Yesterday’s Thoughts

February 2, 2005

Negative Assurance

Dave Pollard is blogging on the ten most important ideas of 2004 related to blogs and the Internet. He is thinking big. There is a more recent post here that covers business and the economy. A third post on politics and society is promised.

One point he touches on is a different way that blogs enable us to know.

In professional auditing circles it’s called ‘negative assurance’, and it means that sometimes you believe what you do in the absence of any compelling evidence to the contrary, if a lot of people have had the opportunity to proffer such contrary evidence. Auditors send out letters of ‘negative confirmation’ of account balances to their clients’ customers with the request that they be returned with corrections only if they’re incorrect. This is not as comforting as ‘positive confirmations’ where a written, signed response is required of each customer, but it’s much better than nothing, and usually very effective.

His example is cribbed from David Weinberger, who believes (“sort of, kind of”) that George Bush wore a radio during the first presidential debate. His evidence for this embarrassing belief? No one in the blogsphere has been able to prove that he wasn’t, despite significant efforts to offer up any other plausible explanation.

Three things about this concept are interesting to me.

Firstly, I am interested in the fact that the concept of negative assurance exists. I think that the principles of accounting offer lots of interesting ways to think about the world. I can’t find any online references that define negative assurance, but Goggle shows 3840 hits where “negative assurance” is used, on the first page or so almost all refer to auditing. I want to know more about how the concept is used, what are the acknowledged limits of the concept, and if there are other related concepts. I presume that double entry bookkeeping is a touchstone of the accountant’s epistemology, but maybe that is just folklore.

Secondly, I’d like to explore how commonly we use negative assurance in everyday life. No non-blog examples occur to me now, but that is because every time I try to trace on to the ground, there seems to be some other way of knowing involved. Maybe I don’t understand the concept well enough. Maybe it is important for the concept that there is this vast cloud of probabilistic information that is required as a background and most things that we know in everyday life have at least some small factual basis.

Thirdly, what are the limits of negative assurance. Can these limits be quantified? Weinberger believes that Bush was wearing an earphone, based on five or six pieces of inconclusive information, from sources that don’t look disreputable, and no definite contradiction. I believe that Karl Rove was behind getting the fake Bush National Guard memos into the hands of Dan Rather and CBS for more or less the same kinds of reasons, but my belief is weaker than Weinberger’s. Why is it weaker? Was there less attempt to refute my belief? Was there less traffic to the rumor? Have I done less research?

More to follow. I’ve made “Ways of Knowing” a category.

Sorry, comments for this entry are closed at this time.