Experimental production of pebble tools


Content of this post:
1. Experiments on  raw material samples from the Belgian Kempen area, similar to the raw materials used in the artifact -assemblages ( dec 2011) to be updated
2. Some experiments on quartz  from the river Tarn (2010 France
3. Experimental making of small pebble tools from the river Moselle (March 2010. France) experiments by Hervé Beaudouin, archaeologist from Lorraine ( weblog: Préhistoire en Lorraine)
4. Some experiments on a very limited pebble clast from the surface at Niels -bij  -As : the production of very small tools

1. Experiments on  raw material samples from the Belgian Kempen area, similar to the raw materials used in the artifact -assemblages

Small hints on artifacts provide information about the  use of the tool, like the small fracture, visible ( blue point)  at the tip of the point of this assumed ''pic '. But is thus really the truth? How can we distinguish  natural fractures  from use wear of intentional retouch? Experiments are helpful, especially in understanding techniques applied on local raw materials.

For the experiments, the raw materials have been collected from the find sports, to have a better comparison between experimental made tools and prehistoric  tools.

Tool properties at the main site cluster of the MA - project are the best studied part, so this is the main starting point for the man-made prehistoric artifacts control group :
( characteristics of the artifact group,  found at MA4:)

1. brown, mostly egg- shaped, fine grained pebbles were chosen, without  inclusions
2. the angle of the flake removal is different , but the low angles prevail in the assemblage
3. the number of flake removals is  between 1 and 7 /8 depending on the size of the pebble and the desired function.
4. though no anvil is found ( yet), it is most likely tools were made on a flat anvil, using hard hammer technique ( looking at the scars on almost every pebble tool)
5. the choice of raw material is the determinant factor in the final tool design: round pebbles show different flaking patterns ( rounded, shell shaped) compared to quartzite blocks with quartz veins, red sandstones, conglomerates, etc.
6. micro -choppers and chopping tools are present in the assemblages. This requires a  more sophisticated precise technique
7. Angles of processed edges usually below 75 degrees

Refit of naturally broken pebble

Naturally broken pebbles show steep angles and a lack of bulb complexes and  real impact points.
Naturally broken pebbles, like in the images above demonstrate totally different breaking/ fracture patterns on the pebbles. For example the edge angle in naturally broken pebbles often lies between 80 and 90 degrees.  These examples are helpful in understanding local debitage techniques.

Some simple experiments showed some important facts about the technique of pebble tool production:
1) free hand flaking almost impossible. it's like you take a piece of metal and slam it with a hammer, painful for the hand; hammer- anvil technique is very fruitful
2) hammer size 2- 3 times bigger than stone that is to be flaked
3) shape of hammer stone less important than weight, but round pebbles preferred
4. choppers are easily made when placed in the right angle on the anvil (low angle)
5. decortication strikes on the quartzite pebble surface only are produced under high angles ( "steep retouch") and also could be interpret as 'mistakes', like steps and hinges.
6. Tool production in this manner gives no wave patterns, impact points not always visible, no expressive bulb, or even "inversed bulbs"(with the strike the whole bulb section has splintered).

Quick tool making at the Helle brook at Eupen, Belgium on unselected pebbles from the brook bed; simple choppers were made in less than one minute, but a pre - shaped rounded edge was required to make them. During winter time, the pebbles are very cold, they are hardly to hold and therefore are very difficult to be processed.
The Helle, (Hillbach)small brook  near Eupen in Belgium, Hautes Fagnes/ Höhes Venn

Small gravel  beach at the present brook Helle ( Eupen), showing different  dimensions of gravels and different types of gravels together.Early man could just pick up pre- selected  pebbles to make tools. 

Before the striking (unsharp picture, but no other available)

After the striking, the cutting edge is rather sharp. The percussion impact lines are slightly visible in the quartzite surface of the flake removal, the bulb- complex  is visible at the left.
Quartzite pebbles ( granular) do not show any percussion waves or bulbs when stroke.
Percussion impact was at blue arrow's point, leaving a useful quartzite flake and a useless core. The flaking could be regarded as tool preparation, to become a secondary striking platform. when tool makers leave the core like this ( like often occurs) it is clear that the flake is desired, or the result of flaking was disappointing, so they discarded it.

Start of this project: januar/ februar 2012
End of project : autumn 2012

Main questions in the experiments
Pebbles might get naturally broken, giving patterns that are similar or different from man-made, prehistoric pebble tools. Characteristics of man -made tools are frequently described ( Bordes 1961, Whiteacker 2003, Hardaker at earthmeasure ) so this is not the objective of the experiments.
Early hominids had the availability of exposed  river gravels, making their tools from these gravels. In the Kempen- area two types of gravels were used: the late Elsterian fine- very fine gravels containing black flint and the much more crude gravels originating from early Elsterian/ late Cromerian period as in the original stratigraphy this gravel type lies directly below it. Both gravel types have their limits and possibilities, depending on purity, grain size, dimensions and typical predetermined fracture patterns.
Main purpose of the experiments is to find answers to the following questions:

1) Using the anvil technique, what angles have been used?
2) What differences are there using different types of hammer stones?
3) How did they make the quartzite micro- choppers/ micro- chopping tools?
4) How did they make the bifacial retouched small rounded flint pebbles
5) What are the production methods of flakes  without prolonged bulbs?
6) How does the deliberate stepped fraction work?
7) How can we distinguish naturally patterns from man- made patterns, to be sure we deal with man-made artifacts on the local gravel types?

Local raw materials to use in the experiments ( gravel)

1. round to oval brown- brown grey colored pebbles, fine grained structure
2. grey quartzites containing veins of quartz
3. brownish sandstones
4. dark grey clastic shales
5. red sandstones
6. phyllite
7. quartz
8. conglomerates

last update  January 30, 2012
(To be continued ! )


2. Some experiments on quartz from the river Tarn (2010 France
The author busy with the experiment at the Tarn river (F)
Pebble tools, made from stones from the river gravel bed could be made from quartz (1). Such tools are not always easy to recognise for the (amateur -) archaeologist. One way to find out the processing and the limits of quartz used for tool production, is to make them experimental.  Reports have been made on experimental  processed quartz tools with classification (2) but they are not always easy to read.

Choice of raw material
In certain regions of Europe  good quality flint to make tools is not available, so other materials were used for tool production. In big parts of   Southern France  and South- East Europe quartz and quartzite was an alternative ressource (3).In the river Tarn in the French region of the Cévennes, quartz is availible ( 4)  as I could determine in very low percentage, I estimate less than 1 % ( main part is composed of granite).

The main objective to make some tools from selected quartz from the present river terrace gravelbed was to see how  it would be to make such tools, and to get examples of quartz with processed edges. Thus, I could compare prehistoric tools with experimental made tools. Another objective was, to see how easy it would be to make a disposable tool from a quartz, since this material was the only raw material found in the riverbed  gravel to make a tool with a cutting edge. Cobbles of granite were amply present in the right ( rounded) sizes and shapes, to be used as a hammerstone.

The experiment ( 20 – 06- 2011, 21 -06 2011 and 24 -06 2011)
(a) Searching the proper materials and a working spot
At the channel edge of the river Tarn, north  from the community of Cocures I found two different spots to search for quartz. It took some time to find the right quartz to be processed. Lots of pieces of quartz were damaged somehow by frost ( frostridges) and these stones were not suitable, like with common flint. I found several pieces of nice white – dull quartz, sometimes with brown-yellow patinas om it. Rounded quartz was absent.  To find the hammerstones was very easy, rounded cobbles of  all practical sizes were available. It was also  easy to use a large flat rock at the riverside to serve as an anvil.
(b) The experiments
I took a piece of quartz with dimensions 9x7x4 cm and striked it with a piece of (by the river-) rolled granite on the edges to get a handaxe type form( cordiform /pointed oblong). I realised, the blank* could only provide a pointed tool without cutting edges. After strong strikes on the edges, the debitage showed flakes, some of them were useful to get further adaptation ( dependent on the thickness, e.g. to serve as a possible mini- quartz points ). After this free- hand striking, I placed the object on a flat rock and carefully striked the edges to get the desired shape. I made a ( very) sharp point at one side. The result was a possible useful  quartz handtool, with a solid, handshaped, ergonomic grip, with at the other side a very sharp point, to drill, press, etc. The tool was made  in ten minutes.
Front side of the pointed quartz
Back side of the pointed quartz 

A second piece of triangular shaped quartz  ( ca. 7 x 7 x 4 cm) was only adapted to get two (convergent) sharp edge's to become a handtool such as to clean skins,scrape meat from bones , etc. The edges were sharpened by hammer-anvil technique, striking and pressing at the backside of the edge. The edge became rounded, but sharp, at least sharp enough to be a possible scraping tool.

A third tool was made from a quartz cobble with patina's ( dimensions ca 12 x 8 x 3 cm). The original shape was good enough to try to make a rough handaxe with point and one sharpened edge. This was partially successful. The rough shape was ready in 10 minutes.
A fourth tool was made by hard striking with a round granite cobble on a larger quartz piece,  a large blade- like flake was produced in this way and later it was  secondary retouched. This tool was very sharp, easy to use and easy to be secondary retouched.

It does not look nice, but it is a quartz flake. Obvious not taken form a rounded quartz pebble. The cutting edge is very sharp.

Main remarks in ad- hoc processing the quartz to become a ( disposable) (hand-) tool:

  • Frost ridges and other invisible damages are ( as expected) important for the quality of the quartz; the moe white/ translucent the quartz, the better quality
  • Small pieces of quartz ( debitage waste) are frequently good enough for further adaptation, especially when they are thick ( ca 2-3 cm)Lots of small pieces however are not re-usable, so handtool production seems logic (4)
  • (Rounded) granites are very suitable as hammerstones
  • Edges are not difficult to sharpen, and do not break easy
  • Patterns of striking are recognisable on the tools edges as small, overlaying edges of  shallow pits
  • Secundary retouche with smaller hammerstones on pre- adapted edges is useless
  • Reducing in size is easier than reducing thickness.
  • Retouch on quartz artefacts is visible as rounded, towards the edge's point processed adaptation, made by striking and pressing techniques
  • Processed quartz – as a tool- is visible by a 3D adaptation, where more than one part of the quartz has been adapted , so the adapted side has been – re- adapted in another direction
  • Pointed retouchoirs of granite could help to get better notches or rounded forms on edges
  • rolled quartzite was absent in all experiments at the Tarn
The pointed tool and the Tarn river

* blank: a  raw shape of the quartz block, non -defined, but denouncing the limits of the process

(1) Quartz must not be confused with quartzite, which is metamorphosed sandstone.
(2) K. Driscoll ,2008 ,IDENTIFYING AND CLASSIFYING VEIN QUARTZ ARTEFACTS: AN EXPERIMENT CONDUCTED AT THE WORLD ARCHAEOLOGICAL CONGRESS, , publ. 2011; L.S. Barham, 1987, The bipolar technique in Southern Africa, a replication experiment;
(4) Roger Armbsruster , Robert C. Antoine Note préliminaire sur quelques stations lorraines a  quartzites paléolithiques. Bulletin de la Société préhistorique francaise  lien   Année   1955   lien Volume   52   lienNuméro   8   pp. 467-479
(5) See: Rabhi M. (2009). Etude de l 'Industrie Lithique du Niveau  de Ain Hanech: Approche Expérimentale.
Athar, Revue Scientifique d'Archéologie et du Patrimoine, Institut d'Archéologie, Université de l'Alger, 8, 13-37.; espacially at page 8 we can see how laminate ( blads) are absent in the concentrations of quartz tools.


Knutsson, K. 
1985.  SEM-analysis of Wear features on experimental Quartz tools. Early Man News 9/10/11. Tübingen.

Knutsson, K., Dahlquist, B. and Knutsson, H. 
1988.  Patterns of tool use; the microwear analysis of the quartz and flint assemblage from the Bjurselet site, Västerbotten, Northern Sweden. In Industries lithiques: tracéologie et technologie. B.A.R. International Series 411(1), 253-294.

Nicholas Toth and Kathy Schick :Stone Technology  Origins of Stone Technology, Investigating Prehistoric Stone Technologies, Modes of Stone Technology ( article)
Justin Pargeter Human and cattle trampling experiments in Malawi to understand macrofracture formation on Stone Age hunting weaponry (in"antiquity 85/ 327, 2011"

Practical primitive: the making of pebble tools

Stone Tool Making 101: The mind of a modern knapper (Science magazine 3-2009)

Shipton, Petraglia & Paddyya
2008 Stone tool experiments and reduction methods at he Acheulean site of  Isampur Quarry, India -PDF ; Antiquity 83 - 769 - 785

3. Experimental making of small pebble tools (March 2010. France)

The pebble tools, found in the Belgian Kempen area were made in bipolar technique, using an anvil and direct percussion technique and in a hard hammer - anvil technique, using an anvil and place blows at one edge of the pebble. To prove the existence of such ( small) tools , it is normal to make them in an experiment. The pebbles, used in this experiment origin from the Moselle. 
Mr. H. Beaudouin, administrator and driving power behind  the French Forum about regional Archaeology in Lorraine,  "Forum de la Prehistoire en Lorraine" has made some experimental 'tool- forms' just to find  the possible applied techniques of the Kempen pebble tool assemblages. Such comparisons are possible, since the fluvial depositions from the Belgian Kempen area date from a period before  the Toul captation,.

What appeared immediately is the outcome depends on the type of anvil, the type of hammerstone ( another pebble) the angle during the percussion and the used technique: bipolar technique gives a different tool type compared with a simple hard hammer/ anvil technique.
Several experiments showed, it is possible to make tools that look similar like the tools found in the Kempen. but it is rather difficult to estimate the right pebble properties and failures are common

 Some examples show the possibilities and limits in tool processing:

This tool type, a small chopping tool ,  is commonly found in the Belgian Kempen. It was made by a well longitudinal  placed straight 90 degrees blow in the bipolar technique, that ended in a short  flake removal at both sides, simultaneously. The rather small working -edge looks useless, but  it seemed to serve very well for wood processing ( see below)

Bifacial blow

Bifacial blow 2
It appeared this tool type was very convenient for the decortification of  small wood  branches, such as e.g. the making of  wooden lances, like the ones they found at Schöningen (Germany) 
To prevent battering of the fingers, a split twig serves very well to keep small pebbles in place. Later, Mr. Beaudouin told me a rectangular block, placed on the pebble was even better.

Adapted quartz, suitable to serve as a tool. It appears to be rather easy to make a  working edge like this, suitable for scraping activities. by  simple hard hammer percussion .
This side chopper with two typical blows forming a "w" , photographed together  with the debitage debris , was made only after half an hour of practicing; these tool types and the debitage waste both were found in the Kempen as well.
End struck of the pebble, caused by a burin/ technique, where the pebble has been placed over the anvil´s edge, the other part shows a typical impact point which is concave  and produced two small flakes. Similar objects appear normally in the Kempen assemblages.
Various tool forms made from one pebble, ranging in size from very small till a core -` tool' similar found in the Kempen assemblages, as  shown at this weblog.
Hard hammer technique provides several failures, especially when beginning in the technique applied on small to very small pebbles.
Typical chopper , like found in the Kempen area. The two separate blows are very well visible, as well as the fact, the ribs between them are not very expressive. This overlap in blows causes a smoother surface, during alteration, where ribs could disappear completely,  leaving less indications for the number of strikes at the pebble
This is a real tool , ( no experiment!)  found in Lorraine in the Moselle region of Northern France. We distinguish at the right upper side of the artifact,  two different overlaying blows, leaving a small rib between them, the similarity with the previous, experimental made pebble is great. The left side of the tool has been broken, a natural event during hard hammer percussion on an anvil block. Image: Jean-Yves Ringenbach

An unifacial blow at a pebble showing the impact pint and rather straight 'percussion'lines , with a typical broken step- fraction in the middle, caused by the  opposite forces of the anvil. The impact point is light colored. The impact points at the Kempen tools are sometimes visible, but do not appear as bright as we find them at  present made tools .

Comparable artifacts from the Belgian Kempen region:

The artifacts, published and discussed at this weblog show big similarities with those made in experiments.
Of coarse, more experiments are necessary to compare applied techniques. In this way, it's already clear the angle of the position of the hammerstone will have been very variable, depending on the desired shape of the tool, the raw material and the capability of the knapper.
Images below are just for an idea how the by experiments made  tools resemble the paleolithic tools from the Belgian Kempen region.

1 A ( bipolar) split pebble with a single retouched edge ( hard hammer / anvil). Prehistoric tool from the MA4 site in the Belgian Kempen.

2 Quartz tool, looks very much like the quartz tool from the experiments. Cortical striking platforms are the most commonly applied features in the processing of pebbles in the Kempen region.

3. Originally, there were two blows placed in lower angle at the side of a transverse split pebble. The alteration made the rib almost disappear.

4. The original size of the pebble was much larger than the remains of the tool, maybe by use of the tool.

5. A quartz with separate working edges, made by trimming reduction of the edge, in stead of flaking. Trimmed edges are less nice than edges originating from flaking, but the processing is more certain.

The series of test pieces and the additional debris from the cores ( experimental pebble tool production, H.B, 03 -2012:

 1 Hard struck off pieces made by bipolar technique, leaving lots of fragmented pieces

1 a Main core and two useful flakes

2 Quartz, two separate blows make a workable side chopper, leaving small debris pieces in the form of small flakes

3 Very fine grained quartzite, leaving the right oval flake pattern, that is typical for the finds from the Kempen; the flake broke into two pieces, and, visible, this has been caused by the impact force, visible by the whit dot. The fresh white surface and concave fracture instigation is different from those found in the Kempen, indicating the prehistoric flaking of the artifacts

4 The flake removals left appear 'right placed' but anvil techniques often are causing chaotic percussion patterns, like fragmentation of ( parts of) the pebble

5 Simple chopper made by a simple strike. 
6 Refit of a pebble that has been fragmentated by a hard impact point 

6a Three useful parts of the pebble, made of a very fine quartzite , providing fine cutting edges

7 Transverse split pebbles are often broken by the anvil technique like ' burin techniques'where the position of the pebble is over the anvil's edge.

8 Experimental made chopping tool, with successful bifacial flaking technique, by simply switching the pebble after the first blow. In the debris we spot a flake with the typical fracture instigation.

8a Front view of bifacial flaking pattern, showing lowa angles at both sides of the core.

8b The flake, like the core, is a possible tool ( scraper, cutting tool), but it is not certain they used the small flakes too.

9 Side chopper with the debitage debris.

With special thanks to Mr. Hervé Beaudouin, Lorraine, France, for the demonstration of the experiments, and for the use of the pictures of his experiments  in this article.
With thanks to Jean-Yves Ringenbach for the use of some images of finds   from Northern France

Website of Mr.  Hervé Beaudouin  La prehistoire en Lorraine  about experimental archaeology, reproductions of archaeologica, forum - website in French, recommended


4. Some experiments on a very limited pebble clast from the surface at Niels -bij  -As : the production of very small tools 

In the Kempen area we find very small pebble tools, made on small pebbles less than  3 cm cross- cut.
The question is: how are these made and why make tools from such small sized gravels?

At Niels-bij -As, not far from As( Belgium), so at the highest part of Kempen-plateau, in the sandy area gravels might be exposed at eroding surfaces in the woods. In these contexts, far from the Maas/ Meuse river, pebble tools were not found.
The locally found gravels are very variable in size and composition, depending on the deposition we deal with - depending on what has been exposed by  erosional processes.
The gravels in the forest at Niels- bij As were small sized ( <5 cm) and vary from quartz, quartzite, sandstone and shale.
In the experiment, random gravels were placed on a flat anvil and were battered with a hammerstone  to see what shapes were possible.

Image above: (left to right) a partially processed quartz, the used anvil with experimentally made chopper, hammerstone. It is very important to find a good anvil, as if this is too little the countershock
is not right- causing stubbing edges. For this reason it good be necessary to place the anvil half in the ground, to reduce vibrations during tool making. 

The production of  useful tools/ flakes was very simple. The large hammerstone was used in direct percussion  of the small pebble at an anvil, making split fractures where useful cutting edges were made.
In the search for convenient small pebbles, quartz seemed to be very useful, giving quick result when battered on the anvil; but the form was not a success.
This was better with the small quartzite and sandstone pebbles ( < 3 cm) where good sharp edges were easily made.

This resulted in the following tools:

Image above: (left to right)  simple end chopper made of a sandstone, top = quartz fragment with ( very ) sharp edges, grey quartzite side chopper,  partially adapted quartzite, and far right a split quartz.Looking at these stones it is hard to believe they are real tools. For this reason other experiments are necessary, to see how productive the tools are for several tasks.

Image above: placed in the right angle and by the use of a twig to prevent battering the hand,  it was possible to make a flake ( right ) with dimensions ca 2, 5 cm long / 2 cm wide and 0, 6 mm thick.

The experiments were made in a bipolar technique, using a rather large hammerstone found at another place ( ca 8 cm cross cut) and very small pebbles (< 3 cm cross cut). One big  strike was sufficient to make tool objects.

The question why make such small tools maybe is answered. It is hereby suggested, the early hominids were migrating into unknown areas. Using all types of pebble clasts is an advantage, especially when larger tools are absent. This is only the fact during interglacial conditions.  On the other hand, there is a possibility these small tools were convenient for wood working purposes, to produce more large wooden tools.

Image above:  what looks like two separate blows is in reality one blow; the angle at this experimental made artifact is wrong to become a good cutting tool