Category Archives: Musings

Romani Recursion – Curiosities and Trivia

“[…] However, Devanagari came into being after the Romani left India.” (from Wikipedia).

There seems not a lot to be added to the recurring popular topic of a possible Brahman / Hindu origin transported via Rom & Dom migration in VMs research, however the connection resembles a promising starting point for further exploration, as so many other things related to VMs research.

A more recent attempt to construct a Romani alphabet based on Devanagari was not successful. The essay “Romani Orthographies”1 outlines some of the difficulties contemporary standardization attempts were & still are faced with. The necessity of concepts like archegraphemes or morpho-graphs does not make the task of constructing a writing system appear particularly simple.

There is a ca. 1515 account of written Romani, as outlined in a 2011 german article on hypotheses2, which in turn refers to Knauer 20103. The abstract of the latter:

“Thirteen unpublished lines of a Latin–Romani vocabulary in a manuscript in Munich represent the earliest document recording efforts to put words of Romani, ‘primarily an oral language’ (Matras 2002: 238), into writing. The Benedictine compiler and scribe of the list was familiar with important contemporary German scholars, a fact that may enhance the authenticity of his numerous excerpts and explain the almost scholarly approach to such exotic languages as Romani. It can be assumed that the ‘interviewer’ got his information in Vienna around 1515, preceding the often adduced vocabularies of Borde (1542), Ewsum (before 1570), Vulcanius (1597), Çelebi (1668), and Marsden (1785). He organized the results of his questioning neatly in groups, heavenly bodies, humans and animals, food, and cardinal numerals.”

Why Johannes von Grafing’s account in Opuscula Cod.graec. 582 a4 lists only seven of the zodiac signs remains curious.

There may be a few interesting points to note, firstly the singularity of the event. As far as currently observable the work seems not to have been influential in any way, thus underlining the possibility of similar finds.
The aforementioned almost scholarly approach seems to be, at least to some amount attributable to education, hinting at “schools of thought”.
Johannes von Grafing’s use of diacritic ü seems noteworthy as well. Interestingly, the proposed use of diacritics has been rejected by participants in a 1990’s university survey of spelling preference by speakers.5

If we allow for a brief moment of speculation, let us imagine a learned6 scribe from around XV. with a similar, yet slightly more ambitious project: the standardization and compilation of ancient Romani tradition from oral and (hypothetically existing) older written sources in Aramaic, Greek, Arabic, Glagolitic (or more exotic) scripts, writing all sorts of Para-Romani languages and different Romani dialects from different ages. It seems kind of a burden to take on for our scribe, and the book may very well turn out a “babel manuscript”.

A group of Romani people also called Erromintxela has a particularly interesting story: formerly referred to as “[…] ijitoak ‘Egyptians’, ungrianok ‘Hungarians’, or buhameak ‘Bohemians'”, arrived in Basque lands in XV., integrated with Basque society, and developed a Para-Romani language using Basque grammar.
“The initial E- is the Basque prosthetic vowel, added because no Basque word may begin with an R-, and the final -a is the absolutive case suffix, used when citing a name. If this etymology is correct, it is a rare case of a native Romani name for themselves (an endonym) being borrowed by another language.”
A form that jumps to eye is the verb “ajin / najin” for “to have / not to have”.7

Trivia: In 2012 german gangster rapper “Haftbefehl” (warrant) charted a hit with “Chabos wissen wer der Babo ist”. The track was praised and criticised for its language in mainstream media, at the same time.
When finding “puer schabo” in the “Collectanæa” it clicked for the author that the rapper might have employed the Romani term “chavô” for kid/lad/friend, and indeed a quick search turned up that Haftbefehl also used “Manische Sprache”,8 a Romani sociolect of the Frankfurt a. Main area.
“Babo” subsequently was voted for as “german youth word of the year” in 2013. The etymological roots of the word point back to Zaza-Language, the rest of the lyrics is composed of a polyglottal mix of German, English, French, Turkish, Kurdish, Arabian and Serbian phrases.


The Voynich Challenge

A Strange Proposal

dungeon map

Photo by Dru!

I wrote the following text a while ago with the idea in mind to put it up as a challenge on a place like Quora but then it got too lengthy and also too implicative. Reading it now it sounds more like the intro to a text based adventure game.



Let’s suppose you receive a strange proposal. Your task is to create a large body of text, split into several parts varying in length from single sentences to several pages, around 200.000 characters in all. The actual content is not important but it should appear meaningful. The text must appear to be written in an ancient, yet unknown language. This is crucial, it needs to be absolutely indiscernible from a real, existing ancient language, even to experts on the field and should hold up under scrutiny for at least a few decades. The reward is tempting so you don’t ask questions and agree to the assignment. Your timeframe is about 12 months.
You receive a batch of papers containing texts as a source of inspiration and a few hints like a list of keywords to use in different contexts, that it would do no harm if fragments of some words and sentences would appear here & there and the suggestion to encrypt parts of it to your liking. You could sometimes make it look lexical or otherwise copy the structure of some of the texts.
Which prove as a random collection of all sorts of different things, single words, names, lists of all sorts, quotes, poems, hymns, meditations, recipes, tractates, short essays, you name it. Mostly they are in latin, some in english. Some are in languages you only know of like ancient greek and some in what you can only guess to be aztec or ancient hebrew. A few remain undiscernible but this does not matter because they are nicely transcribed into latin characters, sometime with markings for extra characters of the originating language which do not have a latin equivalent. It comes in handy that you are asked to deliver your text in latin characters.
Of course there are some limitations. Your use of tools is restricted to pen & paper, dice, slide-rule & abacus (or digital simulations of such). Now here’s the real challenge: You can only use scientific knowledge that has been around until 1910 (it’s best you imagine living at that time). For example, Markov published in 1906 but it is unclear if you can use Markov chains because you do not have a computer. At the very most you have access to a Hollerith machine (or rather the virtual pendant).
How would you go about it?

I would like a really good Voynich game, I guess I’m not alone. So if you’re a talented MUD developer with lots of time on your hands, let us know!

The Voynich Bombe

How would Alan Turing approach The Voynich?

Alan Turing Sculpture

Photo by joncallas

Would he come up with a logical Bombe, like he did for the Enigma? A metaphorical brute-force attack, this time machined into software? Would he try to formalise all problems and develop a Voynich Test, maybe to be held annually, where every year we would see theory after theory fail the test..

I’m afraid there won’t be an actual machine readily cracking The Voynich Code for us. Rather we are looking at a Voynich Complex with a Voynich Blackbox in the center, all of which seem rather unquantifiable.. unless we had some good AI (wishful thinking).

It may be a little acknowledged fact that while Turing devised a generalised approach and greatly improved the concept, the Bombes’ original inventor is yet another genius named Marian Rejewski of the Polish Cryptography Office. Examining the Enigma’s patterns he found a cryptological weakness that enabled him to reconstruct the apparatus. His “Bomba Kryptologiczna” successfully cracked the Enigma model of 1933.

Marian Rejewski Sculpture

Photo by Pete Reed

The Bombe represents a fascinating concept. It is a reverse engineered reconstruction of the Enigmas inner workings, an approach that could prove useful in attacking The Voynich. While this has already been facilitated in the past (see Hyde and Rugg) with promising results it seems that a more generalised approach keeping the larger picture of The Voynich in mind would be required.

The British & US Bombes were sophisticated electro-mechanical devices that developed together with the problem they were designed to solve, the Enigma. They only made sense when deployed in large number. Rejewskis undertaking sounds more rustic:

A machine called the “bombe” is used to expedite the solution. The first machine was built by the Poles and was a hand operated multiple enigma machine. When a possible solution was reached a part would fall off the machine onto the floor with a loud noise. Hence the name “bombe”.1

A different version goes that Rejewski named his baby after a certain ice cream creation sold in vicinity of his office.

Read more on the history of the Enigma and the Bombe.