The fresco hypothesis
Over the last centuries, determination of the painting techniques used in Roman wall paintings has been subject to numerous controversies and conflicting results1 . However, during the last decades of the 20th century, the theory that fresco was the common painting technique in Roman murals gained general acceptance. However, in recent years this theory is being increasingly questioned due to the identification of organic paint media in Roman wall paintings and also to arguable hypotheses that would require full review:
1. Fresco painting techniques require painting on giornate (limited surfaces of mortar that the artist is able to paint over in one day of work)2 . It is currently accepted that the oldest wall paintings executed with giornate date form the second half of the 13th century 3 . In consequence, wall paintings prior to that time could hardly be considered frescoes.
2. In Roman, Early Byzantine and Romanic wall paintings, lime plaster is not divided into giornate, but in pontate, great expanses of mortar whose heights correspond to the levels of scaffolding. Current studies on Roman painting consider that “the painting is executed on a fresh intonaco that is always applied in pontate” 4 . To adequately assess this hypothesis, it is important to bear in mind that, quite to the contrary, pontate are “an unequivocal sign of a painting executed on dry plaster” 5 , and that not a single experimental study or technical treatise on fresco can corroborate the feasability of painting with fresco on pontate 6 .
3. Frescoes reveal a difference in tones across older and newer giornate section joints, demonstrating the difficulty in matching tones of a section being painted with those of an already dry giornata, as color intensity decreases during drying 7 . The absence of tone variation seen around pontate joints in Roman wall paintings indicates that the plaster was dry when the painting was executed.
4. Pliny 8 and Vitruvius 9 , when writing of white pigments, do not mention lime, which is the most important pigment in fresco painting. In Roman wall paintings, white pigments are basically composed of carbonated minerals: cretes, aragonites, marbles 10 . These pigments are excluded as white pigments in fresco treatises because they are transparent when mixed with water and do not allow the artist to see the effect of white brushstrokes during application, as they become opaque only after drying 11 .
5. Roman wall paintings often make use of pigments which cannot be used in fresco because they degrade on contact with lime: white and red lead, malachite, azurite, cinnabar, orpiment and organic pigments. The presence of these pigments indicates that these paintings were not executed on a fresh lime rendering 12 .
6. The paragraphs in which Pliny and Vitruvius comment on the use of pigmented lime renderings have sometimes been taken as proof that Roman painters knew of the fresco painting technique. However, fresco is not the only technique that uses pigments on wet plaster. Colored stuccoes, which do not allow for painting figured decorations, are also executed this way. Pliny´s comments do not allow deduction of the technique used. Vitruvius is more explicit, and his words seem to correspond to colored stuccoes, not to frescoes 13 .
7. Occasionally, the notion is expressed that the presence of calcium carbonate in colored areas of Roman wall paintings indicates that a fresco technique was used 14 . However, it was common among Roman painters to mix colored pigments with powdered crete, marble, or calcite 15 , which would explain the detection of calcium carbonate independent of the painting technique used.
8. The stratigraphic study of Roman wall paintings shows no penetration of the pigments into the lime mortar, indicating that the painting was not done in fresco, but on a dry rendering.
9. Sometimes, chemical studies of Roman wall paintings have not identified the presence of organic matter in the paint layers, leading authors to conclude that the painting must have been done with fresco. But this chemical result could as easily be due to the difficulty of extracting and characterizing the organic binders in ancient paint samples, which can produce widely divergent results between separate chemical studies of a single work 16 .