The contribution that makes the edge of an analysis something someone answers for. Every analysis leaves people and things out. This one asks who they are, and who carries the cost of having been left out.
Who and what is outside our frame, and who bears its costs?
Responsibility for framing treats the drawing of a boundary as an accountable act. Cilliers showed that every model of a complex system must leave something out, so choosing a frame is an ethical act and not a technicality. Ulrich’s boundary critique, Jasanoff’s technologies of humility, Funtowicz and Ravetz’s extended peer communities and Escobar’s question about whose world a design serves all build on that point. Complexity science is normatively thin. This contribution supplies a discipline it lacks: an analysis must account for what and whom it leaves out, and for who bears the cost of its framing. What it adds is legitimacy. It is weakest when it shrinks into a consultation exercise that changes nothing.
Outside school
A town redesigns its bus network. Success is measured by average travel time, and the new network wins. The routes are straighter, the stops fewer, the trips faster. The figures are honest. But the people in them are the people who still ride the bus.
Outside the figures are the people who stopped riding, or never could. The parent with a pushchair whose stop was cut for being too close to the next one. The man who cannot walk the extra four hundred metres. The night cleaner whose shift ends after the last bus. None of them was asked, because the consultation was an online survey of passengers, and they were no longer passengers. The faster network was partly paid for by them.
Cilliers argued that every model of a complex system leaves something out, so choosing a frame is an ethical act, not a technicality. The travel-time measure was not wrong. It was a boundary, drawn by someone, and it decided whose trips counted. Ulrich’s boundary critique turns this into questions anyone can put to a plan: whom it serves, who decided, and who is affected without having a say.
The everyday habit is small. Before judging a plan by its measure, ask who sits outside the measure and what the plan costs them. Then ask them, at a time and in a place they can actually reach. Escobar pushes the question one step further, from who is missing to whose world the design assumes in the first place.
The term
Every page follows the same Singapore secondary school through the same semester, Term 1 and Term 2. The school is changing how it assesses, both of learning and for it: fewer weighted assessments, more feedback, and ratings that are meant to say something. The five people and the events of the semester stay fixed. Only the contribution reading them changes.
All five, interconnected
The five threads interleaved across the twelve weeks, so the hand-offs between them are visible. Pick a person to follow their term on its own. The people are illustrative, and the same five appear on every contribution’s page.
Before Term 1
Mr Ravi, reading the plan
Two years ago a triple-marking policy drove two teachers out of English. It was written by people who did not mark under it, and its cost landed on the people who did. Mr Ravi reads “every marked piece of work carries a written next step” and does the sum for a science teacher with five classes of forty. Nobody in the working group has done it.
Term 1, Week 3
Mrs Lim and Mr Ravi, at the first review meeting
The review group is the group that wrote the plan: Mrs Lim, the HODs of English Language and Mathematics, and the teacher who manages the school’s assessment data. Mr Ravi, added late, reads the plan for who it names. The cluster superintendent’s office and the tracking company appear by name, students appear as a scale, and parents appear only as readers of the app. Nobody objects, because nobody outside the room knows there is a room.
Term 1, Week 4
Ms Tay, sending out the twice-a-term return
Ms Tay sends the twice-a-term ratings return to twelve schools: one row per level, five columns from Emerging to Exceeding. She inherited the form from her predecessor and has never asked who it was built for. This week it adds science, rated against the end of the year, to English, rated against the current unit, as if the two meant the same thing.
Term 1, Week 6
Mdm Rohani, at 3am on a break
Mdm Rohani reads the class parents’ chat group, where families are signing up for tuition centres that set practice papers. She looks up the price, sets it against the month and closes the tab. The plan took weighted assessments out of school, and the families who could pay put them back at home.
Term 1, Week 8
Mr Ravi, after the first book check
The book check rates humanities “inconsistent”, and one department loses a weekend to longer comments. Mr Ravi hears it from a colleague who was in English two years ago. The plan had a target for books and no line for hours, so the cost of being wrong about workload landed where triple marking’s had. He counts his own department’s marking time and finds whole-class feedback the only thing keeping it down.
Term 2, Week 2
Aisyah, in the fortnightly quiz
Aisyah likes the quizzes because they do not count. Getting the energy question wrong costs her nothing, and the next lesson starts from it. One class asks for the quizzes to count. Aisyah does not, because she has worked out who pays when a test counts.
Term 2, Week 2
Mrs Lim, with the allied educator for learning support
The allied educator shows Mrs Lim that students still building their English are rated Emerging in nearly everything, because the ratings judge written explanation. The scale was built to be consistent, and it is, at their expense. They are the students least placed to argue with a rating, and their families go home with the same word in every subject. The allied educator joins the review group.
Term 2, Week 4
Aisyah, at the first checkpoint
Aisyah freezes in the timed English paper and is rated Emerging, after a semester of strong work in class. Nobody asked her form teacher before the rating went home. The format was chosen for consistency, and a timed paper suits students who are calm under a clock. She carries the cost of a choice she did not know had been made.
Term 2, Week 4
Mdm Rohani and Mrs Lim, through the app
Mdm Rohani complains through the app after a night shift. Mrs Lim reads it the same week she learns that a teacher has quietly rated borderline students up to avoid more complaints. The students moved up are the ones whose parents complained. Most families of students still building their English do not use the app, so their children stay where the scale put them.
Term 2, Week 5
Ms Tay, reading the cluster newsletter
The newsletter cites the school as an exemplar of “feedback culture”, with 96 per cent of books compliant. Ms Tay has read the school’s notes on the ratings and the book check, and she asks of her own return whose world it was designed for. It fits the cluster’s report upwards. It has no place for a student who freezes under a clock, or one writing in her second language.
Term 2, Week 8
Mdm Rohani and Mr Ravi, on a Saturday morning
The Meet-the-Parents session has a Saturday morning slot, and Mdm Rohani comes for the first time in two years. Mr Ravi shows her Aisyah’s quiz record, and she sees learning the Emerging rating did not show. Nothing in the ratings could have told her that. By Monday, two parents who work Saturdays have written to say they now cannot come.
Term 2, Week 10
Mrs Lim, writing the end-of-semester note
Her note adds a list nobody asked for: who wrote the plan, who joined the review and when, and who is still outside it. The allied educator joined in Term 2, Week 2, and shift-working parents were heard for the first time in Term 2, Week 8. Still outside are the families paying for tuition, the parents who work Saturdays, and whoever is writing the guidance on reporting for lower secondary, promised for this year and still not arrived. The scale parents see is cut to three points, and the one staff see is not.
Term 2, Week 10
Ms Tay, drafting a proposal
Ms Tay proposes that the cluster stop collecting twice-a-term ratings. She cannot yet say what should replace them, and says so. Her draft asks that any new return be designed with schools, students and parents in the room. The cluster superintendent’s office still wants the compliance figure.
The plan was drawn round the people who wrote it, and the cost of its being wrong landed outside that line: on a student who freezes under a clock, on students rated in a language they are still learning, on a nurse who could not reach a weekday Meet-the-Parents session, on the teachers who wrote the book-check comments. Each time the boundary moved, the school learned something its ratings could not show it. None of the moves was free. The Saturday slot let Mdm Rohani in and shut other parents out, and the ratings that moved after complaints moved only for families who complained.
Implications
The same five dimensions for all seven contributions. Read them across on the compare page.
From treating analysis as neutral to treating the drawing of boundaries as an accountable act. Every analysis has an edge. What is inside it gets counted. What is outside becomes context, or disappears. Usually the edge is drawn by whoever is doing the analysis, for convenience, and then forgotten.
Cilliers made the argument that matters most here. Complex systems are incompressible, so any model of one must leave something out, and what it leaves out can matter in ways nobody can know in advance. If exclusion cannot be avoided, it has to be owned. Choosing a frame becomes a responsibility, not a technicality.
Most ways of thinking take the boundary as given. Analytical thinking works on the parts inside it. Design thinking asks what the user needs, once someone has decided who the user is. Systems thinking knows a boundary is a choice, but tends to make the choice once and then model inside it. Critical systems thinkers have long warned that system maps depoliticise, because whose purposes define the system, and whose knowledge counts in modelling it, drop out of view.
Complexity science does not fix this on its own. It is descriptively strong and normatively thin: it explains how patterns arise and is silent on which ones are desirable. This is where complexity thinking parts company with complexity science. It turns epistemic humility, the admission that every frame is partial, into an ethical discipline with a question attached. Who bears the cost of the frame?
Legitimacy. An analysis that has accounted for its exclusions is more defensible when it is contested, because the objection has usually been heard already. It is also more likely to hold when authority itself is in question. Funtowicz and Ravetz argued that when facts are uncertain, values in dispute, stakes high and decisions urgent, quality assurance has to move from peer review to an extended peer community of those affected.
It adds information too. The people outside a boundary often know things the people inside cannot. A plan judged only on its own measure looks after the measure, and the costs it pushes outward come back later as complaints and workarounds, filed under unintended consequences.
A list, kept up to date, of who wrote the plan, who has been consulted, when and how, and who has not. A meeting at five on a Tuesday consults the people who are free at five on a Tuesday. An online survey of users reaches users. The method sets the boundary before anyone has asked a question.
It also asks for a distributional check on every significant decision: who gains, who pays, and who would never show up in the figures either way. And it asks the school to bring in knowledge it does not have, from students, families, the allied educator for learning support and the teachers who do the marking, and to be honest about the bodies outside its reach, such as a national body whose guidance on reporting sets the terms for everything inside.
The consultation ritual. A survey goes out, a forum is held, a stakeholder map is drawn with every group on it, and nothing in the plan changes. The boundary looks wider and is the same. The test is simple: name one decision that went differently because of the people you added. If there is none, they were consulted but not included.
There is a second failure, in the other direction. Boundary critique can grow until nothing can be decided, because someone is always outside. Every boundary excludes. The discipline is to draw it on purpose, say who it leaves out, and redraw it when the costs show up. It does not settle whose interests should win. It only makes sure the question is asked in the open, and I am not sure that is enough.
Methods
The repertoire for this contribution, with an evidence rating for each method. Strong means tested in more than one field. Emerging means promising, and mostly case studies.
Boundary critique
ModerateUlrich’s questions put to a plan: whom it serves, who decides, and who is affected without a say. Most useful asked early and asked again.
Extended peer review
ModerateThose affected review the analysis alongside the experts, as Funtowicz and Ravetz proposed for problems with high stakes and disputed values.
Participatory framing
ModerateAffected people help set the question, not only answer it. Slow and costly, and the only route to some kinds of knowledge.
Distributional checks
ModerateWho gains and who pays under each option, including people who never appear in the figures.
Reading
Start with the first on the list.
Cilliers, P. (1998). Complexity and postmodernism: Understanding complex systems. Routledge.
The argument that every model of a complex system excludes something, so choosing a frame is an ethical act.
Ulrich, W. (1983). Critical heuristics of social planning: A new approach to practical philosophy. Haupt.
Boundary critique: the practical questions that make a plan’s boundary judgements explicit.
Funtowicz, S. O., & Ravetz, J. R. (1993). Science for the post-normal age. Futures, 25(7), 739–755.
Why, when stakes are high and values disputed, quality assurance must extend to the people affected.
Jasanoff, S. (2003). Technologies of humility: Citizen participation in governing science. Minerva, 41(3), 223–244.
Technologies of humility: institutions that admit the limits of expert knowledge and make room for citizens to take part.
Escobar, A. (2018). Designs for the pluriverse: Radical interdependence, autonomy, and the making of worlds. Duke University Press.
The question every design answers whether it asks it or not: whose world is this for?
Jackson, M. C. (2019). Critical systems thinking and the management of complexity. Wiley.
Critical systems thinking on power, and on how system maps can depoliticise the choices behind them.