Building bridges between archaeology and agent-based modelling: Interview with Andreas Angourakis

What is the first word that comes to your mind when you think of Cambridge and why?

Community. Thinking of the university rather than the city. This is a place where I feel embraced by both people and institutions, which is not often accomplished in high education/research centres. Though work is intensive, there is active care for employees’ mental health and general welfare, from coffee/tea breaks to formal administrative procedures. The continuous activity of lecture rooms in Cambridge offers a great intellectual stimulus for researchers in any field. I believe this community environment to be crucial to balancing the introspective nature of the academic profession. Regrettably, I am yet to be initiated in the ultimate-Harry-Potter-style community feature in Cambridge: college affiliation!

Can you tell us a little about you, and your research interests? As an archaeologist working with agent-based models, I imagine you spend time working with two very different types of researchers. What is the hardest thing to explain to archaeologists about agent-based models, and what is the hardest thing about archaeology to explain to agent-based modellers?

This is a tricky question! A proper answer would probably fit the content of an article. A short version would be: (1) archaeologists struggle with how human behaviours can be modelled and simulated as computer algorithms; (2) computer scientists and modellers from other disciplines generally do not comprehend the specificities of the archaeological evidence, including its temporal scale and fragmentary nature.

In my experience, it is easier to introduce ‘modellers’ into archaeology than archaeologists to computer modelling and simulation. Agent-based modelling (ABM) in particular can be challenging to many archaeologists because it requires two skills: programming and anthropological imagination. The first is increasingly being overcome by the integration of ‘computer skills’ in the archaeologist curriculum. There are nowadays many fields within archaeology that rely on code developed by archaeologists. The latter, in my opinion, manifests a much deeper problem that comes with the specialisation of the archaeological practice. Because archaeology deals with materials so intensively, we often forget that, by studying the material remains/context of human activity, we intend to understand human activity rather than the materials themselves. The many hours spent in fieldwork, sample preparation, and laboratory protocols come with a price!

The heuristics of ABM, as in other simulation approaches, is the same followed in experimental studies. We try to ‘explain’ one trajectory (the one observed) by reproducing a variety of possible trajectories in controlled conditions. A ‘virtual laboratory’ of sorts. Both conditions and trajectories of the model can be linked to the archaeological evidence, but the backbone of any model will be the definitions and assumptions involved. Even if we are inveterate empiricists, archaeologists must fit the pieces of evidence together, even if only tentatively, building models that use abundant references to something that we can never see in the archaeological record: processes. Processes are represented in our models as chains of hypothetical mechanisms, or ‘narratives’, that often involve decision-making. To build and improve narratives in archaeology is impossible without a type of imagination that can only be trained by learning (and reflecting on) the variety of human behaviour. For instance, we may interpret from archaeological evidence that a settlement was abandoned; however, we should further consider the many possible processes leading to abandonment, some of which may be tested as hypotheses—did people died in the site? Did they move in bulk to another location? Were conditions too dire? Did many young individuals decide to move away?

ABM gives the opportunity of formalising models or narratives of human behaviour and interrogating them through observing simulations in light of quantitative and qualitative data. In this sense, ABM is just a method for better controlling the logic of the models we already use. Most archaeologists I have worked with are surprised by the number of definitions and conceptual mapping required to bring ‘textual/graphical’ models into the ABM framework. Modelling with ABM is as painstakingly theoretical as it is humanising, bottom-up and out-of-the-box thinking.

What was your path to your postdoc at Cambridge?

The turn leading to my postdoc in Cambridge was unexpected, but, retrospectively, it was consistent with my convoluted academic carrier.

I was born and raised in Sao Paulo, Brazil. My father is Greek, hence the name! I attended a bilingual school, Colegio Miguel de Cervantes (Portuguese-Spanish), which was culture-focused rather than profession-focused. When first considering my carrier options, I was aiming for film studies and audio-visual production. I spent a lot of time drawing comics and was driven to storytelling and world creation. Once I got access to a university in Sao Paulo, I quickly grew tired  of the emphases given to business and marketing. Fortunately, given my education, I had the idea of studying in Spain. I ended up enrolling in the ‘Humanities’ career at the Autonomous University of Barcelona, which was a perfect mixture of disciplines, beyond what is normally considered Humanities. I treasured those studies. However, I still felt I was missing a lot of Social Sciences, so I decided to pursue a second degree, ‘Political Science’, switching to ‘Sociology’ one year after. It was in the Faculty of Sociology that I had my first experiences with ABM.

At that point, I was working full-time in a hotel and studying at night. The second big step of my academic trajectory was to spend my last year as a full-time, ‘funded’ exchange student in Tenerife, Canary Islands. There, besides the excellent food and company, I enjoyed an intense phase of learning, enrolling in subjects such as Ecology, Biophysics, Ethology, Mathematics for Biologists, and Prehistory. My time in Tenerife drove me to sign up to a Master in Prehistoric Archaeology, back in Barcelona, and focus simulation models applied to archaeology under the supervision of Juan A. Barceló.

A few months after obtaining my MA, I received an offer to work as a research assistant at the University of Barcelona, within a large multi-institution Spanish project called SimulPast. My role in this project was to work in collaboration with researchers from different fields to create agent-based models within one of the project’s case studies: oasis construction in Central Asia. The aim was to investigate how sedentary and mobile people living in and around oases interact to generate different land use patterns. I officially started my PhD at the University of Barcelona and received a grant to engage in another project, CAMOTECCER, focused on multivariate analyses of archaeometric data on ceramics. It was during my PhD that I really gained a more in-depth knowledge of both computer programming and archaeology.

For different reasons, the final stage of my PhD took more time than expected, and I found myself in 2018 writing it while searching for future funding. With incredible timing, I found the call for my current position while browsing on Twitter (thanks to Enrico Crema!). I was fascinated by the work done in the TwoRains project (https://www.arch.cam.ac.uk/research/projects/two-rains). Although living my-own-personal-PhD-hell, I put the extra effort and submitted my cv. Best decision I could ever make!

You have lived in both Brazil and Spain previously, countries with climates and cuisines that are arguably quite distinct from that of the UK. Aside from the weather (just kidding), what are you most excited to experience while living in England? Are there any places you want to visit, foods you want to try, or new hobbies you want to experience?

Well, I did visit the UK numerous times before moving last year, even spent three months in London in 2015, so I already experienced several exciting places. However, I do have several places to check on my list: Hadrian’s Wall, Winchester, Cornwall, Wales, and Scotland. The food, I must thank my colleagues in Cambridge because I already have tasted plenty of different foods. Eating crocodile at The Geldart here in Cambridge is also pretty high on my list.

Although the UK weather may be… suboptimal, I love how the landscapes here are so green and full of (non-human) life. I also like how people here are respectful, tolerant, and welcoming—there are exceptions, of course. I am definitely a fan of British humour!

What do you do in your spare time?

Plenty, but not nearly enough! Nowadays, I am balancing pubs, sightseeing, board games, and video games. I like running, but sports are not really my strong suit!

Know more about Andreas’ work by visiting his institutional profile in the Department of Archaeology at the University of Cambridge, his GitHub repositories, or by following him on Twitter (@AndrosSpica).

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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