[split] Authority and the Appeal to it.

[Jorge_Stolfi]

RadioFM: if I ever wanted to do any research that really depends on the scribes, I would simply call them five hands and treat them as hypothetical. Unless I develop a Theory™, the difference between five people and one person will not matter to me. I can't imagine one good reason why this question should come into play for statistical analysis.

I can think of two reasons why it might affect some analyses:

• If one is trying to use evolution of statistics and/or handwriting to determine the order in which the Scribe(s) transcribed the material to vellum, and one suspects that there were multiple Scribes, then one should analyze the pages presumably scribed by each presumed Scribe separately from those of other Scribes.  Because the evolution of the handwriting or Scribe-dependent statistics (like glyph spacing, abbreviations, errors) for each Scribe would have no relation to that of other Scribes.  Such analysis simply cannot tell the order in which different Scribes worked.  They may well have worked in parallel.
  
• I believe that the Scribe of the Stars section often confused "in" with "r" and "d" with "k" when reading the draft, because they looked similar in the Author's handwriting.  If one is trying to correct such presumed Scribal errors in order to make word statistics more meaningful, one may need to have a different "spelling corrector" for each Scribe.
  

About older experts: science is generally better now than it was 100 years ago.

Science today is humongously better.  But are the scientists better, on the whole?  Can we trust today's "experts opinion" more than that of the experts of 100 years ago?

In some areas, sure.  In astronomy, even a high school amateur astronomer today is more trustworthy on most questions than the best astronomers of 1926.  Ditto for biology -- although in some specific fields, like identifying species of insects or fungi, today's experts may be less reliable than the old ones.

But in electromagnetism, chemistry, mathematics?  I wonder.. the old experts may have had better grasp of the fundamentals than the modern ones, because these tend to specialize on narrow obscure fields.  And I still cannot wrap my head around how Gauss, Euler, and several other mathematicians of 200+ years ago could produce more revolutionary results than 10000 tenured professors could do.

An then in areas that depend heavily on human evaluation and judgement, like history, linguistics, architecture, art, etc -- and paleography, and handwriting analysis -- it is not obvious that today's experts are more competent than those of 100 years ago.  

Even because today many experts fall for the temptation of "advanced computer methods" that they don't quite understand.  Like collecting a bunch of measurable statistics of dubious quality and significance, and feeding them to PCA/MDS/k-Means/LSQ etc...

And what you mention wasn't a peer-reviewed article. I would love to see another paleographer publishing their findings formally, then we could compare.

But was the Five Scribes claim, specifically, reviewed by other handwriting experts who looked at the VMS images?  Or was the article reviewed only as a computer science / data analysis one?

All the best, --stolfi