Land ownership, displacement, informal housing and political contest — the things the simulation abstracts away, and why knowing your model's edges is design literacy.
17 May 2027
The argument#
The land-value curve the studio has been raising since Week 1 is an accurate output of a model that has no landowner, no tenure, no eviction, no informal housing and no objector in it — and a designer who cannot list those omissions is not qualified to cite the curve.
What you will learn#
- State, in writing, four specific things the studio’s model does not represent, and what each omission does to a claim the plan wants to make.
- Distinguish the four forms of displacement in the literature, and identify which of them the model’s household and homelessness figures can and cannot count.
- Trace the eleven-week land-value rise in the Kerrow corridor to the public investments that produced it, and say who would capture that uplift outside the model.
- Write the omissions appendix required in the capstone defence memo, to the four headings given in slide 17.
Theory into practice#
- James C. Scott, Seeing Like a State (1998). Administration requires legibility, and legibility is achieved by thin simplifications — the cadastral map, the standardised unit — which then become the only reality the administration can act on.
- Peter Marcuse, “Gentrification, Abandonment and Displacement” (1985). Displacement has four distinct forms — direct last-resident, direct chain, exclusionary, and displacement pressure — and three of them leave no mark in any count of who currently lives somewhere.
- Ananya Roy, “Urban Informality: Toward an Epistemology of Planning” (Journal of the American Planning Association, 2005). Informality is not a sector sitting outside the plan; it is a mode of urbanisation the state itself produces through what it zones, services and tolerates.
- Douglass B. Lee Jr., “Requiem for Large-Scale Models” (Journal of the American Institute of Planners, 1973). Large urban models fail characteristically by being comprehensive — too coarse at every point to be right anywhere, hungry for data nobody has, and complicated enough to hide their own assumptions.
- Naomi Oreskes, Kristin Shrader-Frechette and Kenneth Belitz, “Verification, Validation, and Confirmation of Numerical Models in the Earth Sciences” (Science, 1994). Models of open systems cannot be verified; agreement between model and observation is confirmation, and confirmation is not proof of correctness.
The deck walks through where each of these lands, slide by slide, against the model’s own omissions.
Simulation mapping#
| Theory move | Cities: Skylines II mechanic | What students should watch happen |
|---|---|---|
| Scott’s thin simplification | The land value info view — a computed field on a tile, with no owner attached to it | Land value rising for eleven weeks with nobody in the model to whom the rise accrues |
| Georgist uplift from public investment | Land value responding to transit stops, park entries, service coverage and trunk upgrades the player funded | The 2027 curve tracking the studio’s own capital programme almost exactly |
| Marcuse’s direct last-resident displacement | Household wealth, rent, and households leaving a building whose rent has passed their means; the district homelessness figure | The corridor’s low-wealth household count falling while total households rise, and the corridor’s livability reading going up as it happens |
| Compulsory acquisition and the objection | Bulldozing an occupied residential building; zoning change applied instantly by one actor | No hearing, no notice period, no compensation, no objector, and no delay |
| Roy’s produced informality | Unzoned, unserviced land | Nothing is built there, ever, under any pressure of demand |
| Lee’s hypercomprehensiveness | The simulation’s single integrated economy, traffic, service and land-value solve | Every claim you make being downstream of one set of coefficients you cannot inspect |
Evidence standard. Named interface fields are direct readouts. Calculations shown from those fields are derived measures. Every fixed scenario value on this page is a worked hypothetical, not a captured run. See resources and evidence for access arrangements and worksheets.
The decision this week#
Mobility performance against livability. The forced decision is the last four blocks at the southern end of the Kerrow corridor, currently medium density, currently housing the largest remaining concentration of low-wealth households in Halstead Reach.
- Upzone them. Residential density at the corridor’s southern terminus is what the transit spine’s ridership has been short of since Week 5. Upzoning closes the gap. Mobility performance rises, and it rises measurably.
- Hold them. Ridership stays below target and the spine’s operating subsidy stays where it is.
The complication, and the reason this decision belongs in this week rather than in Week 5: if they upzone, livability reads as rising too, because the corridor’s livability score is computed over the households that are there afterwards, not the ones displaced. Both metrics improve on the readout. The tradeoff is real and the model cannot display it.
So the studio requires the trade to be recorded by hand: a student who upzones must write what mobility performance was bought with; a student who holds must say what the spine’s subsidy costs fiscal performance. The move is defensible either way; the unrecorded move is not.
Into the studio#
- Arrive with the running Halstead Reach save, and with the land value overlay and the household and rent readouts for the Kerrow corridor already recorded for populated Week 4 and the current state. Week 1 was unbuilt: it supplies a land baseline, not household counts. Missing historical figures stay missing.
- The session’s first hour rebuilds slide 8’s table for each student’s own corridor. Tables may differ in magnitude and direction: explain the comparison rather than forcing it to match the hypothetical.
- The second hour takes the four-block decision and drafts the memo sentence that records it, in the metrics’ own words. Draft it in the session, because it is the sentence students otherwise write at midnight next Sunday.
- The omissions appendix is drafted to its four headings before anyone leaves. Desk crits on Tuesday will read it.
Assessment connection#
Feeds the Comprehensive City Plan (Week 12, 50%, holistic) directly and is the only week that adds a required section to the memo. Holistic marking means there is no line item for the appendix; what it means in practice is that a plan whose claims exceed its evidence is marked down across the whole document, and the appendix is the cheapest available protection against that.
Retrospectively, this week is the reason the two earlier assessments were marked on stated criteria rather than on the model’s raw numbers. Worth saying so out loud: the marking scheme has been a claim about what the metrics can carry since Week 1.
Teaching team
- Marisol Quaye — convenor