The visually most obvious threat from humans is the production of graffiti over rock art. In Australia, a combination of legislative protection prescribing heavy fines, and public education about the significance of the indigenous cultural heritage have widely eliminated the incidence of graffiti at rock art sites. Graffiti removal practices are vigorously pursued in some countries, such as the United States and Russia, but less enthusiastically in Australia — although excellent projects do provide some notable exceptions (e.g. Thorn 1991). The well-worn cliché that graffiti breed more graffiti may be true, but rock art also ‘breeds’ graffiti, and a great deal of what we describe as rock art began its existence as graffiti. Graffiti at rock art sites are a form of human response to rock art, and while clearly undesirable, the removal of existing graffiti may well do more damage to the scientific potential of the site than the graffiti being removed. While extant graffiti pose no major threat to the scientific potential of the rock art they disfigure, graffiti removal treatment does present such a threat, particularly where it involves the application of chemicals (Bednarik 1992c). Moreover, it has som-times been observed that the laborious removal of huge graffiti prompted the almost immediate production of new ones in the area rendered ‘vacant’ (Bednarik 1992b). The laboriously ‘reinstated’ panel merely offered a fresh slate to graffiti artists.

It is also an important consideration that graffiti have been used most successfully in scientific work, for instance for the calibration of dating methods (Bednarik 1992d, 2001), to determine formation rates of carbonate speleothems, and in social studies of site visitors (Morwood and Kaiser-Glass 1991). Bearing in mind that rock art graffiti have been made for almost as long as rock art itself (since the Upper Palaeolithic; Clottes et al. 1992), they provide a record of human reactions to existing rock art whose study is as important as that of the rock art they comment on. There are numerous graffiti from Pharaonic and Roman times. Other examples are the reactions of various religious movements to the rock art of the religions they sought to replace. They can be found, for instance, in much of the distribution area of Islam and in Spanish South America (Bednarik 1988b, 1992e). Few researchers would suggest that the religious Spanish graffiti at Andean sites ought to be removed, even though some were quite intentionally placed directly over earlier rock art motifs. Similarly, we have yet to hear American demands to remove the numerous graffiti at El Morro, New Mexico, where no less than 26 Spanish parties left their inscriptions between 1605 and 1774. There is no intention of removing the many historical graffiti in Rouffignac, France, dating from the early 19th century. In Australia, with its very short history, graffiti of that period are regarded as historical treasures, although many occur at rock art sites and were probably prompted by the presence of the earlier art (Gale and Jacobs 1987: Fig. 2). Even dated graffiti of the early 20th century are considered valuable historical documents, especially in the most recently colonised northern parts of the continent (Palmer 1991: 116). Thus the value of graffiti is derived from their age, certain social constructs and historical contingencies. It would seem that if the lewd inscription of today’s graffiti artist should escape the attention of the graffiti remover long enough, it will in due course become a sociopolitical statement worthy of scholarly attention.

The definition of graffiti itself needs to be examined before we can attempt graffiti removal, otherwise we cannot know what is to be removed. One could argue that if the rock art was made by indigenes it is not graffiti. Is the recent Hopi inscription ‘Tico and Liz’ at Oraibi, Arizona (Ritter 1991: Fig. 3) to be removed? If it is to remain, how would a conservator decide which ‘non-traditional’ graffiti are by indigenes? Around the world there are traditional rock art sites where traditional societies continue painting today, but now paint slogans, motor cars, football teams or green frogs. The question of graffiti removal is not as clear cut as if might appear to be. Rock art conservation is itself a sociopolitical phenomenon: the cultural remains of extinct or marginalised cultures are appropriated by the dominant culture in the name of preservation. Haskovec (1991) has warned against perceiving rock art conservation as self-evidently, intrinsically ‘good’, pointing out that it can also be a cultural ossification factor. Neither he nor I would ever advocate that graffiti practices be encouraged, but it is the role of the scientist to seek understanding, including of the dynamics of rock art conservation.

Moreover, the first condition of being able to prevent new graffiti is to understand the motivation of the graffiti artist, yet oddly enough the rock art conservation industry has not conducted sustained research into this issue, focusing on symptoms rather than the disease. Loubser (2001: 92) reports that in South Africa, pilgrims attending religious ceremonies in the wilderness write their names sometimes over rock art. He established that such names signify membership with credit unions and burial societies of the communities in question. Masao (1996) has reported similar observations in Tanzania. To illustrate the complexity of the issue, I cite an anecdote. When I observed the arrest of a graffiti artist at the Taj Mahal I requested to have him interviewed before police led him away. He had come from a village far away, at great expense, to declare his love for his sweetheart. The Taj is regarded as a monument to love and he was perfectly aware of the consequences of his action. Indeed, his apprehension and jailing was part of his declaration of commitment. Thus the fervour of the graffiti-phobia also needs to be understood, its reminiscence of the puritan lichen-phobia mentioned elsewhere is obvious. Because such blind beliefs prevent rational analysis and understanding of phenomena they can themselves become obstacles to sound rock art preservation policy.

Nevertheless, in preparing rock art sites for public viewing it may be advisable to remove existing recent graffiti. As in rock art, additive (pigments) and reductive (carvings) versions occur, and need to be treated differently. In the first type, a great variety of graffiti materials have been observed, some of which can be readily identified; others require chemical or microscopic analysis. Wax crayon residues are fairly unstable when exposed to the sun and deteriorate, but if they need to be removed, poultices of cotton wool and sepiolite soaked in organic solvents are effective, as well as paint strippers (Rosenfeld 1985). Most paints can be removed with commercial paint strippers or mixtures of cellosolve acetate and acetone. A detailed description of paint removal at a granite site is provided by Thorn (1991), who used acetone poultices. If working close to rock art paint residues, care must be taken not to damage them. Well-bonded iron oxides and hydroxides are the most resistant, while white pigments are usually the most soluble. Graffiti of charcoal or lead pencil are also common, they are best removed by careful brushing with glass fibre brushes and swabbing with alcohol. However, graffiti can be stabilised by silica and other mineral deposits, which renders their removal much more difficult. It is also to be remembered that most chemicals used in graffiti removal will compromise the research potential of any rock art affected.

The removal of rust stains, which can be found at sites equipped with steel grids or cages, has been described in detail by Finn and Hall (1995). Many graffiti have been made by removal of rock mass, usually by means of steel tools or angular stones. The repair of such petroglyphs or lacunae with grouting materials has been attempted, but results have been unsatisfactory (Bednarik 1992b). Such materials have exfoliated within one or two years, for instance at Mootwingee, Australia. The alternative is to mask the effect of incised, carved or pounded areas, which can be done by one of several methods. Shallow scratchings can be camouflaged by brushing off crushed rock particles and establishing a ‘patina’ matching adjacent surfaces (e.g. simply by rubbing with soil). The graffiti can be removed entirely by abrasives, such as the use of a dry air abrasive unit. This consists of a supply of compressed air, hose and nozzle, and a means of introducing abrasive powder into the air stream (Ford 1995). The abrasive can be aluminium oxide powder (17-27 um), glass beads (50 um) or sodium bicarbonate (100 m). This is particularly effective in masking engraved graffiti and has also been used on otherwise intractable painted graffiti, but its use must be restricted to the minimum as it also removes rock surface. On carbonate rock, hydrochloric acid has been used to mask engraved graffiti, followed by thorough cleaning of the chlorides formed.

Rock areas treated with graffiti removal techniques tend to be of ‘fresh’ appearance and need to be ‘aged’. Where the adjacent surfaces are coated by rock varnish, the method developed by Elvidge and Moore (1980) can be considered. Initially applied at Newspaper Rock and Petrified Forest National Park, Arizona, it consists of the application of a preparation encouraging the precipitation of iron and manganese oxides. (It is rather similar to the chemicals probably used in preparing the Piltdown fakes in 1912.) The desired colour can be achieved by manipulation of concentrations, ratios and application (Elvidge 1979). However, Bock and Bock (1990) have reviewed this work a decade later, and report many problems. For instance the artificial coating has significantly darkened with time, which suggests that in such work natural colour should not be matched, but the repair work should initially be much lighter. Most certainly the method should only be used by experienced conservators, and only on small areas that are in need of repair.

REFERENCESBibliography of Rock Art Conservation