Calgary engineer believes he's cracked the mysterious Voynich Manuscript

[Ahmet Ardıç]

[quote="Ahmet Ardıç" pid='60488' dateline='1720704264']
[In dialects where there are no words starting with the K sound, this sound can usually be replaced by the G sound and sometimes the X sound.

Fair enough, but there is no "X" in modern Turkish and no word starting with "ks" except some prefixes loaned from Greek like xeno, xylo, so I'm not sure what you mean: which dialects are there that have no word-starting K sound and have word-starting X sound?

To convince people of the "perfect" phonetic match you should compare frequencies of bigrams and trigrams to prove that there is no big mismatch and also pinpoint the closest (known) dialect. Have you done this study yet?

Another thing that would be persuasive is to show how any grammatical rule is applied consistently in the VMs.]



Dear Nablator,

Looking for overlaps in bigram and trigram structures does not necessarily mean that there is a 100% exact match between the two languages. However, it can indicate a degree of phonetic and structural similarity.

Indeed, to establish a robust linguistic link, it is crucial to translate full sentences with more words and eventually full pages. Analyzing full sentences and paragraphs helps to understand the syntax, grammar, and context, providing a deeper insight into the possible linguistic connections.

If a linguist can demonstrate overlaps between two languages using full-sentence and full-page analyses, where the words correspond closely to known dictionaries, and a significant percentage of these words have retained their phonetic structures over centuries, then the necessity to focus on bigram and trigram structures diminishes. This comprehensive approach offers a stronger and more thorough method of establishing linguistic connections.

Showing overlaps in bigram and trigram structures can suggest some level of similarity, but it has its weaknesses. These overlaps might be coincidental or due to common phonetic patterns shared by many languages. Bigrams and trigrams alone do not account for grammatical rules, syntax, and meaning, which are crucial for accurate translation and understanding.

As been historical linguistics, the linguists who deciphered ancient scripts such as Egyptian hieroglyphs, Sumerian inscriptions, Hittite, and Hatti texts primarily relied on broader linguistic methods. They used comparative analysis, contextual interpretation, and extensive knowledge of related languages and scripts. There is no significant historical record indicating that these linguists used bigram and trigram analysis as a primary method. Their breakthroughs were often due to broader contextual understanding, such as the Rosetta Stone for Egyptian hieroglyphs, which provided parallel texts in multiple languages.

The researcher is free to choose methods of different difficulty levels. Instead of examining double and triple words, we proceeded with a method of examining overlaps in full sentences and full pages. This represents the highest level of progress that can be achieved in this type of work, and similar ones have been recorded in the past in the field of linguistics.

Identifying bigrams (two consecutive words) and trigrams (three consecutive words) in the Voynich Manuscript (VM) and comparing them to modern Turkish can help confirm structural similarities.

However, since it has already been shown that sentences of 6 to 14 words (for example at the reading page 33v study), instead of two or three consecutive words, have a complete correspondence with Turkish as a whole sentence structure, including the positions of the subject and predicate, there is no need to show the harmony by dividing the words into groups of two or three.

For example; the study already shows that every word of every sentence of an entire page (33v) consisting of 10 sentences of 6 to 14 words is a complete harmony between Turkish and Voynich along with the overlaps in sentence structures. In this case, it becomes meaningless to look for such overlaps for words divided into groups of bigrams and trigrams which are less than a sentence.

Linguistics can look for such bigram and trigram overlaps, but if the complete translation of page 33v of the VM texts into Turkish has shown that all sentences consisting of 6 to 14 words are in full overlap on the page, there is no need to show additional small overlap parts such as less than a sentence long as bigrams & trigrams.

This is like asking whether a vehicle with a carrying capacity of 14 tons can carry 2 and/or 3 tons of load. For this reason, searching for bigrams and trigrams at the current data overlap level would mean attempting to interpret the whole with a meaningless retrospective effort. If it has been shown that a complete sentence structure overlaps with all its words, it is an equally absurd request to also want to show double and triple word overlaps. Page 33v has been translated into Turkish in its entirety and the harmonies in the sentence structures have been fully demonstrated. Looking at these is enough to understand the result.

Examples from VM's page 33v could already display these overlaps. Our work indicates phonological and structural parallels between VM and Old Turkish.

Sentence Structure Overlaps:
Analyses suggest that the syntactic structure of VM resembles that of Turkish—both exhibit agglutinative properties (words formed by linear assembly of morphemes). A full-page and full-sentence analysis can illustrate these overlaps, reinforcing the Turkish hypothesis without solely relying on bigrams and trigrams.

Necessity of Bigrams and Trigrams:
Full-sentence and page analysis already convincingly show the structural resemblance to Turkish.
Bigrams and trigrams are supplementary but not essential for proving this relationship. They can enhance the evidence (maybe for some people who don't understand how to analyze a linguistic paper like ours sure) but are secondary to syntactic and phonological analyses which provide more substantial proof of the language linkage.

The full-page (33v) readings already show complete sentences consisting of 6 to 14 words arranged side by side (in a Turkish sentence form overlapping within Turkish to VM) by finding these words in dictionaries. It means that it already shows this overlap in a complete sentence on a full page written with 6 to 14 words, far beyond bigram & trigram word structures. In other words, It is not strictly necessary to find bigram and trigram overlaps, as the broader sentence structure and page-wide analyses already robustly support the hypothesis of the VM text correlating with Turkish.

The study has been done which provides more detailed broader & clearer information efforts to understand the language in the content by analyzing full sentence structures to prove that an entire sample page shows a clear overlap between VM and Turkish. For this reason, if a study has been conducted that provides broader and clearer information, it is pointless to go into detail looking for overlaps in a narrower area.

If this work is done for a full page of 10 sentences of 6 to 14 words and 11 lines in length, much further than the bigram and trigram scale (beyond examining double and triple word overlaps), it means that a more difficult job has been carried out.
That's what we did and demonstrated the following.
1- We showed that the VM sentence structure and Turkish sentence structure overlap.
2- We showed that all the words on the full page read were found in known dictionaries (by showing that 21% of them did not change their phonetic value in 600 years and that the old forms of the remaining words were found in old manuscripts and dictionaries and in the dialect of a certain geography).
3- We defined the words by showing that their roots and suffixes are familiar in Turkish.
4- We evaluated the phonetic equivalents of the words that can be read close to each other in all possibilities and showed that each different possibility provides overlaps between the compared VM language and Turkish without disrupting the Turkish sentence structure and semantic integrity.
5- We translated the entire VM 33 v page into today's language, and while doing this, we also showed in which geography the old and new forms of the words were used and in which dictionary or linguistics article they were mentioned.
6- We did not stop with these, we showed more than 100 random sentences selected from random pages and more than 1000 randomly selected words in dictionaries. In addition, while doing this work, we showed that 112 of the drawings made in this manuscript overlap with some words and pronouns on the same pages.

The researcher is free to choose the research method. If we had proceeded through the bigram and trigram structures and shown clear results, these results would still not prove that the VM content was in Turkish, and at the next stage, we would have been asked to translate full pages consisting of full sentences into today's language. For this reason, we skipped this unnecessary step and made the full-page translations directly, and we achieved a job far beyond bigram and trigram overlaps. Therefore, although our basic method includes what you want to see, it has gone to the last point that linguistics can reach in this type of study, and has achieved more difficulties, and the results have been announced to linguists through international symposiums.

What you are asking here is equivalent to asking us to retest whether a vehicle that has been proven to carry 14 tons can carry three tons, and we don't have to change our working method according to a person's personal preference when the results are obvious. However, when we have more time, I do not see any harm in showing that a vehicle carrying 14 tons can also carry 3 tons. First, try to understand that full sentence analysis and full page analysis have been done and that it is understood that there is a complete overlap with Turkish.