Currently one of the big discussion over on the Erdnase thread of the Genii forum involves Dollie Seely (wife of Edwin Sumner Andrews) being related to Louis Dalrymple. It is probably going to take the passage of some time before I feel that I understand the significance of the recent developments, which are mainly due to the work of Richard Hatch and Bill Mullins. However, below are some reflections that are not set in concrete.
If the authorship of The Expert at the Card Table were a matter of ordinary, day-to-day importance (that is, if it were not very important), I probably would be inclined to say that the case had been resolved adequately, and that it might be time for people to move on.
However, there are several problems with such a conclusion, even if we were dealing with a fairly unimportant matter.
First, there are several other candidates with cases that could be called “pretty good,” and this suggests that, to feel confident about Edwin Sumner Andrews, one probably needs more proof — better proof — MUCH better proof — than that which now exists.
In my most recent (2015) book on the Erdnase authorship issues (Rethinking S.W. Erdnase), I went into quite a bit of detail concerning the views I held regarding Edwin Sumner Andrews. Overall, I indicated that the case favoring him was actually pretty thin. The strongest argument for him being Erdnase was the fact that his name perfectly reversed into S.W. Erdnase, which is not the case for Milton Franklin Andrews, and is not the case for W.E. Sanders.
And when THAT is your strongest argument, you have a really weak case! (However, you will recall that David Ben and others hold a decidedly different view.)
Andrews lived in Chicago at “the right time,” but so did a lot of people.
And among the things that make it hard to accept him as Erdnase is the fact that (as just mentioned by Richard Kaufman) there has been no showing that Andrews was “an experienced amateur or professional magician.” This is a good point. After all, a huge part of the book dealt with card magic.
Or maybe you don’t like that requirement. But shouldn’t there be something a little more than the Pippins bit to show skill at cards?
I do think it’s likely that the Pippins article is the tip of the iceberg, so to speak, but when you get down to it, it doesn’t add a lot of clout to the Andrews case.
What I guess I am saying is that if you attach a great deal of importance to the Dalrymple/Dollie relationship, then your case is made. Andrews was Erdnase, and it’s “case closed.”
But maybe — for whatever reason or reasons — you DON’T attach a great deal of importance to the Dalrymple/Dollie relationship. Maybe you think that, definitionally, a third cousin, once removed, is not close enough to help. Or maybe you think it’s a pretty close relationship, but you don’t think Andrews would have been aware of Dalrymple and who he was. Or maybe you think, “Well, when you look at those two lines of the family tree, the relationship seems pretty close, but I’m guessing that Dollie had 150 third-cousins on her side, and another 150 on Edwin’s side, and NOBODY can be aware of that many people.”
Some might say, “Yes, but Dalrymple was famous, so he would be been well-known to Andrews.” Yes, but that’s just a counter-argument. It doesn’t mean that no one is allowed to think the things I just said.
So, as I said, maybe you don’t think the Dalrymple/Dollie relationship is important. If that’s the case, then there isn’t a whole lot left over, as I discussed at length in Rethinking S.W. Erdnase.
I said, in that book:
“If it turns out that Dollie Seely (or Dolly Seely) was closely related to Dalrymple, then the chances of Edwin Sumner Andrews being Erdnase would immediately skyrocket. Even though I do not place much significance on Marshall D. Smith’s recollections regarding Erdnase, the possible coalescence of Smith’s statements about Dalrymple and observed facts as to Edwin Sumner Andrews is highly intriguing.”
I doubt that many would contend that the two were “closely related.” On the other hand, in relationship to the connections of the other candidates (exception: W.E. Sanders) with Dalrymple (basically zero), the connection is relatively close.
But I am not yet certain that the relationship is “close enough.” Or if “close enough,” it is not clear what the conclusions should be.
One might want to look at matters from the following point of view. Say Dalrymple and Andrews were very closely related. Say Dalrymple was Andrews’s brother-in-law, and that they shot pool together every weekend.
But then assume that Andrews was not Erdnase. Remember, this is just an assumption, arguendo.
Could two such apparently inconsistent assumptions coexist happily?
Oh, I think so, and I may go into this in a future post.
But here is something else which makes it difficult for me to be super-enthusiastic about the latest findings regarding Dollie and Dalrymple. It has to do with the commonness of the surname Andrews. Granted, if you take the typical guy named E.S. Andrews, he is quite unlikely to have a close, or medium-distance, relative named Louis Dalrymple. But if you take a guy named Louis Dalrymple, it does not strike me as especially unlikely that he would have such a relative named E.S. Andrews, or several such.
After all, we are already loosening things up so that we (some of us) like the third cousin, once removed. What about fifth cousins and sixth cousins, once, twice, or three times removed? Where do you draw the line? We didn’t have any a priori standards set up.
It would be nice if, before the fact, someone had said, “I will accept any relative within such-and-such boundaries. They can be uncles, nephews, great great great grandparents. They can be adopted. They can be half-brothers. They can be Dollie’s great-aunt’s grandfather’s half-brother’s twin-sister’s great-granddaughter’s great-grandniece. But they do have to be within the limitations I am setting forth.”
The limitations might end up allowing only 50 relatives, or they might include 500, or 1,000, or 10,000. I don’t know, and I don’t know what a “reasonable” number would be. And it isn’t just the number that is important, because certain circumstances would make it more likely that remote people would be known to Andrews.
If you “count” a third cousin, once removed, and everything closer, I have very little idea of how many are within that envelope. I suspect that it is a large number, the vast majority of whom Andrews would not have had any knowledge of.
If you set forth the criteria in advance, at least you can be certain that you are not back-fitting your criteria to fit your results.
October 19, 2017