The CC Theory

About the debate
Map of the debate web pages
Log of updates

Arguments FOR the CC Theory

Arguments AGAINST the CC Theory

Rebuttal to arguments for the CC Theory

(Rebuttal written by opponents of the CC Theory)

Rebuttal to arguments against the CC Theory

(Rebuttal written by proponents of the CC Theory)


A greatly shortened digest of the CC Theory debate

NOTE: All the original newsgroup discussions can still be read on Google. Go to and type the subject line into the search box.

Date: Dec. 9, 2000
Subject: Why does kong have precedence over chow

This is when the discussion began. Dan Lau asked why, in HKOS, a claim for a discard to make a pung should take precedence over a claim for the same discard to make a chow. He specifically requested a historical point of view. Alan Kwan's answer used Chinese Classical to explain the origin of the thinking. Alan said HKOS had evolved out of CC.
Cofa Tsui retorted that it was just as likely that CC evolved out of HKOS.
I provided a long list of evidence that indicated Alan's theory was more likely.
Cofa fought us on all evidence, saying "evidence isn't proof." We agreed that evidence isn't proof, but we pointed out that this fact didn't make the evidence worthless. These arguments bounced right off Cofa.

Date: Oct. 2, 2001
Subject: Nine tiles warning
The debate flared up again when Martin Rep raised a question about the 9-pieces warning in HKOS. Again, Alan used CC (specifically, a Millington rule) to explain the logic or origin of the penalty on Oct. 9. Again Cofa questioned what CC had to do with an HKOS rule, and again Alan and I tried to explain.

Sidebar: On the challenge of debating anything with Cofa
Some people enjoy debating, just for the enjoyment of arguing. I'm not one of those. I used to work for a guy for whom every conversation was a contest. Perhaps Cofa is like that former boss of mine.
Cofa wore down his opposition by twisting what they said, by arguing them down on their choice of words in a statement, and at times by introducing some arguments that didn't even make sense to me.
A frequent tactic he would use would be to insist that the rules described by one author couldn't be remotely related to the rules described by another author - as though the two authors didn't describe the same variant. He was still using this tactic 6 years later.
He often hinted or said outright that writings by non-Chinese were suspect or invalid.
He decried the use of the word "rulebook" for any books other than his own.
Cofa fought us on all definitions - he has even insisted that the word "mahjong" (by whatever spelling) cannot be defined.
He frequently objected that the term "Chinese Classical" (CC) wasn't used in the 1920s; we explained it's just a term we use to refer to 1920s mahjong, but he fought against it. Six years later, he still acted as though he didn't know what CC was -

I found it very frustrating to debate with him - we couldn't even agree on terminology, so every time Alan or I said something, Cofa could simply challenge us on a term we used (sidetracking the conversation and thus stop all discussion of the main topic). He seemed to conveniently forget or misunderstand most things we said, or twist our words to mean something other than what we'd been trying to convey. This was still a tactic 6 years later. Julian Bradfield (present throughout the entire history of all this) chided Cofa for this in 2006:

Cofa never bothered to obtain copies of any of the 1920s books, even though they are readily available on eBay and Amazon. Twice I offered help. Early in the debate I informed him of a book available on eBay. And in 2006 I even offered to sell him a couple of duplicates from my collection. He replied:

A later post by Thierry Depaulis [Dec. 11, 2006 - "Analysis of rules - A game of the year 1903"] mentioned something Babcock wrote, which Cofa later noted as significant. Cofa complained that I'd been been guilty of omitting this information. But this thing Thierry mentioned was not in Babcock's 1st edition (only in the hardcover). Cofa's reasoning as to which Babcock edition he would regard as worthy of having was woefully uninformed (as usual). And it would behoove him to increase his library if he's going to debate a thing.
In contrast to Cofa's method of debating, Alan's gentlemanly, exacting, cogent, and patient style earned my respect and admiration. I may not have agreed 100% with everything Alan said, but I liked him. Still, sometimes I had to let him fight Cofa without me.
Sorry for the digression. It was necessary to illuminate what occurred next...

Date: Oct. 9, 2001
Subject: Nine tiles warning

Cofa said he didn't understand a particular thing Millington had said.
I decided to reply. But by this time I had become extremely weary of trying to explain things to Cofa, given the way he usually seemingly didn't understand explanations, or fought the terms used in them. This weariness is evident throughout my Oct. 9, 2001 post. In my reply to Cofa's not seeing the point of the particular part of the Millington quote, I said:

This prompted Cofa to insult me:

Date: Nov. 6, 2001
Subject: The Ongoing Debate (long)
It seemed likely that the debate would be sparked into life yet again due to a response I'd written to a post about proto-mahjong (aka "ur-mahjong") by someone named Kevin. Having grown weary of the way it had gone before, I wrote a post trying to civilize the discussion and undo the tactics Cofa used to derail anyone else's line of thought.
On Nov. 8, I said (possibly not for the first time), that I wasn't going to argue for Alan's assertion that CC was the original rule set (proto-mahjong). I was more interested in making it clear that CC predated HKOS (meaning it was not only possible, but also likely, that HKOS evolved from CC, which is why CC rules illuminate HKOS rules).
Nevertheless, when I created the CC Theory Web page, I enlisted Alan's aid in drafting the page. This aid necessitated incorporating Alan's arguments in full. I started this work in November of 2001 and it was online in April of 2002. During that period, I had a chance to go to Hong Kong, where Alan lives, so I met up with Alan in person. Photo at

Date: April, 2002

Moving the debate to the Web in 2002 (see the pink and green table above) did make for a much less contentious newsgroup. I was able to put the whole sorry episode behind me for a few years. Over time I lost touch with Alan (he stopped writing on the newsgroup). I made a personal vow to never respond to Cofa again.

Date: Oct. 26, 2006
Subject: Millington's 3 types of kong
It was unavoidable, though, that the topic would flare up again eventually. This time the spark was Julian Bradfield, expressing dissatisfaction with something Millington said.

Date: Nov. 15, 2006
Subject: A Millington critique(fairly long)
Julian's post prompted Michael Stanwick to write. One of the points of Michael's thesis in this post was that Millington referred to several variants in the 1920s but without providing any details, names of sources, or anything. So we started talking about 1920s variants - it was inevitable that this would lead again to the question of whether CC or HKOS predated the other since CC was the prevalent variant at the time, and it'd be impossible to keep that fact behind the curtain throughout the discussion.
I reexamined my vow to never respond to Cofa in light of then current events. I saw a parallel behind my vow and the deplorable refusal of Pres. George W. Bush to talk to North Korea or Iran - I didn't want to think of myself as being like the hated Bush in any way. So I responded yet again to Cofa when the time came.
In discussing variants, I proposed that a "taxonomy" of mahjong variants would need to be created to enable discussion in detail, classifying variants and families and classes. Cofa said that anytime anybody played using a different table rule, a different variant was created. I said table rules would have to be subtypes under a "genus" (variant) in order to make the taxonomy manageable. Cofa said we should name all variants according to the author - and that when an author described more than one variant, those have to be listed separately too. Essentially, in a typical time-honored tactic of his, he argued the idea of taxonomy to death.

Date: Nov. 28, 2006
Subject: Earliest Chinese reference to "ma que"
Then the CC Theory really started to fray at the edges. This post by Thierry Depaulis revealed details of a pre-1920s variant.

Date: Dec. 8, 2006
Subject: Analysis of rules - A game of the year 1903
In the course of this, Cofa threw words from my own Web page at me. I hadn't looked at the CC Theory Web page in a long time, and when I did look, I was horrified to see that Alan's claim that CC was the original rule set was still there. And I discovered some definitions that had grown stale over the years (as I'd learned more about the origins of the game thanks to Michael Stanwick's articles in The Playing Card). So on Dec. 11, 2006 I went into the CC Theory Web page and modified some definitions and deleted Alan's claim that CC was the original rule set. By Dec. 13 it was clear to me that the 1903 rules described by Thierry, and some 1915 rules described by Mauger, and even alternate rules mentioned by Babcock in his hardcover book, while not disproving the hypothesis that CC led to HKOS, did hint at a possible parallel development.
I totally rewrote the front page of the Web page.
On Dec. 14, the next morning, Cofa exulted:

It was perhaps unwise of me to go so far as to support Alan's thesis that CC was the original rule set. I should have stayed with CC being the "parent" variant of HKOS (as I'd said on Nov. 8, 2001). But especially, I should have given up trying to convince Cofa to agree with anything. His opinions are not important to me.
But for the record, despite Cofa's boast above, not all of his original claims have been proven to be correct. Evidence still supports the theory that HKOS did not yet exist yet in the 1920s. So it is certainly not likely that CC evolved from HKOS, as Cofa once suggested. A recent analysis of the pre-1920s and 1920s rules (which I worked on together with Thierry Depaulis via email and can be seen at indicates that it is indeed likely that HKOS evolved later (possibly directly from CC, or possibly from a parallel variant contemporary with CC).
Alan's initial answers to Dan Lau and Martin Rep, that the HKOS rules they asked about had their origins in CC, are not disproved by the new information. Regardless, it was silly of me to continue working so hard to try to convince one person, and I regret having let myself get carried away with it as far as I did.

"Never argue with an idiot. Onlookers won't be able to tell the difference." - Unknown