Long-time lurker, first post - got here through a few YouTube deep-dives (Voynich Talk) like probably half the new accounts on this forum. I want to be upfront: I haven't translated anything and I'm not claiming to have solved the manuscript. What I've got is a hypothesis about why and how it was made, and a guess about where a translation key could still be sitting undiscovered. Would genuinely like people to poke holes in this.
The medical knowledge gap that started it
Early 15th century European medicine was still mostly bloodletting, prayer, and humoral theory. Meanwhile the Islamic Golden Age and Persia had already produced serious empirical medicine - Avicenna's Canon of Medicine is the obvious example, with pharmacology and surgical knowledge way ahead of anything being taught in European universities at the time. That knowledge wasn't sealed off from Europe - it was leaking in steadily through Venetian trade networks and the Crusader contact zone, and Venice's trade routes ran straight up through the Alpine passes into Northern Italy and Southern Germany. Which, notably, is the same general region most researchers place the Voynich's origin.
The Church didn't just discourage foreign or "pagan" science, it had criminalized it outright - Pope John XXII's 1326 bull Super illius specula explicitly equated unauthorized astrology and alchemy with demonic heresy. So if you're a physician in 1410s Alpine Europe with Arabic-derived medical texts, you're not looking at a fine, you're looking at a stake.
And the Church was actively enforcing this in exactly the right place at exactly the right time. The Council of Constance ran 1414 to 1418 - the largest assembly of clergy, physicians and scholars in Europe at that point - right on the edge of the Alps. And it wasn't a passive gathering: on July 6, 1415, Jan Hus was burned at the stake there for heresy, in front of the assembled Church establishment. If you're a physician holding "infidel" medical texts, that's about as unambiguous a warning as you're going to get.
This wasn't only about survival
My theory is this was also about money. Medicinal bathing - balneotherapy as i understand it - was seriously lucrative in this period. Physicians and guilds who had superior, more effective treatments (baths, herbal preparations, dosing regimens derived from more advanced Eastern pharmacology) had a real competitive edge over rival practitioners and rival towns. So I think what we're looking at is a physicians' guild or syndicate with two overlapping reasons to hide this material: they'd be executed if the Church found "infidel" science in their possession, and separately, they didn't want competing physicians or guilds stealing their treatments and undercutting their income. Trade-secret protection plus heresy protection, at the same time, from the same document. That dual motive is, I think, a better explanation for why they didn't just bury the source texts under a floorboard - hiding isn't enough when you also want to keep using the knowledge yourself without a rival guild copying it.
By this point cryptography was already a real profession - papal secretaries and royal courts had codebreakers doing frequency analysis on substitution ciphers. If you just Caesar-cipher a Latin translation of an Arabic text, and it gets cracked, you're dead, and so is your guild's monopoly on the treatment. So instead of encrypting a known language, I think they built something closer to an invented notation system (William F. Friedman) - an a priori language organized around categories rather than spoken vocabulary. This actually matches the modern statistical work: the text follows Zipf's law and has real word-structure and entropy patterns consistent with meaningful information - but it doesn't match the underlying structure of any known spoken language. That's exactly what you'd expect from a constructed notation system rather than an enciphered natural language - no natural-language "key" to find because there wasn't a natural language sitting underneath it in the first place.
Why the words are so short
Something that convinced me is the shape of the words themselves. 240 pages of decent vellum works out to something like 14 to 15 calves, which isn't nothing - vellum was a real expense, and a finished multi-volume set of that size would also be a pain to physically hide or move around if you ever needed to. If you're trying to cram several treatises worth of material (herbal/medical stuff plus a whole separate astronomical section) into one book you can actually carry and conceal, you're going to need some form of serious shorthand. That's not a new idea for the period either - Tironian notes had been around for centuries at that point, and the whole concept behind them is a single stroke standing in for an entire word, prefix, or common phrase. Scribes copying legal or ecclesiastical documents used systems like this all the time to save time and material.
What's interesting is that Voynichese looks like it's doing something similar. The words are strange in a very specific way, almost none of them run longer than 8 or 10 characters, there's basically no doubling of letters, and you don't get the kind of irregular, messy variation you'd expect from a word in an actual spoken language. It behaves more like a constrained set of building blocks getting recombined over and over. To me that's more consistent with each "word" acting as something like a compressed tag or instruction rather than a word in the normal sense - closer to how a note-taking shorthand collapses "the patient should boil the root for three hours" down into a handful of marks, rather than how someone would actually write out a sentence.
Why the text changes character between sections (Currier A/B)
The herbal section deals with descriptive/botanical data, while the pharmaceutical section handles preparation and dosage. The astronomical section deals with positional/cyclical data - degrees, houses, timing. If you're building a notation system for compression, you can't use the same symbol logic for "root" and "planetary degree" - the underlying data types are just too different. So I think the Currier A/B split isn't two languages, it's the same compression system straining differently against two different kinds of source material, plausibly worked on in parallel by different specialists within the guild (the manuscript does show five distinct scribal hands, and there are no corrections or erasures anywhere, which reads more like fluent transcription of already-organized material than someone inventing content on the fly).
Timeline as I see it
- 1404–1438: vellum C14 range
- 1414–1418: Council of Constance - possibly the point where physicians/scholars from different areas converged and pooled source material under cover of a legitimate gathering
- July 6, 1415: Hus's execution: the likely trigger that turned "let's hide this eventually" into "we need this encoded now"
- 1415–1420ish: the actual writing
- Fashion and architecture in the illustrations (hairstyles, hats, tunics, and specifically the swallowtail/Ghibelline merlons on the castles - a known regional signature of North Italian/Alpine construction) all point to pre-1430, consistent with the C14 window and consistent with the scribes drawing what was physically around them at the time.
The actual point of this post, where I think the missing piece is
None of the above gets anyone a translation, and I think people sometimes oversell "cracking the logic" as equivalent to cracking the text. What I think is actually missing is a physical key (Rosetta Stone) some kind of reference document the guild would have needed internally to keep their own notation consistent across scribes over multiple years, especially given the two-domain (medical/astronomical) split. A guild encoding valuable trade secrets wouldn't operate purely from memory; there'd almost certainly have been an internal reference list, even an informal one.
Given the Constance connection, my guess is that if anything like that survived, it's more likely sitting uncatalogued in a regional Church, guild, or municipal archive near Constance/Lake Constance than in a well-known collection - the kind of document that wouldn't have been recognized as significant unless someone was specifically looking for Voynich-adjacent notation. The other possibility I keep turning over: the Habsburgs apparently acquired a lot of documents from this region, maybe there is some related material - including a key, or reference notes, or even correspondence about it - could been archived and could still be sitting somewhere in the Viennese court archives uncatalogued as because nobody was looking for it.
That's really the actionable part of this theory, if there is one: has anyone here done archive work specifically in the Constance regional archives, or in the Viennese Habsburg court records, looking for guild/physicians' notation systems or references to a "hidden" or "coded" medical text from this period? That seems like the piece that would actually move this from plausible narrative to testable claim, rather than more statistical work on the manuscript itself.
Happy to be told I'm missing something obvious, genuinely posting this to get it stress-tested, not to declare victory.