Payola Diplomacy

When the variant was first designed, it was called "Mercenary Diplomacy," and to playtest it, I created a game on the USEF judge called payola. The only reason I didn't set out to call the game "mercenary" was the Ken Lowe judge's eight-character limit on game names.

When submitting the original rules of "Mercenary Diplomacy" for inclusion in the variant banks, I learned that (owing to the pre-existence of a Lew Pulsipher variant with the same name) the variant would become known as either "Mercenary Diplomacy III" or "Mercenary Diplomacy IV." Yuck. To quote The Dubliners, "I've never fancied being numbered."

So at this point, I undertook to rename the variant, and I asked for suggestions from all quarters, including the players of the test game. Some of the suggestions I received were:

• Soldier of Fortune Dip (or "SOF-Dip")
• Francly-My-Dear-I-Don't-Give-A-Damn Dip
• There's-Gold-In-Them-Thar-Swiss-Hills Dip
• Greed Diplomacy
• Auction Diplomacy
• Hi Bid Diplomacy
• Double Dip (as in the practice of receiving pay, uniforms, and supplies from your home country while taking money and orders from another), which I thought had potential, though I felt that it too was probably already in use for a variant involving two boards or something.

Although no one suggested naming the variant "Payola", I suppose that I had gotten used to associating the now unnamed variant with the game being played. As a result, in the end, I decided to name the variant after its first playtest game, claiming that "Payola Diplomacy" rolls off the tongue better than any of the other suggestions. And so, as it turned out, it was fortuitous that the original name — "Mercenary" — had one too many letters to become a judge gamename.

I must admit that, when going over my saved mail to write these annotations, I was surprised to find that the variant was named after the first game to be played, and not vice-versa. I remembered, of course, that the variant was originally named "Mercenary" and that it was then renamed, but I had forgotten that the decision to name it "Payola" happened so late.

I have often been asked how the idea of Payola Diplomacy originated. Actually, this is a very interesting story.

The most common assumption is that the bribery concept was taken from the game Machiavelli. For those readers unfamiliar with it, Machiavelli was Avalon Hill's first application of the basic Diplomacy ruleset to a different game; Machiavelli is set in feudal Italy and adds bribery (though very much unlike that in Payola), garrisons, assassination, and such chance events as plague, floods, and famine. Personally, I am not a fan of Machiavelli, as I consider the introduction of chance elements to be a ruination of what I think we all consider to be the perfect ruleset of Diplomacy (basically, if a game uses dice, I consider it intellectually inferior — this is just me). I only played Machiavelli a few times, and this was so long ago I that no longer remember the mechanics of the game. So — although I knew that Machiavelli added bribery in some form — the thinking behind Payola Diplomacy neither began with nor centered around Machiavelli.

Rather, the Payola concept came from a different Diplomacy variant, the two-player "Intimate Diplomacy". For those who are unfamiliar with this variant, two players each assume leadership of one of the seven Great Powers, as in the standard game, and then the players bid against each other for control of the units of each of the five other nations for each game-year. The bidding is one round, and control is decided on a country by country basis (not, as in Payola, on a unit by unit basis).

The rules for Intimate Diplomacy were posted to the Diplomacy newsgroup in mid-to-late 1993, with the notice that a pair of games of the variant would be started on the USEF judge and that "dummy" players were needed to provide e-mail addresses for the five "inactive" powers. I volunteered to become one of these "dummy" powers, and at the same time I roped John Woolley into playing a quick game of Intimate against me to familiarize ourselves with the variant.

Admittedly, the couple of games John and I played were a small sample, but from them, John and I found Intimate Diplomacy unsatisfying. It seemed that any lead was basically impossible to overcome. One other thing I remember is that we considered the penalty for overbidding to be especially harsh and unforgiving.

So we set out to improve the game. I came up with the "let each of the two players offer bribes to each unit rather than to each power" idea, and then — before even playtesting the two-player Intimate variant this way — I quickly applied the idea to the full seven-player game. I'll just say here that although I do think that this rule would, perhaps minimally, improve Intimate Diplomacy, much of the beauty of Diplomacy — the whole alliance and "chase the leader" concepts — are lost on a two-player game, simply by its nature. This is not a knock on Intimate, and again, I surely have not played Intimate enough to pretend to pronounce judgement on it.

Another important feature that Payola provides which is not provided by Intimate Diplomacy is secrecy of control. In Intimate Diplomacy, both players know, when writing their orders, exactly which units will be controlled by the other player. By keeping this information secret until orders are processed (as Payola does), Intimate would be made more challenging. Indeed, in a multi-player game situation, this secrecy of control is one of the most important features of Payola.

To resume our story, though, puma was the name of the Intimate Diplomacy game for which I volunteered to act as a "dummy" power. After the very first "bids" for control of the five neutral powers were processed, the winner of the game was decided (in other words, there was a forced win even before the units were actually ordered). Before and while all this was taking place, I entered into a broadcast conversation about the faults that we "dummy" powers saw with the variant, initially intending to offer some mechanism for the five "dummies" to obtain some result from the game themselves. It was during this discussion, which occurred coincident with my own Intimate games with John Woolley, that the idea for what became "Payola" first hit me. From the history of puma comes this broadcast of 3 January 1994 which describes "Payola" at its genesis.

Okay, fellas, get a load of this. In discussing the problem of activating the inactive players in judge Intimate Diplomacy games, he [John Woolley] and I ended up aboard a train of thought that put that problem on the back burner for a minute. But here's a variant idea that just sounds absolutely fascinating. I hope it sparks some discussion from you guys. I would LOVE to play it.

Standard seven player game. All units are basically mercenaries, and if you don't pay your units enough to keep them following your orders, they'll look to pad their own pockets and follow someone else's....

At the beginning of the game and after each winter, each player gets $5 (or 5 pounds, 5 francs, 5 marks, 5 rubles, 5 lira, etc., whatever you want to call them) for each controlled SC. So for Spring, 1901, each country has $15 (except Russia, of course, which has $20).

Then in each Spring and Fall, every country may issue orders for any and all units on the board, whether owned by that power or not. With each of these orders, the power submits how much money it is willing to pay to get that unit to follow the order. Each unit then follows the order which would pay it the most money, all bids considered, with bids for like orders summed together.

If a unit receives two or more orders that would pay it the same, the owner of the unit decides which of these orders his unit will follow.

All money offered for a followed order is spent, gone either as pay to your own soldiers and sailors or as bribes paid to the soldiers and sailors of other powers. All other (losing) bid money is retained in each country's treasury.

The power whose order is followed during movement also controls that unit during retreats, if necessary. If more than one power contributed money to issue the order that the unit follows in the movement phase, the owner of the unit chooses which of these powers will control the unit during the retreat phase.

If a power "wins" control of so many units that he spends more money than he has in his treasury, all his bids are lowered by one dollar, and every unit that would have accepted the order he issued it decides anew, based on the new bids, which order they should follow. If a player is still in this position, his bids go down another dollar, etc.

Builds and removes are entirely the business of each power, acting alone and in its own best interest.

Players may not give money to other players. Well, I guess they could, but it would be a pretty stupid thing to do, and just an extra headache for the adjudicator, so, like I said, they can't.

So, as you can see, the ideas in Payola were nearly fully formed on its very first day of existence. In fact, acceptance lists and retained control by the unit owner in retreat phases are about the only key concepts missing from the above broadcast. Also, at this early stage, the concept of secrecy — not knowing exactly who is controlling each unit — is not to be found.

The interesting part, though, is how John and I tested the Payola principle. When we gave the idea some thought, we realized what the addition of bribing for the right to make every move — both yours and your opponent's — does to normally deterministic games. It is a fascinating change! To test the idea, we applied it to tic-tac-toe ("naughts and crosses" outside the United States), giving both the "X player" and the "O player" 100 units of currency each to begin with. Both the players then secretly write down how much they will pay to place the first "X", and the high bidder places it, and his treasury is reduced by his bid amount. Then the players bid for the right to place the first "O". It is absolutely amazing what an interesting game tic-tac-toe becomes; tie games are much more rare, both players seem to have an equal chance to win, and a player always has no one to blame but himself if the outcome is against him.

Oh yes, tic-tac-toe will never be the same again. From now on, every tic-tac-toe player will have my name on his or her lips. Tic-tac-toe will certainly soon become a professional sport, rivalling soccer ("football" outside of North America) in the hearts of the world.

Next: applying Payola to Chess! Can you imagine?! The winning bidder makes the move for whosever turn it is — any legal move...!!!

Payola is, to pat myself on the back, a revolution in gaming, since it is a variant that can be applied to basically any game. Diplomacy is called the chameleon game because it is uniquely "variantizable," while other games are not. However, Payola's "bid for the chance to make a move" idea is a legitimate "variant" for just about any game in the world.

I played a few games of Payola tic-tac-toe (as discussed above) with my father to introduce him to the concept, and he and I realized that one necessary element of the Payola concept (one that John Woolley and I did not use when first playtesting Payola tic-tac-toe, but which has been used in Payola Diplomacy since almost the beginning) is the neutral third party (GameMaster) who is required to award the right to make each move and to privately (rather than publicly) adjust the treasuries for the players appropriately. Such a party is necessary so that neither player knows how much money the other has, since having this information often enables one player to simply use money management to force a win after the first few tic-tac-toe moves.

It was obvious from the beginning, even under the rules to Payola Classic, that Mastering a Payola game would be more difficult than Mastering a standard game of Diplomacy.

By the time the first playtest game had gotten underway, I had written a computer program to resolve the offers as they came in to the Master. This "Payola adjudicator" has been kept strictly up-to-date. Indeed, it leads the continuing development of the variant, since new changes to Payola have never been considered official until the adjudicator properly handles the modification. (Or, in a couple cases, improperly, since some bugs in the program unfortunately surfaced during play and were thereupon fixed.)

After a while, the Payola adjudicator was given a Web page front-end, so that it was no longer necessary for me, as the GameMaster, to copy and paste offers into an input file and manually run the adjudicator and then manually send its output to the Ken Lowe judge or error notices to the players — all that became automated. At that point, there was the adjudicator (written in C) and the Web pages (written in Python) that fed input to the adjudicator and read its output and delivered that to the Web screen for the player.

Eventually, the offer adjudicator itself was rewritten in Python and it became part and parcel of the Web interface. The whole set-up now also supports a number of different maps and subvariant rules.

The final result is The Payola Place Website, where players can enter their offers, update their acceptance lists, transfer money from one to another, enter retreat and adjustment orders, and see the complete game map. As a result, Payola games are now run with absolutely no manual intervention by the GameMaster. Hallelujiah and good for me!

[Addendum: As the reader almost certainly knows, the programming effort described above continued from here and became the full, non-Payola-centric DPjudge that has now run thousands of games over the years.]

From the beginning of the game, it was decided that there would be no bribing for adjustment phase orders. In a discussion on this point, John Woolley and I agreed that the ability for an allied coalition to force an enemy build to be waived was too great an advantage to allow. Because the adjustment orders occur at the end of the Payola fiscal year, whether a build or remove would be controlled by an enemy could become a matter of how much money each player had left at the end of the year and at a time when the various tax incomes for the next year were known values. Since an uncontrolled remove or a waived build can decide the game, John and I ruled out the possibility of bribery during the adjustment phase. We justified this with statements like "you can't bribe factory workers (or at least not the taskmasters who keep them busy)" and "units may be corruptible, but the bureaucrats who decide which units get maintained each year are not."

The retreat phase was another issue. We also ruled out bribery almost immediately, this time mostly for reasons of expediency, and in the end, of course, total control over retreating units was granted to the unit's owner. However, the thought process that led to this decision was much different than that for the similar decision concerning the adjustment phase.

As can be seen in the initial ruminations on Payola, the initial plan for the retreat phase was to reward the right to retreat a unit to whichever power "bought" it in the preceding movement phase. This plan ran into trouble right away when we realized that it would often be a coalition of two or more powers that successfully bribed a unit. Our quick solution was to grant to the owner of the unit not the right to retreat his piece but instead the right to choose, from among the contributors to the successful bribe, the player to be given control in the retreat phase.

One would think that this short-lived rule would have been killed by the decision to make secret from a unit's owner the contributors to each bribe. However, the decision for secrecy had not yet been made, and the rule on retreat phase control was changed once and for all to allow a unit's owner total control for quite a different reason. Simply, it was decided that the rule should be changed out of interest in the speed of the game. We envisioned that a game would be held up if two "orders" had to be given to each retreating unit (first, the "order" specifying to whom the right to retreat should be given, and second — from what could be a different player — the retreat order itself).

In retrospect, it seems that we should have realized that the current rule was the proper way to go all along, drawing parallels between uncontrolled removal and the disbanding of a retreated unit by an enemy who wrested control of the unit from its owner — perhaps by as little as a single silver piece — during the movement phase. But, as you just read, that's not how it happened.

It seemed a happy coincidence that Switzerland is both impassable in Diplomacy and also the haven of monetary intrigue, and so Swiss currency was the first (and a natural) choice for the currency of Payola.

Francs, of course, seemed a bit small — and also too real-world — so the purposely vague "Swiss mega-franc" was invented.

As the players of the first few Payola games that I Mastered know, the taxes from owned SC's were actually paid yearly in the currency of the various European nations. For example, the yearly reward for controlling Bulgaria was paid in lev. (Actually, it should have properly been paid in mega-lev — an oversight of mine, I suppose.) And, as was apparent from the bank statements received by these same players, it just so happened that — strangely enough — the exchange rate for every different unit of currency with the Swiss mega-franc was permanently fixed at one-to-one.

I suppose some of these things grew to bother me, and I decided to rename the basic unit of Payola currency the "piece of silver." The only thing that initially held me back was the thought that the reference might be lost on players who may be unfamiliar with the Christian tale of Judas Iscariot's betrayal of Jesus in return for payment of thirty pieces of silver.

This was a minor concern, though, and so the shift from mega-francs to pieces of silver took place. I planned on replacing the currency abbreviation "MF" with "SP", but J. Andrew Lipscomb suggested the far better "AgP" abbreviation instead.

As you can see from the original rules to the variant, the initial scheme for calculating yearly income to each bank account involved paying the fixed amount of seven silver pieces for each supply center owned.

Seven was chosen for a couple of reasons. Having decided that income would be paid annually, and not before both Spring and Fall movements, it was felt that an odd number would make decisions on how much to spend for each of the two movement seasons that much more interesting for the player. The initial number was five silver pieces, but this was increased to seven in order to provide more flexibility to the player, and also since seven seemed a nice number, given the number of Great Powers in the game.

At the beginning, a concern that I had about the variant was that the extra income that Russia receives at the beginning of the game would constitute an advantage that would be difficult to overcome. I had nothing to back this up, and the first couple games got underway using this tax scheme.

The first-ever game, "payola", saw a strong Russian emerge immediately, and he seemed destined to win the game going away. Almost before the other powers had time to breathe, Russia was up to something like 15 supply centers. This confirmed in me that the tax income scheme needed to be re-adjusted.

However, as it happened, the game benedict actually finished (a Russian victory, though not a runaway) before payola, and in the End Of Game statements, I solicited from the players their opinions on the rules and their suggestions for improvements. It was at this time that I made public the "decreasing sequence" idea I had conjured up. The idea was well-received by the players (a bit less enthusiastically by the solo winner, who said jokingly that the game is hard enough to win) and so it was decided to try this new payment plan for the next game.

The new income scheme works very well, and it addresses the issue I hoped it would. A power with 17 SC's finds himself with less income than all the other players combined. It seems to encourage the "catch the leader" aspect of the game, and a large power with a chance to win and who knows how he intends to go about doing so is offset by a stronger coalition working against him.

Interestingly, the game "payola" is, at the time of this writing, still in progress, with Russia having been cut down to size by an alliance of France and Turkey, and so it seems that the "seven silver pieces per supply center" rule used by Payola Classic is not at all the impediment to play that I was concerned it would be. Payola Classic seems to be just as viable a game as is the grown-up Payola.

In fact, the mature Payola is sometimes played with just the single change to use the "seven AgP per center" tax income rule from Payola Classic. Such games are called "flat-tax" Payola games.

[Addendum: The game "payola" ended after the game-year 1922 in a 20-center victory for France.]

In many respects, the Master of a Payola game can be looked at much as a "banker" in a game like Monopoly. For this reason, and given the initial choice of "Swiss mega-francs" as the currency of the game, I adopted the persona Swissbancmeister. I knew, of course, that I was butchering the German language with this choice, but it had a ring to it.

When a Swiss resident (Mick Zwahlen, who is sadly missing from the net as of this writing, due, I understand, to sad circumstances) joined the fourth-ever Payola game, I took the opportunity to ask him to correct my pidgin German. Here, from a message he sent me, is his response:

If you want to write it German you should say "bank" instead of "banc." However, "Bankmeister" is a word that definitely is not German. "Zahlmeister" would be someone who pays ("zahlt") money to other people.

The original leader of the Swiss national bank is called "Präsident des Nationalbankrates" (president of the national bank's council) or even "Vorsitzender des Verwaltungsrates der Schweizerischen Nationalbank" (chairman of the governing board of the Swiss national bank which is similar to a CEO but more delegating than executing).

I decided to stick with Swissbancmeister.

The concept of transferring money from one account to another was a part of the variant from the beginning. It was a natural extension of the "each power manages something approaching a bank account" idea. However, the ability to transfer money to another power was always just mentioned as an afterthought, and I actually thought that it served little purpose and that it should be removed. After all, if two players wished to accomplish the same objective, both would be better served submitting their offers separately rather than together, since if one player handed his money to another, saying "attack our common enemy with this as you see best," he would be basically setting himself up to be stabbed with his own money.

When the first Payola game was getting started, I solicited the opinions of the players. Everyone seemed to agree that the transfer power would be used sparingly if at all, but didn't see any harm in leaving it. I believe that my decision at the time was that I would leave it in for this first game and that if transferral was never used in this game, the rule would be removed.

It seemed that in discussion, no one could list a single good reason to transfer money from one power to another. However, under game conditions, occasions arose that showed that it would have been a mistake to remove this ability from the rules. Money transferral is indeed a rarely used rule (though not as rarely as I initially thought), but it can provide some useful tactics. At this point, I can list five situations in which the monetary transfer rule can be advantageous.

1. After Elimination. One feature of Payola is that every power continues to maintain his bank account and do with it as he sees fit, even after having been eliminated from the game. A player who has been eliminated may wish instead to completely terminate his involvement in the game, and one way to do this would be to transfer all his remaining money to one (or more) of the other players, probably to those who the eliminated player feels are most likely to use the money to exact revenge for his fate.

2. Blackmail. I must admit, this reason took me by surprise. In the game graft, Russia was the victim of a 1901 stab at the hands of Turkey. Of course, due to the nature of Payola, the Russian player couldn't be sure that Turkey actually intended to take Sevastopol, but this technicality was disregarded and Russia demanded satisfaction from his invader. In a broadcast message after the Fall 1901 moves, Russia demanded that Turkey transfer a specified amount of money (more, if I recall correctly, than the income which Sevastopol would bring the Turk) into the Russian bank account immediately, and that if this did not happen, Russia would devote his entire game to the eradication of the Turk. Just to finish the story, Turkey resisted and Russia did indeed fulfill his pledge, harboring a gamelong grudge against the Turk, rallying to his side all who would listen. Every single one of the Russian bribes was spent against Turkish units, disregarding the invading German (who gained Warsaw, Moscow, and St. Pete at bargain prices), which eventually hastened Russian elimination.

3. Attesting Innocence. One effective use of the transfer power involves atonement for the deeds of one of your units. If a unit owned by one power takes a center owned by an allied power, this does not need to signal the end of an alliance. Since at least the monetary benefit of the center need not be kept by the power that took the center, the allies could laugh the whole thing off after a transfer from one ally to the other. This seems to be an effective way to hide a fake war while keeping the involved parties a bit more secure about their endeavor. And perhaps the war is not fake. In Payola, there are actually two other possibilities. First, the invasion could have been engineered by bribes offered by a third party. In this case, secretly transferring the tax income gained from the unexpected center back to its original owner can make things right. Second, the move may actually be a true, conscious attack on one ally by another. In this case, the transfer would serve as "evidence" that the invasion was not intended. It would seem to be a valid promise to make that when an alliance is formed, each power would agree to transfer any money gained from centers taken accidentally. Whether, for how long, and exactly why this agreement would be honored are of course other things.

4. Silent Partnerships. As I wrote above, it seemed that transfers for the purposes of combination spending was foolish, since all bribes from all quarters for the same order are combined anyway. The only explanation I could think of was an occasion when one power was unsure how to spend his money, but knew that he wanted it spent in concert with another power, and so he would simply transfer the money to this other power in full faith that it would be spent in his best interest. However, I saw any person who was in this position as being someone who wasn't paying attention to the game. I realize now, though, that there are indeed times when a wish to combine money with another power would be met by an inability to know exactly how to spend that money. These situations are no-press and no-partial-press games. Transferring money from one account to another would seem to be an excellent way to propose or confirm alliances in these types of games.

5. Buying and Influencing Friends. It's hard enough to make friends in Diplomacy, and the added subterfuge of Payola makes it even harder. After all, the player who professes the greatest friendship for you can actually be the one who is paying your units to do those evil things to their own cause, and you would never know it. One way to help make an ally more secure about the genuineness of your friendship is to send him some money. Making gifts contingent on another player showing his friendship ("Remove Clyde and I'll send you five silver pieces") is also an effective mechanism.

It was when the blackmail threat occurred in graft that I realized the transfer rule was here to stay. Just the fact that I myself hadn't envisioned anything like this ploy told me that the rule was a good one.

For a long time, there was an addendum to the transfer rule to the effect that no player could transfer money into an account of another player who, on the same turn, had previously transferred money into the first player's account. You see, I realized, when beginning to Master Payola, that the ability to transfer money from one account to another was one more burden for the GameMaster to bear. In fact, though I knew it was likely a light burden, this was the reason I contemplated removing the transfer rule. When the decision was made to keep it, I decided to regulate it with this "no returning transferred money" restriction. Doubting it would ever happen didn't keep me from thinking of the poor Master who had somehow, someway, gotten on the bad side of two of his players and for their revenge, they decided to transfer money back and forth thousands of times on each turn. Adding the restrictive phrase would at least give him refuge in the written rules from this headache.

This prohibition was lifted when the Payola Place Website was inaugurated. That site automatically handles transfers and never gets a headache.

Note that in some Payola games, a similar prohibition (to wit, that a player may only transfer money into each other account once per turn) can be used as a subvariant. For example, if used in a no-press game, this discourages players from using transfer amounts to communicate in coded messages, but it still allows for transfers to be used for damage reparations, offerings of thanks, and confirmations (or, rather, promises) of peaceful intent.

This rule was a later addition to the Payola rules, and was prompted by the extensions made to the rules to adapt Payola to the Blind variant.

When Payola and Blind came together, it was realized that the ability to offer zero silver piece offers to faraway units would allow a player to determine the locations of units he could not "see" without cost. By simply offering a zero silver piece gift bribe to a unit in every possible location on the board, and then waiting to see which units accepted the bribes, much of the blindness of blind would be gone.

Knowing that the ability to discern unit locations by offering bribes was still an important feature of Blind Payola, it was decided to simply disallow zero silver piece offers made to foreign units. A player could still determine the locations of units he could not "see", but it would cost him at least one silver piece.

After implementing this restriction in the Payola automated adjudicator, I decided that it deserved to be applied to all Payola games — not just Blind Payola games. I figured that in real life an army wouldn't really sell out for nothing, and the possibility of a zero silver piece offer deciding what a foreign unit would do was very remote anyway.

The bribe cost plateau rule grew quickly (though very late — 19 April 1999), but over a winding path. The ability to specify a minimum amount to which an offer would be reduced was proposed on the Payola Mailing List. The subject of discussion was whether to outlaw duplication of orders within the same bribe, and the conversation had gotten around to players saying, "if we could specify that a certain bribe would be reduced by two or three instead of one, it would be helpful". (This is indeed possible — see the annotation on order duplication mentioned above.) In trying to address this request, I proposed various things to myself, and eventually realized that a minimum bribe amount, in combination with the capability to list duplicate orders separately, would add much flexibility. For example:

5 : A MUN - RUH
5 : A MUN - RUH

5 : A MUN - RUH
5#2 : A MUN - RUH

The minimum soon evolved, however, to where it is no longer a hard minimum. This happened when it proved too bothersome, frankly, to programmatically determine the sum total of a player's minimums in order to see if he is still vulnerable to overexpenditure even if every bribe was reduced to its minimum. Accordingly, the minimum became a "plateau" to which the bribes reduce and remain as long as possible, but which disappears if all bribes reach their plateau without seeing the overexpenditure remedied.

This rule has a tortured history. Initially, in what became known as Payola Classic, the rule did not exist. I figured that if someone successfully bribed an army to walk into the sea, so be it. However, I had neglected to consider that the automated Ken Lowe judge program (which runs e-mail games, and to which the output of my initial Payola adjudicator was sent) makes some checks on the validity of all orders issued. So I found out the hard way (when someone tried to have an enemy unit do something impossible) that this was an issue to be resolved.

My initial plan was to convert all impossible orders to be issued into HOLD orders when transmitted to the Ken Lowe judge. There were some problems with this plan, though, since a unit ordered to move, even impossibly, is unsupportable, but a HOLDing unit is supportable. So I quickly decided that I could only accept what the judge accepts, and I broadcast this fact to my players. At this point, my Payola adjudicator was dumb when it came to the legality of the orders given it, so I relied on myself to catch these orders.

It was when the direct (colon) offer type was augmented by arrival of the newer offer types that even this "take what the Ken Lowe judge takes" rule became untenable. As you are probably aware, the Ken Lowe judge does not completely validate the orders sent to it, and some orders that are guaranteed void can indeed be issued. For Payola games, this was not good enough, and it became necessary to restrict the orders for which bribes could be offered to only those that are legal and valid. As a result, I undertook a long effort to add code to the Payola adjudicator to enable it, first, to understand what the orders being fed to it meant, and second, to disallow any order that could only be rendered "void." I can say with complete certainty only now — after looking hard at and updating my own code when the subject of no-press Payola came up — that the loopholes are all closed and the Payola adjudicator is tight as a safe.

For the logic behind the need for this strict "only valid orders are acceptable" rule, I include a section from my correspondence with Stephen Beaulieu concerning no-press Payola and what types of communication would be possible through Payola orders.

Say someone wants to convoy into Greece, but the Turk has an army there. The best way to do it, the invader figures, is to pay the Turkish army to move. He doesn't care where the army moves, though, because Bul, Alb, and Ser are all vacant and will probably (or better yet assuredly) stay that way. So he puts in an offer like 10 > A GRE - ALB ("ten to move to Albania or anywhere else for that matter"). All is well and good, and this ten silver piece offer is the best the Turkish army gets. That is, until some joker on the other side of the board decides to spend one silver piece trying to play some malicious joke on Turkey, say by putting in the offer 1 : A GRE - PAR. Now all of a sudden, the Grecian army gets eleven silver pieces to move to Paris, a bribe that bests all others. Of course, the end result is that "A GRE - PAR" is (*void*), so the army will end up staying in Greece — not moving at all! The would-be attacker is out ten silver pieces (which he promised to pay only if Greece moved) for a HOLD order and has only a bounce to show for the expense.

Now, of course, the joker on the other side of the board could still — even with the current strict rules — screw up the invasion with a single silver piece by ordering Greece to convoy out via the fleet that is carrying the incoming attack. The (*no convoy*) result achieves the same purpose as (*void*) and cannot be disallowed.

Despite this, it was felt that, more often than not, an offer containing a duplicated order is mistakenly entered, and that it should be rejected rather than accepted for consideration. Therefore, in order to specify quicker reduction of bribe amounts in case of overexpenditure, players must split a bribe into two or more smaller bribes, which will each be reduced by one.

An optional repetition count can be specified with an offer to cause it to be repeated automatically. This convenient shorthand provides for easy specification of this quicker reduction. For example, the offer:

Although I initially balked at the new prohibition against repeating offers in a chain of alternatives, the arguments made by the members of the mailing list were convincing that it's more often than not an error, and the rule was added. Later, I came to realize that only by instituting this rule can negative bribes be made as consistent in their semantics as possible with the other bribe types. Repetition of an order in a negative bribe was and would forever remain completely superfluous. To allow that orders could be meaningfully repeated in one type of offer, but have no meaning if repeated in another type of offer was an inconsistency that was easily and best avoided.

To illustrate the various types of bribes, we will set up a hypothetical situation. Let's say that France has launched an attack on England, and has surrounded the English army in London with two fleets — in the English Channel and the North Sea. Let's also say that an English fleet sits in Yorkshire and an English army occupies Edinburgh. At this point, France would like to capture London. This can obviously be done in a number of ways; we will choose only one.

Let's say that France would like to move his Channel fleet into London with support from the North Sea. After consulting his pocketbook, he decides to pay both fleets the price of seven silver pieces if they obey these orders. So he enters the following two direct offers:

7 : F ENG - LON 7 : F NTH S F ENG - LON

Notice that much depends on his choice of seven silver pieces as the bribe amount. If either fleet receives a different offer that would pay it more than seven silver pieces, the whole attack could fail. Also notice that even if the two bribes both go through, London may not be captured. Much still depends on the actions of the English army in London and the English fleet in Yorkshire.

The Frenchman could, of course, issue direct offers to each of these two English units, detailing what each of them should do to assist in the French invasion of their country. However, perhaps another bribe type is preferable. Indeed, as we shall see below, France has other forms of bribery in mind.

Continuing the discussion begun above regarding the planned French attack on an occupied London, we recall that France has issued two direct offers to his own fleets which, if accepted, will result in a supported attack on London. To ensure that this attack is sufficient, France would need to make sure that the English army in London is not supported in place by the Yorkshire fleet.

It is the case that all of the different offer types can be written as direct offers, but that the other types serve as useful shorthand. For example, if the French player wishes to make sure that the Yorkshire fleet does not support the army in London, he may issue a series of direct offers to that fleet, asking it to issue each order other than this SUPPORT order. That is, if the French player has decided that five silver pieces should be sufficient incentive for the Yorkshire fleet, he could issue the following set of direct offers:

5 : F YOR H 5 : F YOR - EDI 5 : F YOR S A EDI 5 : F YOR - LON 5 : F YOR - NTH 5 : F YOR S F NTH 5 : F YOR S F ENG - LON

This list, then, constitutes an offer to pay the Yorkshire fleet five silver pieces for any order it could issue other than the support for London. The astute reader will note that the Frenchman should remove from this list the offer to pay for F YOR - NTH, since this would cut the necessary support for the attack on London. So the list of direct offers should properly be written (this time, using the alternate syntax):

5 : F YOR H | - EDI | S A EDI | - LON | S F NTH | S F ENG - LON

However, there is a much easier way to express this same offer, and this is to use the negative offer type. Consider that what is really wanted is to promise the Yorkshire fleet five silver pieces if it does anything other than support London or move to the North Sea. This can be written as follows:

5 ! F YOR S A LON | - NTH

Notice that the negative offer above must be written on a single line, using the vertical bar. Consider the effect if the two parts of this offer were split into separate offers:

5 ! F YOR S A LON 5 ! F YOR - NTH

The first of these offers will pledge payment of five silver pieces for any order that is not a support of the London army. That is, if the Yorkshire fleet is ordered to the North Sea (which will cut the support for the attack on London), the French player would still pay it the five silver pieces. This is not what he wants to do. Likewise, the second offer means that if the Yorkshire fleet does support the London army, the Frenchman will also be on the hook for five silver pieces. Additionally, if the Yorkshire fleet issues something other than both these orders, the Frenchman will pay it ten silver pieces — five for each one of the two accepted offers.

Notice here that I have chosen this situation carefully. The number of valid order choices for a fleet in Yorkshire is much more limited than, say, an army in Galicia (or even in Yorkshire). The usefulness of the shorthand is much more apparent in most other situations. Indeed, even in the situation chosen, many potential valid orders to the Yorkshire fleet were left unmentioned in this example, a full list of orders being forbidden by the negative offer also includes:

The player who would rely entirely on direct offers is likely to find that when the orders go through, he has overlooked listing one of the many possible orders for the unit in question (to support some adjacent unit, to support some attack originating from a distant location, to convoy somewhere, to HOLD, etc.), and that this order spells doom for his campaign (not to mention to his pocketbook, since he will pay for the unit to issue the overlooked order).

We have been, in the notes above, planning a French attack on occupied London. We've come up with three bribes that the Frenchman will be offering to assure the capture of London. Indeed, these three bribes make the attack foolproof, but only if all three of the bribes are accepted by the units to which they are offered (and the North Sea fleet's support remains uncut). The Frenchman would be wise, however, to bulletproof his attack by offering a few more bribes in case one or more of the three we've written for him so far are not accepted.

The offers, as they stand, are for the English Channel fleet to move to London with support from the North Sea, and for the English fleet in Yorkshire to permit this supported attack to go off without interference.

The Frenchman, however, is probably unsure that the five silver pieces he's offering to the English fleet in Yorkshire will be sufficient to bribe that unit, and so he could be afraid that this fleet could act against his wishes and either support the London army or attack the French fleet in the North Sea. To account for this possibility, the Frenchman decides to offer some money to the London army. Consider that if this army moves out of London, the support from the North Sea will be unnecessary, and the Channel fleet is allowed entry (meaning that it is no longer important that the Yorkshire fleet not attack the North Sea). Consider also that even if the army attempts unsuccessfully to move out of London, any support given to London by Yorkshire would be rendered void (meaning that it is no longer important that this support be avoided).

Given this, the Frenchman decides to ask the London army to move, offering it four silver pieces to do so. The following direct offers are prepared for it:

4 : A LON - WAL 4 : A LON - YOR

Notice that once again, I have chosen the location of this situation so as to minimize the choices that the Frenchman would have to list. Were we working elsewhere on the board, the French player would have had to list upwards of a dozen different possibilities. In either case, the Frenchman has at his disposal the useful shorthand of the move offer, which keeps him from listing (or even attempting to list) all the possible locations to which a unit could be ordered to move. In this case, the offer to be issued to the London army is:

4 > A LON - WAL

which means that the Frenchman will pay four silver pieces if London is ordered to move to Wales (which is his preferred site) or to anywhere else for that matter.

The alert reader will note that Wales and Yorkshire, as listed above, are not the only destinations for a moving London army. To properly expand the move bribe offer we built ("4 > A LON - WAL"), we would need to list not only the direct offers to move to Wales and Yorkshire, but also to convoy to each and every possible location reachable via the North Sea and/or English Channel fleets. In the case of such convoy orders (since these orders, the Frenchman hopes, will fail due to lack of cooperation by his fleets), and also in the case of an order to move the London army to Yorkshire, the army will remain in (or may bounce back to) London, and the French attack will fail if the fleet in Yorkshire does manage to cut the support offered by the North Sea fleet. So in this case, the Frenchman might be wise to avoid the move bribe and simply issue the direct bribe 4 : A LON - WAL.

Just as the move offer type provides a useful shorthand to indicate that any movement order at all is acceptable to the offerer, the hold offer type provides the same shorthand for orders that leave the unit in place.

We've been discussing, in the above notes, how France would go about capturing London from England. Let's continue this discussion to explore how England may choose to defend himself. Of paramount importance to him could be that the London army not be moved, since if this happens, either of the two French fleets could enter London unopposed. So he could decide, hoping that three silver pieces are sufficient, to offer a direct offer to the London army as follows:

3 : A LON H

If the French offers are as described above, then three silver pieces are not sufficient — it would take at least four to prevent the army from moving to Wales. But of course the English player doesn't know this when putting his offers together, and anyway, he may have a limited amount of cash at his disposal.

Now, to contrive an example, let us say that the Turkish player, far distant as he is, has decided to have a little fun with the English player's units by asking them to perform useless but otherwise harmless actions. (Perhaps the English player is allied with the Turk, is aware of this practice, and has even condoned or suggested it, in order to make the English pieces look as though they are being bought off, and thus to drum up sympathy for the embattled island nation.) In keeping with this, Turkey has made the following toss-off direct offer:

1 : A LON S F YOR

By itself, this order will have no effect, since the four silver piece French offer outweighs it. If it could be combined with the English three silver piece offer, however, the French plan to move the London army would be foiled. If only the English player had thought to offer his direct bribe for more than just a simple HOLD, but for anything that would keep London in place. In other words, the English bribe would be better written as a hold offer:

3 @ A LON H

This way, the result will be that London will support Yorkshire, which is a much better eventuality for the English player than is moving to Wales and leaving London open for the French landing.

Let's go one step further. Rather than simply defending himself, England may see an opportunity to turn the tables on France. The English player might choose to issue the offer as follows:

3 @ A LON - ENG - BRE

This means that three silver pieces will be paid if either London convoys (or attempts to convoy) to Brest or stays in place. Although (if the other offers are all as listed above) the army will not choose to convoy, but instead to uselessly support the Yorkshire fleet, I have included this example to make the point that the order (or orders) listed in a hold bribe or move offer do not need to be only those offers that either keep the unit in place (in a hold offer) or those that call for the unit to move elsewhere (in a move offer). The offer above is a complete equivalent to the two offers:

3 : A LON - ENG - BRE 3 @ A LON H

As discussed above for the move offer, the hold offer is completely equivalent to a series of direct offers listing each qualifying order. The shorthand is useful for the very reason that the number of qualifying orders can be very large and difficult to completely list. (For example, ordering support to every army that could possibly be convoyed into each adjacent space, etc.)

"Why in the world would I want to just give my money away?" is a common reaction to the gift offer type. Indeed, the usefulness of this bribe type is specialized, to say the least. However, I can think of two reasons why this bribe type would come in handy.

The first use for the gift bribe is in Blind games. Much about the combination of Blind and Payola is described in other notes, but one feature of this game is the ability to obtain (via bribes) information about units that a player cannot "see". For instance, if the Russian player — confined to the eastern side of the board — would like to know if there is a unit in Portugal, he could offer such a unit a one silver piece gift offer, then wait to see if it is accepted. By using a gift bribe, the Russian is almost guaranteed not to affect in any way the order that will be issued to the unit (since his bribe amount is added to every other bribe received by that unit).

The second use for the gift bribe applies to the standard (non-Blind) game as well as to the variants. It involves a true apathy about what a given unit does, combined with an ardent desire to profess interest. Let's take an example of a player who, in true diplomatic style, is courting the affections of two warring nations. In the course of his artful dodging, this player finds that both sides in the conflict are interested in obtaining his financial support for their conflicting aims. Let us say that it is of paramount importance to one player that a certain unit is moved, and that the other player considers it imperative that it holds in place. Our hero, on the other hand, is only concerned with his diplomatic standing with both of the enemy nations, and honestly could not care less (despite all his verbiage to the contrary) what actually happens to the unit in question. So he issues the unit a gift bribe, knowing that whatever order it issues, he will be able to claim a share of the credit. He can accurately claim to each player that he lent his assistance, and can enjoy the celebration of a shared triumph with one player while agonizing with the other over their shared failure. Indeed, the owner of the unit (if he is the one whose order was accepted) will have concrete proof — in the form of the list of bribe amounts accepted by each of his units — that his own bribe was in fact augmented by the amount claimed by our hero. Conversely, if the enemy's bribe goes through, the owner of the unit will be certain that his own bribe, combined with the amount claimed by our hero, was not sufficient to gain control of the unit (he simply will not know, of course, that the same amount was contributed by our hero to the winning bribe as well). Using the gift bribe (or some combination of other offers like it) can ensure that no matter what happens, the fence-sitting player can almost certainly maintain his friendly position with both the combatants.

Like the move offer and hold offer, the gift offer is simply a shorthand for listing (in a series of direct offers) every possible order that could be issued by the unit. As discussed above, the shorthand is useful since that list of orders is almost always very large and difficult to completely list.

The gift offer is also a shorthand for a set of offers that is much less cumbersome, however. A gift offer is exactly equivalent to a direct offer to the unit to HOLD followed by a negative offer to the unit to HOLD, both having the amount specified in the gift offer.

Bruce suggested the addition of the functionality as a variant rather than as part of the standard game, but I felt that there was no reason not to incorporate it into the base Payola rules. The additions to the Payola code were relatively simple, although the changes to support this feature were much more widespread than were those of any previous release.

Bruce further suggested that players might be allowed to direct that their available monies be split into multiple "accounts," from which separate bribing campaigns could be conducted, each account able to be given its own separate savings requests. This, however, seemed to me to be a bit complex for the standard game, so if implemented, it would be as a variant.

When the first Payola game started, the observers indicated that their positions were a bit disappointing. They wanted to see the bribes that were given, or at least the amounts paid out. I submitted this request to the players, but it was vetoed, and so Payola observers shall forever remain in the dark until the end of the game.

Basically, initially the problem was logistical, having to do with the fact that Payola games were being run on Ken Lowe judges. Any press sent to observers only (through the Ken Lowe judge) will appear in the game history, and anyone (including the players) could ask the judge for the history and have all the beans spilled onto his plate. Apart from that, there is always the fear that a personal friend of one player could signon as an observer and pass information to the player.

Interestingly, however, I later added a variant to the Payola adjudicator that will cause generation of a broadcast message listing the total bribe amounts received by each unit on each turn. So it seems that the original set of Payola observers may yet get at least part of what they asked for, providing that the Payola games they observe are using this new (non-standard) variant.

The history of secrecy in Payola is much broader and deeper than the few paragraphs above. In fact, the original concept of the game had no secrets being kept by the Master at all. Not only would the total amounts paid to each unit be revealed to all, but the origin of the money as well. Support for this position eroded slowly and gradually, and credit for the present rule goes to the players of the first game, who voted on the proper way to handle the issue. At the time, I wasn't sure, but on reflection, it is very obvious that they made the absolutely correct call.

John Woolley is the brain behind the acceptance list. The need for some tie-breaking mechanism became obvious when we decided it would be too inconvenient to ask a unit's owner to choose between tied orders everytime such a situation occurred. John's solution, the acceptance list, is an elegant, simple, no-bother list that could even be submitted once and forgotten, and yet which guarantees that the orders for all units can be resolved from any set of offers. And it maintains the feature I like best in Diplomacy — "if things go wrong, you've got no one to blame but yourself."

My own (temporary) contribution to the acceptance list was far less impressive. In fact, I'm not even proud of it. I gave the thing its original name. I decided to call the list a "reference list" and immediately began a constant search for a new and better name. I chose "reference list" to try and keep the "bank account" theme going, intending to draw a parallel between the list and a list of credit references kept on file at the bank. But from the beginning, "reference list" was understandably confused with "preference list" (a well-known term to Ken Lowe judge players), and many Payola players would submit "preference lists" to me. Anyone familiar with the Ken Lowe Diplomacy judge knows that the words "preference list" already hold special meaning, and to re-use this descriptor would (did) play havoc.

I've since realized that the list is more like an "exchange or black market currency rate". It can be looked at as the ordered list of currencies that are practically or morally acceptable by a unit of a particular nationality. The reason why Turkey may be second on the Russian list is because the commanders of Russian units, whose nation enjoys an alliance with Turkey, can easily convert Turkish money to Russian money without suspicion (or without appreciable loss in the exchange). This is perhaps stretching things, but it does make sense that some currencies would be more readily accepted by some armies and not by others, and this nicely explains the nice, tidy tie-breaker list.

However, I was at a loss to come up with a short, proper name to convey this concept, and so it remained "reference list" for a very long time.

Enter Mike Connaghan, who suggested "acceptance list". Needless to say, I found this term quite appropriate and very fitting, so it was adopted. To quote from Mike's suggestion: "[the list represents an ordered sequence of] the currencies that are most acceptable to the commanders. Another way to look at this concept is that, sure, commanders will take money to convoy enemy armies to the homeland, but they have a little bias for friendly countries as long as they're not too stingy to come up with an equal offer."

Thanks, Mike!

Mario Huys created the Void variant, a variant with a whopping 36 powers. The acceptance list for Payola play of this variant would be ugly to look at and potentially heck for a player to maintain. Mario suggested that players should be able to provide an acceptance list that does not include all the powers, and that those powers not listed be placed automatically at the end of the list in random order. I decided to expand upon this idea to allow the player to use a single question-mark, not just at the end of the list, but anywhere, to represent a random sequencing of all omitted powers.

I liked this so much that when I implemented it on 5 September 2004, I modified Rule 4.2 to state that the initial default value of each player's acceptance list (at the beginning of the game) is simply that specific power followed by a question-mark. As examples, the initial acceptance list for England is E?, for Russia is R?, and so on.

As a historical note, before this change was made, the initial (game-start) acceptance list for each power consisted of all seven letters, in alphabetical order (wrap-around) beginning with the power in question. That is, England's initial acceptance list was EFGIRTA. While it has always been incumbent on a player to adjust his acceptance list ("if anything goes wrong, it's your own fault"), I always knew it was wrong to assume that unless the player changed his default, every Englishman's second choice was French money, and every Russian's second choice was Turkish money, etc. The question-mark shortcut lets me feel much better about the default, since truth be told, very few players seem to pay much attention to their acceptance list.

The first and by far the most common method of determining which order will be issued to a unit is to simply discover that more money would be paid by the player(s) for one order than for any other. In other words, there is no tie-break involved. To illustrate this situation, consider the following offers:

Offers Made to the French Fleet in Brest

French Offers	5 : F BRE - ENG
English Offers	4 ! F BRE - ENG
German Offers	2 : F BRE - GAS

In this case, the fleet in Brest would receive five silver pieces (from the French player) to move to the English Channel, but six silver pieces to move to Gascony instead. So the unit is ordered to move to Gascony.

Occasions obviously arise when the bribes for more than one order to the same unit sum to the same total, and each of these amounts to more than any other of the offers. In this case, the "acceptance list" must be consulted to determine which order should be issued. Consider the following example:

Offers Made to the French Fleet in Brest

French Offers	5 : F BRE - ENG
English Offers	4 ! F BRE - ENG
German Offers	1 : F BRE - GAS

In this case, the fleet in Brest is in a quandary, having received two offers that total five silver pieces each. To break the tie, the French "acceptance list" must be consulted. Let us assume that this acceptance list is the following sequence of powers — FRITAGE (meaning that French offers are to be regarded as preferable to Russian offers, which are in turn preferable to Italian offers, etc., etc.) From this, we can easily see that the French fleet sets sail for the English Channel, since French money (the most preferred to French units, according to the acceptance list) is involved in this order.

Consider now, by way of contrast, the following set of offers to the same unit.

Offers Made to the French Fleet in Brest

French Offers	1 : F BRE - MAO
English Offers	4 ! F BRE - ENG
German Offers	6 : F BRE - ENG
Italian Offers	2 : F BRE - GAS

This time, we see that the fleet is given four choices: to HOLD (for four silver pieces of English money — this is the "hidden" order behind the English negative offer), to move to the Mid-Atlantic Ocean (for a total of five silver pieces contributed by France and England), to move to the English Channel (for six silver pieces, offered by the German), and to Gascony (for six silver pieces from the English and Italian accounts). The HOLD and the move to the Mid-Atlantic are both outbid by the other offers, and so they are not considered as options. Both the remaining two orders (to move to either the English Channel or Gascony) would pay the unit six silver pieces, and so the acceptance list of the French player is consulted. Assuming this acceptance list is as given above, and since none of the orders involve French (the most preferred) currency, the order backed by Italian money (to move to Gascony) is preferable to any order that is not. So the unit is ordered to Gascony.

In Payola Classic, in which only one offer may be given to each unit by each power, the need to resort to any tie-breaker but the first is obviated. However, when this restriction was lifted, the single tie-breaker became insufficient. Consider the following simple example.

Offers Made to the French Fleet in Brest

French Offers	5 : F BRE - ENG | - GAS

Here, we see that the French fleet would be paid five silver pieces to move to either the English Channel or to Gascony, and that, since this money would be paid by the same power, consultation of the acceptance list would not help to decide which order should be issued.

The solution was to decide that the sequence of the various offers given by each player is significant. In this case, the fact that the order to move to the English Channel is mentioned before the order to move to Gascony decides the matter. The unit is ordered to move to the Channel.

Note that this rule has no regard for the amount of the various offers made by any particular player — only the position of the offers in the list. Consider the example below:

Offers Made to the French Fleet in Brest

French Offers	1 : F BRE - ENG 3 : F BRE - MAO
German Offers	2 : F BRE - ENG

Here, the tie between the two choices (both of which would pay three silver pieces) is broken by this "position in the list" rule. The fleet in Brest will be ordered to the English Channel despite the fact that the first power in the unit's acceptance list would pay more for a move to the Mid-Atlantic than it would for the move to the Channel. All that matters is the position of the offers.

The second tie-breaker can be a confusing one when negative offers are involved. To assist, each negative offer should be considered to be two offers, the first being the instruction to HOLD if no other offer is given, and the second being the instruction not to perform the specified action(s). The sequencing of these two orders (one of which is implied) can be important.

For the hand-adjudicator, it is important to note, then, exactly how each negative offer should be transcribed. Consider the negative offer "4 ! F BRE - ENG." This is converted into the following two offers: first, an offer of zero silver pieces to HOLD, and second, an offer of four silver pieces to do anything other than move to the English Channel.

Note that the reason this first offer is for zero silver pieces is that the offer to expend four silver pieces for a HOLD is implicit in the other part of the offer. Note also that the prohibition against zero silver piece offers being made to foreign units does not apply to this case.

Similarly, the negative offer "4 ! F BRE H." would be expanded into (first) a zero silver piece offer for the unit to HOLD, and (second) a four silver piece offer to pay for anything other than a HOLD, Despite the directive not to HOLD, the offerer also is put on record as asking the unit to HOLD (without promise of payment).

Because of this "expand the negative offers" rule, a unit that receives no offer except for a single negative offer will still have a direct offer (the implicit HOLD) to consider (and accept).

The third and final tie-breaker is not necessary in Payola Classic. Just as the second tie-breaker became necessary due to the addition of multiple bribes from one power to a single unit, this third tie-breaker is required because of the addition of the negative offer type. Basically, the situation that calls for this final tie-breaker is exemplified by the following:

Offers Made to the French Fleet in Brest

French Offers	5 ! F BRE - GAS
English Offers	1 : F BRE - PIC
German Offers	1 : F BRE - ENG

Here, France (who is positioned at the top of his own "acceptance list") has offered five silver pieces to the Brest fleet to do anything other than move to Gascony. This gives the unit two competing six silver piece choices: to move to Picardy or to move to the English Channel. The fact that France occupies the first spot in the "acceptance list" in question is of no help here, so subsequent powers down the acceptance list must be consulted. In this case, if we assume that the French acceptance list is FRITAGE, the German offer is preferred over the English offer, so the unit is ordered to move to the English Channel.

Notice that had France also submitted a one silver piece direct offer for the unit to HOLD, then this would make a three-way tie of six silver piece offers, and the unit would be ordered to HOLD (a decision made at the first tie-breaker).

The interesting thing is that this third tie breaker, in combination with the method of expanding negative offers, guarantees that there will be a single order issued to every unit. No further tie-breaks are necessary.

I can't take credit for this gem of a rule. John Woolley came up with this mechanism for preventing overexpenditures. To me, it seemed like he tossed it off as a whim; likely he thought about it first — maybe even quite a bit. Either way, I consider it a stroke of genius.

One of our complaints about Intimate Diplomacy was the penalty for overbidding; I don't recall exactly what it is, but overbidding seemed to be a death knell. We didn't want anything like this in Payola, yet we knew, of course, that there must be some provision or procedure for cases where a power would find himself with more money promised than he could pay out.

I don't recall any consideration of deficit spending, nor of limiting the amount that could be offered to the amount on hand. We considered both of these to be inferior solutions, and we knew that the ability to offer more money than you had was an important one — one that required that more unconstrained thought be put into each offer. As I say, it seemed to me like John came up with this elegant solution ("reduce all offers by the offending player by one silver piece and repeat if necessary") in the wink of an eye.

I can't imagine any better way to resolve this issue. To illustrate the principle, consider the following set of offers:

Offers Made to All Units

French Offers (Bank Balance: 20)	10 ! F BRE - GAS 15 : A PAR - BUR
English Offers (Bank Balance: 20)	8 : F BRE - GAS

Here, the Brest and Paris units initially jump at the ten and fifteen silver piece offers. However, it becomes obvious that the French player cannot honor both of these. One can imagine the French leader sheepishly re-approaching the leaders of these two military units asking them to accept less than they were promised. Instead of ten silver pieces, the fleet in Brest is offered nine, and instead of fifteen, the army in Paris is offered fourteen.

At this point, the French offers are still the best offers that these two units are given, and so both would be accepted. Once again, however, the French bank balance will not cover the planned expenditure, and so the process is repeated and the units are re-offered eight and thirteen silver pieces respectively.

Yet again, the French money would be accepted (the money for the fleet via tie-breaker), but it cannot be (since this still would bankrupt the French player). As a result, the French offers are reduced once more, this time to seven and twelve.

Now France would be happy (and finally able) to pay the money offered. However, the fleet in Brest now finds that it can get more money from another source, and so it takes the eight silver pieces offered by the English player and heads for Gascony. Paris is ordered to Burgundy at a cost of twelve silver pieces to France.

The situation that required this particular wrinkle in the rules to be decided and described happened in a Tin Cup Payola game in early 2010. To see the effects of this rule, consider the following example, in which (as the final result) the Russian army in Warsaw will be ordered to Galicia.

RUSSIA
ACCEPT R?TAE
2 : A WAR - LVN

TURKEY
9 : A WAR - GAL (but this Turkish bribe reduces to 0 due to overexpenditure)

AUSTRIA
1 : A WAR H

ENGLAND
2 : A WAR H
3 : A WAR - GAL

Notice that the Turkish bribe still determines the outcome of the tie-break (as per rule 5.2) despite the fact that it has reduced to a zero AgP bribe (which could not have been initially offered as such).

Blind is a Diplomacy variant in and of itself. In Blind Diplomacy, the players are not told where all the units on the board are — only where their own units are, what is in the adjacent spaces, and what is in their own home country and the spaces adjacent to it. That's simplified a bit, but it covers it.

This rule was added during the 25th anniversary year of the publication of these rules, which goes to show that you really can never think of everything. A bug reported by a player in a (non-blind) Payola game being run on 19 May 2020 was the cause for this rule. The bug concerned the behavior of the order validator when a unit type was not given in a HOLD order that was (accidentally) offered to a unit that didn't exist. In non-blind Payola, this is an error that must be corrected, and the adjudicator did indeed say so, but it reported a misleading (okay, flat-out wrong) cause for the rejection of the offer. While fixing the bug, it was realized that calling out the proper error (i.e., "hey, there's no such unit there") would definitely not be good to do in Blind variant games, since this would allow a player to simply experiment with his offer sheet until he knew exactly which units existed on each location on the map. It seemed the best thing to do in cases where, for example, a non-existent TUN H was offered money (without any unit of either type in Tunis) would be to have the validator smile to itself and go on validating the offer without reporting any error to the player. However, it soon became clear that the orders in the offer could not be validated without knowing the unit type for which it is intended — for example, TUN -TYS will be ruled invalid if there actually is an Army in Tunis.

The solution, of course, was to require that Blind Payola players be explicit about the type of unit to which they are offering money. If you want an unseen unit that may or may not be in Tunis to move to North Africa, you will need to issue two offers: one to A TUN and the other to F TUN.

Actually, a better solution was identified that would have allowed unit types to be omitted, but coding it in the DPjudge proved harder than I expected, and so until I get more energy, it stays this way.

In standard Payola, the message sent to each player upon payment of bribes from that player's account includes an itemized list of payments. Each outgoing bribe is detailed down to what order the accepting unit will issue.

In Blind ("Tin Cup") Payola games, this proved to be too much information. Consider the case of a French player who has no tangible information as to the location of units in the Balkans. Before this rule was adopted, the French player could have — by simply making the gift bribe offer "1 & A SER H", and at a cost of only one silver piece — received a message not only telling him whether there is a unit in Serbia but also what that unit did.

Obviously, this took the "blind" out of "Blind," and so the rule limiting the information returned for offers in Tin Cup games was instituted. Note that despite this restriction, bribery is still a powerful intelligence weapon in a Tin Cup game. For example, for a player with no knowledge of Balkan unit positions to determine whether there is a unit in Bulgaria, the following three offers would suffice:

1 & A BUL H 1 & F BUL/sc H 1 & F BUL/ec H

With these three offers, the offering player will spend only a maximum of one silver piece, and will be guaranteed to learn whether or not there is a unit in Bulgaria, and what type of unit and on which coast it began the turn if a fleet. By this rule, though, the offering player is not told exactly what order was issued by a unit that accepts this single silver piece bribe.

To make up for a bit of what was seen as overcompensation, the nationality of the unit accepting a bribe is also reported to the power paying the bribe. It just made sense that covert intelligence operatives would be able to send back a bit more than "Well, I see an army here, but I have no idea whose army it is." So not only does the above set of three offers return the information about whether or not such a unit exists, but also who owns that unit.

This rule does not make it impossible to also find out what order the unit actually issues; it only makes it more difficult and more expensive. This also makes real-world sense — a spy can be paid to look through a fence and obtain information about a unit's existence and type and ownership at far less cost than it would take to pay this spy to actually infiltrate the unit and learn its plans for the coming season. For example, if a player having no knowledge of the Balkan situation wished to determine not only whether there is a unit in Serbia but also — if there is — something about what it is ordered to do, this player could enter something like the following hefty set of bribes.

1 @ A SER H 2 : A SER - TRI 3 : A SER - GRE 4 : A SER - ALB 5 : A SER - RUM 6 : A SER - BUL 7 : A SER - BUD

With this set of offers, the player will know whether there was an army in Serbia, and, if there is, he will also know (by cross-referencing the offers he made with the amount of his money accepted by the unit) whether the army was ordered to stay in place or, if not, the location to where it moved (or attempted to move).

This is still incomplete information, since without issuing a similar — actually, an even more extensive — set of offers to everything else in the region, the player cannot know if the move succeeded or bounced, or even if the Serbian army itself was dislodged. There is also the disadvantage not only of cost, but that these intelligence gathering offers would actually decide the order issued by the unit — perhaps to the detriment of the offering player. However, even the ability to get (purchase) this amount of guaranteed correct information is a unique addition to Blind games, and this rule seems to provide the proper mix between paid intelligence services and the work-in-the-dark atmosphere of a Blind game.

Frankly, I've found Tin Cup Diplomacy to be the most fun way I've ever played. In it, you obviously need to use your diplomatic skills to get information about the placement of other pieces, and you also have a treasury at your disposal to help you do the same. The espionage and counterespionage options available in the game are tremendous. It seems to force a level of strategic thinking, deliberate planning, intelligence gathering, counterintelligence dispersal, intrepid diplomacy, calculated risk-taking, and careful execution which (I imagine) closely mimics real-life warfare.

In the tradition of the online auction, the idea here is that the amounts being offered are maximum offers. The winning offer is automatically reduced to just a small amount (one AgP) above the highest non-winning offer.

For example, if England's offers include:
     10 : F LON - ENG
and France's offers include:
      3 : F LON - NTH
then in eBayola, London will move to the Channel but England will pay only four (one silver piece better than the second-best offer) instead of the ten he offered.

Savings are awarded to players in acceptance list order. For example, England offers:
     10 : F LON - ENG
France offers:
      3 : F LON - NTH
Germany offers:
      5 : F LON - ENG
London will move to the Channel — England will pay zero (his bribe being completely reduced away and the offered amount still sufficient to better the bribe to go to NTH) and Germany will pay four (his bribe having been reduced the one more it took to get the Channel bribe down to "one silver piece better than the second-best bribe").

It is important to note that the "second-best" bribe made to a unit (which determines in eBayola the amount to which the winning bribe will be reduced) is not decided upon at the same time the winning bribe is determined. Here is an example that shows why an extra bit of analysis is necessary. The only bribes that Fleet London gets come from England:
     10 : F LON - ENG
     10 : F LON - NTH
The "second best" bribe is the North Sea bribe, but unless a little more thought is given to the matter, England will still pay ten in eBayola for the Channel move (because reducing the Channel bribe by even one would put it "at or below" the second-best bribe).

You can tell by looking at things, though, that England shouldn't pay the full ten. If the only person he is "bidding against" is himself, that should not be held against him, keeping his offer high. To avoid this, the following occurs:

1. The winning bribe is determined.
2. Any contributors to the winning bribe then have any contributions that they made to other non-winning bribes to the same unit reduced by up to the amount that they offered to the winning bribe.
3. Then the second-best bribe is determined.

In the example above (with F LON receiving only the two orders from England), the North Sea bribe is reduced to zero. It is still the second-best bribe, of course, but the Channel bribe reduces down to one instead of staying at ten.

Another example — England offers:
     6 : F LON - ENG | - NTH
France offers:
     2 : F LON - YOR
     1 : F LON - NTH
Germany offers:
     2 : F LON - YOR
The winning bribe here is the North Sea move, this time at an (original) cost of seven (six offered by England, one offered by France). Without the extra step, the "second-best" bribe is the Channel move (six AgP). But before the "second-best" bribe is determined, all non-winning offers to F LON are reduced by (up to) six English AgP and one French AgP (the make-up of the winning offer). After doing so, the second-best offer is found to be the YOR move, which totals 4 AgP. Therefore, the cost assessed for the North Sea move is 5 AgP — four paid by England and one paid by France.

So, even though England had offered to pay up to six AgP for an order (to move to the Channel) that was not issued, the end result is that he is charged four AgP for the order that was issued (which accepted bribes totaling five AgP). Without an understanding of this point of the rules, then on receiving his financial report after the move is adjudicated, the English player might say, "this is wrong! I would have paid even more for my first-listed (preferred) move; to the Channel!"

The second subsurface eBayola implementation detail concerns the interaction of bribe plateaus with the eBayola bribe-cost reductions.

Essentially, bribe plateaus are obeyed...except for any plateaus given by the owner of the unit. (Unless a non-standard variant rule is in use which causes the total bribe amounts accepted by all units to be disclosed to all players; in this case, even the plateaus requested by the unit owner are obeyed.)

An example should clarify. England offers:
     6#4 : F LON - ENG
France offers:
       3 : F LON - NTH
Germany offers:
     6#6 : F LON - ENG
The result is that London moves to the Channel. England doesn't pay a penny for it (because F LON is an English unit, England's plateau is ignored and his offer is reduced from six to zero) and Germany pays six (his plateau is obeyed, and even though his bribe could have been reduced to four, it is not).

As mentioned above, if the game is being run in such a way that the total accepted bribe amounts are being disclosed publicly, the English plateau would have been obeyed as well, which would mean that the Channel move would be made at a cost of four to England and six to Germany.

Obviously, using plateaus in eBayola to pay more than you really need to pay can cause the unit owner to mistakenly believe that his units were the subject of a heated bidding war, perhaps inducing him to begin devoting more of his budget toward that unit in the future.