Noesis

 

The Journal of the Mega Society

Number 134

August 1997

 

Acting EditorChris Cole

P O Box 10119

Newport Beach, CA 92658-0119

 

 

IN THIS ISSUE

EDITORIAL

RESULTS OF MEGA SOCIETY ELECTION by Jeff Ward

CHESS PROBLEMS by Jeff Ward

MY TUPPENCE WORTH ABOUT TEN BALLS by Robert Low

A NOTE ON CONFLICT by Robert Low

A SHORT (AND BLOODY) HISTORY OF THE HIGH I.Q. SOCIETIES

by Darryl Miyaguchi

 

 

The results of the election are in, and they are clear in some respects and unclear in others (see Jeff Ward’s report for the numerical results).  It is clear that the membership desires to ratify the concept that the Mega Society is open to anyone with one-in-a-million test scores, and that Ron Hoeflin’s tests are capable of distinguishing intelligence at this level.  Thus, we can conclude that the Society’s historical focus on using these tests is ratified.  In particular, Paul Maxim cannot be admitted on the basis of the test scores he has currently submitted.

 

We also have a volunteer for Editor, and since there was only one, there is no need for an election.  Thanks to Kevin Langdon for volunteering, and the next issue (September, #135) will be edited by him.

 

What is unclear is what the bylaws of the Society will be.  The voting on this was almost evenly divided across the proposals, with a lot of abstentions.  Therefore, I think we need a period of time for discussion of the various proposals, followed by another vote.  Since I’ve already argued for my simplified Bylaws, I’ll keep quiet until I hear from the people who voted for either the original Bylaws or for the Langdon modifications.

 

While the membership voted overwhelmingly that the Mega and Titan tests are appropriate vehicles, they did not vote on the exact raw scores to be used.  I’m told that the old Mega Society voted to accept a score of 43 on the Mega Test and 175 on the LAIT.  Since a raw score of 43 corresponds to the one-in-a-million level on Ron Hoeflin’s latest norming of the Mega Test, this seems appropriate.  However, Kevin Langdon has called this norming into question.  As I understand it, Ron is using an adjustment factor near the top end of the test to adjust for “ceiling bumping.”  Kevin questions whether this factor is justified.  I think it would make sense to see that issue debated in these pages also, again in preparation for a subsequent vote of the membership.  Perhaps a committee of members could be formed to evaluate the raw data and make a recommendation. 

 

We need some criterion for admitting people to the Society.  For the time being, I suggest that we continue to use these scores for admission.  Perhaps if Kevin becomes convinced that the “ceiling bumping” adjustment is legitimate, then 175 on the LAIT will correspond to the one-in-a-million level.  Also, since the Titan Test was normed in the same way as the Mega Test, I suggest that we use the one-in-a-million level on its latest norming, which would be a raw score of 43.  All of this is temporary, until a vote.

 

Chris Langan has sent around a newsletter claiming to be Noesis, The Journal of the Mega Society.  Speaking for Jeff, Kevin and myself, and indeed all the other members of the Mega Society, let me make clear that while Chris can send around any newsletter he desires, he cannot claim to be publishing the Journal of the Mega Society.  His newsletter has no association with the Society whatsoever, other than that Chris himself is a member.  Also, needless to say, all claims that he has admitted Paul Maxim to the Society, that he has called off the election, etc. are illegitimate.  In particular, subscription fees to Noesis should not be sent to Chris.

 

Subscription fees for Noesis are $2.00 per issue, and should be sent to the address given above, made out to “Noesis.”   This $2.00 covers the cost of production and distribution.  However, in an attempt to encourage subscribers to submit quality material, I will extend a modified form of the previous policy giving credit for published material, to wit: if the Editor decides to publish a submission, the submitter will receive a free copy of that issue.  Thus, if your subscription runs out at issue 135 (which you can tell by examining your mailing label), and something you submitted is published in issue 135, your expiration date is automatically extended to issue 136.

 

So, please, gather together some interesting ideas, write them down, and send them to our new Editor:

 

Kevin Langdon

P. O. Box 795

Berkeley, CA 94701

(510) 524-0345

 

My tuppence worth about ten balls

 

I’ll get my disclaimer in first: in the following I will criticise what I understand to be Chris Langan’s position on the ten balls problem that has been discussed somewhat in the pages of Noesis. I’m pretty sure Chris will feel this misrepresents his position, so I’ll also get an apology in right here: Chris, I’m sorry if I’ve misunderstood you. But I’ve no objection to having the error of my ways pointed out in public if you still have the energy for it.

 

So, to state the problem (yet again). A closed box contains ten balls, each of which is either white or non-white. In keeping with my culturally offensive background, I shall refer to the non-white balls as coloured...oh, why quibble, I’ll call them black. Right then, the box contains ten balls, each of which is either white or black. Four times you sample a ball at random from the box and return it. Each time, the ball sampled is white. What is the probability that all the balls in the box are white?

 

Answer: it depends on the distribution. We suppose the box is (in principle) taken from a collection in which the probability of choosing a box containing n white balls in p(n).

This may mean that the box is in fact taken from a large collection, or that the balls are put into the box randomly according to some distribution: it doesn’t matter--but without some assumption equivalent to this the problem is not well-posed It is simple then to work out the probability of four successive observations of a white ball given n white balls in the box, and Bayes’ theorem allows us to work out the probability that the box did in fact contain ten white balls given such a collection of observations. The answer depends on the values of p(n), and in the special case where all the p(n) are equal, we obtain the result of about 0.67. Different prior distributions give different results.

 

However, Chris wants to argue that in some sense 0.67 is still the answer when we know nothing about the initial distribution. As far as I can see, his argument is that knowing nothing about the initial distribution entitles to make the assumption that all numbers of balls are equally likely: except that since white balls have been observed, we know that they can’t all be black, so we assume that all numbers of white balls are equally likely except for zero. Then using Bayes’ theorem gives the required result. I think there are two problems with this: the first is that there is an element of having one’s cake and eating it. Using the observation of white balls to restrict the distribution and then claiming that nothing is known apart from the fact that they aren’t all black is not consistent. The second is that I don’t follow the step from ‘we know nothing’ to ‘we can assume equal probabilities’.

 

But there is a way of approaching the problem which attempts to do what Chris claims to do. The set of all possible distributions can be modelled as the collection of points in eleven-dimensional space whose co-ordinates are all positive and whose co-ordinates sum to 1: the point (p0 ... p10) represents the distribution where the probability of there being n white balls in a box chosen randomly from the distribution is pn. For each such distribution, one can calculate the probability of the box containing ten white balls given that four samples are white; then one can integrate over the surface to find the expected value of this probability. Roughly speaking, what Chris has done is to work out the probability of ten white balls for the distribution at the centre of gravity of the surface of distributions, rather than work out the probability for each distribution and then average them. The problem is that the function taking you from initial distribution to probability of ten white balls is not linear, and so a different result will be obtained.

 

Now, I’m pretty sure that Chris claims the following. If you repeat many times the prescription ‘fill a box with white and black balls according to a randomly chosen initial distribution, sample it four times, and retain those boxes which gave a white ball on each sample’, then in the limit, the proportion of those boxes you have retained which actually contain ten white balls will be approximately 0.67.

 

The problem is that because of the nonlinearity, this averaging process gives a different result. (I don’t know what it is: my brain is too small to do the integral---for all I know, the answer could actually be 0.67, but if it is it’s a huge coincidence.)

 

There’s another, deeper, problem, namely the choice of measure on the surface that describes all possible distributions. Uniform measure induced by the choice of co-ordinates above will put the ‘average’ distribution at equal probabilities. Other choices of measure will give different ‘best guesses’. It depends on how you split the universe up into exclusive events.

 

And finally, I know of nobody who says that the law of large numbers doesn’t apply to balls in a box. If I have a box of balls, and repeatedly sample one ball from it, and the proportion of times I get white is about 0.4 after thousands of samples, I’d be pretty confident that there were 4 white balls in there. But I don’t know what that has to do with the problem in hand...

 

Robert Low

email: r.low@coventry.ac.uk

 

 

A Note on Conflict

 

Something that’s bothered me for some time is why it is that conflict between groups seems to be particularly vicious when the two groups are culturally similar. One possible answer is just that it catches the attention more when a couple of groups who seem similar start fighting, but I don’t think that that is the answer.

 

My own suspicion is that this response to slight difference may be rooted in a fundamental psychological need of humans, namely that of distinguishing ‘me’ from ‘not-me’.  If you’re the sort of critter who makes a living by making the environment adapt, rather than by adapting to it, then there is a clear evolutionary incentive for such a trait. This need is pre-rational, and drives a considerable amount of our early development. It strikes me that there may just be some carry over into cultural identity. If so, it is particularly plausible that cases where there is more potential for a mistake should be regarded with greater hostility than cases where the distinction is obvious. Thus, if for some reason boundaries are being drawn up between groups, the more culturally similar the groups are, the less tolerant of slight difference will each group be, and the more savagely will they treat outsiders.

 

There are various parallels to this. One of the most obvious is the reaction in the south of the US not so long ago to Negroes. A visibly black Negro, while treated with contempt and with scant regard to his rights, would be treated far better than a relatively fair-skinned one who had attempted to pass for white.  Again, in religion: a fundamentalist Protestant sect, while taking it for granted that Roman Catholics are the spawn of Satan, will reserve its serious criticism for a group who splits away because of minor doctrinal differences.

 

This may sound defeatist. It isn’t intended to be. Acceptance that some aspects of our behaviour may be influenced by genetics does not obviate the notion of moral responsibility. The brute fact that I may have a genetically determined propensity towards a certain type of behaviour does not refute the fact that I also have a choice about whether to follow my instincts or my conscious morality.

 

Robert Low

email: r.low@coventry.ac.uk

 

A Short (and Bloody) History of the High I.Q. Societies

Maintained by Darryl Miyaguchi

Last updated: September 4, 1997

See bottom of page for Change History

6/28/97: The history is now as complete as I intend to make it. Future revisions will be logged. Most of this material is from the pages of In-Genius or Oath (i.e., Mr.

Hoeflin has been a good source of information—any mistakes in

translation should be attributed to me); a little has come from Marilyn vos

Savant’s book, Omni I.Q. Quiz Contest. Kevin Langdon has also contributed

his comments. Some of the information presented here may be considered

inflammatory, especially since I can’t divine with certainty the underlying

purposes of people’s actions; if I have committed any inaccuracies, please

contact me for corrections.

Some might wonder what relevance this soap-opera-ish tale has to the stated

goals of the high-IQ societies. I would argue that in order to understand

what these societies are about, one should understand their history, including the very human motivations that drove their foundings.

This history is in roughly chronological order.

 

The Chinese Mandarin Class (1 out of 100; 1 out of 10,000; 1 out of 1,000,000)

According to an article published in the Bulletin of the International Test

Commission, and retold by Christopher Harding of Australia (founder of

several high-IQ societies), intelligence tests were invented by the Chinese

in the 7th Century A.D. The Mandarins who ran China for centuries were chosen by examinations that tested for memorization and understanding of the Confucian classics and, in so doing, screened for intelligence. Then Mandarin class was said to have three levels: the public service (top 1 percent of all candidates), the Mandarins (top 1 percent of the public service), and inspectors (top 1 percent of the Mandarins!).

 

High IQ Club with unknown name (unknown admissions requirement)

Christopher Harding writes that he has come across evidence from two

different sources that a high IQ club existed in London, England in the

1890’s. This predates the Binet, though not the Cattell. Harding suspects

this club is associated with Sir Francis Galton.

 

The High IQ Club (1 out of 100)

Begun in 1938 by Dr. Lance L. Ware, a scientist and lawyer, at Oxford

University; this club appears to be the forerunner of Mensa. Their

requirement was the 99th percentile on the Cattell Verbal Test. It was

somewhat informal and produced no literature and became inactive after 1939

(during World War II).

 

Mensa (1 out of 50)

Founded at Oxford University in 1946 by Roland Berrill, a barrister, and

Dr. Lancelot Ware, who later also became a barrister. The original aims

were, as they are today, to create a society that is non-political and free

from all racial or religious distinctions. Mensa welcomes people from every

walk of life whose I.Q. is in the top 2% of the population. Mensa’s primary

emphasis is social. Some see this as one of the major attractions of the society and a key recruiting tool.

There are others who are disappointed with what Mensa has and has not

become. At a 1996 convention celebrating the 50th anniversary of Mensa’s

founding, Dr. Ware (now 81 years old) voiced hope “that Mensa will have a

role in society when it gets through the ages of infancy and adolescence

... but at least it has satisfied its members.” Dr. Ware seemed

disheartened by the Mensan’s seeming inability to focus beyond

self-gratifying pursuits and apply their collective brain-power to problems

facing the world today. “I do get disappointed that so many members spend

so much time solving puzzles,” Ware said. “It’s a form of mental masturbation. Nothing comes of it.”

 

The Berkeley High IQ Society (Admissions requirement unknown)

Admission to this society, founded 3 months after Mensa was founded in the

U.K., was based on College Admission tests to the University of California

at Berkeley, which was similar to the American College Admission exams later taken by American students across the USA in the late 1940’s.  Defunct.

 

Tenta (1 out of 10)

Founded in 1959 at the 90th percentile, Tenta has been defunct for many years.

 

MM Society (1 out of 2,500 nominal, 1 out of 1,000 actual)

The MM Society (also known as “Double M”) was founded in 1966 as a Mensa’s

Mensa, with the intent of accepting at the top 50th of the top 50th

(one-in-2500) percentile. However, MM’s actual qualifying scores were at

almost exactly the one-in-1000 level. It does have the distinction of being

the first of the “higher IQ” societies. After its founder died, it was

taken over by Robert Kaufmann, who treated it as a joke, for which he got

interviewed by Tom Snyder on national TV once. Hoeflin lists this as an

inactive or defunct society as of the early 1980’s. The society is said to

have published an interesting journal.

 

Intertel (1 out of 100)

Intertel, which was originally known as the International Legion of

Intelligence (members are still known as “Ilians”), was founded in 1966 by

Ralph Haines and now has about 1700 members in over thirty countries. Its

theme is “participation and excellence” both within the organization and in

public life.

 

The Hundred (1 out of 100)

Founded in Melbourne, Australia by John Walsh in 1970 and became defunct in

1977. They had a 99-percentile admissions requirement on the Cattell higher

form III (verbal scale) form b (supervised test) only. None other was

considered as far as Chris Harding, who is the source of this information,

knows.

 

The International Heurist Association (Admission based on high-IQ and proven creative ability)

Founded by D. H. Ratcliffe of Western Australia in 1970 and survived until

1973. It never had more than 19 members, and finally disbanded for lack of

interest. Most members were above the 98th percentile in IQ and none were

below the 95th percentile. All had proved creative ability—the basis for

their selection was certification of an original idea by Professor I. J.

Good. Chris Harding, who was a member, recalls this as an unusually

productive group, writing that at least three members had major theories

published around the time of the society’s existence. This society became

the inspiration for Chris Harding’s own International Society for Philosophical Enquiry.

 

The Near Mensa (1 out of 20)

Founded in 1970 by a woman whose name Chris Harding doesn’t recall; became

defunct by 1972. With an advertising slogan that was apparently, “Failed Mensa? Join the Near Mensa,” it’s unsurprising that they went under.

 

The International Society for Philosophical Enquiry (1 out of 1,000)

In 1974, Australian Christopher Harding founded a society called MENS

(Latin for the Mind) at the 99.97th percentile to “one-up” the MM society,

which at the time had the highest requirement at 99.96 [nominal]. Mens later dropped its requirement to 99.9 and called itself “The Thousand,” which in turn later adopted the name “International Society for Philosophical Enquiry” (1976).

The group presents itself as the high-achievement society that invites and

expects creative contributions of its members. The society accepts scores

at the 99.9th percentile on standardized tests and designated unsupervised

tests for admission. People join as Associates, on the basis of their

potential; thereafter, they can attain the level of Member, Fellow, Senior

Fellow, Senior Research Fellow and Diplomate by accumulating specified

numbers of various ‘achievement,’ including such things as earning academic

degrees, publishing, corresponding with other members, etc. The highest

title, Philosopher, is awarded via election. Associate members, who

represent about two-thirds of all ISPE affiliates, are not allowed to vote

in ISPE elections.

The ISPE is directed by a Board of Trustees consisting of three to seven

senior members. A former member of the society criticizes the members of

the Board who “make decisions for the society and are answerable to no

one.” This person also objects “that contested elections are a rarity, with

the decisions of the leadership routinely rubber-stamped, that no dissent

is permitted in Telicom [the society’s journal], and that the ISPE [Board

of Trustees] continues to expel people without affording them the

opportunity to present a defense and without recourse to a vote of the

membership.” As far as I can tell, as an outsider, this assessment appears

to be supported by the events of the ISPE’s history.

ISPE used to use a 70-item vocabulary test called the Vocab A and a

136-item vocabulary test called the Vocab B. The original Harding

Skyscraper test had a 10-item vocabulary test that [Hoeflin believes] was

later called the Vocab C. When the ISPE required a 99.9 percentile score on

both an I.Q. test and one of these vocabulary tests, it concluded that a

person who could pass both tests would be about one-in-2000 in AQ (“Ability

Quotient”). The vocabulary test requirement was dropped in 1989 since most

IQ tests already test verbal ability; moreover, it was deemed unfair to

non-English speakers to discriminate on the basis of an English-language

vocabulary test. Another factor in the change was that there was no way to

control cheating on the vocabulary tests.

The ISPE Vocabulary test ‘B’ can be found in its entirety with answers and

percentile rankings in the book, The Ultimate iQ Book, by Marcel Feenstra,

Philip J. Carter, and Christopher P. Harding, 1993 (ISBN 0-7063-7148-8). I

have been informed that the ISPE Vocabulary test ‘A’ can be found

(presumably in its entirety with answers and percentile rankings) in a book

by the same authors, The Ultimate iQ Challenge. This was published in maybe

1994 or 1995.

The ISPE used to accept Hoeflin’s Mega Test scores for admission, but

dropped its acceptance of that test in 1992[1]  The society also doesn’t

accept Kevin Langdon’s LAIT. Christopher Harding’s own W-87 is accepted,

though, despite being unsupervised, heavily dependent on vocabulary, and

subject to cheating since it prohibits reference aids. The W-87 does,

however, have the advantage of being normed under the supervision of an

“accredited psychologist,” according to an ISPE representative. The

disadvantage is that an adequate report on its norming has never been

published. When the Triple Nine Society Psychometrics Committee asked

Harding for data on the norming of his tests he said that he had discarded

it. It is also unclear to me whether or not the accredited psychologist presiding over the W-87 norming was actually Chris Harding himself.

Kevin Langdon’s response to the ISPE’s official rationale is this: “What

many people, even in the highest-level societies, do not realize is that

psychometrics is a science, though a relatively inexact one. The relevant

question with regard to scientific work is whether its methodology is correct, not whether it is performed by a member of the priesthood.”

 

401 Society (1 out of 10,000)

A “secret” society founded by Chris Harding in 1975 for the 3 or 4 people

who had managed to reach or exceed the one-in-10,000 level on his Skyscraper test. The society is now defunct.

 

Four Sigma Society (1 out of 30,000)

The Four Sigma Society was founded by [then] ISPE member Kevin Langdon in

1977. The society was active for about six years (1977 - 1983). Kevin

edited four issues of the society’s journal Sigma Four, with an average

interval of two months. George Koch edited eight issues from 1980 to 1983,

with an average interval of six months. The society accepted only one test,

the Langdon Adult Intelligence Test (LAIT), on which an I.Q. score of at least 164 was required (later, other Langdon tests were also accepted).

When the LAIT was published in Omni, in the April 1979 issue, it was taken

by over 25,000 people, resulting in many new recruits for Four Sigma.

Unfortunately, the large volume of responses to his test (which is no

longer scored), coupled with Kevin’s propensity for tardiness, also

produced numerous complaints of late or non-existent score reports. Omni

eventually sued Kevin for one million dollars (which they never collected).

Kevin did eventually score the backlogged test answer sheets.

During the late 80’s, the society was briefly revived, but it is now defunct again.

 

Triple Nine Society (1 out of 1,000)

The Triple Nine Society was founded in 1979 as a more democratic

alternative to the ISPE by Richard Canty, Ronald Hoeflin, Ronald Penner,

Edgar Van Vleck, and Kevin Langdon, who was the driving force. At that time

a small group of early members of the ISPE, largely under the direction of

C.R. Whiting (ISPE’s first elected president), had suddenly introduced an

autocratic setup that would perpetuate their control of the society, which

up to that point had been set up more democratically[2]. Whiting evidently

resented Kevin for “upstaging” the ISPE’s king-of-the-hill status with its

99.9th percentile minimum requirement by founding the Four Sigma Society in

1978 with its one-in-thirty-thousand minimum requirement. Whiting’s response to the establishment of the Triple Nine Society was immediate:

all five members stopped receiving the ISPE journal, Telicom, and they were

informed six months later that they had been expelled from the society by a

secret “Ethics Committee,” whose members’ identities are still unknown

nearly twenty years later. Hoeflin writes that his own infraction was

apparently that he agreed to serve as ombudsman for the new Triple Nine

Society, which the ISPE’s leader construed as an attempt to “destroy” the

ISPE. Expulsion procedures have been a consistent source of criticism directed at the society by former members (see also entry for Cleo Society).

Joe O’Rourke, at the time editor of the ISPE journal, Telicom, refused to

be a party to the actions of Whiting and company, but didn’t want to

embroil himself further. He wrote a scathing denunciation of the ISPE

leadership and resigned from the editorship and the society—but he was

not one of the founders of TNS, as I have written earlier.

Ronald Hoeflin served as Editor for 63 of the first 100 issues of the

Triple Nine Society’s journal, Vidya. From around September 1985 to January

1989, he managed to eke out a living from that job. At the time when

Hoeflin became Editor, the society was having a hard time finding anyone

willing to do the job. Hoeflin presented the society with a proposal under

which he would be paid a flat amount per issue of Vidya produced.

At TNS election period 1987, Hoeflin supplied advance copies of writings by

those with views opposed to his own (submitted for the election issue of

Vidya) to their political enemies, who were thus able to reply in the same

issue. He published this election issue after he was ordered by the TNS

Executive Committee to withdraw it until it had been substantially revised.

For this action, the Committee decided to replace Hoeflin as Editor.

The election resulted in two Executive Committees, each claiming

legitimacy. When TNS’ funds were turned over to the Financial Officer

(Barry Zalove) belonging to the new faction, they continued to pay Ron to

produce Vidya. Later, that committee fired Ron as Editor. Ron dropped out

of the society altogether.

By the time of Hoeflin’s removal, he says he could no longer earn a living

this way anyway, since constant squabbles and infighting had reduced

membership from a peak of 750 to a bare 400. Continued contention in TNS

has led to continued membership decline. Current membership is about 160.

Ironically, the Triple Nine Society no longer accepts Kevin’s own tests for

admission because of Paul Maxim’s campaign (see entry under the Mega Society). Kevin tried to exert his influence upon the current admissions officer to keep listing his LAIT as an acceptable test, but to no avail.

 

The High-IQ Society (1 out of 10)

Announced in the early 1980’s with a 90th percentile requirement like Tenta; used a mailing list supplied by Kevin Langdon of people who had tried his LAIT, but this group did not get off the ground.

 

The 606 Society (6 out of 1,000,000)

The 606 Society, founded by Christopher Harding, was originally named the

501 Society, which was founded in 1980. This latter society had a 99.999 (1

in 100,000) requirement. Later the requirement was raised to the 99.9994

percentile (6 per million) and the society was renamed 606. Still later,

all members of the 606 Society were inducted into the Mega Society (1 per

million requirement) when the latter was formed in 1982. The names of Chris

Harding’s various societies (606, 501, 401) are derived from the various

admissions requirement: the minimum rarity level for 401 is one in ten to

the fourth, for 606 is six in ten to the sixth, etc.

Evidently, the name “606 Society” caused some heartburn. “Formula 606”

refers to an early, pre-penicillin cure for syphilis based on a compound of

arsenic, as indicated in the classic 1940 movie, “Dr. Erlich’s Magic

Bullet,” which is a bio-drama about the inventor of this cure starring

Edward G. Robinson. Thus the 606 Society seems to suggest that the members

are people who were cured of syphilis using Formula 606!

 

The Mega Society (1 out of 1,000,000)

The Mega Society was founded in 1982 by Ronald Hoeflin. The society was

initially set up as an experiment to see if a society with a

one-in-a-million requirement could be achieved. Neither Christopher Harding

nor Kevin Langdon thought such a high entrance requirement psychometrically

feasible; nevertheless Harding agreed to supplement the Mega Society with

members of his 606 Society (a 6-per-million group), and Langdon allowed

Hoeflin to use his list of the highest LAIT scorers, to help Hoeflin get

his society off the ground. Hoeflin occupied the position of Administrator.

Unsurprisingly, the Mega Society’s formation did not happen without

conflict. As Hoeflin tells it, Kevin Langdon resented Hoeflin’s “upstaging”

of his Four Sigma Society, and started a campaign to undermine the

society’s status as a one-in-a-million society. Kevin wonders why Ron would

take this position after accepting Kevin’s help in founding the society in

the first place! What Kevin did question was Ron’s norming of the Mega and

Titan Tests, which placed the ceiling at 190+. He has written that there is

evidence that the ceilings of Ron Hoeflin’s tests are no higher than 180,

such that the society’s requirement on these tests (43 right) is less than

the one-in-a-million level. Ron has rebutted Kevin’s claims, but neither has ceded his position.

In the society’s journal, Megarian (issue #6, Oct 1982), Johannes Veldhuis,

Mega’s Recruitment Officer, proposed that three test scores combined

according to a certain formula[3], be required for admission in the future

and that, as only five of Mega’s 18 members at the time met this new

criterion, the remainder of the membership be relegated to “honorary”

status. The rationale for this proposal was the need to substantiate the

claim of the Mega Society’s one-in-a-million admission criterion for

listings in the Guinness Book of World Records and the Book of Lists. In

Megarian #11, Hoeflin proposed a set of rules under which the Mega Test

would be the only exception to the three-test rule and Hoeflin would have

exclusive executive power in the society. A vote was taken of the Mega

membership, and Marilyn vos Savant announced the results in Megarian #15.

The members overwhelmingly supported an undifferentiated membership list.

In Megarian #21 (June 1984), acceptance of a set of bylaws establishing