The Word in the Image: an Epiconographic Analysis of Reformed Mosaics in Rome (Twelfth-century)

Full text

* I am grateful to the Pontifical Institute of Me-
diaeval Studies of Toronto and to the Getty
Foundation of Los Angeles which generously
supported my research, as well as to the Medie-
val Academy of America for providing a travel
grant (funded by John Goelet) which allowed
me to communicate my research at the MAA
Annual Conference in Vancouver (2008). I am
also profoundly grateful to the Scuola Normale
Superiore of Pisa which recently offered me a
grant to work on epiconography. I again benefi-
ted from important discussions and suggestions
with Prof. Herbert Kessler. Finally, my thanks to
Lucy Donkin, David Defries, Anne Dunlop and
Louis Hamilton with whom I discussed my work
and who helped me with my English. Obviously,
any errors remain my own responsibility.
1 On the relationship text and image, see: Ric-
cioni 2008a, 465-469; Riccioni 2010/2011, 326-
327, with bibliography.
STEFANO RICCIONI
The Word in the Image: an Epiconographic
Analysis of Mosaics of the Reform in Rome*
Abstract
Using a new methodological approach, called epiconography, the present paper examines the most
significant monuments produced in Rome during the Gregorian reform: the mosaics of S. Clemente
and S. Maria in Trastevere. This approach situates the works in the historical and cultural contexts
of their production, and attempts to resolve the divide between “visible” and “legible”, figure and
text, which has been present for so long in art-historical studies. The two mosaics are exemplars of
a practice of “visual composition” that followed the rules of medieval rhetoric in order to emphasize
an ecclesiastical message. This strategy was typical during the Gregorian reform, since S. Clemente
was composed under the direct influence of the initial ideals that gave impetus to this ecclesiastical
reform, while S. Maria in Trastevere constitutes the apex of “Gregorian art”, showing the triumph of
the reformed Church. In both mosaics, the scripts guide the viewer by the type of letters used, the
placing of texts and images, and their colors, depending on the position of the viewer in the archi-
tectural space of the church. Gregorian art demonstrates how traditional models of decoration are
reinterpreted to create new schemata, closer to the contemporary message of the Church. Reformers
created a visual rhetoric based on the display of closely connected scripts and images, which sys-
tematized knowledge for the benefit of the beholder.
Historians of art and architecture have traditionally displayed little interest in
including textual components in their analyses. When they do discuss texts, they
focus mostly on defining a methodological dichotomy between text and image,
with the result that they treat even inscriptions displayed on monumental works
as something “other”, that is, as elements symbolically and aesthetically out-
side the composition. 1 This attitude can be traced back to Giorgio Vasari, who
defined inscriptions in works of art as “gofferia” (awkwardness or clumsiness),

86 STEFANO RICCIONI
2 Vasari, Le vite, Barocchi & Bettarini (eds.) 1967,
II, 171.
3 Riccioni 2008a, 465, with bibliography.
4 On the signature in art, see: Chastel (ed.) 1971;
Klotz 1976; Claussen 1992; Claussen 1981; Di-
etl 1995; Fraenkel 1992; Dietl 2009. For a more
comprehensive project on signatures in works
of art, see: Donato (ed.) 2003a [2008]; Donato
2003b [2008].
5 The bibliography is vast. See Riccioni 2008a.
More recently, interest in text has moved to-
wards a new methodological approach, see: Do-
nato 1997; Méhu 2006a; Kessler 2007b; Thunø
2007; Gardner [forthcoming].
6 Kessler 2000; Castelnuovo & Sergi (eds.) 2004;
Ganz & Lentes (eds.) 2004; Bouché & Ham-
burger (eds.) 2005; Hahn 2006.
7 Illich 1993; Cavallo 1994; Barreto, Cerman,
Soubigou & Toutain-Quittelier (éds.) 2007.
8 Reflections along these lines inform the major-
ity of the medieval debate about images, the
‘word’, and text (whether biblical or liturgical),
see: Gombrich 1969; Gombrich 1982; Riccioni
2008a, 465. On the function of Christian art,
see: Duggan 1989; Duggan 2005.
9 Gregorius Magnus, Registrum Epistularum, 9.209,
768, lines 12-14; Gregorius Magnus, Registrum
Epistularum, 11.10, 874, lines 23-26. Kessler 2006.
considering them on par with jokes. 2 Thus, inscriptions (called tituli) have been
treated using two main interpretative approaches: in the first, their relationship
to the images is ignored in favour of their apparent value as literary texts and
historical sources; in the second, they are treated as complements to and de-
scriptions of the images. 3 Although recent studies, especially those that focus on
a particular type of text in art – the signature of the artist 4 – have begun to revise
our understanding of the relationship of text to image, 5 the traditional attitude
still prevails in most scholarship. This conservatism has had a particularly del-
eterious effect on our understanding of medieval art because, in the Middle Ages
especially, images were treated as controlled and selected visual transpositions
of texts, while displays of the written word on visible monuments (even when
not read) indicated authority (auctoritas). Furthermore, words were interpreted
as visible signs of an invisible referent. 6 In fact, because of its predominantly
Christian nature, the medieval image is connected to the Word Incarnate and
therefore to the definitions of the words “visible” and “legible”. 7 Even though it
cannot be established that images could be “read” more easily than words, me-
dieval image-making was intended to be a narrative and didactic art addressed
especially to the illiterate: it was intended to be the “literature of the laity” (lit-
eratura laicorum). 8 These strands converged in Pope Gregory the Great’s (d. 604)
iconic statement that “pictures are the scripture of the uneducated,” an axiom
that was repeated with many variations in later discussions of the function of
art. 9 From this perspective, the division between text and image is a matter of a
specific literacy rather than ontology. On the one hand, this view made images
available for exegesis according to the allegorical hermeneutics used for sacred
scripture. On the other, it meant that the rhetorical arts of classical antiquity
and monastic discourse could be applied to the composition of images.
Working within this medieval perspective to advance and celebrate their ide-
als, the Gregorian reformers created a rhetorical art by “composing” new im-
ages. During the twelfth century, reform theologians accorded an important role

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 87
10 On the existence of a theory of Art in the Grego-
rian reform, see: Kessler 2007a; Kessler 2007b;
Riccioni 2010/2011, 328, with bibliography.
11 Petrus Damianus, De picturis, PL 145, 589B-
596B. Leclercq 1960, 233; Riccioni 2006, 20-21;
Kessler 2007a, 25, 33-35; Speciale 2009.
12 Kessler 2007a, 33-34. Peter Damian’s poems
and sermons contain several interesting verses
that could have been composed (or simply
used) for picture captions, see: Petrus Damia-
nus, Carmina sacra et preces, PL 145, 917-985.
13 On the concept of ‘heresy’ in the Middle Ages
and more specifically during the Gregorian re-
form, see: Lambert 2002 [1992].
14 Bruno Signinus, De sacramentis ecclesiae, PL
165, 1089B-1110A; Bruno Signinus, De figuris
ecclesiae, PL 165, 875A-902B; Bruno Signinus,
De ornamentis ecclesiae, PL 165, 901B-942D;
Honorius Augustodunensis, De gemma Animae,
PL 172, 541-738. Hamilton 2005; Hamilton
2007; Quintavalle 2003.
15 The foundations of the Church are Jesus and
the prophets: Bruno Signinus, De sacramentis
ecclesiae, PL 165, 895C; the columns are the
apostles: Bruno Signinus, De figuris ecclesiae,
PL 165, 896A; the windows are the Fathers of
the Church: Bruno Signinus, De figuris eccle-
siae, PL 165, 896B; the walls represent the con-
gregation of saints: Bruno Signinus, De figuris
ecclesiae, PL 165, 896A. On Bruno of Segni, es-
pecially on De Laudibus ecclesiae, its symbolical
meaning and its importance for church dedica-
tion, see, Hamilton 2010, 162-211.
16 Bruno Signinus, De figuris ecclesiae, PL 165,
896B.
17 Bruno Signinus, De ornamentis ecclesiae, PL
165, 940C.
18 Bruno Signinus, De figuris ecclesiae, PL 165,
886C-886D.
to church decoration, and it is possible to argue, with Herbert Kessler, that they
created a kind of “theory of Art”. 10 For example, in the eleventh century, the
monk Peter Damian wrote a letter to Abbot Desiderius of Montecassino con-
cerning images that show Christ flanked by St Peter and St Paul, a common ar-
tistic composition. 11 He also wrote picture captions that directed viewers to the
higher, spiritual meaning of what was visible. 12 The words were meant to pre-
vent viewers from interpreting the images as simple descriptions of reality and,
more importantly, to help them escape the danger of idolatry and “heresy”. 13
Reformers also interpreted the architecture and decoration of churches allegori-
cally as a reflection (image) of biblical and liturgical knowledge (texts). In the
writings of Bruno of Segni and Honorius Augustodunensis, the church building
is a sacred space, a microcosm designed to reflect the macrocosm. 14 Bruno of
Segni described the church building as a symbolic universe, and suggested a
comparison between architecture, liturgical furniture, decoration, biblical sym-
bolism and the living members of the Church. 15 In his writings, the unity of the
church building is based on concord and peace, as an example to the faithful. 16
Bruno wrote: “In the house of God nothing must appear fatuous or foolish, ugly
or impure”; 17 and in the De figuris ecclesiae he declared: “In the temple nothing
is idle; whatever is written or carved is written for our instruction. The walls
themselves teach us and, in a certain way, speak to us”. 18
Exegesis of the visual image logically implied a concern for its rhetorical com-
position. The Gregorian reformers created a practice of visual composition that
followed the rules of contemporary rhetoric in order to emphasize an ecclesias-
tical message. Much of this composition was based in what Mary Carruthers has
called “monastic rhetoric”, which cast rhetoric as an orthopraxis designed, like

88 STEFANO RICCIONI
19 Carruthers 1998, 1-6, 60-115.
20 In the Middle Ages, most scholars attributed the
Ad Herennium to Cicero. Freedberg 1982, 87-97;
Murphy 2005, 1-26; Ward 1995; Ward 2006.
21 Camargo 1991, 17-28; Bolgar 1982. In the De
inventione, Cicero gives the “five parts of rheto-
ric” as: invention (inventio), arrangement (dis-
positio), style (elocutio), memory (memoria) and
delivery (pronuntiatio); Murphy 2005, 1-2.
22 Murphy 1974, 89; Carruthers 2008 [1990],
chap. 6.
23 Yates 1966; Mary Carruthers demonstrated that
the techniques of rhetoric, used during the clas-
sical period, were adopted by monks to create
a map of loci, a kind of mnemonic archive. See
Carruthers 2008 [1990], 154; Carruthers 1998,
81-82. On composition of letters and books, see
Baldwin 1928; McKeon 1942; Murphy 1974;
Parkes 1976.
24 Alberic, monk and teacher at Monte Cassino,
composed the Dictaminum radii [Albericus
Casinensis, Flores rhetorici; Alberic of Monte
Cassino, “Flowers of Rhetoric”]; on Alberic
but not specific on ars dictandi, see: Radding
& Newton 2003. He was among the first medi-
eval philosophers to connect classical rhetoric
to the writing of letters, helping create the ars
dictaminis. On the origins of the ars dictaminis,
see: Murphy 1971a; Bloch 1972; Murphy 1971b,
56-64, 67-69; Larmon Peterson 2003. On Alberic
and the birth of the ars dictaminis in Bologna,
see: Licitra 1977.
25 The literature is vast; see, for example, Toubert
1990, 93-138, 193-238 [2001, 73-102, 143-175];
Riccioni 2006, 11-15 and Riccioni 2010/2011, 321-
324, with bibliography. Grégoire 1971; Kitzinger
1972a; Cowdrey 1986 [1983]; Bloch 1986; Speci-
ale 1991; Avagliano (ed.) 1997; Cowdrey 2000.
asceticism, to shape character and was the dominant mode of rhetorical train-
ing from the fourth through the eleventh centuries. 19 A new interest in rhetoric
emerged in the eleventh century, based on a revived interest in the principles
enunciated in “Ciceronian” texts, especially the De inventione and the anony-
mous Rhetorica ad Herennium. 20 This interest resulted in the ars dictaminis, a
kind of theory of composition 21 given authority by the antiquity of the classi-
cal books (auctoritas antiquitatis) from which it drew. 22 This rhetoric influenced
various enterprises – poetry, preaching, the arts of memory and the writing of
letters and books. 23 Arguably, it also influenced the composition of images in
works of art. Alberic of Montecassino (d. 1088), a supporter of Gregory VII and
an influential rhetorician introduced the ars dictaminis in the school of Monte-
cassino. 24 The close relationship between the Gregorian reformers in Rome and
the abbey of Montecassino 25 helps to explain some of the novelties in Roman
church decoration in this period.
Using a new methodological approach, the present paper examines the most
significant monuments produced in Rome during the twelfth century, the mosa-
ics of S. Clemente and S. Maria in Trastevere. The two mosaics are exemplars of
the monumental and rhetorical art of the Gregorian reform, since S. Clemente
was composed under the direct influence of the initial ideals that gave impetus
to this ecclesiastical reform, while S. Maria in Trastevere constitutes the apex of
Gregorian art, showing the triumph of the reformed Church.
The Epiconographic Methodology
As an art historian, I am interested in understanding the visual discourse gen-
erated by the interaction between images and the written word, especially as ex-
pressed in monumental works displayed to the public. When displayed, writing

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 89
26 Petrucci 1986, XX (the English translation, Lap-
pen 1993, does not contain the methodological
introduction quoted).
27 On this argument, see: Favreau (ed.) 1996;
Méhu 2006b; Treffort 2006.
28 Favreau 1979, 16; Favreau 1997. For a recent
methodological approach to medieval epigra-
phy, see Treffort 2008, 3-10.
29 Petrucci 1994, 819.
30 Campana 1984, 363.
31 Campana 1967; Campana 1976, 85.
32 Riccioni 2008a; Riccioni 2008b.
is an image and an object, visible and tangible, even before it is a text to be read.
In purely iconographical terms, images are signs that bear an intrinsic mean-
ing. This is particularly true of inscriptions presented in monumental contexts. 26
Working together in a complex reciprocal relationship, the scripts and the other
images give the monument/icon a unified and highly articulated meaning that
cannot be divorced from the physical impact of the image on the public. In fact,
the inscriptions best reveal the complexity of the visual message. If the work
of art is recognized as an ensemble of diverse semantic vehicles, which include
both images and texts, inscriptions are essential for defining and understand-
ing the compositional process and the function of the object. Inscriptions allow
viewers to understand the modes of communication of the work of art as a result
of a coherent cultural system. Indeed, when words and images appear together
in the same context, the figural and linguistic systems of communication derive
meaning from one another and generate in their turn new semantic content.
To reflect on these themes is to confront the constitution of the medieval im-
age, understood as a complex discourse in which epigraphy played an important
role. 27 In the discipline of epigraphy some of these themes are now being ad-
dressed; it is no longer limited to the study of inscriptions in hard substances
such as stone or bronze, but is open to the function of “universal and lasting
publicity”, 28 a certain solemnity of purpose, and a slow execution that employs
particular techniques of script. 29 In this way, epigraphy shares with the history of
art the same objects of investigation and the same material approach. It pays at-
tention to style and form; the sign and the drawing constitute the graphic nature
of the written word. Inscriptions provide a multi-layered testimony comprising
three fundamental aspects – text, script, and monument 30 – and therefore can-
not be divorced from the context to which they are linked organically and with
which they form a single “complex monument”. 31 Still, epigraphy lacks the aes-
thetic, iconographic and stylistic means to treat the study of texts in images as a
unified whole. That is to say, what is still missing is a precise methodology that
considers the inscriptions as visual objects functioning in various ways in the
work of art, to be investigated in the totality of their aesthetic components. I call
this methodology “epiconography”. 32
Epiconography aims at filling the gap between text and image, reading and
seeing, by studying textual components in images, their visual disposition and
their function in the visual narrative, but also by treating scripts as images. The
term epiconography combines the original meanings of epigraphy (™pi-gr£fein =

90 STEFANO RICCIONI
33 This section is taken from my book. The literature
on the mosaic is vaste, for a comprehensive bib-
liography see: Riccioni 2006; Croiser 2006a, 209-
218; Riccioni 2007a; Riccioni 2010/2011, 341-344.
34 On the consecration of S. Clemente in 1118-
1119, see: Barclay Lloyd 1986; Barclay Lloyd
1989, 43-51; Claussen 2002, 303; Riccioni 2006,
3; Quintavalle 2009, 409, 437, note 41. Quin-
tavalle suggests the mosaic was finished in
around 1110. The hypothesis is plausible, al-
though the scholar does not provide any new
information.
35 On inscriptions in S. Clemente, see: Nilgen
1996, 160-162; Favreau 1997, 233-236.
to write on) and iconography (e„konograf…a = figurative representation), without
forgetting, however, the other possible meanings of gr£fein (to write, but also to
draw, paint, depict) on (™p…) the image (e„kèn), where the script itself is an icon. It
addresses, therefore, a way of seeing that is not simply the combined results of an
epigraphic and iconographic examination, but an interpretative system (both de-
scriptive and analytical) that considers the artifact as a “irreversible compound”,
overcoming the Vasarian dichotomy between text and image. This approach situ-
ates the work in the historical and cultural context of its production, and attempts
to eliminate the division between visible and legible, figure and text, which has
been present for so long in art-historical studies. In this way, the history of art
can widen its investigative horizons and claim a central role in the examination
of visual, and also verbal forms, establishing a point of departure for future re-
search, not just in the humanities. Indeed, the discipline of art history is currently
advancing towards themes of wide relevance from the perspective of the history of
images and perception. These approaches extend the history of art from the field
of anthropology and socio-cultural study, to the study of cognition and neurosci-
ence, exploring issues of visibility, including interior, invisible visions.
S. Clemente
Examination of the compositional methods of visual discourse, extended to
include the style of inscriptions and images, reveals the narrative and icono-
graphical function of the apse mosaic of S. Clemente in Rome, 33 commissioned
by Pascal II (1099-1118) and finished before 1119. 34 The mosaic includes in-
scriptions which are all integral to the work and closely related to the images
(FIG. 1). 35 The scripts guide the viewer through the type of letters, the placing
of texts and images, and the colors, depending on the position of the viewer in
the architectural space of the church. From a seventeenth-century drawing by
Ciampini, we can see that the nave was originally divided into two main parts:
one part that included the area of the apse and the extended chancel known as
the schola cantorum, which ended at the two pilasters of the nave, and the area
open to the laity, beyond those pilasters (FIG. 2).
The focal point of the mosaic is the Crucifix in the apse conch. The Crucifix is
aligned on an axis with the bust of Christ, who is shown with one hand raised in
blessing within a roundel at the center of the surrounding arch. The inscription,
which is placed around the lower edge of the arch, frames the apse and joins the

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 91
36 Translation: “Glory to God in the highest seated
on His throne and on earth peace to men of
good will”.
37 Is. 6, 1: “In anno, quo mortuus est rex Ozias,
vidi Dominum sedentem super solium excelsum
et elevatum”.
two parts of the decoration. The text is an antiphon from the Christmas Mass, and
reads: gLORiA iN EXCELSiS DEO SEDENTi SUP(ER) THRONUM ET iN TERRA
PAX HOMiNiBUS BONE VOLUNTATiS. 36 Displayed in an elegant and balanced
manner, in a semi-circular arrangement within a golden frame on a blue back-
ground, the inscription is interrupted at the top of the arch by the image of Christ.
To the left and right of the arch, the prophets (FIG. 3) Isaiah and (FIG. 4)
Jeremiah, dressed as ancient Romans, are identified by inscriptions: ISAIAS
and HIEREMIAS, and by quotations on their scrolls written in black capitals
on a white background. Isaiah’s reads: VIDI DOMINUM / SEDENTEM SU-
PER SOLIUM, and Jeremiah’s: HIC EST D(EU)S N(OSTE)R ET N(ON) ESTI/
MABIT(UR) ALIUS ABSQ(UE) ILLO. Isaiah’s scroll contains an explicit quota-
tion from the beginning of Isaiah 6, after the Song of the Vine. 37 The text was
Fig. 1 – S. Clemente, Rome. Mosaic in the apse and on the surrounding wall. Photo: S. Clemente.

92 STEFANO RICCIONI
Fig. 2 – S. Clemente, Rome. The interior. Drawing by G. Ciampini
(Vetera Monimenta, I, tav. VIII).

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 93
38 Sinding-Larsen 1978; Sinding-Larsen 1984, 47-50.
39 On the symbolic meaning of papal throne dur-
ing the Gregorian reform, see Gandolfo 1981.
For a broad discussion of this argument, see
Riccioni 2006, 23-34, 80-81.
read during the first part of the Christmas Mass introducing the canon 38 and is
closely connected with the inscription of the Gloria, where the phrase sedenti su-
per thronum is interpolated. Moreover, the reference to a throne (either as thro-
num or solium) is connected with the ideological relationship between Ecclesia
and the papacy. This connection, so important to the Investiture Controversy,
is made explicit by the papal throne at the base of the apse and symbolizes the
authority of the Church over the Empire. 39
Fig. 3 – S. Clemente, Rome.
Mosaic on the apse arch. The
prophet Isaias. Photo: S. Ric-
cioni.

94 STEFANO RICCIONI
40 Bar. 3, 36: “Hic Deus noster non aestimabitur
alius adversus eum hic adinvenit omnem viam
disciplinae et tradidit illam Iacob puero suo et
Isrhael dilecto suo post haec in terris visus est
et cum hominibus conversatus est”.
41 Quodvultdeus Carthaginiensis ep., Contra Iu-
daeos, 11.7: “Hic est [...] Deus noster, et non
aestimabitur alius absque illo, qui invenit om-
nem viam scientiae, et dedit eam Iacob puero
suo et Isrhael dilecto sibi. Post haec in terris
visus est, et cum hominibus conversatus est”.
Thomas 1950; Favreau 1997, 227.
Jeremiah’s text is a reference to Baruch 3, 40 excerpted from the sermon Con-
tra Iudaeos, Paganos et Arrianos, 41 a polemical text against heretics and the Jews
that was written in the fifth century by Quodvultdeus, but ascribed to St. Augus-
tine during the Middle Ages. It was eventually developed into a “prophet play”
used during the Christmas Mass, and also employed in several epigraphic con-
Fig. 4 – S. Clemente, Rome. Mo-
saic on the apse arch. The prophet
Jeremiah. Photo: S.Clemente.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 95
42 Sepet 1867; Sepet 1868; Sepet 1877; see also:
Sepet 1878; Young 1921; Thomas 1950. On li-
turgical drama and its influence on art, see:
Forsyth 1972, 49-60; Favreau 1997, 225-227.
43 Glass 1991, 203-221, esp. 213-219.
44 Newton 1976, 42-43; Newton 1999, 315-317.
45 See manuscripts: Montecassino, Biblioteca
dell’Abbazia, Casin. 99; 100; 103; 106; 107; 115
and 462. Iguanez 1915, 101-103, 103-109, 119-
123, 130-137, 137-141, 180-183, III (1941), 96-
100; Avagliano 1970, 304, no. 6, 305, no. 7, 10,
13, 14, 306, no. 22, 315, no. 91; Loew & Brown
1980, II, 65-67, 69, 74, 87-88.
46 Glass 1991, 215.
47 Riccioni 2006, 18-20.
48 On the Salerno mosaic, see: Kitzinger 1972b;
for a new interpretation of the mosaic and its
chronology, also connected with the Gregorian
reform, see: Iacobini 2005.
49 See the description of Del Pezzo, Contezza
dell’origine, aggrandimento e stato dei seggi della
città di Salerno, original manuscript in the Li-
brary of the Badia of Cava dei Tirreni, Arca XIII,
ms. 142; Braca 2003, 115. According to Braca
the description of Del Pezzo demonstrates that
the inscription was not placed at the base of the
apse conch. The oldest transcription of these
verses was made by Marsilio Colonna 1580, 77;
Mazza 1681, 42. The edition used here is that
published in Acocella 1966, 27: “Da Mattahaee
Pater Patris hoc det et innuba mater / ut pater
Alphanus maneat sine fine beatus / Ecce Dei na-
tum sine matre Deum generatum / praedicunt
vates nasci de virgine matre / sic Christus natus
nostros removendo reatus / vivit cum patre in
coelo et cum virgine matre”. See also: Kitzinger
1972b, 152, note 11. The verses were replaced
on the mosaic during a restoration campaign in
1950. On the prophets in Salerno mosaic, see:
Pace 1997, 195-196.
texts. 42 The quotation was often ascribed to Jeremiah because of his associa-
tion with the doctrine of divine incarnation; and it was often used in polemics
against the Jews. 43 Moreover, because the sermon Contra Iudaeos, Paganos et Ar-
rianos was ascribed to St Augustine, the text was well known at Montecassino. 44
Manuscripts found in the monastery’s library in the eleventh century, entitled
Sermones et homiliae diversorum Patrum, contain portions of it, 45 and annota-
tions like In vig(ilia) natalis Domini and In nat(ivitate) Domini confirm that the
text was read at Christmas. 46 The iconographical origins of the prophets on the
apse arch are probably to be found in the dialogue between the Roman reform-
ers and the monks at Montecassino. 47 The same iconography of Isaiah and Jer-
emiah may have been used in the almost lost mosaic on the apse arch of Salerno
Cathedral 48, based on what on what is known of the inscription along the frontal
frame of the apse conch. 49 In any case, the iconography of the prophets with
written scrolls extending the length of their bodies, which later became quite
common, may be regarded as an “epiconographical” novelty created during the
Gregorian reform to signify speaking images teaching the congregation of the
faithful the correct interpretation of the artistic program.
On each side of the arch are two saints: St Peter and St Clement on the right;
St Paul and St Lawrence on the left. Each pair has an inscription under the
saints’ feet. The text under St Lawrence and St Paul is a kind of commentary,
external to the images: DE CRUCE LAURENTI PAULO / FAMULARE DOCENTI
(FIG. 5). It is displayed on two lines, and written in white letters on a black
background. In contrast, the inscription under St Peter and St Clement is inter-
nal to the visual composition and makes visible the voice of St Peter speaking
to St Clement: RESPICE P(RO)MISSUM / CLEMENS A ME TIBI CH(RIST)UM

96 STEFANO RICCIONI
50 Bonizo Sutrinus, De sacramentis, PL 150, 860B:
“Post beati Petri apostolorum principis in-
clytum martyrium quod uno eodemque die cum
beato Paulo doctore gentium sub Nerone Cae-
sare suscipiens, gloriosam Romanam fecit Ec-
clesiam, Clemens natione Romanus Romanum
suscepit pontificatum. Qualiter vero primus sit
per electionem Petri, et tertius in gradu […]”.
On Bonizone of Sutri and the influence on the
Church reform, see: Berschin 1992 [1972]. Dur-
ing the eleventh and twelfth centuries reformers
thought that St. Clement had succeeded St. Pe-
ter, but he was actually the third pope after St.
Peter, see: Scorza Barcellona 2000.
(FIG. 6). This inscription is written in two colors: the first line is in white char-
acters on a black background, while the second line has yellow characters on
a red background. It functions in the same manner as rubrics in manuscripts,
‘highlighting’ the beginning of the text, and has a rhetorical value as a guide for
readers. The inscription was specifically emphasized to exalt the pope as succes-
sor of Peter, thus demonstrating the primacy of the Church of Rome. 50
Fig. 5 – S. Clemente, Rome.
Mosaic on the apse arch. St
Paul and St Lawrence. Pho-
to: S. Clemente.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 97
51 Other distinctive elements are the A with the
broken central bar and horizontal line at the
top of the two converging strokes; and the X
made up of two rounded Cs back to back.
If we focus our attention on the style of the inscriptions, especially the Glo-
ria and those on the scrolls of the prophets, we find further elements for anal-
ysis. The script of the Gloria is in golden capitals of monumental type and
displays some Greek graphic elements, which have not previously attracted at-
tention from scholars: for example, each letter O in the Gloria underneath the
prophet Isaiah and (FIG. 7) in hominibus displays the characteristic “gem” or
ornamental mark found in Greek inscriptions, and each M includes a cross-
stroke where the two oblique strokes meet. 51 Such graphic ornamentations,
Fig. 6 – S. Clemente. Rome. Mosaic on the apse arch. St Peter and St Clement. Photo S. Clemente.

98 STEFANO RICCIONI
52 See Supino Martini 1987, 156-159, pl. XXX,
170-173, pl. XXXV. From a palaeographic
point of view, the rare examples of characters
displaying Greek influence in Rome and the
areas under Roman influence are the result of
a familiarity with the Beneventan script. For
an overview of epigraphic script in Rome, see
also: Supino Martini 2001a; Supino Martini
2001b.
which are also present in the caption for Jeremiah, were not native to Ro-
man book production or monumental epigraphy, whether painted or inscribed
on stone. 52 Rather they come from Byzantine southern Italy, where the Latin
alphabet of display scripts had been modified under the influence of elegant
Greek graphic models. Between the end of the eleventh century and the begin-
ning of the twelfth, this precise geo-cultural area saw the elaboration of a Latin
minuscule rich in Byzantine elements. It was used widely in epigraphy, espe-
cially in the southern Italian province of Apulia and the city of Salerno, ruled
by the Normans, in the Benedictine communities of Campania, in Bari, Sicily
Fig. 7 – S. Clemente, Rome. Mosaic on the apse arch, letters O and M in hominibus. Photo: S.Riccioni.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 99
53 Significant “Greek” elements are found in the
script of liturgical manuscripts and Exultet rolls
from the end of the tenth century and the begin-
ning of the eleventh (the Pontifical of the Bib-
lioteca Casanatense, from Benevento, and the
Exultet 1 from Bari), as well as in the majuscule
and capitals of manuscripts written in Beneven-
tan script at the Abbey of Monte Cassino, dating
from the second half of the eleventh century and
the beginning of the twelfth. The same elements
are found in monumental works executed during
the abbacy of Desiderius (1058-87), such as the
inscriptions incised on the bronze doors brought
to Monte Cassino from Byzantium or the inscrip-
tions in the church of S. Angelo in Formis, where
the painted inscription in the apse below the
Pantokrator displays the characteristic M and O.
54 On bronze doors, see: Seloni (ed.) 1990; Ange-
lucci 2003 [2008]; De Spirito 2003 [2008]; Iaco-
bini (ed.) 2009.
and the abbey of Montecassino (FIG. 8). 53 In Rome, this graphic style was rare.
Not coincidentally, however, it can be found on the bronze doors of S. Paolo
fuori le mura, which were made in 1070 in Byzantium and brought to Rome
via Amalfi, when Hildebrand, the future Pope Gregory VII, was a monk at the
basilica. 54 The presence of this graphic style in the mosaic of S. Clemente dem-
onstrates particular contacts with authors or patrons connected to cultural en-
vironments influenced by Byzantium. Considering the close relationship be-
Fig. 8 – S. Angelo in formis. Inscription in the apse painting. Photo: S.Riccioni.

100 STEFANO RICCIONI
55 Meyvaert & Devos 1955; Meyvaert & Devos
1956. See also: Riccioni 2006, 18-20.
56 In classical and medieval rhetoric, schemata
and diagrams were used to organize concepts.
For mnemonic purposes, schemata serve as
fixed points for memory storage and as a cue
to start the recollective process, see: Carruthers
2008 [1990], 332. The schema of a composi-
tion guided the ductus, its logical and narrative
‘flow’, see: Carruthers 1998, 77-81.
57 The Rhetorica described three levels of rhetori-
cal style, see: Cicero M. Tullius ps., Ad C. Her-
ennium, 4.13.55: “Sunt igitur tria genera, quae
genera nos figuras appellamus, in quibus om-
nis oratio non vitiosa consumitur: unam gra-
vem, alteram mediocrem, tertiam extenuatam
vocamus. Gravis est quae constat ex verborum
gravium levi et ornata constructione. Mediocris
est quae constat ex humiliore neque tamen ex
infima et pervulgatissima verborum dignitate.
Adtenuata est quae demissa est usque ad usita-
tissimam puri consuetudinem sermonis”.
58 On the new church of S. Clemente and its can-
onry build for reformed canons, see: Barclay
Lloyd 1989, 20-35; Riccioni 2006, 7-10.
59 Translation: “We make the Church of Christ
similar to this vine that the law makes arid but
the cross makes luxuriant”.
60 Translation: “The wood of the cross and the
teeth of Jacob and Ignatius are placed in body
of Christ written above”. Note the term supra-
scripti, used in the sense of ‘displayed’, ‘painted’
or ‘made in mosaic’ but literally meaning ‘writ-
ten above’.
tween the church of S. Clemente and the abbey of Montecassino – for example,
the ex-librarian of the abbey, Leo of Ostia 55 was called to write Clement’s Pas-
sio – it may be taken to confirm that the workshop commissioned to execute
the mosaic included artisans from the cultural area of Benevento and Cassino
who were familiar with Greek culture and script.
When the images are analyzed, in terms of shape, style, or meaning and in
relation to the graphic and textual elements, and when the scripts are themselves
considered as images, a schema emerges in the mosaic – an organization of the
figural discourse – composed according to rhetorical principles borrowed from
the ars dictaminis. 56 For example, the rhetorical style, form and display of the in-
scription of the Gloria on the triumphal arch match the monumental dimensions
of the figures on the arch, all easily legible from afar, corresponding to the ornatus
gravis or solemn style in the rhetorical terminology of the Ad Herennium. 57 Mean-
while, the inscription at the base of the conch, in a smaller format, matches the
tiny figures almost hidden in the acanthus/vine scroll, and corresponds to the or-
natus adtenuatus or humble style. The strategy divided the viewers into those who
could gain entrance to the choir (that is to say the reformed canons 58) and could
thus see the small text and images in the tendrils from close up, from those who,
beyond the schola cantorum (in other words, the laity), could only make out the
inscriptions and images in the monumental format.
The inscription of the conch itself is, moreover, made up of two texts:
1) ECCLESIAM CHRISTI VITI SIMILABIMUS ISTI || QUAM LEX ARENTEM SET
CRUS FACIT E(SS)E VIRENTEM; 59
2) DE LIGNO CRUCIS IACOBI DENS IGNATIQ(UE) REQUIESCUNT IN SUPRA-
SCRIPTI CORPORE CHRISTI. 60
The first indicates that the pictured acanthus tendrils should be interpreted as
the vine, a symbol of the Church; the second mentions the relics – the wood of

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 101
61 On a more comprehensive interpretation of the
apse inscription see Riccioni 2006, pp. 65-75,
with discussion of earlier bibliography. The hy-
pothesis that the last section of the inscription
was changed during the papacy of Innocent
II (Stroll 1988; Stroll 1991, 118-131, esp. 126-
131) should be revised after the restoration by
Istituto Centrale del Restauro, which has dem-
onstrated that the mosaic was made during
one single campaign of works, see Anselmi &
D’Angelo; Basile 2002. Furthermore, the paleo-
graphical examination reveals that the inscrip-
tion was coherent with epigraphic script in me-
dieval Rome, see Riccioni 2006, 68, note 26.
the Cross and St James’s tooth – that were preserved in the church. 61 The second
text is embedded within the first, creating a visual correspondence between texts
and images. The reference to the relics is found at the center, on the same axis as
the Crucifixion, while the two parts of the leonine verse referring to the Church
extend to the ends of the apse conch, corresponding with the acanthus/vine ten-
drils above. This type of composition, unlike traditional processes of reading a
linguistic message which demand an uninterrupted progression of text, is cha-
racteristic of figural messages and corresponds to the iconography of the apse
conch. In the gaps between the tendrils, the large acanthus/vine scrolls contain
tiny figures of animals, especially birds on the upper level of the decoration,
amorini, and groups of people (FIG. 9). Moving from the details to the design
as a whole, it becomes clear that the acanthus tendrils, symmetrically arranged
Fig. 9 – S. Clemente, Rome. Mosaic in the apse conch. The owl. Photo S. Clemente.

102 STEFANO RICCIONI
62 In rhetorical terms, ornatus is closer in mean-
ing to furnishings or equipment than its Eng-
lish cognate ‘ornament’. On these concepts as
they relate to the mosaic of S. Clemente, even if
from a different perspective, see Bonne 1997.
63 Carruthers 2008 [1990]; Carruthers 1998.
in five roundels on five levels on each side, form a schema that functions in
rhetorical terms taken from manuscripts. It is, in effect, a diagram, inspired by
the trees of medieval mnemonic and preaching techniques, which contain fi-
gures functioning as exempla. In these terms the composition of style and color
follows the precepts of medieval rhetoric found in the study of ornatus, 62 and
the scripts themselves must be considered as images, in which iconicity and tex-
tuality are matched in an integrated, rhetorical machina. 63
As part of this rhetorical strategy, St Ambrose, with his red halo, is dis-
tinguished from the other Fathers of the Church, who have blue halos in the
mosaic (FIG. 10 and 11). In fact, the red halo functions in the same manner as
the inscription under St Peter and St Clement, “rubricating” the image of the
saint because it has a rhetorical value as a guide for viewers indicating that the
Fig. 10 – S. Clemente, Rome. Mosaic in the apse conch. St Ambrose. Photo S. Clemente.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 103
64 On this symbolic interpretation see Riccioni
2006, 41-64.
65 On the use of spolia with inscriptions in Re-
form churches of Rome, see Gandolfo 1974-
1975; Gandolfo, 1981; Kinney 1996; Riccioni
2005.
saint is the key to understanding the allegorical significance of the apse conch.
Reading Ambrose’s writings, in particular the Hexaemeron, we can find fur-
ther clues to the interpretation of the images-exempla contained in the tendril-
diagram. Above the image of St Ambrose are birds (the animals closest to the
sky and thus to God) and symbols of heaven; the world is at the same level and
under the saint, we can see a group of humans and fighting animals, which
together symbolize the people of the Church and their ascetic struggle against
temptation in order to reach God (FIG. 12). 64 Finally, not a part of the mosaic,
but symbolically connected with it, is the papal throne, which is located at
the base of the apse conch (FIG. 13). The back of the throne was made from
a marble spolium that displays the inscription: MARTYR. 6 5 This spolium was
selected from the debris of the early Christian church of S. Clemente because
Fig. 11 – S. Clemente, Rome. Mosaic in apse conch. St Augustine. Photo S. Clemente.

104 STEFANO RICCIONI
the inscription makes visible the connection between the martyrs St Peter (the
first pope), St Clement (thought to have been Peter’s successor), and the pope
sitting on the throne (originally, Pascal II). As we have seen, the highlighted in-
scription under their feet draws attention to this pair of saints. Thus, without
integrating the two means of communication, figural and textual, it is impossi-
ble fully to understand either the mosaic’s rules of composition or its message.
This reading of the S. Clemente mosaic reveals the complexity of the message
and the way in which the audience (both idiotae and literati) was directed by a
sophisticated combination of texts and images. The complexity of the visual nar-
rative functioned as an invitation to the observer to explore, elaborate, and cre-
ate individual meaning as a stimulus for contemplation. This particular strategy
seems to have been developed during the Gregorian reform and ultimately used
to celebrate the triumph of the Church after the Investiture Controversy.
Fig. 12 – S. Clemente, Rome.
Mosaic in the apse conch.
Heron fighting against a liz-
ard. Photo S. Clemente.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 105
66 This section is taken from my dissertation for
the Licence in Mediaeval Studies of the Pontifi-
cal Institute of Mediaeval Studies of Toronto,
see: Riccioni 2009 (under revision for publica-
tion). The bibliography on the mosaic is vast,
for a comprehensive bibliography see Parlato
& Romano 2001, 60-75; Romano 2006; Croiser
2006b; Riccioni 2010/2011.
S. Maria in Trastevere 66
In 1130, after the death of Honorius II (1124-1130), Gregory Papareschi was
elected pope, taking the name Innocent II (1130-1143). In the same year, how-
ever, Peter Pierleoni was elected to the papacy as Pope Anacletus II, creating a
schism. Forcing Innocent II to flee, Anacletus governed Rome until his death
in 1138, at which time Innocent II returned to Rome. The close connections
between the Pierleoni family and the Trastevere quarter, as well as the fact that
Fig. 13 – S. Clemente, Rome.
The papal throne in the chan-
cel. Photo: P. Zolli.

106 STEFANO RICCIONI
67 Krautheimer 1980, 212-214, 217-218, 226.
68 Di Carpegna Falconieri 2004, 410.
69 Mansi (ed.) 1901-1962, XXI, 535. All acts, decisions
and activities of the antipope were invalidated.
70 Kinney 1975b, 190-222, esp. 215-216. Dale Kin-
ney suggested that the church was finished by
the time Innocent II died: Kinney 1975a, 42-53.
See also: Krautheimer, Corbett & Frankl 1967,
65-71; Parlato & Romano 2001, 60-75, esp.
61-64, 73 (on the date of the mosaic); Croiser
2006b; Kinney 2006, 200-201.
Peter Pierleoni was cardinal, have led scholars to ascribe the first rebuilding
of S. Maria in Trastevere to him. 67 Since Innocent II, who came from the “de
Papa” or Papareschi family, the other powerful Roman family in Trastevere, 68
had the damnatio memoriae of Anacletus II ratified in the Second Lateran Coun-
cil when he had regained possession of the city, 69 we cannot exclude that the
evidence for the correct attribution was destroyed. The mosaic of S. Maria in
Trastevere, however, was probably done under Innocent II, after the end of the
schism, closely related to the historical events and probably finished in 1143. 70
Fig. 14 – S. Maria in Trastevere, Rome. Mosaic in the apse and on the surrounding wall. Photo:
S.Riccioni.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 107
71 LP, II, 384: “ecclesiam beatae Dei genitricis
Mariae tituli Calixti totam innovavit et con-
struxit”. Benedictus canonicus also wrote in his
Liber politicus (1143) that Innocent II had re-
built the church and decorated it with mosaics,
see: Liber censuum, II, 169: “Innocentius papa
II, dominus meus [...] Ecclesiam sanctae Mar-
iae Trastiberim novis muris funditus restauravit
et absidem ejus aureis metallis decoravit”. See
also Fabre 1892, 13, n. 3. The Liber censuum
was probably written no later than 1143, be-
cause it is dedicated to Cardinal Guido de Cas-
tello, elected pope Celestine II in 1143.
72 Kinney 2002, 22-24.
73 On Mary as wife of Christ and queen, see: Wellen
1966; Kitzinger 1980; Verdier 1980, 40-47.
74 Nilgen 1981.
75 Apoc. 4, 6-8; Apoc. 4, 5.
76 On the conservation and restoration of the mo-
saic, see: Matthiae 1967, I, 421-422, II, Grafico
del restauro di S. Maria in Trastevere, s.n.; Lotti
1996; Tiberia 1996, 187-197, esp. 187-192, fig. 84.
Various twelfth-century sources credit him with the work on the church and its
decoration. 71
The apse mosaic is based on a symmetrical schema: each image corresponds
with another on the other side of the central axis. If the arch is seen to serve as
a frame, a rhetorical exordium or introduction to the main theme of the narra-
tive, the conch contains the core message of the discourse. The composition of
the mosaic is organized around a central vertical axis and symmetrically com-
posed horizontal levels. 72 The central axis is marked by a vertical succession of
motifs, all of which refer to Christ. At the apex of the arch is a gold cross with
pendant alpha and omega; in the soffit of the conch, a christogram combin-
ing the first two letters of Christos; just below the christogram, a tiny cross;
in the center of the conch, the figure of Christ wrapped in a gleaming gold
mantle; and, at the bottom, a cross-nimbed lamb. Just as at S. Clemente, the
papal throne is located at the base of the apse conch outside the mosaic and is
symbolically connected to the meaning of the mosaic (FIG. 15). Four different
levels of images intersect the vertical axis horizontally: the upper level of the
arch, the central level of the conch and the arch, a monumental inscription,
and a frieze of lambs. Although Christ is in the very center, the central image
of the apse is extended to his right to include the Virgin Mary. They are seated
on the same throne and Jesus embraces the Virgin as if she were his wife (FIG.
16). If we understand the pair as the central image of the conch, the main mes-
sage is the enthronement of and marriage between Mary / Ecclesia / sponsa
and Christ. 73 This, however, leaves the vertical axis as being of less importance.
Ursula Nilgen has even suggested that this composition was the result of a
change in the original iconography after the death of Anacletus II, 74 who prob-
ably initiated the re-building of the church. As we will see, this hypothesis can-
not be accepted.
At the top of the arch is a row of apocalyptic symbols: the four winged crea-
tures identified with the four evangelists, and the seven candlesticks (FIG. 17). 75
The names of the evangelists are given in plaques with polychrome frames – it
should be noticed that only the inscription below the eagle is entirely original
(FIG. 18). 76 The mise en page of the text is completely novel, even compared

108 STEFANO RICCIONI
Fig. 15 – S. Maria in Trastevere, Rome. The papal throne in the chancel. Photo: S.Riccioni.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 109
with the mosaic of S. Clemente. The use of frames to include the names of the
evangelists, and to isolate them, is previously unknown in Roman (and not only
Roman) mosaic decorations. To the left and right of the arch are the prophets
(FIG. 19) Isaiah and (FIG. 20) Jeremiah, dressed as ancient Romans. The figures
are identified by labels in colored hexagonal frames: ISAIAS P(RO)PH(ET)A and
HIEREMIAS P(RO)PH(ET)A, and by quotations on their scrolls written in white
capitals on a red background. Isaiah holds the words: ECCE VIRGO CONCI /
Fig. 16 – S. Maria in Trastevere, Rome. Mosaic in the apse conch. Christ and the Virgin. Photo:
S.Riccioni.

110 STEFANO RICCIONI
77 Is. 7, 14: “Virgo concipiet et pariet filium, et vo-
cabitur nomen eius Emmanuel, quod est inter-
pretatum ‘Nobiscum Deus’ ”.
78 Tertullianus, Adversus Marcionem, PL 2, 338A-
B; Tertullianus, Adversos Judaeos, PL 2, 616C-
617A; Zacchaeus Christianus, Consultationum
Zacchaei christiani et Apollonii philosophi libri
tres, II, 4. Quid Judaeis respondendum sit, PL
20, 1114B; Evagrius monachus, Altercatio inter
Theophilum christianum et Simonem judaeum,
PL 20, 1170D; Hieronymus Stridonensis, Ad
Pammachium, PL 22, 574; Hieronymus Strido-
nensis, Liber hebraicarum quaestionum in Gene-
sim, PL 22, 973D-974B; Agustinus Hipponensis,
Contra Faustum manichaeum, PL 42, 282; Pau-
linus Aquileiensis, Contra Felicem Urgellitanum,
PL 99, 371C.
79 Quodvultdeus Carthaginiensis ep, Contra Iuda-
eos, 11.6.
80 Bruno Signinus, In Matthaeum, PL 165, 76C-
76D, 169C-170A; Petrus Cluniacensis, Adversus
Judaeorum, PL 189, 585D-587C; Petrus Clunia-
censis, Adversus sectam Saracenorum, PL 189,
703A-703D.
81 Gislebertus Crispinus, Disputatio Iudei et Chris-
tiani, 44, 45, 55, 59.
82 The hypothesis that both S. Clemente and S.
Maria in Trastevere contain anti-Jewish mes-
sages is emphasized by Mary Stroll, see Stroll
1991.
83 Thomas 1950; Quodvultdeus Carthaginiensis
ep, Contra Iudaeos, 11.7; Favreau 1997, 227.
PIET ET PARIET FILIUM, and Jeremiah: CHR(ISTU)S D(OMI)N(U)S CAPTUS
E(ST) / IN PECCATIS N(OST)RIS.
The iconography of the prophets on the arch comes from the apse mosaic of
S. Clemente, with some important exceptions. The scroll of Isaiah contains an
explicit reference to the conception of Christ, taken directly from the Vulgate. 77
Since the early Christian period, this formula had also been used in polemical
texts against heretics and the Jews, 78 such as the sermon Contra Iudaeos, Paga-
nos et Arrianos discussed above. 79 This quotation became frequent between the
end of the eleventh century and the beginning of the twelfth century 80 as can be
seen in the Disputatio Iudei et Christiani of Gislebertus Crispinus, which uses
it four times. 81 Before this text can be assigned a polemical value, 82 as com-
ing from the “prophet play”, however, the quotation on Jeremiah’s scroll must
be taken into account. Normally, after Isaiah’s quotation-dialogue Jeremiah res-
ponds: “Hic est Deus noster, et non aestimabitur alius absque illo (or adversus
eum)”. 83 The quotation in Jeremiah’s scroll does not correspond with the answer
given by the prophet in the drama.
Fig. 17 – S. Maria in Trastevere, Rome. Mosaic on the apse arch. Alpha/Omega and the four Beasts
of the Apocalypse. Adapted by S. Riccioni from A. Giorgetti in Andaloro & Romano 2006, 305, no. 1.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 111
84 Hesbert & Prevost (eds.) 1963, I, 4-5, n. 1. The
text of the sermon also is used in the Office for
the Wednesday and Saturday of the third week
of Advent (Hesbert & Prevost 1968, III, 191,
no. 2557).
85 Hesbert & Prevost (eds.) 1968, III, 191, no.
2558.
86 Tertullianus, De carne Christi, PL 2, 781C; No-
vatianus, De Trinitate, PL 3, 901A-905C; Ambro-
sius Mediolanensis, In psalmum David CXVIII,
PL 15, 1309B; Hieronymus Stridonensis, In
Isaiam, PL 24, 107B, 144B; Agustinus Hippon-
ensis, De consensu evangelistarum, XXVI. Idolo-
latria per Christi nomen et Christianorum fidem
juxta prophetias eversa, PL 34, 1061; Gregorius
Magnus, In Ezechielem, hom. I, PL 76, 786B,
792 C.
87 Bruno Signinus, In Lucam, PL 165, 342C-343B;
Bruno Signinus, De laudibus Beatissimae Virgi-
nis Mariae, PL 165, 1022B.
88 Honorius Augustodunensis, In Cantica cantico-
rum, PL 172, 351D; Rupertus abbas Tuitiensis,
In Cantica canticorum, PL 168, 871A-909A.
89 Bernardus Claraevallensis, In adventu Domini,
Sermo II, PL 183, 41A-41C.
90 Bernardus Claraevallensis, In Assumptione B.V.
Mariae, PL 183, 433B-433D.
91 Lam. 4, 20: “Spiritus oris nostri Christus Domi-
nus captus est in peccatis nostris”.
92 Augustinus Hipponensis, De Civitate Dei,
XXXIII. De Christo et vocatione Gentium quae
Jeremias et Sophonias prophetico Spiritu sint
praefati, PL 41, 591-592; Hieronymus Strido-
nensis, Adversus Pelagianos, PL 23, 560C; Hie-
ronymus Stridonensis, In Ezechielem, PL 23,
105C-105D; Hieronymus Stridonensis, In Za-
chariam, PL 25, 1515C-1515D; Isidorus Hispa-
lensis, De fide catholica, XXIII. Comprehensus
est, PL 83, 479B; Amulo Lugdunensis, Contra
Judaeos, PL 116, 179B; Rabanus Maurus, De
laudibus Sanctae Crucis, PL 107, 253C; Rabanus
Maurus, In Jeremiam, PL 111, 1259B-1261A;
Paschalis Radbertus, In Lamentationes Jere-
miae, PL 120, 1229D, 1230C, 1231A, 1231B.
The text in the scroll of Isaiah was used as a liturgical formula in the an-
tiphon for the first Sunday of Advent 84 and, with a variation, in the office of
the Annunciation. 85 The text was also frequently used in exegetical comments to
provide typological support for the incarnation of Christ and the mystery of the
Trinity. 86 Bruno of Segni quoted the text in this sense in his De laudibus beatis-
simae Virginis Mariae; 87 Honorius Augustodunensis and Rupert of Deutz both
used Isaiah in their commentaries on the Song of Songs. 88 Finally, St Bernard
of Clairvaux quoted the text in two situations: as a sign of the Advent of Christ 89
and in his work on the Assumption of the Virgin. 90
The quotation on Jeremiah’s scroll is from Lamentations (Threni) 91 and was of-
ten interpreted as a reference to the Passion of Christ who takes upon himself the
sins of all, to expiate them through his sacrifice. 92 During the eleventh and twelfth

112 STEFANO RICCIONI
93 Petrus Damianus, Contra Judaeos, PL 145,
56C: “Et Hieremias: ‘Spiritus, inquit, oris nos-
tri Christus Dominus captus est in peccatis
nostris: cui diximus: In umbra tua vivemus
in gentibus (Lam. 4)’. Et per beatum Job ipse
Dominus in passione positus conqueritur [...]”.
Petrus Damianus, De Quadrigesima et Quadri-
ginta duabus mansionibus Hebraeorum, VII,
Quare tentatio virtuti admisceatur, PL 145,
557B; Petrus Damianus, In Vetus Testamentum,
XIX. In epistola ad Hildebrandum, PL 145,
1061C-1061D.
94 Rupertus abbas Tuitiensis, De Trinitate et operi-
bus eius, PL 167, 1418D-1419A; Rupertus abbas
Tuitiensis, In Apocalypsim, PL 169, 845A.
95 Gerhohus Reicherspergensis, In psalmos et can-
tica ferialia, PL 183, 1277C-1278B.
96 Hieronymus Stridonensis, Adversus Pelagianos,
PL 23, 560C; Isidorus Hispalensis, De fide ca-
tholica, XXIII. Comprehensus est, PL 83, 479B;
Amulo Lugdunensis, Contra Judaeos, PL 116,
179B; Petrus Damianus, Contra Judaeos, PL
145, 56B-56D.
97 Riccioni 2006, 68-71.
centuries, we find confirmation of this interpretation in the writings of Peter Da-
mian, 93 Rupert of Deutz 94 and Gerhoh of Reichersberg (a student of Rupert of
Deutz). 95 Even if Peter Damian quoted the text in a treatise against the Jews and
there are three more quotations in his polemical works 96 we cannot assume that
the triumphal arch bears a specific polemical message against the Jews. In a gen-
eral sense, both the Isaiah and Jeremiah texts were used to support the Church in
its struggle against heresy and incorrect interpretations of the Bible. As in the mo-
saic of S. Clemente, 97 the texts and the images of the prophets recall the authority
of correct interpretation but are adapted to convey different messages.
In monumental art, the scrolls held by saints or prophets usually employ black
letters on white backgrounds. At S. Maria in Trastevere, in contrast, the inscriptions
on the scrolls held by the prophets are written in white script on a red background.
As we have seen in S. Clemente, this choice recalls the rubrics in manuscripts and
has a rhetorical function, because the texts on the scroll are emphasized to intro-
duce and guide the right interpretation of the new iconography of Christ (as both
Fig. 18 – Rome, S. Maria in
Trastevere. Mosaic on the apse
arch. Framing of the inscrip-
tion under the symbol of Luke.
Photo: S.Riccioni.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 113
Fig. 19 – S. Maria in Trastevere, Rome. Mosaic on the apse arch. The prophet Isaias. Photo:
S.Riccioni.

114 STEFANO RICCIONI
Fig. 20 – S. Maria in Trastevere, Rome. Mosaic on the apse arch. The prophet Jeremiah. Photo:
S.Riccioni.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 115
98 Ambrosius Mediolanensis, De mysteriis, 7.34-37.
99 Ladner 1984, 301.
100 Honorius Augustodunensis, In Cantica cantico-
rum, PL 172, 415B: “Hujus genae sunt verecun-
dia, qua erubescit peccare, vel peccasse. Quae
sunt ut fragmen mali punici, quia exterius ru-
bent charitate, interius albescunt castitate [...]”.
101 Libellus de cerimoniis aulae imperatoris (Graphia
aureae Urbis Romae): Schramm (ed.) 1969, 344
and CT 3, 101: “Imperator ferat camisum, ex
subtilissimo et candidissimo bisso contextum,
cum aurea bulla, ornatum a pedibus ad mensu-
ram brachii in circuitu de auro frigio”. Schramm
dated the Graphia aureae Urbis Romae to 1030,
but the Graphia must have been written after the
death of Anastasius IV (13th December 1154), be-
cause his sarcophagus is mentioned in chapter
16. See Schramm 1969, 353-359; Bloch 1984; Pa-
ravicini Bagliani 1994, 119-20, 68, note 100.
102 Paravicini Bagliani 1994, 138, note 62.
103 Rupertus abbas Tuitiensis, In Cantica cantico-
rum, PL 168, 920: “Candidus sanctitate, rubi-
cundus passione”. This is probably to symbolize
the white lily of the confessor and the red rose
of the martyr. I am grateful to David Defries for
this suggestion.
104 Adamnanus Hiensis, De locis sanctis, 1.3, 232,
lines 16-19; Beda venerabilis, De locis sanctis, 2,
305, lines 2-4; Hermann 1969.
105 Deér 1959, 146-154. On the color of porphyry
and its meaning in ecclesiastical history, see:
Steigerwald 1999; Longo (ed.) 1998; Filoramo
1998, 233-241.
106 As suggested by Kessler & Zacharias, it sur-
vives a fragment in the Lateran cloister, see
Kessler & Zacharias 2000, 27-28, fig. 23.
But, in the opinion of Herklotz, all the frag-
ment are lost, see Herklotz 2001 [1985],
147-148, note 67 and note 70, 151-153.
107 It seems plausible that Innocent II wanted to use
the imperial signs quoted by the Dictatus papae
as specific to the bishop of Rome. The reuse of
the sarcophagus appears to have made a great
impact on contemporaries; it was mentioned in
the Mirabilia Urbis Romae, CT 3, 46-47; in the
Graphia Aurea Urbis, CT 3, 86; by Iohannes dia-
conus, Liber de Ecclesia Lateranensi, CT 3, 348;
by Petrus Mallius, Basilicae veteris Vaticanae
filius and sponsus) and Mary (not only as mater but also Ecclesia and sponsa) on
the same throne, displayed in the apse conch. The use of both red and white could
also have been inspired by political considerations related to the papacy since a va-
riety of writings suggest that these colors had both ecclesiastical and imperial con-
notations. In his De mysteriis, for example, St Ambrose had described the Church as
dressed in white in the pure garments of innocence. 98 The white and shining tiara
in the Donation of Constantine is clearly a symbol of the resurrection of Christ. 99 In
Honorius Augustodunensis’s commentary on the Song of Songs, white and red are
related to the Church. 100 Describing the garments of the emperors, the Libellus au-
lae imperatoris (written sometime before 1154) lists, after the chlamys, a shirt woven
from “the finest and whitest” linen. 101 The origins of this garment lay in Byzantium,
where the emperor dressed in a tunic of white silk under the purple chlamys. 102
Moreover, white and red together were the colors of Christ. In his commentary on
the Song of Songs, Rupert of Deutz described Christ as “pure (snow-white) with
holiness and red with the Passion” 103 and exegetical tradition reported that the Holy
Sepulcher was decorated in white and red, symbolizing divinity and martyrdom. 104
Innocent II was clearly conscious of the significance of these colors, being
the first Pope to use a sarcophagus made of porphyry, the imperial stone par
excellence. 105 The sarcophagus, which was transported to the Lateran from Cas-
tel S. Angelo, had the added attraction of being the same as the one used by
the Emperor Hadrian. Unfortunately, only a fragment is still to be found in the
Lateran cloister, 106 but it is known from numerous testimonies. 107 Although the

116 STEFANO RICCIONI
descriptio, CT 3, 431; and in a biography of In-
nocent II written by bishop Boso. Subsequently
Anastasius IV transferred the sarcophagus of
Helen, also made of porphyry, to the Lateran.
108 Innocentius III, Mysteria evangelicae legis et sa-
cramenti eucharistiae, De sacro altari mysterio
libri sex, PL 217, 801. During the two important
feasts in honour of St. Peter and St. Paul (29th
June) and of the Chair of St. Peter (22th Febru-
ary), the Pope had to wear liturgical garments
in these two colors: red (during the feast of 29th
June) and white (during the feast of 22th Febru-
ary). See: Paravicini Bagliani 1994, 47-48.
109 Galbreath 1930, 1-5; Erdmann 1933-1934, 46.
That these colors were explicitly used by the
Church is mentioned only from the thirteenth
century, but their symbolic meaning was pres-
ent earlier in the ecclesiastical texts.
110 Paravicini Bagliani 1994, 120.
111 The French origins of the iconography sug-
gested by Emile Mâle and Guglielmo Matthiae
were convincingly refuted by Verdier and are
no longer accepted. See Mâle 1953, 182-185,
esp. 184; Matthiae 1967, 305-314; Verdier 1976;
Verdier 1980, 40-47.
112 On Virgin as Ecclesia during Gregorian reform,
see: Russo 1996, 232-249; on the marriage be-
tween Mary/Ecclesia and Christ, but in thirteenth
century, see: Aronberg Lavin 2009, 160-164.
113 On Mary as Queen, see: Nilgen 1981; Osborne
2009; Themelly 2009, 118-134.
114 Mâle 1942, 200-209, esp. 201; Kitzinger 1980,
8-9. For the composition of the Assumption/
synthronos in the conch see also Tronzo 1989.
first evidence of the association of red and white as symbols of the pope is found
in a passage of the Cerimoniale of Gregory X (1272-1273) concerning the im-
mantatio and coronation, there was a long tradition of having these two colors
signify the relation of the Church to the papacy and papal power. The Mysteria
evangelicae legis et sacramenti eucharistiae of Lotario dei Conti di Segni, the fu-
ture Innocent III, demonstrates a clear awareness of this symbolic value: “Licet
autem in apostolorum Petri et Pauli martyrio rubeis sit utendum, in conversione
tamen et cathedra utendum est albis” . 108 Red and white thus became the colors
of the banner of the Roman Church, 109 and a visible sign of the imitatio imperii
and of the plenitudo potestatis of the Roman pope as vicarius Christi. 110 Hence,
the scrolls held by the prophets can be seen to display the colors of Christ and
of the banner of the Church and Papacy, functioning as a kind of monumental
“coat of arms”. The visibility of papal power was a fundamental concern during
the papacy of Innocent II, who continued the policies of Gregory VII, exalting
the temporal power of the pope and assuming the symbols of empire.
The central image of the apse is Christ with the Virgin Mary (FIG. 21). They
are seated on the same throne and Jesus embraces the Virgin. To either side are
a host of saints, martyrs and popes, all identified by colored inscriptions under
their feet: on the left, Innocent II, St Lawrence, and Pope Calixtus (FIG. 22); on
the right, St Peter, Pope Julius I, Pope Cornelius, and St Calepodius (FIG. 23).
This arrangement conforms to the rule of symmetry based on the rhetorical or-
ganization of visual speech.
The first problem involved in analyzing the group of the Virgin and Christ
is to determine the origin of the iconography. 111 It seems to be the first known
example of the union of two specific themes: Christ and Mary as spouses 112 on
the same throne, and the Virgin crowned as Ecclesia. 113 According to Emile Male
and Ernst Kitzinger, the apse mosaic in S. Maria in Trastevere is a monumen-
tal representation of the Feast of the Assumption. 114 This theme is clear from

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 117
115 Translation: “Come my chosen one, and I will
place you on my throne”.
116 Cant. 4, 8.
117 Compiègne Antiphonal, Paris, Bibliothèque Na-
tionale, ms lat. 17436.
118 Hesbert & Prevost 1963, I, 125 (Feast of the
Assumption), 368, 369, 371 (Communion of
virgins).
the inscriptions displayed on the book held by Christ and the scroll unrolled by
Mary. Christ’s book reads: VENI / ELEC/TA MEA / ET PO//NAM IN / TE THRO/
NUM / MEUM. 115 The text is a paraphrase of a verse from the Song of Songs:
“Veni de Libano, sponsa mea, veni de Libano, veni, coronaberis”, 116 which was
used in the responsory of the liturgy for the fiftheenth of August, as given, for
example, in the ninth-century Compiègne Antiphonal. 117 The text is also found
in the manuscripts of the cursus romanum used during the liturgy for the feast
of the Virgin’s Assumption and for the communion of virgins: “Veni electa mea,
et ponam in te thronum meum, quia concupivit Rex speciem tuam”, 118 in a pas-
sage which also draws on Psalm 44 (Vulgate 45), verse 12. The Virgin’s scroll
reads: LEVA / EIUS / SUB CA/PITE ME/O ET DEX/[T]ERA IL/LIUS AM/PLESA/
Fig. 21 – S. Maria in Trastevere, Rome. Mosaic on the apse conch. Photo: Stefano Riccioni.

118 STEFANO RICCIONI
119 Translation: “His left hand is under my head,
and his right hand will embrace me”. Regarding
this inscription, De Rossi noted that the restorer
changed the term AMPLESABiT(UR) with the abbre-
viation meaning ur to AMPLESABiT, without the
horizontal stroke over the letter T, making it lose
the abbreviation, so it currently reads AMPLESABiT,
see De Rossi 1899, f. 140v. Recent restoration
did not remarked this change, see Tiberia 1999,
190, fig. 83. Mary Stroll argued that the substitu-
tion of S for X in amplesabitur could have been a
“foreign usage” related to the French influences
imported by Innocent II, see Stroll 1991, 130.
It seems more likely that the version amplesabi-
tur had been caused by a linguistic idiotism of
Italian vernacular, as the word crus for crux in
the apse inscription of S. Clemente, see also De
Rossi 1899, f. 140v; Riccioni 2006, 68, note 26.
120 Cecchelli 1933, 93; Hesbert & Prevost 1963, I,
286.
121 Astell 1990; North, 1998; Debergé & Nieuviarts
2002.
BIT(UR) ME. 119 These words are a literal quote from Cant. 2, 6, 10-11; 8, 3, and
were sung during the feast of the Assumption. 120
The exegetical tradition regarding the Song of Songs is extensive, 121 and dur-
ing the Gregorian reform it was added to by a number of commentaries which
Fig. 22 – S. Maria in Trastevere, Rome. Mosaic in the apse conch, left side. From A. Giorgetti in
Andaloro & Romano 2006, 308, no. 4.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 119
122 Ohly 1958; Wirth 1999, 426-430.
123 Hieronymus ps., De assumptione beatae Mariae
Virginis, PL 30, 122C-145B, esp. 138B: “Nunc
autem circumndant eam [Virgin Mary] flores
rosarum, indesinenter eius admirantes pulchri-
tudinem inter filias Jerusalem in qua posuit rex
thronum suum, quia concupivit eius speciem ac
decorem”. Juge 1944.
124 Bruno Signinus, In Cantica canticorum, PL 164,
1233-1288.
accentuated an ecclesiological dimension. 122 Most of these were based on the
interpretation of pseudo-Jerome, used since the Carolingian period in texts read
for the Feast of the Assumption. 123 During the eleventh century, Bruno of Segni
dedicated a long treatise to the Song of Songs, where an explicit association be-
tween the sponsa and the Virgin/Ecclesia is introduced for the first time 124 In the
first book of his Sententiae, the description of the Church as bride of Christ, also
connected to the bride of the Song of Songs, comes partly from the Apocalypse
of John and partly from Psalm 44; and she sits on the throne near Christ as his
Fig. 23 – S. Maria in Trastevere, Rome. Mosaic in the apse conch, right side. From A. Giorgetti in
Andaloro & Romano 2006, 309, no. 5.

120 STEFANO RICCIONI
125 Bruno Signinus, De figuris ecclesiae, PL 165,
888-891, esp. 890B: “Coronabitur igitur, et illa
corona coronabitur de qua superius dicitur:
‘Et in capite ejus corona stellarum duodecim
(Apoc. 12, 1)’. Possumus tamen per illam quae
de Libano venit eos intelligere, qui ex Judaeis
crediderunt [...]. Haec est autem illa mulier, sive
potius illa regina (coronata enim est, quod regi-
narum proprium est) de qua Psalmista loquitur,
dicens: ‘Astitit regina a dextris tuis in vestitu de-
aurato, circumdata varietate. Audi filia, et vide
et inclina aurem tuam, obliviscere populum
tuum, et domum patris tui; quoniam concupivit
rex speciem tuam [Psal., 44 (45), 10]’ ”.
126 Bruno Signinus, De laudibus beatissimae Virgi-
nis Mariae, PL 165, 1021A: “Non incongrue ergo
virgo Maria civitas Dei appellatur, quam virgini-
tas mentis et corporis, quasi murus ita ex omni
parte vallavit, ut nullus unquam libidinis acces-
sus adesset, et omnis inimicus a suae virginita-
tis corruptione deesset”.
127 Bruno Signinus, De laudibus beatissimae Virginis
Mariae, PL 165, 1022A: “Unde Salomon in Can-
ticis canticorum gloriam istius virginis decantat,
dicens: ‘Hortus conclusus soror mea, sponsa,
hortus conclusus, fons signatus: emissiones tuae
paradisus (Cant. 4, 12)’. Virgo quippe Maria fuit
hortus, in quo varii flores virtutum erant, et
conclusus, quia undique virginitate munitus”.
128 Is. 7, 14.
129 Jer. 31, 22.
130 Bruno Signinus, De laudibus beatissimae Vir-
ginis Mariae, PL 165, 1022B: “[...] id est virgo
Maria portabit in utero Christum, quem totus
mundus capere non potest”.
131 Kitzinger 1980, 11.
132 Tiberia 1996, 86, 96-101, 202 no. 18; Thérel
1984, 197-199. On Mary as mediator, see: Rau-
gel 1935, 152-161; Limongi (ed.) 1980, 43-52.
133 On Mary as advocata, see: Belting 2001 [1990],
381-403; Tronzo 1989, 176-179; Russo 1997.
Wirth 1999, 435-440; Parlato 2002, 61, 64.
134 Bittremieux 1929.
135 On the symbolism of scripts, see: Petrucci 1976.
queen. 125 In the fifth book, the Virgin Mary is compared to the city of God on the
grounds that her virtues are fortified as if by walls. 126 For these qualities Mary is
regarded as the hortus conclusus, the bride of the Song of Songs. 127 Reading on,
the association becomes more explicit, with Bruno quoting both Isaiah: “Ecce
virgo concipiet et pariet filium”, 128 and Jeremiah: “Faciet Dominus miraculum
super terram, femina circumdabit virum”, 129 that is the Virgin Mary who will
bear Christ. 130 St Bernard developed this exegetical tradition in his commentary
on the Songs of Songs. Ernst Kitzinger was the first to identify Bernard’s com-
mentary as an inspiration for the iconography of the apse conch. 131 Vitaliano
Tiberia has further related the mosaic to Bernard’s concept of “Mary as media-
tor”, 132 an association that in the mosaic is supported by the gesture of Mary
as advocata. 133 The teachings of St Bernard regarding Virgin Mary as mediator,
enjoyed a wide diffusion from the twelfth century onwards, and were accepted
by theologians and lay people alike. 134
The different voices in the dialogue between the Virgin and Christ are made
visible through the different media, scripts, and layout of the texts they hold. The
following discussion focuses on the “epiconographical” appearance of the scripts
on the book and the scroll. On the book, the text is written in black book hand, as
demonstrated by the three uncial Us (FIG. 24), 135 while the scroll has gold classi-
cal capitals on a blue background (FIG. 25). The letters are framed by the scroll,
but whereas text is usually written parallel to the long side of a scroll, in this case
it is written as if on a book, parallel to the short side of the scroll. Most important-
ly, the different media (book and scroll) have different visual impacts and sym-

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 121
136 Busia 2000a. 137 Cavallo 1997a; Cavallo 1997b; Busia 2000b.
bolic significance. The scroll held by Mary is technically a volumen. The volumen
symbolized intellectual activity in Antiquity, coming more specifically to signify
Christian doctrine in the early Christian period. 136 At that time, when books re-
placed scrolls, the book or codex was identified with the Bible. 137 The book held by
Christ is a codex. In this visual composition, the juxtaposition of volumen and co-
dex, written in capitals and book hand, symbolises the shift from old to new: from
scroll to book, and from the Old Testament to the New Testament. The different
graphic typologies and the different colors of the letters and backgrounds further
distinguish the two images and reveal an artistic program that assigns a precise
iconographic and symbolic value to script and its graphic space. One may there-
fore presume that the contrasting treatment of the two media and their respec-
tive graphic forms is meant to signify a sort of integration of the old law (the Old
Testament) and the new (the Gospels) through the embrace of the two figures. As
Fig. 24 – S. Maria in Trastevere, Rome. Mosaic in the apse conch. Christ’s book. Photo: S.Riccioni.

122 STEFANO RICCIONI
138 Kitzinger 1980, 13; Tronzo 1989, 177-178.
139 Enckell 2004; for a comprehensive bibliogra-
phy, see: Enckell & Romano 2006, 340.
is apposite given their didactic connotations, the book and the scroll teach those
who can read (literati) the meaning of the images of the apse arch. The Virgin
Mary, as sponsa and Ecclesia, is intimately connected with Christ, and also with
the pope, given the similarities, noticed by Kitzinger and Tronzo, with the icon of
Christ (acheropita) in the Sancta Sanctorum. 138
As suggested previously, the apse arch functioned as an introduction to the
main theme of the mosaic, and was adapted to fit the central message of the
apse conch. This becomes clear if we compare the images of the prophets and
their scrolls with those in other mosaics of the Gregorian Reform in Rome –
those of S. Clemente and those which once decorated the triumphal arch of S.
Maria Nova. 139 The latter, now lost, were recorded in two drawings by Antonio
Fig. 25 – S. Maria in Trastevere,
Rome. Mosaic in the apse conch.
The Virgin’s scroll. Photo: S.Ric-
cioni.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 123
140 Osborne & Claridge 1986, 212-218, nos. 84-96.
141 Julie Enckell has suggested 1165-1167, when
Pope Alexander III was in Rome; Enckell-Juil-
lard 2004, 28-33; Enckell & Romano 2006, 340.
142 Favreau 1997, 153-155, fig. 30.
Eclissi, 140 and can be dated to around 1160, probably between 1165 and 1167
(FIG. 26 and 27). 141 The iconography, identical as far as the figures of the proph-
ets are concerned, even to the extent of the colors of the scrolls in the drawing of
S. Maria Nova, is changed by the partial modification of the texts. While Isaiah
has the same verse at S. Maria in Trastevere and S. Maria Nova, he holds a scroll
with a different text at S. Clemente. Jeremiah has a quotation from Baruch at S.
Clemente and S. Maria Nova, but not at S. Maria in Trastevere. The difference
reflects the different iconographic messages of the respective apses: S. Clem-
ente’s apse shows the Crucifixion and S. Maria Nova displays Maria hodegetria
with the Son (FIG. 28). This observation leads to two conclusions. First, if the
theme of the apse conch of S. Maria in Trastevere originally had been intended
to be the Madonna of Mercy, as Nilgen suggested, we would expect the text on
the scroll of Jeremiah to be the same as that at S. Maria Nova. As it is, the
current iconography, with the enthronement of Christ and the Virgin Mary, is
heralded by the scrolls of the prophets and must therefore have been planned
from the beginning, solving the apparent problem of the alteration of the rules
of symmetry discussed above. Second, the comparison opens up other possible
ways of interpreting the apse of S. Maria Nova. When the texts of Isaiah and
Jeremiah are quoted together, they reveal the influence of Quodvultdeus’ Contra
Iudaeos, Paganos et Arrianos, in which the two prophets testify consecutively to
the coming of the Messiah, and orient the interpretation towards anti-Jewish
and anti-heretical polemic, which is not the case at S. Maria in Trastevere.
On the triumphal arch, the deliberate use of colors is also evident in the in-
scriptions below the evangelists and the prophets. In the case of the evangelists,
the frames of the inscriptions, which recall the works of goldsmiths, 142 are com-
posed of variously colored bands. From the outside to the inside, these are red,
sky-blue and white. The letters are white on a blue background. The frames of
the captions under the prophets use a greater number of colors: red, sky-blue,
green, white, gold and again green. The background is blue and the letters are
gold. This compositional approach, which isolates the scripts from their context
and highlights the name of the person, is an innovation. There is nothing com-
parable in earlier mosaics or other monumental images with captions. In both
the western and eastern visual traditions, and especially in the Roman mosaics
of SS. Cosma and Damiano, S. Maria in Domnica, S. Prassede, S. Marco, S. Ce-
cilia and S. Clemente, the inscriptions that identify the figures are not framed.
The choice of colors is also significant. In the frames of the evangelists the col-
ors correspond to those of gems mentioned in the passage in the Apocalypse de-
scribing the throne of Christ: red, white and green aquamarine. Commenting on

124 STEFANO RICCIONI
143 Bruno Signinus, In Apocalypsim, PL 165,
626A-626B: “Super hanc autem sedem Domi-
num sedere vidit, cujus species erat similis
aspectui lapidis jaspidis et sardinis. Jaspis
viridem habet colorem, sardius vero clarum
et igneum. Tali ergo colore Dominus noster
apparere voluit; ut nobis insinuaret quid ap-
petere debeamus. Habet enim colorem jaspi-
dis; quia semper viridis, semper vivens, semper
immortalis est, et nunquam ad siccitatem per-
veniens. Habet autem et colorem sardinis; siq-
uidem Deus noster est ignis ardens”; Honorius
Augustodunensis, Sacramentarium, PL 172,
751A-751B: “Sedes est Ecclesia, super quam
Dominus sedet. Color jaspidis aquae judicium,
color sardii judicium ignis”.
the passage, Bruno of Segni and Honorius Augustodunensis both interpreted these
colors ecclesiologically: jasper (green) and sardonyx (red and white) are specific to
the Church, the sedes on which God sits. 143 Moreover, red and green, with a par-
ticular gradation tending towards sky-blue (caeruleus, et quasi aqua viridis), are the
Fig. 26 – S. Maria Nova, Rome.. Mosaic on the triumphal arch, left side. Watercolour by Antonio
Eclissi, Windsor (RL 8976, n. 1), from Osborne & Claridge 1986, 214, no. 84.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 125
144 Bruno Signinus, In Pentateuchum, PL 164,
184A: “Et quia prius per aquam judicatus est
mundus, iterum autem per ignem est judican-
dus; ideo duo principales colores in arcu appa-
rent, viridis scilicet, et rubeus; et viridis quidem
aquam, rubeus vero ignem praetendit”; Bruno
Signinus, In Apocalypsim, PL 165, 626B-626C:
“Et iris erat in circuitu sedis, similis visioni
smaragdi. Smaragdus enim quasi herba viri-
dis est, per quam immortalitatem intelligimus,
quae semper virens nunquam ad siccitatem
pervenit. Iris autem, id est arcus coelestis, duos
principales colores habet, quorum alter est
igneus, alter caeruleus, et quasi aqua viridis; et
rubeus quidem martyrium, viridis autem bap-
tismum designat. Nemo ergo sedi Dei appro-
pinquabit, nemo viriditatem et immortalitatem
suscipiet, nisi aut per martyrium transeat, aut
baptismate abluatur: sint ergo iris, qui volunt
esse in circuitu solis”.
most important colors in the rainbow that appeared after the Flood. 144 Could the
frames of the evangelists’ inscriptions be an explicit reference to the arcus coelistis
described in the Apocalypse of John? The same colors appear in the frames of the
Fig. 27 – S. Maria Nova, Rome. Mosaic on triumphal arch, right side. Watercolour by Antonio
Eclissi, Windsor (RL 8974, n. 3), from Osborne & Claridge 1986, 215, n. 85.

126 STEFANO RICCIONI
prophets’ inscriptions, but here there is a different chromatic range. In addition
to bands of red, white and green (in two different bands, outside and inside of the
frame), there are bands of gold and sky-blue. These are the main “apocalyptic” col-
ors (green, red, gold, sky-blue, violet) which describe the Heavenly Jerusalem, as
suggested by Bruno of Segni in his commentary on the Apocalypse. 145
Turning now to the connection between colors, scripts, and images, I propose
that the mosaic uses colors to symbolize the ideals of the Church following the
Gregorian reform, but also to indicate to observers the correct way of reading the
images. The author of the Ad Herennium mentions three essential styles: elegantia,
compositio and dignitas. According to medieval rhetorical theory, the last must be
realized using exornationes or images, called colores from the eleventh century on. 146
145 Apoc. 21, 18-21; Bruno Signinus, In Apocalyp-
sim, PL 165, 719-729D.
146 Cicero M. Tullius ps., Ad C. Herennium, 4.18.4:“Di-
gnitas est quae reddit ornatam orationem varie-
tate distinguens. Haec in verborum et in sententi-
arum exornationes dividitur. Verborum exornatio
est quae ipsius sermonis insignita continetur per-
politione. Sententiarum exornatio est quae non in
verbis, sed in ipsis rebus quandam habet dignita-
tem”. Murphy 1974, 22-23.
Fig. 28 – 28. S. Maria Nova, Rome. Mosaic in the apse conch. Maria hodegetria. After G. Piccini
ICR from Andaloro & Romano 2006, 335, no. 1.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 127
The background of the apse is gold. As the most radiant color, gold was
traditionally a metaphor for light, the symbol of the incarnation of God in
Christ according to the saying in the Gospel of St John: “Ego sum lux mun-
di”. 147 Moreover, the medium of gold mosaic was a vehicle for expressing the
metaphysical idea of divine illumination. 148 For the reformers, gold also had
ecclesiological meaning. Bruno of Segni described the personification of the
Church as wearing golden garments. 149 Gold is an ornamentum ecclesiae, the
symbol of the ecclesiastical virtues, because it represents wisdom, integrity and
purity, without which no quality has value. 150 For Bruno the colors and the or-
namentum had ethical significance and were associated with goodness. More-
over, in the Sententiae he accorded a fundamental role to colors connected to
the liturgy, and urged readers to “see” and “understand” them. 151 In the mo-
saic, the clothes of the Virgin Mary and Christ are primarily golden, although
those of Mary display a variety of colors. The clothes of the saints are all dif-
ferent but executed with carefully chosen colors. 152 Pope Cornelius (251-253)
and Pope Calixtus (218-222) are dressed in blue/violet chasubles; Pope Julius
I (337-352) is dressed in emerald green; the priest Calepodius in scarlet red;
Pope Innocent II in purple with yellow highlights; and the deacon St Lawrence
wears a mantel with sky-blue and green highlights over his dalmatic. This dif-
ferentiation based on the colors of garments represented another innovation
in Roman apse decoration. As we have seen, the same colors are also used
on the frames of the triumphal arch. It is probable that they had an ecclesio-
logical (and liturgical) meaning, since they are associated in Bruno of Segni’s
De ornamentis ecclesiae and De Confessoribus with the four cardinal virtues:
hyacinthus - violet/blue - Wisdom; 153 coccus - red - Fortitude; 154 byssus - white
147 Ioh. 8, 12.
148 Especially on the metaphor of light in roman
mosaics, see Borsook 2000; Thunø 2003; Thunø
2005.
149 Bruno Signinus, De figuris ecclesiae, PL 165,
890B-890C. Bruno Signinus, De ornamentis ec-
clesiae, PL 165, 940A-940B: “Haec regina ipsa est
de qua loquimur, Ecclesia Dei […]. Multa sunt
ejus vestimenta, multa sunt ejus ornamenta;
et ideo varietate circumamicta perhibetur; sed
nullum ornamentum habet, quod vel aureum,
vel deauratum non sit. Nullum sine auro est ei
ornamentum. Saepe jam diximus quod per au-
rum, aut sapientia, aut vitae integritas, et puritas
designatur, sine quibus nullum ornamentum pul-
chrum est, et quodcunque his non decoratur vile
et abominabile est”.
150 Bruno Signinus, De ornamentis ecclesiae, PL 165,
940C-948D, esp. 940D: “Merito ergo in vestitu
deaurato Ecclesia pingitur, cujus omnia orna-
menta tam purissimo auro intexta sunt”.
151 Bruno Signinus, De Confessoribus, PL 165,
1064A-1067B, esp. 1066B-1066D: “Valde utile
est, hos colores semper inspicere, et quid signifi-
cent intelligere”.
152 On clerical clothing, see Braun 1924; Reynolds
1999. On liturgical colors, see: Leclercq 1914;
Reynolds 1999, 10-15.
153 Bruno Signinus, De Confessoribus, PL 165,
1066B: “Et hyacinthus quidem, quia est aerei
coloris, ad sapientiam te provocat, quae de sur-
sum est; quia ‘omnis sapientia a Domino Deo est’
”.
154 Bruno Signinus, De Confessoribus, PL 165,
1066D: “Coccus autem qui rubri et sanguinei
coloris est ad fortitudinem nos invitat, qua sancti
martyres armati occidi quidem potuerunt, vinci
non potuerunt. Est autem coccum bis tinctum,
quia duplex est martyrii genus, quoniam non so-
lum illi martyres sunt, quos tyranni interficiunt;
verum et illi, qui carnem suam cum vitiis, et con-
cupiscentiis crucifixerunt”.

128 STEFANO RICCIONI
- Temperance; 155 purpur - purple - Justice. 156 They were considered necessary
“to adorn” the ecclesiastical building, 157 the garments of its clergymen 158 and
in particular the Pope, suggesting that the saints were dressed in these colors
in order to visualize the virtues of the Church.
Very unusually, the scripts and backgrounds of the inscriptions identify-
ing the saints employ various colors which suggest associations between the
figures. Calepodius and Cornelius are connected by yellow scripts on a blue
background, probably because they were both martyrs and buried in the same
cemetery. 159 The inscriptions identifying Lawrence and Calixtus employ the
same golden script, but on different background: gold for Lawrence, blue for
Calixtus. The use of the gold could be explained by the importance of both
saints: Lawrence as a traditional saint of Rome and Calixtus as the founder of
the church of S. Maria in Trastevere. 160 St Peter and Innocent are associated
by texts written in white characters on a black background. Thereby a con-
nection is created between the first pope and the new founder of the church,
Innocent II, thus affirming him as the legitimate pope of the Roman Church.
Only Julius II remains isolated with his sky-blue script on a blue background.
This connection through color can be understood to follow the techniques of
rhetoric and the arts of memory regarding the need to create visual connec-
tions between images. 161
At the bottom of the mosaic, above the frieze of the lambs, is a monumen-
tal inscription written in gold capitals on two lines with different colored back-
grounds: green/sky-blue and blue. The text is composed in leonine hexameters
and was inspired by the early Christian inscriptions which exalt the beauty of
155 Bruno Signinus, De Confessoribus, PL
165,1066D: “At vero per lini candorem tempe-
rantia figuratur, quae semper laeta et asperum
nihil ostendens, omnia ad concordiam, et pa-
cem trahere conatur”.
156 Bruno Signinus, De Confessoribus, PL 165,
1066C: “Purpura vero justitia est, quoniam et
purpura, et justitia ad reges pertinet, qui et le-
gum conditores sunt, et purpura specialiter
induuntur. Quibus etiam dicitur: ‘Diligite justi-
tiam qui judicatis terram (Sap. 1)’ ”.
157 Bruno Signinus, De ornamentis ecclesiae, PL
165, 915C-915D: “Quatuor illi colores, quibus
vela et cortinae tabernaculi intextae erant,
qui ubique simul et nunquam separatim po-
nebantur, quatuor istas virtutes significabant.
Hiacynthus videlicet, et purpura coccus bis
tinctus, et bissus retorta de quibus in Exodum
sufficienter diximus: Nunc autem loquendi
solummodo materiam ministramus. Illis ta-
bernaculum, istis vero Ecclesia ornatur et de-
coratur, quia sine sapientia, sine justitia, sine
fortitudine et temperantia nulla anima ornari
vel decorari potest. Haec sunt illa ornamenta
de quibus dicitur: ‘Astitit regina a dextris tuis
in vestitu deaurato, circumamicta varietatibus
(Psal. 44, 10)’ ”.
158 Bruno Signinus, De Confessoribus, PL 165,
1066B: “Quatuor autem colores, quos ibi vides,
quatuor sunt virtutes principales; prudentia, ju-
stitia, fortitudo, et temperantia. His quatuor vir-
tutibus regitur mundus quae ita inter se conjun-
ctae sunt, ut una sine reliquis esse non valeat.
Ubi enim sapientia est, ibi justitiam, fortitudi-
nem et temperantiam esse oportet. Similiter au-
tem ubi justitia est, ibi sapientia est, fortitudo et
temperantia. Sic est in aliis”.
159 On Calepodius, see: Amore 1963; Furlani 1949.
On Cornelius, see: Gordini & Aprile 1964.
160 Cecchelli 1949; Ferretto 1963.
161 Carruthers 1998, 79-81.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 129
the decoration of the Church. The text is composed in three parts and is the key
to interpreting the mosaic:
HEC IN HONORE TUO PREFULGIDA MATER HONORIS /
REGIA DIVINI RUTILAT FULGORE DECORIS //
IN QUA CHRISTE SEDES MANET ULTRA SECULA SEDES /
DIGNA TUIS DEXTRIS EST QUA(M) TEGIT AUREA VESTIS //
CU(M) MOLES RUITURA VETUS FORET HINC ORIUNDUS /
INNOCENTIUS HANC RENOVAVIT PAPA SECUNDUS. 162
In the first verse the Virgin Mary is celebrated as the mother of God using
three terms (praeclara, rutilat, fulgore) which all allude to her brightness, a qual-
ity she holds in common with the building. Similar texts expressing the idea of
divine light were common in ancient inscriptions on apse mosaics in Rome, 163
but here it may also suggest a reference to the commentary on Song of Songs
written by Bruno of Segni. Bruno used the same verb rutilat to describe the shin-
ing quality of the Church, on account of which she was “adorned” by the virtues
(iam decore virtutum rutilat Ecclesia). 164 This part of the text also confirms the
iconographical interpretation of the mosaic as the union between Christ and the
Virgin Mary symbolizing the church of Trastevere and the Universal Church. 165
The Virgin was celebrated for her qualities as Ecclesia and regina because of the
association between the prefulgida mater and the shining regia. Moreover, the
Church shone because it was “adorned” with the cardinal virtues represented
by the various colors of the garments worn by the saints, while the crystalline
brightness of the polychrome clothes of Mary evokes the celestial Jerusalem. As
we have seen, gold was not only a metaphor for light and the incarnation of God
in Christ, but also symbolized the ornamentum ecclesiae.
In the second verse, the reference to the Apocalypse is emphasised. The employ-
ment and repetition of the term sedes, which is both a verb and a noun, evoke the
Biblical verse in which sedes means both the “apocalyptic throne”, on which God
sits, and the action of sitting. 166 To the right of Christ is Mary, wrapped in golden
garments, as in the descriptions of the celestial Jerusalem and the Queen of Psalm
44. The association between the Virgin’s qualities of mater, sponsa, Queen and Ec-
clesia becomes explicit in this part of the inscription.
162 Translation: “In your honour, shining Mother,
this palace of godly honour glows with the
brightness of beauty. / Where you sit, Christ,
will be a seat beyond time; worthy of your right
hand is she enveloped by the golden robe. / As
the old building was threatening ruin, Pope In-
nocent the second, originating from here, re-
newed it.” For a previous interpretation of the
inscription, see: Nilgen 1996, 162-163.
163 Borsook 2000.
164 Bruno Signinus, In Cantica canticorum, PL 164,
1248A.
165 For the political interpretation of Mary as the
Universal Church, see: Nilgen 1981; for the
interpretation of Mary as the local church of
Trastevere, see: Wirth 1999, 435-441.
166 Apoc. 4, 2: “[…] et ecce sedes posita erat in ca-
elo, et supra sedem sedens”.

130 STEFANO RICCIONI
Finally, the third part of the inscription connects the decoration and its sym-
bolic meanings to the celebration of Innocent II, restorer of the church and vic-
torious in the political controversy with Anacletus II. As has been observed, the
inscription sheds light on the iconography of the mosaic suggesting not only the
celebration of the Church as the Virgin but also the association of S. Maria in
Trastevere with Maria Ecclesia. Therefore, in reconstructing the destroyed building
both physically and morally (moles ruitura), Innocent II restored the regia, giving
her brightness and symbolically restoring the dignity of the Church (lost during
the papacy of Anacletus II), which Christ will maintain in eternity (ultra secula).
This part of the inscription contains a political message. The Virgin Mary, in the
recognizable garments of the Queen, symbolizes the Church (and the church of
Trastevere) that, thanks to Innocent II, obtained a new brightness, and, after the
Schism, returned to the arms of the pope, represented by the image of Christ.
We have seen that the mosaic is articulated by vertical and horizontal succes-
sions of motifs. The former refers to Christ, while the latter refers to the Church.
As was also the case at S. Clemente, the horizontal reading is more complicated
than the vertical one, revealing the complexity of the overall message by using a
sophisticated combination of texts, images and colors. Still, even if the mosaic of
S. Clemente was the model for the apse decoration of S. Maria in Trastevere, the
visual programmes were conceived with completely different aims. At S. Maria,
the horizontal reading has a kaleidoscopic character based on the relationship
between the Virgo / Ecclesia / Mater / Sponsa and Christ, as it is interpreted in the
exegesis of Bruno of Segni and St Bernard of Clairvaux, adapted to glorify Pope
Innocent II. The colors of the garments all have a particular meaning and the
scripts and frames of the inscriptions identifying the saints employ various colors
that encourage the viewer to make associations between the figures. The fourth,
and last, level of the horizontal composition is the frieze depicting the lambs that
proceed from Jerusalem and Bethlehem towards the Agnus Dei. The Lamb of God
has an unusual red nimbus, which should also be connected with a particular way
of interpreting the mosaic using the colors as a rhetorical guide. The colors and
the inscriptions make connections between the different parts of the narrative in
order to underline the central theme of the mosaic. From the red scrolls of the
prophets to the red garments of Innocent II and Calepodius, and the red-haloed
lamb below, the composition is symbolically unified in the image of God and in
the main theme of Christ’s marriage to the Virgin.
The mosaics composed during the Gregorian reform, from its original ideals
(S. Clemente) to its “triumph” (S. Maria in Trastevere), demonstrate how the re-
interpretation of traditional models of decoration creates new schemata, more
apposite to the contemporary message of the Church. Reformers created a new
language based on the display of closely connected scripts and images, which sys-
tematized knowledge for the benefit of the beholder. This language was arranged
in compositions using a kind of visual rhetoric, which borrowed from the rules of
the ars dictaminis and monastic rhetoric in order to teach and make memorable

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 131
the message of the reformed Church. Epiconography aids our understanding of
the various “iconographical” and decorative elements, especially the iconographi-
cal aspect of scripts used in images. The complexity of the visual narrative dif-
ferentiated the audience (idiotae and literati) based on their knowledge and un-
derstanding of its sophisticated combination of texts and images. Furthermore,
it functioned as an invitation to the observer to explore, elaborate, and create in-
dividual meaning as a stimulus for contemplation. This strategy, created at the
beginning of the Gregorian reform to teach reformed canons how to instruct il-
literati, was ultimately also used to celebrate the supposed triumph of the Church,
after the Investiture Controversy, during the short Roman papacy of Innocent II.
Stefano Riccioni
Scuola Normale Superiore di Pisa
Piazza dei Cavalieri, 7
I-56126 Pisa, Italia
BIBLIOGRAPHY
CCSL = Corpus christianorum, Series Latina, Turnhout,
1954-.
CSEL = Corpus Scriptorum Ecclesiasticorum Latinorum,
Wien, 1866-.
CT = Codice Topografico della città di Roma, R. Valentini &
G. Zucchetti (eds.), 4 vols., Rome 1940-1953.
Liber censuum = Le Liber censuum de l’Église Romaine,
publié avec un introduction et un commentaire par P.
Fabre et L. Duchesne, Tables avec la collaboration de G.
Mollat, 3 vols., Paris, 1889-1952 (Bibliothèque des École
Française d’Athènes et de Rome. 2a Série. Registres et
lettres des papes du XIIIe siècle, 3.7).
LP = Le Liber Pontificalis, ed. by L. Duchesne, 3 vols., Pa-
ris, 1886-1892 (new ed. by C. Vogel, I vol., Paris, 1955-
1957).
PL = Patrologiae cursus completus. Series Latina, ed. by J.
P. Migne, Paris 1844-1864.
Acocella N. 1966: La decorazione pittorica di Montecassino
dalle didascalie di Alfano I (sec. XI), Salerno.
Alberic of Monte Cassino, “Flowers of Rhetoric”, transl.
by J.M. Miller, in Reading in Medieval Rhetoric, ed. by
J.M. Miller, M.H. Prosser & T.W. Benson, Bloomington,
Indiana, 1973, 131-161.
Albericus Casinensis, Flores rhetorici, ed. by D.M. Iguanez
& H.M. Willard, Montecassino, 1938.
Ambrosius Mediolanensis, De mysteriis, ed. by B. Botte,
Nouvelle édition revue et augmentée de L’explication du
symbole, Sources Chrètiennes 25 bis, Paris, 1961.
Amore A. 1963: “Calepodio”, Bibliotheca Sanctorum, III,
Rome, 669-673.
Angelucci S. 2008: “Committenti, artefici e materiali delle
porte costantinopolitane d’Italia”, in M.M. Donato (ed.),
L’artista medievale, Pisa, 100-109.
Anselmi E. & D’Angelo C. [year]: “Relazione tecnica di
restauro”, Archivio dei Restauri 499/3.
Aronberg Lavin M. 2009: “Maria-Ecclesia and the mean-
ing of marriage in the late 13th century”, in L. Hodne &
S. Sande (eds.), Mater Christi (Acta ad archaeologiam et
artium historiam pertinentia 7), 153-169.
Astell A.W. 1990: The Song of Songs in the Middle Ages,
Ithaca – London.
Avagliano F. 1970: “I codici liturgici dell’Archivio di Mon-
tecassino”, Benedectina XVII, 300-325.
Avagliano F. (ed.) 1997: Desiderio di Montecassino e l’arte
della riforma gregoriana, Montecassino.
Baldwin C.S. 1928: Medieval Rhetoric and Poetic to 1400,
New York.
Barclay Lloyd J. 1986: “The Building History of the Medi-

132 STEFANO RICCIONI
eval Church of San Clemente”, Journal of the Society of
Architectural Historians XLV, 197-223.
Barclay Lloyd J. 1989: The Medieval Church and Canonry
of S. Clemente in Rome, Rome.
Barclay Lloyd J. 2005: “A New Look at the Mosaic of San
Clemente”, in A. Duggan, J. Greatex & B. Bolton (eds.),
Omnia disce. Medieval Studies in Memory of Leonard
Boyle, O.P., Aldershot, 9-27.
Barreto J., Cerman J., Soubigou G. & Toutain-Quittelier V.
(éds.) 2007: Visible et lisible. Confrontations et articula-
tions du texte et de l'image, Paris.
Basile G. 2002: “Il mosaico absidale di S. Clemente a
Roma”, in E. Borsook, F. Gioffredi Superbi & G. Pa-
gliarulo (eds.), Medieval Mosaics. Light, Color, Materials,
Milan, 149-155.
Belting H. 2001: Il culto delle immagini. Storia dell’icona
dall’età imperiale al tardo medioevo, Rome.
Berschin W. 1992: Bonizone di Sutri, Spoleto.
Bittremieux J. 1929: De Mediatione universali B. M. Virgi-
nis quod gratias, Bruges.
Bloch H. 1972: “Monte Cassino’s Teachers and Library
in the Middle Ages”, in La scuola nell’occidente latino
dell’alto medioevo (Settimane di studio del Centro Ita-
liano di Studi sull’Alto Medioevo, 19), Spoleto, 593-594.
Bloch H. 1984: “Der Autor der Graphia aureae Urbis Ro-
mae”, Deutsches Archiv für Erforschung des Mittelalters
40, 55-175.
Bloch H. 1986: Monte Cassino in the Middle Ages, 3 vols.,
Rome – Cambridge.
Bolgar R.R. 1982: “The teaching of Rhetoric in the Middle
Ages”, in B. Vickers (ed.), Rethoric Revalued, Bingham-
ton – New York, 79-86.
Borsook E. 2000: “Rhetoric of Reality: mosaics as expres-
sions of a methaphysical idea”, in Mitteilungen des Kun-
sthistorischen Institutes im Florenz XLIV/1, 3-18.
Bouché A.-M. & J. Hamburger (eds.) 2005: The Mind’s
Eye: Art and Theological Argument in the Middle Ages,
Princeton.
Braca A. 2003: Il Duomo di Salerno. Architettura e culture
artistiche del Medioevo e dell’età moderna, Salerno.
Braun J. 1924: Die Liturgischen Paramente in Gegenwart
und Vergangenheit, ein handbuch der paramentik, Frei-
burg.
Busia M. 2000a: “Rotolo”, in F. Bisconti (ed.), Temi di ico-
nografia paleocristiana, Vatican City, 274-275.
Busia M. 2000b: “Libro”, in F. Bisconti (ed.), Temi di ico-
nografia paleocristiana, Vatican City, 203-204.
Camargo M. 1991: Ars Dictaminis. Ars Dictandi, Turnhout.
Campana A. 1967: “Paleografia oggi. Rapporti, problemi e
prospettive di una ‘coraggiosa disciplina’ ”, Studi Urbi-
nati 61, 1013-1030.
Campana A. 1976: “Le iscrizioni medievali di S. Gemini”,
in U. Ciotti (ed.), S. Gemini e Carsulae, Milan, 81-132.
Campana A. 1984: “La testimonianza delle iscrizioni”, in
Wiligelmo e Lanfranco. Il duomo di Modena, Modena,
363-373.
Carruthers M. 2008 (1990): The Book of Memory. A Study
of Memory in Medieval Culture, Cambridge.
Carruthers M. 1998: The Craft of Memory. Meditations,
Rhetoric and the Making of Images, 400-1200, Cam-
bridge.
Castelnuovo E. & Sergi G. (eds.) 2004: Arti e storia del me-
dioevo, III, Del vedere: pubblici, forme e funzioni, Turin.
Cavallo G. 1994: “Testo e immagine una frontiera ambi-
gua”, in Testo e immagine nell’Alto Medioevo (Settimane
di studi del Centro Italiano di Studi sull’Alto Medioevo,
XLI, Spoleto, 15-21 Aprile 1993), I, 31-62.
Cavallo G. 1997a: “Buch”, in Der neue Pauly. Enzyklopädie
der Antike, II, Stuttgart, 809-816.
Cavallo G. 1997b: “Codex”, in Der neue Pauly. Enzyklopä-
die der Antike, III, 50-53.
Cecchelli C. 1933: S. Maria in Trastevere (Le chiese di
Roma illustrate, 31-32), Rome.
Cecchelli C. 1949: “Callisto”, in Enciclopedia Cattolica, III,
Vatican City, 386-393.
Chastel A. (ed.) 1971: L’art de la signature, (Revue de l’art)
XXVI.
Cicero M. Tullius ps., Ad C. Herennium. De ratione dicendi
(Rhetorica ad Herennium), translation by H. Caplan, ed.
by G.P. Goold, Cambridge - London, 1981.
Claussen P.C. 1981: “Früher Künstlerstolz. Mittelalterli-
che Signaturen als Quelle der Kunstsoziologie”, in K.
Clausberg & al. (eds.), Bauwerk und Bildwerk im Hoch-
mittelalter. Anschauliche Beiträge zur Kultur und Sozial-
geschichte, Gießen, 7-34.
Claussen P.C. 1992: “Nachrichten von den Antipoden
oder der Mittelalterliche Künstler über sich selbst”, in
M. Winner (ed.), Der Künstler über sich in seinem Werk
(Internationales Symposium der Bibliotheca Hertziana,
Rom, 1989) Winheim, 19-54.
Claussen P.C. 2002: Die Kirchen der Stadt Rom im Mittel-
alter 1050-1300, A-F (Corpus Cosmatorum II, 1), Stutt-
gart.
Cowdrey H.E.J. 1986 : The age of Abbot Desiderius: Monte-
cassino, the papacy and the Normans in the eleventh and
early twelfth century, Oxford.
Cowdrey H.E.J. 2000:Popes and church reform in the 11th
century, Aldershot.
Croiser J. 2006a: “I mosaici dell’abside e dell’arco absidale
della chiesa superiore di San Clemente”, in M. Andaloro
& S. Romano (eds.), La pittura medievale a Roma. 312-
1431 (Riforma e tradizione), IV, Milan, 209-218.
Croiser J. 2006b: “I mosaici dell’abside e dell’arco trion-
fale di Santa Maria in Trastevere”, in M. Andaloro & S.
Romano (eds.), La pittura medievale a Roma. 312-1431
(Riforma e tradizione), IV, Milan, 305-311.
De Rossi G.B. 1899: Musaici Cristiani e saggi dei pavimenti
delle chiese di Roma anteriori al secolo XV. Tavole cromo-
litografiche con cenni storici e critici del commendatore
Gio. Battista de Rossi, con traduzione francese, Rome.
De Spirito G. 2008: “Appendice epigrafica”, in M.M. Do-
nato (ed.), L’artista medievale, Pisa, 110-121.
Deér J. 1959: The Dynastic Porphyry Tombs of the Norman
period in Sicily, Cambridge, MA.
Di Carpegna Falconieri T. 2004: “Innocenzo II”, Diziona-
rio biografico degli italiani, LXII, Rome, 410-416.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 133
Dietl A. 1995: “Italienische Bildhauerinschriften. Selbst-
darstellung und Schriftlichkeit mittelalterliche Künst-
ler”, in H. Gersiepen & R. Kottje (eds.), Inschriften bis
1300. Probleme und Aufgaben ihrer Erforschung. Referate
der Fachtagung für mittelalterliche und frühneuzeitliche
Epigraphik Bonn 1993 (Abhandlungen der Nordrhein-
westfälischen Akademie der Wissenschaften, 94), Opla-
den, 175-211.
Dietl A. 1997: “Die Reliquienrekondierung im Apsismosa-
ik von S. Clemente in Rom“, in R.L. Colella, M.J. Gill,
L.A. Jenkins & P. Lamers (eds.), Pratum Romanum.
Richard Krautheimer zum 100. Geburstag, Wiesbaden,
97-111.
Dietl A. 2009: Die Sprache der Signatur: die mittelalterli-
chen Künstlerinschriften Italiens, 4 vols, Berlin.
Donato M.M. 1997: “Immagini e iscrizioni nell’arte politi-
ca fra Tre e Quattrocento”, in C. Ciociola (ed.), “Visibile
Parlare.” Le scritture esposte nei volgari italiani del Me-
dioevo e Rinascimento, Naples, 341-396.
Donato M.M. 2008: “Il progetto Opere firmate dell’arte
italiana/Medioevo, ragioni, linee, strumenti. Prima pre-
sentazione”, in M.M. Donato (ed.), L’artista medievale,
Pisa, 365-413.
Duggan L.G. 1989: “Was Art really the “Book of the Illiter-
ate”?”, Word and Image 5, 227-251.
Duggan L.G. 2005: In Reading Images and Texts. Medieval
Images and Texts as Forms of Communication (Third
Utrecht Studies in Medieval Literacy), Turnhout.
Enckell J. & Romano S. 2006: “I mosaici dell’abside e
dell’arco absidale di Santa Maria Nova”, in M. Andaloro
& S. Romano (eds.), La pittura medievale a Roma. 312-
1431 (Riforma e tradizione), IV, Milan, 335-343.
Enckell J. 2004: “Santa Maria Nova (Santa Francesca Ro-
mana) ou la ‘navicula’ d’Alexandre III”, Cahiers de civili-
sation médiévale 47, 17-35.
English E.D. 1998: Reading and Wisdom. The Doctrina
Christiana of Augustine in the Middle Ages, Notre Dame.
Erdmann C. 1933-1934: “Kaiserliche und päpstliche Fah-
nen im hohen Mittelalter”, Quellen und Forschungen
aus italienischen Archiven und Bibliotheken, 25, 1-48.
Fabre P. 1892: Étude sur le «Liber censuum» de l’Église
romaine, Paris.
Favreau R. 1979: Les inscriptions médiévales, Turnhout.
Favreau R. (ed.) 1996: Épigraphie et iconographie (Actes
du Colloque tenu à Poitiers les 5-8 octobre 1995), Poi-
tiers.
Favreau R. 1997: Épigraphie médiévale (L’atelier du médié-
viste, 5), Turnhout.
Ferretto g. 1963: “Callisto I”, Bibliotheca Sanctorum, III,
Rome, 680-689.
Filoramo G. 1998: “Variazioni simboliche sul tema della
porpora nel cristianesimo antico”, in O. Longo (ed.),
La porpora. Realtà e immaginario di un colore simbolico
(Atti del convegno di studio, Venezia, 24 e 25 ottobre
1996), Venice, 227-242.
Forsyth I.H. 1972: The throne of Wisdom. Wood Sculptures
of the Madonna in Romanesque France, Princeton.
Fraenkel B. 1992: La Signature. Genèse d’un signe, Paris.
Freedborg K.M. 1982: “Twelfth Century Ciceronian rheto-
ric: its doctrinal development and influences”, in B.
Vickers (ed.), Rhetoric revalued. Papers from the inter-
national Society for the History of Rhetoric, New York,
87-97.
Furlani S. 1949: “Calepodio”, Enciclopedia Cattolica, III,
Vatican City, 372-373.
Galbreath D.L. 1930: Papal Heraldy. A treatise on ecclesias-
tical heraldy, Cambridge.
Gandolfo, F. 1974-1975: “Reimpiego di sculture antiche
nei troni papali del XII secolo”, Atti della Pontificia
Accademia Romana di Archeologia. Rendiconti, III, 47,
203-218.
Gandolfo F. 1981: “Simbolismo antiquario e potere papa-
le”, Studi romani 29, 9-28.
Ganz D. & Lentes T. (eds.) 2004: Ästhetik des Unsichtba-
ren. Bildtheorie und Bildgebrauch in der Vormoderne,
Berlin.
Gardner J. [forthcoming]: “The placement of inscrip-
tions on painting and sculpture in Italy c. 1250 - c.1350:
contexts and status”, in II Colloquio Internacional de
Epigrafia Medieval, León 11 - 15 Septiembre 2006, forth-
coming.
Gislebertus Crispinus, Disputatio iudei et christiani et
anonymi auctoris. Disputationis iudei et christiani conti-
nuatio, ed. by B. Blumenkranz, Utrecht - Antwerp, 1956
(Stromata Patristica et Medievalia, 3).
Glass D. 1991: Romanesque Sculpture in Campania. Pa-
trons, Programs, and Style, Pennsylvania.
Gombrich E.H. 1969: “The evidence of image”, in C.S.
Singleton (ed.), Interpretation, Theory and Practice, Bal-
timore.
Gombrich E.H. 1982: The image and the Eye. Further stud-
ies in the Psychology of Pictorial Representation, Ithaca.
Gordini G.D. & Aprile R. 1964: “Cornelio”, Bibliotheca
Sanctorum, IV, Rome, 182-189.
Grégoire R. 1971: “Le Mont-Cassin dans la reforme de
l’église de 1049 à 1122”, in Il monachesimo e la riforma
ecclesiastica (1049-1122) (Miscellanea del Centro di stu-
di medieovali, 6), Milan.
Hahn C. 2006: “Vision”, in C. Rudolph (ed.), A Compan-
ion to Medieval Art: Romanesque and Gothic in Northern
Europe (Blackwell Companions to Art History), Oxford,
44-64.
Hamilton L.I. 2005: “To Consecrate the Church: Ecclesias-
tical Reform and the Dedication of Churches”, in Decor
et decorum: Reforming the Episcopacy in Bruno of Seg-
ni’s De laudibus ecclesiae (Eleventh Century), Aldershot,
87-137.
Hamilton L.I. 2007: Decor et decorum: Reforming the Epis-
copacy in Bruno of Segni’s De laudibus ecclesiae (Elev-
enth Century) LMS diss., Pontifical Institute of Mediae-
val Studies, Toronto.
Hamilton L.I. 2010: A Sacred City: Consecrating Churches
and Reforming Society in Eleventh-Century Italy, Man-
chester.
Herklotz I. 2001: “Sepulcra” e “Monumenta” del Medioe-
vo. Studi sull’arte sepolcrale in Italia, Naples.

134 STEFANO RICCIONI
Hermann A. 1969: “Farhe”, Reallexikon für Antike und
Christendum, VII, Stuttgart, 432.
Hesbert R.-J. & Prevost R. (eds.) 1963-1979: Corpus an-
tiphonalium officii, 6 vols. (Rerum ecclesiasticorum do-
cumenta, series maior, Fontes, 7-12), Rome.
Iacobini A. 2005: “Immagini, ideologie, storiografia: il
mosaico absidale del duomo di Salerno e l'arte della
Riforma gregoriana”, in A.C. Quintavalle (ed.), Medio-
evo: immagini e ideologie (Atti del V Congresso Interna-
zionale di Studi, Parma 23-27 settembre 2002), Milan,
288-301.
Iacobini A. (ed.) 2009: Le Porte del Paradiso. Arte e tecno-
logia bizantina tra Italia e Mediterraneo (Convegno in-
ternazionale, Istituto Svizzero di Roma, 6-7 dicembre
2006), Campisano.
Iguanez M. (ed.) 1915-1941: Codicum Casinensium Manu-
scriptorum Catalogus, 3 vols., Montecassino.
Iogna-Prat D. 2006: La maison Dieu. Une histoire monu-
mentale de l’Eglise au moyen-age (v. 800-v. 1200), Paris.
Illich I. 1993: In the vineyard of the text: a commentary to
Hugh’s Didascalicon, Chicago.
Juge M. 1944: La Mort et l’Assomption de la Sainte Vierge,
Vatican City.
Kessler H.L. 2000: Spiritual seeing: picturing God's invis-
ibility in Medieval art, Philadelphia.
Kessler H.L.& Zacharias J. 2000: Rome 1300: on the path
of the pilgrim, New Haven.
Kessler H.L. 2006: “Gregory the Great and Image Theory
in Northern Europe During the Twelfth and Thirteenth
Centuries”, in C. Rudolph (ed.), A Companion to Medi-
eval Art: Romanesque and Gothic in Northern Europe
(Blackwell Companions to Art History), Oxford, 151-
171.
Kessler H.L. 2007a: “A Gregorian Reform Theory of Art?”,
in S. Romano & J. Enckell Julliard (eds.), Roma e la Ri-
forma gregoriana. Tradizioni e innovazioni artistiche (XI-
XII secolo),Rome, 25-48.
Kessler H.L. 2007b: Neither God nor Man. Words Images
and the Medieval Anxiety about Art, Freiburg i.Br - Ber-
lin - Wien.
Kinney D. 1975a: “Excavations in S. Maria in Trastevere,
1865-1869: a Drawing by Vespignani”, Römische Quar-
talsschrift für christliche Altertumskunde und Kirchenge-
schichte 70, nn. 1-2, 42-53.
Kinney D. 1975b: S. Maria in Trastevere from its founding
to 1215, Ph.D, New York University.
Kinney D. 1996: “Making Mute Stones Speak: Read-
ing Columns in S. Nicola in Carcere and S. Maria in
Aracoeli,” in C.L. Striker (ed.), Architectural Studies in
Memory of Richard Krautheimer, Mainz am Rhein, 83-
86.
Kinney D. 2002: “The Apse Mosaic of Santa Maria in
Trastevere”, in E. Sears & T. K. Thomas (eds.), Reading
Medieval Images. The Art Historian and the Object, The
University of Michigan, 19-26.
Kinney D. 2006: “Rome in the Twelfth Century: ‘Urbs frac-
ta’ and ‘renovatio’ ”, Gesta XLV/2, 199-220.
Kitzinger E. 1972b: “The first Mosaic Decoration of Saler-
no Cathedral”, Jahrbuch der Osterreichischen Byzan-
tinistk, 21, 149-162.
Kitzinger E. 1980: “A Virgin Face: Antiquarism in Twelfth-
Century Art”, Art Bulletin 62, 6-19.
Klotz H. 1976: “Formen der Anonymität und des Indivi-
dualismus in der Kunst des Mittelalters und der Renais-
sance”, in P. Blum (ed.), Essays in Honour of Sumner
Mc Knight Crosby (Gesta XV/1-2), 303-312.
Krautheimer R. 1980: Rome, profile of a City. 312-1308,
Princeton.
Krautheimer R., Corbett S. & Frankl W. 1967: Corpus ba-
silicarum Christianarum Romae, III, Rome – New York.
Ladner G. 1984: Die Papstbildnisse des Altertums und
des Mittelalters. Addenda et Corrigenda. Anhänge und
Exkurse, III, Vatican City.
Lambert M. 2002: Medieval Heresy: Popular Movements
from the Gregorian Reform to the Reformation, Oxford
2002.
Larmon Peterson J. 2003: “The Transmission and Re-
ception of Alberic of Montecassino’s Breviarium de
dictamine. The Question of Alberic of Montecassino”,
Scriptorium LVII, n.1, 27-50.
Leclercq H. 1914: “Couleurs liturgiques”, Dictionnaire
d’archéologie chrétienne et de liturgie, III, Paris, 2999-
3004.
Leclercq J. 1960: Saint Pierre Damien ermite et homme
d’Église, Rome.
Libellus de cerimoniis aule imperialis (Graphia aureae Ur-
bis Romae), v. P.E. Schramm, Kaiser, Könige und Päpste.
Gesammelte Aufsätze zur Geschichte des Mittelalters. III.
Beiträge zur allgemeinen Geschichte, Stuttgart, 1969,
338-353.
Licitra V. 1977: “Il mito di Alberico di Montecassino ini-
ziatore dell ars dictaminis”, Studi Medievali, ser. 3a,
XVIII, 609-627.
Limongi P. (ed.) 1980: San Bernardo di Chiaravalle. Gli
scritti mariani, Rome.
Loew E. A. & Brown V. 1980: The Beneventan Script. A
History of the South Italian Minuscule, Rome.
Longo O. (ed.) 1998: La porpora. Realtà e immaginario di
un colore simbolico (Atti del convegno di studio, Vene-
zia, 24 e 25 ottobre 1996), Venice.
Lotti L. 1996: “Il restauro dei mosaici di Santa Maria in
Trastevere”, Alma Roma 37, 171-180.
Mâle E. 1942: Rome et ses vieilles églises, Paris.
Mâle E. 1953: L’art religieux du XIIe siècle en France. Etude
sur les origines de l’iconographie du Moyen Age, Paris.
Mansi J.B. (ed.) 1901-1962: Sacrorum Conciliorum Nova
et Amplissima Collectio, ed. by J.B. Mansi, Florentiae-
Venetiis, 1759-1768, (reprint facsmile, 54 vols., Paris).
Marsilio Colonna M.A. 1580: De vita et gestis Matthaei
apostoli et evangelistae, Naples.
Matthiae G. 1967: Mosaici medievali delle chiese di Roma,
2 vols., Rome.
Mazza A. 1681: Historiarum Epitome de Rebus Salerni-
tanis, Naples.
McKeon R. 1942: “Rhetoric in Middle Ages”, Speculum
XVII, 1-32.

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 135
Méhu D. (ed.) 2006a: Mises en scène et mémoires de la
consecration de l’église dans l’occident médiéval, Turn-
hout.
Méhu D. 2006b: “‘Historiae’ et ‘imagines’ de la consecra-
tion de l’église au Moyen Âge”, in D. Méhu (ed.), Mises
en scène et mémoires de la consecration de l’église dans
l’occident médiéval, Turnhout, 15-48.
Meyvaert P. & Devos P. 1955: “Trois énigmes cyrillimé-
thodiennes de la ‘Légende Italique’ résolues grace à un
document inédit, Translatio corporis santi Clementis
martyris ac pontifics”, Analecta Bollandiana LXXIII,
374-461.
Meyvaert P. & Devos P. 1956: “Autour de Léon d’Ostie et
la ‘Transaltio sancti Clementis’ ”, Analecta Bollandiana
LXXIV, 189-240.
Murphy J.J. 1971a: “Alberic of Monte Cassino: Father of
the Medieval Ars Dictaminis”, American Benedictine Re-
view XXII, 129-146.
Murphy J.J. 1971b: Medieval Rhetoric. A Select Bibliogra-
phy, Toronto.
Murphy J.J. 1974: Rhetoric in the Middle Ages: a history of
rhetorical theory from Saint Augustin to the Renaissance,
Berkley - Los Angeles.
Murphy J.J. 2005: Latin Rhetoric and Education in the
Middle Ages and Renaissance, Aldershot.
Newton F. 1976: “The Desiderian Scriptorum at Monte
Cassino. The ‘Chronicle’ and Some Surviving Manu-
scripts”, Dumbarton Oaks Papers XXX, 35-54.
Newton F. 1999: The Scriptorium and Library at Monte
Cassino (1085-1105), Cambridge.
Nieuviarts J. & Debergé P. (eds.) 2002: Les nouvelles voies
de l’exégèse. En lisant le Cantique de cantiques (XIXe
congrès de l’Association catholique française pour
l’étude de la Bible, Toulouse septembre 2001), Paris.
Nilgen U. 1981: “Maria Regina - ein politischer Kultbildty-
pus?”, Römisches Jahrbuch für Kunstgeschichte 19, 1-33.
Nilgen U. 1996: “Texte et image dans les absides des XIe -
XIIe siècles en Italie”, in R. Favreau (ed.), Épigraphie et
Iconographie, Poitiers, 153-164.
North W. 1998: In the Shadows of Reform: Exegesis and
the Formation of a Clerical Elite in the Works of Bruno,
Bishop of Segni (1078/9-1123), Ph.D. dissertation, Uni-
versity of California at Berkeley.
Ohly F. 1958: Hohelied-Studien. Grundzüge einer Geschich-
te der Hoheliedsauslegang des Abendlandes bis um 1200,
Wiesbaden.
Osborne J. & Claridge A. 1986: Mosaics and Wall paintings
in Roman Churches (The Paper Museum of Cassiano
Dal Pozzo Series: Early Christian & Medieval Antiqui-
ties Ser. a. Pt. II, Vol. 1), London.
Osborne J. 2009: “The cult of Maria Regina in early me-
dieval Rome”, in L. Hodne & S. Sande (eds.), Mater
Christi (Acta ad archaeologiam et artium historiam per-
tinentia, 7), 95-106.
Pace V. 1997: “La cattedrale di Salerno. Committenza pro-
gramma e valenze ideologiche di un monumento di fine
XI secolo nell’Italia meridionale”, in F. Avagliano (ed.),
Desiderio di Montecassino e l’arte della Riforma gregoria-
na, Cassino, 188-230.
Paravicini Bagliani A. 1994: Il corpo del papa, Turin.
Parkes M. B. 1976: “The Influence of the Concept of ‘Or-
dinatio’ and ‘Compilatio’ on the Development of the
Book”, in J.J.G. Alexander & M.T. Gibson (eds.), Medi-
eval Learning and Literature. Essays Presented to Richard
William Hunt, Oxford, 115-141.
Parlato E. 2002: “Le icone in processione”, in M. Andaloro
& S. Romano (eds.), Arte e iconografia a Roma, Milan,
55-72.
Parlato E. & Romano S. 2001: Roma e il Lazio. Il romani-
co, Milan.
Petrucci A. 1976: “Aspetti simbolici delle testimonianze
scritte”, Simboli e simbologie nell’alto medioevo (Set-
timane di Studi del Centro Italiano di Studi sull’Alto
Medioevo, XXIII, Spoleto, 3-9 aprile 1975), Spoleto,
813-844.
Petrucci A. 1986: La scrittura. Ideologia e rappresentazione,
Turin (English translation, Public lettering. Script, Pow-
er, and Culture, translated by L. Lappen, The University
of Chicago Press 1993).
Petrucci A. 1994: “Epigrafia”, in Enciclopedia dell’Arte Me-
dievale V, Milan, 819-825.
Quintavalle A.C. 2003: “Riforma gregoriana, scultura e ar-
redi liturgici fra XI e XII secolo”, in E. Castelnuovo &
G. Sergi (eds.), Del costruire: tecniche, artisti, artigiani,
committenti (Arti e storia nel Medioevo II), Turin, 235-
266.
Quintavalle A.C. 2009: “Giotto architetto, l'antico e l'IIle
de France”, in A. Tomei (ed.), Giotto e il Trecento : “il
più Sovrano Maestro stato in dipintura” (Catalogo della
mostra, I, Saggi), Milan, 389-437.
Quodvultdeus Carthaginiensis ep., Opera tributa. Contra
Iudaeus, Paganos et Arrianos, ed. by R. Braun, CCSL 60,
Thurnout, 1976, 228-258.
Radding C.M. & Newton F. 2003: Theology, Rhetoric and
Politics in the Eucharistic Controversy, 1078-1079. Al-
beric of Monte Cassino against Berengar of Tours, New
York.
Raugel A. 1935: La Doctrine Mariale de Saint Bernard, Pa-
ris.
Reynolds R.E. 1999: “Clerical Liturgical Vestments and
Liturgical Colors in the Middle Ages”, in R.E. Reynolds
(ed.), Clerics in the early Middle Ages: Hierarchy and Im-
age, Hampshire, 1-16.
Riccioni S. 2006: Il mosaico di S. Clemente a Roma.
‘Exemplum’ della Chiesa riformata, Spoleto.
Riccioni S. 2007a: “Segni epigrafici e sistemi illustrativi
‘alla greca’ nel mosaico di S. Clemente a Roma”, in Me-
dioevo Mediterraneo: l’Occidente, Bisanzio e l’Islam dal
tardoantico al secolo XII (Atti del VII Convegno interna-
zionale di studi, Parma, 21-25, settembre 2004), Milan,
371-380.
Riccioni S. 2008a: ‘L’Epiconografia’: l’opera d’arte come
sintesi visiva di scrittura e immagine”, in Medioevo: Arte
e storia (Atti del X Convegno internazionale di studi,
Parma, 18-22 settembre 2007), Milan, 465-480.
Riccioni S. 2008b: “Epiconographie de l’art roman en

136 STEFANO RICCIONI
France et en Italie (Bourgogne/Latium). L’art médié-
val en tant que discours visuel et la naissance d’un
nouveau langage”, Bulletin du Centre d'études médié-
vales d'Auxerre, 12, 267-282, (http://cem.revues.org/
document7132.html).
Riccioni S. 2009: The rhetoric of visuality in the mosaic
of S. Maria in Trastevere in Rome, LMS diss., Pontifi-
cal Institute of Mediaeval Studies, Toronto.
Riccioni S. 2010/2011: “La décoration monumentale
à Rome aux XIe et XIIe siècles: révisions chronolo-
giques, stylistiques et thématiques”, Perspective. La
revue de l’INHA 2, 319-360.
Romano S. 2006: “La chiesa trionfante (1100-1143
ca.)”, in M. Andaloro & S. Romano (eds.), La pittura
medievale a Roma. 312-1431 (Riforma e tradizione),
IV, Milan, 163-182.
Russo D.1996: “Les representations mariales dans l’art
d’Occident. Essai sur la formation d’une tradition
iconographique”, in D. Iogna-Prat, E. Palazzo, & D.
Russo (eds.), Marie. Le culte de la Vierge dans la so-
ciété médiévale, Paris, 173-291.
Russo D. 1997: “Sede Christi, Madonna Avvocata, à
Rome vers le milieu du XIIe siècle. La mosaïque d’ab-
side de la basilique Sainte-Marie du Trastevere”, in
La Vierge à l’époque romane. Culte et représentations,
(Revu d’Auvergne 542), 91-105.
Schramm P.E. (ed.) 1969: Kaiser, Könige und Päpste. Ge-
sammelte Aufsätze zur Geschichte des Mittelalters. III.
Beiträge zur allgemeinen Geschichte, Stuttgart.
Scorza Barcellona F. 2000: “Clemente I”, in Enciclope-
dia dei Papi, I, Rome, 199-212.
Seloni S. (ed.) 1990: Le porte di bronzo dall’antichità al
sec. XIII, Atti del convegno, Trieste, 2 voll., Rome.
Sepet M. 1867: “Les prophetes du Christ. Étude sur les
origines du théàtre au moyen âge”, Bibliothèque de
l’École de Chartres XXVIII, 1-27, 211-264.
Sepet M. 1868: “Les prophetes du Christ. Étude sur les
origines du théàtre au moyen âge”, Bibliothèque de
l’École de Chartres XXIX, 105-139, 261-293.
Sepet M. 1877: “Les prophetes du Christ. Étude sur les
origines du théàtre au moyen âge”, Bibliothèque de
l’École de Chartres XXXVIII, 397-443.
Sepet M. 1878: Les prophetes du Christ, Paris.
Sinding-Larsen S. 1978: “Some Observations on Litur-
gical Imagery of the Twelfth Century”, Acta ad Archae-
ologiam et artium historiam pertinentia VIII, 201-204.
Sinding-Larsen S. 1984: Iconography and Ritual. A
Study of Analytical Perspectives, Oslo.
Speciale L. 2009: “Memoria e scrittura: ‘Tituli’, pro-
gramma, scelte d’immagine da Montecassino a
Sant’Angelo in Formis”, Medioevo immagine e me-
moria (Atti dell’XI Convegno internazionale di studi,
Parma, 23-28, settembre 2008), Milan, 144-153.
Steigerwald G. 1999: Purpurgewänder biblischer und
kirchlicher Personen als Bedeutungsträger in der früh-
christlichen Kunst (Studien zur Alten Kirchenge-
schichte, 16), Borengässer - Bonn.
Stroll M. 1991: Symbols as power: the papacy following
the investiture contest, Leiden.
Supino Martini P. 1987: Roma e l’area grafica romanesca
(sec. X-XII), Alessandria.
Supino Martini P. 2001a: “Aspetti della cultura grafica a
Roma fra Gregorio Magno e Gregorio VII”, in Roma
nell’alto medioevo (Settimane di studi del Centro Ita-
liano di Studi sull’Alto Medioevo, XLVIII, Spoleto, 27
aprile - 1 maggio 2000), Spoleto, II, 921-968.
Supino Martini P. 2001b: “Società e cultura scritta”, in
A. Vauchez (ed.), Storia di Roma dall’antichità ad oggi.
Roma medievale, Bari, 241-265.
Themelly A. 2009: “Immagini di Maria nella pittura e
nei mosaici romani dalla crisi monotelita agli inizi
della seconda iconoclastia (640-819)”, in L. Hodne &
S. Sande (eds.), Mater Christi (Acta ad archaeologiam
et artium historiam pertinentia 7, 107-138.
Thérel M.-L. 1984: Le Triomphe de la Vierge-Eglise.
Sources historiques, littéraires et iconographiques, Pa-
ris.
Thomas L.-P. 1950: “Quatre systèmes de rubrication
dramatique. Dans le ms Paris latin 1139”, Speculum
IV, 107-110.
Thunø E. 2003: “Decus suus splendet ceu Phoebus in
orbe. Zum Verhältnins von Text und Bild in der Apsis
von Santa Maria in Domnica in Rom“, in B. Janowski
& N. Zchomelidse (ed.), Die Sprache der Bilder. Zur
Korrelation und Kontradiktion von Text und Bild im
Wirkungskreis der Bibel, Stuttgart, 147-164.
Thunø E. 2005: “Materializing the Invisible in Early
Medieval Rome: The Mosaic of Santa Maria in Dom-
nica in Rome”, in G. de Nie, K.F. Morrison & M. Mo-
stert (eds.), Seeing the Invisible in Late Antiquity and
the Early Middle Ages, Turnhout, 265-289.
Thunø E. 2007: “Looking at letters: ‘living writing’ in S.
Sabina in Rome”, Marburger Jahrbuch für Kunstwis-
senschaft 34, 19-41.
Tiberia V. 1996: I mosaici del XII secolo e di Pietro Ca-
vallini in Santa Maria in Trastevere: restauri e nuove
ipotesi, Todi.
Toubert H. 1990: Un art dirgé. Réforme grégorienne et
iconographie, Paris (Italian translation, Speciale L.
2001: Un’arte orientata. Riforma gregoriana e icono-
grafia, Milan).
Treffort C. 2006: “Une consecration «à la lettre». Place,
rôle et autorità des textes inscrits dans la sacralisa-
tion de l’église”, in D. Méhu (ed.), Mises en scène et
mémoires de la consecration de l’église dans l’occident
médiéval, Turnhout, 219-251.
Treffort C. 2008: Paroles inscrites. À la découverte des
sources épigraphiques latines du Moyen Âge (VIIIe-
XIIIe siècle), Rosny-sous-Bois.
Tronzo W. 1989: “Apse decoration, the Liturgy and
the Perception of Art in Medieval Rome: S. Maria
in Trastevere and S. Maria Maggiore”, in W. Tronzo
(ed.), Italian Church Decoration of the Middle Ages
and Early Renaissance, Bologna, 167-193.
Vasari G., Le vite de più eccellenti pittori e scultori ... nel-
le redazioni del 1550 e del 1568, ed. by R. Bettarini,

THE WORD IN THE IMAGE: AN EPICONOGRAPHIC ANALYSIS 137
commentary by P. Barocchi, Florence, 1967.
Verdier P. 1976: “Suger a-t-il été en France le créateur
du thème iconographique du couronnement de la
Vierge?”, in P. Blum (ed.), Essays in Honour of Sum-
ner Mc Knight Crosby (Gesta XV/1-2), 227-236.
Verdier P. 1980: Le couronnement de la Vierge: les ori-
gines et le premiers développements d’un thème icono-
graphique, Montreal.
Ward J.O. 1995: Ciceronian Rhetoric in Treatise, Scho-
lion and Commentary, Turnhout.
Ward J.O 2006: “The Medieval and Early Renaissance
Study of Cicero’s De inventione and Rhetorica ad He-
rennium. Commentaries and Contexts”, in V. Coox &
J.O. Ward (eds.), The Rhetoric of Cicero in its Medieval
and Early Renaissance Commentary Tradition, Leiden
- Boston, 3-75.
Wellen G.A. 1966: “Sponsa Christi: het absismozaïek
van de Santa Maria in Trastevere te Rome en het
Hooglied”, in Feestbundel F. van der Meer: opstellen
aangeboden aan Prof. Dr. F. G. L. van der Meer ter gele-
genheid van zijn zestigste verjaardag op 16 november
1964, Amsterdam, 148-159.
Wirth J. 1999: L’image à l’époque romane, Paris.
Yates F.A. 1966: The Arts of Memory, London.
Young K. 1921: “Ordo Prophetarum”, Wisconsin Acad-
emy of Sciences. Transactions, XX, 1-82.