Beyond nine cities
Separate occupation phases, building episodes and ceramic phases. A new label need not mean a new population.
A place repeatedly rebuilt. A sequence repeatedly reinterpreted.
Research checked 25 September 2026Citadel and immediate surroundings · not the full lower town
Separate occupation phases, building episodes and ceramic phases. A new label need not mean a new population.
Keep an excavated sequence distinct from the calendar assigned to it. Approximate boundaries differ between published chronologies.
Damage, rebuilding and changes in storage have multiple possible causes. Their relationship to epic tradition is a further argument.
This teaching adaptation updates the interpretation of a published plan; it is not a new excavation survey. Broad architectural overlays are retained where the source does not distinguish subphases. No new walls or exact findspots have been invented for Troy 0, Proto-IV, or the 2026 agora report.
The default VI–VII chronology follows Pavúk’s 2020 synthesis. Alternative labels and dates are compared explicitly. Earlier and later periods use approximate teaching ranges; they do not form one uniformly dated statistical model. “Provisional” marks a proposed correlation or preliminary report, not a measured probability.
Original architectural geometry and selected point locations: ARCHAEOLOGY magazine / Archaeological Institute of America, Uncovering Troy. All rights in the source plan remain with its rights holders. This is an independent teaching adaptation, not an AIA publication or endorsement. New interface, chronology comparison and annotations prepared for HIST 304.
The original III–VI overlay is shown only as an architectural reference for VI; it does not provide distinct III, IV or V plans. The VII overlay cannot independently map VIIa, VIIb1 and VIIb2. Shared architecture is context, not evidence that every wall was built or occupied throughout the selected interval. The VIII–IX overlay is likewise composite.
Research cutoff: 25 September 2026. The 2026 agora report is kept separate from published stratigraphic models. Model date ranges are approximate and should not be read as calibrated confidence intervals unless expressly identified as such.