Music in The Legend of Zelda: Ocarina of Time, part two

April 9, 2008

[Part one | Part two | Part three | Part four | Part five | Part six]

This post continues analyzing the songs in Ocarina of Time and theorizing about Koji Kondo’s intentions with them.

Before reading, I encourage you to check out the comments on my last post. Hopefully the discussion between Ben Abraham and I can clarify what I was saying, as well as point out where I might have gone wrong. I’m grateful to have the perspective of another music student on this, and I’d like to thank Ben for engaging with me.

Now, on to business! We have three more songs to look at from the “child Link” portion of the game. Here’s the solo ocarina version of “Zelda’s Lullaby”:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

And here’s a harmonized version, transcribed from the Hyrule Castle Courtyard background music (thanks, Kyle, for catching a typo):

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

The harmony is, in my opinion, very well done — there’s a great chromatic bassline from C down to A in measures 4-7, seamless modulations between G major and C major, and a somewhat rare (for Kondo) tritone substitution in the penultimate bar. But that’s not what I want to focus on here.

With the conspicuous absence of the iconic “Hyrule Overture,” “Zelda’s Lullaby” is the closest thing Ocarina of Time has to a theme song. Link is required to play it more often than any other piece, and the melody can be heard during several plot points throughout the game.

Significantly, it’s also the only ocarina song that isn’t an Ocarina of Time original — “Zelda’s Lullaby” was in A Link to the Past as well. That means it’s the only piece that we know predated the five-note restriction.

Why is that significant? Given its prominence, I think it’s very likely that “Zelda’s Lullaby” dictated — at least in part — which pitches were chosen for the five-note palette.

My guess is that Kondo knew he would at least need A, B, and D to work with for “Zelda’s Lullaby,” and added the lower D and F so that he would have the minor triad D F A to work with also. It’s just my pet theory, but it would explain why five of the twelve ocarina pieces are written in some form of D minor harmony.1

Next let’s look at “Song of Time”:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

There’s nothing too complex going on here; in fact, this phrase functions very much like “Song of Storms.” The trigger motive is an inverted D minor triad (A D F) which firmly establishes the harmonic area, and the answer motive continues in the same vein. Let’s move on to the Temple of Time background music, which features the “Song of Time” melody:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

There are three things to note here, and two of them are obvious just from looking at the score. First, there’s no harmony. In music theory we call this monophonic texture — a single melody without accompaniment. Second, the phrases are of irregular length. You’ll note that I transcribed the piece without meter, and if you count up the number of beats in each phrase you’ll see why — there are no consistent metrical divisions. (I used bar lines to separate phrases.)

The third item of interest is the strong sense of modality. While in “Song of Storms” we could only deduce the D Dorian mode through analysis of the harmony, here it’s placed front and center in the melody. Every time a B♮ appears, Kondo has approached it by half step from above. The upshot is that the Dorian raised sixth is emphasized, and any sense of modal ambiguity is obliterated.

The monophony, irregular phrase length, and strong modality are all characteristic of a genre of music called plainchant, of which Gregorian chant is the best known variety. I think Kondo has captured the spirit of the style quite well; he really shows off his range with this one.

Finally, let’s look at “Saria’s Song”:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Like “Epona’s Song,” this is a bit ambiguous without harmony. The trigger motive outlines a tritone (F-B), and the answer motive cadences with an arpeggiated E minor triad (B G E). Here’s the Lost Woods background music, which features the “Saria’s Song” melody:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

This is one of my favorite pieces of Kondo’s. Harmonically, it’s dead simple.2 The only real surprise is the turnaround on the III chord at the end, and even the III-IV progression isn’t so unusual.

It’s the melody that makes this one of the catchiest songs in the game. The F-B tritone over the F major chord is a great hook, and the insistent “ti-ti ta” rhythmic pattern is arresting. There’s a remarkable amount of intensity generated in a very short piece. I don’t have any greater point to make here, but Darunia is right — what a hot beat!

Next time we’ll take a look at the songs Link learns as an adult.

[Part one | Part two | Part three | Part four | Part five | Part six]


  1. Having two octaves of the same pitch available — D, in this case — is also compositionally useful. Two of the ocarina songs are in D major, which means that nearly three quarters of the pieces use D as a tonic.
  2. So simple, as it turns out, that several of the chords sound to me like they’re missing thirds. I’ve reflected that in the transcription, but have assumed diatonic harmony in the analysis for the sake of convenience.

9 comments

Hawt beats! Awesum!

I think your unpacking of (admittedly) simple tunes is really valuable – and revealing. I guess you are revealing the why they are so memorable.

And TRITONES! om nom nom

If I were any good at Jazz, I’d play with Tritone substitutions *all* day… or at least until I got bored and moved on to noise music. =)

Can I ask how long it usually takes you to transcribe the songs? I’m actually rather horrid when it comes down to the crunch… :S

by Ben Abraham on April 10, 2008 at 12:42 am #

It really depends on the piece. I have more trouble with harmony than melody, so Iknocked off the Lost Woods transcription in 10-15 minutes but got hung up on Lon Lon Ranch because of those odd augmented triads.

At some point, though, I just want to get the damn things up, so I’ll stop futzing around with the chord voicings and inversions once I feel like I’ve gotten the basic idea.

Glad you’re enjoying the series!

by Dan Bruno on April 10, 2008 at 8:18 am #

u did so well on the songs.
the hermony is perfc with everything.
out 10 u get a 10

by trevor on November 9, 2008 at 2:58 pm #

First of all, I’d like to say thank you for putting together a really interesting collection of essays about a great interactive score. I have to do a presentation on this game’s music in a couple hours and this has been great food for thought.

That being said, I disagree with your analysis of “Saria’s Song.” I hear the tune as being the F lydian. While not easily analyzed with the traditional classical roman numeral system, this idea does the piece a bit more justice, and I feel that you should’ve mentioned it. If you look at the melody, the tonic F modality is really beaten into your brain, and the Cmaj7 harmony functions just like a dominant would, just with a raised 7 due to the lydian modality. Note how the phrase ends on an e natural, the leading tone and then resolves the tonic in the next measure.

I should mention that I go to Berklee and we do a lot with jazz theory hear, which lends itself to these kinds of modal tendancies, and this may be a “tomato-tomawto” argument. But, considering the tones you mentioned that were available to Kondo when composing each of these themes (which I imagined dictated the direction of each composition), I think when Kondo chose F-A-B, he was more likely thinking “do-mi-fi” than thinking “fa-la-ti.”

If it were me (and how I wish it was), I would’ve probably gone in the same direction, since the lydian mode is, in my opinion, the most dream-like, and creates a perfect mood for “The Lost Woods.” Either way Kondo, did a great job with this game.

Let me know what you think of all this.

by Ben Bromfield on December 18, 2008 at 11:46 am #

Hey Ben! Thanks for coming by and offering your comments. It’s great to have perspective from other musicians on this.

You’re absolutely right that Kondo uses elements of the Lydian mode to his advantage here, but I think it’s problematic to analyze the entire piece that way.

The F-A-B melodic hook would be 1-3-#4 if we assumed an F tonic, but since that outlines a tritone it obfuscates the tonicity. In other words, it’s very hard to hear a tritone as a resolution, as you’re proposing here — that’s why Locrian mode is so rarely used as a key center, and why the tonic chord is so often avoided when it is. If Kondo had used the tritone in passing between consonant intervals, I think a strong case could be made for Lydian mode — compare Danny Elfman’s Simpsons theme, which uses 1-#4-5. But as it stands, I don’t feel the melody supports the argument that F is the tonic.

The C chord, meanwhile, feels very much like a resolution. The E that you mention could be analyzed as the leading tone, but it makes more sense to me as just the third of the C chord. (Note the inverted turn-like gesture in bar 4 which reinforces the E as a melodic destination). Also note the clear ii-V-I in C major during the bridge (bars 9-10 and following) — working backwards, it feels odd to analyze that section as modulatory. I like what you’re doing, but I think I’m going to stick with my initial analysis on this one.

In any case, thanks for your input — you’ve given me more to think about here!

by Dan Bruno on December 21, 2008 at 5:17 pm #

Wow, the nostalgia hit me like a ton of bricks with Zelda’s lullaby and the Temple of time song…very nice reminders of my childhood!

re: the tonality of Saria’s song, I see both points, especially what with the ii-V-I in the bridge in C major (tritone subs and ii-V-I? Kondo is JAZZ) but I still feel a strong resolution to F…as well, the piece has a certain “lydian character” to it—I have heard it said that there’s a dreamy feel to the mode. I wish I had some decent harmony/theory to back me up, I feel a bit…less erudite than the author and reader base…but oh well. this is how we learn, after all!

by Sam on January 19, 2009 at 3:15 am #

wow that was so cool. i know for a fact that what Kondo does is what I want to do with my life cause uhhh… this middle school band thing isnt too fantastic.

looking at song of time, though, It sounds like its in four with a pickup quarter. Re-express the fermata whole notes?

by Maximillian Montiel on April 26, 2009 at 11:22 pm #

I have also found it unusual how many peices from the game actually are written in some form of D minor tone. I would conclude the song of storms as Dorian because of the 6th scale degree in the bassline. Is that correct?

by Michael Fields on June 10, 2009 at 1:00 am #

@Michael

Dan first mentions the Song of Storms in Part 1 with D dorian.
http://cruiseelroy.net/2008/04/ocarina-music-1

He also mentions it later in Part 4.
http://cruiseelroy.net/2008/04/ocarina-music-4

So yes, that’s correct. You’ll even find his short discussion of D dorian on page 1.

by Rafi on June 10, 2009 at 8:56 am #

Leave your comment

Required.

Required. Not published.

If you have one.