On Streets
When Rem Koolhaas and his Harvard Project on the City team visited Lagos in 2001, they made a provocative claim: “Lagos has no streets.” When this was said, it referred to Lagos’ resistance to the neat cartographies and categories used to understand the so-called “global cities,” nodes of command and control (Sassen, 1991). What they saw instead were curbs, gates, barriers, and hustlers pouring through open and closed spaces. Even Lagos’ superhighways, they observed, appeared as porous landscapes with bus stops, markets, workshops, and industry bleeding into and across them.
Since then, people have responded in two broad ways. Some criticized Koolhaas’ statement as another stereotype—why would yet another foreign architect come to Africa and say we have no streets, representing Lagos as pure chaos without order? Still others have grappled with the idea that conventional planning's fixed categories cannot adequately capture the city's urbanism (Fourchard, 2011; Godlewski, 2010). Now, more than two decades later, I would like to revisit this idea. Of course, there is no question as to whether Lagos has streets as infrastructure, but really whether the very idea of the street as an ontological device is insufficient. To understand mobility in contemporary Lagos, I argue that we should instead imagine rivers.
On Rivers
The street as an ontological device means it is more than just physical infrastructure but also the basic unit of urban space through which people experience a city. The Global Designing Cities Initiative describes it as a multi-planar corridor, bounded by property lines, ground, and sky (Global Designing Cities Initiative, 2023). In this sense, the street is a rectangular void: a measurable, stable space. The problem with this definition is that Lagos’ streets do not fit into that box. They sprawl, they collapse into compounds, they bleed into plots. The jagged edges and contours inscribed on the urbanscape require too many approximations to be seen as planar.
I suggest we think of Lagos’ mobility system not as one connected by streets but by rivers. Lagos is, after all, a marine city—surrounded by water, cut through by lagoons, pressed against by the Atlantic Ocean, and resting on average just 1.5 meters above sea level (Higuera et al., 2022). Furthermore, annual floods that impartially inundate rich and poor remind us every year that we only temporarily occupy the riverbed. Yet, even without the water, the ‘river’ is alive with a great plentiness of vendors, hawkers, vehicular and human traffic, organic and inorganic waste, flowing and mixing rapidly and vigorously. To walk across Egbeda, from Akowonjo Rd onto 5th Avenue, is to submerge oneself in the viscosity of the air, to strut along with the tides, to take it all in—the mosque prayer calls as well as the bells from the roadside white-garment preacher. To walk that 400 m stretch is not to experience the city, but to become part of it, making ripples with every step you take.
The Waka Lagos Project
If Lagos did not have streets, then how might we describe and understand what we see when we step out of our homes? Waka Lagos, a project I recently led to evaluate pedestrian access to the newly commissioned metro system in Lagos, sought to do just that: to jump into the river, per se, and report what we saw.
Using Google Street View datasets, we applied computer vision to classify whether images were paved or unpaved, with or without sidewalks. We used a segmentation model (ResNet50) to identify trees, sidewalks, and clutter. Regarding visual clutter, we developed an edge analysis to measure how dense or obstructed a particular view was in comparison to the rest of the images in the vicinity. These tools allowed us to ‘see’ the urban landscape as it presents itself, not as it should be.
If there are no streets in Lagos and only rivers, then we realize that it would be constantly changing, shaped by the collective imaginations of those who inhabit them, in ways that master plans could never truly hold still. So we convened ministries, agencies, transport authorities, community development associations, and everyday residents [Images 1 & 2]. We asked, what should we be looking at? What qualities matter? Which details belong in our collective ethnography of these rivers?
Mapping the River
Yes, our work from the workshops eventually produced maps [and occasionally still refer to “streets”]. Not as ideal representations of the city form, but as incomplete descriptions—fragments along a centerline that tell thick stories: here the surface is unpaved, here the sidewalks are absent, here the clutter index is high, here the greenery index is low. Take Lagos Island as an example. We modeled pedestrian flows between the new train station at Marina and the surrounding one-kilometer walkshed. Assuming every building’s residents would walk toward the station, we estimated how many people would traverse each segment. The median was about 1,300 pedestrians per street segment. Overlaying this with our classification of paved/unpaved and sidewalk/no sidewalk, we could identify precisely which corridors would bear the heaviest burden of increased pedestrian activity due to the newly commissioned metro station [Image 3].
This serves as quite a useful tool for communities seeking investments in the pedestrian infrastructures of their neighborhoods. In essence, they now have empirical data to approach leaders with to say, "Look: these streets, already carrying thousands of pedestrians daily because of the metro station, need urgent investment in sidewalks or trees." Or they could flip it the other way and say, “Hey, your train stations could see so much more ridership if you capture the land value around these areas by investing in walkable urban designs.” To be honest, I can't really imagine all the possibilities of how people might use these insights, but I can see at least three benefits from this reframed perspective for urban studies.
First is the doing away with the ideal of a “good” or complete street. When we brought all stakeholders together, we realized that conceptions of what was good were so far apart that it was pointless to try to resolve. For some, it was smooth pavement; for others, shade; for others still, the presence of vendors. Poll results in Image 4 illustrate this chasm resulting from the lived positionalities of residents from different neighborhoods (Ikeja vs Lagos Island) and institutional priorities (State Ministry vs Community Development Association).
Secondly, with rivers, we can accommodate divergent visions to coexist. We make space for turbulence rather than forcing a single line of best fit. This means when mapping the city, urban scholars become kayakers, spotting changing currents and eddies, constantly reflecting in action. Thirdly, and perhaps more controversially, by connecting to big-ticket projects such as the new metro system, communities can begin to subversively divert resources and attention to the chronically neglected pedestrian infrastructures in the neighborhoods.
Perhaps your city too has no streets. Perhaps its life emerges not from the voids between property lines but from the submarine connections across spaces of deep difference. If so, then you must open space for new ontologies—rivers, archipelagos, rhizomes—that better capture life in your city. You cannot keep looking on from the map’s surface. You must step in, feel the currents, and learn to swim.
The study was supported by the Sidara Urban Seed Grant (2024-2025) of the Leventhal Center for Advanced Urbanism at MIT. I am also grateful for the leadership of our PI, Cong Cong, and the contributions of my colleagues Mikel Berra Sandin and Javier Ricardo Diaz.
References
Fourchard, L. (2011). Lagos, Koolhaas and Partisan Politics in Nigeria. International Journal of Urban and Regional Research, 35(1), 40–56. https://doi.org/10.1111/j.1468-2427.2010.00938.x
Global Designing Cities Initiative. (2022, October 27). Case Study 1: Calle 107; Medellin, Colombia - Global Designing Cities Initiative. https://globaldesigningcities.org/publication/global-street-design-guide/streets/streets-informal-areas/recommendations/case-study-1-calle-107-medellin-colombia/
Godlewski, J. (2010). Alien and Distant: Rem Koolhaas on Film in Lagos, Nigeria. Traditional Dwellings and Settlements Review, 21(2), 7–19.
Higuera, O., J. Jack O’Connor, Taiwo Ogunwumi, Ihinegbu, C., Reimer, J., Sebesvari, Z., Eberle, C., & Koli, M. (2022). Technical Report: Lagos Floods. https://doi.org/10.13140/RG.2.2.28097.43364
Sassen, S. (1991). The Global City: New York, London, Tokyo. Princeton University Press.