Using Script Against Undesirable Readers: The Coded Messages of Mihovil and Antun Vrančić

Abstract

During the period of their correspondence, the brothers Mihovil and Antun Vrančić would occasionaly write messages in code[1]: Antun in a letter written while on a diplomatic mission in Paris in 1546, and Mihovil in four letters written in Šibenik some twelve years later. While conducting the necessary research required for our investigation, we discovered that one letter had, until now, remained unknown. We succeeded in deciphering Mihovil's system of signs by comparing the frequency of signs in the coded parts of the text with the frequency of signs in those parts of his letter composed in Latin script. The majority of the signs are derived from the Latin script and only their quality has beeen changed. It seems that this was influenced by the Polygraphia, a work by the then contemporary cryptographer Iohannes Trithemius. In addition, a certain number of Arabic signs as well as signs from other scripts have been included.

Most of the hidden content deals with investments in real estate and other business ventures, thus the purpose of this secret system of writing was evidently to prevent the competition from gaining a possible advantage over Antun and Mihovil. At the time, after a four-year diplomatic embassy position in Turkey, Antun had been awarded by the Emperor Ferdinand and his income had increased significantly. The need for caution and discretion is evident in two messages in which we find Mihovil warning his brother to be wary of two other brothers who have a dubious moral reputation.

Antun’s extract remains unsolved as he wrote using a different system. Additionaly, the text itself is quite brief so an analysis of the frequency of signs is of little use.

Keywords: Mihovil Vrančić, Antun Vrančić, correspondence, cyphers, decyphering, Šibenik, investments, Renaissance diplomacy

The Corpus

It was in 1863 during the publication of the seventh volume of the collected works of Antun Vrančić that Laszlo Salay informed the scientific community that the two oldest and closest brothers of the Vrančić family had a habit of using a secret code in their correspondence. As a footnote to three of Mihovil’s letters from 1558 and 1589, the editor had cautioned (in Hungarian) that certain passages of the text were written in a secret script (Verancsics 258, 290, 336)[2]. He neither attempted to reproduce these passages nor did he attempt to interpret them.

Almost a century and a half later in his comment on the unpublished papers of the Vrančić family (which are stored among the collection of manuscripts in the National University Library of Zagreb), Darko Novaković pointed out an interesting fact that there is a passage written in code in Antun’s letter which was sent from Paris in 1546. He pointed out the type of code and the reasons for secrecy: “The code is based on the characters of the bosančica script, which the brothers obviously assumed would pose too great a cryptographical challenge for the many curious agents of the secret service who might chance along the way between Paris and Šibenik” (161-162). Besides this, he also pointed out that on several occasions Antun had expressed certain concerns to his brother over a sack containing his most important (and thus encrypted) writings (162-163, 166, 168).

The co-author of this paper, while collecting material for her dissertation on the familial corespondance of Antun Vrančić from the Sczéchenyi library in Budapest, obtained a photographed copy of Mihovil’s handwritten manuscripts. Now one can gain further insight into this topic and the types of letters written by both brothers.

Although it is evident from the archived correspondence that Antun and Mihovil exchanged letters for over thirty years, from 1538 to 1539[3], there is a paucity of letters written in code. There are only five in all: Antun’s letter from 25 January 1546, approximately two lines of text (NSK lr-2v) and four letters written by Mihovil (the first from 6 November 1558, a nota bene at the end of the letter comprised of five and a half lines (Séchenyi 236r); the second from 27 December 1558, five extracts comprised of five then one, two, four and seven lines respectively (Séchenyi 241r-243v); a third from 11 January 1559, with a postscriptum of eight lines (Séchenyi 262r-263r) and a fourth - a note which belongs to a letter from 9 February 1559, comprised of nine lines (Séchenyi 262v).

Mihovil’s letter from January 1559 had managed to evade Szalay and Wenzel and so it was not published in the series of books Monumenta Hungariae Historica. After perusing photographed copies which had been sent form Budapest, it is clearly evident that it cannot be found among the corpus of letters marked under 1681 fol. Lat. Besides this, one cannot find any of the markings which were most likely used by Szalay and Wenzel while collecting this material for publication.[4] On the basis of these findings one can easily point out that one must be prepared for new discoveries when attempting to systematise and classify the epistolary corpus of Antun Vrančić and his correspondents. Other than this, it is perhaps possible that someone might find the sack that Antun cared so much about, as it seems that it contained other letters that were written in code. Although he referred to these letters as his “writings” on three occasions[5], there is a passage in a letter from Vienna, 17 October 1558, in which he refers to them as his letters (NSK 6, lr):

Quum huc adueneris, referas mihi torbam istam cum scriptis meis, quam ex Turcia ad Ioannem et ad te destinaueram: sunt in ea quędam scripta mea ziffris exarata, quę interpretari desiderarem, ut tandem meę epistolę in ordinem redigantur.

Once you have arrived here, bring me that sack with those writings of mine which I sent from Turkey to you and to Ivan; in it are some of my writings that are marked in code which I would like to interpret, so that my letters can finally be set in order.

Unless he was using the term epistola in its wider meaning, one may surely assume that Antun is referring to the confidential diplomatic mail that he had sent to Vienna. It will remain unknown whether the ziffrae he is referring to are identical to the ones found in the letter from Paris.

Mihovil’s Letters

Deciphering

In our endeavour to understand an elusive text, we thought that, at first, we should ask assistance from our colleagues who research old Croatian letters. Meanwhile, the response was that no sense could be made out of them, and that it was impossible to conclude in which language they had been written[6]. What remained was to reconcile ourselves to the fact that there would be several lacunae in the prepared critical edition. Another option was that we put to use our acumen in deciphering the elusive content.

The challenge itself was indeed tempting. As a start, we compared only Antun’s coded extract (Figure 1), seeing that it is the oldest and contains signs derived form the bosančica script. The results were non-existent because we could not uncover any discernable meaning. There was a second attempt with the aid of the glagoljica script but this was also unsuccesful. Upon closer inspection it had become evident that the numerous signs used by Antun did not correspond to those of the bosančica script (because of this, we will be using the term bosančica with certain reservations, in quotation marks). Greek characters can also be found and some of them are of the same value as their corresponding Latin characters. In addition, certain letters do not seem to belong to any alphabet that we could recognise. Most interesting of all is the fact that the signs, used by Antun and Mihovil respectively, correspond only in minor details.

Figure 1 - An extract from Antun Vrančić's letter from 25 January 1546 (NSK 1, 1r)

The Shortest Passage

Our next task was to analyse Mihovil’s more numerous and longer extracts. To begin, the shortest passage was selected (Figure 2), with the assumption that, as it is the shortest passage, it would be the easiest to surmise its meaning by relying on the context of the letter itself.

Figure 2 - An extract from Mihovil Vrančić's letter from 27 December 1558 (Széchenyi 241r)

In order to make the analysis easier, the coded text was transcribed into its corresponding signs that can be found on almost any computer. The cited passage, along with its Latin context, would look something like this:

As is evident, the Latin portions of this message are also rather cryptic. If we have understood this correctly, mention has been of a group of people that Mihovil holds in low regard. They had decided upon something, but had changed their decision, yet may even forfeit this decision upon the return of an obviously influential member of the group, who will most likely become a member of its governing body. In this instance, the reference is to a monastic community in Šibenik.

Because this extract begins with the preposition de, the ablative case must be present in the question. It is for certain that, beside this ablative, we will find a word in the genitive case (as in De coniuratione Catilinae or as in De humilitate... Christi ). It was a fortunate circumstance that the first two signs (g3) were exactly the same as the last two, only that they are facing in the opposite direction (3g). The duogram HH, if it is not a geminata, is most likely a border between two words, which would certainly make deciphering easier. Furthermore, the same sign is repeated two more times in the later parts of the text, and in addition, the sign p appears in the fourth and the seventh part.

We continued by combining the case endings for the genitive and ablative, and contrariwise, of all five declensions in the singular and the plural. We hoped that the shortest case endings were not the only ones used, as this would barely be of assistance to us. We gained little from this. We were still faced wih a great deal of aporia. It is worth noting that only later it became apparent that the “poorest” case endings were the most prelavent: the a-declension for the ablative singular and the o-declension for the genitive singular.

Frequency of Signs

After our initial disappointements, we came up with a new idea: to construct a table of the frequeny of signs from the bosančica script, the Latin signs from the remainder of Mihovil’s letter and of the Latin signs from a Croatian text from the literary remains of both brothers, which, truth be told, does not belong to Mihovil but to Antun. We hoped that, seeing that the other sources were lacking, they would serve their purpose to reveal in which language the messages were written. We are referring here to Antun’s prayer which was published in the Navk Karstyanski (44-47; facsimile in Horvat 196-200), which is in turn a translation of Bellarmini’s Dottrina christiana breve. It was only then that we realised that, by calculating the frequency of signs, we had made our first step in deciphering the language of this text. The Arabic scholars had already used such means in the Middle Ages (Weber 7) and today one can easily find the tools for such a task on the Internet. These are the results:

It is not difficult to see that there is a greater congruence between the passages in code and the Latin script than in the coded passages and the Croatian prayer. In the Latin script and in the passages in code the signs in the first and the second part are quite close, as are the signs in the third and the fourth part, whereas between the second and third part there is a noticeable gap. In the prayer there appears to be a greater breach between the first and second, third and fourth parts, and a greater closeness to the numbers in the second and third parts, which is the definite opposite of what we found in the Latin script and in the coded passages. This was our first sure sign that these camouflaged passages were written in Latin.

Thanks to this insight, we continued by combining the case endings for the genitive and ablative with even greater ambition, and yet the text still seemed to escape our understanding. The reason being that our exemplum text was found to be insufficient to make a more certain calculation of the frequency of signs. The closest we got to a correct combination was a variation of i-n-s-uao-?-e-?-?-?-r-n-i, in which the first three signs (ins-) and the last three signs (-ni) were defined correctly, as later became apparent. Besides this, the first of the three proposed signs in the fourth part (uao) had also proven correct, but we could get no further than this. Also unsuccessful was our attempt, based on the phrase gregem illum, to discover which family from Šibenik (Šupuk 126-162) this phrase in the text is referring to. One must note that a geminata is incorporated into their surname.

The Number 400

After another defeat, a detail from the second extract of the same letter (Figure 3) attracted our attention. Among a series of incomprehensible signs, we noticed the number 400, written in Arabic numerals.

Figure 3 - An extract from Mihovil Vrančić's letter from 27 December 1558 (Széchenyi 241v)

Transcribed, together with the Latin text that surrounds it, the extract appears as follows:

We surmised that before every number we would need to find the term used for a certain currency in the genitve case, especially due to the fact that we found y in seven signs before a number in the second part, and in the fourth a (h y b a 3 d t). It seemed to us that this might mean ducator(um) or perhaps the Croatian variant dukatov, but there seemed to be no other possibilities. Likewise, we did not succeed when we used floreni, another type of currency of which Mihovil makes mention in his letters, for e.g. in the letter from 16 November 1558 (Széchenyi 236r; Verancsics 1, 257). This time round, fortune did not smile upon us, but was cruel, as florin was actually the currency mentioned in the text, but it was in the Latin ablative. In order to get to the correct reading, we needed to include one more sign in our calculations, the eighth sign before the number 400.

Cryptography

After yet another defeat, we thought it was time to try something else. We researched the history of cryptography, which had been developing with greater intensity in the Occident at the very beginnings of Humanism (Weber 5-13)[7]. The reason why certain messages were encrypted was to protect mercantile and political interests. When we consider the times in which they lived and Antun’s social position, it becomes clear that there is nothing strange in the fact that there are enigmas in the correspondence of the two Vrančić brothers.

It is of greatest interest that at the beginning of the 16th century the German priest Ionnes Trimethius published the Libri polygraphiae VI[8], in which he perfected a system similar to Caesar’s. Towards the end of the book, he included several writing systems (Norman, Frankish...), among which one can find signs that correspond to those used by the two Vrančić brothers.

»Qu« saves the day!

Suddenly, while continuing an unsuccessful search with minds almost powerless, were we struck by the fact that in Latin there is a (semi)indispensable conjunction between two signs, and this is -qu-. If one is dealing with Latin, among the encrypted phrases, one is certain to find a repeated combination of two signs, in which only one sign will be found in the conjunction, while the other will be found on its own too, in the company of other signs. And certainly, it quickly became apparent that they are the signs marked as -8p-. Their appearance together in the text is tenfold[9]. p appears frequently without 8, and in combinations with other signs, it is uneccesary to cite examples. It is interesting that the sign marked as 8 (ie. q) appeares once (Szechenyi 236r) without p (ie. u). The reason for this being that the word itself is shortened behind the sign marked as 8 (in the phrase uobiq(ue), where one can notice that the final s from the dative uobis has been left out deliberately).

Geminatae

With a renewed enthusiasm, we came up with the idea that led us toward recognising the first word. We constructed a comparative table of the frequency of signs that were doubled in the Latin and in the coded text. We took into account the possibilitiy of superposition in the Latin script in which a previous word ends with the same sign with which the next word begins. We placed instances such as these in brackets. Here are the results:

We began with the premise that the most prevalent geminata in both writing sytems would match, meaning that the hh and the tt from the bosančica script would be represented by ss and tt in the Latin script, or prehaps on the contrary, by tt or ss. We took into account the signs that surrounded the geminata: two signs to the left and right of each other. We immediately noticed two identical geminatae in the extract ?yttattgy (Figure 4).

Figure 4 - An extract from Mihovil Vrančić's Letter from 27 December 1558 (Széchenyi 241v)

One must recall that a and g in the bosančica script are signs of the first and second rank and, the counterparts of which are e and i in the Latin script. Once they have all been placed in the series of ?yttattgy, one will find four different variants: ?y-ssessi-y / ?y-ttetti-y / ?y-ssisse-y / ?y-ttitte-y. The first variant appeared as the one that would provide us with the most definite result, as it seemed possible that it might mean po-ssesi-o. We were, in fact, correct. What followed was much easier.

By this stage, we had seven recognisable signs: 8 = q, p = u, t = s, a = e, g = i, ? = p, y = o. Besides this, other geminatae began to appear like the Latin (e.g. in pahhađa (Széchenyi 242r). Based on the frequency of signs, it seemed most likely that the equivalent of hh would be dd, cc and ll, and would provide us with uedette, ueccete, uellete). Thus, aid was given to us by characters that we had had recognised, a context that also became more palpable by the comparative frequency of signs. We managed to decipher Mihovil’s entire alphabet![10] The results are as follows:

>

The Origin of These Signs

It would seem perfectly natural to ask what kind of writing sytem Mihovil had, in fact, used. For a moment it seemed that there were multiple sources. The greatest number of signs, more than a third, are equal to the Latin signs as can be found in the remainder of letters (s, n , a, g, h, o, p). Although they are not written in his handwriting, H, y, x and q are certainly of Latin origin, which, when added up, is more than half. Three arabic numerals (3, 7, 8) are also of equal value in both of the alphabets that he used[11].

It is also entirely unneccesary to look for the origin of T outside of the Latin script. Although it is regularly written as a letter in lower case, therefore in the midrange, (as is not the case with H, which is placed in the upper range) it would be reasonable to consider the influence of the greek letter ?.

The most unusual symbol, +, which we noticed in Capelli’s Dizionario (418) as the medieval symbol for the Roman numeral 50, is in fact the same as the upper symbol (T), only that it has been inverted. What supports this conclusion is its proximity in the alphabet: T represents f, and + represents g.

The sign for the letter e looks a little different in the manuscript as the greek letter ?[12], which means that this is most likely its origin, although it shows some similarities to the same letter in the bosančica script.

The sign for ž in today’s cyrillic script (?) is equal, as it appears in the manuscript, to the number 600 (occasionaly it is used for 700) in the bosančica script (cf. Zelić-Bučan 20 and Table II), to the sign for the number 30 in manuscripts written in the glagoljica script (Zelić-Bučan Table III) but also to the sign for m in the Old Frankish alphabet (Trithemius 586) and to the g in the series of abbreviations devised by Cicero (Trithemius 601).

The sign for the letter t in the manuscript is more rounded at its lower end which makes it similar to the jat (j) in the bosančica script (Zelić-Bučan 20 and Table I).

We are still unsure of the origin of b, originally conceived as a circle in its fundamental field from which a vertical line ascends, which is crossed by a horizontal line of shorter length. It reminds one of the sign for 100 in the glagoljica script (Zelić-Bučan Table III).

Nor could we ascertain the origin of đ. One should not entirely disregard that this was of Mihovil’s own invention. He wrote the letter d in Latin script in much the same way, so it is quite likely that he simply added a short horizontal line to it.

The last three signs are all quite similar (t, b and đ) and when we began, we asked ourselves if there really was any difference between them, or if they were merely the work of an unsure hand. The author himself was surely aware of the difference as he placed them next to each other (they thus represent, r, s, t), and he most certainly wanted to make the possibility of deciphering more difficult by doing so.

The foundation for this code is therefore Latin. The use of Greek, bosančica and glagoljica script is almost non-existent. The trace of a possible influence can be seen in the fact that the quality of an unignorable number of Latin letters converges with the value that can be found in the book by Trimethius! This is as follows (587, 591, 592): H (=a), s (=c), 3 (=n), y (=o), q (=x). In addition, it is indicative that this concerns the first three signs of the alphabet, as well as two from the middle and a sign from the end of the alphabet as well.

It seems that the cryptographer from Šibenik has arranged these signs, to a certain extent, in a systematic fashion. As this is the case, the signs e, g, and h are moved three spaces to the right (and thus represent h, i, and l), p and q have been moved five spaces (they represent u and x), and n (=e) and y (=o) have been moved eight spaces to the left. The movement of other signs are more particularised: a (=e) has been moved four places to the right, s (=c) has been moved fourteen spaces to the left, and o (=m) has been moved two spaces.

The fact that there is no completely unilateral system in this code, in its choice of signs or in the arrangement of its symbols, is the greatest value of this system and also its greatest safeguard.

Deciphered passages

When one changes the signs of the coded text in Mihovil’s letters for the signs they represent in the Latin script, one will receive a clear text in Latin despite the occasional mistake made by its author, which was most likely the result of either a lack of concentration or a lack of practice. We will now present to you all the coded extracts with their surrounding context, and will place a transcription and a translation beneath them.

Mihovil crammed into this same piece of paper what he had only hinted at earlier on in the letter and had promised to soon make clearer (in its lower left corner). He had already written that the “business in Trogir” did not progress, but that something “similar” had been offered in Šibenik by Petar de Bassano, the father of a young man who had recently visited Mihovil in Hungary[13]. One may discern from the secret text that negotium refers to an investment, and the phrase “a clearer notification” refers to a definite buisness offer, the price and the yearly earnings of the abbey.

Petrus de Bassano was also the name of a contemporary notary in Zadar (1531-1570)[14], so it is difficult not to identify him with the bidder from Šibenik. Mihovil mentions him again in the next letter, written a month and a half later (27 December 1558), which we will comment upon a little later. In the letter he writes about the same offer, only in greater detail and with an appeal for a reply. It seems that he found this business offer quite attractive. Mihovil also writes about Petar’s son Ivan with much elation at the end of the letter, as Ivan congratulated Antun on his father because he had greeted him hospitably (Verancsics 296), and again on 9 February in the same fashion (335-336).

This is probably the most puzzling of all the passages in code. After much effort and consultation with others, we concluded to interpret this as though Mihovil had offered a joint task to an influential personage in Split, but that the said personage had refused this as he had nephews. Due to familial obligations he has reserved these tasks for them, otherwise he would most willingly carry this out with the Vrančić brothers[15].

It is possible that this task is the same thing as to bequeath the episcopacy, which depends upon whether or not one can assume that either of the two adverbs ‘already’ and ‘otherwise’ are present (thus “to one of whom he has already decided – or: otherwise decided – to bequeath the episcopacy”). It seems, though, that there are two different matters at hand. It is as if the episcopacy has been mentioned as an afterthought, as an example of a familial obligation which is a hindrance to some other business, and this other business remains unnamed to ensure a greater security. (Mihovil mentions the business in Trogir[16] allusively, as well as the business in Dubrovnik).[17]

What aids such an understanding of this passage is the context in which the younger brother provides an account of his tasks, in a letter composed two months later (Verancsics 243-244):

Quod reliquum est, ea quę perficienda fuerant mihi iniuncta per dominationem vestram reuerendissimam, partim confecta sunt, partim vero iis proximis diebus diuino auxilio conficientur. Vt puto, domus conductio, hortus, insula, negotium Spalatense et similia.

Of what is remaining, that which your most honoured lordship has assigned me to complete, is completed in part, and part will be completed, with the help of God, in the days ensuing. I am thinking of, for example, the lease of the house, the garden, the island, the business in Split and similar things.

Taken together, these are all investments so it is certain that the “business in Split” can be included among them, as was the case with the abbey in Šibenik. Antun returned from a four-year embassy position in Constantinople the year before and for this was awarded with honours by the bishop of Eger (Novaković 132). Besides this, his earnings increased and certainly led to a “wave of investments in Dalmatia” in 1558 and in 1559, immediately after Mihovil’s return from Vienna. He obviously received accounts concerning this while he was still in the North.

It remains a matter of conjecture with whom he had spoken in Split. The care for the nephews and not for one’s own children and the bequeathing of the episcopacy all indicate that it was with a member of the clergy.

We are unsure of the meaning of “to bequeath the episcopacy.” Does this refer to the episcopal see or to the management of the bishopric, or to a lease?

Haec provincia et alia most certainly refers to the Venetian and Habsburgian part of Dalmatia, although other interpretations cannot be entirely excluded, prehaps including Croatia with Dalmatia.

Antun and Mihovil had received a reply that was polite and honest, but ultimately negative. It is not entirely obvious whether or not the gift from Constantinople was of any assistance, which was definitely more than a souvenir, as Mihovil had also written a record of this in code. [18]

For a moment we thought that Antun and his nephews were mentioned allusively in the reply. If this were the case, the meaning of this passage might be that an important personage in Split agrees with Antun’s choice to select one of his own nephews as succesor based upon his abilities, only that this would not be the correct move due to mutual blood relations. At the time Antun actually had two nephews, Faust born in 1551 and Kazimir born in 1557. But after careful analysis we decided to let go of this variant, as it is illogical that, whoever it was in Split, would have bearing on Antun’s reply on who he would most willingly and with the most security take as a business partner. Besides this, Faust was only seven years old at the time and it is difficult to conceive that Antun would already be making serious consultations about his future career, although his father had congratuled him on his future prospects at the end of the letter.[19]n>

The island of Krapanj is therefore the answer to the shortest of Mihovil’s puzzles, which we placed under closer inspection at the beginning of our article, but this is not enough. Taking all into consideration, as we noted earlier, a community plays a part in this. It is perhaps a monastic community, which most likely has rights to, or property on the Island of Krapanj, in which Mihovil has some interest. One remains unsure whether he referred to this evasively because they had changed their decision or because they might have had a bad reputation.

His report on the failed negotiations to purchase a house (which was, without a doubt, in Šibenik) is in much the same fashion and the reasons are much the same as regards the island of Krapanj that we have seen previously:

Although Mihovil, further on in the letter, amongst other news from Šibenik, writes that Ioannes Longus alias Misich has died, this is not directly connected to the Misić widow so we cannot be sure if it has something to do with same family.

In Šibenik there are two houses known to bear the emblem of the Vrančić family (Kurelac 2008, 67-68, 78-79; 2011, 100-105), so the widow’s house must have been in the vicinity of one of them.


Before the rather strange tamen at the beginning of the second sentence it seems that one should assume the following: “I am not one of those with a plenitude of assets. Yet (I will mention:) amongst other things...”

Of the two properties on the island of Prvić one was in the possession of the bishop of Nin (it is possible that this concerns the property of the Divnić family, to which the bishop of Nin, Juraj, and his succesor, Jakov (died 1558) belonged. The other property was in the possession of one Petar Parisoti. It seems that the Vrančić brothers bought at least one of them, as one can read in Mihovil’s letter from January 1564. In this letter Mihovil, no longer using code, gives a report to Antun concerning two investments - the purchase of two barns for cattle and the purchase of a garden in Crnica near Šibenik[20] as well as the purchase of the Lukočić property on the island of Prvić which was right next to the Vrančić property[21]. The Lukočić property was of particular value, so Mihovil claims, first of all due to the fact that it was next to the existing property and because this house was the birthplace of their father Frane Vrančić.

It seems that a third property on the island of Prvić was also considered. In a letter written on 1 October 1558, Mihovil includes a short passage written in Hungarian, in which mention is made of a house with a vineyard which was in the possession of Petar Skruanović, and for which he does not have the funds to complete the purchase.[22] For this reason we are still unsure as to the property next to the Lukočić property, which was purchased at a later date.

It is interesting that the surname Lukočić belonged to two brothers who had quite a bad reputation, and it is to them that Mihovil dedicates his last two messages. But seeing that he does not connect them to the property of the same name, we should not either, although we are tempted to think that he took advantage of the sale of this property at a much lower price.

Another interesting fact is that Mihovil used Hungarian when writing about his first intended purchase. It is certain that he did not feel safe using it so he incorporated code in his letters some time later. Yet after the purchase had been completed, he no longer keeps it a secret, but mentions it openly. That the sending of mail entailed a great risk at the time was something that Antun’s friend complained about to Mihovil from Venice 1558 (Verancsics 7, 224): Quod plus est, neque scribere ad dominationem vestram reuerendissimam per eorum nuncios audet, nam littere aperiuntur eius et accepto succo ex ilis pro suis mittunt. (“What is more, he dares not write to your most honoured lordship through his own messengers; for his letters are opened and the core has been taken out of them, they send them as their own.”)<br />


We once again encounter indiscernable particulars in a message of rather clear foundations.


If one of them has rushed off without the letter (which was most certainly adressed to Antun), does this mean that somebody else has taken it? The last sentence is especially puzzling. First of all we are uncertain if the second eius (sc. animum) in this letter refers to Antun, or perhaps to Šimun which would be more difficut to comprehend. As it is, it seems that Mihovil held his older brother in such a high regard that he had refused to give him any advice so that it would not seem as if he were imposing something on him, which was something which Antun obviously could not tolerate.

Taking all into account, it seems that Mihovil found the entire situation rather discomforting, as he repeated the message again two weeks later, in a special addenda in the following letter.

In this reprisal we find out why Šimun had left Šibenik so suddenly. Next to the sins of wantoness, bigamy and apostasy, acts of larceny are added. The mention of Rankolin adds a new dimension to this. Seeing that Mihovil mentions him without any other attributes, it is certainly the name of a public personage on the other side of the law. Perhaps mention is being made of social unrest, as had occured twenty years later (Stošić): “in the year 1582 the people rebelled, who were led by Ivan Ručić and his brothers, Nikola Rankolin and Luka Rajčević. It seems that this was a political uprising.”

We can only surmise what the connection was between the Rankolin mentioned in this letter (1564) and the Nikola Rankolin mentioned in the cited article, and yet some of those with the surname Rankolin, according to Mihovil’s views were not among the respectable inhabitants of Šibenik.[23]

Antun’s Letter

We were taken by no small surprise when we attempted to apply the quality of Mihovil’s characters to Antun’s coded passage. It is not the same alphabet! Should one attempt to locate the first sign used by Antun, for example, (which reminds one of the greek letter ?), it is impossible to find it. The same applies to the third sign (the greek letter ?), to the fifth, the ninth (the greek letter ??), the eleventh, the twelfth, the sixteenth, and in the second line to the ninth and tenthFigure 5 (=Figure 1): Antun's passage.

There is a convergence only between a smaller group of signs, seven of them, some of them might come close, but a good portion of them are simply different. Figure 6 - Passage from Mihovil's letter from 28 December 1558>

Only a smaller number of them are derived from the Latin script, and there are several new signs whose origin still awaits investigation. It is no wonder that Antun’s letter cannot be read by using Mihovil’s system as a guideline, despite several attempts to find different convergences between the two series of signs. Therefore, either the shape and the value of these symbols had changed in the meantime, from Antun’s letter in 1546. To Mihovil’s in 1558-59, or the two brothers had their own respective systems.

Calculating the frequency of signs is also of little assistance, obviously due to the brevity of the passage, only 45 signs in all, half a sentence.

The context of the message is rather clear:<br />


As much as we can gather from all this, Antun presented a proposition to Isabela the Queen of Transylvania which proved fruitful, yet it took some time to come to a final decision (if we have interpreted the meaning of the verb consulere correctly). Meanwhile hatred between the two princes proved a hindrance to this. If one of them was the King of France Francis I, the other would have to be Karl V (1500-58, emeperor of the Holy Roman Empire since 1589), who was his rival for dominance in Europe during the course of his life. We cannot be entirely sure whether Francis’s preoccupation with the conflict with Karl was what hindered the expected result or whether the emperor was directly aganist it.

As a diplomat, Antun, with some caution, is awaiting the semingly assured success in his embassy service to Francis I (the King of France since 1515), only that we do not know what this success entails. In a long and warm letter to his friend Carlo Capelli he recounts the same task in hindsight, just after he had completed it, in almost the same words, yet with such generality and inscrutability that it is impossible to discern the exact content (Verancsics 6, 186).

Scito igitur, me ex Galliis incolumem Venetias rediisse, et hoc biduo, publicis rebus exigentibus, in Transsylvaniam promoturum. Omnia in eo regno mihi non omnino inprospere successerunt, nostraeque factioni, fortasse etiam adversae, brevi idque facile consultum iri sperarem, si odia sua duo principes ad catastrophen venire jam aliquando paterentur...

You must know, thus, that I have returned form Gaul to Venice safe and sound, and that I will in two days, as this is demanded of me by public service, to continue into Transylvania. All has succeeded for me and my side almost according to all expectations, prehaps for the other side as well, so I am hoping that all will be swiftly decided with ease, should the two princes finally allow their hatred to come to an end...[24]

The theme of the Transylvanian-French discussions might have concerned Antun seeking assistance against the Turks; towards the end of the 1530s, his uncle, the Transylvanian bishop Ivan Statilić (1528-1542), found himself in the same place with the same task (Novaković – Vratović, 46). Yet it is equally possible that what was discussed was assistance for the Transylvanian initiative towards independence from Austria and its ruler, Karl’s brother Ferdinand of Habsburg. One must also keep Protestantism in mind, which had become a problem, not only in Karl’s Germany but also in Francis’s France and Antun’s Transylvania.

In hope of an answer, what remains for us is to dedicate ourselves with more time and imagination to unraveling this mystery.

Conclusion

After four and a half centuries, the veil has been removed from the secret messages of Mihovil Vrančić. Attention was directed toward them for the first time after the publication of the Hungarian edition of the collected works of Antun Vrančić in the 19th century. In Croatia, attention has been directed only recently to the fact that, in the National University Library in Zagreb, one of Antun’s letters contains a hidden sentence written in code. Also, it has been suggested that it might be written in the characters of the bosančica script. While preparing the scholarly edition of the familial correspondence of Antun Vrančić, which was also part of her doctoral dissertation, the co-author of this paper discovered a letter by Mihovil, which likewise contains a passage in code and which had remained unknown until now. When she came upon the discovery she turned to her supervisor for assistance – the second author of this paper.

Our attempt to decipher this passage with the aid of the bosančica script gave no results. We concluded that the bosančica script does not play a part here, and that it is certain that Mihovil used a different system of signs than Antun. Attention was directed to the shortest of Mihovil’s passages, which provided us with the hope that by calling on the context of these letters and Latin case endings for assistance we would be able to recognize certain signs. However, this proved unsuccessful as did our attempt to recognize the term for a certain currency in front of the Arabic numeral 400 in the second passage.

Only a systematic approach provided us with the solution: by calculating the frequency of signs in the coded passages as well as those in the parts of the text written in the Latin and Croatian alphabet. A comparison confirmed that the secret portions of the text were most probably written in Latin. We grasped something that would prove conclusive when we realized the fact that, in Latin, there is a semi-compulsory connection between the signs q and u and also by having found the corresponding combination among the signs in code. We had yet to calculate and combine the frequency of double letters in the “bosančica” and the Latin script to secure our foundations for recognizing the first word. It turned out to be possessio hidden in the code ?yttattgy. The remaining text almost deciphered itself.

The system of encoding and its probable paradigm have also been described. The majority of signs originate from the alphabet, including some Arabic numerals, some Greek signs and some from the bosančica script; only that their phonetic quality has been changed. It is interesting that all except one sign can be found in the Polygraphia, a book by the German cryptographer Trithemius dating to the beginning of the 16th century. It is even more interesting that the phonetic quality of a third of Mihovil’s signs concur with the qualities found in Trithemius’ alphabet. It would not be surprising that the Vrančić brothers used this or a similar guidebook, seeing that cryptography was a compulsory means of communicating diplomatic and business matters in the 16th century, and it was certain that it comprised a part of Antun’s diplomatic education.

The content of these secret messages was translated and commented primarily with the aid of certain facts from the further correspondence of the brothers from Šibenik. Mihovil would report to Antun about economic opportunities for investing in and purchasing real estate in Šibenik and also on the islands of Prvić and Krapanj. It is unclear what the business matters in Split were, in which Mihovil apparently negotiated with a person in the upper echelons of the church. The reason why he made recourse to code was obviously to keep this business matter a secret, and certainly to get there first before any other interested parties. Mihovil’s letters to Antun were sent from Šibenik at the end of 1558 and the beginning of 1559. The investments began after Antun was promoted to the position of Bishop of Eger, by Ferdinand I of Habsburg for his long and difficult embassy service as ambassador with the Turks in Constantinople.

In two of the messages, personal safety is the predominant theme. Mihovil warns his brother that the two Lukočić brothers are now heading toward him in Hungary, and they have the reputation of being utterly dishonorable and dangerous men.

Antun’s letter concerning his expectations of diplomatic success at the court of the French King Francis I remains a mystery, where he, at the beginning of his career, was an ambassador of Isabela the Queen of Transylvania. The Vrančić brothers were so cautious that each brother used his own system of code. Unfortunately, having cracked Mihovil’s code in no way assists in uncovering Antun’s message, which is itself too brief for any other means to be of assistance.

Other than this challenge, there is much space open for further illuminating the theory and practice of cryptography in the early modern period and how the code of the Vrančić brothers relates to it. As far as the content is concerned, we need to place the realia from Mihovil’s passage in a more precise economic and social context, so that we can gain the larger picture of an episode in the life of this illustrious family from Šibenik.

<strong>Appendix </strong>8 - Note added to Mihovil's letter from 9 February 1559 (Széchenyi 262v)<strong>Appendix </strong>7 - Mihovil's letter from 11 January 1559 (Széchenyi? 57v)<strong>Appendix </strong>6 - Mihovil's letter from 11 January 1559 (Széchenyi? 57r)<strong>Appendix </strong>5 - Mihovil's letter from 27 December 1558 (Széchenyi 242r)<strong>Appendix </strong>4 - Mihovil's letter from 27 December 1558 (Széchenyi 241v)<strong>Appendix </strong>3 - Mihovil's letter from 27 December 1558 (Széchenyi 241v)<strong>Appendix </strong>2 - Mihovil's letter from 6 November 1558 (Széchenyi 2363)<strong>Appendix </strong>1- Antun's Letter from 21 January 1546 (NSK 1, 1r)

Works Cited

Capelli, Adriano. Dizionario di abreviature latine ed italiane. Milano: Ulrico Hoepli, 1912.

Bellarmino, Roberto. Dottrina christiana breve. Roma: 1661.

Cocai, Merlin (Folengo, Teofilo). Le maccheronee. A cura di Alessandro Luzio. Bari: G. Laterza, 1927.-1928.

Horvat, Vladimir. “Molitva Antuna Vrančića.” Zbornik o Antunu Vrančiću. Šibenik: Gradska knjižnica “Juraj Šižgorić”, 2005. 191-200.

Katona, Julianna. “Puteanus’ Cryptographia Tassiana.” Humanistica Lovaniensa, XLIX (2000). 293-300.

Kurelac, Iva. “Potraga za istinom duga četiri stoljeća: Šibenski humanistički pisac Dinko Zavorović – prvi povjesničar Dalmacije.” Hrvatska revija 11.1 (2011): 100-105.

---. Dinko Zavorović: Šibenski humanist i povjesničar. Šibenik: Gradska knjižnica “Juraj Šižgorić”, 2008.

Navk Karstyanski, sloxen Po Prisvitlomu Gospodinu Rubertu Bellarminu Kardinalu S. R. C. I prinapravglen Po zapovidi S. Skupsctine od razplogenya suete Vire. Romae: Typis S. C. de Prop. Fide, 1661.

Novaković, Darko, and Vladimir Vratović. S visina sve. Antun Vrančić. Zagreb: Globus, 1979.

Novaković, Darko. “Neobjavljena ostavština Antuna, Mihovila i Fausta Vrančića: R-5717 u rukopisnoj zbirci Nacionalne i sveučilišne knjižnice u Zagrebu”. Zbornik o Antunu Vrančiću. Šibenik: Gradska knjižnica “Juraj Šižgorić”, 2005. 155-181.

Karstyanski, Navk. Xloxen Po Prisvitlomu Gospodinu Rubertu Bellarminu Kardinalu S. R. C. I prinapravglen Po zapovidi S. Skupsctine od razplogenya suete Vire. Romae: Typis S. C. de Prop. Fide, 1661.

Sorić, Diana, and Milenko Lončar. “Pismom protiv nepoželjnih čitatelja – kodirane poruke Mihovila i Antuna Vrančića.” Colloquia Maruliana, 22 (2013): 49-75.

Stošić, Krsto. “Bune u Šibenskoj prošlosti.” Novo Doba 94 (21. IV. 1935).

Šupuk, Ante. “Šibenski Liber baptizatorum (1581-1590), njegova antroponimna građa i osobitost te antroponimije”. Radovi Instituta za hrvatsku povijest 7 (1975): 126-162.

Trithemius, Ioannes. Libri Polygraphiae VI…, Argentorati: sumptibus Lazzari Zetzneri Bibliopolae, anno M. DC. XIII; http://books.google.hr/books?id=QLRAAAAAcAAJ&printsec=frontcover&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false; 12 Dec. 2012.

vol. 10, ed. Szalay László, Wenzel Gusztáv, Pest, 1871. (= Monumenta Hungariae historica, Scriptores XXV).

---. 11, ed. Wenzel Gusztáv, Budapest, 1873 (= Monumenta Hungariae historica, Scriptores XXVI).

---. 12, ed. Szalay László, Wenzel Gusztáv, Budapest, 1875. (= Monumenta Hungariae historica, Scriptores XXXII).

---. 6, ed. Szalay László, Pest, 1860. (= Monumenta Hungariae historica, Scriptores IX).

---. 7, ed. Szalay László, Pest, 1865. (= Monumenta Hungariae historica,Scriptores X).

---. 8, ed. Szalay László, Wenzel Gusztáv, Pest, 1868 . (= Monumenta Hungariae historica, Scriptores XIX).

---. 9, ed. Szalay László, Wenzel Gusztáv, Pest, 1870. (= Monumenta Hungariae historica, Scriptores XX).

Weber, Ralph Edward. United States Diplomatic Codes and Ciphers, 1775-1938. First Transaction paperback printing 2011 (Originally published: Chicago, 1979).

Zelić-Bučan, Benedikta. Bosančica u srednjoj Dalmaciji. Split: Historijski arhiv – Split, 1961.

Ruseckaja, Irina Alekseevna. Kriptografija v Zapadnoj Evrope v rannee novoe vremja. Moskva: 2003. Naučnaja biblioteka dissertacij i avtoreferatov disserCat. http://www.dissercat.com/content/kriptografiya-v-zapadnoi-evrope-rannee-novoe-vremya#ixzz2EpJIKdg 12 Dec. 2002.

NSK, R-5717, folio 2, number 2.

Séchenyi Library, fol. Lat. 1681, vol. 3.