Many facts must be learned through intermediaries with special expertise or ac- cess to information, such as law enforcers, scientists, journalists, and public officials. This paper considers whether society can learn about such facts when intermediaries are devoid of ethical motives and make sequential public announcements. The answer depends on the degree of information attrition affecting the amount of discoverable evidence about each fact. Information attrition is nonexistent in fields based on repro- ducible scientific evidence but can affect the evidence in criminal and corruption inves- tigations. Applications to institution enforcement, social cohesion, scientific progress, and historical revisionism are discussed.