IS SAM SMITH’S song Stay With Me based on Tom Petty’s I Won’t Back Down? What about Marvin Gaye’sBlurred Lines and Pharrell’s Happy? Are those too similar to be considered unique songs?
Let’s look back at a much older case. In 1976 there was a case against George Harrison for his song My Sweet Lord being plagiarized from the song He’s So Finecomposed by Ronald Mack. I don’t want to go over all the details of the case – but you can read a pdf of the court document here. Instead, let me focus on the parts I think are important. Note – I am not a lawyer. I think that’s obvious.
Both songs contain the same short musical phrase of “sol-mi-re” repeated 4 times. This is followed by 4 repetitions of “sol-la-do-la-do”. (this is straight from the court document)
It seems reasonable that George Harrison and others were aware of He’s So Fine since it topped the charts for 5 weeks.
I guess if there was some unknown (or not well known) song, it would be difficult to convince the court of plagiarism.
It is likely that George Harrison did not deliberately copy the song – but that doesn’t matter.
Perhaps this part of the testimony (or whatever it is called) is the most important (for me):
So maybe it is just these three notes (sol-mi-re) that caused the plagiarism. Maybe.
HOW MANY THREE NOTE SONGS?
Suppose that just three notes are like a fingerprint for a song. If that’s the case, how many songs are there? I am going to go with the same convention from the court case. Only the notes matter, not the timing of the notes. Using the “do-re-mi-fa-sol-la-ti” method of naming notes, there are only 7 possible notes. Yes, I don’t know what I’m talking about here – I am just going by this Wikipedia page on Solfege.
How many combinations of notes could you have? Let me start with a different example. Suppose I had a set of 10 notes. Maybe I can label these notes with the numbers 0 through 9. Here are some sample combinations of these three items.
Yes, those look just like actual numbers. The smallest number would be 000 and the largest 999 for 1000 total possible combinations. It might be easier to write these combinations as:
Ok, now replace the 10 digits with just 7 notes. How many combinations are there?
There are just 343 different 3-note combinations. But wait – I already know you want to complain. You are saying that there aren’t just seven notes. I understand your complaint. Really, this is like saying there are 7 colors in the visible spectrum. Really, the visible spectrum is a continuous range of light with different wavelengths. The same is true for sound. However, humans find it best to break these notes into different chunks called notes.
There are some other ways of grouping music into a different scale. Some of these music scales have as many as 12 notes. With 12 notes, you could get 1728 different sets of three notes. However, this is not the most common Western scale. And this post is not about the number of different songs, it is about the number of different songs that could cause a lawsuit. So, I am sticking with the 7 notes.
But what about the beat and rhythm of this short musical segment? According to the court document, this didn’t matter. But what if it did? In the world of music, there is a whole note. This is a note that lasts 4 beats. You can also have a half note (2 beats) a quarter note (1 beat) an eighth, a sixteenth, and a thirty-second note. That means there are 6 different “rhythm” types for each note in the sequence. That means that if you want a note, it could have 7 different “tones” and 6 different “rhythms” (yes, I am probably using the wrong terms here) for a total of 13 varieties.
With this, there could be 13 to the third power of combinations or 2197 different song melodies. Really, I think that the rhythm thing is over rated. I suspect you could break this down into just 3 note lengths in order to not sound too similar. Really, how many people can tell the difference between a quarter note followed by a sixteenth note vs. a quarter note followed by a thirty-second note? I doubt that I could.
Ok. One more silly point. In Western music, a 4/4 beat has 4 beats in a measure (I think that’s what it’s called). That means that in one measure of music, you couldn’t have 4 whole notes. This assumes this three note melody would fit into one measure and that the song would be in 4/4 time which of course neither have to be true.
In the end, I am going to go with 7 different notes with no distinction due to rhythm. This means that there will just be 343 different songs. But of course, I am not the first one to try to calculate the number of different songs. Rachel Hall (a music professor) used 16 beats (7 notes each) to get a value over 30 trillion. I think that is indeed a valid estimate, but I am still going to use the 343 value.
ONE SONG TO RULE THEM ALL
Here is my plan. I am going to write one song. This one song will have every possible 3 note sequence. If this song then makes it to the top of the charts, every following song will be based on my One Song and thus owe me royalties…..profit.
If I go with the “sol-mi-re” structure repeated 4 times, I can estimate how long it would take to play my chart topping song. Each “sub song” will be 3 notes repeated 4 times for a total of 12 notes. If each note lasts 0.2 seconds (any shorter and it doesn’t sound like a melody) then 343 sub songs would take 12*343*(.2 seconds) = 13.7 minutes. That’s a long song, but not crazy out of control long. It’s short enough to make it to the top of the charts. Just as a side note – songs on radio are “around” 3 minutes long. There’s a reason for that.
Ok, now for the song. Are you ready for this? Here it is – I made it just for you (and for me).
I created this song as a program in the MIT Scratch online programming tool. Yes, it’s not much fun working with a graphical interface, but it plays sounds. I was going to make my stick figure animated too, but I will leave that for others to work on.
So that’s it. All I need to do now is to get my song on the radio and make it to the top of the charts.
culled from : www.weired.com
images from : www.weired.com