Marius,

Again, thank you for your thoughtful reply <…>

To give you more background of the case, you can read

about John Magnuson in The Casebook of Forensic

Detection" by Colin Evans.  I believe you look at

these pages on-line by logging into Amazon.com and

searching the book.  Briefly, the misspelling was on a

hand written address on the bomb packaging.  The guess

of being Swedish was done by "rolling the

pronunciation around on his tongue."  The only Swede

in the community lived nearby & he didn't get along

with the victim. John Magnuson gave a  sample of his

handwriting where he made another similar  misspelling

- this time Marsfild.  There was some other minor

evidence against him.  Some felt that John Magnuson

was innocent.

I agree about your assessments about many of the

fictional detectives.  That's why I tried to have my

stories be based on real forensic deductions.

Sincerely,

Alan

--- Marius Hjelle wrote:

> Hello again, and thanks for taking the time to reply

> to my comment. 

> 

> I’m sorry it’s taken me a while to get back to you,

> but I’ve been through a very busy exam period. What

> prompted me to read and comment on you story was, in

> fact, because I was working on a paper on the

> subject of ’Law and Literature’, and/or Rhetorics. Among

> other things I commented on a few rhetorical deductions

> and ”incriminating” circumstantial evidence that was

> used in a trial I went to look at. The real-life examples

> on our syllabus was (naturally) taken from known

> cases of innocents being found guilty (the english term

> for which has slipped my mind this morning), but seeing

> examples also in the random case I witnessed… well,

> my indignation over your story must be seen in the

> light of this. People go to jail on far less proof than

> what Lestrade has on your Mr. Shield.

> 

> Now, regarding your story and email.. You didn’t

> tell me why the misspelling of the word caught the Yule

> Bomb Killer. Did this point the bomber out as a

> Swede?

> Were there examples of  him writing the word this

> way on other occations not related to the case? I assume

> he was already a suspect, or one of the suspects,

> and that they found other evidence and/or got a

> confession out of him. 

> 

> I see how this phonetic spelling can work if the

> writer is not very fluent in English, in other words

> that his pronounciation of the English word is

> incorrect. Still, this would be more likely with a

> word like Marshfield, as the s when following an r

> gets a little thicker and a little closer to the

> english phoneme of the word (that the s is followed

> by the f only makes this more so). I have a harder time

> believing this might happen at the beginning of a

> word like shield. He would have to pronounce the s the

> way it’s pronounced in ’say’ – and that’s a bit far

> fetched (try saying ’sure’ with that phoneme at the

> beginning). He must not have heard or read much

> English at all, and particulary, he can’t have

> thought about the relation between written and spoken

> English. Not knowing how the name (Shields) is written, he

> must have only heard it spoken.

> 

> But in this case, we have the fact that the victim

> is a typist. I don’t know if (or for how) long he’s

> been employed in England, but – event though it’s said

> that he hasn’t been in the country very long, one must

> assume so. He is, after all, referred to as a

> typewritist, not just a typemachine-owner. Even if

> he hasn’t – would he come to England seeking employment

> without any knowledge of the English language? [I

> remember now that the cryptic message was at the

> bottom of an unfinished office memorandum]One would

> assume that anyone dealing with language in such a

> way must have an above average interest in it. A certain

> sensitivity.

> 

> This being said; the man was, after all, dying. And

> the suspect was seen leaving the building (but what

> kind of building was this? Might there not be a

> hundred reasons for him having been there at the

> time?). I would feel a lot better if there was

> indeed more evidence against him. A little charity for the

> author: It is a circumstantial evidence that points

> them in the right direction, yes. But it isn’t

> enough to hang a man.

> 

> ”Sellds” won’t do, though. He might type that (if he

> was far gone), but Sherlock’s deduction won’t work.

> Let’s leave the S as it is if it’s enough that he

> doesn’t know much about English spelling. If not,

> add a j (or a k). There are very rare occations when the

> s might actually be pronounced sh in norwegian and

> swedish, but you want to let Holmes make an argument

> about this, and so add the j. The E might work if he

> is thinking of how it would be pronounced in the

> English alphabet (otherwise, make it an i). The

> double-L must be the dying man’s typo if it is to be

> left as it is. This would definately make the i into

> a short vowel, as in ”bitter”, in other words – a very

> diffrent phoneme from what he’s trying to get at. DS

> is just fine. Him only having heard the name, you

> could even drop the d, couldn’t you?

> 

> Sjilds / dkoød (heh, I have a Norwegian keyboard,

> with

> æøå) 

> Skils /dlod

> 

> If you can indeed drop the d, I like the idea of

> ’Skils’ very much. Ski is pronounced ’shie’ exactly

> as in Shields, and there being only one L makes it a

> long vowel in the word as a whole, too (in Norwegian, at

> least. I can consult my Swedish friends, but there

> wouldn’t be any problems making him a Noregian). If

> you add the d there, it would definately be a short

> vowel, but like we know – he is dying as he types

> this. If you don’t want to leave this slack, add an

> extra i.

> 

> So, regarding Holmes’ deduction you need this:

> The phoneme SH is written sk or sj in Swedish and

> Norwegian. The phoneme of the e, as it is pronounced in the

> english alphabet, is equal to the i in the Norwegian

> and the Swedish alphabet.

> 

> Sorry for giving you such a hard time about all

> this. If it makes you feel any better I mentioned in my

> essay how it was a good thing that the fictional

> detectives hardly ever had to take their cases to

> court without a confession from the misdoer. Without

> these, both A.C.Doyle himself, or, say,  A.

> Christie, would be guilty of putting innocent men and women

> behind bars. Poirot in particular has a bad habit of

> falling in the trap described by Holmes in at least

> three different stories; 

> ”It is a capital mistake to theorize before one has

> data. Insensibly one begins to twist facts to suit

> theories, instead of theories to suit facts.”

> 

> 

> Sincerely,

> -Marius N. Hjelle

> 

> 

> 

> 

> --- alan downing <acmosh@yahoo.com> wrote:

> 

> > Thank you very much for your feedback.  This

> > deduction was based on an actual 1922 case in

> > England.  John Tyrell deduced who the Yule Bomb

> > killer was based on the Swedish killer's misspelling

> > of  "Marshfield" as "Marsfilld".  The Yule Bomb

> > killer wasn't well educated and was doing phonetic

> > spelling.  In the real case, they appeared to have

> > correctly convicted the guilty party.

> > 

> > In the typewritist, I changed "Marshfield" to

> > "Shield" based on the same deductions.  Do you

> > believe that the Marshfield deduction was correct

> > and the Shield deduction is incorrect?  Do you have

> > other suggested words or names that can produce more

> > accurate deductions?

> > 

> > Of course, what was missing from "The Typewritist"

> > is that Lestrade was later able to find

> > incriminating evidence on Henry Shields that proved

> > beyond any doubt that Sherlock Holmes had the right

> > man!

> > 

> > Thanks,

> > 

> > Alan

> >