Drawing in the Land
For archaeology, only the world of experience is left. (Layton 2000b:51)
The Sydney Basin has been occupied since the late Pleistocene, yet it was not until the mid-Holocene that rock art was produced in shelters. It is instructive to question and explore philosophically the implications of an earlier lack of engagement with this form of cultural practice. To do so may pave a way toward an understanding of its first appearance and subsequent change in the archaeological record.
While referring to an earlier absence of image-making in Europe and Africa, Soffer and Conkey’s (1997:6) discussion is of relevance in attempting to understand change and transformation in the production of rock art, and the practice of marking the land. They remark:
What was absent, then, was what we have termed performance, the habitual use of the capacity [for image-making] in patterned behaviour. What was absent were the social and cultural contexts within which such material practice would have been efficacious and would have had appropriate meanings. Such a perspective requires us to begin examining the circumstances under which image making becomes habitual, an approach that can help us understand the temporal and regional discontinuities in the global record.
Soffer and Conkey (1997:5) are concerned with the evolutionary context for the emergence of image-marking at c. 40,000 years BP, associated with anatomically modern humans and thought to signal the ‘advent of [a] fully modern way of life – one embedded in and shaped by culture, dependent upon the making of meaning’. In Australia, it can be assumed that a modern way of life has been in place for as long as people have been present. Consequently, the concern is not related to the origins of symbolic and modern behaviour, but the apparent absence, then making and eventual efflorescence of rock art as praxis. While not framed in evolutionary terms, the aim here has been to explore the historical dimension of change and transformation in regard to image-making. This topic has been approached from the perspective of embodiment and experience, i.e. the performance of rock art as practice. The question as to why people commenced to produce imagery in rock shelters is important because this practice represents a significant shift in people’s engagement with their social and physical world.
The discussion in this chapter is concerned with the manner in which the object world was constituted, the situated experience of this, and how, in turn, this material world may have shaped being and social life. Materiality is situated in the context of people and social relations: ‘bodies of artifacts [sic] … implicate particular cosmologies where the role of materiality may have been central to people’s relationships to each other and to the deities’ (Meskell 2005b:52). One of the material features of rock art is its potential for permanence, and that it may continue to mark the land long after its production. A consideration of this requires acknowledgement of the notion that rock art, irrespective of its original significance, is ‘continually socially reconstituted in the process of its observation and, possibly, its repainting or augmentation’ (Rosenfeld 1997:296). That is, imagery itself, and other marks on rocks, are constantly coming into being (cf. Boast 1997:181). But more than this is the question of whether rock art can have a productive power of its own (cf. Bennett 2010:2).
Two major pigment phases of rock art have been recognised in the Sydney Basin (McCarthy 1988b; McDonald 2008a). In its broad structure, this sequence has been identified to be applicable to the Upper Nepean catchment, although some differences have been defined. The two major temporal phases of pigment rock art have been identified to have their own specific and nuanced formal signatures. In addition, this research has revealed that temporal variability also has a complex locational, environmental and spatial signature, which has not been previously identified in the Sydney region.
At some time, and probably in the early to mid-Holocene, a fundamental change occurred in Aboriginal society in the Upper Nepean catchment and surrounds, when people began to mark stone in rock shelters. The archaeological record in the Sydney Basin demonstrates change in a number of behavioural and technological indices, particularly in the late Holocene. This is a general trend observed across much of the continent (cf. David 2002:150). However, the temporal framework of change in rock art in the Sydney Basin, and correlations with the dynamics evident in other archaeological material, is not well understood. McDonald (2008a) formulated her chronological ordering of the Sydney Basin rock art based on an assumption of the contemporaneity of her defined Phase 3 pigment imagery, and the period of greatest intensity of occupation, inferred from archaeological deposits in rock shelters. However, the empirical basis for this correlation is not accepted by all Sydney Basin specialists (cf. Attenbrow 2004:225).
At this time, the chronological dimensions (the timing and duration) of the Upper Nepean relative sequence are not possible to define. The sequence that is proposed is coarsely conceived, given the limitations of the data. It is likely that there are chronological nuances within the sequence that have not yet been revealed.
In western Arnhem Land, where the sequence of change has been well delineated, Taçon (1993:113) argues that rock art was not produced at a ‘steady, constant rate over time’. Instead, the pattern is one of ‘many outbursts … over time rather than a continuous, non-varying output’ (Taçon 1993:113). He identifies intervals when very little rock art was produced, which punctuate periods in which there was a relatively high rate of production. A similar pattern appears to exist in the Upper Nepean catchment, although here, quantifying the timing of the rock art phases and the intervals is not possible. Nevertheless, the analyses have revealed that each phase is different in its structural character. What is notable about graphic rock art production in Phase 2 of the Upper Nepean sequence is how little of it appears to have been produced. This, and the distributional nature of Phase 2 rock art, indicates that it was generated within a highly organised corporate structure. In contrast, the character of Phase 3 graphic rock art has a much more diversified organisation. The relative abundance and variability in structural patterns, suggests that, in Phase 3, rock art was produced in a greater range of contexts, and that corporate and individual goals and motivations were more diverse.
In Chapter 1, it was proposed that one possible motivation for the marking of rock shelters with imagery is that, within these environments, there exists the potential for durability and permanence (cf. Forge 1991; Rosenfeld 2002). If this notion is invoked to explain the motivation to commence producing imagery in rock shelters, there is the implication that the impetus for the practice of rock art may have been to satisfy a requirement relating to intransience. Rock shelters may have been marked because there was a perceived potential that they may facilitate a process of objectifying an ideology in a material and durable manner. Referring again to the quote from Soffer and Conkey (1997) set out earlier, the question is posed as to why the performance of producing imagery specifically in rock shelters in the Upper Nepean began, and whether or not this may be answered by considering that marking the land was undertaken as a social strategy, not only to materialise people’s concerns and beliefs, but also to do so in an enduring and permanent way.
The corollary of this is that previously this need was absent. This does not imply an absence of beliefs, but instead that the performance of inscripting the land in respect of an ideology was not a practice that was a part of experience and occupation of the world. Fundamentally, it suggests that the need to conceptualise the immaterial, via its materialisation in a permanent and durable form, was absent. This may be read to imply that a land-based ideology and social geography may not have been in place at this time, or was expressed in some other way.
The putative widespread distribution and homogeneity of Panaramitee-style rock art has been interpreted to reflect extensive and open communication networks, the function of which is to support the economic and biological viability of small, dispersed populations (cf. Rosenfeld 1997:290). While Phase 1 rock shelters south of the Georges River are distributed in low density, their geographic range (within a c. 50-kilometre radius) is not extensive, and not greater than would be expected of people to traverse during their seasonal rounds. Sefton (2009b:13) argues that these sites relate to the earliest occupation of the region, that their distribution indicates a partial use of the landscape and, specifically, ‘the rugged sandstone of the Woronora Plateau’ was avoided.
However, the watershed divides and ridges in the Upper Nepean catchment are gently undulating, and do not pose impediment to movement or occupation. Boot’s (2002) investigations of the forested and rugged south coast hinterland reveals occupation of these types of landscapes in the Pleistocene and early Holocene, some of which was of high intensity. In the absence of detailed and adequate archaeological investigation of the Woronora Plateau and Southern Highlands, it may be premature to interpret the occupational significance of the low density distribution of Phase 1 rock art. The absence of rock art need not imply the absence of people, and a low density distribution of rock art need not necessarily reflect low population levels (contra Sefton 2009b:13). Rock art, as a technology to convey meaning, may not have been a practice that required marking the land in a spatially dense manner.
Given the nature of the mechanics for the production of engravings, a suitable material is required to penetrate sandstone, and this could only be stone. Such stone must be of a kind that can withstand the stress of impact, and is likely to be a type with similar qualities to hatchet head stone (i.e. toughness) (Dickson 1978:66–67; McBryde & Harrison 1981:183). Stone with these qualities is not found in the sandstone country of the Sydney Basin, and so it can be inferred that the production of engravings was predicated upon the acquisition of suitable stone from elsewhere. This consideration implies a relational context between people and other places. However, the spatial distribution of Phase 1 rock art does not provide any firm clues as to the arena of social geography in which it was relevant.
The current view in regard to the antiquity of Phase 1 engravings in the Sydney Basin is that they were produced in the mid-Holocene, which is more or less coincident with the stabilisation of the post-glacial sea level rise. The lack of ability to precisely define the contemporaneity, or otherwise, of engraving activity and environmental change precludes any attempt to correlate the two and make behavioural inferences. If there was a correspondence, consideration could be given to whether or not engraving the land with imagery developed as a social strategy motivated by the need to mediate social change in light of reduced land mass and shifts in resource type and abundance.
Phase 1 rock art in the Upper Nepean catchment is comprised of small engraved bird and macropod foot/paw motifs. It is essentially figurative and is an ostensive referential system (cf. Layton 2000b:51). Although sometimes paired, the motifs are not arranged in track alignments and, therefore, do not evoke a narrative depiction. In Chapter 8, it was suggested that, based on a consideration of their formal, iconic qualities, this rock art may be interpreted to have been produced to refer to symbolic values (cf. Rosenfeld 2002:75). In respect of these types of motifs in Central Australia, Rosenfeld (2002:74) distinguishes between those in alignment, and others that are formal depictions, to infer different contextual significances of the locales in which they were situated. Rosenfeld argues that alignments of tracks are common in occupation sites (secular), and while not discounting that these illustrative compositions may evoke mythological values, she proposes that their narrative may have been understood at a lesser esoteric level ‘and hence be the more appropriate visual form in secular contexts’ (Rosenfeld 2002:75). When these motifs occur in single or non-narrative compositions, their iconic form is suggestive that they served to encode symbolic values pertaining to the representational being that was depicted (Rosenfeld 2002:74–75).
It is premature to characterise Phase 1 engravings in the Sydney Basin as being homogeneous in form because a broad-scale, regional analysis of the topic has not been undertaken. In any case, if they were, this need not imply social or cultural unity. Similarity in material culture does not necessarily reflect uniformity at the structural level (cf. Rosenfeld 1999; Thomas 2004:100). While some motif diversity is apparent on the Woronora Plateau (e.g. the inclusion of anthropomorphic images in FRC226 in the Woronora River catchment), all Phase 1 shelters in the local area do contain bird and macropod foot images. The widespread distribution of this limited suite of iconic imagery suggests a pan-regional ideational system that articulated a mythology predominantly associated with the symbolic values evoked by the production of these motifs. The addition of anthropomorphic motifs to this basic repertoire, in the single shelter FRC226 in the middle of the Woronora Plateau, is notable. The inclusion of the anthropomorphs in the heart of the plateau suggests that this area may have been a focal node, with a unique significance. However, the widespread distribution of formally comparable imagery suggests the presence of an ideational mythology that was inclusive and relevant over a broad geographic area. This rock art and its relative uniformity does not conform to a pattern that suggests the existence of a totemic geography (cf. Layton 2000a).
However, the production of engravings in the Phase 1 rock art of the Upper Nepean catchment and its local region signals the initiation of a people/land relationship that was mediated, at least in part, by the practice of inscripting imagery on stone. Their production can be interpreted to reflect the assumed beginning, in the Upper Nepean and its surrounds, of the use of a material technology of marking meaning onto the land. In the practice of taking advantage of the environments that rock shelters afford, rock art as praxis became an enduring feature of being and experience in the Upper Nepean.
The graphic form of the Phase 1 engravings in the seven local shelters in the Woronora Plateau (including the two in the Upper Nepean catchment) is found in subsequent phases of rock art. Macropod paw and emu feet motifs are represented in Phase 2 and 3 rock art, in graphic and/or stencilled form. Likewise, circles, bars and formally comparable human motifs occur as graphics in later phases. Macropod and emu bodied motifs become a significant component of the pigment rock art. While a concern to mark the land, with reference to these animal forms, commenced with the production of the engraved bird and macropod foot/paw motifs, their materiality exerted considerable influence on the trajectory of the symbolic worldview in the Upper Nepean.
The rock art in Phase 2 indicates that the materialisation of the ideological system diversified. It is probable that there are two sub-phases of Phase 2 rock art: an earlier phase of motifs that have their formal counterparts in Phase 1; and a later phase, in which animal depictions were produced and rendered in their full bodily form. As well, Phase 2 includes a gestural category or rock art: human hand stencils and prints, and non-graphic pigment smears. While recognising problems relating to the implications of the contemporaneity or otherwise of different categories of materials (cf. Attenbrow 2004; Pinney 2005), it is possible that Phase 2 rock art was produced more or less contemporaneously with the use of ground-edge stone hatchets and, by association, the grinding grooves distributed across the plateau.
The use of pigment to mark the land provided people with the means for a new and different engagement with their physical and social world. It is not known where red ochre was obtained for the production of Phase 2 red motifs in the Upper Nepean catchment. Pigment may have been sourced from outside, although there are no known significant ochre quarries in the region. Red pigment is available on the plateau in the form of iron-rich seams in sandstone, laterite on ridges, ooze in shelters that is re-precipitated material, and decomposed sandstone, which breaks down behind case-hardened crusts in shelters (Sefton, pers. comm., 2003). Huntley (Ford 2006:87–88; Huntley et al. 2011) found that all paints examined in the Upper Nepean were clay based. She argues that the clay was most likely derived from the Wianamatta group and Camden sub-group shales of the Illawarra Coal Measures. This result implies that clay, used in the preparation of paints for the production of Phase 2 rock art, was sourced from outside the Woronora Plateau, although the likelihood of this is questionable. However, given that this research is in its infancy and requires further work (cf. Ford 2006:96), it is premature to speculate about spheres of sociality and relationships across the region that may be inferred. But more importantly, this issue introduces a new dimension relating to the practice of Phase 2 rock art. It implies not only that people began to mark the land with a suite of novel and diversified range of imagery, but also that they engaged with a new range of social and technological processes, relating to the acquisition of materials and preparation of paint.
In the Upper Nepean catchment, Phase 2 rock art is comprised predominantly of stencils. Forge (1991:40) and Rosenfeld (1999) consider stencils as marks that are individual in their referential content and, hence, that they refer to an individual’s relation to place. This view is in keeping with ethnographic understandings of the motivation for stencilling (see Peterson 1972:16; Layton 1992:75). While marking the land at this time may be interpreted to have been most frequently an individualised and participatory expression, it is likely to be both more complex and interesting than this.
In Chapter 8, the analyses revealed that Phase 2 stencilling activity was normally undertaken in shelters used exclusively for that purpose. In addition, stencilling commonly occurs in shelters that are small and possess no living area. This contrasts with the types of shelter typically used for the production of graphic rock art. Where stencilling and graphics co-occur, they were produced in separate intra-site locales. This trend in the Upper Nepean is not unique. Cole et al. (1995:63) report that in the Laura area of north-eastern Australia, red stencils are concentrated in shelters that have few or no other motifs. Rosenfeld (1999) argues that such trends suggest that they reflect different sociocultural contexts. Accordingly, in Phase 2 of the Upper Nepean sequence, a contextual diversification in rock art production is evident.
The Phase 2 stencilled rock art is comprised of human hands. Stencilling as practice, inscribing images of body parts that are inalienable (cf. Thomas 2002:41), may have been motivated by a concern to emphasise a connection between people and place. If stencilling acts to locate people, who are mobile within the landscape, to place, its expression in Phase 2 rock art may have been motivated by a need to negotiate people and land relationships. Rosenfeld (1999) has referred to stencils in Tasmania produced during the terminal Pleistocene when significant adjustments in subsistence strategies were required. She argues that gesturally marking the land at this time is an expression of person–land relationships being emphasised during a period of severe stress. In the Sydney Basin, the dating of red stencils is not secure enough to correlate their appearance with either climatic, environmental or archaeological change, and the duration of red pigment hand stencilling as a practice is not known. If the motivation for stencilling was constituted in a manner comparable with the scenario Rosenfeld suggests, it may have occurred at any number of times during the Holocene, including, for example, environmental or climatic shifts associated with sea level rise, or the onset of the ENSO pattern. Stencilling as practice is worthy of considerably more attention than given in this current research.
The earliest Phase 2 graphic rock art (Phase 2a) is comprised of motifs, which are formally similar to Phase 1 intaglio engravings, and they typically co-occur with Phase 2 stencils. These motifs were produced in a small number of large, commodious shelters, on major thoroughfare landforms. Thus, their environmental signature is comparable with Phase 1 rock art. However, their restricted location in large shelters is a significant change, and implies a productive and social context that related to abundant space. It was noted in Chapter 8 that the temporal relationship between engravings and Phase 2a pigment graphics is actually not understood in the Upper Nepean catchment. They may be either temporally discrete, in accordance with McDonald’s (2008a) chronology, or contemporaneous. While this poses an interpretative constraint, the production of Phase 2a pigment graphics with stencils, and only in commodious shelters, nevertheless represents the emergence of a shift in the productive context in the practice of rock art.
However, these formally similar Phase 2a motifs are distributed across space in low density, and in a pattern that is comparable with Phase 1 rock art. This suggests that the corporate signature of Phase 2a graphic rock art does not exhibit evidence of significant structural change in social geography. The nature of Phase 2a graphic rock art, and its locational patterning, suggests the continued emphasis of an inclusive ideology, which was relevant over a broad geographic area. The introduction to rock art practice (the production of human hand stencils) is a significant development. This phenomenon, in which individual and land connections began to be emphasised, may be informative in regard to the impetus and motivation that led to subsequent shifts expressed in Phase 2b graphic rock art.
A number of notable formal and locational changes are evident in Phase 2b rock art. The production of rock art occurred more intensively, and in a relatively denser distributional pattern, than in Phase 1 or Phase 2a, although the environmental signature of Phase 2b is similar to earlier locational patterns. The shelters used for the production of Phase 2b imagery are normally located on or near to major thoroughfares in accessible environmental locales. With the exception of some eel motifs, Phase 2b rock art is located on highly visible rock art panels in shelters.
Phase 2b rock art suggests a reconfiguration of social geography, which was accompanied by the formulation of a new ideology. This rock art is dominated by fully formed animal imagery, and these motifs are frequently large. While different animal species occur within this suite of imagery, there also exists variability in qualitative traits and, from this, it is inferred that not all Phase 2b animal imagery was equivalent in its purpose and meaning. This variability is evident in three primary categories: manner of depiction, associations between imagery and spatial distribution.
Certain motifs have a narrative quality, which is evoked by their depiction in movement or action, and by their formal arrangement in respect of natural features in rock-shelters. These motifs conjure an imagined space that is three dimensional, animated and fluid (cf. Taçon & Ouzman 2004; Dobrez 2009). Motifs, such as eels, which may possess these qualities are also those that are abundant and distributed widely across geographic space. The eel motif animates the earth, and the nature of its inscription and distribution suggests a deep connection and relatedness between people across the Upper Nepean catchment (cf. Tamisari & Wallace 2006:216, 218). Eel motifs, and others, suggest their function in the depiction of stories and narration relating to the land and water ways.
Motifs that contrast with this pattern tend to be more formal in their schema, limited in number, occur in groups of pairs or more (e.g. gliders and echidnas), and are present in spatially discrete locales within the landscape. It is the production of the latter group of imagery in focal points in the landscape that suggests the existence of a totemic geography (cf. Layton 1992:77). The Phase 2b rock art has a structural pattern that is generally consistent with a model of social organisation based on clan totemism. According to Layton (2000a:181), totemic rock art is spatially patterned. Animal species are preferentially depicted at sites within the territory of the group for whom it is a totemic emblem. Large numbers of species are represented in totemic rock art, but each occurs with more or less the same frequency (Layton 2000a:181). In the Upper Nepean, Phase 2b animal imagery is patterned in accordance with these criteria.
In his review of Aboriginal rock art, Layton (1992:242) argues that the archaeological patterns in the Georges River rock art, identified by Officer (1984), closely correspond to a totemic model of social geography. The patterns remarked upon by Layton (1992:242) include the highly varied range of motifs, with only a few repeated consistently; the emphasis on certain motifs in individual sites; motifs that were repeated several times in sites; and the functional separation between large and small sites. Given the proximity of the Georges River, it is not surprising that a similar pattern exists in the rock art of the Upper Nepean catchment.
The emergence of clan totemism, which may be inferred from the patterning of Phase 2b rock art, indicates that social group and place relationships were expressed by this practice of rock art production. This may be seen to be a development that materialised from the practice of mediating individual and place connections, as expressed via Phase 2 stencilling. During Phase 2b, the infrequent use of rock art, its formal constraint, and restricted geographic and environmental location are suggestive of rock art functioning to mediate a relatively stable social geography. Given uncertainty in regard to the age of Phase 2b rock art and the absence of an archaeological context (excavated data), it is not appropriate to speculate in regard to what might have motivated this change. It is, however, highly unlikely to have been a single cause, and there is the possibility that transformation was an endogenous, gradual and distributed process (cf. Hodder 2011:18) of being and experience.
In contrast to Phase 2, the patterning of rock art and contextual variables indicate that, in Phase 3, the marking of rock with imagery was produced within a greater number and range of social contexts to that which had previously existed. Graphics continued to be produced in commodious shelters situated on thoroughfares, but the range of that imagery increased dramatically. Some older graphic forms, specifically animals, continued to be produced, and older motifs were redrawn (re-marked) in charcoal. However, a new suite of models and graphic schema were added to the repertoire and the graphic art is considerably more heterogeneous. While some forms were repeated across the land, rare and unique forms were produced. Most graphics and other marks were drawn with charcoal (scratching was also used), but others were highly inscripted with colour and complex infill patterns. Stencils were made with white clay rather than red ochre.
The source of the white pigment is not currently known, and could have been obtained either within or outside the Woronora Plateau. The use of charcoal for the production of the majority of graphic imagery within the Phase 3 rock art is notable. Because it is a readily accessible material and likely to have been available almost anywhere in the landscape as a result of bush fires, land management practices or domestic activities, its use was not predicated upon complex procurement and processing strategies. It is for this reason that the use of charcoal for rock art production is usually considered to be convenient and expedient (cf. Frederick 2000:320). Charcoal is often the material used for the production of rock art in contexts of cross-cultural exchange between Aboriginal people and Europeans (Smith & Rosenfeld 1992:11; Frederick 2000:320). Explanations, which are posited to account for its use, include considerations relating to difficulties and limitations in obtaining ochres, or because of its convenient availability, while people were on the move during contact contexts (Rosenfeld, pers. comm., 2003). Given that the production of charcoal drawings in Phase 3 rock art was undertaken on an unprecedented and prolific scale, the use of charcoal at this time may simply have met a requirement for an abundant, readily available material.
In the Upper Nepean catchment, scratching was a recent method for the production of imagery and making of non-graphic marks. The use of this technique may be interpreted as reflecting expeditious marking of the land, and this hints also at a social context of adjustment and flux, which may have been the colonial encounter. At this time, the nature of the implement/s used for scratching is not known, but it is possible that, given the fineness of the etched marks, these were metal (e.g. nails), and had been obtained from the European colonisers.
The most recent rock art shows a greater diversity in graphic form, and this is accompanied by a significant shift in the locations that were chosen to mark rock. Instead of graphic rock art practice being limited to typically large shelters on major thoroughfares, and in readily accessible locales, the land everywhere became marked with graphic imagery and other gestural marks. In addition, a greater diversity of shelter types is chosen for the production of rock art. Many of these do not contain floors or living spaces, are small and otherwise difficult to move around in, and are located within cliffs and places that are difficult or awkward to access. Rock art and other marks are commonly produced in ‘hidden’ locales within rock shelters.
Many of these physical contexts suggest the presence of one, or only a few, individuals at any one time. Furthermore, they indicate that people were occupying and using remote and out-of-the-way places. Both these contexts contrast with the embodied experience of Phase 1 and 2 rock art production. The production of white stencils away from major thoroughfares, and the significant contrast of this pattern with the earlier Phase 2 stencil sites, further emphasises the changed spatial signature of rock marking in the recent past. The functional and social contexts, in which Phase 3 rock art was produced, appears to have diversified, and both corporate and individualised motivations appear to have been expressed via rock marking. The social geography, as meditated by rock art in relation to country, thus shifted. The significant use of previously unused rock shelters for the production of Phase 3 rock art is suggestive that art practice functioned at this time to achieve a much greater level of mediation between people and land.
A model of the use of rock art during colonial encounters has been constructed by Frederick (1997), based on her work at Watarrka National Park in Central Australia. Her thesis is underpinned by the proposition that Aboriginal people encountered and responded to white society not as passive victims, but as active social agents. The use of a graphic system on rock, within the context of contact, can act as an intervening force during a period of stress and insecurity. The structural patterns in Phase 3 rock art in the Upper Nepean catchment are consistent with Frederick’s (1997) model of rock art produced within a contact context. It is particularly the change in the use of landscape, the diversity displayed in both gestural and graphic rock art, and the nature of the material used (charcoal) that conform most closely to this model.
In the Upper Nepean catchment, and more generally the Woronora Plateau, imagery that overtly represents contact items or themes is rare and otherwise ambiguous. However, a number of other variables directly related to some Phase 3 rock art are relevant to an assumption of its recent age and production within the contact period. The main pigments used during the recent period are charcoal and white clay, both of which are relatively unstable and do not bind strongly to rock surfaces. Imagery produced with these materials is unlikely to be old. Additionally, the graphic imagery is produced with dry rather than prepared pigment, and by scratching rock surfaces. This is suggestive of a more expedient mode of rock art production, which can be expected during a period of conflict and alienation. Other more qualitative attributes of Phase 3 rock art indicate a significant shift in people’s concerns. The emphasis placed on the drawing of a new suite of large and imposing anthropomorphic imagery (with ‘eyes’) exemplifies a different and more outward-looking manner of depiction. These motifs were produced in highly visible locales, and are read as entering actual lived space. This quality evokes a purpose that seeks to transcend or go beyond the earth itself, and this contrasts with the manner with which Phase 2 anthropomorphs and animal imagery were inscripted.
However, the production of new graphic forms, and the diversity represented in Phase 3 imagery, tends to mask a fundamental continuity between older and newer rock art. During the production of Phase 3 rock art, many of the choices, made in regard to motif and location, appear to have been significantly influenced by earlier rock art. While continuity in the drawing of some imagery and re-marking of older imagery indicates the referential relevance of these motifs, the significance of certain locales in which earlier imagery was present appears to have been symbolically charged. It is in many of these shelters that recent gestural marking was undertaken, particularly rubbing and pecking, and this suggests a ritual context.
The historical records that attend to the intellectual and cultural response of Aboriginal people in south-east Australia during the early colonial and post-colonial period, while scant, nevertheless reveal the dynamic nature of people’s concerns and ritual practice. While the impacts of European occupation were devastating at all conceivable levels (e.g. population and land loss, and personal and social stress and dislocation), it is highly probable that, as elsewhere across the continent, in the Upper Nepean catchment, people also responded ritually to the European encounter, and sought to mediate their position vis-à-vis the new colonial geography. The topography of this spatial milieu, being one of general diminishment and impoverishment, of loss of life and land, was one in which the maintenance of a totemic geography became to some extent untenable and, furthermore, not now fully relevant to being and experience. It is inconceivable that occupation within the colonial milieu in the Upper Nepean was not accompanied by a reformulation of ideology. In the suite of Phase 3 imagery, a host of new anthropomorphic motifs indicates the ascendance of novel referential beings that are likely to have ‘served to order social space and meaning in the world’ (cf. David 2002:204). Swain (1993:121–122) argues that a twofold cosmological orientation existed in south-east Australia during the early colonial period, and the diversity in the recent rock marks in the Upper Nepean is consistent with this.
Given its geographic situation away from the hub of colonial settlement, the Woronora Plateau potentially provided the opportunity for Aboriginal people to maintain connections with country. In the negotiation of social relations during this time, the marking of place with individual gestures and religious iconography indicates that, despite devastation and loss, Aboriginal people actively sought to maintain social and moral order. In the words of Swain (1993), ‘Aboriginal people trusted, as they still trust, in the powers of regeneration’.
The land of the Upper Nepean catchment, with its abundance of sandstone, provided Aboriginal people with an opportunity to formulate and enact a visual language for the objectification of their social geography. Now, as in the past, this marked landscape resonates with its visual marks and motifs. The diachronic sequence evident in this body of rock art has revealed a rich and complex history of a dialogue between people and the land, which, brokered by inscription in rock shelters, was mutually influencing and transformative. The continuity over time of many motifs suggests that the placement of certain images within shelters by earlier generations successfully served to fulfil a role of transmitting information generationally. It is likely that in the most recent past, when people were mediating cross-cultural exchange within the colonial period, they drew heavily on the existing technology to convey meaning, and earlier imagery to reaffirm their relationship with their land.
The rock art of the Upper Nepean possesses a diversity of themes and behavioural signatures that have both synchronic and temporal significance. It is likely that previous studies, which have assumed functional equivalence in Sydney Basin sheltered rock art, have failed to sufficiently explore the variability, and the contextual and historical narratives, this may inscribe.
In this engagement, the approach has been to explore rock art as embodied practice. This has focused attention and consideration on notions of experience in space, the manner of producing or crafting marks, and where and how they reside in the land. These are the extant elements of the patterns of how humans experienced and lived in the Upper Nepean, and the material discourse they produced and were created by.