L'INNOCENZA PERDUTA (LOST INNOCENCE):
CONSERVING A CARRARA MARBLE STATUE

Chris Cleere

BSc. Archaeological Conservation
Institute of Archaeology, University College London, 31-34 Gordon Square, London WC1H 0PY, United Kingdom

Abstract

This paper describes the conservation of L'Innocenza Perduta (Lost Innocence), a marble statue by the Florentine sculptor Emilio Santarelli. As the statue is on open display at University College London and is accessible to the public, it has been vulnerable to destructive elements, including airborne pollutants and vandalism. Cleaning was done by steam and poultice. However, during the cleaning process it became clear that the capital and plinth required more complicated conservation treatments due to considerable cracking of the stone.

Figure 1: Full view of the statue before conservation (59K)

Introduction

The aim of this project was to assess the condition of the statue, formulate a conservation plan, and make recommendations for future display in order to protect the object from further deterioration. L'Innocenza Perduta (Lost Innocence) is a Carrara marble statue owned by University College London. It measures 117 cm high, and is mounted on a column-style plinth, oval in shape, which is 115 cm high. The statue is comprised of a white marble capital and base with a serpentanite drum. The figure is of a seated, partially-draped nude female. The left hand holds lilies that are cascading to the ground where they entwine with a snake that is emerging from the rock on which the figure is sitting. The statue is signed "Emilio Santarelli F 1862."

Figure 2: Signature of the sculptor (39K)

Emilio Santarelli (1801-1886) was born in Florence the son of an artist (Giovanni Antonio). Santarelli initially trained with Francesco Carraderi at the Academy of Fine Art in Florence. In 1832 Santarelli continued his studies with Thorwoldsen in Rome. Returning to Florence he was elected Professor of the Academy of Fine Art in 1849. Among the works of Santarelli are the statue of Michelangelo Buonarotti placed under the loggia of the Uffizi, Florence; a series of bas-reliefs at Santa Croce, Florence; a statue of Columbus in Genoa; a half colossal statue of Saint Frances; and statues titled "The Good Shepherd," "Kneeling Magdalen," "Bacchante," "Cupid in Mischief," and "The Prayer of Innocence" (a possible pair to this statue). Santarelli is also remembered for the collection of drawings by old masters that he collected and donated to the city of Florence, "so they may never be dispersed or fall into mercenary hands." This collection is now displayed in the Corridoio Vasariano at the Uffizi Museum.

Figure 3: The statue before conservation (41K)

The UCL collection catalogue describes the statue as: "A partly draped seated female figure. Bequeathed the University College by Louisa Lady Goldsmith." It is more likely that Louisa Lady Goldsmith was, in fact, Louisa Lady Goldsmid, wife of Sir Francis H Goldsmid, son of Isaac Lyon Goldsmid, one of the original founders of University College. Isaac Goldsmid was an extremely rich bullion dealer in the City of London, who donated vast amounts of money to University College. This generosity was continued first by his son Frances and, after his death in 1874, by his widow Louisa. In 1862 Sir Frances commissioned P. C. Hardwick to build his country mansion Rendcomb House, Gloucestershire. The house was built between 1863 and 1865, and the similarity between these dates and that of the statue (1862) suggests that the statue was purchased for the house. If this was the case, it would explain why the statue does not appear in any catalogues of Santarelli's work, as it may never have been on public display.

Previous Treatment and Condition Assessment

The missing area of the nose had been partially filled with a white gap-fill material and the joint between the statue and base had been badly filled with a grey material with the appearance of Portland cement. There was, however, no evidence of previous comprehensive conservation treatment.

Figure 4: Previous treatment of the base using a grey material (27K)

Both statue and plinth were covered with a thick cohesive layer of dirt, comprising of a greasy material containing dirt particles, hair and detritus. There was evidence of liquid having run over the figure, leaving brown stains down the left thigh. However, the dirt layer obscured the surface to the extent that the overall level of staining of the stone could not be determined.

Figure 5: Detail of stains on the thigh of the statue (98K)

The figure had suffered damage resulting in loss to some areas, most noticeably the left hand, the nose, the right foot, the drapery and the head of the snake. The joint between the drum of the plinth and the capital contained no jointing material and was out of alignment by approximately one centimetre. This had produced a gap between the drum and capital resulting in uneven loading to the plinth. The neglected state of the statue and its positioning in a public place had lead to it being treated with little respect. There was evidence of cigarettes having been stubbed out on the base and chewing gum embedded in recesses. It was decided that the statue and plinth should be cleaned; firstly, so that the condition of the stone could be assessed, and secondly, that once clean the statue might be treated with more respect.

Account of Conservation Measures

Due to the lack of laboratory space and equipment capable of dismantling the statue and plinth, conservation work had to be carried out in situ. As the statue was positioned in a public area it was decided that work should be carried out only at times of low traffic, and that no conservation treatments, equipment, or materials that could pose a threat to public safety could be left on the statue when the conservator was not present. It was also decided that the statue should be screened from the public while work was undertaken. It was hoped that screening the statue would have two effects; firstly, it would stop members of the public interfering with the statue during the conservation treatment, and secondly, on unveiling, the finished statue would show a dramatic change rather than a gradual improvement. It was felt that the impact of a dramatic change would show that the statue had warranted conservation work and that this might discourage people from interfering with the statue in the future.

Due to the position of the statue between three doorways, it was impossible to erect permanent barriers as this would have broken fire regulations. It was therefore decided that the statue should be covered while the conservation work was carried out. To achieve this a 'Tyvek' cover was manufactured and taped over the statue when work was being undertaken. 'Tyvek' was chosen as a covering material as it resists tearing and is permeable. This meant that the statue could be covered while still wet from the conservation treatment and be allowed to dry through the cover with no ill effects.

The first stage of the conservation treatment was to remove the dirt layer to reveal the stone surface. Loose dust was first removed with a vacuum cleaner. Swab cleaning and steam cleaning were then considered and steam cleaning was chosen as the most applicable treatment, based on the level of soiling. Swab cleaning was the more cautious option, offering a more controllable treatment, but the level of soiling and time available ruled it out as too time consuming.

Tests with steam cleaning were carried out on small areas of different surfaces (broken edges, polished surfaces, rough surfaces) and a good result with no disruption of the surfaces was achieved in a reasonable time. Steam cleaning was therefore adopted for the initial cleaning treatment. Cleaning commenced at the top of the head and continued down the figure and plinth. This pattern of cleaning was adopted, and although water was constantly blotted from the surface, it was impossible to avoid runs. By working down the statue the problem of dirt from above re-depositing on cleaned areas below was avoided. The previous gap-fill in the nose was removed with 'Nitromors' paint stripper. Steam cleaning the figure successfully removed the dirt layer from the surface and revealed large areas of disfiguring stains . When the cleaning treatment reached the capital of the plinth another problem was encountered. What had looked to be a surface scratch or fault in the stone running down the right hand side was revealed to be an open crack approximately two millimetres wide.

The open crack appeared to correspond with a closed crack on the opposite side that ran from the top of the capital for approximately ten centimetres. What appeared to be stress fractures ran diagonally up from the open crack indicating that the capital had broken and slumped in the centre. This deduction was reinforced by the gap between the capital and drum being out of alignment and containing no jointing material.

Figure 6: Stress fractures on the capital (39K)

The broken capital was seen to be a much larger threat to the continued survival of the statue than the contamination of the surface, since, if the capital had slumped, the weight of the statue would no longer be evenly distributed and could lead to cracking of the statue itself. Cleaning of the statue was consequently suspended while the problem of the cracked capital was addressed. It was decided that the only viable treatment option was to dismantle the statue and plinth. Adhesive could be injected into the crack, but without access to the break surfaces adhesion could not be guaranteed. Furthermore, consolidating the crack would not reverse the slumping of the capital. In order to dismantle the object, the joint between the capital and statue had to be broken. The joint contained a material that resembled Portland cement. The joint was approximately five millimetres wide and the material had cracked in some places. It was possible to remove a section of the jointing material and this revealed that there was very little adhesion between the jointing material and the capital or statue. The jointing material extended to a maximum of seven centimetres from the exterior of the join.

Figure 7: Lifting of the capital (98K)

Eight evenly spaced sections of the joint material (A in Figure 8), approximately five centimetres in length, were removed and replaced with wooden wedges (B in Figure 8). The wedges were then driven in, in order to raise the statue from the capital. The jointing material was then removed from between the wedges, leaving the statue approximately two centimetres above the capital. Webbing straps could then be placed under the statue (C in Figure 8). The statue was raised using a one-tonne chain hoist on a scaffold frame erected around the statue. This allowed the capital to be slid forward using rollers onto a trolley placed in front of the statue. The statue was then lowered onto wooden boards placed across the capital and the scaffold was disassembled.

Figure 8: Diagram showing the steps taken in removing the capital from the statue (20K)

On removal of the capital the reason for its failure became obvious. Inlayed into the top of the capital was an iron disk approximately thirty centimetres in diameter. A fragment of the capital had become detached, revealing that the iron disk was approximately seven centimetres in depth. The iron disk was extremely corroded and the pressure resulting from the subsequent increase in size had caused the capital to fail. The reason for the disk's presence was a mystery as the capital was easily substantial enough to support the weight of the statue without added reinforcement. The capital was moved to the laboratory, and corrosion cleaning was initiated with a view to discovering the role of the disk. Cleaning revealed that the disk was in fact two disks; one on top of the other. A lever was inserted between the two disks and the top disk was lifted. The two disks parted revealing that the disks had been a turntable with a central pivot that extended below the lower disk by approximately fifteen centimetres. The pivot had broken when the lever was applied.

Figure 9: Section of the capital showing the turntable (39K)

The removal of the lower section of the turntable proved extremely difficult due to a mortar that had been used to bed the turntable to the capital. During removal, the crack that ran part way through the capital continued and the capital was reduced to a number of fragments. While breaking the capital was unfortunate, it did allow for the surfaces of the crack to be cleaned and prepared correctly for joining. The corrosion of the turntable left a large amount of staining on the capital. It was felt that if this material was left in situ when the fragments were reassembled there was a possibility that it might become mobile and migrate to the outer surface of the reconstructed capital. To remove the iron stains, a five percent solution of sodium dithionate was applied to the surface in a poultice of sepiolite clay. The poultice was covered with 'cling film' for twenty-four hours, after which the clay was uncovered and allowed to dry. After three applications the iron stains had been removed and the capital fragments were rinsed in running tap water for eight hours to remove any residues of sodium dithionate. As the capital plays a structural role in the support of the statue, the joints in the reconstructed capital had to be as strong as the stone. As the main joint in the capital ran at ninety degrees to the downwards force from the statue, it was decided that the joint should be reinforced. A fifteen millimetre stainless steel (grade 322) rod was chosen as a reinforcing material.

Figure 10: Reinforcement of the capital using stainless steel rods after reconstruction (20K)

To allow for expansion of the stainless steel rods, 'Plastizote' plugs were inserted into the dowel holes before the rods were adhered into place. The adhesive of choice for both the dowels and the stone was a polyester mastic (General®) used as recommended by the manufactures (Pisani Ltd). The mastic was kept to within three centimetres of the surface and not allowed to contaminate the outer surface of the stone as it will yellow in ultraviolet light. The rods were roughened with a hack saw prior to adhering in the dowel holes and the break surfaces were roughened with a stone chisel in order to provide a key for the adhesive. The surface of the void left by the removal of the turntable was first coated with a five percent solution of Paraloid B72 in acetone in order to produce a separating layer between the fill and the stone. The void was then filled with the same polyester mastic used as an adhesive, mixed with marble powder in the proportion of one part mastic to three parts marble powder. The filler was used to reduce the shrinkage that is a property of this type of material. Once reconstructed, the exposed joins of the capital were filled with a twenty percent solution of Paraloid B72 in acetone containing glass micro-balloons and coloured to match the stone using earth colours in Paraloid B72 as a medium. The capital was then replaced on the drum of the plinth with a layer of lime mortar acting as a bedding layer. The statue was then replaced on the plinth again using lime mortar as a bedding layer. When the mortar had dried the joint between the statue and capital was filled with Paraloid B72 and glass micro-balloons in order to fill chips that had previously been lost from the statue base.

Following the repair of the capital, attention was again turned to the soiling of the statue. Steam cleaning had revealed heavy staining over most of the surface. Swabbing with a variety of common solvents had no effect, so poulticing of a test area was carried out. A poultice of Sepiolite clay with distilled water and a small amount of non-ionic detergent was applied to the test area and covered with 'cling film'. The poultice was left covered for seven days, after which the 'cling film' was removed and the poultice allowed to dry. Upon removal, the staining had been dramatically reduced. There were, however, some areas, notably the face, left arm and left leg, where some staining remained. The overall colour of the surface had been lightened and the remaining staining was highlighted against the new background. The remaining stains were lightened with a 20% volume solution of hydrogen peroxide applied locally with cotton wool swabs. By this method the stains could be reduced to a level that blended with the overall patina of the surface without bleaching the surface to pure white. After the cleaning procedures had been completed the statue and base were again steam cleaned to remove residues of the treatments used.

Throughout the conservation treatment, research had been undertaken to try to establish the correct form of the missing elements of the statue. Enquiries in both the United Kingdom and Italy had revealed no reference to the correct appearance of the missing portions of the statue, and for this reason it was decided that no reconstruction should be carried out.

As the statue is on open display and accessible to the public, it was felt that the figure was still vulnerable to the same destructive elements that had caused its previous deterioration -- i.e. airborne pollutants and vandalism. For this reason it was decided that the surface of the statue and base should be coated with a material that would both seal the open porous structure and be reversible in order to aid in the removal of any unwanted additions by the public. 'Cosmaloid 80' micro-crystalline wax was chosen as a coating because it fills the criteria required and produces a finish that enhances the patina without the appearance of an applied coating. The micro-crystalline wax was applied by brush as a ten-percent solution in white spirit. When the solvent had evaporated, the surface was buffed with a soft brush to remove excess material and impart a sheen to the surface. In areas the coating was too thick and had become too glossy. The coating in these areas was reduced by swabbing with white spirit. By this method a uniform finish was achieved over the whole object. Following the culmination of the treatment, the cover was removed from the statue and the object was again displayed to the public.

Figure 11: Statue after conservation (39K)

Conclusion

The aim of the project was to assess the condition of the object, formulate a conservation plan, and make recommendations for future display in order to protect the object from further deterioration. These aims have been achieved, albeit as a far bigger project than was originally envisaged. Recommendations for redisplay have been made and until these are acted on, the object cannot be considered to be safe from any future deterioration. Therefore, although the aims of the project have been fulfilled, the conservation of the object cannot be considered complete.

The methods, techniques, and conclusions found in individual papers are the work and responsibility of the author of the paper, and should in no way be thought to represent the opinion or endorsement of either the Journal of Conservation & Museum Studies, the Institute of Archaeology, or University College London. No liability or contract is accepted or implied by the publication of these data.

Copyright © Chris Cleere, 1997. All rights reserved.
http://www.ucl.ac.uk/archaeology/conservation/jcms/
jcms@ucl.ac.uk