• Options
• Login
• Desktop Style

Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

• Index

  • The Voynich Ninja
  • Voynich Research
  • News
  • [Article] Technical rebuttal: semantic density and the “Naibbe” verbosity constraint
• Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

• Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  Torsten > 08-09-2026, 11:04 PM

  There is a You are not allowed to view links. Register or Login to view. to the editor by Domingo Delgado: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint"
  
  

  Quote:Technical rebuttal: semantic density and the “Naibbe” verbosity constraint." Cryptologia, ahead-of-print(ahead-of-print), pp. 1–2

  
  Delgado also wrote at You are not allowed to view links. Register or Login to view. about the paper:
  
  

  Quote:I’m pleased to share that my paper, “Technical Rebuttal: Semantic Density and the ‘Naibbe’ Verbosity Constraint,” has been accepted for publication in Cryptologia and is now in production with Taylor & Francis.
  
  The paper examines a focused technical question in Voynich Manuscript research: whether short, compact Voynich-like ciphertext units can, in principle, preserve greater local information density than the Naibbe cipher model appears to permit.
  
  Using a deliberately limited toy model, I illustrate how a cipher architecture incorporating positional behavior, nulls, permutation, and abbreviation could encode more plaintext information while retaining compact ciphertext forms.
  
  The paper does not claim a decipherment of the Voynich Manuscript. Its purpose is narrower: to clarify what conclusions can—and cannot—be drawn from the Naibbe model regarding the possible information density of alternative Voynich-like cipher systems.

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  Koen G > 09-09-2026, 08:00 AM

  Good times when people post an AI-generated abstract of their Voynich paper on LinkedIn.

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  nablator > 09-09-2026, 09:18 AM

  (08-09-2026, 11:04 PM)Torsten Wrote: You are not allowed to view links. Register or Login to view.The paper does not claim a decipherment of the Voynich Manuscript.

  
  Unusual for Domingo Delgado.

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  Torsten > 10-09-2026, 12:11 AM

  I haven't got the letter yet, so this is based on the author's own summary on LinkedIn.
  
  Taking the four devices in turn:
  
  Abbreviation is the only one that clearly raises density. But it raises entropy too — abbreviation removes the redundant material that makes characters predictable from context. Voynichese sits well below plain Latin already. The one device that buys density moves h2 further from the target.
  
  Permutation preserves h1 and should push h2 up toward it by destroying sequential regularity. A strictly deterministic permutation leaves structure intact but adds no concealment. Either way, permutation is length-preserving — it can't increase plaintext-per-glyph.
  
  Positional behaviour puzzles me for a different reason. For position to carry semantic load you need the same glyph appearing contrastively in several slots, so that knowing the slot tells you something. Voynichese gives the opposite — knowing the glyph mostly tells you the slot.
  
  Nulls lower entropy by adding predictable padding — but they lower density by definition, since they carry no plaintext. So the list contains one density-raising device (abbreviation) and one entropy-lowering device (nulls), and they work against each other.
  
  This raises the key question: at what rate are nulls inserted, and is density computed per total glyph or per non-null glyph? If per non-null glyph, any density can be reached by declaring enough glyphs meaningless, while the entropy match is bought by the glyphs removed from the denominator.
  
  Underneath all four is a point I'd expect the letter to address: no cipher can recover more information from a ciphertext than the ciphertext contains. The measured entropy of Voynichese is a ceiling on recoverable plaintext regardless of architecture. So was any text actually enciphered with the toy model, and are h1, h2, word-length distribution and type-token growth reported for the result?
  
  Has anyone here got access to the full text yet?

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  Mauro > 10-09-2026, 12:42 PM

  (10-09-2026, 12:11 AM)Torsten Wrote: You are not allowed to view links. Register or Login to view.Underneath all four is a point I'd expect the letter to address: no cipher can recover more information from a ciphertext than the ciphertext contains. 

  
  Strictly, this is not correct. Ie. with a You are not allowed to view links. Register or Login to view. cipher the cyphertext contains zero information, yet the full message can be recovered (in this case, all the information is in the key).

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  Torsten > 10-09-2026, 03:02 PM

  (10-09-2026, 12:42 PM)Mauro Wrote: You are not allowed to view links. Register or Login to view.Strictly, this is not correct. Ie. with a You are not allowed to view links. Register or Login to view. cipher the cyphertext contains zero information, yet the full message can be recovered (in this case, all the information is in the key).

  
  Fair correction, and I'll restate it: the relation is plaintext–ciphertext–key, not the ciphertext alone. 
  
  But I think that sharpens the point rather than blunting it. A OTP ciphertext is maximally high-entropy — uniform, h2 equal to h1, no positional structure. Voynichese is anomalously low on all three. So the one architecture that genuinely puts most information in the key produces exactly the profile Voynichese doesn't have. The key is also message-length, and neither half is readable alone.
  
  Which is worth restating generally: the information has to be somewhere, and every place you can put it has a measurable cost. In the ciphertext, it raises entropy. In the key, it makes the text unrecoverable in principle.

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  magnesium > 10-09-2026, 03:07 PM

  (10-09-2026, 12:11 AM)Torsten Wrote: You are not allowed to view links. Register or Login to view.Has anyone here got access to the full text yet?

  
  Yes. The extent of the treatment of the "toy model":
  
  A useful way to frame the issue is through a toy model. Suppose a short ciphertext token is generated by combining four operations: one sign may represent either a single letter or a common digraph depending on position; one sign may function as a null; one positional marker may trigger the reversal of two adjacent plaintext elements; and one abbreviation rule may compress a frequent Latin sequence. Such a system would remain compact at the ciphertext level while allowing more plaintext material to be packed into a short token than a model that maps most ciphertext units to only one or two plaintext letters. Unlike the Naibbe cipher, which often expands plaintext into multiple glyphs per unit, this toy configuration allows a single token to encode multiple plaintext elements through positional and compressive rules, thereby increasing local information density without sacrificing compact ciphertext form. The point of this toy model is not to claim a decipherment. It is simply to show that the design space includes Voynich-like systems capable, in principle, of greater local information density than the Naibbe cipher appears to permit.
  
  This does not strike me as much of a "technical rebuttal," truth be told. And to Torsten's point, the "toy model" is not quantitatively demonstrated to deliver all these properties. It should also be noted that Delgado's "toy model" is the general cipher architecture he claims underpins the VMS in his book Codex Obscura. (For those who haven't read that book, Delgado presents provisional decipherments of portions of the VMS...without ever revealing the exact text he is purportedly deciphering, let alone the specifics of the cipher.)

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  Torsten > 10-09-2026, 03:31 PM

  (10-09-2026, 03:07 PM)magnesium Wrote: You are not allowed to view links. Register or Login to view.Yes. The extent of the treatment of the "toy model":

  
  Thanks for posting that — it's more specific than the LinkedIn summary, and it makes the four devices easy to sort.
  
  Two raise density and raise entropy. A sign valued as letter-or-digraph by position is a polyvalent sign: sometimes one glyph buys two letters, but a sign whose value varies by context is by definition less predictable from context, which is what higher h2 means. Same for compressing a frequent Latin sequence — abbreviation squeezes out redundancy, and redundancy is what keeps entropy low.
  
  Two lower entropy and lower density. A null is a glyph carrying no plaintext, so it subtracts directly. The positional marker triggering a two-element swap is worse: a deterministic local transposition encodes nothing, it's length- and value-preserving, and it costs a marker glyph. That's a null with extra steps, and it's a net density loss. I'm not sure why it appears in a density argument at all.
  
  Which leaves the letter demonstrating that some cipher can be denser than Naibbe — which nobody disputed — without showing that any such cipher matches the Voynich text, which is what the verbosity constraint was about.

• RE: Technical rebuttal: semantic density and the “Naibbe” verbosity constraint

  magnesium > 10-09-2026, 04:00 PM

  (10-09-2026, 12:11 AM)Torsten Wrote: You are not allowed to view links. Register or Login to view.I haven't got the letter yet, so this is based on the author's own summary on LinkedIn.
  
  Taking the four devices in turn:
  
  Abbreviation is the only one that clearly raises density. But it raises entropy too — abbreviation removes the redundant material that makes characters predictable from context. Voynichese sits well below plain Latin already. The one device that buys density moves h2 further from the target.
  
  Permutation preserves h1 and should push h2 up toward it by destroying sequential regularity. A strictly deterministic permutation leaves structure intact but adds no concealment. Either way, permutation is length-preserving — it can't increase plaintext-per-glyph.
  
  Positional behaviour puzzles me for a different reason. For position to carry semantic load you need the same glyph appearing contrastively in several slots, so that knowing the slot tells you something. Voynichese gives the opposite — knowing the glyph mostly tells you the slot.
  
  Nulls lower entropy by adding predictable padding — but they lower density by definition, since they carry no plaintext. So the list contains one density-raising device (abbreviation) and one entropy-lowering device (nulls), and they work against each other.
  
  This raises the key question: at what rate are nulls inserted, and is density computed per total glyph or per non-null glyph? If per non-null glyph, any density can be reached by declaring enough glyphs meaningless, while the entropy match is bought by the glyphs removed from the denominator.
  
  Underneath all four is a point I'd expect the letter to address: no cipher can recover more information from a ciphertext than the ciphertext contains. The measured entropy of Voynichese is a ceiling on recoverable plaintext regardless of architecture. So was any text actually enciphered with the toy model, and are h1, h2, word-length distribution and type-token growth reported for the result?
  
  Has anyone here got access to the full text yet?

  
  Just to add to Torsten's point here: As I noted in the Naibbe cipher paper, the VMS's type length distribution and conditional character entropy jointly place tight constraints on how much plaintext information any given Voynichese token can feasibly contain, assuming that Voynichese glyphs or strings thereof are meant to map to individual plaintext letters. That means there is not a whole lot of budget for the full suite of operations Delgado proposes. No matter how you define a glyph within Voynichese, there are only so many Voynichese tokens that are eight or more glyphs long.
  
  The permutation principle can in theory lower conditional entropy; the permutation could be alphabetizing every time, for example, and then you could encode which permutation restores the plaintext from the alphabetized letter string. But I have run a quick test of that on Naturalis Historia with a toy-model cipher (a two-letter code encoding which of 24 permutations is the correct decryption of an alphabetized four-letter chunk), and it doesn't lower entropy all the way to Voynich-like levels. Not only that, but it's just a bad cipher. It immediately becomes clear what the game is.
• Next Oldest Next Newest