Methods used

The methods used in the scientific study of rock art can be divided into three basic approaches (Taçon & Chippindale 1998): informed methods, formal methods and general analogy. This dissertation makes use of all three.

1.2.1 Informed methods

An ethnographically ‘informed’ approach refers to the use of ethnographic ‘inside’

information concerning the meaning of a rock art tradition. Such information is only rarely available, and its validity is sometimes contested. The informed approach to rock art interpretation has thus been variously received in different parts of the world in different dec-ades. For example, whereas in Australia native ethnography has always been a self-evident component in the study of rock art (Layton 1992), in South Africa an initial use of /Xam ethnography pioneered by Orpen (1874) was later rejected, only to be rediscovered and re-embraced in the past two or three decades (Vin-nicombe 1976; Lewis-Williams 1981; Blundell 2004; Lewis-Williams & Pearce 2004). In the same way, in the 1960s Native American eth-nography was rejected in interpretations of the rock art of the American South-West (Heizer &

Baumhoff 1962), but has been ‘rehabilitated’

in the last two decades (Whitley 1994; 1998;

2000). Furthermore, while most researchers would agree that the most celebrated rock art of all, the Palaeolithic cave art of Spain and France, lies outside the parameters of ethnographic

‘inside information’, the use of ethnographic analogy in its interpretations has experienced similar ups and downs in the past one hundred years. Used by the early interpreters (Reinach 1903; Breuil 1952), ethnography was later rejected by the structuralists (e.g., Leroi-Gourhan 1968) but has recently been applied by scholars such as David Lewis-Williams and Jean Clottes (e.g., Clottes & Lewis-Williams 1998; Lewis-Williams 2002).

In this study, I argue that although the pre-Christian Saami and Finnish religious tradi-tions never explicitly mention rock art, some sources available to us are highly relevant to understanding its meaning. As already pointed out by Radcliffe-Brown (1948), ethnographic texts are raw data that must be ‘read’ and interpreted. This dissertation includes some readings of ethnographic accounts that, in my opinion, make the informed approach to Finnish rock art valid. The approach is not altogether new. Although the early interpreters of Finnish rock art tended to rely on ahistorical theories of hunting magic, the use of local (mainly Saami) ethnography in interpretations of Finnish rock was already pioneered in the 1970s by Ville Luho (1970; 1971) and, to a lesser extent, by Jussi-Pekka Taavitsainen (1978; 1981). My ap-proach thus represents a somewhat similar return to ethnographic sources that has occurred, for example, in North American rock art research.

1.2.2 Formal methods

Formal methods refer to those methods that depend on no ‘inside knowledge’. These rely on the information that is inherent in the images, associated finds and the location of the sites. We may, for example, gain some understanding from a semiotic analysis of the images (Conkey 2001); from their location in landscape (Bradley 1997); or, for instance, from the study of neuropsychological phenomena related to altered states of consciousness (Lewis-Williams

F^igure 2. A descendant of a Saami shaman (noaidi) studies the Stone Age rock painting of Flatruet, Northern Sweden. Photo taken by Ernst Manker in 1938 (from Manker 1965: 17).

& Dowson 1988). The importance of formal methods generally increases the further back in prehistory we go.

Formal methods used in this dissertation include archaeological excavations and soil geochemical analysis conducted at rock art sites (Paper II), a study of rock art location in the light of the cognitive study of religion (Paper III), quantitative studies of motif dis-tribution and location (in this introduction) and some inferences made using the ‘neu-ropsychological model’ (in Paper I). Although the model is not emphasised in most of the papers, it forms an important, independent line of argument that supports several of the conclusions presented in this introductory es-say. A short review of the model is therefore warranted.

In the past two decades, many archae-ologists have sought support for their inter-pretations of rock art in neuropsychological research (for reviews of this research, see, e.g., Lewis-Williams 2001; Whitley 2006: 109-22).

As presented by David Lewis-Williams and Thomas Dowson in their 1988 article titled The Signs of All Times: Entoptic Phenomena in Upper Palaeolithic Art, the ‘neuropsycho-logical model’ is based on the undeniable

fact that all anatomically modern humans share the same nervous system (Lewis-Williams

& Dowson 1988: 202). We experience the world in a fundamentally similar way, and even our dreams and hallucinations are similar. Neuropsychological research on human cognition is thus cross-culturally applicable and forms a kind of ‘bridge’ to prehistory. The visions and sensations experienced by a 21^st century European in trance are fundamentally similar to those experienced by the rock painters 6000 years ago.

In certain altered states of consciousness, induced, for example, by the use of psychoac-tive drugs, hyperventilation, rhythmic drumming or dancing, it is possible to experience visual phenomena – sometimes called ‘entoptics’ (from the Greek, ‘within vision’) – that manifest themselves in simple geometric forms, always the same regardless of the cultural background of the person. This visual imagery is, according to Lewis-Williams and Dowson (1988: 204), experienced differently in the three cumulative stages through which the deepening trance progresses (Fig. 3). However, perhaps more importantly from the point of view of this work, Chippindale et al. (2000: 72-74) have emphasised that an altered state of consciousness also involves specific somatic experiences that are similarly universal. This includes sensations of metamorphosis, weightlessness, breathlessness and out-of-body experiences, which, in a state of deep trance, are typically interpreted as flying through the air, diving under water or dying.

Like the visual hallucinations experienced in a trance, such experiences may have inspired some of the imagery of shamanistic rock art – including the apparent depictions of metamorphosis and subaquatic travel that I find central to the interpretation of Finnish rock art.

The neuropsychological model has attracted much discussion in rock art research. Although embraced by many scholars, the model has not won universal acceptance (for critical views,

Figure 3. The three stages of trance as they might be perceived by a Westerner (according to Lewis-Williams 2001). Reproduced with the permission of Rock Art Reseach Institute, University of the Witwatersrand, Johannesburg, South Africa.

see, e.g., Bahn 2001; Helvenston & Bahn 2003; Kehoe 2003). In spite of the controversy sur-rounding the model, it has arguably brought a certain sense of direction to Stone Age rock art research. In the words of Whitley (2006: 110), the model is best understood as a formal analyti-cal tool, the purpose of which is “to determine whether a corpus of art portrays hallucinatory imagery”. Conversely, the identification of hallucinatory imagery in rock art does not reveal its meaning, only its origin.

1.2.3 General analogy

As the principles of using analogy in archaeology have been discussed in Papers IV and V, it is not necessary to discuss the subject here in any great detail. It is sufficient to note that although general analogies based, for instance, on the ethnography of the South African /Xam may ap-pear rather far-fetched from the point of view of Finnish rock art, they can provide important insights into the question of how rock art as a phenomenon should be approached. As David Whitley (2006: 86) has written,

[…] rock art ethnography gives us a context within which the plausibility of any particular rock art interpretation can be assessed. Put another way, ethnography provides a series of competing hypotheses that can be evaluated for any empiri-cal case. And, while there is no reason to assume that every prehistoric example will necessarily conform to the origin and meaning of the ethnographic cases, a starting place for analysis is the assumption that a prehistoric case should be reasonably close to the known range of variation in the ethnography.

As mentioned above, detailed records on rock art ethnography are extremely rare worldwide, but some information has been preserved in regions like South Africa (e.g., Lewis-Williams

& Pearce 2004), Canada (Rajnovich 1994), Australia (Layton 1992) and California (Whit-ley 2000). Some Siberian sources apparently also describe beliefs associated with rock art (Okladnikov 1972: 41; Devlet 2004), but I have been unable to find any detailed information on this material.