Atlas: The Archaeology of an Imaginary City

Dung Kai-cheung

4 ideas

  1. History as a reconstructable, misbuildable object

    The past is not retrieved but assembled from surviving fragments, and the assembly can come out wrong while still looking whole. A city reconstructed from its maps can be internally coherent, richly detailed, and entirely false, so coherence is not evidence of truth.

  2. Reading maps as arguments, not mirrors

    Every map selects, names, omits, and orients according to whoever commissioned it, so it records a claim about a place rather than the place itself. Reading a colonial map this way shows the mapmaker's purposes, such as asserting possession, fixing boundaries, and ranking districts, more clearly than it shows the terrain.

  3. Place names as sedimented, unreliable evidence

    Place names outlive the conditions that produced them, so a street name can preserve a coastline, a landmark, or an event that no longer exists. Each name is a fossil that also misleads, because the name persists after its referent has shifted, been erased, or been translated into a different meaning.

  4. Fiction as method for disappearing places

    When a place's political identity is about to be rewritten, as Hong Kong's was at the 1997 handover, the writer imagines a future scholar excavating it as already lost. That projected retrospect makes the present strange enough to inventory. Mixing genuine sources with invented theories and false entries forces the reader to question every source, which enacts the city's own uncertain, constructed identity instead of simply describing it.

Save and mark ideas in the app