The Weight of Water and the Cost of Labour: Reconstructing the Past with Roman Concrete

Roman concrete has been praised by scientists for its remarkable ability to withstand the test of time. It characterises ancient Roman cities and their remains today since without it, there would be no aqueducts, baths, dams or most other characteristic buildings. The ingenuity displayed by Roman infrastructure has led experts to try and recreate ancient materials as well as to question the hidden cost of labour involved in their making. It is this aspect in particular that prompted the practical session that took place at the BCDSS on November 29th, 2024.

That day, we, the 1st-semester students of the master’s program "Dependency and Slavery Studies" had the opportunity to experience the processes involved in the making of Roman concrete. In this blog post, you will read about the purpose of experimental archaeology, the material culture theory we used during our practical session and how we recreated opus signinum. Most importantly, this experiment aimed at offering us a chance to reflect on labour and supply processes of the past and to think through a critical lens about situations of dependency within different contexts.

Diving into material culture: experimental archaeology

Reconstructive experimental archaeology (see Schöbel 2019) is interested in physically recreating past material culture and technology. Through carefully designed, monitored, and documented scientific projects, artifacts are brought back to life using materials and equipment from their specific time period. Therefore, the aim is to reproduce as accurately as possible the historical context of an artifact while following a precise methodology. Reconstructive experimental archaeology is particularly useful in providing a new perspective to scientific analysis. Indeed, it can offer unique insights into artifacts, such as how they were manufactured, how they were used, and how efficient they were. Furthermore, reconstructive experimental archaeology can provide a deeper knowledge of the lived experiences of individuals with the artifacts and the sensory experiences they would have had while engaging with them.

Physical reconstructions can be divided into three different levels based on how authentically close they are to the original context and the methodology used. Level I reproductions are deemed to be non-scientific and non-authentic. Therefore, they may not be referred to as “reconstructive archaeology”. In the context of our experiment, we aimed at reproducing a Level II status, which is an authentic but non-scientific project concerned with understanding the steps and considerations involved on a purely experiential level. We were thus interested in the making of Roman concrete as well as the technical knowledge necessary in every step of the process. Given the practice and time investment needed to successfully execute a scientific and authentic reconstruction, it was not feasible for us to achieve the third, experimental level.

Making concrete as done by Martínez et al.

Our model study, conducted by Martínez et al., aimed at recreating opus signinum, a type of water-resistant Roman concrete. The researchers were interested in the amount of water required in the production of opus signinum as well as the logistical challenges involved with acquiring and handling it.

Opus signinum is a mixture of lime putty, sand and chamotte, which are pieces of crushed ceramics that function as an aggregate. In their approach, Martínez et al. drew upon ancient sources from the Roman world as well as other archaeological experiments. They also consulted master builders familiar with traditional techniques and materials in order to mimic the production process as successfully as possible.

Our concrete journey

For our in-class practical session, we followed the quantities established by the example study. We approached the process on an experiential level, with an interest in the techniques used to make, and the experience of making, Roman concrete.

Our lecturer, Dr. James Harland, spent several months preparing the proper building materials, tools, and safety precautions as well as practicing the procedure. Most tools were acquired at a hardware store, and most building materials were available to order online.

On the day of the experiment, students were provided with safety equipment and tasked with crushing pieces of pottery into chamotte of varying sizes. This step was followed by lime-slaking, a highly dangerous process of mixing quicklime with water at a 1:3,5 ratio to form a thick putty.

While mixing, the temperature of the lime increases rapidly, which causes bubbles and fumes to rise from the hot mixture. Our volunteer students conducted this step in full safety gear with suits, goggles, gloves, and masks. The observers were tasked with continuously reminding students not to touch anything that had come in contact with the corrosive mixture. After mixing, the lime putty was left to cool.

Figure 1 (on the left): Pieces of chamotte
Figure 2 (in the middle): Three students slaking the lime in a bucket
Figure 3 (on the right): Slaked lime putty
(All photos in this post are by Buğra N. Duman and Dr. James Harland)

The mortar was then created by combining the lime putty, crushed chamotte, and sand at a 1:1:3 ratio. We added an additional 1 liter of water to the mixture so the consistency of the mortar resembled the description in our ancient Roman source (Vitruvius, De Architectura II 5). The mortar was then left to rest so that any residual water could evaporate.

We applied the opus signinum to a wooden frame prepared earlier. The application process proved to be technically difficult, and those of us with prior experience in construction were able to demonstrate their skill. Since Martínez et al. also consulted professional builders in their experiment, it seems that previous knowledge and experience are crucial to a successful application of opus signinum.

Figure 4 (on the left): Student applying opus signinum onto the frame
Figure 5 (on the right): Final result of the opus signinum after the drying process

Overall, the process took our group an entire afternoon, during which we only produced and applied about 8 liters of opus signinum. We used a total of seven liters of regular tap water.

While the mixture did resemble the description in our sources during the process, we observed a crack on the semi-hardened surface of the frame during our follow-up examination three days later. Dr. Harland suggested it was likely due to a lack of moisture in the mixture, probably caused by the cold weather at the end of November.

Remembering the precise instructions of this exercise was a feat that involved most of our group at any given moment, constantly shouting around times, quantities, and materials to record. Division of labour was crucial here as the students in safety gear and involved in the mixing of the mortar could neither consult the sources and make calculations nor record the process.

Opus signinum and dependency

Researchers at the Morón lime museum, as mentioned by Martínez et al., suggest that lime-slaking was probably performed by female workers. Unlike the volunteers in our reproduction, they did not have access to the appropriate safety gear and would oftentimes go blind. Having experienced the process ourselves, we realised the immense vulnerability of the women on a Roman construction site, but also their integral role in society.

Figure 6: Students and the lecturer posing in safety equipment, ready to proceed with the lime-slaking!

By engaging in a practical exercise, we were able to uncover “hidden labour” involved in the process. It is those tasks, and more importantly, the people engaged in the tasks, that we tend to take for granted and thus forget. The practical session widened the scope of our analysis from the grand buildings and infrastructure of ancient Rome to the (oftentimes dependent) people who created them. By recreating their tasks and conditions, we learned to ask the right questions, namely questions that could help us understand not only the work of dependent labourers but also their roles in society. We got to ask who it was that had to complete the more dangerous tasks and why. We began to wonder who provided the copious amounts of water needed on an ancient construction site and how challenging that would have been.

As outlined earlier, although our exercise was not quite on the experimental level on which Martínez et al. operated, as students we were able to gain knowledge by employing new methodology and also trained some of the skills crucial for researchers in any field: observation and note-taking.

Our personal reflections

Lisa:
Doing this practical exercise made me appreciate the dangerous process of concrete production and made me think about the people involved in this labour. I also wonder how they would have handled the transportation and storage of water, since it was already tedious and frustrating for us to manage the tap water in 1l plastic bottles...

We were also very precise with the timing of the individual steps, so I am curious how this would have been done on an ancient Roman construction site, where not everyone had a watch on their wrist. I feel like I came out of this exercise with more questions than before, but I am now asking more specific and productive questions.

My biggest takeaway, however, is that recording the process is extremely difficult. While taking notes, I was constantly relying on classmates to keep track of time and materials. I also quickly learned how much the layout of the recording sheet determines the information that is captured on it.

Figure 7: Students collaborating to consult the sources and record the process

Israt:
In my personal opinion, if we could watch some documentaries prior to the experiment on ancient Roman construction and how opus signinum was used, I would have felt a stronger connection to the process. We did this experiment on a previously made brick frame and we were not engaged in building it. If we were involved in that too, I think it would be a more complete experience.

Engaging with this experimental process was really new for me as I have never studied archaeology, but I am always a big fan of ancient architecture on sites and in museums, as it carries the histories of human ingenuity. So, even beyond our class content, this experiment made me gain new insights and an interest into ancient Roman construction.

Claudia:
This experiment made me realise how easy it is to take for granted modern daily life objects in their finished form and forget about the lengthy process that is usually needed to bring them to life. Making Roman concrete was a reminder for me to stay aware of the context in which technology and goods are brought up, but more so, who is behind their production.

More specifically, I believed questioning the manufacturing process was a good starting point to comprehend how vulnerable groups are oftentimes targeted within the labour market. Indeed, they are oftentimes the ones forced to work in precarious conditions and carry out dangerous tasks. In the context of Roman concrete, it was striking to me that women were probably the ones traditionally entrusted with the treacherous job of lime-slaking. Ultimately, this practical exercise allowed me to seek a greater understanding of the dependencies that shape my surroundings, even the ones that can sometimes seem invisible.

We would like to thank our lecturer, Dr. James Harland, for his extraordinary commitment in organising and executing this project with us.