The Impact of the Mongol Conquests on Earthen Cities in Central Asia

TL;DR

The article argues that in Central Asia’s earthen cities, the Mongol conquests’ main impact was depopulation and the disruption of maintenance cycles rather than literal wholesale demolition, because mudbrick architecture decays rapidly without constant care and is difficult to destroy by fire. Using Merv and Otrar, it contends that archaeological evidence shows abandonment, shrinkage, and later recovery more clearly than Mongol destruction.

Key findings

• No convincing archaeological evidence of Mongol-period destruction has been found at Merv or Otrar: citadels and defences remain visible, no fire damage was evident on the tomb of Sultan Sanjar before its 1990s restoration, and no mass graves have been convincingly found. Instead, Merv shows desertion—including ceramics workshops and houses abandoned with some haste—and continued occupation from the thirteenth to fourteenth centuries, while Otrar has abandonment deposits and radiocarbon dating suggests abandonment began more than fifty years before the Mongol armies arrived.
• Mudbrick is nonflammable, and demolishing mudbrick structures would have required significant manual labour, more than the fast-moving Mongol army would have had time or manpower for; historical accounts of razing and levelling should therefore be reconsidered alongside environmental impacts and the flight of populations.
• Earthen buildings required continuous maintenance: ethnographic studies of mudbrick buildings in use in 1960s north-eastern Iran indicate a lifespan of fifty to one hundred years with proper maintenance; Otrar residential phases tend to last around a century, radiocarbon dating shows three overlying mudbrick house phases span a 150-year period, and a palatial structure at Merv was occupied from the mid-twelfth century into the fourteenth.
• Abandonment and city shifts can be pragmatic responses to erosion or water supply: Merv’s shift from Gyaur Kala to Sultan Kala began in the eighth century at least partly due to difficulties providing water, and later settlement moved to Abdullah Khan Kala nearly two centuries after Merv’s decline; Otrar recovered and flourished in the fourteenth and fifteenth centuries.

Methods

• Qualitative synthesis of archaeological, architectural, and historical evidence, focused on earthen construction techniques—mudbrick, pisé, and pakhsa—and urban occupation at Merv, Turkmenistan, and Otrar, Kazakhstan.
• Draws on prior excavations, radiocarbon dating, coins and pottery, microstratigraphic analysis, ethnographic studies, and medieval accounts such as Juvaini, Rashid al-Din, and Ibn al-Athir; the author states that a direct comparison across stone, fired brick, timber, and earthen architecture would require a larger dataset and more complex analysis than possible in this study.

Limitations

• The paper states that a detailed, direct comparison between stone, fired brick, timber, and earthen architecture would require a larger set of data and more complex analysis than is possible in this study.
• The paper notes that mudbrick structures are difficult to identify and excavate, especially across hundreds of hectares and metres of vertical stratigraphy, and that historical accounts are debated while archaeological evidence for destruction and victims is lacking.
• Evident from the text: the argument centers on two case studies, Merv and Otrar, so broader generalization across Central Asia is limited; it relies on existing excavation and historical sources rather than new systematic comparative data.