I’ve been reading the late Mark Perakh‘s book Unintelligent Design on the train into work the last few days. His first chapter lands a long series of hard punches on William Dembski’s neo-creationist glass jaw: it’s a good read, even though pitting a properly sharp physicist against someone who merely mimes mathematics and logic does make for a fairly one-sided bout.
(Oh, and if you don’t know about Dembski, Behe, Johnson et al, they are “Intelligent Design” Christians aiming to ‘prove’ that DNA cannot have been formed by pure chance; that the biochemistry of life cannot have incrementally evolved into its current state; that Darwin and neo-Darwinians was/are all Just Plain Wrong; etc etc.)
In his books, Dembski uses a broad set of structural and logical arguments to try to categorise the kind of thing DNA is, in terms of probability and complexity.
Fairly unsurprisingly, Perakh rips these artificial categorization schemata apart, by demonstrating with numerous examples (particularly the Voynich Manuscript, nicely enough) how real life things fail to fit Dembski’s neatly-made (but false) pigeonholes, as well as how Dembski’s conceptions of probability and complexity simply don’t work the way he seems to think they do.
But for all Dembski’s (numerous) flaws, he does employ one particular analogy which amused the heck out of me, and yet also challenged me to properly think its implications through. (Though not about his hopeful brand of Intelligent Design, I hasten to add.)
One of Dembski’s tricksy categorization hacks involves trying to differentiate between genuine patterns (which he says are the result of what he calls “specification“) and fake patterns (which he says are the result of “fabrication“). His much-quoted example colourfully compares a true archer who causes his arrow to hit a genuinely pre-drawn target (specification) with a fake archer who shoots his arrow anywhere he pleases into a wide wall and then proceeds to paint a target around the landed arrow to retrospectively ‘prove’ his initial skilfulness (fabrication).
I like Dembski’s fabricating archer as an antipattern – a recurring pattern of wrong-headed and/or self-defeating behaviour that, once named, becomes painfully obvious all around you. I mean, haven’t we all met plenty of fabricating archers in our lives? By which I mean people who try to add imaginative ‘fabric’ to their otherwise substance-less and evidence-free arguments.
Perakh also uses the better-known phrase “just so stories” in his Chapter Two, but that’s actually a phrase for ad hoc narratives purporting to explain something that manifestly is the case, such as “how the elephant got his trunk”, or perhaps “how the wooden politician got his long nose”. What I’m talking about here is something slightly more virtual: plausible-sounding narratives concocted to try to justify improbable (or indeed impossible) claims.
The messily rubbish world of Voynich Manuscript theories has long had a glut of these fabricating archers, constructing their post hoc secondary narratives to support a badly chosen and/or emotionally-invested initial position. However you try to pass off this process (‘lateral thinking’, ‘abductive reasoning’, “Ockham’s Razor” or whatever), it really all boils down to nothing more than painting your made-up target on the wall after you’ve shot your little bolt.
Look, (they say), this constellation of secondary stories I made up clearly demonstrates how close I was to the mark in the first place. Oh, and don’t listen to all those other fabricating archers, their post hoc stories don’t have even half the explanatory power of my post hoc stories.
And how many overdressed little bolts masquerading as supposed big shots do you think I’ve seen, hmmm? Perhaps a more difficult challenge would be to list how many Voynich theories you can name that don’t fit this dismal pathology?
Of the recent wonky crop, Tucker and Talbert’s article certainly follows it, as does Stephen Bax’s theory (he seems eager to get into a rebuttal posting war, but what was that American phrase about not getting into a pissing contest with a fire-hose? Spare me from Voynichological fire-hoses, O Lord!) and indeed pretty much all of the others.
However, the disappointing truth is that for all these Voynich theorists’ wobbliness, excessive hopefulness and sparseness of evidence, they still remain rank outsiders in the fabrication department. They’ve been beaten by what can only be described as a class act…
The #1 Voynich Fabricator
Cipher Mysteries regulars will probably have already worked out which particular Colossus stands atop the list of Voynichological fabricating archers (and by a mile): Professor Gordon Rugg. Despite authoring a 2013 book about blind spots in reasoning and research, he manifestly leaped wholeheartedly (and unashamedly) into this foolish epistemological trap back in 2004 or so, and has resolutely stayed there ever since.
For me, Rugg’s hoax argument is nearly the ultimate example of fabricating archery, in that his entire Voynich ‘research programme’ isn’t even remotely about any critique (or indeed meta-critique) of internal or external evidence. Rather, it is about post hoc fabricating a conceptual Cardan grille-style mechanism whereby an existing hoax hypothesis (for, of course, Voynich hoax hypotheses long preceded his entry to the arena) can be ‘proved’ to have been more possible.
The first problem is that you can’t prove something is ‘more possible’. Broadly speaking, an hypothesis is either possible or impossible (issues of constructability aside), and as far as I know nobody ever claimed that an ultra-sophisticated Voynich hoax wasn’t possible. So if Rugg is talking about ‘possibility’, he’s just been kicking at an open door for the last decade. (So let’s assume that he’s better than that.)
The second problem is that Rugg also isn’t talking about probability, because his focus is purely on whether it would have been ‘practically possible’ for a 16th century hoaxer to have produced a simulacrum of a book with the same properties as the Voynich Manuscript’s text – and this focus consciously excludes consideration of all the (fairly obviously, I think) 15th century evidence (e.g. the radiocarbon dating, the art history, the palaeography, etc). Hoax theories are more like meta-theories, in that they try to ‘win’ by sidestepping all the awkward issues of historical probability: any pesky conflicting evidence gets filed away into the ever-fattening “must have been fabricated somehow” folder.
(Yet when Rugg’s computer-fabricated ‘Voynich-cheese’ was passed through Mark Perakh’s Letter Serial Correlation (LSC) tests, it yielded the same type of result that gibberish texts did. The Voynich Manuscript’s LSC test results resembled those of real languages, so even the statistics are against him.)
Possible is not plausible is not probable
But surely, Rugg claims, the existence of his fabricated narratives wrapped around the core claim of a hoax serve to make the whole idea of a hoax more plausible?
Here at last is what Rugg is actually talking about: plausibility. Now, if you have read Nobel prize winner Daniel Kahnemann’s exemplary (2012) Thinking Fast And Slow (which I highly recommend), you’ll know that many otherwise clever people confuse ‘probable’ with ‘plausible’. But they are far from the same thing.
Specifically, plausibility is a story-telling quality, not an evidential quality. A story can be highly plausible but still impossible when tested: a story can be highly plausible but still extremely improbable when tested. Really, you don’t have to read many court transcripts to see that plausibility is no guarantee of either possibility or probability. In fact, if I had tuppence for every plausible-sounding Voynich-related claim I’ve read or heard, I’d surely be able to give every Cipher Mysteries subscriber a 14″ pizza, and still have enough left to buy a helicopter. And a yacht.
For clarity, I’ll repeat that every single notion within Rugg’s palette of historical assertions is fashioned from the same fabricating clay, the same post hoc painted rings around his hoax arrow. Take away this fabric, his set of imaginative reconstructions and post hoc narratives, and there is nothing left: not a page, not a paragraph, not a line, not even a word. His plausible-sounding narrative about an historical fabricator is itself nothing more than a sustained present-day fabrication, with absolutely nothing to back it up beyond his desire for the fantasy to be true, perhaps for the sake of meta-theoretical neatness. Whatever this makes it, what he has done is not history and it is not science.
Hence all I can conclude is that what Rugg has done with the Voynich Manuscript has literally been a waste of a decade, both his own and that of many others.
The Man Maketh The Word, or The Word Maketh The Man?
In closing, I’ve long struggled to find a word or phrase that sums up the pointless anti-historicism (and indeed unhelpfulness) of Rugg’s work: and so I find myself curiously grateful to William Dembski for his fabricating archer antipattern – this has given me the tool I needed to scratch this particular itch.
Yet I can’t help but feel that perhaps Dembski himself is an even bigger fabricating archer than Rugg: the neo-creationist arrow Dembski lodged in the wall is transparent and visible for all to see; while Perakh was clearly in no doubt that all the maths-styled and logic-styled presentation of Dembski’s books was no more than a decoration or distraction to conceal the conceptual vacuum at the heart of his argumentation.
So we end up with a curious pair of ironies: not only of Gordon Rugg writing books about logical fallacies and yet trapped for a decade in Dembski’s fabricating archer antipattern, but also of William Dembski employing an analogy to make plain to the world the same core fallacy that he himself is stuck in.
“For ’tis the sport to have the engineer. Hoist with his own petard.“