Not so many news of Amelia these days, so I would like to come back to this (just two pages above):
Perhaps viewing the operative words vertically (?this might have already been done elsewhere?) would help mediate the bias we sometimes have in attaching sentence structure to potentially disparate words:
- 281
- NORTH
- HOWLAND
- KHAQQ
- NORTH
- MUCH LONGER (seem to go together)
- ABOVE WATER (ditto)
- SHUT OFF (ditto)
Any mathematics probability experts out there?
I think you have it right. This leaves only BEYOND and DONT HOLD WITH US as ambiguous.
I would put the word CALL with KHAQQ in some fashion - CALL IS KHAQQ, ITASCA CALL KHAQQ, etc.
Not a probability expert here, but if I get a moment, I will figure out a couple simple formulas to analyze the remaining ambiguous words. Probably better than just hacking at it with a guess meter. Basically do the same sort of thing as the STENDEC to VALP transformation, but take into account all possibilities - could help figure out what they were trying to send instead of BEYOND, and possibly clean up DONT HOLD WITH US into something that makes better sense.
The good thing about doing it that way is when you type all possibilities, sometimes something you didn't consider immediately pops out.
Interesting Note: Ric's comment about DON'T maybe being WON'T actually is a possibility - when CW ops are very tired, they can accidentally invert the dits and dahs in a letter. D (dah dit dit) is the inverse of W (dit dah dah). This tendency to invert when tired was seen in the early days of commercial marine radio at shore stations where ops were required to work long hours of often tedious code to send; and also in modern Amateur Radio contesting in like the 30th hour of being awake and sending CW.
I completely agree on the idea of a reinterpretation of the "281 N" message on the basis of possible miskeyings of the Morse encoded message. I actually launched about three years ago the hypothesis that what was read as "281 N" could actually have been something else (with OO inside, i.e. twice a triple dash like in 28, and an N at the end, see
A poorly keyed 281 N.
Along the same line, I disagree with the idea that BEYOND and DONT HOLD WITH US are the only words that need be reinterpreted as ambiguous. The trend for the listeners of Morse code to write down the wrong word may be even stronger for those words that seem familiar. Think of HOWLAND and the number of messages these operators would have heard telling "ARRIVED AT HOWLAND", "EXPECTED AT HOWLAND", "NOT SEEN AT HOWLAND" and so on. Whatever sounded more or less like HOWLAND was very likely to be caught like "Oh, a bit miskeyed, but they are still telling about HOWLAND" ... So in the sequence "281 N HOWLAND", HOWLAND may be as questionable as the senseless 281 N we have desperately tried to link to a latitude, even though it points to the wrong hemisphere.
So what ? There was a lot of discussion recently on whether the idea that Fred was injured in the landing was more than a hypothesis and something that could be tracked in the plausible post-loss radio calls. If so, the question must be raised whether one can detect something in the "281 N" message too. My first suggestion, relying on the observation that "281 N" may be heard as a result from a faulty transmission of NOONAN, was that we had Fred's signature here, and that he made something like a fatal error by miskeying his own name. But everybody more or less agrees that Amelia was the one at the microphone and this is actually more likely. So what could she mean sending Fred's name on the air ? If Fred was injured, wasn't it straightforward to dial NOONAN INJURED ?
I won't explain again how easily NOONAN can be miskeyed and read as 281 N (please see the other thread). Let us only do the test : what can INJURED become, with not too many transformations. INJURED: .. -. .--- ..- .-. . -.. With only two switches, one erased dot and two compactions: .... --- .-- .-.. -.. = HOWLD. May be we are, as usual, lead by what we want to find, but I did not think, when I started the test, that it would work so easily.
If you do not like it please take it only as a winter tale. The probability certainly remains less than 50% that the 281 N HOWLAND message actually was a NOONAN INJURED distress call. On the other hand, let us not forget that the purloined letter may also lie among objects assumed of no importance.
Best wishes to Niku VIII.
Christophe