The Rhetoric of the Graticule

A rectangle on a screen looks like a unit of space. Move the same rectangle north on a world map and the promise may fail. The pixels are unchanged; the ground area they suggest is not. My concern begins with that ordinary handoff: a display convention becomes an apparently neutral instrument of comparison.

Snyder’s USGS manual identifies what projections preserve and sacrifice. For spherical Mercator, local linear scale is sec φ, where φ is latitude; local area scale is sec² φ. At 60°, those factors are 2 and 4. These are local differential values under the spherical model, not exact multipliers for an arbitrary country spanning several latitudes. A classroom comparison should state that distinction before asking the reader to judge the picture.

Here is a constructed test. Put two small equal ground patches at the equator and at 60° north, use the same Mercator parameters, and compare their projected areas. The northern patch approaches four times the equatorial patch as their sizes become locally small. That ratio follows from the model. It is not a measured survey, and it does not establish the motive of whoever chose the map.

Esri documents Web Mercator as the default coordinate system for ArcGIS Online web maps and describes analytical tools that use local projections or geodesic distance. The distinction matters: a familiar basemap need not determine the calculation. A workflow can retain a consistent display while using a different geometry for a buffer or area result (Lange, 2020).

This is a documented institutional use, rather than a claim that every web map implements the same measurement policy. EPSG:3857 is commonly called Web Mercator; it uses a spherical projection calculation with WGS 84 coordinates. The ideal spherical scale example above should not be sold as a full ellipsoidal accuracy statement for that implementation.

Imagine a planning team comparing two candidate service areas. If it measures the colored shapes with screen pixels, the display participates directly in the allocation. If it computes ground area with an appropriate method and discloses it, the same basemap can support a defensible comparison. The institutional question is who receives the method, who can challenge it, and which decision depends on the visual impression.

Distortion alone does not prove colonial intent. Establishing intent requires evidence about a particular commissioner, edition, audience and use. Establishing a political effect requires evidence about what people understood or what institutions decided. Geometry supplies a constraint; it cannot supply those histories on demand.

An equal-area projection is therefore a repair for a particular comparison, not an ethical exemption. It can preserve area while producing unfamiliar shapes, exclude settlements from the dataset, or obscure uncertainty with a precise boundary. A conformal map can be an appropriate tool for a local task. The useful criticism names the task and the sacrificed property.

I would put the measurement method beside the result: coordinate system, projection parameters, planar or geodesic calculation, relevant extent and input date. I would provide an equal-area comparison when the argument concerns relative size, and test whether the decision changes. A disclosure buried in a technical footer is less useful than a method attached to the value it qualifies.

The graticule becomes rhetorical when its regularity lends certainty to a claim it cannot support. The answer is neither to abandon grids nor to declare them innocent. It is to make the transfer from picture to judgment inspectable. The reader should be able to tell which part is mathematical, which part is documentary, and which part is my interpretation.