Encyclopedia Ichnographica

Nolli Plan of Rome

2/2



Nolli Plan of Rome


Pagan - Christian - Triumphal Way
3123h 3123i 3123j 3123k
1999.11.21




The place where the bodies of the dead were burned and buried under Hadrian
2001.07.18

abstract for Studium Urbis
2003.01.19 17:26

Mnemonically Delineating Veracity

"Authenticity is one thing, veracity another."
Marguerite Yourcenar, "Faces of History in the Historia Augusta" in The Dark Brain of Piranesi and Other Essays.

An apparent lack of veracity has always been at issue within modern interpretations G. B. Piranesi's Ichnographia Campi Martii (1757-62) despite Piranesi's extraordinary 'scientific' knowledge of ancient Rome and it's remains as evident throughout the four volumes of Le Antichità Romane (1756), as well as throughout Piranesi's other archaeological publications, including the Il Campo Marzio dell'Antica Roma. Contemporary architectural theorists from historian Manfredo Tafuri to architect Peter Eisenman view the Ichnographia as a city devoid of its own history, thus a plan prognosticating autonomous urbanism, yet that is exactly what the Ichnographia Campi Martii is not.

Beginning with comparisons between select portions of the Piranesi's Ichnographia and Giambattista Nolli's Pianta Grande di Roma, it becomes clear that the Ichnographia is an elaborate mnemonic devise. Like the imaginary building plans that Roman orators created in their minds as an aid toward the memorization of their speeches, the Ichnographia is literally an imaginary plan manifest as an aid toward the memorization of virtually all of ancient Rome's history. Thus the Ichnographia is not a fantastical reconstruction, rather, like the art of memory itself, the Ichnographia is a reenactment.

Mnemonically Delineating Veracity concludes with a brief reenactment of how an independent artist from Philadelphia came to discover a heretofore unnoticed initial(?) printing of the Ichnographia Campi Martii.




Campo Marzio discoveries
2004.03.18

2. the circle/square Porticus Neronianae marks the spot of Alaric's Visigoth camp prior to the (primal) break-in at the Porta Saleria. The 45 degree marks of the Porticus Neronianae marks the four cardinal points. Since the compass of the Ichnographia matches the center of the Porticus Neronianae in size, perhaps Piranesi is here indicating the spot where the "winds" shifted--the historical winds. I now know how the Nolli map corresponds with the Campo Marzio area of the Celian Hill/Horti Salustiani. Platner provides an interesting history of the "Domus Valerianus"--nobody bought it in 404 AD; it sold for almost nothing after Alaric 410 AD; it appears to have become a hospital (which makes sense with regard to Melania's mission.






plan superimposition
2008.08.14


Superimposition of a portion of Nolli's plan of Rome (1742) and a portion of Piranesi's Ichnographia Campus Martius (1762) featuring the relationship between the Horti Luciliani and the Spanish Steps.






Bufalini--Nolli--Piranesi 01
2010.09.12

"[Nolli's] debt to another plan-map of Rome, the 1551 woodcut by Leonardo Bufalini is more direct. On sheet 35/36 we find Nolli's tribute to the work of this Renaissance topographer. It is a reduction of the original woodcut with changes only in the orientation (north instead of north-east), in the graphical conventions, and in the decorations around the edges. With respect to all of its Medieval and Renaissance predecessors, Bufalini's map represents a remarkable change. From old collections of buildings drawn in elevation or oblique view, with little topographical consistency, and a nearly total absence of indications of streets in plans and views like those of Strozzi and Schedel, we suddenly pass to Bufalini's meticulously complete plan where every street and city block is clearly drawn. The map is not very precise in some of its angles and distances (e.g. the trivium of streets at Piazza del Popolo are shown converging at unequal angles), but it is the first map since antiquity that enables us to grasp the city in its topographic entirety. The varying width of the street is clearly indicated, with wide straight Renaissance streets contrasting with Medieval ones. In its inclusion of ancient buildings in reconstructed form, Bufalini's plan reflects the contemporary interest in antiquity.

Nolli made no attempt to rectify inconsistencies of the Bufalini map when he prepared the reduced copy. The Popolo trivium still has its unequal angles, and the Circus Maximus is still semicircular at both ends instead of one. He did, however, give greater relief than the original to the densely populated part of the city by shading in the city blocks. This is the same graphical convention he used on his own map. He maintained Bufalini's emphasis on antiquity by copying them exactly as in the original. Indeed he acknowledges this aspect by including views of ancient buildings at the bottom, where the map appears to be scrolled up, so as to reveal the "Colonnacce," a broken aqueduct, and the Pantheon (with Baroque additions) in the lower left corner, and the three ruins of the Forum Boarium in the lower right hand corner: The round Temple of "Mater Matuta," the Arch of Janus (with Medieval additions) and the Temple of "Fortuna Virilis."
Allan Ceen, Rome 1748: The Pianta Grande Di Roma of Giambattista Nolli in Facsimile (New York: J.H. Aronson, 1984), p. IV.

"The Grande Pianta includes all of the projects completed or about to be completed in Rome at the time that Nolli was preparing his magnum opus. [...] Many of these contemporary buildings are illustrated in the small Nolli plan (sheet 33/34 [immediately preceding the reduced copy of Bufalini's map-plan]). Here the cityscape was drawn by G.B. Piranesi whose name, together with that of Nolli, appears on the end of an overturned piece of ancient entablature drawn so as to seem protruding from the picture (a Piranesi trademark).
Allan Ceen, Rome 1748: The Pianta Grande Di Roma of Giambattista Nolli in Facsimile (New York: J.H. Aronson, 1984), p. V.

Bufalini--Nolli--Piranesi 02
2010.09.12

Piranesi acquired a detailed knowledge of Bufalini's Ichnographia Urbis of 1551 via his direct involvement with Nolli's Pianta Grande di Roma of 1748. A decade later, in 1758, Piranesi began his Ichnographia Campus Martius where, in some instances, he utilized Bufalini's map/plan as source material for the redrawing of ancient Rome's urban plan. Bufalini's plan, especially in the open areas all around the built-up section of Medieval and Renaissance Rome, includes 'reconstructions' of the larger ancient edifices like the imperial baths and stadiums, and some temple complexes. There are also near countless unnamed, fragmentary plans of ancient remains; remains, moreover, that, after consulting Nolli's plan, appear to no longer exist in Piranesi's time. It is from a select group of plan fragments on the Mons Pinicus or Collis Hortulorum of the Ichnographia Urbis that Piranesi imaginatively redraws the Horti Luciliani, the Sepulchrum Neronis, a Basilica along the Via Flaminia, the Horti Pincii, and the Monumentum Comitis Herculis.

Piranesi's resultant redrawn plans suggest a methodology whereby the fragmentary plans of Bufalini were used as kernels of ancient fact that, in turn, galvanized newly interpreted redrawings of what once was. As suggested by Dixon, "as there were great gaps in the knowledge of the past, great leaps were then needed to supply the holistic vision of the past which was the aim of scholars--archaeologists and historians--like Piranesi. In the Ichnographia, Piranesi filled the gaps..." Futhermore, from a strictly design point of view, Piranesi used some the fragmentary plans of Bufalini as contiguous elements which, when mirror-copied and multiplied, manifest the beginnings of the new plans.

Besides Bufalini's plan delineations, Piranesi also makes use of Bufalini's labelings. Bufalini labels all his full plan reconstructions of ancient buildings, often labels the fragmentary plans, and even labels blank locations (indicating the spot of an ancient edifice although actual remains no longer then existed). In utilizing the labels within the area of the Mons Pinicus or Collis Hortulorum, however, Piranesi hardly remains faithful to Bulafini's data. For example, where Bufalini positions the Horti Salustiani and Domus Pincii, Piranesi places the Horti Luciliani and, in turn, places the Horti Salustiani and Domus Pincii further east; the street Bufalini labels Via Conlatina, Piranesi labels Via Flamina; where Bufalini positions the Sepulcr. Neronis, Piranesi places the Bustum Caesaris Augusti and, in turn, labels an unnamed fragment along his Via Flaminia Sepulchrum Neronis; a small round structure Bufalini labels T. Solis, Piranesi labels Aula within the Horti Luciliani and, in turn, labels a newly imagined round building further south Delubrum Solis. It is honestly difficult to discern whether Piranesi is here playfully inverting Bufalini's data or actually rectifying Bufalini's "facts" with advanced knowledge of the past. Like Bufalini, Piranesi groups the Domus Martialis, Ludus Florae and the Templum Florae together, but he positions the group further west and moves the Domus Martialis south rather than north of the Ludus Florae. And where Bufalini locates the Sepulch. Falimiae Domiciarum, Piranesi places a very small Sepulcr. Familiae Aenobarb. and a very large Sep. Cnei Domitii Calvini whose plan Piranesi bases on an unrelated fragment of the Forma urbis.



««««

»»»»

www.quondam.com/e28/2823a.htm

Quondam © 2016.04.03