Another language/script puzzle

« previous post | next post »

On reddit.com/r/linguistics, someone recently posted the following query:

At the bottom of a letter written by one John England during the US Civil War (it is not by Bishop John England) is a paragraph in a script and (seemingly) language that I don't immediately recognize:

I think this is Irish Gaelic written in the old Gaelic script, but I'm really not sure. Anyone have any guesses?

I suspect that there are some LL readers who can do better than a guess.



71 Comments

  1. Ben Hemmens said,

    January 30, 2013 @ 2:05 pm

    No it isn't, if those are z's and j's and w's I see; the traditional Irish gaelic alphabet doesn't have the letters j, k, q, v, w, x, y or z. Also not like Irish are the double-dot diacritics. And though my school Irish is a long time ago, I can't recognize any words. A just-possble "agus" (and) as second word, if it's a word, but no more.

  2. john mcferran said,

    January 30, 2013 @ 2:07 pm

    I'm pretty sure it's Manx (Isle of Man) Gaelic

  3. janice said,

    January 30, 2013 @ 2:33 pm

    Elvish

  4. Darryl Shpak said,

    January 30, 2013 @ 3:16 pm

    There are a couple of words that I can make a guess at (like "Deil" = "Devil" in the second line), but just deciphering the script is proving to be challenging. Assuming the author's name is correctly spelled "John", then the lowercase "o" has an oddly square form that doesn't match anything I've been able to find elsewhere…though I have seen examples of lowercase "a" written this way.

    Bear in mind that I'm stunningly uneducated in these matters, since I know no Gaelic or related languages, and have very little experience with old handwriting. So this is just a diversion for me, not anything that I expect to be able to make a significant contribution to…

  5. Lillian said,

    January 30, 2013 @ 3:33 pm

    Turkish?

  6. Daniel von Brighoff said,

    January 30, 2013 @ 3:44 pm

    It does resemble cló Gaelach (making those "z's" actually g's), but it's full of non-Irish features such as the digraph sh in final position and a lot of terminal a's (rare in Irish outside of borrowed nouns). I'm leaning toward Manx myself.

  7. Cindy said,

    January 30, 2013 @ 3:45 pm

    A quick poke around Wikipedia suggests the Gaelic guess is good. The letter forms look a lot like the ones shown here: http://en.wikipedia.org/wiki/Gaelic_type

    Which means that those aren't zs but insular gs… Unfortunately I know no Gaelic, so I'm going to be no help deciphering the actual content.

  8. Howard Oakley said,

    January 30, 2013 @ 3:53 pm

    Although Manx has highly variable orthography, not having undergone the reform that Gaelic and Irish did, 'agus' is Irish or Gaelic: Manx would surely be 'as'.
    The script is Irish in form too – cf. any printed text using a traditional Irish font.
    Howard.

  9. Levni said,

    January 30, 2013 @ 4:01 pm

    I know this is hardly helpful, but I'm curious as to why Lillian would have suggested Turkish.

  10. Svafa said,

    January 30, 2013 @ 4:12 pm

    I'd agree with Daniel von Brighoff that those are g's and not z's. The diacritics are still odd (diaeresis above g?). I'm not familiar with Manx, but I didn't think it used diacritics.

  11. Darryl Shpak said,

    January 30, 2013 @ 4:17 pm

    I've heard of various dots and dashes used in old handwriting to indicate abbreviations, so perhaps the "diaeresis" above the g's indicates that a letter was omitted? My understanding is that that sort of shorthand could be highly variable from writer to writer.

  12. Cindy said,

    January 30, 2013 @ 4:18 pm

    Are the single-dot/double-dot diacritics on the g being used to distinguish its stop and affricate pronunciations? This is a guess based entirely on how he wrote his name…

  13. Seán Ó Briain said,

    January 30, 2013 @ 5:34 pm

    The dot above a letter in Irish Gaelic was traditionally a seimhiú (lenition), but is today represented by the letter h.

    If the second and third word were 'agus' and 'le' (meaning 'and' & 'with') – then they would not be mutated as shown in the letter (aġus / lé).

    Wait for a better reply.

  14. Michael Bauer said,

    January 30, 2013 @ 6:01 pm

    Ok, first we need to remember that Uncial script, while it was common in the Celtic countries for much longer than elsewhere, was not restricted in its use to just Celtic languages.

    But as others have pointed out, beyond the script, it does not really make much sense reading it as Irish/Manx/Gaelic. So… it's marked PRIVATE, so we can assume with some certainty it's a cipher. Looking at the only immediately obvious bit, his signature, we see that the dotted ġ appears where we expect J, a comes in place of o, h and n are the same. In England, the yogh (ȝ) take the place of g and the only other oddity seems the last bit there th seemingly takes the place of d.

    I would say it's a some sort of a replacement cipher. I'd try that alley.

  15. Jerry Friedman said,

    January 30, 2013 @ 6:03 pm

    I think Cindy's suggestion about the dots on the g's makes sense. Now what's the difference between the J with two dots in "John" and the J with one? (If the dots are a code and the letters are just a distraction, it's going to be hard to break.)

    The h-bar proves that the author knew quantum theory avant la lettre.

    And yes, Tolkien must have seen scripts like this before inventing tengwar.

  16. W.F. said,

    January 30, 2013 @ 7:51 pm

    Anybody want to take a stab at transliterating/transcribing this? Some of the letters I am very unsure of, and I'd love to see what people make of Mr. English's spelling of his own name.

    Superficially, this doesn't look like Manx to me at all, and my understanding is that this script isn't associated with Manx.

    Anybody think it's interesting that the capital D characters aren't in the "gaelic script"?

  17. catsidhe said,

    January 30, 2013 @ 8:45 pm

    It looks like Irish, but with private orthography. It looks like the insular 'g' with diaresis represents 'g', but insular g with one overdot represents 'j'.

    "Gardagv* agus le meilyga dcs*aguigi…"

    It's very far from standard Irish, and I can only pick words out here and there: agus (and), le (with), garda (guard, police), maybe moir = 'mór' (big, great). The rest is incomprehensible.

    Interesting that he spells his name "Jahn".

  18. dainichi said,

    January 30, 2013 @ 8:52 pm

    If indeed it is a cipher, why would somebody mark it with "PRIVATE"? I mean, to me that's basically an invitation to try to break it. Is it an indication to the (intended) reader to switch into deciphering mode?

    Or am I over-thinking this?

  19. Dave K said,

    January 30, 2013 @ 9:20 pm

    The "private" seems like an explanation for why the writer suddenly made things difficult for the reader.

    It might help to know who England was writing to. If he was writing to a parent, it's more likely to have been another language–he wanted to show that he hadn't forgotten his ancestral tongue. If it's addressed to a spouse or a friend, odds are greater that it's a cypher the two of them worked out for fun. If it's a sibling, it could go either way.

  20. W.F. said,

    January 30, 2013 @ 9:24 pm

    I didn't even think twice about the Jahn spelling– I was on auto-pilot, figuring he was just correcting for pronunciation.

    Here is a quick transcription, I'm sure it's got mistakes. I used ǧ for the double-dot g, and didn't mark the other diacritics, to focus on forms.


    Ǧardaǧv agusle meilyga desaguigivs segaleinoir O! mair la shjigtin*, maoiǧino le hisgvpauz!

    Ouvacǧas manoiǧa delglmj'sh, Deil ougeaji nioanouga conimjroga! Dinogagaiga mywi fair.

    Ǧahn Englanilh


    One thing I noticed, there's dotted and dot-less I in this, as in what I wrote as shjigtin*, which has one of each. Could just be an error, anybody have any other ideas?

  21. baernty said,

    January 30, 2013 @ 11:04 pm

    1) Why does it have to be one language? It could be Irish mixed with something like German

    2) Maybe it's gibberish.

    3) Maybe it's a mix of one or many languages, as well as gibberish. I'm a novice learner of many languages and I tend to either innovate (ungramatically) or play around with the language. It could be a case of "play"

  22. Max Overby said,

    January 30, 2013 @ 11:16 pm

    Could this be a mix of Gaelic words and phonetically spelled English ones? Sort of the reverse of what humorists of the era and later used to do with Irish, Yiddish or German? (Finley Peter Dunne's Mr. Dooley, for example.)

    I also wondered if the odd diacritical marks were a way to represent letters or sounds not present in Gaelic.

  23. Thomas Widmann said,

    January 31, 2013 @ 4:02 am

    Could it be Old Shelta (http://en.wikipedia.org/wiki/Shelta)?

  24. PaulB said,

    January 31, 2013 @ 5:36 am

    For what it's worth, my transcription, following WF, is:
    __

    Ǧardaǧr agus le meilyga desa Guigirs segaleinoir O! mair la shrigtint, moaǧino le ħisgrpauz!

    Ou racǧas manoiǧa delglms'sh, Deil ougeasi niainouga conimjroga! Dinogagaiga mywi fair.

    Ǧahn Enǧlanth
    __

    He seems to mix up Gaelic and English letter forms for 'a', 'i', 'r', 's' and 't' (I'm reading 'r' where WF has 'v'. I could easily be wrong; it would be helpful to see the handwriting in the rest of the letter.)

    I have no special expertise, but it seems to me that this is a few Gaelic words mixed in with something else, written in a partly Irish Gaelic script. Perhaps the something else is a Gaelic version of Pig Latin.

    I can't make sense of the diacritics. I suppose the accent in 'lè' is not correct in Gaelic. The 'ħ' is unexpected: Maltese is the only language I know of that uses it.

  25. Qaoileann said,

    January 31, 2013 @ 6:14 am

    The ħ might be an attempt to render Gaelic 'ch'; similarly 'v' could be 'bh' or 'mh' (as they are written in modern Irish Gaelic). When I replace those some words seem more familiar, but we'd really need someone expert in the Irish(es) of the mid-nineteenth century… if it is Gaelic or indeed a language not a cipher.

    Ghardaghbh agus le meilyga desaguigibhs segaleinoir O! mair la shjigtin*, maoighino le chisgbhpauz!
    Oubhacghas manoigha delglmj'sh, Deil ougeaji nioanouga conimjroga! Dinogagaiga mywi fair.
    Ghahn Englanilh

  26. Mr Punch said,

    January 31, 2013 @ 1:16 pm

    I find it hard to believe there's any language containing the word "chisgbhpauz".

  27. H said,

    January 31, 2013 @ 1:22 pm

    If it is a cipher, "O" must be A or I. I can't find 'O' as a word in Gaelic.

  28. Jeff B said,

    January 31, 2013 @ 1:36 pm

    I am guessing this is the same John England who wrote these letters to Ellen Hargedon:
    http://www.scribd.com/doc/19329905/1862-Civil-War-John-England-letters-Read-in-Fullscreen-Zoom

    His "PRIVATE" looks like it's in the same handwriting as the block lettering in these letters.

    I haven't succeeded in finding more information on this letter-writing John England, other than the assertion that he was Irish-American.

  29. Jerry Friedman said,

    January 31, 2013 @ 2:37 pm

    Good God, look at the handwriting in those letters!

    Dave K, I agree, and I think there are other reasons that the rest of the letter might help.

    I'm beginning to wonder whether some digraphs starting with "g" should be skipped.

  30. marie-lucie said,

    January 31, 2013 @ 6:34 pm

    The handwriting in the letter does look like a model of calligraphy! I agree that the block capitals look the same in the address at the end of the letter and in the word PRIVATE. In addition to using a cypher of some kind in the PRIVATE document, the handwriting is rather sloppy. This would indicate that the message is addressed to a family member or close friend who shares the secret and with whom the relationship is informal, while with Miss Ellen H. the writer is on his best behaviour. Additionally, it is usually harder to write fluently in a language or script one is not fully familiar with, so that one writes letter by letter rather than in complete words and phrases.

    The address has what looks like the word "Zouaves" in it. I didn't know that the term had been used in the US Army.

  31. marie-lucie said,

    January 31, 2013 @ 6:51 pm

    Help!

    In order to look at the letter I had to follow the instructions to enlarge the page to Full Screen and also to zoom. When I tried to reverse these operations, I managed with some difficulty to leave the site containing the page, but I cannot get out of Full Screen and have access to the toolbar, except for a thin one-line space at the top when I am lucky (I see the LL URL but cannot scroll down to other addresses, and the page I am on – like now – has the text in the middle, including a wide grey margin on the right, and the whole thing is flanked on both sides by narrower pale blue margins. What can I do to get back to normal?

  32. Belial said,

    January 31, 2013 @ 6:53 pm

    F11

  33. marie-lucie said,

    January 31, 2013 @ 6:53 pm

    OK, strolling up, I see that the wide grey margin is the one with all the names of Language Log writers, but apart from that I am stuck on this page.

  34. marie-lucie said,

    January 31, 2013 @ 7:03 pm

    Belial, thanks for the suggestion. I did F11, which just rolled up the page with just a one-inch left at the top. Finally I shut down the computer, waited a few minutes and started again. Things are now back to normal. I guess I was so used to the margins that I never noticed them until the menu bar at the top was gone.

  35. Rod Johnson said,

    January 31, 2013 @ 7:03 pm

    Marie-Lucie: yep, there were Zouaves on both sides during the US Civil War.

  36. Belial said,

    January 31, 2013 @ 7:23 pm

    When in their cups, the soldiers called the Zouaves "chisgbhpauz".

  37. Chris C. said,

    January 31, 2013 @ 7:40 pm

    When in their cups, the soldiers called many things "chisgbhpauz". So do I, sometimes.

  38. Ó Domhnaill said,

    January 31, 2013 @ 8:01 pm

    maybe its a code script? it does say private

  39. David said,

    January 31, 2013 @ 9:10 pm

    My yiddish-speaking mother-in-law said I had a lot of "chisgbhpauz", or something like that, when I first proposed to her daughter.

  40. marie-lucie said,

    January 31, 2013 @ 10:17 pm

    The French Zouaves arose from the conquest of Algeria and they wore a more or less Arab-looking uniform. Did the US soldiers too?

  41. maidhc said,

    February 1, 2013 @ 12:53 am

    Here's a picture of a Zouave uniform:
    http://www.civilwar.si.edu/soldiering_zuoave.html

  42. Jeff B said,

    February 1, 2013 @ 10:13 am

    The John England who wrote the Ellen Hargedon letters was a private in Company I (apparently letter i, not Roman numeral one) of the 9th New York Volunteer Infantry Regiment, known as Hawkins' Zouaves.

  43. KevinM said,

    February 1, 2013 @ 10:22 am

    @marie-lucie. Good instinct:
    My guess is that it says "Ellen is putting on weight."

  44. Qaoileann said,

    February 1, 2013 @ 10:43 am

    @ Mr Punch: clearly you have no orthographic imagination:)

    Coincidentally or not, other minority languages in Ireland in (very limited) use at the time included Shelta/Cant, Yola (Wexford only), and Fingallian. http://www.worldirish.com/story/22176-noos-a-daalyon-6-languages-from-ireland-that-you-may-not-have-heard-of

  45. Tom Emerson said,

    February 1, 2013 @ 11:11 am

    I'm the OP of the inscription: thank you all for your insights. In response to some of the questions raised.

    JohnB is right in that the letter is written by the same John England from the NY 9th. All of his letters are in the hands of the New York Historical Society and the company I work for (EBSCO Publishing, though I am definitely representing them on LL) is collaborating with NYHS on these.

    I got permission to make the full page available:

    http://i.imgur.com/WIwXuef.jpg

    Unfortunately I don't know exactly who he was writing to since I haven't seen the preceding pages. He refers variously to Bill and Jack at various points.

    His writings are very interesting: my wife has read more of them than I have, but I plan to go through them at some point and see if I can find other instances of this "private" script. I've become obsessed with finding out more.

  46. Belial said,

    February 1, 2013 @ 12:00 pm

    Seems pretty clear from the full page posted that he is writing to a friend or relative named Bill, and that "Jack" is himself.

  47. Jeff B said,

    February 1, 2013 @ 1:16 pm

    Looking at the online catalog of the NYHS, my guess is this is one of the letters from John England to William Kelby, longtime librarian of the Society. Kelby was born in 1841 in County Sligo, Ireland, to "Scotch-Irish" parents. Kelby's family moved to the U.S. when he was an infant. This is according to this biography:
    http://archive.org/details/memoirofwilliamk01stev

  48. Jerry Friedman said,

    February 1, 2013 @ 3:14 pm

    Tom Emerson: Thanks for sharing the full page.

    Did you by any chance leave out a "not" after "definitely" when you wrote "the company I work for (EBSCO Publishing, though I am definitely representing them on LL)"?

    After a glance at Kelby's biography, I'm thinking he wasn't likely to have known Gaelic.

  49. Tom Emerson said,

    February 1, 2013 @ 4:35 pm

    Jerry Friedman: Indeed I did leave a word out… I'm *not* speaking for EP here. :)

  50. Gearalt Ua Fathaigh said,

    February 1, 2013 @ 6:22 pm

    Irish is my second language – 'agus le' for the second and third words, meaning 'and with' – seems strange. I propose the 'agus', if Irish, is instead 'abhus' not 'agus'; the second letter being 'b'. abhus means 'here, in this place, on this side'. I'm trying to recognise other non standard spellings and contractions in the text. 'Deil' in the second line looks like the Connemara contraction 'Deile' from 'Cad eile', meaning 'what else'. Could be totally wrong. There is a small genre/cache of letterwriting from 19th century Ireland of people who attempted to write Irish speech with their limited knowledge of Irish spelling melanged with English orthographic conventions (resemblinh Manx, though Welsh has an influence there!)… fascinating…

  51. Gearalt Ua Fathaigh said,

    February 1, 2013 @ 6:28 pm

    something about the word transcribed 'desaguigibhs' looks to me like it is 'do shaighdiú(i)r', meaning 'for/to (a) soldier(s)…

  52. Gearalt Ua Fathaigh said,

    February 1, 2013 @ 6:40 pm

    hang on – i'm flummoxed – take all my suggestions with a pinch of salt; there are too many confounding elements to attribute it as Irish…

  53. Qaoileann said,

    February 1, 2013 @ 7:07 pm

    A Ghearalt,

    Just going through the transliteration and sounding it out, some more may make sense by themselves e.g.:

    Ghardaghbh = comhairdegeas (congratulations)
    chisgbhpauz = uisce beagh (whiskey)
    Oubhacghas = ar/an bhfaca tú… (did you see…)

    Of course that could be entirely fanciful but I like the possible congruence of congratulations and whiskey…

  54. maidhc said,

    February 1, 2013 @ 7:20 pm

    -ibh was the old dative plural ending, wasn't it?

    It could be some kind of cipher. I have a diary from about the same period that uses a partial substitution cipher (not all letters are substituted) whenever he talks about illicit sex.

  55. W.F. said,

    February 2, 2013 @ 1:25 pm

    Maidhc-
    What kind of partial substitutions were used in that cipher? I'd assume that the bulk of codes are people who are mimicking a cipher somebody else came up with, so seeing how robust contemporary strategies were could be illuminating. (e.g. "just substitute the vowels")

    That asked, I'm still not sure I see powerful evidence pointing to a cipher over some bizarre improvised spelling of Irish, or for improvised spelling over a cipher for that matter. If anybody can put forth anything a bit more definitive (maybe with his name?) I think that'd be a big step toward solving this. Currently, I just find it telling that people with some knowledge of Irish are looking at it and saying "well, it's strange" because coming at this from English, my gut says it's definitely not English unless it's a very complicated code.

    "H." above said he can't find "O" meaning anything in Gaelic, but I found in an old Irish dictionary as an ablative preposition, (as ó:) another preposition with meanings like "since" and "as," and an archaic/poetic word for "ear" and several metaphorical extensions. And if this is improvised spelling, it could be any number of words rendered as "o," or even be o or oh as in English if he didn't know how to express a certain sentiment in Irish, and so on and so forth ad nauseam.

  56. W.F. said,

    February 2, 2013 @ 2:18 pm

    I just noticed: the 2nd, 6th, and 7th words as I rendered them originally:
    manoiǧa nioanouga conimjroga, and maybe the next word Dinogagaiga, all seem to have an ending of (V)VGa. (meilyga might fit this too, depending on what that odd y-character is)

    Anybody with knowledge of Irish and related languages know a suffix or common ending resembling this? I see -a and -acha as plural endings, but maybe somebody has a better idea?

    There might be a few more patterns: probably the most robust is the mair/fair/(segalei-)noir set.

    Glancing at a list of prepositions, I see le ("with") a few times, the second line, maybe that starts with O + an = On ("from"), but not much else, and nothing looks much like a pronoun.

    And by the by, deil has several dictionary entries, including "separation," beam or pillar, and probably uselessly: female pig two years old.

    Anybody see anything else we can work with?

  57. Nelson said,

    February 2, 2013 @ 5:50 pm

    Following Gearalt's proposal of "abhus" for the second word, I think that the letter some have been interpreting as an insular g in other places is really a lowered b with a dot over it, which might indicate lenited b (previous transcriptions seem to have lacked b). If you look closely it seems to me the g's are more loopy, the b's are sharper and curl up more. Here's my transcription (I add one h for one dot above the letter and 2 h's for two dots, for the character that looks like a tall "i" I write "I", ?'s I have no idea about):

    Ghhardaghhvh abhus lè nioilygha derhaghuigIvr sebhalcinoir O! mair lha shIig?int, moabhhino le ħisgvpaug! Ow racghhar manoighha delghlmIsh, Deil sugcaji niainoubha conimjrogha! ?Qinobhabhaigha mywI fair.

    Ghahn Enghhloith

    Don't know if that will resemble Irish any more or less…

  58. Qaoileann said,

    February 2, 2013 @ 7:03 pm

    I've played with it a little, using the following assumptions:

    1) it is Irish (though not standard Irish as it is used now)
    2) the writer knew some contemporary orthography of Irish, but not much, or at least, not enough to be particularly consistent (this may also be deliberate and/or unremarkable given contemporary English spelling); he may have been a fluent speaker but not literate in Irish.

    Given that, here is a possible partial transliteration of the passage into modern standard Irish, with original transliteration in [..]

    Comhairdegeas agus le míle deis’agaibh [segaleinoir O!] mar laistigh in, [maoighino] le huisce beagh!
    Thú bhfacas mun óige [delglmj'sh], Deil thú go dti [nioanouga conimjroga! Dinogagaiga mywi fair]
    John England

    Any other Irish speakers think that's likely?

    @W.F.: Irish has several noun declensions; an ending like (V)VGa is possible but probably not common http://www.nualeargais.ie/foghlaim/nouns.php

  59. W.F. said,

    February 2, 2013 @ 7:10 pm

    Qaoileann- looks great! Any chance of a translation?

    Nelson- I like your thinking, I took a stab at it too, and I come up with a few things different. I'm not saying I'm right, or that yours is wrong or worse, just what a different set of eyes came up with:

    Ghhardaghhvh abhus lè nieilygha deshaghuigIvr sebhaleinoir O! mair lha shIigtin?, moaghhino le ħisgvhpaug! On racghhar manoighha telghlmIsh, Deil ougeaIi niainoubha conimjrogha! Dinobhabhaigha mywI fair.

    Ghahn Enghhlonith

    All of my changes and reasoning are at the bottom of this post. If anybody notices I missed an h or two, I'm pretty sure I probably did. Anybody think this is any closer to anything?

    The double-dot g is still a bit of a curiosity (so, we have g gh ghh b bh AND w v and vh?), and the character transcribed as I is making me wonder more and more, especially with the dot next to it occupying it's own space, not really being above it, almost like an apostrophe (esp. in telghlmIsh, ougeaIi, and mywI)

    changes:

    *nioilygha: I'd go with nieilygha, his Os are pretty clearly round and that one is so lopsided and has the tail. I thought it started with M, but later on in "conimjrobha" he has that same is-it-NI-or-M thing again, and I'm going to err on the side of caution with NI. (I'll do the same for that one too)
    * derhaguigIvr: long r is an S in gaelic type, so deshaguigIvr
    [also later on racghhar, racghas]
    * sebhalcinoir: c looks like an e to me, (cf. the first e in the preceding word), sebhaleinoir just strikes me as a little tidier
    * shIig?int: did you mean shIigtin? Gaelic type again, I think the first one is the t, and the latter is the odd one.
    * moabhhino: I think the b is a g because of the incline
    * ħisgvpaug: there's a dot over that v, looks like, so ħisgvhpaug,
    * Ow: I just don't think it's "ow" so I put "on"
    * delghlmIsh: telghlmIsh (look at how he writes his Ds, one swoop, like the Gaelic type, this resembles the more tau like t form, as in shIigtin?)
    * sugcaji : I got ougeaIi, his Ss all seem rather clear, the e/c I'm not sure, but same reasoning as in sebhaleinoir
    * ?Qinobhabhaigha: I'd say D is more likely, I see why you suggested it, but it seems so odd to me. Maybe if we figure out that it's a code I'd be more willing to go with it, but for now, I say it's gotta be something else.
    * Enghhloith: I think you left out an N or something between the o and i

  60. blahedo said,

    February 2, 2013 @ 9:07 pm

    A couple of thoughts from someone who knows no Irish but approached this as a possible cipher:

    * There are a few letters in this that are very clearly distinguished in the written form but collapsed in the transcriptions above. The 'a' that is in his name (also the 2nd letter of the inscription) is written differently from the 'a' that resembles a small capital A (see the 5th letter of the inscription, and the 1st of the second word). He uses both forms numerous times and in each case it is very clear which form is used, which implies to me that they are intentionally different rather than in free variation.

    * Also distinct are the letters being transcribed as 'g'. There are several that seem like a classic insular G (as in his last name and in the 6th letter of the inscription, which appear always with one or two dots; there are several that have a vertical stroke with a partial ring below (as in the first letter of his name and of the inscription), which may appear with or without a bar but always with one dot. Both of these two forms appear several times and appear to be made carefully distinct. There are also two places where the letter appears to be more like a script 'z' (end of the first line, middle of the second), with no dot. These are notable as the only places in the inscription that he uses a script loop, which could either mean a lapse into his obviously excellent English calligraphy, or it could indicate a third distinct letter.

    * Further evidence for the hypothesis that it's some sort of (partial) cipher: his handwriting is elegant, flowing, and connected elsewhere in the letter. Even if he were writing in a foreign language, we might expect a certain amount of similarity, given that the alphabet is basically the same (though with some variations, so it wouldn't look identical). But this writing is picked out letter by letter except for just two or three places, just as you'd expect if he had to think through a cipher.

    * The "capital D"s are written very differently as well.

    PS: anyone still looking at the original image should make sure to take a look at the whole-page version posted by Tom Emerson above, even if you're only looking at the transcription—the resolution is higher.

  61. Nelson said,

    February 2, 2013 @ 10:19 pm

    Hmm, looking at the letters further, there is also a slight difference between the "n"s in his name and in the fist line and the more rectangular ones in the second line — some of them are shaped more like pi's than n's.

  62. Eoin P. Ó Murchú said,

    February 2, 2013 @ 10:31 pm

    Hi, finding this very interesting. I am pretty sure it is Irish spelled in a strange way, either to obscure the meaning or due to a lack of understanding of Irish spelling. Working through it with help from previous posts, reading out loud and trying to piece it together, will put work up when makes some sense!

  63. W.F. said,

    February 3, 2013 @ 2:00 am

    Blahedo-

    I remain unconvinced this is a code– all of what you are describing can be easily attributed to other things, and each of those things are at least as likely the cause being that it is a code.

    1) Yes, it appears that there at least two forms of A in this text, but it could easily be either free variation or context-driven variation, or a mix, the kind I imagine is typical of most people. Just look at the writing elsewhere on the letter, the writer doesn't always bar his Fs and Ts, and has tall s and short s right next to one another i.e. "soulless" or "classical" early on.

    2) I don't think his writing is "very clear" in many other instances of this text, and I'm not sure I could confidently distinguish between these As in every instance, for example, the second A in the second line. Even in the English, look at "soulless" again. It took me a moment to realize the dot was actually a stray mark. Lots of weird things happen in writing: "pack of" has a connector.

    3) Even if I could tell exactly which A is which, and given his uneven, up and down writing, and that we can't distinguish other characters clearly, the idea that a code would rely on distinguishing "bent-bar A" from "non-bent bar A" is suspect at best, certainly without something more to go on. (Same goes for the couple instances of dotless i, which at least is a features of Irish writing)

    4) It does appear there are a few different g type characters, but again, natural variation is at least as plausible as separate letter without additional evidence. Consider: the looped "z" occurs at the end of a sentence, a natural place for a flourish for a descending letter, and the next is the only occurrence of a "g" before an e (or c, or anything that looks like it). I can't predict his use of connecting letters perfectly, because the looped letters are the most difficult to identify, but I don't see anything so-left-field as to suggest his looping technique is indicative of code. While the distinction appears robust at first, a closer look at all of the "g" forms will show more of a continuum than a clear division, such as the g in deshaghuigIvr.

    5) His lack of flow could be for numerous reasons: Assuming he's not very literate or confident in his Irish abilities, he will take longer to be sure of his spelling or have to take time figuring out how to render certain words. I'd also assume more the opposite, that using these different forms would interrupt flow quite substantially (I, for one, can't produce a cursive x without stopping, I do it so rarely). And that's assuming he's not doing it on purpose– maybe he thought very cursive Irish would be unreadable for whoever was meant to read it. Simply, there are plenty of innocuous reasons to write slowly.

    6) The "D"s are different (assuming they are both "D"s, I'm not 100% confident), but so are the "O"s, and numerous other letters, he has top-heavy and bottom-heavy "s"s, very-slanted and not-so-slanted "e"s, "i"s with a slight tail and ones without, and so on. Earlier in the letter it says "Southern States" with two very different S forms. Two different capital letter forms is quite possibly meaningless.

    It's good to brainstorm, and I did notice a couple little things I didn't before because of your post (like I think the fourth letter is a T not a D, based on how his other Ts and Ds are; maybe the h in his last name is a barred-h), but this is wild speculation; and even if you are 100% correct in your observations, I don't think anything you said is actually indicative of a code.

    Anyone feel free to correct me, but only three things point to it being a code, and they don't point very stongly:
    1) The double-dot and some other oddities suggest that the author is linguistically clever if not outright playful, and maybe that personality type is more inclined to use a code, as Baertny suggested

    2) Nobody who knows Irish can confidently point to anything in the letter as being Irish.

    3) If this isn't Irish or some closely related language, then it's not the "obvious" answer and therefore it is something unusual (and I guess it's safe to assume that a code is one of the more likely "unusual" things)

    But this is pretty weak evidence– and if you put an English crytogram or other cipher in front of me (one that I couldn't crack by eyeballing, at least), 99 times out of 100, I'd be able to say without a doubt that it's not "just some weird old english or something like that." It's only my gut, but my gut tells me that the objections to this being Irish should be stronger if it's not Irish, or some closely related language (Look how quickly we dismissed Manx as a possibility)

  64. Qaoileann said,

    February 3, 2013 @ 6:48 am

    @W.F.

    A necessarily partial translation would be:

    {Congratulations and with a thousand chance(s) for you (plural)} segaleinoir O! {because[it’s] within}, maoighino {with whiskey!}
    {If you hadn’t seen my youth} delglmj'sh,{ Devil take you to} nioanouga conimjroga! Dinogagaiga mywi fair
    {John England}

    This is based on one of the earlier transliterations posted; with the newer ones I'd have to look at it again, but the method may work.

    @Eoin P. Ó Murchú, I agree, will be very interested to see what you come up with!

  65. Jerry Friedman said,

    February 3, 2013 @ 7:39 pm

    @W. F.: I agree with most of what you say about irregularities in England's writing, but as far as I can tell, he uses long s always and only before short s.

  66. Andrew Carnie said,

    February 4, 2013 @ 10:42 pm

    Could it be Shelta, the language of the travelers in Ireland in the last century? That would account for the cló-Ghaelach script and the Irish elements,but explain why do much of it is not Irish. I don't know any Shelta myself. But it would surprise me if there were travelers in the us at the time and the still spoke it.

  67. W.F. said,

    February 5, 2013 @ 10:38 am

    @Jerry Friedman
    With respect to the 'S's, that would be a context-driven variation, which I described; generally speaking I only meant that form can be very different without any impact on meaning, to dispute the idea that these "clear" distinctions must be (or are even likely to be) meaningful.

    @Andrew Carnie
    I'm in the same boat you are, I know little about Shelta (or even Irish for that matter), if somebody knows something about it, I'd be more enthusiastic about their endorsement of Shelta. I guess the start would be for somebody to point out that certain features of this text are impossible or very atypical of Irish, e.g. words do not start or end with certain patterns, impossible cluster, things of that sort. Even in this context, I still haven't seen anybody come out and suggest any solid reason this isn't or can't be Irish.

  68. Treesong said,

    February 5, 2013 @ 3:50 pm

    conimjroga = Connemara?

  69. Dara said,

    February 5, 2013 @ 4:49 pm

    Quaoileann-I like where you're going there…..
    Treesong-I think the "Conmyroja" is Spanish – "with my red (wine)"
    and
    the woman in question is the sister (deirifiúr-Irish for sister) of "Síle mór"- big Síle…

  70. Tom Emerson said,

    February 5, 2013 @ 6:21 pm

    All – I've had to delete the full-letter image: I was not authorized to post it and did so without EBSCO Publishing or the NYHS consent.

    Obviously now that it's on the Net the truly dedicated will be able to find it, but I've done what I can to limit its distribution.

    My apologies.

  71. Qaoileann said,

    February 5, 2013 @ 7:10 pm

    @Andrew Carnie. I don't think it is Shelta/Cant/Gammon – not because I'm very familiar with it, but because it was – is – used by a community (i.e.: Irish Travellers) who would have been very unlikely to use it with outsiders. It also doesn't have any written tradition I'm aware of at this time (though that could account for the odd orthography).

    It's not outside the realms of possibility, of course – maybe John England had a relative or friend who taught it to him, and as it's a cryptolect it would have been ideal for a "Private" message, but that is also assuming that his correspondent knew it as well. I think either Irish or a cipher is more likely than Cant, but if anyone reading knows more of it than me, please chime in!

RSS feed for comments on this post