Until the Reformation virtually all Western Christians permitted and even encouraged the use of religious imagery, following in this respect the example of the pagans rather than the Jews. The most famous justification for the practice appears in two letters written by Pope Gregory the Great at the end of the sixth century to Serenus, bishop of Marseille. Serenus had destroyed some narrative paintings (‘stories’) of ‘holy people’, apparently because they were being worshipped by members of his flock. Gregory was concerned that this action had alienated some of the congregation. Instead of destroying the paintings, Serenus should have explained their value in teaching and reminding the illiterate about the subjects depicted, at the same time stressing that worship was due to God alone.
A couple of centuries later Pope Adrian I declared that ‘holy images and painted stories were venerated from early times’. The distinction in art between stories, that is to say representations of events, and images – representations of people, sacred or not, in non-narrative contexts, such as a statue of the Virgin or a saint – remained a consistent feature of Christian writing on religious art over the next thousand years. It is ubiquitous in Italian writing on the topic and can also be found in medieval texts from England, France, the Iberian peninsula and, slightly later, from Germany. Stories provide instruction for those without access to texts, as Gregory had explained, and images are to be venerated but not worshipped, because that would be idolatry. This was the position taken in 1563 at the Council of Trent, where the main concern was to justify the use of painting and sculpture against Protestant criticisms, which were particularly directed at the invocation of saints. Accordingly, the Tridentine decree began with a direct defence of this practice, followed by a section on images and then one on stories, including the familiar Gregorian justification. It concluded with warnings against the illustration of heterodox beliefs and of anything that might arouse lascivious thoughts.
Although the public position of the Church was essentially defensive, in practice it recognised the value of stories and images, and encouraged their production. At least from the Middle Ages onwards they were to be found in vast numbers on the inside and outside of churches. Images of the Virgin and saints also decorated public buildings, city gates, private houses and even street corners. Their ubiquity underlined the idea that the faithful on earth belonged to a much larger community which also embraced the dead in purgatory and the saints already in heaven, not to mention innumerable angels of different types and sizes. The saints were particularly important because their privileged access to God enabled them to act as intercessors for believers, alive and dead. In this respect the Virgin was the most important of all.
It is very difficult to get a clear idea of the interiors of most medieval churches, especially the larger and more important ones, although they were evidently more cluttered than today. As tastes changed, early works of art were gradually replaced by later ones, which in turn were often eliminated in the 19th and 20th centuries, when Baroque art was considered out of keeping with medieval buildings. The result is that many of these churches are now more or less empty, for example the cathedral of Orvieto, which seems like a huge barn, with most of the surviving works it formerly contained now deposited in a nearby museum. Changes in devotional practice introduced after the Council of Trent also had a powerful effect. The monastic choirs that had previously obstructed the laity’s view of the high altar were often moved behind the altar, which was raised higher in order to be more easily visible to the congregation. At the same time, rood screens across the nave of many churches, which had almost entirely obscured the high altar, were eliminated. These changes were intended to give the faithful a more direct experience of the Mass, which previously had been heard rather than seen. The laity were now also encouraged to take communion more than once or twice a year, as had been the norm.
The most familiar examples of religious art specifically designed for churches are altarpieces, mostly painted but in some cases carved, placed above and behind altars. Their production provided an important source of income for Renaissance artists. Since most altarpieces were removable, they now constitute the most impressive pieces of Renaissance art, especially painting, to be seen in museums. One obvious consequence of their presence there, which mostly began in the 19th century, is that they soon acquired titles according to a now widely accepted system of classification, although in earlier times pictures in general were not categorised in such a consistent way. The modern system isn’t perfect, and a famous example of this is provided by paintings often titled ‘Tobias and the Angel’. Ernst Gombrich long ago pointed out that these were painted as representations of the archangel Raphael, who was the object of popular cult, with many churches named after him. Raphael was almost always painted with the child Tobias, whom he helped on one occasion. There was no cult of Tobias, and no reason why anyone would have wished to spend money on a painting in which he was the principal figure. He appears, in effect, as the identifying attribute of Raphael, just as the baby Jesus is the attribute of the Madonna.
Another example is provided by works showing the angel Gabriel appearing to the young Virgin, which are invariably now called an Annunciation. Rather unexpectedly, in texts before about 1800, when most pictures of this kind were still in churches, these works were an Annunciation when they were part of a series of stories of the life of Christ or the Virgin, but the ‘Nunziata’ (short for Annunziata) when the reference was to an altarpiece. In his Lives, Vasari referred to more than a hundred of the latter, but only one Annunciation as such, the first of a series of stories of the life of Christ on the bronze doors of the Florentine Baptistery. Works identified as the Nunziata were regarded as images of Santissima Annunziata, the Virgin Annunciate, of whom there was and is a very important cult, with many churches named in her honour. In other words, the altarpieces were understood as images of a person to be venerated, who was readily identifiable by the presence of the archangel Gabriel, whereas those works identified as an Annunciation were categorised as stories suitable for the instruction of unlettered people. Here the distinction between stories and images was not based on what the works in question looked like, but on their religious function, which was evident from their context. It is understandable that this distinction should have been disregarded by 19th-century art historians. This wasn’t the only instance in which the narrative subject was chosen over the iconic one. The same is true of the Crucifixion, the Deposition of Christ and the Assumption of the Virgin, to give just three examples. This way of categorising works of religious art is now almost ubiquitous. Italian art historians, like their colleagues elsewhere, now normally use the term Annunciation by default. But Titian’s huge altarpiece in the Basilica dei Frari in Venice, today called the Assumption of the Virgin in English-language publications, is still always identified as the Assunta by Italians, never the Assunzione.
It is understandable that David Ekserdjian consistently adopts the modern museum-type categorisation of subjects. His massive and very impressive book is remarkable for the range of altarpieces he considers, covering the whole of Italy from the early 13th century to 1600, many of them unfamiliar to all but a few specialists. As the subtitle indicates, one of his main themes is the process by which altarpieces with rows of panels showing single figures were gradually replaced by large single panels containing many figures, with an increasing stress on drama and interaction between them, as in scenes of martyrdom. This development is clear enough, but its significance is more problematic. Does it mean, as Ekserdjian seems to suggest, that in altarpieces images gradually gave way to stories, or that these two categories were not distinct, or indeed that they were largely imaginary, or that the requirements of patrons changed in some fundamental way over the centuries?
In dealing with such issues, he gives little weight to the persistence of the distinction between stories and images in texts before the 19th century. But his thinking is illuminated by his account of the uses altarpieces were supposed to serve. He singles out five main functions: as a backdrop to the Mass celebrated at the altar; as a means of identifying the dedication of the altar; as a means of instruction for the unlettered; as an inspiration for or aid to prayer and good conduct; or as a way of giving thanks. The first four seem consistent with the aims of the Church as an institution, while the last fits better with the likely motives of the patrons, who were mostly private individuals or groups of lay people such as confraternities.
But none of these supposed functions seem very compelling. There is not and never has been a requirement to have an altarpiece on an altar, and it is certainly not an essential feature for the celebration of Mass, when any members of the congregation present are supposed to concentrate on the actions of the priest. I know of one recorded instance of a priest himself looking at an altarpiece during Mass. This occurred in 1526 in Treviso, where the defacement of an altarpiece of the Virgin Annunciate was the subject of an inquiry. The picture included a likeness of the patron rather incongruously depicted in the far background looking towards the Virgin. During the inquiry a witness reported that a priest had told him that when he said the Memento, the prayer for the souls in purgatory, he looked towards the Virgin in the picture but instead saw the patron, which made him feel ‘thoroughly contaminated’, and he thought that anyone who eliminated that figure would be doing a good deed. That is precisely what happened shortly afterwards. It makes sense that the priest would have looked at the Virgin during the Memento, since she is our principal advocate and intercessor. Otherwise the imagery of the painting had no bearing on the Mass he was celebrating.
There are altarpieces that make clear allusions to the Mass, for example by showing the institution of the Eucharist by Christ at the Last Supper, or by emphasising the body of Christ, the Corpus Christi. These were generally commissioned by individuals or groups with a particular devotion to the Eucharist, and it cannot be concluded, as has often been supposed, that all or most altarpieces contain allusions to the Mass. Equally common today is the belief that an altarpiece is supposed to indicate the dedication of the altar on which it is placed. This idea appears in sets of instructions to priests written about 1290 in France and repeated in 1310 in Germany, in which it is stated that ‘in every church in front or behind or above the altar there should be an image, whether a sculpture, an inscription or a painting, clearly indicating and making plain to any observer the identity of the saint in whose honour and merit the altar was constructed.’ The instruction, whose purpose isn’t entirely clear, seems not to have been repeated and it is obvious to anyone who has visited Italian churches that it was not widely followed except in the case of high altars, in which the titulary saint of the church is often the main figure. But not all altars have altarpieces, and those that do not usually do not have inscriptions either. Even when altarpieces are present it isn’t always clear which of the various saints depicted, if any, is the relevant one. That is probably why in guidebooks and documents references to the name of the altar are very rare. A clear example of disregard of this supposed rule is provided by Titian’s famous Annunciation in the Venetian church of San Salvatore, which was painted for an altar of St Augustine.
Because Gregory the Great’s defence of paintings was specifically concerned with stories, the idea that altarpieces were intended as a means of instruction can have been directly relevant only to relatively few of them, such as the series of four showing episodes in the life of the Virgin, along with local patron saints, that once decorated the east end of the cathedral of Siena. These would surely have been accessible to the laity, although not necessarily during Mass. Ekserdjian concedes that most altarpieces were not painted stories, though he thinks that these became more common over time. Many of them incorporate anachronistic saints who can never have been present at the event supposedly depicted, which would seem to be inconsistent with Gregory’s position. As for the suggestion that altarpieces were an aid to prayer and good conduct, images may well have been relevant to prayer, but not to good conduct, for which stories provided countless relevant examples.
None of these considerations, in fact, tells us much or indeed anything at all about the motives of the patrons. Neither does Ekserdjian’s later comment that they were supposed to be ‘uniquely appropriate to the devotional concerns of the patron or patrons’, which begs the question of what those concerns might have been. His reluctance to explore the issue further has been shared by many art historians, but I think it is not difficult to suggest why so many patrons were prepared to spend very large sums of money on altarpieces, equivalent to at least twice the average annual wage of a working man. In fact, the expenditure involved was much more than that, because the patron normally had to acquire the rights to a chapel, provide an endowment for its maintenance and cover the costs of the construction of the altar itself and the frame. When a number of new chapels were constructed at Santa Croce in Florence as part of a scheme to meet Counter-Reformation recommendations in the 1560s, the altarpiece accounted for no more than about a quarter of the total cost of each chapel.
The ownership of a chapel was attractive to patrons for two reasons. First, it gave the family prestige and a reputation for piety. Hence many chapels in Italy are known by the name of the family who originally owned them, far more than are known by the name of the titulary saint of the altar. Second, and much more important, a chapel provided a burial place for the owners and their descendants. The advantages of burial in church were summarised by Saint Antoninus, the greatly revered 15th-century archbishop of Florence, who explained that the saints honoured by the church would intercede on behalf of the deceased, the faithful coming into the church would see the tomb and pray for the deceased, and the dead would be assured of rest undisturbed by demons. The most desirable location for a tomb was close to the high altar, where Mass was said most frequently. This may explain why church authorities were often able to induce the owners of chapels in this area to decorate them, at great expense, with painted stories. Side chapels with altars of their own were also favoured places for burial. The vast number of them in most pre-modern Catholic churches, such as the 38 in the 15th-century Florentine church of Santo Spirito, wasn’t a response to any liturgical need, but to the demand for private chapels, a demand which not only provided lucrative employment for artists, but also produced funds for the construction and maintenance of the churches themselves. Probably even more than is the case today, many of these chapels would normally have been locked. Ekserdjian claims there was a ‘requirement, during this period, that Mass be said at least once a day, however cursorily, before each and every consecrated altar’, but I have not found any evidence for this. On the contrary, I believe that in many private chapels Mass was said very rarely indeed, since the patrons would have had to pay for it. That is why the wills of wealthy people often included bequests for masses to be celebrated in their chapels on specified occasions.
In virtually all Italian wills of the Renaissance it was standard to begin by entrusting the testator’s soul to God, the Virgin and the celestial court, often naming individual saints. Because most chapels contained tombs it is not at all surprising that their altarpieces usually included images of saints, and especially the Virgin, as intercessors. As one would expect, the Virgin, the principal figure in so many altarpieces, is also often shown on Italian tombs. The role of saints in altarpieces is essentially the same as that of about a hundred statues of saints in niches still to be seen in the chapel of Henry VII in Westminster Abbey, which was constructed according to specifications contained in the king’s will. By honouring the saints in this way, he hoped that they would ease his passage in the afterlife, just as he hoped that the effigy on his tomb would encourage visitors to pray for his soul. In the same way, it was common for patrons in Italy to be shown in altarpieces as donors. Although Ekserdjian discusses donors, he is unforthcoming about why they are included at all. He is particularly hostile to Philipp Fehl’s idea that the donors, as depicted, are ‘wholly unaware’ of the sacred presences around them, but stops short of explaining why he doesn’t think it would be ‘unforgivably presumptuous for donors to imagine themselves in the company of Christ, the Virgin or the saints’.
The simple solution to the problem is that they are included to demonstrate their devotion to particular saints and to encourage others to share that devotion and also to pray for their souls. Instead of donors, many altarpieces of the 14th and early 15th centuries include on the frame an inscription identifying the patron and stating that the work in question has been commissioned ‘for devotion’ or ‘for the peace of his soul’ or similar sentiments. Sometimes inscriptions name the painter, and on occasion even the carpenter responsible for the altarpiece itself. Art historians usually see these as indications of the rising status of the artist, a popular cliché in the study of Renaissance art. But it makes more sense instead to regard them as requests to spectators to remember the individual named in their prayers, just like the inscriptions identifying the patron. After all, most artists shared the religious beliefs of their contemporaries.
One way of thinking about the behaviour of donors in altarpieces apparently looking at divine beings is that they are looking at images of those beings, as we are, because Christians are taught to make a clear distinction between images and those they represent. To take a famous example, Raphael’s fresco The Expulsion of Heliodorus from the Temple includes at the left Pope Julius II observing the event. Clearly Raphael was not implying that Julius ever saw Heliodorus, but indicating that he was mindful of what had happened, as an example of divine intervention on behalf of the church – or in this case its predecessor institution. Of course, confusion between the image and the thing represented by the image is much more likely to occur with painting than with sculpture, because painting is so much more illusionistic. No one would ask where the Virgin and saints are and what they are doing in connection with the famous group of bronze statues by Donatello behind the high altar of the Basilica del Santo in Padua, but art historians often ask where the saints and the Virgin are and what are they doing in altarpieces, especially those, common in Venice, in which the painted space seems an architectural extension behind the frame. Perhaps we should recognise that these figures are images, just like Donatello’s statues, and that such questions are not applicable to them.
An altarpiece showing a combination of saints in a single painted space is now often called a ‘Sacra conversazione’. The term, which was coined in the early 19th century and which has no Renaissance equivalent, could imply that the artist represented something more than a collection of separate images, even though the figures themselves are seldom if ever shown conversing. That altarpieces did increasingly come to look like narrative paintings is undeniable, and the process is charted very well by Ekserdjian. But it doesn’t necessarily reflect a change in function from image to story, as seems to be implied by the subtitle. Instead, it could be a consequence of changes in artistic practice and ideals, as illustrated in Raphael’s School of Athens, which shows philosophers from different periods interacting with one another, although there is no narrative subject and Raphael simply broke with the tradition of showing famous historical figures standing in a boring row. What happened with altarpieces was not very different, although such was the illusionistic power of painting that towards the end of the 16th century even some sophisticated contemporary writers were sometimes confused and critical – not that this had any lasting effect on artistic practice.
For Ekserdjian the presumed development represents a change in subject matter, but that is largely because modern notions of subject matter do not always fit well with those of artists and patrons of earlier periods. If they made a clear distinction, for example, between images of the Virgin Annunciate and representations of the Annunciation we ought to recognise that this mattered to them and to try to understand why. In this kind of inquiry it is important to recognise that the expectations of believers have changed greatly over the centuries. Few today, I suspect, are greatly worried that when they die their rest will be disturbed by demons. But for people who lived during the Renaissance that mattered a great deal, and their hopes and fears provided much employment for artists.