Memory Practices in the Sciences

The problems of perfect memory

from chapter 4: The Mnemonic Deep, p.174

... More deeply, they resolve into theoretical questions of just what d inow about the past, and how do we know it. We do not want a world with perfect memory:

• Perfect memory has a high overhead: It would be nice if we could preserve all external murals in Italy, but this militates against our action in the present (we can't paint our current houses).

• Perfect memory is not what it seems: The Ise Shrine in Japan has been torn

and rebuilt every twenty years since AD 652 using the same tools and skill set; it is recognized as the oldest temple in the country. What is being preserved here is not the ding an sich (which creates a legacy of preservation techniques) but the mode of building (which creates a legacy of organizational forms). The overhead problem of course recurs at this level.

• Perfect menory does not matter if no one listening to your stories: The "archival literature" in science is written as if someone someday will have time to go back and read all this welter of material and make sense of it - assigning priority, determining value, and so forth. This is the secular version of the Last Judgment - and is equally dependent on an Entity capable of massive datata

storage and analysis. There is no evidence that this Entity is in the process formation.