1 See Linda Faye Ferguson, ‘The Classical Keyboard Concerto: Some Thoughts on Authentic Performance’, Early Music 12/4 (1984), 437–445, ‘“Col basso” and “Generalbass” in Mozart's Keyboard Concertos: Notation, Performance Theory, and Practice’ (PhD dissertation, Princeton University, 1983) and ‘Mozart's Keyboard Concertos: Tutti Notations and Performance Models’, Mozart-Jahrbuch (1984/1985), 32–39. See also Walter Lebermann, ‘Zur Frage der Eliminierung des Soloparts aus den Tutti-Abschnitten in der Partitur des Solokonzerts’, Die Musikforschung 14/2 (1961), 200–208; Hermann Beck, ‘Das Soloinstrument im Tutti des Konzerts der zweiten Hälfte des 18. Jahrhunderts’, Die Musikforschung 14/4 (1961), 427–435; Charles Rosen, The Classical Style: Haydn, Mozart, Beethoven, revised expanded edition (New York: Norton, 1997), 192–193; Robert Levin, ‘Improvisation and Embellishment in Mozart Piano Concertos’, Musical Newsletter 5/2 (1975), 5–6; Eva and Paul Badura-Skoda, Interpreting Mozart at the Keyboard, trans. Leo Black (New York: St Martin's, 1962), 199–201; Elwood Derr, ‘Basso Continuo in Mozart's Piano Concertos: Dimensions of Compositional Completion and Performance Practice’, in Mozart's Piano Concertos, ed. Neil Zaslaw (Ann Arbor: University of Michigan Press, 1996), 393–410.

2 Other indications Mozart used were ‘Col Primo violino’, ‘Col 1mo violino’, ‘violino unisono’ and ‘unisono’. Mozart seems to have used these terms interchangeably, and did not decisively switch terminology at any particular point. These indications are not limited to this context, for Mozart and others regularly used phrases like ‘col basso’ and ‘unisono’ in their scores to avoid writing out a doubled line more than once.

3 The Violin Concerto No. 3 in G major, k216, is an exception. For this concerto Mozart leaves the solo staff of the opening ritornello blank (as in his other violin concerto autographs), but fails to write in the doubling indication. It does appear for the parallel passage in the recapitulation, however (bar 94), as well as in other places throughout the movement. The absence of the col violino instruction for the opening ritornello seems to have been an oversight, taking into account the blank staves, the parallel passage and the consistent use of this instruction in the other four concertos. The violin concerto volume (V/14/1 (1983), ed. Christoph-Hellmut Mahling) of Wolfgang Amadeus Mozart: Neue Ausgabe sämlicher Werke, ed. Internationale Stiftung Mozarteum Salzburg (Kassel: Bärenreiter, 1955–1991; hereafter NMA) prints the tutti doubling in the opening ritornello. Henning Bey gives the reasons above as justification in NMA Kritischer Bericht (hereafter KB) V/14/1 (2005), a/24.

4 André and others had the ability to print small-sized notes, and did so to indicate cues. See the examples given in my ‘To Play or Not to Play: The Soloist's Role During Tutti Sections of Mozart's Concertos for Strings and Winds’ (PhD dissertation, University of Minnesota, 2008), 63–77.

5 These instructions were carried out in manuscript parts as well, as can be seen in late eighteenth-century sets of parts for k207, k216 and k218 from the estate of Heinrich Henkel (1822–1899). In all three sets, tutti music appears in the solo violin part during the orchestral ritornellos. This is particularly noteworthy in the Henkel parts for k216 (US-CA: Houghton Library fMS Mus 204 (Haverlin Collection)), because the autograph score was missing the col primo violino instruction for the opening ritornello (see note 3 above).

6 I use the term ‘treble woodwind soloist’ as a matter of convenience, referring specifically to flutes and oboes. Despite the fact that sources point to a similar expectation that it would double the first violins, I do not include the clarinet or its relatives in the present discussion because the particularities of their soloist participation require greater elaboration than space provides (although objections to soloist participation in Mozart's clarinet concerto are cited later in the notes). On soloist participation and the clarinet family see Campbell, ‘Soloist's Role’, 187–241.

7 D-Rtt: Bachschmidt 26, ‘Concerto / à / Oboe Principale / II Violini / II corni in C / Violetta / è / Basso / Del: Sig: Antonio Bachschmidt’. This is one of fourteen oboe concertos by Bachschmidt in D-Rtt.

8 Also D-Rtt: Bachschmidt 26. Each part is in the same hand as the score, and the parts probably also date from 1763. The set contains single parts for solo oboe, violin 1, violin 2, viola, horn 1 and horn 2 and two copies of the basso part. Further details on the score and parts for this concerto may be found in Gertraut Haberkamp, Die Musikhandschriften der Fürst Thurn und Taxis Hofbibliothek Regensburg: Thematischer Katalog (Munich: Henle, 1981), 12.

9 When referring to specific bars for this example, I disregard repeated bars (indicated by repeat signs and/or ‘bis’) and count consecutively.

10 Other examples of the ‘complete verbatim’ type in flute or oboe concertos can be found in works by Johann Stamitz (D-KA: Mus Hs 912), Wendling (D-Rtt: Wendling 2), Fischer (US-AAu: M 1040. B29 C73), Cambini (DK-Kk: mu 6207.2880 U46 (Giedde collection VIII.50)), Righini (US-AAu: M 1020. R57 C7) and Hartmann (DK-Kk: mu 6208.0483 (Giedde Collection VIII.46) and DK-Kk: mu 6207.2896 (Giedde collection IX.6)).

11 D-KA: Mus Hs 913, ‘Concerto. / a / Flaut Travers Principale. / Violino Primo. / Violino Secondo. / Viola. / Due Cornu. / e / Basso. / del Sign: Stamitz.’. This manuscript set contains single parts, all in the same unidentified hand, for flute solo, violin 1, violin 2, viola, basso, horn 1 and horn 2. For further information on this set of parts and the concerto's attribution to Johann Stamitz see Ingo Gronefeld, Die Flötenkonzerte bis 1850: Ein thematisches Verzeichnis, volume 3 (Tutzing: Schneider, 1994), number 824.

12 Other examples of ‘altered complete’ doubling may be found in the opening ritornellos of concertos by Johann Stamitz (D-KA: Mus Hs 914), Hugot (US-AAu: M 1020. H92 C73 1989), Schwindel (DK-Kk: mu 6207.2887 (Giedde Collection VIII.51)), Wendling (GB-Lbl: h.3213.j.16), Scheibe (DK-Kk: mu 6304.2471 (Giedde Collection VIII.36b)) and Scherer (DK-Kk: 37 mu 6304.2469 (Giedde collection VIII.37)).

13 This contradicts Edward R. Reilly, who after pointing out that ‘in a number of Quantz's concertos the flute part is written out as a doubling of the first violin part in the opening ritornello of both fast and slow movements’, concludes ‘the fact that occasionally the violins' double stops are included in the solo flautist's music suggests that such passages were probably intended as cues’. See Johann Joachim Quantz, Versuch einer Anweisung die Flöte traversiere zu spielen (1752), trans. Edward R. Reilly as On Playing the Flute, second reprinted edition (Boston: Northeastern University Press, 2001), 202, note 1.

14 Other examples of ‘incipit’ participation in the flute and oboe literature can be seen in Hasse (B-Bc, photocopy in US-IO (Voxman collection)) and Vanhal (DK-Kk: mu 6207.2882 U46 (Giedde collection VIII.50)). A rare example of ‘excipit’ participation, where the soloist is given only the last few bars of a ritornello, can be found in Graf (DK-Kk: mu 6304.2371 (Giedde collection VIII.22)).

15 DK-Kk: mu 6306.1950 (Giedde collection VIII.51), ‘Concerto / â / Flauto Traverso. / Violino 1mo & 2do / Viola è Basso. Corno 1mo è 2do / Composta dal Sigr Fils’. This set, from the later eighteenth century (see note 16 below), includes single parts by an unidentified copyist for solo flute, violin 1, violin 2, viola, basso, horn 1 and horn 2. While the parts appear to be notated in the same hand (the mismatch between the appearance of the crotchet rests in the flute and violin parts visible in Figure 3 does not happen again in any of the other parts or later in the work; the copyist's usual crotchet rest is as it appears in the flute part), there are inconsistencies in the presentation of the tempo designations. The first movement, for instance, is marked in the first violin part as ‘allegro’ (with a small A) and in all of the others as ‘Allegro’ (capital A). The third movement designation appears as ‘Allegro assai’ (violin 1, violin 2, basso, horn 1), ‘allegro assai’ (flute, viola) and ‘Allegro Assai’ (horn 2). Other details, such as the tail on the letter ‘o’, remain constant. Of greater import, however, is disagreement among the parts regarding the metre. The flute part is notated in 2/4, the string and basso parts alla breve, and the horns in common time (4/4). Erasure of a time signature is evident in the flute, violin 1 and violin 2 parts. Apparently, this set of parts was either copied by similar but different hands, or by the same hand at different times or by a single, remarkably inconsistent copyist.

16 The Giedde collection at the Royal Library of Copenhagen is named after its founder, W. H. R. R. Giedde (1756–1816), and contains 1, 230 items, some 655 of which are printed. The rest are in manuscript, probably purchased from German or Dutch music dealers. None of the copyists has been identified, but the manuscripts were acquired in the later eighteenth century. See Inge Bittmann, Catalogue of Giedde's Music Collection in the Royal Library of Copenhagen (Copenhagen: Egtved, 1976), 5–6. The entire collection is available online at <www.kb.dk>.

17 It is curious that rather than finishing out the phrase so as to end on the first beat of bar 13 (counting the bars as given in the flute part), the descending semiquavers of bar 12 are left unresolved. Despite what appears in the part, in practice a flautist would surely play the missing g¹ and rest thereafter.

18 Manuscript set of parts, DK-Kk: mu 6304.2373 (Giedde collection VIII.25), ‘Concerto / à / Flauto Traverso / Violino 1mo & 2do / Corno 1mo & 2do / Viola / & / Basso. / Dall Sigr: Filtz’. The seven single parts are all in the same unidentified hand. As with all of the Gieddes collection manuscript copies, this is likely to be from the late eighteenth century. Other examples of the ‘split’ can be found in concertos by Fiala (D-Rtt: Fiala 4), Baumgarden (GB-Lbl: H.102 (3)), J. C. Bach (D-Au: HR III 4½ 2º 503), Cambini (DK-Kk: mu 6207.2881 (Giedde collection VIII.50)), Vanhal (DK-Kk: mu 6207.2883 U46 (Giedde collection VIII.50)) and Scheibe (DK-Kk: mu 6304.2472 (Giedde collection VIII.36a)). ‘Unaltered splits’ can be seen in Hoffmann/Haydn (D-B: Mus.ms. 10724/25), Benda (DK-Kk: mu 6301.0976 (Giedde collection VII.1)), Vanhal (DK-Kk: mu 6304.2368 (Giedde collection VIII.10)), Wendling (F-Pn: K.807), Cambini (DK-Kk: mu 6304.2367 (Giedde collection VIII.9)), Graf (DK-Kk: mu 6306.1850 (Giedde collection VIII.50) and DK-Kk: mu 6306.1852 (Giedde collection VIII.50)), Hoffmeister (DK-Kk: mu 6208.0396 (Giedde collection VIII.19) and DK-Kk: mu 6208.0395 (Giedde collection VIII.18)) and Klöffler (DK-Kk: mu 6304.2463 (Giedde collection VIII.30)).

19 The same cannot be said for the earlier half of the century. Vivaldi and J. S. Bach often had important soloistic interjections during the tutti sections, for instance in rv444, rv428, bwv1054 and bwv1056. Examples from the oboe literature can be found in Alessandro Marcello's Oboe Concerto in D minor (1717) and Tomaso Albinoni's Op. 7 and Op. 9 concertos (1715 and 1722 respectively). Albinoni's procedure in his solo oboe concertos, called a ‘devise’ by Michael Talbot (‘The Concerto Allegro in the Early Eighteenth Century’, Music & Letters 52/2 (1971), 171), was for the opening ritornello to be interrupted by a preliminary statement of the soloist's head motive, which is followed by the conclusion of the opening ritornello.

20 DK-Kk: mu 6306.1851 (Giedde collection VIII.50), ‘II. / Concerto. / â / Flauto Traverso / Violino 1mo è 2do / Corno 1mo è 2do / Viola. / è / Basso / Composta dal Sigr F. H. Graaf’. The manuscript set contains single parts, all in the same unidentified hand, for solo flute, violin 1, violin 2, viola, basso, horn 1 and horn 2.

21 Examples of similar simplifications can be seen in concertos by Westerhoff (DK-Kk: mu 6208.0480 (Giedde collection VIII.43)), Hartmann (DK-Kk: mu 6208.0484 (Giedde collection VIII.47)), Heinichen (DMÜu, photocopy in US-IO (Voxman collection)) and Jacques Loeillet (D-ROu, photocopy in US-IO (Voxman collection)). The flute abandons its verbatim doubling of the first violin for a brief time and is treated as an orchestral wind (sustaining a chord with the horns) in J. C. Bach's Flute Concerto in D major (D-Bds: Mus. ms. Bach P393).

22 Adolf Layer, ‘Graf: (3) Friedrich Hartmann Graf’, in The New Grove Dictionary of Music and Musicians, second edition, ed. Stanley Sadie and John Tyrrell (London: Macmillan, 2001), volume 10, 263–264.

23 These scores are shelved together at D-Bds: Mus. Ms. Bach P356.

24 B-Bc: 5520 MSM and B-Bc: 5519 MSM respectively. Michel's parts were prepared in 1792, but the autograph scores are from 1765. For a full discussion see Janet Page's Introduction to Carl Philipp Emmanuel Bach: The Complete Works, series 3, volume 5, Oboe Concertos, ed. Janet Page (Los Altos: Packard Humanities Institute, 2006), especially xiii. In her edition Page follows Bach's autograph instructions but gives the tutti doubling in small notes, and out-of-range pitches appear in brackets.

25 The role of the copyist can be illuminated further by examining all extant solo parts for a single concerto with the aim of evaluating to what extent they agree, and by comparing concerto solo parts by different composers prepared by the same copyist. I am grateful to one of the anonymous readers of this article for suggesting these avenues for further research.

26 Quantz, Versuch, 202.

27 See Horst Augsbach, Johann Joachim Quantz: Thematisch-systematisches Werkverzeichnis (QV) (Stuttgart: Carus, 1997), 143–237.

28 Only part of Quantz's Versuch concerns flautists exclusively; a great deal of the treatise is aimed at a more general audience of instrumentalists, including music directors. The excerpt quoted, however, is from a section specifically addressing flute players (‘What a Flautist Must Observe if He Plays in Public Concerts’), suggesting that soloist participation in tuttis was within the purview of the performers, further highlighting the latitude they apparently had where this practice was concerned.

29 Johann George Tromlitz, Ausführlicher und gründlicher Unterricht die Flöte zu spielen (1791), trans. Ardal Powell (Cambridge: Cambridge University Press, 1991), 323.

30 Paris: L. Hue, 1741. Also London: Walsh, 1750.

31 Rameau also allows the viol to be replaced by a second violin, for which a dedicated part is supplied.

32 See Campbell, ‘Soloist's Role’, 123–128, for a full discussion and examples.

33 Two chords are given as examples, each followed by an arpeggiation. The first chord is within the flute's range and its arpeggiation requires no alteration. The second, however, has c¹ as its lowest note, which Corrette's arpeggiation transfers up an octave (since c¹ would be out of the flute's range, d¹ being the usual lower limit). He then notates a chromatic line descending from d♭² to g¹ (including all enharmonic equivalents), labelled ‘Flute’. On the same staff, directly beneath each corresponding flute pitch, Corrette notates a descending line one octave lower (from d♭¹ to g) and gives it the caption ‘Etendue de la 4^e Corde du Violon’.

34 DK-Kk mu. 6210.2528 (Giedde Collection I.16). The Solfeggi manuscript may be viewed online at <http://img.kb.dk/ma/giedde/gs01-16m.pdf>. See also Bittmann, Catalogue, 15, 98.

35 Winfried Michel and Hermien Teske, Solfeggi pour la flute traversiere avec l'enseignement, par Monsr. Quantz (Winterthur: Amadeus, 1978).

36 Michel and Teske, Solfeggi, ii–iv.

37 Presumably 1782 was chosen as the terminus ante quem because it was in that year the Solfeggi appeared in music dealer Johann Christoph Westphal's catalogue. Augsbach's evidence for overturning Michel and Teske's assertions is presented in Johann Joachim Quantz: Thematisches Verzeichnis der musikalischen Werke, Werkgruppen QV2 und QV3 (Dresden: Sächsische Landesbibliothek, 1984), v–vi. A briefer account is given in Augsbach, QV, xii.

38 Steven Zohn, ‘New Light on Quantz's Advocacy of Telemann's Music’, Early Music 25/3 (1997), 443.

39 See, for instance, Claire A. Fontijn, ‘Quantz's unegal: Implications for the Performance of 18th-Century Music’, Early Music 23/1 (1995), 54–62.

40 A complete list is given in Michel and Teske, Solfeggi, 95.

41 According to Ulrich Leisinger and Michael Rautenberg this concerto is an earlier version of the Keyboard Concerto in D major h416/wq13, composed in Berlin in 1744. To them, ‘the light texture of the accompaniment and the melodic contour of the soloist's part make it clear that the concerto was originally conceived for the flute’. Leisinger and Rautenberg base their modern edition of this flute version on two eighteenth-century copies in the music archives of the Sing-Akademie zu Berlin (S.A. 2584 and S.A. 4845) on deposit at the Staatsbibliothek zu Berlin. One of these sources contained verbatim doubling of the first violin by the flute soloist during the ritornellos, which is retained in their edition. See the preface to Carl Philipp Emanuel Bach, Concerto for Flute, Strings and Basso Continuo in D Major, Wq 13, ed. Ulrich Leisinger and Micael Rautenberg (Wiesbaden: Breitkopf & Härtel, 2003).

42 This passage is used in h416/wq13 as a retransition, setting up the recapitulation at bar 99. On C. P. E. Bach's formal procedures see Jane R. Stevens, The Bach Family and the Keyboard Concerto: The Evolution of a Genre (Warren: Harmonie Park, 2001), 136–147 (note 44 discusses this concerto).

43 Concerto in G major for Flute, k313 (283c); Concerto in D major for Flute, k314 (285d); Concerto in C major for Oboe, k314 (285d/271k).

44 See Campbell, ‘Soloist's Role’, 133–186, for an examination of these sources and the soloist participation indicated therein.

45 In NMA V/14/3 (1981), viii–ix, Giegling specifies only autumn 1778, but the dating is narrowed to November in NMA KB V/14/3 (1986), c/41.

46 See Robert L. Marshall, ‘Clues to Mozart's Creativity: The Unfinished Compositions’, in The Pleasures and Perils of Genius: Mostly Mozart, ed. Peter Ostwald and Leonard S. Zegans (Madison, CT: International Universities Press, 1993), 146–151.

47 GB-Cfm: MuMs. 607. A facsimile of the score is in NMA X/30/4 (ed. Ulrich Konrad (2002)), 37–40 (Fr. 1778c). It is transcribed in NMA V/14/3, 167–174.

48 This additional autograph for bars 62–70, now lost, is transmitted by a facsimile in Oskar Fleisher, Mozart (Hofmann: Berlin, 1900), ii. It contains music for only the first violin and solo oboe, which consists of a two-bar cadential flourish in the violin (punctuating the end of the soloist's first thematic statement, and identical to the flourish that ends the opening ritornello), followed by another thematic statement by the soloist (still in F major). For further detail about this concerto and its sources see Giegling, NMA KB V/14/3, 40–41, and Konrad, NMA X/30/4, 231–232.

49 This opening pair of phrases does not return in what remains of this concerto, so it is impossible to tell what would have become of them.

50 A number of sources that point to the leadership role of a violin concerto soloist are discussed in Campbell, ‘Soloist's Role’, 63–77. For eyewitness accounts of participation by solo violinists in tuttis in the early nineteenth century see Ferguson, ‘Col basso’, 207–209.

51 Quantz, Versuch, 209.

52 See, for instance, Rosen, Classical Style, 191–193.

53 Dexter Edge qualifies his claim that ‘it is virtually certain that soloists in eighteenth-century Viennese concertos had the option of playing and probably were expected to play during tuttis’ by adding that this would not have been the case when the timbre ‘contrasted sharply with the other instruments of the ensemble’. As examples he cites the harp, the mandolin and, ‘to a lesser extent’, the oboe (‘Manuscript Parts as Evidence of Orchestral Size in the Eighteenth-Century Viennese Concerto’, in Mozart's Piano Concertos, ed. Zaslaw, 445). Colin Lawson's view is that ‘one of the distinguishing features of a classical clarinet concerto is that solo involvement in tuttis has a far greater influence on tone colour than is the case with concertos for instruments such as the violin or bassoon’. And, although he concedes that Anton Stadler may have played along with the opening of Mozart's Clarinet Concerto, k622, ‘at the very opening (perhaps eight bars or so), in the concluding bars of the first movement, and at the very end of the work’ (perhaps for leadership purposes), he is firm that ‘solo participation in the adagio tuttis would clearly detract from the dialogue which lies at its very heart’ (Mozart: Clarinet Concerto (Cambridge: Cambridge University Press, 1996), 78). This is despite the fact that the available source evidence for K622 points very clearly to soloist participation (see Campbell, ‘Soloist's Role’, 187–218).

54 Charles Rosen views the tutti doubling indicated in the sources for k622 as standing ‘in flat contradiction to all that we know of Mozart's delicacy and tact in doubling the string parts with wind instruments, [if one takes the tutti doubling at face value] we would have to believe that whenever the clarinet is not playing solo, it incessantly doubles the first violin part throughout; this doubling is, of course, nothing but a system of cueing’ (Rosen, Classical Style, 192). Rosen is correct that colla parte woodwind scoring is unusual for Mozart in other circumstances, but I would argue that soloist participation during tuttis should be considered a separate issue, especially given its practical benefits.

55 Edge, ‘Manuscript Parts’, 446. Edge's focus is on Viennese concertos, but his comments may profitably be taken more broadly.

56 Again, as the sociology and aesthetics of the concerto change, so does its performance practice. In fact, to some writers, soloist participation during tuttis seems to have been distasteful. The practice is parodied in the 1852 article ‘How to Play a Grand Flute Concerto’ by the fictitious author C. Sharp, who, in giving advice on how a concerto ought to go, encourages a solo flautist during the ‘introduction’ (opening ritornello) to ‘occasionally, only occasionally, … put the flute to your lips, and play a bar or two of it, just to show the folks you could play the introduction, if it was not “infra dig” [beneath one's dignity], and you happened to be in the humor; however, let that pass’ (Musical World and New York Musical Times 4/13 (27 November 1852), 196). Original italics.

57 To name one eminent example, this can be observed in the wind concerto volume of the Alte Mozart-Ausgabe (Wolfgang Amadeus Mozarts Werke: Kritisch durchgesehene Gesamtausgabe, series 12, part 2, Concerte für ein Blasinstrument und Orchester, ed. Johannes Brahms and others (Leipzig: Breitkopf & Härtel, 1881)). On page 9 of the 1886 Revisionsbericht for this volume, Ernst Rudorff notes the out-of-range notes in the tutti doubling for K313 found in its source, but, as he finds any ‘practical meaning’ of the notation to be ‘impossible’, soloist participation is omitted from the edition.

58 A vexing question is whether this practice is transferable to modern instruments or is applicable only to period-instrument performances, where orchestras are smaller and the woodwind tone is less brilliant.