In 1952 the Danish archaeologists Johannes Iversen, Jörgen Troels-Smith, and Svend Jörgensen conducted an experiment to better understand how stone axes functioned in Neolithic Europe.  The team replicated wood handles based on the Sigerslev axe—an ancient axe found in a Danish bog—and hafted them with archaeological flint axes borrowed from the National Museum in Copenhagen.  They gave these reconstructions to professional lumberjacks from Northern Europe—experts in using steel axes—and monitored their efficiency in chopping through beech trees.  The lumberjacks swung these experimental stone tools in the same way as their steel axes, and within five minutes all four stone axes, plus the handles, were broken to pieces.

At the same time, on the opposite side of the world, Indigenous people in the highlands of New Guinea were using their stone adzes and axes to cut back forest growth in savannah landscapes they had been managing for thousands of years.  In the lower-elevation forests, horticulturalists prepared single-use gardens by clearing rainforest trees—entirely with stone axes.  

Stone axes and adzes were used daily across this vast region to cut garden stakes and firewood, construct and maintain houses and buildings, and fashion and repair wood tools, canoes, and elaborate carvings.

Although many of us may imagine that stone axes were crude and only nominally functional, the story from New Guinea shows us that this is clearly untrue.

Felling a Tree with a Stone Axe

Using a stone axe to cut down a tree or sapling is very different to how we might use a steel axe.  Steel is unbreakable, so we can swing a steel axe very forcefully.  The danger is that the handle will snap, not that the steel will fail.  

The usual Western method of using a steel axe is to deliver powerful chops into the wood at a steep angle from the top and bottom sides of the cut, creating a V-shaped notch, until the tree weakens and falls.  The angled blows slice across the wood grain, and as the steel axe progresses through the wood, the wedge-shaped axe head forces open the slice and snaps off the thin chip.  The cut is incrementally deepened by chopping first from the top and then the bottom of the notch.

With stone axes, the same basic concept was at play—chips were wedged from the wood by angled  blows that slice across the wood grain.  But instead of the chips detaching automatically, as occurs with a steel axe, they accumulated at the bottom of the angled cut where they were still attached to the tree.  

Periodically the New Guinea stone ax-wielder reached in with their hand and wrenched the chips free, breaking them manually across the grain.  Once the cut was cleared, more chopping followed, and again the partially-detached chips accumulated and were wrenched free.  In this way, the cut was incrementally deepened until the tree fell.  

Because the notch was one-sided instead of V-shaped, fewer blows were delivered in cutting down a tree this way.  The resulting stone-cut stump, with its frayed end, has been compared to a giant shaving brush.

Rather than making the cut at about waist height, as is typical with steel axes, the stone axe cuts were made at about chest height (if the person was standing).  This is so the axe-wielder didn’t have to repeatedly bend over to break off the chips, which was hard on the back.  It resulted in a relatively tall stump.  

Also, rather than dropping the axe to reach in and wrench away the chips, the axe was instead rested on or hooked across the shoulder.  The hafted stone axe or adze was carried like this throughout the day to free both hands for other tasks.  There are many photographs from New Guinea showing people with stone axes or adzes hooked across the shoulder.  Nowadays steel axes are often carried in the same way in New Guinea.

Stone adzes were used in a similar way as stone axes, but usually on smaller-diameter wood.  A blow from a large mortise-hafted axe caused a small sapling to flex, absorbing much of the force of the blow.  Hence these larger axes were used on trees greater than about 10 cm in diameter.  Smaller adzes weigh less and avoid the inertia problem, and the speed of the swing was also greater.  This helped drive the cutting edge through the wood.  In contrast to the large mortise-hafted axes, smaller adzes were used on all sizes of trees, including very large ones.

The archaeologists Pierre and Anne-Marie Pétrequin described the link between the stone tool type and the vegetation that needed clearing in New Guinea.  Smaller adzes were used most often to lop off shoots from stumps, cut small saplings, and to shape various wood tools and building materials.  They were the most-used stone tool by savannah farmers in the highlands.  Larger axes were used to clear large trees from garden plots and were used more frequently by horticulturalists in primary forests or mature secondary forests.  Large axes were used by people in both regions to split tree trunks into firewood and planks for buildings.

Archaeologists and anthropologists working with New Guinea people discovered that it took about the same amount of time to fell a tree with a stone axe as with a stone adze.  A 10 cm diameter hardwood tree could be cut down in about 4 minutes on average, and a 30 cm diameter one in about 20 minutes.  Softwood trees generally took about 30% longer to cut down than hardwood trees.  Compared to steel axes, felling small trees with stone axes took from 2-2.5 times longer.  This increased to 4-4.5 times longer for large trees.  

Cutting times with stone tools  increased exponentially for trees over 30 cm in diameter, as the notch needed to be progressively expanded in length to reach the centre of the tree.  For this reason, very large trees were usually girdled to kill them and were left standing to be burnt later.  Smaller trees were often girdled and left standing and harvested for firewood once they dried.

The Stone Axe Swing

Stone axes were of great value to people in New Guinea, and much social capital was spent in procuring them.  Unlike steel, stone can break, and the user was highly motivated to avoid this.  One adaptation to stone can be seen in the way the axe was swung against the tree, so this is worth looking at in greater detail.  

Western lumberjacks tend to swing a steel axe fairly directly into the side of the tree, and Western carpenters swing steel adzes in a flatter, hoe-like angle.  The adze-swing is closer to parallel to the front-facing part of the wood—more of a ‘skimming’ stroke.  

People in New Guinea swung both their stone axes and adzes at the flatter, skimming stroke angle.  Also, stone axes and adzes were swung downwards from the shoulder with less than maximum force.

This differs dramatically to how we use steel axes in the West.  Steel axes are  swung through hip and upper body rotation to drive the steel as deeply as possible into the wood.  Indeed, the reason the Danish experimental team broke so many of their stone axes is because the lumberjacks swung them forcefully from the hip, as they would a steel axe.  

Among some New Guinea groups a ‘cross-handed’ method was used to hold the handle, with the dominant hand below the non-dominant hand.  This is the reverse of the usual Western holding method.

What might explain the difference in stroke angle  between stone axes and steel ones?  It seems partly due to the forces acting on the stone’s cutting edge when struck against the tree.  Because stone is brittle and relatively fragile, it is crucial that the cutting edge meets the wood in relatively close alignment with the forces conveyed by the direction of the swing.  Problems arise if the force trajectory of the swing is too far out of alignment with the direction the cutting edge is taking through the wood.  

Conversely, a small amount of misalignment is crucial to provide the angle of attack necessary to drive the edge into the wood and across the grain—perfect parallel alignment would cause the axe to bounce off the surface, with no purchase into the wood at all.  If the misalignment is too great—or if the cutting edge is too sharp relative to the strength of the blow—a bend-initiated flake may be torn from the edge of the tool.  

Edge-flaking was a common occurrence in New Guinea that required pauses in the work so that a new edge could be ground-on using a portable abrading tool.  If the misalignment was particularly egregious, the bending stresses might cause the stone axe to snap into two pieces somewhere near the middle.  These types of damage are commonly seen world-wide on axes and adzes in the archaeological record.

Bending Stresses on Stone Axes—A New Zealand Study

In 1977 the New Zealand archaeologist Simon Best performed some experiments into the function of Maori stone adzes.  He borrowed three adzes from the Auckland Museum and lashed them onto modern handles to test how well they worked.  Stone adzes, as defined by archaeologists, have the cutting edge offset towards one face from the centreline of the stone.  Best noticed that the edges of Maori adzes are offset by various amounts, and his experiments were designed to reconstruct how the different edge-offsets on adzes might be related to how they were used.

In addition to the experiments, Best cut profiles of Maoris adzes out of perspex to examine how stresses are distributed through the stone during use.  He applied a static mechanical load to induce stress to the edges of three perspex models—one with a centred edge (as seen on stone axes), one with the edge offset entirely to one face (as seen on stone adzes), and one with the edge offset in an intermediate position between these two.  Mechanical stress was induced along an axis from the model’s cutting edge to the butt.  By shining cross-polarised light through the models, and viewing the results through a polariscope, radiating isobars of stress intensity became visible.

With the centred edge, the axis of stress was distributed down the length of the model, in a direct line from the cutting edge to the butt.  The greatest stress intensity was along this axis, ‘stiffening’ the stone, and making it resistant to fracture.  In contrast, the greatest stress on offset edges was deflected towards the bevelled faces of the models, rather than towards the butts.  The degree of edge-offset increased the intensity of the bending stress towards the face of the adze.  We can infer from this that, all else being equal, the greater the degree of offset, the more vulnerable the stone edge is to edge-damage through flaking, or catastrophic failure through breakage across the middle.

Best applied an idealised mechanical load to his perspex models.  In actual tool use, the direction of mechanical load depends on the orientation of the swing relative to the wood—the mechanical effect of the edge-offset potentially varies according to the orientation of the swing.  

The details of this complex interaction have yet to be explored in engineering terms, but it is safe to say that even the most skilful stone axe-user sometimes misdirected their swing, and therefore the mechanical load.  A misdirected blow could potentially intensify stress fields towards the face of the adze or axe—sometimes further exacerbated by the degree of edge-offset—resulting in damage to the stone.  So while edge-offset could control the stress fields through the stone, the success or failure of a particular stroke depended mostly on the skill of the tool user. 

In New Guinea, edges ranged from centred to almost completely offset, regardless of whether the tool was hafted as an axe (with the edge parallel to the handle) or an adze (with the edge at 90 degrees to the handle).  Adjustments were made to the swing to accomodate the position of the cutting edge and angle of attack.  These technical decisions were also influenced by other aspects of stone axe design, particular the nature of the edge itself. 

Stone Axe Edge Design

Aside from skilful handling of the hafted tool, and the degree of offset of the cutting edge, another strategy to reduce the frequency of failures can be seen in the edge’s shape.  The corners at either end of the cutting edge are vulnerable because if they are too sharp they can dig into the wood and ‘catch’, exacerbating the bending stresses as inertia carries the axe forward.  When the stone fails, a flake can initiate from the corner that propagates across the axe’s face or up its side, often causing extensive damage.  A response was to eliminate the corners by shaping the edge into a curve rather than making it straight.  

Also, work adzes were rarely manufactured with razor-sharp edges.  Such an edge has a tendency to bite too deeply into the wood; if the tool catches in the cut, this can heighten the amplitude of the bending stress induced by the inertia of the swing.  Because of this, razor-sharp edges were not always favoured, and the edges on some stone work axes and adzes can appear quite dull by our modern conventions.  Of course, although a relatively dull edge may be more resistant to flaking by bending, it is also far less effective at cutting through the wood grain.   Design decisions involved various tradeoffs between efficient cutting and preservation of the stone.

Offsetting the Angle of Attack

To better deal with the flatter skimming stroke, New Guinea stone axes were sometimes mounted with the edge rotated at an angle to the handle rather than parallel to it.  This improved the ‘attack angle’ of the cutting edge against the wood in the skimming stroke, and also had the benefit of helping keep the handle clear of the wood face.  

The attack angle of stone adzes was adjusted as well, but the rotation took the edge out of an exact right-angle to the handle.  There is a continuum between the amount of skewing of cutting edges relative to the handles, resulting in some tools described as axe-adzes.  In some regions the stone was mounted into wood sleeve which was tied onto the elbow handle; the sleeve could be rotated to adjust the attack angle to match the woodworking task.

The French photographer Tony Saulnier described a first encounter in 1959 with a group of New Guinea men in the highland foothills of western Indonesia.  The men were investigating the various objects the Europeans carried, and ‘clumsily and apprehensively they tried out an iron axe, but it slipped off the trunk each time.’  This is probably because they were attempting to cut the wood using their skimming stone axe stroke, and the edge of the steel axe lacked the proper attack angle to make it work effectively when swung in this way.

Axes and Adzes Were Used for Many Things

Both axes and adzes were used to process materials other than wood—for instance, in disarticulating pigs, separating pandanus nuts, or pounding sago palm.  Also, both axes and adzes were used extensively for delicate wood scraping and carving.  The tools were held close to the cutting edge and handled more like Westerners  might use a steel chisel or pocket knife.  Stone adzes, in particular, were the go-to stone tools for people going about their daily lives in New Guinea.

One aspect of anthropology is to study the knowledge and know-how of First Nations people who have fine-tuned the design of their tools, and how to use them, over hundreds of generations of research and development.  This knowledge can be used by archaeologists as analogies for generating hypotheses about how forgotten stone technologies may have worked in the deep past in other parts of the world.  Archaeologists call this ‘middle range research’ as it provides a way to link the static artefacts we find in the ground to the dynamic practices that produced them. Although steel tools have since replaced stone ones across most of New Guinea, knowledge about how they were used, and their symbolic roles in society, live on through traditional stories.

References

Best, S. (1977). The Maori adze: An explanation for change. The Journal of the Polynesian Society, 86(3), 307–337.

Blackwood, B. (1950). The Technology of a Modern Stone Age People in New Guinea. Pitt Rivers Museum Occasional Papers on Technology 3. Oxford: Oxford University Press.

Gardner, R., & Heider, K. G. (1968). Gardens of War: Life and Death in the New Guinea Stone Age. New York: Random House.

Godelier, M., & Garanger, J. (1973). Outils de pierre, outils d’acier chez les Baruya de Nouvelle-Guinée. L’Homme, 13(3), 187–220. https://doi.org/10.3406/hom.1973.367374

Hardy, K., & Sillitoe, P. (2003). Material perspectives: Stone tool use and material culture in Papua New Guinea. Internet Archaeology. https://intarch.ac.uk/journal/issue14/hardy_toc.html

Jørgensen, S. (1985). Tree Felling with Original Neolithic Fint-Axes in Draved Wood: Report on the Experiments in 1952-1954. Copenhagen: National Museum of Denmark.

Noble, G. (2017). Woodland in the Neolithic of Northern Europe: The Forest as Ancestor. Cambridge: Cambridge University Press.

Pétrequin, P., & Pétrequin, A.-M. (2020). Ecology of a Tool: The Ground Stone Axes of Irian Jaya (Indonesia) (Revised and expanded edition.). Oxford: Oxbow Books.

Saulnier, T., & Bisiaux, M. (1963). Head Hunters of Papua. New York: Crown.

Townsend, W. H. (1969). Stone and steel tool use in a New Guinea Society. Ethnology, 8(2), 199. https://doi.org/10.2307/3772981