Holding up a mirror
I’m revisiting this work to coincide with the launch of Co-Risk Labs’ initiative ‘Responsible Engineering, Science, and Technology for Disaster Risk Management’.
This is the backstory… It is the write up/draw up of an experiment devised by the brilliant Alison Killing (and funded by RIBA). Her hypothesis was that urban design was a blind spot for international humanitarian organisations (IHOs). Our hope was to explore urbanism in humanitarian settings by bringing professionals from different backgrounds to the same starting point for a discussion. So we invited staff from humanitarian organisations, urban planners, engineers, architects and academics to a workshop.
We formed the audience into small groups (mixed deliberately by background) and we used visual material to set the scene inclduing short documentaries filmed in Haiti by Alison and aerial images and plans. This is how the task was described.
… this footage gives you some idea of the backdrop, density and typologies that are at play. What we are going to ask you to do is the first half of a design exercise.
You will have 2 pieces of information: 1) aerial images, one will give you an idea of the damage and an idea of where people have settled. It is set in month two: there have been distributions of plastic sheeting, latrines etc. It is the emergency life-saving phase. 2) you will get a plan of the neighbourhood.
This is an audience concerned with the build environment, we will be thinking about access, roads, rubble water sanitation housing shelter etc. The first task is to put the sketch paper over the base plan and put your ideas over the base plan. The overarching question is what are you going to do? And how to draw this out.
The groups were then recorded as they talked about how to respond to a small corner of an earthquake ravaged city.
So you think you can code?
Then we took the transcripts and did some coding (funded by UCL Grand Challenges).
No - not that kind of coding… the kind of coding that social scientists use to dig deep into text. For this kind of coding, you listen to the audio, transcribe it into text and read over this ‘data’, underlining and labelling words and themes that come out. As you work through what was talked about and the way the talking unfolded, you also have to read around the topic and you have to think about the ideas and stories (the discourse) in which your ‘data’ sits, pondering the tropes, what is missing and what might be surprising.
What came out of this work were contradictions and paradoxes that we bundled under the headings: ways of working, conceptualising the city and fixing stuff. With the help of Veronica Wood, an amazing illustrator, these conversations have been drawn out.
Here or there, us or them
That particular earthquake ravaged city is not the object or the subject of this work. This is about the culture of the people wishing to “make a difference”: IHO staff and the consultants and academics in their orbit. We were not pinning down solutions for humanitarian response in cities, we were showing the quirks of professional interactions and how these shaped the problem. While we regret that this gives these characters disproportionate attention and distracts from the legitimacy and humanity of those already dwelling in and governing real places, this is the limit of what we felt we could learn from our stack of transcribed conversations.
Whiteness
I made a decision with the artist to represent age and gender but only one colour: white. I’m not sure this was the right call but my thinking was:
this was was approximately true of the workshop in Oxford
this reflected a truth about the conversations that happened in Haiti and about Haiti but without Haitian policy-makers, planners or academics
where this was interestingly untrue, drawing attention to it would potentially take away one or two people’s anonymity
there is much better work on this question
Handling the Task: ways of working
The challenge we set was pretty ambiguous and, in the face of this, people expressed both urgency and uncertainty. A couple of differences in approach came out.
There were those comfortable to start without a final objective in mind and those who preferred to write a list and those who started to categorise before drawing.
There were those who interpreted the task to be about working together as a group and those that assumed the task was to decide on solutions in the neighbourhood itself.
There were also different reactions to the images, with some people poring over them and describing information out loud, some by writing and some by drawing.
Some participants ignored the images altogether while others talked about the act of drawing as if it were essential to identifying or translating what might make up the neighbourhood. A French speaker talked about "diagnosis" which might make English speakers think about medicine: intervening in something that already exists, something (like a body) that even an expert cannot fully understand or mend.
Surprisingly, lots of participants downplayed the act of drawing itself. Some people:
dismissed drawing in favour of writing lists
referred to drawing as childish "scribbling" or casual "whacking down"
there were jokes that drawing was [too] "bold" and even "dangerous"
This reluctance to draw seems to overlook its usefulness and overplay its dangers. Listening to the conversation, I heard different ideas filtering through:
Drawing does not generate insights when it is done by “professionals”
There seemed to be an attitude that drawing by ‘professionals’ would not generate insights: professionals (even the ones that draw for a living) rejected drawing as a way of ‘drafting’. But the group that were most vocal about this were policy people.
How is it that professionals who draft written policies without flinching would be horrified by drafting sketches?
Is it because we privilege some ways of knowing over others: do writing and measuring have more prestige than drawing, imagining or craft?
Do we tend to overlook the sensory, experiential and intellectual value of handling stuff?
If we admit we are working with something that already exists and may be unknowable, can we afford what seems to be an unconcerned illiteracy with forms of information other than text?
Drawing does generate insights when it is done by “the community”
A defence of this reluctance to draw was that it should not be done from the top down or by outsiders. Instead, drawing was best suited to “the community", people who would know this place through their lived experience. This would generate insights.
To anyone familiar with the critiques of aid over the last 40 years, this will strike a righteous chord. It took me a while to understand what I found uncomfortable. It was something about the notion of “the community”: it held an almost fetish-like assumption of sameness, of illiteracy, of experience before intellect. But in this room, the enthusiasm for valuing experience over intellect applied to “them” not to “us”.
Are drawing and mapping (if expertly facilitated) were (only?) desirable, suitable and possible as alternative (or second-rate?) ways of knowing for an illiterate and inexpert 'community'?
Did this almost give us permission to avoid facing the articulate, the artists, the leaders, the architects and the engineers in that place?
Design = doing something
The task was framed as a design exercise. The groups assumed they were professionals arriving to do something, rather than nothing. An assumption that it is possible to come up with something, even from the ludicrously limited information to hand, went without challenge.
For most groups, the ‘something’ that came out was defined by objectives, “improvements” and “achievements” in the place. Only one group got stuck on what their role should be: this was the only group that were then compelled to explicitly distinguish themselves from “the community”. Everyone else took it for granted.
Because one group drew attention to this, it helped me to notice that none of the groups acknowledged relative power within the group - age, experience, gender, prestige of discipline – and yet these made a difference to the conversations. [Side note: How many people have been in a meeting of North Americans or Brits where (invariably) a woman apologies for not being an engineer before making a point…. If you are a from other parts of the world with more female engineers, you might not get this reference. Lucky you.]
Again, I thought is this a fetish: the obsession with only one inequality of voice and power – the one between the design ‘professionals’ and “the community”. It is so obvious that it’s barely worth mentioning. It is so dramatic it distracts us from looking at any other power dynamics.
Ways of Working in Groups
To get across the make up of the groups, as the analysis proceeded we gave the groups names that exaggerate and play on who was speaking and what was said…
The Experienced Non-humanitarians tended to tell stories, ask questions and describe interventions, rather than looking at the aerial images. The group comprised 1 urban planner, 1 donor, 1 development-sector researcher, 2 architects; 3 were men and 2 were women; 1 was over 40 with mainly UK work experience; others 30-35. 3 had experience living and working outside the UK; 2 had 1-2 years of experience in charities; 3 architects, 1 anthropologist, 1 geographer. 2 PhDs. All English speakers. Most frequently used words: people, think, land and things (know, infrastructure, needs, camps)
The Curious Generalists who read and talked about the images and focused on the task at hand. 2 were researchers, 1 urban planner, 1 NGO manager; all women; all 30-40; 1 had 5-10 years of experience of the humanitarian sector in the UK and abroad; 2 architects; 2 humanities. 1 PhD. All English speakers. Most frequently used words: like, looks, know, think (road, area, need, information)
The Institutional and Independent Practitioners started by defining objectives and interventions but then talked about the images and told stories. Of the 5 people: 2 were academics, 2 humanitarian practitioners, 1 architect/community activist; 2 were women, 3 men; 3 were over 40, 2 were between 30-35. 1 had 5-10 years of experience of the NG humanitarian sector in the UK and abroad, 1 had 2 years of experience of the NG humanitarian sector outside UK; 1 had 10 years urban design experience outside UK; 3 were engineers, 2 architects. 1 PhD. 3 English speakers.
The International Designers read the images and drew first and then moved to a preoccupation with questioning their own roles. Of the 5 people: 1 was an architect, 1 urban planner, 1 architectural researcher, 1 humanitarian relief worker (in their home country). 2 were women, 2 men; all were between 25-35, 3 had 5 years’ experience as architects or urban planners; 3 were architects, 1 international development. 2 English speakers; Most frequently used words: like, people, make, think, (community (working, know, different)
Tales of the city: ways of conceptualising
The next theme that came up for me was what a group of people, armed with different experiences and disciplines, would make of a small piece of a city.
Stuff of the city: categories or places
The original instructions to participants did not ask that they start by categorizing.
Two groups, however, set about identifying objects, elements and boundaries, finding something to say about the texture of the images, densities (of buildings), surface colours, contours and slopes and shadows. Note that both group names - International Designers and Curious Generalists - are named to reflect the mixture of people with and without experience of handling two dimensional plans.
Spaces were described by the International Designers as "empty", "open" and "clean", rubble-filled and "damaged" or as "camps".
They exchanged stories. The Curious Generalists made assumptions and probably imagined more information than the pictures could really offer, talking about: "untouchable" or "dangerous" areas and an "impassable pattern". They also talked about what might have happened in the past. They suggested to each other that any boundaries they thought they could see were made by people. Neighbourhoods and roads already had names. Fences, cordons and boundaries were in place because of some prior "status" and the function of different spaces.
The Institutional and Independent Practitioners switched between talking, tracing onto the paper and trying to describe physical and social features.
They linked together observations, like the disappearance of green space, with speculation about "camps" and stories about "congregational space" being integral to community life.
One person suggested that the act of plotting out (physical) infrastructure routes might be a useful way to learn more about the boundaries of (social) neighbourhoods and perhaps indicate something that would underpin people’s "sense of ownership".
This group went furthest from categorising and finding types of space. They recognised something 'urban' that was changing and in development; something technical, interconnected and related to all sorts of social issues.
Invisible social city: stories, systems and spaces
The groups with people who had been trained in mapping or technical drawing saw a lot in the images to talk about and seemed to be able to start with some shared questions about the space.
What was amazing was how people dealt with the many things that could not be known from the information at hand. Non-architects were more inclined to express difficulties reading the images and to confront the limitations of a static, two dimensional image but all the groups found that it was often "hard to tell" what was happening.
Things that could not be seen, had to be imagined. For example, the groups speculated as to: whether things were "temporary" or "already existed"; why people might move from one place to another; and whether the boundaries between different spaces had come about because of infrastructure or from a process of social "negotiations".
Two contrasts surfaced.
Firstly, broadly speaking, the people trained as urbanists kept trying to weave together a "whole" "urban fabric" of "relationships". By contrast, the people with humanitarian experience seemed to tease the pieces apart by talking about:
people as "the community" and "the municipality" almost as if they were things
their own interventions as gentle, neutral "processes" and using the passive voice
time and space as "phases", "timeframes" and "terms" during which things in the maps and images could either continue, transition or disappear
Secondly, although all the groups were all able to look at the data and imagine “risks” and demarcate "camps", they placed importance on different ideas:
Risks as physical things "rubble", "hazards", "slopes".
Places "at risk" such as "settlements", "situations" and "vulnerable areas"
People with vulnerabilities, thoughts and hopes or people facing precariousness, "fear" and "violence".
I had to keep rewinding the audio over and over again until I could prise out the two things that were bothering me.
Space
Very quickly an innocent or well-intentioned attempt to mark out an imagined “camp” made this place real.
Do humanitarian organisations need these things to be found? The organising principles of several IHOs focus on displacement camps and many are mandated to target those in need. So, if you are able to find “camps” or risky dysfunctional spaces in a place, you have automatically produced needy targets for your organisation.
[Side note: I did try to talk about this at an event once but I think it sounded as though I was claiming there were no camps or displaced people when I was questioning whether marking out “camps” are the best way to understand what is going on since this way of looking is so disconnected from any wider sense of social or political place.]
Time
It was possible to talk about the urgency, risks and fluidity of a disaster but simultaneously make the “processes” of intervention sound bland, necessary and legitimate and “the community” something fixed, dysfunctional and myopic.
Does something about the way that time was imagined in "phases" and in the separation of “emergency” and the “long term” serve organisational interests more than it serves understanding? Humanitarian organisations are only funded for the “emergency”, other international organisations are funded outside the immediate aftermath.
One group recognised that looking down god-like from space was problematic. But, on the question of time, it was a given that the vantage point of people somehow removed from the situation (experts), allowed them to see ahead.
Is there a weird mis-match between how little we challenge ourselves to understand the ‘before’ compared to how much we expect ‘processes’ to deliver the ‘after’? Is this bias hard-coded into design thinking? Does design thinking assume it is conservative to dwell obsessively on continuity and repair instead of starting afresh, reacting to data and iterating?
[Side note: I do wonder if I am guilty of fetishising the past of others/other places while holding a wariness and scepticism about using past data from my history/my place to estimate the future because all the old biases of the status quo are re-learnt by a new machine!]
Doing something: ways of fixing
The ways people talked with each other shaped the solutions they came up with. The ways they talked as a group shaped how they perceived their own roles and the traits of others.
Patterns of exchange: just talking is intervention
We tried to visualise the conversations as patterns of exchange - a graph of the time spent on different types of talking: ‘flights of steps’ mean there was a lot of back and forth, a long ‘plateau’ means one person holding forth. Colours show what was going on: orange is attention to images; yellow means asking questions; purple means talking about the role of the team.
It seems that when people were reading the images or trying to get to grips with the task itself, the conversation was dominated by short exchanges: nobody spoke for very long and different people contributed. In the two younger groups, similar in age and experience, a participant aware of her greater experience volunteered gently to "shut up" and there was a certain openness to new or mapped information and an inclination not to settle too soon on fixing a solution.
Conversations that started with the images, came back only at the end to discuss the task itself and possible solutions. By contrast, the group that made no reference to the images shared a series of stories, questions and solutions. They quickly asserted and talked for longer about their prior knowledge.
The group that dwelt first on their objective and solutions, was then steered by a balance between the forcefulness of the two older and most experienced participants. This pair had quite different perspectives and this led the group discussion to switch between reading the images, telling stories, asking questions, driving the task forward and proposing solutions. In going back and forth from the messy space to trying to distil "issues", they began arguing for the inseparable connections between space, infrastructure, people and their land (Excerpt 10).
From identifying space to intervening in land
The groups that embarked on telling stories from or about this real place drew the conversation away from categorising the space and towards land. Realising that important social boundaries - the institutional and relational influences on whether people stayed, squatted or moved - could not necessarily be read from the images, even if physically manifested there, one group's solution was simply to find out more (Excerpts 4 and 9, Curious Generalists).
The group that threw doubt on the value of referring to the images, instead asserted their experience and produced stronger, value-laden distinctions between places. They talked about places that implied agency ("self-settlement"), places that had been decided ("planned IDP settlements") and "camps" as places of opportunism: magnets to which people succumb from a distance.
Humanitarians in this group described land in terms of "issues", "processes", "assessment", "claims", while the urban planners talked about "opportunities", "accessibility"; "logical patterns"; "sustainability", "development", "infrastructure", "services" and "rights". The accompanying ideas about what to do included planning "an intervention on a neighbourhood" and different varieties of infrastructure - physical things that would function to demarcate plots and underpin invisible/hard to determine aspects of land such as tenure, claims and rights. All this was discussed with continual references to processes that would "make [landowners] part of the community" or involve "the community" in "the process".
In the group that disregarded the images, land was discussed in terms of "arrangements" and "envisaging situations", "claims" and "opportunism". Solutions emerged as participants used passive constructions, like "there needs to be", to couch normative suggestions including "removal" of people, "acquisition" of land, "corroboration". Together with the “they” - to refer to people on the ground - this had the effect of othering the people from that place (all groups did this) and attributing themselves with the power to intervene while softening the extent of their own power by projecting the responsibility for intervention onto "you" rather than "we".
Acknowledging power through empathy or caricature
The groups had not been told who they were supposed to be or work for. Yet, there were only two moments in the transcripts when people asked "but we are not Haitians, right?" or suggested the group "imagine" being there (Curious Generalists).
The combination of talking heads in the films - foreigners working as planners for IHOs in Haiti - and the impulsion to look down from above implied by aerial images and two dimensional plans, overlaid onto the prior experience and training of these participants created the automatic assumption of intervention from outside and above.
The group that focused on categorising and tracing over the images ended up in a cul-de-sac. They stalled in confusion about the role of their team and uncertainty about what identifying the spaces was for (Excerpt 17 International Designers). They struggled with how they themselves, as a group, should relate to other people and organisations and what they could or should do. Were they close to reality and therefore somehow positioned to make logistical or technical decisions? Or were they "on the ground" not to make decisions but to be a "listening device"?
The group decided they were "obviously" not Haitian and recognised that there is "Haitian power". They become preoccupied with clashing/coordinating with other organisations that were, somehow by virtue of being "technical", not Haitian.
They described an ideal situation where "the community" dictates what should happen but then back-tracked because in "the short-term" this ideal situation would be impractical. In attempting to resolve these dilemmas, the group tried to separate who they were from how they would work and what they would do. This detachment could give them a forward looking, bird's-eye vantage point.
Other groups concerned with how their processes and actions related to the neighbourhood, the community, the municipality. They caricatured urban planners as the poeple compelled to order and clean up a place - urban planners being the people behind the idea that cities need to be cleaned of the messy, slummy, informality of ‘pesky’ people. There seemed to be an idea that planning was something authoritative and long term.
The group that did not want to “play god”, discussed a wider variety of other actors (humanitarians, landlords, decentralised local government, gangs) and relationships (negotiations, situations, arrangements, prior conditions).
They resorted to telling stories from their own knowledge and imagining the motivations of (absent) people. On one hand they expressed concern for the abysmal conditions people faced. On the other hand, what emerged was a story of:
opportunism and connivance on the part of Haitians (Excerpt 16 Institutional and Independent Practitioners);
inaction, failure, slowness or incompetence on the part of humanitarians; and
local power structures that were non-existent, untrustworthy, unreliable or monopolistic (Excerpt 18 Experienced Non-Humanitarians)
This tendency to cariciature was not unique to this group. Other participants ascribed behaviours or traits to uniform/faceless groups of people as chaotic, ignorant and similarly careless everywhere in the world.
The exercise was framed in a way that both constrained and revealed of the ways of working, conceptualising and fixing. The plenary created an opportunity for the participants to reflect on this:
Excerpt 19 Plenary
... I was really struck by everyone talking about roads and rubble. The thing is if you give architects and urbanists some maps, they’re just going to do what they always do. It just occurred to me that half of us deal with nothing but data in terms of the number of people with access to latrines, amount of water people had before the earthquake, how much they have now… I feel we’re being encouraged to look at this thing from above with no data apart from the vision, which is for me an incredibly weird place to start. But for some people that’s what they always do, they always start with a map. I appreciate that humanitarians need to think about space, some people only think about space. So to say that urban planning can deal with this because it’s more about space, when actually it’s about politics, it’s about culture, society and an extremely difficult, dangerous urban environment. It’s not to do with space, it’s to do with so many other things too.
Why now and why us?
In the years since we did this analysis, I went back to working in tech; #metoo happened and untoward behaviour by NGO staff in Haiti and in London HQs hit the headlines; and innovation, failure and design thinking became modish.
These shifts have helped me organise my thoughts about the research.
We failed in managing the research project. The work ran over time and over budget because this method of “discourse analysis” was impossible to break into project activities that could be "outsourced" and then reassembled: ”you transcribe, you analyse, you theorise” doesn’t work. We had to accept that the work could still be compelling without ever being complete.
But so what? Low stakes… Except that it mirrored a tension that ran through the transcripts between weaving and cutting: those with urban planning backgrounds, who tried to interweave a "whole" "urban fabric" of "relationships", and those – including humanitarian staff and engineers - who tried to break down the space, process and analysis into solvable elements.
I felt that we failed in producing films. The intention was to show Haitian urbanists and politicians challenging the action and legitimacy of the organisations that had entered the city. Instead, the stories were told by our associates. Worse still we screened the films for students - they did not see what I saw: white talking heads to a backdrop of silenced Haitians moving through the city. Instead, they asked questions about careers in aid. The monster was made. We’d worked in good faith but we couldn’t fix the consequences of what we wrote, drew, talked about.
Ok. But nobody got hurt…. Except that it showed us how difficult it was not to privilege the familiar. It showed how easy it was to showcase only the people who would talk to us when there was limited time to build trust. Time that was limited, of course, just as it is in the urgency and drive after a disaster.
Sometimes break things, sometimes try to make things unbreakable. Working with an urban designer showed me that making, for people who curate space, edit films, is, necessarily, also a commitment to the possibility of making mistakes. This was not my idea of design. Maybe if I had been a software engineer, it would have been: the beta version of an app will not kill anyone if it crashes. You would never build a multi-storey reinforced concrete building in beta.
Yeah, we get that. But there are rules about that kind of thing…. As engineers, in aid work, we often enjoyed importance and influence as technical specialists that would be unusual at home and generally we don't think of ourselves as simple "followers" of rules. But our professional responsibility is not actually to decide what is "safe" – because that decision is socio-political, it is a deliberation not a calculation. Our duty is to design and check for compliance (with safety rules). This was a near impossible conversation to have on the ground because the engineers I talked to framed this question of high safety standards as a professional/ethical duty to make something safer never as a professional dilemma about what their role in safety decisions should be...
We got that there was an inequality of power, but we focused on the wrong one. In the workshops, professionals working with each other were aware of the distance between elite groups and the excluded. This did not make us look at our own group dynamics. Instead, the task made it necessary to personify absent voices and this swiftly created simplistic and negative assumptions about those who were not in the room. The more people tried to advocate on behalf of “the community”, the more stories were told and the more disparaging the caricatures of others as inconsistent, corrupt or myopic. Suddenly empathy morphed into an argument for intervention by those with a privileged vantage point.
Hmmm. Uncomfortable but it wouldn’t happen if there was better data…. I think faced with data of some kind, engineers and analysts can rush quickly to value-laden speculation and cariacature to find explanations. This possibility is so important and so freighted in this new Age of Inference - where machines don’t just do but decide - that this work uses animation to reverse the cariciature and bring the technicians into the light.
Toward a Critical Technical Practice
When Co-Risk Labs introduced me to the idea of critical technical practice, I thought it was more about the projects we deal with as engineers and how these can have consequences.
But when I did this analysis, I realised it is not just about what we make/do but how we make/do and how we talk about our work.
Imagine critical practice among the professionals and officials of a town that has experienced massive flooding: is there some way - through the use of images and exercises - to:
unearth and confront disparities of power and voice,
to distract the professionals from their own prior expertise and know-how and keep open the space for the ambiguities that they might find unsatisfying and inconclusive?
prepare the people expected to come up with solutions to be bold enough to do nothing or shut up or look sceptically at whose evidence and stories are being marshalled?
suggest ways to resist responding to all things human, psychological or precarious with engineered infrastructure?
No. But srsly. What about ‘proper coding’?
This is one of the hardest analytical tasks I’ve ever set my poor mind to.
As a technophile I have since wondered how much “the other coding” (e.g. natural languge processing) could have helped. If what Bill Gates recently said is true - “computers are so dumb they an’t even sort your email” - then no. It may be too hard to distinguish speaker from speaker for transcription by machine. And once transcribed the dataset of voices is too small and the dialogue too idiosyncratic to find patterns more meaningful than word or phrase counts. By contrast, the dataset of texts - the reports, the papers and the books - is potentially vast and far larger than anything one person could digest in a lifetime. But surely it would matter that the dataset is dominated by one language, a handful of places of publication and a small set of institutions?