Time Magazine recently picked their 100 all-time greatest pop songs, an ambitious goal, certainly, and not anything that hasn’t been done before. But their criteria deviates a little from the usual flat listing:

• All selections are unranked.
• Only one song per artist.
• Only songs written after 1923, the year of the magazine’s founding.

That’s not unsound criteria, and actually, it’s a pretty good list of songs. ¹ But given their rules, I don’t know if “most extraordinary” is the right word. Certainly the list is full of excellent, excellent songs, but are they really the greatest?

Let me nitpick. There’s no “In The Mood”, or “My Generation”, or “A Day in the Life”, or “Good Vibrations”, or “Like A Rolling Stone”. These are not deep cuts. They are regular stars on countless lists, and were it not for the “one song per artist” rule, they would almost certainly be on Time’s list.

The proportions are also fairly interesting:

Decade	Songs
2010s	1
2000s	9
1990s	10
1980s	10
1970s	21
1960s	19
1950s	10
1940s	10
1930s	8
1920s	4

40% of the list is dominated by the 1960s and 1970s. Is that overkill? Probably not; it’s arguable that the uniqueness of the recording medium only came into its own in the 1960s and 1970s.

But I do have a problem with the weight given to the last couple decades. Do we really have so much historical context that we already know the best songs of the 2000s? Think about it: there are just as many songs for the 2000s as there are for the 1990s and the 1980s. Certainly we can’t already be that sure.

On the complicated and highly subjective art of music lists

Two constants:

1. You can’t make a science out of it.
2. Some group of people will find some reason to be unhappy with it.

See above.

Doing an all-time best is especially hard. That’s why lists like Rolling Stone’s generate so much controversy (which, given the free publicity, might be the point).

Most lists reflect their site’s audiences. When Pitchfork did their all-time lists, they opted to sort it by decade: 100 best songs of the 1960s, 1970s, etc. That covers an awful lot of ground — perfect for Pitchfork’s reader base of mostly music nerds — but it’s not very useful for someone who just wants a general survey.

iTunes might strike the best compromise. For each of their iTunes Essentials (iTunes link) selections, they put songs in three categories: The Basics, Next Steps, and Deep Cuts. This is great when you’re looking for specific decades and artists; it gives you the best 25 to start with, then lets you drill down from there. Problem is, the categories aren’t quite what we want, which is just a flat list from a very general pool.

For what it’s worth, my perfect list criteria

Is this:

1. All songs are unranked, sorted by decade (like Time’s list.)
2. Songs are at least 20 years old. Thus, the awkward but appropriate title would be something like “The Top 100 Greatest Recordings That Are At Least 20 Years Old”.
3. All selections refer to specific recordings, not compositions. The difference is key: that means an influential cover version can outrank the original if the cover is more widely regarded. It also means some of Time’s selections, like Johnny Cash’s 1968 rendition of his 1955 song “Folsom Prison Blues”, would be listed as a 1960s song, not a 1950s one.

Now, quick! Someone go out and write this up for me.

---