Was it like this the chert at Melsvik was quarried in the Stone Age? Archaeologist Lars Julsrud of Alta Museum applying brute force with a hammerstone (without touching the ancient surface!), his colleague Martin Hykkerud perhaps taking a more sceptic stance. Photo: Per Storemyr

Update: See this blog post for confirmation of firesetting at Melsvik!

A few weeks ago I had the great opportunity to take a look at the large, recently discovered Stone Age chert quarries at Melsvik, some 10 km to the west of Alta in Northern Norway. Very regrettably, the construction of a new highway will soon cut right through this piece of invaluable archaeology – and thus destroy large parts of it. This is the reason why there is an amazing archaeological excavation going on, by the University Museum at Tromsø. Excavations started last summer and will continue next season. On looking at the work areas, and the quarries themselves, I started wondering how on earth this extremely hard rock was actually extracted. Was it by applying brute force using heavy stone pounders, or could firesetting have been used?

Melsvik is the largest of five known Stone Age chert quarries close to the Alta Fjord in the region west of Alta (Finnmark county). According to Anja Roth Niemi, leader of the excavation project, the quarry seems to have been in use for quite some time, starting in the Early Mesolithic, not long after the big Ice Age ice sheet retreated up north and the sea level was much higher than today. The Mesolithic in Finnmark dates to about 9.500 to 5.000 BC and is often termed the “Older Stone Age”.

Geological map of the Alta region, with chert quarries (red) and rock art sites (blue). The most important geological formations are; in light green – metagreywacke/metasandsandstone; in brown – metabasalt and tuff; in blue – dolomite with occasional chert. Map by Per Storemyr, made with Google Earth, geology based on http://geo.ngu.no/kart/berggrunn

The chert deposits at Melsvik mostly occur as beds associated with dolomite and possibly various shales and sandstones within the Precambrian, so-called Kvenvik Greenstone Formation. The formation has been metamorphosed and subjected to folding and considerable tectonic stress throughout the geologic history. This implies that the chert is of very uneven quality, with pronounced cleavage, and full of cracks and fissures. Colour range from white to dark greyish and purplish.

Chert above lighter dolomite at Melsvik. This is one of the quarried spots. Photo: Per Storemyr

Thin, folded beds of chert and dolomite by the Melsvik quarries. Photo: Per Storemyr

Generally, chert consists of extremely fine-grained (microcrystalline, cryptocrystalline) quartz deposited with the aid of organisms or developed by replacement (silicification) of other rocks. It may occur as nodules in limestone or as thicker and thinner sedimentary beds. Nodules within chalk and limestone are often called flint, but there is hardly a good distinction between flint and chert and for many people these notions are synonymous. For others, chert may refer to flint of inferior quality when used for tools.

Chert is quite rare in the Norwegian bedrock, and the Melsvik quarry is of importance in order to get a grip on Stone Age tool procurement strategies. Archaeologist Bryan C. Hood investigated the four other, smaller quarries around Alta the 1980s, concluding that the tools that were made may have ended up in the far inland to the south – as well as at the outer coast – and thus they were part of regional exchange systems. Hood also speculated whether there could have been a social and ideological link between chert procurement and the making of Stone Age rock art at the bottom of the Alta Fjord (see list of Hood’s work below). For it is here the famous World Heritage Alta rock art is situated.

The main extraction spots at Melsvik are centred around a hillock 60-65 metres above the current sea level. The excavated flats just below the quarries, at both sides of the hillock, were clearly used as work areas – for primary reduction of irregular blocks taken from the quarries, production of cores and further reduction to at least half-finished scrapers and other tools. Importantly, the hillock would have formed a promontory just by the fjord in the Early Mesolithic, implying that the work areas would have been literally at the beach.

Excavated quarry spot along the hillock and tool production areas in front. Photo: Per Storemyr

According to blog posts by the excavation project, there seems to be distinctions between work areas, or tool production sites, and habitation areas. Thus, people lived here, at least seasonally when the work went on. Moreover, there is an occurrence of black chert tools that may possibly not originate from the Melsvik quarries, but rather from some of the quarries that Hood investigated in the 1980s. This would mean that people brought their tools to Melsvik, for use and perhaps replacement.

Debitage and small blocks from the Melsvik work areas showing the chert’s range of colours. Photo: Per Storemyr

Arrow heads from dark chert, found during excavations of habitation sites (not Melsvik), now exhibited at Alta Museum. Photo: Per Storemyr

Until now two larger extraction spots in chert bedrock have been found around the central hillock. Also, many heavy, naturally shaped pounders or hammerstones were strewn along the hillock, implying that they certainly would have been used for the work. But in what kind of quarry work were they actually employed? There can be little doubt that they were used to break off available corners and edges as the work started, in particular at places where the chert is almost “draped” around underlying dolomite. Subsequently, heavy blows of hammerstones would also have been applied for primary reduction of the blocks that were achieved.

Hammerstone with clear marks of use in the Melsvik chert quarry. Photo: Per Storemyr

Probable traces of “stepped extraction” (left), taking advantage of the pronounced, sub-vertical cleavage to produce slabs. The image also shows the quality of the chert in the most massive, quarried deposits, with alternating light and darker, almost purplish (jasper-like) layers Photo: Per Storemyr

In addition there is evidence of the making of “terraces” into solid bedrock. Here it is much more difficult to figure out techniques employed since the rock is hard and usually relatively smooth on the surface, with few or no remaining corners to knock off. But on closer inspection it can be seen that the quarriers may have taken advantage of the pronounced cleavage of the rock. They may have produced thin slabs or plates of chert by careful direct or indirect blows with hammerstones along a line perpendicular to the cleavage. A fracture will thus have opened and the slab probably automatically loosened along the cleavage plane. In the end a pattern looking like fine steps emerges from such quarrying – a pattern that is known from other Stone Age quarries, for example in a quartz deposit at the Western Isles of Scotland, investigated by Torben Bjarke Ballin, who calls the technique “stepped extraction”.

Bigger steps at Melsvik, perhaps evidence of a variety of the “stepped extraction” technique. Photo: Per Storemyr

It is possible that varieties of this technique were employed also to extract thicker slabs and even sub-rectangular blocks at Melsvik. This is because there are traces of bigger steps at the “bottom” of one of the terraces. Of course, it is not unthinkable that bone wedges were used in the work, in addition to hammerstones.

Yet, it does not seem to be possible to explain all the quarrying at Melsvik with blowing off corners and edges with hammerstones, and the use of the “stepped extraction” technique. Another careful look may give additional hints. For at the back of one of the terraces there are smooth, concave cut marks – marks that look familiar to those who have worked with firesetting, either in old mining or in stone production. Such marks are very hard to explain without introducing the use of fire. But wouldn’t the chert become damaged or even “explode” when subject to heat? Yes, there is always a danger that quartz-rich stone may deteriorate upon strong heating. But it is usually not necessary to heat much in order to create strong temperature gradients and enough shear stress to break loose stone. With small, controlled fires it is even possible to create good fractures for extracting huge blocks of granite, like it is still done today in India. And in Ancient Egypt extremely quartz rich rocks (silicified sandstone) were routinely quarried by the use of fire setting for the production of grinding stone.

Possible traces of firesetting at the Melsvik chert quarry: Concave (and some convex) scars on the vertical face at the back of a quarried terrace. Photo: Per Storemyr

Consequently, fires lit along the vertical face at the back of the terraces at Melsvik may very well have been able to break up flakes and small blocks which could then be reduced to tools. It may have been necessary, though, to apply careful knocking with hammerstones or even bone wedges to aid in the final loosening of such flakes and blocks from the cracked-up bedrock. According to excavation leader Anja Roth Niemi no charcoal has been found at such spots, but this may be explained by poor preservation conditions and the use of just minor fires (or it has just not been found yet).

At several sub-horizontal spots on the Melsvik quarry hillock are also concave and convex scars in the chert that would be explicable with the use of fire. Maybe such scars are evidence of testing the chert rather than serious extraction attempts. The scars are only superficial and further work at such spots seems to have been abandoned.

Possible traces of firesetting at the Melsvik chert quarry: Convex, superficial scar at sub-horizontal surfaces. Photo: Per Storemyr

Globally, firesetting in ancient tool quarries is, of course, nothing new. According to an overview by Weisgerber and Willies it is known since the Palaeolithic, but there is often much controversy surrounding its application since it may be difficult to find good evidence in the form of the characteristic, concave scars, charcoal or direct scientific proof, i.e. changes in mineralogy, structure/texture or magnetic properties of the rock in question. There is one well-known example in Norway, in the Mesolithic/Neolithic greenstone axe quarries studied by Bruen Olsen and Alsaker at Hespriholmen off the south-western coast (more info and video here). This is a quarry that has much in common with the famous Neolithic “axe factories” at Great Langdale in Cumbria (England), where there also seems to be evidence of firesetting, reported by Bradley and Edmonds. But indications are also reported from elsewhere in Norway, seemingly also from the far north, in the Neolithic quartz quarries at Skjaanes in East Finnmark. And in South Finland there is a Mesolithic quartz quarry at Kopinkallio where evidence of firesetting has been found from characteristic microfracturing. This work of geologist Kari A. Kinnunen has also enabled a rough estimate of firing temperature, which obviously did not exceed 480 degrees centigrade.

Traces of firesetting in the Neolithic axe quarries at Great Langdale in Cumbria. Photo: Per Storemyr

In all these examples firesetting went hand-in-hand with the use of hammerstones, to loosen cracked-up bedrock, for reduction purposes and quite certainly, whenever possible, also for extraction without firesetting.

The Melsvik quarry may become another addition to the list of Stone Age quarrying by firesetting. Excavation and other investigations will continue by Tromsø University Museum next season; maybe they will help in getting a better grip on the hypotheses presented here. And perhaps simple experiments with small fires could be a way to explain the extraction techniques – to get a fuller understanding of the chaîne opératoire – at this wonderful site. Before large parts of it will fall victim to the new highway and our modern crave for rapid transportation.

Thanks to excavation leader Anja Roth Niemi of the University Museum at Tromsø for information, and archaeologists Martin Hykkerud, Lars Julsrud and Are Skarstein Kolberg of Alta Museum for guiding me to the Melsvik chert quarries.

Sources not linked to in the text:

Publications by Bryan C. Hood:

• 1988: Sacred Pictures, Sacred Rocks: Ideological and Social Space in the North Norwegian Stone Age. Norwegian Archaeological Review, 21 (2): 65-84
• 1992: Chert Sources and Distribution Patterns in the Stone Age of West Finnmark, North Norway: A Preliminary View. Acta Borealia, 9 (2): 69-84.
• 1992: Prehistoric Foragers of the North Atlantic: Perspectives on Lithic Procurement and Social Complexity in the North Norwegian Stone Age and the Labrador Maritime Archaic. Unpublished Ph.D. Dissertation, Department of Anthropology, University of Massachusetts, Amherst.
• 1994: Lithic Procurement and Technological Organization in the Stone Age of West Finnmark, North Norway. Norwegian Archaeological Review 27 (2): 65-85.
• 1999: Chertbrudd fra steinalderen i Alta. Ottar, 225: 25-32.

Map