Archaeology of Lake George

ALLISON BYRNE: Good evening everyone, welcome to this Friends of the Museum and CAS [Canberra Archaeological Society] great night. What a wonderful turnout. We’re not even offering Tim Tams or anything, I just think it’s marvellous. Anyway, on behalf of the society I would like to acknowledge that we meet on the lands of the Ngunnawal, Ngunawal, Ngarigo and Ngambri peoples. We respect their knowledge and commitment to this country. I’d also now like to introduce Amy Way. Amy is going to speak on Lake George and the background to this is that it’s a massive resource rich upland fresh water lake that is unique in Australia. Amy Way examines the human occupation and environmental prehistory of the area in an attempt to discover more about the prehistoric foraging.

For her PhD, Amy used an event based approach to examine the stone artefacts of the area and they provided a high level of detail into the hunting and foraging and gathering of the peoples here in the Aboriginal Australia. I’m sorry that I couldn’t read that very well, my eyes are all blurred from those lights, so I hope you don’t have any problems with it. I’d also just like to welcome you if you’d like to join the Canberra Archaeological Society. We are having our AGM on the 29 November at the Fellows Bar. If anybody would like to come along to that meeting, hear what we’re on about, maybe offer some suggestions about what could happen next year. I know Amy gave a terrific walk around Lake George before she put her PhD in. There was so many people that just so thoroughly enjoyed it.

You’re most welcome to come along with your ideas and come along to the AGM at the Fellows Bar at ANU [Australian National University] at 6 o’clock on Wednesday the 29th. Anyway, welcome Amy.

[applause]

AMY WAY: Thank you everyone for coming out. I have to admit, this is the largest crowd I’ve spoken to, so it’s very exciting to see you all here. My name is Amy and I have just submitted my PhD on this topic on the archaeology of Lake George. It’s a pleasure to now be shifting from studying to talking about my findings. I just want to start, before I go anywhere by saying that there are two ways of understanding the past. One is through the cultural framework. To speak about cultural meaning and cultural significance you need to be a cultural knowledge holder, for which I’m not, so I’m not going to talk at all about the cultural significance of these finds. I’m situated wholly within the scientific framework and speaking as an archaeologist about the artefacts.

At Lake George, it’s a very harsh environment, so the only evidence we have that is surviving are the stone artefacts. There would have been in the tool kit, many more objects that were made of organic materials, textiles and so forth, but these simply haven’t survived, to the point that I thought, in this landscape that it is currently, mostly farmland. You have animals dying with the bones being left on the surface. I thought that during the excavation I would have at least excavated a sheep or a rabbit or two, but nothing. Not a single piece of bone or anything else organic survived. All we have in terms of archaeology are the stone artefacts. That’s what my whole thesis was, was on the stone artefacts that were excavated.

I just want to tell you a little bit of a back story about how archaeologists construct meaning. The principle way in which we do it is through establishing a relationship between objects. One of the main ways we do this is we look at the stratigraphic profile and then we look at the artefacts, which are found together at the same layer and at the same depth. Then we make assumptions after many tests and analysis of contemporaneity then between those objects. Those artefacts are then considered as a sum group and then their meaning is established through the relationship between the objects from each temporal layer.

But at Lake George, we have a problem [laughter]. On the left we have what is a fairly typical cave section and on the right we have what we find at Lake George. We just have this uniform sand body with no evident layering. That’s problem number one. Problem number two is that it’s sand, so all of the excavations were conducted in sand bodies. Something that happens when artefacts are in sand is that they move vertically. This isn’t unique to Lake George. [shows slide] This is a site in Belgium, which is also a sand dune site. You can see the artefacts from one occupational event have been dispersed by 50 centimetres. A very similar thing is evident at Lake George. We have artefacts that have conjoins so we know that they are from a single knapping event that then become vertically dispersed through time by sometimes up to 70 centimetres.

These two problems led me to think, well how am I going to analyse these assemblages because a traditional analysis in these environments is by depth, but what happens if you take a slice by depth? What happens is that you’re getting some artefacts from one knapping event, some from another and some from enumerable number of different knapping events then some artefacts are combined from each into what effectively is a meaningless data set. There is no relationship then to the original activities, which produced the artefacts. I thought that is a much better unit of analysis.

The thing that I really wanted to be able to say at the end of my thesis is what were the people doing at the time? To do that, I had to make sure my unit of analysis aligned with the action, which actually produced the artefacts. In the end it became a methodological thesis and I’ve termed it event based analysis where we shift the unit of analysis from being divided by temporal depth into by action.

Tonight, I’ll start by just giving you some background about the excavations and then I’ll talk about the archaeology itself and finish with some conclusions. This is Lake George, I think you’re all probably quite familiar with the landscape [shows slide]. These were the five open air excavations. I had two here, which were on a little creek called Bridge Creek, which runs into Butmaroo Creek, which ultimately ends in the lake. When I first started I thought, ‘I’m going to look at the whole of the Lake George basin,’ which is about 1000 square kilometres, and then I realised that the pits I was planning to dig were a quarter of a square metre and I realised that I was probably a little ambitious. I narrowed then my field of study to just Butmaroo Creek which runs in on the south eastern side. The first two sites were positioned either side of this little creek here, to here which were on a ridge next to – there’s a lagoon formation there and one up on this elevated ridge line.

The other thing I think many of you will know about Lake George is that the level has fluctuated through time. [shows slide] These grey deposits here are actually old shore lines to the lake. As the lake level comes up, a new shore line’s formed, the lake level drops and that elevated feature remains in the landscape. This is a look at the changing lake levels through time. You can see that way back here, around 30,000 years ago, the lake was much, much higher than what it is today. It was actually to the point where when you’re driving on the freeway from Canberra to Sydney, and you come off the escarpment, just at that point where you go down through Gearys Gap, that was the level of the lake, around 30,000 years ago. Since then its dropped but it’s also fluctuated. This is the period of time that I was really interested in. You can see that the lake level is fluctuating through that period, which brought the shoreline right up to the edge of the sites, at that point, and then around 1000 years ago it dropped back to around there.

During this period of occupation, the last 5000 years, the lake shoreline would have shifted back and forward between this zone. That means that even though today that all of the sites are about five kilometres from the current lake shoreline, they would have been within 500 metres at the time of occupation.

This is the first area, this is Bridge Creek [shows slide]. You can see the little creek running through there. I had one excavation on this side of this creek, just over here and one up there where that car is parked. The first one was around here and the second one here. The creek has cut this lovely section here through the deposit and this photo is a close-up of it. The top 900 is the Aeolian sand and that’s the bit that I was interested in, that’s the bit that I dug down to. It’s underlain by this bedded cobble layer. The nice thing about that was that it provided a really clear point of correlation for the excavations on either sides of the creek. That’s what the section looked like once it was excavated. You can see the cobalt layer adhering just at the base there.

The dates, I had both OSL [optically stimulated luminescence] and carbon dates. The ones on the left are the OSL, so that’s when the – I don’t understand it in its entirety, but basically you can tell when a grain of sand was last exposed to sunlight. That gives you the age they were last exposed to sunlight, so the effective age of that part of the deposit. What that gave was a basal age for the deposit.

This is the second area, Wrights Creek lagoon [shows slide]. These lakes here, they sometimes had water in them and sometimes didn’t as we excavated. You can see they just had water in them at this point. To start with, I dug all these test pits with the help of all these great people. Two of them were then selected for open air excavations and one of them up on the crest is this one, and then the one here further down. This runs down here, is this one here. This is a close-up of the section there. You can see again, these are the OSL dates and these are the C14 [carbon-14] dates. This site was quite exciting because it had several hearths at it. They provide really good points of reference in an excavation. There were two quite deep pit hearths and then there were a couple of these shallower hearths there as well.

This is the fifth site, and it was up on this elevated ridge line here, which you can see has magnificent views to the north over Lake George [shows slide]. There we are excavating and again we have the OSL ages and the carbon ages. This was the only site where there was any real discrepancy between the two sets. We’re looking at this date a bit further at the moment, and so this may change, this number here. This figure here was taken from a hearth, so at the moment, the simplest explanation is that this hearth was dug into older sands but that might change, that age.

Before I get into the way the artefacts were analysed, I just wanted to give you an overview of what they looked like when they first came out and I did my first spatial knapping. I just plotted the artefacts by depth and raw material and they are all producing this distributed, vaguely approximating a normal curve vertically. But when they are plotted spatially, they created these really nice, distinct concentrations of artefacts, which said to me that even though there appeared to be a lot of this vertical displacement going on, the concentrations didn’t appear to be very affected spatially. They seemed to retain their coherence. This was evident across all the sites. We have this vertical displacement but really good lateral bounding to the concentration. This is actually a massive course event here that I know Jen in the corner helped excavate. Again, there’s the bedded cobble layer at the base of that excavation.

At the lagoons, this was the only site where there wasn’t any apparent concentrations of stone artefacts, they were much more evenly dispersed and presented with a more uniform distribution. I’ll return to the reason why later. [shows slide] This is the other Wrights Creek lagoon site, and it did have again, the really clear concentrations by raw materials, that’s a very dense silcrete knapping event. The quartz and the chert were more evenly or uniformly distributed. Here you can see that again, even though there’s the vertical displacement, it wasn’t an entire mixing because we’ve got one raw material being positioned here and then another being positioned below it. It was some kind of action, which was just churning the deposit, then you wouldn’t see the separation between the two curves by raw material. I could see that some level of coherence was being maintained vertically but there was also the vertical displacement.

Bulla camp Hill, again [shows slide], we had a really clear concentration of silcrete here and down the bottom a very dense quartz knapping event. Here again, the artefacts – so even though they’re distributed vertically, there’s some kind of separation being maintained between this upper level and this lower level with many more of the silcrete and the chert artefacts up the top, and the lower level being dominated by quartz.

In the background I was thinking about how to analyse these. I’d come up with this idea about analysing them in terms of knapping events. This is how I did it. I started with the assemblage and then I divided them into raw material units. Each of these units is effectively a single piece of stone. What I wanted to be able to do is to say what all the artefacts were. To reunite all the artefacts that were removed from a single piece of stone that’s brought into a site. Each RMU [raw material unit] is effectively one rock. The first bit is identifying, which of all those artefacts belong to each unit. Then the next step is looking at what the composition of each of those units is, how many artefacts are in it, what levels of retouch are involved, how much production debitage there is, and part of that involves the attribute analysis and part of it also involved refitting.

Then we come into the life history framework. This is really a famous archaeologist named [Michael Brian] Schiffer, this is his life’s work. What I wanted to do was to take his idea that by doing this you can look at what form the stone was brought into a site, what actually took place on site. What reduction happened, what was discarded, then lastly, whether anything was transported on or whether it was all discarded at that point. The final step, I wanted to be able to then position each of these events in a sequence because the big thing that’s lacking in open sites, particularly one’s like Lake George where we have this vertical displacement, is that it’s very hard to then construct prehistoric sequences. Because of this, most of Australia’s prehistory is written from cave deposits. I really wanted to test this idea as to whether these open air sites and particularly open air sandy sites, could be brought back into the sequence issue. I think they can. The first thing that has to happen is you have to do the disturbance assessment to make sure this whole scale disturbances so forth hasn’t taken place. After that, I think that they can still be positioned, the units can be reunited and positioned in a temporal sequence.

What it effectively involves then is a very moment in time event. Something that’s happened within hours or within days being positioned then within this very coarse temporal scale. We’ve got the two scales operating in unison, one that’s within a human day or a couple of days and then we have a thousand years. Somewhere, that very discrete moment took place but positioning it exactly is not possible; but I think it is possible to position them relatively to each other and then to group them by raw temporal periods to look at change through time.

This is the first part of the analysis, the raw material unit analysis [shows slide]. There’s a lovely microscope at Sydney University that has this split screen function. At the base of it you can hold your reference piece here. You can see at the bottom of these four images that’s the same image. That would be the reference piece for that particular unit. Then I could pass all the artefacts that I think belonged to that unit under the microscope and compare the matrix and the inclusions to see whether they were indeed of the same geological source. This was particularly useful in cases like this. Colour is sometimes used as an attribute that can say whether or not artefacts are from the same rock. If they’re heated as has happened here – this is a single flake that’s been split at the point of manufacture and half of it has flown into the fire, been heated and changed colour. It means when you’re trying to catalogue them, you can’t really rely on colour to say whether or not they’re from the same piece of rock. It is much better to look at them microscopically and then analyse the matrix and the inclusions which don’t change with heat or weathering.

These were my final numbers. I had nearly 7000 pieces of non-quartz artefacts. ninety-one per cent of them were assigned to one of these raw material units. That shifted my unit of analysis from these artefacts to these units. It was the units which formed the central body of my thesis in the way that I analysed them.

This is the second part [shows slide], and it involves the attribute analysis which just looking at the characteristics of each of the artefacts and also the refitting. You can see there, I am with some of my trusty helpers refitting. We refitted 1250 artefacts. There were three types of refit that we formed. This is a stacking refit. What’s happened here is the core would have been here, and this flake was removed first and then this one and then this one and then this one. This was a really interesting case here. What they’ve done is they’ve prepared the platform by removing material before the flake struck. Originally the core would have been up here and as the platform is being prepared, the core length has reduced through the sequence. This is a breakage refit and then we also had retouch refits. This is a flake obviously broken as well. This is a flake that’s then been retouched. This little flake was removed off this flake and then refitting could then tell the story of how it would have been removed.

This here is an example of a single unit [shows slide]. This unit had one, two, three four, five flakes. That’s the ventral and the dorsal of the same flake. You can see that little platform preparation flake has been refitted onto the back here. What that tells us is that when this core was brought in, only a very small number of flakes were removed, and then the core isn’t present. That means that the person who brought it in, or someone subsequently then carried that core on for use at another place – what that tells us is that this piece of stone had a long life history as it was moved across multiple sites and worked across multiple sites in the landscape. It also tells us that what people were doing was not intensive reduction at this stage. They were probably just preparing the core, just removing a few flakes to get the core ready for transport and use elsewhere.

These are just a couple more examples of these units that I formed [shows slide]. This here was a really great example. If you can see here, that’s a platform and that’s a bulb. This piece here was a very thick flake, which was subsequently rotated and then knapped as a core itself. Onto its ventral surface refitted this elongated flake which has been snapped transversely and then the top of the proximal pit section has then been reworked as a backed artefact. On the dorsal side of this large flake which became a core, there was some cortex which tells us that when it was brought in, it probably hadn’t been very well worked before hand. This might have been one of the first sites that that stone was worked. Then the person removed a large flake and another elongated flake. Then they reworked the flake as a core and produced a backed artefact and then they carried the original core on for use elsewhere.

[shows slide] This is from the same site and that contrast with an artefact like this, which is a single, large, handheld scrapper which had no associated reduction debris. From that I can say that it wasn’t created at this site. That scraper was created somewhere else, it was formed, it was brought in as a fully formed object. It was then retouched or resharpened on site and we have one resharpened sharpening flake refitted there. Then it was discarded and so nothing was carried on.

After all the units had been identified and described and so forth, then I came to this question of sequencing. I was very mindful of needing first and foremost to rule out whole scale disturbance. Luckily there’s quite good methods for ruling that out. The first is by examining the assemblage for very small artefacts. If an assemblage has been affected by wind or water, the very small artefacts will be removed by those processes, but there were many, many very small artefacts, so I could rule that out. The second is by looking at the lateral coherence and the knapping events had retained really good boundaries. I felt confident in saying that there hadn’t been any whole scale disturbance, even though I was seeing the vertical shifting.

My next concern was looking at patches of disturbance and particularly burrowing. We know in this landscape there are wombats, foxes, rabbits and so forth. The last form of disturbance is really a form of disturbance which happens as the assemblage is formed. That is occupational trampling. That’s as the people who were making the artefacts are using the site and walking around and camping – because it’s sand, that action of just simply being there pushes the artefacts into the deposit and causes this vertical displacement. These days there have been a lot of experiments on the signature of trampling looks like, and it looks like a normal curve. What you have is you have the artefacts that are quite large not being as mobile as the very small artefacts and the artefacts that are larger maintaining position at the level of deposition with the smaller artefacts which are more mobile shifting above and below it through time.

Burrowing, within the space of the burrow itself, it causes effectively whole scale churning. That action shifts both the large and small artefacts. This was a lovely case, I knew that there was a burrow because I excavated and had seen the change in colour and I knew that the artefacts here were from a burrow. When I plotted them by depth, we’ve got mostly the artefacts positioned within a normal curve, which you’d see from the occupational trampling, and then we have a couple of large ones up here in the burrow zone. I felt like that was confirmation of the experimental literature that I read that said that the trampling signature could be distinguished from a burrowing signature.

With that in hand, I went forward and positioned the artefacts, so each of the knapping events was then positioned in a sequence. I removed the outlying small artefacts and then produced box plots of the remaining. The threshold was 30 square millimetres. Everything smaller than that was removed and then the box plots illustrated the spread of the remainder. The really nice thing that this did, was it decoupled artefact density from this idea of occupational density, because now I could plot, rather than looking at artefact density, I could look at the number of events over time, the number of knapping events. This has really illustrated very well at this site where we have many more artefacts in this upper level and fewer in this lower level [shows slide], but if we look at the number of events, we have many more events in the lower level. It’s just that each of these only involved the removal of a limited number of artefacts. This looks like an increase in activity over time but really, its’ a reduction in the number of knapping events over time. It’s just that in this period, people were removing hundreds and hundreds of artefacts in any single event.

This is back at the lagoons site [shows slide]. This was the biggest conjoined set that was produced through the project. It actually had this second set but I just could not get it into it for all the trying in the world, which says that there must have been some removal of material as well that prevented the refitting. All of these artefacts were effectively made from this one action, this one importation of stone and extensive knapping where more than a thousand artefacts were removed in one knapping event. [shows slide] That compares to this site which is again, at the lagoons but further up the hill. Remember I said this one didn’t have any evident clustering and the reason it turns out, is that this side had by far the greatest number of events but each of the events was quite small. Rather than this site which only had a limited number of events but were very productive, producing this well bounded concentration, here we have many, many knapping events but they’re all much smaller. The net result is that you don’t get any evidence of spatial clustering.

This was where the two deep pit hearths were as well. It had the most events and the two pit hearths. I think that what was happening here is that a lot more of these actions involved just the limited removal of flakes or the preparation of cores, or the maintenance of tools. It wasn’t a primary tool production site, it was more a place where people repaired and maintained their tool kits. This was the only site where it didn’t really change over time. The number of events were fairly constant. The lagoons must have been a fairly attractive point through the whole period as a focal point of the landscape, which meant that we were capturing that repeated use of a place. That contrasted very nicely with Bridge Creek junction, where in the earliest period, they were only two events and then there was this explosion around 3000 years ago, with a huge number of knapping events here. Here it was principally the production of backed artefacts.

This was a really nice example [shows slide]. I think what’s happening here … so we have all of these artefacts and these ones which refit and the crosses are all of these ones. Overlying them were all of these backed artefacts. These two pieces of stone were very similar geologically. This one had a red stripe through it, but apart from that, the matrix and the inclusions were very similar. What it looked like to me was that they had been brought in from possibly the same quarry and at the same time. They’re about a metre apart. It could be, if I was brave, I could say that it was two people sitting next to each other. One of them was producing backed artefacts and the other one wasn’t, there weren’t any backed artefacts in this set here. There were however, a lot of missing flakes. I think that this person was focused on the production of elongated flakes. This person was focused on the production of backed artefacts.

We know from other studies in south eastern Australia that these tools are often hafted. The idea is that they’re very regular in their shape because it’s much easier to make the stone piece when it gets exhausted than it is to make the wooden haft. As the stone got exhausted, all of these ones at the top had a heavy use where [shows slide], as they’re exhausted, they’re removed, new ones are manufactured and they’re replacing the haft.

This was one of the really interesting trends which appeared through time was the change in the way people were transporting artefacts. In the earliest period before around 3000 years ago, mostly people were transporting cores, and then around 3000 years ago, there was a massive change and people started carrying more regularly, just implements. That could be isolated implements like these [shows slide], or small sets of implements. These have been formed off site, selected and carried in. Rather than being produced on site, they’ve been carried across the landscape in their largely finished form. There’s also a great increase in the complexity in the transport strategies around the same time. The main thing, if you just notice here, there’s not much going on down here, and then there’s many more things going on after around 3000 years.

This was paralled by the change in the way people were using and carrying cores. In the early days, they were mostly worked before hand, that means the cores had been heavily used and many flakes removed before they were brought in to the site. Then they were principally prepared on site. That means only a small number of artefacts were removed and then they were always carried on. This says that the raw itself was needed to be conserved because people were using it across many sites, not removing much material and then always conserving it and carrying it on. That changes around 3000 years ago when the cores often have cortex on them, they’re much more heavily reduced and then they’re often discarded. That’s a signal that says to me that people at this point have a lot more access to new sources of raw materials, so they can be more liberal with their use of stone.

The other thing that happened around 3000 years ago as well as this explosion and access to high quality material is that people are heat treating it as well. If you heat the stone it improves the flakability of it, and this was why raw material quality was key. I had two high grade materials, and then a medium and then a low. Before around 3000 years, mostly people were using fairly low grade material. They just didn’t appear to have any access to the higher grade material that suddenly became possible around 3000 years ago.

This is just a bit of a summary then through times. We have that early period, there had been no backed artefacts, the raw material quality was low, cores were conserved, reduction was limited and there appeared to be limited access to new material. The middle period, we see this proliferation in the production of backed artefacts, access to high quality raw materials and an extensive reduction. In the most recent period, there were fewer stone working events, the raw material quality declines, but this doesn’t seem to be related to the issue of conservation as it was in the earliest period. We also have appearing in the most recent time, we have axes and grinding stones. They weren’t found in any of earlier periods.

This was the conclusion in my thesis that the palimpsest or the mixing of artefacts in different assemblages from different times and activities, it can be disentangled and then their constituent events can be sequenced. We can shift from either a whole of assemblage analysis or a level assemblage. We can shift to an analysis, which focuses on the event. The really good thing that that lets you do then is that it’s a much stronger platform from which to make inferences about the actions and the activities of the people.

I obviously couldn’t have done this without the help from many, many people. First and foremost is the land owners, both the Aboriginal and non-Aboriginal land owners, they all gave me access to their land, and without that, I simply couldn’t have done anything. I have several supervisors, and I think that Philip and Brad and Wilfred all might be here tonight. It was really great to have access to such quality advice during my thesis. I also had money from various places and Wilfred also was very kind in helping out with the field work. Of course, I also couldn’t have done it, with the more than 80 students who gave up their time to volunteer with me. I particularly couldn’t have done it without these people who just came back time and time again to excavate with me. In particular, Peter White, I don’t think he missed a single day in 18 months to volunteer his time. He’s a superb archaeologist and I thank him very much.

Thank you.

[applause]

ALLISON BYRNE: Would anybody like to ask any questions to Amy?

QUESTION: That was very good. I’ve got one observation, you talk about the vertical spread of these artefacts, which I agree with in this case. But you need to make it clear that not all sand bodies suffer from the same problem and that what you’ve demonstrated to give to people like us is that you have to take each site on its own basis. One of the two things is though, in a few of those sites you have hearths or the traces of hearths, but you also have tremendous surface movement of artefacts due to trampling. How do you reconcile the fact that the hearths have survived despite the fact that the artefacts have moved up and down, but the hearths have somehow survived?

AMY WAY: Yes, I have thought a lot about this. I think it’s probably just because of compactness of the charcoal. Just the firmness of the sediment at that point is just stopping it from shifting where the sand, where the hearths are, it’s just so much loser and so much more mobile that the artefacts are shifting there. But if you have any suggestions?

QUESTION: No, no. That now makes sense. Just one comment of course is that, what you’re really saying in the sense is that if you don’t analyse these kinds of sites the way you have done it, what’s the point of digging them?

AMY WAY: Yes, I think it’s much more limited what you can say in the end if you do either a whole of assemblage analysis or you do the divided analysis. I think your ability to then comment on what that meant for activities and actions in the past is very limited.

QUESTION: Could you just explain again what the word knapping means?

AMY WAY: Knapping, yes. To make stone artefacts, you have a core, which is the original piece of rock. You have the hammer stone, and you strike the core with the hammer stone and then the flakes come off the core. That action of producing flakes is knapping.

QUESTION: Thank you Amy, it’s Carol, that was so wonderful. It was so interesting having followed a bit of the work that you’ve done, but it was very exciting the conclusions I thought. I love that middle period where that – exotic is probably the wrong word – but that better quality stone appeared and was also being heat treated. How does that fit into the grander scheme of Australian prehistory and work that’s been done elsewhere? Do you think it reflects trade in stone to a material? I know that it’s not close to the 3000 years ago, isn’t close to the ethnographic contact period, but are there hints from the ethnographic literature or the historical literature about people in that area trading in stone or carrying stone with them? Because I thought that was just so interesting.

AMY WAY: Yes, well I had a good look around Lake George for sources of stone and couldn’t find any. I think, except for the quartz, the quartz is local, all the other material is coming in some form. The alignment with the eastern regional sequence is fairly good. The timing for that period of backed artefact proliferation and access to high quality materials and heat treatments and so forth, that aligns fairly well with most of the eastern region of Australia. The interesting thing is that we then see a decline in that backed artefact production in the last 1000 or so years. The difficulty then with making ethnographic correlations is that there’s just too much of a temporal gap then back to that 1500 year period.

QUESTION: It looked from your graphs as if the quartz featured was more dominant earlier on and it gave away to silcrete. It didn’t look as if quartz came back again later. How does this relate to other sequences that have been identified locally and a little bit further to the east?

AMY WAY: Yes, I think that what’s happened is that my sequences are so dominated by single knapping events, that it’s very hard to then comment on the other excavations at Lake George that saw the two operating in unison, or silcrete first and then quartz second. The issue is that all of the excavations are so small that at one point you capture someone knapping in quartz, then 3000 years later in silcrete and then 500 metres down the road is the reverse. I don’t think enough work has been done yet to say whether it’s a real trend or just these bursts of knapping that have been dominating the sequences.

QUESTION: Have you any idea why they knapped in those particular places?

AMY WAY: The lagoon site has this natural attraction to it. I think that what actually is happening is that people were knapping everywhere. It’s simply that, that’s where I happened to excavate that I found those artefacts. I think if I had excavated anywhere else I would have found a similar abundance of artefacts. I think the whole landscape’s full of stone artefacts and it’s just a sampling issue.

QUESTION: Do you have any idea of the numbers of people involved at each site?

AMY WAY: It’s so difficult. I steered away from making any predictions about population. The difficulty is that you just don’t know what the back story is. There might be one person knapping and they’ve got with them a group of 100 people who didn’t touch stone that day. Or it might be one person with a piece of stone and that’s all that there was to the story. I just thought there were too many unknowns to extrapolate from knapping events to numbers of people.

QUESTION: You didn’t find any burnt remains?

AMY WAY: Bone remains, no I didn’t find anything.

QUESTION: No burnt ashes or anything?

AMY WAY: Just a couple of hearths at the lagoons site. Yes, they were small camp fires really.

QUESTION: Hello Amy, it’s Peter. I’m just mind-boggled by your conjoining. It seems to me you’ve got this huge big pile of rock and you’re trying to do a three dimensional jigsaw puzzle. How do you actually do that? Can you tell us how much time you spent on some of those ones where there are lots and lots and lots of pieces all being conjoined together? I just can’t imagine how you can possibly have done it. What was involved?

AMY WAY: After I had identified which pieces were related geologically, I spread them all out on a table and they each sat on top of their bag, which had their identification labels on them. Then you would pick up one and you would walk around and you would compare it to every single other one on the table, be that 100 or 1000, and then if it joined together, that became a unit. Then you took it for a walk again until there could be nothing else stuck to it. Then you moved to the next one and it took six months. I didn’t go insane though, which I feared I might at the onset.

QUESTION: I was going to ask, we’ve got this present exhibition here in the Museum on songlines, what work is being on, or is anyone else in the community working on songlines in this area? You might go down to the coast or Ginninderra Falls or Namadgi [National Park] or what have you. Why haven’t we mentioned the word Lake Werriwa? To what extent are local Aboriginal tribes involved in all of this?

AMY WAY: I haven’t done any work on the songlines. I really steered away from doing anything that involved contemporary Aboriginal knowledge. I was very mindful of misappropriating knowledge or speaking on behalf of people. I just didn’t want to do that, so I didn’t go into any contemporary cultural knowledge, but I did consult heavily. I have a lot of registered Aboriginal parties on this project with whom I’ve been consulting since 2013 regularly. They’ve all had an opportunity to come out whenever I’m out there and if they want to talk about anything, they have done so with me.

QUESTION: Hi Amy, I’m Liz Williams an archaeologist and I know Phil and Marjorie and Wilfred and Carol. It might be a question for Phil and Marjorie – knowing that there’s the different lake levels of Lake George that you can see and people have worked on them – is there a way to access the really older parts of Lake George? More than 5000 years old, because you’ve gone down to the rubble level and that might be a –

AMY WAY: Yes, only at one of the sites did I hit the rubble though. At the others, we put augers into them, and the sand deposit went down for another couple of metres. It was really a question I had to ask earlier on was whether I was going to look widely or deeply. I went with widely, but there’s no reason why there wouldn’t be more evidence below that depth and going back in time. Philip and Marjorie worked on one of the oldest – so I wasn’t on the very old strand lines. One of the older strand lines up at the north near Collector Creek, they worked on found artefacts.

QUESTION: Hi, up here in the dark. You briefly mentioned the expectation maybe finding some sheep bones or whatever. Could you just talk a little bit more about why you didn’t find any organics and it wasn’t expected or what?

AMY WAY: Yes, I thought I’d find some from the last 20 or 50 years, but the reason is that the environmental conditions are just so harsh. The sun and the wind and the water just means they don’t survive any length of time, they just perish on the surface. They don’t enter the deposit.

QUESTION: [inaudible]

AMY WAY: No, there was nothing. I was also interested in hopefully getting some pollen or some phytoliths from that period, but not even phytoliths were surviving.

QUESTION: Thank you for the talk, I’m so glad you didn’t go mad [laughter]. I wouldn’t like to have a jigsaw with you. We know that the levels of Lake George go up and down, and so were any of your sites every covered in water do you think?

AMY WAY: Yes, they would have been at the really high, high stand around 30,000 years ago, they would have all been under water.

QUESTION: Does that mean that people, first of all, the level might have been lower and that’s why they worked quartz there; and when the level was higher, then they had to bring the stone in because they couldn’t find anything in the district?

AMY WAY: Yes, I definitely think that the changing environment has something to do with the different access to raw materials. One of the ways you get raw materials is cobbles in creeks. As the creeks fill and expand and then become smaller and get infilled, there must be changing exposure in what’s available. The quartz is fairly ubiquitous so it’s high and low. I think through all time periods, there would have been access to quartz. The other raw materials I can’t even say whether they’re local or and in any great distance, whether they were local and exposed at some point and now aren’t visible, whether they were never there to begin with. There was such a variety, 303 different types of stone that trade must be part of the story.

QUESTION: Thank you.

ALLISON BYRNE: Any more questions?

QUESTION: Very elementary question I guess, I’m just wondering about the strength, the hardness of these three different materials that you’ve been talking about. I’m wondering whether or not quartz is as good a raw material and lasts as long as say some of the subsequent materials?

AMY WAY: Yes, the strength has been tested not in my study but in other studies of these raw materials and quartz is harder. It’s harder to flake, but if you flake it and get a sharp edge, that sharp edge will be retained for longer. The silcrete and the cherts, they’re easier to knap but the edges are softer. Tools made on cherts and silcretes are easier to produce but they need to be replaced more frequently. Whereas the quartz you can have really nice quartz that fractures well and the way you would like it to fracture and then you can have much lower grade quartz, which will just shatter and will not give you any usable edges. All types of quartz are present in the Lake George landscape.

QUESTION: Did you find when doing all this work with the depths of the soil that you got – I was watching as you were talking and you seem to have been through your study in contact with a kind of antiquity of the ancient sacred culture you – you know like when you’re working with the quartz and the crystals and all those sacred and precious gems did you get a sense of that from doing this study?

AMY WAY: I did often wonder if the person who made these artefacts could see me now trying to put them all back together, what would they think? I think they would think I was insane [laughter]. It is nice to remember the humanity of the subject.

QUESTION: How long in this – 30,000, 50,000 – how long did the Aboriginal people live in this area?

AMY WAY: I only went back to about 5000 years but I do not think that, that’s a comment on the antiquity. I think because we have sites all around at Bass Point and Burrill Lake and up in the highlands with Josephine Flood’s work, all going back twenty odd thousand years; and we now have the date pushed back in northern Australia as 65,000 and Tasmania, where we’ve got 30 – and 40,000. It had to be much greater than 5000 years. The actual answer I don’t know. I’d be happy to say it’s at least 30 or 40 [thousand years].

ALLISON BYRNE: Thank you very much Amy. It was a very rewarding talk.