Designing Agency in the City




For Margaret Mead, cities had unique characteristics: “A city is a place where there is no need to wait for next week to get the answer to a question, to taste the food of any country, to find new voices to listen to and familiar ones to listen to again” (Partnow, 1993). Jane Jacobs (1961) pointed out that cities “differ from towns and suburbs in basic ways, and one of them is that cities are, by definition, full of strangers.” This is as true of Mexico City as of London.

What this suggests, as with so much in contemporary life, is complexity. Cities, and the issues facing them, are complex, difficult to grasp, through size and diversity and life. They are messy (Haque, 2013). But this complexity is often part of their appeal— from Samuel Johnson’s “No, Sir, when a man is tired of London, he is tired of life; for there is in London all that life can afford”, to Kevin Lynch’s (1960) “Looking at cities can give a special pleasure… At every instant, there is more than the eye can see, more than the ear can hear, a setting or a view waiting to be explored”—writers have demonstrated a fascination with unpredictability, with verve and emergence in the city context.


As Carlos Gershenson points out, complex systems such as mobility within the city are not something where studying the individual components in isolation enables you to say much about how the system functions as a whole, nor to predict with much confidence. The interactions or couplings between parts of the system (including people, in all their diversity), and how those relationships act on each other, mean that any model or simulation needs the ability to adapt, to change itself, and—taking a second-order cybernetic perspective—so do the people creating the models and simulations, in terms of learning how to learn.

Beyond the city scale, humanity is surrounded by, and enmeshed in, complexity which at once causes us paralysis over not being able to take action, and regret over the actions we do take (and continue to take). We simultaneously worry—and yet do very little—about issues such as the military-industrial surveillance state, ageing populations, inequality, war and privatisation of the commons. We fudge our responses to planetary-scale crises such as climate change, pollution or poverty because our understanding of what we are able to do locally does not match our understanding of what is possible at a larger scale. We face a crisis of agency, in the phrase used by Natalie Jeremijenko, Gyorgyi Galik, Zygmunt Bauman and others.


In this sense, planners, designers, and civil servants need not just humility about their ability to enact change, but recognition that their decisions, of what to model, what to measure, and what possibilities are considered, are themselves being influenced by their positions within the system and the history of their previous actions. There are no detached observers: what a government seeks to ‘control’ inevitably ends up controlling it, in turn, just as a thermostat ‘controlling’ the temperature of a room is in turn controlled by the room temperature it leads to (Glanville, 1995).

Complexity can be difficult for governments to handle. Scott (1999) emphasises how so much work of central planning, from Haussmann’s Paris to Soviet collectivisation, has been about trying to reduce complexity, to simplify populations and places, natural and human, to make them ‘ordered’, quantified (compare Domesday Book) and hence ‘legible’ by state apparatus with a paucity of social or cultural sensitivity.

As Gyorgyi Galik explains, the “perfect knowledge” intended through the uncritical quantification of “smart cities” is fraught with political and commercial incentives and decisions which potentially make the result a reductive, unrepresentative and dangerous one. Too often, legibility and quantification go hand-in-hand with control: the echoes of the ‘high modernist dream’, from Pruitt-Igoe (see Laura Ferrarello’s essay) to the ways in which corporations (and even universities) are embedding the ‘Quantified Self’ within the workplace (Moore & Robinson, 2015; Machkovech, 2016; Whitson, 2015), can be read as attempts at aligning the behaviour 55 of populations with a particular model of ‘best practice’, both biopolitical and ideological.

In one of my own areas of research and practice, what has become known as ‘design for behaviour change’ (Lockton, Harrison & Stanton, 2013) has increasingly been applied in ways which embody a deterministic (Broady, 1966), even behaviourist, model of a kind in which people are essentially considered to be components in a system, with known properties, which, if made legible to the system’s controller, whether algorithmic or human (Dutson, Fantini van Ditmar & Lockton, 2015), can be treated as ‘solved’. We are seeing this reductiveness applied in visions of our domestic life (Fantini van Ditmar & Lockton, 2016) and potentially even in visions of human behaviour in ‘streamlined’ futures (Ranner, Lockton, Steenson, Galik & Kerridge, 2016). As reflective, thoughtful, engaged designers, we must challenge this, and open up more pluralistic approaches. People’s lives are not just there to be made ‘legible’ to authorities (or indeed to corporations).

And yet, as we will see, legibility of the system, of policy and politics, is something that can work from the other direction—to empower change by the people, rather than solely from above.


What do we mean by ‘legible policy’? It’s perhaps easiest to start by looking at illegible policy. Using an example from the UK, here is a sign, in a little bit of public space alongside the River Thames, in Surrey (just outside London). That’s the borough council’s logo on the sign, and it gives details such as the name of the space, including the heartening phrase “This public open space is provided for the pleasure of all”—swans included!


If we look closely, on the back of the sign, also hidden by the tree, we find this extra sign:


These are bye-laws, a form of local legislation, enacted by the local council, which in a sense supersede or go beyond the laws which are nationally applicable. For example, it is not generally illegal to climb a fence in the UK, in and of itself (it might be illegal for other reasons), but the bye-laws here (if you can squint enough to read them) make it illegal—a criminal offence, leading to prosecution in a magistrate’s court—to climb any fence within the particular area of the public space, which could potentially include the railings alongside the road. Among other things, the bye-laws also appear to make it illegal to allow your dog to jump into the river in this area (a law which must be breached dozens of times a day), putt a golf ball, or “deliver a public address”, and there are a whole lot of specific rules about playing sports and sharing space with others.

In themselves, these seem like reasonable policies for a public space, but the question of legibility is important: how many of the people who take their dogs out for a walk, and let the dog splash about in the river (often chasing ducks, swans or moorhens— which is also against the bye-laws) know that it’s illegal and they could be prosecuted? Elsewhere this behaviour is permitted, or at least is not banned. But how many of the visitors actually read these bye-laws, printed in tiny text, in complex legal language, on the back of a sign half-hidden by a willow tree? Do any of the policy-makers at the council genuinely believe that visitors will read the text, and understand it, or is it essentially assumed that the public will not care?

Legible policy, then, is partly about making policy (and law) accessible: using language which is easily understandable, making things visible and open and readable, potentially also less complex and more consistent across contexts (for example, as the Common Good movement in the US proposes), and not hiding policy changes. But, legibility is somehow more than this.

One of the extra dimensions of legibility which became apparent through our discussions of mobility policy in Mexico City with Laboratorio para la Ciudad was the idea of making the intentions behind policy much more visible. A policy is essentially an argument put into practice, a stance on an issue which almost necessarily involves some debate and dissent about what the ‘best’ thing to do is (otherwise there would, perhaps, be no need for the policy).

But how much of this argument is apparent in the way that policies are presented and explained to the public? Does the public get to see the rationale, the intended consequences, or the reasoning which presumably was persuasive enough to win over a committee (or at least other colleagues) at some point, within whichever government agency produced the policy? How much are policies even explained to the public? Is it any wonder that misinformation, and mistrust of government policies, spreads easily when there is so little transparency around the intentions? Is this about consultation—genuine consultation, in which the public’s ideas and views are accorded value?


So, as we build our tentative model of legible policy, it seems as though intentions are something else to communicate, and actually to involve the public in. In Mexico City’s implementation of parking meters, for example, to what extent was the perception of lack of consultation (as complained about on this poster in Coyoacán) part of a wider ‘illegibility’ of the policy?

In his essay in this volume, John Lynch suggests that while “a policy… actively discussed, before, during and after implementation… demonstrate[s] positive momentum towards policy legibility”, going beyond this means “true participation in the policymaking process” on the part of the public, and in “iterating policy over time in response to the changing needs of the city.” This implies the need for the public to be able to act, and this is indeed something others have considered in this context. For Abraham Moles (1986), designed legibility is associated with an explicit dimension of potential for action on behalf of the person who is apprehending a ‘sign’ (which can be taken to refer to any kind of presented information in the world). Design is a key tool for “transform[ing] visibility into legibility, that is “transform[ing] visibility into legibility, that is, into that operation of the mind that arranges things in the form of signs into an intelligible whole in order to prepare a strategy for action”. This consideration of (graphic) design as “a project of legibility of the world” is intrinsically linked to agency, to people’s perceived ability to act, and it is to agency that we turn next. But what can it mean, in relation to policy? Is it the ability to challenge policy, to change it, to overturn it—or simply to feel that you are able to be involved in creating the policy in the first place? In summary, three components of legible policy which seem important, on this preliminary reading, are accessibility, transparency of intentions, and the potential for action—and they are all components which design is, I think, able to address.



“The city we experience is, to some extent, a product of a city council’s culture and behaviour, legislation and operational modes, its previous history and future strategy, and so on. The ability for a community to make their own decisions is supported or inhibited by this wider framework of ‘dark matter’, based on the culture of the municipality they happen to be situated within as well as the characteristics of their local cultures.”

Dan Hill, Dark Matter & Trojan Horses, 2012

“For each technology in infrastructure space, to distinguish between what the organisation is saying and what it is doing—the pretty landscape versus the fluid dynamics of the river—is to read the difference between a declared intent and an underlying disposition.”

Keller Easterling, Extrastatecraft: The Power of Infrastructure Space, 2014


To engage with complexity, rather than attempt to destroy it, requires planners, designers, and civil servants to understand society, culture and context better—to understand people’s lives, and appreciate the realities of situated decision making and subjective experience, appreciating the systems not just from the point of view of a fictionalised ‘user’, but from the perspectives of multiple actors, including other civil servants.

We need designers to engage with the invisible ‘dark matter’ (Hill, 2012) of infrastructure, politics and institutions, “the substrate that produces” the world as it is. We need this engagement even though the dark matter may often be experienced—and perceived—by designers as an impediment to action, rather than an enabler. We need designers to understand (and be allowed to deal with) the wickedness (Rittel & Webber, 1973; Conklin, 2005) of the problems we are facing: they will not be understood until ‘solutions’ have been attempted (which will in turn create new problems, as Gall (1975) points out); there will be no stopping rules; there will be no right or wrong answers; and all attempts to deal with a problem will only highlight its uniqueness and contextual peculiarity.


If we take an example such as Mexico City’s transport and mobility systems—and the policies around them—as discussed throughout this book, understanding the complexity of the interactions, the dark matter of infrastructure and organisations, and the everyday decision making of millions of people is a laudable aim for researchers, but it is surely not the ultimate goal, whether for transport planners or members of the public. The real goal is understanding how to enact change. Understanding how to act to change the systems we’re in is arguably the biggest meta-challenge of our age. We need not just information, but agency: tools for connecting our understanding of how things work and how we can act, around everything from cities, the environment, our own bodies, and networked infrastructure to policies in social, civic and political contexts, emerging technologies and plural considerations of futures.

This is design for behaviour change, but is not about planners, designers or civil servants trying to change ‘public behaviour’ as if it were somehow a separate phenomenon. We are all part of the same systems. An approach which prioritised ‘designing agency’ would use ‘design’—in the broadest sense—as a way to:

1. understand the world

2. understand people’s understandings of the world

3. help people understand the world

4. help people understand their agency in the world

5. help people use that agency in the world

…in a progression from understanding to action. But how would we do it in practice? Different techniques would be effective at different levels. Some would be investigatory, some practical, some speculative or critical. Some would give us tools for understanding and learning, some tools for doing, some provocations for reflection.

At Level 1: Understand the world, Laboratorio para la Ciudad’s Mapatón is a fantastic example: designing a system which is essentially a large-scale probe deployed into the world, running a designed experiment to investigate phenomena in the world through gathering data in a way which provides meaningful scaffolding for the next level.

Level 2: Understand people’s understandings of the world, in attempting to ‘understand understanding’ (in Heinz von Foerster’s phrase (2003), would take things a stage further: using activities which practically try to explore the different ways in which people imagine, conceptualise and think about how things work. For example, very basically, it could be about using techniques such as drawing (Bowden et al, 2015), to uncover people’s understanding and mental imagery of a system, or creative games to explore people’s thinking about the world and possible futures (Tsing & Pollman, 2005). In the example concepts we generated with Laboratorio para la Ciudad, one idea in this vein was to ‘Probe the Government’, to put probes inside government offices, to find out how much civil servants actually know about their ‘own’ system: what’s spent, where it goes, and how decisions are made. This is partly about triggering people to think through their own thinking about an issue, making the invisible visible, tangible or legible, from the point of view of people themselves (i.e. what is legible, or not, to them), but also about surfacing people’s different understandings of situations, and how that leads different people to act.

One of the key issues here for legible policy is one pointed out by Leisa Reichelt (formerly of the UK’s Government Digital Service, and now leading service design for the Australian government’s Digital Transformation Office): “people have no mental model of government… Most people don’t know all the parts of government and what they do. Many people think of government as one thing but government generally doesn’t present a single view to end-users” (Reichelt, 2016). Perhaps government is not just invisible dark matter, but perceived or imagined dark matter in this sense, a complex, difficult mass.

So, understanding what people understand at present about a system—within and without government—is an important step towards, at Level 3: Help people understand the world, designing ways which help change people’s understandings of the world and the systems they’re in. This could take the form of new kinds of interface, designed experiences, educational activities—a range of things. Some of the examples collected by Dieter Zinnbauer’s Ambient Accountability project (Zinnbauer, 2015) perhaps fit here, from a transparency perspective, but Level 3 interventions could be about changing mental models, expanding horizons, reframing of situations, or even trying to facilitate empathy for other people within a system. In a sense, the microbus map produced by Laboratorio para la Ciudad as a result of the Mapatón research could be considered in this way—it helps people to understand a system they use every day, in a different way.


In our example concepts generated with Laboratorio para la Ciudad, we considered ideas such as putting a form of ‘price labels’, Sankey diagram or other forms of information visualisation on urban infrastructure such as parking meters, to show where the money raised from parking charges actually goes, since as Sofia Bosch argues, increasing trust in government can come about through transparency. It’s important to be clear here that this isn’t just about ‘correcting incorrect mental models’ on behalf of the public, but about enabling and supporting people to construct and refine their own models of the world, and the government of the cities they live in, experientially, which serve them better, and standards of evidence which satisfy their decision-making.

Level 4: Help people understand their agency in the world and Level 5: Help people use that agency together are where agency comes in more directly: helping people understand what they can do to change things, and then helping people do that. What could this look like? Examples such as DemocracyOS (as discussed by Gyorgyi Galik in her article) are very clear instances of empowering people not only to understand what they can do, but take action. We can consider creative agency as an important component of this—the ability to use creative approaches to participate in, and have an impact on, civil society (Hargreaves & Hartley, 2016; Lockton, Greene, Casey, Raby & Vickress, 2014) including changing the behaviour of the systems around us. Can this be more constructive than ‘fighting back’, and actually be about co-designing policies that behave more effectively, and work better for more people? These could be applied critically, speculatively or provocatively—a what if?—or they could be direct ways of enabling action, empowering people to change the behaviour of the systems in which we live.

In this vein, A. Baki Kocaballi has written very usefully about agency sensitive design, particularly the notion of relationality (recognising that assumptions of neither full technological determinism, nor full social determinism, are useful when understanding agency in context. Kocaballi’s six qualities for agency sensitive design (Kocaballi, Gemeinboeck, Saunders, Loke & Dong, 2012) —relationality, visibility, multiplicity, configurability, accountability and duality—could be a valuable set of considerations to explore in relation to the design of these ‘Level 4 and 5’ attempts to help people understand and use their agency in the world.


In conclusion, the complexity of cities and the policy issues involved means that simple solutions which address phenomena in isolation are likely to miss important relationships. When considering engagement with the public, reflective planners, designers and policy makers must challenge reductive approaches to ‘behaviour change’ and consider opening up policy in ways which improve its legibility, through accessibility, transparency of intentions, and making the potential for action clearer. Legibility of the system, of policy and politics, can empower change by the people, rather than solely from above. Designing agency in this way, in a progression from understanding to action, offers a series of potential intervention types.

At this level, planners, designers and policy makers should be particularly mindful of their roles within the systems we are aiming to help people change. The power dynamics, and our assumptions about the people we are designing with or for, need to be surfaced and questioned.


Parts of this essay draw on Dan Lockton’s blog post ‘Let’s see what we can do: Designing Agency’ published 23 December 2015: https://medium.com/@danlockton/let-s-see-what-we-can-do-designing-agency-7a26661181aa





Bowden, F., Lockton, D., Brass, C. & Gheerawo, R. (2015). Drawing Energy: Exploring perceptions of the invisible. Royal College of Art: London.

Broady, M. (1966). Social Theory in Architectural Design, In Gutman, R (ed.) (1972), People and Buildings. Basic Books: New York.

Conklin, J. (2005). Dialogue Mapping: Building Shared Understanding of Wicked Problems. Wiley: Chichester.

Dutson, C., Fantini van Ditmar, D. & Lockton, D. (2015) The Performance of Nonhuman Behaviour. Workshop at Nordes 2015: Design Ecologies: Challenging anthropocentrism in the design of sustainable futures, 7-10 June 2015, Stockholm, Sweden

Easterling, K. (2014). Extrastatecraft: The power of infrastructure space. Verson: London.

Fantini van Ditmar, D. & Lockton, D. (2016). ‘Taking the code for a walk’. Interactions 23(1), 68-71.

von Foerster, H. (2003). Understanding Understanding: Essays on cybernetics and cognition. New York: Springer.

Gall, J. (1975). General systemantics: An essay on how systems work, and especially how they fail. General Systemantics Press.

Glanville, R. (1995). ‘A (cybernetic) musing: Control 2’, Cybernetics & Human Knowing 3(2), 43–46.

Haque, U. (2013). Messiness will inevitably arise in spite of smart cities. Wired, Retreived from July 2013 from: http://www.wired.co.uk/magazine/ archive/2013/07/ideas-bank/in-praise-of-messy-cities

Hargreaves, I. & Hartley, J. (eds.) (2016). The creative citizen unbound: How social media and DIY culture contribute to democracy, communities and the creative economy. Policy Press: Bristol.

Hill, D. (2012). Dark Matter and Trojan Horses: A Strategic Design Vocabulary. Strelka Press: Moscow.

Jacobs, J. (1961). The Death and Life of Great American Cities. Random House: New York.

Kocaballi, A. B., Gemeinboeck, P., Saunders, R., Dong, A. & Loke, L. (2012). Embracing Relational Agency in Design Process. Design and Semantics of Form and Movement DESFORM 2012, Wellington, New Zealand.

Lynch, K. (1960). The Image of the City. MIT Press, Cambridge, MA.

Lockton, D., Greene, C., Casey, A., Raby, E., & Vickress, A. (eds.) (2014). Creative Citizens’ Variety Pack: Inspiring digital ideas from community projects, London: Royal College of Art.

Lockton, D., Harrison, D. & Stanton, N.A. (2013). Exploring design patterns for sustainable behaviour. The Design Journal 16(4), 431-459.

Machkovech, S. (2016). Evangelical university requires Fitbit ownership, data syncing for freshmen. Ars Technica, Retrieved 1 Feb 2016 from: http:// arstechnica.co.uk/gadgets/2016/02/evangelical-university-requiresfitbit-ownership-data-syncing-for-freshmen/

Moore, P. & Robinson, A. (2015). The quantified self: What counts in the neoliberal workplace. New Media & Society, Retrieved from: doi:10.1177/1461444815604328

Partnow, E. (1993). The new quotable woman. Headline Books.

Ranner, V., Lockton, D., Steenson, M.W., Galik, G. & Kerridge, T. (2016). ‘Plans and Speculated Actions’. Workshop at DRS 2016: Design, Research, Society: Future-Focused Thinking, 27-30 June 2016, Brighton, UK.

Reichelt, L. (2016). GOV.AU is a ‘mental model’ of government. Australian Government Digital Transformation Office. Retrieved 16 March 2016 from: https://www.dto.gov.au/blog/gov-au-is-a-mental-model-for-government/

Rittel, H.W.J. & Webber, M.M. (1973). Dilemmas in a General Theory of Planning. Policy Sciences 4, 155-169.

Scott, J.C. (1999). Seeing Like A State: How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press: New Haven.

Tsing, A. & Pollman, E. (2005). Global Futures: The Game, in Rosenberg, D. & Harding, S. (eds.), Histories of the Future. Duke University Press, Durham, 107-122.

Whitson, J.R. (2015). Foucault’s Fitbit: Governance and Gamification, in S. Walz and S. Deterding (eds.), The Gameful World. MIT Press: Cambridge, MA.

Zinnbauer, D. (2015). Crowdsourced Corruption Reporting: What Petrified Forests, Street Music, Bath Towels, and the Taxman Can Tell Us About the Prospects for Its Future. Policy & Internet, 7(1), 1-24.