One who writes an overview of a field of study, however general or specific, is not a neutral observer. By the very act of his writing, he becomes a participant himself. An actor between research and reader, he selects what is worthy of inclusion and what isn’t. He might even add a critical remark or two, a word of warning. By describing the landscape, he cannot avoid shaping it as well.

A landscape.

This is not necessarily a bad practice since, admittedly, libraries full of nonsense have been written about the Voynich manuscript. Newcomers will surely feel less overwhelmed with some pointers at their disposal. But when the guide is engaged within this contentious field, he must keep the reader aware of that fact – by speaking in the first person, by referring to his preferences, by explaining why he prefers one opinion over another, and by not presenting his views as if stating facts

I feel that there is a problem, and judging by the comments on my earlier post, this feeling is shared by others as well. Some of the most prominent Voynich researchers maintain an air of objectivity, while they actually represent just one specific cluster of views on the manuscript, which I call the “dominant paradigm”[1]. Certain studies, discoveries and “expert opinions” are systematically favored at the expense of others, guided by the author’s own convictions and experience.

This may not be entirely exceptional within scientific debates nor, I believe, is it done with bad intentions. But where in an ideal scenario the author will be explicit about his vested interest and the frame of reference he favors, the study of the Voynich has seen the dominant paradigm become synonymous with “common sense”.

That is why I have decided to write a short history of Voynich studies. Not necessarily to prove something about the manuscript. Rather, I want to demonstrate that even if an overview sticks to the facts and reports objectively about the findings and opinions of others, a narrative is created purely by the process of selection.


17th century

The earliest Voynich theory we know of was formulated by the manuscript’s first known owner, Georg Baresch, who had the manuscript in the early 17th century. Around this time, Jesuit scholar Athanasius Kircher was showing interest in the Egyptian script. Baresch wrote several letters to the self-proclaimed “Oedipus of Egypt” about the mystery manuscript sitting on his shelf, hoping that Kircher could identify the script.

Kircher had himself drawn as Oedipus teaching the Sphinx a lesson.

We don’t know for sure how and when Baresch obtained the manuscript some two centuries after its creation. We also don’t know if any information about its contents or origin accompanied the artefact when it came into his possession. We do know, however, what he thought it was all about. In a letter to Kircher, he writes:

In fact it is easily conceivable that some man of quality went to oriental parts in quest of true medicine (he would have grasped that popular medicine here in Europe is of little value). He would have acquired the treasures of Egyptian medicine partly from the written literature and also from associating with experts in the art, brought them back with him and buried them in this book in the same script. This is all the more plausible because the volume contains pictures of exotic plants which have escaped observation here in Germany. [2]

In short, Baresch believed the script and general contents of the manuscript to be non-European, wisdom gathered by a traveler written down in a foreign script. Baresch was convinced he could not read the manuscript because it was gained from a non-European culture.

20th century

The manuscript appears again when Wilfrid Voynich reveals it to the public in 1915. Voynich’ own theory was that his “Roger Bacon cipher manuscript” had been brought to Prague by John Dee. The Bacon authorship was already considered unlikely or even impossible soon after Voynich’ death [3].

One of the first scholars to examine the manuscript was art historian Erwin Panofsky. He believed that the VM was made in the early 15th century (remarkably accurate) in Spain or somewhere southern. As for the culture, he names Jewish and Arab influences, as well as “some Spanish or anyhow southern qualities”. Ethel Voynich, in her notes on a conversation with Panofsky, adds that there were “learned Jewish and Christianized Arab scholars” in Spain during the early 15th century and that hence there is “nothing inherently impossible in this suggestion”.[4]


In short, Panofsky dated the artifact accurately at first sight, and saw a cultural blend of “Spanish/southern”, Jewish and/or Arab influences. He was later influenced to change most of his views, guided by a number of dubious claims about sunflowers and marginalia. Carbon dating of the manuscript has since revealed that his earlier assessment of the date range was correct.

Cryptology expert Captain Prescott H. Currier presented his findings about the Voynich manuscript during a seminar in 1976. Today still, some of his insights are relied upon: based on statistics, he discerned two “languages” in the text. Additionally, after analysis of the handwriting, he estimated that “six to eight scribes (copyists, encipherers, call them what you will)” penned the main text. For him, there was only one possible conclusion: “It must be a copying job.”[5]

Recent opinions

About the Material

In a recent interview, Dr Raymond Clemens, curator of manuscripts at the Beinecke library, talked about the tests that have been performed on the manuscript’s materials. While we can be confident about the time of manufacture (early 15th century), we know much less about the location. The inks are mostly vegetable based, and according to Clemens those were “largely cheap and readily available” throughout much of the Mediterranean basin. [6] In other words, we still have no way to determine where exactly the manuscript was fashioned.

About the language

The 1990’s and early 2000’s too saw a number of academics taking interest in the manuscript. Among them computer scientist Jorge Stolfi, whose previous work included computational linguistics projects [7].  He argued that, based on the properties of the text, the possibilities of what it could be are extremely limited: either an excessively complex code, or a monosyllabic language.

He preferred the latter, and demonstrated that Voynichese could be a language like Chinese. Since at the time cipher theories were popular, such insights were not always welcomed. The following quote by Stolfi (2002) paints a picture:

I am aware that many quite reasonable people, like Glen and Philip, find a non-European origin so unlikely /a priori/ that they would rather believe in impractically complicated codes, Byzantine decoys, and secretive communities of herbal conspirators, just to avoid it. [8]

Another linguist who had taken interest in the manuscript was Jacques Guy. In January 1992, he offered an overview of his opinions:

The manuscript was written by several authors. […] The text is meaningful. […] The language is real, natural, it is neither invented,nor artificial, nor philosophical […] Needless to say, there is no cipher there, and I am in good company asserting that: William Friedman had come to the same conclusion.

He then suggests, albeit with less certainty, that the manuscript was written in the 12th century, in Italy, Southern France, or Northern Spain “in a language now extinct, indigenous to Europe, but non-Indo-European.”

Guy adds that phenomena usually found in spoken language can be reflected in written language, citing Sanskrit as an example [9]. He argues that similar phenomena can explain the statistical oddities found in Voynichese.

Imagine now a language with such tendencies as Sanskrit; where words are run together, except, naturally enough, where a pause occurs in speech. As you, a scribe, write in this language, I dare say that you will tend to pause long enough at the end of a line so that sandhi rules cease to apply. The statistical properties of what you write will, I think, be quite similar to those observed by Currier in the Voynich manuscript. Letter frequencies at the beginning and at the end of lines will differ from anywhere else.

Similar ideas are found in the work of Emma May Smith, who also considers sandhi effects and speech-to-writing uncertainties as important factors [10]. She believes Voynichese is unlikely to be an Indo-European language and thinks of Turkic as the most likely language family (without excluding other possibilities). [11]

About the images

The most comprehensive analysis of the manuscript’s imagery was published by Diane O’Donovan on her blog. She is one of the few experienced iconography specialists to have studied the Voynich for a sustained period of time. To summarize her conclusions in a short paragraph (as I must) is difficult, but they include the following:

  • The manuscript’s imagery has its roots in Hellenistic sources.
  • “In their method of construction, the plant pictures in the manuscript show they are not related to the western tradition in herbals.”
  • “The Voynich imagery shows remarkably little sign of [adaptation to a Western environment]. Its non-Latin, and indeed non-European character remains clear.” [12]

In short, the manuscript cannot be the product of a 15th century European “author”.

Many other qualified specialists have offered their opinion, disconnecting the Voynich manuscript from known European traditions. A selection:

Ethnobotanists Guy Mazars and Christophe Wiart are specialized in the subjects of East Asian and Indian plants. They believe the manuscript’s botanical sections are “une pharmacopée d’origine indienne” and identified several plants as medicinal species from the Indian subcontinent. [13]

Christophe Wiart with some plants.

Agricultural researcher George Turnbull offered his opinion on the manuscript’s plants, calling the lack of common European medicinal plants like dandelions “disturbing”. “Perhaps this is a pharmaceutical manual from another location that was copied and prepared for a European audience.” [14]

The VM, especially Q13, is at times likened to imagery of the alchemical tradition. However, two renowned specialists on the subject, Adam McLean and Jennifer Rampling, have both stated that they see no connection. [15]

Jennifer Rampling
Jennifer Rampling with the Voynich manuscript.

Finally, professional historian of science Jordan D. Marche’ II, Ph.D. noted similarities between the Greco-Roman period Egyptian Dendera Zodiac and the Voynich “Zodiac section”. He suggests transmission through Hermetic sources.[16]


What I have done here is select those specialist opinions and statements which are compatible with my own views on the manuscript. Proponents of the dominant paradigm do this as well, but they are less forward about the subjective nature of their selection, because their opinion equals common sense. As Jorge Stolfi already knew, there is an a priori rejection of any explanation which dwells too far outside of European borders.

In general, there is more tolerance for “exotic” language proposals (though not too much), but when it comes to cultural aspects of the imagery, those who look beyond the wrong borders are not taken seriously. The dominant paradigm embraces the manuscript as the work of a European genius, scientist or sex-obsessed quack, despite overwhelming evidence to the contrary, supported by specialist opinions.

This post is not meant as a provocation, though it might be taken as such. Rather, it is an invitation to reconsider what we believe, and why we believe it. The foundations of our beliefs about the manuscript, where do they come from? Who built them? And especially, how sturdy is their construction?


//I expect there will be some mistakes in this post – do not hesitate to suggest corrections.

// I would like to thank Sam G, Diane O’Donovan and Emma May Smith for answering my questions about their respective research areas.

[1] For an animated discussion on the dominant paradigm, I refer to the comments on my previous post. There is much more to be said, but this should give you an idea about the various points of view on this issue.

[2] Taken from Philip Neal’s translation:

[3] See

[4] See

[5] All Currier quotes taken from the transcription of Currier’s work by Jacques Guy and
Jim Reeds, available in pdf form here:

[6] For a transcription of the interview, see

[7] “At UNICAMP Jorge also worked with C. Lucchesi and T.Kowaltowski on finite state transducer technology for spell checking and other natural language processing tasks.” (wiki)

[8] See As to why Stolfi rejects any cipher theories, he summarizes: “I offer the word structure, Zipf plots, and label structure as arguments against character-based ciphers; and the size of the text as argument against codebook-based ones.” The rest of the page is also an interesting read.

I also recommend the following page, where Stolfi explains why he believes the VM is an attempt to write a tonal language in a non-native script:

[9] An interesting aspect of Guy’s contributions to the mailing list is that he often resorted to exotic languages like Chinese and Sanskrit to explain the features of Voynichese, but, to my knowledge never committed to the possibility of the manuscript’s actually being in such a language.

[10] Emma’s Transformation Theory is certainly one of the most promising current developments in Voynich-related linguistic research.

[11] Pers. comm.

[12] See

[13] From an article in Actualites en Phytotherapie available in pdf format here:

[14] Turnbull’s opinion is quoted on without reference of the source.

[15] Rampling wrote an essay in the Yale edition. McLean’s opinion can be read here. The editor in the mailing list added the following note to McLean’s reply in 1998: “Adam McLean is probably the world’s greatest expert on alchemy. Therefore, this note is probably the last word on the subject.” It seems like there are no last words in Voynich studies.

[16] See