I did my usual trick last night of putting CNN on to help me adjust to an eight hour time difference. I find this works well as you never sleep deeply under those circumstances. So when you wake up at 0130, but your body is expecting its Sunday morning Bacon & Egg Butty and a large Americano your mind starts to think about the days tasks, or the previous days disappointments (How did the Blues and Ospreys fail to win? When am I going to find the time to finish this article? Why didn't I knuckle down to my accounts last week? Why had I agreed to subject myself to BA's lack of service?) You know the sort of thing. Having CNN or the BBC 24 News (better as it repeats) on in background gives you something to engage your conscious mind and you slip back into sleep.

The one danger is that a news item can grip you, this time two did. FIrstly a bear pit of a Democratic candidate debate which was devoid of content and a poor advert for democracy. Secondly and in some ways more interesting a report that the Governor of Georgia's attempt to achieve rain through the power of prayer had been successful. This was debated at length on CNN this morning and with no sense of irony. In fact it got more serious consideration than the state of American Democracy.

Now if you do a google search on "prayers answered" you get about two million hits (a lot of them telling you how to achieve this for a fee. If you type in "prayers unanswered" then you get about a million and they are all selling systems or books to create success. While this may be an interesting twist on Weber's The Protestant Ethic and the Spirit of Capitalsm, it is more interesting for its concept of proof, and I am finally getting to the point here.

The basic argument seems to be that if Prayer works, then it proves the existence of God. If it does not then you, the supplicant are at fault. This is not difficult as the conditions for acceptance seem harsh and I find it difficult to belief that any State Governor satisfies all of them, or any other human for that matter.

Now for the connection. I hear a lot of case studies at KM conferences and elsewhere, and read about them in various popular and academic journals. From all I can see they follow the same form. Successful case studies are held up as evidence of an approach, method or software tool. If unsuccessful case studies are reported (and they rarely are) or presented then they are used as illustrations of how the approach, method or software tool was not used properly.

My general argument is that the number of reported successes is normally within the range of statistical probability for success by chance. For example I hear about many successes on CoP, I also hear about more failures but the former are reported not the latter. There is also the tendency to interpret the facts to achieve success.

The only real validity is if you predict an outcome and achieve it, or base the method or tool on scientific principles where the situation is complex with too many factors in play. Of course there is a place for faith in human evolution and for prayer/meditation, but its not the subject of empirical proof. Mind you in respect of drought and tempest alike it may be all the next generation (and possibly mine) has to fall back on, which brings us back full circle to American democracy.