Printer Friendly

Kimberley points and colonial preference: new insights into the chronology of pressure flaked point forms from the southeast Kimberley, Western Australia.

KEYWORDS: Pressure flaking, Kimberley points, chronology, collections, technological change


Based on recent archaeological research in the southeast Kimberley this paper argues that while bifacially pressure flaked points are clearly present in the archaeological record from c. 14001000calBP, Kimberley point manufacture reached its zenith only after AD1885, when there was both an intensification in the numbers of points produced, and increasing formalisation of point design. Changes in the numbers and formalisation of the shape of points produced after AD1885 are linked both to post-contact social changes and the dialectic role played by non-Aboriginal collectors in selecting for particular point forms. It is suggested that 'Kimberley points' as defined by Akerman and Bindon (1995) should be seen as an extreme formal variation on the range of pressure flaked point shapes that were manufactured over the last millennia in the Kimberley, developed to meet the particular preferences of colonial collectors. This study of formal changes in artefacts as a result of colonial contact in Australia is set within a framework of other research in colonial New Britain and the Admiralty Islands on the influence of collectors on the formal qualities of Indigenous objects.


The bifacially pressure flaked 'Kimberley point' is probably the most easily recognised formal stone artefact type from Aboriginal Australia, although ironically, this 'stone' artefact is most widely known as a flaked bottle glass artefact, as it was in this form that it was widely traded and collected in colonial Australia and elsewhere. Bifacially pressure flaked points were collected in large numbers by antiquarians and museums throughout the world during the late nineteenth and first half of the twentieth centuries, and Aboriginal people took advantage of the availability of bottles and sheets of plate glass to manufacture large and intricately worked pressure flaked pieces. During this time points became important material symbols associated with masculine social identity and status for the Aboriginal people who made and received them through trade, as well as the antiquarian collectors who played

a dialectic role in developing the formal aesthetic qualities of the Kimberley point by creating a market for particular point shapes. This paper considers implications of recent work at excavated and open archaeological sites in the southeast Kimberley region of Western Australia for understanding the chronology of the appearance of pressure flaked points in the Kimberley, and distinguishes their first appearance during the late Holocene from the more recent phase after c.AD1885'--in which the manufacture of Kimberley points intensified and point form became increasingly formalised. The paper concludes with a review of other studies of the effects of colonial trade on the formal qualities of Indigenous objects.

Kimberley points defined

The use of the term 'Kimberley point' to describe a number of different point forms, ranging from non-invasive marginally retouched uniface points which were manufactured in the Kimberley region (Figure 1), to finely worked symmetrical invasively pressure flaked points with serrated margins, manufactured in areas as far away from the Kimberley as Rottnest Island, has suggested the need for clearer definitions to distinguish between what appear to be temporally and spatially distinct artefact forms. Akerman and Bindon (1995) have addressed this problem by defining 'Kimberley points' with reference to their marginal treatment (1995; Akerman el al. 2002). They argue that the term 'Kimberley point' should be reserved only for invasively pressure flaked points with denticulate or serrate margins. On the basis of differences in the marginal treatment of points occurring in the broader Kimberley region (the Kimberley and west Arnhem Land), they distinguish four distinct classes of pressure flaked points:

* Wanji bifaces, which were manufactured in west Arnhem Land by percussion flaking stone materials with clearly defined cleavage planes, and in the recent past were made by percussively or pressure flaking glass window panes. Wanji bifaces are not invasively flaked and were only pressure flaked in the contact period, and can be clearly distinguished from glass Kimberley points of the contact period in that the pressure flaking is short and occurs only at the margins to provide plan symmetry to the artefact.

* Northern Territory triangular points, which occur only in the precontact period on stone, are bifacial and invasively flaked using precise percussive flaking, with minimal pressure flaking to form straight (not serrated) margins, and are shaped into the form of an isosceles triangle.

* Kimberley points, which are invasively pressure flaked points manufactured on blanks formed by percussive flaking, in which each successive series of pressure flakes 'is taken along one margin and refines half of each face alternately' (1995: 92). Kimberley points have serrate or denticulate margins, with small projections or 'teeth' separated by notches of similar (serrate) or smaller (denticulate) width.

* Kimberley dentate points, which are invasively pressure flaked biface points, where the 'teeth' formed by pressure flaking on the margins are separated by notches wider than their teeth (Akerman et al. 2002). The marginal working on dentate Kimberley points is considered to be qualitatively different to the (interchangeable) serrate or denticulate margins of Kimberley points as they are defined above.

What distinguishes this set of points from other northern Australian point forms (Davidson 1934; Smith and Cundy 1985; Attenbrow et al. 1995; Allen 1997) is the use of pressure flaking (ie the use of an indentor to remove flakes from an artefact) in their manufacture. However, it must be noted that not all pressure flaked points from the Kimberley display the characteristics of 'Kimberley points' as they are strictly defined above. I would argue that the emphasis of these categories on extreme stylistic variants masks the significant variation in the form of pressure flaked points in the Kimberley and northern Australia more generally. In this paper, the use of the term 'pressure flaked point' should be understood to refer to any point which has been manufactured using pressure flaking techniques, as distinct from the term 'Kimberley point', which refers only to those points as defined above. A further confusion arises from the occasional use of the term 'Kimberley point' to describe percussively flaked points manufactured in the (geographical) Kimberley region. In this paper, such points will be referred to simply as 'percussively flaked points', once again to distinguish them from those defined as above. It is important to note that a range of morphological variants on the above stylistic extremes are often encountered as part of a single assemblage, a line of reasoning to which I will return later in this paper.

Chronology of points

Dating the appearance of the Kimberley point has generally proven somewhat problematic. Most authors have accepted that, to some extent, the finely crafted Kimberley points that made their way into the hands of adjacent Aboriginal groups, antiquarians and archaeologists during the late nineteenth and twentieth centuries were the product of recent (post-contact) developments, due to the relative absence in archaeological excavations of finds of finely crafted bifacially worked pressure flaked points with serrate or denticulate margins (e.g. Balme 2000; McCarthy 1976: 44; O'Connor 1995, 1996, 1999: 71; Vietch 1996, 1999; Wallis and O'Connor 1998; White and O'Connell 1982: 112). However, Akerman and Bindon (1995) highlight the illustration, by British explorer Philip Parker King, of a stone Kimberley point collected in AD1810 from the north western Kimberley coast as evidence for the existence of finely worked bifacial pressure flaked points, complete with denticulate margins, prior to the sustained European settlement of the Kimberley (and hence the appearance of the trade market which was said to have stimulated their production). Nonetheless, the general absence of these large pressure flaked biface points with serrated margins from archaeological deposits and surface sites dating to the last few thousand years has generally been seen as evidence for the late development of Kimberley points as they are defined above. Akerman and Bindon's (1995) assertion that dentate Kimberley points should be considered pre-contact in age, while Kimberley points are most likely both pre and post-contact in age, has not been able to be tested stratigraphically for the same reasons. Walsh and Morwood (1999: 53ff.) note the occurrence of images of spears with what appear to be representations of dentate points associated with animals of the Wandjina rock art style, which they suggest must date to after 3800BP on the basis of the oldest current direct AMS determination for Wandjina style motifs (not a direct date on a point motif). Spear points associated with Wandjina rock art possibly include post-contact metal points, such as that illustrated in Walsh and Morwood (1999: 55). The continuation of painting Wandjina style rock art into the historic period means their association with the style, in the absence of direct dates on point motifs themselves, provides at best a terminus ante quem for the stone artefact type.

Most authors would agree with a general chronological schema for the Kimberley which saw the appearance of percussively flaked uniface and biface points in the mid Holocene, and the appearance of pressure flaked points, including Kimberley points, either in the early contact period (ie c. 1885AD) (eg. White and O'Connell 1982:112) or some time immediately preceding it (Akerman and Bindon 1995; O'Connor 1999:71). Recently, Akerman et al. (2002) published an extensive review of published and unpublished accounts of Kimberley point manufacture, and suggested the existence of several different functional categories or types amongst museum, private and archaeological collections of points, grouped on the basis of the results of use-wear analyses. Their paper raises important questions regarding accepted chronologies for Kimberley points. They note that Aboriginal people themselves often distinguish between 'early' point types, as both unifacial and bifacial points which lack invasive retouch, and 'late' point forms, which are often invasively pressure flaked, with flaking extending beyond the midline of the point. However, they suggest that their data does not support this assertion, as symmetrical points with serrate or denticulate margins and invasive pressure flaking are found in both (pre-contact) archaeological contexts as well as in 'early' and 'late' museum collections. On this basis, they contest the notion that finely worked Kimberley points are a product of post-European forces (Akerman et al. 2002: 29).

Archaeological data obtained from excavations in the Keep River region is cited to support this argument. The earliest points from excavations at Jinmium are unifacial points that occur in C1/II spit 20, unit 20, dating to between 1020 and 1760BP based on radiocarbon determinations cited in Roberts et al. (1998). The earliest biface point is from C 1/II spit 9, with optical age estimates from 540-1200 years. Akerman et al. (2002: 29) note that this early biface point is similar in age to the earliest points with serrated margins from C1/I spit 6. Recent analysis of several other excavated stone artefact assemblages from the Keep River region, with the exception of one possible outlier artefact, indicates a post- 1000BP date for the introduction of pressure flaking in this area (Boer-Mah 2002: 39, 55). These dates find general agreement with O'Connor (1999: 71, 76) who estimates the appearance of pressure flaking technology in the west Kimberley by around 1000BP.

As I have implied earlier in this paper, one of the problems with previous work on the origins of Kimberley points is that it fails to discriminate adequately between the appearance of pressure flaking as a technology for the manufacture of points, and the appearance of Kimberley points as they are defined in the strict sense above. Certainly, the illustrated points from Jinmium in Akerman et al. (2002: 16), although showing evidence of pressure flaking, appear qualitatively different in terms of their symmetry and invasiveness of working to Akerman and Bindon's (1995) definition. It is best to consider these precontact Keep River points as 'pressure flaked points' in the terminology adopted in this paper, rather than 'Kimberley points' under the Akerman and Bindon (1995) definition, as most display either invasive working or serrated margins, but generally not both. While the data cited from the Keep River region supports a c.1000BP date for the onset of pressure flaking in the Kimberley, I would suggest that it does not assist with providing a date for the development of an industry of formal 'Kimberley point' manufacture.

Wilinyjibari rockshelter

Archaeological excavations undertaken in 1998 at Wilinyjibari, located approximately 50 km to the south of the town of Halls Creek (Harrison and Frink 2000), are relevant to discussion regarding the origins of pressure flaked points in the Kimberley. Within a small rock overhang a single 1 x 1 m pit, excavated in 50x50m quadrats, uncovered a rich assemblage of over 7000 flaked stone artefacts dating back to at least 4500 years calBP. Analysis of almost 3000 artefacts from quadrat D has focussed both on retouched stone artefacts as well as associated debitage captured in the 5mm and 3mm sieved fractions of the excavated deposit.

One of the problems faced by archaeologists interested in dating the appearance of pressure flaked points is the low numbers in which they often occur in excavated rockshelter deposits. This analysis sought to overcome this problem by identifying 'pressure flakes' as a distinct artefact type, in addition to pressure flaked points and point fragments themselves. Pressure flaking produces a range of different flake shapes as debris, but a distinctive type are the tiny, flat, elongate flakes with long flake scar arrises indicating the prior removal of other pressure flakes. During analysis of the excavated material these were recorded as separate 'types' in an attempt to detect pressure flaked point manufacture, even in the absence of the points themselves.

I undertook excavations at Wilinyjibari with the assistance and under the instruction of the 'Lamboo mob', a group of Jaru people who had lived and worked on Old Lamboo station since the 1930s, who had extensive traditional and historical knowledge of the places under investigation (Harrison 2000c). The process of Kimberley point manufacture was well known to the Lamboo mob (see also Harrison 2002a, 2002b). Kimberley points are known as jimbila or more rarely yalga. After heat treating the prepared core and selecting a flake suitable for point manufacture (balgi, described as a 'proper slice'), dibid bayarnarni lindij barnu ([you] break it up to pressure flake). Pressure flaking was carried out with a long hardwood pressure flaker called a mungarda. The waste flakes that are left behind from lindij (pressure flaking), are called bulba or a more generic term for flakes, inga. The general term jimbila was used to describe a wide variety of points including unifacial percussion flaked points as well as bifacial pressure flaked Kimberley points. Pressure flaked points with prominent dentate teeth were called lirra (teeth) jimbila, but were considered to be a stylistic variation, such that they were functionally equivalent to other point forms (both with and without pressure flaked margins). As discussed in Akerman et al. (2002), some of the Lamboo mob made comment on the differences between ngamungamu (old time) jimbila and jalanijarra (this time) jimbila based on differences in the extent to which points were invasively flaked and the apparent symmetry and form of the point.

The earliest points at Wilinyjibari occur in excavation unit 18, associated with an OCR date of 2289-68calBP (see Harrison and Frink 2000) immediately above which is a radiocarbon date of 2100_140BP (Wk 6645). The points are one unifacial and one bifacial point, both made on short blade flakes, and neither of which show evidence of pressure flaking (Table 1). Pressure flaked points and pressure flakes themselves only occurred in the upper and middle layers of the excavation at Wilinyjibari (Table 2). The earliest pressure flakes and pressure flaked point fragments (two non-invasively flaked quartz point tips which appear to have broken during manufacture or use, one bifacial the other with evidence of unifacial pressure flaking only) were recovered from excavation unit 14, at a depth of 35-40cm below surface level (Figures 2 and 3). Two OCR determinations were obtained from this level, of 1398 [= or -] 41calBP and 1088 [+ or - ]32calBP respectively (Harrison and Frink 2000). The appearance of pressure flakes and pressure flaked points in the deposit occurs at a time of technological change, indicated by the steep jump in the number of small flaked stone artefacts less than 5mm in maximum length (Figure 4). Such a jump in small debitage would be consistent with the first appearance of pressure flaking, which produces many small flakes and pieces of debris (Tindale 1985). The graph of numbers of pressure flakes through the deposit shows a truncated bell curved distribution, with peaks in numbers of pressure flakes in excavation units 10-13. There appears to be a drop in numbers of pressure flakes produced in the upper layers, although this is likely associated with an ethnographically observed change in site use and function (Harrison 2002c). On the basis of this data, the appearance of pressure flaking associated with point manufacture at Wilinyjibari is considered to date to some time between 1400calBP (AD550) and 1000calBP (AD950).

Pressure flaked points and 'Kimberley points' in the southeast Kimberley

Although the sample of excavated points is relatively small and fragmentary, it is worth making the observation that none of the point fragments from Wilinyjibari have serrated margins (with the possible exception of the fragment illustrated as Figure 2d), although points with serrated and denticulate margins are common on known post-contact archaeological sites in the study area, such as the pastoral workers' encampment associated with Old Lamboo pastoral station (Harrison 2000). It is also worth noting that in an analysis of stone artefacts from eight surface sites undertaken as part of my research, only one point recorded in a sample quadrat at a pre-contact stone artefact scatter in the study area had denticulate margins, while on the sites associated with the pastoral workers' encampment at Old Lamboo (occupied after c.AD1903), between 11% [Lamboo surface collection 2, total number of points=30] and 16% [Lamboo surface collection 3, total number of points=41] of the points had denticulate margins. This data would suggest that while the manufacture of points with serrate and denticulate margins probably sometimes occured in the period prior to the sustained European settlement of the Kimberley (c.AD1885), that there was an increased preference for the manufacture of points with serrate and denticulate margins in the post-contact period. Although dentate points were rare on both pre and post-contact sites in the study area, dentate points certainly were present in the surface scatter of artefacts associated with the pastoral workers' encampment at Old Lamboo Station, suggesting that in this area at least, there was some manufacture of dentate points in the historic period.

Other work at known post-contact open sites in the region suggests that Kimberley points make up a larger proportion of the retouched artefact assemblage at post-contact sites when compared to pre-contact archaeological sites in the study area (Harrison 2002c). (2) Table 3 summarises the results of counts of artefacts from eight 2x2m and 5x5m surface collections at surface artefact scatters in the study area. Sites are divided into pre-contact sites, post-contact pastoral station sites and post-contact non-pastoral station sites. Note formal 'Kimberley points' represent a much larger proportion of the retouched assemblage at post-contact pastoral station sites than at either post-contact non-pastoral station sites or pre-contact sites. This data accords well with Head and Fullagar (1997), who note an increase in the production of points through the pastoral station period of occupation of Marralam outstation (Figure 1) by Aboriginal people.

It is important to note that only particular kinds of post-contact archaeological sites demonstrate this pattern of increased point manufacture, and this relates to the different mechanisms which I have suggested are responsible for the post-contact intensification of pressure flaked point manufacture, and the development of the formal aesthetic of Kimberley points. Elsewhere I have outlined the role that points played in the evocation of hybrid masculinities, and the relationship between the increase in point manufacture on pastoral properties and changing pastoral labour regimes (Harrison 2002a). On Kimberley pastoral properties where male Aboriginal pastoral labourers found they were unable to undertake particular tasks associated with masculine status and authority, points were invested with discursive power as similes for the masculine hunting pursuit by both their Aboriginal creators, as well as the non-Indigenous collectors that stimulated their increasing formalisation and rate of manufacture as trade items in the post-contact period. And yet on contemporary non-pastoral station encampments, the data cited above suggests that the manufacture of pressure flaked points is clearly less important than during the pre-contact period. The two post-contact non-pastoral station sites discussed here are holiday camps that were used predominantly by non-pastoral station workers in the period after AD1885. As such, they were separated from the context in which points had most discursive power in relation to the competition between Aboriginal men, and where pastoral station managers could act as middle traders in the acquisition of points for non-Aboriginal collectors. These differences in the numbers of points and Kimberley points produced in the two different kinds of post-contact Aboriginal archaeological sites supports the argument about the mechanism for changes in point form proposed in Harrison (2002a). The data cited in Table 3 clearly suggests an intensification of Kimberley point manufacture on those post-contact archaeological sites in the study area associated with pastoral settlements.


Given that, at least in the southeast Kimberley, there do appear to be fewer Kimberley points as defined by Akerman and Bindon (1995) in the pre-contact period when compared with particular kinds of post-contact archaeological sites, it is worth returning to the data cited by Akerman et al. (2002), discussed earlier in this paper. In the broad sense, data in the two tables published as Tables 2-3 in Akerman et al. (2002: 30-35) could be read as representing two chronological periods: post-contact (AD1895-AD1949[n=84]) and contact/pre-contact (from as early as AD160 to ?AD1950 [n=41]). These observations are summarised in my Table 4. By discounting the excavated points from station period BOAB excavations In=7] in the contact/pre-contact sample, the remaining points in Akerman et al. (2002) Table 3 are generally thought to be pre-AD 1895 in age (R. Fullagar pers comm. 2002). Out of 84 complete post-AD 1895 points from museum collections (Akerman et al. 2002 Table 2), 68 [81%] are bifacial invasive serrate points (ie 'Kimberley points'). Out of 32 (predominantly pre-AD1895) excavated points (Akerman et al. 2002 Table 3), 8 [25%] are bifacial invasive serrate points (ie 'Kimberley points'). Although these samples could never be considered statistically similar, it is worth noting the preference in the post-AD1895 museum collections for bifacial, serrated invasively pressure flaked points. The excavated samples representa much broader range of point types, grading from unifacial points with marginal working, to invasively pressure flaked bifacial points. The preference for symmetrical points with denticulate and serrate margins on the part of the non-Aboriginal collectors who formed a new trade market for these objects in the period after AD1885 is critical to understanding the development of formal Kimberley point design that occurred during the post-contact period.

There can be no doubt that non-Aboriginal collectors played an important role in the development of the formal aesthetic qualities of post-contact Kimberley points. Henry Balfour, a keen collector and ethnographer of point manufacture who played a key role in the distribution of Kimberley points to both Australian and overseas museums, noted (1951: 274) that:
   I sent some biface points made by the Worora tribe to the
   British museum and the reply came back that they were
   the most beautiful spear points made by any natives in the

The craze for collecting Kimberley points amongst European, American and settler Australian collectors and museums, and the ensuing spread of these objects outwards from their geographic area of manufacture was both widespread and rapid. Only some 18 years after the sustained settlement of the Kimberley region, Balfour (1903: 65) was able to write:
   The spear heads made with such skill by natives of N.W.
   Australia from broken glass bottles, telegraph insulators,
   and the like, have long been familiar objects in museums
   and private collections, and need no description here. As
   is well known, many of these spear heads are really
   beautiful objects ...

In the work of antiquarians and ethnographers who described Kimberley point manufacture, particular attention is paid to the description of the serrated or denticulate margins of Kimberley points, and their method of serration or denticulation. The discussion of the use of bone and wire indenters or pressure flakers forms a central part of such descriptions. For example, in his book Over the Range, Idriess devotes several pages to describing the process of pressure flaking a glass Kimberley point.
   Each gentle scratch, each careful sawing left a tiny
   needle-point along the razor edges of the glass. All
   terminated in the delicate point of the finished spearhead,
   the serrations producing a truly terrible cutting
   instrument (1938: 61-62).

A similar obsession with describing the margins of Kimberley points in detail can be found in other ethnographic, antiquarian and archaeological descriptions of Kimberley points (see Balfour 1903; Spencer 1922: case 3; Basedow 1925: 367-370; Spencer 1928:510-11; Love 1936; Elkin 1948; Mitchell 1949: 64; Petrie 1954; Tindale 1985). The result of the establishment of a trade market for intricately worked, invasively pressure flaked points with denticulate and serrate margins was the increased formalisation of point design that occurred in the post-contact period.

Archaeological data from the southeast Kimberley cited here would suggest both a proportionately higher number of points in general on post-contact pastoral station sites, along with a greater preference for the manufacture of invasively pressure flaked points with serrate or denticulate margins (ie 'Kimberley points'). Finer detail on the chronological dimensions of this preference comes from a comparison of glass artefacts produced on pre-AD1930 and post-AD1930 glass at Lamboo, dated on the basis of glass colour and manufacturing technique. Of 243 pieces of flaked glass from the two Lamboo collections associated with the pastoral workers' encampment, the post-AD1930 assemblage contains 6.2% points and 2.8% point blanks, while the pre-AD1930 assemblage contains 13.4% points and 4.5% point blanks (Harrison 2002c). This suggests that the 'hey-day' for point manufacture at Lamboo occurred between c.AD1903 (when the station was established) and c.ADI930, during the period in which the demand for points by collectors was highest, and when Jaru people were coming to terms with the implications of European invasion and their role in the post-contact pastoral economy (Harrison 2002a, 2002c).

Other archaeological studies of changes in the formal quality of Indigenous objects in colonial contexts

Recent work by Gosden and Knowles (2001) in colonial New Britain, and Torrence (1993, 2000) in the Admiralty Islands, Papua New Guinea, provide comparative material in the role of collectors in changing the formal qualities of Indigenous objects within the context of colonial exchange. In a detailed quantitative study of a range of different trade objects collected by five collectors in New Britain over the period AD1910-1992, Gosden and Knowles found few changes in the formal qualities of trade objects through time. Important trade objects such as Siassi bowls, Pig's-tusk ornaments and barkcloth show some variation, but none of this variation appears to be directional in the sense in which objects changed in relation to the demand of collectors (2001: 174). The only major change noted was a change in the decoration of shields, which although remaining uniform in size and shape throughout the colonial period, altered so that the back of the shield came to be undecorated. This change is consistent with a shift from shields as functional objects (when decoration on the reverse of the shield would have been visible to the warrior in battle) to aesthetic objects, primarily manufactured to be viewed from the front. Importantly, Gosden and Knowles (2001: 175ff) link these formal changes in shields to a decline in their importance as markers of masculine identity.

Torrence's work in the Admiralty Islands involved a quantitative analysis of over 1,500 obsidian spear and dagger blades manufactured between the early nineteenth century and AD1990, now held in museum collections across the world. She documents the ways in which people on Manus modified both the form and decoration of obsidian blades to suit the tastes of European collectors. Eventually obsidian blades came to be mass produced, with a corresponding decrease in the quality of the bulk of those manufactured; however, some manufacture of very large finely worked points continued to supply a 'serious' collector's market (1993: 477, 2000: 131). By the end of World War 1, primary manufacture of obsidian blades had almost entirely ceased, and scavenging of previously discarded blades, along with unexacting retouch techniques, meant that blades became smaller, while decoration of the haft or handle also became, in general, simpler in form (2000: 130-131).

While at first glance the 'direction' of change in the case of Kimberley points appears to be different to both of the studies cited above, what needs to be kept in mind is Australian colonial collectors' obsession with the level of 'fine working' on Kimberley points. The areas of similarity lie in the way in which mass production for a colonial curio market led to formal changes in the objects themselves which can be read as 'artefacts' of colonial taste or preference. I think it would be fair to say that collectors 'created' Kimberley points as much has Kimberley Aboriginal people did, in the sense that it was their selection for particular formal qualities, in negotiation with a range of new factors within Aboriginal societies which changed the context of manufacture, which led to the stylistic formalisation of Kimberley point form after c.AD1885. 'Kimberley points' should be seen as an extreme formal variant of pressure flaked points manufactured throughout the Kimberley region after c.1000-1400BP, developed to meet the particular taste of colonial collectors in the contact period. Kimberley points represent the result of nexus (after Gell 1998) of Australian colonial relations. These colonial relations were often enacted through the movement of objects that served to link both colonists and colonised together, such that the objects themselves can be seen to have a form of agency (Kopytoff 1986; Gell 1998; Harrison 2003). These colonial relations can be better understood through an analysis of the peculiar set of social and aesthetic qualities that are represented by these changes in the formal qualities of Kimberley points after AD1885 (after Gosden and Knowles 2001: 22; see further discussion in Harrison 2002a).


The data discussed here suggests that points collected from pastoral station and other Aboriginal post-contact contexts (such as missions and prisons, but possibly not from those post-AD1885 sites that are not associated with contact with Europeans) are more likely to be Kimberley points. The collected museum examples indicate antiquarians' preferences for collecting invasively pressure flaked points with denticulate or serrate margins. While pressure flaked points with serrate, denticulate and dentate margins certainly do appear to have existed in archaeological contexts in the Kimberley since before AD1885, the preference for particular point shapes by collectors in the post-contact period appears to have had a direct influence on the prevalence of large, invasively pressure flaked points with serrate and denticulate margins in post AD1885 archaeological sites associated with Aboriginal pastoral workers' encampments. Kimberley points as defined by Akerman and Bindon (1995) should be seen an extremely formalised morphological variant of the pressure flaked points that were made prior to AD1885 in the Kimberley, developed to meet the demand of colonial collectors and other Aboriginal groups who received them through trade. Kimberley points were developed as part of a dialectic relationship between the Aboriginal people that manufactured them, and the colonial collectors that created a market for them (see also Bradshaw 1990), and existed in the contact period alongside a range of other less formal 'functional' pressure flaked point forms.

I am conscious of the degree to which the standard technological attributes used in these analyses are not sensitive to my own qualitative observations that post-contact points are generally more finely worked and more formalised in terms of shape and design. This suggests the requirement for more targeted research that focuses on the formal characteristics of Kimberley points, as opposed to standard technological attributes used by Australian archaeologists to describe what are in general less intricately worked stone artefact types, to address the issue of the chronological relationship between shape and invasiveness and proportion of retouch on Kimberley points further. There is also clearly a requirement for further targeted excavation that addresses the issue of the chronology of particular Kimberley point forms, in particular the manufacture of dentate Kimberley points as opposed to denticulate and serrate Kimberley points, to address this problem adequately.

Recent archaeological work in the southeast Kimberley would suggest that there is a chronological progression from early percussively flaked unifacial and bifacial points dating to before c.2300calBP, to pressure flaked points after c.1400-1000calBP. This accords relatively well with the data presented by Akerman et al. (2002) from the Keep River excavated sites, which show the appearance of pressure flaked points in layers with optical age estimates of between 540 and 1200 years before present, and data cited by O'Connor (1999: 71, 76), which suggests the appearance of pressure flaked points in the coastal west Kimberley at around 1000 years before the present, and percussively flaked points from as early as 4500BP (Bowdler and O'Connor 1991). Preliminary analyses suggest that early pressure flaked points were highly divergent in shape, while post AD1885 points, particularly those manufactured on pastoral camps and fringe settlements, became increasingly formalised in terms of the degree to which points were invasively worked and the occurrence of serrate or denticulate marginal treatments. The mechanisms associated with this formalisation of design developed as a result of post-contact social relations, and the mutual self-definitions that occurred between the Aboriginal people who made points, Aboriginal groups who received them in trade, and non-Aboriginal collectors (Harrison 2002a).


Funding for multiple field seasons of research in the Kimberley over the period 1997-2000 was kindly granted by the Centre for Archaeology at the University of Western Australia and the Australian Institute of Aboriginal and Torres Strait Islander Studies (Research Grant (397/6071). I am indebted to the many individuals and corporations who provided support to me over this time, in particular Josey Farrer, Doris Fletcher and members of Yardgee community, the Lamboo mob [especially Jack Jugari, Stan Brumby, Pattercake Imbelong, Doris Ryder, Jack Ryder, Barbara Imbelong and Jerry Woodhouse], Henry Achoo and other members of Mardiwah Loop community. Thanks must also go to the Kimberley Land Council and Kimberley Language Resource Centre for facilitating this research and to Kate Golson, Joseph Blythe and Mary Anne Taylor for logistical and personal assistance in the field. My field team of Kathryn Przywolnik, Danny Tan, Genievieve Clune, Stewart Morton, Joe Blythe and Ashley Johnson worked tirelessly through several long field seasons. Kim Akerman and Richard Fullagar read and provided useful comments on an earlier draft of this paper, while Richard Fullagar and Tessa Boer-Mah allowed me access to unpublished data from the Keep River which allowed me to refine the paper substantially. Peter White and an anonymous reviewer provided comments that helped me to focus the argument, for which I thank them both.
Table 1: Points recovered from Wilinyjibari and associated dates

                   Approx age    Number
Unit                (calBP)     of points   Description

Upper    surface            0       1       1 point tip
            1             100       2       2 pressure flaked point
            2        170-200        2       2 pressure flaked point
            3             215       1         Pressure flaked point
            4        225-280        2       2 pressure flaked point
Middle      9             600       3       2 pressure flaked tips,
                                              1 percussively flaked
           10             750       4       1 butt and 3 tips,
                                            3 pressure flaked and
                                            1 percussively flaked
           11             870       7       4 butts percussively
                                              flaked and 3 tips
                                              pressure flaked
           12             950       3       Two butts and one point,
                                              percussively flaked
           13            1000       1       Pressure flaked tip
           14      1000-1400        2       Pressure flaked tips
                                            [oldest pressure flaked
                                              point tips]
Lower      18            2300       2       2 point butts,
                                              percussively flaked
total                              30

Table 2: Numbers of pressure flakes made on different
raw materials in 5mm and 3mm sieved fractions from
quadrat D at Wilinyjibari

                   White   Other
          Unit     chert   cherts   Quartz   Total

         surface     0       0        0        0
            1        0       1        1        2
            2        4       0        0        4
            3        3       2        1        6
            4        5       2        0        7
            5        0       0        0        0
            6        2       0        0        2
            7        6       2        0        8
Upper       8       13       2        3       18
            9        9       1        2       12
           10       19       8        7       34
           11       20       6        5       31
           12       13       4        3       20
           13       19       5        9       33
Middle     14        1       1        2        4
           15        0       0        0        0
           16        0       0        0        0
Lower      17        0       0        0        0
           18        0       0        0        0
           19        0       0        0        0
           20        0       0        0        0
           21        0       0        0        0

Total               114      34       33      181

Table 3: Numbers of points as proportion of retouched stone artifacts
from analysed open sites, southeast Kimberley [includes point blanks]

                                                Stone       Points
Collection                Period            artefacts (n)    (n)

Junda                  Pre-contact                78           0
Jawa                   Pre-contact                58           2
Marnyjal 1             Pre-contact               214           6
Marnyjal 2             Pre-contact                42           0
Lamboo 2           Post-contact station          723          36
Lamboo 3           Post-contact station         1076          55
Lamboo outside   Post-contact off-station        131           1
Marlinyana       Post-contact off-station        476           1

                 Kimberley      Retouched     Points
Collection       points (n)   artefacts (n)    (%)

Junda                 0            14            0
Jawa                  0             6           33
Marnyjal 1            1            39           15
Marnyjal 2            0             4            0
Lamboo 2             10            60           60
Lamboo 3             14            70           79
Lamboo outside        0             6           17
Marlinyana            0            56            2

Table 4: Summary of material that appears as Tables 2 and 3 in Akerman
et al. (2002: 32-35). BFI = Bifacial, invasive, serrate;
BIM = bifacial, some edges marginal, some invasive; BM = bimarginal;
BP = backed point; UI = uniface, invasive flaking; UIM = uniface, some
edges marginal, some bifacial; UM = unifacial, marginal flaking.
2 points not listed for their marginal treatment and 7 points from
BOAB excavations omitted.

Point types   Museum Points   Excavated Points
              (post AD1895)     (pre AD1895)

    BFI         68 (81%)          8 (25%)
    BIM          8 (10%)          5 (16%)
    BM           6 (7%)              0
    BP              0             4 (12%)
    U1           1 (1%)           5 (16%)
    UIM             0             4 (12%)
    UM           1 (1%)           6 (19%)

   total           84               32 *

(1.) This date has been chosen to represent the beginning of the period of sustained non-Aboriginal settlement of the inland Kimberley. It represents the year after which the first Durack pastoral stations were stocked in the east Kimberley and the time of the first gold rush following the discovery of gold by Hall and Slattery at Halls Creek. Further discussion of the significance of these events can be found in Harrison (in press).

(2.) In the absence of other chronological markers (for example the presence of knapped bottle glass artefacts) to distinguish between pre and post-contact archaeological sites in the study area, a series of open stone artefact scatters thought by informants not to have been used as camping places in their and their parents' lifetimes formed the 'pre-contact' archaeological site sample. The methods of site selection are discussed in detail in Harrison (2002c).


Akerman, K. and P. Bindon 1995. Dentate and related stone biface points from Northern Australia. The Beagle, Records of the Museums and Art Galleries of the Northern Territory 12: 89-99.

Akerman, K., R. Fullagar and A. van Gijn 2002. Weapons and wunan: production, function and exchange of Kimberley points. Australian Aboriginal Studies 2002/1: 13-42.

Allen, H. 1997. The distribution of long Nades: evidence for recent changes in Aboriginal ceremonial exchange networks. In P. McConvell and N. Evans (eds) Arcbaeology and Linguistics: Aboriginal Australia in Global Perspective; pp. 357-376. Oxford University Press, Melbourne.

Attenbrow, V., B. David and J. Flood 1995. Mennge-ya and the origins of points: new insights into the appearance of points in the semi and zone of the Northern Territory. Archaeology in Oceania 20: 105-120.

Balfour, H.R. 1903. On the methods employed by the Natives of N. W. Australia in the manufacture of glass spear heads. Man 1(5): 65.

Balfour, H.R. 1951. A native toolkit from the Kimberley district, Western Australia. Mankind 4(7): 273-4.

Balme, J. 2000. Excavations revealing 40,000 years of occupation at Mimbi Caves, south central Kimberley, Western Australia. Australian Archaeology 51 : 1-5.

Basedow, H. 1925. The Australian Aboriginal. Preece, Melbourne.

Boer-Mah, T.S. 2002. Beyond the dripline: Comparison of Rockshelters and Excavated Open Sites in the Keep River Region, Northern Territory, Australia. Unpublished B.A. (Hons) thesis, University of Sydney.

Bowdler, S. and S. O'Connor 1991. The Dating of the Australian Small Tool Tradition, with new evidence from the Kimberley, WA. Australian Aboriginal Studies 1991(1): 53-61.

Bradshaw, W. 1990. European Materials and Aboriginal tool technology, East Kimberley. Unpublished BA (Hons) thesis, University of Western Australia.

Davidson, D.S. 1934. Australian spear-traits and their derivations. Journal of the Polynesian Society 43: 41-72; 143-162.

Elkin, A.P. 1948. Pressure flaking in the Northern Kimberley, Australia. Man 48:110-113.

Gell, A. 1998. Art and Agency: An Anthropological Theory. Oxford University Press, Oxford.

Gosden, C. and C. Knowles 2001. Collecting Colonialism: Material culture and colonial change. Berg, Oxford and New York.

Harrison, R. 2000. Challenging the authenticity of antiquity: contact archaeology and Native Title in Australia. In I. Lilley (ed.), Native Title and the Transformation of archaeology in a Postcolonial World; pp.35-53. Oceania Publications, University of Sydney, Sydney.

Harrison, R. 2002a. Archaeology and the colonial encounter: Kimberley spear points, cultural identity and masculinity in the north of Australia. Journal of Social Archaeology 2(3): 352-377.

Harrison, R. 2002b. Australia's Iron Age: Aboriginal post-contact metal artefacts from Old Lamboo station, southeast Kimberley, WA. Australasian Historical Archaeology 20: 67-76.

Harrison, R. 2002c. Ngarranganni/Ngamungamu/Jilini-jarra: 'lost places', recursiveness and hybridity at Old Lamboo pastoral station, southeast Kimberley. Unpublished PhD thesis, Centre for Archaeology, University of Western Australia.

Harrison, R. 2003. 'The magical virtue of these sharp things': mimesis, colonialism and knapped bottle glass artefacts in Australia. Journal of Material Culture 8: 311-36.

Harrison, R. in press. Contact archaeology and the landscapes of Pastoralism in northern Australia. In T. Murray (ed.) The Archaeology of Contact in Settler Societies. Cambridge University Press, Cambridge.

Harrison, R. and D. Frink 2000. The OCR Carbon Dating Procedure in Australia: new dates from Wilinyjibari rockshelter, southeast Kimberley, Western Australia. Australian Archaeology 51: 6-15.

Head, L and R. Fullagar 1997. Hunter-gatherer archaeology and pastoral contact: perspectives from the northwest Northern Territory, Australia. World Archaeology 28(3): 418-428.

Idriess, 1. 1938. Over the Range: Sunshine and Shadow in the Kimberleys. Angus and Robertson, Sydney.

Kopytoff, I. 1987. The cultural biography of things: commoditization as process. In A. Appadurai (ed) The social life of things: commodities in cultural perspective; pp. 6491. Cambridge University Press, Cambridge.

Love, J.R.B. 1936. Stone-Age Bushmen of Today: Life and Adventure among a tribe of Savages in North-Western Australia. Blackie and Son Ltd, London and Glasgow.

McCarthy, ED. 1976. Australian Aboriginal Stone Implements. Second edition. The Australian Museum Trust, Sydney.

Mitchell, S.R. 1949. Stone-Age Craftsmen: Stone tools and Camping places of the Australian Aborigines. Tait Book Company, Melbourne.

O'Connor, S. 1995. Carpenter's Gap Rockshelter 1:40,000 years of Aboriginal occupation in the Napier Ranges, Kimberley, WA. Australian Archaeology 40:58-9.

O'Connor, S. 1996. Thirty Thousand years in the Kimberley: Results of Excavation of three rockshelters in the Coastal West Kimberley, W. A. In P. Veth and P. Hiscock (eds), Archaeology of Northern Australia: Regional Perspectives; pp. 26-49. Anthropology Museum, The University of Queensland, St. Lucia.

O'Connor, S. 1999. 30,000 Years of Aboriginal Occupation: Kimberley, North West Australia. Terra Australis 14. Department of Archaeology and Natural History and Centre for Archaeological Research, The Australian National University, Canberra.

Petri, H. 1954. Sterbende Welt in Nordwest-Australien. Limoach: Braunschweig.

Roberts, R., M. Bird, J. Olley, R. Galbraith, E. Lawson, G. Laslett, H. Yoshida, R. Jones, R. Fullagar, G. Jacobsen, and Q. Hua 1998. Optical and radiocarbon dating at Jinmium rock shelter in northern Australia. Nature 393: 358-362.

Smith, M.A. and B.J. Cundy 1985. Distribution maps for flaked stone points and backed blades in the Northern Territory. Australian Aboriginal Studies 1985(2): 32-7.

Spencer, B. 1922. Guide to the Australian Ethnological collection exhibited in the National museum of Victoria (3rd Edition). Albert Mullet, Government Printer, Melbourne. Spencer, B. 1928. Wanderings in Wild Australia. MacMillan, London.

Tindale, N.B. 1985. Australian Aboriginal techniques of Pressure-Flaking Stone Implements: some personal observations. In M.G. Plew, J.C. Woods and M.G. Pavesic (eds), Stone Tool Analysis: Essays in Honour of Don E. Crabtree; pp. 1-34. University of New Mexico Press, Albuquerque.

Torrence, R. 1993. Ethnoarchaeology, museum collections and prehistoric exchange: obsidian tipped artefacts from the Admiralty Islands. World Archaeology 24: 467-81.

Torrence, R. 2000. Just another trader? An archaeological perspective on European barter with Admiralty Islanders, Papua New Guinea, in R. Torrence and A. Clarke (eds), The Archaeology of Difference: negotiating cross-cultural engagements in Oceania; pp. 104-41. Routledge, London.

Veitch, B. 1996. Evidence of mid-Holocene change in the Mitchell Plateau, Northwest Kimberley, Western Australia. In P. Veth and P. Hiscock (eds), Archaeology of Northern Australia: Regional Perspectives, pp. 66-79. Anthropology Museum, The University of Queensland, St. Lucia.

Vietch, B. 1999. What happened in the mid-Holocene? Archaeological evidence for change from the Mitchell Plateau, Northwest Kimberley, Western Australia. Unpublished PhD thesis, University of Western Australia, Nedlands.

Wallis, L. and S. O'Connor 1998. Residues on a sample of stone points from the west Kimberley. In R. Fullagar (ed.) A Closer look: Recent Australian studies of Stone Tools, pp. 149-178. Sydney University Archaeological Methods Series 6. Archaeological Computing Laboratory, University of Sydney, Sydney.

Walsh, G.L. and M.J. Morwood 1999. Spear and spearthrower evolution in the Kimberley region, N.W. Australia: evidence from rock art. Archaeology in Oceania 34(2): 45-58.

White, J. P. and J. F. O'Connell 1982. A Prehistory of Australia, New Guinea and Sahul. Academic Press, Sydney.
COPYRIGHT 2004 Blackwell Publishing Limited, a company of John Wiley & Sons, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Harrison, Rodney
Publication:Archaeology in Oceania
Geographic Code:8AUST
Date:Apr 1, 2004
Previous Article:Modelling hunter-gatherer settlement patterns: an Australian case study.
Next Article:Late Lapita occupation and its ceramic assemblage at the Sigatoka Sand Dune site, Fiji, and their place in Oceanic prehistory.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |