This paper discusses rock art in southern Scandinavia as a multisensory format, where both sight and touch would have contributed to the comprehension of the images. From a structural semiotic point of view, we suggest that rock art can be construed as an organised set of features, such as visual and tactile elements, organised into heterogeneous unities with dynamic relations between elements that can change over time with respect to how they are experienced. We argue that in order to understand the rock art medium, it is crucial to take into consideration the multisensory interaction between the perceiver and the qualities of the rock art surface. The reason for including tactile elements in our interpretation of the conception of rock art is the way it was created: by hands interacting with tools and rock surfaces, as well as the spontaneous human tendency to explore the physical world through touch. One can identify key features in the images that would arguably facilitate tactile recognition, as well as be better explained from a multisensorial perspective. This includes the position of the images on horizontal outcrops, the moderate size of the images, the application of an orthographic perspective, the use of ‘tactile markers’ (ie crucial features having a strategic function for understanding images by touch), and the occurrence of incomplete images. A multisensorial perspective on rock art furthermore has semiotic implications. Incomplete images, for example, can be understood as indexical stand-ins for the whole imagined picture, ie as iconic indices. A multisensorial approach to Scandinavian rock art thus allows for new explanations for certain design choices, as well as a new understanding of how the images could relay meaning to a perceiver.

The Scandinavian landscape is littered with postglacial outcrops, many of which carry engraved motifs. Although drawings of ships are most often discussed, this paper focuses on representations of feet. In Northern Europe ship motifs are often associated with cosmologies based on the movement of the sun. This paper investigates whether drawings of feet could have been associated with the same worldview. A number of interpretations are offered of the images at two sites in different parts of Sweden: Järrestad 13:1 and Boglösa 138:1.

A survey and new perspectives of Celtic or La Tène art from Britain is presented. Following Spratling, Celtic art is defined as ‘ornament or pattern and animal/human/supernatural images in metal and other media’. Regional and temporal variations in the type and usage of decorated artefacts are summarised. Three case studies, made of different media, are presented: metal scabbards, bone and antler weaving combs, and pottery. By asking the question ‘why decorate?’ it is argued that the decision to decorate an artefact can affect its life history, marking it out from undecorated artefacts of the same type. Rather than serving a single function, decoration was employed to serve multiple social goals throughout the Iron Age. Different forms of social expression, such as feasting, elaborate display, or weaving, are significant at any one time or place. It is argued that decorated artefacts often played a significant role in these different social arenas. Contrary to many past discussions, decorated artefacts in media other than metal are demonstrated to have been important in negotiations of social power and cosmology.

The rock carvings at Knarrbyn in Dalsland, Sweden, lie high on a rocky ridge and consist almost entirely of multiple circles in various groups. In both location and imagery the panels at Knarrbyn provide a contrast with the varied figurative images on a majority of large rock carving sites of the Bronze Age in southern Scandinavia, which are mostly set low and near the contemporary seas. The paper aims to explore both shape and place of the Knarrbyn discs, with new recordings and landscape assessment. In contrast to the general opinion that Bronze Age carvings of circles had some close relationship to concepts of the sun, the Knarrbyn discs, by their unusual internal shaping and their position here on the high rocks, offer an alternative concept, that these particular carvings had close physical and ideological relationships with Bronze Age burial cairns in this isolated part of Dalsland. These and other late prehistoric monuments are mapped in the paper and include several previously unrecorded sites on the Knarrbyn ridge. The precise landscape location of the carving sites suggests they were part of a sacred passage in the Bronze Age, leading southwards to the major cemetery and rock carving sites at Tisselskog; such a passage may have foreshadowed and influenced the emergence and orientation of the Pilgrimsleden that traversed the same landscape some 2000 years later.

Reappraisal of an early 20th century excavation at the Cronk yn How round barrow near Ramsey in the Isle of Man suggests that a stone pair was demolished during the 3rd millennium BC to make way for a round barrow with a single central burial. It is suggested that one of the stones from the original pair was decorated with a series of motifs before being incorporated into the barrow. Some of the motifs used find parallels amongst later Neolithic incised rock art on the walls of tombs and houses, and on stone plaques. Other motifs, including what appear to be representations of deer, serve to expand the repertoire of known designs and highlight the potential of this kind of this rather understudied category of rock art. Parallels for the zoomorphic motifs can be found in Scandinavia. A review of other rock art within the Isle of Man revealed more than 70 recorded panels at 55 individual sites making this one of the more densely populated rock art landscapes in the west of Britain. Two main styles are represented, the passage-grave style, which includes the Cronk yn How Stone, and the cup-mark dominated style, or Galician Style. The latter accounts for more than 95% of recorded sites which accords well with what is known of the Isle of Man's cultural relationships during the 4th and 3rd millennia BC.

A series of undoubtedly Palaeolithic engraved figures have been recorded for the first time in the United Kingdom in Church Hole Cave, Creswell Crags. The first recorded images were thought initially to be two birds and a large ibex. This paper presents the preliminary results of the first systematic survey of the caves for engravings which identified a total of 16 figures. On closer examination with correct lighting the ibex was seen to be a stag, accompanied by a bison and another herbivore. The presence of the birds was confirmed and at least one other bovid and several other images, including possible vulvas, were identified.

We report on the recording and analysis of 17 petroglyph or pictograph sites, containing a total of 168 glyphic units, carried out as part of a large scale survey of the ancient district (moku) of Kahikinui, on south-eastern Maui Island, Hawaiian Islands. In contrast with previous studies which have tended to view Hawaiian petroglyphs as divorced from their larger archaeological context, we analyse and interpret this corpus in terms of a landscape-level settlement analysis. The Kahikinui petroglyphs exhibit a regular and limited range of motifs, with certain styles of anthropomorphs and zoomorphs (especially dogs) dominating; petroglyphs dating to the early post-European contact period are characterised by Roman lettering reflecting early missionary efforts at literacy. Petroglyphs are strongly associated either with an early historic-period trail, or with rockshelters and cliff faces where there is evidence for freshwater springs or seeps. In the arid environment of Kahikinui, freshwater was a precious resource, and the petroglyphs may have served as territorial markers, or signs of individual ownership or rights of access. Excavations at three rockshelter sites with petroglyphs provide indirect evidence for dating these petroglyphs to the late prehistoric era (16th to 18th centuries AD). Comparisons with petroglyph sites on other islands in the Hawaiian archipelago indicate the existence of distinct regional variations.

A small panel of mobiliary rock art containing two cup-mark motifs discovered in north Pembrokeshire in August 2002 is described and compared with other finds of rock art from Wales. Although the sites with passage-grave style rock art in north Wales are well-known, the more widespread yet less impressive cup-mark dominated panels found mainly around the upland fringes of the country have received relatively little attention. A provisional corpus of 33 rock-art sites comprising more than 37 panels is provided.

Elaborate, religiously sanctioned relationships between people and place are one of the most distinctive features of Aboriginal Australia. In the Australian desert, rock paintings and engravings provide a tangible link to the totemic geography and allow us to examine both changes in the role of individual places and also the development of this system of relationships to land. In this paper we use rock-art to examine the changing history of Puritjarra rock shelter in western central Australia. The production of pigment art and engravings at the shelter appears to have begun by c. 13,000 BP and indicates a growing concern by people with using graphic art to record their relationship with the site. Over the last millennium changes in the surviving frieze of paintings at Puritjarra record fundamental changes in graphic vocabulary, style, and composition of the paintings. These coincide with other evidence for changes in the geographic linkages of the site. As Puritjarra's place in the social geography changed, the motifs appropriate for the site also changed. The history of this rock shelter shows that detailed site histories will be required if we are to disentangle the development of central Australian graphic systems from the temporal and spatial variability inherent in the expression of these systems.

The paper re-examines arguments concerning the early development of early Celtic (La Tène A) art and its relations with the Mediterranean world. Taking as a focus the recent spectacular discoveries below the hillfort of the Glauberg north-east of Frankfurt, the Celtic approaches to representations of the human form are analysed and the complex question of meaning and the relationship of early Celtic art to contemporary belief systems discussed. As a brief coda, the striking changes which took place in many parts of the early Celtic world after c. 400 BC are referred to in the context of the major population movements of the time, notably those which brought settlers from north of the Alps to Italy.

Rock art in Morocco is widespread and varied, but little known in English-speaking circles. The present paper aims to present a broad outline of this art — almost entirely represented by engravings — as it is known today. A brief survey of past research is given. The distribution of sites is indicated, showing the country to be roughly divided into two areas: the High Atlas mountains and the sub-Saharan regions to the south. Four major groups of engravings are identified, according to theme, technique, and, to a certain extent, style. In the absence of radiocarbon or other datings, the only clues to the chronology of the engravings come from datable material objects or events. None of this art is thought to be older than Neolithic and the most recent engravings certainly date from early historical times. Problems of conservation are discussed, along with measures being taken both to protect the sites and to extend research on this informative aspect of Morocco's past.

Anthropomorphic hiked short swords were a rare and specialised weapon whose size, style, and human symbolism altered little over some 300 years. A small proportion of these short swords are engraved or stamped with designs inlaid with gold or other metals which are suggested to represent different stages of the lunar cycle. The measurement of time in nights is evidenced by later Gallo-Latin calendars written in Gaulish and this is endorsed by literary evidence for Late Iron Age France which points to the association of this knowledge with a specialist religious class. It is suggested that the swords were used in practices or ceremonies associated with making and keeping the time by counting nights, and determining what was a propitious day.

The interpretation of prehistoric rock art has posed some intractable problems, but recent studies have sought to integrate it within a more broadly based landscape archaeology. They emphasise the special character of this material, not only as a system of distinctive motifs, but also as a source of information employed by people engaged in a mobile pattern of settlement. This paper investigates the character of the rock art of south-west Scotland, comparing the positions of the petroglyphs with two series of control samples in the surrounding landscape. The carvings seem to have been situated at viewpoints. They may have been directed towards the coastline and the Galloway hills and commanded a significantly wider field of vision than locations in the surrounding area. There is some evidence that differences in the size and complexity of the motifs are related to their placing in the local topography, with the simpler carvings around the edges of lowland ‘territories’ near to the shoreline, and the more complex compositions in upland areas, especially around shallow basins and waterholes. The changing character of the designs may reflect differences in the composition of the audience who viewed them.

Discovery of the cave of Altamira in 1879 introduced a new aspect to ideas on human evolution, requiring a fresh approach to previous notions of life and customs among prehistoric groups. In the final years of the 19th century the prevailing view of prehistoric life was still coloured by the prejudice and concepts which had marked early discussion on the origin of man. From what was known of their tools and art forms, early men were presented as living in simple communities which were considered equivalent to the ‘primitive peoples’ being described, often with a Euro-centric superiority, by contemporary ethnographers. The already classical volumes of the period, such as the Reliquiae Aquitanicae recording Lartet and Christy's discoveries, were full of interesting parallels to Eskimos and Bushmen. Up to this time, known Palaeolithic art was confined to engraved bone, some of it of high technical competence as at Chaffaud and La Madeleine. Consequently when the discovery of Altamira was presented at the Lisbon Congress it was received with some degree of caution.

The ‘art’ of the early Upper Palaeolithic refers to the images and symbols of the hunters of horse, bison and mammoth during the last European Ice Age that began more than 32,000 years ago. It is therefore interesting that it is not the image of the animal that develops in complexity and variability across Ice Age Europe for the next 20,000 or so years, but the image of the female. The image and the concept of the Ice Age female has perhaps been the subject of more intense emotional debate in the last century than the image of the animal (Ucko 1962; 1969; Clottes and Cerou 1970; LeroiGourhan 1965; Ucko and Rosenfeld 1972; Stoliar 1977–78). The debate continues, both with new finds and new ideas (Rosenfeld 1977; Delporte 1979; Rice 1981; Gamble 1982; Cerda 1983; Guthrie 1984; Praslov 1985; Sonneville-Bordes 1986; Lorblanchet and Welte 1987; Gvozdover 1989; Duhard 1989).

In 1902 Emile Cartailhac published his Mea Culpa d'un Sceptique. His acceptance of the high antiquity of prehistoric art in western Europe followed Capitan and Breuil's convincing discoveries in Font de Gaume and Les Combarelles and reflected a widespread change of opinion. Despite previous scepticism, researchers were beginning to allow that the parietal as well as the mobile art did indeed date back to the Upper Palaeolithic. But this swing in scientific opinion opened up an even more baffling problem: why did Upper Palaeolithic people make these pictures? In the year following Cartailhac's turn-about Salomon Reinach tried to answer this question by developing an analogical argument based on ethnographic parallels. He could see no other way of approaching the problem: ‘Our only hope of finding out why the troglodytes painted and sculpted lies in asking the same question of present-day primitives with whom the ethnography reveals connections’ (Reinach 1903, 259; my translation, his emphasis).

To describe, let alone explain, the paintings, engravings and sculpture of the Upper Palaeolithic as ‘adaptations’ may sound absurd. These are products of the human mind — a world of symbols and dreams, myths and fantasies. So to suggest that this art can be understood in an ecological framework may strike one as facile. Upper Palaeolithic art is one of the great cultural achievements of human kind. It testifies not only to an immense technical skill but to the human capacity for expressing emotion through the use of line, form and colour. Although we cannot know the meaning of the art, through it we can begin to share the sensitivities of the Palaeolithic hunters to their natural world and the animals of the chase. Like all great art, it transcends the boundaries of time and space to say something fundamental about the human condition — though that ‘something’ is forever elusive. The paintings, engravings and sculpture of the Upper Palaeolithic are indeed the epitome of human creativity. So when faced with either the great bulls of Lascaux or just a scratch upon a broken pebble, surely it must be trivial to invoke notions of adaptation and ecology. After all, is not adaptation solely about the more basic features of human life — the selfish struggle to survive and reproduce — hardly the basis for the fine arts.

Art is unique amongst the categories of evidence which survive from the Palaeolithic period in that it was originally structured to communicate. It is true that style is an aspect of tool-making, and the particular forms adopted in the manufacture of tools may be intended as practical innovations or as recognizable types that define distinct social groups (Wiessner 1983) but art is only communication; it has no practical use. Its purpose is to convey a message, simple perhaps, as with ornamentation, or more complex as is suggested by the repertoire of painting and engraving in caves. In a culture without a literate tradition music, mime, recital and art are the vehicles of information. Only one of these remains from the Palaeolithic and, in isolation, the visual evidence is to a certain degree illuminating and to a certain degree frustrating in that certain deductions can be made from its intrinsic features, but not an explanation of its meaning.

My aim in this paper is to put the question, ‘Where was the centre of the Upper Palaeolithic world?’. I promise no answers and I will confine the discussion to Europe.

The purpose behind raising such a seemingly irrelevant problem is that its investigation leads us directly into issues of palaeolithic society, the interpretation of art in such contexts and the often opposed views of how art contributed to adaptation and change.

At the outset let me declare that I am not dealing with the origins of art and its association, or not, with language. Art for me is, however, a system of communication and includes a wide range of mediums and messages. As an act of social communication it is defined by style which, as Wiessner (1984, 191) argues, has its behavioural basis in a fundamental human cognitive process; personal and social identification through comparison. Consequently style is not just a means of transmitting information about identity but is an active tool used in building social strategies (ibid. 194). It has a role in negotiation which, as I shall argue below, is the basis for defining palaeolithic society.

The Magdalenian sites of Andernach and Gönnersdorf are located in the central Rhineland at the northwestern end of the Neuwied Basin (fig. 1). The first investigation of the Andernach-Martinsberg site was carried out in 1883 by H. Schaaffhausen (1888); between 1979 and 1983 a new campaign of excavations took place at the same site (Veil 1982a). The site of Gönnersdorf was investigated between 1968 and 1976 (Bosinski 1979). The two sites are located directly facing each other above the Rhine, which at the time of the occupation was much wider than at the present day; Andernach is sited on a Middle Pleistocene lava flow, while Gönnersdorf is on the Middle Terrace of the Rhine. Both sites are assigned by pollen analysis to the end of the Bølling interstadial; the great similarities in their archaeological material, down to individual details, suggest that the two sites were contemporary.