Spanish Wit in Cosmopolitan Dress – on de Falla's El sombrero de tres picos

Manuel de Falla (1876–1946), composer of El sombrero de tres picos. Public domain · Wikimedia Commons

I. The Opening Bar

I have spent much of the past year sitting with the score of El sombrero de tres picos, and every time I begin again, I find myself stopping at the opening bars.

The work opens with a gesture that immediately tells the listener where we are. A short fanfare in trumpet and horn, with hemiolas and a dance pulse – rhythmically and modally unmistakably Spanish – and then, suddenly, a mezzo-soprano voice singing "¡Olé!". It is an opening that feels like a show prelude, pure atmosphere, pure invitation. With closer listening and reading of the score, I have gradually become aware of something else: that every detail is economically placed, that nothing has been added for effect. The fanfare is short. The cry is short. The orchestration has exactly what is needed to set the scene, and not a single instrument more than absolutely necessary.

It is this duality that makes El sombrero de tres picos the work it is, and it is what I want to write about here. The work is at once a Spanish ballet-festivity and an internationally advanced compositional work. It is simultaneously the folkloric and the modernist. And for me as a conductor, it is at once the invitation to let myself go, and the reminder that I am not permitted to.

That is probably why I have kept returning to this opening bar. It is like a small contract between composer and conductor, written into the first bars: this is a celebration, yes, but it is a celebration that demands precision. If I break that contract, I lose the work.

II. The Work in Its Time

For me it is not possible to work on El sombrero de tres picos without bringing Ballets Russes into the rehearsal room. From 1909 until Diaghilev's death in 1929, Sergei Diaghilev's ballet company was the single most influential institution in European modernism. It was not merely a dance theatre; it was a curatorial enterprise that engaged the foremost composers, artists, and choreographers across nations and combined their work into productions that had an immediate international effect.

What has struck me over the years is how much of what makes El sombrero the work it is lies in Diaghilev's assembly. No single one of the artists could have created the ballet alone. Picasso alone would have made a modernist-Spanish set, but without the music and dance that make it something more than a work of art. De Falla alone would have written a fine concert work, but without the visual and choreographic amplification. Massine alone would have choreographed a splendid dance, but without the modernist frame that lifted it from "Spanish ballet" to "Spanish ballet within the Ballets Russes tradition". Diaghilev's genius was to see which three voices had to meet for the whole to become more than their sum. It is a kind of curatorial artistic decision almost entirely absent from our time, and one I miss.

On 22 July 1919 an audience sat at the Alhambra Theatre in London and watched the premiere. Ernest Ansermet conducted. Léonide Massine was chief choreographer and danced the leading role. Set design and costumes were by Pablo Picasso. The music was newly composed by Manuel de Falla. The piece was Le tricorne – in Spanish El sombrero de tres picos – and it was one of Ballets Russes's new Spanish ballet projects in a season that also included works by Stravinsky and Ravel.

The performance was a triumph. Le tricorne circulated in the following years through Europe and South America, and it established de Falla's international reputation for good. It is a light, witty, colourful work, full of dance and parody – the bright side of Spain, presented to an international audience, and one of the most immediately charming ballets in the whole Ballets Russes repertoire.

But the lightness is not simple. It is a work that demonstrates something specific about what Spanish music could be within European modernism in 1919: Spanish folklore treated not as ethnographic curiosity or exotic effect, but as a fully cosmopolitan musical language on par with French impressionism or Russian neoclassicism. That is the side of the work the conductor has to hear first – for if one hears only the party, one loses de Falla.

The background for this being possible lies in Diaghilev's curatorial hand. He had earlier drawn on Russian material (Stravinsky's early ballets), Egyptian material (Cleopatra), Greek visual material (several Nijinsky choreographies), and from around 1916 it had become clear that Spain was the next great cultural territory.

Diaghilev approached de Falla. The composer had returned from Paris to Spain and was living in Granada. He had made his international mark with earlier works but had not yet written a large stage work for the international market. Diaghilev offered him precisely this opportunity: a ballet for Ballets Russes, with Picasso as visual designer, Massine as choreographer, and the full Diaghilev apparatus behind him.

The material de Falla chose was Pedro Antonio de Alarcón's 1874 novella El sombrero de tres picos – a picaresque story about a corrupt corregidor (a local Spanish magistrate, recognisable by his three-cornered hat) who tries to seduce the beautiful wife of a miller, and about how the miller and his wife outmanoeuvre him. It is a story in the classical Spanish picaresque tradition – cynical about power, warm about ordinary people, full of humour and vitality.

De Falla composed the music between 1917 and 1919. He used authentic Spanish dance types as structural elements: fandango for the opening of the evening, farruca for the miller's dance (a farruca is a Gitano flamenco dance traditionally performed by men), seguidilla, jota for the closing dance. Each individual dance is stylised to concert quality, but retains its folk-musical character.

It is this doubled approach that makes El sombrero special: it is at once an authentic Spanish ballet-festivity and an internationally advanced compositional work. Picasso designed a set that was radically modernist (geometric, abstract forms in strong colours), yet at the same time recognisably Spanish. Massine choreographed dance that took the flamenco and bolero tradition and translated it into the vocabulary of ballet without losing its character. De Falla's music did the same on the sonic side.

III. Folklore as Stage Art

My reading of the work – and it is a reading, not a truth – is that El sombrero de tres picos is a demonstration of how a modern composer can take folklore and turn it into stage art without losing either the folklore or the stage art.

Many twentieth-century composers used folk-musical material in orchestral works, and the results were mixed. The simplest approach – take a folk song and orchestrate it, or write new music "in a folk style" – often produced works that were neither real folk music nor good concert music, but a blurred mixture. De Falla's approach was different. He used folk music's dance types, modal scales, and rhythmic characteristics as structural elements in a modern composition. The folk music is not being decorated; it is the underlying grammar.

That means that each dance in El sombrero is recognisable as Spanish dance to a Spanish listener, while at the same time functioning as an advanced orchestral work for a listener from any tradition. The miller's farruca is an authentic flamenco form, but it is also a formally developed orchestral piece that stands comparison with any advanced orchestral movement from the same decade. The final jota is an authentic Aragonese dance type, but it is also an orchestral finale that gives the audience the cathartic joy every functioning ballet finale ought to give.

That is, as I see it, one of the most important aesthetic positions in European modernism just after the First World War. Several composers were working in the same direction in those years – Stravinsky with Russian folklore, Bartók with Hungarian and Romanian, Vaughan Williams with English. Now de Falla did it with Spanish. It was an international movement in which local culture, translated with precision and respect, became part of the new international modernism.

In the rehearsal room this forces me to treat El sombrero with the same technical precision I would give any other modernist major work from the same decade. The rhythms are often complex – Spanish dance types have hemiola effects, polyrhythmic layers, asymmetric accentuations that require considerable orchestral discipline. The modes must be intonated cleanly. The character of each dance must come through without becoming caricature.

It is a temptation I know well: to let Spanish music become "very Spanish" – to exaggerate tempi, to let tempo changes become large, to let dynamic contrast become melodramatic. I have gradually found that this is precisely what this work does not need. It needs the same refinement de Falla has himself written into the score, no more and no less. It is the duality that must be held – the Spanish character and the modern precision, at the same time, without one of them being allowed to slip past the other. This is also one of the places where I have to do less of what I first want to, and more of what the score asks for.

IV. Three Moments That Open

Three moments in the ballet make this clear to me.

The first is the introduction itself. The work opens with one of the most characteristic gestures in the whole Spanish repertoire: a trumpet-and-horn fanfare immediately established as Spanish through its rhythmic gestures (hemiola, 6/8 dance pulse) and its modal colours. After the fanfare comes a short mezzo-soprano line singing "¡Olé!" – the famous Spanish cry that has been on Spanish stages since the origins of flamenco. It is a gesture that establishes everything at once: we are in Spain, we are at a show, we are going to enjoy ourselves. It is also an opening that does something unusual – a singing voice in a ballet production, something one does not encounter in many places outside the Ballets Russes tradition.

I have often thought that this is an opening de Falla consciously wrote in order to be misunderstood. I used to think myself that it simply had to be played as a show opening, with a big fanfare and a dramatic ¡Olé! With closer listening and study I now hear something else: the mezzo should not raise the roof, she should set the frame; the fanfare should not be pompous, it should be clear. As a conductor I therefore believe I am obliged to resist the temptation I feel to "decorate" the opening. The duality begins already here, in the first bar: Spanish wit and modern economy, inseparable.

The second is the miller's Farruca. This is often the dance number that stays in a listener's memory after a performance. The miller's dance is his character-drawing: he is a wiry, clever, hard-working man (in the ballet version often danced by a virtuoso male dancer, originally Massine himself), and his dance is a male flamenco form – serious, intense, focused, far from the ornamentation of female flamenco.

De Falla composes the dance with a precise rhythmic structure. It opens with castanets and a low string line that establishes a foundational pulse. Over this pulse comes a melodic line (first horn, then strings, then full orchestra) that builds gradually. What is characteristic of the farruca as a dance type is a particular rhythmic profile – accentuations at particular points in the bar, sudden stops and starts, a built-in intensity that demands absolute rhythmic precision.

What makes this dance especially demanding to conduct is precisely those sudden stops. A farruca has built-in pauses where the whole orchestra breaks off – not as a ritenuto, not as a fermata, but as clear, concise silence – before the dance resumes with renewed intensity. If the stops do not sit absolutely together, the dance loses its character. If they sit, but I as conductor do not have the insight to breathe them right, the silence becomes mechanical instead of charged. This is one of the places where an understanding of the flamenco roots must have sunk deeply enough into my body that the beat simply sits – not as a conductor's count, but as a dancer's rhythm I am myself woven into.

And the orchestra must play with focused rhythmic precision, not with expansive Romantic phrasing. This is one of the places where the duality becomes concrete: precision in the service of character, not in opposition to it.

The third is the closing Jota. The final dance is an Aragonese dance type – the jota is one of northern Spain's most characteristic folk dances, faster and more acrobatic than the southern flamenco forms. De Falla uses it as a finale: all the characters dance together, everything is resolved in celebration, the corregidor is humiliated, the miller and his wife are happy, and the audience receives the communal catharsis a good ballet finale is supposed to deliver.

Musically the jota is an orchestral high point. The orchestra is fully mobilised – strings, winds, percussion (including castanets, tambourine, bells), a whole choir of instruments. The tempo is fast and drives forward. It must be one of the most spectacular finales in the entire ballet repertoire from this period.

The most important thing here is to keep the tempo in check – like a rider holding the reins tight to keep the horse from bolting. It is tempting, for me as well, simply to let the orchestra tear loose in a wild celebration, but de Falla's music demands rhythmic structure even in its most ecstatic form. If the foundational pulse becomes unclear, one loses the danceable quality that is the jota's whole condition. This is the duality at its most exposed: the festivity's tempo that must not become the festivity's chaos, the precision that must not become the tamer of the festivity.

V. Coda

El sombrero de tres picos is a work that has held up wonderfully in the more than one hundred years since its premiere. It is one of the few Diaghilev works still in regular repertoire both as a ballet and as a concert piece – de Falla's Suite No. 2 from the ballet is often played in concert programmes. It is still charming, still witty, still alive.

It is not a heavy work. It is not about mortality or consciousness or existential questions. It is about a deceitful corregidor and his follies. It is a farce in music. But it is a farce written by one of the era's finest composers, designed by one of the era's greatest artists, choreographed by one of the era's most important choreographers. It is a farce of the highest quality, and that is what makes it what it is.

And it is perhaps the finest demonstration I know that modernism and celebration are not opposites. De Falla showed us a possibility: that a new musical grammar can serve an ancient art form of festive celebration, and that the celebration does not need to be reduced to cliché in order to make room for the new. The duality holds. It still does.

As I now stand before the work again and am to perform it, it is this duality I hold in my hands. The fanfare in the first bar, the silences in the farruca, the reins in the jota's finale – all of it rests on the same grip: to give Spanish wit its full place, without letting go of the precision that makes the wit hearable in the first place. That is where I have to be when the music begins.