Page 1 of 2 12 LastLast
Results 1 to 15 of 20

Thread: So Nice (Summer Samba)

  1. #1
    Registered User
    Join Date
    May 2012
    Posts
    23

    So Nice (Summer Samba)

    Hello guys,

    Well I injured myself a while back and I'm slowly picking up playing again. I figured this would be the right time to dive more into theory, so here's an analysis I've just done of So Nice. I've searched but I think it's never been talked about here before.

    Fmaj7 | % | Bm7 | E7
    I -----------|II/III | V/III

    Bbmaj7 | Bb6 | Eb9 | %
    IV ------| IV --| V/IV----

    Am7 | D7b9 | Gm7 | Em7b5-A7#5
    III --| V/II --| II ----|II/VI-V/VI

    Dm11 | G7 | Gm7 | Db9-C9
    VI ----| V/V | II ---| #V-V

    Let me know what you think!

    Thanks

  2. #2
    balladeer page's Avatar
    Join Date
    Nov 2005
    Location
    the windmill area
    Posts
    5,244
    Hi mm2703,
    can you explain what you mean with the %? Maybe you could indicate where the measures are? Like with | - - - - | ? If it's not too much trouble that is.
    I'd like to understand while I'm learning theory myself. Thanks.

    Regards, page

  3. #3
    Registered User
    Join Date
    May 2012
    Posts
    23
    Hi page,

    % means the chord is repeated. I believe it's the standard way of indicating it. I'll try to reformat my analysis right now. I had indented the chords with spaces but they have been trimmed when posting.

  4. #4
    balladeer page's Avatar
    Join Date
    Nov 2005
    Location
    the windmill area
    Posts
    5,244
    Ah right! I didn't think of that. It is much clearer now, thanks.

  5. #5
    Piano/Compose/Arrange engelbach's Avatar
    Join Date
    Dec 2005
    Location
    Pátzcuaro, Michoacán, México
    Posts
    7,678
    Roman number analysis can be done a number of ways.

    For example, one might consider BbΔ not IV of F, but a modulation to IV, which would make it I/IV. The melody gives no clue, as it doesn't touch on the notes that differentiate the keys of F and Bb, the E and Eb.

    One also has to include chord qualities where relevant. II of F is ambiguous. Do we mean diatonic II, which is just a Gm triad, or IIm7, or II7, which is a dominant G7?

    Eb9 is not a V chord here. It's a IV7 of Bb. Or, if we decide that we're still in the key of F, a bVII7 of F.

    The Am7 and D7 have a dual function, as the IIm7-V7 of Gm, and as the IIIm7-VI7 of F. Since there is no real modulation to Gm, I prefer to define them in terms of F.

    The same thing goes for Eø-A7. We're going through a cycle that's inevitably leading us back to F, with no real modulation to Dm.

    The Db7 would normally be notated as bVI7 rather than #V7.

    So I would analyze it as follows:

    Fmaj7 | % | Bm7 | E7
    | IΔ --- | IΔ --- | IIm7/III --- | V7/III --- | (or | #IVm7 --- | VII7 --- |)

    Bbmaj7 | Bb6 | Eb9 | %
    | IΔ/IV --- | I6/IV --- | IV7/IV --- | IV7/IV --- |

    Am7 | D7b9 | Gm7 | Em7b5-A7#5
    | IIIm7 --- | VI7 --- | II7 --- | VIIø - III7 - |

    Dm11 | G7 | Gm7 | Db9-C9
    | VIm7 --- | II7 --- | IIm7 --- | bVI7 - V7 - :||

    Cheers,
    Jer
    Jerry Engelbach, piano/arrange/compose
    Engelbach Music
    Weaver of Dreams
    Artwork

  6. #6
    Registered User
    Join Date
    May 2012
    Posts
    23
    Hi Jer,

    Thanks for the help. I think I was a bit afraid to use un-diatonic chords, that's why I've tried going a bit too much into other degrees. It had one advantage to me, though: I could notate Em7-A7 as a iim7-V7, which it clearly was to me. And I still think that viim7-III7 isn't as easy to understand. What's your opinion?

    Thanks

    Maxime

  7. #7
    Piano/Compose/Arrange engelbach's Avatar
    Join Date
    Dec 2005
    Location
    Pátzcuaro, Michoacán, México
    Posts
    7,678
    Quote Originally Posted by mm2703 View Post
    Hi Jer,

    Thanks for the help. I think I was a bit afraid to use un-diatonic chords, that's why I've tried going a bit too much into other degrees. It had one advantage to me, though: I could notate Em7-A7 as a iim7-V7, which it clearly was to me. And I still think that viim7-III7 isn't as easy to understand. What's your opinion?

    Thanks

    Maxime
    I prefer to notate a non-diatonic progression in relation to the key we're in, unless we're actually modulating to a new key.

    But as I say, it's a matter of preference.

    If in a IIm7-V7-I non-diatonic progression the I stops definitively on a new tonic, then it's a modulation. But if it's just going to another IIm7-V7 it's usually just part of an overall movement towards a resolution in the key of the piece.

    One gets used to progressions like VIIm7-III7. If one treats every non-diatonic IIm7-V7 as a modulation the overall picture is lost.

    Here's Jerry Coker's notation (Improvising Jazz) for the first ending to Love Is Here to Stay:

    |IΔ---|VIIø-III7-|VIm6---|II7---|IIm7---|V7---:||

    One is expected to understand that VII-III-VI-II-V-(I) is a progression around the circle of fifths.

    Here's John Mehegan's notation (Jazz Improvisation 4: Contemporary Styles) for the beginning of Stella By Starlight:

    || bVm | VIIx | II | V | Vm | bV | IV | bVIIx | (Mehegan actually uses "/" as a bar line, but I find that confusing.)

    Mehegan used an unorthodox system wherein all chords are considered diatonic unless otherwise notated, and he uses "x" to indicate a dominant. The Vm - bV - IV (bV is a triton sub for Ix) is functionally the same as IIm7/IV - V7/IV - I/IV. But Mehegan assumes that we understand that and avoids the complicated notation of slash chords.

    Cheers,
    Jer
    Jerry Engelbach, piano/arrange/compose
    Engelbach Music
    Weaver of Dreams
    Artwork

  8. #8
    Piano/Compose/Arrange engelbach's Avatar
    Join Date
    Dec 2005
    Location
    Pátzcuaro, Michoacán, México
    Posts
    7,678
    I wanted to edit my post but the Edit function is not working on the Forum.

    Change my third paragraph to:

    If in a IIm7-V7-I non-diatonic progression the I stops definitively on a new tonic, then it may or may not be a modulation. But if it's just going to another IIm7-V7 it's usually just part of an overall movement towards a resolution in the key of the piece.

    Add to my last paragraph:

    Coker notates his version of Stella with the same numerals.
    Jerry Engelbach, piano/arrange/compose
    Engelbach Music
    Weaver of Dreams
    Artwork

  9. #9
    Piano
    Join Date
    May 2007
    Location
    So.Cal.USA
    Posts
    952
    Quote Originally Posted by mm2703 View Post
    Hi Jer,

    Thanks for the help. I think I was a bit afraid to use un-diatonic chords, that's why I've tried going a bit too much into other degrees. It had one advantage to me, though: I could notate Em7-A7 as a iim7-V7, which it clearly was to me. And I still think that viim7-III7 isn't as easy to understand. What's your opinion?

    Thanks

    Maxime
    Hi Maxime,

    Lots of different systems of roman numeral analysis, so I'll give you my somewhat different take on it ....

    First of all, I think there are two fundamentally different approaches to using roman numerals ... one I call "descriptive", the other "functional". They serve two different purposes.

    Descriptive roman numeral analysis pretty much numbers every chord with reference to the tonic. So, for example, the Db7 in this tune would be identified as bVI7 (or #V7 if you prefer), because its root is the flat-six (or sharp-five) of the tonic F. And a 3-6-2-5-1 would be IIIm7 - VI7 - IIm7 - V7 - Ima7, with every chord referenced to its position in the F scale. I think this is similar to the approach of the Nashville Numbering System.

    A huge virtue of descriptive roman numeral analysis is that it can be easily used to play and sight transpose a tune ... it's quickly readable. If you're playing, you may not particularly care about the technical harmonic function of Db7, you just want to quickly get to that flat-six chord in any key.

    Functional roman numeral analysis, on the other hand, focuses on harmonic function, which can make it unwieldy for use in playing and sight transposing, but very useful for understanding what's going on harmonically. For example, as I see it, the harmonic function of Db7 in this tune is as a tritone substitute for G7, which is V7/V in the key of F. So in functional roman numeral analysis in the jazz context the Db7 would often be shown as "subV7/V". Of course, if you're playing the tune and trying to transpose it on sight, that's a lot of "bits" of information to visually process; "bVI7" would register much more quickly in your brain.

    Same with 3-6-2-5-1. Descriptively, it's IIIm7 - VI7 - IIm7 - V7 - Ima7, with each chord identified simply by its scale degree. Functionally, though, it's IIm7/II - V7/II - IIm7 - V7 - Ima7, which clearly identifies the secondary dominant function of the second chord in the sequence, and the related IIm7 function of the chord that precedes it.

    Which brings me to those slashes ("/"). I think Jerry's view is that a slash suggests that a modulation is occurring. Personally, I view the slash as signalling the opposite, i.e., that no modulation is occurring, because the slash is used to express the harmonic function of the chord in question relative to one of the diatonic chords of the key. Most often, the chord in question is a secondary dominant, such as V7/II (descriptively, VI7). That secondary dominant can be viewed as temporarily tonicizing the chord that follows it, but that's not the same as a modulation.

    As one theory book puts it: "Some analysis systems use I7, II7, III7, and so forth, to analyze these secondary dominants. The problem with this method is that it only indicates the chord's position in the key and quality; it does not say how the chord is functioning. Function is all important. The analysis used will indicate that the secondary dominant chord is dominant (V7) of a diatonic chord (II, III, IV, V, VI). To abbreviate this, a slash is used to mean 'of.' V7/II means V7 of II." Nettles & Graf. By definition, a secondary dominant has a non-diatonic chord tone, but it has a diatonically related function and the slash is used to show that function.

    In the system I'm familiar with, modulations are shown with colons (":"), not slashes. So, for example, if a tune modulates to the key of the IV of the original key, the start of the modulated section would look like this:

    IV: Ima7 (or whatever) - etc. (So, in a tune written in the key of F, this would be understood to mean that everything following the "IV:" is in the key of Bb, until otherwise indicated.)

    When it comes back out of the modulation, it would look like this:

    I: Ima7 (or whatever) - etc.

    Something else that complicates analysis is that a chord sometimes can be heard to have one function "on arrival" but another function "on departure". That is, it will be heard one way when it first hits, because of its relationship to the chord that precedes it. But then the next chord hits and the chord in question may take on a different "meaning." This may be because the chord is being used to pivot between keys. In that case sometimes two different roman numerals might be shown on the same chord, one in each key, to show the pivot function. I think there's a good argument that the Eb7 in this tune has a dual function.

    Anyway, that's just my take on it. There are many different r.n. systems and approaches.

  10. #10
    Registered User JoeB's Avatar
    Join Date
    May 2009
    Location
    Virginia, USA
    Posts
    166
    This is one clever tune, both harmonically and melodically. I agree with Mike A and took a first shot (my take anyway) at a functional analysis based on Berklee harmony back in the day when Barrie Nettles was a Harmony course author at the school.



    The solid bracket underneath the chords indicates a II-V relationship. The solid curved line with ending arrow indicates a dominant resolution down a perfect fifth. The dashed line with ending arrow indicates a substitute dominant resolution down one-half step. Dual function chords are indicated in parenthesis.

    The melody must be factored in to derive a correct harmonic analysis. Given the first two measures are tonic chords, things immediately get harmonically interesting in measure 3. We have a fairly unusual non-diatonic rooted #IVm7 chord. This chord would be easier to analyse as a tonically related #IVm7b5, and I have seen various charts and recordings that do use the m7b5 chord here, but let's stick with the m7 chord. It is fresh sounding, surprising, and indicates we are likely going to be taking an interesting path to get back to tonic.

    Given the following E7, we can add the II-V bracket. The related IIm7 of a dominant chord typically has the melodic tendencies of the IIm7 chord in a major key, but there are exceptions. One is VIIm7, which wants to sound like IIIm7 of whatever key it is the third diatonic structure of. This tune has made me realize that this can be the case for #IVm7 also. In this specific key, we have Bm7 wanting to sound like IIIm7 in G. The melody confirms this with the C note. I indicated this using the dual function parenthesis in combination with the : notation, which is likely not be the formally correct way to indicate this but is the best I could come up with for now.

    Another way to think of this is that sometimes when the harmony shifts to another tonal center the melody stays the same, overriding a possible alternate melodic direction. That is precisely what happens in measures 1 and 3.

    In measure 4 we likely hear the E7 as a secondary dominant V7/III. It is important to differentiate secondary dominants from special function dominants as that can change the expected melodic choices, chord extensions, and resolution tendencies. For example, VII7 usually cadences to I and likes natural chord extensions (9 and 13 for example). We are not likely at a point where we would return to I here though, and the melody changes only one note from measure 2, modifying the A down to G# (the 3 of the chord) but keeping the C note (b13 of the chord). We'll come back later to why I have the parenthesis around (subV7/VI).

    In measure 5 however we have a deceptive resolution, or more likely a delayed resolution as we consider later measures. I like to use diatonic chords as much as possible in an analysis, so have listed the Bbmaj7 as IVmaj7. The measure before sounded E notes as the root of the chord and on stronger beats of the melody than the Eb note sounded in the chromatic line, so I like to hear the E note ring over into the Bb chord, indicating IV. So in the absence of melodic clues, when do listeners tend to like to hear IV sound like a I? Two possibilities: when IV is preceded by its V7, or followed by IVm. We clearly do not have the first possibility (that would be F7 --> Bb). The second possibility gives us Bb to Bbm, which I'll come back to after looking at some other things in measure 7.

    So that brings us to measure 7 and 8's Eb7. We likely first hear this a special function bVII7, and so have that in parenthesis. However, if we look back to measure 4 we had that listed as V7/III. That would normally resolve to Am7 in this key. If it resolved to A7 then the E7 would be a extended dominant. The A7 would then be the secondary dominant V7/VI. Using the substitute secondary dominant of A7 gives us Eb7 as subV7/VI. We can thus now draw a dashed curved line from E7 to Eb7 to indicate the chromatic extended substitute dominant resolution, and put parenthesis around the V7/III indication. Again we'll come back to why I have the parenthesis around (subV7/VI). Also, Eb9 can be a disguised IVm6, in fact I hear some Summer Samba recordings that do this. Would that then perhaps give a tonic flavor to the preceding Bb chord?

    In measure 9 we have Am7. Back in measure 7 and 8, we likely first heard Eb7 as bVII7, the backdoor dominant that likes to resolve to I. Am7 as IIIm7 in F is a tonic substitute, so I have it listed as a dual function chord in parenthesis, and from bVII (the Eb7) would be a deceptive resolution. However, I have not drawn a curved line to the Am7 because it also is in a II-V relationship with the following D7, thus the bracket. We had the Eb7 as subV7/VI which would resolve to Dm7. Now that we see the D7b9 in measure 10, we can add parenthesis around the subV7/VI and can draw an extended substitute dominant dashed curved line from Eb7 to D7b9.

    Whew, mm2703, you picked a really interesting tune to dive back in on. I have to run out for a while; I'll finish up the analysis in a bit.

  11. #11
    Piano/Compose/Arrange engelbach's Avatar
    Join Date
    Dec 2005
    Location
    Pátzcuaro, Michoacán, México
    Posts
    7,678
    Mike A is correct, and I stand corrected.

    Slash chords denote a secondary dominant function, and the use of a colon to indicate a true modulation is the better way.

    I also agree with dividing the two types of notation. It's the same as what I've written before about ordinary chord symbols: there are analytical symbols that reflect what is actually being played, and performance symbols that are a more simplified sketch for the player.
    Jerry Engelbach, piano/arrange/compose
    Engelbach Music
    Weaver of Dreams
    Artwork

  12. #12
    Registered User JoeB's Avatar
    Join Date
    May 2009
    Location
    Virginia, USA
    Posts
    166
    It is interesting to examine tunes like this that were wildly popular in the past (the 60's in this tune's case), and still show up in popular culture like movies (Austin Powers in this case), various TV shows, and computer games. You almost take tunes like this for granted until you re-study them and realize just how well they are written, and in this respect reminds me of The Girl From Ipanema. So I'll continue with a little more analysis and if you are thinking that all this is very complicated, I'll try to show at the end how it can simplify things.

    So we were at measure 10 and identified the D7b9 as a secondary dominant V7/II, and in measure 11 it does resolve to Gm7 (IIm7) so we connect them with a solid curved dominant resolution line. In measure 12, we have another diatonic chord Em7b5 which is VIIm7b5 but more usually functions as the related II of V7/VI which it does here, so we bracket it with the A7#5.

    In measure 13 the expected resolution occurs to Dm7 as VIm7, so we draw the solid curved dominant resolution line to it from the A7#5. Measure 14 has G7 V7/V, so we bracket it as a II-V with the previous Dm7 and enclose the VIm7 in parenthesis to indicate that it has dual functionality.

    In measures 15 and 16 we see the V7/V resolves to V7 with some intervening chords, so we draw a solid curved line to indicate dominant resolution from the G7 to the C7. The Gm7 is the related II of C7 so we add the solid bracket. The Gm7 sounding between the G7 and the C7 in jazz is referred to as an interpolated IIm7 (kinda cool, huh). The Db7 before the C7 is the tritone substitute of G7, and so is labeled subV7/V.

    So how does all this simplify things? One example is that the curved and dashed dominant resolution lines show that the main harmonic activity to get to measure 11 is a good old turnaround to II, with some added "interpolation" and augmented harmonic rhythm. Basically we have:
    F ---> E7 ---> (interpolation) ---> Eb7 ---> D7 ---> Gm7.
    It might not be a stretch to consider the Bb chords tritone subs for E7 (the root movement kind, not the tritone sharing kind).

  13. #13
    Piano
    Join Date
    May 2007
    Location
    So.Cal.USA
    Posts
    952
    Very interesting analysis, JoeB! The more you dig into this tune, the more you find.

  14. #14
    Registered User
    Join Date
    May 2012
    Posts
    23
    Quote Originally Posted by engelbach View Post
    I also agree with dividing the two types of notation. It's the same as what I've written before about ordinary chord symbols: there are analytical symbols that reflect what is actually being played, and performance symbols that are a more simplified sketch for the player.
    I do, too! I wasn't aware of this division but it makes a lot of sense. Thanks, Mike.

    I'll have to read Joe's analysis again a few times though because this is the first time I meet some of those concepts. But that's the best way to see precisely what I've missed in my analysis. Thanks!

  15. #15
    Registered User
    Join Date
    May 2012
    Posts
    23
    Joe, is there a reason you didn't add a dashed line between D7b9 and Db7?

Similar Threads

  1. jazz + samba
    By funk freaka in forum General Music Discussion
    Replies: 10
    Last Post: July 23rd, 2013, 04:01 PM
  2. Duduka Da Fonseca Quintet: Jazz Samba-Samba Jazz
    By RSS Feed Guy in forum RSS Feed Articles
    Replies: 0
    Last Post: November 18th, 2012, 01:30 AM
  3. Samba Da Volta
    By raymb1 in forum Music Theory and Analysis
    Replies: 9
    Last Post: December 25th, 2011, 08:37 AM
  4. So Nice (Summer Samba)
    By BrazJazz in forum New to Jazz? Have a Question?
    Replies: 2
    Last Post: March 26th, 2007, 09:47 PM
  5. Samba e choro
    By Luis Guilherme in forum Musician 2 Musician
    Replies: 9
    Last Post: October 25th, 2003, 05:39 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  




Support the All About Jazz website and forum. Make a donation today!



Download the Jazz Near You iOS app

Download the Jazz Near You iOS app - Free!

Never miss another jazz concert again! Jazz Near You is a simple yet powerful way for fans to discover who is playing where and when. Access local jazz events by date, by distance, by venue, by musician or by festival; map to venues, set reminders, and get detailed information about musicians. Jazz Near You is your complete guide to jazz music near you! Download it now.



Visit All About Jazz at Twitter   Twitter Visit All About Jazz at Facebook   Facebook Use the All About Jazz content widgets on your website or blog   Widgets Subscribe to the All About Jazz RSS feeds   Feeds


All About Jazz | Jazz Near You | Jazz Musician Directory | Jazz News | Jazz Photo Gallery