How a South African data art project landed at the Venice Biennale
When Caroline Wanjiku Kihato first wrote her book on migrant women in Johannesburg, she had no idea her work would land at the Venice Biennale. Or that she would have to learn data visualisation in a single week.
Kihato’s 2013 book, Migrant Women of Johannesburg: Everyday life in an in-between city, became an interdisciplinary project when she started collaborating on a musical interaction of the book with composer Clare Loveday and pianist Mareli Stolp. After a performance of You Will Find Your People Here at Johannesburg’s Centre for the Less Good Idea, founded by artists William Kentridge and Bronwyn Lace, the curator of La Biennale Architettura 2023 asked if they would bring the performance to Venice.
There was a catch. Because the festival aims to slash its carbon footprint and be a zero-waste event, the film needed to be accompanied by artwork that was responsibly produced and could be reused. They decided it would be accompanied by an artistic interpretation of migration data printed onto traditional African fabrics.
As an urban sociologist who works in migration, gender, governance, and urbanization, collecting data was always part of Kihato’s work. But by transforming it into an artwork that people could relate to and keeping the integrity of both the data and the people it represented, she had a steep learning curve. And a tight deadline.
We spoke with her in Oxford, where she is a visiting fellow, about her work, the project and data’s role in it. This is an edited version of our conversation.
Q: You have been involved in different forms of storytelling as a way to communicate research and the sometimes complex ideas behind it. Why is that important to you?
I have worked both in academia and the nonprofit/think-tank sector most of my adult life and have become increasingly disillusioned with the fact that research doesn’t have the kind of social justice impact that it should in our communities.
I founded Frame45 to try and bridge this gap. Our hypothesis is that we need to tell research-based stories that inspire people to work towards building thriving communities and a just world. We believe that in addition to helping organisations reposition their strategies, research can actually have greater transformative value.
Q: How did the project start?
With my collaborators, Clare and Mareli, we wanted to approach women’s migration to cities in a way that surfaces their humanity and dignity. Our point was not to speak on behalf of women who move, but to offer a reflection of their worlds through our craft — words, music and embodied performance — and reflect on migrant women’s testimonies, through music and performance.
Q: How did it evolve into a data project?
After we were invited to turn the performance into an exhibition, we needed to think about another medium of telling the story of migrant women’s mobility to Johannesburg. We thought about putting together a series of maps which both contextualised the film and also responded to the intimacy of women’s lives and words. We wanted something beautiful, something that matched the sensitive telling of the film — soft, hard, gentle, harsh, comforting and unsettling. We threw a lot of concepts around and came up with what we think fits the bill.
Our data would be printed as khangas or kitenges — African fabrics worn and used by women across the continent. What’s special about these fabrics is that their bright and wonderful patterns carry a specific meaning. They celebrate an event, carry a proverb, or a secret code shared amongst women in the know. When Julius Nyerere died, there was one and when Obama became US president there was one that commemorated the event. So these textiles are important data transmitters, ones we could use to represent the data we wanted to highlight in the exhibition.Q: Where did you get the data points and what were the key points that you wanted to map?
We used data collected by the African Centre for Migration and Society at the University of the Witwatersrand. They surveyed human mobility in Johannesburg to understand how migration shaped everyday lives and politics in Katlehong, Berea and Diepsloot, three suburbs in the city where migrants live and work.
We wanted to understand three things: how many members of respondents’ families lived in another part of the city, country or the world; how diverse these spaces were — we used languages spoken as a proxy; and respondent’s aspirations, where they hoped to live, bring up children, invest, in the short, medium and long term.
Q: You taught yourself a lot of data visualisation. What did you have to learn and what tools did you use to get there?
Yes, I did! It was at once the most frustrating and fulfilling experience. So we knew we wanted the data to speak to the film, but we were not satisfied with showing just bar graphs and bubble maps. So using rawgraphs.io, I created the graphs, you know, the usual bubble graphs or tree graphs. And then we identified khanga or kitenge patterns that would best communicate the data. My favourite tool is definitely rawdata.io, but I’m just loving the Adobe suite, particularly Illustrator and Design.
Q: Who else did you collaborate with on the cloth artworks and the data analysis and processing?
To produce the maps we collaborated with Ghanaian artist Awo Tsegah. We struggled to find someone who could match our expectations for the data exhibition part of the project. We thought we wanted a GIS specialist, data geek, and artist all rolled into one. And they never came! We were about to throw in the towel when Awo was introduced to us by a mutual friend. That’s when we got it; the artist did not have to be the data geek or geographer, just a person who could work closely with them. So Awo built on work done by a team of people including Kabiri Bule, Juan S. Moreno, Dare Brawley, Laura Kurgan, Loren Landau and Tom Asher.
Q: What was the most difficult part of the data analysis and visualisation process?
Keeping the data and the visual form true. Once we identified how we wanted to tell our data story, the challenge was how to ensure that the data representation was accurate and that the pattern was recognisable as kente, kitenge or khanga. In other words, how could we render the information in a way that didn’t compromise on both forms? This takes a whole lot of back and forth because, in the end, we are creating an artefact that is larger than the sum of its parts.
More about the exhibition can be found here.
Disclosure: Tanya Pampalone, who initially interviewed Kihato for this Q&A for Media Hack Collective, later assisted with the editing of project materials related to the exhibition.
Media Hack Collective publishes data journalism on The Outlier, offers data journalism training across the African continent, and publishes a number of data-backed newsletters.
