Spatial Vision, Vol. 21, No. 1–2, pp. 3–18 (2007) Koninklijke Brill NV, Leiden, 2007.Also available online - www.brill.nl/sv

Piranesi and the infinite prisons

SERGIO RONCATO ∗

Dipartimento di Psicologia Generale, Università di Padova, Via Venezia 8, Italy

Received 14 July 2006; accepted 24 December 2006

Abstract—Piranesi was an extraordinarily talented artist who came to be considered the best knownengraver and etcher of the 18th century. He spent his lifetime recording the magnificent buildings andruins of ancient Rome. In his earlier work, he developed architectural fantasies and dark visions ofimaginary prisons, the Carceri d’Invenzione (Imaginary Prisons), which have fascinated people eversince they first appeared. They also made an impact in later artistic and literary contexts.

Today they are considered the best expression of bewilderment, of the world’s massive oppressionof man. How did Piranesi generate such a powerful source of meaning? I have reviewed a numberof contributions by art essayists in the search for what lies behind the fascination of the ‘Carceri’.Analysis of these texts provides a rich set of aesthetic and psychological attributes as well as detailedindications of where to seek the source of the meanings. I suggest here a fruitful integration ofthis analysis, developed by considering various particular perceptual and geometric cues. I discussthree possible sources of meaning: the combination of architectural elements out of proportion; thedifficulty of building a unitary space; and the ‘bottom perspective’ that gives rise to a sensation offloating.

Keywords: Space perception; art perception; feature integration; impossible figures.

OVERVIEW OF PIRANESI’S WORK

Giovanni Battista Piranesi was born in Mogliano near Treviso in 1720 and died inRome in 1778. Son of a Venetian stonemason, his first teacher was an uncle, MatteoLucchesi, who taught him to draw and design architectural structures as well astheatrical sets. His uncle also imparted an appreciation of Palladium’s work. In 1740he left for Rome where he remained almost all his life. He became a draftsman inthe entourage of the Venetian ambassador Marco Foscarini, but the city itself offeredhim the opportunity of developing an interest in classical antiquities. In these yearsin Rome, he also studied perspective and stage design under Giuseppe Valeriani,and engraving under Giuseppe Vasi.

∗E-mail: sergio.roncato@unipd.it

4 S. Roncato

He returned to Venice briefly in 1744, hoping to establish a career as an architectbut failed and returned to Rome in 1747, setting up as an engraver and printpublisher and finding an immediate outlet for his talent. Eighteenth century Romehad become one of the main destinations for visitors on the Grand Tour of Europe;souvenirs most appreciated were the vedute (topographical views) and capricci(imaginative compositions, a Venetian tradition). Seizing the moment, Piranesiopened a gallery for the sale of prints — chiefly his own — of his etchings ofpoetic ruins and of Renaissance Rome. The international reputation he had gainedgrew rapidly.

Over his career he produced around 1000 etchings. There were three main series:Vedute di Roma (Views of Rome); Antichità Romane (Roman Antiquities); and theCarceri d’Invenzione (Imaginary Prisons), this last a group of 16 etchings. A firstset of 14 plates of capricci (Invenzioni capric(ciose) de Carceri) were publishedduring his first stay in Rome.

In a republishing in 1761, all the etchings were reworked and two new ones added.In this second edition, Piranesi heightens the dark/light contrasts in an architectureof increasing complexity.

In Fig. 1 three etchings are reported, the complete series can be seen at the web

(a)

Figure 1. The Carceri for the second edition of 1761. (a) Plate III (The round Tower).

Piranesi and the infinite prisons 5

(b)

(c)

Figure 1. (Continued.) (b) Plate VI (The Smoking fire). (c) Plate XIII (The Well).

6 S. Roncato

(d)

Figure 1. (Continued.) (d) Plate XIV (The Gothic Arch). The arrow indicates volumes impossible toextend in depth: an anticipation of Reutervärd, Escher, Penrose impossible figures.

sites listed in the References.His powerful images of the Roman ruins fascinated his contemporaries; in

particular they had a strong influence on Neoclassicism and Romanticism. Themenacing, nightmarish atmosphere of the Carceri inspired the Romantic artists fora century after his death, and exerted strong influence on the Surrealist movementin the 20th century. This influence extends even to cinema, Eisenstein beingparticularly struck by the Carceri (Nesbet, 2006). In an article from 1946 (reprintedin Oppositions 11, 1997), he analysed and described an imaginary film sequencebetween two engravings by Piranesi: the Carcere Oscura (Dark Prison) of 1743and Plate XIV of the Carceri d’Invenzione of 1761 (see also Tafuri, 1987).

REACTIONS TO PIRANESI BY THE LITERARY WORLD

The imaginary interiors of Piranesi had a profound effect not only on stagedesigners, but also in literature, as clearly demonstrated by eminent writers suchas Goethe, Stendal, Byron, Coleridge, De Quincey, Canetti, Huxley, Yourcenar,who made significant commentaries. I have chosen extracts from two essays —

Piranesi and the infinite prisons 7

written six decades apart — that demonstrate the persisting interest in Piranesi andhis continued relevance.

(i) Prisons by Aldous Huxley. This essay originally formed an introduction to a1949 edition of Piranesi’s famous etchings.

All the plates in the series are self-evidently variations on a single symbol,whose reference is to things existing in the physical and metaphysical depthsof human souls — to acedia and confusion, to nightmare and angst, toincomprehension and a panic bewilderment. The most disquietingly obviousfact about all these dungeons is the perfect pointlessness which reignsthroughout. Their architecture is colossal and magnificent. One is made tofeel that the genius of great artists and the labour of innumerable slaves havegone into the creation of these monuments, every detail of which is completelywithout a purpose. Yes, without a purpose: for the staircases lead nowhere, thevaults support nothing but their own weight and enclose vast spaces that arenever truly rooms, but only ante-rooms, lumber-rooms, vestibules, outhouses.And this magnificence of Cyclopean stone is everywhere made squalid bywooden ladders, by flimsy gangways and cat-walks. And the squalor is forsqualor’s sake, since all these rickety roads through space are manifestlywithout destination. Below them, on the floor, stand great machines incapableof doing anything in particular, and from the arches overhead hang ropes thatcarry nothing except a sickening suggestion of torture. Some of the Prisons arelighted only by narrow windows. Others are half open to the sky, with hintsof yet other vaults and walls in the distance. But even where the enclosureis more or less complete, Piranesi always contrives to give the impressionthat this colossal pointlessness goes on indefinitely, and is co-extensive withthe universe. Engaged in no recognisable activity, paying no attention to oneanother, a few small, faceless figures haunt the shadows. Their insignificantpresence merely emphasises the fact that there is nobody at home.

. . .

Considered from a purely formal standpoint, the Prisons are remarkableas being the nearest eighteenth century approach to abstract art. The rawmaterial of Piranesi’s designs consists of architectural forms; but, becausethe Prisons are images of confusion, because their essence is pointlessness,the combination of architectural forms never adds up to an architecturaldrawing, but remains a free design, untrammelled by any considerations ofutility or even possibility, and limited only by the necessity of evoking thegeneral idea of a building. In other words, Piranesi uses architectural formsto produce a series of beautifully intricate designs — designs which resemblethe abstractions of the Cubists in being composed of geometrical elements, butwhich have the advantage of combining pure geometry with enough subjectmatter, enough literature, to express more forcibly than a mere pattern can do,the obscure and terrible states of spiritual confusion and acedia.

(ii) Extract from Marshall Berman’s Notes from Underground: Plato’s Caves,Piranesi’s Prisons, and the Subway (2001).

8 S. Roncato

The underground world of Imaginary Prisons is dramatic, but it is an ominousspectacle. In these enormous rooms, it’s hard to find a stable place to orientourselves. We are overpowered by gigantic structures, but we can neither tracetheir foundations nor figure out their functions. Cables and stairways slashthrough the sky, but it’s difficult to say where they are going or why. If wemove our eyes only slightly, we are forced to shift our perspective radically,and we may find it hard to shift back. All the visual elements of this worldseem to clash and contradict. This must be the purest “negative space” everenvisioned. The light is weird: how can such vast darkness yield such clear,precise images of details?. . . The way structures are layered, cross-hatched,and intertwined in this underground world, the way light flashes from aboveonly to be swallowed up by the darkness again (this is especially true in the1761 second-state engravings), the absence of entrances or exits, the way theheaviest piers are piled on each other and seem to press us down — all thisgives this place an aura of disaster about to happen. Any minute, it seems, thewhole world could crash. . . Piranesi seems to gather up all the forms that menhave created to break out of Plato’s cave, but he then puts them together insuch a way that we seem to be trapped in that cave more deeply than ever.

And yet, somehow, those of us who love Imaginary Prisons don’t feel trapped.Instead, we experience this work as a parable of rising, of striving, ofovercoming, of breaking out of the cave and into the sunlight.

ARCHITECTURAL ATTRIBUTES OF PIRANESI’S WORK

The work of Piranesi has been subject of deep enquiry by specialists in historyof art and architecture (Tafuri, 1980; Wilton-Ely, 1978). Very recently, Kantor-Kazovsky (2006) published an extensive analysis of the whole body of Piranesi’swork, and to her I am indebted for the main suggestion to analyse in perceptualterms the 16 Carceri. In the analysis of Prima Parte di Architetture e Prospettiveshe claims: “Piranesi uses the principle of repetition of forms, sometimes rathermonotonous. While Juvarra’s aim was to combine many motifs in one frame,Piranesi concentrates on a single simple colonnade, pier or arch. His chosen motifmay be isolated or repeated several times in depth or in width. A slow developmentand display of similar forms, mostly on a horizontal or slightly diagonal plane,allows the viewer to get a full idea of individual elements and the ways of theircombination” (pp. 184–185).

In Piranesi’s etchings, a fantastic architecture combines naves, beams, arches,towers, halls to create an environment that stirs up a rich array of impressions andemotions. Exploration of some of the numerous web pages dedicated to the Carceriallow many of these comments to be gathered and classified. The most frequentattributes are the following:

(A) Depth or abyss (“dungeons, with us at the bottom, like inmates of hell’s deep-est level”, “underground world, enormous subterranean world”, “structureslayered, cross-hatched, and intertwined in this underground world”).

Piranesi and the infinite prisons 9

(B) Pointlessness (“galleries, ladders, bridges leading nowhere, endless, to dead-ends”, “the perfect pointlessness, emptiness of purpose”, “insignificant bridgesnonetheless, trapped for eternity”, “ interlocking mysterious compositions (flatplanes, deep diagonals), “endless stone stairways and arches”, “untraceablefoundations, inexplicable functions”, “bridges to nowhere extending to oneside of the structure to the next, but leading only to dead ends not doorways”,“bizarre galleries leading nowhere”).

(C) Cyclopic architecture (“architecture is colossal and magnificent”, “Massivearchitectural forms loom above darkly shadowed spaces”, “the immensityof the architecture seems to embody the workings of a great supernaturalpower”, “combining these imaginary buildings together as closely as possible,alternations of scale and juxtaposition of elements enhance the sense ofgrandeur”, “immense gloomy arcades” , “invention of extraordinarily spaciousbuildings such as baths, amphitheatres”).

(D) Voids (“immense arcades”, “vastness”, “vast spaces”, “opening up of spacetowards infinity”).

(E) Vertiginous walkways (“cavernous spaces traversed by vertiginous walk-ways”, “staircases rising to incredible heights”, “wooden ladders, vertiginouswalkways”, “stairways lead nowhere”).

(F) Dark and shadow contrasts (“dramatic contrasts between lit spaces and deepshadows”, “dark and shadowy prison with no exit”).

(G) Horror (“sinister and fantastic views of imaginary prisons”, “awful horrorsuggested by sinister machinery of cables, pulleys, and levers”, “cosmos ofterror and disintegration”, “iron rings”).

(H) Recession ad infinitum (“colossal pointlessness continuing indefinitely, co-extensive with the universe”, “sequences of vaults, arches and stairs recedingto infinity introduced beyond the arches and bridges by Piranesi”, “spectatorsdrawn into proliferating labyrinths of staircases, bridges, and passagewaysthat seem to lead into infinite depths left, right, and centre”).

(I) Insignificant people (“diminutive figures seemingly doomed to climb endlessstaircases without hope of arrival; human figures barely noticeable, tinyhumans travel these stairs”).

(L) Confinement (“absence of entrances or exits”, “bridges leading to dead-ends not doorways”, “fantastic immeasurable dungeons”, “limitlessness as aprison, with no escape”).

(M) Oppression (“apparent pressing down of the heaviest piers, piled on eachother”, “arches swinging above the observers heads”).

It is very difficult, perhaps impossible, to find a further example in history of anartist whose etchings stirred up such a richness of impressions and such deep andcomplex meanings. The leading motif might be summarised as an organisationof underground immense space with pointlessness but whatever the definition, the

10 S. Roncato

variety of impressions and phenomenological attributes inherent in the Carcerirender Piranesi a singular artistic figure indeed.

CUES TO MEANING

So what are the pictorial and architectural cues that produce such a rich set ofmeanings? The most frequent attributes cited by Carceri’s art critics are confinementand immensity. Piranesi represents these two attributes in vedute and in monumentetchings; the confinement is almost claustrophobic in etchings such as Antichitàd’Albano (water reservoir), the immensity of vaults is widespread in many etchings(Adriano bridge and mausoleum). Etchings and engravings do exist in whichdreariness and neglect emerge and pervade, but none has been taken as the symbolof fear, bewilderment and anxiety that the Carceri have become. Where, then, isthe source of the meanings? According to some essayists, it is the paradoxicalsimultaneous presence of confinement and immensity that disturbs. The immensevaults and naves are perceived as a wide space enveloping itself, a continuousexperience in which infinity and finiteness alternate and cancel each other out.

Arcades and vaults are almost ubiquitous in the 16 etchings, but it is difficultto assign to these architectural forms — despite their specific representations —the source of the attributes. If we look at Raphael’s School of Athens nothinganalogous to the oppression of Piranesi’s halls can be found in the perspective ofthe vaults; indeed, there is quite the opposite impression, one of openness. Theweird combination of confinement and immensity itself does not generate anxiety.In Figs 2 and 3, two examples demonstrate this combination: the immensity of thespaces and the ‘shell’ that encloses them are both striking: the impressions givenby the vastness of the spaces and of their enclosure are equally powerful, this latterheightened by the light that appears or filters through from openings high up. Buteven so, the impact of this impression is very much less than the sense of uneaseand claustrophobia evoked by the Carceri etchings.

Walls and arches — made up of colossal blocks often arranged irregularly —may evoke feelings of being hemmed in, of inescapable barriers. Kantor-Kazowsky(2006) claims that the Tuscan way of building fortifications with ‘huge irregularblocks of stones’ arises out of a reaction of terror seen in Renaissance architecturalthought (Leon Battista Alberti). A wall of massive blocks is undoubtedly impressivebut the sinister spell of the Carceri is so pervasive throughout the 16 etchings that itcannot derive from the ‘texture’ of a simple wall.

Some researchers shift attention from the architectural forms or details them-selves, and onto the layout, which has bewildering characteristics. According toHuyssens (2006), “Instead of viewing limited spaces from a fixed-observer perspec-tive and from a safe distance, the spectator is drawn into a proliferating labyrinthof staircases, bridges, and passageways that seem to lead into infinite depths left,right, and centre. It is as if the spectator’s gaze is imprisoned by the represented

Piranesi and the infinite prisons 11

Figure 2. Piranesi, Giovanni Battista. Veduta interna dell’antico Tempio di Bacco, oggi Chiesa diSanta Costanza [interior of the ancient temple of Bacchus]. Original etching from the “Vedute diRoma”.

Figure 3. Piranesi, Giovanni Battista. Veduta interna della Chiesa della Madonna degli Angioli dettadella Certosa che anticamente era la principale Sala delle Terme di Diocleziano. [S. Maria degliAngeli. Interior. Formerly the central hall of the Baths of Diocletian]. Rome. 1776. Original etchingfrom the “Vedute di Roma”.

12 S. Roncato

space, lured in and captured because no firm point of view can be had as the eyewanders around in this labyrinth” (p. 18).

In the search for a typical feature of Piranesi’s Carceri, we cannot concentrateon the labyrinthine layout represented. Some images and representations aremore labyrinthine than others, and a source of wandering — technical drawings,urban landscapes — but they have not the pregnancy of meaning of the Carceri.Conversely, we may register a sensation of wandering in an environment with quitethe opposite characteristics of the Carceri, for example a modern building with glasswalls and bright intersecting corridors.

The conclusion is that the acute sense of oddity that the Carceri convey hasmore than one origin. Not only does one perceive a labyrinth, but also a seriesof dissonances, incongruities, a scene with a fleeting spatial organisation. Thereare no absolute reference points or directions for ‘inside/outside’, ‘above/below’, sotheir order reverses as scene inspection proceeds. The axes and directions of spatialorganisation are not steady; Piranesi reaches this result principally by violating twomain constraints, namely dimensional coherence and the unit of perceived space.The complex of these incoherencies is sometimes associated with a perspectivewhich views from below, generating a particular sensation of objects floating. Letme illustrate this violation of constraints, and the sensation of floating.

VIOLATION OF CONSTRAINTS

Dimensional and topological

Figure 1a (Plate III, ‘The tower’) is a good example of the perceived effects ofviolation of architectural units proportions.

At the bottom of the figure, staircases lead to the base of a bastion; from itstop, traversed by a terrace, the arches of a nave lead out. A bastion is in factpart of a fortification that projects outwards from the curtain or wall corner. Inthe etching, it plays the role of a column that supports arches forming an immensenave. The bastion is simultaneously the outer side of part of a building, and thestatic component of an inner space. The ‘exterior’ attribute of the bastion wall isenhanced by the bridge that joins it through an appropriate wall.

The various architectural forms we have analysed here — stairs, grating, stones,arches, terraces — and the small figures do not have compatible (or consistent)dimensions. Note that the bastion has the dimensions of a tower when referred tothe stairs, but the stones and grating are out of proportion with respect to the stepsthemselves and the figures. At the bastion top, a terrace replaces the capital, againlarger and out of proportion. The arches departing from the top are as large as theterrace.

What the observer can view resembles the illusory Ames room, where dimen-sional incongruences are distributed in both depth and vertical direction.

Note that three scenes are piled up on top of each other: staircases, bastion,arches and naves. They preserve the perceptual sizes they have from different

Piranesi and the infinite prisons 13

distances and are combined ‘pictorially’. The staircases and bridge are drawn asthough perceived from further off compared with the bastion, as are the arches. Thephenomenological effects are clear: the tower appears magnified. Dürer in his ‘Viewof Arco’ (1495), in which a giant face is carved out in the rock, can be considereda precursor of these effects. Vogt-Göknil (1958) has observed that in some plates,the dimensional spaces evolve into two-dimensional planes; dimensions of depthare pulled apart; and dimensions of breadth are being made to shrink (Escher’sengravings, Carceri, Plate XIV).

Space fragmentation

What we perceive is the result of a series of processes by which fragments areintegrated into wholes. Psychologists have singled out and analysed several aspectsand manifestations of these completion processes, one of the best known beingamodal completion. These processes are harder to explore when complex visualscenes are considered. The Carceri offer a singular chance to examine theseaspects of visual integration or completion. What we experience is the imaginingof a fragmented world; what is missing is not the unity of an object, but ratherthe unity of space and the coherence of the parts in which it articulates. Thevarious architectural elements are pictorially joined with great skill to guaranteespatial and constructive continuity. But the final result appears phenomenally odd,sometimes at local level, often at global level. A possible source for the perceptionof incongruence at the local level has been considered in the section above onincoherence. Here I aim to show how spatial integration can be thwarted, or lead toillogical solutions.

The lack of spatial unity of the Carceri scenarios can be likened to the effect wecan observe when we reproduce polyhedral shapes on a plane surface. The threepolygons in Fig. 4a are perceived as a block (parallelepiped) (or more precisely‘interpreted’, since they can as the interior of an open block). If we draw patternsof this type in conjunction with a lateral edge, we produce various differentphenomenological outcomes. In Fig. 4b, two blocks contact along an end edge,while in Fig. 4c they have a common side edge. In this latter case, the contemporarypresence of the two blocks in the phenomenological world seems incompatible. Thedrawing appears unstable: if we perceive one parallelepiped, the other disappearsand vice versa, and it is certainly impossible to perceive a unitary volume.

Over the last century, many artists have dicovered the fascination of the ‘chimeric’shapes produced by particular combinations of surfaces and volumes reproduced intwo-dimensional space: shapes that are complete only in part, are unstable, andpossess elements that are impossible to fit together to form a real three-dimensionalobject.

These curious effects have been widely illustrated in the drawings of the Swedishpainter Reutervärd, then by Dutch engraver M. C., Escher and English mathemati-cian Roger Penrose. In Fig. 4d, three parallelepipeds reproduce the basic pattern

14 S. Roncato

Figure 4. (a) parallelepiped. (b) and (c): the two volumes have opposite orientation in depthand appear as ‘impossible pairings’. (d) Volumes drawn following the Reutervärd rules, the wholepattern appears as an impossible figure. (e) An impossible figure that is not perceived as such, theimpossibility becomes apparent by exploring the block surfaces (f).

with which Reutervärd drew the first impossible figures of cubes. If the three paral-lelepipeds are brought together so as to touch, it creates the Penrose triangle.

The three shapes in Fig. 4d appear impossible to fit together to form a wholeobject. The instability of the percept here consists in the impossibility of connectingall the shape at once, since one of the three areas of connection has a block thatuncouples; if we try perceptually to hook up the ends of the blocks, the couplingsjust made vanish, and so on. In this case, the resistance to a perceptual integration ofthe three blocks reflects the impossibility of physically (geometrically) connectingthem in the direction illustrated in the drawing. Nevertheless, the physical-geometrical constraints do not seem inviolable for the perceptual system. Figure 4eillustrates a case of physical impossibility that is not perceived as such, unless theobserver is required to explore the world defined by the blocks. An observer askedto imagine walking along the direction of the arrows in Fig. 4f will notice at the endpoint that he is close to one extremity of the left-hand vertical parallelepiped butabove the start point, despite the fact that he was walking on a plane surface.

In this case, the spatial exploration refutes the phenomenal percept, an eventneither rare nor exceptional in daily life or experimental enquiry. The strangeness isenhanced when the paradox is made directly visible. Escher illustrated this paradoxin a startling way in his engravings entitled: Belvedere, Ascending-Descending andWaterfall.

Piranesi and the infinite prisons 15

The scenarios created by Escher have spatial arrangements that reproduce thebasic organisations illustrated in Figs 4c and 4d; their illogicity is made evident byimages of people walking or climbing and by water-flow from which the observersees that the motion is impossible because the space extension in height and depthis illusory, or else through a drawing where distance between surfaces is altered.

The space in the Piranesi etchings also presents serious difficulties to an observerwho imagines walking within them. An interesting essay by Stoppani (2006) offersvaluable suggestions for analysing the ‘movement’ in the Carceri (“The nature ofspace is . . . determined by the movement that operates in it”). Nevertheless, thereasons for which this space is difficult or impossible to explore are richer and morecomplex than just those taken into consideration in polyhedral volumes.

Let me illustrate two conditions of difficulty of local scene integration.In Fig. 5, I have drawn an example of a possible solution to a process aimed

at integrating a cube and a staircase. The stairs lead to a level below that of thesurface with sides A and B; if the phenomenal field were a reproduction of thesegeometrical indications, we would be induced to perceiving a cube without a baseand a staircase suspended in empty space. This solution is discarded in favour ofa perceptually simpler one: the staircase begins at a higher level than A and leadsdown to a solid surface. The solution is illogical, but the visual system prefers itbecause it is ‘less disturbing’ (the viewer has a weak impression of space torsion).

There can also be volumes that ‘resist’ the process of integration. In Fig. 6, aperspective view depicts a space divided into three adjacent cubes; two walls areinterrupted by corridor openings. At local level, the solutions are easy becausethe depth cues are clear. Globally, the integration of the volumes — of which a

Figure 5. A staircase suspended in empty space or leading to the A × B floor?

16 S. Roncato

Figure 6. Intersecting corridors (see text).

partial view is drawn — is difficult. Integration can occur along two different spatialdirections, as indicated by arrows. Along the first, surface completion leads to theperception of cubes (or three rooms); in the perpendicular direction, completiongives rise to corridors through the rooms. The corridors and rooms are interlacedbut the solution is not easy to perceive for observers unable to complete the volumesof corridor and rooms at the same time. They appear to interfere and cannot beintegrated into a unitary volume.

These two examples above illustrate possible sources of difficulty in achievinga perception of a whole congruent space in the Piranesi Carceri. Furthermore, itshould be noted that the scene gives no clear indication of how to explore its spaces;rather, it seems to impede exploration or take the explorer off route. The peoplescattered throughout the etchings do not allow the observer to understand where theycome from, nor do they demonstrate intentionality or destination points. Piranesidraws all types of devices — stairs, passages, ropes, chains — inventing connectionand interlacement of lines, surfaces and volumes, to complicate the perceived space.In summary, any attempt to come through the space of the Carceri is frustrated bya double failure: impossibility of creating for oneself a plane of exploration, andimpossibility of inferring, from the motions represented, a direction to proceed oreven an auxiliary device such as Ariadne’s thread.

The Carceri are a collection of challenges for the visual system, which has tosolve problems of surface and space completion at different levels. The scenariosare made up of intersecting solid and empty volumes, and concave and convexsurfaces, in a variety of ways that severely hinder the search for basic patterns ofspatial incongruity as can be carried out in the Escher engravings.

The perceptual solution is unsteady and illogical, like an unconvincing solutionof a problem. Perhaps the true fascination of the Carceri lies in the uncertainty ofthe perceptual results, that is, in the elusiveness of many of its contradictions thatsolicit a continuous revision of the hypothesis of their odd world.

Piranesi and the infinite prisons 17

Figure 7. The same pattern from two different perspectives.

Floating patterns

In Fig. 7, the same series of parallelepipeds has been reproduced from a top anda bottom perspective. Note the difference of phenomenological impression. Fromthe top, the complex is steady; viewing from below gives a vivid impression thatit is floating and dynamic. Massive form, vastness and perspective from below: allthese attributes contribute to generating a perception of unstable equilibrium, easy todisrupt and collapse on the observer. Piranesi uses this device directly by drawingsuspended wooden passages, but the interlacing arches and bridges viewed frombelow contribute to giving the overall scene the appearance of a floating building.The bottom perspective is not Piranesi’s usual point of view, according to Vogt-Göknil (cited by Kantor-Kazovsky, 2006): the Carceri represent an exception thatdemonstrates how impressive the objects may become when depicted from thisunusual direction.

This brief overview clearly demonstrates that Piranesi’s Carceri constitute verysuitable material for research into vision and scene perception, providing a uniqueopportunity for study. Their variety might indeed form a fertile basis for an all-round investigation of the procedures by which the visual system achieves itscoherent representation of the world in its complex manifestations. Figuratively andotherwise, Piranesi certainly challenges us with problem-solving at many levels.

REFERENCES

Berman, M. (2001). Notes from underground: Plato’s caves, Piranesi’s prisons, and the subway,Harvard Design Magazine 15.

Eisenstein, S. M. (1997). Piranesi, or the fluidity of forms, (Translated by Roberta Reeder),Oppositions 11, 84–110.

Huxley, A. (1949). Prisons. With the ‘Carceri’ Etchings of G. B. Piranesi. Zeitlin and Ver Brugge,Los Angeles, USA.

Huyssens, A., (2006). Nostalgia for ruins, Grey Room 23, 6–21.Kantor-Kazovsky L. (2006). Piranesi as Interpreter of Roman Architecture and the Origins of his

Intellectual World. Olschki, Firenze, Italy.Nesbet, A. (2006). ‘The building to be built’: Gogol, Belyi, Eisenstein, and the architecture of the

future, The Russian Review 65, 491–511.

18 S. Roncato

Stoppani, T. (2006). Voyaging in Piranesi’s space: a contemporary re-reading of the beginnings ofmodernity, Haecceity Papers 1, 32–54.

Tafuri, M. (1980). La sfera e il labirinto. Avanguardie e architettura da Piranesi agli anni ’70. Einaudi,Torino, Italy. English Translation (1987): The Sphere and the Labyrinth: Avant-Gardes andArchitecture from Piranesi to the 1970’s. MIT Press, Cambridge, USA.

Vogt-Göknil, U. (1958). Giovanni Battista Piranesi: ‘Carceri’. Origo Verlag, Zurich, Switzerland.Wendorf, R. (2001). Piranesi’s double ruin, Eighteenth-Century Studies 34, 161–180.Wilton-Ely, J. (1978). The Art and the Mind of Giovanni Battista Piranesi. Thames and Hudson,

London, UK.Yourcenar, M. (1984). The Dark Brain of Piranesi and Other Essays (translated by Richard Howard).

Farrar, Straus, and Giroux, New York, USA.

Web sites reporting the Carceri d’Invenzionehttp://cgfa.sunsite.dk/piranesi/index.htmlhttp://rubens.anu.edu.au/htdocs/bytype/prints/piranesi/display00052.html