Domenico Scarlatti - the Recording

Domenico Scarlatti, Velasco, ca.1740 Domenico Scarlatti was born in Italy in 1685, the same year as Johann Sebastian Bach and Georg Frideric Handel. What an incredible year for music! He moved to Portugal in 1719 to become music master to the young Princess Maria Barbara; when she became Queen of Spain in 1729, he followed her there. Respected in Italy as an extemporizer on the harpsichord, and for his dazzling technique, he did not begin to formally write his keyboard music down until 1738, when he was knighted by Portugal and composed a volume for presentation. A few years later, he collected a number of his older pieces into two more volumes. But then, ill health and gambling debts galvanised him into finding his voice. During his last 6 years 1752-7, he transferred his keyboard skill to paper in the form of some two hundred suites which he called sonatas. They combine pure joyous harpsichord sounds with the taut rhythms of Spanish dance and the harmonic brilliance of his Italian heritage to a degree that places him among the greatest musicians of all time.

Download them as a zipped file (1.4 MB).
You may be able to use my sound font to hear them as I do.

"Jacket" Notes

For a performer, there is always a conflict between saying as much as one can with each individual piece, and being faithful to the lifetime-built philosophy of the composer. Intellect produces complexity, but feeling demands simplicity. Early on, I played a Rameau piece for Kenneth Gilbert. There was a pause while he searched for something polite to say, then he commented that "I suppose it is a complex piece of music"!

Most performers, on encountering the range and quantity of Scarlatti's music, quickly choose a few pieces and interpret him as a capricious mannerist (or, worse in my opinion, as a romantic). This tendency is exacerbated by the characteristics of the piano, to which Scarlatti's sounds do not transfer well. (His breathtaking technique can transfer for some pieces, as Vladimir Horowitz amply demonstrated. He and Scarlatti would have had a ball together.) Here, I attempt the opposite - to present the cumulative achievement of a great musical colourist on the instrument which was his canvas.

Music is a language, physically, spiritually, culturally and technically. Playing Debussy on the harpsichord is as frightful as translating any language to another word for word. No other composer built his musical language around the sounds and dynamics of the harpsichord as Scarlatti did. No other music benefits so much from being embedded in harpsichord sound.

My recording follows the numbering of Kirkpatrick, whose study of Scarlatti is the base from which a modern player must build ("Domenico Scarlatti", Ralph Kirkpatrick, Princeton, 1953). To start, there are the 30 "exercises", as Scarlatti called them, of 1738. This is where Scarlatti introduces us to his incredible facility of playing any note anywhere on the keyboard with either hand. Many of the pieces Kirkpatrick lists as no. 31-93 were presented to the Queen in a volume in 1742, and as 94-147 in another in 1749. The pieces in these two volumes almost certainly predate 1738. They include some works for solo and continuo, which I have recorded as though they were for harpsichord pupil and teacher, in the Flemish tradition.

Then, with number 148, we continue with the Sonatas proper, the pieces that were presented to the Queen as they were composed, between 1752 and 1757. 545 of the sonatas were transcribed for piano by Alessandro Longo in 1906, and are still available from Ricordi (Italy) and Kalmus (USA). I thank Mark Newbold for generously making Ralph Kirkpatrick's facsimile of the sonatas not published by Longo available to me to record the remainder.

Most of the sonatas are built of hierarchical pair patterns - pairs of sounds paired in turn with other pairs, which in turn can be paired with other pair sets in French rondeau fashion. The primary formal structure of almost all of the sonatas follows two pairwise symmetries: tonalities are mirrored about a central double bar, and thematic material repeats after the double bar (although not always in exactly the same order). For example, K1 begins in D minor, progresses to A major at the double bar 14, and ends in D minor bar 31; thematically, bar 1 matches bar 14; 2-5, 22-25; 7, 17; 9, 18; 13, 31. In addition, almost all the later sonatas are written in formal pairs, several with explicit marking that they are to be played together. I have included silences at the end of each sonata such that, if the sonatas are played in numeric order, this pairwise arrangement on which Scarlatti obviously placed considerable importance will be heard. (Regrettably, these silences were omitted in the MP3 versions posted by Michael Hart.)

Harpsichord actions have a tiny inertia compared to that of modern pianos - a well-voiced harpsichord can be played much faster. Scarlatti obviously enjoyed having the fastest fingers in Europe, and explicitly noted some passages far faster than I can play them. (Burney quotes Thomas Roseingrave, no mean keyboardist himself, on a Scarlatti performance in 1714 as "ten hundred devils at the instrument - he had never heard such passages of execution and effect before".) Nevertheless, modern players unfamiliar with old instruments and old performance surroundings often play harpsichord music faster than it would have been played at the time. (One of my correspondents described one such concert as, "he played the fugue very very quickly, like at a gallop toward the victory, but the harpsichord maked like a great din"!)

Although harpsichords have no sustaining pedal, playing any note on good Italian instruments, such as Scarlatti played on, re-excites into sound all other undamped strings, thus sustaining a tonality for as long as one has fingers available to hold down the relevant keys. The Spanish royal quarters were veritable echo chambers compared to today's concert halls. Scarlatti did not mark precise tempos, but just noted a word or two concerning the way the piece was to feel (mostly Allegro, "get going"). These recordings are an attempt to produce on modern wavetable cards sounds of the musical character of which Scarlatti was a master - those of a powerful Italian instrument in rooms typical of the Spanish court. I have strictly restricted the techniques I use to those that were available to Scarlatti on his instruments.

Many of Scarlatti's works are centered upon the visual drama of his technique, which must be absent from a recording. Nevertheless, these recordings still display, I hope, some of his incredible brilliance. First, there is the consistency of Spanish dance rhythms as the foundation of his sound. To me, these rhythms are not polyphonic, but elaborated percussive solo accents, and as such are entirely consistent with the precision striven for by most recording musicians of today. And, when Scarlatti's phrases are repeated with no variations of sound, as he mostly explicitly wrote them, they build structure and power upon a sustained rhythmic foundation, rather than on a phrase-oriented vocal one. I have therefore eschewed melodic inflections and rubato for the most part (probably to a degree that overcompensates for the tendency of most performers to take the opposite approach).

Scarlatti introduces musical ideas in such profusion that, in most cases, if conventional phrasing attention is paid to them, the music becomes totally fragmented. The rarity with which Scarlatti actually notes pauses or breaks between apparently-disjoint phrases becomes justified when his work is studied overall - the silences he marks explicitly become more effective, and the phrases take their place as his development of melodic sequences, using sounds rather than just notes. The harmonies of these sequences are based on tonalities, and multiply in the manner of Italian toccatas, while the melodic lines proper continually expand into multiple voices that blend into harmony. In the Italian style of his training, it is mostly pure sounds, almost totally free of extra-musical allegories.

A twelve-note scale can not have all intervals in tune at the same time. MIDI systems default to equal tempering, where only octaves are really in tune. This tuning was not musically acceptable to keyboard musicians of Scarlatti's time, who restricted the keys they played in so that more of the musically-important intervals could be in tune. They also valued the variety of characters that differing keys have when all intervals are not equal. I used a technique of consonance analysis to aid me in finding the tuning that Scarlatti used most commonly, since no records of this survive other than the music itself. These recordings use the best tuning I have found, one very similar to one published by d'Alembert in 1752. With it, Scarlatti displays a harmonic sureness that is, to my ears at least, lacking with Italian tunings of the period, which historically one would expect him to have used.

What would I do differently with the sonatas if I were to start over? 555 things different! After all, I was playing from Longo knowing that it was a heavily edited text, and later had access to the original. I did re-record K175 - it blew Longo's mind. What an incredible sound! A muscular gal with jet black eyes flashing, a skirt meters in diameter swirling, feet imperiously stamping, bells in her ears and castanets all over her fingers, her hubby and brother standing at the side each making like a full orchestra with their guitars and feet. And, all invoked with ten fingers and a few harpsichord strings. Then there is K.555 - the last breath of Scarlatti. Of course I was tempted to play it with reverence, dwelling on each note as the last of a lifetime. But, I've always skipped the funerals of people who made a difference in my life. I want them to continue to live, in my mind since no other way is possible. And so, I recorded K.555 as if there could be another 100 sonatas to come. I hope you understand.

Three decades later, many of my tempi sound so furious,so determined. But, in my 40's, when these recordings were done, I valued flying fingers and exploding sound. I play more detaché now. I spent so many formative years with Artur Garami, "Full bow! Every inch!", then so long working on legato thirds & octaves with one hand and such, hanging onto the sound ...

I'll always keep my focus on tonality and tone colours - it's so special to the harpsichord. I want my sound to be clear and powerful - I'll keep everything that contributes to that. You play very differently in a living room than in a large concert hall - you have to play differently yet again in cyberspace, the world of pure disembodied information. You have to recast the whole way you approach music, compared to live performance - your appearance counts zero (no reviving listeners with spectacular hand crossing variations), audience interaction is gone (no coughing is great, but no breathing isn't), suspense, surprise and other excitement pale by the third playing. (Listen a few times to K.164, where I have left in one 'surprise' pause, for an example.) There's an old saying, "friends die off but enemies accumulate" - if you are going to record music, you'd better believe it.

But mostly, I'll keep trying to grow, keep exploring the fascinating beauty of sound. All kinds of little details will change - that's life. To me, music is the essence of the instinctive and total struggle of life against non-life. But most of all, it's love of sound.

John Sankey
other notes on harpsichord playing

I thank Prof. Francesco Bertoldi for providing me with a copy of the Velasco portrait, the only one I know of that shows Scarlatti's hands.