A decryption challenge

[Risla]

So I've been keeping a journal in code since 2010-ish, and more or less finalized the form of it in 2012. This was originally the conscript for a long-abandoned conlang, but I adapted it to write English and have used it for such ever since. I'd been considering making a post explaining how it works, but thought it might be more fun to upload a sample and see if people can figure it out. I figure you're all pretty damn unlikely to ever have access to my journal (and thus my private thoughts)!

A few notes about it:

• It's got a number of quirks, and especially function morphemes are not always what you expect them to be.
• It's not 100% consistent—there can be multiple ways to write a given word, and I tend to mix them up a bit.
• Although I've read over it a couple times, it's possible that I've missed a small mistake or two (especially missing dots). The text contains a handful of mistakes which I've scribbled out.

So without further ado, here it is! Feel free to ask any questions or ask me to rewrite any segments, but to keep it as challenging as possible I won't give any explicit information how it works…yet.

[Yalensky]

I can tell that it's written left-to-right.

The little uppercase "gamma" begins each paragraph of the journal entry and can be isolated from other characters by interpuncts, so I'm betting it represents "I", whether the sound [ai] or the letter <i> or only the 1^st person pronoun I don't know. (Assuming interpuncts separate words, which in the absence of spaces makes sense to me. And then again, maybe the gamma just means "new paragraph".)

EDIT: actually I see now you don't mention whether this sample was created for the ZBB or was taken directly from your journal, so hmmmm

[Risla]

The sample was created for the ZBB. It talks about some of my recent experiences.

[WeepingElf]

I can't say anything about its decryption now, but the script is beautiful and reminds me of Carsten Becker's Tahamo Hikamu.

[Risla]

I can't say anything about its decryption now, but the script is beautiful and reminds me of Carsten Becker's Tahamo Hikamu.

Thanks!

I've discovered a couple of mistakes—missing dots (as expected, oy) and a missing letter. I'll fix them in the morning…

[masako]

I can't say anything about its decryption now, but the script is beautiful and reminds me of Carsten Becker's Tahano Hikamu.

I concur.

[alice]

Perhaps each line begins with a date, or perhaps "on this day"?

I detect something about cigarette papers, so perhaps you don't distinguish s and z, with whatever brackets are appropriate?

It's very nice, in any case.

[Kuchigakatai]

This is very nice. The overlong descenders remind me of some the more flourished styles of Guitarplayer's Tahano Hikamu too, but the letter cores themselves seem like the children of Hebrew and Hangeul.

I don't think I want to spend time deciphering it today or this weekend, but maybe I'll try a little bit the days after.



As preliminary hypotheses go, as alice said, I wonder whether the first three glyphs of each paragraph is a date. Although it could be the first glyph is just how you write "the".

The tokens separated by middle dots seem strikingly short considering this is English, which makes me think that either the dots separate morphemes as opposed to words, or perhaps this writing system works partly as an abjad. You mentioned you have quirky glyphs for functional morphemes, but eye-balling it, the inventory of glyphs seems pretty low for its length (and the glyphs themselves are not complex). A closer examination may show I'm very wrong about this though.

[Risla]

To be clear, the overlong descenders are actually just a quirk of my handwriting.



(yes, my lowercase g is always this fancy)

It's very possible to write with much shorter ascenders and descenders:



I just usually don't!

[bradrn]

It's very possible to write with much shorter ascenders and descenders:



I just usually don't!

If you do end up rewriting this to fix the missing dots and missing letter, do you think you could use this format? It would make it much easier to read! (I must say, that is some beautiful handwriting though… much nicer than mine!)

[alice]

I'll second this: it would be helpful to be able to distinguish the graphemes a bit more easily.

But you have very lovely handwriting; it reminds me of Tolkien's mother's, for some reason.

[Risla]

Okay, I'll do that tonight! Should be up in a couple of hours.

[Risla]

Here you go.



I was pretty careful with this one. I did absentmindedly slightly change the wording in one place, and made a couple of stylistic changes. Not every change in here is mistake correction…

[bradrn]

Thank you! I haven’t looked at it in detail yet, but it already looks a bit easier. (e.g. there were some places where what I thought were single letters turned out to be two letters instead; and it’s much more clear now that dots always occur between letters, whereas before some of the dots looked like they were inside a letter.)

[alice]

Has anyone made any progress with this? I can identify the end-of-sentence markers, but that's about it.

[Moose-tache]

I'm very interested, but I've not been able to accomplish much. I had an easier time identifying vowels in the Voynich manuscript.

[quinterbeck]

Has anyone made any progress with this? I can identify the end-of-sentence markers, but that's about it.

I hadn't realised that's what they were! I'm curious about the doubled and tailed variants. Maybe that means the angle brackets are quotation/name marks; there's a combination of both in the middle of the second paragraph.

More: show

I'm in the process of making a frequency list of glyphs ignoring diacritics - so far I reckon this is an abugida with vowels represented by combinations of (de/a)scenders and diacritics.

[alice]

I think we'll need to petition Miss Burns for further help...

[Qwynegold]

I'm in the process of making a frequency list

Are you still on this?

[quinterbeck]

I'm in the process of making a frequency list

Are you still on this?

I'm still working on the decryption, but the frequency list is done now

I've determined the basic shapes (or series) and the correspondences between x-height chars and chars with ascenders/descenders (grades). They make a four grade table of 29 series, including chars with bar diacritics as separate series (7 of those). The table only has a few gaps. I've tallied the occurence of each char in the table, ignoring the dot and grave accents, which seem pretty evenly distributed.

Based on that, I think it's a sort of abugida with the series indicating a consonant (although it's 5 more than english has), and the vowel indicated by a combination of grade and accent.

I've also listed all the words in the text and counted their frequencies. So far, guessing based on frequencies hasn't yielded much, so I'm encoding the text into excel, hoping that being able to make quick substitutions will speed decryption.

All my tables and tallies are on paper, but I'm happy to scan some of it and add to the thread if anyone wants