JENNIFER NEWSOM + TOM CARRUTHERS

SCANPATHS RECALLING MEMORY

Scanpaths recalling memory
Motion paths bearing an image
A re-presentation of Power in Renaissance Gardens


Professor Bryan Fuermann
History of Landscape Architecture
Yale School of Architecture
December 26, 2003
Thomas B. Carruthers


This paper initially proposed to investigate the relationship between conventions of pictorial representation and the spatial organizations of renaissance garden design.  There is an idea that Perspective was seen as symbolic system of organization that would function in both two and three dimensions.  In two dimensions, the relationship between the viewer’s eye and the object of view mathematically arrayed the peripheral objects upon a picture plane placed between the two.  Three dimensional garden design, beginning with Bramante’s Belvedere Court deployed perspective as a symbolic and totalizing organizational system.  Subject and Object were specified in space along an axis of view that represented a clear delineation of power.  Just as painted perspective occurred as a limited visual phenomenon within a frame so too was three dimensional perspectival spatial organization framed in plan.  These gardens were understood to be limited experiments in the establishment of logical social hierarchies.

The reason the representations of these gardens share certain characteristics such as the elevated axial view is to make clear that this type of organization represents a hierarchy of power.  The importance of the creation of these gestalt views must have been the felt need to have a source of power communicated clearly and immediately. One’s ambulatory experience of such a garden would be understood to be always functioning with respect to that gestalt view.  The reason the stroller, the walking subject of the garden, is never allowed to travel along that central axis, excepting certain moments, is precisely that they would occupy a position of gestalt power and secondly, they would interrupt the core relationship between the subject and the object.  Water flows down the axis to provide discrete objects of pleasure, eye-catchers, and to express the flow of power from on high to Earth.

In light of scientific research into the physiological programming of human experience and the creation of memory it is of great interest now to look again at the renaissance garden as a representation of order because of its reliance upon pictorial conventions of composition.  The paper by Weber, Choi and Stark suggests that one experiences a 2d or 3d scene the same way.  The eye of a subject will behave reliably in certain ways and will be attracted by certain effects.  These result in scanpaths, “a repetitive sequence of saccades and fixations, idiosyncratic to the viewer and to the picture.”  One would not experience these gardens in a gestalt fashion .  Instead there may be an analogy to be made between the scanpath of the eye and the ambulatory path of the stroller in the garden.  Where these two would coincide is in the creation of memory.  

Storing and retrieving memory are important components of visual learning and pattern recognition.  Hence, the memory system of the brain must contain an internal representation of each pattern that is to be recognized.  Familiarizing oneself with a pattern maybe considered as the process of constructing such representations.  Conversely, recognition of a pattern may be viewed as the process of matching it with its stored internal representation.  Such a non-Gestalt view suggests that the internal model consists of component features which are matched step-by-step with the pattern during recognition.  This serial recognition process is supported by several studies on object and pattern recognition that the eyes seem to visit the features of the objects or items in the scene cyclically, following stereotyped regular scanpaths or search paths rather than random sequences. . . object representations are composed of sensory memory traces recording the features and motor memory traces of the eye movements from one feature to the other.

Just as the eye uses these motor memory traces to recreate a memory so to could a stroller use the pattern of his ambulatory experience to recreate, in sequence, the Gestalt reading of the garden space.   Just as a memory is first created and later relived through the patterned totality of sensory and motor inputs from the eyes, so to does the stroller by experiencing the garden re-inscribe that hierarchical organization back upon themselves.  In this sense then, there is a harmony between perception and reality.  The gestalt view of a garden can be read as a memory, that is recreated by an idiosyncratic series of movements.  This is a potentially powerful idea, that the representation of power functions in analog to the representation of memory.  This makes a good deal of sense if one believes that the renaissance garden was a repository of social memory, of mankind’s place within theological, cosmological, historical and intellectual modes of thought.  That is to say that while we are hard-wired to experience and record our world in certain sequences and orders we are capable of recreating from those core processes the necessary concepts to communicate something else just as clearly.  At a fundamental level then, the Italian renaissance garden  is an architectural extension and growth of these native concepts.

This paper will develop this concept by looking at the way the eye creates these scanpaths before looking at how paths, axes, subjects, objects and viewers operate within gardens.  Given the interest of renaissance architects in the Roman antique this paper will set a baseline by analyzing the Houses of Jason and Tibertimus in Pompei.  The purpose of this will be to demonstrate two types of axial design: one where the visitor can occupy the axis and one where the viewer walks alongside the axis.  The intent is to establish the Roman understanding of pictorial space and how that related to the larger organization of their architectural space.  Following this, the paper’s attention moves to both Bramante’s Belvedere Court, and Vignola’s Villa Lante.  The effort here will be to demonstrate the points of similarity and difference between Roman and Renaissance Garden space.  The Villa Giulia will be considered last as a further development of the tendency of perspectival/axial constructions towards scenography.  The Renaissance analysis will differ because there are also pictorial representations of the gardens, not merely ruins to look at.  Thus there is a great interest in the reciprocity of information contained within these two cultural modes of the organization of symbols.   

In the tradition of Uvedale Price, John Barrel introduces “landscape” as a term that introduces, “whether we want [it] to or not, notions of value and form which relate, not just to seeing the land, but to seeing it in a certain way –pictorially.” (Barrell 1)  Interestingly, Weber, Choi and Stark state that, “in relation to the effect of two- versus three-dimensional representations of the same composition of architectural forms and spaces, little difference seemed to appear in the primary-fixation zones between the three-dimensional models and their two-dimensional images.” (Weber 66)  This understanding of landscape, as a phenomenon that has inherently pictorial, or two-dimensional compositional qualities is given scientific grounding then by a scientific study.  In fact, what the study believes, is that an eye will move over a picture of a scene just as it would move over the scene itself.  

Both authors discuss the specific nature of this eye-movement further.  Weber, Choi and Stark believe that the eye’s movement, “does not trace shapes completely, but focuses on the overall arrangement of visual centers, major masses and on objects with distinct fomal differences from the overall set.”  However, the eye will specially favour, “elements indicating spatial depth, such as vistas. . . the left area of a space. . . redundant elements draw less attention than solitary shapes,” and, “obliquely oriented shapes.” (Weber 57)  In light of this information regarding how an eye moves and what it prefers to focus on let us look at Barrell’s analysis.  “A landscape by Claude employs, in the first place, a fairly high viewpoint –high enough, that is for a distant horizon to appear above any rising ground between it and the viewpoint: and the first impression which everyone must receive, I imagine. . . is one of tremendous depth.”  “The eye, attracted by an area of light usually set just below the horizon, travels immediately towards it over a long and often steeply contoured stretch of intervening land. . .The initial movement in all Claude’s landscapes is this one, from the foreground straight to the far distance. . . a rapid movement. . . an immediate response to the way the picture is organized.” (Barrell 8)  Barrell’s argument here is that the spectator’s pleasure in viewing a Claude is, once the eye has been immediately drawn to the far distance, figuring out how it was done!  Barrell enumerates a number of devices: an alternation of light and dark planes that culminate in the highlighted eyecatcher discussed above; a darker third plane overshadowed by trees and the use of foreground elements to frame the view to the distance.  The net result for both artist and viewer is the development of, “a specialized vocabulary, and a grammar, as it were, of landscape patterns and structures. . . A landscape was integrated into the established set of landscape patterns, and so became part of the universal landscape, which included any tract of land the connoisseur chose to examine.” (Barrell 8)  Claude created a method of visual organization that so accurately used the eye that he created not only a pattern whereby people saw landscape, but is also a visual pattern where by people created and recalled memories of landscape.

It is possible now to take a leap and say that the composition of a picture is carefully calibrated to excite the eye in specific ways.  That in fact that composition is pleasing because there is a resonance between the eye’s natural desires and the artist’s control of the sequence of their satisfaction.  However, a crucial distinction between the picture and strolling within the Renaissance landscapes is precisely the nature of that movement within when compared to the elevated gestalt image of its design.  Do two dimensional compositional elements translate into three dimensional experiences?  

Pliny is quoted in the book Landscape Design as saying that, “he could gaze at an

Immense amphitheater which only nature could create. . . The broad plain is ringed by mountains on the crown of which are ancient stands of tall trees, and various kinds of hunting may be found thre. . .  It is a volulptuous experience to look down on this scene from the mountain. You seem to be seeing not real land but rather a painted scene of exceptional beauty, and wherever the eye runs it is refreshed by its variety and precision. 

(Moore et al 88)

The Romans had the ability to conceptualize a view of terrain in terms of pictorial conventions.  While we today would understand this phenomenon as a transposition of a set of pictorial conventions applied as a means of perceiving terrain I suspect that the two were one and the same.  In this sense, the transposition at work was spatial.  The concept of rus in urbe was a common means whereby city-dwellers re-presented their understanding of the idealized landscape with-out the city.  The design of urban Roman houses demonstrated how, “in some cases, an axial view from the atrium into the peristyle garden was prolonged by trompe-l’oeil paintings on a rear wall.  Attached columns framing these delightful garden scenes made them appear to be natural views seen from a portico.”  (Moore et al 85)  This use of axial view and illusionary prospect is a dramatic counterpoint to the shaping of the inward-turning Roman house, “around ritual activities and the concomitant use of axial composition and symmetry to enforce notions of order stability and place.” (Moore et al 85)

The House of Jason in Pompei, in Bettina Bergmann’s account, makes use of an inhabitable axial composition.  “Immediately upon coming into the house, one could have turned left and, with doors opened, looked through a succession of framing doorways to glimpse a seated female figure painted on the far back wall.  Moving through the antechamber, the viewer would see a further figure to her left. . . Upon reaching the threshold, an older boy kneeling on the ground.” (Bergman 202)  This room, this eye-catcher, in fact held three paintings, 

their compositions clearly present them as analogous.  In each, a dominant three-figure group with a child is disposed at different levels by sitting, standing and kneeling positions, creating a dynamic play of diagonals within  the square frame and against a grid of columns, pillars, partitions and doors.  This formal unity engages the spectator in a coherent visual rhyming scheme, and that scheme expresses a conceptual unity. . . Each picture was originally “enframed” illusionistically by an elaborate aedicule with a cassette ceiling shown in perspective, so that the architecture within the picture was recessed several planes back from the wall, thereby multiplying the planes between the viewer and the represented scene.  This layering of frames and deepening of space, first experienced physically in the actual doorways leading to the room, ends visually in the depicted columns and doors behind the figures in the panels.  In this way, fictive space became continuous with actual space. . .The most effective coordinating device in the room was the perspectives within the panels, which were adjusted to the oblique viewpoint of the person stopping at the threshold. (Bergmann 207)

So here you have an axial composition within a domestic household.  “All three panels present the key juncture in a plot when a wife is about to uproot herself as keeper of the house and destroy the pillars of patriarchy.” (Bergman 207)  What impressed me was the extent to which both the two and three dimensional elements of the scheme are malleably altered in terms of their organization in order to represent the synergy between Medea, Phaedra and Helen of Troy to a moving subject in space.   Perspectives, layers and space are deployed as a series of veils in order to blur the distinctions between fiction and reality.  However, while the axis guides the viewer to a progressively dislocated scene it is not activated symbolically beyond being a means of guidance.

The concept of rus in urbe extended during Hellenistic times to the expansion of the peristyle garden.  By the time of Augustus, during whose reign both of these houses were built, the garden of Tiburtinus had expanded to occupy the length of a Pompeian block.  A euripus with pergolas on either side connected a gate leading to the amphitheater through a series of three fountains, a nymphaeum and ultimately to the frescoed triclinium of the house.  In front of the Triclinium and running perpendicular to the main axis was a cross-axis connecting a biclinium with zoothecula.  This organization symbolizes these water filled axes as main views, for viewing and not for traffic.  The cross-axis links a space of eating and service to one of repose.  The main euripus channel, perhaps more interestingly, uses manipulations of perspective to place and dimension fountains in order to visually extend the garden further than it really did.  Here the triclinium is the location of the subject, while the fountains and gate present a series of objects, or eyecatchers receding in space.  One can see clearly here the organization of a space according to the pictorial concerns of a single location along with the symbolic isolation of the main axes of view.  Such a display of organization clearly represents not only the ability of the owner to construct such a garden but also the symbolization of the power of that owner’s eye.  

The spatial world consists not of instantaneously created units, but of processes of becoming, indefatigable transformations of spatial configurations.  Nature forms; lowers, trees, rocks, mountains, cloud formations, animal or human bodies as well as man made forms; buildings or implements, are only temporarily configurations in the perpetual flux of becoming and disappearing.  (Kepes 186)

Nature, whether man-made or imitated by man, persists through change as the inevitable backdrop for the representations of landscape architecture.  It is precisely due to this flux that the re-presentation of power in garden form makes sense.  The reason for this focus is best stated by Moore et al, “because the Renaissance was initiated and carried forward by a small elite of royal and noble patrons influenced by ideas fostered by scholars and expressed by artists, our discussion is necessarily focused upon power centers.” (126) These scholars and artists were focused upon forming a humanist project in contrast to the medieval one preceding it.  At the core of the humanist project was a redefinition of mankind’s place as the intermediary between God and Nature.  This resulted in the view that, “the Renaissance universe is hierarchical, with God at the summit, human beings in the center, nature below, and each part related to the other. . .” (Lazzaro 8)  

However, the gardens had a more specific linkage to the economic source of power of the royals and nobles.  Mirka Beneš asks the question, “since pastoral art has always been the invention of city dwellers, not of the herders themselves. . . : How . . . do urban societies that are pastoralist in their economic base relate to the pastoral art that is invented for them?” (Beneš 91)  Beneš continues in her argument to identify two transformations that justify the transpositions involved in the making of Gardens.  Firstly, she notes the sale of Palestrina in 1629 to the Barberini family from the Collonna family as indicating a larger shift in status and money from old families to new.  Perhaps driving the first transformation is the shift from an agricultural to a pastoral economy.  In effect the power base of the Renaissance movement was relatively new to power.  They nonetheless sought to define their power in the terms of the baronial families from whom they succeeded.  Stability, order, and prestige remained defined largely in the medieval terms of fiefs and farms. “The prints. . . tell us what mattered to the villa owners. . . [they] allowed a group of elites to display both material and cultural wealth while simultaneously rationalizing power and privilege at a time of social and economic destabilization.” (Harris 179)  The concept of landscape is embodied in a garden as the representation not only of a humanist enterprise but more fundamentally as a sign of an economic source of wealth.  In this logic gardens have,

a double message about power.  First they bespoke the power of the owner, who could afford (like Nero in the Golden House) to have large rural lands close to the city center.  Second, they spoke of the string of landholdings of the owner farther away from Rome, in the country side: These transpositions, materialized in the forms of the villa gardens, were like a synthetic representation of these countryside properties –farms and fiefs –placed right at the exit point of the city, in the first ring of vineyards, where the journey to these properties began. (Beneš 106)

Hand in hand with economic and political super-positions, are likewise a series of representational juxtapositions that culminate in the overall organization of a renaissance garden.

Claudia Lazzaro addresses the recurrent theme of Art and Nature.  If one considers Art to provide organization and nature the elements what is produced when art imitates nature?  It is not unspoilt nature, nor the 2nd Nature of Palladio and le Corbusier which treats the land as man-made, a tabula-rasa.  Because the Renaissance garden seeks to represent the larger hierarchy, from macrocosm to microcosm and does so in the belief that art is therefore imitating nature, the Renaissance Garden is a third nature.  Lazzaro describes the obsessive nature involved in closely imitating nature.  She quotes, “Vincenzo Giustiniani’s thoughtful account of the requirements for creating and ornamenting gardens . .  . he explains the necessity of naming all the parts of a large garden in order to be able to discuss it, “It will be necessary to give as specific a name as possible to everything. . . That is so that. . . everyone understands perfectly which part is meant; otherwise, there will always be confusion.” (Lazzaro 72)  The function here is that organization is understood as an assemblage of parts.  In this sense, organizing nature was considered to be an ontological exercise in the clear separation and definition of terms.  

This process of division and isolation is evident, for example in the Villa Lante, where nature is first divided into garden –known nature- and hunting park –unknown nature-.  Known nature is further divided in section into three major terraces, each terrace is sub-divided into a layout of compartments or geometric figures. The process continues into smaller units as the detailed etching by Tarquinio Ligustri  of the Villa Lante demonstrates.  What unifies these sub-divisions is the sequential experience of the main axis.  To quote Lazzaro, 

In gardens from the fifteenth through the sixteenth century two primary sets of concerns can be distinguished: units, compartments, separate parts that can be named, measured, and counted, enclosed and hidden spaces, and sequential experiences on the one hand; and on the other the linking of parts, axial organization, directional impulses, vistas, and unity. 

(Lazzaro 70)

So there is a balancing act between two functionally opposite practices:  One being to further objectify each part geometrically and the other to express a directional unity. 

In Renaissance Gardens the effect of this organization was captured in the plan, however the experience was felt in the elevation and the section.  The subdivisions of the garden created major rhythms of experience that operated at different tempos.  For example, the use of multiple terraces could be considered one tempo while one’s experience of the compartments within a terrace would be considered a more rapid tempo.  At the Villa Medici at Fiesole, “The lower garden is, however, cut into the hillside and is sufficiently lower than the upper garden so as not to intrude upon the panoramic view from the upper terrace.” (Moore et al 130)  This particular meter of subdivision is particularly interesting to trace in its trajectory from a sectional device that isolates one visually to that terrace to a scenographic device that locates a sequence of views along an axis, without the section as at the Villa Giulia.  

This, “essentially geometric handling of the landscape imposed on it a strong sense of order.  This made the visitor entering the garden feel that he was in complete possession of its design, for no matter how vast, the garden read as a visually finite space.” (Hazelhurst 373)  By adding the main axis, the movements of the stroller would always be relative to a larger direction of flow that indicated a rhythm of experience.  This, “rhythm cannot be grasped as one isolated visual sensation.  Its very meaning lies in the fact that it is an order of a greater temporal whole.” (Kepes 53)  

This whole, 

a garden, like a theater, machine, or city, was used as a metaphor for a closed but encyclopedic system.  Recently some scholars have suggested that there is a relationship between the garden and memory systems. . . However, the similarity between the two structures lies only in the fact that the garden was conceived as a finite inventory of the natural world to which another significance could be attached.” (Lazzaro 16) 

This other significance is the symbolism of perspective.  The signification of the eye, be it from the papal apartments, the windows of Villa d’Este or merely an elevated view is effectively becomes a referent line that unifies these compendia of nature.

Gyorgy Kepes’ definition of perspective on page 86 of his book focuses upon the tendency of perspective to destroy time and thus the wealth of visual experience.  In looking at the Villas Lante and d’Este it is interesting to see that it is those perspectival axes of water and power that have in fact withstood time while the plantings have obviously moved on.  However, while it will be shown that Perspective was used in the design of Renaissance gardens it did not serve to deny the temporal experience by a visitor of the space.

At the Belvedere Court Bramante deployed perspective in space to create an axis between the Pope’s apartments and the elevated niche-a mask for the cross axis of the Belvedere villa.  Perspectival convergence was heightened by the placement of two loggias on the second terrace.  From the Pope’s elevated point of view this created three stages for action beneath the main axis.  However, the stroller’s experience would be quite different.  The terracing of the site, like at Fiesole, created three distinct platforms of visual activity.  As one processed from the lowest court to the highest up the axially placed staircases one would maintain a mental image, perhaps the view from the Pope’s windows, of the overall organization of space.

Representations of the Villa Lante, the Fresco of 1574-78 and the engraving mentioned above of 1596 demonstrate two dislocations from Bramante’s work.

Firstly, the viewer is given an abstract, elevated point of view instead of grounded one of the Papal apartments.  Secondly, the main axis is no longer given over to circulation but is instead a singularly long fountain where Bramante’s excedral staircase becomes a fountain of lights.  Here, at the Villa Lante that axis is re-conceived as only being symbolic of power.  The placement of the Fresco, inside one of the two pallazine meant that upon entering the garden from the town below one was given the means by which to comprehend the total organization as one traveled through it.  By representing the garden with an abstract eye and then placing that representation near to the entrance of the garden one would be constantly occupying both the garden and that image.  In effect, the fresco is a kind of gestalt image or memory that one would carry with themselves as they travel alongside and perpendicular to the main axis.

At the Villa Giulia the main axis, like that of a Roman atrium house, is occupied upon entrance by a visitor to the villa.  The eye is pulled through a series of loggias towards a small figurative sculpture at the very back wall that looks back at you.  However, one cannot move along or even beside that visual route.  The villa is conceived as a series of overlapping scenes with interstitial spaces that are not introduced or anticipated!  Here the perspectival axis merely describes the potential for the sequential unfolding of spaces without the presentation of a gestalt understanding in advance.  The terraces, initially re-created by Bramante to mount the hill and isolate portions of the plan from eachother, reoccur at the Villa Giulia as something different.  With no image at hand, the axis itself becomes the only point of reference for a moving visitor.  By putting the visitor in the position of power regarding the main axis Vignola et al create a new kind of pictorial representation that can only be played out in three dimensions: the scenography.

The two Roman houses and three Renaissance Villas discussed demonstrate the use of perspectival systems of vision in the organization of three-dimensional space.  In each it is interesting to see where the subject of that main terminus occurs.  As at the House of Tiburtinus and the Villa Lante, the subject moving in the space is displaced from an axis of water.  At the Belvedere Court the Pope may be the subject of that perspectival axis but he may not travel along it simultaneously.  At the House of Jason and the Villa Giulia the space is calibrated for a subject that simultaneously occupies and moves along the organizing axis.  However, in all these houses it is surprising how a method for creating proportional geometric relationships on a canvas translate into the deployment of elements in three-dimensions.  Just as the representations of these spaces are calibrated to order the progression of the eye through the depth of the painting so to are these gardens calibrated to order the progression of a visitor through their actual depth.  It appears then, that in the Renaissance tradition of the continuity from microcosm to macrocosm, that the sequences by which memory is created in two dimensions likewise operate in three.  To quote Vincent Scully, “it is therefore an architecture that is intended to enclose and shelter human beings in a psychic sense, to order them absolutely so that they can always find a known conclusion at the end of any journey, but finally to let them play at freedom and action all the while.” (Bois 44)


Works Cited

Barrell, John The Idea of Landscape and the Sense of Place United Kingdom: Cambridge University Press, 1972.

Beneš, Mirka “Pastoralism in the Roman Baroque Villa and in Claude Lorraine” ed. Beneš and Harris Villas and Gardens in Early Modern Italy and France Cambridge UK: Cambridge University Press, 2001. 88-113.

Bergmann, Bettina “The Pregnant Moment: Tragic Wives in the Roman Interior” Sexuality in Ancient Art Cambridge UK: Cambridge University Press, 1996. 199-218.

Bois, Yve-Alain “A Picturesque Stroll around Clara ClaraOctober 29, Cambridge, MA: MIT Press, 1984.

Harris, Dianne “Landscape and Representation” ed. Beneš and Harris Villas and Gardens in Early Modern Italy and France Cambridge UK: Cambridge University Press, 2001. 178-204.

Hazlehurst, F. Hamilton Gardens of Illusion Nashville TN: Vanderbilt University Press, 1980. 373.

Kepes, Gyorgy Language of Vision Chicago: Paul Theobald, 1944.

Lazzaro, Claudia The Italian Renaissance Garden, New Haven and London: Yale University Press, 1990. 1-108.

Moore, Stainton, and Gallin ed. Landscape Design: A Cultural and Architectural History  New York: Harry N. Abrams, 2001. 58-96, 125-164.

Weber R, Choi Y, Stark L. “The Impact of Formal Properties on Eye Movement During the Perception of Architecture.” Journal of Architectural and Planning Research. 19:1 (Spring, 2002)