L'architettura dipinta da Agostino Tassi a Palazzo Lancellotti in Roma

Palazzo Lancellotti in Rome is the work of Francesco Capriani and Carlo Maderno, who restored several old buildings, giving them aristocratic unity and dignity. In a room on the first floor, created by restructuring the pre-existing medieval fabric, is preserved an extraordinary fresco by Agostino Tassi, depicting a loggia over a portico, covered by elegant cross-vaults in the midst of a fertile and hazy Roman country side. The spectator always has the illusion of depth, from any position in the room. The authors decided to apply the reverse perspective procedure to represent the painted architecture as if it were real and study its relationship with the solid architectural contest. This, however, meant that first of all they had to decipher the painting’s latent geometric code, whence the need to determine the focal point and the vanishing points. It was therefore decided to make a study two of the frescoed walls by close-range photogrammetry, and to extend the images of the parallel symmetrical edges on the model by computer, in an attempt to identify their point of intersection. But, against all expectations, they discovered a whole series of possible focal points, and found themselves faced with the impossibility of obtaining a model by adopting familiar well-proven methods. They accepted the fact that the problem could not be solved and proceeded experimentally, ready to apply statistical methods if necessary. Before formulating a hypothesis, however, they asked themselves what might have been the most likely technique used in transferring the original drawing onto the walt. By examining the accessible part of the fresco under a close oblique light they established that Tassi had mainly resorted to the use of cartoons. It was therefore presumed that, given the dimensions of the larger wall, the original drawing was at least ten times smaller than life size and had been divided into a matrix of more than a hundred elements. This would explain two errors present from the outset: one was generated by changing from one scale to another, and the other when positioning the cartoon on the wall. This led to the formulation and study of five hypotheses, three of which yielded positive results and produced the following conclusions. The painting was definitely preceded by a careful project, which must have materialized in the form of a considerably large original drawing (possibly measuring 1.00 x 1,5 m); in this original drawing the artist respected Leonardo da Vinci's principle of linear perspective, adopting the same width for all the column shafts. It is likely that in the original sketch the apparent distortions were reduced by extending the observation point to the entire central part of the room. It can therefore reasonably be assumed that a certain number of vanishing points were used in the original drawing for the lines in perspective. The sketch was then enlarged (not all of it, however) in the production of a series of cartoons, limiting the operation to one or two overlapping orders, and then making use of the symmetrical elements. This would have produced the errors of transfer and positioning. At this point, the initial project of giving a spatial rendering to Agostino Tassi's architectural painting, was no ordinary affair, and not so much due to the technical difficulties of the enterprise but rather because it involved a philological problem. Indeed, we learn from geometry that no real space can ever be equivalent to simulated space, in the sense that simulated space can evoke shapes and emotions in a way that is unique. So what would be the point of reducing a daydream, perpetuated over the centuries in the art of a dreamer, to an architectural project?