Itinerary of a recording

Claudio Ronco

 

Salvatore Lanzetti, Napolitano,

Sonate op.I  n. 5-6-7-8-9-11-12


Nuova Era Records, 1991.
cod 7048

Claudio Ronco
Stefano Veggetti
Diana Petech
Joachim Held

 

MP3 files

 

 


The man * The interpretation * The language * The Path

 

1- The man.
«... despite its intimist, wise appearance this music may be placed among the extreme arts: it gives expression to a singular, untimely deviant even deranged subject, though with an ultimate touch of sophistication it refuses the mask of madness.»
(Roland Barthes: Le Chant Romantique, in France Culture (12/5/76).

 

There is little to be said about Salvatore Lanzetti that is not to be found in any good dictionary of music: born in Naples about 1710, died in Turin about 1780 (his real name was Lancetti, but in the north of Italy everybody pronouced "violoncello" with a z, "violonzello"). A pupil of the conservatory of Santa Maria di Lorero in Naples, he was a cellist and composer, active first in Lucca (where he played with F. M. Veracini) and then at the court of Turin in 1727/28 and from 1760 until his death. He was in London from 1739 to 1754 or '56 and also played in Sicily, France and Germany. He is considered to be one of the first great virtuoso cellists. He published at least six collections of Sonatas for Cello and Basso Continuo (opus II, howover, is nothing other than a re-working of opus I, transcribed for the flute) and a method: Principes de l'application du Violoncelle par tous les tons, which appeared in 1779 circa in Amsterdam. Other Sonatas remained as manuscripts. Madame Marie Thérèse Bouquet, musicologist at the Sorbonne, very kindly offered me consolation when she confessed that she could add next to nothing to the few details available about this composer: at most she could add a few questions...
She did however point out to me the existence of two wills, drawn up in 1750, separated by no more than a few months: the first during a serious illness, and the second before Lanzetti left on a journey to France, when he was completely restored to health. From these two documents we learn that he possessed a considerable fortune in shares, goods and property; he had a brother, Daniele, a well-known theatrical agent in Italy (Vivaldi met him in Ferrara in 1736), and another brother, Luigi, whose son Domenico was an acclaimed cellist and composer. It is known that in 1737 he married the sister of the Besozzi brothers (the famous Turin oboists) but that in 1748 she filed for and obtained a divorce on the grounds of torture ("sevizie"). I also know that he was greatly appreciated in Turin, and in 1727 received a salary of 1200 lire per annum (as much as a first violinist and conductor), as well as a further 200 lire to provide himself with an appartment. Lastly, it seems that he was the first to perform a cello solo in the Concerts Spirituels in Paris, on the 10th and 31st of May 1736, the year which saw the publication of his first work: the Sonatas of our recording. As I write there is nothing else to be added, except what may be gleaned from a reading of his publication on the frontispiece on which, under his name, he proudly states his origin: Salvatore Lanzetti, Napolitano.
These Sonatas show us that Lanzetti was one of the first virtuoso cellists to lead cello technique forward towards the conquest of the most difficult uses of the bow (balzato ["leaping" - the bow rebounding off the string], picchettato ["knocking"], martellato ["hammered" - with a series of short sharp blows upon the string] etc.), the performance of double-stopped passages, of ingenious two-part patterns that call for bold, complex fingering; and, above all, the mastery of the use of the thumb as a "capotasto"', one of the most important evolutions in string playing that permitted remarkable extension of the cello's range as far as soprano high sharps.
His career may have been restricted by the limited interest that the cello excited in his time: it was, after all, a "bass', and the lower range indicated the plebeian, servant caste. The idea of rising to the Olympus of the higher notes, of being a "prince" or an "angel" for a few short passages, did not, perhaps, even attract the curiosity of those who recognised the sexual ambiguity of this type of instrument: "angelic and sublime" (or morbid) like the song of the castrato. The times were nor ripe: his patrons had not yet had the French Revolution (at most they caught its scent in the air and were worried), and perhaps Lanzetti, like a true Neapolitan -or as we think all Neapolitans should be- aware of the necessity of changing the tastes and prejudices of others, may have opposed a noble, poetic indolence, or given himself over to melancholic, perhaps sorrowing, disappointment. I know nothing more. All that remains is a teaching method of no great interest, the few cello Sonatas that I believe extraordinary, and a portrait in sanguine which brings out all the pride of his "Neapolitanness", where we see a man whose gaze is disillusioned, a little bitter and ironic, but also extraordinarily mad, if indeed madness can have been a part of his trade as Court Virtuoso Musician, just like that provocative, exasperated "gesture" with which he played the cello.


The man * The language * The Path

II- The interpretation

".. Interpretation is seen here as the ability ro read the anagrammes of the text (...), to draw out from the rhetoric of tone, rhythm and melody the net of accents; and the accent is the truth of music, the element that moves the interpretation."
(from R. Barthes, L'obvie et 1'obtus, Essais critiques III, Editions du Seuil, Paris, 1982)

 

In the seventeenth and eighteenth centuries the Sonata invariably appears (with a few exceptions) as two parallel staves: on the upper staff a melodic line furnished with practically everything that the musician needs to know to perform it; and on the lower staff a line of music generally lacking in indications and suggestions about what is needed to accompany the upper line. This accompaniment is more than one thing at the same time: the "Basso continuo" contains, hidden in its melodic line, a "fundamental bass", which one ought to recognise at sight and be able to let the listener hear. The Continuo may or may not have numbers written above the notes indicating the harmony, but in any case only the bare minimum is indicated for the performance of the Sonata - naturally with a great variety of possible solutions - so that, in this part particularly, the composer is inevitably held back from total possession of his creation and from his own clear idea of what he wanted it to be.
Opinions about how the Continuo should be performed are thus controversial: is one to improvise on well-practised patterns that are suited to the style and taste of the composer? Is one to invent ex tempore musical sound qualities, situations and developments? (In this context we should bear in mind the fact that the Baroque Sonata was not generally written thinking of eternity, but was expected to have a life cycle of a few seasons.) Or again should one carefully arrange a well structured accompaniment, perfect in every detail, so that every piece be offered a precise, clear and invariable meaning? Probably all of these: I really cannot believe that for two centuries performers refused to score and prepare the Basso Continuo carefully; I am convinced on the other hand that at times they developed it like a miniature orchestration, just as certain keyboard accompaniments written by Bach, Handel and others lead us to imagine.
In performing these Sonatas I have not aspired to so much, nor would I have trusted in my ability to succeed within the time limits of a recording: a Compact Disc today is little more than an ephemera. Or rather: it is the great number of discs that are produced that make it thus. Yet this music too deserves long study, spread over the various experiences and mentalities of music, just as has been done for those composers who have become myths and models. Beethoven and Mozart have been performed and studied incessantly: their "depth" derives in part from this constant application. Purcell, Bach and Vivaldi had to wait for generations to pass before musicians of a different age recovered their work, and a couple of centuries before attempts were made to reconstruct their authentic language faithfully. For Lanzetti, I have been able to do no more than offer an earnest performance, though one that had been pondered in only a few months and realized in only a few days, for this first publication of the work, in front of microphones that are much less attractive than a real, live audience. I have, however, tried to orchestrate the Basso using it as a potential orchestra, exploiting, that is to say, all the combinations of sounds that I was capable of imagining and producing. The choice of harpsichord and theorba offers textures of timbre that are very different one from another? Fine, they are used to highlight particular sections, or to emphasize rhetorical solutions (for example the melancholy of a development in minor) or again to isolate musical moments in which I am looking for a special sense which I wish to render with greater clarity. In any case the true Continuo instrument for these Sonatas is the second cello: only this instrument is a true second voice, or third, when the solo part duets with itself. I have asked the second cello to perform the Basso, trying to follow me or to contrast me, to join or separate itself from my musical discourse, and also to harmonise as much as possible with double string playing, as was the custom in Lanzetti's time. I have preferred to score these parts rather than to have them improvised, for they complement and enrich the double stopping of the solo voice, and can share in its temerity as if there were an imitator of Lanzetti himself playing the Basso.
In some instances I have chosen to have the second cello play an octave lower, like a double bass (for example in the Allegro of Sonata VI, where the instrument had to be tuned down a tone), in other parts an octave higher (the parts written on the soprano stuff in Sonata XII, and various passages in Sonata IX). I am confident that these are not merely arbitrary decisions: they should be contemplated among the many possibilities that the extreme simplicity with which the Basso is written, leave to the artistry of the performer. Scoring even a single double string in a part that was destined for more than one instrument, would have created confusion, as the addition of ligatures as bowing and phrase marks would have confused the reading of the figured bass. I could cite further little justifications based on practical experience, but suffice it to say that I see the Basso Continuo as a place where we should admit all the means and "artifices" that we might plausibly presume were in use in the specific epoch and school, and I mean by this everything that can be learnt from the solo parts, even the most eccentric, of virtuoso performers.
Bearing this in mind I have accepted a copy of a theorba built in the early seventeenth century, for the Sonatas of a hundred years on, but I know that it was used thronghout the age of the Basso Continuo, and perhaps only for this. The harpsichord in this recording is a copy of a rather late Flemish instrument, but is not unlike the instruments that Lanzetti would have found at the time of these Sonatas; and then -forgive the brutality of my confession- it was the only instrument available at the time of our recording. The cellos, however, have been rigorously restored to what we believe was their original condition, with bows that are (correct) period copies, and gut strings with all the ensuing qualities and problems. The pitch we have chosen is 440, since this was most probably the pitch used in northern Italy in the first half of the eighteenth century (circa 445); the recording studio is modern, but has the same dimensions as a baroque theatre; only two microphones, placed at a reasonable distance, seemed to give a satisfactory reproduction of an ideal listening position.
Lastly my approach to these pieces. I possess the Paris edition which is undated, but is presumably the second, probably printed by Le Clerc in 1738 or '39 (the first edition will have been the 1736 Amsterdam version). This collection begins with four short Sonatas, of medium or little difficulty (the level of difficulty found in Vivaldi's or Marcello's Sonatas), fairly conventional, but quite charming (especially Sonata III, with its delightful Minuet). From the fifth onward the Sonatas become dramatic compositions, and are rich and virtuoso, as if the first pieces were intended for amateurs, and the others for professionals (understandably: at the time there were more amateurs than professionals among the purchasers of musical publications). Two Sonatas, number eight and number ten, were certainly intended for a five-string cello; despite this fact I have performed number eight on the usual four-string instrument (and number ten is not included in the CD ). In the other Sonatas even the frequent sharp passages are conceived for a normal cello, with advanced use of the capotasto thumb, indispensable even if they are performed on a five-string cello. The style follows a formula that was quite common at the time: a mixture of French and Italian taste, but with several moments (e.g. the Adagio in Sonata VIII) which sound as though they were written in the years of Boccherini and Haydn (or in imitation of certain German music like C.P.E. Bach): the galant gesture, the "lightness" of the phrases, their linearity seem already "classic", but their are no doubts about the dating of the works. Elsewhere the scoring is bizarre in the asymmetry of its rhythms (Minuetto of Sonata V; central development section of the Largo of VI) or in the frequent interruption of melodies with pauses, or with arpeggio chords, whose execution is not clarified.
Ritornellos are present in less than half of the pieces, though never in slow or cantabile movements. The Adagios, Andantes and Largos, in fact, follow emotive directions that are clearly evidenced in the scoring and in the invention of the phrases: I do not think they would tolerate any repetition not founded in their poetical design. Nor would there be any scope - either in the Adagios or in the Allegros - for inserting ritornellos with significant diminutions or embellishments: I believe that Lanzetti left this sort of thing to Corelli's music, and that his idea of composing was quite different from all this. He writes, naturally enough, as a cello virtuoso; such that all his ideas seem to spring from his instrument; I imagine, therefore, that he must have taken delight in attributing precise values to certain sounds in certain positions on the fingerboard; for example: the central B flat on the first string (a slightly unnatural note, since the hand of the baroque player is habitually looking for the A that precedes it) is a note which recors invariably in almost all the movements of his Sonatas, always tending to generate tension, dramatic stress, both when it is to be "shouted" and when it is to be "whispered"
Again, in the numerous passages for two voices in the solo part, a note is often doubled by the Basso. One would almost like to have it played by the other instrument, so as to avoid the risk of producing unpleasant unisons if not perfectly in tune. Yet (playing them in tune) the effect is marvellous, thanks to a well-hidden virtuosity. The truth is that the greatest difficulties in Lanzetti seem to hide from superficial ears, rather than shining out in "special effects"; it is a good sign, but not for Lanzetti, who lived in an age that too often was superficial, at least as far as music was concerned.
Indeed, it was by playing this music, rather than by reading it, that I perceived its nature: it is music made of gestures, movements of the body, pulsations, calm or troubled breaths, of the desire to "be" and to "speak". It is dance, song, vision, everything that a body can create. This is why I gave up looking for the "logical sense" in the phrases of its structural development, in their rhetorical valency, in the mesh of agogic accents, and I sought out only the value of my gestures in the act of performing the written notes: there I found that real, living «body», that moves in a manner perfectly coherent with its nature, pronouncing phrases and fragments of great poetry. I looked for the "actions" of that body that lie hidden behind the codes of its written form, within the weave of its accents, all those dense, strong sounds, charged with "intentions", that generate movement. To all these I dedicated my capacity for concentration, my emotivity, my desire to express something that was born within me. And I believe I have interpreted Lanzetti.


the man * The interpretation * The Path

III. The language

" .. for if only you wish to move back a little to its (music's) origin, you will find a living, enduring, inexhaustible source of philosophical cognitions that are not common." (Decio Agostino Trento: preface to "Trattato di Musica..." by G. Tartini, Padua, 1754).

 

Perhaps I am incapable of seeking, in the arts, anything other than the signs of unrest. Perhaps because those signs stir my desire for art, or perhaps they are simply elementary, universal signs of an attention that moves towards the unconscious, like the act of listening which is directed automatically towards the content and meaning of a discourse and then, at other times, only to the abstract sound of the voice, with its unending chain of significances.
When we attempt to find these signs of unrest, no historical period is better or worse than another: the most bewigged baroque will do quite as well as the most tormented romanticism. For -as we know- a work of art is similar to the human being who observes it: body and soul, the conscious and the unconscious; subject to the time in which it articulates its own existence: changing, communicating, expressing meaning; in short: living and summing experiences.
In music, the experience of listening may follow a linear progression in time as in narration, where the present is variously charged with the past and the future, but is proceeding towards a specific end (and this is its manner from the ideation of Sonata Form, throughout the entire Romantic period); or again it may develop in a "circular" time ideally unmoving, where a supreme synthesis of things is offered in apparent immobility, -whose symbol, or artifice, is cyclic action- similar to that of an allegorical statue absorbed in the timelessness of its messages. In the latter case it is only the individual observer that can make it dynamic, translating it in his own intellectual and emotional activity. Thus it is -in varying degrees- in every art form; or at least it seams to me that only thus can its decodification and fruition be commenced: from the abstract to the concrete, or vice versa, producing those results that our culture and sensibility can achieve. Such then is music too: ineffable in its extreme abstraction, in the fact that it shares in the idea of a space-time collocation that is peculiar to the gods and not to mortals, or concrete, "physical" in that it is similar to a pulsating, rejoicing, suffering body.
Music too eventually treads the path that leads to the opposite of its initial condition: in the absence of a "tale" structured in time and in the "material" of humanity (a Bach Fugue, for example), there emerges an inner plot, a "voyage", an approach to emotive events that have something to say. When the music develops situations that can be identified (it is "descriptive" by means of rhetorical effect: associations {horns/hunting] become allusions or metaphors; similarities [violin tremolo/shudder, bass tremolo/terror] are cases of paronomasia..., all in all chaotic, uncontrollable chains of potential equivalencies - or of figures in which we can read more than one meaning, held in control only by the music's general rhetorical structure or by a title) or that can be derived directly from a concrete reality (such as certain guttural sounds produced by the jazz saxophonist, which move their listener violently, since he can feel the sound vibrating in his own throat), we may find ourselves roaming in the natural abstraction of "narrating" in music, towards a pure, absolutely free form of wandering; in Vivaldi just as in Shostakovich. This because -though the fact is not widely known- the work of art is similar to the search of the artist, to his awareness (or lack of awareness) of the fact that Art is first and foremost the act of seeking, of awaiting the ideal or perfect artistic event.
When I study the shifts and paths of musical language through the centuries, I have the impression that I am awaiting something; the fragments that I seem to find are always stimulating, for in a sense they belong to me intensely; all of this whispers a simple secret to me: everything is within me, and is susceptible of decodification, like a microcosm. I must recognise my language as being similar to those steps forward and digressions found in history, for I too am "structured" in that manner: it is the result of the regular sedimentation, of the constant overlapping of events and experiences in the history of our civilisation; each of these still belongs to us and is a part of a whole, that is at one and the same time, chaotic and organised; every fragment may be isolated here, but may continue to act within the whole, without being consumed. Every history -collective or personal- is "transparent" in its overlapping with another. Observing the "metamorphoses" of a Vivaldi into a Beethoven, then is really the same as looking at oneself in a mirror (with all of the vanity that this entails).
And Lanzetti? Lanzetti is a character of his time -of History, irretrievable in his totality- but also of my time, as long as the "mask" I don to perform his work (authentic instruments, authentic language) does not annul me as an individual. A mask may not be transparent (it would be annulled even in its ritual value), it can nonetheless generate new codes, which when they are decodified, do not lead us back to a world, an "external" place, be it historical or symbolical, but to an "interiority", where every meaning is possible: the Theatre which shifts its focus from the mythological individual, to the "concrete" individual in his totality.
Lanzetti is indeed a theatrical narrating musician. He has his tale proceed in a time that is marked out by the perception of life, and that is consistent with the rationality of the body and the irrationality of the soul or the mind. His music is dionysiac (from man towards god) not apollonian (from god towards man): it calls out in the listener a panic in the Greck sense: the anxiety, the uncertain fear that the Ancients attributed to the presence in nature of the god Pan. It seems to seek expression in an almost literary form, developing and arranging ideas that resemble complex rhetorical figures (one example may suffice: Antanaclasis or Reflexio: we encounter this figure a number of times, but we need only listen to it in the Adagio Cantabile of Sonata V: a brief, concise musical phrase is expressed, and then disappears. It emerges again at the end of other phrases which develop their own, separate discourse. As it reemerges, its "sense" is perceptibly altered, and the listener wanders: what does it mean? what is happening? why does it return and express things which seem to be so different?). I really cannot say how well schooled Lanzetti was; studying was no easy matter in those days, but music was in great demand, and so much work was done and much repetition was used: this means that many "cultured" aspects of his music may well be there quite by chance, or rather by virtue of an excellent instinct for imitation. Yet I would like to think that his choices (even if unconscious) were horn of a desire to narrate in tragic fashion through the ambiguity of his manner (Italian/French, galant/restless, playful/tragic) in an attempt to counter those who made music only for pleasure: to entertain. To seek out then, and to choose that which would be the "crisis" in the romantic sense of the word; what the painter Füssli would describe in a splendid aphorism some decades later as "the central moment, the moment of expectation, the crisis; this is the moment that counts, full of the past, charged with future...". If so it was, then this is why Lanzetti wrote for his own hands and arms that could play the cello: body and soul closely bound together, seem suspended in that absolute expectation that is the unique and inappellable condition of being human. This is the twin reality of the mystic; he desires nothing more.
Lanzetti is indeed a Virtuoso of the eighteenth century, whose profession causes him to seek to render natural to himself what must sound transcendental to his audience. Clearly his work may not be too unlike the work that his century imposes, and to communicate his language must resemble the rhetoric that is in currency. His way of being a cellist is then simply derived from the fact that he plays the cello in those years: he accompanies instruments and voices in the theatre or in Chambers; he follows, flatters and converses with the soloist; he represents scene setting, space and measure for the singer who is executing the Recitative or the Chamber Aria with Continuo. He may duet with the bass and tenor voices, even with the Sopranos: he is capable of becoming their substitute, and inevitably he will eventually appreciate the immense expressive strength, that lies hidden in the "substitution" of a voice made of bone, cartilage, muscle, gut, humour and tissue -all in vibration, all sharing in a sound which is heard in its entirety by other similar bodies- with the voice of a vaguely anthropomorphic instrument, if the latter is handled with skill. The body recognises the sense of its own sounds, and on this occasion the game of substitutions has tremendous potential, for, as Barthes puts it: "if sensible phenomena do possess a meaning, it is always to be found in movement, in substitution; in short: in an absence which is more clearly seen" (op. cit.). This then produces an unending chain of other meanings, which a man of the theatre, with his intuition and experience, can easily recognise and exploit. It is evident then, that the virtuoso instrumentalist will in the end create music on his own sound and personality, and thus will compose things that resemble his own mental and physical persona: selfportraits (once again, the element of vanity...).
Thus if Lanzetti's cello "sings", it is not merely because it has singing as a model, but is principally because, having been brought up among the finest theatrical singers, like these singers it does not articulate sounds in musical phrases, but pronounces a text, founded on sounds; in other words, his cello "speaks". And it speaks because, in a sense, it has become a living, throbbing body, complete with an "idiom" of its own with which it can express itself: its style, that is to say. This cello and the music written for it, through a rhetorical language which is still perfectly effective, enter into a parallel "world", which has its own Nature and Culture, sure that their messages can be decodified, since they are expressed in a similar, though metaphysical, reality.
The ear tries to catch "signs": "I listen as I read, that is to say, on the basis of certain codes", Barthes remarks. Thus it is that if Lanzetti's cello "speaks" to tell its tale, then his music too -his "tale"- becomes a body, which moves, acts, listens and moves on gathering experiences. While all this is going on, we perceive its truth, for everything can be recognised in the authentic onward motion of life, that can be identified in its natural directions. Thus it is moreover, that the performing artist must offer himself alone as an individual being: authentic, unique, perfectly human: not sublime or divine. The person who performs guides the person who listens by means of his own experience of the things that belong to a shared reality, as in hyper realist theatre: "love with me, for I love. Share in my suffering, for I suffer", he might say.
I attempt to "pronounce" every phrase that I play, I try to imagine the sound as a "body" that bends, stretches, leaps, contracts, speaks; for -as lovers of Dance will understand- in a manner that I feel intensely and that fascinates me, every movement of the body is immediately "translated" into movement of the soul.


The man * The interpretation * The language *

IV- The Path

"The unconscious, structured as a language, is the object of a particular and exemplary way of listening: the way of the psychoanalyst." (R. Barthes; op. cit.)

 

I have reflected at length over the decision to publish the following (many are irritated by the notion of reading music through images, and would like to feal free of it), but even the God of the Bible, immaterial and without form as he is, needs to use hands, eyes, ears to be understood and followed... and then, more than anything else this is my personal, interior, almost secret path, often hidden behind a performance that is apparently less dramatic; but these are the ideas I used to make the musicians who accompanied me understand me, and thanks to these "tracks" we shared in a musical experience. I shall publish then, and if I am wrong to do so, I shall take the blame.
Greek vocabulary is a useful aid in ordering the succession of events during the performance on this disc (this too is a suggestion of Barthes). There is a fact, pragma; a chance occurrence, tyche; a solution, telos; a surprise, apodeston; an action, drama. Everything comes about as if we were in the theatre: we take the disc, insert it into the player, sit in front of the loudspeakers, the curtain rises: we follow the events, receive the messages; when it has finished, we feel that something has happened within us. The pragma, the fact, is I playing my cello, my colleagues, their instruments. The tyche, the chance occurrence, is the manner in which the interpretation of the Sonatas begins to uncover directions, correlations, "signals" of meaning: the baroque Sonata is aleatory enough to turn even into the opposite of what one intended to propose. It is with a great deal of chance that I find suggestions on how to direct musical events; seeking them in rational ways might be merely deception. The telos, the solution, comes about by necessity: in this, musical ideas find their concrete form a finality: what I hear now has a more precise sense, for to give it a sense, now, is no longer a desire: it is a necessity; I cannot accept the abstract, if I have not attained it through the concrete. The apodeston, the surprise, is the destabilising event, the first to generate the "crisis", to drive me violently towards action; but it also resembles the re-awakening of the consciousness, the exciting revelation: it is unfurled, and opens out new horizons; the desire is: to go on, to become involved in the facts, to act. The drama the action, can now be experienced wholly, accepted without restrictions, shared in the first person. The first rule in music is, perhaps, to excite in the listener the "desire" for music, and perhaps this very desire is the "action". All of this happens in any case, both in the act of playing and in the Sonata itself. What follows then, is the path I have taken through the seven Sonatas on this disc.
The rhythm of baroque music contains something urgent, unrelenting, as though it were trying to represent the time to which every living being is subject. In short, if music is rhythm and melody, then the melody is the soul: free, potentially unconditioned, but temporarily bound to a body, of its own nature a prisoner of time: the rhythm. The Baroque might never tire of playing with this idea, ordering it in varying manners in rhetorical figures: Allegory, Simile, Comparison, Metaphor Thus rhythm -or the absence of rhythm- becomes the fundamental element of the music (whereas in romanticism the "body" might also be the melody: the theme). And in the Adagio Cantabile of Sonata V we do indeed begin without a real rhythm: two pauses written on the bass line dissolve its initial rhythmic sketch into nothingness. It is as though it had no body but ouly a "desire" for a body: I have no form yet, I am only an idea: I rise up in search of a heaven, and the Basso calls me back -rhythmically- to earth.
After the second pause sign, there are four trills, like nervous shudders: I feel as though I were shaking off my torpor: now I have a body, and I enter the rhythmic "machine" with energy. With anger too: here is the first of the high B flats, which I must play as though suffocating a scream in my throat; immediately after this, I walk almost in pain, my head is bent: now I have a body, and it moves through the fatigue and suffering of living; it moves on, accepting and rejecting, stopping and then running. It discovers that it is proud and desirous of love, of kindness. It lets itself be drawn along (with ouly a minimal attempt at holding itself back) by a movement from the Basso, a movement that is majestic, impetuous, emphatic. It cedes again, picks itself up again, walks on in pain, its head bent; it rebels, and once again the pauses reappear: two, as before, but now that body exists, and I cannot but accept it.
There is a pause: a silence which sounds alarmed to me; then I read: "Volti" [turn the page] (in other places I find "Segue" [next section follows without a break] or "Volti subito" [turn the page quickly]) and I continue. I realise that it is the marked strict rhythm that is disturbing, and that it is the pivot on which rests the entire dramatic effectiveness of the piece. The Allegro brings it back in the same function, but at first it seems to be clearly separated from the preceding movement, and it is here that I detect the first truly significant signal of Lanzetti's restlessness: it begins like a brilliant, virtuoso movement; throughout the first part it develops like a light comedy: a charming dialogue between lovers, a succession of sparkling and sweet incitements; then, in the second part, it becomes tormented: the same incipit, now in C major (a third higher and thus more emphatic; in rhetoric hyperbole) becomes aggressive, agitated. All the phrases become painful or more perturbed; trills repeated on the same note come back again. Tricks of the bow which, at the beginning, were joyful acrobatics, are now performed on the outer strings, with anger, effort; everything seems to be exasperated and seems not to reach any conclusion: only the return of the initial key with a phrase of convenience, to annul, to cancel a discourse that was becoming intolerably serious. It is as though it finished in nothingness, in emptiness: there is an ellipsis of thought (reticence, aposiopesis) and we feel it quite clearly: silence dictated by modesty. At this point the score calls for the Ritornello of the second half too, but I could never repeat it; I am convinced that it is there only as a sign of respect to symmetry, but nobody would want it. In this non-conclusion I am left dangling on a thread: I can only fall straight down into the Minuet. It seems to begin from nothing, or to have come out of another silence; I want the two cellos alone, for all the first part. The harpsichord enters in the major, to confirm its joyful colour A little later I play a duet alone, with the Basso gracefully chasing after me. In the Da Capo everything seems to fall into its rightful place: a galant Sonata, in elegant livery, for a high society audience, well served and entertained.
The sixth Sonata seems galant too, when I begin the Allegro. Galant and calm, like a walk in a park amidst calm, smiling middle-class strollers. It calls to my mind pieces of celebratory music by Elgar: "unproblematic", self-assured, with a clear conscience.
I move amids fond phrases: the music I play reassures me, pleases me, tells of pleasures. But immediately the following Largo is an intensely melancholic melody; it proceeds a while, then it is suspended a moment (there is a pause marked), and suddenly it is replaced by a totally different figure: a chromatic descent, dancing on regular, dotted quadruplets, light -calling to my mind a coach driving along calmly- yet full of subtle agitation: undefinable, somewhere between a sense of weariness, of disillusionment, and a gesture of extreme emotion, almost of sorrow. It finishes, and the melancholy theme returns; now it is tormented: it writhes, it seems to have no end, but it lives in its own sublime sensuality. I should never want to leave it: to forget my destination. Another pause, and the coach rolls on again, with a rhythm like an obsession, a trance. Have I been riding in the coach all the time? Dreaming at times, and moving away from the place where she whom I love was present, she who made my life pleasing by being at my side? A pause mark suitable for a cadenza brings these questions to mind, just before the last note of the Largo.
The harmonic path of the Sonata is simple and effective: B flat major the first movement, and G minor the Largo, which closes on its dominant, very gentle in concluding the final movement with a return to B flat; Lanzetti uses this pattern almost always. The finale is a Gavotte with six variations: here, on an elementary Basso, always repeated in the same form, I decide to play on an obsession: in these variations I can find nothing other than bitterness, and I accept it as it is: an obsession, or at most a monologue on an obsession (how indeed could it be a dialogue, if the journey has led me away from my beloved?) I reach the conclusion without having shaken off my obsession: I shall continue to hold it within myself: obsessionately.
Sonata VII also begins with a galant movement (it is rhetorical: it is called Captatio Benevolentiæ...); it is rich in ideas, surprises, brilliant and curious gestures. Yet it has no conclusion: it flows (or is transformed) into an impassioned fugato: as I play it, I feel as though I were being sucked into the whirl of a frenzied dance. Halfway through it suddenly changes: out of nothing there emerges the minor key, with a phrase that sounds like a tongue-twister in Neapolitan sorcery (this is why I play it with "notes inegales": not to make it French, but to produce that effect of popular music, that certain "swing" that is a little magic and a little primitive), then it takes on darker shades, like a Sabbah painted by Salvatore Rosa, and the return to the major key is only a reversal of the lights: nothing could stop the rush of notes: I cannot yield, or hold back, except in a chord that once again is suspended in emptiness: before me, the Largo in E minor and the inertia of the "fuga" squeezing me from behind. Here there is a long note, while the Basso begins a regular, slow, expanding movement; then a trill (it must gradually increase in tension: carry the voice), then a melody, that is so intense, so impassioned that it must be shouted with one's whole body, until blood pulses with the pulsing of the Basso (the dynamic mark is Forte: it is to be taken literally: with physical strength). Then straightaway it is all interrupted: we read: "piano, Arpeggiato", and the same sequence of pulse beats is repeated in emptiness, in the desert, in the startling absence of the melody. The first time I played it, I felt a sense of bewilderment, then I came to love that absence better than any reappearance of the melody, splendid though it is; when it disappears it seems to be all the more present, and its secret charms me. Is this a dialogue with nothingness? Or is it an deliberate conscious act of substituting presences with absences, to rise still higher, to touch with one's hand an abstraction that is even more sublime than music itself? The trace -the artifice of that void- gradually loses effect as the Largo proceeds, and it too becomes present, concrete, sentimental. It is with this shifting element (at the beginning it is the Basso line, then it becomes the absence of a melody, lastly it is played by the solo cello, in two parts) that we reach the conclusion: a pause of great intensity. After this I would have the Rondò explode, and it is an almost satanic Rondò: here we have the trance, the obsessive beat of primitive dance: it has no beginning and no end: I listen to only a fragment: it has been and continues to be. It is a dance that is as old as man: it represents the eternity of the gods. Yet is it also nearly a revelation of what the fugato -behind the apparent austerity of its form- was hiding: the magic of its own origins?
Sonata VIII is a subtle game: its is suspended on the delicate thread of the bow stroke in the rapid triplets, in seeming contrast with the indication Allegretto, considering the calm pulsation of the harmonies and musical ideas. In fact, there is something serene about the latter, something agitated about the former: an agitation which, while it may be joyful excitement (Nature as friend expectation of the beloved) is also a subtle form of uneasiness (Indian music has a very similar Raga: Lalit; imagine a prince at dawn on the day of his wedding). The second movement is an Adagio; majestic at the beginning (triumphant entrance of celebrating Nymphs?) and tender and loving in its development. I chose to perform it like a love duet with the harpsichord. It seems to me to be like a dialogue of tender, sensual gestures; the harpsichord manages to find something feminine, innocent, silvery. Those long syncopations with their sweet embellished resolutions are like long lovers' embraces. And in the Allegro finale the agitation returns, only this time a little more dramatic: it is torment, excitement: it is the desolation and desperation of the abandoned lover. I have tried to find other things in it, but have been obliged to yield to this violent anticlimax.
I find this desolation and desperation again in the Sonata IX: a heavy sensation of pain and weariness. It seems to be a part of the Adagio, raising it to the point of tears, to a scream of anguish. I proceed in the Allegro with something like the rage of one who is trying to drive out the demon of torment: trying, even weeping over myself to relieve the pain, but finishing by screaming it out aloud, abandoning myself to lamentation. Paradoxically, this act of screaming out against the fear of death is a shout of life, and in it we may recognise all human dignity. It seems to find no path to follow: only broken, acephalous sentences, like funeral columns. To conclude, Lanzetti here writes a very brief, stylised phrase, so strongly contrasting with everything that has gone before, that it sounds sarcastic (I have toned it down by playing it twice, for I could not bear its brevity; but in rhetoric this is Antiphrasis: it means the opposite of what it affirms) I perceive the probable reason for this choice: a clear sharp cut is needed to end this movement, and this calls for a language that is quite different: the actor has let himself go too far, he has become a real man, has really soffered; now he is an actor again; he takes his bow, and the scene changes. It is a Rondò full of sweet melancholy, agitated to the point of madness. At the point in which the cello voice is to rise up to the high sharps, I wanted to have the harpsichord play alone an octave higher: I felt that in this way I could make the musical effect less ferocious, though I have repented a little: when I left the Basso "a loco" and the second cello played too, the impression of solitude in my part was harrowing; but I found it excessive.
Then, tenderly comes the Da Capo. If one has to repeat the beginning of the Rondò, then one might as well recuperate a sense of serenity and peace in it (although, if this contains a rhetoric figure, it is Dubitatio: let the audience choose the sense of this finale).
The opening of the Sonata XI sounds like a bold, clear, sonorous symphony. It seems to imitate the interplay of solo and tutti of a chamber orchestra. The development is lively and serene, rather like in Sonata VI. Once again such a reassuring, even conventional movement goes on to startle the listener: a bizarre, brief, intense Adagio, fragmentized by arpeggio chords, repeated in Forte, pauses and highly contrasting dynamics. I decided to render it similar to an almost recitative, underlining this idea with brief cadenzas on the three pause signs. This "reciting" led me to a profound, insuperable, even violent sense of solitude. The final phrase of the Adagio closes in a penetrating, intense pathetic suspension, and the cello solo begins piano an Allegro fugato that is quite different from the one in Sonata VII. The melody that I have to play here mezza voce, sounds infinitely sweet and full of sadness: now I think that my intuition was correct: there is a great sense of solitude. Here again I want the two cellos: alone. I ask the two other instruments to wait for the return of the theme before playing together with us. Then everything passes into another Adagio, very short this time: it seems only to wish to repeat the finale of the preceding Adagio (it is almost the same melody), in search of a new development. This time it develops into a Rondò, almost grotesque in the way it becomes heavier and tries to be majestic. Convention would have it repeated after every variation, but in this instance I do not think this is what Lanzetti wanted: the variations are written out in succession, and only at the end do we read "Segue il primo Rondò sino al segno" [repeat the first Rondò as far as the sign]. The justification may be weak, but all the variations seem to come from a "world" that is quite different from the first Rondò, and to be seeking each other in turn. Thus, amid languid, drunken affections with the sweet rhythm of the melody (it sounds more like a Sicilienne than a Rondò) we reach the minor key: and it is the most pathetic, the most heart-rending, the bitterest that I imagine could be written for the cello. The return to the first - massive, heavy, almost clumsy- Rondò is of so little importance that I discover that I still hold within myself all the feelings that came before it, all still intact: the finale closes like a coffer made of iron and solid wood, full of precious things.
Sonata XII opens with triumphant fanfares: the key of D major helps: it is sonorous and vital as no other key on the cello. A bold festive phrase (I have been told that it is the same as one in a cello concerto by Boismortier; who knows who copied?...), played in unison, renders the idea of an orchestra, and the soloist plays above it with splendid virtuosity, amidst flashes of light. This is the Sonata that makes most use of the soprano voice, in brilliant, ingenious manner, played with humour, charm and taste. Everything is concluded without a change of tone and without excessive severity, sounding like the Sinfonia before the curtain is raised. Indeed, I am certain that this was its intention, for the Andante cantabile tbat follows really does sound like a pastoral scene. It is so full of descriptive elements, all set out in order and clear in their rhetorical meanings, that I cannot help yielding to the temptation of portraying it like a painting: set between two large, leafy trees, close by an amenable hedge, two young lovers; she seems to abandon herself on the grass, inviting love; he reaches his hand out to her, as though yielding to her desire, but his gaze is directed towards the background. Here we see a distant valley, through which flows a sinuous river. On one side two horn players raise their instruments, while two deer run away in fright; they are followed by a pack of fierce dogs and hunters on horseback. Now I no longer know if he, the lover, asks her to get up and watch the fate of the deer (with sadness? Lovers think of death with scorn), or if she asks him not to look, to forget the struggles of life and abandon himself to the carefree joys of love. As I play, I seem to hear the horns calling the dogs running, the deer trying to get away, their fall overwhelmed by the noises of satisfaction and victory. I seam to see the lover get up or bend down, calling or being called by the sweet sensuality of the woman. Everything is motionless, yet alive, dynamic in the secrets of its allegories.
There now follows the most amorous Minuet in the entire collection of Sonatas: full of a light that is so serene and pleasant that I could not do other than orchestrate it for all our instruments. Each has his turn and plays with his own loving phrases, as fair as the harmony of life that music delights in imitating. Or, perhaps, as fair as madnes, for, as Barthes remarks: "the musician is always mad, unlike the writer, who may never be mad, for he is condemned to meaning".

Post scriptum

All the work on this disc was born and grew out of a reading of Barthes. It was therefore inevitable that this should have flowed into this text, even though mainly unconsciously. Where Barthes is quoted, he is quoted in a sense that is quite unfaithful to the original meaning, in a transverse reading that he would surely have appreciated. I hope my readers will also have appreciated it.

 

Claudio Ronco, Venice, july 1990.

 

(English translation by Timothy Alan Shaw)


back to the top * The interpretation * The language * The Path


claudio ronco