During the UNESCO WHIPIC Fellowship, visits to Doyang Gongso and Daejuk Gongso in Andong brought a different limitation into view. Through photogrammetry, close observation and available documentary information, much of the surviving physical fabric could be recorded: architectural form, material condition, thresholds, devotional objects and traces of earlier use.

What remained far less accessible to me as an outside researcher were the social relations through which those places had once been known. The buildings remained available for observation, but many of the everyday associations, memories and spatial narratives connected to their former communities could not be reconstructed from physical evidence alone.

The distinction became important: documenting what survives is not the same as recovering how a place was socially understood.

Later visits to gongso where community members were still present shifted the question again. There, memories and accounts could enter the documentation process through conversation and situated engagement. The comparison helped establish the methodological problem that SNIP would later test.

The question was therefore not whether community memory had simply “disappeared,” but what a researcher can responsibly know and interpret when the relationships through which that memory circulated are no longer readily accessible.

That experience became especially important because it exposed a distinction that would later become central to my thinking: to document a place is not the same as to interpret its heritage. The gongsos could still be modelled, measured and visually analysed. Their material condition, devotional objects, thresholds, interiors and traces of earlier use could all be recorded. What physical documentation could not provide on its own were the relationships through which these places had been known: histories of attachment, everyday patterns of use, local memories and the meanings associated with particular spaces.

In that sense, the Andong visit became more than a field exercise within the fellowship. It became a methodological precursor. It confronted me with the limits of documentation when physical fabric survives but the social relationships through which its meanings circulated are no longer readily accessible to the researcher. The question was no longer only how to record architecture more completely. It was whether heritage interpretation could be adequate if it relied primarily on the surviving physical record.

This tension stayed with me. Later visits to other gongsos where community members were still present shifted the problem again, showing what became possible when accounts of memory, use and attachment could enter the research through situated conversation. In Andong, however, the problem appeared in a particularly clear form: what can a researcher responsibly interpret when a place remains physically accessible but the relationships through which it was socially understood are not?

That encounter became foundational to the methodological question that SNIP would later test: how might heritage documentation bring situated knowledge into relation with the physical record without treating either as a complete account of the place?