|  
  
 
Research over the past twenty years  
	has revealed Giuseppe Valentini (1681–1753)  to be a figure of considerable importance in  early-eighteenth-century Italian music.(1)  As both a violinist and a composer (of operas, oratorios  and cantatas, as well as instrumental pieces), he was evidently one  of the most prominent musicians active in Rome  from around 1700, achieving substantial success  there with his sonatas and concertos even  
before the death in 1713 of Arcangelo Corelli.  Besides his several published collections issued  in the period 1701–24, numerous instrumental works  survive in manuscript, including nine preserved in Manchester  (in a collection 
described below), twelve in the archive of the  Scuole Pie di S. Pantaleo, Rome, and six in Dresden.(2)  Working often in a freelance capacity, Valentini received  patronage in Rome from various churches and important persons,  many of whom were connected with the Arcadian Academy.  Most pertinent to the concerto 
presented here is his association,  notably in 
the 1720s, with the  court of Cardinal Pietro 
Ottoboni (1667–1740).(3)  
To the best of my knowledge, the only known source  for the present work is a set of eight manuscript parts  within a large collection of Italian music now  known as ‘The Manchester Concerto Partbooks’.(4)  Since the history and contents of this significant collection,  representative of the heyday of the Italian concerto, are  assessed in detail elsewhere, only a summary needs to be given here.(5)   The volumes contain sets of separate parts for 95 compositions,  mostly concertos, that came into the possession of Charles Jennens,  well-known as the librettist for Handel’s Messiah and other works;  later they passed to the music libraries of the earls  of Aylesford and Sir Newman Flower.  It was Jennens who had acquired the music from Italy and  had it bound in its existing volumes.  Earlier, in its unbound state, the collection had almost  certainly been part of a much larger corpus:  the music amassed over many years at Cardinal  Ottoboni’s court, sold off after the illustrious patron’s death in 1740.  The diverse contents of the concerto collection suggest that  Ottoboni’s musicians acquired and performed music from artistic centres  elsewhere (notably Venice and Bologna) as well  as works composed locally.
Like most pieces contained in  the partbooks, the source for the present work is a  utilitarian manuscript intended to be used in performance  rather than to grace a library shelf. Catalogued as item 28,  it belongs to the largest 
subset of the collection:  43 works copied on 
Roman music-paper by scribes  who are likely to have worked at Ottoboni’s court,  in and around the mid-1720s.(6) The eight parts’ original  
nomenclature and location within the partbooks are as follows:  
First recorder  
& oboe:		Flauto, et obuč Primo  
		volume xiv, ff. 17–18 
Second recorder 
& oboe:		Flauto, et Obuč 2.o 
		volume x, ff. 15–16 
First violin of 
the concertino:	Violino Primo Concertino 	
		volume i, ff. 58–61 
Second violin of 
the concertino:	Violino 2.o Concertino 	
		volume ii, ff. 42–45 
Violoncello of 
the concertino:	Violoncello del concertino 
		volume vi, ff. 49–52 
First violin of 
the ripieno:	Violino Primo di ripieno 
		volume iii, ff. 36–37 
Second violin of 
the ripieno:	Violino 2.o di ripienov
		volume iv, ff. 38–39	 
Basso of 
the ripieno:	Basso di ripieni 	
		volume v, ff. 15–16 
 
The wind parts are of special interest in that  each calls for both “flauto” (here meaning the standard flauto dolce,  an alto recorder) and “obuč” (oboe). This reflects the common practice  of the time whereby woodwind players were expected to perform competently on more than one type of instrument, alternating  between them as required. On rare occasions this would involve  swapping one instrument for another even within the same piece,  as the second movement of the present work demonstrates. Cues given  in the manuscript (quoted in the Textual Notes, below) indicate  the choice of instrument at every stage. If suitable personnel  are available, the solution of engaging oboists who are  capable also of playing the recorder parts naturally remains the  best one today.  
The Roman terminology (“del concertino”, “di ripieno”,  etc.), which would typically indicate the concerto grosso  disposition of an 
ensemble, does not in this case reflect the reality of  Valentini’s musical design.  (This is a traditional nomenclature  doubtless used by the Roman scribes out of habit.)  Discounting  minor errors, there is no textual distinction 
between the concertino  and ripieno string and bass parts, and indeed no solo passage to  be found, except during the third movement.  Consequently, for  movements i, ii and iv in the present edition the various parts are  combined as first violin, second violin and basso (including cello).   In those movements continuo realization of the bass part is presumably  intended, even though the violoncello del concertino and basso di  ripieni parts contain no bass figurings or any other features to  confirm that conclusion.  In the case of the third movement,  where the basso di ripieni part is silent throughout, we may be  reasonably certain that the bass phrases appearing in the cello part  (bars 1-8, 18-23 and 38-44) are not to be realized.  
Only the part for the first recorder and oboe identifies  the composer and gives a title for the work, having been  designed as a folder into which the remaining parts could be inserted  when not in use. The music itself begins on the second page (f. 17v),  following a title-page which reads: ‘Flauto, et obuč Primo |  Concerto Con V.V. [violini] obuč, e Flauti | Del Sig.RE Gioseppe  Valentini | Fogli 5,4 ÷’. Those words, like most of the manuscript’s  text, are in a notably competent, presumably professional,  hand; the same copyist was responsible for several of the collection’s  other compositions by Valentini (items 25, 51, 64 and 65). A second  scribe was responsible for writing out the Violino 2.o di ripieno  and Basso di ripieni parts. Since the manuscript employs three  particular varieties of five-stave music-paper that appear elsewhere  among the collection’s Roman manuscripts, there can be little  doubt that the two scribes were associated with the cardinal’s establishment.   
Dating almost certainly from the mid- or late 1720s, the  present concerto in A major and its close contemporaries items 51  and 64 (a concerto in F major and a sinfonia in D major,  respectively, available in this series as HH 023 and HH 025) are  among the most mature and most innovative of Valentini’s surviving  compositions. Common to all three are the absence of a viola part  (a factor not uncommon in music for Ottoboni’s establishment after  c.1720) and an inventive juxtaposition of wind and strings. They are  further characterized by their galant expression and remarkably  experimental designs: sectional movement-structures, often  exhibiting, within the framework of binary form, the  tendencies of incipient sonata form.7 Showing only a minimal use of  solo-tutti contrast and the traditional Roman concerto grosso  disposition of forces on which the composer’s earlier instrumental  works had relied, this was, in its day, a new kind of ensemble  music that anticipated the classical symphony.   
Editorial policy and matters of performance 
In this critical edition, original key signatures, time  signatures and clefs are retained. For the most part, text missing  from the source or otherwise added by the editor is shown within  square brackets. Editorial slurs and ties are given in dashed  form, except for slurs that are understood to link appoggiaturas  to their main notes, which, when missing in the source, are  added tacitly. (Such appoggiatura slurs are in fact entirely  lacking in the manuscript for the present work.) In other cases of  editorial intervention, including instances where such intervention  cannot be distinguished in the main text by typographical means,  the original readings are described in the Textual Notes.  The original notation follows the normal convention of the early  eighteenth century whereby an accidental governs only the note it  precedes and any immediate repetitions of that note, whether  barlines intervene or not. The conversion to modern notation  has thus entailed the tacit suppression of accidentals that are  redundant in today’s usage (where an inflexion,
occurring earlier in  the bar has not been cancelled) and the tacit addition of others (after  a barline, when an inflected pitch continues to apply). An  accidental omitted from the source in error is recorded in the  Textual Notes if an accidental occurring earlier in the bar remains  valid, by modern convention, for the pitch in question; it is  otherwise restored in square brackets in the normal way. Editorial  cautionary accidentals are given within round brackets; cautionary  accidentals included in the source are reproduced without brackets.  
As was mentioned earlier, the bass in movements i,  ii and iv is surely intended to be realized, even though the source  provides no figures for the continuo part. An entirely unfigured, or  only sparsely figured, bass is not in itself unusual;  professional musicians of Valentini’s day were typically expected  to realize basso continuo extempore, without the aid of figures –  especially in the case of Italian music such as this in which the  harmonic progression is mostly straightforward. Naturally, the  editorial figures included in the present edition do not provide  the only possible harmonic solution in every instance; their  purpose is to ensure that a complete text is presented and to assist  continuo players inexperienced in realizing unfigured basses.  (A version of this score with a written-out continuo realization is  available separately.) The size and complexion of the continuo  group (which may of course include, besides cello and harpsichord,  a double bass playing at sixteen-foot pitch and an archlute or  theorbo) is best determined in relation to the number of players  allotted to the violin parts. Concertos such as this were typically  intended for fairly small forces. Although an interpretation with  one player per part is in some cases the most appropriate  choice, performances of the present work that employ at least  two or three players on each string part are more likely to realize  the powerful vigour of this music. Such doubling is in any  case implicit in the fact that the present manuscript provides  pairs of virtually identical parts for first violin, second violin  and basso. The wind parts, naturally, are not meant to be  doubled – but neither are they to be regarded as ‘solo’ instruments  to be accompanied.  
The source is quite specific with regard to dynamics,  trills, appoggiaturas, slurring and staccato, presumably in  reflection of the composer’s own wishes. Accordingly, editorial  markings concerning interpretation are added mostly by analogy  with original ones. In works such as this, forte (intended to  mean a ‘normal’, full-bodied sonority) is typically assumed rather  than explicitly marked, especially at the beginning of a fast  movement and for ritornellos. Accordingly, piano means a relative  reduction from that norm, not soft in an absolute sense; only pianissimo,  when specified, requires a particularly hushed sound. Further  dynamic variation may, of course, be applied from  
moment to moment. Even so, the ideal communication of the  music of Valentini’s time 
depends far less on dynamic contrast than  on subtle variations in tempo, articulation and embellishment.  The cello solo featured in the third movment may be appropriately  graced by the player, and here, too, one finds at bars 18,  
31 and 37 opportunities for insertion of short 
cadenzas if desired.  
Paul Everett 
Cork, March 2001
 
 
1  For detailed information on Valentini’s life and works, see Michael Talbot, ‘A Rival of Corelli: the Violinist-Composer Giuseppe Valentini’, in Sergio Durante & Pierluigi Petrobelli (eds), Nuovissimi studi corelliani (Olschki, Florence, 1982), pp. 347–65; and Enrico Careri, ‘Giuseppe Valentini (1681–1753): Documenti inediti’, Note d’Archivio, 5 (1987), pp. 69–125.
2  See Enrico Careri, Catalogo dei manoscritti musicali dell’Archivio Generale delle Scuole Pie a San Pantaleo (Torre d’Orfeo, Rome, 1987).  Incipits and other details of the manuscripts (five concertos and one sinfonia) in the Saxon State Library, Dresden, are given in Paola Pozzi, ‘Il concerto strumentale italiano alla corte di Dresda durante la prima metŕ del settecento’, in Albert Dunning (ed.), Intorno a Locatelli.  Studi in occasione del tricentenario della nascitŕ di Pietro Antonio Locatelli (1695–1764) (Libreria Musicale Italiana, Lucca, 1995), pp. 953–1037: 1027–9.
3  The most recent evaluation of evidence linking Valentini with Ottoboni’s court is Stefano La Via, ‘Il Cardinale Ottoboni e la musica: nuovi documenti (1700–1740), nuove letture e ipotesi’, in Dunning, Intorno a Locatelli, cit., pp. 319–526; especially pp. 340, 361–3 and 499.
4  Thirteen volumes exist in the Central Library, Manchester, shelfmark MS 580 Ct 51.  A further partbook which completes the set, referred to below and in former literature as volume xiv, survives in the British Library, London, shelfmark RM.22.c.28.  We are grateful to Manchester Public Libraries and the British 
Library for permission to use this source for the present edition.
5  See Paul Everett, The Manchester Concerto Partbooks (Garland, New York & London, 1989).
6  See Paul Everett, ‘A Roman Concerto Repertory: Ottoboni’s “what not”?’, Proceedings of the Royal Musical 
Association, 110 (1983–84), pp. 62–78.
7  The technical features of these works are explored in Everett, Manchester Concerto Partbooks, cit., pp. 325–42.
 
  
 |