds me of old swans with a different vocalist.
-bill

they remind me more of grief, but the swans comparison is apt as well.

[Leaf]

interesting, very interesting. keep it coming. I dug this up with Google - makes quite interesting reading
http://en.wikipedia.org/wiki/Song

But what do I think? Well, there are three basic building blocks to music for me, and those are rhythm, melody(pitch) and tempo. But what came first - someone banging a drum or someone blowing a digeridoo? Don't know, have to do more research. But song refers to something vocal, melodic and I guess in the beginning they were songs of praise. But praising what?

Uh, sorry. I'm dozing. Have to pick this up tomorrow.

I always thought of the three elements as being (in order of importance) rhythm, melody, harmony.

Tempo is a good addition to that, however, I view it as an element of rhythm.


A song to me has vocals. Everything else about it supports the vocals.

That's why when I write a musical meandering piece that has vocals placed on top off it , Blue gets on the wire and says "where's the song?"...


Although, I really don't give a shit about the definitions of "song" or "tune" or "intrumental" or "sonata" or "piece" or "movement" or "symphony" or whatever in terms of being some kind of limitation. That's other people's perogative. Spelt properly though.
If your question is geared towards self-definition of your own material... that's up to you. If it's in response to critizimazization... get a spine (kidding Mr. green.... kidding) and if it's simple curiousity... get a dictionary!!! (then scan it and email me the link... I could use one too!!!!)

[jb]

We've been through this before.

Labels, in this particular case, are a tool for communication. Often people get upset when you challenge their label of something that isn't a "song". But if I call a cat a dog, how does that help you know what you're dealing with?

A "piece" can be any reproducible musical item. Or a "work". You usually reserve the term "a work" for something that's on a larger scale, such as a symphony or concerto, or an opera (which contains specific types of songs). Beneath the label "piece" you should put all of the instrumental types of music, such as sonatas, etudes, concerti, etc.

Knowing this sort of thing will help you and I to have a discussion about music. You don't necessarily have to be totally sure of yourself if, say, you don't know what a thing is exactly ("hmmm, I'm not sure if it's an etude or what.") But by expressing that doubt, we'll still both be on more or less the same page.

The urge to label everything under the musical sun a "song" strikes me as obnoxious intellectual laziness, and I tend to lose respect for people who insist upon doing so even after they should know better.

Just because "instrumentals" are allowed on SongFight doesn't make them "songs".

[j$]

You know, JB, i agree with you and disagree with you on this at the same time. Let me explain ...

You are absolutely right on definitions. A song is not a tune. I don't know what a nocturne is exactly, but if I heard a piece of music that someone had composed within a certain set of 'rules', or whatever you want to call them, I would want to appreciate the craftmanship as well as the art.

My background is literature, so forgive my imperitence if I compare it to writing strict forms of poetry. To take a group of words and applying a set of criteria to it, pummelling them into a certain shape, is a great skill, and involves no small amount of artistic endeavour as well. It desrves to be recognised.

But, to continue the metaphor, two things come to mind. One, which you recognise, is that just because you know it's a sonnet, or whatever, doesn't mean that your intended audience will be able to appreciate that skill, unless they too are skilled in the same form. I personally don't think one should expect it. Which leads to the second point.

If your audience doesn't know a certain form, then the content (what a lot of people mean when they say 'song' I guess) must be strong enough to interest them on a different level. In poetry, i have always believed that those considered the 'best' managed to make their strict forms sound relaxed and everyday. The depth was there as a next level to discover. In other words, as I once rather badly expressed in a poem about Tennyson, 'A problem, as he wrote, arose / the strongest stanzas read like prose.' Hey, I was only 17, so give me a break

I guess the point I'm trying to make is an inherent paradox in the composition process. If you write an incredibly complex verse form, or peculiar time signature or whatever, then people are either going to get sucked into its internal rhythm and not notice the effort put in; or they're not going to be able to get into it at all, and feel like it's studied.

A bit of a lateral shunt from your topic, I appreciate, but just wanted to take the conversation on...

j$

[jb]

You know, JB, i agree with you and disagree with you on this at the same time.

I'm sorry, I couldn't follow that, so I'm not sure where you're disagreeing with me.

I think the only time the label matters is when something is under discussion, or when you're trying to gather a set of similar things together. And in the latter case there's always going to be gray areas where you have to make a decision as to whether a particular item qualifies for inclusion in the collection.

In the case of a sonnet, it only matters that it's a sonnet when that fact becomes relevant to a conversation. As an inherent fact of "art" or "entertainment" it doesn't matter at all what you call something.

But if I want to read a collection of sonnets, I don't want to find haiku in there. And if I'm reading a collection of "traditional japanese haiku" I don't want to find any of Kerouac's "haiku" in there. (I adore Kerouac's "haiku" btw.)

[Jim of Seattle]

J$, you prompted a thought.

Part of knowing the different forms of something is that it adds a level of appreciation that is lost if you don't. For example, some people have pointed out that Radiohead's Paranoid Android is in sonata-allegro form. I don't know whether it is or isn't, because I've never listened to it with that question in mind. But if you know what sonata-allegro form even is, and can recognize it when you hear it, then discovering that this Radiohead song adheres to it presents you with another level of appreciation.

If I went to see a Chinese opera, I'd come away thinking, "Huh, that was weird. So that's what Chinese opera is like.", while my companion, well-versed in Chinese opera, could say "That was an extremely unusual Chinese opera because the second half of the third stanza of the opening of the second act wasn't a duet." And I could say "Oh, so that's why the whole audience gasped in horror at that moment?" "Yes. It's very non-traditional, and almost heresy." So there's a whole new level of communication available to the artist and audience merely through this sort of dance that can be choreographed between a given piece and the expectations based on the piece's assumed form.

[erik]

Labels, in this particular case, are a tool for communication. Often people get upset when you challenge their label of something that isn't a "song".

Okay, I'm probably going to screw this up, because of my lack of musical education.

There are two rather different definitions for the word "song". One definition, can almost certainly be agreed upon by people who are educated in musical concepts: a melodic vocal line, accompanied or not by music. The other definition is the one used by the population at large, to mean "a short piece of music with words, or with no words, but with a strong melodic line". The general populace may be adhering to a definition that is wrong, but it's a commonly held definition.

The problem is not one of who is right and who is wrong, but rather, how do you carry on meaningful discussion with that knowledge that both definitions exist?

I've come across the same problem in teaching mathematics: the words "some" and "most" have very different connotations in a mathematical context than they do in a general populace context, especially if that population is exclusively young kids. "Some", generally speaking, does not mean "any positive number". If you said "I have some money saved up" and it turned out you had saved up a penny, most people would feel you had been dishonest. Or if you said "Some of the songs in this week's fight are really good" and then it turned out you meant one. The definition for "some" used in common language is something like "a small percentage, considerably less than 50% of the total, but considerably large enough to not name the number by name". Same thing with the word "most", except that one gets even trickier.

To me, if you really want to have a discussion where the meaning of a commonly misunderstood word is important to the discussion, then it's best to flat out state what the word means so no one can try and apply their own nontechnical definition to the mix.

[Eric Y.]

ok at risk of making this discussion even more off-topic than it was...

i'm pretty sure "some" and "most" don't mean the same thing. "most" typically refers to more than half -- i.e. if most people behave a certain way, that would mean more do than don't.

[jb]

Labels, in this particular case, are a tool for communication. Often people get upset when you challenge their label of something that isn't a "song".

There are two rather different definitions for the word "song". One definition, can almost certainly be agreed upon by people who are educated in musical concepts: a melodic vocal line, accompanied or not by music. The other definition is the one used by the population at large, to mean "a short piece of music with words, or with no words, but with a strong melodic line". The general populace may be adhering to a definition that is wrong, but it's a commonly held definition.

I could see that, if it were true, but lots of the population at large say "song" when they mean "Beethoven's 5th Symphony".

"Hey dude! Yesterday in class we listened to this awesome song by Beethoven!"

*thinking to myself* "Hmmm, probably something from Fidelio (Beethoven's only opera)."

"You know, it's that famous one! You know, DADADADAAAAAAAAA! That one!"

"Ohhhhh you mean you listened to a movement from that symphony. That's not a song, dude."

I can accept generalizations of "techno song" or "rap song" and not get upset. I mean, hell, *none* of it gets me *upset*. But if we're trying to talk about stuff, you gotta agree on terminology before you can actually communicate ideas effectively. (Even cavemen had to agree that two grunts meant "club" and one meant "sharpened rock".)

When a term's definition becomes too broad, it loses its value. In the case of "song", if you let the definition become too broad, it starts to just mean "music", and we already have a word for that.

[HeuristicsInc]

i think the general populance thinks "song" == "some chunk of music that (would take up a track on a cd or would be played on the radio)"
classical and jazz are the only instrumentals played on the radio, for the most part, so people don't even really need the vocabulary to talk about things they won't be hearing.
-bill

[erik]

ok at risk of making this discussion even more off-topic than it was...

i'm pretty sure "some" and "most" don't mean the same thing. "most" typically refers to more than half -- i.e. if most people behave a certain way, that would mean more do than don't.

I didn't say that they meant the same thing, I was using both words as examples of terms that have both technical and popular definitions. One definition of "most" is, like you said, more than 1/2. But if you put 100 cards (51 black, 49 red) in front of a bunch of 2nd graders and ask them "Are most of these cards black?", you will not get a resounding "yes". You can even let them count the cards, and tell them the "real count" so they don't have to worry that they counted correctly. They will all agree that there are more black than red, but will not agree that "most of the cards" are black. The word "most", on a general level has a colloqiual definition that implies more than just "over 1/2". If I said I had most of the Simpsons episodes on tape, and people come over to my house and I only have 8 out of 15 seasons, there will not be a consensus as to whether I have "most" of the seasons, even though by one strict technical defintion, I do. To go back to the 100 cards, if I start trading out red cards and replacing them with black cards, there will be some point around the 2/3 to 3/4 range where everyone will agree that "most cards" are black. The general definition is somewhat different from the technical definition.

[Leaf]

This conversation reminds me of this arguement a buddy and I had that dragged on for years: when I first joined his cover band in 1993, I did not listen to "classic rock" or blues and that stuff. I was intohardrock and jazz mostly...

We jammed twice, but they had a very busy gig schedule, so I told him "just tell me if the tune is a shuffle, or straight rhythm and that'll do for tonight..."

Well, it comes time to play this particular tune which I can't remember the title of! and Ray turns around and says "it's a shuffle"... through the whole tune these two guys keep looking back at me like I'm insane... I don't know the tune, but after Ray says "what the hell was that? I said a shuffle". I said, "that's what I played". Turns out he thinks that shuffle means that chuck berrish power chord classic rock riff that has the fifth move to the sixth, seventh and back thing that tons of guitarists play....

His arguement was "this old blues guy called this pattern a shuffle...."


So, terminology for the sake of clear communication: important.

Terminology to discredit artist work based on definition: pointless.

That's merely my opinion.

[Jim of Seattle]

JB, I always really enjoying reading your posts. You're enviably good at articulating your thoughts.

We used to have an argument over the order of likeliness of the following sentences:

It's likely going to rain.
It's probably going to rain.
It's surely going to rain.
It's certainly going to rain.

We also argued among the quantities implied by these words:
several
some
a few
a couple
a handful
a smattering

Has anyone ever done a study where the term "four or five" is used (or "two or three" or "one or two" for that matter), and determined whether, when someone says "four or five" if it is in fact more likely to actually be four vs. five? I think someone should get on that right away, since I'd find that a useful statistic.

[j$]

But if we're trying to talk about stuff, you gotta agree on terminology before you can actually communicate ideas effectively. (Even cavemen had to agree that two grunts meant "club" and one meant "sharpened rock").

To use the caveman example, they could the following week decide they need different grunts to describe different length clubs, but if they then used the 'two grunt' to describe any club, it might be wrong, but they would both get the message. For me language is first and foremost about communcation. As Laurie Anderosn sang 'Language is a virus' - it evolves and mutates so if 'song' has become a catch-all word for lots of types of music, then we should embrace it. Other words do not need to be forgotten as a result.

In short, I'm agreeing with you, I think!

Jim, I hadn't heard that about Paranoid Android. i wonder if radiohead knew/intended that when they wrote it. Probably. A thought back at you, one which i honestly don't have an answer for - do people who couldn't recognise a sonata-allegro (me included) get less from the song by not having access to that level?

j$

[Jim of Seattle]

Yes, definitely. But in that case admittedly not very much less. In some cases a LOT less, as in my Chinese opera example.