For anyone who wants to get their teeth into decrypting an English cipher that's somewhat easier than Risla's, here's one for you!
I devised this a few years ago and don't use it very much (you can see had to do some surgery in places). The only non-English word in the text is the name of the cipher itself. It does have a few quirks, and a number of abbreviations of common words. It's not as elegant as Risla's but it works fine.
EDIT: fixed some more errors
Last edited by quinterbeck on Fri Sep 18, 2020 10:45 am, edited 2 times in total.
I did a transcription of this. E stands for the kind of [ looking character where the upper stroke is arched downwards. I'll count the characters soon.
ıπ c£LTE ùΔʌ w Δ̈TḾ ƽ π̈ c̈ƽTE ȷ̇ıπ wo |ı̈T ōΔȷM̄ɔ π̀ȷT̈TE .
w oL̈ β u иȷ̈£ʌ p̄ıc иċ Δw u oπ̄ʌ̇ȷTE ƽ |ċ|ɔ Δ̈ıʌ Lc̀ΔMo
ȷ̀ıcTE uc̀TE T w Δ̈TM.
иȷ́ıβo pc̄ɔ Δw w иȷ̄Tѥo ƽ w ʌȷ́ıβ Lȷ́ūıʌTE u ѝcM Lȷ́ıT |ċʌ̇ȷTE ƽ ćıƽo .
ı LT ó u Lȷ̀ T w ȷ̇ı| ƽ u Ēȷ̈ζ ūʌ w ӣL - ΔT ƽ w L̄|́ɔβ M̀ΔToʌ́ȷβ
ɔ̇βo uʌ ūβ oʌ̀ȷ̈ζ .
ü| ı ʌ̄ıL w ӥTo ùΔʌ Δ̈ıc ẃɔȷ ı̀pT uʌ ucΔ̄ɔβ óıπo ʌ и u̇Ṫʌ́TMM
TM w ȷ̀Δβ ƽ Mȷ̈TLo ѝʌco T w źcƽo ū ucπ̀ıoʌ ūoLTE ʌ и T̈Lʌ .
иʌ ıπ T̀ v́ı| Ṁop̈ıʌ s̀Δ ı́βɔ uʌ ΔM и ʌ öπpcɔ ʌ̄ıL .
uʌ óıπo wʌ ıƽ Èʌ ɔ̇ıoʌ ʌ T̈ıo LΔ̈uʌ Δ́ıĹTMo.
и̇|̀ȷ w c̀LM̀ΔT ıM ɔ̇ıχ̇cɔ È ʌ u ζ̄ṗT́ıβ Lc̄o иʌ T̀Δ ẃȷβ Ṫv̈TE ȷ́Eɔ̇ċȷ
ʌ uʌ́TM - up̄ȷʌ |ȷ̀π ѥȷѥ T ȯTM̄ɔ .
T u |ɔ̇ π̈Ṫʌo ıc È ı̀ƽȷ ʌ w ĒȷMT óTʌȷ ʌ ӥ u pc̄Tʌ |̇ π̈ M̄Mo иȷvM̄ɔ
Δ̀To w ı̈иɔ̇ıoȷ̀|́T ı ɳ̇oʌ ʌ£L s̄β ṀcM w p̄ıT T π̈ |£ʌ.
uʌ Δβ ÉʌTE и́ʌȷ иʌ uʌo |ćȷM up uÉT ʌ̇M̄ɔ.
ı zM ȷ̈cɔ Éʌ uȷ̀ΔTM ʌ óıTE w M̀Lʌȷ иʌ ı Ĺıp p£ʌTE uʌ ù| и́L̀β ƽ w
s̄oc ƽ ȷ́ɳ̈oʌ̇ȷTE ūʌ u Tɔ̇ pȷ̄Lʌ̈o
Δ̈ıc ı Δ̄ıʌ |̇ wʌ ıπ ȷ̈ıʌTE wo ùΔʌ T T̀ƽ́TEc̀ - Δ́c ūLѥ̇cɔ Mȷ̄|ʌTE
w ʌ́Loʌ - ıc ȷ̈ıʌ uʌ T |c Δ́ ı Éʌ ӣL.
ı sƽ ʌ È ȯıT и̇|̀ȷ w ĒȷMT óTʌȷ Lc̀ıβo
ʌ 75
T 66
ı 56
ȷ 47
o 47
u 40
c 37
M 31
L 29
E 29
Δ 29
w 29
и 24
ɔ 22
| 20
ƽ 18
π 17
β 16
p 13
£ 5
s 5
ѥ 4
ζ 3
v 3
z 2
ɳ 2
χ 1
27 base characters
á 40
ā 38
à 37
ȧ 35
ä 34
5 diacritics
. 11
- 5
2 interpunctuation
There are five diacritics which would correspond to five vowel letters? And maybe there's also an inherent schwa? But all the diacritics are almost equally frequent, which is really weird.
ƽ occurs 18 times, out of which 11 are just the character ƽ alone. This makes me think it's the word "the" when alone, and /θ/ otherwise.
The sequence TE occurs 17 times, and always word-finally. It must be some suffix.
w occurs 29 times, out of which 21 is just that character alone. I'm thinking maybe it's a silent vowel carrier, and when it appears alone it stands for the word "a".
At first I was thinking ѥ was a ligature because it's so uncommon and it looks like ı and c. But then I realized that the sequence ıc also occurs.
Oh crumbs, I've slipped into an old version partway through, although just for one character - the one you've labelled ɳ should be the same as ζ. I'll fix the text tomorrow
Qwynegold wrote: ↑Thu Sep 17, 2020 3:15 pmThere are five diacritics which would correspond to five vowel letters? And maybe there's also an inherent schwa? But all the diacritics are almost equally frequent, which is really weird.
ƽ occurs 18 times, out of which 11 are just the character ƽ alone. This makes me think it's the word "the" when alone, and /θ/ otherwise.
The sequence TE occurs 17 times, and always word-finally. It must be some suffix.
w occurs 29 times, out of which 21 is just that character alone. I'm thinking maybe it's a silent vowel carrier, and when it appears alone it stands for the word "a".
At first I was thinking ѥ was a ligature because it's so uncommon and it looks like ı and c. But then I realized that the sequence ıc also occurs.
Some good guesses there! Only off track in a few places
Although most of these seem to be dead ends, this makes me think that cɔ is also a suffix: it appears to occur five times (or possibly six, depending on how diacritics work), always word-finally.
There's also the interesting case of T̀ƽ́TEc̀, where the TE suffix occurs with...another suffix? But c̀ isn't very common (it only occurs five times), and this is the only case where it occurs word finally.
There's also one sequence of words that both have the TE suffix:
ȷ̀ıcTE uc̀TE
I'm reasonably certain that w and u are the articles: a and the (I don't know which is which). They notably frequently occur in sequences following other short words (one or two characters), which seem likely to be prepositions:
T w (3)
T u (1)
ƽ w (3)
ƽ u (1)
Δw w (1)
Δw u (2)
TM w (1)
ʌ w (1)
ʌ u (1)
ūʌ w (1)
ūʌ u (1)
β u (1)
ʌ ӥ u (1)
They also don't seem to occur prior to other short words, which backs up my suspicion that they're articles.
I'm also suspicious of π̈, which also seems to pattern in a similar way, but which there's not much data on.
A couple of other observations:
β takes no diacritics, and always occurs word-finally.
It also seems potentially notable that the sequences ʌ и and и ʌ both occur.
quinterbeck wrote: ↑Thu Sep 17, 2020 4:37 pm
Oh crumbs, I've slipped into an old version partway through, although just for one character - the one you've labelled ɳ should be the same as ζ. I'll fix the text tomorrow
ıπ c£LTE ùΔʌ w Δ̈TḾ ƽ π̈ c̈ƽTE ȷ̇ıπ wo |ı̈T ōΔȷM̄ɔ π̀ȷT̈TE .
w oL̈ β u иȷ̈£ʌ p̄ıc иċ Δw u oπ̄ʌ̇ȷTE ƽ |ċ|ɔ Δ̈ıʌ Lc̀ΔMo
ȷ̀ıcTE uc̀TE T w Δ̈TM.
иȷ́ıβo pc̄ɔ Δw w иȷ̄Tѥo ƽ w ʌȷ́ıβ Lȷ́ūıʌTE u ѝcM Lȷ́ıT |ċʌ̇ȷTE ƽ ćıƽo .
ı LT ó u Lȷ̀ T w ȷ̇ı| ƽ u Ēȷ̈ζ ūʌ w ӣL - ΔT ƽ w L̄|́ɔβ M̀ΔToʌ́ȷβ
ɔ̇βo uʌ ūβ oʌ̀ȷ̈ζ .
ü| ı ʌ̄ıL w ӥTo ùΔʌ Δ̈ıc ẃɔȷ ı̀pT uʌ ucΔ̄ɔβ óıπo ʌ и u̇Ṫʌ́TMM
TM w ȷ̀Δβ ƽ Mȷ̈TLo ѝʌco T w źcƽo ū ucπ̀ıoʌ ūoLTE ʌ и T̈Lʌ .
иʌ ıπ T̀ v́ı| Ṁop̈ıʌ s̀Δ ı́βɔ uʌ ΔM и ʌ öπpcɔ ʌ̄ıL .
uʌ óıπo wʌ ıƽ Èʌ ɔ̇ıoʌ ʌ T̈ıo LΔ̈uʌ Δ́ıĹTMo.
и̇|̀ȷ w c̀LM̀ΔT ıM ɔ̇ıχ̇cɔ È ʌ u ζ̄ṗT́ıβ Lc̄o иʌ T̀Δ ẃȷβ Ṫv̈TE ȷ́Eɔ̇ċȷ
ʌ uʌ́TM - up̄ȷʌ |ȷ̀π ѥȷѥ T ȯTM̄ɔ .
T u |ɔ̇ π̈Ṫʌo ıc È ı̀ƽȷ ʌ w ĒȷMT óTʌȷ ʌ ӥ u pc̄Tʌ |̇ π̈ M̄Mo иȷvM̄ɔ
Δ̀To w ı̈иɔ̇ıoȷ̀|́T ı ζ̇oʌ ʌ£L s̄β ṀcM w p̄ıT T π̈ |£ʌ.
uʌ Δβ ÉʌTE и́ʌȷ иʌ uʌo |ćȷM up uÉT ʌ̇M̄ɔ.
ı zM ȷ̈cɔ Éʌ uȷ̀ΔTM ʌ óıTE w M̀Lʌȷ иʌ ı Ĺıp p£ʌTE uʌ ù| и́L̀β ƽ w
s̄oc ƽ ȷ́ζ̈oʌ̇ȷTE ūʌ u Tɔ̇ pȷ̄Lʌ̈o
Δ̈ıc ı Δ̄ıʌ |̇ wʌ ıπ ȷ̈ıʌTE wo ùΔʌ T T̀ƽ́TEc̀ - Δ́c ūLѥ̇cɔ Mȷ̄|ʌTE
w ʌ́Loʌ - ıc ȷ̈ıʌ uʌ T |c Δ́ ı Éʌ ӣL.
ı sƽ ʌ È ȯıT и̇|̀ȷ w ĒȷMT óTʌȷ Lc̀ıβo
ıπ ӣL - ʌ̈ıπ ʌ ȷ̈ıʌ wo Δw u ċƽcɔ L̇o ƽ ʌ́ T s̄TM
ʌ 75
T 66
ı 56
ȷ 47
o 47
u 40
c 37
M 31
L 29
E 29
Δ 29
w 29
и 24
ɔ 22
| 20
ƽ 18
π 17
β 16
p 13
£ 5
s 5
ѥ 4
ζ 5
v 3
z 2
χ 1
27 base characters
Hmm, I checked and E does not occur word-finally unless it is preceded by T. What suffix is really common in English? I was thinking -ed, but I was assuming E stands for a consonant. Maybe vowels aren't always diacritics?
Risla wrote: ↑Thu Sep 17, 2020 8:10 pm
I'm reasonably certain that w and u are the articles: a and the (I don't know which is which). They notably frequently occur in sequences following other short words (one or two characters), which seem likely to be prepositions:
They also don't seem to occur prior to other short words, which backs up my suspicion that they're articles.
Aha. Well spotted!
Risla wrote: ↑Thu Sep 17, 2020 8:10 pmβ takes no diacritics, and always occurs word-finally.
Correction: we also have -βo word-finally (three instances). One word is just the character β.
Qwynegold wrote: ↑Sat Sep 19, 2020 3:11 pm
Hmm, I checked and E does not occur word-finally unless it is preceded by T. What suffix is really common in English? I was thinking -ed, but I was assuming E stands for a consonant. Maybe vowels aren't always diacritics?
The suffix -ing. The suffix -ed would be unlikely already, given that there's a sequence of two words with it—but that's okay with -ing! Alternatively, it could be the suffixes -(e)s (one or both of them) but my intuition is telling me no.
Correction: we also have -βo word-finally (three instances). One word is just the character β.
Whoops. I think my search box had a space in it… (I was counting the independent one as word-final, though, because it is! )
Qwynegold wrote: ↑Sat Sep 19, 2020 3:15 pm
Aha! I counted only two instances of u that are not word-initial. This is a vowel carrier with inherent /a/!
This seems like a big conclusion to jump to. I'm cautious of the idea that diacritics mark vowels, just because of frequency.
I'm going to direct at quinterbeck the same question directed at my own code: is this phonemic, or is it based on the English writing system?
Risla wrote: ↑Sat Sep 19, 2020 5:19 pm
I'm going to direct at quinterbeck the same question directed at my own code: is this phonemic, or is it based on the English writing system?
It's mostly phonemic, but does preserve a few elements of written English.
Consonants are almost completely phonemic.
For vowels, a few aspects of English spelling have been incorporated.
Some vowels have more than one representation.
There's one or two irregularities for common morphemes.
After looking at this a bit more, I'm inclined to say that w is "the" (and likely the phoneme /ð/) and that u is "a", for the reasons Qwynegold mentioned—I agree with the logic that it's probably a vowel carrier, although I wouldn't go so far as to say it has an inherent /a/.
I think w is probably /ð/ because, looking at distribution, it seems to frequently occur in these four words that seem likely to be function words:
w (18)
wo (3)
wʌ (2)
Δw (3)
These are short, high frequency words, and there are only two (?) other words with w, out of 29 instances of this character in the text. This is consistent with how the phoneme /ð/ patterns in English.
In that context, I'm inclined to think that Δw is very likely to be the word "with": it only occurs prior to the apparent articles, which makes me think it's a preposition, and what other frequentish prepositions would have /ð/ at the end? (IMD it's /wɪθ/ but I know that's not true for everyone).
This means wo and wʌ are potential candidates for the word "that", which is usually also very frequent. Given that /t/ is one of the highest frequency consonants in English and the fact that the character ʌ also occurs as an apparent preposition ("to"?), I will tentatively hypothesize that "that" is wʌ.
I'm also suspicious of the character β. The fact that it doesn't take diacritics (which I also suspect to mark following vowels, hence the use of a vowel carrier character that generally occurs initially) and is almost exclusively medial or final makes me think it may be /ŋ/, but then I'm confused about the one case where this character occurs independently.
Questions: Can the character β occur with diacritics?
I'll take a whack at it soon.
)
