ERDNASE

Discuss general aspects of Genii.
Richard Hatch
Posts: 1857
Joined: January 17th, 2008, 12:00 pm
Location: Providence, Utah
Contact:

Re: ERDNASE

Postby Richard Hatch » November 2nd, 2017, 6:43 pm

Brad Henderson wrote:could charlies be the result of someone who had heard it pronounced (char-lee-ais) but just didn't know how to spell it correctly? (char-li-es)

perhaps our earth nose doesn't know how to properly parse parley vous?


Seems to be a simple typo, as he spells it correctly on page 192 of the original edition (in the "Acrobatic Jacks" trick explanation). Curiously, in the original mention, he says "This is known to conjurers as the "Charlies [sic] Pass" and we presume was invented by the famous magician of that name." (p. 128 of the original edition), but in the page 192 mention, he calls it the "Charlier Shift", combining the conjuring and gambling nomenclature.

Leonard Hevia
Posts: 1441
Joined: January 17th, 2008, 12:00 pm
Favorite Magician: Dai Vernon, Frank Garcia, Slydini, Houdini,
Location: Gaithersburg, Md.

Re: ERDNASE

Postby Leonard Hevia » November 2nd, 2017, 10:10 pm

Richard Hatch wrote:Seems to be a simple typo, as he spells it correctly on page 192 of the original edition (in the "Acrobatic Jacks" trick explanation).


Is it a typo where the typesetter accidentally substituted the "s" for the "r" or did the typesetter misread the manuscript and deliberately used the "s" because that is what it looked like to him on the page?

Richard Hatch
Posts: 1857
Joined: January 17th, 2008, 12:00 pm
Location: Providence, Utah
Contact:

Re: ERDNASE

Postby Richard Hatch » November 2nd, 2017, 10:54 pm

Leonard Hevia wrote:
Richard Hatch wrote:Seems to be a simple typo, as he spells it correctly on page 192 of the original edition (in the "Acrobatic Jacks" trick explanation).


Is it a typo where the typesetter accidentally substituted the "s" for the "r" or did the typesetter misread the manuscript and deliberately used the "s" because that is what it looked like to him on the page?

Who can say at this point? But if one is claiming that the typesetter made the error by misreading the author’s handwritten manuscript, because his “r” looked liked an “s”, it seems likely he would have made the same mistake on page 192, unless we have a different typesetter, or different handwriting.
I find it more curious that Erdnase characterizes Charlier as a “famous magician”. I don’t think he was ever famous, or even well known, outside a small circle of magic aficionados in London. To mischaracterize him that way seems to indicate that Erdnase was not part of the magic community, but someone who might have gotten that impression about Charlier from reading about that sleight.

User avatar
magicam
Posts: 861
Joined: January 28th, 2009, 8:40 pm

Re: ERDNASE

Postby magicam » November 3rd, 2017, 8:59 am

^^^ Or (if Erdnase's text was handwritten) a clearer "r" in the second reference, or both references were typeset as "Charlies Pass" but only the typo on p. 192 was caught in the proofreading stage? As you say, who knows! :)

Topics are often inadvertently "recycled" in this humongous thread, so my apologies in advance if this is a rehash observation/query. But to Dick's point about the "famous magician" statement, Erdnase's reference reads as follows:

    This is known to conjurers as the “Charlies Pass,” and we presume was invented by the famous magician of that name.
The wording of that passage allows the inference that Erdnase's knowledge of the pass and Charlier himself were from different sources. In other words, he learned the pass mechanics simply as the "Charlier Pass" (with no further information on who or what Charlier was), and then assumed it was named after Charlier based on his subsequent (or perhaps preexisting?) knowledge about Charlier the man and his prowess with magic.

Would this help to pinpoint Erdnase's sources? Many moons ago on this thread, the books listed below were identified as containing references to Charlier (there may be others that I missed). Do any of those books describe the "Charlier pass" mechanics without providing any other information on Charlier? Which of those books are especially praiseworthy of Charlier's conjuring skills? Etc.

    Robert-Houdin's Secrets of Conjuring and Magic (1878)
    Hoffmann's More Magic (1889)
    Bertram's Isn't it Wonderful? (1896)
    Roterberg's New Era Card Tricks (1897)
    Howard Thurston's Card Tricks (1901)
    Hoffmann's Tricks with Cards (1889) was also mentioned, but perhaps this was an excerpt from More Magic?
The foregoing assumes that Erdnase's knowledge about Charlier and his pass was from magic books; as Dick noted earlier in the thread, Erdnase does mention learning from the "literature of conjurers." But it doesn't seem inconceivable that Erdnase's knowledge came from a person instead of books (which might also explain his misperception about Charlier's fame as a magician?).

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 3rd, 2017, 10:36 am

Leonard Hevia wrote:Is it a typo where the typesetter accidentally substituted the "s" for the "r" or did the typesetter misread the manuscript and deliberately used the "s" because that is what it looked like to him on the page?

Clearly it was an accidental substitution, because if it was based on the authors handwriting we would need to see several other instances where an s has been erroneously substituted for an r. I think it is simply a typo and it doesn't really matter who introduced it, the author or anybody who may have typed up a handwritten manuscript, or the typesetter.

I think it is pretty clear from where Erdnase has the Charlier information. He has it from Hoffmann's "More Magic". Hoffmann names Charlier 16 times in this book and he refers to him as: "the venerable wizard", "Professor Charlier", "Charlier System of Card Marking" - which he describes as "complete and admirable system" (perhaps that is also where Erdnase has his "Erdnase system of ..." which he uses several times) Keep in mind that Erdnase uses in his book a rather peculiar expression from "More Magic". In the description of the cards up the sleeve (Erdnase calls them the Traveling Cards) Hoffmann describes the trick depending on "dextrous card-palming supplemented by unflinching audacity" Erdnase almost replicates the sentence saying it depends on "Masterly feats of Palming and Unflinching Audacity". That is proof that he read "More Magic".

Since Erdnase read "More Magic" he most likely also read "Tricks with Cards" which is not simply an excerpt of the card sections of his other books. There he mentions Charlier 17 times and writes "the most skilful card conjurer I have ever met with, M. Charlier,..."
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 3rd, 2017, 10:44 am

Recall also the speculation that the magician who traveled in America under the names "Carabaraba" and "St. Jean" was in fact Charlier. For reasons I should write up some time, I don't believe this magician was Charlier. But if he was, Erdnase could have met him and learned the sleights directly from him.

More Magic is online, if you want to read it.

Tom Sawyer
Posts: 307
Joined: January 7th, 2012, 6:44 pm

Re: ERDNASE

Postby Tom Sawyer » November 3rd, 2017, 4:19 pm

Hi All,

This post has to do with the s-for-r substitution.

Chris above indicates that it is clear that the s-for-r "was an accidental substitution." The reason stated was that if it were NOT accidental, there would be other examples within the book.

That might be so, but if it is, it appears to me that it such examples would have to be those rare instances in which an "r" can be substituted for an "s" and still make perfect sense.

You might say, "Well, without an apostrophe, 'Charlies' does not make sense." But to a person who has never heard of Charlier, "Charlies" would look relatively good.

Then you have Clay's "type case" argument, that (depending on what case or cases were used) the lower-case "r" is pretty far from the lower-case "s." If you want to explore "type-case lays," the following might be a good place to start: "Type Cases, David Bolton."

Some arrangements were undoubtedly more popular than others, but in the ones I looked at, the "s" was generally far from the "r." This is consistent with what Clay said.

In thinking again about the "Charlies" situation, one of the questions is, "How important is that argument, one way or another?"

The s-for-r substitution, as discussed in Marty Demarest's Genii article, has always seemed to me to be one of the very good arguments favoring W.E. Sanders. It's only a tiny part of the picture, of course, but when you throw in the Dalrymple argument (re the political cartoon) and the Del Adelphia argument, and many others, the Sanders case remains one of the strongest cases overall.

--Tom Sawyer

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 3rd, 2017, 5:30 pm

Speaking of Dalrymple, Heritage Auctions currently has a piece of his art, used in Puck, up for auction.

Some of his art resonates even today.

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 3rd, 2017, 5:56 pm

Tom Sawyer wrote: but when you throw in the Dalrymple argument (re the political cartoon)


Tom, from your work on the Godwin blog, I know you are a student of illustration. Don't you think (as I do) that the "Montana" character in the Doomed cartoon from Puck is merely a personification of the mining interests of the Western states? I don't think it is intended to represent Sanders, any more than the Washington, Idaho, or Dakota characters are meant to represent real politicians from those places.

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 3rd, 2017, 7:46 pm

A comment about comparing writing style. On Tom Sawyer's blog in a recent post he wrote about Oscar Teale and his book "Higher Magic". https://erdnasequest.wordpress.com/2017 ... all-cases/ He mentioned for example that "Higher Magic" also has occurrences of '(?)'. I guess his argument is that even though Oscar Teale is very likely not Erdnase, the occurrence of a linguistic feature doesn't mean he is.

However, when we take a more careful quantitative look it turns out this actually speaks against Teale, even though he has '(?)' in his text. Here is the data:

- Erdnase: (?) appears 3 times in a text of 52k words.
- Teale: (?) appears 58 times in a text of 85k words.
- Gallaway: (?) appears 0 times in a text of 30k words.

As Mosteller-Wallace have argued words can essentially be modeled statistically as a Poisson process. So for Erdnase the average (lambda) is '3 per 52k words'. If we plug these numbers into a Poisson distribution (k=3, lambda=3) we get as probability 0.224. Now lets scale lambda to the longer Teale book. 3 * 85/52 = 4.9. So for the Teale book we would on average expect about 5 of the (?). But we see 58! Plugging this into a Poisson process comes out to 3e-41 or in other words essentially zero probability. If we scale the lambda to 30k for Gallaway, 3 * 30/52 = 1.73 and then we plug this into Poisson (k=0, lambda=1.73) we get a probability of 0.177. In summary the likelihoods are:

Erdnase: 0.224
Teale: 0
Gallaway: 0.177

In other words, given the observations of "(?)" in Erdnase the fact that in another 30k words we do not find any (Gallaway) is not particularly surprising or statistically unexpected. However, it would be highly abnormal to see 58 of them.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Tom Sawyer
Posts: 307
Joined: January 7th, 2012, 6:44 pm

Re: ERDNASE

Postby Tom Sawyer » November 3rd, 2017, 9:53 pm

Bill:

That is a difficult question.

I actually think there is a decent chance that it is supposed to represent W.F. Sanders. But see below.

In Rethinking S.W. Erdnase, I said:

Personally, I think it is supposed to be Wilbur Fisk Sanders, for various reasons. To me, it looks pretty much like him. And the flag pitched nearby says: “Republican Senatorial Battle Ground.” And Dalrymple plainly did not show the faces of two people to the right of the Montana man. If Dalrymple didn’t want people to think it was Wilbur Fisk Sanders in the Montana hat, he did not need to show a face.

Of course, responses to that could include the following. I think at least a couple of these have been discussed elsewhere on this thread:

    Essentially zero people would have recognized the elder Sanders, so why would Dalrymple have depicted him?

    Several of the people are labelled with their names. If Dalrymple had wanted to picture Sanders, he would have labeled Sanders.

    The guy does kind of look like a stereotypical grizzled-miner.

    And looking at the image as an example of art, the Montana man (with several others) is definitely in the shadows. This tends to support the idea that the precise identity of any of the people in that section of the picture was not important.

Overall, it looks that the case for that cartoon depicting Sanders’s dad is somewhat weak.

If the cartoon does not depict the elder Sanders, then the Dalrymple connection of the younger Sanders with Dalrymple is pretty weak, but perhaps strong enough to meet the "related" requirement. At some point, it becomes pretty subjective.

As a matter of fact, this falls into the category of a fact that is used to support a candidate, when the original fact itself has not been well-established. Of course, there are tons of "facts" in the Erdnase arena that fall into that category.

I guess this is a good place to mention that when I proved that my own daughter was Erdnase, part of the evidence was as follows:

As to that particular Puck cartoon by Louis Dalrymple, well, she is not wearing a hat with the name of her state. But her actual name — Sawyer — is plainly spelled out on the page, as a “shout out” to her from Dalrymple.

And for those not familiar with the cartoon, the name Sawyer is there, no joke.

--Tom Sawyer

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 3rd, 2017, 11:16 pm

lybrary wrote: As Mosteller-Wallace have argued words can essentially be modeled statistically as a Poisson process.

In Teale and in Erdnase, "(?)" is not a word, but is punctuation used to indicate that the preceding word is being used ironically. Why do you proceed as if a statistic of word usage would have any relevance at all?

To me, the takeaway is that both Erdnase and Teale use the same technique to denote irony, and Gallaway does not use it; thus, Erdnase is more like Teale than he is like Gallaway. Yes, Teale uses it much more than Erdnase on a per-word basis, but that may be a function of topic. "(?)" may be highly context-dependent. Regardless, the usage rate of Erdnase (5.8e-05) is much closer to that of Teale (6.8e-04) than it is to Gallaway (0.0) when viewed on a logarithmic scale, which is what you should have used instead of a linear scale.

Tom -- I agree there is a chance that the grizzled Montana character was meant to depict Sanders Sr. I think there is a much greater chance that it wasn't, though. Marty did a lot of good research in his Genii article and the follow up in Montana magazine, but this is one of the weaker links in the chain, and I don't think it strengthens his case.

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 4th, 2017, 5:56 am

Bill Mullins wrote:
lybrary wrote: As Mosteller-Wallace have argued words can essentially be modeled statistically as a Poisson process.

In Teale and in Erdnase, "(?)" is not a word, but is punctuation used to indicate that the preceding word is being used ironically. Why do you proceed as if a statistic of word usage would have any relevance at all?

The same statistical model applies to punctuation. I have actually plotted the probability distribution of many function words and punctuation from Erdnase, and they can all be surprisingly accurately be modeled with a Poisson process. I think Mosteller-Wallace were spot on with that observation.

Bill Mullins wrote:To me, the takeaway is that both Erdnase and Teale use the same technique to denote irony, and Gallaway does not use it; thus, Erdnase is more like Teale than he is like Gallaway. Yes, Teale uses it much more than Erdnase on a per-word basis, but that may be a function of topic. "(?)" may be highly context-dependent. Regardless, the usage rate of Erdnase (5.8e-05) is much closer to that of Teale (6.8e-04) than it is to Gallaway (0.0) when viewed on a logarithmic scale, which is what you should have used instead of a linear scale.

Then you are ignoring math. The takeaway is that using so many (?) in a magic book is much less likely Erdnasian than using none of them. That is exactly the Mosteller-Wallace framework I am applying here. I recommend that people read their article which was for good reason a milestone in stylometry.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Leonard Hevia
Posts: 1441
Joined: January 17th, 2008, 12:00 pm
Favorite Magician: Dai Vernon, Frank Garcia, Slydini, Houdini,
Location: Gaithersburg, Md.

Re: ERDNASE

Postby Leonard Hevia » November 4th, 2017, 11:11 am

Bill Mullins wrote:
Tom Sawyer wrote: but when you throw in the Dalrymple argument (re the political cartoon)


Tom, from your work on the Godwin blog, I know you are a student of illustration. Don't you think (as I do) that the "Montana" character in the Doomed cartoon from Puck is merely a personification of the mining interests of the Western states? I don't think it is intended to represent Sanders, any more than the Washington, Idaho, or Dakota characters are meant to represent real politicians from those places.


I disagree here, those are the real politicians we see in the cartoon. The Dalrymple Puck cartoon that Bill linked showed a close up and not a complete panorama of it. The reproduction in that Genii issue depicts the whole illustration. Looking at the cartoon with a magnifying glass, I can see a flag in the center next to a tent. There are words on the flag that say:

Republican Senatorial Battleground

Those are republican senators we are looking at.

Roger M.
Posts: 1244
Joined: January 17th, 2008, 12:00 pm

Re: ERDNASE

Postby Roger M. » November 4th, 2017, 3:24 pm

Dalrymple makes it very clear to the viewer that:

    If the character has a name - it's a person.
    If the character doesn't have a name - it's not a person, it's a "thing" or a "situation".

Such a practice is still extremely common in political cartoons in 2017.

Leonard Hevia
Posts: 1441
Joined: January 17th, 2008, 12:00 pm
Favorite Magician: Dai Vernon, Frank Garcia, Slydini, Houdini,
Location: Gaithersburg, Md.

Re: ERDNASE

Postby Leonard Hevia » November 4th, 2017, 3:49 pm

Roger M. wrote:Dalrymple makes it very clear to the viewer that:

    If the character has a name - it's a person.
    If the character doesn't have a name - it's not a person, it's a "thing" or a "situation".

Such a practice is still extremely common in political cartoons in 2017.


Yes--the men on the front lines of that battlefield may not be actual senators and instead are symbolic representatives of their respective states. But that may not have stopped W.E. Sanders from looking at it and believe the fighter from Montana was his father. The individual representing Montana does look like Wilbur Fisk Sanders, a Republican senator, and there are Republican senators on that battlefield.

Tom Sawyer
Posts: 307
Joined: January 7th, 2012, 6:44 pm

Re: ERDNASE

Postby Tom Sawyer » November 4th, 2017, 10:09 pm

Leo, that (in my opinion) is an excellent point on the Dalrymple situation. I had forgotten that argument (that Sanders could well have thought the picture depicted his dad, whether it did or not), but if I had remembered it, I would have mentioned it in my post above on this issue.

But when you get down to it, what one really wants is a nice, solid, close relationship, not a relationship that one has to do any "explaining" on. That Dalrymple-cartoon argument needs quite a bit of explaining.

On the other hand, I think that the Edwin Sumner Andrews argument regarding Dalrymple ALSO requires a moderate amount of explaining, and as I sit here, I have no real comprehension of how close or how far the "third cousin, once removed" is in the ESA case. It sounds sorta close, and it seems kinda close, but it is not obvious to me that it really IS close.

Overall, I think ESA's arguments re Dalrymple are better than WSA's.

--Tom Sawyer

Leonard Hevia
Posts: 1441
Joined: January 17th, 2008, 12:00 pm
Favorite Magician: Dai Vernon, Frank Garcia, Slydini, Houdini,
Location: Gaithersburg, Md.

Re: ERDNASE

Postby Leonard Hevia » November 4th, 2017, 10:33 pm

Thank you Tom!

We don't know what Erdnase exactly told Smith. If we take the family connection at face value, then researchers look for marital or blood ties. If Erdnase had said "I have a family connection to Dalrymple," then Smith could have taken it to mean that Dalrymple was a blood relative. If Sanders was Erdnase, he was not going to tell Smith that Dalrymple drew his father in Puck magazine. That would have been giving away too much personal information.

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 5th, 2017, 1:32 am

lybrary wrote:However, when we take a more careful quantitative look it turns out this actually speaks against Teale, even though he has '(?)' in his text. Here is the data:

- Erdnase: (?) appears 3 times in a text of 52k words.
- Teale: (?) appears 58 times in a text of 85k words.
- Gallaway: (?) appears 0 times in a text of 30k words.

As Mosteller-Wallace have argued words can essentially be modeled statistically as a Poisson process. So for Erdnase the average (lambda) is '3 per 52k words'. If we plug these numbers into a Poisson distribution (k=3, lambda=3) we get as probability 0.224. Now lets scale lambda to the longer Teale book. 3 * 85/52 = 4.9. So for the Teale book we would on average expect about 5 of the (?). But we see 58! Plugging this into a Poisson process comes out to 3e-41 or in other words essentially zero probability. If we scale the lambda to 30k for Gallaway, 3 * 30/52 = 1.73 and then we plug this into Poisson (k=0, lambda=1.73) we get a probability of 0.177. In summary the likelihoods are:

Erdnase: 0.224
Teale: 0
Gallaway: 0.177

In other words, given the observations of "(?)" in Erdnase the fact that in another 30k words we do not find any (Gallaway) is not particularly surprising or statistically unexpected.


You're doing it wrong.

The only way you can compare the relative probabilities of features in Erdnase and Gallaway's writing like this is if you assume a priori that the distribution of "(?)" in Gallaway's writing is dependent on the distribution of "(?)" in Erdnase's writing. The probability of "(?)" showing up 0 times in Gallaway's writing is 0.177 if and only if Gallaway writes (statistically) like Erdnase. But that is the question being investigated, so you can't assume it as an axiom.

When you calculate the PGallaway = 0.177, you are using λErdnase, when you should be using λGallaway. But we don't know what λ is for Gallaway, so it's foolish to make any kind of quantitative analysis. But the statement "Erdnase and Teale both use "(?)" and are therefore alike in that respect" is a qualitative comparison, and is valid.

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 5th, 2017, 1:03 am

The Dalrymple cartoon has, on the left side, named depictions of Republican senators who, for various reasons, supported the Free Silver movement. They were allied with Western mining interests who profited from the mining of silver. I don't think there is any great reason to assume that the image of the Montana miner was meant to depict, overtly or otherwise, W. F. Sanders.

Consider the images of the senators, and how accurate they are:

Frank Hiscock
Image

George Hoar
Image

Philetus Sawyer
Image

Matthew Quay
Image

Dalrymple is a pretty good cartoonist -- these are excellent likenesses.

Now look at the Montana miner, and compare it to Wilbur Fisk Sanders:

Image

There is a superficial resemblance, but the cartoon is not a likeness of Sanders. The nose in the cartoon is too hooked, and the eyebrows are too bushy. The miner’s brow ridge is more pronounced than that of Sanders, his eyes are deeper-set, and his lower jaw is recessed compared to that of Sanders. If Dalrymple had wanted to make it look like Sanders, it would have looked just like him.

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 5th, 2017, 5:56 am

Bill Mullins wrote:You're doing it wrong.

You should tell this Mosteller-Wallace and the entire stylometry community who has followed their lead. Bill Mullins the know-it-all.
When one starts with Erdnase and asks how likely is it that Erdnase is Gallaway, my calculation, which follows Mosteller-Wallace, is perfectly fine.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 5th, 2017, 8:17 am

Great juxtaposition of caricatures and photos. To me the Montana fellow looks like Wilbur Fisk Sanders. Dalrymple could have drawn the Montana guy in many other ways, but he didn't. However, the part of the argument I can't follow is this. Why would WE Sanders say he is related to Dalrymple when Dalrymple merely drew his father? Just because somebody draws somebody else doesn't mean they are related.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 5th, 2017, 10:17 am

lybrary wrote:
Bill Mullins wrote:You're doing it wrong.

You should tell this Mosteller-Wallace and the entire stylometry community who has followed their lead.


I'm not saying that Mosteller and Wallace were wrong. I'm saying that you are wrong. The data about "(?)" doesn't support the conclusion you are drawing.

Leonard Hevia
Posts: 1441
Joined: January 17th, 2008, 12:00 pm
Favorite Magician: Dai Vernon, Frank Garcia, Slydini, Houdini,
Location: Gaithersburg, Md.

Re: ERDNASE

Postby Leonard Hevia » November 5th, 2017, 10:18 am

lybrary wrote:Great juxtaposition of caricatures and photos. To me the Montana fellow looks like Wilbur Fisk Sanders. Dalrymple could have drawn the Montana guy in many other ways, but he didn't. However, the part of the argument I can't follow is this. Why would WE Sanders say he is related to Dalrymple when Dalrymple merely drew his father? Just because somebody draws somebody else doesn't mean they are related.


Chris--did you read my post directly above? We can't be sure that Erdnase told Smith that he was actually related to Dalrymple. He could have merely told Smith that he had a family connection to Dalrymple. Smith could have taken that at face value to mean that Dalrymple was a blood relative. The Dalrymple subject was most likely mentioned in passing as the two men were conducting business with the sketches and not the focus of their discussions.

Bill Mullins wrote:There is a superficial resemblance, but the cartoon is not a likeness of Sanders. The nose in the cartoon is too hooked, and the eyebrows are too bushy. The miner’s brow ridge is more pronounced than that of Sanders, his eyes are deeper-set, and his lower jaw is recessed compared to that of Sanders. If Dalrymple had wanted to make it look like Sanders, it would have looked just like him.


Great images Bill! Agreed that the image of the Montana miner is only a superficial resemblance, but it's close enough that Sanders could have interpreted that character as a caricature of his father. If the character of the Montana miner was a heavyset and clean shaven fellow, then it is sensible to say that Sanders would not have confused that character with his dad. Political cartoons and comic book illustrations are stylized art and not meant to be lifelike renditions.

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 5th, 2017, 10:30 am

Bill Mullins wrote:I'm not saying that Mosteller and Wallace were wrong. I'm saying that you are wrong. The data about "(?)" doesn't support the conclusion you are drawing.

I am doing it exactly as Mosteller-Wallace. So yes, you are saying they are wrong. You seem to not understand simple conditional probabilities. Given the observation of 3 in Erdnase, and the Poisson distribution, we are asking what is the probability of the observation of 0 in another text (Gallaway). That calculates to 0.177. Then we are asking what is the probability to observe 58 (Teale) and it comes out as essentially zero. Simple conditional probabilities. It is not that hard Bill. You can do it.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 5th, 2017, 10:37 am

Leonard Hevia wrote:We can't be sure that Erdnase told Smith that he was actually related to Dalrymple.

Oh, I see. This argument requires for Smith to have incorrectly remembered something. What a novel thought! I will not argue against it. But I find the alleged comment rather unlikely in this situation. If Dalrymple would have actually drawn a nice full page portrait of his father then I guess you would have a point. But here his father was depicted as a tiny part of a larger image. It is silly to think that this would have been subject of a conversation.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Richard Hatch
Posts: 1857
Joined: January 17th, 2008, 12:00 pm
Location: Providence, Utah
Contact:

Re: ERDNASE

Postby Richard Hatch » November 6th, 2017, 12:41 pm

Looks like this Erdnase kickstarter campaign will meet its goal. Thoughts?
https://www.kickstarter.com/projects/110191763/the-expert-at-the-card-table-with-photographs-from

User avatar
Richard Kaufman
Posts: 23864
Joined: July 18th, 2001, 12:00 pm
Favorite Magician: Theodore DeLand
Location: Washington DC
Contact:

Re: ERDNASE

Postby Richard Kaufman » November 6th, 2017, 12:45 pm

Selling yet another copy of the original text with no new text ... one can only ask why.

If you're going to add photographs, why not add 200 (certainly a more reasonable number) instead of 100?

$200?
Subscribe today to Genii Magazine

Richard Hatch
Posts: 1857
Joined: January 17th, 2008, 12:00 pm
Location: Providence, Utah
Contact:

Re: ERDNASE

Postby Richard Hatch » November 6th, 2017, 1:21 pm

Richard Kaufman wrote:Selling yet another copy of the original text with no new text ... one can only ask why.

Apparently it includes some new text: "an introduction by Joe Crist". " In the introduction, Mr. Crist will disclose one of the most closely guarded revelations in the history of playing cards. Known by just a handful of card experts worldwide, the 180 year old secret of the practice board is lauded as the greatest tool in developing "overall card handling ability" of all time."

User avatar
Richard Kaufman
Posts: 23864
Joined: July 18th, 2001, 12:00 pm
Favorite Magician: Theodore DeLand
Location: Washington DC
Contact:

Re: ERDNASE

Postby Richard Kaufman » November 6th, 2017, 2:15 pm

Yes, I did note that. But it has nothing to do with Erdnase.

So is this really about charging people $200 for the secret of "The Practice Board"?
Subscribe today to Genii Magazine

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 6th, 2017, 3:01 pm

Leonard Hevia wrote: Political cartoons . . . are stylized art and not meant to be lifelike renditions.
Except in this case, the cartoons of real people are lifelike renditions. (add to the list John James Ingalls, who is sprawled on the ground in front of the cannons.)

Dalrymple did accurate likenesses when he was depicting real people. "Montana" is not an accurate likeness of Sanders, so it wasn't intended to be W. F. Sanders.

W. E. Sanders presumably knew what his father looked like. He would not have perceived "Montana" to have been his father, and there's no reason to think that a decade later, he would have said to Smith that the cartoon referred to his father.

The rest of this post is math-heavy, boring, and probably should be skipped. TL;DR: I disagree (once again) with Chris.

lybrary wrote:I am doing it exactly as Mosteller-Wallace.
Not at all.

What MW did (paper online here) was examine several of the Federalist papers whose authorship was unknown, but presumed to be either Madison or Hamilton, and compared them to works which were known to have been written by Madison, and by Hamilton.

So MW had disputed works, and works of known authorship. We have a disputed work (Expert), and works of known authorship for comparison (Gallaway and Teale). So far, so good.

MW did this:
1. Derived a list of function words that would be used in the analysis. Cross-checked and tested that list to ensure it was valid.
2. Measured the usage rates of these words in the disputed papers (individually), and in the collected works of Madison, and the collected works of Hamilton.
3. Calculated the probability (for the words in the list, as they appear in each of the disputed papers) that an author with Madison's "native" usage rates would have used the word as is was used in each disputed paper, and then did likewise for Hamilton.
4. Look for patterns, draw conclusions.

You did this:
1. Used a single punctuation marker instead of a list of words. Did no tests or cross-checks to determine if this marker is subject to the same statistical patterns as words are, or if it is otherwise valid.
2. Measured the usage rate of the marker in Erdnase (3/52k), Teale (58/85k), and Gallaway (0/30k).
3. Calculated the probability that Teale would have used the marker 58 times if he had Erdnase's native usage rate, and the probability that Gallaway would have used it 0 times if he had Erdnase's native usage rate. (see where you did it differently?)
4. Draw conclusions.

You have worked the problem from the wrong direction. MW calculated to what extent the unknown author wrote like Madison and Hamilton; you calculated to what extent Teale and Gallaway wrote like the unknown author.
You used λErdnase = 3, and then for Teale and Gallaway, you scaled it so λTeale = 3*85/52 = 4.9 and λGallaway = 3*30/52 = 1.72 (that should be 1.73, BTW).
You calculated
P(58)Teale using λErdnase scaled upwards ≈ 0; and
P(0)Gallaway using λErdnase scaled downwards = 0.177.

If you had approached the problem like MW, you would have calculated
P(3)Erdnase using λTeale (= 58*52/85 = 35.4) = 3.12e-12; and
P(3)Erdnase using λGallaway (= 0 ) = 0.

So, the probability that a person who uses "(?)" 58 times in 85k words (Teale) would use it 3 times in 52k words is small: 3 parts in 10^12.
The probability that a person who never uses "(?)" (Gallaway) would use it 3 times in 52k words is nonexistent.

The takeaways should be:
1. Using "(?)" as a measure of stylistic similarity is unverified, and doesn't really get you anywhere. MW spent a large portion of their paper (pp. 279-286) just determining what markers to use, and how much to weight each of them. Nothing like that was done with "(?)"; we just used it because it stood out to a casual reader.
2. If it is a valid marker to use, however, the probability (based solely on the usage rates of "(?)") that either Teale or Gallaway wrote Expert is vanishingly small, and zero.
3. But in those extremes, Teale (who uses the marker) is more likely than Gallaway (who doesn't use the marker) to have written a book that occasionally uses the marker (Expert). Teale beats Gallaway.

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 6th, 2017, 3:21 pm

One can ask the question of authorship in either direction. You can start with Erdnase and then ask how likely is it that he would have authored another text. Or you can start with a candidate and ask how likely did he write Expert. Both answer the same question - Who is Erdnase? - and both fall within the Mosteller-Wallace framework. Since the Poisson distribution is not defined for lambda = 0 one has to start with Erdnase in the case of (?), otherwise you cannot answer that question.
Bill Mullins wrote:The probability that a person who never uses "(?)" (Gallaway) would use it 3 times in 52k words is nonexistent.

We do not know if Gallaway never used it, because we do not have all of Gallaway's writings. We only have a sample of his writing. That is exactly why starting from Erdnase and asking how likely is he the author of Estimating makes a lot more sense than asking it the other way around where one cannot use that method of analysis.

Bill Mullins wrote:Using "(?)" as a measure of stylistic similarity is unverified, and doesn't really get you anywhere.

I have simply pointed out why I disagree with Tom Sawyer's argumentation. I have not used (?) in my own linguistic analysis. However, punctuation does follow a Poisson distribution, as do most function words. And some stylometrists do use punctuation as a stylistic feature.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Bill Mullins
Posts: 4501
Joined: January 17th, 2008, 12:00 pm
Location: Huntsville, AL

Re: ERDNASE

Postby Bill Mullins » November 6th, 2017, 5:51 pm

lybrary wrote:One can ask the question of authorship in either direction.

You said that you did it exactly like MW, when in fact what you did was opposite to MW. That's what I was showing.

You can start with Erdnase and then ask how likely is it that he would have authored another text. Or you can start with a candidate and ask how likely did he write Expert.

But the standard in author discrimination is to compare the unknown text to each of the candidates. You can't just invert that and assert that you've answered the same question.

Can you point to any respected studies where the author(s) have done it in your order?

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 6th, 2017, 6:28 pm

Bill Mullins wrote:You said that you did it exactly like MW, when in fact what you did was opposite to MW. That's what I was showing.

That is like saying they wrote a+b so it can't be used as b+a. Of course it can be used that way, too, because it is equivalent. The authorship question is symmetric. Saying "A has likely written B" obviously and trivially also means "B likely wrote A". The MoWa paper does not say you must start at the known texts and then calculate the probability to the unknown. It is simply a method to compare two texts regardless of who is the unknown. (see further down for more on that)

Bill Mullins wrote:
You can start with Erdnase and then ask how likely is it that he would have authored another text. Or you can start with a candidate and ask how likely did he write Expert.

But the standard in author discrimination is to compare the unknown text to each of the candidates. You can't just invert that and assert that you've answered the same question.

You still haven't understood the MoWa paper. You can apply the math but you don't comprehend what it actually does. The fact that you think applying it the other way around is wrong demonstrates that you do not understand what MoWa actually calculates.

The reason why MoWa started from the known texts by Hamilton and Madison and then calculated the probability to the various disputed papers, is the difference in text volume. The average length of a disputed article is about 3000 words. The known text volume for each author was a lot more. In this situation it makes a lot more sense to start with the known text, because you have a larger set of function words to choose from. Even if the unknown text does not have that word at all, the model still works. Poisson(k=0, lambda>0) is well defined and nicely models the actual situation. On top of that they had several disputed papers. To allow for a uniform set of function words, which can be applied across all disputed papers they had to start with the known texts. So two reasons why the did a+b and not b+a: size of text and multiple unknown texts

However, with Erdnase the situation is entirely different. For many candidates we have less words than Erdnase wrote. Erdnase wrote 52k words. For Gallaway we only have 30k words. For MFA we only have 3k words, etc. On top of this we only have one unknown text we are analyzing. Starting from Erdnase is therefore the much better way to analyze it, because it allows a larger set of function words and allows a consistent and uniform set of function words across all candidates, regardless how long any particular candidate's text sample is.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Roger M.
Posts: 1244
Joined: January 17th, 2008, 12:00 pm

Re: ERDNASE

Postby Roger M. » November 6th, 2017, 6:53 pm

Richard Hatch wrote:Looks like this Erdnase kickstarter campaign will meet its goal. Thoughts?
https://www.kickstarter.com/projects/110191763/the-expert-at-the-card-table-with-photographs-from


This is exactly what David Ben and Julie Eng have already done to perfection with The Experts at the Card Table.

I'm not sure what possible improvement could be achieved beyond what David and Julie have already done so well?

Maybe I'm missing something?

User avatar
Richard Kaufman
Posts: 23864
Joined: July 18th, 2001, 12:00 pm
Favorite Magician: Theodore DeLand
Location: Washington DC
Contact:

Re: ERDNASE

Postby Richard Kaufman » November 6th, 2017, 7:06 pm

Roger, you have only seen the first of the three volumes of David's work on Erdnase. I keep nudging David to finish the next two volumes.
Subscribe today to Genii Magazine

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 6th, 2017, 7:43 pm

Roger M. wrote:
Richard Hatch wrote:Looks like this Erdnase kickstarter campaign will meet its goal. Thoughts?
https://www.kickstarter.com/projects/110191763/the-expert-at-the-card-table-with-photographs-from


This is exactly what David Ben and Julie Eng have already done to perfection with The Experts at the Card Table.

I'm not sure what possible improvement could be achieved beyond what David and Julie have already done so well?

Maybe I'm missing something?

The reason why some think this is a particularly interesting book is due to the fact who Joe Crist is. Joe Crist's teacher was Joe Artanis from Bottom Deal fame https://www.lybrary.com/artanis-bottom-deal-p-242.html and the teacher of Joe Artanis was a Kaldarash gypsy who was friends with Erdnase.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Tom Sawyer
Posts: 307
Joined: January 7th, 2012, 6:44 pm

Re: ERDNASE

Postby Tom Sawyer » November 6th, 2017, 8:07 pm

Hi All,

I'm sure that this post will scroll-away quickly, and maybe that is for the best.

I did discuss Teale on my Erdnase blog, but I am not sure what my "argumentation" was that was referred to by Chris.

It's plain to me that Chris approached the problem from the opposite direction from the study regarding The Federalist. On that much, there is no dispute.

But this case is structured differently from The Federalist case from the get-go. This is so, at least this respect: To the average analyst, Galloway and Teale are just two guys whose text he wants to analyze. The typical analyst would not have as a premise the concept that Galloway probably wrote The Expert at the Card Table. In The Federalist case, they had two candidates, and they knew that one of them wrote one or more of the articles, and that no one else did. They just did not know which one wrote which.

The way I presently see it, the main problem with Chris's analysis on the "(?)" is that it is not an effective approach if the rate of use of "(?)" in the three texts does not accurately reflect what you might call the "long-run" usage of the term for each author.

If you assume that the "(?) rate" is the correct rate in each of the three texts (Erdnase, Teale, and Gallaway), you are left with what Bill said: Gallaway never uses the "(?)" expression.

The remaining two guys do use the expression, so based on that one measure, as Bill says, Teale is closer to Erdnase. And this would be so even if Teale had used the "(?)" expression a thousand times.

I have a suspicion that this reasoning will not appeal to everyone.

But even if I just look at the matter informally, without specific reference to statistics, I could say, well, both Erdnase and Teale used "(?)." So they have something in common that Gallaway does not.

To me, it is not appealing to say, "Gallaway never used the expression, and Erdnase rarely used it, so those two have something in common that Teale does not."

Boiled down, to me, the (from what I gather from this thread) complete absence of "(?)" in known Gallaway writings tends (very weakly) to show that Gallaway was not Erdnase.

Whether the frequent use of that expression by Teale is an argument in Teale's favor, well, actually I don't think it means much one way or the other. Since Teale LOVED the expression, you would think that if he was Erdnase, he would have used the term SOMEWHAT more than three times in the Erdnase book. But as Bill Mullins mentioned, the nature of the subject matter can explain a greatly reduced rate of usage in the Erdnase book.

By this time, I think the two basic points of view (Chris's and Bill's) been pretty thoroughly explored on this thread. Of course, that does not mean that the discussion is over, when one considers the nature of the thread.

I prefer Bill's method for the present set of facts.

--Tom Sawyer

User avatar
lybrary
Posts: 844
Joined: March 31st, 2013, 4:59 pm
Contact:

Re: ERDNASE

Postby lybrary » November 6th, 2017, 9:09 pm

Tom Sawyer wrote:Since Teale LOVED the expression, you would think that if he was Erdnase, he would have used the term SOMEWHAT more than three times in the Erdnase book. But as Bill Mullins mentioned, the nature of the subject matter can explain a greatly reduced rate of usage in the Erdnase book.

That is again an argument in favor of Gallaway. A book on print estimating is subject wise much father removed from Expert than "Higher Magic". On top of that, the distance, statistically speaking in terms of probabilities, is shorter from 3 use cases to zero use cases, than from 3 use cases to 58 use cases. The underlying stochastic process is a key component. That was one of the important contributions of Mosteller-Wallace. If we want to have an intelligent conversation about stylometry then we have to include the basics the field is built on. But I guess there is little desire for an intelligent discussion on that subject.
Lybrary.com https://www.lybrary.com/s-w-erdnase-m-11.html
preserving magic one book at a time

Leonard Hevia
Posts: 1441
Joined: January 17th, 2008, 12:00 pm
Favorite Magician: Dai Vernon, Frank Garcia, Slydini, Houdini,
Location: Gaithersburg, Md.

Re: ERDNASE

Postby Leonard Hevia » November 6th, 2017, 10:18 pm

Bill Mullins wrote:
Leonard Hevia wrote: Political cartoons . . . are stylized art and not meant to be lifelike renditions.
Except in this case, the cartoons of real people are lifelike renditions. (add to the list John James Ingalls, who is sprawled on the ground in front of the cannons.)

Dalrymple did accurate likenesses when he was depicting real people. "Montana" is not an accurate likeness of Sanders, so it wasn't intended to be W. F. Sanders.

W. E. Sanders presumably knew what his father looked like. He would not have perceived "Montana" to have been his father, and there's no reason to think that a decade later, he would have said to Smith that the cartoon referred to his father.


Bill--
The faces of the senators in Dalrymple's cartoon are accurate and lifelike, but that is a stylized work of art we are seeing. It's a cartoon with realistic looking faces on disproportionate bodies. The senators running around are not proportioned correctly. Senator George Hoar resembles a dwarf, which I doubt he was.

Perhaps "Montana" was not intended to be W.F. Sanders, but it does look at least a little like him, and anyone who glances at a photo of W.F. Sanders and looks at that cartoon may believe that it is indeed him--including W.E. Sanders.

I defy you to display a photo of W.F. next to the Dalrymple cartoon to family and friends and ask them if "Montana" resembles W.F. I eagerly await the results.


Return to “General”