+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Voynich Talk (https://www.voynich.ninja/forum-6.html)
+--- Thread: Critique wanted (/thread-4756.html)
Critique wanted - KJ now - 19-06-2025
Title: POLER-D Method: Transparent Translation Framework from a Curious Independent
Hello everyone,
I’d like to take a moment to share the methodology I’ve been developing and refining over time. A phonetic (oral) framework I call POLER-D, short for Phonetic-Oral Linguistic Encoding Reconstruction-Diachronic variant. Recently I began posting translation snippets and findings from this process over on X (Twitter) under @VoynichUnleash for critique, but wanted to offer an overview here on Voynich Ninja. Both for transparency and to welcome any feedback from those more experienced in historical linguistics, cryptography, or manuscript studies.
First and foremost this isn’t a commercial venture, nor a gimmick. I’m not selling anything. I’m driven by deep curiosity and respect for the mystery of the Voynich Manuscript. I believe it's good to explore any method (however unconventional) that helps unlock the possible history within it. I understand that skepticism is healthy and necessary, and I welcome it. But I also want to be clear that this is not a “solved it” post. This method is a work in progress, and one that has already challenged many of my own original assumptions. In fact, most of my early theories have been overturned through application and cross testing. I’m not afraid of being wrong, but it would be a shame not trying.
What Is POLER-D?
POLER-D is built around the idea that the Voynich Manuscript uses a disguised phonetic encoding system. Drawing from spoken (and often unwritten or drifted) dialects from Northern and Eastern Europe and nearby regions as they existed after 1300AD.
Here’s how it works:
We begin with EVA (European Voynich Alphabet) transcriptions of the manuscript text. Then carefully align with standardized formats available in public datasets.
Each word is then broken down phonetically, not visually, using oral sounding reconstructions as if the scribe was encoding sound, not meaning.
For example:
shedy > skaidan / skēd (Germanic root: to speak)
kai > kaido / kaijan (Baltic: keeper/guardian)
chol >chwal / sol (Slavic/Hebrew: soul)
From there, we test drifted linguistic matches against known post-1300 forms in a selected group of languages (listed below), adjusting for diachronic shifts.
We also match results against imagery on the same folios like plants, figures, cosmological diagrams, etc. This is not cherry picking, it's about whether what we decode reflects what we see.
Each translation is checked against a stable pool of phonetic and written languages, with no additions unless justified. Results that don’t fit are discarded or flagged for review.
Languages We Use (Post-1300 Parameter Tightening)
To avoid the common pitfall of “too many possible matches,” we’ve restricted POLER-D to languages and dialects that were actively spoken or transitioning after 1300, including:
Proto-Germanic derivatives: Middle High German, Old Norse, Old Saxon
Celtic: Gaulish remnants, Middle Welsh, Irish phonetics (oral influence)
Slavic: Old Church Slavonic, Old Czech, Old Polish
Romance: Vulgar Latin, Middle French, Occitan, Italian dialects
Greek: Koine and transitional forms
Hebrew: Classical/Biblical Hebrew (as a known scholarly base)
Finnic/Uralic: Karelian, Hungarian
Latin alphabet: All reconstruction uses era-consistent Latin script, no anachronisms
As a base study we tested historical languages throughout the world with POLER-D and none worked beyond the northern and eastern European tribal areas.
Note: Etruscan and other unrelated ancient scripts were excluded deliberately. This isn’t about guessing or using every ancient language. It’s about limiting our pool to reasonable cultural and temporal candidates. We continue shrink parameters when possible.
Current Findings
We’re seeing internal consistency across a number of tested folios. Including those from the herbal, cosmological, and so-called "recipes" sections. Some patterns emerging:
Ritual herbalism, not just plant identification, but instructions, applications, and contexts
Ritualistic bathing, including symbolic language around purification and preparation
Astronomical cycles, likely ritual calendar components tied to lunar or planetary phases
Hermetic influence, recurring themes of balance, fire/water opposites, ascension, and transformation
Interpretations match imagery surprisingly well so far. We continue to apply this method across random folios (e.g., f1r, f67r, f54r, etc.) to test for stability.
I’m documenting the process in real time, including changes as they happen, at @VoynichUnleash on X. You’ll find raw phonetic samples, before/after comparisons, and posts noting where translations are being adjusted. This is about open research, not a closed theory.
Sample Entry
Here’s a real breakdown using POLER-D from f1r:
EVA: otedy shedy laram ychor
POLER-D Phonetic: ot-ed-ē / skēd / lar-an / i-khor
Possible Translation: “To speak the flowing essence (divine fluid)”
Matched Imagery: A root based plant drawn as if exuding fluid, possibly used in ritual speech or invocation.
POLER-D is evolving. We expect some variation in future translations as our parameters improve and mistakes are corrected. But the method is holding strong so far under repeated tests. If you’re skeptical, that’s good. This project is self funded and critique is free.I welcome engagement, suggestions, and yes, critique. I’m not a credentialed linguist or cryptographer, just a persistent researcher who’s stumbled my way into a tool that appears to be uncovering something meaningful.
Now strongly suspected that the Voynich Manuscript was authored by multiple classically trained scribes, not a single individual. This theory is supported by earlier work from paleographers like Lisa Fagin Davis, who identified at least five distinct scribal hands based on letter formation, writing angles, and stylistic patterns across the manuscript. Our findings align with hers, subtle shifts in phrasing, vocabulary range, and glyph formation all suggest collaboration. These scribes likely shared training in Latin, Greek, and Hebrew, but also picked up oral phonetic dialects through travel, allowing them to encode their knowledge in ways that reflected both elite training and folk wisdom.
We believe the Voynich Manuscript represents a collection of ritual practices, likely from preChristian or pagan traditions. Encoded not merely to protect “secrets,” but possibly to shield the content from persecution or prejudice. The use of a disguised phonetic script, rather than a true cipher, supports this idea. About 6 months ago AI was brought in to use to speed up the process and is now showing amazing potential. It took time to enter in all the parameters used, but once AI learned everything it began returning word by word and phrase by phrase to be assessed. I now feel it's a proper methodology to share on multiple platforms and began several days ago.This has been countless hrs spanning yrs of work prior to AI’s introduction. I want that to be cclear.i welcome all critique.
Thanks for reading,
KJ
(@VoynichUnleash on X)
RE: Critique wanted - RadioFM - 19-06-2025
So are you trying to match and map the EVA transcription (e.g. "otedy shedy") to other languages based on similarities, or was that just an example? Because if so, that'd undermine the whole thing IMO
RE: Critique wanted - KJ now - 19-06-2025
(19-06-2025, 01:33 AM)RadioFM Wrote: You are not allowed to view links. Register or Login to view.So are you trying to match and map the EVA transcription (e.g. "otedy shedy") to other languages based on similarities, or was that just an example? Because if so, that'd undermine the whole thing IMO
Hey, totally fair question and I appreciatethe thoughtful pushback. Also, love the glyph on your profile. I ran it through our POLER-D framework, and it seems to say “it is Sacred word” or “the essence is declared”. Would that be an accurate interpretation on your end?
To your main point though, we don’t just match EVA glyphs to other languages by surface similarity. That would undermine everything. What we’re doing instead with POLER-D is:
Phonetic drift modeling across post 1300 phonetic languages. EVA> phoneme groups> Latin>English
Consistent syllable parsing, guided by diachronic sound rules (exampl: how "k" > "ch" > "sh" in Germanic/Slavic shifts)
Latin as a neutral anchor. It helps maintain structure and limits cherry picking
AI implementation, now that our parameters are set, every line goes through the same system without guesswork or intuition. That removes confirmation bias from the equation.
When ritualistic or botanical patterns show up across multiple unrelated folios, we’re not projecting meaning, we’re surfacing it.
Let me know what you think. I’m always open to better ways of testing this system.
RE: Critique wanted - oshfdk - 19-06-2025
Hi and welcome!
It's hard for me to understand the details of your system, it looks very complex.
Is it OK if we do a simple test instead?
Can your system decode the meaning of the following snippet? Let's try this without context, to avoid cherry picking and introducing any bias.
RE: Critique wanted - RobGea - 19-06-2025
Phonemes sound good
However your response to RadioFMs post #2 does not fill me with confidence.
Why would you think that any EVA letter, say EVA-k for example, has a phonetic similarity to any sound whatsoever ?
EVA is a transliteration, you can read about it on voynich.nu >> Text Analysis - Transliteration of the Text >> You are not allowed to view links. Register or Login to view.
RE: Critique wanted - Rafal - 19-06-2025
I would have a simple question.
You say things like 'it seems to say “it is Sacred word” or “the essence is declared”' so it looks like you are getting some meaning from Voynich Manuscript.
What language is Voynich Manuscript written in then?
Do you have any longer sentences translated or just single words?
RE: Critique wanted - Mauro - 19-06-2025
The problem is EVA is just a transcription, the 'k' 'a' 'p' etc. are just characters assigned more or less at random to each Voynich glyph. They do not pretend to represent 'real' phonemes and, even if VMS is a phonetic transcription of some language (which it might actually be, who knows) it would be an incredible coincidence if EVA happened to choose the right phonemes by sheer luck.
Ie. EVA 'shedy' - 'chedy' - 'opchedy' could have instead been transliterated 'puast' 'guast' and 'erguast', or 'klira' 'blira' 'soklira' or anything else really, so I fear that trying to build on EVA 'phonetics' is a waste of time.
(Try thinking EVA as a series of numbers rather than 'pronounceable' letters', ie. 's' = 1st symbol, 'h' = 2nd symbol, 'e' = 3rd symbol etc.. What's the phonetical 'pronunciation' of the sequence 1 2 3 4 5 (shedy)? It's clear one cannot tell, each number could represent any possible phoneme).
RE: Critique wanted - kckluge - 19-06-2025
(19-06-2025, 12:52 PM)Mauro Wrote: You are not allowed to view links. Register or Login to view.The problem is EVA is just a transcription, the 'k' 'a' 'p' etc. are just characters assigned more or less at random to each Voynich glyph. They do not pretend to represent 'real' phonemes and, even if VMS is a phonetic transcription of some language (which it might actually be, who knows) it would be an incredible coincidence if EVA happened to choose the right phonemes by sheer luck.
Ie. EVA 'shedy' - 'chedy' - 'opchedy' could have instead been transliterated 'puast' 'guast' and 'erguast', or 'klira' 'blira' 'soklira' or anything else really, so I fear that trying to build on EVA 'phonetics' is a waste of time.
More "less" than "more" at random -- a 'y' kinda-sorta looks like a '9' without a closed top, a 'd' kinda-sorta looks like an '8' with the upper loop squished, an 'l' kinda-sorta looks like a cursive lower-case 'e' but stretched, an 'e' kinda-sorta looks like a 'c', but turned into an 'e' so it's a vowel, etc.
Unfortunately the original sin here is the reason for the kinda-sorta's, which was a questionable belief on the part of EVA's creators that there was some sort of virtue in making EVA pronounceable. Per Rene @ You are not allowed to view links. Register or Login to view., "The basic Eva alphabet was chosen such that the transliterated text is almost pronouncible. This excellent idea from Gabriel was not to be able to 'speak Voynichese', but to make it very easy for the human brain to recognise and remember transliterated words." Ignoring the bludgeon adjective "excellent" there, while it may make it easier to *remember* transliterated words I question whether it makes it easier to *recognize* them. Unless you're an experienced user of EVA, I really doubt it's easier to recognize "opchedy" on a scanned page image than it is to recognize "oqjctc89" (the Frogguy equivalent).
The outcome of that design decision was this latest in a whole string of such misunderstandings of what the EVA transcription is/means. Had they just stayed with Frogguy as an alternative to Currier/Bennett/etc., I doubt anyone looking at "oqjctc89" would have been lured down this garden path.
RE: Critique wanted - Mauro - 19-06-2025
(19-06-2025, 02:13 PM)kckluge Wrote: You are not allowed to view links. Register or Login to view.The outcome of that design decision was this latest in a whole string of such misunderstandings of what the EVA transcription is/means. Had they just stayed with Frogguy as an alternative to Currier/Bennett/etc., I doubt anyone looking at "oqjctc89" would have been lured down this garden path.
I'm no fan of Frogguy, but you're undoubtely right.
RE: Critique wanted - kckluge - 19-06-2025
(19-06-2025, 03:31 PM)Mauro Wrote: You are not allowed to view links. Register or Login to view.(19-06-2025, 02:13 PM)kckluge Wrote: You are not allowed to view links. Register or Login to view.The outcome of that design decision was this latest in a whole string of such misunderstandings of what the EVA transcription is/means. Had they just stayed with Frogguy as an alternative to Currier/Bennett/etc., I doubt anyone looking at "oqjctc89" would have been lured down this garden path.
I'm no fan of Frogguy, but you're undoubtely right.
I'm a huge fan of Frogguy for its intended purpose, which was to provide a vanilla-ASCII scheme for representing Voynich Mss text in a way that visually approximates the script on the page as closely as possible given that constraint. As a transcription scheme the learning curve is negligible. For analysis purposes it obviously requires post-processing, but then again so does EVA.