Summary of my views on the Voynich Manuscript

About this site and my views

I am a professional academic and linguist, whose ‘normal’ professional work you can read about here. However, this website has largely been taken over by discussion of the Voynich manuscript (VM).

In 2014 I set out a partial, provisional theory about the manuscript which you can read about here and watch on Youtube via this page here.

Below, to the right, and in the menu above, you can see links to other pages and to the views of a host of other contributors, whose insightful contributions I gratefully acknowledge. In response to requests, however, let me here set out in brief where I currently stand on the Voynich manuscript (VM).

Not a hoax, and probably not a cipher

Before my 2014 paper many people who had researched it for years believed the manuscript to be either a sort of code/cipher, or a hoax. I do not believe it is a hoax (see here). I don’t consider it alchemical, nor diabolical, and probably not a mysterious cipher designed to conceal knowledge. My work was attacked rather aggressively by some who hold particular views, such as Nick Pelling here, who believes it to be a cipher or code designed to conceal knowledge, and you can see my response here.

I accept the carbon dating evidence which suggests that the manuscript was produced in the early 15th century, and the ink analysis which found nothing to contradict a 15th century date.

With apologies to conspiracy theorists, I (boringly) consider the VM most probably to be an attempt at an encyclopedia or ‘summa’ aiming to encompass contemporary knowledge of plants, astrology/astronomy and related areas, exactly as the wonderful Occitan manuscript which I discuss here does.

It was probably intended for a specific linguistic community, for which the script was specially devised. In my view the script had probably been developed previously, not solely for this manuscript, but if so, evidence of that development is now lost. In short, the VM just happens to be in a script we cannot yet decipher, but is probably in a language which is known or can be reconstructed.

My approach is to work on it carefully and with an open mind, mainly on the linguistic elements of the script and language, but taking account of other research on the imagery, without any preconceived ideas about where it came from or its genre.

My current view is that the manuscript was possibly written by or for a community in eastern Europe, the Caucasus, or western Asia – sorry not to be more specific – and has the traces of Arabic, Persian. Mongol, Indian ideas and even language which would not be unexpected or surprising at the time in, say, the area of the Black Sea, the Caucasus or eastern Turkey.

However, this site is open to all approaches – including analysis of images – so feel free to post your views.


  1. Daede

    I haven’t read any mention of the mistake on f69r. I find this error particularly interesting given the attempt at and ultimate failure in correcting the error by the scribe.

    The symbol begins at the inner ring with a six pointed star with single characters labelling the void space between each point. The second ring provides long form words for each of the points themselves bifurcating the circle from 6 to 12 branches. The third ring where the error occurs is clearly supposed to consist of 22 word groups each separated by a blank spoke; there should be 44 spokes all told but a simple error in the drawing leads to an extra 2 spokes, 46 total, one of which is then shaded out leaving the odd and clearly unintentional number of 45 remaining. This error propagates into the fourth and final ring which mathematically should consist of either 12 or 22 word groups actually being incorrectly divided into 16 word groups. You can tell this division is incorrect not just based on the obvious intended symmetry of the image but by the scribes own bifurcation of a word in the top left corner of the symbol.

    Given the error propagation it can be logically inferred to indicate that this symbol at the very least was copied poorly from another source. More importantly given the failure to correct the error it suggests that it was copied by someone who was not well versed in the subject being copied.

    The further implications we can draw from this inference are even more interesting not least of which is very credible evidence against the hoax theory or at the very least against this being an original hoax, as it could still be a copy of a hoax unwitting or otherwise.

    There is always the possibility of a lack of care on the part of the scribe but that seems unlikely given the attention to detail elsewhere in the manuscript coupled with the attempt to correct the error. Could the failure to correct imply a corrupt or otherwise obscured source material, perhaps verbal or remembered?

    Ultimately the greatest concern may be that what prevents the realisation of our hope to one day see this manuscript translated into some sort of meaningful treatise is a compounding of errors in spelling, grammar and basic understanding by the author of the subject matter itself.

    Image below highlights Original Error, Compound Error and resultant Geometric Inconsistencies in a hopefully blindingly clear and suitably concise manner for immediate comprehension.

  2. Dear Professor Bax! It seems to me that I solved the question of the principles of coding the text of the manuscript ( I would be glad to know your opinion and answer questions, if such (hopefully)) arise.

    • MarcoP

      Hello Alexandr,
      here are my comments about two of the words you discuss:

      3.1 EVA:dcholday Which you seem to transcribe “prianpaus” and interpret “p[a]ri[licus] an p[h]a[r]us” (“Parilicus or the Lighthouse”) – I guess you interpret the ending -9 as the common Latin abbreviation for -us.

      3.2. EVA:otolal Which you transcribe “akanun” and interpret “a[strum] K[ull]a[t] Nun”

      Here are some observations from a fellow amateur with no particular competence.

      A) You write that EVA:a reads ‘u / v’, ‘t’, but in 3.1 you seem to read it as ‘a’ EVA:day=pharus, while it should be phurus? Maybe I am misinterpreting, but I don’t think you made clear what your letter-by-letter reading is.

      B) The kind of abbreviation you devised is too unconstrained (in the past, Stephen has pointed out this problem in other similar approaches). For instance, 3.1 could also read
      “tri[gonon] an t[riang]u[l]us” (you write that EVA:d corresponds to the letters ‘b’, ‘t’, ‘p’)
      “pri[mum] a[ster] n[oc]tu[rn]us” (the first star of the night i.e. Venus)
      “pri[mus] a[mor] n[on] p[ar]vus”
      “bri[t]an[nicus] ba[c]us” (Roger Bacon?)

      C) 3.2 is interpreted in a radically different way from 3.1. EVA:ol in the middle of the “word” is not interpreted as a disjunction. The name is interpreted as Babylonian, rather than Latin. Of course, this is a problem for many ideas about the manuscript (I have often mixed different languages in my ramblings). But if you combine “abbreviations” freely interpolating 60% of the text with arbitrary languages, you can really read anything anywhere.

      D) an would not be used in Latin when listing alternative names. For this, you will find vel or sive. But this really is a detail.

    • Peter

      I think you have a small mistake on your side.
      The g in Taurus is actually an o, it just looks like this, but is a fold in the parchment. (Toaro) (Taura)

      • It’s not a mistake)) I didn’t hold the parchment in my hands, didn’t see it with my own eyes, but Rene could see.

        • Peter

          You’re right. It should really be a “g”.

  3. Would you consider the images in the Voynich Manuscript to be in any way heretical? Given the times they were drawn.

  4. Michael calin

    Hi there. I had six months of interesting events bumping into this person
    And that article and have a theory to look into.
    I’m thinking VM was in fact a copy of an original, but the copier wasn’t good with
    The language they were copying even though they understood it.
    I came across an Armenian guy who said his language was the oldest in the world
    and wanted to see his language. Currently there Are 6 variations and have been as
    many as twenty in the last couple thousand Years.
    I compared letters which all matched (reading right to left) and I believe
    They are packed with measurements, Armenian uses 20 numbers to depict
    Scale and if it were in a unit of measure, then you would see something
    Like the metric gram weight scale into the thousands which unless you
    Realized that, you would think they were only words. This would explain
    The constant repetition of similar “words” because they are identifying
    A count rather than a definition.
    If you review the art and illustrations in many Armenian texts the design
    Is remarkably similar in drawing styles even more than simply the period.
    Please let me know your findings.

  5. orun

    i agree with this idea.
    firstly the crosbowman in horoscope pages is a drawing of eastern person. İn historic art, from İstanbul to tokyo, all civilizations draw humans in slant-eyed form.
    Another point is blonde hair women drawings. there are only 2 human race in east that have blonde hair: Russians and cuman(kipcak) turks. but russians are orthodox, in VM i see only catholic crosses. Cumans had Tengri faith but we know many missionaries worked at Cuman steppes and after 1227 huge part of cumans became catholic, because they were running away from mongol hordes and they refuged to catholic Hungary kingdom.
    Cumans were nomadic people so i think they didnt create written literature. İ believe VM was written by a european person, maybe a missionary. This book “aimed at Cumans by an catholic missionary” theory is consistent in my opinion

  6. Katie

    I am from Georgia, next to black sea, which is considered east Europe, or Caucasus, I am not linguist but we have 3 kind of writing, Asomtavruli, Nuskhuri and Mkhedruli.we had already our script 3 BC. when I looked this as they call manuscript it is so much similar with Georgian, I was sure that I can read if I zoom in 🙂

    • Stephen Bax

      Thanks Katie, I too have wondered about links with aspects of the Georgian script. However, despite my investigations and comparisons I can not find a strong relationship, unfortunately.

  7. Sheryl Skoglund

    Tauros Os is ancient name of a god in Old Norse. Tauras bull of Orion. Ref in Silmarillion and Lord of Ring by Tolken. Thank you for your research.

    • Klara

      I’m sorry, I can’t find any sources on that. Do you mind sharing yours?

  8. Michael Dixon

    Very intresting video by someone who clearly has done a lot of work on the manuscript. Unfortunately, it is fairly anonymous. There is also a part 2.

    • MarcoP

      Hi Michael,
      the author of the videos is Derek Vogt.
      See here.

  9. espi

    with many words beginning “o”, and the mix of arabic, latin and perisan words, could the language be based on portuguese?

    • Holde

      could it be possible that this book was written by a Crusader that learned to write in Arabia, used there alphabet but writing from left to right ? A book like this would have been very helpful for them.

      • Stephen Bax

        Thanks – but alas there is no evidence for that. The ‘Could Be’ page for the Voynich manuscript is very very long indeed. The ‘Evidence’ page is rather shorter, sadly.

  10. Frederick Herrmann

    Just watched your YouTube video. Great work! Thank you.

  11. Emre Yılmaz

    I am fascinated by Voynich manuscript. I always loved languages, how they change over time, how they are connected with each other… I think VM is just so mysterious. This is what makes it special. I will start a research on this book. I hope our humanity can solve this big mysterious book. Wish you good luck, Mr. Bax.

  12. Michiel Ockeloen

    Dear Stephen Bax,

    Amazing and interesting to read about your research here.

    Just a few remarks:
    – reminds me of this research, i’ve recently read about in Italian;
    – the typical swallow tail shapes of the battlements of the depicted city or fortress, had had a political meaning in Italy, as i learned last summer on holidays, and occurred only in Northern Italy at the time, as stated in an American documentary about VM.
    – good luck succes with your work!

    • Stephen Bax

      Thanks, though apparently they did not appear only in Italy…

      • Peter

        However, I am aware of no before 1500 north of the Alps. At least not found.
        White but also have disappeared in stronger walls a lot because of changes.

      • james w

        Keep going my good man! Really impressed so far.

  13. Jason Hall

    For: Stephen Bax concerning the “Voynich Manuscript”
    Have you considered the use of a linguistic savant to associate written language leads? I hope this does not sound so trivial, but a savant named Christopher might see something others might not? He is noted as having a knack for associating written languages with spoken languages. Perhaps he will see something in the text that others could not? Here is one suggested starting point.

    I don’t want to seem presumptuous or a sciolist. I only hoped to help.
    Thank you.

  14. Zoila Brashears

    Mr. Bax, I have read many of the comments on here, but none seem to match my speculation. This manuscript is definitely of Franciscan origin, circa third voyage of Columbus. Many of the plants seem to be of tropical origin, Southern Mexico through Central America. The illustration on page 45 is a jalapeño plant, and several plants come under the classification of wild greens or quelites. This is a book on edible/medicinal plants, how and when to find them, and how to cook with them.
    The Monks would have extensive knowledge of several languages and definitely would have made an alphabet for the Nahuatl language as they had no textual one. So, borrowing from Greek, Latin, Arabic, Germanic, or any other language they knew would be natural to them.
    Also, since it was easier to learn the native languages than to try and teach Spanish, it would follow that they would have students helping them illustrate edible plants and write in the explanations of use afterwards.
    Your conclusion of extinction is also correct, as the native people slowly forgot their original languages and adapted to the Spanish invasion.
    As for the astrological charts, I believe they also refer to planting and harvest times, coinciding with their then current calendar system. I believe some gentleman has deciphered how to read them (on YouTube). Pretty ingenious, really.
    One last thing – the Nahuatl people believed the stars were female, a possible explanation for all the illustrations of women….

  15. Natalia
  16. Natalia

    Dear Mr. Bax,
    and what is about the data an the medieval paintings?

    1. Hans Schuechlin 1479
    2. Meister of 1477
    3. Martun Schongauer 1469
    3. Martun Schongauer 1486
    => Ulm and Augsburg, Germany

    Its just a suggestion.

    Best wishes,

  17. Professor Bax –

    You mentioned that looking for patterns is a lot of work.

    What if:

    – All the VM characters were assigned numeric codes, including spaces
    – A data entry operator typed the VM text as a string of the numeric
    codes just assigned
    – Computer whizzes were to scan the text looking for linguistically
    significant patterns in the numbers?

    Eleanor White, retired engineer
    Ontario, Canada

    • Stephen Bax

      Thanks Eleanor – such things have been tried, alas. However, what do you mean by ‘linguistically significant patterns in the numbers’? Unfortunately, if we don’t know the underlying language, we don’t know which language family could furnish us the appropriate information! It is that kind of problem which faces us.

      • Dr. Bax,

        New to this topic and the various approaches or history. However, assigning the numbers to each common or similar character or symbol and then doing pattern matching would make sense.

        All language has repetition and commonality within its own language parameters. By using the pattern matching, across the various images, you may be able to find common “themes”. For example, on many of the flowering plants, there may be “common patterns” indicating some type of similar descriptions, characteristics, or uses for those “similar” objects. You could look for patterns in objects which might all be food or fruit bearing plants. Patterns in any of the objects with human personalities attached.

        In other words, similar to what you have done, but looking for common threads across multiple related objects in the manuscript. The greater the number of pattern matches across objects that seem to have some rational relationship would indicate some common descriptors or characteristics.

        This method could be used across objects that appear to be related, and then more tangentially across objects which are not directly related. Even a complex language with many variations for words (e.g. Greek), would most likely still have some common root patterns (whether prefix or suffix patterns).

        This exercise could be done by looking for larger and larger pattern matches. And then looking for the greatest number of common pattern matches across objects. Large numbers of individual pattern matches across different objects (whether they appear related or not) would likely reveal some clues about the character of the content.

        One last thing occurs to me. Is it possible this was a person’s “personal shorthand” similar to what used to be taught to clerks and legal secretaries? So that there might not be a direct language comparison but an indirect one with that individuals “shorthand” applied? The pattern matches should still show some of that indication, but it may be more attenuated.

    • The entire manuscript text has long been entered in computer-readable form, and there are numerious studies that have tried to find evidence for linguistic patters. When I started putting together my web site in the 1990’s, my primary aim was to bring together all these studies in an organised manner. This was never very successful, and I spent more effort on the history of the MS than anything else.

      Recently, I have come back to my original aim, and bringing all this information together is now in progress. Several publications are available online, and links may be found in this section:
      and primarily in this page:
      (As I said, in progress. There is still a lot to do).

    • that’s funny when i say i’ve decoded the VM and noboby seems to trust me but VM is exactly what i said and you’ll probably never find linguistics shemes in it… you find more mathematical items as lines , matrices, angles and vectors than any significant word… this code only gives the shape of the draw on the pages besides…i know, it’s difficult to believe that but that is…

      • Stephen Bax

        Eric, maybe one reason people are not convinced is because your idea (like so many others 🙁 ) does not account for and explain large parts of the manuscript in any satisfactory way. Too many people take a few elements – some pictures, some letters, some shapes – and then build elaborate explanations, insisting that they have the whole answer.

        People will only be convinced when many, or even most, or all, of the elements of this manuscript, including the script and language, are explained as a whole. Sorry, but I think that’s the fact.

      • Marios

        hi Eric ! no one will believe you since you have not decoded the manuscript . what i mean give a word with exact meaning and exact illustration which matches with exact language . that what called decoding or decifering . till now no one did 1 word of it .

  18. Stephen,
    It looks to me as if part of the astronomical imagery probably came from the general region around Marāgha (mod. Maragheh).

    That would explain the Bactrian residue, and connect well to the plainly similar imagery informing the “ladies” and forms in MS Sassoon 823 – similar to imagery from a non Latin corpus gained from Persia. (for some reason I can never remember the number the Sassoon ms was given when the collection was broken up).

    The botanical and lading sections, I think, had a different region of evolution before coming to the west, pretty late.

    I mention it now because I thought you might be interested to know that Kircher seems to have been speaking of that region when, he tells his readers that Egyptian hieroglyphs were carried to China.

    Of course the historical and epigraphic evidence doesn’t exactly support that notion – but of more interest is the way Kircher thinks the matter travelled.

    …as a secondary writer has said of him: “dwelling on the origin of Chinese writing… saying that the third son of Noah, Cham, immigrated with his tribe from Egypt to Bactria and settled “in the vicinity of the kingdom of Mogor” through which the knowledge of hieroglyphics penetrated to China proper.

    There’s certainly clear evidence of persons re-located from Egyptian territory to that region, and among them the Carians, who always served as border guards for Egypt. Their script has been discovered there, a copy of the Delphic maxims. This is apart from the Bactridae etc. whom the Persian relocated on their eastern borders.

    What fascinates me is where Kircher got this idea. Doesn’t sound like a flight of pure fancy, nor does it reflect any classical text that I know. Could it have been a legend maintained by people of the region? If so, who told Kircher? I’m inclined to think it’s a garbled version of the origin of printing, the early stone-rubbing printing circulated from Buddhist centres.. but the region was certainly multi-lingual and we have examples of one language being written in the script of another, and all sorts of fun.


  19. Jack L.

    What about Maltese?

    • Hi Jack, I’ve actually looked into the possibility of Maltese. I think the historical context is pretty good: part of the European cultural and intellectual sphere with an emerging native literacy in the 1400s, and with good reason to shun existing scripts (an awareness that Maltese was different to other European languages, yet other Semitic scripts such as Hebrew and Arabic had religious baggage).

      If the script was not fully vowelled (common for Semitic languages) it would make sense of the small number of characters in the script and relatively short words. That could also explain repeated words (they would thus be derivations of the same root but with different, though unwritten, vowel sounds).

      The character [o] would almost certainly be some kind of ‘l’ sound and stand for the definite article. However, that would make ‘l’ far too common in the text as a whole, as it often appears in the middle of words and not only at the beginning.

      There are also problems regarding the rigid internal structure of words which cannot be squared with the (presumed) density of information that each character should have. Even though derivations of the same root would have similar or same structures, different roots should often present completely different ordering of characters. Yet the order of characters in any given word is often very similar.

      • Jack L.

        Thank-you for answering! However, I still wonder about it being Maltese with maybe Greek loanwords, as with the decipherment of “Tauros” and “Coriander” by Stephan Bax. Maybe there’s some constructed grammatical feature or omission of others within the text.

      • D.N. O'Donovan

        May I ask why you don’t think the “o” the Greek definite article? I do agree, though, that is the letter’s function in that position.

  20. Vaska Paris

    Have you possibly considered the Roma language of Eastern Europe. They migrated from India, have picked up all the languages you mention and still use the language today although inclusive of the likely inflections of the local languages where they live today (i.e. Hungarian, Serbian, Spain, etc…) plus they are very well known for their herbology and mysticism.

    • Stephen Bax

      Yes, thanks – many people have suggested that possibility, but the problem is to get evidence for it! Thanks

      • Tim

        Dr. Ian Hancock is a linguist for the Romani language, he may be a good source to ask about how the language in the VM compares. There are different dialects to Romani though.

  21. It’s old Fenno-Norse. Nick’s jaded, so he tends to go on the attack. Kudos to you on remaining civil.

  22. Suzanne Redalia

    I’ve got a translation of Folio 108 posted on Quora, on my blog ‘Fun With Jiroft Script.’
    The Voynich writing system is Byblos script, which was used by Italo-Celtics, and is a syllabary related to Linear B. The language is a Celtic-Indic pidgin, very terse, no articles, and phrases are separated by spaces, while words are usually not. The text of Folio 108 is a prayer to Shiva (li), Viraka (bi-ra-ka), Cytherea (ku-to-ra) and the angel Raphael (ra-pho) absurd as this sounds.
    Please take a moment to look at the blog entry and comment, if you are not too busy.

    • Stephen Bax

      As a courtesy to readers, can you please tell us the full web address? Thanks.

      Also, the Byblos script is very different….

      • Suzanne Redalia

        The article on Voynich script that I’ve written most recently is titled ‘Piacentinu and Saffron Cake’, and it features Folio 40v, a plant identified as Crocus by Dr. Edith Sherwood. I no longer believe that any of the text has prayers to Hindu deities, rather, they address Gaia and Jove in the Venetic language, asking for cattle, chickens, food for them, calves, sheep, edible plants and so forth.

        • Darren Worley

          Hi Suzanne – I find it interesting that you’ve proposed a language that is local to the NE of Italy, however, I find it problematic that the Venetic language is dated to a period 1500 years before the VMS. Is there any historical evidence suggesting it was spoken during the medieval period?

          There are other languages native to this region, which are still spoken, although nearly extinct. These seem to be more probable candidates, if you suspect that Voynichese might originate from this region. (Are these surviving languages related to Venetic?)

          Examples include the Mócheno language and the Cimbrian language.

          Cimbrian is particularly interesting as it was “discovered”, or at least first reported, in the 14th century by Italian humanists. I find it interesting that scholars in 14th-century were becoming curious about little-known languages in this region.

          • MarcoP

            Thank you, Darren! I was not aware that the “Mocheni” moved to Italy so early, and I had never heard of the Cimbrian language before. Very interesting 🙂
            I think it unlikely that the main body of the VMS is related with German, but the color annotations (as confirmed by Touwaide) and the marginalia seem to be.

            • Notula

              the Cimbrian language is an old Bavarian-Tyrolian idiom spoken in some very isolated villages in northern Italy. This is in fact very interesting but not in relation to the VMs because there is – to the best of my knowledge – no evidence of written Cimbrian in the 14th/15th or even earlier. The first Cimbrian text was a catechism translated from Italian into Cimbrian and printed in 1602. There is, of course, no such thing as Cimbrian writing either.

              By the way: Cimbrian dialects sound really peculiar and are hard to understand, even for Bavarians. Kind of living middelages.

              • Darren Worley

                Thanks Notula for the additional information about Cimbrian.

                In my opinion, the fact that there is no known written examples of Cimbrian from the 14th/15th or even earlier, makes it an even more interesting candidate!

                Marco – I do recall a (female) VM researcher claiming to have identified several German word in the VM. It wasn’t recently. However, I can’t find any links – can anyone recall who it was?

                • Notula


                  it’s possibly me you are referring to? It was a couple of weeks before when I made some comments about my research in German, but I never claimed anything. Not yet. The first part of my study will be pubslihed in a few days and will include my general view on the codicology, paleography and the origin of the VMs as well as the complete transcription of all marginalia. More details will then follow as work in progress.

                  Coming back to Cimbrian and other local languages and dialects in the poorest alpine regions, we face one big problem that counts for all of them. The number of native speakers was so small and almost all of them were illiterate that time, so nobody would invest in an expensive codex with no reader at all. Furthermore, if there was any book for that group of speakers it would have been a religious one. The idea of a codex in Cimbrian may sound attractive but is very far from reality in the early 15th century. People from that region may have sold their calfs for the production of parchment, but they surely never saw a book made of it.

                  • Peter

                    Considered from the plants, Northern Italy is probably the most probable choice.
                    Linguistically: in Switzerland alone with more than 200 dialects, 120 already extinct, I still prefer the side of Austria to dialects without end, and only in German. Every 20 km a new dialect.

                    Even with the sign for encryption issues raises questions.
                    Look where the signs come from, and you’ll be amazed.


                    • Darren Worley

                      Peter & Notula, Thanks for your replies.

                      I recall reading the article about a German connection several years ago and it was written several years before that. I remember the author tentatively identifying 12 (or so) German-derived words.

                      I’m less agreeable to an Alpine origin, instead I favour an origin centered around Venice and Padua. Reading into the origins of the Cimbrian language we learn about a migration of Bavarians to Verona c.1050.

                      I’ve been curious why many German printers migrated to the Venetian region in the latter 15th-century, perhaps in addition to it being a favourable intellectual and commercial centre, there was already an earlier German-speaking population settled in this region. Perhaps this also made it an agreeable destination. I can imagine this earlier population might have spoken a language mixing an older archaic German with Latin/Italian influences. (I intend to find out more about this migration.)

                      This mix seems to describe what I perceive in the VMS – the Zodiac iconography showing a South German influence and the VMS script incorporating symbols from Latin.

                    • MarcoP

                      Thank you, Peter!
                      It’s interesting to see EVA:k in another medieval alphabet!

                    • Peter – I don’t understand why it’s not just called the “Chaldean” alphabet.

                      This has been considered.. it was mentioned on the first mailing list by Knox Mix, and the thread is worth looking at.

                      I’m not sure if it’s mentioned by d’Imperio, but if it is you can be sure, altogether, that some fairly high-powered cryptographers have looked at that ‘Chaldean alphabet’ and decided it is not the cipher we want.

                    • Notula

                      “Alphabetum Kaldeorum” is, and was always, the standard term for this diplomatic chiffre that was allegedly invented by Rudolf IV, duke of Austria. There is proof (X-ray analysis) for him to have written a prayer/poem in that alphabet on a donation document (14.06.1360) but there is also evidence for the use of that chiffre long before Rudolf IV.

                      Apart from that its deriving from “Chaldean” is pure fantasy. It was only its form and appearance that was supposed to be a lookalike of something “chaldean” – which doesn’t exist as a single writing system. For paleograhers the term “Alphabetum Kaldeorum” is the only correct one because it has nothing to do with a chaldeic/semitic origin. And it has, of course, nothing to do with Voynichese.

                    • Darren Worley

                      Notula – regarding your comments that “[the Alphabetum Kaldeorum] deriving from “Chaldean” is pure fantasy”.

                      Whoever coined the term “Alphabetum Kaldeorum” must have thought they were similar, in the same way that the pseudo-kufic script sometimes seen in European artworks represents what kufic was though to be.

                      Similarly, the Alphabetum Kaldeorum may be an “artists’ impression” of the Chaldean script.

    • Martine Tiramani

      My ancestry is from that region of italy and what you say is very interesting. The Scots colonized the po Valley around the year 1200 so italy celtic makes lots of sense. What I thought was interesting about the script is that some of the script ressembles some of the eastern scripts.your reference makes sense. Local dialect.

  23. With more distance, i still say that i’m right in decoding VM … that’s a construction language which details and follows the shapes at the versus of the page… thies language is majoritary made of items for construction !!!

  24. Have you considered the Banasik Manchu theory?

    Using Banasik’s glyph-to-phoneme matrix applied to the Glenn Claston digital transcript, I was able to find many common verbs, nouns, particles, and small phrases that correspond to Manchu words (or Jurchen, the earlier related spoken dialect from the 1400s). The output reads like word salad or glossographia, but it seems to me there’s a surprising amount of signal compared to noise.

    Best wishes and thanks for your contributions to the VMs research efforts,

    • D.N. O'Donovan

      To William Porquet and Stephen Bax,
      Information about Jurchen, and Stolfi’s work is very difficult to come by, and I would certainly be interested to read an essay offering an evaluation of it. (This is quite apart from a recent jeu d’esprit post of mine).

      Would William or Stephen be willing to post an essay like that?

  25. ec

    Looking at the script, it looks very much like an “anglicized” sindhi/indian script — e.g., if i were to align the script along the bottom of the letters and removed the headstroke several indian languages would visually look quite similar to this. Especially given the content (e.g., possible connections to ayurvedic healing and astrological information) and other evidence for indian influence as you mention above. Has this possibility been examined? This would explain some of the confusion in trying to decypher since it most likely would be a syllabary rather than alphabet with use of many conjuncts.

    Sorry if this has been discussed thoroughly already — please let me know if it has I am very curious!

    • Stephen Bax

      Some have indeed sought to suggest resemblances to Indian scripts. However, the problem is that various letters share similarities with letters in many other scripts, not only Indian scripts.

      The only way to make progress is to establish significant and systematic patterns of similarity, and that means a lot of work!!!

      • ec

        Thanks for the reply — that is indeed quite interesting!

        Do you think there is some evidence for syllabics instead of a pure alphabet though? For example, alphabets tend to have approx 20-30 unique symbols while a syllabary might have more like 100 (and is different from glyphs which could have 1000s). I guess I am thinking it could be a syllabics system (consistent with the indian language hypothesis) which at first glance looks like an alphabet but is indeed much closer to a syllabary except that syllables with common sounds have similar symbols.

        Much hard work is required either way! Thanks for spearheading the effort and indulging “novices” like me. 🙂

        • Stephen Bax

          Yes, people have given thought to the syllabics possibility, but to tell the truth the number of signs suggests that it is probably closer to an alphabet/abjad, maybe with some partly syllabic signs. But as yet nothing should be ruled out!

  26. D.N. O'Donovan

    I do wish Thomas Spande were still about. After I’d mentioned the role of Armenians in establishing the kilns in Nusantara, he took up the theme and spent a couple of years (or so) looking at Armenian manuscripts and developing his ‘Armenian thesis’. He wasn’t paid too much attention at the time, but I’m sure he’d be very glad to see how enthusiastically this idea is being taken up now.
    He did a lot of work – much of which he reported on Nick Pelling’s blog.

    Others interested might like to see what they can turn up there under his name.

    • Stephen Bax

      I agree – I was always interested to hear more about his ideas. The Armenian link has always been of interest….

  27. As i often said you, i persist to say that’s only a memory of an art very similar of sewing… the sign on verso are traductions of moves to do the draw on the recto… and i deeply think that there’s no more code there… but the draws, as you remarked, are interresting because they come from discorides, spanish court art, arabic astronomy, and many encyclopedies’ knowledge… then the analysis of the code will lead nowhere else to a final artistic object… but the studies of origins may be (and is) fascinating.

  28. Darren Worley

    Stephen – I’m surprised that you’ve omitted a possible Greek influence in your summary. Was this intentional?

    The absence of a Greek influence would be quite surprising given the geographical areas you’ve suggested.

    The Black Sea region, for example, had a continuous Greek presence from 700 BC until 1922.

    Apparently, the Greek spoken here was much closer to the language of Homer than modern Greek. Surprisingly, some of the settlers here even began a nomadic existence, rather than the reverse (nomadic>settled) that might be expected.

    A surviving example of the language spoken by these nomadic peoples is Ossetian, and there are records of this being rendered in Greek with special diagraphs.

    I suspect the VM represents the mingling of Greek with an Near East/Asian culture – this has occurred many times in various places over the past couple of millennia. (So there are many possibilities.)

    So this mixing of Greek with Central Asian steppe dwellers represents another example (a couple of others being Alexander the Great’s expeditions into India/Pakistan; Hellenistic Jews migrating into Central Asia and India; the Sabians – possibly a mix of Greek and Babylonian Jews etc.etc.)

    In later periods, from mid-13th century to the late 15th century, this Black Sea region also included Genoese and Venetian colonies.

    I think the Greek influence in the VM is fairly evident eg:

    f1v – depicts the Wild Olive (kotinos), and the lions paw is possibly an allusion the killing of a lion with a club made from wild-olive by Heracles.
    f2r – depects Diffuse Knapweed (Centauria Diffusa). This plant also has a Greek mythological connection in that is named after the Centaur Chiron who was, co-incidentally, “killed” by Heracles.

    Furthermore, Derek has proposed several stars following the Greek names.

    • Stephen Bax

      I didn’t omit the Greek possibilities deliberately, and it is useful to be reminded of them – thanks. I went a few months ago to Rostov in Russia and visited the site of a Greek trading colony, which again reminded me of how far Greek communities travelled and how wide their influence was.

    • Darren,
      [re your post of July 14, 2015 – 12:02 am] which I’ve just re-read.

      – without prejudice – (i.e. without necessarily agreeing that the plant is an olive and so forth), the unusual combination of lion+olive is one I’ve come across recently, and I’m racking my brain trying to recall where. … Given what I’m working on at present, I expect it may be arms of an Italian banking company during the late thirteenth to late fifteenth century. Possibly Florentine. No, probably Florentine. The combination is unusual. So if you were correct, that might be an important historical clue.

      But… looking at folio 1v again…… nnnot really sure that you are. 🙁

    • D.N. O'Donovan

      re your post of July 14, 2015 – 12:02 am.

      I’m amazed to see someone else looking closely at these matters, including the Black Sea colonies founded by Greek-speakers, the interaction of Greek and Asian, and even the role of Alexander and his men in the east. Perhaps if you want to work up the idea, it might save you some time to hunt keywords in my two blogs, where you’ll find fairly detailed essays and bibliographic references already on offer about all these points, except perhaps Ossetian. They also occur as commentary on specific folios and the imagery contained in them.

  29. MarcoP

    Hello Stephen, the last section of this 1420 ca Armenian manuscript (LJS 443) contains circular diagrams that seem to me comparable with some of those in the Voynich ms.

    The description says: Collection of commentaries, treatises, tables and diagrams concerning the calendar, by authors from the 7th to the 15th centuries, the most recent being Hakob Ghrimetsʻi, who compiled his commentary in 1416, probably within a decade of the copying of this manuscript. …. Written in a bolorgir script

    • Stephen Bax

      Thanks – to my mind this is potentially a very interesting discovery, given the date and content, as well as the resemblances. I’m off to learn Armenian!

      • Stephen Bax

        It looks as if the most interesting diagrams in the manuscript which Marco identified, including the diagrams he attached to his comment, were from f.145v-212r, which according to this page consist of a “Commentary on the calendar” by “Hovhannēs Vardapet (also known as Hovhannēs Sarkawag)” whose dates are approx 1025-1129.

        • Darren Worley

          Marco – thanks for alerting us to this image you’ve found. I look forward to reading a transcription/translation of this page. Hopefully, one of the visitors to this site might be able to assist.

          I recall that you previously posted a very similar image to your pinterest channel about 6 months ago. Here’s the link.

          Title: Eight phases of the moon corresponding to the Lunar gravitation. By 7th century Armenian Mathematician Anania of Sirak.

          As you noted, this originally appeared here :

          I vaguely recall you posted another similar image too, but the quality of original scan was quite poor. Could you re-post it here?

          • MarcoP

            Thank you Darren! I had completely forgotten about I often put stuff on Pinterest as a bookmark for a future more in-depth reading that I rarely actually do 🙂 I would be cool to get in touch with the author! The discovery of the parallels between Armenian diagrams and the Voynich ms should certainly be credited to him!

            I am sorry, but I don’t remember the image you would like re-posted. Was it another “nebula” diagram with curved lines?

        • MarcoP

          Thank you for the additional information, Stephen! So the most interesting section is not the last one: thank you for the correction!

          I understand very little of these diagrams. f207r apparently includes the symbols of the seven planets, with the Sun occurring twice. The Moon is easily recognizable, I think the Sun and Jupiter follow clockwise. A similar Sun symbol appears in this XVI Century Greek manuscript.

          I find Armenia interesting also because it fits so well with what you wrote about the place of origin of the Voynich ms: “the manuscript was possibly written by or for a community in eastern Europe, the Caucasus, or western Asia – sorry not to be more specific – and has the traces of Arabic, Persian. Mongol, Indian ideas and even language which would not be unexpected or surprising at the time in, say, the area of the Black Sea, the Caucasus or eastern Turkey”.

          • Darren Worley

            Hi Marco – is there any chance you could annotate your ideas on the meaning of f207r. I’m not sure that I quite follow which symbols map to which planet. Do you think the sun appears twice, or could this be Venus? (once as the Evening Star and once as the Morning Star, perhaps?)

            Which diagram(s) in the VM do you think it most closely parallels?

            Regarding your link to the related Greek manuscript – I think the more interesting aspect is that it appears in a Cabalistic treatise i.e. a Esoteric Jewish text. The fact that its in in the Greek language is of lesser relevance, in my opinion.

            Here is the image title: Influence of the moon on reading the signs of the cabala (kabbalah), miniature from the Cabalistic treaty, Greek manuscript, 16th century. Venice, Biblioteca Nazionale Marciana

            • MarcoP

              Hello Darren, I hope the attached image clarifies what I previously wrote: I have highlighted the Moon in blue (labeled “selene” in Greek), the Sun in red (labeled “elios” in Greek) and Jupiter in Green (labeled “zeu[s]”). I am sorry, but since I understand so little of the Armenian manuscript, I currently have no definite opinion about the possible correspondences for LJS 443 f207r in the Voynich manuscript.

              I don’t think this is particularly relevant, but I have some doubts about the correctness of the caption for the Greek image. I think this diagram is about Hermes / Mercury (see the red symbol at the top left). Since the diagram is divided into two halves each having 12 segments, I think it could assign each of the 24 hours of the day of Hermes (Wednesday) to one of the seven planets. The inner circle uses Greek letters to number the 24 segments. So, while I can see that this illustration is Greek, I cannot see anything cabalistic in it.

          • Darren Worley

            I also agree that Armenia is an interesting topic for further research – Stephen has written about possible links with Armenia (and Amirdovlat Amasiatsi) in the past.

            My particular interest relates to its early Judeo-Christian settlers. Armenia was the first country to convert to Christianity in c300, also also large numbers of Hellenistic Jews settled in Armenia in the 4th century (although at this time the Kingdom of Armenia extended beyond its present borders.)

            This article on the Jews of Armenia, describes the recent re-discovery of abandoned medieval Jewish settlements in Armenia, which I think again are a plausible candidate for the origin of the VM (although I think there are better examples). These medieval communities seem to have been literate, and in existence for a considerable time before they were abandoned and the inhabitants migrated elsewhere or were wiped-out.

            Lastly, regarding the manuscript written by the philosopher, mathematician, astronomer, geographer and alchemist, Anania of Sirak; it demonstrates that this type of diagram dates from about 7th-century, so these diagrams do have very ancient origins.

            This website has some novel VM animation – and briefly mentions this same diagram.


            Quote: There’s an even earlier astronomical drawing in the same spirit credited to the Armenian scientist Anania Shirakatsi (AD 610–685), […] from a reproduction in the Soviet Armenian Encyclopedia. The caption identifies it as illustrating “phases of the moon,” but if that’s true, it does so in a way very unlike the previously seen examples. Robert H. Hewsen (“Science in Seventh-Century Armenia: Ananias of Sirak,” Isis 59:1 (Spring 1968), pp. 32-45, at 38) writes that Shirakatsi attributed moon phases “to the fact that the constant movements of the sun and moon cause them to change their positions in regard to one another, which thus results in the change of contacts between the light of the sun and the moon’s surface.”

            Do you know in which manuscript this diagram originally appeared? Where is it, and if it digitized?

            The low-quality image I referred to seems to have been a poor quality scan from the Soviet Armenian Encyclopedia. Marco’s pinterest image is an improvement.

            It can be seen here, together with some related information (in Armenian) – so use Google translate.


          • Darren Worley

            Another good reason to consider Armenia, is that during the medieval period Armenian traders conducted significant amounts of trade with India. This would possibly explain the Indian influence that has been reported in the VM recently.


            This line-of-inquiry would imply that the VM might be written in a mercantile script.

            These scripts, that were used by merchants and businessmen, appear to have been commonly used in Northern India. e.g.

    • David Wiffler

      Hello everyone, I thought you all might like this little curiosity I found. On the fourth page of the calender manuscript ( there is a symbol that is interestingly similar to one in the voynich ms(f42v I think?).

      Anyhow, it is probably just a common medieval motif but I thought I’d point it out just for fun.

      • Darren Worley

        Thanks David. Well spotted! What do you think this sign/symbol represents? Is it a monogram of the scribe that wrote/copied the manuscript, or perhaps an owners monogram?

        What do you make of the writing above it? Cursive Armenian?

        Do you really think this sign is a common medieval motif? It seems quite unusual to me.

        I’d be interested to know of other examples. Is there a specialist term for this kind of symbol/writing?

        Lastly – I came across this blog post with a lot of links to other early Armenian manuscripts.

        • MarcoP

          Hello David and Darren. I think doodles like these are not very common: most XV century manuscript have nothing like that. Similar decorations from the Swiss manuscript COD. SANG. 754 (e.g. p.153) have been discussed by Brian Cham.

          I think that, since it is not part of the main text, the Armenian flourished monogram might be a later addition.

          • Darren Worley

            There seems to be a similar-ish Georgian tradition for these calligraphic monograms – they’re called Khelrtva

          • Darren Worley

            In my opinion, one of the most unique and identifiable aspects of the VM calligraphy are the loops containing dots.

            This is clearly seen in the opening letter of f42v, as David Wiffler remarked earlier.


            I’ve come across 2 further similar examples, and these originate from Switzerland and Southern France respectively, so perhaps
            this calligraphic style might be Western European, after all.

            The first example, is taken from diagram describing the “Dynasty of the Dukes of Gruyere (c1000 – c1270)” (Dynastie des Comte de Gruyere).
            The diagram is unattributed – but I would hazard a guess that this calligraphic monogram is a modern reproduction taken from an original manuscript from this period and location.

            This lead me to focus on France/Switzerland as a source for further examples and I quickly located this better image:


            The origin of this signature, is a little unclear, but its being used to illustrate “Signatures, écritures et pouvoirs autour de Gaston III comte de Foix dit Fébus”
            In English : Signatures, records and authorities around Gaston III, count of Foix or Fébus.

            Hopefully, someone with better French or investigative skills can clarify and/or locate the precise origin of this image.

            This image appears to be the word “Febus” with loops and dots, just as found in the VM.

            This individual named Gaston III lived from 1331 – 1391, and was duke in Foix the capital of the county of Foix. Foix is closely associated with the Albigensian Crusade that occured a century earlier.

            Quote from Wikipedia: Gaston III/X of Foix-Béarn, also Gaston Fébus or Gaston Phoebus (1331 – 1391) was the 11th count of Foix, and viscount of Béarn (1343–1391). Officially, he was Gaston III of Foix and Gaston X of Béarn.

            Can anyone find other similar monograms from these regions?

          • Darren Worley

            I’ve discovered a set of medieval documents that contain several examples that are very similar to the VM calligraphy.

            I’ve attached examples below – these are signatures found in medieval charters from Catalonia.

            Here is the original :

            Source: Biblioteca Universitària de Barcelona, Pergamins C (Sant Pere de Casseres) núm 20

            This originates from the Sant Pere de Casserres Benedictine monastery and here

            And is well described and transaction in this blog post – here

            The author of the blog post queries the script used to sign this charter.

            Quote: […] however that this one’s especially good for the range of scripts; all these people seem to have signed for real and they none of them show the same hand, which is interesting, because so much of the writing we have is in formal scripts, it’s fascinating to me that that isn’t what people use when they sign their names. Are they going up a register so as to stand out? Or are they reverting to their ‘usual’ hand? I don’t know.

            Could the VM script be a vernacular script from the region of Catalonia towards the border with France? Like the other example I gave earlier (the signature of the Gaston Febus) the County of Foix is fairly near to the Sant Pere de Casseres monastery (approx 200km apart).

            Perhaps more importantly, I think this also suggests a useful source of documents for further comparison. I don’t think the VM really belongs to the ecclesiastical manuscript tradition so looking at medieval charters, legal contracts and maps from this area would be a useful direction for further study.

          • Darren Worley

            And here is a link to another Catalan medieval charter containing signatures similar to VM calligraphy.


            Source: Arxiu Capitular de Vic, Calaix 9, I, no. 102

            ..and its described here

            • Darren,
              re yours of August 4, 2015 – 10:26 pm

              what a coincidence, eh?

              I happened to link to exactly the same blog, and the same image, and the owner of the blog was kind enough to explain his header-picture too.

              Let’s see… I think that was part of the revisionist study of the Sagittarius figure.. anyway, interesting co-incidence again – as always.

          • I’m not sure how widely people look, apart from the small parochial set of manuscripts which might support their preferred ‘Voynich story’. Such pen-flourishes are not at all uncommon, and I’m not surprised that they appear (among so many other regions) in manuscripts now in st.gall.

            The interesting thing is that their heyday (if you want to call it that) is not in the fifteenth century as I recall, but rather earlier. I have a nice collection from Jewish and Latin manuscripts, ranging from the eighth (not so many) through to the time the Vms is thought made i.e. 1438. And that’s just in manuscripts. When you start looking at some ceramic works, and especially at engraved metal, you find that the “flourish” was a real sub-category on its own. Perhaps someone, somewhere has written a book on it. Still, nice to know that st.gall didn’t miss out. 🙂

            • Brief random thoughts from an artist/writer:

              Have any of you ever done calligraphy? When I was learning calligraphy, doing loops like that is one of the first things I did before I began making letters. It’s a move that gets the ink flowing well, and gives you a good idea how the ink will flow on a particular type of paper–also let you know how much the dot will bleed. The modern equivalent would be shaking or tapping your pen, then scribbling before using it to get the ink flowing. It was like a warm-up before I began lettering. Almost a nervous habit. I can’t imagine doing calligraphy without doing that first.

              Artistically, it echoes the way the rows of women in the manuscript link hands if you see the bottom of the loops as linked arms, and the tops as the women’s heads (if it was someone’s It’s a very natural shape to make. I’m not saying all the marks are unintentional, but rather suggesting that some of the marks may just be doodles that would ordinarily have been put on a piece of scrap paper. When paper was more precious, perhaps there was no such thing as “scrap paper.” If it is a “Khelrtva,” then they might have slipped the shape (consiously or not) into the illustrations as well.

    • D.N. O'Donovan

      some of the diagrams are nearly-fractal in their impact.

      I’ve made Fol. 203r [=204r

      my new desktop.

      Thanks –

    • Lee

      Hello Stephen,

      I was just curious to find out more about the V Ms when I stumbled across your site.

      For some reason, I seemed to find visual similarities with the astrological illustrations and an image I found online at it has something to do with electromagnetic torsion fields.

      There may not be any correlation at all, I’m just a curious person who likes to explore strange and peculiar things.

      Anyway, good luck on your continuing work.

      Kind regards


    • Jan Kammerath

      This to me seems to underline my assumption that the Voynich Manuscript might actually be a copy or translation and the author is not the primary source of information and not the creator of the original document.

      I also find Stephen’s theory of an extinct language very reasonable and would even go so far to assume that a wealthy person ordered a universal scholar to capture information that were only available as spoken language.

      As the scholar might not have been capable of writing or writing of the source language might not have existed at all, he could have used his existing knowledge of languages to try to improvise or create the writing of this language.

      This would also explain to me the weird images of bathing women which might be the pictorial representation of a complicated explanation or instruction for bathing or showering.

      The use of Caucasian language influence together with Arabic, Persian, Indian perfectly relates to the Georgian, Armenian, Iranian region. This area was also very close to major trade routes.

      My bet (and I’m completely speculating here!) is that it is a 15th century Beauty Guide with herbals, baths, astronomic and esoteric sections. I wish it’s the holy grail, but I think that’ll remain wishful thinking.

  30. Derek Vogt

    Not a bad question, just asked too soon. Translation can’t begin until we know the sound system and what other known language(s) it’s related to (if there are any). I’m optimistic that we’re close to getting the former done, but not quite yet, and the latter is almost completely untouched so far. Without them, it’s just to soon to even think of translation yet.

    • Valérie-Anne

      Ok, it’s interesting. I try to find my way without reading this place’s researches. (I used to play video games when I was younger and I don’t want to be spoiled.) So when I find something, I check and sometime it’s already found.
      I just found the coriander on 41v. Not the same way as mister bax did, but, I guess I’m on the right way… 🙂

  31. Valérie-Anne

    Ok I guess it was a wrong question

  32. Valérie-Anne Bertin

    Hello Mister Bax, I completely agree with your point of you, I just cannot believe it!
    Have you began the translation of the text?

Leave Your Comment

Your email will not be published or shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>