Introduction

These notes were made while transcribing ¹ the studio recording of the guitar and bass parts of Marc Ribot’s cover of La Vida Es un Sueño. The transcription, recordings, etc. are here

Who is responsible for this music ? The original music and lyrics are by Arsenio Rodriguez, who wrote it in the early 1940’s. There is a good story, which may even be true, that says he arrived in New York from Cuba and the doctor told him that it was impossible to restore his eyesight, and shortly thereafter he dictated the song to his brother. The recording I worked from is on Marc Ribot y Los Cubanos Postizos (1998 Atlantic), which can be purchased at marcribot.com. It was recorded at Water Music, in Hoboken; it has that inimitable Hoboken sound. In an interview, Ribot gave great artistic credit to the producer JD Foster. Ribot plays the guitars. Brad Jones plays bass (included in the transcription). John Medeski plays organ. Percussion is by Robert J. Rodriguez and E J Rodriguez.

Notes on the Music

The song (the recording of the song by Los Cubanos Postizos) has an A B A B A A structure with variations in all repeated parts. Each of A and B is a chord progression. Bars 1-11 are A and consist of solo guitar, with tremelo and not too much distortion and probably finger picked. Bars 12-23 are B and add spoken vocals. Bars 24-31 are A played a bit differently and with sung vocals. Bars 33-45 are B and are the guitar solo. The arrangement here uses distorted guitar, electric bass guitar, organ (or something like that) and some kind of stick, maybe claves. Bars 46-53 and 54-61 are A done twice with the same instrumentation as the previous section but added sung vocals and maybe some other noises. Bars 62-64 are an outro consisting of solo guitar.

I play bars 1-32 mostly with my thumb and fingers when necessary. I use the side of the thumb and a bit of nail for normal or louder or more percussive notes and rotate the thumb and use more pad for the soft notes. I play it on a telecaster with the neck pickup and electronic rotating tremelo. I don’t know what Ribot uses. It sounds like he arpeggiates, at least slightly, every chord, which makes me think he may be strumming somehow. I play the rest of the song with a pick and more distortion and no tremelo.

Notation used in the transcription

This is my first transcription, the first time I have written down music in any notational system. I learned most of the music theory using the “this guy on the internet” method. My choice of notating enharmonic pitches may be wrong.

Chords and Frets:

I began putting fret diagrams on the score simply to notate easily where to put my fingers, because in several sections Ribot plays a chord or broken chord or arpeggio with an immobile left hand. I began to put chord symbols (ie, names G, Am, etc.) over the fret diagrams that corresponded to obvious chords. Then I began to try to find the chords for all of the music. The result is that the chord symbols don’t have a single purpose. They may be a literal interpretation of the fret diagram. They may represent the chord that I think Ribot and Jones may have in mind. They may represent what I think is the basic chord in the progression without a substitution.

Instead of fret diagrams, I sometimes use an approximation of standard classical guitar notation by putting arabic numerals above note heads to indicate left hand fingering. Likewise I sometimes use italic roman numerals to indicate left hand position, and circled arabic numerals to indicate the string. A zero with a tail is borrowed from cello notation to indicate fretting with the thumb. I would like to indicate barres and the duration of positions more accurately, but all of this is laborious in LilyPond. Unfortunately, it is not possible include the standard notation for string bends in transcribed blues guitar. I have tried to indicate some of this with the word “bend” and text notes below. I have used a cup above notes to indicate a quarter bend. These are never really quarter bends, but indications to listen carefully at that bar for how the guitarist bends the note. Slurs indicate hammerons and pulloffs. Glissandos are indicated by straight lines (sometimes too short or misshapen, depending on how much effort I was able to put into fighting LilyPond), and represent slides.

Barres are not notated, but should not be hard to figure out.

Articulation:

I use the tenuto symbol, a bar over a note head, to indicate that a note is held; a staccato symbol, a dot, to indicate that a note is shortened; and a portando symbol, a line and a dot, to indicate that the note is shortened a bit more than if playing legato or normally. But they are a rather crude approximation of the articulation and rubato on the recording and are merely reminders to listen carefully to the articulation at that postion in the piece.

Engraving System

I used the LilyPond music engraving (i.e. typesetting) computer program. I spent some hours fighting various details of the arcane LilyPond markup system and, as is evident from the score, frequently got the worst of the fight. LilyPond works a bit like writing a TeX document. You create a plain text file representing the music notation and then the computer program processes the file, generating a score in pdf format and a midi file.

There are two LilyPond source files for this project, lveus.ly and lveusbass.ly. Run lilypond on lveus.ly to produce the midi and pdf score (the midi will be wrong; see below.)

Currently the goal of separating structural commands from layout commands is not in sight in LilyPond. Collisions between printed symbols are the rule, and a mish-mash of means that change with software version number are employed to try to fix them. I left a large number of collisions in the score just because I don’t want to invest the time to fix them.

Midi

The midi is not at all for performance. It is useful for hearing wrong intervals and rhythms. However, if you listen to it expecting to find music, you will find anti-music. Generating the midi was rather complicated: The midi that LilyPond will generate from lveus.ly will be wrong dynamically and the the bass and guitar will go out of sync. I generated the midi from a modified version of the source file lveus.ly (I didn’t upload this modified source file). I stripped the dynamics from lveus.ly with a perl script; otherwise the midi is silent in many places. Glissandos to and from nowhere must be marked by making glissandos to hidden grace notes. The note is marked grace so that it occupies no time in the bar. It is marked hidden so that it doesn’t print. But the software that converts to midi, not only does not hide grace notes, but lets them occupy time. So I stripped these commands as well.

Details of Transcription

• bar 4   I need to check this again. It is seems that the space of five eight notes are occupied by seven even beats. The notes are all in the G maj scale except for Bb. The first five notes form a C maj7. I wonder to what extent the choice of the last two notes fits into a practical theoretical idea.

• bar 5   This maybe something like a C#m chord. It is probably related to the last dyad of bar 4.

• bar 7   The first five notes are E7. Almost the same as bar 27

• bar 8   The (1, 2, 3b and 5) from A maj scale are here. Very similar to bar 28. In bar 8 dyads consisting of minor thirds above each melody note are present (lowered an octave). In bar 28, the same notes are present, but with minor thirds below each melody note.

• bar 16   If the bass were here, it would be walking chromatically down from A to F#. One might consider the root to be on A the whole time, rather than walking down. I think the chromatically changing root is probably the correct way to think about it. Anyway I marked the chords both ways.

• bar 25   This chord has the notes C# G Bb E and can be interpreted as 1 3b 5b 6 with any of the four notes taken as the root (with inversions). Since the chord is arpeggiated, the ambiguity over which is the tonic may be enhanced.

• bar 33   Starts with 3,5 of G chord. The 3b appears as well. These plain major and minor chords (and distortion and volume) produce a sudden rocky feel compared to the piece up to this point.

• bar 34   G maj, or G min ? Not clear whether this is a major or minor or blues scale. Other than (1,3,3b,5) only 2, the A appears. The bass plays (1 3 5) of G maj. An alternate position, that I prefer, for the notes after the rests is the tenth position.

• bar 35   This measure fits well with a strummed B or B7. The bass player plays 1 and 5 of B. The last note in the measure is 3b, but it does not give a minor character to the bar. The guitar enters with 7b and 4 of B. Most following notes are in the G M sc. The 4th and 5th of B are present.

• bar 36   Starts with arpeggiating an Am triad, then walks the bass down.

• bar 39   This software can’t do notation for bends in blues guitar. Both notes in this bend (written as a slur) are struck. The first note is struck as a C and bent quickly to a D. After holding the D, it is lowered quickly and struck as a C. There is standard notation for this extremely common blues licklet.

• bar 41   Hits all of and only G maj scale. This makes an obvious contrast with the blues scale in the previous bar.

• bar 42   The notation for the rhythm here is a bit awkward. The triplet counts for one eighth note. The actual rhythm steals somewhere between a 32nd and a 64th from the eighth note. I wonder if this is what Ribot meant when he talked about Monk’s rubato, I have not listened that closely to Monk.

• bar 43   Gu. enters playing (1,3) of B. Descends chromatically to B and descends here with (1 5+ 4 7) then ascends chromatically to E for the next bar. Bass plays (1,3,5,7b) of B.

• bar 46   Can barre with 3rg f. An alternate fingering is to play g-3 and barre c-4 e-4

Back to posts