Can one "read" a work of visual art?

Polonius: What are you reading, my Lord?

Hamlet: A painting.

The purpose of this article is to offer points of view about the value of the commonplace metaphorical reference to the "reading" of works of visual art as if they were, like language texts, composed of an underlying linguistic structure. In order to deliberate the question posed in the title of the article the first part will deal with an historical issue in Western art. The implications of the statement by Pope Gregory the great (590-604) that depictions of religious narratives on the walls of churches represent the Bible for the illiterate will be treated in some detail by discussing a specific example of medieval art in which knowledge of a written text as well as other literary information is necessary for an understanding of its representation and meaning. In the second part the conclusions drawn from the discussion in the first part will be followed by a brief overview of contemporary theories which claim or deny that visual works of art can be "read".

PART ONE: IMAGES FOR THE ILLITERATE

The example chosen for discussion represents the relaying of a message via a messenger. The messenger as signifier is the angel in the Annunciation panel of the Hildesheim bronze door at present inside the St Michael's Church, designed and executed during the last years of office of Bishop Bernward who died in 1022. (1) The sixteen panels of the Hildesheim door form a complex narrative unity based on Old and New Testament scenes. Each panel is a complete pictorial image in itself, cast in high relief with some of the figures almost fully sculpted. To be brief, only the panel will be discussed on which an angel relays a message to an astonished woman. This example is chosen because the iconography of the Annunciation scene follows a more or less fixed pattern during the Middle Ages and Renaissance and has various features that make it suitable to an understanding of visual communication for the illiterate.

The messenger, the Archangel Gabriel, is represented in purely visual terms, without an explication, since this may not have been necessary for a medieval viewing public familiar with the Biblical narrative of the Annunciation in Luke 1: 6-32. The presence of this messenger in the depiction belongs to a New Testament cycle as visualised by Bishop Bernward and his sculptor, but the message he communicates will be interpreted from a modern vantage point (because for the present writer no other is possible), linking up with semiotic theory in a way that presents the image as a literal manifestation of signifier and signified.

Contrary to postmodern theory in which criteria for the interpretation of written or spoken texts are applied to visual works of art, amounting to a "reading" (2) of the visual image as a language utterance, an attempt will be made to treat the message that the angelic messenger in the Annunciation panel communicates in visual terms by applying the definition of semiosis formulated by Christopher Collins (1991: 7) as "the relay of a message via a messenger, that is a signified via a signifier". The Hildesheim angel relays his message by means of gestural rhetoric to a figure who has to decode its meaning. In his or her turn the viewer, as addressee, who stands in real space outside the scene on the bronze door, has to decode the meaning of the encounter between the represented figures.

Collins completes the statement quoted above by explaining: "The messenger carries burdens, verbal and perhaps also objectal, that within the social context are interpretable by all the addressees as meaningful." Obviously this statement is meant to be valid in relation to literary texts. In the case of visual images, even quite literally in the case of the depicted messenger relaying his message, there is no verbal communication. Both the messenger, the angel, and the virgin as the recipient of the message are mute. Furthermore, Collins (1991: 1) notes that in what is termed "imaginative literature", "we prize the verbal skill of poets and novelists to 'portray' persons and settings so vividly that we seem to view them with what some have called an 'inner eye'." While literary artists are lauded for portraying visual settings vividly, one may also be tempted to say that traditional Western narrative painting and relief sculpture often communicates so vividly with viewers that they "hear" the message relayed by the figures represented symbolically in visual settings with an "inner ear". However, notwithstanding all metaphorical analogies, the rhetoric of images is silent, while language consists of audible words.

Clearly, the angels in the Hildesheim panel and myriads of other medieval, Renaissance and later Annunciation scenes impersonate a narrator or rhetor. We know that the Archangel Gabriel, according to the passage in Luke referred to above, visited the virgin, who was taken by surprise at his appearance, and announced to her, according to Christian belief, that she would bear a son, which astonished her even more. The exact moment of the Annunciation is the moment of the Logos--the Word made flesh--that is, the conception of Jesus, God incarnate.

While the iconography of the Annunciation is more or less standardised in Christian art, the various artists' representations include or omit details relating to setting or symbolism in order to convey through gesture and other means what cannot be explained verbally. Indeed, in all Annunciation scenes the messenger is himself a visual message-the focal point of the composition.

The Hildesheim Annunciation panel is the opening motif for the New Testament cycle depicted on the door of the church of St Michael. Those viewers acquainted with the gospel of Luke in which the meeting of the Archangel Gabriel and the virgin Mary is related, as well as with the iconographic conventions for the representation of biblical figures, will recognise the winged figure which appears to the right of the panel as an angel and the figure to the left as the virgin. In this particular representation some details are unique. Of note is the open door to the left and the empty seat which feature quite prominently in the composition. Both the virgin's empty seat and the door open for the King have an allegorical meaning. Even more unique is the palm branch in the virgin's hand which is intended as an attribute to identify her. Only distant parallels for the inclusion of a palm branch in Annunciation scenes exist. On the door of Monreale Cathedral a palm branch is depicted in a vase placed between the virgin and the angel, and in only four Trecento and Quattrocento paintings does the angel carry a palm branch. In other Annunciation scenes either a lily or an olive branch is substituted. While in various later Annunciation scenes the angel hands the lily to the virgin, the Hildesheim Virgin holds the palm branch in her right hand.

The unique details of the Hildesheim Annunciation are understood better when seen in relation to the Creation of Eve panel, situated on the opposite side of the door. As a redemptive figure the Virgin supersedes Eve as the Second Eve. According to Church dogma the Virgin twice became a bride: when conceiving Christ and at the birth of the Church, both moments coinciding at the Annunciation. The folding stool from which the virgin arose to stand in the king's open gate can be seen as a representation of the virginal lap and a symbol (reminiscent of a throne) of the new leadership (Christ) announced by the angel. The King will be the head of the church in which the Old Testament prophecy of salvation comes to fulfilment. Thus the palm branch represents a distinct sign of New Testament salvation. It is the sign of Christ as the conqueror of death, and Gabriel's annunciation is symbolically represented as the last link in the chain of salvation. As interpreted by St Augustine, the body of the virgin had to be transformed into that of a bride in a twofold way: the Logos had to recreate human nature and by taking on human nature it became the bridegroom and Head of the Church (Gallistl 1987: 163). The palm is a sign of the chain's completion; God's becoming flesh through Mary. It represents the completion of salvation in the fullness of time, as promised in Ps 9 :13, Phil 3:14 and Revelation 7: 9. The virgin, the one chosen among all living beings, carries the palm branch which derives from a tree in Paradise and signifies the triumph of immortality through Christ, the Word or Logos awakened in her lap.

Clearly, a purely visual interpretation of a medieval representation of the Annunciation is inadequate. Without supplementary knowledge based on biblical evidence and other sources, only a very literal interpretation of the scene is possible. Collins (1991: 17) explains that "the audient spectator of an oral performance must know how to look and what to see; but the reader of a written text [and one may add: painted image] must do all this through the mediation of graphic symbols and, moreover, contextualize these verbal cues with supplementary, extra textual details."

As explained, the Hildesheim Annunciation panel relies on extra textual details which contribute to the meaning of its visual representation. This reopens the debate about whether a medieval narrative cycle based on Scripture, of which the Hildesheim bronze door is an example, was indeed a "Bible for the illiterate" as Pope Gregory the great claimed. (3) The English version of his credo (translated by L G Duggan, 1989: 227) reads as follows:

   Pictures are used in churches so that those who are ignorant of
   letters may at least read by seeing on the walls [of churches] what
   they cannot read in books (codicibus).

   What writing (scriptura) does for the literate, a picture does for
   the illiterate looking at it, because the ignorant see in it what
   they ought to do; those who do not know letters read in it. Thus,
   especially for the nations (gentibus), a picture takes the place of
   reading ... Therefore you ought not to have broken that which was
   placed in the church not in order to be adored but solely in order
   to instruct the minds of the ignorant.

In his own time Pope Gregory's statement was interpreted in a variety of ways, and art historians are still preoccupied with its implications. However, not all of them can be dealt with in this paper. An analysis by Lawrence Duggan, an art historian, suggests that the analogy between "reading" verbal texts and visual images can be misleading. It is difficult to decide in what way the viewing of images by illiterates corresponded to a "reading" of them, unless one is clear about what the Pope meant by "read by seeing" in the context of visual art. Duggan (1989: 243) concludes that "pictures cannot be 'read' as books can." In this he concurs with GC Coulton (1953), A Henry (1987), Ernest Gombrich (1982) and F Schier (1986). With the exception of Coulton the art historical research done by those mentioned above revises the accepted meaning of Pope Gregory's statement, as well as the interpretation of medieval narrative art.

Gombrich (198 : 155-157) notes in his discussion of Pope Gregory's statement that religious art cannot function without the aid of a specific context, an inscription or title, and codification which refers to conventional iconographic treatment. By itself an image cannot convey an unknown narrative to the viewer. However, images serve their purpose very well as illustrations of the stories conveyed in sermons. Schier and Gombrich emphasise the fact that illustrations, contrary to natural language, do not have any Grammatical rules. (4) One can recognise the content of images but one cannot "read" this content. Duggan (1989: 243-4) states that "while Leonardo [da Vinci], [Michael] Baxandall [1985] and others rightly insist that pictures can present in a coup d'oeil what words can do only at length, if at all, the other side of the coin is that pictures as instruments of precise communication fall far short of words, that a mark of the disparity is that pictures must be made intelligible in words to the intellect (but not necessarily other parts of the psyche), and that pictures cannot be 'read' in the same way, or as fully, as books."

The ambiguity contained in Pope Gregory's quoted statement will most probably never be resolved to the satisfaction of all art historical researchers, but Duggan (1989: 248) insists that it is an error to assume that images can do more than remind one of what one already knows and deepen such knowledge. In the medieval context A Kibedi Varga refers to L'Abbe Du Bos (1988: 194-208) who "tells painters to choose well-known subjects, that is, those in the Bible and in mythology: the image is not a second way of telling the tale, but a way of evoking it." Pope Gregory most certainly understood this, as well as the didactic potential of the visual arts, but it is clear that he overestimated the ability of the masses to understand visual art. (5) In this respect, and with a view to the multidimensionality of contemporary Art History and visual Culture, the debate over Pope Gregory's statement should be taken up again. Duggan (1989: 251) concludes his argument by stating: "Do we really have anything to fear if we at last admit that Gregory and his many disciples erred in regarding art as the book of the illiterate?" (6)

Without admitting as much, B Bruns (1992: 9) who wrote a scholarly book on Bishop Bernward's door, is one of Pope Gregory's disciples who believes that: "Die Bildsprache der Bernwardstur kann wie eine geschriebene Sprache erlesen werden" (i.e. its contents can be acquired by reading). And Bruns adds: "Die theologische Botschaft der Bernwardstur is zu Bernwards Zeit Verstanden worden, weil die darmaligen Theologen die beziehungsreiche und subtile Symbolsprache lesen konnten, in der sie verkundet wurde." Even though Bruns does not refer to Pope Gregory's statement explicitly, his approach to the interpretation of the narrative represented on the door nevertheless conforms in spirit to the Pope's insight. However, Bruns does emphasise that, first and foremost, the message of Bernward's door was directed at theologians. Their task was to elucidate the message to the populace in order to strengthen their faith (Bruns 1992: 9). In the same vein Michael Camille (1985: 33) reminds us: "The audience before most twelfth-century images would have been 'dumb', since they were unable to read the tituli or inscriptions, which like the text in ... [illuminated] manuscripts are crucial in interpreting the meaning of the picture."

The above analysis of only one image and interpretations of Pope Gregory's ideal for church art seems to suggest that--when the populace is illiterate--the main reason for using visual narratives is that it is the most effectual didactic way of communicating. I therefore concur with J Anthony Blair (2004: 53) concerning the visual expression of moral arguments: "Besides giving [a] moral argument a permanence, its visual expression communicates something unavailable to the verbal version, whether it is communicated orally or in writing. [...] It is one thing to hear a description [for example of the Annunciation] ; it is quite another, far more vivid and immediate, to see [it] with your own eyes." It therefore comes as no surprise that Desiderius Erasmus, a Renaissance scholar, reiterated the superiority of painting to "speech": "Painting is much more eloquent than speech, and often penetrates more deeply into one's heart." (7)

PART TWO: THEORIES FOR THE LITERATE

The above analysis strongly suggests that art historians who follow the postmodernist paradigm should admit that it is highly problematic to designate all works of visual art as "texts", demanding of the art historian as recipient to relay the message that the artist encoded to the viewer. In this respect, history offers a parallel in Christian narrative art which needed an intermediary in the person of a theologian to explain the meaning to illiterate believers. At present works of art are explained to a different kind of illiterate viewing public--a theory-illiterate public. This is understandable if one agrees with the way in which Tom Wolfe (1975: 6) parodied Leo Steinberg's famous dictum that: "Whatever else it may be, all great art is about art", by stating that the abandoned ideals of traditional art have been replaced by the "painted word", that is, works created for the sake of theory. Also Neil flax (1984: 2) points out: "[T]he artist in modern culture is acutely dependent on art criticism to explain his innovative work to a puzzled public." He furthermore points out that Michael Fried's Absorption and Theatricality (1980) "is a powerful demonstration that the history of art cannot be told in isolation from a critical reflection on art criticism." (Flax 1984: 1). But the problem with Fried's model of "a mutuality between pictures and language is that it accepts as given that what is precisely at issue in any modern critical account of the learned disciplines." (flax 1984: 1). The fact that modern conceptually based art (installations, performance, multimedia graphics, etc.) became self-referential has caused a situation which Paul Crowther (1993: 180) characterises as follows: "Broadly speaking, the artwork is what the artist designates as such, on the basis of some theory about art."

At a later date Crowther (1999: 128) explains the postmodernist dilemma:

   Unless the artist explains the intention and significance of the
   object, its meaning is unavailable. The road is thus clear for the
   critic to step in. And this is the supreme irony. Of all artistic
   idioms it is the conceptually based ones which affirm the hegemony
   of that insidious, priestly class of curators, critics, and art
   historians, who dominate the art scene.

We have come full circle in "reading" an art object--having its meaning explained to the viewer verbally by a "priestly class" instead of being able to interpret them visually. Crowther (1997: 2) thus reiterates what he said previously about the "colonization of twentieth-century art by those whose purpose it is to analyse it in theoretical and historical terms. Art historians, curators and theorists are now in effect the managers of meaning."

Art historians and literary critics interpret works of art or texts in order to understand them, but it is a prerequisite that they should understand the work of art or text in order to interpret it--a process which establishes a circularity that confounds hermeneutics. One cannot avoid reading a literary text because its medium is language. But when one sets out to "read" a visual artefact, the assumption seems to be that this is possible because such works emulate literary texts.

Since the "mutuality between pictures and language" is at present so deeply entrenched, it raises many questions, such as that asked by Michael Kelly (1995: 695): "Why is theory prominent in art history at this time?"--which he answers himself by asking more questions: "[T]he linguistic turn in philosophy has its counterpart in art history in the form of a semiotic turn; but then why did this turn not take place when Meyer Schapiro wrote about the semiotics of art? Or is theory an intrusion from some other discipline (e.g., literary theory) ..." (8) It may be the latter, since the concept of "visual literacy" (defined by David Rosand [1973/74: 443] as "the ability to read and respond to pictorial structures" is deeply entrenched in curriculae at academies where Art History is taught under the influence of postmodern paradigms, no matter how harshly they are critiqued." (9) In opposition to the acceptance of models such as that proposed by fried, Barbara Stafford, a practitioner of the discipline of visual Culture who concerns herself with the "intelligence of sight" (1996: 4), states that "we need to disestablish the view of cognition as dominantly and aggressively linguistic. It is a narcissistic tribal compulsion to overemphasize the agency of the logos and annihilate rival imaginaries." (1996: 7). Even though we agree that the logos is entrenched in our "tribal" memory, in representations like the Hildesheim Annunciation, it is also entrenched visually. visual representation as an aid for the illiterate, assuming visualisation to be a secondary intelligence, has also become contentious in current studies of visual images.

There is a self-evident parallel between medieval and modern figural art. While most medieval people who attended services in cathedrals were illiterate, most people who presently visit museums of art are literate. Even so, the modern "high art" of the museums is inaccessible to all but a privileged minority whose taste is mediated by art critics, art historians, aestheticians and museum curators. A new situation of power seems to have arisen with the role of medieval priests taken over by art critics. The effect of the comprehension or incomprehension of works of art was explained by Jose Ortega y Gasset (1966: 6-7) as long ago as 1926:

   When a man dislikes a work of art, but understands it, he feels
   superior to it; and there is no reason for indignation. But when his
   dislike is due to his failure to understand, he feels vaguely
   humiliated and this rankling sense of inferiority must be counter-
   balanced by indignant self-assertion.

Since the late 19th century modern "high art", much of which deviates from naturalistic representation, has evoked indignation among the average viewing populace. However, with the more "pictorial turn" that popular culture has currently taken according to WJT Mitchell (1994), the visual image as a medium for the transmission of information or ideas has finally been emancipated from its dependence on language as the mediation medium of scholarly interpretation. But has it? Or is the comprehension and interpretation of visual works--even of popular images--inescapably tied up with language as the medium of explication and interpretation?

There is a vast difference between interpreting a visual work of art and reading a literary text. However, it has become common practice for art historians to refer to the act of interpreting a work of art as an act of "reading." (10) Nelson Goodman's influential Languages of Art (1968) has, perhaps more than any other work on art theory, given rise to the fashionable practice to regard non-verbal representations, such as we find in drawings, paintings, sculptures and in some respects also architecture, as explicable only in the light of a linguistic or quasi-linguistic interpretation. In the above mentioned work (1968: 14) he states categorically: "Pictures in perspective, like any others, have to be read; and the ability to read has to be acquired."

It is no longer acceptable to evaluate works of art in higher and lower categories as Giorgio vasari had done in the sixteenth century, but to interpret them according to various theories. Such theories have proliferated to the extent that they cannot all be listed here. (11) Suffice it to say that art theory has become, at one extreme, an understanding of hermeneutic methods and, at the other extreme, a formal approach that focus on the work of art as an autonomous aesthetic and self-referential object. So where does the notion of "reading" a work that is visually expressed in terms of formal artistic ideals fit in? Most obviously where theory posits the visual work in terms of language. However, the critique of semiotics converted into a theory that turns painting into a visual text is summed up by Christopher S Wood's (1995: 678) statement in his review of Hubert Damish's The Origin of Perspective (1994): "Certainly the metaphor of 'reading' a painting, associated with orthodox structuralism and especially Louis Marin is jejeune."

What else can one do with a book than to read it? If the walls of the Romanesque churches are mentally to be transformed into a "book", then the pictures forming narratives are forms of script. It has indeed become so common that academic courses on art theory are called "visual literacy", implying that the methods of literary theory--however implausibly--can be applied to visual works of art. Furthermore, implying an acceptance of the postmodern dictum that all works of art are a "text" is problematical. As Donald Brook (1997: 44) points out, "resemblance-driven, non-verbal, representation is both prior to language and logically independent of language."

However, Crowther (1999: 135) offers a remedy for the contemporary malaise: "We need to clarify the symbolic structures of specific media, noting, in particular, the epistemic conditions of their legibility, i.e. the way in which such symbolic structures acquire a communicable meaning which is not tied to accompanying explanations from the artist or critic."

Clearly Crowther's concern is the invasion of one territory by another--such as is dealt with in this article: the invasion of the visual by the non-visual. The symbolic structure of language is its grammar by means of which its statement function is encoded. But can one speak about the "grammar" of an image by means of which information is transmitted? In this regard Gombrich (1982: 138) clearly states the differences between various symbolic forms of communication: "Looking at communication from the vantage point of language, we [...] shall see that the visual image is supreme in its capacity for arousal, that for expressive purposes it is problematic, and that unaided it altogether lacks the possibility of matching the statement function of language." To repeat: to understand a visual image the viewer needs to understand the codification which refers to conventional iconographical treatment. This implies a theoretical literacy necessary for the decoding of visual configurations that serve to convey meaning, but linguistics, particularly grammar, cannot be applied to visual configurations. (12)

This leads to two final remarks. The first is based on an insight by Ludwig Wittgenstein (1967: 27): "Everything is what it is and not another thing." Thus a work of art as a visually presented symbolic structure is not like a literary text. The second utilises Aristotle's (1941: 1457b 6-7) prototypal definition of metaphor, that it "consists in giving the thing a name that belongs to something else" to conclude this discussion about the ambiguity of "reading" a visual work of art.

REFERENCES

Aristotle, The Basic Works of Aristotle, R McKeon, ed. (New York: Random House, 1941).

M Bal, Reading "Rembrandt": Beyond the Word-Image Opposition (New York: Cambridge University Press, 1991).

Roland Barthes, Camera Ludica (New York: Hill and Wang, 1981).

M Baxandall, Patterns of Intention: On the Historical Explanation of Pictures (New Haven: Yale University Press, 1985).

J Anthony Blair, The Rhetoric of visual Arguments, in Charles A Hill and Margaret Helmers (eds), Defining Visual Rhetorics (Mahwah, New Jersey & London: Lawrence Erlbaum): 41-61.

D Brook, On Non-verbal Representation, British Journal of Aesthetics, July 1997, 37(3): 132-45.

B Bruns, Die Bernwardstur (Hildesheim: Bernward Verlag, 1992).

M Camille, "Seeing and Reading: Some visual Implications of Medieval Literacy and Illiteracy", Art History, March 1985, 8(1) 1985: 26-49.

W Cole, "The Book & the Artist: Rethinking the Traditional Order", Word & Image, Oct.-Dec. 1992, 8(4): 378-82.

C Collins, Reading the Written Image: Verbal Play, Interpretation and the Roots of Iconophobia (University Park, Penn: Pennsylvania State University Press, 1991).

P Crowther, Critical Aesthetics and Postmodernism (Oxford: Clarendon Press, 1993).

P Crowther, The Language of Twentieth-Century Art: A Conceptual History (New Haven and London: Yale University Press, 1997).

P Crowther, "Art and the Reconfiguration of Contemporary Experience", Filozofski Vestnik (Proceedings of the XIVth International Congress of Aesthetics, Ljubljana 1998, Part I), 1999: 127-35.

DA Dondis, A Primer of Visual Literacy (Cambridge: MIT Press, 1973).

LG Duggan, "Was Art Really the Book of the Illiterate?" Word & Image, July-September 1989, 5(3): 227-51.

GC Coulton, Art and the Reformation (Cambridge: Cambridge University Press, 1953).

J Elkins, On Pictures and the Words That Fail Them (Cambridge: Cambridge University Press, 1998).

N Flax, "fiction Wars of Art", Representations, Summer 1984, 7: 1-23.

M Fried, Absorption and Theatricality: Painting and Beholder in the Age of Diderot (Berkeley: University of California Press, 1980).

B Gallistl, "Die Tur des Bischofs Bernward und ihr ikonographisches Programm", in S Salomi (ed.), Le porte di bronzo dall'antichita al secolo XIII (Roma: Instituto della Enciclopedia Italiana) 1987: 145-82.

EH Gombrich, The Image and the Eye: Further Studies in the Psychology of Pictorial Representation (Ithaca: Cornell University Press / Oxford: Phaidon, 1982).

N Goodman, Languages of Art: An Approach to a Theory of Symbols (New York: Bobbs Merrill, 1968).

M Kelly, "A Review of Works on Art Theory", The Art Bulletin, 1995, LXXVII (4): 690-5.

A Kibedi Varga, "Stories Told by Pictures", Style, Summer 1988, 22(2): 194-208.

H Krause-Zimmer, Bernward von Hildesheim und der Impuls Mitteleuropas (Stuttgart: Verlag Freies Geistesleben, 1984).

Alberto Manguel, A History of Reading (London: Flamingo, 1996).

Estelle A Mare, "The Life and Work of Bishop Bernward of Hildesheim (993-1022), With Reference to Narrative Art as the 'Bible for the Illiterate'" [written in Afrikaans], South African Journal of Art History, November 1993 (11): 80-92.

Estelle A Mare, "The Relaying of a Message via a Messenger: An Analysis of the visual Communication Between the Angel and the virgin in the Annunciation Panel of Hildesheim Bronze Door, and a Reevaluation of Images as Texts", in VP Ryzhov (ed.), Information Paradigm in the Human Sciences (Proceedings of the International Symposium, Taganrog, Russia), May 2000: 183-9.

WJT Mitchell, Picture Theory: Essays in Verbal and Visual Representation (Chicago: University of Chicago Press, 1994).

J Ortega Y Gasset, The Dehumanization of Art (New York: Anchor Books, 1966).

David Rosand, "Art History and Criticism: The Past as Present", New Literary History, 1973/74 (5):435-45.

F Schier, Deeper Into Pictures: An Essay on Pictorial Presentation (Cambridge: Cambridge University Press, 1986).

D Sherer, "Tafuri's Renaissance: Architecture, Representation, Transgression", Assemblage, December 1995, 28: 35-45.

BM Stafford, Good Looking: Essays on the Virtue of Images (Cambridge, Mass: The MIT Press, 1996).

Mitchell Stephens, "By Means of the visible", in Lester Faigley et al, in Picturing Texts (New York & London: WW Norton): 56-67.

J J Sweeney, "Art History and the Artist", Art Journal, 1960/61, xx (3): 143-4.

G Vasari, Le opere di Giorgio Vasari, 9 volumes, G Milanesi, ed. (Florence: GC Sansoni, 1906).

L Wittgenstein, Lectures and Conversations on Aesthetics, Psychology, and Religious Belief, C Barrett, ed. (Berkeley and Los Angeles: University of California Press, 1967).

C S Wood, A review of H Damish, "The Origin of Perspective", J Goodman, trans. (Cambridge, Mass.: MIT Press, 1995) and Hubert Damish, "Le Jugement de Paris" (Paris: Flammarion, 199), The Art Bulletin, 1995, LXXVII(4): 677-82.

(1) For more extensive treatments of the Hildesheim bronze doors and the Annunciation panel respectively, see Estelle A Mare (1993) and (2000). For a complete image of the doors, see Helen Gardner, Art Through the Ages (London: G. Bell, 1959), 226.

(2) Strangely enough, the phrase "reading a painting or work of art" cannot be translated meaningfully into various languages, for example German, Dutch, Afrikaans and French. This common usage in English has surely been established by popular definitions of painting such as the following by James Johnson Sweeney (1961/62: 143): "Painting, as an art, is a language of conventions or symbols. This language is modified--even altered by each user." In a much more serious vein Nelson Goodman (1968: 38) states: "Realism is relative. We may find an ancient Egyptian representation unrealistic because we have not learned to read it."

(3) In all fairness one should point out that Gregory the Great was not the first to conceive of images in churches as comprising a book for the illiterate. In the last few years of the fourth century Saint Nilus of Ancyra [Ankara, Turkey] founded a monastery near his native town and "imagined the illiterate faithful coming to these scenes [from the Old and New Testament painted by the hand of a gifted artist] and reading them as if they were the words of a book" (quoted from Manguel 1996: 96).

(4) One may add the following insight by William Cole (1992: 381): "Language and image convey information in such radically different ways that it is simply impossible that either one could, in any meaningful sense, duplicate--and thus obviate the other."

(5) During the Renaissance two types of art coexisted: art for the docti, inspired by Classical art, and art for the indocti. According to Savonarola the latter kind of art, represented by images in churches, was the "bible" for women and children.

(6) It may or may not be due to the influence of Pope Gregory the Great that, after the discovery of printing, metaphors such as that used by Galileo activated scientific thinking: "The Book of Nature is written in the language of mathematics." In Christian religion the faithful were told that nature is god's creation that should be "read" like the Bible. However, Manguel (1996: 168-9) points out that book metaphors may actually have begun in Classical Greece and persisted in Christian times and later. For example, Saint Augustine was of the opinion that angels don't need to read the book of the world because they can see the Author Himself and receive from Him the Word in all its glory.

(7) Quoted from Stephens 2004: 56.

(8) The first third of On Pictures and the Words That Fail Them (1998) by James Elkins is an extensive attack on theories that present pictures as rational systems of signs, especially the so-called semiotic theory of art history which was prominent in the early 1990s.

(9) See Donis A Dondis (1973).

(10) There have been attempts to relate music to language, for example by Deryck Cooke, The Language of Music (London: Oxford Univ. Press, 1959), and Werner Jauk, "Sprache und Musik: der angebliche Sprachcharacter von Musik, International Review of the Aesthetics and Sociology of Music, 1995, 26(1): 97-106. From mid-1960 onward Manfredo Tafuri's work on the Renaissance is informed by the problem of language, as summed up by Daniel Sherer (1995: 37): "Thus architecture, by constitution a mute art, is prompted by the historian to speak of the conflicts that attended its genesis ..." See also Carlpeter Braegger, Architektur und Sprache. Gedenkschrift fur Richard Zurcher (Munchen: Prestel-Verlag, 1982). Not only Nelson Goodman (1968) refers to the "language" of art in the title of his book, but also Paul Crowther (1997), while the title of Mieke Bal's work on Rembrandt is Reading "Rembrandt" (1991).

(11) See M Kelly, "A Review of Works on Art Theory", The Art Bulletin, 1995, LXXVII (4): 690-5.

(12) One may note that Roland Barthes (1981) accepted defeat of his semiological perspective on visual images, but he did point out the necessity of examining the inner relations of images.

Estelle Alma Mare is professor emerita of the University of South Africa, and a research fellow in the Department of Art History and visual Culture Studies at the University of the free State in South Africa. She was awarded her third doctorate in 2003, for a thesis entitled "El Greco's Achievement of his Personal Maniera".