Writings and essays about flamenco

Category — Manolo de Huelva’s Views on Flamenco

Flamenco Guitarist Manolo de Huelva on Flamenco – article in Guitar Review by Virginia de Zayas – Part 2

Note by Brook Zern:  In the mid-1970’s in Seville, I contacted Virginia de Zayas, in whose large house the legendary and secretive flamenco guitarist Manolo de Huelva was living.  I knew that her husband Marius had arranged for the historic 1936 Paris recording session that documented the solo art of the great Ramón Montoya, and hoped to somehow obtain recordings or written versions of Manolo de Huelva’s playing.

I was the Flamenco Editor of Guitar Review magazine, an elegant and authoritative publication dedicated primarily to the classical guitar — Andrés Segovia was the Chairman of the Advisory Board and was a regular visitor to the offices, and I arranged for his only interview about flamenco to run in the magazine.

Ms. [somehow, "Madame" seems more appropriate] de Zayas was interested in explaining Manolo de Huelva’s views and telling his stories, and offered to write an article for the magazine.  I agreed, and the long article appeared in three parts spread over time.  This is the second part, from issue 45 dated Spring 1979.  The first and third parts have already appeared in this blog.

At one point in our conversations, I reluctantly corrected Ms. de Zayas on a point of fact, despite her vastly greater knowledge.  She may have known that I was right, but she said something interesting.  ”Mr. Zern, please do not correct me on flamenco matters.  When I talk to you, I am speaking as Manolo de Huelva — that is, I am repeating what he has told me over the years.  That is my value — that I can speak for him.  It really doesn’t matter so much whether what I say is correct or incorrect; what matters is that it is what Manolo de Huelva thinks.”

Well, she had me there.  I actually had to agree — because I desperately wanted to know what he thought, and not what Ms. de Zayas felt at the moment.

And for that same reason, this is an important article and I want to bring it to the attention of the few people — mostly Spanish authorities — who will be interested in these old stories and arcane information, assuming someone is motivated to translate this.

Everyone else is excused — but hey, if you’re serious about flamenco, you just might find something illuminating here.

Remember — this article takes us into the heart of flamenco creation, as witnessed by one of the greatest guitarists in the history of the art.  It comes from a now distant past, where seventh chords were considered far-out and radical innovations.  It is source material for any advanced course in flamenco history and aesthetics.  And while many authorities today cast aspersions on the worth of unprovable and malleable recollections, and while the exact words may not be Manolo’s, it is yet another example of the value of oral history in understanding flamenco, or any subject that involves human beings.

But I never got to hear Manolo de Huelva, or obtain recordings of his playing at its best. Shortly before he died, Mrs. de Zayas invited me to visit in hopes that he might give me something, but it was too late.  He was determined to keep his music from being learned by other players, and regrettably, he succeeded.  He is heard as the accompanist to some noted singers on numerous old 78’s, often reissued in new formats; but as he said, he never revealed the amazing abilities that made him the favorite player of countless great artists and knowledgeable aficionados of his era.  Ms. de Zayas later released a double LP which contained, in addition to excellent material by Ramón Montoya, some cuts that once again failed to reveal Manolo de Huelva’s true genius.

Foiled again.  Here’s the article:


by Virginia de Zayas


Except for alegrías and some bulerías which are played and sung in our major scale with tonic and dominant chords, the song or guitar piece does not modulate to their modes.  The harmony of the flamenco scale is different, primarily because it is a descending, not an ascending, scale.  Therefore, the harmony points are reversed.  The note which leads the ear to expect a progression down to the final E is the F, reinforced by harmonization on that note.  This is the tonal  center of gravity, and is essential for the cadences.  The upper E, the beginning of the scale, is where melodies and falsetas often start.  The traditional simple harmony uses only the notes of the mode, except for the G sharp in the final chord.  All chords, except that on the dominant A, are obliged to be major.  The dominant A chord, of course, is obliged to be minor.

Because we think of the scale as a minor one (counting from the bottom of the octave), these chords seem to produce a delightful clash with the melody, one of the subtle surprises of flamenco music.

The polos are accompanied by the downward progression A G G E, with passages in C, in voice and guitar.  The C is part of the harmonic structure, appearing in passages in the soleares which are called apoláApolá means “as in a polo”, because of this passage in C.  This does not mean that they should follow a polo, as some have thought.  They should not follow a polo because  the essence of flamenco is contrast.  The soleá apolá is a soleá de cierre; to terminate a group of soleares.  There are four of these left.

Viewed from the perspective of a descending scale, the dominant on A is the important note, the fifth down from the beginning of the scale.  The C is the major third and the G is the sixth.  These notes are the structural notes, with the cadence F E.

Soleares are closely related to the polos, but because their harmonic cadence is a little more complicated it may be that soleares accompaniments as they exist today are later than polo accompaniments.  This is in line with the tradition held by old men in 1910 (the polos are older).  Their cadence is A G C F E.  The G and C occur on beats seven and nine, on the rasgueado, which gives a quick jump from G to C.  This is called el cambio (the change).  The soleá continues on the C chord until the end of the twelve beat measure; then, on beat one of the next vuelta (measure), it changes to the F of the cadence.  Singers not accustomed to singing to guitar accompaniment, as were most Gypsies, and who beat out the rhythm on the table with their knuckles (or use their walking stick, when it was still in fashion), do not know that they must signal the chord change on or before beat six.  If they fail to do so, the rule is that the guitar must wait until beat seven of the following measure of twelve beats, regardless of what the singer chooses to do.  Good guitarists refuse to accompany such a singer because the audience would place the blame on the guitarist.

As heard by the casual listener, counting as he or she would from the bottom of the scale, the harmony notes seem to be on the fourth, A, the third, G, and the sixth, C.  This peculiarity of modal harmonization has led many musicians to have the whole thing upside down because of the “mirror music” effect.  The so-called major third in the E scale leads them to feel that the polos and soleares are “sad songs.”  As one taxi driver told me: “Es que no saben” (They don’t understand).  Actually, polos and soleares are serious songs.  The seriousness of flamenco songs has been impressed upon me by Manolo.  The words are serious.  They are pieces of life.  It is true that some speak of unrequited love, others of the contrary.  Here is a typical one:

Aquello cuatro puntale
que mantiene a Triana
San Jacinto y lo Remedio
La O y Seña Sant’Ana.

(Those four pillars/ which hold up Triana/ San Jacinto and Los Remedios/ the O and Lady Saint Anne”)  Remedios and O are different surnames for the Virgin Mary, while Anne is her mother’s name.  There are four parishes of Triana and this song is sung to a rousing melody.  One could almost say it is a patriotic song,  The people of Seville, where Triana is, are very much attached to their parish church and its feast days.

A typical soleares accompaniment would be the one which goes with the above words:  E major, A minor,  F (CI), E (cadence); then follows A minor, G seventh, C, F (CI), E (cadence.  This last part is repeated.  The changes of chord occur at the beginning of each measure of twelve beats, and at the rasgueado on beats seven and nine.  At the opening of the song of the guitar sounds the E chord, and then changes the chord if necessary.  This is because the guitarist never knows what the singer is going to sing, unless especially told.

Most flamencos do not know our names for musical notes or chords, so a singer will tell a guitarist to play, “por el medio,” in the middle (the chord of A major), or he will say, “por arriba”, upwards (the chord of E major).  These expressions refer to positions on the fingerboard.


The G sharp in the final E chord is used in flamenco music in all E chords, which are never minor.   It is found  in medieval polyphonic music, in Mode III, the (medieval) Phrygian.  “The notes proper to the mode were used, save for this introduction of the major third into the final chord.” (Grove 1954: II, 11).  Did the flamenco guitar take this harmonization from polyphonic practice?  It is significant that the chords used to accompany the flamenco scale are all major except for the one minor chord on the dominant A.  As to the history of flamenco chords, apparently the Spanish guitar has been played in chords, rasgueado (strummed) with the fingernails, since time immemorial.  Ramón Menéndez Pidal (in his Poesía Juglaresca y Juglares, pub. By Espasa-Calpe, 1942) wrote that Spanish guitarists, playing rasgueado, went to the court of the Duke of Normandy, France, in the 14th Century.

The E major chord on the final note of the cadence is called the Picardy Third, or Tierce de Picardie, in French.   (Picardy is in northern France.)  Grove tells us:  The name, the origin of which is obscure, is used for the practice of ending a polyphonic composition in a minor key with a major third above a tonic bass in the final chord.”  It is noteworthy that much printed German organ music of the 15th and 16th centuries consisted of Spanish dances (many specimens may be found in Wilhelm Merian’s Der Tanz in den Deutschen Tablaturbüchern, Breitkopf and Hartel: Leipzig 1927).  This suggests that perhaps ecclesiastical practice was derived from the Spanish guitar chord.  The Church would have taken only the single chord of E major, and did not use major chords on the other minor modes, those which had minor thirds when counted from the lowest note, such as D, A and B.  This is a matter which requires thought and investigation.  We might suppose that the Spanish guitar took this chord from the Church and then applied it  to all their harmony, excluding the A chord, but then we overlook the fact that the flamenco scale is not a minor scale like the Phrygian ecclesiastical mode (which also has its dominant on A and its fundamental on E).  The flamenco scale is a major one, counted from the top, and therefore is properly accompanied in major chords.  Besides, all its chords except for the final one employ the notes of the mode.  In the 16th Century tablature books  religious pieces are given for guitar and vihuela.

In polyphonic music the voice sings the major third in the final chord only.  In flamenco, it is th guitar which has this note, not the voice.  The voice sings G natural with the E major cHord except when a G sharp appears as ;an; ornament in a rising phrase.  There is never an E minor chord.


Both Arabian and Andalusian music, as well as other musical systems, used rhythmical periods of a varying number of beats, and it is worthwhile to ask whether the Andalusians derived theirs from the Arabs.  However, this is scarcely possible because the Arabian rhythms are falling ones, as shown in their treatises on music, and twelve-beat periods, for example, will have the accent on the first beat (R. d’Erlanger: La Musique Arabe, Paris, P. Geuthner, 1959, vo. VI pp 67-74).  The polo-soleares rhythm is a rising rhythm with the accent on each third beat.  The principle is different.  [Note from B. Zern:  most flamencos and experts perceive the polo rhythm the same as the soleares rhythm:  a twelve beat pattern with accents usually on the third, sixth, eighth, tenth and twelfth beats.]

Indeed, as Aben Hazem the Cordovan philosopher said, the zejel, or polo, is of Spanish origin.  It would have been used by the Andalusians before the 9th century, date of the early Arab collections of zejeles.

The Andalusians seem to look for the origins of flamenco among the Arabs and Gypsies because thy are the most recent invaders of their land.  They now think that flamenco was the invention of the Gypsies, disregarding the fact that no Gypsies anywhere else have such a polo rhythm, and that all other Gypsies sing or play the music of the country they live in.

Besides the accent on the third of every three beats, the striking thing about the soleares rhythm is the syncopation accenting every two beats at the end of the twelve-beat period, with guitar rasgueado.  This is not a poetical rhythm, it is rhetorical.  It brings each period to a close with a rhythmical cadence.  The Greeks and Romans punctuated their oratory with such closer spaced feet at the ends of sentences, and the Andalusians, who spoke Latin in the Roman period, went to the great rhetorical schools of Baetica in western Andalusia, the home of flamenco (William J. Entwhistle:  The Spanish Language, London: Faber & Faber 1936, p. 75).  The people heard in their theaters and stadiums public speakers trained in these schools.

The polo-soleares rhythm is therefore a native Andalusian invention of well before the 5th century, applied to guitar music, a native technique.


The siguiriyas rhythm was probably ultimately derived from the Latin popular saturnian meter which would have become popular in Spain in Roman times.  An example of this meter is:

prím(a) incédit Céreris     Prosérpina púer.

This line of verse has the first part with three accents and the second part with two accents.

We still find this division in the following seguidilla:

Tóda vá de vérde
La mí galéra
Tóda vá de vérde
de déntro a fuéra.

Francisco de Salinas, writing in 1577 (De Musica Libri Septem, published in Salamanca) on p. 342 gives the following meter, which may be compared with the saturnian meter given above.  Salinas used a staff of three lines instead of five, and this is a C-clef.

Example to be inserted.  The first bar shows three whole notes between the middle and and lower line.  The second bar shows a whole note on the middle line, followed by a half note on the middle line, followed by a whole note on the lower line, followed by a half note between the middle and the lower line.  The third bar, like the first, shows three whole notes between the middle and lower line.  The fourth and final bar shows a whole note on the middle line, followed by a half note on the middle line, followed by a whole note on the lower line.

Salinas calls it “a very ancient and simple melody” (p. 346).  The first foot is a molossus followed by two trochees, that is to say a rhythm of five beats in two bars.  It was a meter used for singing romances or ballads, such as the Conde Claros (also used for variations played on the vihuela) and the Retrayda está la Infanta.

These two ballads had more syllables than the notes of the basic melody, so the third whole-note was divided into two half-notes, giving an extra note where needed,  When this was done, the molossus became an ionic “a majore” (Salinas, p. 346).  Salinas printed the melody and the words separately and I have joined them.

[Example to be inserted]

The difference between the two melodies is that the second one has an extra short syllable where the first one would have had arrest in performance.  The verbal accents would fall in different places according to the poem.  In the many examples given by Salinas, notes could be added to or subtracted from either end of the meter, presumably to fit the number of syllables.

This basic ballad melody has five beats in two bars just as the siguiriyas rhythm has five beats, as the flamencos count it, divided a little differently, as they start on the second beat of the old meter.  The siguiriyas begins with a spondee, followed by trochees and an extra beat.

The siguiriyas rhythm is:  [half note, half note, bar, half note followed by dot, half note followed by dot, bar, half note]

[Note from B. Zern:  This is a logical rendition of the siguiriyas rhythm which is heard by flamencos as five accented beats as follows:  ONE and TWO and THREE and a FOUR and a FIVE and…]

Salinas tells us (p. 342) that a meter which has the first syllable removed is epiploca (a Greek word), and a meter which has a syllable added is called hypercatalecticum (a Latin word) (p. 344).


In Spain, as among Arabs, before rhythmic periods the basis for music was poetic feet, or rhythmic modes.  These were used by the ancient Greeks as well as by  the later Byzantine Greeks, by the Romans, Arabs and Europeans.  They are still found as late as the 13th century in the manuscript of Montpellier, France.  The rhythmic modes are iambs, trochees, spondees, etc., and the music is glued to the poem, a long syllable having a long note, and a short syllable having  a short note.  There was a break in the music corresponding to the break at the end of the poetic line.  These rules prevented the music form having long, independent phrases.

The Arabian zajal in Andalusia originally had two characteristics:  “An accentuation identical to that of melodic rhythms,” and “The use of … popular language [rather than literary]”  (R. d’Erlanger, La Musique Arabe I, 636).  This musical accentuation was an dis “based on the antithesis of strong accents (Dum) and weak accents (Tak) balanced by silences…  This accentuation is freed from the principle of equivalence in quantity (between the note and the syllable) which is the rule in strictly quantitative classical (Arabic) prosody, which knows only the antithesis of long and short syllables.”  The Dum is a beat which sounds lower in pitch, produced by striking the center of the drum.  The Tak sounds higher and is produced at the rim of the drum.  These are the basic Arabian accents, but there are rhythmic drum sounds with other pitches and functions.  D’Erlanger asks:  “Because isn’t it in Andalusia that was born the art of the Tawsih (the literary form) and the Zajal, this new form of poetry which revolutionized the traditional prosody of the Arabs by giving more freedom to prosodic rhythm by the organization of the poem in stanzas where the (single line) verse plays no more than a secondary role?” (I, 141).

Therefore, we see the Arabs first learned to free their music from poetic feet upon listening to Andalusian popular music and the zejel.  I think that Europe too was freed from the bondage of the rhythmic modes when they first heard the Andalusian music which accompanied this poem made to be sung, a poem organized by accents instead of syllables.  The zejel spread all over Europe, even into Germany, Italy, Scotland and Ireland, where there were songs in praise of the Virgin Mary.  There was much travel and communication in the early middle ages.  In the 13th century the Castillian King Alfonso X, the Wise, himself wrote, or collected a book of songs, Las Cantigas de Santa María, Songs to Holy Mary, many of which are in the zejel form,  Note that in the macho of Tobalo’s polo occurs the phrase “Queen of the Heavens!”  (See Part One, GR 45, page 16.)

The Andalusian zejel, therefore, was measured by the number of accents instead of the number of syllables.  It was the first European musical rhythm in poetry, taking the place eventually of the former prosaic rhythm.  This means that the number of accents are counted, as in English poetry (see the poems of Gerard Manley Hopkins), and that there can be more or fewer syllable in a line of verse.  This is akin to our music, in which we count the number of beats and not the length of notes.  This began in the later 11th century, first inspiring the troubadour music of Occitania (Southern France).  The troubadours, too, wrote poems in zejel form; one of the first to write them and compose the music was Guillaume, Duke of Aquitaine.  The rote love songs, descriptive songs, etc., as did the Andalusians.  This was the breakthrough for free, long phrased music in Europe, and it began in Andalusia.

To return to the question of origins, the Dorian-Andalusian scale and the position of its dominant and leading note is the revers of our music.  The ancient Andalusian rhythm is the reverse of our rhythm because in the voice the accent falls basically on every third beat.


The word “guitar” is derived from the Greek “kythara,” as is the shape of the instrument to some degree.  The kythara was a lyre with a sound-box and a fingerboard added to it.  Like the guitar it was flat-backed.  We find guitars represented in Egyptian art of the end of the 2nd millennium B.C. and on Hittite bas-reliefs of the beginning of the 1st millennium B.C.  The Greeks had a tradition attributing the kythara to Asia Minor – the land of the Hittites.  It is very probable that the Greeks brought the guitar with them to Spain, as it was already there when the Arabs and their Berber horsemen invaded the Iberina peninsula in 712 A.D.  (This is by my son Rodrigo de Zayas, who wrote it for the text to go with his LP records to appear in France in 1979.)


Although the flamenco guitar is basically a harmonic-percussion instrument, for introductions and falsetas (variations), punteado (plucked stings) is used with the fingernails.  Traditionally played with the thumb, it is now played by some solo guitarists and others with the classical guitar technique of thumb for the bass notes and fingers on the other strings.  Manolo plays almost everything with his thumb.  He tells me that if one has a precise thumb, and practices, the thumb keeps the rhythm far better and is also steadier for the accents.  It is also the traditional school.

Manolo says, “The contrast between the guitar part and the voice is one of the charms (encanto) of flamenco.”  It may be that centuries ago the Spanish guitar was only played rasgueado and that the introduction of falsetas was influenced by the Arabian lute.  The lute was monadic, with an occasional octave, fifth or fourth, and it was played with a plectrum.   Ziryab, the famous 9th century singer who started his career in Baghdad and ended it in Córdova, played his lute with an eagle’s plume, to the envy of his rivals, for which reason, as well as for the superior thick wood construction of his lute, his colleagues in Baghdad obtained his banishment (R. d”Erlanger, op. cit.).

In the middle ages there were two kinds of guitars played in Castilla, the guitarra castellana (Castilian guitar) which was played rasgueado, and the guitarra morisca (Moorish guitar) which was played punteado.

These punteado guitar parts are variations on a harmonic sequence or set of chords, or rather, the main note of a chord.  Such sequences to be played with variations may be seen in the little book Tratado de Glosas by the Spaniard, Diego Ortíz:  “his famous book of divisions for bass viol and cymbalo, compositions of great beauty, appeared in Rome in 1553 in Spanish and Italian versions” (Grove).

The fingernails are not only used for plucking the strings, but also for the rasgueado (strumming) in its various forms.  The simplest are as follows: there is a single up or down stroke on a beat, and also an up and down stroke on a single beat.  These are often accompanied by the golpe, or nail tap.  The rasgueado proper is played with four c-a-m-i fingernails struck downwards, one after the other on a single beat, followed immediately on the next beat by an up stroke with the index (i), or else, in the cadences, with a down stroke with the thumb.  In polos, soleares and bulerías, it is always played on beats seven, eight, nine and ten.  Soleares and bulerías should begin with six palpis (see below), the traditional rule, and not a confused up and down.

The rasgueado exists to set the rhythm for the listener or performer.  It is also a moment of increased rhythmical tension, and should not be syncopated as is sometimes done now.  The thumb is used for punteado, or for an upstroke, in the termination of a falseta.  The up stroke is called alza púa, púa meaning plectrum.  I have used arrows on the note stems to indicate the direction of the strokes.  The effect of a chord is different if played up or down, because the high notes come first or last.

Just before a cadence, the third beat in the vuelta of twelve beats is often played with a palpi.  The palpi curls the index finger tightly into the thumb hollow, and uncurls it suddenly, striking three or even four or five strings with force, downwards.  If the singer is able to make the proper note and the word stress coincide with this accent, it makes a very flamenco effect, especially if the singer cuts off the sound a brief rest) for a moment after this accent.

The last six beasts of a bulerías vuelta (measure of twelve notes) are played as in soleares, with two rasgueados.  Many modern guitarists in bulerías mistakenly play only the second of these two rasgueados.  Manolo calls this soso (insipid).  This also makes it more difficult for the singer to follow the rhythm.

Andalusians have a preference for periods which can be broken down into triple time.  Even when they begin to dance double time, such as in the farruca, they will soon change to triple time, which seems to offer more scope.

A practiced singer will introduce many slurs and accents, even on sixteenth notes.  In a triplet, all three notes will be slightly accented.

Flamenco guitarists particularly notice the quality of sound given by a player’s nails.  The player uses his nails on the strings and on the tapa (soundboard) of the guitar, on a shield called golpeador (tapping plate).  This shield protects the wood and used to be made of wood or celluloid, now usually of transparent plastic.  Manolo remembers the nails of all the good players he has heard, especially whether they were dull sounding or inaudible, or whether they give a “crystalline” sound.  Although he has never mentioned this to me, his nails have the desired crystalline sound.  Bulerías used to begin with a confused up and down rasgueado before starting the bulerías al golpe, those with two nail taps followed by an up stroke with accent on three strings.  Such bulerías can be heard on most records, the rasgueados coming in the appropriate spot, as in soleares.  It was Manolo who initiated starting to play bulerías al golpe with the nail taps.  No sooner had the record started than the listening guitarists said “él de Huelva’ (he of Huelva)..  It became almost his trademark on the few records he made before the Spanish Civil War.

The guitarists with dull-sounding or inaudible nails would substitute palpis in bulerías where percussive effects are necessary (an example is Javier Molina).

Guitarists with weak and brittle nails avoid palpis and golpes, and even rasgueado as much as possible. Ramón Montoya had very soft nails, so he would also avoid playing for dancing, except in special cases such as when he accompanied La Argentina in alegrías at the Opera Comique in Paris in 1937.  But instead of beginning this dance with the necessary palpis, Montoya would play quick turns, with the chords.  Also, instead of ending soleares with two palpis, as has always been done to lend finality to the piece, Montoya would play an arpeggio downwards (harp-like effect, played fairly slowly).  That is, toward the upper strings.  The lower strings are “upwards” when viewed from the player’s perspective.  This is called the “modern” way to finish off soleares.  It is quite incorrect because two palpis are indispensable.  It is to be hoped that they will be reintroduced.  Montoya was contracted by a record company during the first part of this century, and he accompanied so many singers that other players imitated him, including his weaknesses.  These peculiarities of his, which destroy the very individual rhythmic effects necessary to the flamenco guitar, may have contributed to influence the present all-pervading arpeggio at the end of every vuelta of soleares throughout the accompaniment.

All of these lacks are increased in prominence when a modern player uses a classical guitar which does not, like the flamenco guitar, have an incisive sound.

In polos, Caña, soleares and bulerías, a good player accompanies some chords or notes with a golpe on the tapa.  This is played, in the first of these three toques (guitar parts) on the third beat with the fourth fingernail, when there is no palpi on the chord.  There are also other golpes, those for bulerías being complex.  Such taps are a continuous accompaniment to good flamenco playing, and the guitarist may choose to put in more or fewer.  Nail-breaking rhythmic effects are understandably avoided by those with weak nails.  It is regrettable that Manolo de Huelva did not make more records, for he is an expert at all these flamenco effects.  It would have counterbalanced the overpowering effect of Ramón Montoya on the record market.

This percussive aspect of the flamenco guitar does not suit those who like their music softly played, nor does it suit guitarists with weak and easily broken nails (the latter should use several coats of transparent nail varnish (sold in the United States under the name “Hard as Nails”).  When heard “live”, the guitar does not seem as loud although played with full sound; but nevertheless modern records record the guitar more softly, subordinating it to the voice.  However, it does not sound this way in Andalusia, except in siguiriyas and regional malagueñas (when correctly played) which are rather soft.

The traditional guitar is played with the thumb and with ligados (ligated [tied] notes).  There are two types of ligated notes: the apagado (stopped with the left hand but not sounded with the right hand) and tirados (with the string pulled with the same left finger that stops the note when it first sounds).  Arpeggios can result in a higher note or a note on another string.  Tirados result only on a lower note.  Tirados are played on the first of a group of two notes, or on the first two of a group of three notes.  An apagado must not begin a group.  By a group I mean the notes on one quarter-note beat.  Ligated note passages can rival in speed the plucked note scales of a virtuoso and have far more gracia (charm) and flamenco flavor.  They are also much easier to play and require little practice compared to classical guitar scales.   The sound of a string plucked, apagado, or tirado, is different and produces three shades of sound.  It results in a contrapeado (syncopation) of sounds which displaces the accent and gives light and shade to true flamenco playing which is absent from the modern school of speed and invariably plucked strings.  A whole piece can be played thus with the left hand only, the right hand not striking a note.  I heard the young and very flamenco Gypsy Sabicas play a spectacular solo thus in 1936, in Madrid.  It is hoped that ligated notes will be revived.

Besides the above mentioned ligados, Manolo told me that the traditional guitar often used campanelas (chimes), and he puts many in his falsetas.  The guitarist must hunt for these.  Campangelas are common in 17th century guitar technique.  See Gaspar Sanz (Guitar Review 40).  They are notes of the same pitch on different strings, and are played consecutively.  The sounds are different due to the varying thicknesses of the strings.  One of Manolo’s best falsetas has repeated campanelas.  One can see that the flamenco guitar did not seek to play for frequently defective microphones in a stadium.  Flamenco is an intimate gathering.


In addition to making the word accent and the guitar accent coincide, it is important that the words be understood  In our music, the first syllable of a word may be sung and the second syllable may occur several notes later,  The flamenco rule is that the words are sung in little groups.  All the syllables of a short phrase in a sentence must be sung together.  One cannot vocalize, “Mary – - – - – sang.”  It must be, “Mary sang – - – - -in the garden.”  This is much easier in Andaluz because all the syllables end in vowels (they drop off Castilian final consonants), so the vocalization is done on this final vowel.  The words must fall on more or less adjacent notes and thus be understandable.  The main verbal accent should generally fall on the rhythmic accent, but not always.  For example, I had set words to a soleá, observing the group rule and the accent rule and following the lead of previous words.  Manolo turned the whole thing upside down.  The song had begun with A B and A D on the second  and third beats, the accent falling on the third beat  He increased the notes to five quarter notes (rest, DDDDC) on the first three beats, and all the strong accents of the first two lines of verse – on the first two vueltas of guitar music – were placed on the weak rhythmic spots and were not accentuated.  The whole is martelé, that is, with all the beats and half beats accentuated.  He did this to the first soleá of Triana, a song he must have heard hundreds of times.  Fifty years ago it was sung as the first of a group, and all the singers would sing it, one after the other, with different words, until they had sung it six to ten times.  In one version, where the verse begins with a strong accent on the first syllable, there are two rests in the music, and the song begins on the third beat, on D.


At the end of a group of soleares, the guitarist will play an extra vuelta (twelve beats), or tow, to ensure that the singer does not prolong his final note beyond the tenth beat.  This means that, unless the guitarist knows his singer’s individual sytle, or the singer knows where to stop, the two do not usually end simultaneously.  In the siguiriyas gitanas, written in alternating 3/4 and 6/8 rhythm, the singer will signal the close with a cadence and a corte (break) of two beats, so they can finish together.  Unlike the polos and the soleares, in the siguiriyas, the guitar usually waits for the singer throughout the song, as in our music.  He waits at the end of the 6/8 bar.  The singer does not follow such a steady beat as in the soleares.

Basically the guitar rhythm in the soleares has ten beats followed by two rests.  When the singer continues to sing, without the corte, the rests are filled by the guitar.  When the tercio (the line of verse) ends, particularly the first one in a song, the corte is usually observed, sometimes by the guitar also, and the two rests are counted to make up the twelve beats.  These two rests are a rhythmical ornamentation, similar to those used in other cyclical music, such as Arabian and Hindu music.  Singers who have no guitar accompaniment, and who beat the rhythm on a table with their knuckles, usually leave out these two silent beats.  They are even omitted by some guitarists, for example, the Gypsy guitarist Manitas de Plata, in his performance of soleares at Carnegie Hall, New York 1967.  This is quite wrong.  Accompanying dance is the best master for the beginning flamenco player.

These two rests are also important to observe when the guitarist plays a cierre (cadence) as an introduction for the singer.  This cierre is called marcar la entrada (signal the entrance).  I have been accompanied in soleares by good guitarists, and although I instructed then beforehand, one of them could not remember to make the corte to let me begin.  So I just sat and waited as he continued playing, until finally it dawned on him.  When we did siguiriyas, I asked him to do the usual two rasgueados to show me when he had finished playing (so I would not disturb his part of the performance).  Although he teaches other guitarists, he was so inflexible in his bad habit that he could not remember to play those rasgueados.  Because he did not give me the entrance signal, I entered at the correct place in the rhythm, without receiving his llamada – something I had heard modern singers do.  I sang five or six siguiriyas, continuing to enter with new estilos (songs) without giving him a chance to display his talent.  He finally seized an opportunity and burst out with some falsetas.  My friends smiled.  I tell this anecdote to illustrate the wisdom of the few old rules.  Guitarists who ignore them have only themselves to blame if the singer does not know exactly when he or she is meant to begin.  Nowadays the singer seldom gets the llamada (entrance).

The other guitarist, who had played for dancers, remembered what  I had asked him to do, but he played a syncopated rasgueado for the siguiriyas entrance.  He did not realize that the rasgueado is played to establish the rhythm, not to disturb the singer.  Although I did not say anything to him, it did prove to me that a singer has to know not only what the guitar should do, but also how to describe it in terms the guitarist will understand – something which requires considerable thought,

It is difficult for the flamenco singer to fit the word accents to the guitar accents, so that both fall on a suitable note,  If he knows the song well, and has learned the sequence of all the notes, the singer will run the notes ahead of the rhythm, or hold them back.  If done in moderation, this is considered acceptable.  But because each melody had a composer who placed the notes within the rhythm as he wished them to be sung, a signer should make an effort to find the best possible versions of each estilo (song).  A very gifted singer is excused if he makes an outstandingly good variation on some estilo, as did Antonio Chacón and the Gypsy, Tomás Pabón.  A singer who can ligar los tercios (join the twelve-beat measures without having to stop to take a breath) is very much admired.

Gifted singers will increase the number of notes, singing as many as three to a quarter note in soleares, and occasionally, even four.  In malagueñas, which is a somewhat slower type of song, some singers can fill the whole song with extra notes, singing four to a quarter note, continuously.  Antonio Chacón, for example, did this on a record when he was still fairly young, as did Manuel Centeno on a tape he made for us.  I transcribed these versions and sang them.  They are both the same malagueña by El Canario Chicoo (the Young Canary).  I was more pleased with the original, given to me by Manolo.  It is a beautiful song, with all the feeling expressed by the composer,  True flamenco has few ornaments or extra notes.  Famous singers add far too many notes to display their voices.


A singer will sometimes choose to arrange his words and notes a little differently.  “We do not sing a blueprint each time,” Manolo says.  But the guitar has its rules and the singer should learn them.  The singer must sound the note for the chord change before the guitar can change its harmony.  The clash is a signal to the guitarist to change his chord.  Short clashes are part of the tradition.   They are a kind of harmonic (not rhythmic) appoggiatura.  An absolute rule is that the guitar may change the chord only on certain notes in the rhythm.

An example which is closer to our own music than true flamenco, is the malagueña – a regional song, now part of the “flamenco” repertory.   It is in triple time and is a slow song,  Because the guitarist can only change his chord on the first beat, the singer should change the harmony on the third beat.  The singer can change on the second beat and the guitar on the following first beat.    The singer can even change oon the first beat, but this leaves rather a long clash and the guitar cannot change sooner, as it is already embarked on the first beat.

Polos and soleares change chords on beats one, seven, and nine.  The singer signals the change of harmony of the first six beats of the vuelta (period) of twelve beats, on beats eleven and twelve of the previous vuelta.  The cadences come on beats seven and nine (two rasgueados), so the singer must signal for this on or before beat six.  If the singer makes his change when the guitar is already playing beat seven, the guitar must continue on its course – no matter what the singer chooses to do – until the following vuelta, that is, until it again reaches beat seven.   Star singers with a long fuelle (breath), or with the ability to sing many extra notes with celerity, will lengthen a single vuelta to two vueltas, using the same line of verse.   Occasionally, they will lengthen a single vuelta into three.  The great Gypsy singer, Tomás Pabón, in pre-war records, gives examples of this.

Singers who have la voz dura (an inflexible or difficult voice), are sometimes unable to make their voice obey quick changes without stretching a tercio or verse (one or two vueltas).  Rafael Pareja, who also sang before the war, is an example of this; he was a success because he sang many good songs with much alma (soul)

Formerly, singers woud use few letras (lyrics),  Now they have forgotten so many of the old melodies without  having invented new ones, that, more often than not, they sing the 19th century ones, trying to make up for the narrow scope of the modern repertoire by singing more and more letras to a single melody,

Because it is difficult to fit the words properly to the tune and to the guitar accents – especially because they usually do not understand the principles – they change the notes.  This would seem to be the reason for the present day changes in  the melodies and the lack of observance of the rules; also lack of sufficient study,  The guitarist has to do the best he can for the singer; as a result, there are more and more rehearsals today, something which was formerly unheard of.  Personally, although I may know an estilo (melody) thoroughly, most of the time I am unable to recognize it when the singer is singing.  Manolo, however, with his vast experience, recognizes even the most deformed estilos, and tells me afterwards.

The singer may even insert little runs down to the bass note in places where there is no cierre (cadence), as Tomás Pabón did on one of his records, aided by his vocal facility.  Aurelio de Cádiz also did this in one of his songs.  Manolo used to argue with them about this, but they continued to do so, thus contributing to the breakdown of the rules.  If a singer matches his word accents exactly to the guitar accents, he is said to cantar pegado a la guitarra (sing glued to the guitar).

Nowadays, much of the old tradition has been lost, and singers usually try to start at the beginning of the vuelta, but after that they drift away from it, paying attention only to where they should end.  Singers don’t usually know what the guitar is playing.  Even many good ones before the war did not know that.  This makes listening to flamenco music confusing to those not familiar with it, because they are unable to perceive any clear-cut rhythmic pattern.  Andalusians themselves are confused, and have often said to me (even taxi drivers who listen to flamenco on their car radios): “I cannot explain flamenco, but I can feel it.”

The idea that modern singers drift or “float” on the guitar accompaniment was pointed out to me by Manolo de Huelva, who has been a full time professional accompanist since 1910.  He has always refused to play guitar solos, saying that flamenco music is a “conversation”.;  He has always followed the strict old rules through which a guitarist and a dancer or singer could come together for the first time and go on stage without a rehearsal.  As described above, the signals are primarily rhythmic: the guitar will cortar (cut) for the two beat rest, after a cierre (cadence), to allow the singer to enter.  Or the dancer will give one, two or even three heel taps at the cierre so that the guitarist will know he has finished that section of the dance and wishes to pass to the next section (beats 8, 9, 10).  Or, the singer will say to the guitarist, “en ésta” (in this one), meaning that the next song will be the last of the group.

Before, good singers could martelé and also accentuate along with the guitar, following the Triana tradition, considered to be the oldest.  The influence of four soleares from Alcalá (Seville) began about 1910.  Sung in an even, martelé style, they started a fashion: the word accents of these soleares are usually disregarded; the singers get lost in the rhythm and do not know where to place the word accents.  This has become the modern style of singing, making it even more confusing for the flamenco newcomer.

Over a hundred years ago, one had to cross the river to Triana over a picturesque bridge made of boats strung together, on which planks of wood were placed,  Triana was called “the mother of song,” because the oldest songs were preserved there.  Even the composers’ names are lost to memory.  The  Gypsies lived there until 1941; their forges on San Juan street are now deserted.  They were not professional singers and did not own guitars: several of the best singers lived there and Manolo often went to listen to them and learn their songs,

For centuries, all the bread for Seville was made in Alcalá de Guadaira, and brought on mule back to the city, at dawn.  For this reason, the village was also called Alcalá de los Panaderos (Baker’s Akcalá).  The Gypsy, Joaquín de la Paula, who sang those songs, lived in a cave by the side of the road,  People would drive out to listen to him sing.  The center for flamencos in Seville was the Alameda de Hércules (Hercules’ Promenade).  (See Guitar Review 43 p. 12).  At either end of the Alameda there are two gigantic Roman pillars surmounted by statues.


I want to stress that the two beats at the end of the soleares vuelta (as in the bulerías vuelta, too) are very important to the singer and the guitarist.  Beat eleven should be played with a downward strike, and beat twelve with an upward stroke.  Many modern guitarists in Seville play these last two beats as an arpeggio, which is disturbing to the singer, who no longer knows where he is in the vuelta, especially if he ha trained himself to listen to the guitar.  Of course, he can sing with this intrusion, but it detracts from the singer’s feeling of security, because this arpeggio ends on the highest guitar string, leaving the listener suspended on a single note which calls attention to itself, rather than with a series of solid rhythmical chords.

The only time this final arpeggio is used correctly is at the end of a falseta (variation), never in the accompaniment.  This is a bane which should be abolished; it does not help the singer and it also breaks one of the fundamental flamenco rules, which is that the guitar should not intrude during the song.  Being a triplet followed by a single note, it comes as a surprise in the middle of what should be the rhythmical base of a song.  This arpeggio may partially account for the fact that most modern singers do not sing together or in concert with the rhythm of the guitar.

Manolo tells me that this arpeggio played at the end of every vuelta comes from the increased number f modern solo players, and began to appear about thirty or forty years ago.  A guitar solo does not consist of falsetas alone: it also has interspersed lines of plain chords and rhythms, and these should not end in an arpgeggio.  Because many of these guitarists have not played accompaniments, they have not had to learn the basic rhythm.  They do not know how impotent it is for the singer, or, even, to themselves.


Flamenco songs are divided into three groups: de entrada (to begin), intercalada (intermediary), and de cierre (to close).  If you sing eight songs, particularly soleares and siguiriyas, sing them in that order.  There are very few de entrada songs, but many more of the other two groups.  You alternate the songs from either group.  A principle to follow in choosing songs for a program is contrast,  It is better to group the songs according to their place of origin or their composer.  Sing the low songs first and the ones with higher notes later.  Some people call a siguiriya de cierre, a siguiriya de cambio (change), but this is an error because a siguiriya de cambio is one in our major scale of A: it changes from the flamenco scale to our scale. I have six of these, two sung by Manuel Cagancho himself, a great Trina Gypsy singer.

The curious thing about flamenco music today is that while the guitar music has changed so much, the songs are still the same traditional 19th century songs, performed with very little variation.  Most of the old songs have been lost to the flamenco repertoire.  Consequently, singers will sing a group of four songs of one type, instead of six or more.

One time, a singer who came to my house sang the same estilo eight or more times with different words.  Manolo finally stopped playing when he saw that no more estilos were forthcoming.

End of the second of three parts.  The other parts can be found elsewhere in this blog.

Brook Zern

March 8, 2014   No Comments

Flamenco Guitarist Manolo de Huelva on Flamenco – article in Guitar Review by Virginia de Zayas – Part 1

Note by Brook Zern:  A few weeks ago, I added Part Three of Virginia de Zayas’s very long article to this blog.  It was preceded by my combination introduction, explanation and warning about the article, which I’m inserting again right here:

In the mid-1970’s in Seville, I contacted Virginia de Zayas, in whose large house the legendary and secretive flamenco guitarist Manolo de Huelva was living.  I knew that her husband Marius had arranged for the historic 1936 Paris recording session that documented the solo art of the great Ramón Montoya, and hoped to somehow obtain recordings or written versions of Manolo de Huelva’s playing.

I was the Flamenco Editor of Guitar Review magazine, an elegant and authoritative publication dedicated primarily to the classical guitar — Andrés Segovia was the Chairman of the Advisory Board and was a regular visitor to the offices, and I arranged for his only interview about flamenco to run in the magazine.

Ms. [somehow, "Madame" seems more appropriate] de Zayas was interested in explaining Manolo de Huelva’s views and telling his stories, and offered to write an article for the magazine.  I agreed, and the long article appeared in three parts spread over time.  This is the third and final part, from issue 47 dated Spring 1980.  The others will appear soon in this blog.

At one point in our conversations, I reluctantly corrected Ms. de Zayas on a point of fact, despite her vastly greater knowledge.  She may have known that I was right, but she said something interesting.  ”Mr. Zern, please do not correct me on flamenco matters.  When I talk to you, I am speaking as Manolo de Huelva — that is, I am repeating what he has told me over the years.  That is my value — that I can speak for him.  It really doesn’t matter so much whether what I say is correct or incorrect; what matters is that it is what Manolo de Huelva thinks.”

Well, she had me there.  I actually had to agree — because I desperately wanted to know what he thought, and not what Ms. de Zayas felt at the moment.

And for that same reason, this is an important article and I want to bring it to the attention of the few people — mostly Spanish authorities — who will be interested in these old stories and arcane but often valuable information, assuming someone is motivated to translate this.

Everyone else is excused — but hey, if you’re serious about flamenco, you just might find something illuminating here.

Remember — this article takes us into the heart of flamenco creation, as witnessed by one of the greatest guitarists in the history of the art.  It comes from a now distant past, where seventh chords were considered far-out and radical innovations.  It is source material for any advanced course in flamenco history and aesthetics.  And while many authorities today cast aspersions on the worth of unprovable and malleable recollections, and while the exact words may not be Manolo’s, it is yet another example of the value of oral transmission in understanding flamenco, or any subject that involves human beings.

But I never got to hear Manolo de Huelva, or obtain recordings of his playing at its best. Shortly before he died, Mrs. de Zayas invited me to visit in hopes that he might give me something, but it was too late.  He was determined to keep his music from being learned by other players, and regrettably, he succeeded.  He is heard as the accompanist to some noted singers on numerous old 78’s, often reissued in new formats; but as he said, he never revealed the amazing abilities that made him the favorite player of countless great artists and knowledgeable aficionados of his era.  Ms. de Zayas later released a double LP which contained, in addition to excellent material by Ramón Montoya, some cuts that once again failed to reveal Manolo de Huelva’s true genius.

That was my introduction to Part Three.  For this Part One, let me simply reemphasize that we are entering Mrs. De Zayas’s world, which to a large extent reflects Manolo de Huelva’s views – views formed in the early part of the Twentieth Century.  So when Mrs. De Zayas uses the term “modern school of flamenco”, for example, she is talking about flamenco that was becoming outmoded sixty years ago – well before the radical changes wrought by the great guitarist Paco de Lucia and the great singer Camarón de la Isla changed the art forever, and nearly eradicated the once-immutable sound of flamenco as it had been for several generations.

I happen to believe that flamenco didn’t emerge suddenly as a completely new kind of music – instead, I think some of the key primordial song styles were forged over time, possibly more than a century, within the confines of certain Gypsy families.  If they were not precisely described in books or newspaper reports, that doesn’t mean they did not exist; it is possible that they were not bandied about in public in an era when royal decrees made it legal to kill Gypsies.

But while these songs – notably the few so-called cante jondo or deep song forms – may have a lengthy history, there is no reason to believe Mrs. De Zayas’s romantic notion that anything resembling flamenco existed back when the Romans did indeed bring dancers from Cádiz to the empire’s capital around the time of Christ.

Clearly, some of the views here are those of Mrs. de Zayas – Manolo de Huelva probably didn’t talk about ancient musical modes, or Arabic zejels, or things that may have happened in Roman times.  But it’s not too hard to separate the her views and theories from his ideas and recollections.

The idea that the polo is the oldest flamenco song – as Manolo was told by the oldest flamenco people he knew – was accepted seventy years ago, but is no longer fashionable.  The polo is now viewed as a sort of one-off – not as the progenitor of the solea.  Likewise, the caña (the “mother of the soleá”, as the polo was called “the father of the soleá”), is now considered “autoctonous”, a word that implies little relation to other styles.

Her remarks about the flamenco scale seem very informative.  She focuses on what I consider the most important distinction between flamenco music and our familiar Western music – that it is properly viewed as drawing its power from a scale that descends toward the tonic, instead of one that rises.  This reversal of a scale’s direction and the different aesthetic that results from an incessant “downwardness” is the key to understanding the pull, the gravitas, the attraction that characterizes the entire art.  Our major scale inherently goes up – C D E F G A B C.  Their scale doesn’t use the white notes of the piano from C to shining C — instead, it uses the white notes from E to E.  But it is best conceived as going down — E, D, C, B, A, G, F, E.  The importance of the descending nature of the scale is underlined by the so-called Andalusian Cadence — when E is the tonic note or root note or chord, the descent is A minor, G major (or G7), F major and E major.  (It’s interesting to note that the tonic E chord has a G sharp note within it — though the G sharp is not part of the scale itself, except as an accidental.)

(What about their happy music, like the alegrias (the word itself means happiness)?  Easy, it’s not in the modal scale or natural (Phrygian) mode that defines flamenco, but rather in our own major key.)

Also note, at the end, Manolo de Huelva’s insistence that the most flamenco guitar falsetas are the “más monótonas“.  It seems that the word here doesn’t mean monotonous in the usual English sense, though it’s related; it means that they are mono-tonal; they have a limited range, they aren’t “fancy”, and they don’t use complex chordal conceptions.  This aesthetic may be the one that attracts some people to the music of Diego del Gastor, who also relied on thumb-driven playing.  Many of them seem “all alike”, but the subtle distinctions are a key to their brilliance.

The article is followed by a three-page transcription of the polo sevillano by Virginia de Zayas — a guitar introduction and then the vocal line over the guitar accompaniment chords; I hope to put it onto this website soon.  (For anyone with a guitar, the characteristic “chorus” for the sing is identifiable as a chordal rise and fall: E F G F E).

My interjected comments are in brackets — the parentheses are Mrs. De Zayas’s.

Here is Part One:

Although flamenco is the traditional music of Andalusia, southern Spain, a deep fascination for flamenco – guitar, songs, and dance – has increased the popularity of its rhythms and harmonies throughout the world.  Flamenco was a success as far back as two thousand years ago, when dancers were brought from Cádiz to ancient Rome.

I shall write of the origin of the oldest flamenco songs, the derivation of the flamenco scale, flamenco’s traditional rhythms and their possible sources; in other words, the origin of the modern school of flamenco guitar music.

Polos, Caña, soleares, bulerias, siguiriyas, martinetes, and Cádiz dancing songs [i.e., the alegrías family] are all considered to be pure flamenco.  I use the word flamenco to designate the music and the people who perform it.

The word flamenco means Flemish.  Flanders, now northern Belgium, was the birthplace, in1500, of the future Emperor Charles V, who was also king of Spain.  When the young king arrived in Spain he was accompanied by many Flemings, including Flemish singers from his chapel.  Carlos Almendros has settled the much discussed question of why flamenco singers are called flamenco by finding early 16t century Spanish music in which the words flamenco and first flamenco – placed at the beginning of the staff – means cantor or singer. (Flamenco magazine, July 1975, p. 39.)

While in France, during the Spanish Civil War, I met a number of the best flamenco performers.  I was full of admiration for them, for their artistic perfection and precision.  The one I found most interesting, because his answers to my questions were so clear (perhaps because  he himself has asked questions all his life) was the flamenco guitarist Manuel Gomez, “Manolo de Huelva.”

Donn E. Pohren, author of “The Art of Flamenco”, writes in his book “Lives and Legends of Flamenco” (1964, p. 278):  “How does one begin to talk of the wondrous Manolo de Huelva?  Perhaps by stating that he has quietly, semi-secretly reigned as flamenco’s supreme guitarist for half a century?… Andrés Segovia became so inspired, in fact, that he devoted the major part of a thesis to Manolo de Huelva…When he [Manolo] becomes inspired his playing drives aficionados to near-frenzy, striking the deepest human chords with overwhelmingly direct force.”  Speaking of Manolo’s “blindingly fast and accurate thumb,” Pohren continues, “his manipulations of the compás (rhythms) are fabulous… Manolo’s left hand has been marveled at by Segovia… He [Manolo] is flamenco’s most original and prolific creator.”

Manolo de Huelva was born in 1892 in the ancient mining town of Rio Tinto, near the Atlantic port of Huelva, Andalusia.  Besides a brief apprenticeship with him in 1937-38 and during my intermittent trips to Spain between 1953 and 1958, I have been studying with him continuously since early 1966.  Little by little, I have found Manolo to be a walking encyclopedia who knows not only all the authentic flamenco songs but also, the traditional guitar music.

As with folk traditions everywhere, in Andalusia information is passed on by word of mouth.  Books and articles have been written about flamenco singers or the lyrics of flamenco songs, rather than about flamenco music itself.  My husband and I taped and bought records by the best singers of the 1953-1966 period,  I have transcribed many of these tapes as well as several pre-war records, and have then sung these versions to Manolo’s accompaniment.  But it was only by coming back again and again to details that I discovered the older versions, the ancient tradition.

Traditional rules become clear by studying the details which influence the artist’s performance; details such as whether the guitarist had strong fingernails, or whether the sound made by his nails was clear enough so that he would not have to resort to less flamenco substitutes.  Furthermore, by transcribing guitar accompaniments and then singing the lyrics, I could see how the singer matches the song to the guitar accompaniment.  This was made especially clear to me, because I was working with such a precise player as Manolo.

The Polos

I asked Manolo which are the oldest flamenco songs.  He told me, “The polos are the oldest songs.  All the old men said this when I was young.  I heard it first from the oldest singer I have known, Antonio Silva “El Portugués,” a Spaniard from the province of Seville, though Silva is a Portuguese name.  I met him in Huelva, where I had just finished learning to be a tailor.  My father brought him to the house, and Antonio came with his guitar.  That was when I first heard the polos.

“When I arrived in Seville in 1910, and became a professional guitarist, there were three Sevillian singers from the Triana district, and they sang the polos.  There names were Pepe Villalba, Fernando el Herrero (the blacksmith), and Rafael Pareja – none of them Gypsies.  The people who listened to them would ask for the polos.  Others who sang the polos were Antonio Chacón and Diego Antúnez, a Gypsy singer from Sanlúcar de Barrameda.  By playing for these older men, I learned how to accompany these cantes (songs) with their exacting rhythm.  But after about 1920, the new generation of singers no longer sang the polos: they turned to different cantes.

Origin of the Oldest Songs

What about the origins of the polo?  Although there are no written records about flamenco music before 1830, oral tradition tells us that the three oldest singers whose names are still remembered lived at the end of the 18th century.  We can learn much about the origin of the oldest songs by examining the literary and musical structure of the polos as they exist today and comparing them to the structure of a popular song which existed in Andalusia in the 8th century (and probably even earlier).

As far back as the ninth century, the Arabs in Spain wrote books of songs which, as they acknowledged, had a popular Andalusian origin.  These songs were called zajal in Arabic (zejel in Spanish), a word meaning “to raise the voice” (as in song).  (Baron Rodolph d’Erlanger, La Musique Arabe, 6 vols., Paris, 1936-1959.)  The Arabs wrote their songs in Arabic, while the Andalusians sang theirs in Romance (ancient Spanish), occasionally using Arab words.

The zejel is generally begun with two lines of verse which state the theme or subject of the whole poem.  These two introductory lines are called markaz in Aarabic – and the literal translation of this word into Spanish is poloMarkaz and polo mean center or pivot, as “pole” does in English.

The markaz or polo (also called estribillo in Spanish) was followed by four-line stanzas which comment on the subject as stated in the two-line polo.  These stanzas are called machos (males) by the flamencos; perhaps as one of a pair, at a stage when each polo had one macho.  The markaz or polo was sung by the principal singer with the other singers, and. at times, even members of the audience joined in.  The markaz (polo) was repeated in chorus after every macho stanza.  In the Arabian zejel, the rhyme of the markaz was repeated in the last line of all the stanzas.

Polo is now a flamenco word used to describe the ancient zejel.  Although we do not know when it was first used, the word or its equivalent may be assumed to have existed before the Arabs came to Spain.  The zejel runs throughout early Spanish literature; its “polo” (center, pivot or theme) was usually composed of two lines.  A beautiful example of the two-line Spanish refrain (or estribillo) as it was used in the 16th century has been published by Rodrigo de Zayas [son of Mrs. de Zayas] in Guitar Review 38 (1973), En la Fuente del Rosel (At the Rosebush Fountain).

According to Rodrigo de Zayas, the zejel is still sung in Arabic-speaking countries, and may be heard any day on the Damascus radio. This is not surprising, because since the eighth century Arab singers frequently traveled from Cordoba in Andalusia to Damascus in Syria, to Baghdad in Iraq.  Because the zejel was a poem written to be sung, European musicians soon popularized it as quickly through Europe.

The mixed Spanish and Arab population in Andalusia is reflected in the following excerpt of an Andalusian zejel which combines Romance and Arabic words.  The Romance, or Spanish words, are italicized:

Ya, Mutarnani Salbato,

Tu’n hazin tu’n penato

tara al-yaumaa wastato

Lam taduq fih geir luqema

(Oh, my crazy Salvado, you are sad, you suffer, you will see the day wasted without tasting more than a little.)

One Arabic writer of the 11th century has described this mixture of poplar Arabic and Romance as being: “the aroma of the zejel, its salt, its sugar, and its musk”  The zejel became a literary form in Arabic, and the great 11the century Cordoban philosopher Aben Hazem wrote: “Among the excellent qualities of the Spaniards… was their invention of the zejel. (Ramón Menéndez Pidal)

Although the zejel (and the polo) are originally of Spanish influence, many have been preserved in books by the Arabs since the 9th century.  The influence of native Andalusian music has been underestimated: the native population clung to its own traditions, in spite of the arrival of the Arabs.

There are two polos still known:  the Sevillian polo and the polo of Tobalo.  The songs appeared before the 19th century, although the date is not known,  Following is an example of the way the Sevillian polo is sung when it is performed:

Ere el demonio, romera, que

Que me viene a tentá


No soy el demonio ni el diablo, que

Que soy tu mujé naturá


(You are the demon, traveler, who comes to tempt me.  I am not the demon or the devil, I am your true wife).  This dialogue is the culminating stanza of the Romance del Conde del Sol, the “Ballad of the Count of Sol.”  As in the old English ballads, the word “ballad,” here, has the meaning given in Grove’s Dictionary:  “a piece of narrative verse written in stanzas and occasionally followed by an envoi or moral.”

The above romance or ballad was printed in 1847 by S. Estébanez Calderón.  It tells of the Count going off to fight in the war with Portugal, leaving his very young wife with her father.  After fifteen years, she follows him and finds him on the day he is to marry another woman.  She puts on her velvet gown and goes forth to ask him for alms.  Needless to say, the ballad ends happily for her.

Estébanez Calderón wrote his article about a night in Triana, Seville, spent listening to flamenco music.  He printed the entire ballad and changed the above stanza to read as follows:

Sois aparición, romera,

Que venisme a conturbar?

No soy aparición, Conde,

Que soy tu esposa leal.

(Are you a spirit, traveler, who has come to disturb me?  I am not a spirit, Count, I am your loyal spouse.)

The words “spirit”, “disturb” and “spouse” (aparición, conturbar,  esposa), are probably substitutions, following the 19th century custom of making popular texts more literary.

The polo of Tobalo has several different letras (verses).  The following polo was written by Rafael Pareja, who sang it for my husband and me.  Pareja was a good folk poet, as well as being a singer.

En er queré no hay sabé, que

Compañera mía, lo tengo experimentao


De lo que siempre he juío, que

Compañera mía, un Devé me ha castigao,


(In love there is no wisdom, Wife, I know it from my experience, as I have always thought, Wife, God has punished me).  The fact that the words “que” (that) and “compañera mía” (wife, companion) are obligatory, indicates the traditional origin of this polo.  Devé, or, more correctly, Devel, is the Gypsy word for God, and is close to the ancient Sanskrit word.  The above spelling is taken from a book about Gypsies, written by a college-educated Gypsy, Juan de Diós Ramírez Heredia.  Gypsies who sing flamenco originally came from part of Pakistan, and have a language of their own, akin to Sanskrit, called caló.

The only macho left to us is one belonging to the polo of Tobalo.  As has been said above, the macho must use the main idea expressed in its polo.  For example, a polo whose theme is “God has punished me,” would have the following words for its corresponding macho:

Harza y viva Ronda!

Reina de los cielo,

Un Devé A…un Devé

Me ha castigao…

(Hail and hurrah for Ronda!  Queen of the Heavens, God, A… God has punished me.)  The praise of the Queen of the Heavens (the Virgin Mary) reminds us that the zejel was used for songs written to her during the 12th century, in Castillian, and even in Scottish.  The long vowels are sung to the notes of the modal cadence.  The occasional Gypsy words are, again, a reminder that Arabic and ancient Spanish words were mixed in the same verse,

Today, most people think there is only one polo, because modern singers have not had the opportunity to hear both polos sung correctly.  There is even a record on which parts of the two polos have been combined.  Was this a Gypsy joke originally?

Manolo tells me that the old singers had never heard the polo choruses sung by anyone except the principal singer, singing alone. Yet there must have been a time, long ago, when other singers joined in, because this is the way the choruses were sung in another ancient song related to the polos and still remembered today: the Cante de la Caña.

The Cante de la Caña

El Cante de la Caña (the Song of the Cane) is so called because a cane was used to mark the rhythm of the song.  The people who sang this song usually did not have guitars.  To beat the rhythm they would use a short length of dried bamboo, especially prepared.  The soft center part would be cleaned out and the hollow cane split about two-thirds of its length.  I have seen the tribal Filipinos in the mountains of Luzón use a similar instrument.  Many years ago they made a gift of one to me, and I noticed that the split is not a mere slit:  it must be a little wider so it will give a good sound swhen it is struck against the palm of the hand.

I once saw two performers in a flamenco night club who had heard of the bamboo but evidently did not know how to use it.  They had a piece of green bamboo (still uncured), which was much larger than a hand instrument.  They had it placed vertically, fixed on a stand, and they twirled it with a piece of string or wire – making a humming sound!

Old men told Manolo de Huelva that the correct way to use the cane or bamboo is to strike it against the palm of the hand to keep the rhythm.  These old men also said that they knew the chorus of the Caña had formerly been sung by everyone present – but they themselves had never heard it performed that way.  However, we have verification of the oral tradition which tells of the chorus of the Caña being sung by the whold group in the article by the journalist S. Estébanez Calderón, who heard it performed that way.  He says that the chorus of the Caña was sung by all the singers, to a guitar accompaniment.  He does not mention the bamboo – probably because there was a guitar present, and no canes were used.

The structure of the ancient Caña is a little more complex than that of the polos.  The Caña begins with a melodic phrase sung to the exclamation Ay! – which sets the mode (scale) in which the song is to be sung.  Following the Triana tradition, this is followed by a chorus sung on the vowel “A”, and then the polo, or main theme, is begun.  This structure seems to date back to the time whent the audience ceased to join in the singing of the polo itself, and began, instead, to sing a cadence using only a single vowel sound sung to chords.

A modern variation has developed in which the exclamation Ay! Is sung in the chorus, instead of the vowel “A.”  This custom was begun by Curro Dulce, a Gypsy singer from Cádiz, from whom Ignacio Espleleta, himself a Gypsy from Cádiz, learned it.  La Argentinita learned it from him and popularized it.  It is not correct.

The Caña has no macho.  The famous flamenco singer Antonio Chacón (1869-1929), known especially for his singing of malagueñas (songs from Málaga), used to sing the Cante de la Caña, adding to it at the end, the macho belonging to the polo of Tobalo.  But this was just his own idea, as he admitted to Manolo, who accompanied him in performance.  He did follow the ancient tradition of repeating the subject of the song in the macho.

As far as we know, the polos and the Cantes de la Caña were not danced, because both either discuss a subject or tell a story.  The famous dancer La Argentinita was the first to dance the Caña, followed by here sister, Pilar López.  The Caña has now become a “mummified” relic for night clubs and the stage.  Recently I saw the famous dancer, Antonio [Soler] dance the Caña, turning it into a dramatic and humorous display intended to make people laugh.  For his act, the singer sang only part of the song, to identify it to the public.

The Caña should now be ripe for a serious revival.

Flamenco and the Arabs

In a discussion of the flamenco scale, how this scale is harmonized, and the oldest flamenco rhythms, it will be necessary to refer to Arabian music (since the Arabs occupied Andalusia for eight centuries), as well as to regional songs and dances.

Both flamenco and Arabian music are not folk, but art music.  Andalusians generally sing regional songs, such a fandangos, sevillanas, verdiales (from the mountains above Málaga), and granadinas (from Granada), not flamenco.  In ancient times what is now known as Andalusia was the most civilized region of Spain and must have had its art music.  (See the writings of the Greek geographer Strabo, 2nd century A.D.)  Something of this art music was bound to remain in flamenco music; its scale and rhythmical accentuation did survive in flamenco.

Accentuation of regional songs is the exact opposite of flamenco, although the same scale is used in both.  Flamenco is accentuated on the third beat of a group of notes, while the regional songs are accentuated on the first beat of a group of notes.  Our own music and most Arabian music are both accentuated in a manner similar to that of Andalusian regional music, suggesting that the rhythm of Andalusian regional songs may have been derived from Arabian music, or from music from other regions is Spain.  This is why most Andalusians, when they clap (palmas) for bulerías, start on the wrong beat.

Since flamenco became a public spectacle about one hundred years ago, malagueñas , tarantas, etc., began to enter its repertory, but these additions do not obey the rules of pure flamenco.

It has been suggested that flamenco music was brought to Andalusia with the Arab invasion of 711.  Certain similarities such as rhythmic cadences may have been derived by both Andalusians and Arabs from as far a way as India.  However, rhythmical guitar effects, which (combined with characteristic harmony) constitute the basis of flamenco playing, are more likely to have been developed by guitar players, that is, by Andalusians.  The guitar is basically strummed in chords which give the meter and harmony..  The Arabian lute, on the other hand, is played in single notes with a plectrum (pick), with an occasional octave, fifth or fourth.  It is significant that Spaniards did not adopt the pear-shaped lute [oud] which the Arabs always preferred.  Spaniards continued to cling to their flat-backed guitar, with its own technique.

When we speak of the influence of Arabian music we must remember that before the spread of Islam the Arabs lived in cities like Mecca and Medina as well as in the desert.  Later they conqueres many lands and were brought under the influence of the (late Greek) Byzantine music of Syria, as well as the music of Persia, brought through Iraq, and containing even influences from India.  Such influences could have reached Andalusia through the trade routes, before the Arabs spread out of Arabia.  Thus, we must turn to a period earlier than the Arab invasion of Spain and speak of the principal ancient Greek scale, which is the basis of flamenco.

The most fascinating thing about flamenco is the strange scale (mode) with its cadences.  The combination of this scale with our major chords on the guitar produces an unusual clash, because on the beginning chord of a song and to end a cadence, the scale may have a natural note in the voice, while the guitar has a sharp.  This is combined with equally strange rhythms and accentuation, with percussive effects on the guitar.

The Flamenco Scale and the Greek Dorian Scale

Here is a comparison:  Our own scale is a rising one: C D E F G A B C, with the leading note being B, and the important note being G (the fifth above the tonic).    To make the comparison easier, let us transpose this C scale to the key of  E major:  E F# G# A B C# D# E.  Here the leading note is D# rising to the E at the end of the scale.  The important note is B, the dominant, the fifth above the tonic.

The flamenco scale is the opposite: it is a descending one, beginning with the upper E.  The notes are, in descending order, E D C B A G F E.  The leading note is F, which leads down to the E at the bottom.  The important note of this modal scale is A.  The note is a fifth below the top E, contrasting with our own dominant, B, a fifth above the tonic.

This flamenco scale is the same as the ancient Greek Dorian scale, the principal Greek one.  In fact, our scale and the flamenco Dorian scale are as a mirror, in which the functions of the notes are reversed.  Our scale leads upwards, and theirs leads downwards.  The dominant note of the Dorian scale, the A, was called mese (the middle note) by the Greeks.  In the Dorian, as well as the flamenco scale, there is a B flat which is sometimes introduced, producing a heart-rending effect which makes our musicians think the music has modulated.  This is not so, as the guitar proves by continuing with its same harmony.  This B flat was recognized by the ancient Greeks and was introduced into a song in a group of four notes called a “tetrachord symnanon”,.  It is interesting to note that Manolo de Huelva tunes his second string, the B, slightly lower than do classical guitarists.  (This observation was made by Rodrigo de Zayas.)

The practical result for flamenco listeners is that we must learn to think of the music as leaning or falling downwards, instead of reaching upwards.  What seems like a minor scale (the Dorian), with a minor third when counted upwards from the lowest note of the octave, is really a major scale with a major scale when counted downwards from the highest note of the octave.  We may view this scale as minor and sad; but the Greeks, in their treatises, said that theirs was a happy music – just as we may say that our major scale is a cheerful scale.  We must change our thoughts about flamenco music: it runs the gamut from sad to serious to joyous, musically speaking.

The ancient Greek descending scale changed direction during the later Byzantine Greek period with the result that Byzantine scales are rising, beginning on the lower G of a two octave extension.  As the Arabs adopted this rising scale when they came in contact with Byzantine civilization (Baron Rodolphe d’Erlanger, op. cit), they cannot have influenced the Andalusian scales and melodies, which take the opposite direction, even though Arabic philosophers bases their musical treatises on the ancient Greek ones.  Early European ecclesiastical scales are also rising, as given by Boethius.

There are many discussions about the practical application of this ancient Greek scale, but in flamenco music we find the same scale, with its downward pull, combined with our major chords with their (to us) upward pull on the guitar.  Flamenco melodies are mostly in conjunct motion.  Singers will fill in the spaces.  The Greek melodies left to us have larger intervals, but these may well have been filled in with ornaments, just as Italian 16th and 17th century melodies were.  This filling in was left to the singers.

The Oldest Flamenco Rhythm

Strange and exciting rhythms are found in flamenco, rhythms which stir up foreigners as well as many Spaniards.  The oldest flamenco rhythm is that of the group which includes polo, Caña, soleares, Cádiz dancing songs [i.e., the alegrías and several related major-key songs], and bulerías.  What makes this rhythm strange to us it that it is “martelé”, French for “hammered.”  Every beat and often half-beats are accentuated, especially in the cadences where the eighth-note accents are multiplied.  Arabian and other oriental music is martelé, as was European music until harmony and bass notes developed.  Then the harmonic accent was placed on the first of any rhythmic group of notes, so the harmonic and rhythmic accents coincided.  The first note of any group came to have  a stronger accent, which is how we usually play music today.

When the guitar plays for this polo-soleares rhythm it continues straight ahead as if for dancing.  Indeed, it was traditionally danced.  The guitar does not wait for the singer; he must fit the melody and accents to the words of the accompaniment.  The measure, also called the rhythmic period or rhythmic cycle [or compás], is a long one compared tour usual shorter measure: it has twelve quarter-note beats.  Orchestral players count our measure of 12/8 in groups of three beats, accentuating slightly on the first of each group of three.  Arabian music also accentuates the first one of almost any group of notes (see examples in R. d”Erlanger, op. cit.).

The polo-soleares rhythm  is composed of twelve quarter-note beats, accentuated thus:  one two three four five six seven eight nine ten eleven twelve.  This rhythm is just the contrary of ours (accentuated on the first [of each three].  Again, it is like a mirror with everything reversed.  The verbal accent should generally be placed on these strong beats.  However, the accent on beat nine shifts back to beat eight, probably under the influence of the guitar accentuation, but also to be nearer to the last syllable on beat six (the syllables of a word must not be too far apart).  If the poem [verse] has too many accents or if it sounds awkwardly fitted to the melody under the usual rules, the verbal accent is placed on a weak beat or on the half beats so as not to disturb the rhythm.

[Note:  Guitarists and dancers who learn to count the accentuated beats in a given compás or 12-beat unit almost always stress the third, sixth, eighth, tenth and twelfth beats.  Mrs. de Zayas seems to be thinking in terms of the singer’s concept of the rhythm, and then theorizes that the stress on eight rather than nine results from the guitarist’s influence.  But nowhere does her concept leave room for the stress on beat ten; so from a guitar or dance standpoint at least, it doesn’t make sense.]

The ending of a tercio, or line of verse, on two longer notes (for example after several eighth-notes two quarter-notes will be sung if ther beats are quarter-notes as they are in flamenco) must be an old Spanish tradition.  In 16th century Spanish villancicos [Christmas songs] the verse ends frequently on two half-notes.  This ending is even found in Italy in that period.  The two longer notes, preceded by notes of smaller value, mark a kind of rhythmic cadence.

I may be asked how it is that I know that the measure begins on beat one, whether the first two notes before the accent are not an anacrusis, and why I do not place conventional bar lines before the third beat as in our music.  My answer 1) because this is what the flamencos say, 2) because I know from singing these songs since 1937, 3) because of the syncopations, 4) finally, because this is the way the long measures, rhythmic periods or cycles, are written for oriental music.  I place the bar lines for our convenience (as I might for Palestrina’s music) and they do not imply an accent.  There are dotted bars for this rhythm.  I place an ordinary bar to mark the end of each rhythmic period of twelve beats.  This is the way the flamencos think their music and the way in which the dance steps are arranged.

To perform flamenco music one must learn to feel and think in periods of twelve beats, with their proper accentuation, and not in short measures of two or three beats, as is generally the case in our music.  In bulerías there are passages of six beats, especially desplantes (passages of syncopation).  The guitarist Pedro Elías told me that he has heard measures of three beats inserted,  but I have never seen Manolo do this, and he is a stickler for rules.

When the guitarist plays falsetas, the ligated [hammered on or pulled off by the left hand alone] give light and shade of three strengths.  This gives an effect of syncopation which traditional flamenco elaborated on purpose.  Wen the scale-like passages are all pulsated [played by plucking the strings with the right hand, or picado] these values disappear completely.  All gracia (grace or wit) is lost, and everything takes on a mechanical sound, full of speed to show that the modern flamenco guitarist can play as well as the best classical guitarist.  He especially shows that he can play exercises with the utmost velocity.  He loses sight of the fact that exercise are for the studio, not for the stage. The rejection of true flamenco explains why so many players like well-exercised race horses reach top billing.  They astonish us at the iron willpower shown.  There will always be an audience ready to be astonished, but the few perceptive listeners may have little opportunity to hear true flamenco if it is never played (listen to the excellent Sabicas).

Many passages are evenly stressed, the highest or lowest note standing out.  The tenth beat is particularly accentuated because this is where the base not falls.  It is followed by an arpeggio to the third, second, or first string, in falsetas (variations).

[Note:  Indeed, most compases resolve to the tonic note not on the final or twelfth beat, but on beat ten, with beats eleven and twelve holding that thought by arpeggiating that tonic chord.  Note that for Mrs. de Zayas (and therefore also for Manolo de Huelva), this arpeggiating should not be done following a strummed or rasgueado cycle – instead the tonic chord should be strummed.  Few guitarists insist on that convention today.]

In accompaniments, the guitar particularly accentuates beat three, which is the pivot note on which voice and guitar should coincide.  [Note: In our Western music, the voice and guitar would normally coincide throughout.  It is a baffling peculiarity of flamenco that while the guitarist maintains a steady beat and strong chords, the vocal line frequently seems to become unglued from the strict guitar rhythm – except at certain points, notably including beat three as Mrs. de Zayas states.]  Beats seven, eight, nine and ten are evenly stressed in the rasgueado and balance the accent on beat three, thus providing equilibrium to the rhythm.  Manolo never accentuates beat six, but I noticed that the late Diego del Gastor did.  However, Manolo accentuates beat six as well as the rasgueado when he plays alegrías.  [Note: It may be unusual for a guitarist today  to differentiate between the alegrías and the soleares/caña/polo family since they are often viewed as sharing the same rhythmic pattern.]  Heavy accentuation ir required when playing for dancers.  After listening to Manolo play soleares for me over the last ten years, I would almost venture to say that he puts the four rasgueado stresses in the stellar spot, relating them to beat three.

Beat twelve, while receiving the word accent in this rhythm, is distinguished from beat eleven.  On beat eleven the direction of the stroke is down, toward the upper strings while on beat twelve it is up, towards the lower strings.

A difference may be heard in sound, not in intensity.  Beat twelve is also a principal one of the bulerías, the beat at which the first strong word accdent of the beginning a a bulerías falls; the singer begins the bulerias on beat twelve.

I have heard young boys at their games such as hide-and-seek or hop-scotch in the streets count as follows: one two three four five six seven eight nine ten.  If the count were to be extended to twelve, ten would not be accentuated .  This rhythm has been ingrained in the Sevillian people for centuries.

As can be noted in his records, Tomás Pabón would begin his soleares on the third beat, although of course he knew better than to do this.  Perhaps he was influenced by the siguiriyas gitanas in which the song usually begins on this beat.  Perhaps he thought that this manner of singing soleares was “more Gypsy.”  (Pabón often spelled Pavón.)

Rests are an important part of rhythmic music and should not be forgotten, as they frequently are.  Much use is made of rests in the cycles of Arabian and Hindu music  As in flamenco, half-beats and even quarter-beats are often accentuated.  What is the rasgueado but four equally stressed beasts struck with four fingers down on a quarter-note beat followed by a quarter-note up beat played with a single finger up?  Such rests and stresses are an integral part of the rhythm and sometimes a rhythmic ornament.  In song, the accents are multiplied at the cadences, falling on eighth-notes and ending with two quarter-note stresses.  This seems to be the pattern which soleares follow.  Manolo pointed this out to me, so he is conscious of it..

It is the accent on every beat (martelé) which makes the music sound flamenco.  In the quarter-note passages of soleares, every beat is strongly stressed and must be sung with accuracy and firmness, showing a command of the rhythm.  I am now notating the alegrías of [the great Jerez guitarist] Javier Molina, as played by Manolo de Huelva.  They are in G major, with some falsetas in G minor.  Manolo stresses all the notes between the beats, and some passages are like trumpet calls.  Manolo, who has not specialized in dance music, learned this alegrias of Javier when he was  young, captivated by its beauty.  He played it in the film danced by La Argentinita, screened a number of times at New York’s  Metropolitan Museum and at the Museum of Modern Art in the fifties.  The film was initiated, artistically directed and produced by my husband, Marius de Zayas.  I cut the film because the person whom we engaged to do the cutting could not understand the music.  I had to go to Joinville, near Paris, to the cutting room to go over the film in great detail with the girl who cut negatives, impressing upon her the importance of observing my marks.  It was essential to have the steps of the foot coincided with the musical beat.  The alegrías is the most difficult dance, for man or woman.  I cut this dance once at ordinary speed and then repeated it  with portions in slow motion.  I was able to do this because of the discipline of the periods of twelve beats and the fact that slow motion was four times slower than normal speed.  The film was made in the spring of 1938 and a little later Ginger Rogers and Fred Astaire made a film in which they danced dream-like passages in slow motion, but for which the music was specially composed.  We had intended to enter our film in the Cannes festival of 1939, but events [probably WWII] overtook us.

Accidental Notes and Transpositions

In monodic phrases, both in guitar and voice, besides the aforementioned B flat, other accidental notes sometimes occur.  The G, c, and D are often sharp when the phrase is about to rise, becoming natural when it descends.  This is also found in the music of the 16th and 17th centuries in Europe and in the singer’s oral tradition.  The flamencos call thes semi-tones “tonos menores” (minor tones, as opposed to whole or major tones).  These chromatically altered notes may be an influence from our music.  I have noted down an older form of siguiriyas without these sharps and a version with the sharps inserted by Manuel Torre, another great Gypsy singer.  Tomás Pabón was also partial to these sharps.  However, we must not overlook the possibility the the sharps are remnants of the chromatic mode used by the Greeks.

The flamenco guitar is played, with few exceptions, either in E or A.  All the falsetas of Paquirri, the great source of traditional guitar music, were transposed into either key.  The siguiriyas gitanas and the chuflas (bulerías) as well as the cantiñas (alegrías) , and other songs for dancing, are played traditionally in A.  The A major chord sounds its bass note that is the fifthe string, in the rasgureado.  The E chord does not sound its bass note, or sixth string, in the rasgueado, because the thumb is held against that string.  Therefore the guitar sounds quite different in these two tonalities.

The chuflas (bulerías) are rarely played in E because the fingerboard of the guitar does not give much field for the harmonico-rhythmical effects peculiar to that toque (type of rhythm).  Voice and guitar in some bulerías are in our major or minor keys.

The cantiñas (alegrías) are usually in our A major key, accompanied by our dominant and tonic chords.  The change of chords occurs on the seventh beat of a measure of twelve beats (this will be discussed more fully later).  Javier Molina accompanied them in G.  These cantiñas are said to have been brought to Cádiz by Spanish soldiers from the north who came to fight Napoleon’s invasion.  He did not take Cádiz, which is on a narrow peninsula.  These soldiers would have brought the jota of Navarra or Aragón.  The people of Cádiz put these chords to their own polo rhythm and at first these songs were called “jota of Cádiz.”  I have one very old melody which is reminiscent of the jota, but most of the older ones are lost.  People now sing the cantiñas of Enrique el Mellizo.

The Ancient Guitar and Chord Inversions

This brings us to the tuning of the flamenco guitar and its effect on chord inversion.  Rodrigo de Zayas has described guitar tuning in a historical perspective to be published in a future issue of Guitar Review.  I shall confine myself to a supposition of the popular guitar some centuries ago.  To begin with, in the modern guitar the lowest of the six strings gives a bass note to the cadences in E, which ar played with the thumb.  The polo and soleares are played in that key.  The low E string does not always sound because the right thumb is held against the sixth string [i.e., it rests on the sixth string] in rasgueado accompaniment.  It is loosely held and thus thumb and string part company slightly at times.  All six strings sound only in final cadences.  Therefore, the chords are usually inverted, unless an occasional one should be based on E on the fourth string.  The inversion of chords and frequent lack of a bass note is strange until one becomes accustomed to it, but it gives a characteristic flamenco effect and especially suits “mirror music.”  (Also, chords are often incomplete and these are especially appreciated.)

On an ancient guitar with the lower string A, if it were played in A, the above would be true, but if it were played in E the effect would be even stranger.  Inverted chords, lacking a bass note, were probably characteristic of the popular guitar since ancient times.  One thinks of Rodrigo de Zayas’ description (in Guitar Review 40) of some of Gaspar Sanz’ chords.  Ambiguity alternating and contrasting with clear cadences is the basis of the flamenco guitar.

On Manolo de Huelva’s Playing

If the ancient flamenco guitar had the first as well as the sixth course of strings missing and remained with only the four inner strings, it might possibly account for the way Manolo de Huelva plays almost exclusively on the four center strings of his modern guitar.  The first and sixth strings are rarely used except in the cadences of falsetas and sometimes on the finals of falsetas and, of course, when he uses the bass or treble strings in a falseta intentionally and not mechanically.  He is more inclined to the bass strings, and frequently composes very flamenco repetitious phrases.  Even his falsetas often end on the third or second string rather than the first.  Manolo thinks that playing on the four middle strings is “more flamenco.”  Because he constantly uses the traditional flamenco technique of playing with the thumb, this concentration of playing on these four strings requires a precise and practiced thumb.  In the rasgueados and accompaniments, he stops the high E string from sounding by holding his right fourth finger against it and on the soundboard.  This steadies his hand and permits greater rhythmic exactness.  His four fingers are held straight, with his nails against the tapa (guitar surface), where he has almost made the beginnings of a hole in his practice guitar because his fingers are so much longer than the wooden or plastic guard made for shorter fingers.

Sevillians for whom I have played one of Manolo’s tapes of bulerías notice his general avoidance of the first high string unless he especially wishes it.  They know little of flamenco, but they missed the constant modern sounding of the first string, the high E.

As a youngster, Manolo began by learing traditional flameanco as well as some eighteen classical pieces.  He had speed and, as he is a composer too, played in the style of the modern school.  But he gradually turned more and more to developing the traditional school, with thumb and ligados (ligated notes), leaving aside the contributions from the classical guitar, the arpeggios, scales, and tremolos.  He has a lightning thumb, very long, with a strong nail, and bent backwards.  He said he would not take a million pesetas for his thumb.  He must have practiced a lot to develop such a thumb.  He is particularly outstanding in the siguiriyas, with curious dissonant chords.  In bulerías he has complete control of the very difficult rhythm, upon which he improvises rhythmically and melodically with tremendous variety and inspiration.  Due to his dislike of making records—and because his falsetas are purposely very difficult, beyond the ability of most flamenco guitarists – he has not had much influence on the modern school.

Although it is possible that this technique of the thumb resting on the lowest string was not used in the older flamenco guitar, when we remember that the flamenco guitar is more than anything percussive and rhythmic, it follows that the thumb-resting technique has as its primary purpose the development of steadiness in the rhythm.  One can still see today poor guitarists, accompanying themselves in fandangos and sevillanas (songs of Seville), who use a free sweeping strumming.  This is easier than subjecting the thumb.  Manolo de Huelva says that the rhythm is very much steadier with the thumb resting on the lowest string, and steady rhythm is the most important thing for a flamenco guitarist.   In any event, the thumb technique was used by Paquirri at the end of the 18th century.  Manolo always keeps the rhythm going.  Even if he wants to repeat when practicing a new falseta, he does not simply break off, but ends the vuelta so as to make the periods or cycles continuous.  Then he begins again (these vueltas, or measures, are twelve beats in the soleares).

Manolo thinks ahead as he plays or, perhaps, his hands act with muscular continuity, accustomed as they are to the sequences of positions on the finger board.  His left hand forms the next choir almost before he has left the last one.  He plays with fully formed chords even when only a couple of its notes are pulsated.  When I write down music, he often tells me: “this is the chord,” and I must explain that in our musical notation we only write the notes actually sounded.  Of course, this does not apply very much in his lightning punteado (plucked string) passages [i.e., picado passages].  From his early playing of classical guitar he has retained command of the chords of the upper reaches of the fingerboard, and he is quick to transpose or find the same chord in alternative positions.  He also accentuates strongly, using these accentuated notes upon which to pivot the sequence.  He will tell me: “These accentuated notes, whether in the song or the guitar part, are what gives sense to the phrase… Accents make the music come alive.”  This is true in singing as well, and gradually he has made me accentuate songs more and more strongly.

Manolo also has a showman’s technique of making the end of a phrase very much softer before beginning a brilliant falseta and then attacking it strongly.  Thus for such reasons I imagine that one could break down his playing into sections, divided rather than in a continuous flow as some Wagnerian opera.  Continuous flow is given to flamenco until the end is reached.  One might call it essentially a rhythmical rather than a harmonic music, in this sense.

Cuando má monótonos son las falsetas, más flamencos son,” says Manolo (the most monotonous falsetas are the most flamenco).  “Many times they are almost the same notes but with different values.  These are the enchantments of the flamenco guitar.  This is why musicians have gathered together with flamenco guitarists to find themes for their music.”

To be continued.

End of Part One.  Part Three is already on this blog, and Part Two will be added in the near future.

February 7, 2013   No Comments

Flamenco Guitarist Manolo de Huelva on Flamenco – article in Guitar Review by Virginia de Zayas – Part 3

Note by Brook Zern:  In the mid-1970′s in Seville, I contacted Virginia de Zayas, in whose large house the legendary and secretive flamenco guitarist Manolo de Huelva was living.  I knew that her husband Marius had arranged for the historic 1936 Paris recording session that documented the solo art of the great Ramón Montoya, and hoped to somehow obtain recordings or written versions of Manolo de Huelva’s playing.

I was the Flamenco Editor of Guitar Review magazine, an elegant and authoritative publication dedicated primarily to the classical guitar — Andrés Segovia was the Chairman of the Advisory Board and was a regular visitor to the offices, and I arranged for his only interview about flamenco to run in the magazine.

Ms. [somehow, "Madame" seems more appropriate] de Zayas was interested in explaining Manolo de Huelva’s views and telling his stories, and offered to write an article for the magazine.  I agreed, and the long article appeared in three parts spread over time.  This is the third and final part, from issue 47 dated Spring 1980.  The others will appear soon in this blog.

At one point in our conversations, I reluctantly corrected Ms. de Zayas on a point of fact, despite her vastly greater knowledge.  She may have known that I was right, but she said something interesting.  ”Mr. Zern, please do not correct me on flamenco matters.  When I talk to you, I am speaking as Manolo de Huelva — that is, I am repeating what he has told me over the years.  That is my value — that I can speak for him.  It really doesn’t matter so much whether what I say is correct or incorrect; what matters is that it is what Manolo de Huelva thinks.”

Well, she had me there.  I actually had to agree — because I desperately wanted to know what he thought, and not what Ms. de Zayas felt at the moment.

And for that same reason, this is an important article and I want to bring it to the attention of the few people — mostly Spanish authorities — who will be interested in these old stories and arcane information, assuming someone is motivated to translate this.

Everyone else is excused — but hey, if you’re serious about flamenco, you just might find something illuminating here.

Remember — this article takes us into the heart of flamenco creation, as witnessed by one of the greatest guitarists in the history of the art.  It comes from a now distant past, where seventh chords were considered far-out and radical innovations.  It is source material for any advanced course in flamenco history and aesthetics.  And while many authorities today cast aspersions on the worth of unprovable and malleable recollections, and while the exact words may not be Manolo’s, it is yet another example of the value of oral history in understanding flamenco, or any subject that involves human beings.

But I never got to hear Manolo de Huelva, or obtain recordings of his playing at its best. Shortly before he died, Mrs. de Zayas invited me to visit in hopes that he might give me something, but it was too late.  He was determined to keep his music from being learned by other players, and regrettably, he succeeded.  He is heard as the accompanist to some noted singers on numerous old 78′s, often reissued in new formats; but as he said, he never revealed the amazing abilities that made him the favorite player of countless great artists and knowledgeable aficionados of his era.  Ms. de Zayas later released a double LP which contained, in addition to excellent material by Ramón Montoya, some cuts that once again failed to reveal Manolo de Huelva’s true genius.

Foiled again.  Here’s the article — I’ll add some comments later on.


by Virginia de Zayas


Formerly, there were no solo guitar players: the only name which is still remembered and mentioned (and this, before the Spanish Civil War) is that of Paco de Lucena (not to be confused with the modern Paco de Lucía).  Paco de Lucena lived at the end of the 19th Century and he was not considered to be an outstanding player.  The following story told about him illustrates one of the rules of flamenco.

Once, while he was accompanying a song for the great singer Silverio [Franconetti], Paco suddenly introduced some fast notes, thus calling attention to himself.  Because of this, he broke the spell cast by the singer, who refused to be accompanied by him after that.  From this, we can see that traditional flamenco is a conversation.  First, the guitar plays, laying the foundation of the rhythm, and doing everything he can to display his talent.  Then the singer has the audience to himself and should not be interrupted or distracted, if he is to convey his song and its meaning.  Then the guitarist is given another chance to show what he can do, in the falsetas, which are played between each song while the singer is resting.  Then the singer is heard again.  They continue to alternate until the singer has exhausted his repertoire, or chooses to stop.

This idea of “conversation” is carried right into the guitar playing, in which the falsetas (variations) are divided into “questions” and “answers”.  The questions are played on the treble strings and the answers on the bass strings; the conversation finally ends on the lower strings.  The songs, too, are divided into a first section, which is sometimes called the question, and the second part, with or without repetition, which is called the answer.  Most modern guitarists do not know about these subtleties of flamenco as they would have before the war.


The oldest flamenco guitarist remembered, whose falsetas and accompaniments were played until recently, is Paquirri, “El Guanté” from Cádiz.  The meaning of his nickname is not known.  He was born around 1780 and he died young, in Madrid.  A woman fell in love with him, but he did not love her, so she avenged herself by poisoning him.  To save the woman, the doctor said that it was an embolism (blood clot).  Paquirri sang, played and danced with equal perfection.  His school of guitar playing was continued faithfully in Cádiz by his followers, notably by Patiño, who was too young to have known Paquirri himself.  At least a generation separated them.  These falsetas are much admired by Andrés Segovia and Antonio Chacón, the famous flamenco singer,  They are beautiful in their simplicity and “llaman al cante”, call for the song, for the singer to start.

About 1870-1880 Sevillian guitarists, with an inferior tradition, went to Cádiz to learn this school by listening to Patiño.  They were Juan el Jorobao (The Hunchback) and Maestro Pérez.  This is why the school is called the “school of Patiño”, although Patiño never composed a note.  Patiño also came to play in the new cafés cantantes in Seville, where they had flamenco song and dance.  Manolo [de Huelva, the source for Virginia de Zayas’s information in these articles] often talked with the old men in Cádiz, including “El Pollo” (The Chicken), a contemporary of Patiño’s, whom he was just in time to hear.  Flamencos used to have long conversations about these things, because almost none of them were professionals in the old days and because they had plenty of time to talk.  Afterward, when more flamencos became professionals, they would have to wait in the cafés and taverns until a client came, and during these waits there was much time to talk.  A flamenco’s hours would last until seven or later in the morning, and sometimes for three nights and two days, without sleep.  Formerly they would only drink wine and a little aguardiente (brandy); there is something about flamenco which sustains the participants

The famous Café del Burrero was the first to be remembered in Seville.  El Burrero sold donkeys (burros)  Silverio, the great singer, was associated width this café, both financially and as artistic director.  Here they established the flamenco program, beginning with the lighter malagueñas, sung by La Rubia de Málaga and La Africa, then proceeding to soleares, which are real flamenco and more serious.  Then, after several singers of soleares, the curtain fell and then rose to reveal Salvadorillo, the singer of siguiriyas.  The date of this café, I suppose, might have been around 1875,  Our information about the period comes from the writer, Demófilo [Antonio Machado y Álvarez] who was a friend of Silverio Franconetti’s.  He speaks a lot about Silverio’s desire to make flamenco better known, to bring it out of its hiding places in taverns and at family festivals.  Demófilo was opposed to this, saying that the demands of the public would make flamenco lose its quality.  He was a prophet.  In 1880 Silverio had just closed his café and was opening another.  He used to bring Patiño and the singer [Enrique] El Mellizo (The Twin) from Cádiz and, at times, he himself would sing.  He was called “el rey to los cantaores” (the king of singers).  Of Italian parentage, born in Seville, his family were tailors.  Silverio left this occupation so as to dedicate himself to flamenco.  When flamencos say that this or that song was “of [de] Silverio”, they mean that it was sung by him.  This is a great distinction.

This habit of taking the singer of a song to be its composer has caused great confusion in attributions.  The reason for it is that singers and guitarists, who used to perform above all in private gatherings, would be asked who was the composer of this song or that falseta.  They wanted to give an answer comprehensible to their public, so they would supply names of performers not long dead.  For example, I would ask Manolo de Huelva who composed a falseta, and he would reply “Patiño.”  It took me quite a while to discover that Patiño was not a composer and that he faithfully played the compositions of Paquirri.


Miguel Borrull (father), a Gypsy from the province of Valencia, was the first to introduce dissonances into flamenco music; these were particularly dissonances of the seventh.  However, I must point out that Andalusians had their own type of dissonance which consists of incomplete chords, which are not sevenths, also reversals and repetitions of the same noted at different octaves, the note which is repeated being part of an incomplete chord which sounds clashing, and other such devices.  Alonso (a town in the province of Huelva) has a fandango which is characterized by the incomplete, dissonant sounding chords.  What Borrull introduced, then, were the simplest chords of the seventh from classical music.

The origin of this was a blind Gypsy from either Valencia or Barcelona, named El Sisqué.  He knew a little music.  This was at the end of the 19th century: he had a donkey cart with a pianillo (harmonium) in it and a small boy to lead the donkey.  He would stop before bars and cafés in Barcelona and ply the songs of the eastern coast of Spain, especially tarantas, introducing chords of the seventh.  His also is the innovation of playing the tarantas in F# (previously it was played in G#).  Borrull had opened a café in Barcelona, after a very successful career in Madrid.  El Sisqué would stop before Borrull’s café  and Borrull listened and introduced these two innovations into the flamenco guitar.  He returned at times to Madrid and Seville where the other guitarists copied him.

Borrull was the first Gypsy guitarist of whom we have name and music.  He played in a very authentic flamenco style, and he also composed, with original ideas, but always following the traditional method of playing. He had a fine flamenco thumb, and played with much alma (soul).  Borrull in the north of Spain and Patiño in the south of Spain were the two outstanding guitarists of whom we have record who either introduced innovations or, in Patiño’s case, preserved Paquirri’s school for us.  Flamenco history begins with these three names.


When Manolo was a boy his dream was to go to Seville to hear the two most famous guitarists in Andalusia: Juan Gandulla, “Habichuela” (Stringbean) of Cádiz, and Javier Molina of Jerez.  When Manolo got to Seville, he immediately listened to Gandulla every night, and within a short time had memorized all he played.  Gandulla had learned Paquirri’s falsetas from his master Patiño of Cádiz, whose personal pupil he was.  Paquirri was born about 1780.  Gandulla is thus an important link in the transmission to us of Paquirri’s guitar school, the traditional 19th century school, preserved by the masters of Cádiz.

Gandulla had a marvelous left hand, Manolo tells me, and he played with authentic style and gracia (wit), and much alma (soul).  His right thumbnail was very macho (male) and hard.  He played the authentic pieces of Paquirri in all their beautiful simplicity, serving the music and not trying to put himself forward, or trying to show off on his own account,  Later on he came to play a few chords of the seventh from hearing Javier so much; they often played in the same café.

Javier’s thumb was more precise than Gandulla’s, but he never had his guitar well tuned.  Gandulla’s guitar was perfectly tuned after an instant’s testing.  The four fingernails in both of them were soft and almost inaudible in nail taps, so Javier resorted to many palpis (see above [a prior installment]), in bulerías to compensate.  (A thumbnail can be hard and the four fingernails soft.)

Javier could read music and wrote a book of his life.  After some years in Seville he returned to stay in Jerez.  He extended the use of the seventh chord and a few other dissonances in soleares, siguiriyas, bulerías and alegrías.

Manolo learned these dissonances from listening to Javier, Javier having learned the seventh from Borrull during the last years of the century.  Javier played brilliantly and was a fine composer, with many new ideas and a personal style, but always within the traditional rules.  His falsetas of soleares had a slightly different character from his siguiriyas, whereas Paquirri’s were all fitted to the different rhythms and played in both keys, E and A [Phrygian or Natural].

There is a long guitar introduction with a three note tremolo by Paquirri, but it is exceptional.  It is called Las Cuarenta (The Forty).  But Javier was really the first to play falsetas of more than two vueltas (cycles) [compases].  These long falsetas are properly played only as introductions.  Javier played falsetas of one or two vueltas only, between songs, and this is still the correct way to do it so as not to cool down the singer’s inspiration.  This was an acceptable innovation of Javier’s giving more scope to the guitar.  The point is that although revolutionary from the guitarist’s view, it did not take away from the singer and so obeyed the traditional rule.  It is a good example of a fine innovation.  Would that all innovations were as good and reasonable.


Rafael Marín, born in El Pedroso (Seville province), began the great revolution which made the flamenco guitar the hybrid that it is today.  He first learned flamenco, then the classical guitar and how to write it.  He wrote a method of flamenco guitar music (Madrid, 1902).   Instead of observing the ancient rules, especially those of playing with the thumb and ligados, he introduced arpeggios, scales and four-note tremolos [i.e., thumb note on a low string, followed by index, ring, middle and index on a higher string, or piami – the initial index finger stroke is the innovation], none of which are pure flamenco.  Rafael Marín went to Madrid and it is there that Ramón Montoya [widely seen as the progenitor of the virtuoso flamenco guitar] listened to him.

Ramón Montoya was born near Toledo and belonged to the musical tradition of the Gypsies of Madrid, Barcelona and Valencia.  This is not Andalusian flamenco.  Montoya went to Seville and there learned the traditional school correctly.  He began to study with Pepe “el Rojo” (the Red).  Then he listened to Marín and to Miguel Borrull (father) in Madrid.  He introduced Marín’s effects, and also the four-note tremolo.  But his heart was in the taranta and the rondeña, in which he could lengthen the tercios for more and more notes.  The taranta is originally from Cartagena and is sung by the miners of the hills above, at La Unión.  Montoya learned these songs from his first cousin, Basilio, and he would call them either taranta or minera.  Basilio and Ramón were cattle and horse dealers, an old Gypsy occupation, near Toledo.  The rondeñas he played were simple songs, apparently from the Gypsies of Barcelona.  He gave me two which he used in his compositions.

Montoya’s playing was perfectly even, no part being louder than the other.  He practiced at least four or five hours a day.  He was not interested in rhythm, and because of his weak fingernails he would play delicately but sonorously.  He preferred an arpeggio to a ligado.  Montoya’s mineras record contains the oldest taranta melody, played in G#, the old murcianas key.  His tarantas are in F#.  (* See editor’s note at end of article).}

Montoya had very excellent small hands.  He overflowed with notes, so much so that in the traditional toques (guitar parts), he sometimes played too many notes, and even too many beats.  In Paris in 1936-38 he stayed in the boarding house of another flamenco guitarist, Amalio Cuenca.  When Montoya composed a new falseta he would have to ask Cuenca to beat out the time, and then cut out the extra beats from the rhythm.  They would both laughingly tell us about this, illustrating the process, when they came to stay with us.  Ramón Montoya made some solo records at this time (in 1936), which are known to all guitarists.  They were initiated and financed by my husband Marius de Zayas, the caricaturist and painter, a lifetime lover of flamenco, which he played on the guitar  He was the father of Rodrigo de Zayas, whose writings have appeared in Guitar Review issues 36 and 38.

This is the origin of the modern school of flamenco guitar playing, beginning with Borrull.  Solo playing came into style, and today the flamenco player must rival the classical  guitarist and practice for hours because, now, speed is essential.  Ramón Montoya was ten years older than Manolo de Huelva (born 1892) and Borrull was older than Montoya.


We now turn to two traditions: the spoken, among the flamencos themselves, in which Paquirri and Frasco el Colorado, a Gypsy, are outstanding, and the literary, beginning with Richard Ford essentially.  Ford wrote his guide book for Spain in 1845.  Next comes S. Estéban Calderón, a journalist who happened to attend a flamenco festival in Triana sometime before 1847, at which he heard the singers El Planeta and El Fillo.  Then comes Demófilo (1881).  Since then small books by “flamencologists” have gradually proliferated.  These experts are chiefly interested in poetry and personalities.

[Editor’s note:  Richard Ford (1796-1858) write his Handbook for Travellers in Spain in 18445, and Gatherings from Spain in 1846.  Demófilo was the pseudonym of Antonio Machado y Álvarez (1846-1893) who published his collected and annotated Cantes Flamencos in 1881.  Long out of print, this book has seen two new editions in the 1970’s.]

El Fillo gave his name to a type of voice called afillá. This is a low falsetto voice used by baritones and basses.  For example, our usual falsetto is high, and when the singer goes too low and out of his falsetto range, he will sing with a natural voice.  This was the voice of Antonio Chacón, inappropriate for true flamenco.  Chacón sang malagueñas instead of soleares and siguiriyas.  In the low falsetto the singer, instead of singing in his chest voice, sings with falsetto in his chest range.  When he gets up out of this range, he will sing with a natural voice.  Conversely, when a singer sings with a natural voice and enters his chest range, he or she can sing in falsetto.  The medium range and low voices can have a few low afillá notes.  Pastora Pabón “La Niña de los Peines” (The Girl With the Combs), the great Gypsy singer of the period 1910-1940 who has made many records, is one example.  She made 78 rpm records, some of which have been re-recorded from 78 rpm to LP.  The late Aurelio de Cádiz had these low notes (his last records are not so representative of him), as did the late Perla de Cádiz (the Pearl of Cadiz).  Aurelio’s low notes sounded like a bassoon.  He was half Gypsy and La Perla was Gypsy and was especially good in bulerías.  Manolo says that no women have afillá voices.

The full low falsetto was possessed by many singers of the middle of the last century, among whom Silverio Franconetti was outstanding.  Manolo is a minute observer of voices, and he tells me that the last such singer was Diego Antúnez, a Gypsy from Sanlúcar de Barrameda, a great artist whom Manolo often accompanied.  He was old when Manolo was young.

This low falsetto voice must be an old tradition in Spain because the Pope had such Spanish singers brought to Rome to sing polyphonic music in the Sistine Capel.  This is described in an early edition of Grove [Music Dictionary].  When the Pope could no longer obtain singers with low falsetto voices, then began the reign of the castrati, in the 17th century.

In addition to the afillá voice, there are often found hoarse or bad voices.  People who have never heard an afillá voice call hoarse or bad voices afillá, for example the late [Manolo] Caracol.  I have often discussed this with Manolo [de Huelva], and it would seem that the uninitiated call afillá any voice with “mucho alma” (much soul), one which transmits [emotion], as did Caracol.

El Fillo himself was called “pollo ronco” (hoarse chicken) by his girl friend and this may have helped to begin the confusion.  Afillá and hoarse are not synonymous.  Hoarse or bad voices are not an obstacle to flamenco fame provided the singer sings “con alma” (with soul).  But the same singer would acquire far more fame if he had a good voice.

There is also the voice called rajá (rajada in correct Spanish).  Manolo tells me that the old Gypsies in Triana spoke of rajá voices, but he says that the correct word is desgarrada, or desgarrá in Andaluz.  Desgarrada means ripped, torn, broken, while rajá means cracked or rent.  Manolo describes this to me as if he were speaking of a cloth.  The only singer he has known with this kind of voice was Aurelio de Cádiz, who had a very flamenco, sonorous voice with an occasional fluttering, not a vibrato.  However, rajá is another word which seems to be used indiscriminately in describing a voice, often that of a singer with much alma, who transmits.

A touch of rajá was a characteristic of flamenco singing since ancient times and, among the Arabs, something similar was much appreciated.  The clear bel canto voice can be used only in malagueñas, provided it is sung with expression and flamenco style (i.e. martelé).

Since the middle of the 19th century the first to be remembered as singing with a natural voice is Manuel Molina, a Gypsy of Jerez.  He has also left two fine siguiriyas.  The low falsetto voice was thought to be very flamenco and manly, for of course all such singers were men with low voices.  It is very difficult to learn.  Very flamenco effects can be made with the low falsetto part of a voice because it is very flexible, even in women.  For example, the records of La Perla, in bulerías.  The voice also has much fuelle (breath) and is louder than the singers natural voice.

These effects are made with a play with the larynx, high or low.  This brings us to the “lloros de la voz” (sobs in the voice) which a singer will employ when his voice allows him to do so.  The throat must not be tight and the larynx must be lowered suddenly and then immediately raised, much as in yodeling, although less exaggerated.  I have tapes of a very excellent, pure flamenco singer Luisa La Pompi (The Behind), a Gypsy from Jerez.  By playing a 7½ at 3¾ speed, one can hear every detail.  One can also transfer the record to a tape at 7½ speed  and play it at 3¾.  La Pompi introduced many of these lloros, throughout her range, especially on descending conjunct or disjunct notes,  They do not stand out, but are a discrete ornament.  She also uses many slurs, especially in rising pairs of notes.  The slurs appear in fandangos de Huelva as well.


These are the two oldest singers of whom we know something.  The oldest is Paquirri “El Guanté”.  In addition to creating the traditional guitar school he also left three soleares which were still sung by Aurelio de Cádiz.  One is quite exceptional:  “Metío entre cañaberales / lo pájaro son clarine / que cantan al sol que sale.” (Among the canes, the birds are clarions, who sing to the sun which rises.)  This is one of the lovely nature pictures characteristic of the zejel in 9th century Andalusia.

The soleá is unusual today because it covers an extension of nearly two octaves: on octave and a sixth, attacking the upper C on the repetition, falling to establish itself on the lower E.  For this reason, not everyone can sing it. When he sang it for our tape recorder, Aurelio used some of his low falsetto notes to obtain this extension (incidentally, twenty-year-old tapes often spoil and are scarcely “for posterity” unless re-recorded often: the guitar comes out badly, sometimes sounding like a continuous organ tone).

Paquirri composed his grand soleá before leaving Cádiz, where it is traditional.  On his long journey to Madrid he would have passed through Seville and would have heard the singers there, especially Frasco el Colorao (Frank the Redhead).  This may have been about 1820-1830.  My thought is that after hearing Paquirri sing, Frasco composed his soleáCorreo de Vélez, Se cayeron cuatro gota / se mojaron los papele / Te tengo comparaíta / con el correo de Vélez.”  (“The mail from Vélez, four drops fell, and wet the papers.  I compare you to the mail from Vélez”).  It covers an octave and a seventh, and has other similarities.

We can also imagine that Frasco learned from Paquirri something of the condition of flamenco in the south.  Before the introduction of railways, towns had their own songs, as mountain villages still have their own fandango,  There was little interchange until travel became easier in the 19th century.  In Paquirri’s time only Cádiz had any real flamenco tradition, so far as we know, among the towns to the south of Seville.  So Frasco proceeded to action.  He went south to teach flamenco.  The old men of Triana told Manolo that the Gypsies of Jerez, of Puerto de Santa María, and Isla de San Fernando only knew the bulerías.  Frasco taught the Gypsies in those places to sing siguriyas.  So far as we know, the soleares did not catch up with them.  Since Frasco’s trip, a number of very fine siguiriyas have been composed in the south.  The names of the composers are remembered.  The songs of Triana are said to be so old that the composers are forgotten, and the songs are attributed to their singers.  Silverio’s trip in the south would have taken place well after Frasco’s.

I asked Manolo when he had first heard of Frasco.  He said, “As soon as I arrived in Seville I went to the home of the great Gypsy singer Tomás Pabón [Tomás Pavón] and his sister Pastora.  I had met her two years before in Huelva.  They immediately told me about Frasco, and so did other flamencos.  He was the greatest composer and singer they knew of.”


The Gypsies arrived in Spain before 1447, at which time an edict was published against them in Barcelona, where they arrived from France.  Recently Carlos Almendros (in the magazine Flamenco, July 1975, p. 39) discovered another document showing that there were Gypsies already there in 1425.  It has always seemed to me possible that some Gypsies arrived from Egypt (whence their name) through north Africa.  But if any did come from Egypt, they seemingly brought no Egyptian or north African music with them.

Did the Gypsies invent flamenco?  This has been a much-discussed question.  Undoubtedly they contributed greatly to preserving this music: they are composers and have a special genius for performing it.  They are especially gifted for the rhythms, in the way that Blacks are in the New World.  Today they and their admirers often seem to think that they are the almost exclusive progenitors of flamenco.

It is possible that Gypsies did not distort flamenco as much as they did ancient Hungarian melodies, about which Béla Bartok wrote so emphatically, rejecting the unforgettably enchanting Gypsy style with its excessive rubato, strong accentuation and variations on the violin (similar to the variation by Corelli, published by Walsh).

One might hazard the the flamenco Gypsy laments prefer a small extension, such as a fifth or sixth.  The polos, El Cante de la Caña, the serrana and some other old songs do not seem to have as much Gypsy influence as some others.

We have already answered the question of whether the Gypsies or the Arabs originated flamenco by showing that it existed before their arrival in Spain.  The structure of the polo goes back at least to the 8th century, if not earlier, particularly in its simplest form, that is the triple time accentuated on the third beat, which is probably indigenous to Spain, as well as the percussive guitar effects combined with harmony, and the rhetorical rasgueado rhythm.

[*Editor’s note:  The striking effect of Ramón Montoya’s use of the key of C# here is due in part to the intriguing sound of the tonic chord with the first and second strings played open,  Reading upward from the sixth string, the notes of the chord are G#, D#, G#, C, B (open), E (open).  Montoya’s tarantas, played in F#, derive their unique tonal signature in the same way.  The chord (from the sixth string upward) is F#, C#, F#, A#, B (open), E (open).]

End of article.

[The two prior parts of this Guitar Review article by Virginia de Zayas will appear in this blog soon.]

January 12, 2013   No Comments