The burden of proof is usually described as the combination of two things:
(1) the burden of production; and
(2) the burden of persuasion.
The second element simply states that the status quo will be accepted unless and until the proponent of any change/policy/idea shows that sufficient evidence supports it. This is the classical "he who asserts must prove" element.
But the first part of the burden also requires a proponent to
produce evidence. The proponent of a new or original idea/policy is the one charged with coming forward with evidence.
In this case, "prove someone committed fraud 100 years ago" is not only a pretty unreasonable request from the start, but is also a reversal of this burden.
The problem with Ms. Piper is similar to that encountered by Tony Youens and Stumpy when dealing with the "Jacqui Poole" case, and that one was only 18 years old, as opposed to almost a century. The biggest problems are that the witnesses are dead and there is no way to set up a controlled test anymore.
Gardner wrote about Piper some time ago and gave some information that he considered indicative of fraud -- primarily (IIRC) "fishing" inquiries from the transcripts. He also pointed out some possible avenues of receiving information on the part of Ms. Piper, both through her husband and servants. Likewise, the fact that "unkown" sitters were brought was considerably less impressive when a number of them were prominent in the community and could have easily been recognized (and a thick file could have presumably been collected on such persons from public sources quite easily). Without the source documents, though, I have no way of knowing how sound these criticisms are.
Also, while several investigators are called "skeptical," most people identified as skeptics were only skeptical of the mediumship claims -- their preferred theory was some sort of telepathy. This is not quite the same type of skepticism as most readers might expect.
Given the difficulties in (or impossibility of) making anything resembling a thorough investigation; given the lack of source materials (in my home city, at least); given the time and expense necessary in trying to locate and review such materials; and given the very likely scenario that few would pay attention to -- much less be convinced by -- an analysis done a century (plus) after the fact, I just don't feel too enthused about being given a homework assignment of "proving" fraud that may have occurred 120 years ago.
It is not the burden of anyone to prove Ms. Piper a fraud. If something outside the normal realm of accepted physical laws occurred -- whether PSI or mediumistic -- it is the proponent's burden of proof.
Unless someone makes such a study worth my while -- say paying for a leave of absence so I can travel and spend 6 weeks on the original source documents, I decline.
N/A
PS: Garrette had some crticisms of the Gault book that were referenced here (there is a link to his original post, I think):
http://www.randi.org/vbulletin/showthread.php?s=&threadid=30101&highlight=piper