The Knowledge Illusion

Steven Sloman and Philip Fernbach examine the gap between what individuals understand and the knowledge available in their surroundings and communities. Their publisher's description introduces this account of shared knowledge. A useful reading question is where your own explanation ends and reliance on someone else's expertise begins.

Try explaining a familiar system

In this original exercise, take a library reservation. You know how to click “Reserve,” and you expect a message when the book is ready. Before reading instructions, describe what you think happens between those two steps. Does the system reserve a specific copy? What happens if it is overdue? How is your place in the queue decided?

You may know the procedure well while being unsure about its operation. That gap is not a failure to be embarrassed about. It tells you which questions require documentation or someone who maintains the service.

Make a knowledge map

Divide your notes into three parts: what you can explain, what you can verify in a document, and what needs a knowledgeable person's answer. For example, the website may explain collection deadlines while a librarian needs to clarify how multiple branches share copies.

Then revise one sentence. Replace “I know how the reservation system works” with “I know how to reserve a book and where to check the collection rules; I do not know how the queue is implemented.” The second statement is less sweeping and more useful when someone asks for help.

Dependence is not the same as unreliability

Nobody needs to understand every layer of a service to use it responsibly. The practical question is whether the people or records you rely on are appropriate for the question. Three friends repeating one rumor are not three independent sources. A current operating policy may be a better source for a deadline than a confident recollection.

Choose this book for reflecting on the boundary between familiarity, explanation, and shared expertise. For checking a specific claim you find in an article or discussion, continue with How to Check a Research Claim Before You Trust It.