Yes, in large part, and the answer has been in print since 2018. Levente Zoltan Kiraly and Gabor Tokai published a working reading of the Rohonc Codex in the journal Cryptologia, and the book that comes out of it is not a lost gospel or a heretical sect's secret. On their reading it is a late-sixteenth-century Catholic devotional compilation: a long Passion narrative, parables, Marian prayers, saints' legends, and a scattering of stories nobody has yet matched to any other source.
The reason this is not treated as a closed case is narrower than it sounds. A decipherment of an unknown script is only checkable once the dictionary and the grammar are published in full, and that part is still in preparation. What exists in the meantime is a character-by-character digitisation and an online tool the two researchers built so that anyone can test their claims against the text.
So the interesting material is no longer whether the book can be read. It is what the object itself records, and how much of the argument has always been decided by the paper rather than by the writing on it.
Key Takeaways
- A reading of the Rohonc Codex was published in Cryptologia in 2018 by Levente Zoltan Kiraly and Gabor Tokai. The signs are not a letter substitution; they do not track the sound of words at all.
- On that reading the book dates itself in a colophon-like entry, counting from the Ascension rather than the Nativity. Converted by the author's own arithmetic, that is 1593.
- The paper was dated to the sixteenth century in 1842 by Janos Jerney, who matched its anchor watermark against several thousand charters in his own collection.
- Three authorities have counted the drawings and got three different totals: about 28, then 57, then 87. They were not counting the same thing.
- A monochrome facsimile of the whole manuscript has been openly downloadable from the Academy library's digital archive since 2015.
Has the Rohonc Codex been deciphered?
The central finding is that the script was never a cipher in the sense everyone spent a century assuming. Kiraly and Tokai describe a code system in which the signs carry meaning directly and do not encode the sound-shape of any spoken word, closer in behaviour to Chinese characters, which readers of different languages can share on the page without agreeing on a single pronunciation.
That is why substitution attacks failed so completely. There was no sound layer underneath to recover.
Their first published step was less glamorous and more useful: working out the order of the pages. Reading the text for sense told them which leaf followed which, and the sequence that emerged does not match the manuscript's current binding. The opening page of the text, by the numbering the manuscript now carries, is the reverse side of the fourth leaf.
With the order fixed, the shape of the book comes out clearly (Kiraly 2023, pp. 365-367). The first quarter, running to the front of leaf 54, covers primordial history, a short summary of Jesus's life, and then a long and detailed Passion. The second quarter, to the front of leaf 137, runs from the resurrection to Pentecost and takes in parables and teachings. After that it grows miscellaneous: Marian prayers with the years of indulgence attached, legends, more miracle stories. The last pages are barely legible, worn by handling and thick with what Kiraly calls peculiar vocabulary; on his reading they probably concern the period in which the book was written.
Here is a line a reader can check. On leaf 137, verso, lines three and four, the text carries a version of the prayer Ave Sanctissima Maria. Read in the manuscript's own direction, right to left, the signs gloss as: you-Mary, conceive, Jesus, without, sin; then Mary-bear, lord-creator. Against the Latin, Tu concepisti Jesum sine peccato and Tu peperisti creatorem, that is a close fit. The oddity is what stands where the pronoun and the verb ending should be: the sign for Mary's name, used consistently in their place. A dictionary alone will not get you through a sentence built like that, which is why the grammar has taken longer to publish than the word list.
What the book says about its own date
There is exactly one place where the manuscript refers to the circumstances of its own making, and it is in the upper lines of leaf 223, verso. On Kiraly and Tokai's reading it gives the date of writing or completion as 1560 years after the Lord Jesus's going to the Father.
That is not a typographical slip for 1560 in our reckoning. The author counts from the Ascension, not the Nativity, and elsewhere in the book dates the resurrection to 33 years after the birth. Adding the author's own offset gives 1593.
The same calendar runs through everything after the New Testament. The siege of Jerusalem is placed at 24, the Exaltation of the Cross at 26, Augustine at 60. The Last Judgement is placed at 2000. Kiraly leaves that one unconverted; applying the author's offset in the same way the 1593 figure is derived would put it in 2033, and that arithmetic is ours rather than his.
Older events get a different treatment. The author counts 5166 years from Adam's fall to the birth of Christ, where the standard medieval calendar counts 5199 from the creation of the world, and the two reconcile only if you assume Adam was 33 at the fall.
Inside that frame the internal arithmetic goes visibly loose. Seth returns to the gate of Paradise in year 47, Noah and the flood fall in year 50, Abraham is placed at 120, the Elijah stories at around 536, and then Moses arrives with a jump to a barely legible 3570 or 3670. Kiraly reads this, along with the frequently mistaken biblical citations, as the work of somebody who knew the stories well but by ear, and who was not in a position to check them against a library.
Some of the details have no known parallel anywhere. In the parable of the talents, at leaf 173, the three servants receive five, three and one, where Matthew 25:15 gives five, two and one. The codex says Jesus wept five times, one of those occasions being at the destruction of "the five cities", which no one has yet identified, nor even worked out which five cities are meant. At leaf 100 the author quotes an unnamed saint to the effect that the Lord created Elijah before the sun and the moon.
Why the argument over the Rohonc Codex has always turned on the paper
The first published examination was read to the Hungarian Academy's small assembly on 18 July 1842 by Janos Jerney, and printed two years later (Tudomanytar, Ertekezesek vol. 15, pp. 25-36). It is also where the manuscript got its name: Jerney proposed calling it the Rohonc codex, after the place it came from, and it stuck.
He recorded that the book had already been round Vienna, where scholars and antiquarians had pronounced it unreadable, and that the Academy's secretary had brought it forward in 1840 with a proposal to publish a few pages in facsimile so that somebody else might have a go.
Jerney's own contribution was physical. He noticed watermarks in the paper, mostly broken up by the folding but showing whole on one leaf. The mark was an anchor inside a circle. He had spent years collecting Hungarian charters, several thousand of them, and he went through the collection looking for a match. He found paper carrying the same mark and the same quality on documents from the sixteenth century, and on none from later. Cross-referencing a printing history of 1808, he concluded the stock came from an Italian mill.
He also read the wear as evidence. The cover was extraordinarily worn and the paper notably dirtied in places, which he took as proof of long and frequent use, and therefore, in his words, that the work was by no means fabricated to deceive the later world, or made in play.
The attribution that later hardened into consensus rests on much less than that. In 1866 the historian Karoly Szabo swept the codex in with a batch of exposed Hungarian forgeries, writing that if he was not mistaken it too had come from the hand of a known fabricator. He offered it as a supposition, hedged in that exact form, with no argument attached. Kiraly notes that the guess nonetheless became the standard scholarly position, and that the manuscript shows no sign of having been made to imitate the runic writing Szabo was actually cataloguing.
The Academy library's own printed catalogue still carries the tension in a single sentence. Its 1973 entry describes the manuscript as probably a nineteenth-century forgery, on paper from the beginning of the sixteenth century (Csapodi 1973, p. 109).
Why no two people have counted the same number of pictures
Three published counts of the drawings exist, and no two agree.
| Authority |
Year |
Count |
What was counted |
| Janos Jerney |
1844 |
about 28 |
Only vignettes filling a whole page or most of one; he left the smaller class unnumbered |
| Csapodi's MTAK catalogue |
1973 |
57 |
Pen drawings, with no split by size |
| Kiraly and Tokai |
2018 onward, figure reported 2023 |
87 |
Ink drawings, of which about a quarter show a scene identifiable from the picture alone |
None of these people were being careless. They were counting under different definitions, at different levels of access, and before and after the pages had been put in a readable order.
The signs have the same problem. Kalman Nemati, working in 1889, distinguished around 800 of them. Jerney had already run into the reason the totals scatter: he arrived at his own impression of the count by tallying the letters' variant forms alongside their base shapes (Tudomanytar 1844, p. 29), and the number you get depends on whether you do that. Nemati's own method is not recorded. His alphabet of the codex is held in the same manuscript collection as the codex itself, catalogued separately as Ms 884.
It is a useful reminder for any unread manuscript that the headline figures quoted about it, including the ones quoted about the Voynich manuscript, are usually somebody's editorial decision about what counts as one unit.
What the drawings were tested against
Jerney tried to place the author from the architecture. He was struck by how many of the towers in the drawings are topped with crescents, and combined that with the robes and headgear of the figures to suggest an Eastern Orthodox author.
That hypothesis has now been tested by counting rather than by impression, and it does not hold (Kiraly 2023, p. 366). Crescents account for 36 percent of the tower ornaments. Knobs account for 18 percent and twisted crosses for 10 percent, and the remainder is a mixture of crosses, flags, triangles, closed semicircles and plain horizontal lines that carry no denominational signal at all. Before the eighteenth century, biblical scenes were routinely drawn in contemporary dress and contemporary buildings, so the likeliest reading is an attempt to render the Holy Land as it then looked.
What the pictures do carry is a trail of copied compositions. The Entry into Jerusalem on leaf 14 compresses two episodes into one frame and follows Byzantine composition of the kind found in the twelfth-century mosaics at Monreale and the Cappella Palatina in Palermo. The Adoration of the Kings tracks Pietro Cavallini's late-thirteenth-century work at Santa Maria in Trastevere. Angels catch the blood of the crucifixion in chalices, as in Giotto's Scrovegni Chapel fresco. Two leaves carry a semicircular cross composition that echoes the apses of late antique churches such as Sant'Apollinare in Classe at Ravenna.
The small details point somewhere else again: Gothic speech scrolls, a fluttering loincloth of a type first seen in Netherlandish painting, a banner whose cloth swings from a free-turning rod in the northern manner rather than being fixed to the pole. Art historians consulted separately put the halberds and the tower-nave-sanctuary church plan in the sixteenth or seventeenth century and mostly in Central Europe.
Taken together, the reading is a Central European artist who had seen Italy and copied what was publicly on view there. That is a long way from an unbreakable secret, and closer to the working life of an ordinary devotional book than any of the theories that surround it. Anyone drawn to that gap between the reputation of a puzzle and its evidence will find the same pattern on the Dorabella cipher.
Can you look at the Rohonc Codex yourself?
Yes, and this is the part most accounts leave out. The manuscript is held in Budapest, in the Manuscript and Rare Books Collection of the Library and Information Centre of the Hungarian Academy of Sciences, under the shelf mark K 114. Its older shelf mark was Magyar Codex 12mo, no. 1, and it is the last item in the series of Hungarian codices that runs from K 31 to K 114.
The catalogue description is worth reading beside the mythology. Paper, 120 by 100 millimetres, 224 leaves, script unknown, language not established, binding a wallet-style leather cover kept in a half-leather case, place of origin Hungary with a question mark. Two items are kept with it as enclosures: a postcard sent from Innsbruck in 1885 about attempts to decipher it, and two photographs of the watermark with a dating of the paper, made by Kalman Nemati later in the nineteenth century. A microfilm has existed for decades.
The Academy library placed a scan in its open digital archive in 2015. The high-resolution colour version is restricted to library staff, but a monochrome preview of the whole book, 227 scanned images, is downloadable under an open licence for academic and educational use. A separate image collection carries five high-resolution photographs made in 2016, where the catalogue description is admirably blunt: a manuscript on sixteenth-century Venetian paper, written in an unknown sign system, of undetermined content.
That is an unusual position for a famous unread book to be in. The object is smaller than a modern paperback, its paper was dated from its watermark in 1842 and again later in the same century, its pictures have been traced to identifiable Italian models, and a reading of its text exists in a peer-reviewed journal. What is still genuinely missing is a name for the person who wrote it, and a published dictionary that would let everybody else check the reading line by line.
Frequently asked questions
Is the Rohonc Codex the same kind of puzzle as the Voynich manuscript?
Not really, and the difference is in the script rather than the reputation. The Voynich text still resists a reading that other researchers can reproduce. The Rohonc Codex has a published reading in Cryptologia, a reconstructed page order, and an identified genre, because its signs turned out to carry meaning directly instead of standing in for sounds. What it lacks is the full published dictionary and grammar that would let the reading be independently verified end to end.
What year was the Rohonc Codex written?
On Kiraly and Tokai's reading, the manuscript dates itself on its second-to-last leaf, but it counts from the Ascension rather than from the Nativity, so the number in the book is not the number in our calendar. Applying the author's own offset of 33 years, which the book states elsewhere, gives 1593. That sits inside the sixteenth-century date assigned to the paper from its watermark by Janos Jerney in 1842 and again by Kalman Nemati later in the same century.
How was the paper dated?
By watermark comparison, and by an unusually direct method. Janos Jerney found an anchor inside a circle in the codex paper, then searched his own collection of several thousand Hungarian charters for the identical mark. He found it on sixteenth-century documents and on nothing later, and concluded the stock was in general use in Hungary during that century. Two photographs of the watermark, with a dating of the paper made by Kalman Nemati later in the nineteenth century, are still stored with the manuscript, and Kiraly reports that a subsequent analysis of the ink does not appear to contradict a sixteenth-century date.
What is actually in the Rohonc Codex?
On Kiraly and Tokai's reading, devotional material, mostly rewritten from the New Testament. The largest single block is a long Passion narrative, followed by a section running from the resurrection to Pentecost, and after that a mixture of Marian prayers, saints' legends and parables. Several episodes have no known parallel, including a version of the parable of the talents in which the servants receive five, three and one rather than the five, two and one of Matthew.
Can I see the pages without travelling to Budapest?
You can. The Academy library published a monochrome scan of the entire manuscript in its open digital archive in 2015, and it can be downloaded for academic and educational use; the colour scan is restricted. The two researchers who published the reading also built a public site pairing their dictionary with their character-level transcription of the text, so that the claims they make about it can be checked.
If the pattern of a famous puzzle turning out to be an ordinary object holds your interest, the Beale ciphers and the Somerton Man are the next two to read.
Sources
- Janos Jerney, "Nemi vilagositasok az ismeretlen jellemu rohonczi irott konyvre", Tudomanytar, Ertekezesek vol. 15 (1844), pp. 25-36. Read to the Academy on 18 July 1842. Physical description, the anchor watermark and the charter comparison at pp. 26-31; the count of full-page vignettes and the argument from wear at pp. 27-28; the crescent-tower inference at p. 28 - retrieved 10 August 2026: full scan, Library of the Hungarian Academy of Sciences
- Csaba Csapodi, A "Magyar Codexek" elnevezesu gyujtemeny (K 31 - K 114), Catalogues of the Manuscript Collection of the Library of the Hungarian Academy of Sciences 5 (Budapest, 1973), p. 109. The formal catalogue entry for K 114: dimensions, leaf count, 57 pen drawings, binding, the forgery note and the enclosed watermark photographs - retrieved 10 August 2026: full scan
- "Rohonc Codex", Library of the Hungarian Academy of Sciences, manuscript archive record for K 114, deposited 29 June 2015. The open monochrome facsimile linked from that record runs to 227 scanned images - retrieved 10 August 2026: record and facsimile
- "Rohonci kodex", Library and Information Centre of the Hungarian Academy of Sciences, image archive record 2216, digitised by Klara Lang in 2016 - retrieved 10 August 2026: record
- Levente Zoltan Kiraly, "A Rohonci kodex teologiai karaktere", in Hagyomany, Identitas, Tortenelem 2022 (Budapest, 2023), pp. 363-376. Contents and page order at pp. 365-367; the tower-ornament statistics at p. 366; the iconographic parallels at pp. 367-370; the calendar and the idiosyncratic passages at pp. 373-374; the Ave Sanctissima Maria gloss at p. 365 - retrieved 10 August 2026: full text
- Levente Zoltan Kiraly and Gabor Tokai, "Cracking the code of the Rohonc Codex", Cryptologia, volume 42, issue 4 (2018), pp. 285-315. The article itself sits behind the journal's paywall and was not opened for this piece; it is cited here from the authors' own account of it in the openly available 2023 chapter above, which places the page-order reconstruction at pp. 288-291 of the original.
- The researchers' public dictionary and transcription tool - retrieved 10 August 2026: Der Rechnitzer Kodex
