The “Carlo G manuscript”

New light on early seventeenth century

accompaniment and diminution practices

Elam Rotem

Introduction

The newly surfaced “Carlo G manuscript” constitutes an exceptionally

important finding in the fields of early continuo and vocal ornamentation
alike. Containing around 300 pages of written-out keyboard accompani-
ments for highly ornamented monodies and duets, this manuscript is prob-
ably one of the most concrete and substantial sources that survives from that
time.

The so-called “Carlo G manuscript” was bought around 15 years ago in a flea
market near Vienna for circa 60 euros (!). Later, after its importance had
been recognized by Roman Chlada, who subsequently wrote his bachelor
thesis about it,1 the manuscript was scanned and sent to Sotheby’s for sale.
Iain Fenlon was assigned to examine the manuscript and to write a short
description of it, and in 2007 the manuscript was resold for 65,300 pounds.2

1 Roman Chalda, “Die Begleitung am Tasteninstrument bei Carlo G.”. The work can
be downloaded on this link: www.romanchlada.com/download/BA-Arbeit.pdf (10 January
2017).
2 Sotheby’s, London, “Music”, 4 December 2007, lot 34: “Early seventeenth-century
Italian motets”. According to Iain Fenlon, the manuscript contains “321 pages, folio
(ca. 33×23.5 cm), including the 2-page autograph index signed, the double-page openings
numbered 1–158, bull’s head watermark, with 4 smaller fragments loosely inserted, con-
temporary limp vellum, vellum tabs, Italian provenance (probably Rome or the Papal Sta-
tes).” Further findings by Fenlon and the team from Sotheby’s (2007) are available on this
link: www.sothebys.com/en/auctions/ecatalogue/2007/music-l07408/lot.34.html (31 May
2017).
402 Elam Rotem

The manuscript is currently inaccessible in private (anonymous) hands;

research on its content is only possible thanks to the scans made prior to the
auction.3
The manuscript contains 89 liturgical and para-liturgical pieces in Latin
for one or two highly ornamented vocal lines and organ accompaniment (see
Ex. 1 showing the first pages of music in the manuscript). There are 55
motets for one voice and organ (four of which are repeated, with the organ
part replaced by a chitarrone tablature), 23 motets for two voices and organ,
one motet for two voices and basso continuo, and one motet for four singers
and two organs (two groups of performers each with two voices and organ).
In addition to the vocal motets, there are five instrumental toccatas which
are meant to serve as “preludes” for specific motets (e. g. toccata per il mot-
tetto …). See the complete index in the appendix of this article, Table 1.
75 of the pieces in the manuscript are compositions by Carlo G; his
name appears fully only on the index of the manuscript, but unfortunately a
stain hides his last name (see Ex. 2). In the rest of the manuscript, only the
initials C G are used (e. g. see the top left part of Ex. 1). Alongside composi-
tions by Carlo G, the manuscript also includes a few pieces (or arrangements
of pieces) by Giulio Caccini (1551–1618; named Giulio Romano in the
manuscript), Luca Marenzio (1553–1599), Paolo Quagliati (c. 1555–1628),
Girolamo Giacobbi (1567–1629), and Bartolomeo Barbarino (c. 1568–
c. 1617). Excluding the piece by Marenzio (see below), thus far none of the
other pieces in the manuscript have been found in other sources; all the
items in the manuscript are unica.

3 Unfortunately, several photos are missing from the original scanning, and this is
evident in the index (see Table 1 in the Appendix of this article). As we do not have
access to the manuscript, we cannot tell if only photos are missing or if some pages are
physically missing in the original. However, the few missing photos seem insignificant in
the context of the manuscript as a whole.
 The “Carlo G manuscript” 403

Ex. 1: Carlo G MS, Tota pulchra es, fols. 1v-2r.

404 Elam Rotem

Ex. 2: Carlo G MS, title on the table of content.

Historical context

Since we do not know the identity of Carlo G,4 we may learn about his
manuscript’s historical context through related material, like pieces men-
tioned in the manuscript that are known to us, composers that are men-
tioned in the manuscript, and lastly, musical sources that are musically sim-
ilar to the “Carlo G manuscript”.
My research has identified only two pieces mentioned in the manuscript
which can be found in the historical literature. The first piece is the motet Sic
parasti cor meum (no. [32]), which bears the title “Se bramate ch’io mora di
Luca Marenzio”. The mentioned madrigal, originally written for six voices, is
found in Marenzio’s fourth book of madrigals (Venice, 1587).5 Therefore it
would seem that the motet presented by Carlo G is a contrafactum and
arrangement of a secular Italian six-voice madrigal into a sacred Latin motet
for two highly ornamented voices and organ accompaniment. Interestingly,

4 The only suggestion for the identity of “Carlo G” was made by the team of Sotheby’s
while preparing the manuscript for sale (see no. 2). They suggested the name Carlo Gra-
tiani based on an evidence presented in: Tibor Tallian, “Archivdokumente über die Tätig-
keit Stefano Landis in Rom in den Jahren von 1624 bis 1639”, in: Studia Musicologica
Academiae Scientiarum Hungaricae 19, fasc. 1/4 (1977), 267–295 (“[…] A spese dette
s[cudi] 68.80 m[one]ta in cro come sopra pagati con mand[at]o 3023 al Sud[ett]o
S[igno]r Carlo [Gratiani] p[er] pagarli a diu[er]si musici […]”). Fenlon believes that
“while it is of course possible the evidence is not strong” (private correspondence, 1. 12.
2014).
5 Luca Marenzio, Il quarto libro de madrigali a sei voci, Venice: Giacomo Vincenzi
1587.
 The “Carlo G manuscript” 405

while the original madrigal was written in a set of high-clefs (G2, G2, C2, C3,
C3, F3), this arrangement is transposed a fourth lower, using the practice
nowadays referred to as chiavette;6 this feature, in addition to the written-out
ornamentation and written-out accompaniment, demonstrates the practi-
cality of the manuscript. A comparison of the original Italian text of Mar-
enzio’s madrigal with the Latin contrafactum in the Carlo G manuscript fol-
lows:

Original 1587: Contrafactum in C. G. manuscript:

Se bramate ch’io mora Sic parasti cor meum
Gli è van credete à me vostro desire et semitas meas sic direxisti
Che chi vita non ha non può morire quod iram inimici non timebo
Co’ bei vostri occhi voi o Iesu virtus mea
Datemi vita e poi gloria mea et salus
Havrà ciascun di noi quel che desia ostende mihi lucem et satiabor
Voi la mia morte & io la vita mia. et quiescam semper in vita mea.

The second piece which is mentioned in the manuscript and found elsewhere
is placed in the Alleluia section of the motet Mater Hierusalem (no. [37]).
On the upper margin of fols. 70v–71r, above the score, the copyist/scribe
wrote: “Passo d’Oratio Vecchi del madrigale quella ch’in mille selve à 5”
(“passage from Orazio Vecchi’s madrigal quella ch’in mille selve for five voi-
ces”).7 Indeed, a comparison proves that Vecchi’s passage was copied and re-
used by Carlo G in what is, apart from being transposed by one tone, an
original composition (see Ex. 3).
In terms of related musical style, a source that should be mentioned is
the Arie Devote by Ottavio Durante Romano, which was printed by Simone
Verovio (Rome, 1608). The motets in that collection, for one voice and basso
continuo, have much in common with Carlo G’s monodies; both are highly
ornamented pieces that use similar diminution figures and ornament signs.
The same publisher, Verovio, also published Luzzaschi’s famous madrigali

6 A lot was written about the issue of high clefs combinations and their transposi-
tions. For a general overview see Patrizio Barbieri, “‘Chiavette’ and modal transposition in
Italian practice (ca. 1500–1837)”, in: Recercare 3 (1991), 5–79.
7 Vecchi’s madrigal is found in Madrigali a cinque voci di Horatio Vecchi […]. Libro
primo, Venice: Angelo Gardano 1589.
406 Elam Rotem

Ex. 3: Comparison between a) Orazio Vecchi, last section of quella ch’in mille selve (Madri-

gali a 5vv, 1589), and b) Carlo G MS, a section from Mater Hierusalem.

(Rome, 1601), which with its keyboard accompaniment also has much in
common with the “Carlo G manuscript”. Another publication with some
connections to the “Carlo G manuscript” is Barbarino’s second book of
motets from 1614,8 where for each of the book’s monodies in the book Bar-
barino didactically supplies two versions, semplice and passaggiato (without

8 Il secondo libro delli motetti di Bartholomeo Barbarino, Venice: Bartholomeo Magni

1614/5.
 The “Carlo G manuscript” 407

and with ornamentation). As Carlo G’s diminution style is quite similar to

Barbarino’s, and since he included in his manuscript one piece by Barbarino
(no. [5]), it is possible to hypothesize some connection between the two
composers. Another source which has similarities in style to the “Carlo G
manuscript” is the Mottectis et dialogis by Quagliati,9 yet another composer
mentioned in the manuscript.
Considering the references to pieces from the late 1580s (Marenzio’s and
Vecchi’s madrigals), the general flourishing period of the other composers
mentioned in the manuscript, and the similarities with other musical sour-
ces, it seems possible that the “Carlo G manuscript” was scribed and in use in
the period of around 1600 to 1620. As there are several related sources from
Rome (more Roman connections highlighted below), one may suggest that
Carlo G was somewhere in Rome or in an institution with Roman con-
nections.

Notation

The vocal line is written on a standard 5-line staff, while the keyboard part is
written on two 7-line staffs, one for each hand. The notation of keyboard
parts at the time – the intavolatura – was not standardised, and in different
sources we find a variety of numbers of lines on the staffs.10 The music writ-
ten in the intavolatura is an intabulation of several voices, but as Alexander
Silbiger explains, the keyboard intavolatura of the late Renaissance and early
Baroque is not a score in the modern sense where one could see the original
details of the counterpoint; it is closer to a tablature notation where the
musician is merely instructed as to which key to strike and when.11 Practi-
cally speaking, while a modern keyboard score can express the details of
counterpoint using different stem directions for each voice, consistent usage
of pauses for each voice, and extra lines to show where each voice goes, the

9 Paulo Quagliati, Mottectis et dialogis Rome: Baptistam Robletum 1620.

10 For example, this is evident in the keyboard parts found in the different publications
of Simone Verovio (1586–1601).
11 Alexander Silbiger, “Is the Italian keyboard ‘intavolatura’ a tablature?”, in: Recercare
3 (1991), 81–103.
408 Elam Rotem

Ex. 4 : Costanzo Antegnati, L’Antegnata intavolature de

ricercari d’organo (Rome, 1608), fol. 1v.

Italian intavolatura cannot; it simply does not employ these notational

means, and therefore provides the reader with less information than a mod-
ern keyboard score. Thus, it is not surprising that most of the polyphonic
keyboard repertoire from that time was published in partitura – full scores –
rather than in intavolature notation. This is an important point to acknowl-
edge before reading and assessing Carlo G’s keyboard parts.
Carlo G’s keyboard parts contain very accurate, skilful counterpoint.
What might seem to be parallel octaves and fifths in Carlo G’s keyboard
parts is actually one of the features of playing an arrangement of several voi-
ces on a keyboard, either when adapting an existing polyphonic piece or
when playing or improvising in polyphonic textures. One can find such
apparent parallels (which are normally an outcome of voice crossings) in
any intabulation of polyphonic music.12 Example 4 demonstrates where, in a
ricercar by Costanzo Antegnati, parallel octaves and fifths are visible to the
reader (see asterisks), yet are surely not integral to the original fabric of
counterpoint.

12 This topic is mentioned by Lodovico Viadana in his famous preface to cento concerti
ecclesiastici Rome: Giacomo Vincenti 1602, point 9, and implied also in the treatise of
Francesco Bianciardi Breve regola per imparar’ sonare sopra il basso con ogni sorte d’istru-
mento, Siena: [s. n.] 1607.
 The “Carlo G manuscript” 409

An interesting feature of the “Carlo G manuscript” is the addition of

optional variants for certain vocal passages. In many pieces, there are one or
more ossias; these appear typically, but not always, on the final cadence, sug-
gesting different variants for certain diminutions. These optional variants are
not consistent in their nature: sometimes they offer an easier and shorter
passage, but sometimes a longer and more virtuosic one. They are mostly not
labeled with text, yet when they are, it is along the lines of: “passaggio altro”
(“another passage”; fol. 34r), or “per chi vuol variare” (“for those who would
like to vary”; fol. [159r]).
In the motet Et dilectus meus (no. [67]), one of the optional variant pas-
sages bears the text: “à S. Ma Vitta hò dato q[ues]to”. This might suggest
that this passage was sung by or written for a certain Maria Vittoria.
Unfortunately, the name Maria Vittoria is rather common and no historical
candidate has been identified. The solo singing of women in church was not
a completely foreign concept during this time. Along with Caccini’s monody
(no. [6]), which is labeled “per monache” (“for nuns”), we know of the
famous Vittoria Archilei who sung with Emilio de’ Cavalieri in the Holy
week of 1585 in Rome, and of 1599 in Pisa.13

Ornamentation

In addition to having written-out virtuosic diminutions, the “Carlo G manu-

script” is swarming with ornament signs and performance instruction signs
unlike in any other contemporaneous source; the vocal and keyboard parts
have in total circa 700 signs. Although there is no table or other means of
explanation in the source itself, it is possible to recover the meaning of the
signs with a high degree of certainty. The most common sign is the plus sign
(“+”; circa 480 signs); less common sign is the “g” sign (circa 200 signs);
and rather rare is the “t” or “tr” signs (circa 20).
The “+” sign has two usages/functions. The first and most common
occurs as it is found on the last note of passaggi. It could be found both in
the voice and in the accompaniment, but it is more common in the voice,

13 See Elam Rotem (ed.), Emilio de’ Cavalieri – lamentations and responsories for the
Holy week, Wilhelmshaven: Florian Noetzel Verlag 2014, note 10.
410 Elam Rotem

and often occurs before the introduction of a new syllable in the text. We do
not know of another source that employs the “+” sign in this way;14 never-
theless, there are two Roman sources from 1615, both printed by Niccolò
Borbone, which may shed light on the matter. Francesco Severi’s Salmi
(Rome, 1615) employs an “F” sign on the last notes of diminutions, a similar
context to the common usage of the “+” sign in the “Carlo G manuscript”.
According to Severi, the “F” sign is an abbreviation of “fermare” (“stop”),
and was used as a performance instruction to avoid confusion between pas-
saggi or at the end of a passaggio.15 It could be added that these little stops
ease the coordination between the singer and the accompanist when per-
forming a lot of diminutions. See Ex. 5 from Severi’s publication, where the
“F” sign is located at the end of each of the four passaggi (there is also one ‘t’
[trillo] sign). Comparing Severi’s use of the “F” signs with Carlo G’s “+”
signs, we can see that both are used in a similar way; in Ex. 6 we can see the
“+” signs at the end of each of the four passaggi in the upper voice, and once
in the second voice. The similar usage of Severi’s “F” and Carlo G’s “+”
increases the plausibility that both signs have the same meaning. Moreover,
the “F” and the “+” are typographically similar, and it could be that Carlo
G’s “+” arises from a transformation of Severi’s “F”. The second Roman
source that describes a brief articulation stop between quick passaggi is Fres-
cobaldi’s first book of Toccatas (Rome, 1615).16
Beyond these references, there are certain points where it seems that
Carlo G realized the “+” sign himself. In the organ version of Convertisti
planctum (no. [24]) the “+” sign is used on the first measure, whereas in the
version with the chitarrone accompaniment (no. [51]) the sign is missing yet
the rhythm is different, allowing a written-out short break before the next
measure. See Ex. 7 for a comparison between the two versions.

14 In later French music the “+” sign can represent a trill ornament; this is not the
case in the “Carlo G manuscript”.
15 Francesco Severi, Salmi passaggiati per tutte le voci nella maniera che si cantano a
Roma […], Rome: Nicolò Borboni 1615, fol. [IIv]: “Che si fermi un poco chi canta dove
ritrova la lettera .F. e questo tanto quanto non paia di cantare seguitamente l’un passaggio
con l’altro, non interrompendo la voce […].”
16 Girolamo Frescobaldi, Toccate e partite d’intavolatura di cimbalo […] Libro primo
(Rome: Nicolo Borbone 1615). See preface, point 4.
 The “Carlo G manuscript” 411

Ex. 5: F. Severi, Salmi (Rome, 1615), fol. 55.

Ex. 6: Carlo G MS, Miserere, fol. [160r].

Ex. 7: Carlo G MS, comparison between the beginnings of

the two versions of Convertisti planctum: a) from no. [24];

b) from no. [51]).

The second usage of the “+” sign is in the context of a series of four crome
(quavers), where the “+” is found on the second and fourth notes. Out of
approximately 480 “+” signs, circa 50 are pairs of signs in such a context.
There is no direct evidence for the sign’s meaning, but a possible inter-
pretation is that it represents a transformation of the four equal crome into a
412 Elam Rotem

Ex. 8:From G. Caccini, Nuove musiche (Florence, 1602),

preface, facsimile and transcription.

Ex. 9: Carlo G. MS, Peccavi super numerum (no. [2]), voice part, mm. 30–32. Upper line:

as written; lower line: possible realization.

‘lombardic’ rhythm of short-long, short-long (that is, biscroma, dotted cro-

ma, biscroma, dotted croma: see Ex. 8, the version “with more grace”). These
transformations of rhythms from “square” to “sprezzatura” are described by
Giulio Caccini in text and in music alike. See Ex. 8 from Caccini’s famous
preface to his Nuove Musiche (facsimile and transcription) where he demon-
strates how to achieve “più grazia” (“more grace”) when performing a series
of four crome. Although this rhythm is highly common in early Baroque
monody, it is not found in the “Carlo G manuscript” in standard notation.
Thus, it could be that while Carlo G did not clearly notate the common
“lombardic” rhythm, it is nevertheless present in the source, notated in a spe-
cial way with the help of the “+” signs. See Ex. 9 for a possible realization of
these ornament signs in a motet by Carlo G.
 The “Carlo G manuscript” 413

The other ornamentation signs, the “g” [gruppo] and the “t” [trillo], are
found also in other sources from the beginning of the 17th century and have
a rather clear meaning.17

Original pieces or arrangements?

Ten out of the 89 motets in the manuscript are indicated by the scribe to
have originated in polyphonic pieces for six, seven, or eight voices.18 That is,
these ten motets for one or two ornamented voices and organ accompani-
ment were arranged from polyphonic sources. Maybe also the rest or at least
several of the pieces in the manuscript are not original monodies or duets,
but were arranged from polyphonic compositions.
For example, the analysis of Veni dilecte mi (no. [33]) reveals that this
motet was most probably originally polyphonic. Throughout the motet, the
voices in the accompaniment can be easily fitted with text underlay. In origi-
nal monodies this is often not possible due to long notes in the bass that
cannot carry all the text’s syllables. Looking at the text of the first part of the
motet, we find further indication that the piece was originally of a poly-
phonic source:

[Song of Songs 7:11–12]

Veni dilecte mi egrediamur in agrum Come, my beloved, let us go to the countryside;

commoremur in villis let us spend the night in the villages.
Mane surgamus ad vineas Let us rise early to go to the vineyards,
videamus si floruit vinea to see if the vines have budded,
si flores fructus parturiunt to see if their blossomanuscript have opened,
si floruerunt [mala punica] if [the pomegranates] are in bloom,
ibi dabo tibi ubera mea. there I will give you my love.

17 For example, both in Emilio de Cavalieri’s Rappresentatione di anima, et di corpo,

Rome: Nicolò Mutij 1600, and in Giulio Caccini’s Le nuove musiche, Firenze: Giorgio
Marescotti 1602, the gruppo and the trillo are explained.
18 See in Table 1, nos. 7, 21, 22, 32, 35, 38, 61, 77, 81, and 86.
414 Elam Rotem

Ex. 10: a) Carlo G MS, section from Veni dilecte mi; b) Reconstruction of a possible origi-

nal motet in six voices.

I have placed the words “mala punica” (“pomegranates”) in bold and in

square brackets because they do not appear in Carlo G’s arrangement. In the
place of the missing words, an instrumental interlude takes place. Upon close
inspection, it becomes clear that the voices in the accompaniment can bear
the complete text, and probably did in the original polyphonic version. See
 The “Carlo G manuscript” 415

Ex. 10 : (continued).

Ex. 10 for the relevant bars (a) along with a reconstruction of a possible orig-
inal in six voices (b). This explains the omission of the words “mala punica”,
which renders the verse nonsensical.19 It seems that this detail was over-

19 It is true that portions from Song of Songs texts were regularly edited and used free-
ly; this is clearly evident in the other Song of Songs settings found in the “Carlo G manu-
416 Elam Rotem

looked by the arranger in the process of adapting the motet for two voices
and instrumental accompaniment.
With this finding, it is safe to say that the motet Veni dilecte mi probably
originated from a polyphonic source, despite not being indicated by the
scribe. Most importantly, it suggests that there might be other pieces in the
“Carlo G manuscript” with a similar history.
Proving that monodies in the “Carlo G manuscript” were originally
composed as monodies and are not an arrangement of a polyphonic source
is rather difficult. An exceptional example is Caccini’s monody Benche sovra
le stelle / Deus dominus meus (no. [6]); as we know other monodies by Cac-
cini, and since this motet is practically identical to them in style, there is no
reason to think that this piece was not a monody in the first place. Another
piece that very much seems like an original monody is Sub umbra illius (no.
[70]). This is due to some typical seconda prattica features found at the end
of the motet, on the words “quia amore langueo” (“for I am sick with love”).
On the vocal line where this text is first presented, there is a slow chromatic
alteration from c ’’ to c♯’’, then an irregular downward leap of an augmented
fifth to f ’, which is then followed by a resulting unprepared seventh with the
bass (g ’ against A). In the remaining bars of the piece there are no less than
four cases of unprepared sevenths between the voice and the bass on the
words “amore” or “langueo”. Such writing is typical of the expressive mono-
dies for voice and continuo at the first two decades of the 17th century; it
therefore seems likely that Carlo G’s Sub umbra illius was conceived as a
monody. Several other pieces might also have originated as monodies;
unfortunately, their origins cannot be traced definitively, leaving open the
possibility that they were originally polyphonic pieces that were arranged
into monodies.
We have seen that Carlo G did not only borrow and arrange pieces; he
also ‘contrafacted’ them (as is evident in Marenzio’s madrigal and Vecchi’s
passage). Caccini’s monody (no. [6]) is presented with two texts in different
languages – Italian and Latin. Perhaps the Italian text is from the original

script”. However, in a context where complete verses are used (the second part of the
motet also uses complete verses) there is no reason to believe that the text was edited in
this case.
 The “Carlo G manuscript” 417

piece (which unfortunately is not known to us from other sources) and the
Latin text is a contrafactum by Carlo G. There are also reasons to believe that
the pieces by Quagliati and Giaccobi (nos. [7] and [11] respectively) are
contrafacta; both their texts are unique and are not found in liturgical sour-
ces. Thus, the pieces by Marenzio, Caccini, Giaccobi, and Quagliati might
have been borrowed, “contrafacted”, and arranged by Carlo G for his needs.
Since these pieces are contrafacta, finding their original sources based on the
music itself is difficult, should they survive at all.

Basso continuo?

Assuming that the “Carlo G manuscript” contains at least some original

monodies and duets, the next question that arises is whether those pieces
were composed first with basso continuo, and only later realized and intabu-
lated. Unfortunately, unless we find one of Carlo G’s pieces in another source
with basso continuo line rather than with an intabulation, we cannot answer
this question with certainty. However, there are still several points that tie
the “Carlo G manuscript” to the study of early continuo.
Firstly, the “Carlo G manuscript” constitutes a significant new link in
the study of the late Renaissance accompaniment. It proceeds along similar
lines to the Verovio prints with written-out keyboard accompaniment
(1586–1601; the most famous among these is the Madrigali by Luzzaschi
mentioned above), as well as publications like Mikołaj Zielenski’s Offertoria
totius anni (Venezia: Giacomo Vincenti 1611), which supplied a full score as
the organ accompaniment part. However, the “Carlo G manuscript” is dis-
tinguished from these sources as it evidences the new style of the early bar-
oque in terms of both ornamentation and composition.
Secondly, as mentioned above, there are several other publications con-
taining liturgical monodies and duets with basso continuo which are very
similar musically and stylistically to the monodies of Carlo G. If we were to
re-notate Carlo G’s monodies, retaining only the bass line from the keyboard
intabulation, it will be almost impossible to distinguish between Carlo G’s
music and original contemporary monodies with continuo. Therefore, since
the monodies of Carlo G ultimately seem so similar to original monodies
with continuo, there is little reason to doubt that at least some of them were
418 Elam Rotem

conceived with basso continuo, and that the intabulations represent just
another method of notating practically the same music.
Thirdly, it seems that Carlo G’s accompaniments are in line with early
continuo features represented in other sources: doubling the skeletal lines of
the vocal parts, equal distribution of the voices between the hands, incon-
sistent number of voices, apparent parallel fifths and octaves, small imitative
sections, ornaments, and small passaggi between sections. Carlo G’s accom-
paniments provide us with many pages of written-out examples, confirming
much that was formerly known about early continuo.
To summarize, it has been shown that the boundaries between an
arrangement and an original monody, as well as those between Carlo G’s
accompaniments and basso continuo realizations, are vague. It seems that
the accompaniment of a polyphonic piece, of an arrangement of a poly-
phonic piece, and the realization of basso continuo in a monody, are practi-
cally similar in this musical style; these processes are all characterized by
similar features.
It should be added that there are two pieces written with a ‘real’ basso
continuo part in the “Carlo G manuscript”: item no. [83], a toccata for violin
and two figured basso continuo lines (in the style of pieces with two choirs),
and the last item, Miserere mei (no. [89]) for two voices and unfigured basso
continuo.

Instruments and Instrumental toccatas

The following instruments are mentioned in the “Carlo G manuscript”:

organ, chitarrone, lirone, violin, basso di viola, and a consort of viols. The
main accompaniment keyboard instrument seems to be an organ. The
manuscript includes two references to the instrument: once in the toccata
before Florete flores (no. [39]): “Toccata per Florete flores con due organi”;
and once at the end of Panis angelicus (no. [80]): “non si faccia cadenza con
l’organo quando suona il violino”.
The next accompaniment instrument is the chitarrone, which in the
manuscript is referred to with the following abbreviation: “Chittar.e” or
“Chitt.e”. The meaning of the abbreviation is evident when in the table of
contents, as just before the motets with the chitarrone tablatures, it is written
 The “Carlo G manuscript” 419

“Per il Chittar.e”.20 These pieces were already presented in the manuscript

with keyboard intabulation, but in this section are repeated with chitarrone
tablature. In these cases it does not seem that the organ and chitarrone were
meant to play their respective versions simultaneously.21 Further references
to the chitarrone appear in the title of the motet Quasi stella matutina
(no. [27]), during the motet Quam dilecta (no. [60]), where a small ossia
with diminutions appears below the organ part, and in two toccatas
(nos. [64] and [79]). Thus, the chitarrone, according to Carlo G, was not
only a possible replacement for the organ; it occasionally played together
with the organ, as well as with other instruments. Regarding the kind of
instrument, the tablatures indicate a 13 or 14 course lute with 8 courses on
the fingerboard. The tuning seems to be without re-entrant tuning: finger-
board: g ’, d ’, a, f, c, G, F, E♭; diapasons: D, C, B ’, A ’, G ’.22
The violin, apart from being mentioned in the context of two toccatas
(nos. [79] and [83]), is also mentioned in the context of three motets. In all
of these motets (Mater Hierusalem civitas, Non turbetur cor vestrum, and
Panis angelicus; nos. [37], [38], and [80] respectively), the violin is to
replace one of the voices. The violin parts only contain partial text underlay,
and at the points where the part is either too fast, too high, or has too long
passages for a singer to sing, the text underlay disappears entirely, proving
that this is primarily an instrumental part. At some points, possible bowings
markings are seen; see Ex. 11 for an excerpt from the violin part in Panis
angelicus (no. [80]). Notice how the text underlay stops and that possible
bowing markings are included.
The lira is mentioned twice in the manuscript in the context of two toc-
catas (nos. [64] and [79]). A tablature is supplied, and in both toccatas the
instrument is shown to play slow chords, as opposed to the chitarrone, which

20 It seems then that Carlo G spelled the name of the instrument with two t’s: “chittar-
rone”. Variations in spelling were highly common at the time, and do not in any way
suggest that the meaning was other than chitarrone.
21 It will certainly not be possible using only the surviving copy of the manuscript, as
the two versions are found on different pages. Furthermore, the accompaniments of the
two versions sometimes contradict each other.
22 Many thanks to Ori Harmelin and Ryosuke Sakamoto for their insights concerning
the chitarrone.
420 Elam Rotem

Ex. 11: Carlo G MS, fol. 145r, from Panis angelicus.

Ex. 12:Carlo G MS, “Toccata per il mottetto sopra scritto Haurietis aquas con lira, Chitt.e
et basso di viola”, fol. 126v.

plays bass diminutions. Carlo G’s lira might have been a lira da braccio; how-
ever, due to the high positions and its musical function, it seems more likely
to have been a lirone. See Ex. 12 for a toccata with basso di viola, lirone, and
chitarrone. As there are but few sources concerning such instruments, it is
not surprising that there is no historical tuning fitting the Carlo G lira/lirone
tablatures. Based on the tuning found in the treatise of Cerreto Scipione
(1601),23 a possible tuning might be: e ’, a, d ’, g, g, c ’, c.24

23 Scipione Cerreto napolitano, della prattica musica […], Naples: Iacomo Carlino
1601.
24 Many thanks to Elizabeth Rumsey, Brigitte Gasser and Baptiste Romain for their
insights concerning the lirone. More about the Lirone can be read in Imke David, Die
sechzehn Saiten der italienischen Lira da gamba, Bissendorf: Orfeo 1999.
 The “Carlo G manuscript” 421

The basso di viola appears in two toccatas together with the lirone; this
bass instrument (probably what is nowadays referred to as viola da gamba)
might have been meant to complement the missing bass line of the lirone.
The Toccatas in the “Carlo G manuscript” have a special feature that is
rarely found in other sources; each of the toccatas is connected with a certain
motet, and meant to be played before it. Furthermore, there is certain musi-
cal connection between the toccata and its motet. There are no two toccatas
in the manuscript with the same instrumentation, and each is unique: one
for organ (no. [75]); one for two organs (no. [39]); one for bass viol, chi-
tarrone, and lirone (no. [64]); one for violin, bass viol, chitarrone, and lirone
(no. [79]); and lastly, one for violin and two basso continuo lines (no. [83]).

Conclusions

The “Carlo G manuscript” provides a vivid testimony of liturgical and para-

liturgical musical life from the early 17th century. Both the vocal lines and
instrumental parts are notated in detail, giving us a glimpse into what was
actually played and sung; no prior knowledge of counterpoint or diminu-
tions is needed, all is written-out for the performer.
The “Carlo G manuscript” might be seen as a sort-of hybrid between
Renaissance traditions and early Baroque style, both in music and its nota-
tion: Musically, the manuscript demonstrates that the motet is still very
much connected with its polyphonic origin, yet at the same time the motet
gets closer to the expressive and “modern” early 17th century monody. The
notation, as in other sources from the beginning of the “basso continuo era”,
is experimental; it represents a temporary link in the process that will end
with the standardization of the basso continuo notation. The vague bounda-
ries between an arranged polyphonic motet and an original monody are
prominent, and the fact that the written-out intabulation exemplifies the
characteristics of what we know of early continuo may sever to enrich our
understanding of that period and its performance practices.
While the manuscript has been dated rather convincingly between 1600
and 1620, and located in Rome or with Roman connections, Carlo G remains
largely unidentified. As the musical forces implied by the manuscript are
rather small, we could imagine that he was not working for a large musical
422 Elam Rotem

establishment. Beyond this conjecture, we hope that further studies of the

manuscript will reveal more. Apart from finding concordances, a possible
direction of research might be in examining the specific variants of the litur-
gical texts found in the manuscript, as they might lead us to Carlo G’s specif-
ic church or institution. An edition of selected pieces from the manuscript is
available online,25 and a recording was released on 2017.26

25 Elam Rotem, Carlo G Manuscript – Selected piece: http://imslp.org/wiki/Special:

ReverseLookup/454312 (10 February 2019).
26 Profeti della Quinta, The Carlo G Manuscript: Virtuoso liturgical music from the ear-
ly 17th century (Glossa GDC 922516).
Appendix
Table 1: Index of the “Carlo G manuscript”
In the comments: text which is originally found next to the piece, or next to the title of the piece in the original table of content [TOC] is in quotation
marks.

No. Composer Title Setting Folio Paratext / Comments

[1] C. G. Tota pulchra es 1v, organ 1v

[2] C. G. Peccavi super numerum 1v, organ 2v

[3] C. G. Sidus de nube 1v, organ 4v “In festo S. Thomae de Aquino”

[4] C. G. Adiuro vos, filiae Hierusalem 1v, organ 5v

[5] Bartolomeo Barba- Cantate Domino canticum novum 1v, organ 6v “Del Barbarino”;
rino [TOC:] “Bartolomeo Barbarino”
c. 1568–c. 1617

[6] Giulio Caccini Benché sovra le stelle / 1v, organ 8v “Giulio Romano. Per monache” (‘for nuns’); [TOC:] “Giu-
1551–1618 Deus Dominus meus lio Rom.”

[7] Paolo Quagliati Alma mater pietatis 1v, organ 9v “Paulo Quagliati à 6”;
c. 1555–1628 [TOC:] “Paulo Quagl.”

[8] C. G. Magnificat 1v, organ 11v

[9] C. G. Virgo prudentissima 1v, organ 14v

[10] C. G. Candidi facti sunt 1v, organ 15v

[11] Girolamo Giacobi Luce gratiae tuae 1v, organ 16v “D. Girol.o Giacobi”
1567–1629 [TOC:] “D. Girol.o Giacobi”
The “Carlo G manuscript”
423
 424

No. Composer Title Setting Folio Paratext / Comments

[12] C. G. Induit me dominus [incomplete] 1v, organ 17v Incomplete due to a missing photo. However, an additio-
Elam Rotem

nal version with chitarrone is complete (no. [48]).

[13] C. G. Ego flos campi 1v, organ 18v A version with chitarrone is in no. [49].

[14] [no attribution] Domine non sum dignus [I] 1v, organ 19v

[15] C. G. Domine non sum dignus [II] 1v, organ 20v

[16] C. G. Ecce cor meum 1v, organ 22v

[17] [no attribution] Salve Regina 1v, organ 23v

[18] C. G. Congratulamini mihi 1v, organ 26v

[19] C. G. Audite celi 1v, organ 27v

[20] C. G. Confiteor Deo [I] 1v, organ 29v

[21] C. G. In te Domine speravi 1v, organ 31v “In te Domine speravi.

Il Coro è à 8 voci”

[22] C. G. Ecce nunc 1v, organ 33v “Ecce nunc. Il Coro è à 7 voci”

[23] C. G. Vidi speciosam 1v, organ 35v This item is followed by a transposed version of the same
motet “una quarta alto” (a fourth higher).

[24] C. G. Convertisti planctum 1v, organ 37v A version with chitarrone is in no. [50].

[25] C. G. Deus canticum novum 1v, organ 39v

[26] C. G. Ardens est cor meum 1v, organ 41v

No. Composer Title Setting Folio Paratext / Comments

[27] C. G. Quasi stella matutina 1v, organ 43v “Per il Chitte” [chitarrone]
the organ intabulation contains only a diminuted bass line

[28] C. G. Defecit gaudium 1v, organ 45v

[29] C. G. Laudate pueri dominum 1v, organ 46v “Laudate pueri dominum”

[30] C. G. Nisi Dominus 1v, organ 49v “Nisi Dominus”;

Incomplete due to a missing photo

[31] [no information Tibi laus tibi gloria 2v, organ [51v] Only one photo (p. 52r) is extant from this piece. The title
due to missing is available from the TOC.
page]

[32] Luca Marenzio Sic parasti cor meum 2v, organ 53v “Se bramate ch’io mora di Luca Marenzio”;
1553–1599 [TOC:] “Luca Marenzio”

[33] [of unknown Veni, dilecte mi 2v, organ 55v “d’incerto”; [TOC:] “incerto”.
author] Second part: “Vulnerasti cor meum”

[34] C. G. Vulnerasti cor meum 2v, organ 60v

[35] C. G. Sancti tui Domine 2v, organ 63v “à 8”; [TOC:] “à 8”

[for two choirs]

[36] [of unknown Amor Jesu dulcissime 2v, organ 66v “d’incerto”
author]

[37] C. G. Mater Hierusalem civitas 2v, organ 68v for voice and violin

[38] C. G. Non turbetur cor vestrum 4v, 2 organs 71v “à 6”; [TOC:] “a 2 cori”;
The “Carlo G manuscript”
425
 426

No. Composer Title Setting Folio Paratext / Comments

for two choirs:

Elam Rotem

(I) two singers and organ

(II) singer, violin, and organ

[39] [C. G.] Toccata per Florete flores con due organi 2 organs 78v Toccata for two organs for the following item.

[40] C. G. Florete flores 2v, organ 79v

[41] C. G. Pone Domine 2v, organ 83v

[42] C. G. Gaudet anima mea 2v, organ 85v

[43] C. G. Jucundum sit Jesu 2v, organ 87v

[44] [C. G.] Florete flores [transposed] 2v, organ 90v “abbassato una voce”
A transposition of a tone lower of the piece above

[45] C. G. Magnificat 2v, organ 93v

[46] C. G. Exurgam diluculo 2v, organ 98v Incomplete due to missing photos

[47] C. G. Cantate Domino 2v, organ 102r

[48] C. G. Adiuro vos, filiae Hierusalem 1v, chitarrone 103v Chitarrone version of no. [4]

[49] C. G. Induit me dominus 1v, chitarrone 104v Chitarrone version of no. [12]

[50] C. G. Ego flos campi 1v, chitarrone 105v Chitarrone version of no. [13]

[51] [C. G.] Convertisti planctum 1v, chitarrone 106v Chitarrone version of no. [24]

[52] [no attribution] Salve suavissimum cor 1v, organ 107v

No. Composer Title Setting Folio Paratext / Comments

[53] C. G. Confiteor Deo [II] 1v, organ 108v

[54] C. G. Egressae sunt mulieres 1v, organ 109v

[55] C. G. Nigra sum 1v, organ 111v

[56] [no attribution] Astitit regina 1v, organ 113v

[57] [no attribution] Gaudeamus omnes 1v, organ 114v

[58] C. G. Congratulamini mihi omnes 1v, organ 115v

[59] — –- — –- Incomplete material due to missing photos

?

[60] C. G. Quam dilecta 1v, organ 118v A small diminution for “Chitte” [chitarrone] is included.

[61] C. G. Benedictus Deus 1v, organ 119v “A 8”; [TOC:] “A 8”

[62] C. G. Magnificat 1v, organ 122v “2di toni”

[63] C. G. Haurietis in aquas 1v, organ 125v

[64] [C. G.] Toccata per il mottetto sopra scritto Hau- lirone, chitarrone, 126v Toccata for the motet above. Incomplete due to a tear in
rietis aquas con lira, Chitte et basso di viola and bass viol the patch of which the end of the toccata was written on.

[65] C. G. Domine audivi auditum tuum 1v, organ 127v

[66] C. G. Exultabimus in te 1v, organ 128v

[67] C. G. Et dilectus meus 1v, organ 129v

The “Carlo G manuscript”
427
 428

No. Composer Title Setting Folio Paratext / Comments

[68] C. G. Defecit gaudium [transposed] 1v, organ 130v [TOC:] “abbassato”. A transposition of a tone lower of the
Elam Rotem

piece in no. [28]. The accompaniment is incomplete.

[69] C. G. Iste est qui ante Deum 1v, organ 131v “Per S. Bernardo”

[70] C. G. Sub umbra illius 1v, organ 132v

[71] C. G. O pretiosum et admirandum 1v, organ 133v

[72] C. G. Nunc dimittis 1v, organ 135v

[73] C. G. Anima mea liquefacta est 1v, organ 137v

[74] C. G. Inebriabuntur ab ubertate 1v, organ 139v

[75] [C. G.] Toccata di Stabat mater dolorosa organ 140v Toccata for the motet in no. [78]

[76] C. G. Iam hiems transjjt 1v, organ 141v “Per la Madalena”

[77] C. G. Num quem diligit 1v, organ 142v “A 8”; [TOC:] “à 8”

[78] C. G. Stabat mater dolorosa 1v, organ 143v “la sua toccata è à 141” [no. 75]
“Seconda parte è à 117” [the second part incomplete as
the photos of 116v–117r are missing].

[79] [C. G.] Toccata al mottetto Panis Angelicus con Violin, lirone, chi- patch Toccata for the following motet; written on a patch.
violino Chitte et lira, et basso di viola tarrone, and bass [145r]
viol

[80] C. G. Panis angelicus 2v, organ 145r For violin and voice.
No. Composer Title Setting Folio Paratext / Comments

[81] C. G. Sicut sponsus matris 2v, organ 147v “A 6 per sonar quattro viole et cantar due soprani . le parti
sono su’l libro […?]”;
[TOC:] “à 6 due voci et 4 viole”

[82] C. G. Salve puella 2v, organ 149v

[83] [C. G.] Toccata per decantabat che è à car. 53 Violin, 2 BC 150v Toccata for the motet in no. [85].

[84] C. G. Haec est virgo 1v, organ 151v

[85] C. G. Decantabat populus israel 2v, organ 152v “la toccata di questo mottetto e[?] à car. 51”

[86] C. G. Benedicite Deum coeli 2v, organ 153v “A 7”S; for two choirs.

[87] C. G. Ego dormio et cor meum vigilat 2v, organ 155v

[88] C. G. Iubilate Deo 2v, organ 156v

[89] C. G. Miserere Mei 2v, BC [159r] Written on a patch.

The “Carlo G manuscript”
429