Zoning and the Myth of Separation

SLOP2418 — Urban Design Studio
Week 3 · Marisol Quaye · 8 March 2027

direct readout · derived measure · worked hypothetical

The question this week

You have a block layout. You now have to say what goes in it.

  • Every cell you paint names a use. There is no neutral option in the palette.
  • Separating uses removes a nuisance and adds a distance.
  • Mixing uses removes the distance and returns the nuisance.
  • The question is not which is correct. It is which cost your district is better able to carry.

1898: separation as public health

  • Howard’s garden city arranged uses in concentric rings, with the factories on the outer ring and the housing held clear of them.
  • The diagnosis was correct for the conditions: coal smoke, untreated effluent, and no mechanism to control either at source.
  • Separation was not, at that point, a preference about neighbourhood character. It was a way of not dying.
  • Attribution: Ebenezer Howard, To-morrow: A Peaceful Path to Real Reform (1898).

1926: separation becomes law

A nuisance may be merely a right thing in the wrong place, like a pig in the parlor instead of the barnyard.

— Justice Sutherland, Village of Euclid v. Ambler Realty Co., 272 U.S. 365 (1926)

The court upheld comprehensive use zoning as a valid exercise of the police power. Almost every zoning code written since is a descendant.

What the Euclidean code actually did

The classic arrangement, set as a descending list, most protected at the top:

DistrictPermitsExcludes
U-1 residentialSingle dwellingsEverything below
U-2 residentialDwellings, two-familyCommerce and industry
U-3 commercialShops, offices, and the aboveIndustry
U-4 light industryLight manufacture, and the aboveHeavy industry
U-5 heavy industryEverythingNothing

The code is a pyramid. Protection accumulates upward, and so does distance from work.

1929: separation at neighbourhood scale

Perry's neighbourhood unitschool¼-mile radiusarterials carry traffic around

Traffic stays outside; school and open space sit at the centre.

1961: the case against

  • Jacobs’s first condition for diversity: a district must serve more than one primary function, so that people on different schedules are present at different hours.
  • A district with one function has one schedule. It is full for two hours and empty for twenty-two.
  • The consequences she names are livability consequences — surveillance of the street, the viability of small enterprise, whether anybody is about at nine in the evening.
  • Attribution: Jane Jacobs, The Death and Life of Great American Cities (1961).

2000: regulate form, not use

  • Duany, Plater-Zyberk and Speck argue that codes should control what a building does to the street — height, setback, frontage, parking placement — and leave what happens inside it largely alone.
  • The claim is that most of what separation was protecting was really form, and that use was a crude proxy for it.
  • The claim’s limit: form-based codes are weakest exactly where separation was strongest, which is airborne and waterborne nuisance.
  • Attribution: Suburban Nation (2000).

The palette you actually have

  • Residential by density — low, medium, high — plus row housing and mixed housing, which puts commerce beneath dwellings in one cell.
  • Commercial, low and high density. Office. Industry.
  • Cells attach to road frontage to a limited depth, so the block layout chosen in Week 2 has already decided how much of each you can paint.
  • Districts are painted over the top, and scope tax rates and policies to an area rather than to a use.

This is a use-based code with one mixed-use exception. Euclid would recognise it.

Two schemes for Halstead Reach

Both schemes use the same block layout and house the same population.

MeasureScheme A — separatedScheme B — adjacent
ResidentialThe terrace, north of the rail lineThe terrace, north of the rail line
CommercialOne strip on the collector, at the quarter’s edgeMixed housing on the two main frontages, plus corner cells
IndustryOmbersley Flats, 3.2 km south-east, across the KerrowEastern edge of the Reach, 0.8 km from the nearest dwelling
New infrastructure requiredOne river crossing, one arterial to itNone beyond the quarter’s own streets
Home-to-work trip, typical3.2 km, one crossing, one route0.8 km, several routes
Pollution exposure at dwellingsEffectively none — downwind, downstreamMeasurable ground and air footprint on the eastern blocks

Where the pollution goes

Two industrial sites and their pollution pathsReach counterfactualPellow cellsWNW wind ↘housing

Siting changes exposure; zoning colour alone does not.

Where the trips go

The same two schemes, read on the traffic volume overlay at 08:00.

MeasureScheme A — separatedScheme B — adjacent
Work trips crossing the collector mouthAll of themAbout a third
Routes available for the work tripOne, via the crossingThree or more, depending on Week 2’s block dimension
Peak volume on the busiest linkHigh and concentrated in two windowsLower and spread
What fails firstThe crossingNothing yet; Week 6 will find it
Fixable later byTransit — Week 5Not obviously anything

Before Tuesday, and Assessment 1

  • Build both schemes. Use a sandbox save for the second one; sandbox saves are never marked and are the fastest way to find out how a system behaves.
  • Bring four screenshots: ground pollution and traffic volume, at 08:00, for each scheme.
  • Bring the home-to-work distance for one named dwelling under each scheme. One dwelling, named, both schemes.
  • Commit to one scheme in your assessed save before Week 4.
  • The Assessment 1 brief is issued this week. Neighbourhood Unit Plan, due Week 4, 25% — fiscal 30, mobility 30, livability 40. Read Perry again before you read the brief.
  • Save as <student ID>-w03. Desk crits Tuesdays 13:00–16:00, Studio 2.14.