Inpainting: Compensation Goals/Philosophical Issues

Back to the Paintings Main Page

Back to Inpainting Chapter Index

The information presented on the Paintings Conservation Wiki is the opinion of the contributors and does not imply endorsement or approval, or recommendation of any treatments, methods, or techniques described.

Compiler: Catherine A. Metzger

Historical Background

Approaches to the Reintegration of Paint Loss: Theory and Practice in the Conservation of Easel Paintings

Introduction

The inextricable links between materials and meaning in a work of art have ensured that issues of retouching have retained a prominent place in the theory and practice of paintings conservation throughout its history. Damage to the paint layer can result in a change or devaluation of the aesthetic, spiritual, historical, or cultural meaning of the work. Consequently, the method used to integrate paint losses, or the decision not to reintegrate them, strongly affects how the painting is understood and appreciated. A comprehensive knowledge of the underlying rationales that guide conservators in a particular approach is, therefore, very important.

For certain approaches, specifically Italian visible retouching techniques, comprehensive theoretical frameworks have been developed. This includes the writings of Cesare Brandi (2000) and Umberto Baldini (2001, 1997). Imitative retouching, on the other hand, is based on a longstanding tradition that is guided be practical principles rather than an explicit theoretical foundation. It is, therefore, necessary to consider discussions of the relative merits of and justifications for specific approaches to loss compensation that are dispersed in conference proceedings (Association of British Picture Restorers 2000, ARAAFU 2002, Garland 2003, Leavengood 1994, Meiss 1963) and published case studies of important conservation treatments, particularly those where significant areas have been reconstructed (Dunkerton 2003, Wallert, Tauber, and Murphy 2001) or where instead large losses have not been reintegrated. (De Bruyn Kops 1975, Seymour 1972)

Three important international conferences merit special mention, as their associated publications reflect contemporary discourse at key moments in the history of conservation, and indeed, are cited frequently throughout this review. The “International Conference for the Study of Scientific Methods for the Examination and Preservation of Works of Art” was held in Rome in 1930 and is considered to be a seminal event in the establishment of the modern discipline of conservation (Clavir 1998, Von der Goltz 1996). Although the meeting was not specifically devoted to issues of retouching, the subject held a central place in the discussions, indicating that it was a critical issue in conservation at the time. The findings of the conference were disseminated in documents and articles (Ruhemann 1931, Van Gelder 1931) and eventually edited into a handbook, Manuel de la Conservation et Restauration des Tableaux (1939), followed by the English edition, Manual on the Conservation of Paintings (1940). The text was intended as a guide for curators responsible for overseeing conservation work in museums. It was compiled by Helmut Ruhemann and George Stout, and is valuable as an expression of international opinion at the time.

In 1961, the “Twentieth International Congress of the History of Art”, held in New York, dedicated a session to the philosophical and practical issues of loss compensation: “The Aesthetic and Historical Aspects of the Presentation of Damaged Pictures” (Bomford and Leonard 2004, Meiss 1963). The principle speakers were museum director Philip Hendy and art historians Richard Offner and Cesare Brandi, who expressed widely divergent views about loss compensation, advocating for complete reintegration, no reintegration, and visible reintegration, respectively. Other participants included art historian Millard Meiss and curator Charles Seymour, conservators Sheldon Keck and George Stout, and conservation scientist Paul Coremans.

The third important conference was the symposium “Early Italian Paintings: Approaches to Conservation” hosted by Yale University in 2002. This meeting focused on the restoration history and treatment of the collection of Early Italian paintings at Yale University Art Gallery, but broader issues of retouching philosophy were also discussed. The proceedings comprise papers by an international group of conservators whose papers on the history and development of Italian retouching methods provide an unprecedented level of access and understanding of the non-Italian reader.

Several important publications provide critical assessment and historical overviews of retouching approaches. These works, however, are not intended to be comprehensive literature reviews (Bomford 1994, Daly Hartin 1990) and often focus on restorations carried out in specific historical periods (Partridge 2003, 2006) or on a single retouching approach, for instance Italian visible retouching techniques (Olsson 2003, Ramsey 2000, Reifsnyder 2003, Schadler-Saud 1986, 1999b). Despite the crucial impact that retouching has on the appearance and understanding of the artwork, a comprehensive review and contextualization of theory and practice is lacking in the English language literature. This review aims to address the most important contributions on the subject. It should be noted, however, that despite the impossibility of considering retouching theory in isolation from its practical application, the more technical aspects of the methods and materials are outside the scope of the paper. Similarly, philosophical issues specific to the restoration of modern and contemporary paintings are not separately explored.

Complete Reintegration

A loss in the paint layer represents a negative alteration that diminishes the value of a work of art, whether aesthetic, devotional, or commercial. Complete reintegration, also referred to as “imitative” or “mimetic” retouching, aims to reconstruct the missing parts of the image by emulating the appearance of the original paintings as closely as possible. Complete reintegration is the traditional approach to loss compensation in easel paintings conservation. It was used in the earliest restorations carried out by artist-restorers, and continues to be applied to the majority of paintings in public and private collections. Alternatives to complete reintegration were rare until the 1930s and 1940s. Nonetheless, there is a long history of discussion on approaches to reintegration, from both philosophical and practical perspectives, and the issues involved have played a pivotal role in the evolution of modern conservation theory. The first part of this section deals with the practice of complete reintegration from a historical perspective, covering approaches and commentaries on restorations carried out before the early decades of the 20th century. The second part focuses on modern conservation and current practices of imitative reintegration.

Historical Overview of Complete Reintegration

Pioneering works on the history of restoration and conservation by art historian Alessandro Conti (2007, 2002, 1981 and art historian and conservator Roger Marjinissen (1967) draw upon documentary sources to provide a survey of approaches to restorations carried out by Renaissance painters to repair and update devotional images, to restorations in the 18th and 19th centuries that reveal a growing respect for the artist’s original work and avoidance of general overpainting. This latter tendency did not preclude the occurrence of more extensive, “aesthetic” reintegrations, which were undertaken for both museums and private collectors, indicating that different approaches to restoration have always been practiced concurrently. Of particular interest in the writings reviewed by Conti and Marjinissen are cases that reflect attitudes that seem to anticipate 20th-century standards, such as the international use of an easily removable, water-soluble retouching medium by artist-restorer Carlo Maratta in the late 17th century restoration of Raphael’s frescoes in the Psyche Loggia of the Villa Farnesina. The Getty Conservation Institute (GCI) publication Issues in the Conservation of Easel Paintings includes excerpts of several important historical texts from the 19th century and earlier that provides rare documentation and criticism of early restorations, as well as viewpoints on retouching expressed in 19th-century restoration manuals (Bomford and Leonard 2004).

Condemnation of excessive and poorly executed restorations did lead to calls for no reintegration, as well as proposals for stricter retouching practices, well before the 20th century (Conti 2007, Darrow 2000 2002, Levi 1988, Marjinissen 1967, Partridge 2006, Tranquilli 1996). Already in the early 19th century the restorer Pietro Edwards recommended that retouching be limited strictly to areas of loss, and stressed the importance of the stability and removability of the retouching material (Conti 2002; Darrow 2000, 2002; Marjinissen 1967; Tranquilli 1996). In 1877 Giovanni Battista Cavalcaselle, art historian and arts administrator for the Italian State, issued a document outlining guidelines for restoration in which he advocated minimal intervention in retouching. In certain cases of fresco restoration, neutral watercolour toning of losses was carried out under his supervision (Conti 2002, Levi 1988, Partridge 2006). Instances of an approach to retouching more in line with modern standards, however, were relatively isolated and a disconnection between theory and wide scale practice persisted. Restorers frequently resorted to overpainting, which could damage or remove original paint, or which deliberate attempts to falsify the condition of the work or alter the original composition (Dwyer Modestini 2003, Marjinissen 1967).

More recent studies on the history of restoration explore how the aesthetic values and cultural expectations of the time have influenced the retouching and reworking of images (Anderson 1994; Braham 1975; Gould 1974; Hoeniger 2005, 1995; Sitwell and Staniforth 1998). Art historian Cathleen Hoeniger’s research into early renovations of Tuscan devotional and civic painting discusses the religious and social motivations for the reworking of these images. For examples, works were often rejuvenated through retouching to ensure their continuing efficacy or to update them for new generations of worshippers (Hoeniger 1995). Restoration work commissioned by Sir Charles Eastlake, first Director of the National Gallery in London, has been the subject of much critical attention. These restorations sometimes involved the repainting of areas to correct perceived deficiencies in the image, or retouching to alter the image for reasons of propriety so that it would be acceptable to the 19th-century viewer Anderson 1994, Braham 1975, Conti 2002, Gould 1974, Partridge 2006).

Imitative Reintegration and the Establishment of Principles for Conservation

The 1930 Rome conference occured at the time when conservation was establishing itself as a professional discipline, and the manual that was compiled by an international committee following the meeting represents an early attempt to formulate internationally accepted guidelines for practice (Manual on the Conservation of Paintings 1997). The opening pages, devoted to general principles for the conservation of paintings, focus largely on issues of retouching rather than varnish removal or structural treatments, indicating that it was considered to be a central concern. These principles, which serve to distinguish modern imitative retouching approaches from the deceptive practices of the part, include limiting the retouching to the areas of loss (inpainting rather than overpainting), ensuring that the materials used remain easily removable, and thoroughly documenting the treatment. It was also stressed that, prior to treatment, the conservator should gather as much information as possible about the work of art and how it has changed over time, consulting with curators and art historians for their perspective.

While significant attention was given to visible retouching techniques current at the time, the principles formulated at the Rome Conference allowed for a range of practical solutions, including complete reintegration. The aim of complete reintegration, however, was no longer to achieve a pristine finish in the restored work. Greater consideration was given to the impact of the loss on the painting and some signs of age and alteration came to be accepted. Alternatives to imitative reintegration were considered when the loss was extensive or occurred in an important part of the composition, or when there was insufficient information about the missing areas to inform a reconstruction. With the organization of the conservation profession over the following decades, these principles came to be codified through the drafting of professional codes of ethics (Murray Pease Report 1964, Murray Pease Report/Code of Ethics for Art Conservators 1968) and were further promoted in newly-established formalized training programs and texts on the theory and practice of restoration (Brandi 1963, Ruhemann 1968).

It is perhaps because imitative retouching has been the traditional approach to loss compensation, and because its development has been more practical than philosophical, that there has been no perceived need to articulate a theory of complete reintegration. Although complete reintegration continued to be widely practiced in the first half of the 20th century, there was not much published defense or theoretical discussion of this approach at the time (Manual on the Conservation of Paintings 1997, Stout 1941). However, the emergence of visible retouching techniques and philosophies of non-intervention in the 1940s and 1950s no doubt compelled a response from proponents of complete reintegration. These alternative approaches were, in part, a critical reaction against the excesses and empiricism of some 19th-century restorations, but they also reflected philosophies about the impact of paint loss and the function of the restoration that differed from that of supporters of modern imitative reintegration (Baldini 2001, 1997; Brandi 2000; Seymour 1972). The attention received by the published theories supporting these alternative approaches seems to have prompted the need for a defense of complete reintegration. This occurred in 1961 at the New York Congress, which provided a venue for discussion of the relative merits of the different approaches. Philip Hendy, Director of the National Gallery in London, a damaged state was antithetical not only to the artist’s intention but to the very essence of the work of art, and argued for restitution of the unity of the image through imitative reintegration (Hendy 1963). Hendy also expressed concerns over how members of the public would respond to the display of unrestored works. His intention was not to conceal the true state of the painting; in fact, he suggested that this information be made available separate from the artwork, in a collection catalog, for example. Hendy had implemented this approach in the 1930s when he wrote the catalog for the Isabella Stuart Gardner Museum, Boston (Hendy 1931). Subsequent commentators have raised similar arguments for complete reintegration (Dwyer Modestini 2003). Ruhemann, who practiced imitative retouching in most cases, was open to alternatives. He expressed concerns, however, about the distracting effect of both unrestored losses and poorly executed visible retouching (Ruhemann 1968).

No Reintegration

When no reintegration of losses is carried out, it may be part of a systematic policy that is related to an extreme minimalist view of restoration, but more often it is considered for specific cases such as paintings with extensive paint loss that would require inventive reconstruction, or where damages are deemed to carry some historical or cultural significance. Since the work is presented without any pictorial restoration, the appearance of the painting depends entirely upon the condition of the original materials.

There is a long history of commentary on the negative effects of retouching, as well as proposals that paint losses should be left exposed, particularly in Italy (Bomford 1994, Bomford and Leonard 2004, Ciatti 2003, Offner 1963, Partridge 2006, Schädler-Saub 1986). This point of view is often expressed in the context of criticism of a specific restoration where it is judged that a work of art diminished by poor-quality additions. For example, Giorgio Vasari’s disapproval of the restoration of a fresco painting by Luca Signorelli in the Church of San Francesco in Volterra let him to suggest that, “it would be better, sometimes, to keep the things made by excellent men half-damaged than to have them retouched by someone of lesser skill” (Conti 2002, 58). Writing in the 19th century, Cavalcaselle argued that it was better to have “a painting deteriorated or missing some part, than a painting finished or refreshed by the restorer that ends up being neither old nor modern work” (Paolucci 1986, 15). The prominent Italian art historian Roberto Longhi, in his critique of restoration practices in Italy in the early 20th century, stated that it was preferable to leave the losses exposed, “so the eye...can set them aside without effort, and restore, but only ‘mentally’...that which is missing” (Longhi 1940, 124).

Early Italian Paintings and the “Archeological” Approach

A non-interventive approach is often described as “archeological” because the object can appear like a ruin or fragment, and because it often involves the removal of subsequent restorations in a kind of “excavation” aimed at retrieving the original image. As a methodology, the archeological approach is most closely associated with the treatment of Early Italian paintings in the decause around the mid-20th century. During this period it was applied to paintings in important collections in Italy (Ciatti 1990), France (Mognetti 2003), and the United States (Aronson 2003, Hoeniger 1999, Seymour 1972). With the aim of revealing the original image, later restorations and additions were systematically removed, regardless of their aesthetic or historical significance, and the paintings were presented with no, or in some cases minimal, reintegration.

Perhaps the most well-known and discussed application of this purist approach involves the collection of Early Italian paintings in the Yale University Art Gallery. Between 1951 and 1972, conservator Andrew Petryn and his colleagues undertook a conservation campaign, under the direction of curator Charles Seymour, which involved about one hundred and fifty paintings in the collection (Aronson 2003). In the majority of cases, following the removal of previous restorations, a philosophy of no retouching as employed, and paint losses and abrasions, which in many cases were substantial, were left exposed. Seymoud described the conservation program in the 1961 New York Congress (Seymour 1962, 176-178), and in a 1972 exhibition catalog, which documented the most important treatments (Seymour 1972). Seymour stressed the recovery of the original image, which had been obscured by restoration, invoking concepts of “truth” and “authenticity”. Offner, a specialist in the history of Early Italian painting, supported the conservation program at Yale. At the New York Congress, he argues that “the rejection of all but its original elements is the first and the final condition of an adequate restoration of a painting” (1963, 159, Contreras de Berenfeld 2003).

The purist philosophy was driven in part by a desire to distinguish between the work of the original artist and later additions, an important requirement for connoisseurship and art historical study. Extensive restorations had long been considered an obstruction to the study of the work, posint an impediment to attribution and authentication (Partridge 2006). Some advocated for exposure of the actual state of the painting from an aesthetic perspective; that is, the original painting was felt to be best appreciated without the imposition of restoration, even if the work was in a highly damaged condition (Offner 1963; Partridge 2006; Seymour 1972, 1963).

Hoeniger notes that the function and reception of Early Italian paintings, combined with the condition in which many have survived (often fragmented, decontextualized, and badly damaged) has meant that they have often been subjected to heavy or multiple restorations (1995). The frequent presence of reworkings, extensive retouching, and overpaint on these works caused some commentators to draw a parallel between complete pictorial reintegration and falsification, as has been discussed by Hoeniger (1999) and Catalano (1998). Offner, for example, argued that “any addition whatever introduces irrelevant matter and serves to instil a false impression in anyone who sees the restored work...easy removal fails to justify it either on moral or aesthetic ground” (offner 1963, 157). Similarly, Meiss insisted that “mere knowledge of what is authentic and what is added is not sufficient. The two, in my experience, inevitably merge. The original is polluted by the fake” (1963, 165).

The correlation between restoration and forgery in the treatment of Early Italian paintings was, in fact, ont entirely philosophical. With the critical reevaluation of paintings from this period in the late 19th and early 20th centuries, there emerged a new appreciation for these works on the part of foreign tourists and art collectors. This created a heavy market demand for genuine paintings, as well as a flourishing industry around the restoration and forgery of trecento and quattrocento works (Mazzoni 2001). This resulted in heavy restorations of seriously degraded original works - sometimes to deceive potential buyers, sometimes at their request - as well as the production of outright fakes (Joni 2004, Mazzoni 2001). In some cases, the line between a heavy restoration and a forgery can be ambiguous (Muir and Khandekar 2006). With the prevalence of unscrupulous restorations of Early Italian paintings in the early decades of the 20th century, perhaps a systematic policy of nonintervention for works in public collections seemed to provide the only guarantee of an unadulterated presentation of the damaged artwork.

The conservation campaign in Yale has been widely criticized for its strict adherence to a methodology that failed to recognize the unique requirements and histories of individual works (Aronson 2003, Christiansen 2003, Hoeniger 1999, Walden 1985). With the implementation of principles for retouching, as well as the possibilities offered by visible techniques, there is no longer much support for an archeological approach to restoration. Quite separate from a purist ideology, however, is the rejection of retouching in specific cases where the damages are considered to be historically or culturally meaningful (Bergeon 1990, De Bruyn Kops 1975, Mognetti 2003, Muir 2007). Ségolène Bergeon has referred to this as a patine d’utilisation (1990, 194-98) that reflects the history and function of the work.

Visible Retouching

Visible, or differentiated, retouching techniques aim to reintegrate the image by reducing the visible impact of the loss while ensuring that the restoration is clearly recognizable as such. This approach is often seen as a compromise solution because it reachers thing of a middle ground between imitative retouching and nonintervention, namely aesthetic reintegration of the damaged image in a way that neither competes with the original nor conceals all evidence of the actual condition and history of the work.

Examples of visible retouching are known from well behind the 20th century. For example, it is documented that Carlo Maratta used a hatched (tratteggio) technique in his restoration of Rachael’s Psyche Loggia in the late 17th century (Hoeniger 2005, Varoli-Piazza 2002). Conti mentions the use of pointillist techniques in the 1800s (2001). In the early part of the 20th century, there was a considerable amount of experimentation with alternative retouching approaches. In 1914, the German restorer Victor Bauer-Bolton argued that retouching should be clearly identifiable and proposed a method whereby the loss was filled and retouched at a slightly lower level than the original painting (2004). Significantly, he also recognized that the retouching approach largely depended on whether the painting was considered primarily as a work of art, a historical document, or a commercial object.

In 1922, Professor Max Doerner described the use of a hatched retouching technique (Doerner 1962). As early as the 1920s, visible techniques were used at the Fogg Art Museum (Cambridge, Massachusetts), under the direction of Edward Waldo Forbed (Bewer 2002, Contreras de Berenfeld 2003, Mongan 1971). At the 1930 Rome Conference, Ruhemann, a practitioner of complete reintegration in most instances (Jessel 1976, Ruhemann 1968) supported the use of visible retouching in cases where losses were extensive and central to the composition (Ruhemann 1931). He subsequently described seventeen visible retouching methods in The Cleaning of Paintings: Problems and Potentialities (1968). Ruhemann’s recommendation that imitative retouching should be visually distinguishable from the original paint, however, was not always easily translated into practice. As MacBeth and Spronk (1997) have reported, it was Ruhemann’s expressed intention to avoid too pristine a finish in his 1932-33 treatment of Rogier van der Weyden’s Saint Luke Drawing the Virgin from the Museum of Fine Arts, Boston. Nevertheless, his retouching was brought to a high degree of finish, including recreating the craquelure of the original paint in some areas.

Cesare Brandi and the Istituto Centrale del Restauro

Visible reintegration did not achieve a true methodological status until the mid-1940s when the practice of tratteggio was developed by Cesare Brandi, director of the Istituto Centrale del Restauro (ICR) in Rome. Brandi formulated a theory of restoration that guided a number of important conservation projects and was influential in the training of conservators from Italy and abroad. He developed his theoretical framework in a number of articles and lectures between the 1940s and 1960s. In 1963, a collection of these papers was compiled as Teoria del Restauro (1963). His impact on Italian conservation policy is evidenced by the 1972 Carta del Restauro (1963). This official document, which was largely based on Brandi’s theoretical ideas, has had considerable influence on the conservation of cultural heritage in Italy.

Brandi conceived the work of art as an entity possessing both an artist and a historical dimension. It was, therefore, necessary that the reintegration should aim to reestablish the aesthetic aspect, its “potential unity”, while respecting the traces of the artwork’s passage through time. Brandi used the ideas of Gestalt psychology to describe the impact of the lacuna on the perception of the work as a whole. He argued that the loss was not only a disturbing factor because it represented a gap in the image but because it could assert itself as a form within dominance of the loss in order to allow the viewer to concentrate on the image. He advocated strict limits for retouching to prevent it from descending into invention or excess and argued that the retouching should always be recognizable under close viewing. Brandi arrived at tratteggio through experimentation with other visible retouching techniques, which ultimately proved to be unsatisfactory (Brandi 1954, Contreras de Berenfeld 2003).

Tratteggio consists of the application of many vertical strokes of various colors that are gradually built up to approximately the appearance of the original from a normal viewing distance, while remaining discernible under closer examination. Depending on the assessment of the impact of the loss, tratteggio could be used to reconstruct the missing forms of the composition or as a general toning of the lost areas. The technique was intended to provide a precise formula that would mitigate subjective influences and personal idiosyncrasies in the execution of the retouching.

Brandi’s philosophical ideas were further explored by Paul Philippor, a Belgian art historian and teacher, whose ideas on retouching were elucidated in two important papers co-authored with his father, the conservator Albert Philippot (Philippot and Philippot 1959, 1960). Issues of reintegration were also discussed in the chapter “Problems of Presentation” in Conservation of Wall Paintings, which Philippot coauthored with Paulo Mora and Laure Mora (1984). The practical aspects of tratteggio were developed by the Moras, who were early students of Brandi, and conservators at the ICR. This chapter gives a detailed explanation of the technique. Tratteggio was developed in direct response to the profound damage sustained by Italy’s artistic heritage during the Second World War. It was first applied to the frescoes by Lorenzo da Viterbo in the Mazzatosta Chapel of Santa Maria della Verità in Viterbo (Brandi 1996) and those by Andrea Mantegna in the Ovetari Chapel in the Church of the Eremitani in Padua (Brandi 1954), which were both heavily damaged in bombings. The approach was soon after applied to panel paintings (Bonsanti 2003).

Influence of Brandi Outside of Italy

Brandi’s practical solution to loss compensation has been adopted in many European institutions and training programs, and his philosophical framework continues to guide the work of paintings conservation (Bergeon 1990; Masschelein-Kleiner 1989-91; Mognetti 2003,1996 ; Philippot and Philippot 1959, 1960). However, with few exceptions (Lignelli 2003, 1997), tratteggio has not had a strong influence in English-speaking countries. It has been argued that Italian retouching methodology came to be seen as a more ethical approach to loss compensation (Meiss 1963) and many conservators experimented with the hatched techniques in the 1960s and 1970s (Bomford 1994, Olsson 2003). One of the main criticisms, however, is that, when poorly executed, visible retouching can become distracting and draw attention to itself (Bomford 1994). It has also been argued that the technique is only aesthetically suitable for use on frescoes or early panel paintings that do not reply on illusions of volume and three-dimensionality (Ruhemann 1968). Problematic interpretations and poor comprehension of Brandi’s theory, particularly with respect to the application of tratteggio, have been noted (Bomford 1994, Conti 2001, Philippot 2000).

The impact of Brandi’s theoretical writings in English-speaking countries has been debated. In the preface to the first full English translation of Teoria del Restauro, it is suggested that his texts have been inaccessible both because of the lack of English translations and because of their philosophical density (Brandi 2005). Salvador Muñoz Viñas, who characterized Brandi’s theory as outdated (2002), describes Teoria del Restauro as an “unnecessarily obscure text” (2005, 6).

Although the first full English translation did not appear until 2005, it is important to note that Brandi’s thought was actually disseminated, at least in part, much earlier. Brandi’s 1951 article on the restoration activities of the ICR, which included discussion of the reconstruction of the fragmented Ovetaro Chapel frescoes, contained an English language translation (1954). A few years later, he provided the entry on “Restoration and Conservation” in the Encyclopedia of World Art published in 1966, where he discussed the historical development of the concept of conservation and touched upon many of his theoretical principles. The topic of the session devoted to the “Aesthetic and Historical Aspects of the Presentation of Damaged Pictures” at the 1961 New York Congress, where Brandi presented a paper (1963), is significant and Catalano suggests that the subject matter is related to an awareness of Brandian thought at the time (1996). Laurence Kanter, on the other hand, contends that Brandi has been overlooked in America and the United Kingdom because acrimonious professional relationships between Brandi and important English and American historians of Italian art (including Offner and Meiss), a group that had significant influence on the development of conservation practices in Europe and America. He argues that Brandi’s presentation at the New York Congress, which was translated and summarized by Paul Coremans at the meeting, was largely disregarded (Kanter 2007).

Excerpts from Teoria del Restauro were translated in the 1996 GCI publication Historical and Philosophical Issues in the Conservation of Cultural Heritage (Stanley Price, Talley and Melucco Vaccaro 1996). These excerpts are accompanied by useful contextualizing essays. In one essay, Alessandra Melucco Vaccaro argues that “nearly all contemporary methods can be traced more or less consistently and more or less consciously” to Teoria del Restauro and the precepts of Brandi (1996, 328). Recent conferences and symposia celebrating the centenary of Brandi’s birth highlight the enduring relevance of Brandian philosophy, especially in English-speaking countries (ARAAFU 2002, Contreras de Berenfeld 2003, Garland 2003, Kanter 2007).

Influence of Brandi Outside of Italy

The philosophical approach developed by Umberto Baldini in the 1970s and 1980s has guided the restoration of numerous artworks in state-owned collections in Florence and surrounding areas. Baldini served as the head of the Florentine Restoration Laboratories and then as a Superintendent of the Opicio della Pietre Dure. His theory of restoration owed much to Brandi’s Teoria del Restauro. However, he considered tratteggio to be unsatisfactory, largely due to misinterpretations of Brandi’s theoretical framework (2000). Baldini developed his theory of “methodological unity” based on principles of optics and perception, and introduced two new visible retouching methods: astrazione cromatica (chromatic abstraction) and selezione cromatica (chromatic selection) (Baldini 2001, 1997). Like Brandi, Baldini’s philosophical writings can be challenging to read; only short excerpts are available in English (Bomford and Leonard 2003; Stanley Price, Talley, and Melucco Vaccaro 1996). The practical aspects of astrazione cromatica and selezione cromatica were developed by conservator Ornella Casazza, who provided a lengthy, written description of their application with illustrated examples (1999).

The practical application of these techniques depends on a critical assessment of the loss and its impact on the painting. If the loss is limited and can be reintegrated without invention, selezione cromatica may be used. This involves the build up of individual strokes of various colors selected from a limited palette, applied using a hatching technique. The overall color of the retouching is based on that of the adjacent original paint and results from the optical mixing of the colors of the individual strokes. This method differs from Brandi’s tratteggio in that the brushwork is not restricted to a vertical orientation, but rather follows the directionality of the original forms. When the loss is significant and cannot be reintegrated without invention or without overpowering the remains of the original, then astrazione cromatica is employed. In this case, the hatched brushstrokes do not reconstruct forms, and the color of the retouching is derived from the overall chromatic character of the painting. The abstraction is intended to create a “neutral link” between the passages of well-preserved original paint (Baldini 2001, 1997; Ciatti 2003).

Astrazione cromatica was first used in Cimabue’s Crucifixion from the Santa Croce Museum, which had endured serious damage and paint loss in the 1966 Florence flood (Baldini and Casazza, 1983). Due to the large losses that interrupt, most significantly, the face and body of Christ, it was felt that reconstruction of the missing compositional areas was unjustified. To some extent, the color of the retouching is based on the adjacent paint, and in this sense, the application begins to approach selezione. The restoration, in fact, was carried out before Baldini’s theory of restoration was fully articulated and the reintegration of the Crucifixion should be seen as a first realization of the approach. According to Baldini, the restoration was designed to reestablish the Crucifixion as a work of art, while ensuring that the restoration “has no independent action” on it (Baldini and Casazza 1983, 49-50). Others have questioned whether the restoration was successful in its aims (Conti 2001, Marijnissen and Kockaert 1995, Talley 1996, Walden 1985). The method, however, continues to be refined, and modifications and new developments of the techniques proposed by both Brandi and Baldini are evident in recent Italian restorations (Ciatti 2003).

Conclusions

The literature on retouching falls at the intersection of theory and practice. Because of the multiplicity of values that works of art can possess and the challenges of reconciling the work’s original appearance with changes that have occurred over time, retouching represents an ongoing dilemma for the paintings conservator. The fact that this aspect of conservation has been debated for centuries attests to the impossibility of imposing any kind of dogmatic solution. Whether a damaged painting is presented in a fragmentary state or is restored to a relatively high degree of finish depends on a combination of factors whose relative weight may change over time. Many different factors can influence restoration choices, and articulated theories are not always straightforward to implement. In current practice, different approaches to loss compensation can be manipulated to give a range of degree of finish, and conservators etnd to adopt more flexible approaches than were evident around the mid-20th century when methodologies were more rigidly applied.

Current trends in conservation research, which show greater interest in the subjective, culturally contingent factors that shape conservation theory and practice, have stressed the importance of an awareness of the historical and philosophical foundations of the profession (Avrami, Mason, and de la Torre 2000; de la Torre 2002; Muñoz Viñas 2005, 2002; Stanley Price, Talley, and Melucci Vaccaro 1996; Villers 2004). Knowledge of the most significant and fundamental sources in specific areas, such as the theory of retouching, can contribute to a more comprehensive understanding of the contexts of past treatments, as well as offering direction in the complex task of conservation decision-making.

Author’s note: This paper, titled “Approaches to the Reintegration of Paint Loss: Theory and Practice in the Conservation of Easel Paintings”, originally appeared in Reviews in Conservation, 10 (2009), pp. 19-28.

Kim Muir

Submitted January 2010

Retouching Paintings in Europe from the 15th Century through the 19th Century: Debates, Controversies, and Methods

---

Introduction

Loss compensation in paintings has generated discussion in Europe for centuries. This paper will attempt to provide an overview of debates surrounding loss compensation and show where possible the materials and techniques used, or at least recommended, for retouching. In my research, I have been helped by some extremely useful literature, especially Alessandro Conti’s Storia del Restauro published in 1988 as well as more recent studies by Andrew McClellan, Jaynie Anderson, Cornelia Wagner, and Elizabeth Darrow, to name just a few scholars, on specific moments in the history of conservation.

Retouching has been criticized in art historical writing as far back as Giorgio Vasari. In his biography of Luca Signorelli, Vasari discussed a painting where the head of the Christ Child has been damaged and was repainted about fifty years later by Sodoma. Vasari was not particularly impressed with Sodoma’s work, writing,

“At Volterra, he [Signorelli] painted a fresco over the altar of a confraternity in the church of San Francesco representing the Circumcision of Our Lord which is considered beautiful and a marvel, although the Child has suffered from the humidity, and the restoration by Il Sodoma is much less beautiful than the original. In truth, it would sometimes be better to keep works done by excellent men in a semi-damaged state than to have them retouched by some one less skilled.” ²

Vasari’s implication, of course, was that genius could not be reproduced. This sentiment against retouching would be restated periodically in the art historical literature.³ At the same time, there were other authors (often restorers) who, in the 18th century, began to evaluate the specific complexities of loss compensation.

This paper will examine the historical criteria for successful retouching. It will look at the beginnings of retouching in the style of the original artist in the Renaissance, the occasional use of reversible retouching materials in the 17th and 18th centuries, and the search for stable materials in the 18th and 19th centuries. By the end of the 18th century, specific training and skills were suggested for inpainting, including the knowledge of old master techniques and chemistry. In the 19th century, a “less is more” principle was articulated, an apparent response to a purist aesthetic that suggested leaving losses visible.

Retouching in the Style of the Original Artist

Retouching in the style of the artist seems to have begun surprisingly early. For example, in Simone Martini’s Guidoriccio da Fogliano fresco in Siena (1320s), a large area of the castle of Montemassi was damaged and recreated in the original style either seventy-five or one hundred and fifty years after the fresco was painted. ⁴ This practice, of course, was quite unusual at this date. More commonly, damaged and even undamaged paintings were renovated in a contemporary style to ensure the work’s continuing vitality. A Christ Child was added about one hundred years later to a Bernando Daddi triptych of the 1330s (J. Paul Getty Museum) representing a Madonna with saints. Art historians speculate that the piece was commissioned for a funerary chapel and that the Madonna was gesturing toward a sarcophagus, interceding for the soul of the deceased. At some point the paintings was probably moved, the interceding gesture no longer made sense, and the Child was added.

Despite “overpaint-to-update” practices, certainly by the late Renaissance there are examples of faithful reconstructions of the original composition. A famous case is Andreadel Sarto’s Madonna of the Harpies (1517). The painting was damaged in the Florence flood of 1557, and approximately the bottom fifth of the paint layer was destroyed. The painting was considered to be a work of genius and was reconstructed in a manner close to the pre-damaged state.

Reversible Retouching

Although intense preoccupation with reversible retouching seems to have begun in the 20th century, there were 18th-century documents concerning the importance of reversibility, among them the well-known discussions of Carlo Maratta’s projects. Carlo Maratta (1625-1713), an important Roman painter, was responsible for a number of restoration projects in Rome, including Raphael’s Stanze in the Vatican and Psyche Loggia at the Farnesina Palace. Being such major works, the Stanze and the Loggia have been restored and restored many times, and it is difficult to perceive what of Maratta’s work is still extant. The Psyche Loggia project (1693-4), however, was particularly well documented by contemporary and near contemporary writers. It was a complicated intervention involving frescoing areas in the vaults left unfinished by Raphael and undertaking structural work to deal with the consequences of the Loggia having been open to the elements. Also, according to Maratta’s biographers, when Maratta retouched Raphael’s frescoes, he used pastels bound in gum arabic so that his work could be easily removed if a more worthy restorer were found.⁵ There is another story of Maratta being asked by Pope Innocent XI to add a veil to Guido Reni’s Madonna del Cucito fresco in the Quirinal (1609-11). The pope thought the Madonna’s neckline was too low. Maratta added the veil, but with pastel so that he would not be irrevocably changing the work of another artist.⁶ Nothing remains of Maratta’s 17th-century reversible pastel retouching as far as I know. However, a restoration campaign of the 1980s in Anne of Austria’s apartment in the Louvre discovered a very late 18th-century pastel modification in a fresco (painted about 1655) by Giovanni Francesco Romanelli. When the frescoes were restored in 1799 during the French Revolution, the allegory of Faith was changed to an allegory of Victory with the chalice and cross hidden by a pastel fasces and laurel crown.⁷

About one hundred years after Maratta, the Venetian Pietro Edwards (1744-1821) became a tremendously important figure in the history of restoration. He was Director of the Paintings Academy in Venice and Director of Restoration for pictures belonging to the state. Edwards and his team were responsible for restoring over 700 paintings. He wrote a number of proposals and papers on caring for paintings, including Restoration of the Public Pictures Approved by a Decree of the Senate on September 3, 1778: Articles for the Restoration of These Pictures Proposed by Professor Pietro Edwards Preservation and the Best Maintenance of the Public Pictures (1786). Edwards’ recommendations sounds incredibly contemporary and included requests for condition reports. He was concerned about environmental effects on paintings, training for restorers, supervision and accountability, and removing old restoration that compromised paintings.⁸ Pietro Edwards felt that retouching should be reversible. In his rules outlining restorers’ responsibilities, he wrote, “They may not use on pictures materials which cannot be removed. Instead, everything they use shall be removable from the art whenever it is necessary”.¹⁰ In his Proposal for a School for the Restoration of Paintings (written after 1819), Edwards even suggested that students be required to copy “head-dresses, wings, feathers, and foliage” of various painters using varnish colors.¹¹

The Search for Stable Materials

While both Maratta and Edwards were influential figures and their concerns were known and discussed, reversibility was not the major concern in the 18th and 19th centuries. Instead, the real preoccupation in historical inpainting discussions was stability. Worries about inpainting darkening or lightening go back at least as far as Filippo Baldinucci’s 1681 Vocabulario, a dictionary of art terms. Under the heading “restaurare” or “to restore,” he wrote, “Many people think that the best pictures should not be retouched even by someone who is skilled since it becomes possible to recognize over time a restoration however small...”. ¹²

The actual search for stable materials seems to have begun in the mid-18th century. Antoine-Joseph Pernety (1972) in his art dictionary of 1757 recommended either tempera or wax retouching as highly stable. Under the heading “repeindre” or “to repaint” he wrote,

“Many people repaint damaged sections of paintings with the intention of repairing them, but it is very difficult to ensure that the new colors do not create stains...Oil will darken and create marks. One could repaint with tempera to avoid this problem...You can also read in the works of Messrs. de Caylus and Majault on painting in encaustic and wax, page 131, that colors prepared in wax work much better than oil colors to restore old paintings. M. le Lorain, a painter of the Academy, has tried this with singular success. He has repaired old paintings so that it is almost impossible to find the repaired and repainted areas.” ¹³

My favorite example in the quest for stability comes from France during the Revolution. As with Pietro Edward’s Venetian period, this nearly contemporary moment in France is fascinating for the history of restoration. The state was finding itself responsible for a huge number of art works as a result of the confiscation of aristocratic and Church property and Napoleon’s looting. The Parisian art and museum communities had to justify to a very skeptical Europe their ability to care for all this confiscated material. To respond, numerous museum commissions were appointed, including commissions on restoration. Among the appointees were a number of important restorers, chemists, and painters including the painter David.¹⁴ One of the committee members was Jean- Baptiste Pierre Le Brun (1748–1813), who was an art dealer, painter, critic, and husband of the painter Vigée-Lebrun. In 1794, demonstrating a true Enlightenment faith in chemistry, Le Brun invited French chemists to invent a new medium for retouching that would not discolor. Le Brun was on the committee charged with designing an examination for restorers. He thought that retouching should be judged six months after it had been executed to see if the colors had shifted and added, “This leads me to invite chemists, in order to pursue perfection in Art, to discover a way to ally dry pigments with a liquid that would replace oil, without oil’s disadvantages which include rising to the surface and distorting colors by absorbing their brilliance”. ¹⁵

The search for stability led, in fact, to an increasing concern over retouching in oils. The early 19th-century art press praised Pietro Palmaroli (1778–1828), an Italian restorer working in Dresden, for using watercolor on Raphael’s Sistine Madonna. His campaign was compared to an 18th-century effort where the retouching in oil had significantly darkened.¹⁶ We also find oil retouching cautioned against in restoration literature, a new genre that first appeared in the 19th century. The first books exclusively devoted to the restoration of paintings were detailed handbooks with bibliographies, discussions of stable pigments, precise descriptions of inpainting methods and materials, and multiple approaches to particular inpainting problems. The writers did recommend oil retouching since they felt oil paint was easy to manipulate, but they were deeply concerned about colors shifting.

When oil retouching was recommended, it was with caveats. Christian Kӧster (1784– 1851), a German painter and restorer, wrote the first book on the restoration of paintings, Ueber Restauration alter Oelgemälde, in three small volumes published between 1827 and 1830. Kӧster believed that you could use oil safely if you used only the purest materials and retouched over absorbent chalk fills.¹⁶ Ulisse Forni (c.1820–1867), the official restorer for the Florentine museums, published the Manuele del Pittore Restauratore in 1866. He was completely against the use of oils on tempera paintings, but felt that they were the easiest to use on large losses in oil paintings. ¹⁸

Whenever oils were recommended, furthermore, authors gave alternative materials and methods. Kӧster wrote that retouching with tempera and pigments ground in mastic were both possible substitutes.¹⁹ Henry Mogford recommended retouching in oil in his 1851 Hand-book for the Preservation of Pictures; Containing Practical Instructions for Cleaning, Lining, Repairing, and Restoring Oil Paintings. He added, however, that

“It has been suggested that the use of body colours with a water medium would be the safest to prevent the change of colours in the repairs. It might perhaps answer in the bright painted skies in landscape, water, and so forth, but it requires great artistic skill to manage its exact tone: there is no question of it being unchangeable, and this seems its only advantage. There is another method adopted in the restoration of damages, which consists of the use of powder colors with copal varnish, or varnish megilp.” ²⁰

Finally, Giovanni Secco Suardo (1798–1873), a northern Italian restorer and professional rival of Forni’s, wrote Il Restauratore dei Dipinti (1866–73). He described age testing a variety of drying oils, trying to find one that would not yellow. Since he did not succeed, he devised a drained oil inpainting system with a mastic/copal mixture as the medium. Secco Suardo felt that this system was very stable and wrote of it being unchanged on paintings he had treated twenty years previously. He acknowledged, however, that some of his colleagues did not approve of even drained oils. If one encountered this type of objection, Secco Suardo recommended tempera underpainting with only very thin drained oil glazes on top.²¹

Specific Training for Retouching

As is surely becoming clear, by the end of the 18th century, restoration was emerging as a distinct profession. For example, the Danish painter Jens Peter Mӧller was sent in 1810 by the Royal Danish Academy of Fine Arts to study restoration for three years at the Louvre in preparation for becoming the first official restorer for the Royal Paintings Collection.²² One of the results of professionalism was that specific training and skills became required for inpainting. Although a restorer was, of course, still expected to be a good painter, new requirements demanded a range of virtues, from patience, to humility, to subordination of the ego to the work of art. More important, though, Pietro Edwards and the Louvre commissions considered knowledge of old master techniques essential.²³ Jean-Michel Picault, a restorer and one of the commissioners, wrote, for example, that

“The restorer’s art and the painter’s art are in no way similar…a painter who can produce a masterpiece, will ruin masterpieces by another artist in trying to restore them…The most famous painter will substitute his style for Raphael’s…and the result of his retouching will be the creation of a monstrous work which will surely cause the painting to lose value…Successful restoration requires a specific course of preliminary study…The restorer, who studies every master and school is not prepared to be, and should not be prepared to be, like a painter with his own style. He has sacrificed his own ideas to submit to the ideas of another; he no longer exists for himself.” ²⁴

This relationship between understanding old master techniques and successful inpainting was refined in the 19th-century literature. Kӧster’s book contained sections specifically on retouching tempera painting,²⁵ and in this context he included a long chapter on tempera technique by guest author Jakob Schlesinger, a restorer of early tempera paintings. The back of Kӧster’s book even contained an advertisement for a translation of Cennino Cennini’s book. Kӧster’s approach was taken much further by Simon Horsin Déon (1812–1882), a restorer at the Musées Nationaux in Paris. The last section of his book, De la Conservation et de la Restauration des Tableaux (1851), was an overview of the techniques of European painting by school, country, and period with commentary on particular restoration problems and the best methods of approaching them.²⁶

Recommendations for studying chemistry began a bit later, in the 1820s. By the time Forni was writing in the 1860s, he recommended not only books on the chemistry of artists’ materials (i.e., G. A. Chaptal’s La Chimica Applicata alle Arti, 1820), but also technical publications such as the Complete Handbook for the Color Manufacturer and Varnish Maker (1850 and 1862). ²⁷

Development of “Less Is More” Principle

Despite (or perhaps because of) increasing professionalism, the 19th century was also the period of major restoration controversies. The most famous were the cleaning controversies at the Louvre and the National Gallery in London. Another contested area, however, was retouching, and conscious decisions were made to leave losses visible. Works of art were increasingly valued as historical documents, in part due to the rise of archaeology and the creation of national museums. Already in an 1830s restoration campaign of an Amico Aspertini fresco in Lucca, the local arts commission instructed the restorer to leave large lacunae visible if these areas could not be reconstructed accurately.²⁸ In the 1870s, the art historian Giovanni Battista Cavalcaselle (1819–1897) was responsible for restorations of art works belonging to the Italian State, and he consistently wanted structural stabilization only. In work on the Giotto frescoes at Assisi only neutral toning was allowed.²⁹ His 1877 regulations for restoration work undertaken by the State declared, “It does not matter if you recognize a restoration, in fact, you should be able to recognize it, since what is necessary is respect for the original work at least for works belonging to the State. A lie, even a beautiful lie, must be avoided. Scholars should be able to recognize in a restored picture what is original and what is new.”³⁰ Around the same time and in a similar vein, the chemist Max von Pettenkofer (1818–1901) in a quest for a “scientific restoration” also wrote against retouching, saying that, “Whoever would treat a collection of documents — such as our picture galleries should be in the future for the history of art — in such a way that even the most trained and acute eye cannot recognize what is authentic and what is imitation, deserves prison or the galleys because they would be guilty of forging documents.” ³¹

In response to writing and directives against retouching, the formulation of a “less is more” principle began to emerge. The authors of the 19th-century instruction books repeated this again and again. Kӧster wrote, “Retouching is the last step. The less work necessary, the better. And if nothing is necessary, do nothing.” Horsin Déon stated, “The best restoration is that obtained by a light, transparent touch which allows the master to appear wherever original still exists.” Forni concurred, writing, “The best restoration is obtained from economic work which leaves intact all the original parts.”³² Finally, the director of the Louvre wrote an article on restoration at the museum in the 1860s and clearly stated reasons to avoid excessive retouching: “Our mission is to show the masters as they are, to honor them with their strengths and weaknesses…discretely filling losses created by flaking paint without spreading color over the edges of the damage, this is all that restoration allows.” ³³

Conclusion

There is a long history of paintings restoration in Europe. There is also a long history of discussions, both philosophical and practical, on approaches to restoration. We often think of retouching from the past as being sloppy, heavy handed, and unsympathetic to the painting, making wholesale changes for reasons of taste and marketability. In fact, there are many examples of this type of work, but this is not the whole story. I hope I have shown that there were sensitive approaches to retouching and sophisticated discussions about retouching. I think this knowledge provides a certain depth and perspective to our own work as conservators.

Author’s note: This article, “Retouching Paintings in Europe from the 15th through the 19th Centuries: Debates, Controversies, and Methods,” originally appeared in AIC Paintings Specialty Group Postprints, Vol. 16, 2003, pp. 13–23.

Acknowledgments: I would like to thank the many people who helped and encouraged me with this project, especially Keith Christiansen, Dianne Dwyer Modestini, Catherine Metzger, Sue Ann Chui, and Carolyn Tallent, all of whom have heard and read multiple versions of this paper. I would also like to thank the staff at the Intermuseum Conservation Association for all their support during the final stages of this paper.

Wendy Partridge

Submitted December 2003

Tradition of Retouching Practices in America

Retouching, compensation, or inpainting as a professional practice did not fully take root in America until well into the 19th century. Prior to that, there are 18th- and early 19thcentury references to restoration by artists and teachers who restored, retouched, or repainted works, but they did not distinguish themselves solely as professional conservators or restorers until later.³⁴ Even then, the practice seems to have been on an individual basis. Several known itinerant professionals were contracted by various new museums as they sprung up around the United States in the mid- to late 1900s. The practice of conservation in America, or more specifically in this case of retouching or inpainting, was intrinsically tied to Europe.

Late 18th Century and First Half of the 19th Century

Charles Willson Peale and other artists of the 19th century treated their own pictures, often attempting to correct problems with defective materials. For instance, Peale makes mention of paints that have faded and are too cool.³⁴ Benjamin West, Thomas Cole, John Trumbull, and Rembrandt Peale all seemed to be addressing severe paint cracking, possibly from the use of asphaltum. References indicate that artists were overpainting these areas in an effort to obscure the problematic traction crackle.

In 1806, Benjamin West retouched some of his earlier paintings, apparently because of problems with cracking. On one of these paintings, West’s retouching and his second inscription clearly go over and into a pattern of wide traction crackle, indicating that concealing the disfiguring crackle was the likely motivation for the retouching.³⁵ In the late 1820s, both Thomas Cole and John Trumbull were doing the same and for the same apparent reason.³⁶ Between 1835 and 1840, Rembrandt Peale mentions having retouched his paintings three or four times during tours of the country.³⁷ In an early mention of inpainting as we know it, around 1849–52, he stated in his Notes of the Painting Room,

“I witnessed the mode practiced by a distinguished Professor at Florence repairing the Paintings belonging to the National Gallery. After cleaning the picture, the Cracks were neatly filled up with putty. When this was dry, the white lines were coloured by mixing the Colours, which were in impalpable powder & of suitable tints, with Copal Varnish — a little difficult to use but preferable to oil paints, which grow darker; with this varnish little or no change takes place.” ³⁸

Hudson River School artist William Rickarby Miller used a mastic and gum-water mixture.³⁹ Rembrandt Peale chose a copal varnish medium for inpainting.⁴⁰ Both Miller and Peale criticized the use of oil paint for inpainting because it stained and darkened over time.

In another passage from Notes of the Painting Room (1849–52), Rembrandt Peale implies that repairing a painting involves improving it:

“A good artist may submit to be employed to repair a bad picture, which comes from his hands with beauties that were not conceived by the Original Painter; not only in the retouching of forms, painting out defective parts, but with the magic results of toning & glazing. I have thus seen the celebrated Wertmuller repairing & improving (if I may venture the opinion), a damaged Wouvermans belonging to Wm. Hamilton of the Woodlands; and I think, in my own experience, I have several times been similarly successful.” ⁴¹

Second Half of the 19th Century

Frederick Church was documented as having retouched his paintings. In 1885, he was contacted by William Wilson Corcoran⁴² that his Niagara (1857) had vertical streaks through the sky; he responded that the “obnoxious” streaks were the result of the Winsor and Newton canvas where they used sugar of lead (lead acetate drier) to speed the drying of the ground layer. He said that the Winsor and Newton “Roman” canvas had these problems and that many years before he had repainted the sky of Niagara to cover streaks. He recommended taking the painting to Mr. Oliver in New York City to repair.⁴³

But a restorer had more respect for artists’ intent. George Howorth, a restorer and dealer working in the Boston area, wrote a book in 1859 titled Restoration of Oil Paintings: With a Few Practical Hints to the Owners of Pictures. Howorth complained that other painting conservators “daub, either leaving a picture with hideous patches of new paint, or proceeding, step by step, in the vain attempt to secure something like uniformity until they have actually covered up the last square inch of the original handiwork.” ⁴⁴ In contrast, he did not “paint over an old picture for the purpose of restoring it, unless it be in places where the original paint is gone.” Howorth and his son John are known to have treated a number of pictures belonging to James Jackson Jarves, the New England-born collector, whose collection became the foundation of Yale University’s early Italian painting collection. In a broadside, dated 1867, an endorser praises Howorth for being a “scientific man, aware of the latest developments in the restoration of paintings.” ⁴⁵

Jarves’ collection of early Italian paintings was purchased by Yale University and became the first important collection of early Italian paintings in America. Jarves himself apparently restored some of these works, as did a Greek restorer, Georgio Mignati, although details are sketchy.⁴⁶ He also weighs in on the debate over cleaning and retouching practice. In his 1861 book Art Studies, Jarves criticizes restorers of the day in this way: “Inadequate to replace the delicate work he has rubbed off, the restorer, to harmonize the whole and make it look fresh and new, passes his own brush over the entire picture, and thus finally obscures whatever of technical originality there might have still been perceived after the cleaning…Each restoration displaces more of the original, and replaces it by the restorer” and “[t]he true occupation of the restorer is to put the work given to him in a condition as near as possible to its original state, carefully abstaining from obliterating the legitimate marks of age, and limiting himself to just what is sufficient for the actual conservation of the picture.” ⁴⁷ By 1869, Jarves had softened his criticisms of 1861 by stating that the issues he had discussed were “more applicable to past than to present” and that “systems for a reform, founded on true artistic principles, [were] everywhere beginning.”. ⁴⁸

In 1869, Jarves published Art Thoughts, in which he described three modes of restoring pictures: “First, the only sensible one of such local repairs as are necessary to confine it to its original canvas or panel, or to transfer it bodily to a new one, leaving the age of the picture to speak for itself, and every drag of the artist’s brush intact.” The second was “the foolish mode of wholly repainting the original surface in various ways, the worst of which is by stippling, to the utter loss of those conditions of the painting which were a guarantee of its authenticity, and replacing them by the rarely better and almost invariably worse treatment of another hand.” And the third, “the dishonest method of restoration, done with the sole intent of deception. The original picture in this instance is either of no merit or it has been entirely ruined, with but few touches of a bare outline remaining. In the former event the intention is to make the work of an inferior hand or an old copy appear to be a veritable original of a great master, and, in the latter, to pass off the wreck of a fine picture for a perfect, intact one.” ⁴⁹

In 1878, Jarves wrote an article that appeared in the New York Times, in which he expressed an understanding of both sides of the debate over restoration; those who valued the early Italian paintings as aesthetic, whole objects, and those who viewed them as artifacts “best appreciated stripped of the accretions of time, dirt, varnish, and repainting.” ⁵⁰ He clearly understood both sides of the debate, advocating restraint in the face of works that had achieved “harmony with age, repaints and discolored varnish” ⁵¹ and yet he knew that every museum had paintings that would benefit from the removal of dirt and retouching, even if those paintings were to suffer an aesthetic loss from the process. ⁵² Jarves was aware of the professional practice of restoration, advocating the use of “varnish colors, because it would be easy to remove.” ⁵³

First Half of the 20th Century

While the debate over retouching continued, the 20th century saw the beginning of organized discussions about the conservation field in general and, more specifically, about inpainting. Some artists continued to feel they alone were qualified to effectively retouch works of art, leaving structural work and other stabilization to the restorer or craftsmen. The Fogg Art Museum at Harvard University was the first center for discussion of many conservation issues, including retouching. Technical Studies in the Field of Fine Arts was published by Harvard University, from 1932–42, serving as a forum for these and other discussions. Other important voices and practitioners emerged, both in the United States and abroad, as the century developed.

Starting in 1908, the New York City Business Directory listed Stephen Pichetto “as a restorer with an establishment on East 28th Street.” ⁵⁴ Pichetto worked as a consultant to the Metropolitan Museum of Art as well as the National Gallery in Washington, but he maintained a successful private practice, with clients as distinguished as Mellon, Lehman, Heinemann, Dale, Walters, Warburg, Lewisohn, and Guggenheim. He was also connected with Joseph Duveen, but most notable was his association with the collectors Samuel and Rush Kress, with whom he had a relationship for more than twenty years (until his death in 1949), advising them and treating many works in their collection.⁵⁵ Pichetto and his studio, in which he employed at least three other restorers, were exceedingly influential in the art world of the time. “Declaring that he ‘did not want to camouflage the damaged portions rather to retouch the missing portions with local color,’ Pichetto relied heavily on his three inpainters whose method was to apply colors in Winsor and Newton watercolors or in egg tempera, coat with French varnish (shellac), and glaze with dry colors in dammar varnish. Inpainting palettes included only seven colors, and varnish was done with dammar.” ⁵⁶

In 1914, Henry Ward Ranger, an American Tonalist painter, said,

“The mechanical part of the work, such as relining, transferring, removing varnishes, &c., should be done by one who has thorough knowledge of, and continual practice in, these processes. Such a man soon becomes more adept in the art than opportunity permits the painter to be. But, such parts of the work which call for a higher artistic sense, such as reglazing, retouching, etc., should be done only under the supervision of the best available artist of the school to which the picture is closest related. I am confident that no artist would withhold his services from such a worthy cause.” ⁵⁷

In other words, it was acceptable to leave structural work to another professional, but any retouching required an artist.

The artist Abbott Handerson Thayer had the most negative appraisal of retouching. In 1916, painting at the Freer, he left two decisive inscriptions on paintings: “This picture is never to be retouched — not one pin-point.” ⁵⁸

Harry Augustus Hammond Smith — who was a restorer for the Metropolitan Museum of Art, Chicago Art Institute, Cleveland Museum of Art, Minneapolis Art Institute, Worcester Art Museum, the New York Historical Society, the Frick Collection, the Morgan Library, Harvard, and the Yale University Art Gallery — apparently treated twenty-seven paintings in the Jarves Collection in the summer of 1915. He was reputed to be the “first man in this country to raise the work of restoring paintings to an art in itself, second only to that of the original genius.” ⁵⁹ Handwritten condition/treatment notes on Yale’s collection indicate insight and sensitivity of the work. He clearly treated paintings in a thoughtful, restrained manner, confining his retouching to scratches and minor losses where possible. He makes reference to “spots on the draperies and in the background which required retouching.” ⁶⁰ But where there was tremendous loss (Yale’s Botticelli), he clearly felt the need to retouch more extensively to restore aesthetic integrity. He wrote:

“The color had apparently been intentionally scraped off from the old ground, the remains of this color showing that it had been done with a sharp pointed instrument of some kind….The original drawing, the outlines, could be seen on the priming of the panel under a white ground and these where distinguishable were followed in restoring the hands. The darker shadow in the child’s left arm was also damaged by these same scratches. The sky was also much damaged on the right side requiring much retouching.” ⁶¹

Little conservation work was done to the Yale Collection between 1916 and 1949, except specific projects carried out by itinerant restorers.

In early 20th-century Italy, Cesare Brandi, the first director of the Istituto Centrale del Restauro, founded in Rome in 1939, codified a philosophy that retouching should attempt to integrate the work of art as a whole, but remain visible when viewed from near: “Restoration must aim to reestablish the potential unity of the work of art, as long as this is possible without producing an artistic or historical forgery and without erasing every trace of the passage of time left on the work of art.” ⁶² His ideas gave rise to modern theories of restoration. Paul Philippot, the former director of the International Centre for the Study of the Preservation and the Restoration of Cultural Property (ICCROM) in Rome, an art historian by training and an influential member of the conservation community with a comprehensive vision of conservation issues, made Brandi’s Teoria del Restauro accessible to professionals outside Italy. While conservators were developing and espousing theories and practices, it was the 1950s that saw the development of the profession as a distinct tradition in the United States, not merely an appendage to art criticism and the like. ⁶³

Paolo and Laura Mora, the most notable specialists in the philosophy and techniques of Cesare Brandi, described it as follows:

“Tratteggio is a system of small vertical lines averaging one centimeter in length. The first lines, which indicate the basic tone of the retouching, are placed at irregular intervals equal to the width of one line. Next, these intervals are filled with a different color, and then with a third color, in order to reconstitute the required tone and modeling by means of the juxtaposition and superposition of colors which are as pure as possible. Each line in itself should be of weak intensity, the desired intensity of the whole being obtained by the superposition of glazes of transparent lines rather than by strength of color, which would cause the retouching to lack the vibration indispensable for a good integration.” ⁶⁴

The Department for Technical Studies at the Fogg Art Museum, Harvard University, was founded by Edward Waldo Forbes in 1928. He, along with George L. Stout, Rutherford J. Gettens, and Alan Burroughs, as well as others, developed general conservation standards, new methods of treatment, and the theoretical foundations for treatments. R. Arcadius Lyon, painting restorer at the Fogg, introduced visible inpainting techniques in the United States. In the mid-1930s, Forbes states that all compensation should be noticeable. In areas of loss, he imitated the original tone and modeling with hatch strokes similar to those nearby. Because the tone was so accurately matched, the result was something one would not notice at first glance, but because the strokes were less fine and had less subtle transitions, the restored areas could easily be noticed by the observant viewer.,

Lyon discusses a concern related to wax lining that points to the use of watercolor for retouching: “The use of wax for lining presents another difficulty for the restorer, should he desire to retouch some of the wax infiltrated areas with water color. There is a grave danger of the peeling off of the retouchings, even if one has been successful in obtaining a seemingly good adhesion for the colors over wax.” Lyon also discusses the use of UV light as an aid to restorers to determine the extent of retouching, often when they are so skillfully done as to be nearly invisible to the eye in normal light.

Maximilian Toch in his 1931 treatise, Paints, Painting and Restoration, distinguishes repainting from inpainting per se by writing,

“It has been conceded by those who know, that no painting must ever be repainted but that the restoration should be confined to the removal of extraneous substances which have collected on the surface and after this has been done, the picture is varnished. Yet there are noteworthy exceptions to this practice, such as pictures which have suffered from fire, pictures that have been neglected to such an extent that the paint film has dried and fallen off in places, and these must be restored by repainting the damaged parts. Under circumstances like this the restorer is warranted in filling in these bad spots and restoring the picture back to its artistic condition. But the average painting, which has not been partially destroyed or damaged, must under no circumstances be repainted, otherwise the work of the master is altered and interfered with.”

By the 1930s and early 1940s, the debate was fully articulating itself. George Stout, who represented the United States at the International Conference on the Examination and Conservation of Works of Art described four procedures for resolving damages on paintings:

“One way, and much the easiest, is to do nothing at all. The painting, though it may be much changed by overpaint, is left alone and untouched. A second way is to remove from the original paint any crust of darkened varnish or overpaint that would cloud or hide it and to stop there, the blemishes left to appear as they may. A third is to put into blemished areas tones that do not make so sharp a contrast as the losses themselves with what is adjacent. A fourth is to renovate the whole thing, to make a new painting out of an old one. In this procedure signs of age are covered over, dents and cracks are filled, and the original paint serves merely as a kind of tonal reference for the new coatings that are laid on it.”

Stout believed that the first and fourth choices do not give the public a fair chance to see the painting itself. In this context, he refers to the eminent art historian, Max Friedländer:

“What shall the restorer do about the sadly fragmentary condition which he has, of course, not produced, but which he has brought to light? If he does nothing, he runs the risk of being accused of damaging the work and diminishing its value. Dare he replace what is lost by filling in and repairing?... The work of art is a document for the scholar, a source of pleasure for the amateur and an object of value to the owner. That the scholar will answer with a sharp “No” there can be no doubt. In his eyes, every restoration that goes beyond cleaning, preserving and uncovering is a piece of counterfeiting — whether successful or not…He wants to see as much as has been preserved of whatever the artist created…The connoisseur judges from a less firm point of view…He…fears that faulty spots, holes, conspicuous defects and disturbances of the continuity may spoil his enjoyment of it. In this dilemma, [he can] probably recognize the right of the restorer to fill in and repair, providing that he works in the spirit of the old master and is able to recreate what is no longer there. The task is technically, and from the point of style, unsolvable. Memling worked with certain pigments and materials; the restorer with others. And even if the restorer could proceed with exactly the same technique as Memling, he would still not be in a position to repair what is preserved of the original because time has been at work on it since Memling. That which is made is never exactly like that which has grown.”

George Stout goes on to explore what works well, using the second and third outlined procedures. He acknowledges that the issues of compensation are complex in nature and vary with the size and scope of the loss. He suggests that smaller losses, though great in number, may be less disturbing than larger losses, which interrupt the overall visual effect of the work. Stout, working with Alfred Jakstas and consulting with Morton Bradley (Bradley 1950, describes various losses and their impact on surrounding originals. He also comments that solutions might be differently appropriate, depending on the setting (public or private) for the work of art.

In 1942, Helmut Ruhemann put forth a procedure for examination of works of art and made this observation: “Experience has shown at least one advantage of retouching.” He said this in the context of one’s interest in truly understanding the technique of the artist, feeling that the technique of retouching should align with the technique of the artist and the period in which the work was painted. Once the use of ultraviolet light came into common practice for detecting retouching, Ruhemann felt that what had been viewed as an unethical practice of invisible retouching could no longer be viewed as such. He was explicit in his description of his materials and techniques: “The sur face [of the fill] should be slightly lower than that of the surrounding paint to allow for the thickness of the inpainting…Inpainting medium should look and behave like the original medium, but must not darken with age.” He appears to have used tempera for retouching and practiced invisible retouching, citing three important attributes necessary to successful imitation of artistic technique:

1. Luminosity: imitation of the surface to reflect light, as if holding it within.
2. Cool transitions: light paint over dark paint, the “turbid medium effect.”
3. Texture: on every painting, depending on the direction of brushstrokes and viscosity of each paint layer (Jessel 1976, 1–8).

The word inpainting is used for the first time in an American source by Richard Buck and George Stout. They define it as “compensation of losses with no new paint put over that which is old and original.”[ Inpainting is distinguished from overpainting, which goes “beyond mere compensation in areas of loss”; overpaint may have a “great deal of free invention.” Buck and Stout discuss how poor execution of the inpainting, even when confined to losses, can obscure the beauty and fine handling of the original. Buck and Stout believe that the critical question in the discussion of retouching is whether the present condition of a work of art reflects the artist’s intent, or whether his idea has been “marred or destroyed.” This has to do with distinctions to be found between paint put on by the artist and that of the restorer.

William Suhr — who was conservator at the Detroit Institute of Art from 1927 until 1945, who worked simultaneously at the Frick Collection in New York from 1935 to 1977, and who was conservator to many other museums and private clients in the 1940s—took photographs after cleaning and prior to retouching. He wrote, “Restorations should preserve the original and attenuate losses in a manner that will permit the observer’s eye to pass over gaps in the paint without distress. It is difficult to define techniques to achieve this end. Each undertaking requires a different method and ultimately success will depend upon the craftsman, his native ability, his experience and, especially his understanding of and sympathy for the style and quality of the original.”

Mario Modestini earned an international reputation for, among other things, exquisite, imitative retouching. His career spanned more than seventy years and the records of his complete or full restorations became part of the individual paintings and institutional records.“ (Dwyer Modestini 2003). In 1948, the Samuel H. Kress Foundation in New York called on Mario Modestini to conserve its vast collection of Old Masters.” (Bergeon 1990).

Second Half of the 20th Century

The second half of the century brought with it tremendous development in the field of conservation and the training of professional conservators. Discussions about retouching were in the forefront of concern. Conferences that probed research into materials, philosophies, and techniques were exceedingly important, as controversy was also ubiquitous.

At Yale University, mid-century brought the inception of a campaign of restoration, using “clear, philosophical guidelines” that dictated honest presentation, without the encumbrance of repaints. Working with Andrew Petryn (a graduate of the Yale School of Art who trained briefly in restoration at the Fogg, the Boston Museum of Fine Arts, and the Metropolitan Museum of Art), Charles Seymour Jr. directed a campaign of “modern” conservation practice. In 1952, the Yale Art Gallery mounted a small exhibition of thirteen recently restored works. A small, accompanying catalog titled Rediscovered Italian Paintings was published, and the New Haven Register, the local newspaper, reported that many aids of science were employed during the project. Each painting was x-rayed and then subjected to ultraviolet and infrared light to determine original paint, the extent of past damage, and the necessary steps for restoration. Retouches done after revarnishing were meant to be visible and reversible (Aronson 2002). After the 1952 exhibition, the project continued under Seymour’s direction, but the philosophy continued to evolve. When Antonio Pollaiuolo’s Hercules and Deianira was sent for conservation to the Museum of Fine Arts in Boston, it was returned cleaned and retouched, supposedly according to the 1951 model, but in actuality, three separate systems of retouching were employed: invisible retouching on Nessus’s arm, a system of crosshatching on a band of loss in the center, and simple toning and speckling on a larger loss. No treatment record is extant, but in 1961 Seymour spoke at the Princeton sponsored symposium, “The Aesthetic and Historical Aspects of the Presentation of Damaged Pictures,” where he explained his restoration policy, presenting four pictures from Yale and describing the method aimed at doing the minimum needed to preserve the visual unity of the artwork. Sheldon Keck, long-time painting conservator at the Brooklyn Museum who was subsequently director of the New York University’s conservation training program, sharply criticized Seymour for his views. But the campaign continued until 1971, shortly after the publication of a catalog of the collection. Many of the paintings were, in the end, left unretouched.

In 1959, Modestini, in collaboration with the National Gallery of Art and Dr. Robert Feller, began testing synthetic retouching media. PVAc AYAB was chosen as the best; Modestini felt that its low glass transition temperature indicated greater longevity (Dwyer Modestini 2003). He had devised a system for manipulating plastic and, after performing a series of natural aging tests, he was fairly confident that it would stand the test of time. “In fact, this restoration technique has not altered in over fifty years…and its effects are still easily reversible.” (Dwyer Modestini 2003).

Shortly after a fire at the Museum of Modern Art, in 1958, a conservation department was established. Jean Volkmer was the first staff conservator. “Our philosophy has been influenced by the people who trained us. Sheldon and Caroline were influenced, in turn by Alfred Barr, who was then the Director [of MOMA] and one of the founders and great scholars of modern art. Alfred was very adamant about having paintings look to the eye as the artist intended. He did not believe in leaving fillings that are unpainted, nor did he like the idea of crosshatching.” Volkmer went on to describe her choice of materials: “The one difference, of course, is that we never use the same medium of the artist, with possibly one exception, and that would be pastel…For paintings, we use our own pigments ground in methacrylate or in poly (vinyl acetate). Recently, we’ve also used Magna (Bocour), a commercial paint of pigments ground in methacrylate. It sometimes gets pretty impossible to duplicate colors, and we sometimes modify [by] dry color. But in no instance have we ever resorted to using oils. Anyone looking at our pictures under ultraviolet would detect the inpainting.”

The pivotal 1961 conference, sponsored by Princeton University, gathered other art historians as well as conservators, including Sheldon Keck, Philip Hendy, Richard Offner, and George Stout. A wide spectrum of ideas on the ethics of “deceptive” inpainting, practical matters involved in exposing losses, and so on was presented.

Philip Hendy, Director of the National Gallery in London, promoted vigorous training in the methods of artists and techniques of past traditions. He recognized the importance of the professional restorer as “essential to all historians of the art of painting.” (Hendy 1931). He continued by saying, “If we are going to have retouching at all, we must have it most of the time; and those who do it must go through all that rigorous training in the exact matching of tints and textures, which is needed for good restoration…We have to consider the context of every work of art, the purpose which it is serving, the nature of the public concerned with the result…I am going to postulate as a first principle in this question of retouchings that the public should be told at least in broad terms where they are, how extensive they are, and why, if it is not obvious in the context, they are necessary.” (Hendy 1931, 142-143). He went on to say, “No retouching must be allowed to overlap any original paint surface…the past practice which had only one end in view: to cover up the fact that the picture was damaged. This led to much overpainting of original paint…It is not only overpainting, of course, which has to be removed…The excessive overpainting in the past led to a reaction in the opposite direction, first manifested, as far as I know, in the twenties: that it is unethical to do any retouching at all…But there are many possibilities between the two extremes…, ultimately advocating bringing the picture to the point where it can be appreciated as a whole.” (Hendy 1931, 145). Hendy remarked, “Much of the more conscientious retouching of the past has been done with an oil and varnish medium (with varying proportions of each) and this has usually darkened seriously with time. It has to be removed and the retouching done again in some medium less subject to change. In other words, the picture has to be cleaned, and cleaned completely, so that the new retouchings may be matched exactly to the original paint. This is still by no means universal practice.” (Hendy 1931, 143).

At the same conference, noted art historian Richard Offner countered the argument for any retouching as follows:

“The chief reason for the problem is to be sought in the universal greed for completeness…But the work of art will suffer even more if the missing part is replaced. Generally, the replacement is without specific artistic intention and performs no positive artistic function other than that of covering up unsightly lower layers of a painting, or simply that of making it appear that the mutilation was not there. In most instances the restorer feels under obligation to supply the missing parts by guessing what was previously represented where the gaps now occur…Any restoration, therefore, that introduces paint or shape within its boundaries, even if the restoration be limited to the missing portions alone, must prove intolerable…Repainting adds the factor of personality — and possibly of intention — extraneous to the original work. Not the least objection to such additions is the resulting dissonance of color, since the pigment of the restoration is generally of a chemical composition different from that of the preserved portions, and its color tends in time to drift away from that of the preserved color.”

Millard Meiss, a protégé of Offner and noted scholar, then rendered his opinion, coming around a bit and referring to the work of George Stout, Cesare Brandi, and A. and P. Philippot, but suggesting that further investigation and discussion are needed. He stated, “The unreconstructed lacunae remain, of course, obtrusive, ugly shapes. Indeed, they must, I believe, retain a considerable degree of discreteness; otherwise, in the process of perception, they will merge with the surviving original areas and contaminate our response…It is obviously desirable to reduce the obtrusiveness of lacunae and their emergence as figures, but it is still more desirable to avoid their absorption in the image…Treatment of a loss in any given case should seek as recessive a surface as is consistent with the maintenance of its independence from the surviving areas at normal viewpoints. Its very un-assimilability is a measure of the integrity of the original. A damaged painting does not offer a good and a bad alternative; only a better and a worse.”

At the same conference, Paul Coremans, noted Director of the Royal Institute for the Study and Conservation of Belgium’s Artistic Heritage, outlined a thorough, five-pronged approach with guidelines addressing the issues:

A. Precautions to Be Taken Beforehand

B. Aesthetic versus Historical Aspect

C. Aesthetic Aspect and Formal Reconstruction

D. Impermanence and Vulnerability of Reconstruction

E. Possible Deductions and Suggestions

Under A: Coremans encouraged differentiation between the original and later additions as clearly and completely as possible. He felt that a distinction also had to be made between the original state and the actual state of a painting, the actual state being the original state but somewhat altered, patinated by time, with a definitive determination of the exact chronology of any additions. Under B: Coremans acknowledged art historians whose first consideration in regard to an old painting is its significance and importance as a historical witness, mainly concerned with the knowledge and the teaching of objective truth. For them, according to Coremans, the painting should be shown in its actual state without attempting a reconstruction that can be arbitrary: later additions should be removed and the picture should be exhibited with all its mutilations and losses. For many others — the creative artists and numerous gallery directors, for instance — the painting remains, above all, a work of art. As a consequence, Coremans asserts, the aesthetic beauty and plastic unity must be reconstructed.⁷¹ Under C: Coremans defined the work of art as a unity of parts, a totality that cannot be broken. Reconstruction is an act of restoring formal continuity and aesthetic structure.⁷² Under D: Coremans spoke of the fashion of a period for a certain solution, in contrast to what may be deemed appropriate later on. Here, he clearly advocated the concept of reversibility. Finally, and not surprisingly, Coremans concludes by underscoring the need for a complete scientific understanding of a work of art, as well as complete documentation, whatever the chosen course of treatment.⁷³

Sheldon Keck’s remarks at the conference were particularly significant. “Works of art, although constructed of matter, are imbued by the artist with an immaterial content variously ascribed as aesthetic, emotional, intellectual, or spiritual. They have, therefore, both objective structure and subjective significance…The completeness of the visual communication between artist and spectator depends also on the totality of the painting as originally designed by the artist. Damage or alteration of the physical structure modifies the character of its communication.” ⁷⁴ Keck further justifies retouching, alongside the work of the art historian:

“Whether or not compensation for losses is made, as well as manner and extent of the compensation, depends on the present or possible future use to which a painting is being put. The kind of compensation depends on the technique and style of the painting, the relative size of the losses and their position in the design…The restorer who removes later accretions or compensates for losses, like the art historian, makes a critical comment on the painting. The restorer’s comment is directly on the painting. As critical comment, cleaning and compensation are more subjective than objective, more an art than a science. No two restorers would produce exactly the same result.” ⁷⁵

Keck’s remarks on the practice of ethical inpainting, later incorporated into the AIC Code of Ethics and Guidelines for Practice, are particularly significant, stressing the importance of reversibility and documentation:

“What seems abundantly clear to me as a conservator is that the extent and kind of compensation depends on the requirements of the individual painting, its style, technique, type of damage, and present use. A wide variety of compensation is available. He who makes compensation, conscious of his responsibility and limitations, must not only keep it to a minimum but allow for its future removal. A photographic record of the painting’s actual state before compensation is essential.” ⁷⁶

Sue Sack, who was trained by Sheldon and Caroline Keck at the Brooklyn Museum, said the conservators at the museum operated on the premise of the “aesthetic whole.” The inpainting could show close up, but one needs the aesthetic experience. They would inpaint areas of bad traction crackle and then would stand back to judge how far to go. They would inpaint in a lot of light and, given that gallery conditions had less light, they would satisfy the principle of the viable aesthetic whole. Reversibility was always a premise. Sack related Keck’s vision that general toning that might be acceptable on a fresco was not so on an easel painting. More is required when the work is to be viewed at eye level.⁷⁷ Sack said they worked with a lot of living artists and curators. In correspondence with Caroline Keck, she said that the publications of Julius Held were very influential in the conservation world. He published a book in 1963 titled Alteration and Mutilation of Works of Art. Sack said that for inpainting, they would use pigments ground in dammar or other media (PVAc AYAA, AYAB, or AYAF). They would build two or three layers of retouching, rather than underpainting, sandwiching layers with appropriate resins. Magna Colours were not preferred. The palette was considered limited and the texture “stringy.” ⁷⁸

John Brealey was Chairman of Painting Conservation at the Metropolitan Museum of Art for fourteen years, from 1975 until 1989. His effect on attitudes and practices in the United States was transformative for more than a generation of painting conservators. According to Stephen Kornhauser, who worked with Brealey from 1980 to 1982, “John seemed to relate inpainting to the picture rather than just color matching. He forced you to look at the whole painting, as well as the losses. You were pulling the picture together rather than filling holes with paint. He believed that inpainting was ruled by forms even more than color and that once the forms were completed [matched] your eye could be tricked even if your color was slightly off. Both with cleaning and inpainting he’d say that the picture would tell you what to do.” ⁷⁹ Kornhauser went on to describe materials used:

“We used extruded oils and a titanium white ground in dammar/turps and a little egg yolk [to give a crisp line]. We underpainted in watercolor, gouache and at times even egg tempera [compacted with a burnisher at times]. With gouache, we’d underpaint lighter and cooler and thinly glaze with the oil colors. For the most part we stayed away from synthetics though we varnished with Winton Picture and Retouch Varnishes. Though we worked with oil colors, he [Brealey] said there were times that you might have to use synthetics. He liked the oil colors and natural resins because he said you couldn’t get the richness and handling in synthetics that you needed for old master paintings.” ⁸⁰

The 1976 AIC Conference in Dearborn, Michigan, included an important panel discussion on inpainting. Among those presenting were Peter Michaels, Elisabeth Packard, Lawrence Majewski, Robert Feller, Ursus Dix, Bettina Jessell, Mary Lou White, and Louis Pomerantz. Elisabeth Packard said,

“Before synthetic resins began to become available in the 1930s, inpainting mediums, used separately and in combination, included egg tempera, watercolor gouache, gums, glues and casein; beeswax and natural resins such as mastic, dammar and even shellac. The most popular, of course, were drying oils, linseed, poppy seed and walnut. More enlightened restorers chose aqueous mediums or mixtures which would enable the retouchings to be easily removed. Then, as now, much depended on the skill and judgment of the conservator. Although overpainting was prevalent, the report of the 1930 Rome conference indicates that museum conservators like Helmut Ruhemann recognized the importance of preserving the integrity of the original.” ⁸¹

Lawrence Majewski described stability tests that he performed on certain inpainting media, through rapid aging. Pigments were painted on Masonite™ panels on lead white and linseed oil grounds. These “paintings” were damaged and then given to nine conservators to fill and inpaint, recording their treatments. These were then exposed to changes in relative humidity and UV light. Changes in varnishes as well as the inpainting were reported.⁸²

Robert Feller spoke of the problem of whitening of retouches. Mario Modestini and his associates Henry Hecht and Claudio Rigossi prepared test panels using retouching systems with “dammar varnish, egg tempera, dammar and egg tempera and poly (vinyl acetate) AYAB mixed with several colored pigments and three whites: white lead, zinc white and titanium white…Results pinpointed dammar and zinc white as being particularly prone to whitening…The exposure tests also clearly demonstrated the superiority of the polyvinyl-acetate retouching system that has been used by a number of conservators for more than two decades.” ⁸³

Ursus Dix described the practice of inpainting with egg tempera, standard in many conservation studios in the 1950s. He detailed the recipe and technique:

Pigment, dry or ground in water, or reliable egg temper colours in tubes (Neisch, Dresden — unfortunately no longer available)

Egg yolk

Water

Shellsol 715, or comparable mineral spirit

Other egg tempera recipes include drying oils or varnish solutions (dammar, MS2A, n-butyl methacrylate).⁸⁴ Bettina Jessell presented Helmut Ruhemann’s inpainting techniques, listing specific materials and sources.⁸⁵ Mary Lou White described tratteggio: Brandi’s theory, materials used, sources and techniques.⁸⁶ Finally, Louis Pomerantz outlined his working methods:

1. Voids are filled with zinc white and other dry pigments (to match closely surrounding color areas), mixed with Elvace #1874 (Dupont) acetate copolymer emulsion.
2. Varnish is applied to entire painting by brush, with Soluvar Gloss/Matte (Permanent Pigment Co.) acrylic resin varnish in mineral thinner solvent.
3. Inpainting is executed with Magna (Bocour) methacrylate resin paint suspended in mineral spirits, placed in aluminum weighing dishes and allowed to dry. Xylene is then used as a diluent. (Magna uses Rohm and Haas Resin F10).
4. Isolating varnish for inpainting is Magna Varnish (Bocour), acrylic in alcohol, applied by brush locally, or as a continuous coating over the entire surface if necessary. (Magna Varnish is Rohm and Haas resin XR31).
5. Final varnish of Soluvar is brushed on.⁸⁷

Author’s note: The author is grateful for Stephen Kornhauser’s support during the writing of this chapter. He arranged for and participated fully in the interview of Susan Sack. In addition, Stephen provided key editing assistance.

Patricia Sherwin Garland

Submitted April 2010

Italian Differentiated Inpainting Techniques: A Review

Introduction

---

In early 20th-century Italy, new approaches to art restoration were influenced by several social, political, and cultural factors. One important new restoration approach concerned reintegrating paint losses in a way that could be differentiated from the original paint surface. Differentiated methods of inpainting developed in Italy were largely based on the philosophies of Cesare Brandi (1906–1988). The Roman restorers Paola and Laura Mora were among the first to bring Brandi’s theories of restoration to practice and developed the technique of loss compensation called tratteggio in the 1940s and 1950s. Later, in Florence, Umberto Baldini and Ornella Casazza created derivations of tratteggio called astrazione cromatica and selezione cromatica. Today, many practitioners of differentiated inpainting have expanded their techniques and their choice of materials to customize their approach as necessary.

Background

Several factors set the stage for new approaches to the reintegration of paint losses in Italy. The development of art history as a field of study in the late 19th and early 20th centuries, and the growing connoisseurship of early Italian art, enabled historians to become increasingly aware and critical of past heavy-handed restorations. These restorations often included reworking that confused or completely concealed the artist’s intent. Both the strong interest in the study of early Italian art and the drive to keep restoration distinguishable from the hand of the artist were generally consistent with aspects of the social and political environment of Italy in the 1920s and 1930s. The post-World War I environment was fueled by strong nationalism as well as what painting conservator Joan Marie Reifsnyder describes as “a new Italian consciousness, aware of its tradition and resolute in its civil mission” (2003, 25). With its push to establish order and unity after the war, the Nationalist regime in Italy imposed moral and ethical standards on public and private life (Reifsnyder 2003, 29). It was within this environment that Cesare Brandi’s philosophies on art restoration were formed.

Cesare Brandi

In 1939, Cesare Brandi became the founding director of the state run Istituto Centrale del Restauro, which established the first formal state restoration school in the country. The standardized restoration training the school provided was analogous to the rigorous standardization being employed in Italian political, educational, and cultural spheres (Reifsnyder 2003, 29–30). Brandi strongly believed that everyone should have the opportunity to have direct contact with art and stressed that civilization has a “moral obligation” to save works of art for future generations. Brandi’s philosophies concurred with ideas formed in Italy in the late 19th and early 20th centuries that began to favor minimal intervention and respect for the object as a historical document. From this framework, Brandi began to develop and expand principles to guide the restoration of artworks.

From the 1940s to 1960s, Brandi wrote essays on the restoration of artworks, which were published in his 1963 Teoria del Restauro (Baldini ). Stressing the need for balance between the aesthetic and historic aspects of a work of art, he wrote, “Restoration must aim to re-establish the potential unity of the work of art as long as this is possible without producing an artistic or historical forgery, and without erasing every passage of time left on the work of art” (Bandini). He believed that any integration must be differentiated from the original, but not be disruptive to the potential unity of the piece. Therefore, at the general distance from which the work of art will be viewed, the integration should be invisible, but it should be immediately recognizable at a closer distance. He felt that the previous use of a single neutral tone in the paint losses was not sufficient at reestablishing pictorial aesthetic unity, as it visually would not seem to rest on the same plane as the image.¹ Besides the use of a single neutral tone in losses, sometimes mottled with dots of slightly darker color, other early approaches included a technique called “simplified chromatic volumetric completion” (Ciatti 2003, 195),² graphic reconstructions (Ciatti 2003, 194) and otherwise imitative approaches made differentiable by such methods as a network of lines over the surface or keeping areas of retouching at a lower level than the original paint.

The Moras

Laura and Paolo Mora worked closely with Brandi’s philosophical requirements of reintegration and developed the technique of tratteggio in the 1940s and 1950s. The Moras, with Paul Philippot, wrote, “Reconstruction in tratteggio consists in transposing the modeling and drawing of a painting into a system of hatchings based on the principle of the division of tones” (1984, 309). Tratteggio is meant to be used for reintegration of losses with a neatly defined outline. Traditionally, hatchings in watercolor are applied on a smooth white fill that is level with the original surface. The following are excerpts from the Mora’s book with Philippot, Conservation of Wall Paintings:

“Tratteggio is a system of small vertical lines averaging one centimeter in length. The first lines, which indicate the basic tone of the retouching, are placed at regular intervals equal to the width of one line. Next, these intervals are filled with a different color, and then again with a third color in order to reconstitute the required tone and modeling by means of the juxtaposition and the superposition of colors which are as pure as possible. Each line in itself should be weak in intensity, the desired intensity of the whole being obtained by the superposition of glazes of transparent lines rather than by strength of color, which would cause the retouching to lack the vibration indispensable for a good integration.”

In order to obtain neat hatchings, without discontinuity and without the formation of drops at the bottom, the following procedure is recommended:

1. The brush must be sufficiently loaded to trace a full line without letting any paint run. To achieve this, as soon as the color has been taken, the brush should be wiped over an absorbent material such as slightly wet cotton wool, which has been fixed to the bottom right-hand corner of the palette. This operation controls the load of the brush by discharging it if necessary and, by its spiral movement, gives the brush the perfect point it had lost while preparing the tone on the palette.

2. It must be pointed out that the use of a mahlstick is essential. It allows the correct movement of the hand which, while the upper part of the forearm remains still, must cause the point of the brush to trace an arc intercepting the picture plane along the length of the line, so that the line begins at the top and ends at the bottom in a very sharp point. (309–310)

In Conservation of Wall Paintings, the Moras describe various types of losses that are distinguished by size, location, and depth: the wear of patina; the wear of the pigment layer; the complete losses of pigment layer that are limited in surface area and capable of being reconstructed; the complete losses of pigment layer that should not be reconstructed due to their extent and/or localization; and the losses of considerable extent that nonetheless should be reconstructed (because of architectural significance). They advise first treating minor disturbances such as wear in patina and wear in pigment layer so that larger losses may be better assessed. They recommend reintegrating wear in the pigment layer with localized glazes (velaturas) in watercolor slightly lighter and cooler than the original for the purpose of distinguishing the restoration. Although this publication was written for wall paintings, many of the principles and techniques have been applied to easel paintings as well.

The Moras state that the reconstruction of missing parts ceases to be justified when the design becomes hypothetical, and when a loss exceeds a certain size or when smaller losses are too numerous. Their general guideline for non-reintegratable losses is when they constitute more than 20 percent of the original or when losses are located in important areas of the design, such as facial features. They suggest that other methods of identifiable retouching may be preferable to reduce the disturbance caused by these losses. For unreconstructable losses in wall murals, they recommend imitating the underlying arriccio layer by filling losses to the same level of the original arriccio using materials of the same color and texture. However, sometimes a non-reintegratable loss was left as found with panel or canvas exposed (Ramsey 2003, 11).

Florence

Different approaches to inpainting arose in Florence, although they are also based on Brandi’s theories. Florentine restorers Umberto Baldini and Ornella Casazza developed the reintegration techniques called astrazione cromatica (chromatic abstraction) and selezione cromatica (chromatic selection). These were described in Umberto Baldini’s La Teoria del Restauro e Unita di Metodologia, which was published in two volumes in 1978 and 1981. Astrazione cromatica is used for large losses located in important areas of the painting to diminish the visual disturbance of the lacunae without reconstructing form. Selezione cromatica is used for smaller losses, where the reintegration of design is straightforward. Often both types are used on the same painting as dictated by the types of losses. For each, the fills are made level with the original surface and then inpainted with pure watercolors.

The lines of astrazione cromatica are applied in a crosshatching manner as opposed to the hatching manner of trateggio. Marco Ciatti explains, “This layer is applied in a four-color scheme derived from the additive synthesis of colors (yellow, red, green, and black), carefully graded to attain the mean chromatic value of the entire painting” (Ciatti 2003, 197). Baldini and Casazza (Baldini and Casazza 1983) describe the process:

“The first hatching will be complete in itself. It will cover all of the white of the preceding gesso, and will leave a nearly vertical course. The second hatching will be — since it crosses the first — applied lightly horizontally (this hatching will not be held to the form of small stiff segments, but will flow with free and spontaneous gestures of the hand) separated from the first, which will only overlap on the connecting point of the segment. The third and fourth hatchings will logically bring out the other colours, other patterns, and other interweavings, which mix in the eye and bring out a color tone which originates from an analysis of the colors occurring in the picture. (46).”

This objective approach, intended as a “neutral link” (Baldini and Casazza 1983, p. 43), uses equal amounts of the same chosen colors in each area of loss of a painting. This technique was used on artworks damaged in the 1966 Florence flood and was a standardized approach that could be taught quickly. An example of this technique can be seen on the Cimabue Crucifixion in the Sante Croce Chapel.

The lines of selezione cromatica are applied parallel to each other as in tratteggio but, unlike tratteggio, they can change direction to follow adjacent forms. The colors are chosen as in tratteggio, by optical mixing of colors in the vicinity of loss. This approach allows for a greater variety of effect. Marco Ciatti noted that this technique, in combination with astrazione cromatica, can be seen in the Master of Citta di Castello’s Maestà in the Pinacoteca Civica of Castello (2002, 198).

More Recent Developments

Today, tratteggio is often used as a blanket term to cover all types of discernable Italian reintegration and is colloquially referred to as rigatino by conservators (Ciatti 2003, 197); Lignelli 1997, 97). Some conservators are moving away from fixed theories of the past and incorporating updated versions of rigatino, such as varying size and thickness of the juxtaposed vertical strokes in response to loss size and location, slanting the line to increase visual texture, and employing direct color matches through the use of mixed colors (Lignelli 1998). Other options include using textured fills as well as using resin-based paints either glazed on top of a flat base tone of watercolors or gouache, or used as the exclusive colorant. Sometimes mimetic and differentiated approaches are combined (Olsson 2003).³ Examples of rigatino by conservator Teresa Lignelli can be seen on the Philadelphia Museum of Art’s Pietro Lorenzetti, Enthroned Virgin and Child with Kneeling Donor (see Lignelli 2002, 185; Lignelli 1997, 102). Another approach, more closely related to selezione cromatica, can be seen in areas of restoration in the grass in Sandro Botticelli’s Coronation of the Virgin in the Uffizi (see Ciatti 2002, 201).

These expanded allowances lend more flexibility and greater variation of effect. Painting conservator Nina Olsson observes, “The plurality of choices before historians and restorers today also bespeaks an attitude that allows for varied conditions, historical stratifications, and diverse locations and purposes for the viewing of paintings” (Olsson 2003, 10). Rigatino can be successfully used on paintings from other time periods and originating geographic locations, but it is best on panel paintings and is the least successful on paintings with textured canvas supports (Lignelli 1998). As always, restoration choices are made on a case-by-case basis, and certain factors, such as the rarity of the work and context of loss location, play a part in the choice of approach and material (Ciatti 2003, 202 and 206).

Pamela Betts

Submitted December 2006

Individual Techniques

Bettina Jessell

Introduction

The methods Bettina Jessell used were very close to what she learned from Helmut Ruhemann during the time she spent working with him. She did not change her techniques to any great extent, except for the fact that the painting materials were different and that naturally wrought some changes in her methods. But overall, she was still guided by the idea that it is the conservator’s duty to restore, to the greatest extent possible, the quality of the painting. She believed that the best way to achieve that end was to study the technique of the painter and to match reconstructions of losses to the layer structure created by the painter (Jessell 1976). A conservator must try to restore the way the painter did his painting. If you understand what the painter is working toward, it suddenly becomes easier to inpaint and to match the painter’s colors and technique. It takes some time to come to understand just how the painter achieved his results, Jessell noted, but it is time well spent.

According to Jessell, any discussion of inpainting must consider the state of a painting before that process begins. Many factors, unrelated to choice of materials or actual techniques, contribute to whether or not an inpainting job is successful. The success of inpainting is dependent on the structural integrity of a painting and the quality of the surface. Lacunae in the canvas and other structural problems that interrupt the continuity of the painting’s surface must be successfully dealt with before attempting to inpaint. If necessary, the painting must be lined, mounted on a new stretcher, and keyed-out. If the existing stretcher is in acceptable condition, it may be retained.

Surface Preparation

Although contending with structural concerns is important, Jessell strongly emphasized the need to thoroughly remove a discolored varnish — because only by cleaning a painting completely can a conservator see the painter’s intended colors. And it is based on knowledge of the painter’s colors, combined with an acknowledgment of changes in the colors potentially wrought by time, that a conservator attempts to match colors. Jessell pointed out that cleaning must be done very conscientiously. Too often paintings are not cleaned completely; many conservators are understandably frightened at the prospect of cleaning a painting thoroughly. But unless one does clean thoroughly, one never approaches the painter’s original intention. The wrong colors are simply perpetuated.

The preparation of the surface wherever there may be losses is as important as conscientious cleaning. After applying a preliminary layer of Ketone Resin N varnish, Jessell used an English product called Brummer Stopping to fill losses. She found this mixture of calcium carbonate, glue size, and gelatin better than any acrylic filler she tried. After the filler dried, she shellacked the actual filled area, let the shellac dry, and wiped away the excess fill material spread by the balsa wood. She tried to match the texture of the canvas and paint layer as closely as possible while doing the fill. The texture of thickly painted passages could be matched in this way.

Jessell explained, however, that because the fill material was rather brittle once it dried, it was not suitable for building up any significant impasto. If the surrounding area was one of moderate to high impasto, she built up the impasto with alternating layers of the inpainting medium and localized isolating varnish, allowing each layer to dry before applying the next, and finishing with a layer of varnish. She found this a tedious method, but a successful one.

Choice of Pigments

Once the conservator is satisfied with the surface, inpainting can begin. Choice of inpainting pigments and media affects not only the immediate success of inpainting, but also the long-term success of the job. The use of materials closest to those the artist used is appealing, but the reality is that the materials used by the artist often prove unsuited to the different needs of the conservator — most critically, the need for stability. A compromise must be made.

In keeping with the methods of the old painters, Jessell used ground pigments. She found that there were about thirty different pigments that were useful, but she might only use four or five, in addition to ivory or bone black and titanium white, on any given painting (Jessell 1976). Many of the pigments are the same ones used by the old masters. Others are modern equivalents to pigments that are a great deal more permanent than the pigments they have replaced.

Opaque Pigments	Semi-Opaque Pigments	Translucent Pigments
titanium white	cadmium yellow	Indian yellow
Venetian red	cadmium red	viridian
vermillion	cadmium red deep	cobalt blue
yellow ochre	light red	alizarin crimson
Indian red	ultramarine	bone black
raw umber	cobalt violet
burnt umber	Naples yellow
Prussian blue 1	raw sienna
ivory	cobalt
cadmium orange	Monastral blue
Winsor blue	burnt sienna
chromium oxide
NGL opaque green

The type of palette the conservator uses is also important. Jessell had a fairly coarse ground glass palette and little containers for each pigment, and she had all the pigments in a row around the glass palette. The ground glass ensured that the powdered pigments were thoroughly ground into the paint. Jessell found that if you mixed the powdered pigments into the medium, the result was smoother on the ground glass palette than it would be if you were to use an ordinary glass palette or a metal palette. She mixed her paint with the same brush she used to inpaint.

Choice of Pigments

The choice of inpainting medium is one of the most interesting, and historically problematic, decisions to make. With each medium, the inpainting technique changes in terms of how the paint is applied and how the colors are matched. The conservator must take into account how the paint adheres to the surface of the painting, the extent to which the medium is covering or translucent, and the extent to which colors change as the medium dries and the painting is varnished. Jessell used ParaloidⓇ B-72 for oil paintings and used xylene as a thinner. She found that Paraloid B-72 was the best medium for imitating the translucency of oil paintings (Jessell 1976). She still used egg tempera, which she made herself from yoke and white with wax and ox gall, on tempera paintings (Jessell 1976). The wax added elasticity, and the ox gall served as a wetting agent.

When she first began to work in conservation, with Ruhemann in the summer of 1938, most commercial conservators inpainted with oil paint. Tempera was the other option, but many saw it as difficult to use and very slow. The problem with oil is that it begins to oxidize immediately, and the effect of the discoloration, at first negligible, compounds with the passage of time. Jessell explained that you will still see paintings with brown patches, and you will know that this is where the painting was inpainted because the oil has gone brown.

Because oil inpainting discolors, Jessell and Ruhemann used to use tempera for inpainting oil paintings, because no other available medium was stable. Working in tempera is a time-consuming and difficult process. It takes three or four times as long to inpaint in tempera as it does to inpaint in oil. It is slow because you first fill the loss, then put the ground layer on, and then test it with mineral spirits — because mineral spirits will change the color the way varnish will. And then, if the color is not right, you change it until it matches the ground color exactly. The actual inpainting is a similar process often involving two or more attempts to get the color just right. You put the paint on, allow it to dry, burnish it, try it, see that it is the wrong color, and begin again.

Early Italian tempera paintings on panel are constructed differently and thus require a different inpainting technique. The painter first covered the panel with a white gesso ground and a preliminary layer of terra verte. The modeling was then applied, in thin streaks of paint. Tempera paint is very covering, so all the layers must be visible to achieve subtle transitions. Following the painter’s technique, the conservator reconstructs losses using the same layer structure (Jessell 1976).

Inpainting exclusively in tempera was an extremely slow process, but before acrylics were invented, there was no acceptable alternative. During the last six months that Jessell worked with Ruhemann, in the winter and spring of 1939 and 1940, they began using acrylics on an experimental basis. They were excited because acrylics enabled them to work with greater speed. The problem, of course, was that acrylics are not always sufficiently covering, particularly on old master paintings, to easily obscure the filler or ground. Quite often, paintings are reluctant to accept the acrylic layer because it is rather translucent; it nearly always requires two or three layers. The conservator must try it out to see whether it’s likely to match. Regardless, it’s much quicker to inpaint in acrylics than to inpaint in oil or tempera. And, thus far, acrylics seem to last a long time. A painting inpainted twenty years ago still tends to look acceptable. With acrylic, colors change to some extent, but they do not change nearly as much as they do with oil paint.

Jessell found that there are a number of other good inpainting media, that it’s a personal choice, very much determined by the painting one is working on. Jessell used, on occasion, Maimeri paints. She also used watercolors, particularly to execute fine, raised brushstrokes in early paintings, although watercolor darkens significantly when varnished (Jessell 1976). And she used colored pencils as an aid to inpainting, particularly in areas where overpaint would be too heavy or where a fine line is required (Jessell 1992). Still, she generally used permanent powdered pigments, ground into tempera or Paraloid B-72. She believed that to be the most satisfactory way of achieving a good result. There are, of course, commercially prepared tempera paints available — PlakaⓇ colors come to mind — which are covering and behave like tempera, although they are not the true tempera.

Color Matching

One of the basic skills that a conservator must master is the ability to match the colors used by the painter. Matching the actual colors is not the difficult task. The essential challenge, according to Jessell, is that the painter’s colors have aged and the conservator must match the aged and oxidized paint, while knowing that his colors will also change with time, however slightly. Oil paintings, because of the oxidation problem, present the greater challenge. Oxidation affects the whole aspect of a painting. Modern oil paintings look so much more striking than older paintings because oxidation has only just begun. However much you clean a 17th-, 18th-, or 19th-century painting, it will always be browner than the painter meant it to be. That is where tempera gains because tempera does not oxidize. Tempera paintings look very much as they did when they were painted.

Any color can be matched, and that includes the discoloration caused by the oxidation of oil paint. To match the oxidizing of the painter’s colors, you have to add a touch of red and a touch of green to every color you use; Jessell preferred to use burnt sienna and viridian (Jessell 1976). The addition of burnt sienna and viridian approximates the brown cast caused by the oxidation of the oil in the original paint, allowing for the fact that the colors that the painter originally applied have gone brown. Jessell usually began an inpainting session by mixing a small patch of viridian and burnt sienna on her palette, from which she drew small amounts as she mixed colors.

The other technique she found that ensured a good result and, more important, was central to her approach to inpainting is the concept of matching the layers of inpainting to the layers that the painter used (Jessell 1976). She found that this is really the only way to achieve the correct degree of luminosity, preserve the transitions, and match the texture seen in the original. Every painter uses different techniques to create the visual experience he intends. An understanding of those techniques as well as an appreciation for the visual experience intended by the painter are essential to successful inpainting. In the absence of technical analysis, she suggested that conservators inspect the edges of losses, abraded areas, and the edges of a painting under strong magnification to gain an understanding of the layers used and build up their inpainting in layers the way the painter did. This may seem to be a more cumbersome method than simply attempting to match a color directly, but Jessell said it was by far the most successful way of getting an acceptable result.

After applying a layer of varnish, Jessell applied a coat of Renaissance wax with a velvet cloth (Jessell 1976). The Renaissance wax enhances the varnish surface, making it smoother and more even. The wax also protects the varnish layer, providing a final barrier to humidity and friction and making the varnish less friable. Renaissance wax is an English product — the British Museum invented it, but it is now known all over the world as a good final polish.

Jessell noted that inpainting is still the biggest challenge of conservation and, therefore, the greatest pleasure. Structural work answers the physical preservation needs of a painting, but it is through inpainting that the conservator attempts to preserve the vision of the painter and bring back the quality of the painting.

Author’s note: The content of this article is based on interviews that the author conducted with Bettina Jessell.

Laura Rivers

Submitted October 2004

Perry Huston: A Different Way of Using PVAc Colors

Introduction

---

In my first year of graduate study with Caroline and Sheldon Keck at Cooperstown in 1971, I learned to inpaint using dry pigments ground in poly (vinyl acetate) AYAC (Union Carbide) and remember vividly Mr. Keck’s lecture on the “turbid medium effect.h” Because of the way that light is scattered by the tiny pigment particles in a scumble (which is semi-opaque), the resulting color appears warm or cool depending on whether the paint is applied over a light color or a dark color, in the same way that hazy smoke appears brownish in front of a white building but bluish against a dark hillside. When I began working with Perry Huston in 1980, I was exposed to a greatly expanded version of this concept, one in which warm and cool colors, as well as translucent and transparent ones, are juxtaposed on the palette and used to create special effects. Perry’s palette, which was developed from a system James Roth was using in the 1950s at the Nelson Art Gallery in Kansas City, Missouri, uses PVAc-AYAC, but the same principles may be applied to any medium.

Palette Layout

chrome oxide green warm opaque	cobalt blue special	ultramarine blue warm opaque	pthalo blue cold opaque	Prussian blue cold transparent	indigo blue special
Vermillion red deep Warm opaque	Red earth Warm opaque	Indian red Cold opaque	Alizarin crimson Cold transparent	Burnt Sienna Warm transparent special	burnt umber warm transparent
cad. yellow light warm opaque	cad. yellow medium warm opaque	aureolin yellow warm transparent	viridian green transp., special	cad. orange warm opaque	golden ochre warm transp.
titanium white transparent	lead white opaque	bone black warm	ivory black cool	raw umber special	raw sienna warm opaque

Palette Layout

The palette consists of twenty-four colors laid out in four rows of six, as indicated above. It began with many of the colors thought to be used traditionally, including the earth pigments, blacks, and whites, and was modified by additional light-stable colors. Modern or unusual colors were not included, since they were used only occasionally, but a few “special” colors — those which could not be matched with anything else — were included. The colors are arranged on the palette, in some cases, according to whether they are warm/cool or transparent/opaque relative to each other.

In looking at the layout of the palette, certain patterns emerge. While the colors are arranged generally by family (e.g., blues, reds, yellows), they are also set up in warm/cold or transparent/opaque pairs or sets. For example, the ultramarine/phthalo/Prussian blue group on the top row moves from warm opaque at the left to cold transparent at the right. In the case of the four reds in the middle, the opaque reds are located on the left — one warm and one cold — and the transparent reds on the right — one cold and one warm, depending on the scumble or glaze required. Aureolin yellow is included as a warm transparent (although the modern version is not as transparent as the old) next to the cadmium yellows, which are warm and opaque. Earth pigments that are not strong reds are clustered near the lower right, with golden ochre and raw sienna, both warm, placed together because one is transparent and the other opaque.

Inpainting Technique

Perry mixes his colors on a glass palette over a white blotter. He looks at an area to be compensated in terms of whether the color needed is warm or cool in relation to the surrounding paint. He begins by inpainting all the tiny blemishes surrounding the loss, then applies the closest matching color to the loss itself. In the case of a large lacuna, he first breaks up the area into small sections around the edges, then finishes the center. Slight adjustments in hue, value, and intensity are made by applying thin, opaque scumbles or transparent glazes to modify the initial color applied and match it to the surrounding paint. Color may be modified by adding either the complement — to diminish intensity — or a little white, black, or raw umber — to reduce hue and adjust value. Inpainting on dark fills causes the prepared color to appear cooler than intended, so some adjustment with a small amount of warm color is necessary. Once this correction has been made, however, the technique becomes second nature. One solution to this problem would be to mix the colors on a dark palette, as suggested to me several years ago by Tom Caley. Either way, the multilayered system of thinly applied warm/ cool and translucent/transparent colors works quite well in achieving an accurate match, far better than would a single layer of uniform color. Because it involves scumbles and glazes, it uses very little paint and thereby avoids the problem of cold flow sometimes associated with PVAc colors. Intermediate and final sprays of varnish (usually Paraloid B-72) may be applied over the inpainting until the correct color and gloss are achieved. The difference in solubility between the varnish and inpainting medium allows adjustment of the inpainting without disturbing the varnish. As an alternative to intermediate sprays, a small amount of medium may be added to the inpainting to bring up the gloss (particularly for earth colors like golden ochre and raw sienna, which dry light).

Palette Box Design

Jim Roth’s original palette consisted of a block of wood into which were drilled twenty-four wells, each about one inch deep. Into each well was set a small, one inch-diameter ceramic crucible (made by Coors Brewing Company, of the following dimensions: height 1”; diameter at top 1 1/4”; diameter at bottom 5/8”; the number 04 is impressed onto the side of the crucible) held in place with a drop of molten microcrystalline wax at the bottom. The crucibles were filled with pigments ground in the medium to a creamy consistency. To keep the paint moist, the box was set on a glass surface and covered with an open-bottom wood box with a glass top. Cork was used to seal the upper box against the glass. Perry refined this design by making the box out of cast aluminum. The detachable top was replaced with a hinged aluminum and glass lid, attached with small brass hinges on one side and held closed on the other with a window lock. To ensure a tight fit, a bead of silicone caulking material was pressed into a narrow groove along the 1/2”-wide inner edge of the lid. When this surface was pressed down against the corresponding smooth edge of the base, it created an airtight seal, which prevented the colors from drying out. It was occasionally necessary to add diluent and stir the paints to keep them creamy for use and avoid wear and tear on the brushes.

Author’s note: The content of this article is based on the author’s many years of experience working with Perry Huston.

Helen Mar Parkin

Submitted January 2007

Helmut Ruhemann

Introduction

Early in the chapter on the technique and ethics of retouching in The Cleaning of Paintings, Ruhemann covers the fundamental rules of the process. Ruhemann did not believe that every damage should be repaired; retouching should be kept to the minimum necessary to restore the artist’s intent. Restoration should never go so far as to make a painting look brand new. New paint should not cover even the slightest bit of old, lest any remaining part of the artist’s work be obscured (Ruhemann 1968, 241–242). Above all, Ruhemann advocated imitating the original buildup of layers using knowledge gained from a thorough examination of the painting and a strong understanding of the painter’s methods.

Ruhemann was not dogmatic on the subject of the particular technique of retouching that should be practiced; he mentions more than seventeen approaches to the compensation of losses in The Cleaning of Paintings. Despite the merits he saw in some forms of carefully executed visible retouching, Ruhemann clearly felt that, where appropriate, complete reconstruction was better suited to the goal of restoring coherence to a damaged work and that any ethical objections to matched retouching had given way with the introduction of UV light as a means of differentiating new from old paint (Jessell 1976, 1; Jessell, personal communication, 2000). Matched retouching is the end result of the method he chose to describe in detail in the The Cleaning of Paintings.

Yet Ruhemann did not advocate perfectly matched retouching for every damage, allowing that total reconstruction of such essential parts of a picture as hands and heads should never be attempted on a museum-quality painting (Ruhemann, 241–242). He instead emphasized the importance of using inpainting techniques appropriate to the painting in question and to the techniques employed by the artist or school of painting (Jessell 1976, 1). The complete program for matched retouching, which Ruhemann describes in The Cleaning of Paintings, depends on careful examination of the painter’s techniques, and specifically the very layers with which the painter constructed the painting.

Choice of Medium and Palette

The choice of medium was not a topic on which Ruhemann gave a definite recommendation. He rejected oil as a retouching medium because of its propensity to oxidize, choosing instead to work in tempera at a time when many professional restorers worked almost exclusively in oil or oil-resin. Tempera required more experience, more time, and more patience, but Ruhemann believed it was the medium that most closely met the restorer’s needs. He was, however, interested in acrylics, discussing MS2A resin color and Paraloid B-72 in The Cleaning of Paintings (Ruhemann, 245, 254). Bettina Jessell, a former student, remembers working with acrylics, on an experimental basis, during the last six months she worked with Ruhemann, in 1942. She remembers that Ruhemann was fairly impressed with acrylics and that he felt they had great advantages, most significantly that they did not darken with age and required a less demanding application process than tempera (Jessell 1992).

Ruhemann used a palette composed of approximately twenty pigments. He preferred to rely on the pigments used by the old masters, and felt that a few effects could not be achieved without the old pigments (Ruhemann, 247). In general though, Ruhemann believed that the modern, more stable replacements for many of the old pigments were often equivalent or superior and that original hues could be matched using the newer pigments (Ruhemann, 247).

After the preliminary varnish, Ruhemann would begin to reconstruct missing areas in monochrome brown undermodeling. He considered monochrome undermodeling to be an essential step in the process, irrespective of whether undermodeling was present in the original (Ruhemann, 252). He executed the undermodeling in resin paint, thinned varnish containing a little stand oil, or in a very lean oil paint, using burnt sienna, raw sienna, and black pigments.

Taking this step had a number of significant advantages. Undermodeling enabled the restorer to work out the design thoroughly before applying the color. Drawing in the design brought the painting closer to being whole by minimizing the disruption caused by the losses, making it easier for the restorer to envision the final result. Undermodeling also provided the initial, essential layer necessary to achieve the desired hues after the addition of subsequent layers; Ruhemann mentioned that certain grays and the hues of some skies could not be achieved without undermodeling. Contours were made “almost automatic” by the application of undermodeling. Retouching worn areas and light cracks with the brown undermodeling paint could lessen the disruption caused by those problems without the addition of subsequent layers. In some areas, undermodeling could be all that was required to restore coherence to a painting — the restorer need not go further. Ruhemann was of the opinion that once the undermodeling was finished, the painting would look almost complete in a black and white photograph (Ruhemann, 252).

In the case of paintings with large losses, undermodeling might not be sufficient to restore coherence to the painting and further reconstruction might be required. Ruhemann was emphatic in his belief that a model passage on which to base a reconstruction should, if at all possible, be found within the painting being restored. Alternatively, a painting by the same master might provide an acceptable model. The restorer should work from nature only if no other acceptable source was available. Ruhemann believed that reconstruction from memory was “rarely satisfactory” (Ruhemann, 253).

Following the painter’s layers, Ruhemann would apply one or more layers of opaque or semi-opaque tempera body paint. He built impasto up gradually, allowing previous layers to dry at least to some extent before applying the next layer. It was important that this layer be as completely modeled as possible. In part, the undermodeling contributed to the development of cool middle tones in the appropriate areas when subsequent layers were placed over areas of darker undermodeling. After the application of the body color, the painting should look more or less complete. However, the areas where there were formerly losses appeared lighter and cooler than the surrounding areas (Ruhemann, 253).

Ruhemann would then burnish the retouchings to avoid the absorption of the varnish into the tempera paint and then apply a thin layer of varnish to the retouched areas. The next step would be to apply very thin, light glazes to the retouched areas, using watercolor or wax in MS2B. Ruhemann suggested using lighter pigments or adding kaolin to lighten the glazing. He mentioned using Davy’s grey in preference to black and, instead of brown or yellow, light red and cobalt or cerulean blues (Ruhemann, 254). Finally, Ruhemann would varnish the matte spots, applying a varnish layer of normal thickness.

Author’s note: In addition to the references cited, the content of this article is based on accounts from his student and trainee, Bettina Jessell.

Laura Rivers

Submitted December 2002

Contemporary Conservation Profession Materials and Practices

Documentation of Original Condition Prior to Compensation

Introduction

---

The purpose of this special record is to provide a reference image for documenting the extent to which losses, damages, and other factors that detract from the visual integrity of the painted design are compensated by the conservator, this compensation being made by the addition of materials to the artwork. For the record to be complete, it also requires making an after-treatment image for comparative purposes. In addition, it may serve as a record of the extent and appearance of remaining original surface, even if compensation for loss is not to be carried out.

The recommendation for making this record goes back to the early history of 20thcentury conservation. It is seen, for example, in the International Museums Office 1939 Manual on the Conservation of Paintings; because this publication reflected already accepted attitudes at the time, the importance and need for this record was clearly realized early on in the history of modern conservation. The Manual expressed the recommendation as follows: “Photographs should be taken of the different stages of the cleaning operations, but the most valuable record will be a photograph of the picture completely stripped” (International Museums Office 182). And, “Whatever method [of retouching] may be adopted…, the various systems of recording will always furnish the necessary data for ascertaining the condition of a painting before and after stripping and before retouching and additions” (109).

The Murray Pease Report (1963), which served as the Standards of Practice and Professional Relationships for Conservators for the IIC-AG and, until 1993, for the AIC, required the creation of a “Photograph in ‘actual state’ without compensation” (Section IV, C, 2, b). In a conceptual sense, the term “actual state,” still commonly used today to describe the subject of the pre-compensation photograph, was never well defined. The introduction to Commentary 23 (approved Oct 1997) in the AIC Commentaries to the Guidelines for Practice, however, now provides some clarity to this ambiguity, as follows: “The baseline for determining the nature and extent of loss is the point at which the cultural property was generally accepted as completed, although compensation need not return the cultural property to this state. The original completed state (what the artist/ maker actually did) takes precedence over the artist’s/maker’s original intent in guiding the nature and extent of compensation for loss.”

Thus, it is the extent to which the remaining “original completed state” is revealed that is to be documented. In most cases, the making of this record serves to fulfill all the listed rationales in Commentary 27, Documentation of Treatment, and augments the precepts in Guideline for Practice 23 that compensation “be detectable by common examination methods,” and that it “not falsely modify the known aesthetic, conceptual, and physical characteristics of the cultural property.”

While the written record and diagrams are important adjuncts to this record, documentation here must be photographic and should indicate as clearly as possible the condition of the remaining original surface and the areas of the painting that are to be compensated. In some cases, this may require special photographic techniques, filling or varnishing prior to photography, or other measures. Some suggestions follow.

General Recommendations for Photographic Documentation Prior to Compensation

Relevant Information

The relevant information required in Commentary 24 section B must be included in the photograph:

• Unique identifier (e.g., accession #)

• Date of the photograph

Additional Referents and Information

In addition, as listed under “Recommended Practices” in Commentary 27, it is highly recommended that additional referents and information be included, not only in this photograph, but also in all like photographs of the paintings made throughout the examination and treatment process. These include:

• Gray scales: A gray scale is an essential referent to ensure color neutrality as well as uniformity of exposure and contrast among comparative images. It should be included in documentation photographs whenever possible. The AIC Guide to Digital Photography and Conservation Documentation (Frey 2008) should be consulted for a discussion of gray scale reference targets and their use.
• Size scale
• Light direction indicator
• Other clarifiers, such as:
  • Stage of treatment label
  • Label indicating that lacunae have been filled prior to photography, or that varnish has been applied
  • Label indicating special illumination (e.g., UVA)

Note: In digital photographs, it is good practice to document in the image file’s metadata any special measures taken to enhance or clarify the pre-compensation photograph. Such measures (discussed below) might include varnishing or filling prior to photography, transmitted light photography, or UVA-induced visible fluorescence photography. For the latter, any camera filtration used should also be recorded.

Varnishing

The process of varnish or overpaint removal often leaves residues or other disturbances on the surface that create visual anomalies (variations in sheen, saturation, etc.) that do not reflect the true character of the remaining original, and indeed may confuse the record with respect to the photographic image of the remaining design. On paintings that are to be varnished again, it is advantageous to apply a saturating varnish to the surface before making the pre-compensation photograph. The photograph should be labeled to indicate this has been done.

Suggestions for Clarifying the Visibility of Areas to Be Compensated

Detail Photographs

If areas to be compensated are small and localized, so that they would be difficult to discern in the overall pre-compensation photograph of the painting, detail photographs of the area should also be made. Similarly, areas of abrasion that are to be compensated are often best documented through detail photographs. Detail photographs should be made of the same areas after compensation for the most complete record.

Filling

If losses in the painting are too small to be discerned easily in the overall photograph of the painting (especially in areas where the exposed support is similar in tone to the surrounding paint), it may be advisable to fill the losses prior to photography to provide enhanced contrast. The pre-compensation photographs should be clearly labeled to indicate that this has been done. Obviously, this method requires that the fills contrast well with the surrounding original (e.g., untoned gesso fills).

Fluorescence Photographs

Pre-compensation and after-compensation fluorescence photographs of the painting under long-wave ultraviolet (UVA) irradiation can often provide an excellent record of the extent of compensation. This pair of photographs would be made in addition to the standard pre-compensation and after-treatment photographs. See the AIC Guide to Digital Photography and Conservation Documentation (2008) for information on the making of fluorescence photographs.

Transmitted Light Photographs

Generalized pinpoint losses in canvas paintings can often be clearly recorded with transmitted illumination in combination with low-level front illumination.

Annotation

Graphic annotations can serve as useful adjuncts to clarify areas to be compensated and are easily added to digital photographs.

Dan Kushel

Submitted February 2010

The Ethics of Inpainting

“If my grandfather had an axe and he gave it to my father, and my father replaced the handle, and he gave it to me, and I replaced the head, is it still my grandfather’s axe?... For me it’s still my grandfather’s axe…There’s a constant battle between conserving the idea and conserving the object..” ¹

-Damien Hirst

Introduction

Previous generations had widely differing views regarding the ethical principles of inpainting damaged pictures, but their willingness to openly debate these issues has ultimately facilitated a positive result — general agreement on several key points among conservation professionals today. Reinforced by the foundation and credibility of graduate training programs in the United States and abroad; by an abundance of international symposia, conferences, and workshops; and by the publication of peer-reviewed journals focusing on conservation treatment techniques, scientific analysis, and technical art history, this consensus can be defined as follows:

• Minimal intervention with a focus on preservation
• Strong respect for the conceptual “artist’s intent”
• Objectives that account for the viewer’s response — aesthetically and educationally.

This essay will attempt to articulate the complex issues that the conservator must address to integrate and balance these three objectives into inpainting practices. Subtle boundaries within the ethical discussion, such as unity and authenticity that complement the more tangible policies like reversibility, will be explored in the context of historical views and modern perspectives.

One such historical view is found in the early 19th-century treatises on the conservation of paintings and works of art on paper written by German doctor, pharmacist, and chemist F.G.H. Lucanus. Lucanus (1793–1872) wrote for collectors who traditionally cared for their own collections, recommending that unless they were artists themselves, they seek out a restorer who was qualified to do the work (Lucanus 1828, 41). Prior to this time, books on restoration tended to contain tips and recipes haphazardly presented, whereas Lucanus was among the first to address the ethics of inpainting through the systematic presentation of methods and materials. His ethical justification for retouching incorporated the concept of visual unity when restoring a damaged picture without resorting to artistic additions or overpainting:

“After doing everything necessary to remove the defects without damaging the essential substance, then you will have the pleasure of preserving the paint ing with all its colors and glazes, and you would simply re-varnish the surface. However, if there are losses, even if they are filled with paint or putty, if the contours are unclear, whole passages of paint are missing, or even single glazes abraded or faded, then the harmony of a painting is disturbed and a new element of the conservation takes place: you must retouch.” (129–130)

Reversibility

Lucanus is probably best known, however, for introducing dammar in 1829 to artists as a preferred varnish, not only for its desirable visual properties but also because of its relative ease of reversibility (Wagner 1988, 19). Although Lucanus did not refer explicitly to the reversibility of inpainting materials, it can be inferred from his comments on varnish removal that he was conscious of the value in protecting artists’ original materials: “It is not very likely that you will have to remove dammar coatings. However, if removal does become necessary it is very easy to execute it with a mixture of poppy seed oil and turpentine. The painting will not be damaged” (153). Lucanus’s final bit of advice for preserving the paint film was to attach a note to the painting’s reverse about the varnish type along with instructions for removing it, a further indication of his concern for the potential liabilities of reversing previous treatments (154).

Reversibility is a core conservation principle that is especially important with regard to inpainting for at least two reasons. First, the deterioration of the conservation materials relative to those of the original can eventually result in a visual discrepancy, causing the inpainting to become a distraction and no longer serve its intended function. Second, because a degree of interpretation takes place during execution, inpainting is subject to becoming stylistically irrelevant, and a future generation may deem it necessary to reinterpret the painting for its own time. Consequently, the conservator needs to consider the stability and permanence of the methods and materials used for inpainting, which are highly dependent on what materials are available at the time of treatment. Each generation is challenged with making the best use of currently available materials, developing new materials, and finding new uses for existing materials.

Based on the results of scientific testing and personal experience, conservators aim to predict the degree to which inpainting materials can be removed once applied to a painting. At the same time, each painting’s unique characteristics, such as material composition, age, use, condition, and treatment history, call for individualized treatment solutions. For example, an inpainting medium that can be applied to an oil painting and successfully removed some years later might be an inappropriate choice for use with an acrylic painting due to the different solubility parameters of the paint films. The manner in which a material is used can potentially affect its reversibility, such as when a conservator modifies a proprietary inpainting material by adding dry pigments or another resin to alter its working properties. This practice could have an impact on the treatment’s stability as well.

Insoluble inpainting materials, such as egg tempera, that are applied only to filled losses are technically acceptable within reversibility parameters because they can be removed mechanically. But this practice is less desirable because of the high risk of permanently covering the adjacent original paint margins by unintentionally encroaching onto them. Watercolors and gouache, on the other hand, are considered to be reliably reversible and lend themselves particularly well to early stages of inpainting when used to tone white fills, or later over a varnish layer for subtle glazing effects. Yet, even if a material has been professionally tested and deemed to be chemically resoluble over time, history has demonstrated that there are no guarantees with regard to aging properties. For example, Paraloid. B-67, once considered to be an appropriate synthetic resin for use with paintings, is no longer a preferred option as it can become increasingly insoluble over time. Environmental conditions to which the painting is subjected can also affect reversibility, especially if the restored painting is not within a climate- and light-controlled environment.

Although inpainting treatments of past centuries were commonly executed using materials closely matched to the original, such as oil paint retouching on an oil painting, subsequent conservators tended to replace oil with their favorite conservation-grade alternative and employed it in all cases. With a wider set of options now available, conservators have moved away from the formulaic use of a single inpainting material and have embraced a more humanistic approach when making treatment decisions. Consideration for the painting’s overall harmony and legibility has been incorporate into the material selection process, yet only professionally accepted materials remain as options. Therefore, the conservator is faced with the ethical dilemma of accomplishing the aesthetic goals of the treatment while also optimizing the materials with respect to stability and reversibility. The success of the treatment depends on the conservator’s ability to make compromises that reach a balance between preserving the physical nature of the work and maintaining its artistic integrity.

While varnish is perhaps the obvious and more debatable example of appearance versus reversibility, inpainting materials are also scrutinized and ultimately chosen to best suit the period of the painting, with the underlying prevailing factors being color stability and reversibility. For paintings where the retouching is isolated from the original by a varnish layer, it is important to remember that the solubilities of the varnish and inpainting materials are closely linked — we may consider an inpainting material to be reversible but usually not without the general assumption that it can be removed only during a future cleaning. Therefore, the reversibility of the inpainting essentially depends on the reversibility of the surface coating(s), in that inpainting is rarely removed without disturbing the varnish or overall patina.

Two private companies, Gamblin Artist’s Oil Color and Golden Artist Colors, have recently begun to manufacture conservation-grade materials to meet conservators’ needs for reversible inpainting materials that have desirable working properties. Oregon-based Robert Gamblin claims that Gamblin Conservation Colors are “stable, reversible, and suitable for use with a wide array of painting styles and techniques”. Gamblin Conservation Colors are made with a low-molecular weight resin and pigments that lead to a “lightfast, permanent material with enhanced working and aging properties” (www.conservationcolors.com.). Similarly, New York-based Golden Artist Colors has developed several conservation products, including MSA Colors, an alternative to Gamblin’s Conservation Colors. Golden asserts the reversible properties of their paints:

“GOLDEN MSA Colors are produced with a mineral spirit-borne acrylic resin system. They dry quickly to form very durable films with excellent chemical resistance to acids/alkalis, water, and ultraviolet rays. All of the pigments in GOLDEN Acrylics are chosen for the greatest clarity and permanency within each chemical class. MSA Colors remain soluble in mineral spirits, making them ideal for easy removal without disturbing the layers of paint underneath. For this reason, they are beneficial in art conservation for inpainting techniques. (www.goldenpaints.com).”

The availability of suitable inpainting materials from these and other proprietary sources is surely an advantage to conservators — not only does it make treatment materials more accessible and expeditious, but also with more options, a conservator has a greater chance for a successful restoration and an ethical treatment to coincide. While companies like Gamblin and Golden have formulated and tested their own products, many conservators have opted to fabricate their own paints using various conservation grade resins and dry pigments.

Justification for Inpainting

After the conservator has answered the primary question regarding which materials are available and appropriate for a particular inpainting treatment, two additional fundamental questions must be answered: is the inpainting justified and if so, to what degree should it compensate for damage? Answers to these questions are essential when developing an inpainting treatment plan and have been the source of contentious debate over the years. Inpainting is not usually required to preserve a painting’s structural integrity, and a painting could, in a narrow sense, be conserved without the addition of any color. Inpainting is warranted and necessary, however, because the need for a painting to continue to exist as a complete visual document and accurate record of an artist’s work is essential to the painting’s ability to deliver its aesthetic message to the viewer.

Whether to inpaint, and if so, to what degree may be decided by a team of conservator and curator if the painting is part of a museum collection, by the conservator and owner if the painting belongs to a private collection, or even by the conservator and the artist. For practical reasons, it is sometimes the conservator alone who makes these decisions; however, particularly in the case of a badly damaged work where controversial decisions need to be made, a conservator who seeks out other opinions ensures a more prudent and principled approach. Depending on the extent of damage, the compensation questions for an individual painting may differ, and differ yet again when considerations such as location, type, and size of losses are taken into account. Almost always, compensation decisions are influenced by the environment in which the painting will ultimately be seen because the role of the painting as a viewable artwork will dictate varying degrees of unity and authenticity arrived at through treatment. Consequently, for the treatment to remain faithful to the artist’s original intention but relevant to its current and future audiences, the extent of inpainting and the rationalization behind it must be clearly defined for each treatment.

A painting has a dual purpose that directly affects how it is experienced by the viewer. Sheldon Keck (1910–1993), who was a founding member of two conservation graduate programs — the Conservation Center, Institute of Fine Arts/New York University and the Art Conservation Program, Buffalo State College (formerly the Cooperstown Graduate Program) — gave serious and careful thought to this idea. First, a painting expresses the artistic concepts of its creator and exists as an aesthetic object. Second, a painting serves as a social historic entity because it is evidence of the moment and place of its creation. The purpose of a work of art is also connected to its places of exhibition and may change over time: “When it is first created its dominant importance is in the use to which it is put, but as time passes, its historic significance tends to overshadow its original intent and sometimes its aesthetic function” (Keck 1963, 171). The degree to which the historical significance exceeds the aesthetic value of a painting depends on who is making the evaluation. Everyone, including the artist, historian, curator, director, collector, and layman, perceives a painting differently because each has a different viewpoint, motivation, and purpose. The conservator must reflect on the function each individual assumes and consider the context in which the painting will be viewed after it is restored.

Authenticity versus Unity

When assessing the degree of compensation achieved through inpainting, the conservator must debate the value of retaining a level of authenticity versus the consideration for unity. Authenticity signifies the disclosure of the true and current condition of the work to the viewer, a treatment’s faithfulness to the original condition of the painting and to what remains of the artist’s hand. Unity, as a major principle of aesthetics, prompts questions interrelated with authenticity about how much damage should remain visible and how much intervention is necessary and acceptable. For example, how much can a viewer tolerate missing parts of figures or an interruption of a fluid brush stroke? How does a conservator approach an abraded face that no longer engages the viewer or demonstrates the competence of the artist? Does the risk to the distinctiveness of the artist and the individuality of the work brought about by the interference of a conservator’s inpainting outweigh the need to restore the continuity of the composition for presentation to the viewer? The answers to these questions are subjective, as all pictures change differentially, and even with written accounts or photographs, it is impossible to know exactly what a painting looked like when the artist set down the brush for the final time. However, inpainting must not be considered an invention of an uninformed practitioner that leads to an illusion of completeness at the cost of authenticity (Offner 1963, 156). Instead, inpainting should be viewed as a means of reestablishing the visual communication between the artist and the viewer, a communication that depends on the totality of the painting as originally conceived by the artist and leads to a more authentic, and therefore more gratifying, viewer experience (Keck 1963, 170).

Authenticity in restoration has been debated within the artistic community by artists, art historians, and conservators. French artist Eug.ne Delacroix (1798–1863) kept thoughtful diaries and notebooks that were compiled into a journal, first in 1893 and later edited in 1932 by M. Andr. Joubin, the accepted authority on the artist’s writings. In many journal entries, Delacroix commented on discrepancies between the artist’s intent, authenticity, and the current condition of paintings in collections he visited. He especially noted the issue of pictures darkening over time, most often when describing paintings by Peter Paul Rubens:

“…When we copy a Titian or a Rembrandt we believe that we are keeping the same relationship between lights and shadows as the master’s, but actually we are piously reproducing the work of time, or rather its ravages. The great artists would be most painfully surprised if they could see the smoke-blackened daubs that the pictures which they originally painted have become. The background of Rubens’ Descent from the Cross, for example, must always have a very dark sky, although one which the artist could imagine when he represented the scene, but it has now become so dark that one cannot distinguish a single detail. (Delacroix 1980, 238).”

Delacroix recognized that paintings change over time and that the viewer must adjust his impression accordingly to understand what was actually intended by the artist. However, he had very little tolerance for changes made by a restorer whose goal was to compensate for those changes and expressed that a painting’s authenticity was lost after restoration:

“We are sometimes astonished that nothing remains of the painting of antiquity…We should be even less surprised at their destruction if we reflected that most of the pictures produced since the Renaissance — that is to say, comparatively recently — are already unrecognizable, and that a great number have already perished from a thousand different causes…Clumsy restorations only finish the work of destruction. Many people imagine that they do a great deal for paintings when they have them restored. They appear to think that pictures are like houses that can be repaired and still remain houses, like all those other things, in fact, which time destroys, but which we contrive to preserve for our use by constantly repairing and re-plastering. Women who know how to make-up can sometimes disguise their wrinkles so as to appear younger than they really are, but a picture is something totally different. Each so-called restoration is an injury far more to be regretted than the ravages of time, for the result is not a restored picture, but a different picture by the hand of the miserable dauber who substitutes himself for the author of the original who has disappeared under his retouching. (Delacroix 1980, 238–239).”

As an artist of the 19th century, Delacroix likely observed excessively overcleaned paintings and restorations that embraced unrestrained repainting that today would be deemed unacceptable according to current ethical inpainting practices.

Art historians have also expressed their concerns. For example, German-born scholar of Netherlandish art Max Friedländer (1867–1958) commented on restoration in his writings, defending restorations that involved varnish removal, structural stabilization, and consolidation. However, he had a more extreme attitude regarding loss compensation:

“The activity of the restorer becomes highly problematical the moment there presents itself the question of making up for deficiencies — that is, of filling gaps or revivifying passages which have been rubbed…The historian, to whom the work of art is a record, opposes, from his standpoint, with full justification, that kind of restoration which goes beyond preserving and exposing. He demands to see clearly what is left of the original, but wishes it also not to be concealed from him that something of the original is missing. It is precisely the successful reintegration that is distasteful to him: the unsuccessful one he, of course, detects easily and can make allowance for. (1996, 333).”

Because the term authenticity carries an authoritative connotation that an object is what it is claimed to be, there is also the implication that restoration changes the true identity of the work and therefore its very legitimacy. Friedländer writes further to say that art market value was the main reason damaged pictures were restored despite resistance from more purist scholars. (1996, 333).

Richard Offner (1889–1965), an Austrian-born authority on Italian art and cofounder of the Institute of Fine Arts at New York University, had a similar attitude to Friedländer and argued as late as 1963 that restoration was in direct conflict with the authenticity of damaged pictures:

“Any restoration…that introduces paint or shape within its boundaries, even if the restoration be limited to the missing portions alone, must prove intolerable. The plea, generally advanced by the apologists of restoration, that a work may be repainted in tempera or watercolor which can be cleared away or renewed when desired, neglects the transparent truth that any addition whatever introduces irrelevant matter and serves to instill a false impression in anyone who sees the restored work. The longer the repainting endures, the more it misleads. Whether a restoration deludes the eye or disfigures a work, the fact that it is susceptible of easy removal fails to justify it either on moral or aesthetic grounds. Repainting adds the factor of personality — and possibly of intention — extraneous to the original work. (1963, 157).”

Offner suggested that the moment a restorer changes the present state of a work, it no longer exists as the natural development and evolution of the artist’s hand (157). These opinions imply that as the true connoisseur, the art historian would be expected to overcome the disconnected image and be impressed by the authenticity of the work despite any damage it endured.

In contrast, art historian Cesare Brandi (Italian, 1906–1988) was in favor of at least minimal retouching and reintegration, as he acknowledged that sacrificing unity for authenticity does not serve the painting or the viewer. In 1939, Brandi was appointed president of Italy’s first training institute and school for restorers, the Istituto Centrale del Restauro in Rome, where he served until 1960. He is well known for developing conservation theories that informed the tratteggio style of inpainting based on the principles of nonimitative retouching, a nascent version of which was practiced in the early 1930s at the Fogg Art Museum (Contreras de Berenfeld 2008, 185).² Tratteggio was designed specifically to allow the viewer to distinguish retouching from the original while also attempting to resolve the authenticity dilemma. When considering the impact of damages on a painting, Brandi wrote:

“A lacuna in regard to a work of art is an interruption of the figurative pattern… The lacuna, in fact, will have a shape and color that are not relevant to the figurative aspect of the represented image; it is inserted into the work of art as a foreign body…The image is more than just mutilated, it is also devalued in the sense that what was born as a figure is now reduced to mere background. (Baldini 1997, 341).”

In this statement, Brandi justifies the need for retouching as a primary means for maintaining the fundamental nature of a work of art. The secondary purpose of tratteggio was to retain a level of authenticity within the treatment by “prevent[ing]…any personal expression of the restorer…so that the intervention is structurally unmistakable as a critical interpretation” (Mora, Mora, and Philippot 1996, 351). According to Brandi’s theory, unity is restored to the extent that the viewer perceives a continuous image when standing at a normal viewing distance. The unified picture is closer to the original intention of the artist while authenticity is preserved by limiting the degree to which inpainting makes the damage imperceptible.

Keith Christiansen, chairman of European paintings at the Metropolitan Museum of Art, demonstrated in a 2002 symposium on early Italian paintings at Yale University how a painting can lose its narrative function if it cannot transmit its message through a legible composition. Christiansen used Leonardo da Vinci’s fresco, Last Supper, as a prime example of how a lack of completeness can influence the interpretation of a work of art: “…I fear what we see are no more than fragments of disconnected moments — some preliminary, some half way to the completed image, but all of them compromised by deterioration and damage. Were we to have only the remnants of the mural as it is now, would it be possible to deduce Leonardo’s intentions? I doubt it” (2003, 76). Christiansen suggested that early descriptions of the work and photographs of its condition in the past should be used to help evaluate the artist’s objective. Whether the readability of Last Supper in particular can be veritably retrieved is beyond the scope of this discussion, but it is important to note that Christiansen’s comments justify the act of inpainting while also reinforcing the importance of documenting treatments for future reference.

In addition to art historians and curators, an international group of conservators addressed the challenges of maintaining authenticity while achieving unity throughout a conservation treatment. Working in Belgium, Albert Philippot (1899–1974) and his art historian/archaeologist son Paul (b. 1925), who served as director of ICCROM in the 1970s, developed a model in 1959 for treating paintings that was structured around the concepts of balance and unity: “A work of art as such is not composed of individual parts but constitutes, as an image, a whole endowed with its own unity, realized in the continuity of the form — a unity therefore essentially different from that of the things represented. Any discontinuity, any interruption inevitably disturbs the reading of this rhythm” (1996, 335).

Treatments were based on forming an idea about the original appearance of the painting, and the method was informed by the knowledge of the evolution of materials and the ability to differentiate between actual damage and patina (Modestini 2003, 33). The Philippots addressed authenticity as well by qualifying loss compensation in two ways. First, they described how inpainting must be executed with critical interpretation of the paint layer, a high level of technical execution, and a practical cultivation of the visual imagination that is derived intuitively from the picture so that the conservator’s own personality is suppressed as much as possible (336). Second, they considered the principle of strictly limiting the retouching to within the losses as a sacred one (336–337). Authenticity was achieved without sacrificing unity because the retouching was considered to be an informed reintegration of a damaged painting that could be changed in the future if the interpretation was deemed inappropriate.

Some years later, Paul Philippot collaborated with Rome-based conservators Paolo Mora (1921–1998) and Laura Mora (b. 1923)3 to coauthor an article that stated the following principle: because a work of art is created by an individual at a distinct moment in time, it is unique and cannot be reproduced — to attempt to do so would change not only its character but also its meaning (Mora, Mora, and Philippot 1996, 343). The conservator’s role is to preserve that character and meaning through skilled and ethical treatments. The authors pointed out that multiple losses had the potential for overwhelming a painting to the extent that the image would become secondary to it, thus making it impossible for the viewer to experience the work of art in a meaningful way (343). And because the appearance of a work of art is directly related to the artist’s intent, the condition of the work will affect the comprehension of that intention by the viewer. Consequently, if damage to a painting affects its appearance to the point where it can no longer be understood, then inpainting is necessary to preserve the illusion created by the artist. A conservator who is respectful of a painting’s artistic and historical authenticity is justified in attempting to minimize the discontinuity of form because “from an aesthetic viewpoint, the work of art is characterized by the unity of the form as a whole. The image created by the artist differs from the object which is its physical support in that it is not equal to the sum of its parts and is therefore not divisible” (345). However, if the damage across a painting is so great that the unique quality of the work is at risk, a conservator must be limited to the practical extent that he or she does not make suppositions that would compromise the genuineness of the work. The painting’s uniqueness should be respected, and a conservator should make judgments that are guided by conscience, refraining from invention to the point that a treatment becomes a falsification.

Former Metropolitan Museum of Art conservator John Brealey (British, 1925–2002) articulated a humanistic approach to conservation, where decisions and judgments on cleaning and compensation were made through “an educated ability to look at works of art intelligently and sympathetically in order that the entire significance of the work is understood, both intellectually and emotionally” (Talley 1992, 46). According to Brealey, treatments should be guided by the visual interpretation of a painting and should try to resolve how the artist’s intention has been affected by the painting’s poor condition. While unity is the primary goal in this case, authenticity is also retained by incorporating scholarship that relates to a painting’s origin and context. This is accomplished by studying other works by the same artist; understanding work by the artist’s contemporaries, rivals, and imitators; gaining experience with the painting methods and materials used during the artist’s lifetime; and being familiar with the aesthetic, religious, political, and social ideas that were then current (Tomkins 1987, 45). After reflecting on this research, an informed, successful treatment from a humanist’s perspective depends on the recovery of tonal harmony, internal structure, and a painting’s convincingness as an illusion via technically derived methods of preserving and restoring art.

Mark Leonard, former head of paintings conservation at the J. Paul Getty Museum, elaborated on this approach in a personal reflection on paintings conservation. He wrote, “Allowing myself to be guided by the works of art enabled me to engage in restoration treatments that ultimately brought the individual paintings to a state where the artist’s voice could have its fullest expression through the physical reality of the work of art” (2003, 42). The most important part of this statement is “the physical reality of the work of art” because it signifies the acknowledgment of limits in a restoration treatment. There are limits to the extent of repair that can be reached, there are limits to the acceptable materials that can be used to treat the picture, and there are limits to how well an artist’s intention can be understood in a subsequent point in time as well as from an aged and damaged picture. These boundaries are created by the imperative of preservation that comes first in any ethical code in conservation and are guided by a goal in which a restoration allows for a viewer to experience the full impact of the painting as a work of art without commenting on the extent of the damage.

Viable Approaches

Imitative retouching is a practical solution to the philosophical questions discussed above. Also referred to as invisible or mimetic, imitative retouching aims to reproduce the hue, tone, texture, and reflective properties of the adjacent original paint such that the disruption caused by the damage is imperceptible and the illusion is restored to the picture. Directed by the information remaining of the original to either side of the damage, the conservator aims to reintegrate the loss so that it is indistinguishable from the original. New York-based paintings conservator Dianne Dwyer Modestini5 presented her views on this topic at the 2002 Yale symposium on the conservation of early Italian paintings:

“Imitative retouching is not just cosmetic, but part of an aesthetic, some might say humanistic, approach to restoration. It has philosophical roots in Berenson’s concept of tactile values: “Painting is an art which aims at giving an abiding impression of artistic reality with only two dimensions...I must have the illusion of being able to touch a figure, I must have the illusion of varying muscular sensations inside my palm and fingers corresponding to the various projections of this figure before I should take it for real and let it affect me lastingly,” he wrote. Paintings that do not satisfy this illusion of the third dimension cannot communicate tactile sensations and so are difficult and confusing to most lay viewers. (2003, 209).”

This approach is primarily favored when the painting is in a convincingly good state and only a minimal amount of interpretation is required.

When the treatment involves a picture with extensive paint loss and insufficient information to guide the conservator in his or her inpainting decisions, three basic compromises are available. First, based on art historical knowledge and an awareness of the aging properties of the original materials, and in consultation with other interested parties, the conservator can conceive an imitative reconstruction by making comparisons with similar works in a good state of preservation or historical photographs of the work. However, such a solution can be controversial and should be implemented only in response to an extreme situation where other options are impractical. If the painting is to be displayed in a museum or other public place, wall text should be included to inform the viewer of its compromised condition. Likewise, full disclosure of the extent of restoration, which can be accomplished with conservation documentation, should be obligatory when a picture is available to the marketplace.

If an approach that involves extensive reconstruction means going beyond what is deemed reasonable or desirable by the conservator, curator, art historian, owner, or other interested party, then a second option would be to use a type of discernible inpainting, also referred to as differentiated or evidenced inpainting. The purpose of discernible inpainting is to reconstruct the damages in such a way as to subdue the distracting effects of the losses when viewed from a normal viewing distance, but making them readily visible on close inspection. The tratteggio technique is eminently suitable in such cases. The third alternative is another form of discernible retouching in which losses are toned with a neutral color to subdue the distractions caused by the damages. This allows the conservator to present the work in a respectful manner while also acknowledging its fragmentary state. The value of displaying a work in such a poor state of preservation should be widely debated before this type of treatment is undertaken.

In the case where a painting’s condition varies extensively from one area to another, choosing between the two markedly different inpainting philosophies — imitative or discernible — might be difficult. The conservator would want to optimize the well preserved areas without going beyond the ethical boundaries in the damaged areas. This may lead to a solution where both approaches are used within the same picture. This option, which is particularly well suited to large-scale works such as wall paintings, would signal to the viewer the painting’s fragmentary state but also allow for the appreciation of the intact passages of the picture.

Although seemingly apart in their philosophies, conservators who prefer imitative and those who use discernible approaches to inpainting are, in practice, often less polarized than one might imagine. For the conservator who subscribes to discernible inpainting, the level of perceptibility tends to diminish as the level of preservation increases; the smaller and better preserved the picture, the less apparent is the retouching. Often for the imitative practitioner, the larger the loss and the smoother the surface texture, the more visible the retouching is to the connoisseur. Both philosophies can be ethically employed depending on the condition of the painting, the context in which the restored picture is re-presented to the viewer, and the sensitivity and skill of the conservator.

Sensitivity is especially important when a conservator is faced with the ethical challenge of determining whether a treatment requires retouching that may cover over original paint in order to achieve the objectives of the treatment. In light of the misguided and extensively overpainted treatments that conservators frequently inherit, controversy has surrounded the issue of whether and to what extent one should compensate for areas of abrasion, missing glazes, fugitive and discolored pigments, and tonal shifts in dark versus light passages over time. Consequently, when damage to a paint film goes beyond discrete losses, the ethical guidelines for inpainting become more complex. A conservator may be compelled to use thin glazes to restore three-dimensionality to a forehead or soften highlights on a nose to retrieve a portrait’s tonal balance, for example. It is also not uncommon either for a conservator to use scumbling as a means for tonally lifting remnants of darkened varnish or embedded bits of grime that are trapped in the interstices of a flattened canvas weave.

Conservators and others from past centuries frequently commented on limiting retouching to discrete losses, and, as it is related to the principle of reversibility, a staying within-the-lines approach is widely accepted as an ideal ethical practice. Only a few were willing to acknowledge the practical necessity of scumbling and glazing. For example, Lucanus’s 1832 treatise explicitly addressed the subject of replacing the worn glazes of a painting:

“Glazes form the skin of a painting. For this reason, the glazes are exposed to all occurrences. Furthermore, they are sensitive to all solvents that are used to remove varnish as they are applied over or in the varnish coating. If there are small breaks in the glazing then it is appropriate to complete them. In cases where all the glazes are missing, overglazing would disfigure the picture and take away everything that remains from its authenticity. If you are not quite sure if a painting had any original glazes then you should not do any glazing.”

This example of wrestling with the thorny and sometimes agonizing decision to retouch on top of original paint recognizes certain limits but also the instances when glazing can be warranted and successful.

Helmut Ruhemann (German, 1891–1973) recorded practical solutions to loss compensation in his book that reflected on nearly 40 years of working with the collection at the National Gallery London. He believed that “retouching should be kept to the minimum necessary to restore the coherence in composition and the characters of a damaged painting” and that “no new paint must be allowed to cover the smallest part of wellpreserved original” (1968, 241). Ruhemann carefully chose the phrase “well-preserved original” to allow for exceptions to that rule: when a painting’s surface has “no fundamental loss of paint but only an ill-defined wearing of the original surface, ‘stippling-in’ is required” . Ruhemann also acknowledged that each picture may require a different level of inpainting, admitted to using different inpainting solutions within the same picture if necessary, respected the principle of reversibility, and recognized the value in presenting a picture that is aesthetically pleasing to its audience: “On the whole, once everything has been done to assure the preservation of a painting, its successful existence depends on its presentation. It will fail in its destiny to the extent to which the master’s intention is marred by losses or accretions. The initial meaning will not only be obscured but often distorted”. While Ruhemann’s practical approaches to cleaning paintings have been challenged, his writing on the subject of inpainting was balanced and respectful, based on connoisseurship that could be applied to many treatments today.

National Gallery London conservator Larry Keith’s recent treatment of Aelbert Cuyp’s Large Dort presents us with an excellent example of a well-reasoned approach to retouching. After the cleaning phase of the treatment, Keith was presented with a discrepancy in the spatial resolution between parts of the foreground and the middle ground that disrupted the reading of the picture. Guided by information obtained through technical analysis, he determined that much of the foreground paint had experienced a color change due to blanching. Scientists then conducted extensive analyses to determine the actual cause of the blanching, which helped to clarify which areas were meant to appear gray-green and which were not. Keith’s remaining questions regarding the artist’s intent were resolved by making visual comparisons with other Cuyps known to be in a good state of preservation. Strengthened by a complete understanding of the painting’s con dition, the conservator was justified in addressing the damage by selectively toning the blanched paint “to present a more coherent and properly functioning spatial recession… The intent was to aid the spatial illusion not through reconstructing the colours of the unaltered paints, but through re-establishing tonal relationships that functioned well enough to allow the viewer to appreciate some measure of the picture’s original splendor” (Spring and Keith, 2009 ).

In his 1996 essay, conservator in private practice Steven W. Dykstra illustrated how a conservator might deliberate over whether to apply retouching on top of original paint by highlighting the example of a pentiment. The author pointed out that a pentiment likely was not intended by the artist to be part of a work, as it is the result of natural aging of the paint film, but that it could be visually acceptable and of interest to viewers because it provides insight into the artist’s thought process. On the other hand, in instances where a pentiment interferes with the legibility of a picture, it could be perceived as a disfigurement and is therefore something to be concealed with retouching (Dykstra 1996). Dykstra used a similar line of reasoning to demonstrate how to deal with other unintentional features of a picture, such as a discolored pigment or a glaze that has faded:

“Tradition and practicality seem to determine how this approach is applied. Traditionally in older artworks, some varieties of deterioration are commonly accepted despite their deviation from the artist’s original conception while other instances of decay within the same work are not. In baroque painting, especially landscapes, there is a tendency to concede the appearance of brown paints that we know were originally green, but the pale transparent hues of paints once tinted with fugitive red lakes immediately suggest color reinforcement. In practice, the desire to keep compensation to a minimum tends to allow only the most efficient efforts to unveil aspects of the artist’s intent. In the baroque paintings, there is compelling economy of treatment in a decision to touch up red accents and leave browned [ones] alone.”

At first glance, the decision to retouch one area over another may appear to be arbitrary — why replace the red lake glazes but not return the brown landscape to green? But Dykstra came to a logical conclusion about which areas should be retouched by understanding the practical constraints of the paint layer and applying the ethical guideline of minimal treatment. In the conservator’s judgment, too much opaque overpaint would be required to return the brown landscape to green, but the red glaze could be applied transparently and delicately, allowing the artist’s original brushwork to remain legible beneath it. Ultimately, the conservator must concede to making compromises in such cases.

The rationale for retouching on top of original paint could also originate from a conservator’s desire to postpone a painting’s next varnish removal; applying some retouching above an existing varnish layer to extend the life of the previous treatment can momentarily preclude exposing the painting to the rigors of cleaning. Further challenges may be introduced when a conservator devises a treatment for a previously restored picture where there is good reason for retaining those restorations and incorporating them into the current treatment. One example is a picture that was restored or reworked by the artist himself sometime after it was completed, and the later additions that are now discolored no longer serve their intended purpose. In some cases, the historical importance of these alterations could outweigh the impulse to remove them, and the conservator must consider whether to implement glazing as a means for achieving more acceptable transitions between original paint and later additions.

Finally, a conservator’s approach should not be based on a fundamentalist notion in which aesthetic quality — a primary value of a work of art — is ignored in favor of strict technical rules, for to do so would lead to failure. Keith Christiansen (Christiansen 2003) thoughtfully articulated this concept when he reflected on the conservation history of Yale University’s Jarves Collection:

“The tragedy of the Jarves Collection…resulted from zealously pursuing a methodology: a methodology that sacrificed concern for the appearance of a work of art to its documentary value. Moreover, it confused such moral terms as honesty and truth with the complex histories of works of art and the ambiguities they present both to those who, as a matter of course, oppose modern restoration practices, and those who are committed interventionists.”

Christiansen recognized that a conservator’s work is subjective because of taste, sensitivity, scholarship, and training, but that subjectivity is a positive factor because it leads to an “improved” picture rather than simply a “truthful” picture (Christiansen 2003). Thus, if a conservator is concerned with the aesthetic value of a picture and the notion of artist’s intent, some degree of glazing or covering of original material must be acceptable in certain cases, with the understanding that the method of application is consistent with the ethical imperatives of reversibility and documentation.

It is essential that conservators delineate ethical principles in order to maintain the integrity of paintings, artists, and themselves as professionals. The AIC Code of Ethics and Guidelines for Practice devotes only one short paragraph to the topic of compensation, and it offers no specifics with regard to inpainting. Indeed, it lumps together filling and inpainting under the umbrella term compensation without consideration for the distinctions between the ethical questions each may present. Treatment Section 23 “Compensation for Loss” of the Guidelines for Practice states:

“Any intervention to compensate for loss should be documented in treatment records and reports and should be detectable by common examination methods. Such compensation should be reversible and should not falsely modify the known aesthetic, conceptual, and physical characteristics of the cultural property, especially by removing or obscuring original material..”

Because this statement is meant to serve all conservation professionals, AIC has encouraged the specialty groups to provide additional guidance with a higher level of detail to its practitioners. The purpose of this essay is not to present a step-by-step, instruction manual-type methodology for the conservator to follow, but rather to persuade him or her to adopt a well thought-out and philosophical approach that can be modified to suit each individual painting.

While the principle of reversibility is established and widely accepted as a fundamental statute of any retouching treatment, the concepts of authenticity and unity must also be considered to justify the relative degree to which a painting should be retouched. Scholarship and scientific analysis can further rationalize and validate a retouching decision for a particular painting. The aesthetic, cultural, historical, and educational value of a painting determine the context in which a painting is restored, and each of these must be factored into treatment decisions. Because that context varies for each particular work of art, no definitive approach is capable of delivering an ethical justification for inpainting in every case. However, if the current generation of conservators continues to assess the influential judgments of the past and reflect on how its restorations will be evaluated in the future, the ethics of inpainting will likely continue to be revised as the objectives of restoration treatments and the interpretation of artworks change over time.

Rita Albertson and Winifred Murray

Submitted July 2009

Ethics Glossary: A Reference for Inpainting-Related Concepts

Artist’s Intent: The concept of artist’s intent generally refers to how the artist wanted the work to be seen at completion and sometimes includes projections as to how it should be seen in the future (Hedley 1993, 154).The primary layer of intent consists of measurable, material things — the choice and preparation of media; size, shape, and order of brushstrokes; idiosyncrasies of drawing, modeling, and line — that make up a particular artist’s distinctive artistic style (Dykstra 1996 ).1 The secondary layer goes beyond the technical characteristics of a painting: the psychological insights, social and intellectual purposes, and aesthetic effects (Dykstra).

Compensation: Reconstruction of missing design (Ruhemann 1968). The act of making up for, offsetting, or counterbalancing (Webster’s II New Riverside University Dictionary).

• Approximate compensation: Putting into [lacunae] paint that comes near to matching that of the original which is adjacent and in this paint some tone variations may be introduced (Stout 1941).
• Remote compensation: Putting into the lacunae paint of a flat, neutral tone, said to be without drawing (Stout 1941).

Glaze: A transparent or semi-transparent paint layer, applied either directly over the ground or over an underpaint to modify the color of the ground or underpaint. With the use of appropriate glazes, the tonal range of a color can be extended to give a greater contrast between highlights and shadows. Since the glaze is transparent, the color of the underpaint or ground, which is generally lighter, plays a part in the final optical effect.

The term glaze is often used imprecisely to mean any thinly applied paint layer, particularly the final finishing touches to a painting. However, these can include combinations of pigments and media that are normally opaque or semi-opaque: it is simply the thinness of application that allows the underlying layers to contribute to the optical effect. Properly, thin layers of opaque color applied to modify the underlying colors should be called scumbles.

Glazes are usually made by combining pigments and paint media that have similar or identical refractive indices: that is, they transmit light to a similar or equal degree so the light is not bent or refracted at the interface between the medium and the pigment particle. Media such as egg tempera, glue distemper, casein, gouache, and some paints based on synthetic resins have refractive indices that are lower than those of any artists’ pigments: therefore they cannot be used to make truly transparent glazes. At best, the paint layers can only be translucent. Drying oils, however, have higher refractive indices, approaching those of certain pigments. Their refractive indices can be raised further by the addition of natural resins, usually those used in varnish making — for example, copal, sandarac, pine resins, mastic, and dammar.

Pigments that form transparent or semi-transparent glazes when combined with drying oils fall into three groups. The first contains pigments with refractive indices close to those of drying oils. They include several blue pigments (e.g., natural and artificial ultramarine, smalt, and Prussian blue), but they are not completely transparent because their refractive indices are still slightly higher than those of the oil-based media. The second, and perhaps most important, group comprises the red and yellow lake pigments. The red and yellow dyestuffs from which they are made do not themselves have low refractive indices, but when they are to be used as transparent lakes they are precipitated on to colorless substrates of low refractive index, traditionally alum (hydrated alumina) or chalk (calcium carbonate). The third group consists of pigments that are wholly or partially soluble in the paint medium. Since there is no longer any interface between pigment and medium to refract and scatter the light, the refractive index of the pigment is irrelevant. This group includes copper resinate (verdigris dissolved in an oil-resin medium), such colored resins as gamboge and dragon’s-blood, and the various bituminous brown pigments. (Grove)

The term glaze denotes a thin layer of transparent or semi-transparent color laid over another underlying color. The transparent colors used in glazes have low refractive indices (e.g., lake, ultramarine, copper resinate, Prussian blue). The careful building up of final colors by using layers of glazes was common from the 15th to the 19th centuries until the more widespread availability of stronger, industrially produced paints(Concise Oxford Dictionary of Art Terms).

Inpainting: Term used in the conservation of paintings or objects for the toning or imitative matching of an area of paint loss, without obscuring any original paint. (Grove) Applying new paint on areas where original paint has been lost or abraded (CCI 1994 ). This term first appears in American conservation literature in 1935 in an article by George Stout and more specifically refers to the act of adding color within the confines of a discrete loss or abraded area. Stout defines the “placement of new paint in the lacunae of the old film” as “painting in (‘inpainting’)” (Stout 1935, 210).

• Cosmetic inpainting: Ornamental rather than functional; having little or no significance; superficial (Webster’s II New Riverside University Dictionary). Wholesale repainting with resulting personal interpretation, often deliberate falsification, designed to upgrade attributions and conceal the ruins (Modestini 2003, 209–210).
• Deceptive inpainting: Intended or tending to mislead, dupe, delude, fool, hoodwink, misrepresent, trick (Webster’s II New Riverside University Dictionary).
• Differentiated inpainting: Constituted or showing a distinctive difference in or between; distinguished; discriminated (Webster’s II New Riverside University Dictionary).
• Evidenced inpainting: Present and plainly visible; conspicuous (Webster’s II New Riverside University Dictionary).
• Imitative inpainting: Imitative retouching aims to reproduce the hue, tone, texture, and reflective properties of the adjacent original paint such that the disruption caused by the damage is imperceptible and the illusion is restored to the picture. Tending to duplicate exactly, copy the appearance of, reproduce or mimic (Webster’s II New Riverside University Dictionary).
• Invisible inpainting: Not easily detected or noticed; imperceptible (Webster’s II New Riverside University Dictionary).
• Mimetic inpainting: Using imitative means of representation (Webster’s II New Riverside University Dictionary).
• Neutral inpainting: A type of visible inpainting consisting of a plain color — a so-called “neutral” tone — chosen to blend unobtrusively with the overall tone of the painting (The Oxford Companion to Western Art).

Lacuna: [Latin: gap] Completely blank area on a painting or painted object or manuscript, resulting from any form of damage. On paintings, the cause may be the gradual decay and loss of adhesion of the paint layers, allowing flakes of paint to become detached. Accidental damage can give rise to larger lacunae (Grove Art Online). The plural form is lacunae or lacunas; adjectival form is lacuna.

Loss: Impairment of structural unity. Among these are fractures, splits, or checks, punctures, tears, and corrosion (Stout 1935, 205). Scars or actual lacunae in the paint film. The latter ordinarily fall into one of three types: abrasion or skinning, the superficial wearing or rubbing away of the paint; flaking, the separation and loss of small portions of the film above the level of the ground and without a total loss of color; and scaling, the separation and loss of large parts of the film, usually islands, which leave the ground exposed (Stout 1935, 210).

Losses may be divided into five different types:

1. Wear of the patina...due either to abrasion or to the loss of minute flakes of paint beneath which a part of the pigment layer or at least the original rendering remains.

2. Wear of the pigment layer.

3. Complete losses of the pigment layer...limited in surface area and capable of being reconstructed.

4. Complete losses of the pigment layer...which, because of their extent and/or localization, should not be reconstructed.

5. Losses of considerable extent, which should nevertheless be reconstructed because of their architectural significance (Mora, Mora, and Philippot 1996, 348).

Overpainting: Paint not applied by the artist that covers original paint and that is often an excessive and unnecessary alteration to the image; overpaint hides areas of abrasion, or is used to reinforce the image or make changes to it (CCI, 7). Covering old paint with new (Stout 1935, 210).

Pentiment: [pentimento; Italian: repentance]. Visible evidence of an alteration to a painting or drawing that suggests a change of mind on the part of the artist. In particular, it refers to previous workings…revealed by the change in the refractive index of oil paint that occurs as it ages: thin layers of paint that were originally opaque may become semi-transparent. For example, in Titian’s group portrait of the Vendramin Family (c. 1543–47; London, N.G.), the figure of a young, bearded man on the far left was moved inward. The head of the figure in the original position is now evident as a ghostly image on a patch of sky. The term is also used to refer to such effects where they do not necessarily imply a deviation from the original intention. For example, in Pieter de Hooch’s Interior (London, N.G.), the checkered floor is visible beneath a maid’s dress, confirming that the figure was added after the floor was painted. This may have been necessary, given the precise geometric pattern of the floor and the perspective involved. Pentimenti suggest that painters refined and altered compositions as they worked, and, for this reason, they are often cited as evidence of authenticity; similarly, they are less likely to appear in copies. The term is also used to describe the hesitant preliminary workings that show beneath some drawings. (Grove)

Reintegration: Filling in the loss and inpainting the missing area to create the illusion of the original complete image and reintegrate the area of loss into the visual image of the whole (Caple 2000, 119). Careful retouching of gaps (integration) (Ruhemann 1968, 61).

• Functional reintegration: Where the area is restored to make the object stable (Caple 2000, 119).
• Background reintegration: Where the filled area is given a color and texture to blend with the base or background color and texture of the object; this makes it less visible but does not suggest any decorative scheme, often because information on the exact nature of the original decorative scheme is not available.
• Similar reintegration: The filled area is pigmented to give a crude approximation to what was originally present, although no detail is included and exact coloration is deliberately not achieved ((119)).
• Exact reintegration: The filled area is restored to its original appearance with all elements of the design included and correctly colored; as for all restoration, “exact” reintegration must still be detectable on close examination to avoid faking and deception; to undertake this level of reintegration, there must be clear and detailed information about the design and pigmentation originally present in the area of loss (120).

Restoration: “The practice of treating pictures or other works of art to aid in their conservation or to bring them nearer to their original condition” (Stout 1935, 203). Filling in missing areas; the “painting” or “retouching” of the filled areas or of those parts of the surface of the object where the original decoration or painted layer has been lost (Grove Art Online).

• Fragmentary restoration: Conserving the original material by cleaning, stripping, or removing overpaint, but not by retouching, filling losses, or making reconstructions; the painting is left in a fragmentary state (von der Goltz 1999, 202).
• Documentary restoration: After cleaning, losses and holes are filled to restore the in a neutral gray; alternative to fragmentary with greater concern for the spectator (202).
• Complementary restoration: Treatment is to “complete” the painting and achieve the original effect created by the artist without any creative acts by the restorer; the artist’s intent is the guiding principle (202).

Retouching: The act or process of adding new fine lines, strokes, or tinges of color for correction or improvement (Webster’s II New Riverside University Dictionary). A traditional term that has been used synonymously with inpainting; inpainting is more precise because retouching can also imply overpainting so that original paint is covered (CCI 1994, 8). Coloring a defective area of a painting (Nicolaus 1999, 257).

Scumble: A thin layer of opaque or semi-transparent paint, applied on top of another layer without completely hiding it. It is effectively the opposite of a glaze and gives a broken and hazy effect, toning down the brilliance of the underlying color (Grove). Essentially, the partial application of one layer over another. In painting, a thin layer of opaque or semi-opaque color applied over an area of an oil painting without completely obscuring the underpainting (Concise Oxford Dictionary of Art Terms).

Tratteggio: Transposing the modeling and drawing of a painting into a system of hatchings based on the principle of the division of tones…A system of small vertical lines averaging one centimeter in length. The first lines, which indicate the basic tone of the retouching, are placed at regular intervals equal to the width of one line. Next, these intervals are filled with a different color, and then again with a third color, to reconstitute the required tone and modeling by means of the juxtaposition and superposition of colors that are as pure as possible. Each line in itself should be weak in intensity, the desired intensity of the whole being obtained by the superposition of glazes of transparent lines rather than by strength of color, which would cause the retouching to lack the vibration indispensible for a good integration (Mora, Mora, and Philippot 1996, 351–353).

In different centers in Italy, where visible methods of restoration are perhaps most widely applied, several techniques have been developed and their methods and philosophies codified. In the simplest of these, the restoration can be recognized by the use of regular and equal sized vertical brushstrokes, called in Italian tratteggio or rigatini. Areas of damage may be reconstructed and the colors of the original matched, but the brushstrokes always remain consistent in their direction (Grove).

• Chromatic selection (selezione cromatica): A variant of the tratteggio developed in Florence…The brushstrokes are allowed to follow the direction of the form of the area being restored, for instance the sweep of a drapery fold, or can be applied vertically and diagonally to form an interlaced mesh of brushstrokes…The colors for the restoration are chosen from the three primary and three secondary colors and mixed optically by the juxtaposition and interlacing of stippled dots of color, as in the Pointillist color theories employed by, for example, Georges Seurat. It is argued that, while remaining readily detectable, the technique produces a vibrant effect suited to the intense palette of medieval and Italian Renaissance paintings.
• Chromatic abstraction (astrazione cromatica): Similar optical principles as chromatic selection…also of Florentine origin. This method is employed when the losses are too large to suggest the forms and colors that may have been present and is intended to supply a more satisfactory solution than the so-called “neutral” restorations. The strokes, which are not directional, are painted with the three primary colors and black, the proportions of the colors determined by (or “abstracted” from) the colors of the areas surrounding the loss. If, for example, the loss is principally in an area of blue drapery but also extends into areas of red and green, a preponderance of blue strokes may be applied, but the presence of some red and yellow (suggesting green) strokes will in theory make the restoration visually compatible with the red and green areas as well.

Author’s note: The authors compiled this reference for inpainting-related concepts from nonexhaustive research to clarify terms used in the literature, many of which can be found in the authors’ preceding submission, “The Ethics of Inpainting.”

Rita Albertson and Winifred Murray

Submitted July 2009

Endnotes

¹ Contreras de Berenfeld (2003) notes that earlier, in the 1930s, Offner had a more tolerant attitude toward some examples of visible retouching.

²G. Vasari, The Lives of the Artists. Trans. Julia Conaway Bondanelle and Peter Bondanella (Oxford: Oxford University Press, 1991), 269. The painting, actually on panel, is now in the National Gallery, London. Recent xrays have shown that under the Christ Child attributed to Sodoma are two earlier versions painted by Signorelli. See T. Henry and L. Kanter, Luca Signorelli: The Complete Paintings (New York: Rizzoli, 2002), 175–6.

³One of Bottari’s speakers quotes Vasari on Signorelli, Sodoma, and retouching, p. 242. The antiquarian Luigi Crespi wrote in the mid-18th century, “It is better to see an old painting consumed and ravaged by time because at least the little that you see is pure and chaste…(. preferibile vedere un antico dipinto, dal tempo consunto e corroso, poich. almeno quel poco che vede, vergine il vede ed illibato, che da mano oltre il dovere coraggiosa (per non dire di più) ritoccato e compito, discordante veggendolo, crudo e disformato).” G. Bottari, Lettere pittoriche, 1754–73, quoted in M. G. Vaccari, Restauri a Firenze nel Settecento: Tra Teoria e Prassi. Kermes (1989), Vol. 2, No. 4, 39–44. Filippo Baldinucci wrote in Notizie de’ Professori del Disegno (revised 1767 ed.) that Guido Reni “grew furious when he heard that any painter had dared to touch paintings by the old masters, even if torn and ruined; this was something that he would never undertake. (Dava nelle furie quando sentiva che alcun pittore avesse ardito di toccare pitture di antichi maestri, tutto che lacere e guaste; cosa che egli non volle mai fare).” Quoted in U. Forni, 1866. Mannuele del Pittore Restauratore. (Florence: Successori le Monnier), 9.

⁴ A. Conti, Storia del restauro o della conservazione delle Opera d’Arte (Milan: Electa, 1988), 12, and G. Ragionieri, Simone o non Simone (Florence: La Casa Usher, 1985), 26.

⁵ G. P. Bellori’s Vita di Carlo Maratta (1731) discussed the painter’s use of pastels in the 1693–94 restoration campaign: “with a little lapis and pastel he perfectly restored the contours and the color (con un poco di lapis e di pastello é perfettamente ristorato i contorni ed il colore)”. According to the writers of the Encyclopédie, Maratta did this so that his retouching could be removed if a more worthy restorer were found:

“Carlo Maratta having been chosen the first painter in Rome and working on the ceiling of the Farnese Palace, on which Raphael had depicted the story of Psyche, wanted to retouch only in pastel so that, he said, if one day someone was found who was more worthy than myself to link his brush with that of Raphael, he could erase my work and substitute his own. (Carle Maratte ayant été choisi comme le premier peintre de Rome, pour mettre la main au plafond du palais Farnese, sur lequel Raphäel a représenté l’histoire de Psyché, il n’y voulut rien retoucher qu’au pastel, afin, dit-il, que s’il se trouve un jour quelqu’un plus digne que moi d’associer son pinceau avec celui de Raphäel, il puisse effacer mon ouvrage pour y substituer le sien).”

Jaucourt. 1795. Ecole Romaine (Peinture); Maratte, (Carle). In Encyclopédie Raisonné des Sciences, des Arts, et des Métiers, eds. Diderot and d’Alambert, Neufchastel. Vol. 5, 330.

⁶Bellori wrote in 1695 that Maratta was

“embarrassed, needing on the one hand to obey the pope and on the other not having the impudence to erase even one line of such a great man…He came up with a subtlety to obey the pontiff and to leave the work intact. He took chalk pastels ground in gum and with them painted in such a manner that it would remain durable a veil over the breast of the Virgin as the pope wanted; and whenever some one wanted to remove it with a sponge, the original color would return (confuso, dovendo da un canto ubbidire il papa, dall’altro non avendo ardire di por mano e cancellare né meno un tratto di sì gran uomo,… pensò ad una finezza d’ubbidire al pontefice e lasciar l’opera intatta. Pigliati dunque colori di pastelli di terre macinate a gomma, con essi dipinse il velo sopra il petto della Vergine come voleva il papa, in modo che rimane duribile; e quando si voglia torre con la sponga, ritorna il color di prima). Quoted in Conti, p. 108.”

Bottari gives these words to Maratta in his Dialoghi:

“Maybe with so sacred an authority [a text by St. Cyprian] I would have been able to stop the pope and induced him to relieve me of doing this thing which worried me for a month; however, without telling the pope anything concerning what I had decided to do, I made that bit of veil with chalk pastels ground in gum so that it could be removed at any time (forse con un autorità così veneranda avrei fermato il Papa, e indottolo a dispensarmi a far cosa, che mi tenne un mese strubato; benchè senza dirgli niente di come avea pensato di fare, condussi quell poco di velo con pastelli di terra, macinati a gomma, sicchè si può tor via ogni volta, che un vuole), 245.”

⁷S. Bergeon, “Quelques aspects historiques à propos de restaurer ou dérestaurer les peintures murales.” In Les Anciennes Restaurations en Peinture Murale (Paris: International Institute for Conservation of Historic and Artistic Works/Section Fran.aise, 1993).

⁸Articles in Ristaurazione delle Pubbliche Pitture Assentita col Decreto del Sento li 3 Settembre 1778. Quoted in Conti pp. 164–6, 346. Also, Wolfgang Goethe discussed Edwards’ studio at SS. Giovanni and Paolo, writing,

“The institute has the advantage that all the experimental findings made in this art [of paintings restoration] are collected and preserved. The methods and technique of restoring each particular picture are very different, according to the various masters and the condition of the painting itself…The condition of every picture is first of all examined and assessed before it is decided what to do with it.”

Goethe, W. 1790. “Early Paintings and Recent Restorations in Venice”. In Goethe on Art, ed. J. Gage (Berkeley: University of California Press, 1980), 124–9, 128.

⁹Article XIV of the rules set down for the restoration of the paintings belonging to the Venetian State: “Finalemente s’imegnano di non usare sui quadri ingredienti che non si possano più levare, ma ogni cosa necessariamente adoperata sarà amovibile da quelli dell’arte ogni qual volta si voglia.” From Ristaurazione delle Pubbliche Pitture Assentita col Decreto del Sento li 3 Settembre 1778. Quoted in Conti, 165.

¹⁰Pietro Edwards wrote specifications for the restorer Antonio Florian for the 1816 work on Titian’s Assumption of the Virgin. He stated, “The above mentioned restorations shall be made upon a ground of gesso, tempered with weak colla di ritagli and the parts to be repainted, as well as the indispensable reparation of portions of the old colour, shall be executed with varnish, the use of oil being absolutely excluded…” The bill of expenses accompanying the contract indicated that the prescribed varnish in this case was mastic. Contract transcribed by M. Merrifield, Original Treatises on the Art of Painting, reprint of 1849 edition (Dover, New York, 1967), 876.

¹¹Conti, 187.

¹² “Direbbesi restaurare, o resarcire, o riddurre a bene essere, il raccommodare che si fa qualche volta alcuna piccola parte di pittura anche d’eccellente Maestro che in alcun luogo fusse scrostata o altrimenti guasta, perché riesce facile a maestra mano; e alla pittura non pare che altro si tolga che quel diffetto, che, quantunque piccolo, par che le dia molta disgrazia e discredito. Molti però, non del tutto imperito dell’Arte, sono stati di parere che l’ottimo pitture né punto né poco si ritocchino, anche da chi si sia, perché, essendo assai difficile che o poco o molto, o subito o in tempo, non si riconosco la restaurazione per piccola che sia, è schietta va sempre accompagnata con gran discredito.” F. Baldinucci, Vocabulario Toscano dell’Arte del Disegno (Florence, 1681).

¹³“Bien des gens se mȇlent de repeindre les endroits endommmagés des tableaux, dans le dessein de les réparer; mais rien si difficile à executer de maniere que la nouvelle couleur ne fasse pas des taches. On est obligé de salir les couleurs que l’on couche, pour trouver le vrai ton de l’ancienne; l’huile que l’on emploie noircit et produit ces taches. Il faudroit repeindre à détrempe, pour ne pas s’exposer à cet inconvenient; ce seroit le moyen le plus sûr, si la détrempe pouvoit s’unir intimement avec la peinture à huile. On lit dans l’Ouvrage de Messieurs de Caylus et Majault sur la peinture à l’encaustique e à la cire, page 131, que les couleurs préparées pour la peinture à la cire, conviendroient beaucoup mieux que les couleurs à l’huile pour restaurer les vieux tableaux. M. le Lorrain, peintre de l’Académie, l’a essayé avec un succès singulier. Il a repassé des vieux tableaux de cette manière, de façon qu’il est presque impossible de retrouver les endroits réparés et repeints.” A-J. Pernety, Dictionnaire Portatif de Peinture, Sculpture et Gravure: Avec un Traité Pratique des Différentes Manières de Peindre, reprint of 1757 edition (Geneva: Minkoff Reprint, 1972), 499.

¹⁴ The early committee work defining the proper care and restoration of paintings did not stave off controversy. In 1797, a painter/politician named Anthelme Marin accused the Louvre of storing paintings in stairwells and humid rooms and of damaging paintings through incompetent restoration. The Directory ordered an official inquiry and appointed a committee of 31 artists and experts to examine Marin’s charges. The report was an official government publication: Musée Central des Arts: Pièces Relatives à l’Administration de cet Établissement. Two years later rumors again circulated of misconduct at the Louvre, and the administration responded by appointing four experts from the National Institute to supervise the restoration of Raphael’s Foligno Madonna. The structural work was done by François-Toussaint Hacquin and the cleaning and retouching by Mathias Roeser. The supervising experts were the painters François André Vincent and Nicolas-Antoine Taunay and the chemists Claude-Louis Berthollet and Louis-Bernard Guyton de Morveau. See A. McClellan, “Raphael’s Foligno Madonna at the Louvre in 1800: Restoration and Reaction at the Dawn of the Museum Age.” Art Journal (1995) 54(2): 80–5.

¹⁵ “Ceci me porte à inviter les chimistes, pour la perfection de l’art, à faire en sorte de découvrir le moyen d’allier la poudre des couleurs avec un liquide qui remplace l’huile, sans avoir les mȇmes inconvénients, qui sont de pousser à la superficie et de dénaturer les couleurs en absorbant leur éclat.” G. Émile-Mâle, 1956. “Jean-Baptiste-Pierre Le Brun (1748–1813) (Getty): Son Rôle dans l’Histoire de la Restauration des Tableaux du Louvre.” In Paris et Ile-de-France, Mémoires 6:371–417, 402.

¹⁶ C. Wagner, Arbeitsweisen und Anschauungen in der Gemälderestaurierung um 1800. (Munich: Callwey, 1988), 28.

¹⁷ C. Kӧster, Uber restauration alter Oelgemälde (Heidelberg: Christian Friedrich Winter, 1827–30) Vol. 1, 27–8 and Vol. 3, 16–17.

¹⁸ Forni, 88–9 and 143–5.

¹⁹ Kӧster, Vol. 3, 17–18.

²⁰ H. Mogford, Handbook for the Preservation of Pictures: Containing Practical Instruction for Cleaning, Lining, Repairing, and Restoring Oil Paintings (London: Winsor and Newton, 1851), 63–4.

²¹ G. Secco Suardo, Il Restauratore dei Dipinti, written between 1866–73 (Milan: Ulrico Hoepli, 1918), 500, 536.

²² M. Scharff, “Insight into Early Nineteenth-Century Painting Conservation in Denmark.” In eds. A. Roy and P. Smith, Tradition and Innovation, Advances in Conservation, Contributions to the Melbourne Congress 10–14 October 2000 (London: IIC, 2000).

²³ Edward’s thoughts on this subject are recorded in Progetto per una Scuola di Restuaro written after 1819. See Conti, 187.

²⁴ Picault was criticizing the idea that painters only were capable of directing the national museum. The text in French is as follows:

“Observe, au contraire, ledit Picault, que l’art de peindre et celui de restaurer ne se ressemblent en rien; que le peintre qui peut produire un chef-d’oeuvre, gâtera les chefs-d’oeuvre d’un autre, en voulant les restaurer; que dans un tableau malade et défectueux, le plus célèbre peintre substituera sa manière à la manière de Raphäel, des Carrache et de Titien; qu’il ne résultera de sa retouche qu’un assemblage monstrueux dont l’effet assuré sera de déprécier le tableau…que pour restaurer avec succès, il faut une étude particulière de preparation première, d’opération aussi utile qu’indispensable, de même qu’une habitude consommée des manières et préparations des maîtres de chaque école qu’il s’agit de retrouver, si elles sont perdues, et de conserver, si elles ne sont qu’altérées, et rendre, en un mot, jusqu’aux nuances les plus légères qui caractérisent, et chaque école et chaque maître…le restaurateur qui étudie tous les maîtres et toutes les écoles ne s’est point fait, et n’a point dû se faire comme le peintre, une manière à part. Il a fait le sacrifice de ses propres idées pour se plier aux idées d’un autre; il n’a plus d’existence à lui….”

J-M. Picault, Observations de Picault, Artiste Restaurateur des Tableaux à Ses Concitoyens sur les Tableaux de la République (Paris: Imprimerie de H.J. Jansen, 1793), 37–8.

²⁵ Kӧster reminded restorers to take into account tempera’s delicateness of color, fineness of brushstrokes, and undermodeling of flesh tones, adding, “In truth tempera painting has reached such a degree of delicacy that whoever sees it must be astonished if he only knows oil painting” (Wirklich hatte die a tempera Malerei einen Grad der Feinheit und Delicatesse erreicht, worüber der jenige erstaunen muss, welchem von jeher nur die Oelmalerei im Sinne gelegen). Vol. 2, 18.

²⁶ For example, early Italian paintings tended to be in good condition since they were executed with sound techniques and often remained at the site for which they were commissioned. The problems encountered were usually problems with the supports or the paint layer covered with centuries of dirt, 145–6. S. Horsin-Déon, De la Conservation et de la Restauration des Tableaux. (Paris: Chez Hector Bossange, 1851), 127–214.

²⁷ Forni himself wrote a long section on pigments, describing production, properties, stability, and notable uses by certain artists, 285–427.

²⁸ Conti, 246.

²⁹ D. Levi, Cavalcaselle, il Pionere della Conservazione dell’Arte Italiana (Turin: G. Enaudi, 1988), 337–9, 347.

³⁰ “Poco importa che si conosca il restauro, che anzi lo si dovrebbe conoscere, ma quello che é necessario si é che sia rispettato l’originale della pittura almeno nelle opere appartenenti allo Stato. La bugia, anco detto con bel garbo deve essere tolta di mezzo. Lo studioso potrà conoscere da un dipinto restaurato a questa maniera quello che . originale da quello che é nuovo…” Levi, 350-1.

³¹ From Über Ölfabre 1870, quoted in H. Althӧfer, La Questione del Ritocco nel Restauro Pittorico. Italian translation of 1974 German text (Padua: Il Prato, 2002), 20-1.

³² “Die malerische Herstellung macht allemal den Beschluss. Je weniger Einwirkung dabei nӧtig ist, desto besser, wo keine nӧtig ist, bleibt sie hinweg” (Kӧster, Vol. 3, p. 24). “La meilleure des restaurations est celle qui est obtenue par un travail léger et transparent qui laisse le maître apparaître partout où il existe encore” (Horsin- Deon, 116). “Il restauro migliore é quello ottenuto da un lavoro economico, che lascia intiere tutte le parte dell’originale” (Forni, 153).

³³ “Notre mission, encore une fois, est de montrer les maîtres tels qu’ils sont, de les vénérer avec leurs qualités et leurs défauts… Retoucher en quelque sorte avec une aiguille les trous qui feraient une tache insupportable, combler discrètement les lacunes occasionnées par la chute des écailles sans étaler la couleur sur les bords de la cicatrice, voilà toute que la restauration premise.” F. Villot, “Restauration des Tableaux du Louvre.” In Annuaire des Artistes et des Amateurs. (Paris: Jules Renouard, 1860), 265–274, 271.

³⁴Courtesy of Gay Myers: Early advertisement for restoration in the New York Gazette and Weekly Mercury, May 8, 1769, ad for William Williams, who will undertake general painting, and “He cleans, repairs and varnishes any old pictures of value”; Williams also advertises that he teaches art. Quoted in William Kelby, Notes on American Artists: 1754–1820 (New York: New York Historical Society, 1922) , 7.

³⁵Courtesy of Gay Myers: According to Miller, many of Peale’s early portraits were repainted because they faded and there was a “shallowness” in the flesh tints. C. W. Peale later wrote of works of this period: “The shadows [are] too cold, almost black, having used no red in my shading except lake. The fading of the lake left the black predominant in the middle tints and deep shades …Had I used vermillion or light red, how much better these paintings would have been.” See Lillian B. Miller, The Selected Papers of Charles Willson Peale and His Family, Vol. 1 (New Haven and London: Yale University Press), 103–04

³⁶Courtesy of Gay Myers: Paintings by West that are signed and dated twice include scenes from Hamlet and King Lear. See Helmut von Erffa and Allen Staley, The Paintings of Benjamin West (New Haven and London: Yale University Press, 1986) nos. 207 and 210, both of which are signed, “Retouched 1806,” as is The Battle of La Hogue. The retouching of the two Shakespeare scenes can be documented to the time that they were in the collection of Robert Fulton, shortly before they were sent to America: “Mr. West has been retouching my pictures; they are charming.” See H. W. Dickinson, Robert Fulton, Engineer and Artist, His Life and Works, (London: John Lane, 1913), 200. See also Rembrandt Peale, Misc. papers, American Philosophical Society: “[West’s] Picture of Lear in the Tempest, after being many years varnished, he glazed with Asphaltum, Lake & Blue, & retouched it previous to Fulton’s bringing it to America. Our dry atmosphere by contracting these thick glazings & retouches, which were on a soft varnish, has nearly destroyed it, the cracks being more than the eighth of an inch wide.” We (Mayer and Myers) treated the scene from Hamlet (now in the Cincinnati Art Museum) and believe that Peale was probably wrong: West’s retouching and his second signature clearly go over the wide cracks, indicating that concealing the disfiguring crackle was the likely motivation for retouching it, rather than the cracks having been caused by the retouching. Note that West wasn’t inpainting but was generally “going over” the painting with his retouching. Farington was critical of West’s retouching his paintings years after they were painted, saying that he did them “injury.” See Robert C. Alberts, Benjamin West, A Biography (Boston: Houghton Mifflin Company, 1978) , 383. But West, in a letter to the Royal Academy in 1803, justified the practice, saying that when he signed and dated a painting a second time, he considered it “a new picture” (B. West to the Members of Council of the Royal Academy, April 16, 1803, Historical Society of Pennsylvania, AAA roll P23, frame 225). See also Alberts (1978), 280–85.

³⁷Courtesy of Gay Myers: Claire Barry’s article on Cole. Claire M. Barry, “Technical Note: ‘Painting in Imagination’: The Creation of The Garden of Eden by Thomas Cole,” in Franklin Kelly, Thomas Cole’s Paintings of Eden (Fort Worth: Amon Carter Museum, 1994), 55–66. Barry describes traction crackle in Garden of Eden (now in Amon Carter) and other paintings (especially in dark darks and greens), which she attributes to media containing oil varnish or asphaltum or to under-dried underlayers (p. 56); she also mentions the possibility of him varnishing his paintings too soon; the cave area was overpainted, probably because of serious traction crackle (Cole scraped out the area and painted the cave, possibly late in the process) (p. 56).

John Trumbull retouching his own works because of crackle: In 1829 Thomas Sully visits Trumbull in New York where Trumbull is restoring his paintings of Montgomery and Warren: he coated the backs with wax, “and afterwards carefully filled up with paint all the spaces occasioned by the cracks on the paintings.” Trumbull also showed Sully a copy of Correggio’s St. Catherine he did 40 years before and it was in excellent condition, due to the fact it was painted with simple materials. (Sully, Hints (Yale Ms.) (AAA NYPL Papers Roll N18, frame 125).

³⁸Courtesy of Gay Myers: in 1840, Rembrandt Peale requested space for “unrolling, retouching and exhibiting” the painting. Documents from 1846 and 1847 suggest even more retouching had been done by that time. See L. Mayer and G. Myers, “The Court of Death Through Conservators’ Eyes,” Bulletin of the Detroit Institute of Arts, vol. 70l, no. 1–2 (1996), 8.

³⁹Courtesy of Gay Myers: Manuscript is in Historical Society of Pennsylvania, p. 47 (Hevner’s pagination).

⁴⁰William Rickarby Miller, Hints and Recipes Pertaining to Painting in Oil and Water-Colours and the General Practice of Art. June 1862. Owned by the New York Historical Society. Microfilmed by the Archives of American Art [reel number 801], 264.

⁴¹Courtesy of Gay Myers: Joyce Zucker, “From the Ground Up: The Ground in 19th-Century American Pictures,” Journal of AIC, vol. 38 (Spring 1999): 11.

⁴²G. Howorth, Restoration of Oil Paintings: With a Few Practical Hints to the Owners of Pictures (Boston: Press of George C. Rand and Avery, 1859), 6.

⁴³Howorth Broadside (in YUAG archives), A letter on its verso from John Howorth to Luther Armand Jones, 1867.

⁴⁴Mark Aronson, “The Conservation History of the Early Italian Paintings at Yale.” In Patricia Sherwin Garland, ed., Early Italian Paintings: Approaches to Conservation (Yale University Art Gallery, 2002), 32

⁴⁵James Jackson Jarves, Art Studies: The Old Masters of Italy (New York: Derby and Jackson, 1861), 49.

⁴⁶Mark Aronson, “The Conservation History of the Early Italian Paintings at Yale.” In Patricia Sherwin Garland, ed., Early Italian Paintings: Approaches to Conservation (Yale University Art Gallery, 2002), 32

⁴⁷Ibid., 32-34, as it originally appeared in Jarves. Art Studies, 348.

⁴⁸Ibid., 34, as it originally appeared in Jarves. “Art Mysteries Unveiled: Cleaning Old Master Pictures,” New York Times, April 14, 1878, 10, col. 3.)

⁴⁹Ibid.

⁵⁰Ibid.

⁵¹Ibid.

⁵²Ann Hoenigswald, Historical Papers. Stephen Pichette, Conservator of the Kress Collection. 1927–49, 30–41.

⁵³Ibid., 32.

⁵⁴Ibid.,16–17. Internal quotation: New York University radio broadcast, October 22, 1944.

⁵⁵Courtesy of Gay Myers. Ranger quoted in Ralcy Husted Bell, Art-Talks with Ranger (New York and London: G.P. Putnam’s Sons, 1914), 122

⁵⁶Courtesy of Gay Myers. Abbott Thayer Monadnock April 1916. For in-depth discussion of Thayer’s views against restoration, see Nelson C. White, Abbott H. Thayer: Painter and Naturalist. (Hartford, CT: Connecticut Printers, Incorporated, 1951), 243–252, as well as his 1920 essay from Studio International, “Restoration: The Doom of Pictures and Sculptures” (March 1920), 13–17, reprinted in The Arts, August/September 1921, after his death, p. 10–14.

⁵⁷The Sun, obituary, HUAM archive, Edward W. Forbes, “Teaching Materials” file, H. A. Hammond Smith.

⁵⁸H. A. Hammond Smith, Record, condition/treatment notes in the conservation department of the Yale University Art Gallery, p. 58.

⁵⁹H. A. Hammond Smith, p. 58.

⁶⁰Cesare Brandi, Theory of Restoration, I, Reading 22, 231.

Giorgio Bonsanti’s article “Theory, Methodology, and Practical Applications — Painting Conservation in Italy in the Twentieth Century,” in Early Italian Paintings: Approaches to Conservation, 2002, p. 82, refers to the origins of Brandi’s theories as the embodiment of earlier Italian concepts and theories, among them those of Pietro Edwards, an eighteenth-century painter of English origin, who came to Venice and was charged with setting up a team of artists to conserve the so-called public pictures. Bonsanti credits Edwards with being among the first to exemplify modern principles of conservation. For more on Pietro Edwards, see Wendy Partridge’s essay in this volume and Alessandro Conti’s book History of the Restoration and Conservation of Works of Art recently translated into English by Helen Glanville. In the United States, Elizabeth Darrow wrote her thesis on Pietro Edwards, and she has a paper in British museum Occasional Papers no. 145,2001. Past Practice-Future Prospects.

⁶¹A. M. Vaccaro, “The Emergence of Modern Conservation Theory,” Introduction to Part III. In History and Philosophical Issues in the Conservation of Cultural Heritage (Getty Conservation Institute, 1945), 204–205.

⁶²Giorgio Bonsanti’s article, “Theory, Methodology, and Practical Applications — Painting Conservation in Italy in the Twentieth Century.” In Patricia Sherwin Garland, ed., Early Italian Paintings: Approaches to Conservation (Yale University Art Gallery, 2002), 91.

⁶³Ron Spronk, “Early History of Conservation and Technical Studies at the Fogg Art Museum: Four Pioneers,” Harvard Art Museums Review 6 (1996), no.1, Iff. Also referenced in North American Graduate Programs in the Conservation of Cultural Property, 1999, 22.

⁶⁴Courtesy of Wendy Samet: Elizabeth H. Jones, “Edward Waldo Forbes: Herald of Conservation,” Edward Waldo Forbes: Yankee Visionary, 99–146.

⁶⁵Another type of nondeceptive inpainting possibly done by Lyon is illustrated on page 129 of Volume 8 of Technical Studies, July 1939–April 1940. It is vertical hatching — similar to tratteggio.

⁶⁶R. Arcadius Lyon, “Comments on Relining,” Technical Studies, Vol. II (1933–34): 221.

⁶⁷Ibid., 154-155.

⁶⁸Maximilian Toch, Paints, Painting and Restoration (New York: D. Van Nostrand Co., 1931), 60–61.

⁶⁹Irene Bruckle and Christopher Tahk, eds. North American Graduate Programs in the Conservation of Cultural Property, Histories/Alumni (Buffalo, NY: Petit Printing Co., 1999), 24.

⁷⁰George Stout. 1941. Treatment of blemished paintings. Technical Studies in the Field of the Fine Arts, Vol. X, Oct. 1941, No. 2.

⁷¹Ibid., October 1941, Stout’s footnote: Max. J. Friedlaender, Genuine and Counterfeit, Carl von Honstett and Lenore Pelham, trans. (New York: A. and C. Boni, 1930), 27–30.

⁷²Ibid., “Treatment of Blemished Paintings”, 107-112.

⁷³Helmut Ruhemann, “A Tentative Scheme for Analysis of Painting Technique,” Technical Studies, Vol. X (April 1942): 75.

⁷⁴Ibid., 3.

⁷⁵Ibid., As described by Bettina Jessell, adding wax for elasticity and ox gall for breaking the surface tension.

⁷⁶Bettina Jessel, Journal of AIC, Vol. 17, No. 1 (Fall 1977): 1–8.

⁷⁷Courtesy of Wendy Samet: Richard D. Buck and George L. Stout,” Technical Studies, Vol. 8, No. 3 (Jan. 1940): 122–150

⁷⁸Ibid., 129.

⁷⁹Ibid., 125.

⁸⁰Ibid., 123.

⁸¹OAC, Getty Research Institute, William Suhr papers, collection 870697 (online description).

⁸²William Suhr, “The Restoration of the Merode Altarpiece,” Metropolitan Museum of Art Bulletin, New Series, Vol. 16, No. 4 (Dec. 1957): 140–142.

⁸³Modestini, Dianne, “Approaches to Retouching and Restoration: Imitative Retouching.” In Patricia Sherwin Garland, ed., Early Italian Paintings: Approaches to Conservation (Yale University Art Gallery, 2002), 208–209.,

⁸⁴Lbid., 211.

⁸⁵Berger, Gustav, “Inpainting Using PVA Medium.” In J. S. Mills and P. Smith, eds., Cleaning, Retouching and Coatings, Preprints of the Contributions to the Brussels Congress, 3–7 September 1990, IIC, London, p. 150.

⁸⁶Charles Seymour, “The New Conservation Program for Italian Paintings,” Bulletin of the Art Gallery Associates 19, No. I (January 1951).

⁸⁷Aronson, 42.

⁸⁸Ibid., as it originally appeared in Charles Seymour, “The Aesthetic and Historical Aspects of the Presentation of Damaged Pictures,” Vol. 4, Problems of the 19th & 20th Centuries, Studies in Western Art (Princeton University Press, 1963), discussion section pp. 176–78.

⁸⁹Ibid., 150.

⁹⁰Modestini, Dianne, “Approaches to Retouching and Restoration: Imitative Retouching.” In Patricia Sherwin Garland, ed., Early Italian Paintings: Approaches to Conservation (Yale University Art Gallery, 2002), 208–209.,

⁹¹Volkmer, Jean. FAIC interview by Lyn Reiter, November 4, 1977, 13.

⁹²Ibid.

⁹³Philip Hendy., “In the Presentation of Damaged Pictures,” 1961 Princeton sponsored conference papers reprinted, p. 141.,

⁹⁴Ibid., 142-143.

⁹⁵Ibid., 145.

⁹⁶Ibid., 143.

⁹⁷Ibid., 156-157.

⁹⁸Ibid., 164.

⁹⁹Ibid., 166.

¹⁰⁰Ibid., 166-167.

¹⁰¹Ibid., 167.

¹⁰²Ibid., 168.

¹⁰³Ibid., 169.

¹⁰⁴Ibid., 170.

¹⁰⁵Ibid., 173.

¹⁰⁶Ibid., 175.

¹⁰⁷Personal interview with Sue Sack, spring 2008.

¹⁰⁸Ibid.

¹⁰⁹From an interview with Stephen Kornhauser, Chief Conservator at the Wadsworth Atheneum. He was a Mellon Fellow at the MMA under John Brealey, 1980–82.

¹¹⁰Ibid.

¹¹¹Elisabeth Packard, “Changing Approaches and Materials in Inpainting,” AIC Preprints, Fourth Annual Meeting, Dearborn, Michigan, 1976, p. 124.

¹¹²Lawrence Majewski, “Results of Rapid Age Testing on Various Inpainting Media,” AIC Preprints, Fourth Annual Meeting, Dearborn, Michigan, 1976, p. 125.

¹¹³Robert Feller, “Exposure Tests on Traditional and Polyvinylacetate Retouching Systems and Characterization of Certain Modern Proprietary Painting Media,” AIC Preprints, Fourth Annual Meeting, Dearborn, Michigan, 1976, p. 126.

¹¹⁴Ursus Dix, “Inpainting with Egg Tempera,” AIC Preprints, Fourth Annual Meeting, Dearborn, Michigan, 1976, p. 127.

¹¹⁵Bettina Jessel. 1976. “Helmut Ruhemann’s Inpainting Technique”. Journal of the American Institute for Conservation 17(1):1–8.

¹¹⁶Mary Lou White, “Rigattino: An Alternate Method of Inpainting,” AIC Preprints, Fourth Annual Meeting, Dearborn, Michigan, 1976, p. 133.

¹¹⁷Louis Pomerantz, “Inpainting with Bocour Paints,” AIC Preprints, Fourth Annual Meeting, Dearborn, Michigan, 1976, p. 134.

¹¹⁸Brandi’s students, the Moras, would write, “No tone can ever be neutral within the image/loss context, since it inevitably situates itself on a given plane and at a given depth.” Paolo Mora, Laura Mora, and Paul Philippot, Conservation of Wall Paintings (London: Butterworths, 1984), 312.

¹¹⁹Ciatti uses this term when describing an early 1930s restoration by Vermeheren, who was “summarily indicating the volumes without defining the details.”

¹²⁰Olsson refers to a recent restoration on parts of Duccio’s Maestà in the Museo dell’Opera, Siena.

¹²¹Although Prussian blue is a rather translucent pigment, its unusual tinting strength makes it highly covering.

¹²²Ann Temkin, Contemporary Voices: Works from the UBS Collection (New York: The Museum of Modern Art, 2005), 64.

¹²³Contreras de Berenfeld tells the story of Brandi’s early experience with non-imitative retouching, demonstrating that the origins of tratteggio were the result of a fascinating American-Italian exchange in the years leading up to World War II.

¹²⁴The philosophy and techniques espoused by the Moras are discussed in this volume in greater detail in “Italian Differentiated Inpainting Techniques: A Review” by Pamela Betts.

¹²⁵Quotation refers to a conversation between John Brealey and the interviewer, Calvin Tomkins.

¹²⁶Modestini is Adjunct Professor of paintings conservation, Institute of Fine Arts, New York University, and consultant to the Samuel H. Kress Foundation.

¹²⁷The thought-provoking article of Washington, D.C. private conservator Dykstra presents eleven different interpretations of the term artist’s intent that challenge us on clarity when we use the term.

Continue to Materials and Equipment.

Back to Paintings Chapter List