PART I—
A HISTORY OF ITALIAN MUSIC INCUNABULA

I—
Introduction

Music notation placed such difficult demands on the new craft of printing that its initial appearance in an impression from metal type came a full twenty years after the first printed alphabetic texts As early as 1457, printers in Mainz had designed a psalter to include music, but they dealt with the problem of printing notes and staves by inserting blank spaces where the notes and staves of prescribed plainchant would be added by hand By the 1470s a handful of printers were able to produce a font of metal music type and use it to print music exactly on any line or space of a four or five-line staff By the end of the century dozens of such types for both plainchant and mensural music had appeared in the notational styles of different geographic areas Music incunabula are here defined as those incunabula or fifteenth-century printed books that contain printed notes and staves, printed notes, printed staves, or blank spaces for the insertion of manuscript music While books containing blank space for music might not seem appropriately

classified as music incunabula, that category has often been included in such studies as Meyer-Baer's Liturgical Music Incunabula, because the books may exist in copies with music added by hand or printed in a second shop In any case, the present study is concerned primarily with those books that contain printed music Books with space for music receive little discussion, but they are included in the Descriptive Bibliography in Part III Music incunabula may seem at first to be only a small part of the total estimated 25,000-30,000 incunabula That total, however, includes many small pamphlets and single leaves The 156 Italian music incunabula and the 200 to 300 printed elsewhere in Europe are most often impressive volumes of folio or large folio size, printed in two colors, with woodcut illustrations The largest of them, the 1499 Graduate (D 17), is thought to be the largest book printed in the fifteenth century.' Music was printed in three kinds of incunabula: liturgical books for the Catholic church, books of music theory, and books containing secular music for performance Most music incunabula are found in the first category The major services of the Catholic church were the Mass and the canonical hours of the Divine Office.2 By far the most common music incunabulum is the missal, the book of the textual and musical portions of the Mass to be read and sung by the priest In contrast, fifteenth-century printers rarely took up the challenge presented by books of elaborate plainchant performed

by the choir at Mass (the gradual) or at the Divine Office (the antiphonal) The breviary, the book of prayers of the priest for the Divine Office, rarely contained music Other liturgical books printed with music in Italy in the fifteenth century include collections of several special services such as baptisms and funerals under such titles as Agenda, Liber Catechumeni, Ordo ad Catechumenum Faciendum, Rituali Romanum; books for single services, such as Liber Baptismalis, Manuale Baptisterium; books for particular officiants, such as the Pontificate for use in cathedrals by a bishop; processionals; and psalters Most Italian printed liturgical books followed the Roman rite, established for the diocese of Rome Other music incunabula for monastic or local uses, modifications of the standard rite codified by cathedral chapters or monastic orders, include i Frederick R Goff, "A Few Footnotes to Konrad Haebler's Handbuch der Inkunabelkunde, " in Essays in Honour of V Scholderer, ed Dennis E Rhodes (Mainz: Pressler, 1970), p I78 2 For definitions of kinds of service books, see the glossary.

MAP I Approximate distribution of plainchant notation in Europe at the advent of printing.

fifteen (litanies, missals, psalters, rituals) for the Ambrosian rite celebrated in Milan and some neighboring dioceses, and books for local uses in Italy (Aquila, Messina), France (Auch, Besancon, Clermont-Ferrand, Nantes), Spain (Burgos, Segovia, Toledo, Valencia), Hungary (Pecs, Esztergom), and England (Salisbury). Italian music incunabula were published for use by several monastic orders: Augustinian, Carmelite, Franciscan, Benedictine, and Dominican.3 The major determinant of the design of early printed books was the design of the manuscript copytext. So too for music: the look of early printed products conformed to the appearance of manuscript books. The sizes of liturgical books were usually large or _olio format for the priest at the altar and extra large folio for the choir to share at the music stand, although the introduction of printing was to revolutionize standard sizes and make the small missal and regular folio choirbook common by the sixteenth century. Music bookhands in use at the point of transition from the manuscript to the printed book were selected according to the function of the text and the geographical location in which it was to be read. Scripts for plainchant (the monophonic music of the Catholic liturgy) and for mensural or measured music (liturgical or secular, monophonic or polyphonic) and tablature systems for notating finger positions of stringed or keyboard instruments were all in existence. Red ink for

plainchant staves and for the rubrics of service books was standard at the introduction of printing, as were liturgical points of decoration such as the initials that began services on major feasts and the illustration of the Crucifixion scene that preceded the Canon of the Mass. A review of the major characteristics of manuscript music books will provide a foundation for discussing their imitators in print. Plainchant notation could be roman, gothic, ambrosian, or byzantine. The choice of script was a function of geographic and political boundaries, as we have said. although the inroads of the republic of Venice into the Byzantine Empire and the accompanying establishment of Benedictine monasteries there blurred the eastern boundaries of roman plainchant more than is generally realized. Map i, with its division of Europe into areas using different chant notation, shows clearly enough why most music printers in Italy chose to use roman plainchant type exclusively. It was a strictly commercial decision. The map also pinpoints the Ambrosian enclave around Milan that encouraged music printers of that area to design plainchant type for the local rite.4 Only one gothic plainchant font was used in Italy in the fifteenth century, that of German printer Johann Hamman, used in an Agenda Pataviensis for export to Passau across the Alps in Bavaria. Since the area of "Italy" during the time period under consideration does not conform to that of the twentieth century, it will be useful to clarify its

fifteenth-century political and geographical boundaries. In the later fifteenth century, "Italy" was a term used to denote an assemblage of independent states, joined in an uneasy equilibrium by the Peace of Lodi (i454), with boundaries still subject to fluctuation (see Map 2). There were three main states in the north: the Republic of Venice, the duchy of Milan, and the city-state of Florence, by then a territorial state. The Papal States in the center of the peninsula were a significant power. Below them were the two kingdoms of Naples and Sicily, separate during the second half of the century. In addition there were a number of smaller states, including Mantua under the House of Gonzaga, the two Tuscan republics of Siena and Lucca, and the territories of the House of Este: Modena, Reggio, and Ferrara, the last a feudatory of the papacy. The Republic of Venice was at the height of its expansion on the mainland, extending as far west as Brescia and Bergamo in Lombardy, and in the east into the coastal regions of the upper Balkan peninsula. Since the fall of Constantinople in I453, the Turks had been a constant menace to the island holdings of Venice in the Mediterranean, including Crete and Cyprus, and the Genoese fleet had successfully challenged Venetian naval supremacy in

the Adriatic and the Mediterranean. For the purposes of this book, Italy will be defined as including the territory of the modern nation plus that part of Yugoslavia which belonged to the fifteenth-century 3. For a directory of Latin forms of place-names and monastic orders used in Italian music incunabula for liturgical uses, see Appendix 3. 4. For a definition of the boundaries of that enclave, see the map by Michel Huglo in Fonti e paleografia del canto ambrosiano, Archivio Ambrosiano 7 (Milan: Scuola Tipografica di San Benedetto, 1956), p. iii.

MAP 2 Europe at the advent of printing.

Republic of Venice and in which the Missale Romanum in the Glagolitic language (22 II 1483; see Part III) is thought to have been published The basic neumes (from the Greek neuma, "note" or "sign") of the plainchant notations that were cut into metal type in Italy in the fifteenth century are illustrated in Table i The last column provides a transcription of the neumes into modern notation The Latin names and the shapes of the simple and compound neumes provide a point of reference for the discussion of written or printed neumes in later chapters The information encoded in neumes serves two functions: coordination of melodic inflections with syllables of text, and identification of the directions of melodic movement within the inflections represented by the neumes The earliest notation meticulously observed the function of syllable marker, but since neumes were not drawn on a staff it lacked precision for specifying melodic movement from note to note.6 By the end of the fifteenth century, at the other extreme, less careful scribes began to ignore the primary role of plainchant notation as syllable marker and to separate neumes into isolated virgas that functioned solely as pitch indicators This process would accelerate with the advent of printing While a few exacting printers, especially Johann Emerich in his 1499 Graduale (see Fig 7), managed to maintain the visual appearance of compound neumes and to print them carefully over the appropriate syllables of text, others such as Dionysio de Odo in his Ordo ad Catechumeni Faciendum used two or three type designs and carelessly distributed the notes over the text (see Fig 6i).

Within a single plainchant notation, variation could occur to challenge the designer of music type Each style of plainchant notation has a common note that represents the basic time value of a chanted syllable of text (see Fig 1) In roman notation of the fifteenth century the common note could be one of two basic forms, the stemmed virga used in much of Italy or the punctum used in Spain, southern Italy, and part of France.7 A notehead could be sharply pointed at the top and bottom or smoothly right-angled, straight on each side or gently curved, provided with a long or a short stem or no stem at all, filled in with ink or left hollow The Italian printer working for Spain or France could vary the fonting of his type to increase the number of stemless characters and nearly eliminate stemmed ones Gothic plainchant notation was used in Italy during the fifteenth century only at the far north and east, along the borders of the Republic of Venice, and in such occasional strongholds of disparate liturgical tradition as the patriarchate of Aquileia, centered in the old Roman city at the head of the Adriatic that for centuries rivaled the power of Venice.8 The lozenge was the common note of gothic notation, and its style varied just as did the square of roman notation The missals printed for

Hungary in the fifteenth century with staves and without notes contain a variety of styles of manuscript gothic notation, illustrating the range of styles in use (see Figs 22 and 26) The significance of the majority of neumes is restricted to melodic motion, but certain signs for liquescence within neumes indicate the use of ornaments in performance The decoration signified by liquescent neumes is rarely needed in the syllabic plainchant of liturgical books for the celebrant (missals) but is essential in books for the choir (graduals, antiphonals, processionals) By the late fifteenth century, the degree to which liquescence was actually written out varied from scribe to scribe While many liquescent neumes had disappeared from plainchant manuscripts, the appearance of the cephalicus and epiphonus in fonts of FIG I Graphic forms of the basic time value in roman plainchant 5 A survey of the notation of plainchant is contained in Willi Apel's Gregorian Chant (Bloomington, Indiana: Indiana University Press, 1958), pp 99-132.

6 Recently discussed in Leo Treitler, "The Early History of Music Writing in the West," Journal of the American Musicological Society 35 (1982), 2457 Peter Wagner, Neumenkunde: Palaographie des liturgischen Gesanges, 2nd ed (Leipzig: Breitkopf und Hartel, I912), p 352; Johannes Wolf, Handbuch der Notationskunde, 2 vols (Leipzig, Breitkopf and Hartel, 1913), I I46 8 Francesco Spessot, "Libri liturgici aquileiesi e rito patriarchino," Studi Goriziani 35 (1964), 77-92.

Hungarian notation, gothicized by Austrian influence into a form called Messine-German, spread over an area that coincides with the borders of medieval Hungary See the introduction to Missale Notatum Strigoniense ante I431 in Posonio, ed Janka Szendrei and Richard Rybaric (Budapest, I982), pp 46-67 A name coined here to describe the combination of virga and oriscus that commonly appeared in late fifteenth-century manuscripts and printed books.

FIG 2a-b The virga cum orisco in manuscript form Antiphonal, second half of fifteenth century (Archivio di San Pietro, Perugia, Ms L.) metal type is proof of their viability in certain locales Lack of standardization in Italian manuscript plainchant music notation is also to be observed for the ornamented beginning and final notes of chant melodies Frequently a decorated longa of a kind I have called a virga cum orisco was present in forms that varied according to the mode of the chant Most Italian incunabula types include only one way of printing the design Only one printer, Christoph Valdarfer, cut both an ascending and a descending virga cum orisco; it appeared in 1482 in the first font of ambrosian type for a Missale Ambrosianum The carefully written decorated final notes in a Perugian antiphonal (see Fig 2a-b) show how the shape of the note could indicate decoration appro-

priate to the mode and melodic shape of the chant Fifteenth-century music types contain only one design for the virga cum orisco The elimination of variants may be the result of printers' efforts to simplify type fonts or may indicate that the performance practice denoted by the unusual designs was disappearing It was common for chant pieces to be begun by a soloist who would set the pitch level to be followed by the choir; the virga cum orisco often found at the beginning of fifteenth-century Italian chant pieces would have facilitated raising or lowering a pitch that was found to be too low or too high.9 Modern chant books eliminate the sign altogether Developed to notate the metrical extensions of plainchant, mensural notation uses many of the same note forms White mensural notation was used internationally wherever polyphony was written, but black mensural notation was still required for the popular mensural Credos that were a regular part of the gradual Despite the fact that Petrucci is usually credited with using the first mensural type in 150 , both white and black mensural notations were cut into type in Italy in the fifteenth century (for type specimens, see Part II, M1 and R21[M]) In mensural notation the lozenge became the basic note form, with the punctum serving as the breve (see Table 2) Many additional shapes were required of the typecutter for mensural notation:

stemmed and flagged lozenges, ligatures with upward stems at the left (ligatura cum opposita proprietate), mensuration signs, various durations of rests, and the signum congruentiae Further symbols for longer and shorter note values existed but were less frequently used and were not cut into type in the fifteenth century The interpretation of accidentals in fifteenth-century music is an area of considerable controversy Accidentals were occasionally used in manuscripts and incunabula, but a strong tradition existed for omitting them altogether, thus leaving decisions on raising and lowering tones to the performers, who would presumably have mastered a seldom-agreed-upon set of rules That the practice of musica ficta was common in plainchant can be seen in such theoretical works as the Quaestiones et Solutiones (2 June 1509) by the Venetian Johannes Materanensis and in the preface to the printed Graduale of 1499/1500 In the latter the Franciscan Franciscus de Brugis explained that he was aware that the use of accidentals varied from country to country: Gallici enim nostri: et qui Germanias incolunt omnesque alii eas circa regiones duro maxime delectantur Italici vero et cetere nationes citra illos montes molli magis alliciuntur.

We French, the Germans, and those who live in the neighboring regions delight in B natural The Italians and other nations on this side of the mountains prefer B flat." Accidentals were part of the task of the international committee of musicians selected to decide upon the text to be printed in the Graduale "in the hope of unifying and reforming the chant."'2 That the task was not easy is clear from a comparison of extant copies of the Graduale Many printed accidentals were removed in later states of the printing, while others were added by stamping after the initial printing; in some copies of the books, accidentals were later inserted in manuscript The fluid tradition of musica ficta posed a difficult problem for the printer, and it was certainly to his benefit that in the next century accidentals were generally specified This book is divided into three main parts The first (Chapters I-IV) follows the history of early Italian music printing from the point of transition from the manuscript era through several stages of printing After an introduction and a look at the international picture of music incunabula, Chapter III analyzes the techniques of casting and setting types and staves in printed music books A displaced music type in the I499 Graduale is adduced as evidence that the method of casting into type music designs of varying heights in pieces of more than one body

9 For references to soloistic intonation of chant in Italy in the sixteenth century, see Richard Sherr, "Some Remarks on Papal Patronage, Singers, and Music in the Papal Chapel in the Early Sixteenth Century" (paper read at the American Musicological Society Meeting, Vancouver, I i Nov 1985) o Edited by Albert Seay, Colorado College Music Press Critical Texts 2 (Colorado Springs: Colorado College Music Press, 1977), p 14 11 The preface is printed in Giuseppe Massera, La "Mano musicale perfetta" de Francesco de Brugis dalle prefazioni ai corali di L A Giunta (Venezia: 1499-1504), Historiae Musicae Cultores Biblioteca 18 (Florence: Leo S Olschki, 1963), pp 71-73· 12 Ibid.

Table 2 Mensural notation White Black Modern notation notation transcription (1:4)a Long Breve Semibreve Minim Semiminim Descending ligature Ascending ligature White and black notation Modern transcription Tempus perfectum, prolatio minor Tempus imperfectum Tempus imperfectum diminutum Signum congruentiae Accidentals Custos or direct Rests C clef F clef a From the evolutionary point of view, the semibreve corresponds to the modern whole note However, it is common practice to reduce the time value by half or, as here, by four to avoid an appearance of sluggishness

size was in use in the fifteenth century The chapter ends with a sketch of the role of the professional type designer and founder in the early history of music printing The various methods of printing books with music are treated in Chapter IV A first section describes those books printed with space in which music notation and staves could be added by hand Next, the books printed with staves and without notes are examined and techniques for printing staves are discussed The third section surveys Italian music incunabula printed from woodcuts The final section reviews the music incunabula printed using metal types, the printers and places important for the development of music types, and characteristics of those types Part II is organized around the thirty-eight numbered type specimens that I have prepared for each music type used in Italy in the fifteenth century The headnote explains the scheme of type classification and the method used to assemble type specimens from photographs developed to actual size These specimens provide easy access to the full range of music type designs and sizes of each printer; distinctive characters such as clefs, liquescent neumes, and accidentals can be identified and compared easily With each type specimen are given an account of the career of the printer who used the type and a list of the books in which it appears Considerable biographical detail is given for the first printers, with the intention of verifying

those facts that are significant in the history of music printing and of assessing the roles of printers (owners and directors of printing establishments), publishers (those who normally owned no typographic material but commissioned publications by those who did), type designers, and type founders Part III contains a Descriptive Bibliography of Italian music incunabula with information essential in understanding and supporting the conclusions of the book It provides a list of both those editions formally defined as Italian music incunabula (numbered items)and many editions which are not so defined (unnumbered items) either because of new dating, ascription to a place of printing outside of Italy, lack of space for printed music, or other reasons The Descriptive Bibliography resolves many questions of identity raised by the short title entries of Meyer-Baer's list of Liturgical Music Incunabula and expands that list to cover nonliturgical incunabula as well as the holdings of many more countries-in particular, Italy The description of each music incunabulum attempts to establish the ideal copy of the book as it came from the printer, based on personal examination and correlated with descriptions found in printed lists of incunabula Even where no known copy exists, editions have been included when they have been cited in authoritative

sources; as much information as is available is included in the descriptions of such works Following the descriptions are a chronological index, an index of printers, and a concordance of the Descriptive Bibliography to major incunabula catalogues Appendices include archival documents on Jacomo Ungaro, a Venetian music typefounder; a list of nonliturgical Italian music incunabula; and a directory of Latin place-names and monastic orders, to aid in understanding the titles of liturgical incunabula A glossary provides short definitions of selected important bibliographical, liturgical, and musical terms The paleotypography of music is a fledgling enterprise compared to other areas of the study of incunabula With easily accessible bibliographic descriptions and music type specimens, a beginning can be made in understanding the impact of the printing press on music and the place of music books in the work of early printers and centers of printing.

II—
Italy and the Beginnings of Music Printing

Italy played an important role in early music printing in both quantity and quality of production While current information on the amount and distribution of fifteenth-century music printing in Europe is far from exact, what is known suggests that Italy was an innovator throughout the century and dominated production of music books in the last decade just as she dominated the publishing industry as a whole The first music printed from movable type appeared about twenty years after the first European book printed from movable type, the forty-two line Bible produced in Mainz about 1454 Much as alphabetic printing had been perfected for that Bible, so the technology for printing music was in a nearly perfected state at its first appearance in the gothic plainchant of a gradual of about 1473.1 The circumstances surrounding the first music printing are even more obscure than those surrounding Gutenberg's activities in Mainz No documents are known concerning the printer of the first gradual or the location 'of his press The only evidence is the book itself, a gradual that exists in a single copy in the British Library (IB 15154) and in a fragment of seven leaves in the University Library at Tubingen The book contains no colophon naming a date or printer Because it

uses an alphabetic type identical to that of a breviary (GW 53 I5) rubricated in I473 and therefore printed about a year before that date, the Graduale has been dated about I473 The alphabetic type is known to have been used for only one other book, a psalter that survives in a fragment of two leaves (British Library, IA I5152); the music type was never used again The first music press appears, then, to have been quite short-lived The printing of the Graduale was originally assigned by Proctor to Augsburg because the only known copy of the Graduale was bound there and included services for the feasts of the Bavarian SS Laurence, George, and Ulrich.2 It has been transferred to "Germany, unassigned" in the BMC (II, 401) by Painter, who pointed out that SS Ulrich and Elizabeth are given special prominence, that the latter is not an Augsburg saint, and that the local Augsburg SS Afra and Hilarion do not appear Frequently called the Constance Graduale, it is appropriate to the diocese rather than the city of Constance, whose particular saints Pelagius, Gebhard, and Conrad do not appear Indeed, the attribution to "Germany, unassigned" may be too narrow since the saints included are important throughout the diocese of Constance, a physical entity that includes much of Switzerland as well as southern Germany (see Map 3) In his description of a promised catalog of incunabula printed in Constance, Amelung concluded that the types can only be described as belonging to an unknown printer of the inner Swabish area of the diocese of Constance.3 This

diocese, under the archdiocese of Mainz, belonged to the province of Mainz, the largest in area and population in Germany In 1435 it contained 1,760 parishes with I7,060 priests and 350 monasteries and convents and would thus have provided a good market for liturgical books I For the most recent discussion, see Alexander Hyatt King, "The 500th Anniversary of Music Printing: The Gradual of c1473," The Musical Times, Dec 1973, 1220-23 2 Proctor, vol i, part 1, 132, 1940 3 Peter Amelung, Der Fruhdruck im deutschen Sudwesten 1473-1500 (Stuttgart: Wurttembergischen Landesbibliothek, 1979), p xviii.

MAP 3 The ecclesiastical provinces of Germany, with the diocese of Constance.

The watermarks of the paper of the Graduale promise more tangible evidence of the geographic origins of the book My preliminary investigation of i he watermarks, with the help of Piccard's carefully documented work on watermarks in books in the Stuttgart Archive,4 offered somewhat contradictory evidence for a date and place The watermarks include several variants of the ox head with crown and rosette as well as a few sheets with cross-keys (in gatherings i, k, m, n, p, and q) One of the ox heads-f s3, for example-matches that described by Piccard as ox head XV, 203 from the year i477 to 1482.5 Because the mark appears in paper from Donauworth, Kirchheim unter Teck, Leonberg, Nurnberg, Stuttgart, and Ulm, he classified it as from "southern Germany." The crossed keys (f m5, for example) appear in Piccard, Schlussel, III, 1:35, and are found in printed books from Ulm and Wurttemberg during the earlier period of 1467 to I471, although the mark closely resembles Briquet 3888, where it is cited as appearing in the years 1469 to 1474 in incunabula from Augsburg and Ulm Piccard assigns the keys to Wurttemberg and Reutlingen.6 If the Graduale's ox-head watermark does indeed come from late in the decade, it provides evidence of a later date for the Graduale As geographical evidence, the watermarks are less helpful The origin of the paper in

southern Germany is clear, but that paper was used in a wide area Amelung pointed out that the Reutlingen paper of the Constance breviary was used in Basel, Switzerland, as well as in southern Germany Geldner in his 1978 review of incunabula research still questioned whether the music of the Graduale was printed from metal or wooden type, but an examination of the BL copy leaves no doubt that the music present on nearly every one of the I60 leaves is printed from movable metal type The elaborate neumes of the music type are more refined than the rather crude alphabetic type and are printed exactly on the lines and spaces of the separately printed staves, in contrast to the letters of the text, which are irregularly aligned King suggested "that two printers, of varying skill, may have been working together,"8 a practice suggested by the existence of other music incunabula in states with and without printed music (D 63 and D 68) The Graduale may be the first book with printed music and antedate its Italian dated competitors by a year or so, but its priority cannot yet be finally determined The earliest book with music notes and staves printed from movable type that is actually dated is the Missale Romanum of Ulrich Han, issued at Rome in 1476 with sixteen leaves of roman plainchant That priority cannot seriously be challenged by an earlier dated book that appeared in I473 with only one printed line of music notes without a staff Gerson's Collectorium super Magnificat (Esslingen: Conrad Fyner) contains five identical squares or

"notes" in descending sequence that, by accurate measurement of different copies, have been shown to have been printed from type in the same form as the letter text.9 While Fyner did not tackle the problem of cutting music designs into type, he did handle the challenge of printing a specific design at different heights on a theoretical staff that could be added by hand In order to print music from movable type, music printers had to resolve two fundamental problems First, an unvarying set of lines had to be printed for the staff, and, second, music notes and signs had to be cast in metal type so that they could be printed on any line or space of that staff These problems were resolved in both the ca 1473 Graduale and the Han Missale of I476 by a two-impression process A staff of horizontal lines was printed in a first impression from what were most likely printer's rules cast in metal; next, the musical designs of plainchant, cast in type so that they could be accurately printed at any position on the staff, were printed in a separate impression No other technical problems involved in printing music, such 4 Gerhard Piccard, Die Wasserzeichenkartei Piccard im Hauptstaatarchiv Stuttgart, Veroffentlichungen der Staatlichen Archivverwaltung Baden-Wurttemberg, 15 vols to date, Stuttgart: Kohlhammer, 1961-I987.

5 Piccard, Die Ochsenkopf-Wasserzeichen, vol 2 part I (1962) of Die Wasserzeichenkartei Piccard, p 346 Piccard, Wasserzeichen Schlussel, vol 8 (1979) of Die Wasserzeichenkartei Piccard, p 229 7 Ferdinand Geldner, Inkunabelkunde, eine Einfuhrung in die Welt des fruhesten Buchdrucks Elemente des Buch und Bibliothekswesens 5 (Wiesbaden: Dr Ludwig Reichert, 1978), p 124 8 "500th Anniversary," p 1223 9 For a discussion of the book and a reproduction of the five squares, see Kathi Meyer-Baer, "The Printing of Music 1473-1934," The Dolphin 2 (i935), pp 172-73·

as the use of red and black ink for liturgical music, or the use of mensural notation rather than plainchant notation, or the use of giant-size type for choirbook format, are comparable to the fundamental difficulties mentioned above The technique of printing staves and notes was perfected by the very first music printers Later printers of music cannot claim to have discovered or invented the basic technique of music printing, but merely to have bettered it The superimposition of black notes on red staves'° was considered necessary to the early printing of liturgical music, because the manuscript tradition for roman plainchant and for much of gothic and ambrosian plainchant prescribed such color coding As early as 1457 in Mainz, a printed book had used not two but three colors on the same page; for the Mainz psalters of 1457 and 1459, a method of printing in more than one color had been developed that entailed setting all material for one page in one form to be printed in three colors in one impression." The process used was too slow to be economical and could not work where red and black had to be printed at the same place on a leaf Some of the earliest printed missals left all rubrics, initials, and red or yellow staff lines to be added later (if ever) by hand (for example, one copy of the first printed missal of ca 1472, D 38, and Zarotto's

1474 Missale Romanum, D 39); the ca 1473 Graduale had a printed black staff with one red line added in manuscript In his first attempt at a printed missal in I475, Han printed red initials and rubrics in a first impression and the text in a second, black impression, but left the space for music blank In his reprint of the work in 1476, staves were printed with the rubrics and initials in a red first impression and overprinted with notes in a black second impression By the end of the fifteenth century, five different kinds of music notation had been cut into type So far as we know, they first appeared in the books listed below: i Gothic [ca 1473] Graduale, plainchant [Germany?] 2 Roman 1476 Missale Romanum, plainchant Rome, Ulrich Han 3 Ambrosian 1482 Missale Ambrosianum, plainchant Milan, Christoph Valdarfer 4 White I480 Niger, Grammatica, mensural Venice, Theodor Franck of notation Wurzburg for Johann Santritter 5 Black [I486] Graduale, [Basel, mensural Michael Wenssler and notation Jacob de Kilchen] A sixth notation, tablature, would be added by Ottaviano Petrucci in the first decade of the sixteenth century.

An analysis of current information on the amount and distribution of fifteenth-century music printing throughout Europe reveals the importance of Italy's contribution Table 3 presents figures compiled from Przywecka-Samecka's recent work on music incunabula,'3 verified and revised for Italy, Germany, and Switzerland from my own research Books with music printed from both metal types and woodcuts are included; Przywecka-Samecka omitted books with printed staves and without notes Przywecka-Samecka's work indicated that the greatest number of music incunabula came from Germany, whose total includes twelve imprints from anomalous Strasbourg, which lies within France's borders today Revision of that list to in 10 Though the normal practice was a first impression in red followed by a black impression, the black impression sometimes preceded the red, for some sheets at least An example is Planck's editions of the Pontificale Romanum (1487, I495), discussed on pp 85-86 Stanley Boorman cites as further examples Hamman's editions of the Missale Romanum of i5 X 1488 and i XII 1493 and his Missale Sarisburiense of i XII 1494, as well as the Emerich Graduale Romanum of 1499-1500: "A Case of Work and Turn Half-Sheet Imposition in the Early Sixteenth Century," The Library, 6th ser., 8 (1986): 304 n 8.

i See Chapter 5, "Red Print," in Irvine Masson, The Mainz Psalters and Canon Missae 1457-x459 (London: Bibliographical Society, 1954), pp 28-30 12 For an illustration of the notation, see Mary Kay Duggan, "The Music Type of the Second Dated Printed Music Book, the 1477 Graduale Romanum, " La Bibliofilia 89 (1987): 279, Fig i 13 Przywecka-Samecka, list of music incunabula on pp 103-26.

Table 3 Distribution, by country, of incunabula with printed notes and staves Przywecka-Samecka M-B Country Original Revised Total Printed staves Germany 105 [revised 94a 50 I I Italy 95 [revised 89b 50 9 France 46 32 24 Switzerland 22 [revised 24] 6 4 Spain 14 5 3 England 2 i I Austria I o o Unassigned o I 4 Netherlands o o I Total 285 [270] 145 57 Two additional titles may be added: a broadsheet, Ain schone Tagweis [Ulm, ca 5oo] (RISM, Das deutsche Lied, 5oo°' ), and Is ist dye ynnige geistlich bruderschafft genant S Ursulen schiffelin (Strasbourg: Knoblochtzer, ca 148 I) (RISM, Das deutsche Lied, 1481°' ) The Obsequiale Augustense (Augsburg: Ratdolt, 1499) is also listed as printed in 1489 (p 1I6) The Ordo Infirmum Ord Carthusiensis (Cologne: Landen) was printed after i5oo The Missale Misnense (Leipzig: i Ix 15oo Konrad Kachelofen) is listed three times on pp i 9 (H 11329 and H 1330) and 121 (Nuremberg: Stuchs) The Missale Romanum (Nuremberg: Fratres Ord Eremitarum S Augustini, 1491) (P 120) appears also as Missale Augustinorum de Observantia (p 12 ) Two listed variants of Stuchs's Missale Hildesemense (Nuremberg, 17 IX 1499) are of a single edition The Missale Spirense, 14 III 1487, appears on both pp 117 and 122 The Missale Benedictinum Bursfeldense (p 122) does not include printed music The 1497 edition (p 123) of Joannes Reuchlin's Scenica Progymnasmata is now considered to be a bibliographic ghost The Missale Speciale attributed to Pruss in

Strasbourg (p I 23) does not include printed music The Vigiliae Mortuorum Secundum Chorum Moguntinum, p 123, is usually dated after 500oo The Psalterium Constantiense, p 120, is dated as ca 1504 The Ordo Infirmum Carthusiensis, p 1 19, is dated as after 1500oo h The Processionarium Ord Praedicatorum (Venice: Emerich, 1493) is a bibliographic ghost The Antiphonarium Romanum, GW 2061, is dated 1503-1504 The 1484 Missale Romanum of Girardengo was issued in Venice and reissued in Pavia but is one edition The Missale Ord Praedicatorum dated 29 1 1493 is apparently a misreading of the date (see the Descriptive Bibliography) The new year begins on i March in Venice, so two of Emerich's books, the Missale Romanum of 27 1I 15oo and the Missale Ord Carmelitarum of I 500oo, were printed in i50o and are not incunabula Two items may be added to Przywecka-Samecka's list for Italy: D 16 and rD 1 5 'The Antiphonarium Basiliense (Basel: Wenssler, ca 1488) is entered again as Antiphonarium Constantiense Three items may be added to Switzerland's total: a 1486 edition of Wenssler's Graduale Romanum and two broadsheets by Sebastian Brant, Ave preclara [Basel, ca 1496] (RISM, Das deutsche Lied, 1496° ) and Verbum bonum (RISM, Das deutsche Lied, 1496 ') elude recent research, in areas such as dating, bibliographic ghosts, and multiple titles for single works, presents a picture of nearly equally important roles for Germany and Italy as leaders in music printing Meyer-Baer's figures on incunabula printed with music staves are included in Table 3 to provide a more complete picture of the work of the printers who brought music into printed books Przywecka-Samecka's list and the addition of nonliturgical editions to Meyer-Baer's earlier work on

solely liturgical music incunabula dramatically increase the total of known items.4 Of course, the discovery of copies of unverified titles, or of now unknown editions that contain music, will continue to alter the picture Until collections of fifteenth-century books are well described in printed catalogues that include identification of printed music, any description of the corpus will be inexact An analysis of the production of music incunabula by decade (Fig 3) shows how the technique of music printing spread throughout Europe By the last decade the two early leaders in the production of music incunabula were to share one-third of the editions with other countries Of 14 Meyer-Baer, Liturgical Music Incunabula (London: Bibliographical Society, 1962) Helmut Rosing's even more dramatic estimate of 1,5oo incunabula with printed music seems unlikely Helmut Rosing and Joachim Schlichte, "Die Serie A/I des RISM: Eine Dokumentation der Musikdrucke von den Anfangen bis 800oo," Gutenberg-Jahrbuch (1983): 132.

FIG 3 The production of incunabula with printed music, by decade.

course, the reduction of any data to numbers can be misleading For example, the first decade of German music printing includes a landmark gradual, but the three other items are much less impressive: a book with music printed from woodcuts and two editions of Gerson's Collectorium super Magnificat with a short musical example printed as simple squares without staff lines Italian music incunabula include eighty-nine verified editions with printed notes, seventy-six of which were printed from metal type cast in thirty-eight different fonts (see Table 4) Most were large liturgical books-two very large graduals, two pontificals, and 35 folio missals-with chant printed on a number of pages that ranged from some 40 ro 700 (the latter a gradual, D I7) The music type fonts vary in size from a few sorts (Dionysio de Odo's crude font R3o, used in an Ordo, D I43) to elaborate fonts for melismatic chant (Emerich's R21, used in D i7) While some types were monumental firsts (roman chant type, Ri in D 42; mensural type, Mi in D 142), others are so derivative in style and design as to suggest the work of a common type designer (Venetian roman chant types R4, R5, R8, etc.) The books with

music printed from woodcuts range from the ambitious forty-six pages of chant in Emerich's octavo missal of 1493 (D 93) to one staff of chant in Onate's folio Ambrosian missal of I494 (D 28) Table 4 Italian music incunabula Printed notes and staves (metal) 74 Printed notes (metal) 2 Printed staves 7 Printed notes and staves (wood) 13 Total printed music Io6 Blank space for music 34 Unverified titles I6 Total 156 The books printed with staves and without notes rely on techniques ranging from the use of irregular metal rules that provide a poor foundation for manuscript notation and an impossible one for printed metal type to sophisticated nested cast-metal segments used by printers who included printed notation in many of their other books Printing in Italy in the last quarter of the fifteenth century was concentrated in a few cities, with less important smaller presses in smaller cities and towns, and music printing followed the same pattern Table 5 lists the production of Italian music incunabula by city and technique and affirms the

importance of Venice A large proportion of books printed in Milan with space for music comes from the earliest years of music printing, before develop Table 5 Distribution of Italian music incunabula by place Not Total Printed Printed Printed City Verified Verified Notes, Metal Staves Space Notes, Wood Venice 9 76 50 14 Io 2 Milan 4 30 12 14 4 Rome 2 II 7 2 2 Brescia 6 2 4 Naples i 5 I 4 Bologna 5 2 2 I Pavia 2 i I Verona I I Parma I I Messina i i [Turin] I I [Italy]} I Total i6 i40 76 17 34 13

ment began in Venice Milan, the capital of the powerful Lombardic state and the site of a ducal court that encouraged humanism and music, was second in population in northern Italy only to Venice."5 It was also the second world center of printing,16 and its use of a liturgical rite peculiar to a small geographical area provided a strong incentive for music printing Ambrosian service books account for five of the twelve Milanese editions with printed music The large number of books printed in Venice with staves and without notes may be attributable to an early availability of metal rules and the skill of that city's printers in using them as music staves Another likely reason for the appearance in Venice of books without notes was the exploitation of foreign markets for which roman plainchant fonts were inappropriate, a problem solved not by casting gothic type but by leaving the staves blank, to be filled in by hand Only twenty-four music incunabula were printed in other Italian towns than Venice, Milan, and Rome Four books from both Brescia and Bologna were music theory books with music printed from woodcuts, several to be sold as university textbooks Parma can claim the second dated printed music in the world, the Graduale in roman plainchant notation Girardengo had strong ties to the printing trade in Venice but probably printed in Pavia all of his books with music Naples has been indicated as the probable place of printing of at least one missal with music (D 83); the three missals printed there by Moravus contain space for music, and no copy of the one by Cantono can be found today (D 97*)A comparison by city of the production of Italian music incunabula to the production of all Italian

incunabula can be made by using figures produced by Gerulaitis's analysis of the published portions of the GW (see Table 6).17 Italy was responsible for about 37 percent of all incunabula, approximately the same as its 39.2 percent of music incunabula (the percentage obtained from comparing the io6 verified Italian editions with printed notes and staves to the total of 270 known European editions) Venice led other cities in production Most Venetian music incunabula were printed in the last decade of the century, at a time when European production of printed books was largest in that city Rome, prominent in the early production of music and classical literature, had declined in importance as a printing center by the time music printing began to flourish The size of Milan's share is largely due to the local demand for Ambrosian plainchant The first three cities in Tables 5 and 6, Venice, Milan, and Rome, were dominated by five printing shops which were responsible for over half the total Italian music incunabula issued and one-third of the music type fonts used in those incunabula (see Table 7) Only two books printed in Rome with mu-

I5 Francesco Malaguzzi, La Corte di Lodovico II Moro, 5 vols (Milan: Hoepli, 1913-1923), vol 4 i6 Lucien Febvre and Henri-Jean Martin, L'Apparition du livre (Paris: Editions Albin Michel, 1958), p 278 7 Leonardas Vytautas Gerulaitis, Printing and Publishing in Fifteenth-Century Venice (Chicago: American Library Association, 1976), p 64 Table 6 Production by city of all Italian incunabula compared to Italian music incunabula City % Italian % Music Venice 38 54.4 Rome I6 8.3 Milan io 21.8 Florence 9 o.o Bologna 5 3.2 Brescia 3 3.8 Naples 3 3.2 Others 16 4.5

Table 7 The most important Italian shops printing music incunabula City Editions Metal Notes Printing Shops Editions Metal Notes Fonts Rome ii 7(+I?) Han/Planck 8(+ ?) 7(+I?) I Milan 30 12 Pachel ii 5 3 Zarotto 14 6 I Venice 75 49 Hamman 4 I t 4 Emerich 18 12 4 Total I16 68 65(+ ?) 42(+ ?) 13 sic were issued outside the Ulrich Han/Stephan Planck establishment, and that music was printed from woodcuts Antonio Zarotto and Leonard Pachel printed twenty-five of the thirty incunabula that were issued in Milan, working throughout the period from 1474 to I5oo Johann Hamman and Johann Emerich of Speier, who did not enter the field until I487, still managed to print nearly half of the Italian incunabula with printed notes Their shops were equipped with music type of various sizes and included the single gothic type cut in Italy so that, with their luxury folios and practical octavos, they were able to cater to a broad spectrum of patrons, publishers, and buyers.

An analysis by decade of the statistics on production of music incunabula with notes or staves printed from metal in Italy illustrates the increasing importance of Venice and the decreasing relative importance of Rome (see Table 8) Both Rome and Milan managed a constant small growth in the face of Venice's production Cities other than Venice introduced music printing in the 1470s, a decade that saw the creation of Han's plainchant font in Rome and the Moilli brothers' giant type for the Parma Graduale of I477 In this decade of manuscript imitation, almost all the books printed by the twelve contributing printers left blank space for music By the 1480s Venice was printing most of the Italian music incunabula, with contributions by seventeen printers and publishers, though the only contribution of three of those was collaboration on one missal with space for music (D 59) In the 1490S, twenty-one printers and publishers contributed to the production of seventy-five Italian music incunabula Two Venetian printers, Emerich and Hamman, printed thirty-five music incunabula, one-third of the total Italian and half of the total Venetian production Music printing was well on its way to becoming a business of specialists as it remains to this day.

Financial support of music printing in Italy came from the same general sources as did support for all printed books Antonio Zarotto in Milan issued the first dated missal for a company that included a humanist, a noble, and a priest from Cremona The Ambrosian missal was printed at the re Table 8 Production of Italian music incunabula by decade 1472 479 1480-1489 1490- 500 Total Music incunabula verified i 5 1 74 I 4 Incunabula with space for music 13 17 4 34 Printed notes and/or staves by city Venice o 22 45 67 Milan o 6 Io I6 Rome 3 5 9 Brescia o o 6 6 Others I 3 4 8 Total 2 34 70 io6

quest of the bishop of Milan, financed by a local merchant Nicolo Gorgonzola, a priest and ducal chaplain, was the publisher of Zarottos's Ambrosian psalter of 1496 The treasurer of the Cathedral of Pavia provided the capital and a retail outlet in his house for Francesco Girardengo, another printer of liturgical books When production of music incunabula shifted to Venice after I480, capital was commonly provided by such wealthy merchants as Luca Antonio Giunta and Paganino Paganini, and the clerical role was reduced to that of editor The first book from a Venetian press was issued in 1469; by the i48os, as we have seen, Venice was the world center of the industry Before the end of the century one hundred and fifty printing shops had been established there and had printed a total of about two million copies of books, a remarkable figure in view of the fact that the population of Europe in i450 numbered only about sixty-six million (fifty-five million, according to Braudel), who were for the most part illiterate.'8 Venice was the richest city in Italy, possibly in all Europe, and the most important European commercial center of the i490s, 19 capable of providing printers and publishers with capital and a well-established network for the distribution and sale of their books With a population of about one hundred and twenty thousand,

including substantial colonies of Germans, Greeks, Jews, Flemings, French, Slavs, and mainland Italians, it was also one of the largest cities in Europe, rivaled west of Italy only by Paris Its government and the value of its currency maintained stability throughout the years covered by this study.20 Paper 18 For information on the number of books printed, see Febvre and Martin, L'Apparition, p 278; for information on population, see Julius Beloch, "Die Bevolkerung Europas zur Zeit der Renaissance," Zeitschrift fur Socialwissenschaft, 3 (1900); Fernand Braudel, The Mediterranean and the Mediterranean World in the Age of Philip II, 2 vols., trans Sian Reynolds (New York: Harper & Row, 1972), I: 394-96; Fernand Braudel, Capitalism and Material Life 1400-1800, trans Miriam Kochan (New York: Harper & Row, I973), pp 15-16 The establishment of a population figure demands the reconciliation of complex data both on the boundaries of Europe and on documentation of numbers of people there 19 Fernand Braudel, Afterthoughts on Material Civilization and Capitalism, trans Patricia Ranum (Baltimore: The Johns Hopkins University Press, 1977), p 85 20 Gino Luzzatto, Storia economica di Venezia dall'XI al XVI secolo (Venice: Centro Internazionale delle Arti e del Costume, 1971), p 214 was readily available from Treviso, Padua, and other nearby towns with river systems Venice as the world center of book publishing also enjoyed a

position of eminence in music printing, for 85 of the total of I56 Italian music incunabula were printed there Through international arrangements by local capitalists, the most important of whom was Luca Antonio Giunta, and foreign publishers who brought capital and texts for printing, plainchant books were published in Venice for France, Spain, England, Hungary, the Balkan peninsula, Sicily, and Austria The word "capitalism" has only recently been accepted as appropriate to the fifteenth century but, as Braudel points out, when capital is present-a congeries of easily identifiable financial resources-and is controlled by a group of men who preside over its insertion into the process of production, then "capitalism" is the right term for the manner in which this constant activity of insertion is carried on, generally for not very altruistic reasons.2 The history of music printing has its gifted craftsmen in the printers and typecutters, but music publishers in Italy were apt to be astute businessmen, trained in commercial or political life, who knew little or nothing about music With their backing, liturgical printing in the fifteenth century became big business Petrucci was not able to begin his efforts to exploit the market for secular music until he had the financial support of two Venetian publishers.

The facts that two printers from one city, Venice, were responsible for nearly half of the total of Italian music incunabula and that those books were printed late in the century illustrate the trend toward the selection of a very few urban centers as sites for the specialized trade of producing music books The existence of early specialists in music type design and production in Venice is suggested by music type specialist Jacomo Ungaro's claim in I513 that he had been living in Venice for forty years (see Chapter III) Available capital, raw materials, trade routes, and concentration of population with clerical music editors and customers seem to have acted as magnets for talented music printers, publishers, and type designers in Venice from 21 Braudel, Afterthoughts, pp 46-47.

1480 until the political turmoil of the early sixteenth : century forced the book trade to seek a more stable vase, During that fertile period scores of editions and revisions of local and monastic liturgical books were issued, providing texts that both stimulated and aided the reform movement of the next century The clerical editors who worked to provide liturgical texts for the approval of the Council of Trent began with annotated Venetian printed texts.22 Venice's dominance in liturgical book production ended when Rome chose Paolo Manuzio, the inheritor of the famous Venetian printing house established in 1495 by his father, Aldo Manuzio, to set up a branch in Rome to print the liturgical books of the Tridentine reform.23 The decision of the Catholic church to select an authorized printer for the centralized production of its liturgical books completed the movement toward specialization begun by open-market pressures in the fifteenth century 22 For a discussion of the reform movement see Robert F Hayburn, Papal Legislation on Sacred Music 95 A.D to 1977 4.D (Collegeville, Minnesota: The Liturgical Press, 1979), chap 3 23 Robert M Kingdon, "The Plantin Breviaries: A Case Study in the Sixteenth-Century Business Operations of a Publishing House" in Bibliotheque d'Humanisme et Renaissance, Travaux et Documents, 2d sr., 22 (1960), 135.

III—
Early Music Type and Typefounders

Music Type

Apart from the books themselves, physical evidence concerning printing and type of the fifteenth century is almost nonexistent The first crude printed picture of a printing press does not appear until about 1499 (Danse macabre [Lyons: Matthew Husz]), and the first representation of a typefounder was not printed until 1568 (a woodcut by Jost Amman in Hans Sachs's Eygentliche Beschreibung aller Stande [Frankfurt: Feyerabend]) The unlikely recovery of about two hundred and fifty pieces of early type from the Saone River at Lyons has inspired many times that number of words about their manufacture Some copper punches and lead matrices survive from a Dutch foundry of about I5oo.1 Finally, over the years, two to three dozen examples have been discovered in incunabula of accidental impressions made by displaced types that had fallen onto the locked-up form during the inking process.2 So far as I know, the only piece of physical evidence concerning incunabula music printing is the impression of a displaced type (probably a piece of spacing material) in the copy of the I499/I500 Graduale Romanum (Venice: Emerich for Giunta) in the Music Library of the University of California, Berkeley This happenstance provides important evidence to support a new hypothesis concerning the body sizes of early music type and how it was set in forms.

Printing type has its own descriptive terminology and standard size referents (the arrows in Fig 4 point to measurements of height to paper, x-height, and body size) Kerned and abutting faces, two variants of the standard face on a piece of type, are important for early music types A kerned type is one whose face is wider than its body, so that it overlaps the shoulder of the adjacent type (Fig 5) The letter f is one of the few kerned types still commonly in use, but in the i49os in Venice fonts of cursive Greek and Latin type were being cut that relied heavily on kerned sorts whose faces could be more than twice as wide as their bodies.3 Abutting types are those that have been cast or altered by filing in such a way as to have closely adjoining typefaces on bodies without shoulders Staff segments are examples of music types cast to abut each other.4 Type is the product of a process that begins with the cutting of a punch, involves the striking of i Michael Clapham, "Printing," in A History of Technology, 8 vols., ed Charles Singer, E J Holmyard, A R Hall, and Trevor I Williams (Oxford: Clarendon Press, I954-I984), 3:392.

2 For a review of the literature on displaced type, see Michael Pollak, "Incunable Printing with the Form Inverted: An Untenable Theory," Gutenberg-Jahrbuch (1973): 168-84; Victor Scholderer, "The Shape of Early Type," in his Fifty Essays in Fifteenth and Sixteenth-Century Bibliography, ed Dennis E Rhodes (Amsterdam: Menno Hertzberger, 1966), p 107 3 For a photograph of a kerned type that overlaps an adjacent type, see Philip Gaskell, A New Introduction to Bibliography (New York: Oxford University Press, 1972), p 32, fig i8 4 For an illustration of one of the staff segments of van den Keere's "Grande Musicque" (punch, matrix, and type cast from the matrix), see illus 3a, b, c of H Edmund Poole's "Printing" in The New Grove Dictonary of Music and Musicians (London: Macmillan, 1980), 15: 237.

FIG 4 The terminology of a piece of modern type and its printed image (Philip Gaskell, A New Introduction to Bibliography [New York: Oxford University Press, 1972], p 9.) the punch into a matrix, and ends with the casting of an individual piece of type by pouring hot metal into a mold to which the matrix is affixed The process was first described in a technical manual by Vanuccio Biringucci, an Italian metallurgist of the early sixteenth century: The letters for printing books are made of a composition of three parts of fine tin [stagno fino], an eighth part of black lead [piombo negro], and another eighth part of fused marcasite of antimony The desired quantities of these metals are melted and cast into bars so arranged that they can easily be cut Then a mold [formal is made of brass [ottone] or bronze [bronzo], as true as possible and flat so that ii: may fit into its companion The parts of this mold are adjusted to make the thickness and length of the stems of the letters, and likewise are adapted underneath so that the matrix [madre] can be put in exactly The matrix is the impression of the letter that you wish to make, or rather to cast, made by the steel punch [polzone de l'acciaro]

on a little piece of copper When the matrix has been put in its place and the thicknesses likewise fixed by means of certain little screws which secure and close all the pieces of the mold in their places, some of that composition is melted in an iron pot and the letters cast one at a time with a ladle When they have the desired quantity of one kind of letter, they take out the matrix and insert another, continuing in this way until they have as many as they wish of all kinds Then they cut off FIG 5 a A sixteenth-century font of music type with many kerned sorts, composed for printing a specimen b A kerned virga with a bent stem (Copyright Plantin-Moretus Museum)

the gates [gitti] and adjust them all to the same size with a mold as a little gauge.* With these letters, one by one, the compositors set the press forms for the books in a frame [telaro] made of iron, bronze, or wood, and compress and lock them in with screws in the outer edges * In the second edition of 1550 the sentence is altered to read: "Then they remove the said letters and cut off the foot, that is, the part that is to lie on the bed where the composition is placed for printing when the work is being carried out; first, however, those fins that remain attached to the letters in casting are removed one by one with a knife so that they may remain wholly clean and without any unevenness." The antimony present in Biringucci's type would have increased its hardness considerably Antimony was apparently not added to type metal until the end of the incunabula period, when it theoretically had the effect of allowing finer detail in the design of type.6 Early fonts of softer metal had short lives; many fifteenth-century music fonts appear to have been used only once.

Fragmentary evidence about fifteenth-century music printing and typefounding exists in such documents as contracts (see the discussion in Chapter VIII of Antonio Zarotto), testaments (see the discussion in Chapter IX of Francesco Girardengo and his partner Giovanni Beretta), privileges, colophons, and lawsuits involving those in the book trade Sixteenth-century physical and documentary evidence is more accessible, including the major resources of the collection of type and business records of Christopher Plantin at the Plantin-Moretus Museum in Antwerp Types, type specimens, and inventories of Christopher Plantin ( 520-I 589)gives us the names and sizes of sixteen sets of music punches, twelve sets of matrices, and one music mold.7 Those identified as the products of particular type designers are listed in Table 9 The inventory of Guillaume Le Be's sixteenth-century typefoundry listed matrices and punches for twenty-three music types, six of them plainchant (see Table io).8 ST68 is identified as "Grande Musicque" in the Appendix of Parker et al., "Typographica Plantiniana II," but on p 74 in the I58I inventory as "Grosses Notes de Musicque plain Chant de la taille de Henry du Tour." It is clearly not a plainchant type and was used first in 1578 for the polyphonic masses of

Georges de la Hele The 1581 inventory specifies thirty-nine punches, but the illustration of the extant punches uses only thirty-eight The matrices for ST68, MA9ia, have the same problem with name in the 1581 inventory The answer may lie in 5 Biringucci, Pirotechnia, trans Cyril Stanley Smith and Martha Teach Gnudi (New York: American Institute of Mining and Metallurgical Engineers, 1942), pp 374-76 The Italian words have been added from the original edition printed by Venturino Roffinello in Venice in 1540. 6 Michael Pollak, "The Durability of Fifteenth-Century Type," Visible Language 5 (197 ): 160-6 In his dissertation on Petrucci's printing ("Petrucci at Fossombrone: A Study of Early Music Printing, with Special Reference to the Motetti de la Corona [I514-1519]," University of London, 1976, p 29), Stanley Boorman proposed that Petrucci's strongest claim for innovation in printing history might lie in the addition of antimony to his type metal He cited as major differences between incunabula music type and that of Petrucci the latter's small size, lighter design, and consistent use of hollow diamond-headed notes, but those characteristics can already be seen in the first mensural type of 1480 7 The type specimens of 1567 and ca 1585 are reproduced in Type Specimen Facsimiles II: Reproductions of Christopher Plantin's Index sive Specimen Characterum I567 & Folio Specimen of c 585 Together with the Le Be-Moretus Specimen c.I599, with annotations by Hendrik D L Vervliet and Harry Carter (Toronto: University of Toronto Press, 1972) Only three types appear in Plantin's specimens and they are

all mensural types, apparently single-impression (ca 1585 specimen, nos 80-82) Several early inventories are presented in Mike Parker, K Melis, and H D L Vervliet, "Typographica Plantiniana II: Early Inventories of Punches, Matrices, and Molds in the Plantin-Moretus Archives," De Gulden Passer 38 (1960):-I139 The mold is described in Mike Parker, "Early Typefounders' Moulds at the Plantin-Moretus Museum," The Library, ser 5, 29 (1974): 98-99 8 The inventory of Guillaume Le Be's sixteenth-century typefoundry was made sometime after I617 by Le Be's son and is known today through a copy made by Jean-Pierre Fournier in 1730 when he purchased the typefoundry The inventory was published by Stanley Morison, L'Inventaire de la Fonderie Le Be (Paris: Andre Jammes, 1957), pp 22-26 See Daniel Heartz, Pierre Attaingnant, Royal Printer of Music (Berkeley and Los Angeles: University of California Press, 1969), plates 6-7, for photographs of the sections referring to music type A specimen apparently prepared by the typecutter himself with manuscript annotations includes three of the music fonts made in the I55os for Le Roy and Ballard in Paris: a mensural "Musique grosse" (i554-i555), a double-impression "Grosse tablature d'epinette," and a single-impression "Petite tablature d'epinette," in H Omont, "Specimens de caracteres hebreux, grecs, latins et de musique graves a Venise et a Paris par Guillaume Le Be (1545-1592)," Memoires de la Societe de l'Histoire de Paris et de l'Ile-de-France 15 (1888): plate 3.

Table 9 Music types of Christopher Plantin's shop Extant material at the Plantin-Moretus Museum in Antwerp: ST = punches; MA = matrices; LMA = lost matrices; PM = type; GI = molds Type Designer Type Name Punches Matrices Type Mold Robert Granjon Petit Musique sur le Gros Texte, 22, ST75 25, MAioic (I5I3-ca 1589) 1566 Musique sur 2 R Augustine, 73,ST7xa" LMA36 (single impression) Premieres Notes (double i6, ST64 MA86 impression plainchant) Hendrik van den Keere Grosses Notes d'Espaigne, I574 i6, ST6i 20, MA93 PM222 GI55 (active 1540-1580) Moyennes Notes d'Espaigne, 20, ST62 49, MA8ga PM223 GI55 1573 Grosses Notes de l'Antiphonaire, 30, ST63 MA9o '57' Moyennes Notes du Missel, 1572 35, ST65 41, MA87 Petites Notes pour les ST66 39, MA88 Processionels Petites Notes du Missel, i571 32, ST67 38, MA85 Grande Musicque, 1577 38 [39], 63, MA9Ia PM234 GI55 (single impression) ST68 Grosses Notes ST69 MA9Ib Moyenne Musique, Notes in 4° ST7o 67, MAioo MAio2 Petite Musique sur la ST73 LMAIg Parangonne 2 sets for chant in small missals, 4, MA92d 1571 4, MA92e Daniel Heartz, Pierre Attaingnant, Royal Printer of Music (Berkeley, University of California Press, 1969), p 47 The top two lines of the illustration are impressions struck from the punches, ST7 ia The caption declares the illustration to be "Specimens of

music type made from Granjon's matrices," but the text identifies it as struck off punches; the matrices (LMA36) are lost MA9 I b, eleven matrices of plainchant described by Parker et al (pp 74-75) as an "incomplete set." Would it be complete if it used the designs of MAgia? The incunabula type of Emerich, R2I, is an example of a combined plainchant and mensural font produced to handle the mensural Credo popular at the time., and ST68 may be a similar combination of mensural and plainchant type In addition, Parker et al discredit the large number of matrices by suggesting that they were mixed with the "Grosses Notes d'Espagne" (p 78), but with the manipulation of the punches necessary for staff segments needed as spacing material between, above, and below types, it is quite possible that a large number of matrices was necessary An adjustable mold can do only so much; the contract for this type specified at least three molds, one for staves (for the text of the contract, see p 26) Another inventory made in 1595 of type at the Tipografia Vaticana listed one plainchant type of thirty-two punches and thirty-nine matrices commissioned of Robert Granjon for printing the revi-

Table io Music types of the Le Be typefoundry (ca 1620 inventory) Type Designer Type Name Format Punches Matrices Pierre Attaingnant Notte de gros Plainchant portant ses 2 40 34 (ca I494 550) reglets Philippe Danfrie (ca 1504-I606) Musique des airs 16° 47 Nicolas du Chemin Musique p Messe 2 29 92 (active 154I-1576) Pierre Haultin (active I523-1580) Musique moienne, maigre, vielle 86 Nicolas de Villiers Musique des Nottes de Colin 51 67 Guillaume Le Be Notte de Plainchant Corniches 4° 5 15 (active 1545-I592) Antiph mon pere Robert Granjon (I513-ca 1589) Tablature de lut et Guitarres 85 Musique 4 50 Musique 4° 64 Types from Lyons Musique des airs plaine I6° 55 Musique petite pour airs 70 Musique de Coppie moyenne 34 Musique petite pour psaume 23 Anonymous Musique en blanc de Cicero psaumes 2 18 18 Notte de Plainchant, Grad et Antiph 2 30 27 Notte de Plainchant Process 8° 73 Notte de Plainchant heure notte I6° 62 Notte de Plainchant Missel 8° 60 Notte de gros Plainchant, neant [2°] 39 Notte de gros Plainchant courbez [2°] 30 sion of the gradual undertaken by Palestrina for Gregory XIII.9

Such evidence tells us that sixteenth-century double-impression plainchant fonts varied enormously in size, from the small sets such as Le Be's quarto antiphonal, which required fifteen punches and matrices, and van den Keere's tiny sets of four punches and matrices to quite large sets of seventy-three punches and matrices An average early plainchant type might require from thirty to forty punches and matrices The type specimens of Italian music incunabula suggest that most would require only about fifteen punches and matrices, although some were more elaborate and some were as small as van den Keere's sets Insight into the process of making early music type comes from a contract between the typecutter van den Keere and the printer Christopher Plantin The contract concerns the mensural "Grande Musicque" commissioned by Plantin in 1576 and first used in I578 In a letter of i6 January 1576, after looking over the manuscript leaf sent him by Plantin, van den Keere set his terms for accepting the contract He estimated he would need fifty punches, but thirty-nine proved sufficient and were used to produce sixty-three matrices There would

be "expense beyond the usual, because I shall have to use at least three molds But I hope I could help myself out with your molds by packing [the bodies and jets] as needs be The five staff lines answer exactly to 5 lines of the small, or new, Great Primer, by Garamont, so that its mold will be a useful 9 Alberto Tinto, "Di un inventario della Tipografia Vaticana (I595)," Studi di biblioteconomia e storia del libro in onore di Francesco Barberi, ed Giorgio De Gregorio and Maria Valenti (Rome: Associazione Italiana Biblioteche, 1976), pp 547-49; H D L Vervliet, Robert Granjon a Rome 1578-1589 (Amsterdam: Menno Hertzberger, 1967), p 33-

one."'° Measurements made of the type in a printed sample indicate that the staff for which Garamond's Great Primer mold could be used was 22.5 mm high; segments were cast in widths ranging from 2 to 25 mm." The largest of the three molds survives today at the Plantin-Moretus Museum. It was made in 573 for the "Grande Musicque" chant type cast for a projected Spanish antiphonal. The extant mold has register setting and nick that exactly fit the museum's cast "Grande Musicque" type, although the interchangeable stools are missing that would have been necessary for casting from one matrix design the sorts of type with the distinctive faces needed to print characters on lines and spaces."2 A punch would take a fine craftsman like van den Keere more than a day; even with extra hired help, only five or six matrices could be justified per day. "The punchcutting will take three months, the justification a fortnight, the molds and other preparatory work on the steel and copper a fortnight, making four months. However, with other work which may intervene, you cannot reckon on less than half a year. And it will cost quite I50 florins before you can think of casting a font, or thereabouts."" Ten months later van den Keere's bill of I I November I577 for the completed punches at 4 florins each was 152 florins, even before striking and justifying the matrices and casting the type. For a book like the 1499/1500 Graduale of Emerich, the publisher Giunta would have had to subsidize text and music types to print the manuscript copytext, which had been especially prepared

in roman plainchant notation for this first edition of the complete Graduale Romanum. Additional commissions would have been necessary for the woodcuts, initials, and unusually large paper. If a punch took more than a day and a sufficient number of men were on hand to justify five to six matrices a day, it would take about a year per font to complete punches and matrices for the two text fonts and a plainchant font designed and cut for the Graduale. The Graduale's plainchant font was larger in size and complexity than van der Keere's and was probably produced by a smaller establishment, so it would have taken longer. Michael Clapham estimated that a text font of about one hundred and fifty punches would have taken two years to produce in the early years of printing.4 It is no wonder that three years intervened between the granting of Petrucci's privilege for mensural type and the first appearance of it in print. A different kind of evidence comes from the modern impression of van den Keere's single-impression "Grande Musique" made from the type cast from thirty-eight punches extant at the Plantin-Moretus Museum.' The large type (printed on a staff of 22.5 mm) illustrates clearly the problems facing punchcutters of music type. The music signs range in size from large clefs and time signatures to small accidentals and rests. Signs that extended into the area of text type above or below the staff would also need a kerned version (see the flagged and unflagged lozenges). Signs such as

accidentals that might occur above or below another printed design or spacing material had to be cast on a small body that occupied the space between two staff lines. The direct (Latin, custos) is the symbol placed at the end of a staff to indicate the first note of the next staff. Fifteenth-century type often included an alternate narrow version for use within a line as well as the normal wide version, generally kerned, that would be set in the margin beyond the end of the staff in a double-impression type. Van den Keere avoided the kerned direct by curling back the angular line on top of the sign after the point of definition of pitch. (c The necessity for multiple punches for minims with shortened stems such as that used in Emerich's R2 is avoided by inverting the minim. io. H. D. L. Vervliet, Sixteenth-Century Printing Types of the Low Countries, foreword by Harry Carter (Amsterdam: Menno Hertzberger, 1968), p. 334. i. The measurements of the segments are mine, made from the illustrations in Vervliet, Sixteenth-Century Types, P- 33512. Parker, "Early Typefounders' Moulds," pp. 94-95, plate i6A.

13. Vervliet, Sixteenth-Century Types, p. 337. 14. Clapham, "Printing," p. 385. i5. Vervliet, Sixteenth-Century Types, M8, fig. 257, p. 336. For an example of music set with the types, see Plantin's folio specimen ca. 1585, Type Specimen Facsimiles II, and Vervliet, Sixteenth-Century Types, p. 335.

One problem of single-impression music type apparent in music printed with the "Grande Musicque" type is the need for staff segments of varying widths to be used as spacing material between notes; small segments of 2 and 3-25 mm are frequently inserted between printed notes The fifteenth-century printer who cast staff segments for double-impression type could plan their size(s) so that an easy multiple would fit the column width, without taking into consideration the sizes of the notes, which were printed separately in a second impression Spacing material between notes was necessary in the second impression It would not have been type high-that is, as high as the type cast with a face that was intended to make an impression on the printed page-and in the fifteenth century it may have been made of wood In the same way, the type-caster of a double-impression type had to prepare blank spacing material to insert above and below almost every piece of music type There is every indication that such a process was in use for early printed examples of large type such as the complex R2 used in Emerich's Graduale and Antiphonarium A final point of reference for the understanding of punches and matrices of old music type is the physical appearance of their modern counterparts A typical plainchant type in lower and upper cases in the nineteenth century included 128 sorts.l6 As the polyphonic and harmonic complexities of music increased and the details of performance practice

became specific, the size of the music type font grew to nearly 500 sorts Nineteenth-century music fonts were named by their style (chant, music) in "sizes" (Ruby, Diamond, Semi-Nonpareil), but the types were not cast on the point system In his book on music printing, William Gamble wrote that "music type has a 'point' system of its own, otherwise it would be impossible to get the various pieces of type to line up."'7 Since music types are not set on the same line as text type, their body size need not correspond exactly to the size of the text type Still present in modern fonts are the alternate kerned sorts-clefs, noteheads, flagged stems, diagonals (now used to connect groups of eighth notes but identical in design and compositorial problems to the ligature connection of early plainchant diagonals)-that are necessary for printing above or below the staff and thus intruding upon the text type Illustrations of cases of type and specimen sheets carefully note the presence of kerned sorts (unflagged stems no longer need be kerned, because they are built up from small pieces of type that allow variation in length).18

Useful for understanding music types are the variety and number of spacing material sorts for use above, below, and between type with cast typefaces One nineteenth-century music type required eleven different sizes of quads and seven sizes of spaces to set the type in forms.9 Such physical evidence of the technical difficulties in setting music type may help explain why music in print lagged twenty years behind the appearance of printed letters There was little difficulty in casting the musical signs of plainchant in type metal; the difficulty lay in casting them on bodies that would allow the signs to be printed at the proper place on the music staff According to the first detailed description of music typefounding, published in the eighteenth century by Pierre Simon Fournier le Jeune,20 double-impression plainchant type may consist of whole notes, which have a body size equal to the body of the staff and for which a single mold is sufficient, or piece notes, which require four molds for four body sizes The latter method probably more closely resembles that used for casting Venetian incunabula music type Of the fifteen punches that Fournier states were needed for either method, one, the dot for added value, would not have been used in the fifteenth century; but many more would have been necessary then, such as ligatures of various shapes and intervals Of Fournier's remaining fourteen music sorts, six are cast in the smallest mold

(the smallest body size: * * ) and five in the next size( r ).Only spacing material was cast in the third size, and the fourth and largest size was used for staves and full and par I6 Theotiste Lefevre, Guide pratique du compositeur d'imprimerie (i880; reprint Westmead, Farnborough: Gregg, 1972), p i68 17 Music Engraving and Printing (1923; reprint New York: Da Capo Press, 1971), chap i5, "Typographically Printed Music," pp i68-82; includes several type specimens and a variety of solutions to housing large fonts in typecases I8 Ibid., figs 92-93, pp 179, i8i 19 Ibid., p 179 20 Fournier on Typefounding: The Text of the Manuel Typographique 1764-1766, trans and ed with notes by Harry Carter (London: Fleuron 1930), p 58.

tial barlines Because of the short stems of many of the notes in Venetian incunabula music type, the designs of the first two sizes could have been combined, but a second group of ligatures would have needed a double-size body Staves were cast as abutting four-line segments of standard width, multiples of which would make up a column Fournier advises dispensing with special punches for partial staves, "because shorter sections of the staff are so rarely called for," but that advice was not always appropriate for a fifteenth-century music printer like Emerich, who frequently inserted rubrics of varying length in the staff area, creating irregular blank spaces in the staves Abutting types in Emerich R2 I (for the specimens see Chapter VII) include, in addition to staves, several neumes the typefaces of which may have been altered to abut adjacent text types above or below the line Such abutting variant types had been used by compositors to solve particular problems ever since Gutenberg's first book.21 The first type designers to complete the transformation of complex scripts into type used complex kerned and abutting forms that create compositorial problems (gothic text, Gutenberg; cursive Greek and Latin, Francesco Griffo da Bologna) It is only natural that many instances of variant sorts and altered type are visible in Emerich's roman plainchant font, a font that was only the second attempt to cope with the melismatic music of the gradual in roman plainchant.

Harry Carter attributed the development of music type in such a system of "building up the staff from types of various bodies" to the mid-sixteenth-century typecutters Le Be, Granjon, Nicholas Duchemin, and the two de Sanlecques; they supposedly increased the number of kerns and "arranged the joins after the manner of bricklaying."22 The existing punches and matrices at the Plantin-Moretus Museum show that such a system was functioning in the sixteenth century With the evidence of body size provided by the displaced type accidentally impressed on one leaf of Emerich's Graduale (Fig 6), a system of several sizes of types "joined like brickwork" becomes plausible also as an explanation of the complex plainchant type of the incunabula period Examination of that leaf has allowed me to develop a hypothesis of a system of body sizes for Emerich's R2 I In the copy of the Graduale at the Music Library of the University of California, Berkeley, a piece of type that had evidently been pulled out of the form by the ink balls and had fallen back on the form set with type for the music notes (the staves had already been printed in red) was printed on its side Measurement of the size of the indentation in the paper caused by the type when the sheet was printed indicates that the piece of type protruded a millimeter or two above the faces of the other types The type faces of four pieces-the

preceding clivis and barline and the following virga and cephalicus-were broken by the pressure of the displaced type on the form at the time of impression; the displaced type was apparently resting on or near the shoulders of the broken types at the time of impression A later user has added in ink the first note of the clivis and the missing virga without a stem; still missing are the second note of the clivis, the barline, and the ornamental cephalicus The same leaf without the displaced type appears in the Library of Congress copy (Fig 7; staff 2, with the unbroken clivis, barline, virga, and cephalicus) Since we do not know how deeply the type was broken by the piece of displaced type, the set width of the latter cannot be specified with any certainty However, it is possible to make a fairly accurate estimate with the help of the sixteenth-century artifacts at the Plantin-Moretus Museum Measuring a piece of type (body size of 2 mm) recently cast in a sixteenth-century matrix at the museum gives a distance of 2 mm from typeface to shoulder If this distance is considered approximately normal for the time and is added to the depth of the protrusion into the paper made by the displaced piece of type (roughly I mm), the resulting set width of the displaced type would be no more than 3 mm Since the set width of a virga in Emerich's font must be at least 6 mm, the width of the printed virga face, the displaced piece of type would not be wide enough to be a virga or a cephalicus, the characters that are missing from the impression It must therefore be spacing material The body size of the type is approximately II.5 mm and the height 33 mm (see

Fig 8) I conclude that the body size of the common note of R2 I, the virga with short stem, is I 1.5 mm and that the font would include larger body sizes for 2 I Masson, The Mainz Psalters, pp 42-43 22 The process is described by Harry Carter in "Fell Music Type," Penrose Annual 50 (1956): 73-74.

FIG 6 Graduale Romanum Venice: Johann Emerich of Speier for Luca Antonio Giunta, 1499/1500, f LII (Music Library, University of California, Berkeley fM2I49.V4G7.) such designs as the F clef and the clivis and podatus of intervals larger than a second, as well as a smaller body size for such designs as the punctum, lozenge, and accidentals (see Fig 9 for sketched-in samples of body sizes; note in Fig 9b the flat printed below a note, and the bend of the rule) In at least one place the careless inking and resulting impression of the shoulder of the type below the typeface in relief corroborates a hypothetical body size of I.5 mm On f v of the Graduale, a virga was printed with the lower lefthand shoulder of the type visible next to the kerned stem (see Fig io, stave I) The distance from the top of the virga to the printed shoulder is 1.5 mm In such a typeface, the stem of the virga would be kerned; in fact, there are examples of virga stems bent at the bottom If the hypothetical size of I i.5 mm is correct, some multiple of that number would make up the size of the staff between two lines of text type The distance between descender and ascender for the staff (the text type appears to use a face that covers

FIG 7 Graduale Romanum Venice: Johann Emerich of Speier for Luca Antonio Giunta, 1499/1500, f LII (Library of Congress, Washington, D.C., Inc 1500.c3.) FIG 8 Sketches of the displaced type in Emerich's Graduale, I499 1500, f LII (Music Library, University of California, Berkeley fM2 49.V4G7.) the surface without shoulders beyond ascenders and descenders) is 38.5 mm and the actual size of the staff is 32 mm To speak now of Carter's "brickwork," a vertical group of three pieces of I i.5 mm plus one piece of 4 mm would make up a total of 38.5 mm (see Fig i i) The normal virga or its variants for the top and bottom of the staff are cast on the body so that the sign can be printed on any line or space of the staff or above or below the staff Because of the position of the text type, the virga on the space below the staff has no stem at all size of staff 32 mm size of staff plus alphabetic type body (descender to ascender) 38 mm FIG 9 a Hypothetical body sizes of music type R2 I b A flat printed below a note.

FIG o Graduale Romanum Venice: Johann Emerich of Speier for Luca Antonio Giunta, 1499/1500, f V (Music Library, University of California, Berkeley fM2 49.V4G7.) FIG I i Hypothetical body size of virga in Emerich's R2 I The same punch and matrix may have been used to cast all three virgas The process whereby the same punch and matrix were used to cast variant sorts was described in the eighteenth century by Fournier and was found by Harry Carter to be operative for the seventeenth-century Fell music type.2 A special type mold allowed, in addition to the normal horizontal adjustment, a vertical adjustment of the matrix in various positions; the position of the matrix determined where the design would appear on the face of the type and, as a result, where the musical sign would appear on the staff An earlier example of such a mold, with interchangeable stools to alter the head alignment of the matrices, exists today at the Plantin-Moretus Museum (GI 55) It was made by van den Keere in 1576 to cast 23 Ibid.

FIG. 12. Hypothetical body size of staff type for Emerich's R2 1. the large Spanish antiphonal type, a font which included kerned sorts for notes at the top and bottom of the staff. The system may well have been developed in the fifteenth century. Certainly the identical appearance of such distinctive characters as the porrectus at different positions on the staff suggests the use of a single punch.24 Two kinds of type were needed to print plainchant between two lines of text, one set for notes and one for staves. A strategy involving varying sizes of pieces of type may have been used to print the staff as well as the notes. Occasionally the shoulder of a piece of staff type is unintentionally printed and suggests the use of three equal pieces of staff type (body size of about 0o.5 mm) plus a piece of 7 mm to fill the space next to the text type. If an extra ledger line was needed above or below the staff and reached into the space of the text type (as in Fig. I2, top right, and in Fig. io) an additional fifth piece of staff type would be needed on a body the size of the text type, or 19.3 mm. Kerning was as common in Emerich's text types as in his music types. An unusually large repertory of kerned music types appears in R2 I (see the type specimen, Chapter VII); among the kerned types visible in Emerich's text type for the Graduale are f, the long s (to overlap following vowels), a form of

the t with a long crossbar (to overlap vowels), a form of the r and a form of the c (to overlap vowels and other c's (see Fig. 13). Early attempts to transfer manuscript designs to type relied on kerning to create the visual effect of linked letters (ligatures). To be sure, a font with kerned variants was expensive and time-consuming to design and cut as well as to set; the enormous fonts of the I450s (nearly 300 sorts for a text alphabet, plus frequent filed variants) had gradually been replaced by types with few kerned sorts and ligatures. Emerich was creating one of the first music types capable of setting the complex plainchant of the manuscript gradual and antiphonal, and it was natural that he should make use of complex kerning and variant forms of the same character to achieve his aim. Just as natural was its consignment to obsolescence when new types became available that were easier to set and allowed more music on the page, even though they lacked the capacity for printing complex neumes and liquescence. 24. Parker.. "Early Typefounders' Moulds," pp. 94-95 and plate i6A; see also fig. 3 in Poole's "Printing," p. 237. FIG. 13. Examples of kerned text types in Emerich's Graduale.

Evidence supporting the use of kerned characters in types like Emerich's large plainchant font can be sought in another important music type that, though not technically an incunabula type, was being created in Venice at almost the same time In a dissertation on Petrucci's Motetti della Corona (printed from 15 14 to 15 19 with the music type first used in Venice in 15 0 ), Stanley Boorman prepared a specimen of the music font and proposed a theory of kerned types to explain the intrusion of stemmed signs into the text type.25 A frequently used note in a mensural font is the minim or stemmed lozenge, outlined or filled in depending on the meter In Petrucci's type it has a standard height of I 1.2 mm on a five-line staff of approximately io mm, so the stems frequently intrude on the text type above and below the staff Boorman noted that "the tails appear to be particularly susceptible of being bent at a distance of 3.5-4 mm from the top of the note" and proposed that a reversible sort kerned at that point allowed the typesetter to combine text type and music type with less difficulty.

Emerich's last music font appears to have used a system of music type cast in multiple body sizes from several type molds and containing kerned and abutting variants for use at the top and bottom of the staff and for creating complex neumes by combining sorts The same general system may well have been used for many of the incunabula music types We have seen evidence for this hypothesis in a fortuitous instance of a displaced type, in somewhat later use of the system in the earliest extant music types, and in Petrucci's music Accounts from type designers' contracts and early type manuals, as well as the appearance of fonts in use today, provide evidence of multiple molds and large fonts More detailed discussion of particular fonts and the printed editions that use them can be found in Part II; especially interesting are the many instances of the careless printing of shoulders of type bodies and spacing material that confirm a hypothesis of varying body sizes A review of the limited evidence for the physical appearance of early music type prepares the way for a discussion of the history of the trade of typefounding as it relates to early Italian music type.

Music Typefounders

The trade of making types in the fifteenth century was never clearly described in contemporary treatises but can only be inferred from chance comments in colophons and archival documents of the next century.26 Evidence concerning typecasting in the fifteenth century was meticulously assembled by Konrad Haebler in an attempt to answer criticism of his classification of incunabula text types, a system that rests on the assumption that types were made by and in an individual printing establishment rather than purchased from an agent who had duplicate matrices for sale.27 According to Haebler, "it is scarcely an exaggeration to say that up to the year 1500 there were no two fonts of type the forms of which were absolutely alike."28 Recent scholarship has stressed that, while a font of type in the fifteenth century may have been unique, it might be used by another printer who had inherited it (an alphabetic font of Ulrich Han was used by Lupus Gallus in 1475, and Han's music type by Stephan Planck), sold (Francesco Girardengo sold alphabetic matrices to Giacomino Suigo), or, in unfavorable eco 25 Boorman, "Petrucci," pp 377-78.

26 The best modern accounts are in Harry Carter, A View of Early Typography (Oxford: At the Clarendon Press, 1969), and Konrad Haebler, "Schriftguss und Schriftenhandel in der Fruhdruckzeit," Zentralblatt fur Bibliothekswesen, 41 (1924), 8-I o04, translated as "Typefounding and Commerce in Type during the Early Years of Printing," Ars Typographica, 3 (1926), 3-35 27 Konrad Haebler, Typenrepertorium der Wiegendrucke, 5 vols 28 Konrad Haebler, The Study of Incunabula, trans Lucy Eugenia Osborne (New York: The Grolier Club, 1933), p 104.

nomic times, even rented.2 A publisher or commissioner of types might retain ownership and allow types to be used by different printers There is also some reason to believe that there were common sources for the types of many fifteenth-century printers It is part of the purpose of this chapter to show that by the 149os the typical Venetian printer could depend on free-lance professionals to create the complex fonts of Greek, Hebrew, Cyrillic, Glagolitic, and music type that kept Venice in the forefront of the international book trade The first specialist in typefounding cited by Haebler is one Crafto (Kraft?), who stated in a legal document of Perugia in 1477 that a German named Steffen Arndes assisted him in making punches and justifying matrices ("ad limandum et aptandum punctellos matrices ad limandum dictas matrices"'; for filing down and adjusting punches for matrices and justifying said matrices).31 Crafto also taught Arndes how to "make a suitable mold for casting letters for printing books" ("unum instrumentum aptum ad jactandum litteras ad imprimendum libros") The training of Arndes is of considerable interest since he was one of a handful of printers to print the complicated neumes of the gradual in the fifteenth century (Graduale Suecicum, 1493) Arndes had formed a partnership with Crafto in the early 147os and is said to have written a manual on typecasting.32 At the time of the Perugian litigation Crafto was in Rome, where he had

been hired for ten months "to make punches and matrices for the printing of books" ("facere punctellos et matrices ad imprimendum libros") Thus we learn that almost exactly when the first music type was being made in Italy for Han's shop in Rome, a traveling type designer with his own tools was at work in that very city making a typefont on demand The alphabetic types used by Han have puzzling connections to other printed books that suggest a common type designer at work even earlier, in the I46os, at the time of the first printing in Rome The types used by Sixtus Riessinger in what may have been the first book printed in Rome, The Letters of St Jerome, bear a strong family resemblance to those of Han's second book, a Cicero of 4 XII 1468; the capitals are perhaps exactly the same.33 Although it has recently been stated that "only toward the middle of the [sixteenth] century did craftsmen like Garamond, Granjon and Guillaume Le Be become sufficiently famous and sought after to operate as free-lance punchcutters, and they were the first to do so,"34 it is clear from the example of Crafto that free-lance type craftsmen existed as early as the i470s Important to the discussion of early Italian music type are documents that mention Italian music printers who made type and others that refer to a music typecutter who was at the end of a long career

in 1513 Few details are known of their professional training and activities as fifteenth-century typecutters, so parallels are sought in the apprenticeships and careers of men who cut music type in the sixteenth century In an extraordinary document now in the Bibliotheque Nationale in Paris, the son of an early type designer and punchcutter wrote down his recollections of men in the printing trade, including the music type designers Guillaume Le Be, Robert 29 Meyer-Baer (Liturgical Music Incunabula, p xxvii) described the Wenssler music type as "also found in the books of other houses: in Strasbourg at Gruninger's, later in Mainz at Schoeffer's, and in an anonymous Psalter The forms in use in Strasbourg appear to be identical with Wenssler's; in Mainz they appear to be slightly bigger and clumsier than Wenssler's but apparently cut after his model A comparison of the two fonts, side by side, would reveal whether they are of the same size." It is one thing to speak of the similarity of types; it is another to say they are the same 30 Carter, View of Early Typography, pp 57-58, 60-61, io4-6 3 The source of Haebler's information is a document quoted in Adamo Rossi, L'arte tipografica in Perugia (Perugia: Boncompagni, 1868), p 24 32 M Faloci Pulignani, "L'arte tipografica in Foligno nel secolo XV." La Bibliofilia 1 (1899-1900), 283-90. 33.

A Leaf from the Letters of St Jerome, First Printed by Sixtus Reissinger [sic], Rome, c.1466-1467, ed Gilbert Bennett; historical essay by Jeremy DuQuesnay Adams; bibliographical essay by John L Sharpe III (Los Angeles: Zeitlin and Ver Brugge, 198 ) Sharpe (pp 17-19) suggests an independent typecutter but dismisses the possibility "since the independent type cutter is usually considered a later development within the printing business and dependent upon a larger community of printers than was so recently come to Rome in the 146os." The resemblance to the Cicero is discussed in Mariano Fava and Giovanni Bresciano, La stampa a Napoli nel XV secolo, 2 vols and Atlas, Sammlung Bibliothekwissenschaftlicher Arbeiten 32-34 (Leipzig, 1911-1913), I: 10-27; Victor Scholderer, "Sixtus Riessinger's First Press at Rome," The Library, ser 3, 5 (1914): 320-24 34 Heartz, Attaingnant, p 49.

Granjon, and Pierre Haultin.35 Guillaume Le Be (1525-1598), son of a papermaker in Troyes, was apprenticed in 1539 or 1540, at the age of fourteen, to the scholar-printer Robert Estienne in Paris There for the next five years he was taught all facets of the book trade: bookselling, printing, letter-founding, and lettercutting-the last two under the tutelage of respected craftsmen who have retained a place in printing history.36 The next step of his training was a five-year sojourn in Venice-in 1545 still one of the main centers of the book trade-where he cut types for a number of Venetian printers, especially the Hebrew fonts for which he became famous After finishing his Italian visit with a pilgrimage to Rome, Le Be returned to France, stopping in Geneva to see his master, Robert Estienne In Paris he began a decade of free-lance type design for Parisian publishers that included working in i55 for the typecutter Claude Garamond.37 Products of that decade include seven music types, four of them for Adrien Le Roy and Robert Ballard, to print plainchant, mensural music, and keyboard and lute tablature.38 In his mature years Le Be was an international merchant with an inventory of punches and matrices that included twenty-three sets of punches or matrices or both of music type: seven of plainchant (including one of his own design, a 4° antiphonal), fifteen of mensural, and one of tablature.39 The second French typecutter whose career is known in sufficient detail to contribute to our understanding of the early period of the trade is

Robert Granjon (5 I 3-ca 1589), son of a printer and bookseller Granjon was apprenticed to a goldsmith (probably in Lyons) and "when he had learned that calling, he set himself to work at various faces of type He began in Paris about I545 and even before in 1570 he left Paris and went to live at Lyons Afterwards he went to Rome, which he reached in December 1578, where he worked for the Cardinal de' Medici at Arabic types."40 During the visit to Rome, Granjon apparently also cut a single-impression plainchant font and with it printed Guidetti's octavo Directorium Chori (I582).41 Previously, Granjon had cut three mensural types purchased by Plantin.42 A third typecutter, Pierre Haultin, "may have begun about I500 and finished his apprenticeship about 51 o The first matrices on which he worked were for a Pica or Small Pica Greek He began the punches in 1530."4 The first documentary evidence we have of Haultin is a bond of 1547 by Nicolas Duchemin to pay him for punches and matrices for music type.44 Although Fournier credited him with making "the first [French mensural music] punches about 1525," that distinction is now

35 The document has been published in French and English by Harry Carter with a foreword by Stanley Morison as Sixteenth-century French Typefounders: The Le Be Memorandum, Documents Typographiques Francais 3 (Paris: Paul Jammes, i967) The first part of the document was written by Guillaume Le Be II about 1643 and the second part was copied in the hand of Jean-Pierre Fournier l'Aine about 1730, when the Le Be foundry was acquired by the Fournier family 36 Ibid., pp 32-33 37 Annie Parent and Jeanne Veyrin-Forrer, "Claude Garamont: New Documents," The Library, ser 5, 29 (1974): 86 38 Three of the Ballard types are described and reproduced in the type specimen now in the Paris Bibliotheque Nationale (Dept des Mss., n.a.f 4528) and are printed in facsimile in Omont's "Specimens de caracteres," pp 273-83 and plate 3 A leaf of the "Musique grosse" annotated in Le Be's hand is reproduced in Heartz, Attaingnant, plate 14 The other types are mentioned in The Le Be-Moretus Collection of Fragments c.1599, ed H D L Vervliet and Harry Carter, Type Specimen Facsimiles 2:18 (Toronto: University of Toronto Press, 1972), p 12 39 The inventory of the Le Be Foundry was made sometime after 1617 and is now in the Paris Archives Nationales, Minutier Central des Notaires, etude 65, laisse 229 Morison reproduced the document in L'Inventaire de la Fonderie Le Be; the music material appears on pp 22-25 40 Sixteenth-Century French Typefounders, pp 32 and 44n.39· 41 Vervliet, Robert Granjon a Rome, pp 33-35 42 Punches and matrices of the three types remain today in the Plantin-Moretus Museum; a specimen of two of Granjon's music types made from extant matrices appears in

Heartz, Attaingnant, p 47, and Parker, Melis, and Vervliet, "Typographica Plantiniana II," ST7oa The first is a medium-sized mensural font cut before 1565. The second is a civilite or gothic cursive used in printed books in 1558 and 155943 Sixteenth-Century French Typefounders, p 31 Hendrik Vervliet questions Le Be's date and suggests the early i53os for completion of the apprenticeship: "Les Canons de Garamont: essai sur la formation du caractere romain en France au seizieme siecle" in Refugium Animae Bibliotheca: Melanges offerts a Albert Kolb, ed Emil van der Vekene (Wiesbaden: Guido Pressler, i969), p 493 44 Francois Lesure and G Thibault, "Bibliographies des editions musicales publiees par Nicolas du Chemin," Annales Musicologiques i (1953): 270-71.

thought to belong to Attaingnant.45 Haultin's career, too, combines the professions of the typecutter who made several fonts of music type and the printer who used his own music type In 1514 after finishing an apprenticeship in Paris-perhaps with Philippe Pigouchet, whose daughter he married-Pierre Attaingnant (ca I494-155I) became a bookseller with a printing press for hire He began his career printing dominoes, excommunications, and pardons but soon inherited all or part of the business of Pigouchet, who had died by I518.46 In April 1528 appeared the book that would transform music printing, the single-impression music book Chansons nouvelles, which used a music type with the notes and staff lines engraved on one punch Attaingnant also cut what must have been the first single-impression plainchant type (forty punches), perhaps completed even earlier than the famous mensural type.47 The type was still extant for a type inventory in the seventeeth century, but there are no surviving printed sheets from it At that time it was still the custom to print plainchant in red and black, and therefore a single-impression type was impractical What parallels can we draw for the training of the fifteenth-century music type designer? A standard apprenticeship in a printing shop or with a goldsmith would prepare a professional music type

designer for his craft Italy's international importance in type production is clear from the fact that it was not uncommon for a newly apprenticed sixteenth-century craftsman to make a trip to Italy While the distinction between printer and type designer was often blurred, as for Attaingnant and Granjon, who are remembered today primarily for their types, others would be trained from the beginning as specialists in type design and punchcutting, a mature profession in the sixteenth century There is some evidence that in the fifteenth century music printers produced their own type Stephan Arndes learned the craft in the 1470s for a career that continued into the next century Antonio Zarotto, the first printer in Milan, signed contracts in 1472 that obliged him to cast enough Latin (gothic and roman) and Greek types to outfit seven presses.48 Therefore he must have had the ability to make the small font of Ambrosian plainchant types used in his Rituale Ambrosianum (about I488) and Missale Ambrosianum (i March I488) In 1478 in Basel, the music printer Michael Wenssler joined the guild of moneychangers, goldsmiths, and founders; he had been since 1474 a member of the Safranzunft, the merchants' guild By 1478 he had apparently become his own typefounder.49 Wenssler began printing books first with staves and then, in 1488, with the music type of Richel.50 Other printers were not averse to buying types Christoph Valdarfer, another Milanese printer, attached a

specimen of a roman type to a printing contract in 1472; the agreement stipulated that he would retain possession of the type after the book was completed.5' While working in Basel, he obtained from Richel two fonts of text type that he later used in his own shop.52 He may have commissioned in Basel the first Ambrosian plainchant type, which appeared in his Missale Ambrosianum, 15 March 1482, the first dated music printed in Milan As late as 1492, in the small university town of Pavia, Francesco Girardengo was making and casting his own type On the date ending a thirteen-year partnership with the wealthy merchant Giovanni Antonio Beretta, Girardengo paid 6,ooo ducats and retained the tools of his profession: metal, presses, forms, matrices, files, and letters.53 He then made a 45 Pierre Simon Fournier le Jeune, Traite historique et critique sur l'orgine et les progres des caracteres de fonte pour l'impression de la musique, avec des epreuves de nouveaux caracteres de musique, presentes aux imprimeurs de France (Berne: Fournier, 1765); Heartz, Attaingnant, pp 49-58 46 In his Tresor de la langue francaise of 1606 Jean Nicot defined dominoes as pictures and portraits cut in wood or copper and printed and painted on paper ("dominos, c'est i dire des images et oeuvres de portraicture peintes et imprimees en papier, el: gravees en bois ou cuivre," quoted in Heartz, Attaingnant, p 36).

47 Heartz, Attaingnant, pp 57-58 For reproductions of the pertinent pages of Le Be's inventory, see Appendix 7, plates 6-7 48 Arnaldo Ganda, I primordi della tipografia milanese: Antonio Zarotto da Parma (147i1-507) (Florence: Olschki, 1984), p 34 49 Haebler, "Schriftguss," p 89 5o Kathi Meyer-Baer, "Der Musikdruck in den liturgischen Inkunabeln von Wenssler und Kilchen," Gutenberg-Jahrbuch (1935): 117-26 5i Haebler, "Schriftguss," p 86 52 BMC VI: xxiii 53 Archivio Notarile di Pavia Bartolomeo Strada fq Antonio, io November 1492: "stagna, torcularia, matres, forme,

new agreement to rent space in Beretta's house for a printing shop and living quarters, plus a room upstairs for typefounding Girardengo sold types as well, for records indicate that the Turin printer Giacomino Suigo bought matrices from him.54 Documentary evidence shows that by the I49os there were printers in Venice who commissioned type from technical collaborators who cut punches and cast type In I50o the printer Aldo Manuzio praised Francesco Griffo da Bologna as the typecutter of his elaborately kerned Greek and 55 Latin cursive types Particularly relevant to the history of Italian music type is the group of Venetian documents that mention a typefounder referred to as either Jacomo Ungaro (Hungarian) or Jacobo (Jacomo) Todesco (German), gettator de lettere On 2 November 1498, the diminutive form, Jacobino Todeschino, was used by the witness of a death certificate for Francesco del Prestade Bormi, another typefounder (gittator litterarum).56 In 1506 Aldo Manuzio's first will refers to Jacomo Todesco, gettator de lettere, and in 5 I3 Jacomo Ungaro requested a fifteen-year privilege for mensural music Can we be sure that these documents refer to the same man? In 1889 Castellani first pointed out the likelihood that the German and Hungarian Jacomo were the same person, since Hungarians who came to do business in Venice were treated on a par with Germans and lived at the same house, the Fondaco

dei Tedeschi.57 The boundaries of Hungarian and German territory had been in flux throughout the fifteenth century, as had those between Hungary and Venice.58 The distinction between German and Hungarian must have been hazy for many inhabitants of borderlands Jacomo Ungaro's ties to either Hungary or Germany would have been weakened by long residence in Italy In his petition for the music type privilege Jacomo Ungaro claimed residency in Venice for forty years, verifying his presence in the city at the same time as Jacomo Todesco It seems an acceptable conclusion that Jacomo Ungaro and Jacomo Todesco are one and the same man whose background was seen by Italians simply as transalpine Aldo Manuzio's will gives to Jacomo the appellation magistro, indicating some university education 27 March I5o6 Let there be distributed two hundred and fifty ducats to ten young women of marriageable age, twenty-five (25) ducats each; the said young women being four daughters of my compatre Master Jacomo Todesco, typefounder 59 The term compatre has often been translated as "godfather."60 Since Aldo was born about 1449 in Bassiano, a little town near Rome, such a relationship suggests that Jacomo spent most or all of his life in Italy However, the fact that the will uses the term compatre in reference to several individuals

(for the full document, see Appendix i, no i) makes it more likely that the term designated a close business associate As the father in I506 of four unmarried daughters to whom Aldo was bequeathing small dowries, the typecaster was at least middle-aged and probably not very well-to-do On 26 September 5 13, Jacomo Ungaro submitted a petition to the Venetian Senate requesting a fifteen-year privilege to print mensural music Under the terms of a Venetian privilege, which could run from one to twenty-five years, other printers could be forbidden to print a certain title, or to print or sell a title, to imitate a type, or even to import a certain title into the republic The privilege might be granted for one title, a group of designated titles, or whole genres such as small mis plani, littere, tellaria, et omne et totum quicquid est et operatur ad usum stamparie librorum"; as cited in Tullia Gasparrini Leporace, "La societa Beretta-Girardengo (1479-1492) nei documenti inediti coevi," La Bibliofilia 50 (1948): 45 54 Ibid., p 34 55 Albano Sorbelli, "II mago che scolpi i caratteri di Aldo Manuzio," Gutenberg Jahrbuch (1933): 117-23 56 Haebler, "Schriftguss," p 93.

57 Carlo Castellani, La stampa in Venezia dalla sua origine alla morte di Aldo Manuzio seniore (1889; reprint Trieste: LINT, 1973), p 66n.2; Henry Simonsfeld, Der Fondaco dei Tedeschi in Venedig, 2 vols (1887; reprint Darmstadt: Scientia Verlag Aalen, 1968), 2: 81 58 Gyorgy Szekely, "Les Facteurs economiques et politiques dans les rapports de la Hongrie et de Venise a l'epoque de Sigismond," in Venezia e Ungheria nel Rinascimento, ed Vittore Branca (Florence: Olschki, 1973), pp 41-45 59 See Appendix i for full text 60 Recently by Martin Lowry, The World of Aldus Manutius: Business and Scholarship in Renaissance Venice (Ithaca: Cornell University Press, 1979), p 48.

sals.61 Some privileges have certainly disappeared, and some of the titles that were privileged were never printed In Ungaro's petition he claimed to have discovered how to print mensural music in the city of Venice where he had resided for forty years: Because it is the habit of your most Illustrious Signoria to remunerate those who contribute to this famous city some useful and ingenious invention, therefore your most faithful servant Jacomo Ungaro, cutter of letters and inhabitant of this most excellent city for forty years, having discovered the way to print measured music, and fearing that others, as happens, may reap the fruit of his labors, begs your Excellency that you be pleased to grant him the favor that no one else may print or have printed the said measured music either in this city or in its provinces for the next fifteen years, nor bring books printed elsewhere to sell in this city or subordinate lands, under penalty of losing all the books and ioo ducats for every time that it occurs.62 Ungaro was concerned that others would "harvest the fruits of his labor" after he had "found the way to print measured music" because the previous privilege holder, Ottaviano Petrucci, was now absent from the Venetian Republic Petrucci's twenty-year privilege, awarded in 1498, would have had five years to run, but by April 511 he had moved back to Fossombrone, his home in the Papal States, no doubt because of the turbulence caused by the war being waged against the republic by the

League of Cambrai (1509-1513) Petrucci had probably already applied for the privilege to print music in the Papal States that was awarded to him on 22 October 15I3, verifying his intention to remain in Fossombrone.63 The Venetian Senate awarded Jacomo the privilege on the condition that it not prejudice the previous holders ("cum hoc ne praejudicetur concessionibus, si quae forte factae fuissent antehac").4 The Senate's award indicates their support of his claims of residency as typecutter for forty years and invention of the way to print mensural music There is no evidence that Jacomo, a typecutter, wished to use the privilege to print music books Having lived in Venice for forty years and with four figliole by I506, Jacomo must have been reaching old age It is more likely that the privilege was sought to prevent others from printing or bringing printed music books to offer for sale in Venice or its provinces Rather than state that "Ungaro never availed himself of his short-lived privilege,"65 it is more accurate to assume that the privilege was effective in preventing others from "harvesting the fruits of his labor." Petrucci's type was not reused in Venice Jacomo Ungaro may have died by June 15I4, the filing date of a petition by Petrucci's publishers to renew his former privilege, not just for mensural music as in Ungaro's petition, but also for lute and keyboard tablature Another intimation of Ungaro's death is his absence from Aldo's second will

of January I 515 The petition of 1 514 was made on Petrucci's behalf by publishers Amadeo Scoto and Nicolo di Raffaele, since the music printer was by then in Fossombrone The petition asked that the Petrucci music privilege, which had been unused for four years, be extended for Petrucci's former partners for five years; if still valid, the original twenty-year privilege would have run to 5 i8 It sought to prevent others from printing music and from "carrying or having carried or selling those kinds of books in this country or subordinate lands." Scoto and Raffaele were portrayed as financial backers: 26 June 1514 And because in the printing of the said works [of mensural music] much capital was necessary and was not forthcoming, the said Octaviano unable to provide it being a poor man, he took as partners S Amadio Scoto bookseller 61 Some two hundred can be found in Fulin, "Documenti," pp 84-212, and "Nuovi documenti," Archivio Veneto 23 (1882): 390-405; Richard Agee, "The Privilege and Venetian Music Printing in the Sixteenth Century," Ph.D diss., Princeton University, 1982, focuses on music privileges 62 Fulin, "Documenti," no 189 Catherine Chapman, "Andrea Antico," Ph.D diss., Harvard University, 1964, p 23, gives a translation of the document For the complete document in Italian, see my Appendix i, no 3 Agee's translation of the reference to Ungaro as a woodcut artist and printer seems unwarranted ("The Privilege," p 43).

63 Boorman, "Petrucci," pp 381-82, Appendix 4; Claudio Sartor , Bibliografia delle opere musicali stampate da Ottaviano Petrucci, Biblioteca di Bibliografia Italiana i8 (Florence: Olschki, 1948), p 17 The document as quoted by Sartori lacks two phrases 64 Fulin, "Documenti," no 18 65 Martin Picker, "Petrucci, Ottaviano," New Grove Dictionary of Music and Musicians, 14: 596.

and Ser Nicolo de Raphael who at very great expense, and the most diligent and vigilant industry, printed many and diverse volumes of the said books hoping to gain some profit, but because of the war and turbulence they have not been able to sell the said printed works and their capital is tied up, to their most great danger and loss.66 Despite their earnest protestations of interest in printing music, the seekers of the 5 14 privilege were probably looking to protect their capital investment in the still unsold Petrucci editions Amadeo Scoto, nephew of the fifteenth-century music publisher Ottaviano Scoto (d 1498), had inherited part of his uncle's flourishing publishing business He was not a printer and had no further known connection with music publishing.67 Raffaele, nearly blind in I5I4, was also a source of financial support rather than printing expertise Another reason for concern to reestablish Petrucci's privilege was the appearance on the horizon of Andrea Antico, who had been printing music from woodcuts in Rome with the help of capital from Roman members of the Giunta and Scoto families since i 5 o and who was to come to Venice after Petrucci received privileges in the Papal States for mensural music and tablature Ungaro's 1513 privilege is fundamental to the history of music printing Not only does it award to Jacomo the credit for Petrucci's influential mensural type, it suggests that he, already known to

have been a professional typecutter in Venice for decades, was capable of designing the other music fonts that made their first appearance there Petrucci's privilege had said that he had discovered how to print mensural music "and consequently plainchant much more easily." Only four men are known to have been in the business of making types in Venice by 1513: Jacomo, Francesco Griffo, another Hungarian named Andrea (Sigismund) Corbo (Corwin) who was an "inzisor literarum,"68 and Francesco del Prestade Bormi The time has come to propose that a good share of the success of music printing in Venice at the end of the fifteenth century is owed, if not to Jacomo himself, then to fellow craftsmen like him who provided the music type for the production of the liturgical books that entered the international market from Venice Support for Jacomo's claim to have "found the way to print mensural music" also comes from indications that all of Petrucci's type was cast before he left Venice for Fossombrone Boorman pointed out that in Petrucci's Fossombrone imprints the ligatures grow fewer, and some ligatures are replaced altogether by two sorts butted together There is the occasional appearance of "ugly minims with much heavier tails apparently cast from a new matrix, unless, as seems more likely, the original had become distorted."6 Such evidence indicates that Petrucci was not the owner of the matrices of his famous type and was not personally responsible for their design and production Only a copy of Jacomo's will could tell us whether the typecutter owned the music punches and matrices After

Petrucci's death in 1523, his widow apparently sold some of his type to a Gabriele Ceccolino, who, however, made no use of any music type.0 Evidence of Jacomo's role in the development of Petrucci's mensural type calls for a reexamination of the first font of mensural type, used in Venice in 1480 The craftsman would then have been in residence for seven or eight years A foreigner was allowed to trade freely and was exempt from certain customs duties when he became a citizen after residence for fifteen years or for eight years after marriage to a Venetian.71 If Jacomo had 66 Fulin, "Documenti," no 193; translated in David Gehrenbeck, "Motetti de la Corona: A Study of Ottaviano Petrucci's Four Last-known Music Prints (Fossombrone, I514-15I9), with 44 Transcriptions," Ph.D diss Union Theological Seminary 1970, p 53 For the complete document, see my Appendix i, no 4 67 La musica, ed Guido M Gatti and Alberto Basso, 2 vols in 6 (Turin: Unione Tipografico-Editrice Torinese, I966-I971), 2: Io7I 68 Jozsef Fitz, "Ungarische Buchdrucker des XV Jahrhunderts im Auslande," Gutenberg-Jahrbuch (931) : I 15 69 Boorman, "Petrucci," pp 95-96.

70 Augusto Vernarecci, Ottaviano de' Petrucci da Fossombrone, p 226 Jacques-Charles Brunet tells us that the 1521 Motetti libro primo and Motetti liber quartus of Torresano and the 1520 Motetti noui & chanzoni francoise of Giunta were printed by Andrea Antico in Petrucci's type: Brunet, 3: cols 1426-27 Chapman ("Antico," p 91 and Appendix) concluded that the music was printed from woodblocks cut by Antico himself 7 i Brian Pullan, Rich and Poor in Renaissance Venice (Oxford: Blackwell, 1971), pp IOO-IoI.

married upon arrival, he would have been ready to start a type business in 1480, the year in which six pages of mensural music were printed with the first mensural type in a Grammatica written by Francesco Niger and printed by Theodor Franck of Wurzburg (M1; for reproductions of two of those pages and the type specimen, see Chapter VII) There are striking resemblances between the first mensural font and that of Petrucci: an identical pausa of slightly more than a half circle, the distinctly pointed breve and long, the long stem of the long (in variable sizes like Petrucci's), and the calligraphic distinction of thick and thin sides of the minim Even more surprising is the similarity in size: in both fonts the long has a notehead of about 2.5 mm square, the semibreve is 3.8 mm high, and the pausa is 2.5 mm high The pointed noteheads are not accentuated as in Petrucci's type, and the stems of the long are not quite straight Frequent impressions of edges of type shoulders provide evidence for the technique visible in fonts of the i49os of casting a musical sign on a small body that can be positioned to print at the level of different spaces or lines of the staff by the use of spacing material above and below the sort The craftsman responsible for the 1480 Venetian font could well have been the resident music typecutter who cut the next mensural font at the end of the century for Petrucci Petrucci's awareness of interest in a system of printing mensural music both in and outside of Italy

is indicated by a phrase in his privilege of 25 May 1498: "At great expense and with most vigilant care, he has discovered that which many, not only in Italy but also outside Italy, have long sought, which is to print mensural music in a very convenient manner."72 Did Petrucci's publishers finance the invention claimed by the type craftsman Jacomo Ungaro in his privilege of 5 13 ? If Jacomo were responsible for the music type used in the early 1480 printed mensural music, as well as Petrucci's, he would certainly qualify as one who had long sought a manner of printing mensural music, perhaps both inside and outside Italy The Venetian plainchant fonts that appeared in Venice in the last decades of the fifteenth century include some repetitions of design Some characters reappear in different fonts in what seem identical designs; for example, the narrow and elegant flat of Petrucci's type is like that used by Hamman in his first gothic font of 1498, where it is a bit out of character The appearance of two innovative mensural types in Venice just a year apart-Emerich's black mensural type in i500 (R2i) and Petrucci's white mensural type in I50I suggests a single creative designer, especially since Petrucci's font is linked by name to a professional Since the mensural types in R2 match the body size of the plainchant type and would have been cast in the same molds, would not the same craftsman very likely

have cut the entire font, plainchant and mensural? Ungaro, having certainly cut Petrucci's font and perhaps having had a hand in Emerich's mensural type and by extension his plainchant type, could well have cut other plainchant fonts as well Thirty-eight music types were used in Italy in the fifteenth century, twenty-five of them in Venice It seems likely that many were the products of a very few craftsmen who specialized in music type 72 Fulin, "Documenti," no 81 For the complete document, see Appendix i, no 2.

IV—
Italian Music Incunabula

Early printers had at least five options for the printing of music books First, they could omit music and space for it altogether Second, printers could leave, often above printed lines of text that was to be sung, blank space for later manuscript insertion of staves and notes Of the 156 Italian music incunabula, thirty-five follow that prescription (see Table II), including the first four that were printed (D 38-D 4 i) While most of the thirty-five appeared in the i47os and I48os, printers without music type or without type of the appropriate size for small formats continued the practice into the I49os (D 32, D 9 i) A third option, printing staves and omitting notes, was selected for nineteen Italian incunabula (see Table 12) The quality of the printed staves ranges from miserable to excellent; techniques run the gamut from woodcuts and poorly beaten rules to cast metal lines the width of a staff and, finally, cast small metal segments of one line each Poorly printed staves precluded properly printed notes, so that printers experienced with music type rarely risked using wavy, irregular materials for printing staves A small group of twelve incunabula, nine of which are music theory texts, employed a fourth option, notes and staves printed from woodcuts (Table I5) Finally, the largest group, seventy-six music incunabula, are printed from thirty-eight music types with both notes and staves.

Space for Music

Few printers chose the option of omitting music and space for it entirely Admittedly, it is difficult to identify editions that eliminated the music of a copytext from print, but any missal printed without music can probably be placed in that category.' The two modest octavo missals printed in Italy without space for music may have been aimed at readers unable to write music or too poor to hire someone who could One of these was printed in Venice in 1481 by Franz Renner, a printer who never did attempt to print music The Florence copy (Biblioteca Nazionale Centrale, K.7.53) has no added manuscript leaves, and the copy at Munich (Bayerische Staatsbibliothek, 8° Inc.c.a.9) has an added twenty-four manuscript leaves without chant, but the owner of at least one copy (British Library, IA 19880) inserted before the Canon ten manuscript leaves with text and roman chant notation to replace the four printed leaves (ff m4-m7) he felt must have music There was obviously a market for a small, inexpensive missal printed without space for music, but at least this owner of such a book felt the lack keenly enough to redo a gathering The three folio missals printed by Pachel and Scinzenzeller before I483 had included space for music, but their 1483 octavo missal excluded music.

The absence of music in printed breviaries follows the tradition of manuscript breviaries of the last decades of the fifteenth century The list of fifteenth-century printed Italian breviaries in the i The Glagolitic Missale Romanum of 1483 is an exception Its lack of music probably reflects the state of the copytext, since liturgical music for the Slavonic service has a performance tradition that is still transmitted orally today See Josip Andreis, Music in Croatia, trans Vladimir Ivir (Zagreb: Institute of Musicology, 1974), p 19-

Gesamtkatalog der Wiegendrucke includes none with printed music, and only one (dated 29 Sept 1492) was found with blank space for music (Table i I, D 32) No music types smaller than octavo format were cut in the fifteenth century and most printed breviaries were in smaller formats Many more printers of music books in Italy chose the second option, adhering to scribal tradition by leaving space for the rubrics, initials, and music that would have been assigned to professional rubricators, illuminators, and music scribes rather than text scribes Books with space for music were printed in locations in which music types of the right size were not yet available The exception (D 32) also exists in a state with printed staves, suggesting that more than one printer was involved in the production The single folio format book with space for music from Rome was printed before 1476, when large music type first appeared there; most such books (eleven) were issued in Milan in the first decade of printing, before music type was available in the city In Venice eleven such books were published before their printers owned appropriate music type.2 Four examples were produced in Naples before its first music type appeared about I490 The printers who issued such missals in Verona, Messina, and Turin did not attempt to print more than one The genre had become commercially viable only in the hands of specialists at centers of printing with international distribution facilities.

The first printed books to leave space for music were prepared for buyers accustomed to the manuscript tradition that saved rubrics, initials, staves and notes, and illumination to be added after the textual scribe had completed his task Indeed, the large amount of handwork required for such additions resulted in printed missals that can easily be mistaken for manuscripts The Franciscan monks for whom the first printed Missale Romanum (Central Italy? ca 1472, D 38) was intended may have requested an edition that left to the scriptorium the task of adding rubrics and chant The missal exists in two issues which survive in unique, richly illuminated vellum copies One, from a Benedictine monastery near Spoleto, now belongs to the Newberry Library, Chicago; the other is one of five printed books in the Duke of Urbino's library of manuscripts now in the Vatican Library About one-third of the Newberry copy is left blank for extensive rubrics, initials of major, middle, and minor size, a Crucifixion illumination, and music on forty-seven pages In the Vatican copy, much of the rubrication is printed A casual viewer of the books would judge them to be manuscripts; the Duke of Urbino's copy remained undisturbed in the manuscript section of the Vatican Library until recently, and the Newberry copy was described as a manuscript in an inventory at its purchase The second printed missal, with rubrics for a nonmonastic audience, was produced in the urban environment of Milan by a publishing group apparently motivated by profit.3 The Missale Romanum was printed by Antonio Zarotto in I474 without a

red impression, leaving space for rubrics, initials, staves and notes, and illumination Only on the incipit leaf did Zarotto add a second impression in red, but in a fashion that showed him to be an accomplished technician The omissions would have saved money for the producers of the book by leaving to the purchaser the cost of finishing it by hand The first missal for the Milanese Ambrosian rite was also printed by Zarotto, in 1475, again with space for textual rubrics, music, initials, and Crucifixion (Figs I4-I5) Zarotto's Ambrosian copytexts would have followed a tradition of two-colored staves, a red line for fa and a yellow line for ut or C; the copy of his I475 Missale Ambrosianum in Milan's Biblioteca Ambrosiana (Fig 14) contains inserted music written on two lines drawn in colors, and the copy at the Bibliotheque Nationale at Paris (Fig i5) has rastral staves on which the red line for fa has been darkened and a yellow line for ut has been drawn in Early printers of music apparently never attempted printing a yellow

line on a staff In Milan, where the tradition remained strong, no music was printed until the 148os and then only on the easier, red staves None of the printers who issued books with blank space for music did so after acquiring a font of appropriate music type That a lack of technical 2 Since Beretta is known to have used Venice as a false imprint, 1) 32 may be from his home, Pavia For controversy over the place, printer, and date of D 7 1, see Part III 3 Ganda, I primordi della tipografia milanese, chap ii, "La ricerca di un nuovo mercato: i libri liturgici," pp 59-66.

Table I 1 Italian incunabula with space for music Date Author, Title Place: Printer Format M-B W D [ca 14721 Missale Romanum [Central Italy?] 2 38 6 XII 474 Missale Romanum Milan: Zarotto 2° 852 39 23 III 1475 Missale Ambrosianum Milan: Zarotto 2° 26 23 21 IV 1475 Missale Romanum Rome: Han 2 853 40 26 IV 1476 Missale Romanum Milan: Zarotto 20 1 8 854 41 20 IX 1477 Missale Romanum Venice: Siliprandi 2° 120 856 43 io XII 1477 Officium Imm Concept Rome: Han 4 144 BVM I477 Missale Romanum Naples: Moravus 2° 121 857 44 17 I 1478 Missale Romanum Milan: Zarotto 2 858 45 22 IV 1479 Missale Romanum Milan: [Carcano] 4° 46 i V I479 Missale Romanum Venice: [Jenson] 2° 122 859 47 27 IX 1479 Missale Romanum Milan: Zarotto 2° 123 860 48 16 XII 1479 Missale Romanum Milan: Pachel & 2° 49 Scinzenzeller 27 VIII 1480 Missale Ultramontanorum Verona: [Maufer] 2 I607 127 27 VIII 1480 Missale Strigoniense Verona: [Maufer] 2° 1488a 127 (reissue) 1480 Missale Messanense Messina: Aiding 2° 592 33 31 VII 1480 Missale Romanum Milan: Pachel & 40 124 864 50 Scinzenzeller 23 VIII I480 Missale Romanum Venice: Joh von Koln, 2 865 5i Joh Manthen 8 X 1480 Gaffurio, Theorica Naples: F di Dino 80-— 14 31 VIII 1481 Missale Romanum Venice: Torti 2° 866 52 18 IX 1481 Missale Romanum Milan: Pachel & 4 867 53 Scinzenzeller

8 XI 1481 Missale Romanum Milan: Zarotto 2° 868 54 29 XII I481 Missale Romanum Venice: Scoto 8° 125-26 870 55 16 III 1482 Missale Romanum Naples: Moravus 2° 873 57 2 VII 1482 Missale Romanum Venice: Blavi, Torresani, 4 874 58 & Salodio 1482 Missale Praedicatorum Milan: Zarotto 2° 211 132 29 III 1483 Missale Praedicatorum Naples: Moravus 4° 213 i816 133 [ca 1483] Missale Toletanum [Venice: s.t.] 2° 1529 26 [ca 1485-1490] Missale Romanum [Venice? s.t.] 2° I67 953 7I i VIII 1486 Missale Ambrosianum Milan: Pachel 2° 28 25 28 IV 1486 Magnificat Milan: Zarotto 4° 24 2 [2 leaves] 1487 Missale Romanum Venice: Frankfurt 8° 137 900 74 29 IX 1492a Missale Clarimontense Venice: Beretta 8° 58 277 32 25 XII 1492 Breviarium Praedicatorum Venice: Emerich 16° LB36o 7 II IX 1492 Missale Romanum [Turin? Suigo & 8° 148 921 91 Benedicti?] 5 VI 1492 Caza, Trattato Milan: Pachel 4° 9 a Also exists in copies with printed staves.

FIG 14 Missale Ambrosianum Milan: Antonio Zarotto, 23 III 1475, f [CVIII] (Biblioteca Ambrosiana, Milan, S.P.II.22.) material was the reason for the blank space in Han's Missale Romanum of 20 April 1475 is indicated by the appearance of an almost identical book on 12 December 1476 with printed notes and staves instead, and an altered colophon proclaiming it to be "with music, which had never been done before." A comparison of the two editions at the Biblioteca Vaticana revealed that the preliminary signature (the calendar) is identical, the text of the incipit has been reset but is nearly identical, and the number of pages containing music is the same, with one exception On f m3v of the 1476 edition there is no music because the compositor managed to fit the one sung word on f m3 Both editions, furthermore, contain a leaf with either too many music staves or too much space for them The near-identity of the editions suggests that the 1475 edition was carefully designed to include the music that did not appear in print until the second edition, eighteen months later The design and creation of the first Italian music type and staff material may have taken longer than planned.

FIG 15 Missale Ambrosianum Milan: Antonio Zarotto, 23 III 1475, f [xlvi] (Bibliotheque Nationale, Paris, Res B 1485.) Some books with space for music have had staff lines so skillfully drawn by a raster that they have been mistaken for print The three copies I examined of the I477 missal printed by Moravus in Naples have rastral staves, though the book was described by Meyer-Baer as having printed staves Moravus's second printed missal of 1482 is correctly described by Meyer-Baer as having space for music, but again his Missale Ordinis Praedicatorum is described as containing printed staves when in fact in the two copies I examined they were drawn See Fig i6, especially col 2, staff 8, and Fig 17, especially col i, staves 2-3 (note the three-line raster) for the telltale hesitation at the beginning of the staves where the raster first touches the paper Incunabula that contain manuscript music provide a valuable record of printers' first designs for pages of music as well as examples of the scribal tradition at the moment of transition from manuscript to printed book Several major music printers began their involvement with music by issuing works with space for music, and the decisions made

in those books about the relationship of text lines to staves were often retained in their books with music printed from type Some of the earliest printed books with music (both copies of the ca 1472 Missale Romanum, Zarotto's 1475 Missale Ambrosianum, Moravus's 1477 Missale Romanum) contained two-, three-, and four-line staves that varied according to the melodies to be written on them; two and three-line staves, and variation for the ambitus of the melody were eliminated by printers, as was the two-line red and yellow staff of Ambrosian chant Although they contain no printed music, the first printed books to include music resolved problems of book design critical to the next steps of printing staves and cutting music type.

FIG i6 Missale Romanum Naples: Matthias Moravus, 1477, f [106] (Bayerische Staatsbibliothek, Munich, 2° L impr membr 62.)

FIG 17 Missale Ordinis Praedicatorum Naples: Matthias Moravus, 29 III 1483, f [37] (Bibliotheque Nationale, Paris, Velins 1704.)

Printed Staves

Another group of Italian incunabula were printed without notes but with staves for music (see Table 12) The fact that two of these exist also in a state with printed notes suggests collaboration between printers, possibly in different shops Some of the imprint names listed with the books in Table 12 may be publishers' rather than printers': Frankfurt is not known to have been a printer, and Sessa was probably not one either; Scoto called himself a publisher after 1484, but may have been a printer before that year If these three were not printers, the only Venetian printers known to have used rules for printing staves are Paltasichi, Ratdolt, and Hamman Paltasichi was printing in Venice in I475, Ratdolt in 1478, and Hamman in 1482; any of them could have printed the rules in the books of Scoto and Frankfurt (the rules in Scoto's books were printed along with the text) Since Paltasichi was working For Scoto in 1480 and 1481 and did print his own edition of a missal with ruled staves, he may well have been the printer responsible for the two editions of Scoto listed in Table 12 Several techniques were available for printing staves: rules (type-high strips of metal) of a relatively soft metal alloy; cast metal lines a column wide (probably single lines); smaller cast segments of one to four lines, a number of which together made up the width of a column; short cast metal

segments of a single line nested irregularly to make up a staff line; and woodblocks.4 When joints on all staff lines consistently occur at the same place between segments, it is impossible to know whether the segments are single lines or are cast multiples of two or four lines The use of ledger lines above and below the staff in Emerich's R2I proves that cast single lines were available for that font; a four-line segment the size of that staff (32 mm) would be unmanageable in a handheld mold The techniques for printed staves are not chronologically sequential, but the use of rules does disappear after I49I (see Table 2) and nested small segments do not appear until the late I490s That the music printer did not store material for staves in his cases of music type is demonstrated in the earliest extant cases of music type at the Plantin-Moretus Museum in Antwerp This practice was logical if the staves were set and locked up in forms without music characters, which would later be printed in black in a second impression The technique chosen for printing staves would have depended on the material available to a printer, which included the adjustable handheld metal mold cased in wood, larger metal or stone molds, woodcuts, and brass rules.5 Several of the books printed with staves and without notes were apparently prepared in this way

because they were intended for an audience accustomed to gothic plainchant notation The publication by Emerich of three missals for Hungary with printed staves but without notes in 1498, I499, and I500 (1502?) at a time when he owned two fonts of roman plainchant type, was probably due to his awareness of the geographical limits of the use of roman plainchant notation There are certainly fifteenth-century chant manuscripts in the Budapest Szechenyi National Library in both gothic and roman plainchant notation, but all copies I examined of the missals for Gran printed by Emerich had manuscript gothic chant notation added to the printed staves, as did those in Ratdolt's 1486 missal for Gran The Hungarian publisher from Buda, Johann Paep, may well have advised his Venetian printer to print gothic notation or none at all It 4 Previous discussions of the technology for printing music staves mention cast staves in blocks of metal or separate lengths of rule (Alexander Hyatt King, Four Hundred Years of Music Printing [London: British Museum, I968], p 9) or rules

and woodblocks (Maria Przywecka-Samecka, "Problematik des Musiknotendruckes in der Inkunabelzeit," Gutenberg-Jahrbuch [I978]: 55) Meyer-Baer's description of methods (Liturgical Music Incunabula, p xxvi) adds a technique of "double rules recognizable by the curves occurring often in two lines at the same spot," which I have been unable to recognize in printed examples 5 See the 1580 illustration of stone slab molds in Cyril Stanley Smith, "Metallurgy and Assaying," A History of Technology (Oxford: Clarendon Press, 1956), 3: 38 For brass rules see, for example, Ratdolt's 1480 edition of Rolewinck's Fasciculus Temporum, plate 23 in Ferdinand Geldner's Die deutschen Inkunabeldrucker (Stuttgart: A Hiersemann, 1968).

Table 12 Italian incunabula with printed staves but no notes Date Author, Title Place: Printer Format M-B W D Metal rules12 V 1482 Ramis de Pareja, Musica Bologna: Hyrberia and 4 150 Utriusque Cantus Heinrich von Koln Practica VI 1482 Ramis de Pareja, Musica Bologna: Hyrberia 4° - i5 Utriusque Cantus Practica 28 XI I482a Missale Romanum Venice: Scoto 4 127 877 62 24 XII 1482 Missale Ord. Praedicatorum Venice: Scoto 4 2I2 i8i5 131 1484 Missale Ord. Praedicatorum Venice: Frankfurt 8° 214 I8I7 I34 27 IX I484a Missale Romanum Venice: Paganini & 2 132 889 67 Arrivabene 1484 Missale Romanum Venice: Frankfurt 8° - 892 68 I II 485-I486 Missale Romanum Venice: Paltasichi 8° 136 896 72 i8 III 1486 Missale Strigoniense Venice: Ratdolt 2° 492 121 Io XI 1487 Missale Parisiense Venice: Hamman & 8° 97 700 35 Emerich 1487 Missale Romanum Venice: Frankfurt 8° 137 9°o 74 6 VI 1491 Missale Romanum Pavia: Girardengo & 80 I45 916 86 Beretta Cast metal 14 III 1490 Missale Burgense Venice: Hamman 2° 3 31 X 1495 Missale Strigoniense Venice: Emerich 4° - 496 I23

23 I I495-1496 Agenda Aquileiensis Venice: Hamman 4 31 VII 1498 Liber Catechumeni Venice: Sessa 80 - LB725 19 26 II I498-I499 Missale Strigoniense Venice: Emerich 2° i89 1498 124 24 IV 1499 Missale Quinque Ecclesiae Venice: Emerich 4 - 804 36 31 III 1500 or Missale Strigoniense Venice: [Emerich] 4 90 I499 125 I IV 1502 Also exists in a variant state with printed staves and printed notes. would be interesting to know the reception accorded the folio Missale Strigoniense printed by Hamman in 1494 with roman plainchant. The absence of notes in Hamman's Agenda Aquileiensis, at a time when he owned at least two fonts of roman music type but before he used his gothic plainchant font, indicates that the common notation in the Patriarchal See of Aquila was gothic. Although the town of Aquila is only a few miles north of Venice, its territory extended into Yugoslavia, Austria, and over to Como at the end of the fifteenth century.6

The Missale Praedicatorum of 1484, the first of four octavo missals published by Frankfurt in the i48os, was issued with printed staves and without notes. Of the two copies I examined, the copy at the 6. Spessot, "Libri liturgici aquileiesi," pp. 77-92; see also Archibald King, Liturgies of the Past (London: Longmans, 1959), pp. 1-51.

Bibliotheque Nationale in Paris had no manuscript music and the copy at the British Library had roman notation; another four are in Poland within the boundaries of gothic notation Only one of the ten known copies of this missal is in Italy; eight are in Poland, Germany, and the Netherlands and would presumably have manuscript gothic notation, as does the copy I have seen in Vienna Of the five known copies of the 1487 Missale Romanum, one is from Buxheim and one from Budapest (not all copies have manuscript notation) As an ambitious publisher, Frankfurt may have wished to avoid making the choice between gothic and roman notation in order more easily to sell his books in both the north and the south This reason would have become less valid in the 1490s, when almost all missals were printed with notation and buyers were not likely to wish to pay for scribal additions Frankfurt's final two octavo missals were published with roman notation, printed by Hamman (1493 Missale Romanum) and Emerich (500o Missale Praedicatorum) The crudest technique for printing staves used rules of a soft metal alloy It is not certain when brass was first used by printers for rules In large cities metal forges would have sold sheets of an alloy like brass that could have been cut into rules Thus Biringucci speaks of visiting a brass works near Milan in the early sixteenth century, and presumably brass would have been readily available at such sites as Brescia, the so-called iron capital, which had thousands of people working iron at the end of the fifteenth century.7 Properly beaten or

rolled brass rules are still today more expensive than type-metal rules but are "in every way superior."8 That there were problems in making brass rules is evident from the cautions issued by Moxon in the seventeenth century against using unbeaten or poorly hammered metal provided by "unskilled joiners": take care that the Brass, before it be cut out, be well and skilefully Planish't, nor would that charge be ill bestowd; for it would be saved out of the thickness of the Brass that is commonly used: For the Joyners being unskilful in Planishing, buy Neal'd thick Brass that the Rule may be strong enough, and so cut into slips without Hammering, which makes the Rule easily bow any way and stand so, and will never come to so good and smooth an Edge as Planish't Brass will Besides, Brass well Planish't will be stiffer and stronger at half the thickness than unplanish't Brass will at the whole.9 Poorly prepared brass rules must have been responsible for the irregularity of printed staves in books such as the inexpensive octavo missals printed in Venice (Figs 20 and 21) In the earliest Italian example of music books with printed staves but no notes, the single staff so generously provided by the printer in Ramis de Pareja's Musica Utriusque Cantus Practica (see Fig i8) was clearly inadequate for the music notation required by the text In one case the three-line staff, printed from inferior rules, was scratched off the

paper by a scribe and replaced by a staff of five lines to contain the music example called for in the text (see Fig 19) In fact, two such staves should have been printed to illustrate properly the two lines of text printed for the music example, and Franchino Gaffurio, another owner of this same copy, added them in the margin with the letters naming the notes printed in yet another two lines of text.'° Books published by Scoto and Frankfurt used equidistant metal rules in two columns for a page of music with text (Fig 20) The space (4-4.5 mm) was determined by the size of the text type, as is apparent from the occasional rubrics that are printed with the staves and at first glance seem to overlap them The technique allowed the printer to use a single size of spacing material between both rules and text, but the material must have been inadequate, because the staff lines wave and bend at the ends of the lines The same method of equidistant

7 Biringucci, Pirotechnia, p xix; Braudel, Capitalism and Material Life, pp 180-81 8 Ralph W Polk and Edwin W Polk, The Practice of Printing (Peoria: Bennett, 1964), p 148 9 Joseph Moxon, Mechanick Exercises, ed Herbert Davis and Harry Carter, 2d ed (London: Oxford University Press, 1962), p 27 Io Albano Sorbelli, "Le due edizioni della Musica Practice di Bartolome Ramis de Pareja," Gutenberg-Jahrbuch (1930): 106 Musica Utriusque Cantus Practica was intended as an inexpensive textbook for Ramis de Pareja's music course at the University of Bologna.

4 iIG 18 Bartolome Ramis de Pareja

4 iIG 19 Bartolome Ramis de Pareja, Musica Utriusque Cantus Practica Bologna: Baltasar di Hyrberia, 5 VI 1482, f b6v (Biblioteca del Civico Museo Bibliografico Musicale, Bologna, A 81) * FIG 19 Bartolome Ramis de Pareja, Musica Utriusque Cantus Practica Bologna: [Baltasar di Hyrberia and Heinrich von Koln], 12 V 1482, f b6V (Biblioteca del Civico Museo Bibliografico Musicale, Bologna, A 80.)

FIG 20 Missale Ordinis Praedicatorum Venice: Nikolaus von Frankfurt, 1484, f 06" (Bibliotheque Nationale, Paris, Res B 1515.) FIG 21 Missale Romanum Pavia: Francesco Girardengo and Giovanni Antonio Beretta, 6 VI 1491, f n5v(Biblioteca Nazionale Braidense, Milan, BG Inc 53.) rules was employed by Girardengo and Beretta to print an octavo Missale Romanum on 6 June 149i in Pavia (Fig 21) The rules were cut to exact length and suffer from a moderate amount of bending Girardengo himself had printed a beautiful folio missal (D 65) in 1484 with notes and cast metal staves and later printed a quarto missal (1495, D 30) with the same material His printing of the 1491 and 1492 missals with weak rules and without notes can almost certainly be attributed to a lack of type for the octavo format.

Though he was later to be an important music printer in Augsburg, the only music book Erhard Ratdolt printed in Italy is a missal with printed staves and without notes Neither Ratdolt nor any other printer in Italy in I486 had type for the gothic plainchant that was standard in the See of Esztergom (Strigonium) northwest of Budapest in Hungary The folio Missale Strigoniense (Fig 22) disdains the utilitarian practice of less expensive octavo missals of allowing just enough space between staves for the body of the text type; instead, it uses leads in the area above and below the staff to prevent the text from intruding on the staves Each column of staves is designed to fit the ambitus of the music, varying from columns of eight three-line

staves to seven five-line staves The two copies in Budapest's Szechenyi National Library have manuscript notation in the Messine-German style (see Table i) A final example of the use of rules for staves is the 1487 Missale Parisiense printed by Hamman and Emerich, who later became the most prolific printers of music incunabula in Italy The missal for Paris, their first attempt to print music at a time when neither owned music type, is their only book to use rules They clearly had some difficulty with the technique, for many of their staves are missing a line that was not type-high (Fig 23) The ideal printed staff, pleasing to the eye, easy to read, and exact in its spacing and horizontal direction, would be a staff cast in strong metal in one piece measuring the width of the column, without joints to interrupt the movement of the eye and without wavy up-and-down irregularity to disturb the symmetry of the several lines of the staff Entire lines a column wide would have been too large for a handheld mold and would have had to be cast in a rule mold, an item known to have been standard in printers' shops from the seventeenth century and certainly developed much earlier In the first dated printed music of 1476 (see Fig 3I) Ulrich Han used cast metal lines a column wide, four or five of which made up a staff The strong, wide lines have

two distinct edges, clearly visible under a magnifying glass The absence of breaks in the straight lines of the thirty-three pages of staves and the use of five-line as well as four-line staves speak against the possibility of the medium being wood An impressive use of cast metal lines was made in the folio missal printed by Hamman in which the staves reach across the full page (see Fig 24) The unbroken lines provide a good base for the plainchant type; note the provision of shorter pieces for lines that begin with rubrics or large initials The printer was forced to use some short pieces to finish staff 2, however, in which the joints are unpleasantly wide Staves printed from wood appear rarely to have been used, though they necessarily appear in conjunction with notes cut on the same wood (see next

A FIG 23 Missale Parisiense Venice: Johann Hamman and Johann Emerich of Speier, Io XI 1487, f m3" (Bibliotheque Nationale, Paris, Res B 4663.) FIG 24 Missale Strigoniense 0 Venice: Johann Hamman, i II 1493/1494, f ri (Orszagos Szechenyi Konyvtar, Budapest, Inc 180.)

FIG 25 Missale Romanum [Venice: Bernardino Benali? for Luca Antonio Giunta, ca 1490], f o3v (British Library, London, IA 25 Ioo.)

section, "Music Printed from Woodcuts") The appearance of the printed staves in an undated quarto missal (D 84) suggests that wood may have been the printing medium (Fig 25 and Fig 41) Individual lines vary considerably in width; actual breaks occur that could be explained as the wood having been chipped and broken; and sharp changes in width of lines suggest the beginning of the motion of a cutting tool I have attributed the book to the printer Bernardino Benali because it uses his plainchant type R5 The metal rules used in Benali's missal of 1484 (D 63) had provided a poor foundation for printing the notes squarely on lines and spaces In D 84 the same music type is printed on a regular staff that is reduced in size from 13.25-14 mm to o0.5-11.25 mm; the new staff technique provides lines on which notes can be printed in the correct position Because the music type best fits the staff of the quarto format, one would suppose it to have been designed for the smaller book Perhaps a reexamination would reveal that the quarto now assigned the date of ca I490 was the first to use Benali's music type R5, before January 1484, the date of D 63.

The last technique for printing staves developed in the fifteenth century was nesting, the use of a number of short, single-line metal segments to make up a staff line the width of a column The technique was used in Italy by Johann Emerich for a Missale Strigoniense published for Johann Paep of Buda Segments, usually four pieces of II.7512 mm plus one of 8.5 mm, make up each of the four lines of the staff in what appears to be random succession (Fig 26) The handheld, adjustable FIG 26 Missale Strigoniense Venice: [Johann Emerich of Speier for] Johann Paep, 500oo or 1502, f nii (Orszagos Szechenyi Konyvtar, Budapest, Inc 996.)

mold that was used for casting music and text type was the likely instrument for casting the small segments, which are no taller or wider than a stemmed ligature of the same music type This technique became the common method of printing music staves in the sixteenth century and later Since cast segments set as abutting types often leave irregular gaps, they provide less regularity for eye movement, but they do achieve an exact background for printed notes Books printed with staves and without notes could have been composed in a single form, although for Italian liturgical books they would have been printed in two impressions: first in red for staves, rubrics, and initials, and then in black for the text Music theory books such as the Ramis theory text (Fig i8) use a single, black impression for the staves and text The use of two separate forms for printing the text and the staves is suggested in the first Italian printing of staves in two colors and without notes In 1484 Frankfurt published an octavo Dominican missal with red four-line staves printed from metal rules (Fig 20) These rules, often bent and frequently failing to print at all, create a crude and amateurish contrast to the tightly locked and justified text type, and the fact that the black text between the staves is both too high and too low suggests that rules were not present in the black form to aid the compositor in spacing Furthermore, on f 02 nine consecutive red lines are

printed right through black text, an error impossible using one form Music staves in other Frankfurt missals also appear to have been set in two forms, and the Missale Romanum of 1484 has one staff of five lines that overlaps the text (f q7v) Does the use of an additional form for staves suggest the involvement of a separate craftsman to print them? The form used to print the red and black impressions of text type would have had to have the blank spaces for staves filled with spacing material or wooden furniture If the printer had had material on hand to print staves, it would have saved time and energy (and thus money) to print the red staves together with textual rubrics in the first red impression The postponement of printing the staves to a third impression from a newly locked-up form suggests that a separate craftsman with separate materials might have been required for music staves The involvement of a separate craftsman for music printing is also suggested by the existence of two editions with variant copies, some with printed staves and some with an added impression of printed notes One of those cases is the 1484 Missale Romanum of Paganini and Arrivabene Copies exist with the red staves added from a separate form, as described above, and others with both red staves and black notes printed from two additional separate forms It seems unlikely that the printer of the original form

would have chosen to fill the space with furniture and prepare two additional forms if he had owned either staff material or a chant type font Another printer, already a music specialist, may have been hired for the task The following review of the staff techniques of the two most prolific printers of Italian music incunabula (see Tables 13-I4) makes it clear that individual printers tried many techniques Hamman and Emerich used all the techniques described: blank space for music, printed staves from rules, cast lines a column wide, and cast segments The metal rules, probably of poorly beaten brass, used by Hamman for the irregular staves of the I487 Missale Parisiense (Fig 23) did not provide the accuracy necessary for a second impression of notes To print his first notes in the 1488 Missale Romanum, Hamman used cast metal to produce straight lines of exact size That same material, lines of about 75-76 mm for the width of a column, was used in three other books (nos 3, 4, and 12), another 38.5 mm size was used for four books, and two other sizes appear Initials sometimes required shortened lines but rubrics were often pushed into the margin (Fig 27) Both Hamman and Emerich changed their technique for printing staves in I497, discarding metal the width of the column for small segments a number of which made up a column Cast segments allowed easy revision of the length of

one or all of the staff lines, a practice put to frequent use in Emerich's I499/5oo00 Graduale Emerich, the most prolific fifteenth-century Italian music printer, joined Hamman in 1487 for their first book to use metal rules, printed a second in 1492 with space for music, and produced a third in I493 with music printed from woodblocks (Fig 30) After a hiatus of more than a year, Emerich published his first book to use cast music type and metal staves (Fig 28), a technique he continued to use throughout his career His association with

Table 3 Staff techniques of Johann Hamman Date Title Music Staff Size Font Technique D io XI 1487 Missale Parisiense staves io-io.5 40.5-42 (rules) 35 5 x 1488 Missale Romanum notes 18-i9 R9 75 mm 77 14 III 1490 Missale Burgense staves 75-76 mm 31 13 VII 1491 Missale Romanum notes 18-I9 R9 75-83 mm 87 i VI 1492 Missale Valentinum notes 16-16.5 R9 83 mm 128 [from photograph] I VII 1493 Missale Romanum notes io Rio 39 mm 94 XII 1493 Missale Romanum notes io-Io.5 Rio 38.5 mm 95 II 1493/94 Missale Strigoniense notes 5-line R9 138 mm 122 18.5-19 I IX 1494 Missale Sarisburiense notes 17.5 Rg 23 + 15 mm X 3a 118 I XII 1494 Missale Sarisburiense notes o0-10.25 Rio 38.5 mm 119 i II 1494/95 Missale Praedicatorum notes 9.5-10.5 Rio 38.5 mm 136 23 I 1495/96 Agenda Aquileiensis staves 14-15 75 + 25 mm i 5 VI 1497 Missale Romanum notes 16-i6.5 Rii '135-I4 mm X 3a 104 13 IX 1498 Agenda Pataviensis notes 15 Gi 13.5 mm X 8a 2 "Cast segments.

Table I4 Staff techniques of Johann Emerich of Speier Date Title Music Staff Size Font Technique D [n.d.] Missale Romanum ? 37 io XI 1487 Missale Parisiense staves Io-1o.5 rules, 40.5-42 mm 35 25 XII 1492 Breviarium Praedicatorum space 7 28 IV 1493 Missale Romanum notes Io woodcut 93 13 VIII 1494 Missale Romanum notes 14.5 Ri8 54.5 mm 98 9 X 1494 Processionarium Praedicatorum notes 13.75-15 Ri8 85 mm 147 30 IV 1495 Liber Catechumeni notes 14-14.25 Ri8 85 + 16 mm 18 31 X 1495 Missale Strigoniense staves 15.5 58-58.5 mm 123 i4 VII 1496 Missale Romanum notes I5.5 Ri9 77-5-82 mm ioi 1o 0I 1497 Missale Romanum notes 14.5 Ri8 8 mm X 7, segments 1o6 28 VI 1498 Missale Romanum notes i5.5 RI9 8 mm X io 109 15 x 1498 Missale Romanum notes 10.25 R20 9.75 mm X 4 III 26 II 1498/1499 Missale Strigoniense staves 15.5 7.8 mm X 9 124 24 IV I499 Missale Quinque Ecclesiae staves [film] ? X 5, segments 36 28 VI I499 Missale Messanense notes [film] Ri9 34 1499/1500 Graduale Romanum notes 32 R2I 8, 12.5, 17,24 mm 17 6 III 1500 Missale Praedicatorum notes io R20 9.5 mm X 4 139 31 III 1500 Missale Strigoniense staves 16 11.75 mm X 4 + 8.5mm 125 6 VI 1500 Liber Catechumeni notes 13.75-14 Ri8 7.6 mm X 14 20 24 VII 1500 Missale Segoviense notes 15.5 Ri9 7.8 mm X I 120 5 I i5co0/I0I Missale Carmelitarum notes 15.5 RI9 8 mm X 7-130 27 II ISOO/I5o0 Missale Romanum notes 14.5 Ri8 8 mm X 7-17

FIG 27 Missale Valentinum Venice: Johann Hamman, I VI 1492, f 06 (Biblioteca Nacional, Madrid, Inc 773-) publisher Luca Antonio Giunta began with a Missale Romanum of 1494 that contained fifty-four pages of music with staves printed from cast lines and notes from type It may have been Giunta's capital that enabled Emerich to abandon the woodcut technique permanently Emerich's technique for printing staves soon changed to use of short metal segments The quarto and octavo books printed by Emerich used 8 mm segments; large folios used 12 mm segments; octavos used 9.5-10 mm segments When a note of the music strayed above or below the normal staff, it was common practice in manuscript and printed music to change the clef to avoid the need for a line above or below the staff, but in his Graduale of 499/1500o Emerich often used one of his segments to print a ledger line for notes beyond the staff (see Fig io); he was the only Italian printer to trouble to

FIG 28 Missale Romanum Venice: Johann Emerich of Speier, 3 VIII 1494, f k4 (mis-signed 14) (Biblioteca Comunale, Ferrara, S12.1.14.)

do so Emerich's staff lines consistently fit together so well that the eye often detects no break in the printed line The space is almost invariably without fluctuation, making possible an accurate second impression of the notes The stage of printing music books with staves and without notes proved to be of short duration Primarily a transitional practice of printers who lacked the appropriate type, it lost its validity quickly as various notational styles became available in movable metal type It is generally the later specialists in music type who tried printing staves without notes (Frankfurt and Scoto are publishers rather than printers) Books printed with staves provide an interesting glimpse of several printers' first essays into the field of music printing and illustrate a variety of technical solutions that would be integral to their later work in double-impression printing of music with type.

Music Printed from Woodcuts

Some of the earliest and best-known examples of music printed from woodcuts are found in twelve Italian music incunabula (Table 15).l The first woodcut music was printed in the Breviarium Lubicensis (GW5374) by Lucas Brandis in Lubeck in 1478.1 The second woodcut music was printed in the second edition of Niger's Grammatica ([Basel? ca I4851, Hii857); the first edition of 1480, printed in Venice, had used the first known mensural type (MI) The third woodcut music is the first Italian effort, a theory book with examples of mensural music printed in Bologna in I487 German printing historians classify such woodcut incunabula as "block printing" and do not distinguish between wood and metal techniques (the German term Blockdruck comprises both Holzschnitt and Metallschnitt).'3 However, all known incunabula music illustrations from blocks cut by hand are printed from wood The first application of metal engraving to music is thought to be the Intabolatura da leuto del divino Francesco da Milano, published sometime before I536.14 Not until the reappearance of engraved music in the i58os, first in Vincenzo Galilei's Dialogo della musica antica e della moderna (Florence: Marescotti, 158 i), did the technique become frequently used by printers.

Woodcut music illustrations made possible the early printing of theoretical works on music without the expensive production of mensural music types or fonts of complex plainchant neumes They constitute, therefore, an important part of fifteenth-century music printing Multiple editions of several titles prove the heavy demand for such publications; the large number of extant copies of such books as the folio Practica Musicae of 1496 suggests that publishers printed many copies in anticipation of high sales Not only students but also professors purchased the books: an extant copy of Ramis de Pareja's Musica Utriusque Cantus Practica contains marginal notes by Gaffurio (Bologna, Museo Bibliografico Musicale, A.80, see Figure 19), and a copy of Bonaventura da Brescia's Regula Musicae was owned by Pietro Aron (I507 edition; British Library, K.i.gIo [i]) Most Italian incunabula with woodcut examples-ten of the twelve-are theoretical works that require illustrations of mensural music, com i I For a review of the literature on incunabula with music printed from woodcuts, see Hermann Springer, "Die musikalischen Blockdrucke des 15 und i6 Jahrhunderts," Bericht uber den 2 Kongress der Internationalen Musikgesellschaft zu Basel 1906 (Leipzig: Breitkopf & Hartel, 1907), pp 37-46; Wilhelm Martin Luther, "Die nichtliturgischen Musikinkunabeln der Gottinger Bibliothek," Libris et Litteris: Festschrift fir Hermann Tiemann zum sechzigsten Geburtstag am 9 Juli i959, ed Christian Voigt and Erich Zimmerman (Ham-

burg: Maximilian-Gesellschaft, 1959), pp I31-37; Maria Przywecka-Samecka, "Problematik des Musiknotendruckes in der Inkunabelzeit," Gutenberg-Jahrbuch, 1978, pp 54-55; Wilhelm Martin Luther, "Der Blockdruck," in "Notendruck," Die Musik in Geschichte und Gegenwart, 9: cols I676-680 I2 Przywecka-Samecka, "Problematik des Musiknotendruckes," pp 54-5513 Springer, "Die musikalischen Blockdrucke." I4 An illustration is included in H Edmund Poole's "Printing," p 248.

Table 15 Italian music incunabula printed from woodcuts Date Author, Title Place: Printer References D 30 IV 1487 Burzio, Musices Opusculum Bologna: Ugo Ruggerio for GW 5796 8 Benedetto Faelli RISM 162 IGI 1954 15 XII 1492 Gaffurio, Theorica Musicae Milan: Filippo Mantegazza for GS 10437 I5 Giovanni Pietro da Lomazzo IGI 4115 7 III 1493 Verardus, Historia Baetica Rome: Eucharius Silber IGI 10146 156 H*I594I 28 IV 1493 Missale Romanum Venice: Johann Emerich of W924 93 Speier IGI 6630 1494 Missale Ambrosianum Milan: [Giovanni Antonio W 32 28 d'Onate] for Valentino de IGI 6546 Meregari 30 IX I496 Gaffurio, Practica Musicae Milan: Guillaume Le Signerre GW I0434 12 for Giov P da Lomazzo IGI 4112 H 7407 27 VII I497 Bonaventura da Brescia, Brescia: Angelo Britannico IGI 1952 3 Breviloquium Musicale RISM 1:162 [Regula Musicae Planae} 23 IX 1497 Gaffurio, Practica Musicae Brescia: Angelo Britannico GW 10435 13 IGI 413 H 7408 27 IX I497 Bonaventura de Brescia, Regula [Brescia]: Angelo Britannico GW 4833 4 Musicae Planae IGI 1953 21 IX 1499 Compendium Musices Venice: for Giovanni Battista GW 7263 10 Sessa RISM 2:925 3 IX 1400 Bonaventura da Brescia, Regula Brescia: Angelo Britannico GW 4834 5 [15300] Musicae Planae RISM : 162 IGI 1954 o1 IX I500 Bonaventura de Brescia, Regula Milan: Leonard Pachel for GW 4385 6 Musicae Planae Giovanni de Legnano IGI 1955 plex diagrams such as the Guidonian hand, or representations of the hexachord system that are

difficult to create in type A book printed in Rome in 1493 with woodcut music (see Table I5), however, includes the first secular polyphonic music printed in Italy By using woodcuts to print short examples of mensural music, printers avoided the challenge and cost of cutting mensural music types, thereby postponing their development The first mensural music type, a font in the standard white notation (Mi), had been used in 1480 for Niger's Grammatica in Venice, but white mensural type was not to appear again until the next century, in Petrucci's publications All the examples of incunabula with woodcut mensural music appeared outside Venice, in cities where mensural type was presumably not available.

Incunabula with music printed in Italy from woodcuts were not as concentrated in major centers of printing as were music books printed from type, presumably because the woodcuts were more easily produced by local craftsmen than were music types Bologna's single music book using woodcuts was a textbook printed for use at the local university; Brescia's four editions honored local authors; Rome's edition of Verardus was aimed at the fol-

lowers of the king of Naples Milan's editions include a missal for local use, two theoretical works by the head of the ducal chapel, and a reissue of Bonaventura's popular treatise on plainchant Venice's editions include two of a treatise that closely resembles that written in Venice by Franciscus de Brugis (first printed in 1499 in the Graduale Romanum), as well as the introduction to the reader of the missal printed by Emerich in I493 The twelve Italian music incunabula (Table 15) were printed from only eight sets of woodcuts and represent only eight different titles The four editions of Bonaventura da Brescia's Regula Musicae Planae use the same woodcuts for twenty-three pages of illustrations of roman plainchant In addition, the short, eight-leaf Venetian treatise on plainchant was reissued by the publisher with new woodcuts about I5o5. The two editions of the Practica Musicae of Gaffurio use the same woodcuts.'5 Music appears on very few pages in most of the books with music printed from woodcuts The diagrams in the Theorica Musicae of Gaffurio include actual notes on only one leaf, kiv, in a crude representation of the hexachords.'6 The Missale Ambrosianum includes on its final leaf one three-line staff of Ambrosian plainchant (Fig 29) The woodcuts of the Burzio treatise include illustrations of plainchant, mensural music, and diagrams on five

pages.7 Of the two Italian fifteenth-century editions of the Historia Baetica of Verardus, only that of 1493 can be located today; on the final two pages (ff e7V-e8) are printed four voice parts of an Italian song in praise of Ferdinand and Isabella, the first known publication of a polyphonic Italian song The Venetian plainchant treatise contains diagrams or plainchant music on eleven of its sixteen pages In later editions the music illustrations were set in type Three of the sets of music woodcuts were more ambitious Emerich's first printed missal, of I493, contained music on forty-six pages The style of Emerich's woodcut plainchant is close to that of his music type: a short-stemmed virga with unpointed notehead, lozenges set close together for the climacus, a stemmed diagonal, and an angular F clef The overall appearance of the music, however, is comparatively crude, because of the placement of the stem on the wrong side of the virga and a frequent inconsistency in the balance of the size of noteheads (Fig 30) The treatise on music theory by the Franciscan monk Bonaventura da Brescia contains music woodcuts on twenty-eight pages.'8 The Guidonian hand includes three accidentals: flat, natural, and sharp The chant examples are cut in roman plainchant notation with a long-stemmed virga and a liquescent neume as well as many compound neumes Again the effect is crude, because the

notes are often compressed into too little space to align with the text printed from type, the noteheads are far from square, and the neumes are irregular in size The music written in Bonaventura's own 15 The third staff off gg3 of the 1497 editions has been recut (British Library, K.1.g4) 16 Facsimile editions, Bologna: Formi, 1969; Rome: Reale Accademia d'Italia, 1934 Meyer-Baer describes the music woodcuts of the 1492 Theorica as: "cut in wood, in a wonderfully clear and precise manner The typographical beauty of these examples consists in the harmony of the proportions and in the exactitude of the execution" ("The Printing of Music," p 174), but she is clearly referring to the Practica Musicae of 1496 as her illustration confirms The printed music in the Theorica consists of crudely cut isolated notes in diagrams 17 Facsimile edition, Bologna: Formi, 1969 18 Facsimile edition, Milan: Bollettino Bibliografico Musicale, 1934 S Martinotti, "Bonaventura da Brescia (de Brixia)," in Dizionario biografico degli Italiani, vol 2 (Rome: Istituto della Enciclopedia Italiana, 1969), pp 631-32 FIG 29 Missale Ambrosianum [Milan: Giovanni Antonio d'Onate, 1494], f ¥IO (Biblioteca Nazionale Braidense, Milan, AM-IX-4I.)

FIG 30 Missale Romanum Venice: Johann Emerich of Speier, 28 IV I493, f 13 I (I7,4v) (Biblioteca Angelica., Rome, Inc 68.) manuscript hand in his 1489 text of the treatise displays a lightness and grace that were not translated into the woodcut.19 The masterpiece of early Italian woodcut music is the Practica Musicae of Gaffurio Numerous examples of roman plainchant on a four-line staff and of mensural music on a five-line staff are printed throughout the book The mensural music is in a graceful, well-spaced style that flows as naturally as a manuscript hand, with oval noteheads and flags curving down from the tops of note stems The roman plainchant is well spaced but less pleasing to the eye because of the lack of regularity in the supposedly square noteheads The woodcut music of Practica Musicae is a notable exception to the usual

poor level of craftsmanship of early woodcut music, which falls far below the quality of contemporary woodcut pictorial illustrations and initials The technique of printing music from woodcuts was used in the absence of music type or in 19 The manuscript is at the Museo Bibliografico Musicale, Bologna For a reproduction, see M T Rosa Barezzani, "I Britannico stampatori di musica a Brescia," Commentari dell'Ateneo di Brescia 175 (1970): 159.

competition with printers who had monopolies on music printing from type Woodcuts continued to be used, though infrequently, for printing music into the sixteenth century, reaching a high point in the Roman and Venetian music publications of Andrea Antico, which rivaled their typeset competitors Despite such exceptions, the use of woodcuts to print music has been a peripheral technique throughout the history of music books, employed only rarely and then primarily for short musical examples.

Music Printed from Type

The transformation of the music manuscript into the printed music book was completed in the i47os when, in the first book printed with music, metal type was used to print music notation on the lines and spaces of a previously printed staff The less complete techniques of printing notes or staves discussed earlier in this chapter were later developments Of the I56 Italian incunabula, 76 contain music notation printed from metal type cast in 38 designs Analysis of statistics on printers, their music imprints, and their music types clarifies the importance in the history of music printing of certain individuals-Emerich and Hamman-and of certain cities-Rome in the I470s, Milan in the I48os, and Venice in the i48os and especially the i490s During these decades there was a gradual addition of formats smaller than the early folios and a commensurate development of new small music types Type designers chose differing solutions to the challenge of reproducing in metal the varying stem lengths of manuscript notation and the complexity of compound neumes and liquescence, and some of the resulting types were more successful than others (see Part II for discussions of individual types and printers) By the end of the century, and perhaps much earlier, it is likely that printers delegated responsibility for music type to type professionals,

one of whom is named in archival documents as Jacomo Ungaro The printing of music with metal type became a specialized trade focused in a few cities and practiced by only a handful of craftsmen The alphabetic types of fifteenth-century printers were studied and classified at the turn of the century by Robert Proctor (An Index to the Early Printed Books in the British Museum, 1898 906) and Konrad Haebler (Typenrepertorium der Wiegendrucke, I905), illustrated in the Veroffentlichungen der Gesellschaft fur Typenkunde des XV Jahrhunderts (I907-1939), and further classified by the editors of the Catalogue of Books Printed in the XVth Century Now in the British Museum (i908 to the present) The omission of music type from those studies distorts the contribution of printers who specialized in music It is a serious misrepresentation of the printing career of Johann Emerich of Speier, for example, to describe him as a printer who used eighteen gothic text types and not to mention his four roman plainchant fonts, one extended to include mensural type Emerich has been praised for his fine text types in an unusually wide range of sizes and for his illustrated books, but his music printing has been an undervalued field of accomplishment His ambitious production of a gradual and an antiphonal in

roman plainchant printed from a complex font of type that was designed to handle the intricacies of melismatic and liquescent plainchant notation is a fitting capstone to the often uneven body of work that formed the transition from manuscript to printed book for the music service books of the Catholic church With the development by Haebler and Proctor of a system of alphabetic type classification by printer and size and design of type, printing historians acquired a basic tool for the attribution of incunabula without imprint information While the text types of music incunabula remain the basic tool for attribution, the relatively small number of music types makes them a useful additional aid The identification of the printer of a book's text type may not necessarily identify the printer of the entire book The music in the book may have been printed separately, by a second printer-a theory supported by the existence of some music incunabula in two variants, one with printed notes and staves and

Table 16 Classification of music types in Italian incunabula* Size Total Roman Ambrosian Gothic Mensural Small 5 5 Medium 14 10 2 I I Large I7 16 I Very Large 2 2 * With some exceptions, large missal types are defined as those printed on a staff of from 15 to 20 mm; medium missals, on a staff of I I to i5 mm; and small fonts on a staff of 9 to I mm the other with blank space for music The specimens and classification of music types used in Italy in the fifteenth century provide a tool for attribution of early music books Most of the music types can be assigned to printers and chronological periods A few that were used in books without colophons or with colophons that mention only a publisher can be attributed only to a city and time period or can be attributed to a particular printer with reservations From its types the 1487 Missale Romanum (D 73) published by Paganini can be attributed to Bevilaqua, who did the almost identical reprint for Paganini in 1499 (D I 13) with the same music and alphabetic types (see Fig 42) And the appearance of Benali's music type in the anonymously printed Missale Romanum (D 84) of the British Library is firmer evidence for attribution than a woodcut, an item that may be suspected to have circulated more frequently among printers than did music type A review of Italian music types discloses that nearly all are designs of roman plainchant and nearly half are in the "large" type size used for the

folio format of the missal used on the altar (see Table i6) By far the largest category of Italian music incunabula is that of the 121 missals No missal smaller than f-olio size was issued in Milan, and only two missals smaller than folio size in Rome, the quartos of 1488 and 1494, which used the large music type designed for folio format It was Venetian craftsmen, and particularly the printers Hamman and Emerich, who took advantage of the sizable new audience for small missals by introducing five small roman plainchant fonts Venice produced all the Italian small plainchant types (see Table 17), an innovation that dramatically extended the marketing of service books by lowering their cost and improving their portability Benali was the innovator who added plainchant type to the octavo missals that had been appearing in Venice since 1481 with either no space or blank space for music, or with stoves printed from rules Other small types quickly followed Benali's type used a virga with a very small notehead and an even smaller lozenge that were very difficult to print on the appropriate lines and spaces of a staff that was printed from irregular and wavy rules Novimagio used a type that was more balanced in design

though limited in number of characters, but his printed staves were grossly irregular The next three types, very similar in size, design, and number of characters, were used consecutively by Hamman, Bevilaqua, and Emerich of Speier Hamman Table 17 Small type fonts in Venice Date Printer No of Editions Type Measurement 1483 Benali R5: 10.25:I.252 X 4 1491 Novimagio I R5: 9-II:22 X 5 1493 Hamman 4 Rio: IO-IO.5:I.752(2) X 3.5 1497 Bevilaqua 2 R8: IO-Io.5:1.52 X 3.5 1498 Emerich 2 R20: IO-IO.25:I.52 X 3.5 Type measurement is in mm, staff:virga (notehead by stem).

Table I8 The relationship of book production to stem length of Roman Large Missal types Short Stems Long Stems No of Editions Printer Length No of Editions Printer Length 7 Han (Planck) 7.5 i Valdarfer 15 3 Girardengo 7.5 I Torresani 12 5 Hamman 7 I Ragazzoni 12 I Preller 6.5 i Arrivabene I 1.75 i Pinzi 6.5 I Bonini 11.75 i Emerich 6 I Scoto 1 I Torresani 5.5 i Britannico o1 2 Bevilaqua 12 printed octavos from I493 to 1495 and Bevilaqua printed two octavos in 1497 (and D 141* in 15oo?) Emerich began using a small type in 1498 The fact that in I493 Emerich published an octavo missal with forty-six pages of woodcut music (D 93) suggests that he lacked capital to purchase music type at that time; his first music printed from type appeared in a quarto missal of August 1494 (D 98) While the designs of plainchant type are very similar, the length of the note stem is a variable that dramatically changes the appearance of printed chant The stem lengths of Roman Large Missal types are compared in Table i8 The lightness and

fluidity of the long-stemmed pointed types of Torresani and Bevilaqua make a refreshing contrast to the heavy regularity of the short-stemmed unpointed notes of the three main music printers The most prolific Italian music printers, Han (Planck), Hamman, and Emerich, used types with a short stem; the notes seem to hang in midair rather than rest securely on the staff Designs with short stems need fewer variants to enable them to be printed on any line or space of the staff and therefore take less time to cut, cast, and set Such designs, then, were more practical, but they were farther from the manuscript plainchant models, which frequently had stems that reached from the notehead to the bottom of the staff One early roman plainchant font, that of Christoph Valdarfer, first used in 1482, avoided the problem of casting variable stem lengths by using rules placed below cast stemless noteheads (see Figs 56 and 57) The result is far from satisfying: the stems look weak and wavy and often are unconnected to the noteheads In settling on short-stemmed notes, aesthetic demands may have been sacrificed to the technical requirements of the compositor in the age of printing, imposing a standard on plainchant types that persists today It is interesting to note that mensural fonts followed a different path of development; varying stem lengths have been cast in metal from the first fonts of i480 to today.

The size of the notehead in relation to the space between staff lines and the size of the text type is another important determinant of the stylistic appearance of a font A notehead that takes up three-fourths of the space between staff lines looks heavy (RI2), whereas a note that fills only one-half the space (RI7) is much easier to read A notehead that is smaller than one-half a staff space (R23, RI i) is more difficult to read, perhaps partly because of the difficulty in printing the note squarely within a space or on a line If the notehead is as large as the x-height of the text type (as in D I43, which is printed with R30), the music overbalances the letters Most of the music types presented in this study can be associated with a single printer The only music font to appear in Rome in the fifteenth century was that introduced by Han and kept in use by Planck, who was evidently his heir However, the Medium Missal type introduced by Benali is apparently identical to that used in the only music book by the printer Piero di Piasi (D 69) and that used in

in unsigned missal (D 84) There are strong similarities between Scoto's types, Roman Large R3 and Medium R4, and those of Torti, Medium Ri2, and Ragazzoni, Large RI4 If Scoto was a publisher rather than a printer, the music in his 1482 missals may have been printed by Torti and Ragazzoni, who later used the types in signed books Because Ri2 and RI4 have certain idiosyncrasies that challenge that hypothesis, however, Scoto is assumed to have been the printer until further evidence is forthcoming Giovanni Battista Sessa is another individual mentioned in the imprints of music incunabula who was very likely a publisher rather than a printer Proctor suggested that Sessa relied on outside labor for the printing of his books, and that has proven to be true for the sixteenth-century output (see Chapter VII) Sessa's R24 is very similar to Emerich's R20, and one of the books printed with R24 uses the staff technique employed by Emerich It would not be surprising if Sessa's music type were found to have been crafted by a local type designer and used by hired printers.

Fifteenth-century music types had remarkably short lives While Hamman and Emerich had as many as four music type fonts in different sizes and Pachel had three, most printers had only one size of roman plainchant type Twenty-five of the Italian fonts were used in Italy for only one book At least one printer used his type for a second time outside Italy (Bonini's R3 i, in Lyons) Perhaps one explanation for the short life of types in Venice lies in the monopolies created by privileges either for the types themselves or for the special kinds of books for which they were designed Such privileges were not uncommon The first privilege granted to a printer in Italy went to the founder of the art in Venice, Johann of Speier, in I469, who received a five-year monopoly on all printing in the city; fortunately for the development of the field, he died a few months later In i495 the Venetian Senate granted a privilege for printing Greek to the printer who commissioned the first Greek types there, Aldo Manuzio, and in I498 to the music printer Ottaviano Petrucci for a mensural music type In i 50 Bevilaqua asked for a ten-year privilege to print small missals ("li messaleti picoli"), which apparently was not granted; he soon left Venice to take up the life of an itinerant printer The use of privileges to protect type designers encouraged the development of new designs but discouraged competition, since only one type could be granted official approval.

An attempt to follow the documentation from the Venetian archives for privileges for the first publication of the gradual and antiphonal in roman plainchant reveals a confusing sequence of requests by several individuals: I January 1497 Bernardino Stagnino requested a privilege of ten years to print the "Antifonario e Graduale di canto." (Fulin 62) 5 March 1497 Tommaso, a Venetian, received a privilege to print the "Antifonario Graduale et il Salmista in coro," not yet printed, and for which it was very difficult to find a publisher, because it was so expensive (Fulin 64) 6 December I497 Giacomo Britannico, citizen of Brescia and printer officiossimo, obtained the privilege to print the gradual and antiphonal (Fulin 73) 21 January 1499 Andrea Corbo received a ten-year privilege for capital letters of such a shape, height, and width that they could only be used to print choirbooks ("quod ipse solus facere possit stampare litteras ejusdem formae et grossitiei ac magnitudinis").20 (Fulin 9o)

A gradual was finally printed in Venice by Emerich for Luca Antonio Giunta in three volumes that appeared on 28 September 1499, February I5oo, and I March I5oo (D 17) The title page of the third volume states that it was printed under a privilege that also applied to the forthcoming antiphonal 20 Andrea Corbo is cited as Andrea da Corona (Andrea de Cronstadt, Andreas Burciensis, Corbo, Corvo, Corvus) in Gedeon Borsa, Clavis Typographorum Librariorumque Italiae 1465-I600 (Baden-Baden: Valentine Loerner, 1980) A typecutter and printer in Venice from 1476 to 1484, he was from Transylvania, then a part of Hungary, and was formerly known as Sigismund Corwin: "Andreas olim Sigismundi Corui inxisor literarum stampe (impressor liborum)" (Fitz, "Ungarische Buchdrucker," p I I5).

(1503-1504; unnumbered item after D 2) and psalter: Cum privilegio concesso ab illustrissimo Venetorum domino pro dicto graduali modo impresso et etiam pro antiphonario et psalmista iam immediate imprimendis: With a privilege granted by the most illustrious Signoria of Venice for the said manner of printing of the gradual and also for the antiphonal and psalter immediately to be printed: Despite the apparent expenditure of time and money by several printers and by the Venetian Senate and College of Counselors, Giunta managed to secure the privilege for plainchant service books that would bring his family money for reprints for more than a century It is not clear what happened to the privileges for printing choirbooks that were granted in 1497 to the Venetian Tommaso and the Brescian Britannico; perhaps they were unable to secure financial backing for so expensive an undertaking as was mentioned in Tommaso's privilege The large capital letters printed in red in the Giunta choirbooks may well be those cut by Corbo The alphabetic and music types cut for the choirbooks may have been acquired from Venetian type designers in a similar fashion, perhaps from Corbo's fellow countryman Jacomo Ungaro In the fifteenth

century many Italian printers had been commissioned to print liturgical books with music for foreign cities and religious communities, but by the mid-sixteenth century that business was to be concentrated in the hands of a very few.2' Of the men associated with liturgical music printing in Venice, only those who were active publishers in the fifteenth century managed to establish the family dynasties that were to dominate the trade in the sixteenth century-Giunta, Scoto, Sessa, Paganini, Torresani (with Manuzio), and Arrivabene Of those, Giunta was to be by far the most important.2 The existence of a large number of similar fifteenth-century Venetian alphabetic types has long puzzled bibliographers Proctor tried to explain their presence by suggesting that there might be "type factories independent of the printers" where types could be purchased.2 Remarkably similar plainchant music types also appear in Venice in the last decades of the fifteenth century As we have seen, Jacomo Ungaro, a typecutter who claimed responsibility for Petrucci's first music type, worked in Venice "for forty years" (from about I473 to 15 4), and other craftsmen in Venice in the 1490s called themselves typecutters and typecasters Some music printers are known by their privileges

or equipment to have made or commissioned their own types between 1474 and I492: Zarotto, Valdarfer, Bevilaqua, Britannico, and Girardengo Printers in cities far from the industrial development of Venice (Rome, Parma, Naples, Brescia, Bologna, Pavia) seem to have been more skillful in typecutting and casting in addition to the techniques used for printing books Similarities between the music types of the major music printers, Hamman and Emerich, may be the result of common techniques evolved during the time the two men worked together on music books as partners before they began printing music on their own It is also possible that Emerich was responsible for some of Hamman's music printing (his whereabouts are unknown between 1487, when he printed a missal with Hamman, and 1492, when he issued his first book) or that a separate type designer and cutter was used by both printers It is a fact that type foundries and type designers played an important role in Venice in the 1490s in the establishment of such a well-known printer as Aldo Manuzio, and it appears that the music printer Petrucci relied on type specialists to provide him with music types It would be surprising if other Venetian music printers of the I49os did not avail themselves of such accessible type craftsmen.

21 The surviving contracts for the first Ambrosian missal (D 23) and for a 1484 Florentine breviary provide insight into the publishing process for early liturgical books See Arnaldo Ganda, "La prima edizione del messale ambrosiano (1475): Motive pastorali e aspetti commerciali," La Bibliofilia 83 (1 981 ): 97 12; William A Pettas, "The Cost of Printing a Florentine Incunable," La Bibliofilia 75 (I973), 67-85 Kingdon's study of the economics of publishing a nonmusical liturgical book, the breviary, gives an account that parallels in many ways that of the liturgical books with music "The Plantin Breviaries," pp I34-50. 22 For a review of the trade in Venice in the sixteenth century, see Paul Grendler, The Roman Inquisition and the Venetian Press, 154o-1605 (Princeton: Princeton University Press, i977), pp 3-3223 Proctor, I: 14.

Were Venetian music types designed by type professionals? The first music type in Venice, introduced in I48o, was the first mensural type (Mi, white mensural notation) The second and third mensural types were also introduced in Venice, in 1500oo (Ri9, roman plainchant and black mensural rotation) and i o i (Petrucci, white mensural notation) Jacomo Ungaro claimed responsibility for Petrucci's font, the semibreve of which is nearly identical in size to the music font Mi used in 1480 in a book whose colophon tells us little about the edition's producers (the title was selected by Santritter and the printing was done by an otherwise unknown printer, Theodor Franck of Wurzburg) f Ungaro knew how to cut white mensural type for Petrucci in about i500 (the privilege for the type vas granted in 1498 and it was first used in 50oI), ie was certainly capable of playing a part in the creation of the equally demanding first black mensural type that appeared in the 1499 gradual in the same city If Ungaro contributed to the design or production of the mensural types of R2I, he would probably have produced the plainchant designs that make up the rest of the font If we know that Ungaro cut the second mensural type in Venice in about i500 and that he was living in Venice from about 1473, it is also reasonable to suggest that he might be responsible for the only other white mensural type cut there during that time, the 1480 type used by Theodor Franck of Wurzburg Furthermore, if a music typecutter was active in Venice as

early as 1480, most of the music type introduced there in the fifteenth century may have come from him rather than from the printers themselves The most important role in the history of Italian music incunabula, dominated by the Venetian contribution, may have been that of the music typecutter, without whom many printers and publishers who creatively set the type and marketed the finished product could not have printed music While there are many gaps in our knowledge of the origins and training of fifteenth-century printers, there is a consensus that after the art of printing reached its first full flowering in Mainz in the early I450s, craftsmen who were trained there moved to new localities to set up shops, trained others, and rapidly transmitted the technique outside Germany If music printing follows the process of linear development known for alphabetic printing, the training of music printers should lead back to early masters As would be expected with a linear development of printing from Mainz, the earliest printers in Italy were German immigrants Many of the first music printers in Italy were also Germans (see Table i9) The cast of characters in the history of early Italian music printers and types is smaller than it appears from the multitude of names in the colophons of music books Many who claimed responsibility for music books were publishers rather than printers: Giunta, Scoto, Paganini, Sessa, and probably Torresani and others as well The men who actually printed music books were not always

specified but certainly number fewer than the names listed with music types in Part II, since that list includes some publishers Of the twenty-two men who are known to have used music types in Italy during the fifteenth century, eight were Germans, one Dutch, one Dalmatian, and the other thirteen Italian natives of regions from Calabria to Lake Como German printers introduced music printing in Rome (Han), Milan (Valdarfer), Venice (Theodor Franck), and Naples (Preller) They issued the first signed books of roman, ambrosian, gothic, and black and white mensural types anywhere, though it is not certain that any of the nine printers cut his own types Undoubtedly, the contributions of German craftsmen dominated early music printing in Italy What of the chronological sequence of development of individual music types? The first music Table 19 German music printers in Italy Number Number Name of Types of Editions Han (Planck) i 7 Valdarfer 2 2 Pachel 3 4 Novimagio I i Theodor Franck I I Hamman 4 12 Preller I I Emerich 4 I4 Total 17 of 38 42 of 76

Table 20 Venetian incunabula music type by size and year Roman Small Missal Roman Medium Missal Roman Large Missal Type Number Date Type Number Date Type Number Date R5 1483 R4 1482 R3 1482 Rio, 14 149 1-492 R6 1484 R9, 14 1488-1489 R20, 24 1498 Ri2, I3 I489 RI6 1494 Ri8 1494 RI9, 22,23 1496 RII, 24 1497 R25 1499 type appeared about 1473 in a gradual and the next, in Germany, was used by Bernhard Richel (active I472-1484) for a Missale Basiliense in 1480 or I48i.2 The first Italian music types appeared in 1476 and 1477, followed by a proliferation of types in the i48os, mainly in Venice (see Table 20) Practically nothing is known about the printing of the ca 1473 gradual Richel had printed the first dated book in Basel in 1474 and his first Missale Basiliense, with blank space for music, about I478 (IGI 6552) Richel must have known of the earlier book, since Basel is located just outside the Diocese of Constance (see Map 3), for which the ca 1473 gradual was printed; indeed, he may well have received some training in the shop that produced it.

Did Ulrich Han, the German printer of the first music in Italy (1476 Missale Romanum), know of the ca I473 gradual? Han's dated imprints in Rome begin in I467, but no books were issued under his name alone between 4 October 1470 and 24 December 1474 Capable German printers were present in Rome who could have printed for him, including Stephan Planck, who in 1478 took over his shop Perhaps Han's role in Rome in the early I470S was that of an absent publisher and he was present at the first music printing A small Parma publisher, Damiano Moilli, issued the first gradual printed in roman plainchant type, in April 1477, six months after Han's missal It was the first printed book issued by Moilli, who spent most of his career preparing manuscript choirbooks for the local Benedictine monastery He published only three books after the impressive gradual, among them another landmark, the only incunabulum on calligraphy An acknowledged craftsman and artist, he seems to have been quite capable of designing and cutting the text and music types used for the gradual At his first appearance in archival documents in 1474, he was already listed as a printer, and Donati suggested that he may have been involved with the first printing in Italy.25 Where did he learn to cut types and print-more specifically, to print music?

The circumstances surrounding the music type introduced by Christoph Valdarfer at Milan in 1482 strongly suggest a link to the north Valdarfer, the first music printer in Milan, is known to have traveled from Milan to Basel to enter the employ of the printer Bernhard Richel during I4801481, when Richel issued the Missale Basiliense with music.26 Upon Valdarfer's return to Milan, he issued the Missale Ambrosianum of x5 March 1482, the first printed music in Milan from the first ambrosian plainchant type A few months later ( I September) he issued the Missale Romanum with music printed from the first roman plainchant type used in Milan At Basel under Richel he must have acquired the ability to cut music types or have found someone who could, as well as to use them to print music on staves in a two-color process His solution for music type, cast noteheads with stems apparently printed from rules, was unique Leonard Pachel, music printer in Milan from 1486, was, like Ulrich Han, from Ingolstadt Pachel first appears as the witness of a contract between Valdarfer and Cola Montano in 1474; Montano was the publisher of the first dated Missale Romanum, printed in Milan by Zarotto in 1474 Pachel

24 Arnold Pfister, "Vom fruhsten Musikdruck in der Schweiz," in Festschrift: Gustav Binz, Oberbibliothekar der Offentlichen Bibliothek der Universitat Basel zum 70 Geburtstag (Basel: Schwabe, I935), pp 160-77 25 Lamberto Donati, "Passio Domini Nostri Iesu Christi: Frammento tipografico della Biblioteca Parsoniana," La Bibliofilia 56 (1954): 8I-21526 BMC VI: xxiii.

issued his first book in Milan in i477 and his first.Missale Romanun (with space for music) in 1479 In [476 he would have been about twenty-five years old, connected with printing, and without a shop; night he have become acquainted with music type it the printing of Han's Missale Romanum in 1476 at Rome? Because the earliest Venetian music types cannot be associated with particular craftsmen, there is no way to connect them with the beginning of music printing The first, a mensural type, was used by Theodor Franck of Wurzburg, who is described in the colophon simply as one who labored on the book ("fert opus") The second type was used in books issued by Scoto, a publisher who seems to have preferred to hire others to print his books From Petrucci's 1498 privilege we find it was common knowledge that protracted attempts had been made outside Italy to develop music type for both mensural and plainchant notations: "many not only in Italy but also outside Italy had long sought to print music in a very convenient manner and, consequently, plainchant much more easily."7 As we have seen, there is a possibility that Ungaro was cutting music type in Venice as early as 1480, when the first music was printed there in Niger's Grammatica, and that he was responsible for several of the music types used there.

The marshaled evidence, though far from proof, supports the hypothesis of transmission of music printing in linear fashion to future practitioners The current state of knowledge about music printers, publishers, and typecutters makes it difficult to distinguish between their activities and to allocate responsibility for music types The hypothesis promotes an awareness that the reinvention of music printing in neighboring cities is less plausible than shared knowledge of the technique among printers trained in shops where music printing took place Italian incunabula printed from music type reached a peak of quality in types, printing, and marketing that allowed them to dominate the international market for liturgical books in the late fifteenth century Printing began in Mainz, music printing began in or near the diocese of Constance in southern Germany or Switzerland, and Italy's music printing began in Rome, but Venice provided 27 For the Venetian, see Appendix i, no 2 the site of the prodigious activity of the last decade of the century Italy printed two-fifths of all music incunabula, and Venice alone printed about one-fifth Thirty-eight music types were used in Italy, twenty-five of them used in Venice and probably produced there Persuasive factors attracted the international pool of talent that contributed to

printed music books The political and monetary stability of the Republic of Venice made it a good risk for capital investment at a time when southern Germany was going through upheavals The metallurgy industry in Venice that supported a large shipbuilding industry was competitive with the technology of any other European city At that time it may have been the richest city in the world and had risk capital to finance publications for international circulation The Venetian shipping routes to the west and land routes to eastern European countries were well established A highly developed system of privileges may well have prevented other Italian cities from competing in printing and distributing books Italians such as Petrucci and Aldo Manuzio emigrated from the Papal States to the Republic of Venice to take advantage of its attractions The city fathers made it possible for immigrants to acquire citizenship quickly so that they could compete in the central marketplace of Europe But without the talents of creative individuals, the growth of early music printing could not have taken place The inventor of movable metal music type remains elusive, but identification has been made and the role clarified of many of the small

group of printers, publishers, and typefounders who began to print music books in Italy From lists and classification of Italian editions and types, a new picture emerges of a well-developed industry that brought double-impression music type to a peak by the end of the century Recognizable in Italian music types are solutions to problems of printing music notation and staves that still form the substance of today's few music types Much remains to be learned from examination of other copies of the books described, by seeking the answers to the right questions in archival documents, and by establishing lists of editions and classifying types for other countries in Europe Already indisputable is the major significance of Italy's contribution to the early history of music printing.