Why and how the text could be Bavarian

[JoJo_Jost]

I've have possibly "cracked" a part of the structure of the Cores (formerly C2) — don't worry, not a solution, no plaintext yet, just the structure. But before I lay this out, I need to bring you up to speed.

Something occurred to me. We all know the relationships between ched and shed with y and qo. Less well-known are the relationships of the "e" family in Slot 3 to the "o" after y, and some others. A few things just didn't seem to make sense. 

An example:
Why does schedy get followed by q 51% of the time?

   

And why only 39% for chedy?

   

Why doesn't the value in front of it change when the run gets shorter?

   

Why does kedy then sit at only 28%?

   

Okay, part of the effect has to do with the number of hits—the “n” in the graph.

But the difference between Kedy and shedy—n = 621 / 542—isn't affected by that.

So how do you explain the fact that Slot 2 has an obvious influence across the Space?

And: Isn’t that an indication that the spaces aren’t really spaces, since they all continue running smoothly toward K? We don't know, but it.... 

Or let’s take another example: daiin and aiin

       

They run almost synchronously after the space (the possible divergences are plotted here as percentages)—amazing, isn’t it? What are Spaces???

My solution - Part 1

Well, that puzzled me. And then something struck me: we all know that "qo" is actually two characters (because they aren't fused like ch and sh) and yet they're tied together nearly 100% of the time. If you approach Voynichese from the back, reading line by line in reverse, then q is simply a normal successor to "o". Sure, o is exclusive to q, but not the other way around. And once you approach those chains above from behind, the whole thing suddenly makes sense. The high connections between qy and the E-family suddenly reveal a clear structure — the relationships develop from left to right, against the writing direction.

With this view, a few other things tipped over too:

1. The aiin / air family lines up beautifully with the endings in MHD/FNHD — read in reverse, they show up exactly where they should: in C1, i.e. at the word/morpheme end. Previously I had to somehow reverse-engineer them into C2 as endings, which was very unsatisfying and led to downstream errors in the stream.
   
2. The vowel bridges, read in reverse, strongly conditions the following E-family — meaning they influence it. This is where the actual code of the Cores sits. Running it in reverse now, it works exactly as it should. Previously you had to build the existing connection backwards from the end to the front to construct a matrix — but that's not how humans work, and I didn't like it. Now it runs cleanly in sequence: first the vowel, then the switch, then the E-family. Exactly the way you'd build it up.
   
3. The LEMs become the actual line-start markers, and you need so few of them because the only thing that has to be clarified is whether the line starts with a vowel or a consonant. What the former Line Start Markers are actually doing now — I haven't looked into that yet, but they probably terminate the stream.
   

The whole thing was a bit like the planet effect: as long as you think the Earth is at the center, you can barely make sense of those strange planetary motions, patching them up with little epicycles. The moment you see that the Sun sits at the center, the orbits suddenly become strictly, logically explainable. All at once, everything in the VBM fell into place.

These extreme oddities disappear quite easily when we look at the sequence in reverse:

Here, the sequence is viewed from the back, starting at k (where the symbol ends at the top)—the most possible branches are indicated here. You can see in the last few branches that everything is distributed fairly evenly.

   

Some of the most awkward dependencies become much more natural in mirror direction. K does not influence ch. It is a simple sequence of dependencies.

When the sequence is read normally from left to right, oddities arise that make sense if the current were actually flowing in the opposite direction.

But that’s not all—the structur will become clearer in the next post.