Re: VMs: Re: NSU review of Rugg (2003)...

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Dear Gordon,

At 10:15 18/12/2003 +0000, Gordon Rugg wrote:
> I think a closer analogy would be if someone denied that a computer could 
> write
> poetry, and then someone produced a program which could write something 
> that was
> generally agreed to constitute a poem.
>
> That wouldn't prove whether a particular poem was or wasn't written by a
> computer, but it would show that computer-generated poetry was possible.

There's never been any doubt that a hoax was possible: what you've 
demonstrated is a *mechanism* by which a similar-looking hoax could have 
been carried out (but not currently the method by which the VMs *was* 
produced).

I'll freely admit that, because your approach is able to produce gibberish 
of roughly the same complexity and class of structure as observed in the 
VMs, this makes the hoax argument roughly as plausible (technically) as, 
say the verbose cipher argument.

However, because there is historical evidence which points to (roughly) 
1450-1470, surely the cipher argument seem the safer bet of the two? Even 
so, there are *plenty* of ways in which the VMs' text subtly differs from both.

> The VMS could contain ciphertext, but I
> haven't found any evidence that it does; the VMS could also contain 
> nothing but
> meaningless gibberish.

There is good statistical evidence of deep structure in the VMs, which 
stays fairly consistent across all pages. Have you considered the 
difficulty of maintaining this kind of consistency over multiple (possibly 
50+?) random tables?

> With Occam's razor, the simplest explanation now appears
> to be that the VMS is a hoax containing meaningless gibberish. That 
> doesn't rule
> out the possibility that there's a ciphertext in there, and I'd be 
> delighted if
> someone found one.

No, Occam's Razor refers to the principle that "plurality should not be 
posited without necessity" - and if you are merely prestidigitating the 
bulk of the complexity away into large non-random tables (which themselves 
currently aren't sufficient to reconstruct the VMs), that's an awful lot of 
plurality being posited.

I thought the whole point about your grille-based explanation was that it 
was a production-centric solution to the challenge of hoaxing a 200+ page 
manuscript? Without a parsimonious number of tables, surely the whole 
methodological raison d'etre disappears? OK, if a Language A table and a 
Language B table were sufficient to generate what we see, I'd agree with 
you - but just how many tables do you think it would it have needed?

Cheers, .....Nick Pelling..... 


______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list