Exploring the process of hafting

I have discussed before how many of the learning processes I engage in are also about social relationships. I am keen to haft my Johnstone blade and my friend Simon Harper is someone I enjoy spending time with, and is ‘handy’ when it comes to doing practical stuff. Consequently, we got together last week to spend four hours or so drinking beer and trying and haft my blade (note finger and plaster).

haft 4

Simon is into history and prehistory (among many other things) and his fantastic replica Medieval knife (above) was used to fashion the wooden handle for my blade. I feel the need to confess at this point that no stone tools were used in the making part of this project. Partly because I wanted to do it sooner rather than later, and partly because I wanted to see how Simon would approach the task with his tools. I was learning from him really.


Simon’s first move was to select a suitable piece of wood from his wood burner pile that was approximately hand grip size. We had a blade and handle each and the above is my attempt after sawing to size, creating the slit and shaping with the metal knife.

haft 3

Functionally it is good, ergonomic and it holds the blade very tightly. Plus, I like the look of it. As you can see from the following photograph around 45mm of the blade is embedded within the handle. This provides good area of contact and the tight fit means that it has good strength in the longitudinal plane.


However, it currently has no support that would give strength in the transverse plane. In other words, if I were to put pressure on the edge by cutting it would move the blade sideways in the haft. So the question I am playing with at the moment is how to provide transverse support for the blade. I have two ideas but have not settled yet upon materials. Firstly I think some kind of mastic that is malleable when warm, but hardens when cool. This could be pressured into the slot surrounding the blade and providing good support when it becomes hard. This is the kind of material that Kiefer has producing from resin, charcoal and beeswax. However this knife is a slightly different beast. It is not a replica of anything, but a creation based upon a certain set of principles, which I cannot fully explain. It is becoming an exercise in seeing materials differently, and the toilet cistern certainly fulfils this criteria. So mastic wise I have two main candidates: chewing gum; and tarmac. I am thinking chewing gum currently as it is easy to get hold of and I think will provide the kind of rigidity needed.

Once the gap has been filled a different kind of material will be necessary, one that has a behaviour best described as ‘shrink to fit’. In the prehistoric past sinew has served this function, stretching when wet, and then tightening upon drying. The tightening provides an internal cohesion that would bring together the blade, gum and wooden elements. I still have some sinew left from my adventures at Reaseheath, although Bella, our Lurcher, has eaten most of it. However, I have seen another artist’s work that I would like to try and integrate. Micaella Pedros uses plastic bottles for exactly this ‘shrink to fit’ kind of function. This is a video of her’s on Youtube.

This hafting project has been going on for over a week now, and I carry the knife in its current iteration round in my bag with me. It sits on whatever desk I am using as I work through my different ideas on how to take the next step. The next post (in theory) should be the end result. Let’s see…


Saturday night in the lab

three points

I ended up back in the lab on Saturday evening, inspired by the previous night’s results. I took in some undamaged cistern lids, but focused upon using larger fragments lying around from previous sessions. It was interesting because I could feel the enthusiasm taking over on this occasion, whereas the previous evening had been characterised by precise and controlled thinning. Consequently, the two new points are less refined, but are both long. I seem to be managing endshock well, and the main conscious strategy is to listen to the hits. When the point is sounding stressed I back off. I wont go into the sonic nuances but once you have destroyed a few you can begin to hear the material complain when misused. I now respond well to both sound and the visual clues available, and the reward is longer points.


Whilst both the new points are less refined, I can say that they both have ‘character’. The first one was on a large, but curved, piece of ceramic. This presented the very same problems encountered when flattening the bases of beer bottles and so I applied the same methods. I was largely successful in a brutalist kind of way. Largely, because it is the longest of the three points, brutalist because there is lots of original surface left on the dorsal. As the photo shows, it has still retained some curvature in spite of my efforts. Removing that curvature would have reduced the length and so I have accepted one to achieve the other. I think for a knife blade this curvature is not an issue, but it may be problematic for a spear head as the longitudinal strength would be compromised.


I know you are not supposed to have favourites, however, this is my third one of the set. This one was both intuitive and remedial. I worked away, and then adjusted when it went off piste. Consequently the longitudinal mid-line edges are wavy and very much follow the flow of the process of making. It feels like the physical result of a human material dialogue. I like it because it reminds me of some of the Kimberley points from the Manchester Museum, where the person has worked with the form of the material, rather than attempting to mechanically impose a prerequisite shape. We have started to buy ‘wonky’ fruit and veg from the supermarket, and this my wonky bi-point: long, wide, off centre and still too thick, but an enjoyable interactive process and aesthetic result non the less.

Karen’s been out gallivanting


Friday evening and I ended up going back into Uni. Karen was going on a works do and so after taking the dog out I thought I would go back in and finish some things off. We have been knapping Johnstone during our weekly experimental sessions and slowly but surely I have got better. We started out in an unstructured way and I still can’t say exactly what my method is, but I am becoming pretty consistent and avoiding endshock. Anyway, tonight rather than finishing off what I went in to do, I ended up sat in the lab and decided to play with my thinning.

long point

As you can see it has worked pretty well and I am enormously satisfied with this one. Measurements wise it is still what Bruce Bradley would class as an Early, however I don’t really care too much as it is just pretty good. It has also confirmed to me that this material is actually ideal for knife blades. Kiefer Duffy and myself are going to get together to make a knife, and so I might try hafting this blade this weekend. If it works then we have the basis of a nice workshop we can do together, as he can knap, but also make a mastic glue from birch resin. Watch this space!

‘Endshock’ and flake analysis

Mac 1

We are back from our weeks holiday in Mid-Wales and knapping wise I didn’t have much more Johnstone Solutrean success. Ultimately from the two large sheets of Johnstone I took with me I produced the previously discussed Solutrean point, and lots of smaller blanks that will be suitable for arrowheads. The arrowhead blanks are a bi-product and so not relevant to this discussion, however I think it is interesting to explore where my problems lie. One problem that happened more than once was ‘endshock’ (see above image). When this piece snapped in the middle I was actually working on thinning one of the ends. This is how I ended up with the arrowhead blanks.


A second issue that deserves attention was how I was effectively thinning the pieces. Above are five useful penetrative thinning flakes that successfully removed the original surface, however, I couldn’t produce these consistently.  The above photograph shows the ventral, or inner surface of flakes I preserved for reference, to see what was going on when it worked.

mac 3

The above image is of the first three flakes but the dorsal, or outer face, showing the amount of flat original surface successfully removed. They are organised in size order with the first being the largest and therefore most successful. This first flake is 49 mm in length and the platform, or the part of the flake that received the impact is 9 mm in length and 4 mm in width. So all the energy from the blow was transferred into the body of the blank being worked through that small area. The platform can be considered ‘plain’ in that it has not been worked in any way. Another useful analytical aspect is the ‘Platform Angle’. If the angle between the face of the platform and the dorsal surface on any flake is measured it will usually be 90 degrees or less. At any greater angle the energy of the blow would not be concentrated and therefore ineffective. The platform angle on this piece is about 70 degrees. This tells us that the energy imparted from the heavy soft hammer blow entered the piece at 70 degrees to the dorsal surface through a plain small platform. Onto the second longest flake.



The second longest flake is 42 mm in length and the platform seems to be a small plain section some 4 mm by 2 mm. Emerging from the platform on the ventral surface we can see what is called an ‘Eraillure Scar’. This scar illustrates that a mini flake was removed from the main flake at the same time as the main flake was removed from the blank. For my purposes it acts as a signpost to the point of impact. Platform angle is more like 90 degrees on this one. On to the third flake.

mac 6

This one is 35 mm in length and the large erailiure scar indicates a single raised point of impact. This kind of platform is termed ‘Punctiform’, or point like, probably 2mm x 2mm, and the platform angle is again close to 90 degrees.

So how can this help us, or help me? The endshock I think is in many ways simply a symptom of a lack of attention. Firstly, the sound of a hit tells me if the platform has been correctly produced because I can hear the sound of the flake being taken off. Alternatively, if I can hear the dull thud of unstoppable antler against immovable blank then there is a potential problem. This can be exacerbated if the blank is not supported by the knapper’s body. I was guilty of both these ‘crimes’ when the above example occurred. So prepare a good platform; support the blank (perhaps by resting it on my thigh); listen for problems when I hit it. Regarding the flake analysis, it seems to confirm that a platform angle of between 70 and 90 degrees, and a sticky out point in these instances have led to long removals taking off decent flakes.

It has been busy since we came home and it has taken me a while to finish this blog post. I think I understood most of this stuff already before analysing these flakes, and so what this highlights is my movement between systematic preparation and a more intuitive percussion. Sometimes I pay attention, and sometimes I don’t, and the result is inconsistent flaking. If I can stay focused on the above points then in theory I should be able to move on a stage. I think if I can pay attention and get a better result, then my intuitive percussion will also improve.

This can be described usefully by a four stage learning model. The four stages of learning a new skill are:

  1. Unconsciously incompetent (it looks easy because you don’t know you can’t do it).
  2. Consciously incompetent (you try it and realise you can’t do it).
  3. Consciously competent (you progress by maintaining intense concentration).
  4. Unconsciously competent (you have got it and are in the zone!)

When it comes to Johnstone Solutrean points I am currently just past ‘Consciously incompetent’ and approaching ‘Consciously competent’. I was however in the zone on Saturday, but that is a different blog post!


Solutrean, Johnstone, Holiday

We are on holiday in mid-Wales and I have brought my knapping gear, some bathroom ceramic, and am busy testing out the ‘thinning then shaping’ approach discussed in the previous post. This was yesterdays effort and I like it.


Complete bifacial thinning with no original surface left. I didn’t bring my calipers but my best estimate using a ruler is that it is maximum width 35mm and maximum thickness 10mm which makes the ratio 3.5, and therefore Early.


The profile view below shows how it has been thinned more effectively in the lower section than in upper. If I were to use the lower minimum thickness then it would be 35mm divided by 8mm making a ratio of around 4.4, and therefore approaching a Middle.


The lesson here then is consistency, and whilst there is some way to go, I have my next attempt waiting for me outside, so I better stop typing!

The game is afoot!

I have just re-read a paper co-authored by Bruce Bradley regarding a Solutrean site in France. From the lithic debris recovered the authors were able to reconstruct the production process for large thin Solutrean ‘laurel leaf’ points. The key thing that is useful for me is that the authors posit a formula for recognising the stage of production. They divided the width of the piece by its thickness to produce a figure, and this figure indicated how ‘finished’ the piece was. Their model had four phases: Early – 3.6, Middle – 4.8, Late – 5.7, and Finished – 5.2. What these ratios tell us is that thinning was a primary activity that reached its apogee in the Late phase. Shaping would then reduce the width and therefore the Finished ratio would go down slightly. With this in mind I had another go with the aim of removing all of each surface to reduce thickness. As you can see, still a little original surface left on this face towards the front.


And a tiny, tiny piece of the glossed material left on this second face. I thought that would be enough to at least reduce the thickness. As I was by now losing width I didn’t take it further.


And the bad news? the ratio for this piece is 3.5, so technically still an Early in need of significant thinning. I measured all my previous pieces and the best one was unsurprisingly the widest at a ratio of 4, so again still an Early. 

Solutrean 2

I now have a thickness model to work to that tells me how well, or not, my thinning is progressing, and I find that quite exciting. It is ultimately going to be a game of maintaining width whilst being able to take off long flakes. Interesting stuff.

Thierry Aubry, Bruce Bradley, Miguel Almeida, Bertrand Walter, Maria João Neves, Jacques Pelegrin, Michel Lenoir & Marc Tiffagom (2008) Solutrean laurel leaf production at Maîtreaux: an experimental approach guided by techno-economic analysis, World Archaeology, 40:1, 48-66, DOI: 10.1080/00438240701843538

Exploring the concept and practice of ‘overshot’ flaking

This point was reduced in the same way as the previous example, using a heavy-ish soft hammer. Once it was approximately the right shape I started using my pressure flaker to isolate protruding points. Isolating the protuberances weakens them, whilst at the same time improving the chances that all the energy from the blow will be absorbed by that small area. The result is an increased chance that the blow will take off a longer flake. On the photo below I wanted to attack the ceramic gloss surface to the left of upper centre.


After isolating the platform I took off a good flake that travelled half way across the face and removed a significant portion of the gloss surface. The method works, as I knew it would but the angle of the surface is also a factor that needs consideration.


The process here is the same as the one John Lord introduced me to when making a handaxe. Essentially you use the abrading stone or pressure flaker to sculpt the ideal platform so that all the energy from the hit is concentrated in one place. What this means in effect is that each platform has to be systematically sculpted before each blow. For me this means two things: firstly I need to be more systematic and less intuitive in my reduction process; secondly I need a really good and finely shaped abrading stone to really pick out the platforms.


This is the finished artefact and the better of the two faces. As you can see, one of the removals has travelled two thirds of the way across. That in many ways is an ideal removal as it has changed the angle of the face. I think that is the process in essence, deep removals that reduce the thickness of the piece, and by default the angle of the face. The overshot flaking I think is an earlier stage phenomenon that, when using flint tablets, allows you to get to the thinned stage that I am starting at. I wonder if the results of overshot flaking leave a face that is easier to flake than this artificially tabular bathroom material. I need to do a few more, and generate some more data. I can then move onto flint and compare and contrast. So expect a few more of these.