SSG POST 10 – On Carving

Before our upcoming trip to Whistable, here is a report on our previous experiment in carving from the living rock.

In last February, we were looking for sandstone cliffs close by London, to get closer to the material Lia is working with while researching rock-cut caves in Southwest China.

We found comments left by climbers on the net and headed to the cliffs of Tunbridge Wells, an hour ride South of London.

We started wandering above and among the sandstone boulders around Toad Rock, trying to frame the miniature landscapes left over by water run-off and rope marks.

2high pass_MG_0020 2high-pass_MG_0025


Then, in quest of a more secluded bit of cliff, deeper into the woods, we settled on a narrow sandstone cornice covered in layers of soft (wet) dead autumn leaves.


That was the ideal spot to try out the carving sequence deduced from Lia’s collection of unfinished 2nd century AD rock-cut caves, and think through the mental processes of stone masons, when having to design a recessed door.

Previous to the experiment, a collection of unfinished 2nd century AD caves was gathered during several fieldtrips between the years 2009 until today, in the area South of the Upper Yangzi River, including 5 Provinces of nowadays Southwest China: Sichuan, Guizhou, Chongqing, Hubei and Hunan.

The unfinished caves were replaced into a carving sequence, including variants in certain sections of the chaîne opératoire, resulting in a ‘tree of actions’. This was an attempt to reconstruct genealogies of gestures, or stone working traditions.


We selected a flat enough portion of the cliffs, looking for exposed bits that remained protected from water run-off, under natural ledges… and started right away to draw our miniature doors, just line carving in the soft, sandy and homogeneous surface.


We started from the outer layer, gradually deepening the opening, and carefully leaving layer after layer of uncarved matter to form the recessions. As stone carving is a substractive process, there is no way back.


Four months later, in the mountaineous lands South of the Yangzi River, in Chongqing, Southwest China, we repeated the experiment but this time, life size. A 1 :1 scale cave, as compared to its miniature prototype, is caught in a network of real life interactions between ourselves, the stonemasons, the local authorities and of course, the landscape itself.


This is a view of the small agglomeration of Nanquan, on a Southern tributary of the Yangzi River in Banan district, Chongqing Municipality. The mountain in the middle was one among the many former residencies of General Zhang Jieshi (Tchang Kai-shek) in Republican times (late 30s), including residential buidlings and a few hundred square meters of underground bunker spaces cut into the sandstone. Today, the mountain has become a protected heritage site under the supervision of the Banan Cultural Relics Office. The Office delivered us a permit to replicate a 2nd century AD rock-cut cave in the area, curently undergoing restoration works.


Here, you see portraits of the two stonemasons, Master He, 58 years old, and 63-years-old Master Fu. Both were hired by the Office for the on-going restoration works, so we were able to simply ‘borrow’ them from their current job for a dozen of days. He and Fu are from Guang’an, Sichuan Province, and they have been in contact since their childhood with such rock-cut caves.

Since their teens, they have been working at sandstone quarries, until the mechanization of quarrying. They then switched to more various occupations related to stone carving, joining the hundreds of millions of migrant workers active in China today. However, both of them still possess land and a farm back in Sichuan, where they still attend harvest every year, on a seasonal basis, and where they plan to retire in a couple of years.

They accepted the job offer and were satisfied of its outcome but they were very aware of how much easier it would have been to cut a cave by using modern drills and other mechanized stone working equipment.


A standard chaîne opératoire was chosen for the experiment of replicating a cave. As shown on the sketch below, which was communicated in this format to the stonemasons, carving starts with a shapeless cavity. It then extends diagonally into two trenches about 20 cm wide, enough to let a hand holding a chisel and a hammer in. The matter left between the trenches is then chiseled off, resulting in a widening space that is trapezoidal in groundplan. The same operation is repeated several times until the obtention of a deep enough inner volume. Ultimately, the corners of the space are delineated into a rectangular ground plan.


However, as mentioned above, variants exist in the carving sequence of a cave.

Once the actual experiment started, we realized that in our 1 cubic meter cave, the difference between the width of the cave opening and the width of the inner volume was in fact too small to require the trapezoidal stage.

The stonemasons nonetheless started by following the process described in the sketches, to then angrily explain to us through sketches again how much time and effort was wasted in including this intermediary stage, which remained totally invisible in the final result. This ‘Mistake’ was pretty much the only interference between our pre-established sequence of movements and the one the masons spontaneously adopted. This ‘Mistake’ shows how the masons continuously tested our hypothesis against the resistance of matter and the realities of scale. Along such major events, a variety of other details in the masons’ gestures contributed to our understanding of the carving sequence both at the gesture scale and at the scale of visible progress.

We finished this experiment on a thought.

The image below shows a pair of protective glasses hung on a chisel by Master He, who fixed that chisel in the cliff face every morning to hang his jacket, hat, etc. The following images are from the 2nd century AD caves : chiselmarks abound around the cave openings, generating speculations among archaeologists.

The ‘Ordinary Traces’ theory is a reminder of our tendency to overinterpret data and attribute intentions to mundane gestures.



This was the opportunity to try out our SSG collaboration on new grounds, and relate it directly to archaeological field work. But also, it brought us further into action on the landscape, and involving actors from that landscape.

Both sandstone and chalk are sedimentary, homogenous, soft stone, and there is but one step between the canyons of the Yangzi River gorges and the cliffs of Dover. From the mineral layers of sandstone, we turn back to the organic conglomerate of chalk, once alive.

Tomorrow, we set out to the Kent coast again, ready to translate the recessed gate into yet another context, material and scale.


Join the discussion!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s