Archived from groups: rec.games.diplomacy (More info?)
Inside the Adjudicator's Head and Primary Paradoxes
InsideTheAdjudicatorsHeadAndPrimaryParadoxes.pdf
This article may be read at the following web address:
http://www.geocities.com/diplomacy2007/InsideTheAdjudicatorsHeadAndPrimaryParadoxes.pdf
In many ways, this article is elementary. It's first function is to
help me understand how an adjudicator realizes the orders of the
military units without using an automated computer program. Keep in
mind that there already exist well thought out and sophisticated
algorithms, such as the Diplomacy Players Technical Guide (also
referred to as DPTG).
http://www.amarriner.com/dip/dptg.php?games_id=22
Also please keep in mind that I am still very new to the game of
Diplomacy and may, on occasion, make rather crude errors in my examples
and thinking. This is due to the fact that Diplomacy is a rather
difficult game to come to terms with, but hopefully no such crude
errors will make it past the editing process.
Other related topics, such as primary paradoxes, will surely be
relevant. Let's first specify which paradoxical items we will not be
discussing in this article (though these items probably will be
addressed separately at a future time). We will not at this time be
interested in paradoxical situations and rule book ambiguities and
inconsistencies dealing with the following topics:
i. What the actual operational order is for a military unit and whether
or not it is valid and how this effects other support orders given to
other units. For instance, we will not be concerned about whether a
unit in Norway can or cannot be supported by other units if the unit in
Norway is ordered to move to the planet Mars (the planet Mars is not a
province on the Diplomacy map).
ii. Situations where an army can be convoyed by more than one convoy
will not be discussed at this time.
iii. Situations where an army can be convoyed to a province that it
could also legally walk to will not be discussed at this time.
iv. Situations where an army might be kidnapped by one fleet or another
will not be discussed at this time.
v. Situations involving issues related to self-dislodgement, dislodging
military units that are of your country, supporting the dislodgement of
military units that are of your country, and convoying in troops not of
your country which attack a province either held by a unit of your
country, supported by a unit of your country, or attacked by a unit of
your country.
In short, this is why the title of this article concerns "Primary"
paradoxical situations. This is not to say that the above issues are
not important, for they are; however, I have arbitrarily decided that
they are not the primary concern and focus of my current research
(though some of these topics may be researched in the future).
Concerning primary paradoxes (as defined above), we will be concerned
only with either the 1971 rule book or the year 2000 fourth edition
rule book. Actually, since the year 2000 fourth edition rule book is
freely available on-line from the publisher in Adobe Acrobat PDF
format, why don't we say instead that we will only be concerned with
using the year 2000 fourth edition rule book for the duration of this
article.
There will probably only be a couple rules of primary importance to our
investigation of primary paradoxes and these are now listed from the
year 2000 fourth edition rule book:
QUOTE: Page 10.
[Quotes not shown in newsgroup format, please see my web site]
UNQUOTE
QUOTE: Page 12.
[Quotes not shown in newsgroup format, please see my web site]
UNQUOTE
QUOTE: Page 16.
[Quotes not shown in newsgroup format, please see my web site]
UNQUOTE
Using the above quoted rules, and other more standard rules which most
people are already familiar with, we have adequate material to create
paradoxical positions on the Diplomacy board.
This article is not meant to be definitive; that should, I hope, be
obvious. This article is more introductory in nature, and by writing
and thinking on this topic, I hope to be better prepared to understand
algorithms which have previously been written by others and which are
already in existence.
Finally, I can't emphasize enough how important feedback is in my
learning process. Your comments, suggestions, corrections in reasoning
and logic, and general contextual information is, and has been in the
past, exceedingly important as I try to come to terms with the game of
Diplomacy.
Let's begin!
What does it mean to say that a given game of Diplomacy has entered
into a paradoxical situation? In my first introduction to paradoxical
situations on the internet, the reasoning was given in an English
statement, something like this: "The army was convoyed across and
attacked its destination province cutting the support being offered by
the target province; and, through numerous effects, too long to
mention, it then says, and thus the original convoy could never have
existed. We are at a paradox: how did the army get convoyed across
when the convoy could never have existed?" Actually, the examples I
came across on the internet were far better written than the sample I
just gave.
But, my point is this: for a minute, let's forget about trying to put
it in English. Let's define a paradoxical situation a little more
formally, such as like this: the game of Diplomacy enters into a
paradoxical situation when legal move orders cannot be executed without
violating important, fundamental rules of the game.
So, if an army was convoyed using two fleets, then after you have
executed all the legal moves, none of those two fleets must have been
dislodged, otherwise the convoying of the army could never have taken
place. As another example, if a convoying fleet is supported by
another fleet, and the convoying fleet is attacked with a force of two
by the enemy, then it is essential that after executing all the legal
moves wherein one of the moves represented a convoying of an army using
the convoying fleet, that this said supporting fleet not have had its
contributory support function cut, otherwise the convoying fleet would
have been dislodged, and the convoy could not have taken place. This
paragraph may not be crystal clear; hopefully the examples to follow
and the annotation of the adjudicator's thinking process will make all
this clearer later in this article.
Let's speculate what the function of these important, fundamental rules
are: I believe that these important, fundamental rules ensure that
when all legal moves are executed, it will be as if the army units had
been simultaneously thrown into action, even if in actuality the
adjudicator moves one unit at a time until all legal moves have been
executed on the Diplomacy playing board.
Why is it important to create a game and to create rules where the
simultaneous movement of the military units of the board is important?
I'm not sure at this time. But, as openers, each players order's are
revealed simultaneously. One reason might be that in primary
paradoxical situations, if one player moved before another, the player
that moved first would have a distinct advantage. In situations not
containing a paradox, I don't think it really matters who moves first,
the result should be the same as if all the units moved simultaneously
(but I have not yet thought about this overly much).
What is an important, fundamental rule? Well, pretty much all the
rules having to do with how to move a unit, how to support a unit, how
to cut the support being provided by a unit, convoying rules,
dislodgement rules, and so forth.
By the way, I forgot to mention that we will not be investigating any
cases having to do with what happens after it is determined that a
military unit has been dislodged. This will keep our discussion
tightly knit.
The rest of this article, or parts of it, could be considered rather
elementary as I warned earlier. I will present examples (not all of
them paradoxical), and speculate as to the reasoning process carried
out by the adjudicator. In this way, readers of this article can
correct any misunderstandings I have about this process.
For each scenario or board position and orders given, there are no
military units on the playing board but those explicitly mentioned.
Scenario:
France: Army Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
Piemonte is under attack by two contending military units of equal
strength. The adjudicated result is that no military units move.
Note that even at this stage, if we are not careful, we can adjudicate
this simply scenario incorrectly. Here is an example. The adjudicator
first moves the army in Venezia to Piemonte. Then the adjudicator
attempts to move Marseilles to Piemonte, notices that an Italian force
is already there, and so since the French army did not have a greater
attacking force, determines that the French army did not move. The
result would be that the Italian army moves into Piemonte and that the
French army did not move. This is an example of an incorrect
adjudication.
Note that if we continue to adjudicate incorrectly and start the
adjudication process over again, but this time moving France first,
then the result would be that the French army moved into Piemonte and
that the Italian army did not move.
An implied rule, I speculate, is that the end result of conducting all
the legal orders and moving all the units about must be identical
regardless of which unit is moved first on the playing board (whether
the unit is physically moved or whether the unit is moved in one's
head during a calculation). To do item: I will have to read over the
rules to see if this so-called implied rule is actually a rule which
appears in writing in the rule book.
Assuming that the above stated implied rule is correct, then we have a
simple algorithm to help check our adjudication process: if a
different ordering of how the units are moved about on the board
results in a different final result, then the adjudication process is
in error. By the way, we should add that at this time we are only
taking about situations which do not create primary paradoxical
situations.
Note that the above algorithm uses the wording, "final result."
This could theoretically mean that two different adjudicators each
using his own thinking process and with each thinking process being
different, might potentially end up with the same final result, even
though the process that was used is not the same. To say it another
way, assuming that a process of adjudication could be labeled
incorrect, the incorrectness of that process is not always apparent if
you only look at the final result. If we assume that the incorrect
adjudication process is a computer program, then this effects our
testing, for we may incorrectly conclude that our faulty adjudication
program works because in some particular scenario the final result was
identical to a correctly thought out adjudication.
It will be noted that at my web site,
http://www.geocities.com/diplomacy2007/
there are a number of rule sets or algorithmic systems or adjudication
procedures listed. Some of these rule sets do not come with detailed
examples showing the reasoning process from the very beginning to the
very end of an example. Thus, this makes them harder for me to
determine that I understand the algorithm correctly; and on top of
this it is possible that an incorrect interpretation of a rule set
might in some scenarios result accidentally in a correct final result,
then I would have no way of knowing if I was carrying out the rule set
correctly. So, it is my opinion that it is fundamentally important
that these rule sets have a sufficient number of scenarios with the
details of the thinking process spelled out.
Lets go back to our first, simple example again.
Scenario:
France: Army in Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
Drastically simplifying the rule book concerning standoffs on page 6 to
relate to our simple example, it says that when two unsupported units
both attempt to move into an unoccupied province, this results in a
standoff and none of the units moves.
While I have not at this specific time looked through the rule book
again to see if it explicitly says that when adjudicating a position,
the final result cannot be different depending on which units were
physically moved first on the playing board, I took this as an
operating assumption. However, let's forget about this for a moment
until some mathematician has shown us the proof. So, we are stuck now
with only relying on the rule as I paraphrased it just above.
So, it would appear that the correct adjudicational reasoning would be
annotated thus:
i. The adjudicator notes that two military units have been ordered into
the same, unoccupied province (in this case that province is Piemonte).
ii. The adjudicator now simply looks at the rule book to see the
result, and the rule book says (in perhaps my grossly simplified
manner) that if two units of equal strength attempt to move into an
unoccupied province that none of the two units moves.
By the way, as an aside, consider the following.
Scenario:
France: Fleet in Spain(nc) to Portugal.
Italy: Fleet in Portugal to Spain(sc).
The correct adjudicated result is that no unit moves. This is a
standoff (though the rulebook says that in a similar scenario it is an
example of two units not trading places). I mention this because I saw
a similar example in the rec.games.diplomacy newsgroup, and it is not
hard to imagine anyone simply moving the units and not noticing that
the units are really involved in a standoff (or if you prefer, one
might not notice that in moving the units the units had illegally
changed places), just as you would have in the following.
Scenario:
France: Fleet in Mid-Atlantic to West Mediterranean
Italy: Fleet in West Mediterrranean to Mid-Atlantic
Now back again to our first example.
Scenario:
France: Army in Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
Let's now introduce the concept of an ordered to do list. We will
introduce this concept to see if it might potentially become a way that
we can adjudicate this simple scenario as well as far more complex
scenarios. Think of the to-do list as a stack of index cards on your
desk. To execute the to do list, you take the first card off the stack
and do what it says. To put an item into the list, you write down the
task on an index card and place it directly on top of the to-do list
stack.
So, at this stage, all the legal orders are known for this example.
Scenario:
France: Army in Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
You note that two armies are ordered to Piemonte. You create the first
to-do item: "01--Who takes Piemonte, if anyone?" Because there
are no other units on the board either moving, offering support, of
breaking support, there is nothing else to do; so, you now take your
one to-do item off the list, this being "01--Who takes Piemonte, if
anyone?" You execute this to-do list item by answering the question.
In this case, because the units attempting to move into Piemonte are
of equal strength, none of them move.
Here is the same scenario but with France's army's movement being
supported from the sea. We will work through this again using the
concept of an ordered, stacked to-do list.
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marseilles to Piemonte.
Italy:
Army in Venezia to Piemonte.
We note that there are two armies contending to occupy Piemonte, so we
write our first to-do list item and place it in the stack: "01--Who
takes piemonte, if anyone?" We now look around to see if any other
activity on the board effects the resolution, and we note that the
French support from the Gulf of Lyon must be considered. So, we add
that to the ordered, stacked to-do list: "02--Does the Fleet in the
Gulf of Lyon successfully support the Army in Marseilles to
Piemonte?"
There are no more units on the board to consider. Now let's start
resolving the to-do list items in our ordered and stacked list. First
lets look at the to-do list at note that it really is a depiction of
dependencies, of what needs to be resolved before some other situation
can be resolved:
"02--Does the Fleet in the Gulf of Lyon successfully support the Army
in Marsellies to Piemonte?"
"01--Who takes Piemonte, if anyone?"
Note that in the above list, "02" is at the "top" of the list,
it has the highest number of 02, and it must be resolved before moving
down into the list to resolve the lowest item in the list numbered 01.
We pull off the card from the top (by the way, note that the uppermost
or top card always has the highest numbering, in this case "02"),
and it reads, "02--Does the Fleet in the Gulf of Lyon successfully
support the Army in Marsellies to Piemonte?" Since no military unit
attempts to move into the Gulf of Lyon, then the French fleet in the
Gulf of Lyon does not have its support cut; therefore, it does indeed
support the French army's movement into Piemonte.
We now take our next (and last) item off the to-do list: "01--Who
takes Piemonte, if anyone?" At this stage, we now know that the
French have a force of 2 and the Italians only a force of 1, therefore
the correct adjudication is that the French army moves from Marseilles
to Piemonte.
In short, we say that we cannot resolve the situation in Piemonte
immediately because it is dependent upon whether or not the French
fleet in the Gulf of Lyon has offerred support which was not cut.
Let's define an ad-hoc term: "hot spot." A hot spot or
emphasized province on the map exists when
i. a unit has been ordered to move into a province that was unoccupied
at the very beginning of the turn,
ii. two units have been ordered into each other's province (for
instance, French fleet in Mid-Atlantic to West Mediterranean, and
Italian Fleet in West Mediterranean to Mid-Atlantic),
iii. a unit X has been ordered into a province which at the very
beginning of the turn already contained a military unit Y such that
unit Y did not receive a move order or else upon adjudicating unit
Y's move order it was found that for one reason or another unit Y's
move order could not be executed.
Here is a situation which I have scratched my head over (because it is
easy for a novice to get confused by all the rules).
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marseilles to Piemonte.
Italy:
Fleet in Piemonte to Gulf of Lyon.
The hot spot province is the Gulf of Lyon, because in this particular
instance, an military unit is not moving out of the Gulf of Lyon and
another military unit has orders to move into the Gulf of Lyon. Note
that the Piemonte province is not a hot spot (at this time) because we
have not adjudicated whether or not the army in Piemonte succesfully
moved out of that province or not.
Let's also use the ordered and stacked to-do list concept to see if
that still works. We start off at the hot spot and ask the following
question which will be added to our to-do list: "01--Who, if anyone,
ends up occuping the Gulf of Lyon?"
We then look around the board and determine that no other orders need
to be considered for us to resolve the situation in the Gulf of Lyon.
Therefore, we pull that item from our to-do list and read it:
"01--Who, if anyone, ends up occupying the Gulf of Lyon?" The
answer is that there was insufficient force for the Italian fleet to
dislodge the French fleet, so the Italian fleet stays in Piemonte and
the French fleet stays in the Gulf of Lyon.
There is now a new hot spot on the board: Piemonte. This is because
the French army in Marseilles has been ordered into a province
containing a military unit which currently has no orders to move
(because it tried to move, failed, and is now holding).
We now write up a new item for the to-do list which also is the first
item in the to-do list because the list is now currently empty:
"01--Who, if anyone, occupies Piemonte?"
The answer to this question is dependent upon whether or not the French
fleet in the Gulf of Lyon was able to support the French army's
movement. So, we had this to our ordered to-do list: "02--Did the
French fleet in the Gulf of Lyon carry out its support order (that is,
was its ability to support not cut)?"
There are no other units on the board to consider, so we remove the top
item from the to-do list and attempt to answer it: "02--Did the
French fleet in the Gulf of Lyon carry out its support order (that is,
was its ability to support not cut)?"
Due to our memory, we know that the Fleet in the Gulf of Lyon was
indeed attacked. However, there is a specific rule which states that
if the unit, S, giving support, is supporting an attack into province
P, and an attack from province P into unit S fails, then the support
offered by unit S is not cut. So even though the French fleet was
indeed attacked, in this particular scenario, its support of the
army's movement is not cut.
We now take the last item from our to-do list: "01--Who, if anyone,
occupies Piemonte?"
We know that the holding Italian army in Piemonte has a strength of
one. We know that the attacking French army supported by the French
fleet has an attacking strength of two. Therefore, the correct
adjudication is that the Italian army in Piedmonte is dislodged, and
the French army in Marseilles moves into Piemonte.
Please note that I certainly have not attempted to prove mathematically
that following any set of procedures (such as focusing on the current
hot spot and using an ordered, stacked to-do list) will always
correctly adjudicate all non-paradoxical Diplomacy positions. Has this
proof (or disproof) been made by anyone yet: more general, has it been
proven that a specific set of steps are guaranteed to correctly
adjudicate any scenario that might arise in a game of Diplomacy?
Therefore, these tools (focusing on the current hot spot and using an
ordered, stacked to-do list) are at this time experimental in nature
(at least as far as I am aware).
Let's consider a more complicated scenario next.
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marsailles to Piemonte.
Fleet in Tyrhennian Sea to Tuscany.
Italy:
Fleet in Piemonte to Gulf of Lyon.
Fleet in Tuscany supports Fleet in Piemonte to Gulf of Lyon.
Fleet in Roma to Tyrhennian Sea.
France's tactics, at least to a novice like me, seem sound. France
fields an initial force of 2 into Piemonte (we don't know the actual
force until we consider Italy's moves). France attacks Tuscany in
case the Italians in Piemonte hold and Tuscany supports Piemonte.
Let's pretend that before Italy wrote out its orders, Italy received
one-hundred percent accurate intelligence of all of France's orders.
For some reason, let it be assumed that Italy has decided that it wants
to take the Gulf of Lyon immediately. So, Italy orders the fleet from
Piemonte to the Gulf of Lyon with support from the fleet in Tuscany.
Now, however, let's pretend that Italy is new to the game. Italy
notices that the French orders for the fleet in the Tyrhennian Sea are
to attack Tuscany and thus cut Italy's planned support. In error,
the novice Italian player orders the fleet in Roma to "intercept"
the attacking French fleet from the Tyrhennian Sea prior to it reaching
Tuscany so that Tuscany's support will not be cut.
So, this example is just a reminder that even though it may look like
the fleet in Roma, due to the geography of the map, might
"intercept" the French fleet before it reaches Tuscany, the reality
is that the map contains joined provinces. Thus, nomatter what the
Italian fleet in Roma does, the French fleet will attack the Italian
fleet in Tuscany and at the very least (depending on everyone's exact
orders), cut the support being offered from Tuscany.
Let's see if the scenario given above (and repeated just below) will
create problems for the preliminary, experimental tools (the hot spot
identifier and the ordered, stacked to-do list) we have created so far.
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marsailles to Piemonte.
Fleet in Tyrhennian Sea to Tuscany.
Italy:
Fleet in Piemonte to Gulf of Lyon.
Fleet in Tuscany supports Fleet in Piemonte to Gulf of Lyon.
Fleet in Roma to Tyrhennian Sea.
.. . .
THIS ARTICLE IS NOT YET COMPLETED. But, any comments and corrections
you have on what I have written so far is much appreciated. Thanks.
Inside the Adjudicator's Head and Primary Paradoxes
InsideTheAdjudicatorsHeadAndPrimaryParadoxes.pdf
This article may be read at the following web address:
http://www.geocities.com/diplomacy2007/InsideTheAdjudicatorsHeadAndPrimaryParadoxes.pdf
In many ways, this article is elementary. It's first function is to
help me understand how an adjudicator realizes the orders of the
military units without using an automated computer program. Keep in
mind that there already exist well thought out and sophisticated
algorithms, such as the Diplomacy Players Technical Guide (also
referred to as DPTG).
http://www.amarriner.com/dip/dptg.php?games_id=22
Also please keep in mind that I am still very new to the game of
Diplomacy and may, on occasion, make rather crude errors in my examples
and thinking. This is due to the fact that Diplomacy is a rather
difficult game to come to terms with, but hopefully no such crude
errors will make it past the editing process.
Other related topics, such as primary paradoxes, will surely be
relevant. Let's first specify which paradoxical items we will not be
discussing in this article (though these items probably will be
addressed separately at a future time). We will not at this time be
interested in paradoxical situations and rule book ambiguities and
inconsistencies dealing with the following topics:
i. What the actual operational order is for a military unit and whether
or not it is valid and how this effects other support orders given to
other units. For instance, we will not be concerned about whether a
unit in Norway can or cannot be supported by other units if the unit in
Norway is ordered to move to the planet Mars (the planet Mars is not a
province on the Diplomacy map).
ii. Situations where an army can be convoyed by more than one convoy
will not be discussed at this time.
iii. Situations where an army can be convoyed to a province that it
could also legally walk to will not be discussed at this time.
iv. Situations where an army might be kidnapped by one fleet or another
will not be discussed at this time.
v. Situations involving issues related to self-dislodgement, dislodging
military units that are of your country, supporting the dislodgement of
military units that are of your country, and convoying in troops not of
your country which attack a province either held by a unit of your
country, supported by a unit of your country, or attacked by a unit of
your country.
In short, this is why the title of this article concerns "Primary"
paradoxical situations. This is not to say that the above issues are
not important, for they are; however, I have arbitrarily decided that
they are not the primary concern and focus of my current research
(though some of these topics may be researched in the future).
Concerning primary paradoxes (as defined above), we will be concerned
only with either the 1971 rule book or the year 2000 fourth edition
rule book. Actually, since the year 2000 fourth edition rule book is
freely available on-line from the publisher in Adobe Acrobat PDF
format, why don't we say instead that we will only be concerned with
using the year 2000 fourth edition rule book for the duration of this
article.
There will probably only be a couple rules of primary importance to our
investigation of primary paradoxes and these are now listed from the
year 2000 fourth edition rule book:
QUOTE: Page 10.
[Quotes not shown in newsgroup format, please see my web site]
UNQUOTE
QUOTE: Page 12.
[Quotes not shown in newsgroup format, please see my web site]
UNQUOTE
QUOTE: Page 16.
[Quotes not shown in newsgroup format, please see my web site]
UNQUOTE
Using the above quoted rules, and other more standard rules which most
people are already familiar with, we have adequate material to create
paradoxical positions on the Diplomacy board.
This article is not meant to be definitive; that should, I hope, be
obvious. This article is more introductory in nature, and by writing
and thinking on this topic, I hope to be better prepared to understand
algorithms which have previously been written by others and which are
already in existence.
Finally, I can't emphasize enough how important feedback is in my
learning process. Your comments, suggestions, corrections in reasoning
and logic, and general contextual information is, and has been in the
past, exceedingly important as I try to come to terms with the game of
Diplomacy.
Let's begin!
What does it mean to say that a given game of Diplomacy has entered
into a paradoxical situation? In my first introduction to paradoxical
situations on the internet, the reasoning was given in an English
statement, something like this: "The army was convoyed across and
attacked its destination province cutting the support being offered by
the target province; and, through numerous effects, too long to
mention, it then says, and thus the original convoy could never have
existed. We are at a paradox: how did the army get convoyed across
when the convoy could never have existed?" Actually, the examples I
came across on the internet were far better written than the sample I
just gave.
But, my point is this: for a minute, let's forget about trying to put
it in English. Let's define a paradoxical situation a little more
formally, such as like this: the game of Diplomacy enters into a
paradoxical situation when legal move orders cannot be executed without
violating important, fundamental rules of the game.
So, if an army was convoyed using two fleets, then after you have
executed all the legal moves, none of those two fleets must have been
dislodged, otherwise the convoying of the army could never have taken
place. As another example, if a convoying fleet is supported by
another fleet, and the convoying fleet is attacked with a force of two
by the enemy, then it is essential that after executing all the legal
moves wherein one of the moves represented a convoying of an army using
the convoying fleet, that this said supporting fleet not have had its
contributory support function cut, otherwise the convoying fleet would
have been dislodged, and the convoy could not have taken place. This
paragraph may not be crystal clear; hopefully the examples to follow
and the annotation of the adjudicator's thinking process will make all
this clearer later in this article.
Let's speculate what the function of these important, fundamental rules
are: I believe that these important, fundamental rules ensure that
when all legal moves are executed, it will be as if the army units had
been simultaneously thrown into action, even if in actuality the
adjudicator moves one unit at a time until all legal moves have been
executed on the Diplomacy playing board.
Why is it important to create a game and to create rules where the
simultaneous movement of the military units of the board is important?
I'm not sure at this time. But, as openers, each players order's are
revealed simultaneously. One reason might be that in primary
paradoxical situations, if one player moved before another, the player
that moved first would have a distinct advantage. In situations not
containing a paradox, I don't think it really matters who moves first,
the result should be the same as if all the units moved simultaneously
(but I have not yet thought about this overly much).
What is an important, fundamental rule? Well, pretty much all the
rules having to do with how to move a unit, how to support a unit, how
to cut the support being provided by a unit, convoying rules,
dislodgement rules, and so forth.
By the way, I forgot to mention that we will not be investigating any
cases having to do with what happens after it is determined that a
military unit has been dislodged. This will keep our discussion
tightly knit.
The rest of this article, or parts of it, could be considered rather
elementary as I warned earlier. I will present examples (not all of
them paradoxical), and speculate as to the reasoning process carried
out by the adjudicator. In this way, readers of this article can
correct any misunderstandings I have about this process.
For each scenario or board position and orders given, there are no
military units on the playing board but those explicitly mentioned.
Scenario:
France: Army Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
Piemonte is under attack by two contending military units of equal
strength. The adjudicated result is that no military units move.
Note that even at this stage, if we are not careful, we can adjudicate
this simply scenario incorrectly. Here is an example. The adjudicator
first moves the army in Venezia to Piemonte. Then the adjudicator
attempts to move Marseilles to Piemonte, notices that an Italian force
is already there, and so since the French army did not have a greater
attacking force, determines that the French army did not move. The
result would be that the Italian army moves into Piemonte and that the
French army did not move. This is an example of an incorrect
adjudication.
Note that if we continue to adjudicate incorrectly and start the
adjudication process over again, but this time moving France first,
then the result would be that the French army moved into Piemonte and
that the Italian army did not move.
An implied rule, I speculate, is that the end result of conducting all
the legal orders and moving all the units about must be identical
regardless of which unit is moved first on the playing board (whether
the unit is physically moved or whether the unit is moved in one's
head during a calculation). To do item: I will have to read over the
rules to see if this so-called implied rule is actually a rule which
appears in writing in the rule book.
Assuming that the above stated implied rule is correct, then we have a
simple algorithm to help check our adjudication process: if a
different ordering of how the units are moved about on the board
results in a different final result, then the adjudication process is
in error. By the way, we should add that at this time we are only
taking about situations which do not create primary paradoxical
situations.
Note that the above algorithm uses the wording, "final result."
This could theoretically mean that two different adjudicators each
using his own thinking process and with each thinking process being
different, might potentially end up with the same final result, even
though the process that was used is not the same. To say it another
way, assuming that a process of adjudication could be labeled
incorrect, the incorrectness of that process is not always apparent if
you only look at the final result. If we assume that the incorrect
adjudication process is a computer program, then this effects our
testing, for we may incorrectly conclude that our faulty adjudication
program works because in some particular scenario the final result was
identical to a correctly thought out adjudication.
It will be noted that at my web site,
http://www.geocities.com/diplomacy2007/
there are a number of rule sets or algorithmic systems or adjudication
procedures listed. Some of these rule sets do not come with detailed
examples showing the reasoning process from the very beginning to the
very end of an example. Thus, this makes them harder for me to
determine that I understand the algorithm correctly; and on top of
this it is possible that an incorrect interpretation of a rule set
might in some scenarios result accidentally in a correct final result,
then I would have no way of knowing if I was carrying out the rule set
correctly. So, it is my opinion that it is fundamentally important
that these rule sets have a sufficient number of scenarios with the
details of the thinking process spelled out.
Lets go back to our first, simple example again.
Scenario:
France: Army in Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
Drastically simplifying the rule book concerning standoffs on page 6 to
relate to our simple example, it says that when two unsupported units
both attempt to move into an unoccupied province, this results in a
standoff and none of the units moves.
While I have not at this specific time looked through the rule book
again to see if it explicitly says that when adjudicating a position,
the final result cannot be different depending on which units were
physically moved first on the playing board, I took this as an
operating assumption. However, let's forget about this for a moment
until some mathematician has shown us the proof. So, we are stuck now
with only relying on the rule as I paraphrased it just above.
So, it would appear that the correct adjudicational reasoning would be
annotated thus:
i. The adjudicator notes that two military units have been ordered into
the same, unoccupied province (in this case that province is Piemonte).
ii. The adjudicator now simply looks at the rule book to see the
result, and the rule book says (in perhaps my grossly simplified
manner) that if two units of equal strength attempt to move into an
unoccupied province that none of the two units moves.
By the way, as an aside, consider the following.
Scenario:
France: Fleet in Spain(nc) to Portugal.
Italy: Fleet in Portugal to Spain(sc).
The correct adjudicated result is that no unit moves. This is a
standoff (though the rulebook says that in a similar scenario it is an
example of two units not trading places). I mention this because I saw
a similar example in the rec.games.diplomacy newsgroup, and it is not
hard to imagine anyone simply moving the units and not noticing that
the units are really involved in a standoff (or if you prefer, one
might not notice that in moving the units the units had illegally
changed places), just as you would have in the following.
Scenario:
France: Fleet in Mid-Atlantic to West Mediterranean
Italy: Fleet in West Mediterrranean to Mid-Atlantic
Now back again to our first example.
Scenario:
France: Army in Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
Let's now introduce the concept of an ordered to do list. We will
introduce this concept to see if it might potentially become a way that
we can adjudicate this simple scenario as well as far more complex
scenarios. Think of the to-do list as a stack of index cards on your
desk. To execute the to do list, you take the first card off the stack
and do what it says. To put an item into the list, you write down the
task on an index card and place it directly on top of the to-do list
stack.
So, at this stage, all the legal orders are known for this example.
Scenario:
France: Army in Marseilles to Piemonte
Italy: Army in Venezia to Piemonte
You note that two armies are ordered to Piemonte. You create the first
to-do item: "01--Who takes Piemonte, if anyone?" Because there
are no other units on the board either moving, offering support, of
breaking support, there is nothing else to do; so, you now take your
one to-do item off the list, this being "01--Who takes Piemonte, if
anyone?" You execute this to-do list item by answering the question.
In this case, because the units attempting to move into Piemonte are
of equal strength, none of them move.
Here is the same scenario but with France's army's movement being
supported from the sea. We will work through this again using the
concept of an ordered, stacked to-do list.
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marseilles to Piemonte.
Italy:
Army in Venezia to Piemonte.
We note that there are two armies contending to occupy Piemonte, so we
write our first to-do list item and place it in the stack: "01--Who
takes piemonte, if anyone?" We now look around to see if any other
activity on the board effects the resolution, and we note that the
French support from the Gulf of Lyon must be considered. So, we add
that to the ordered, stacked to-do list: "02--Does the Fleet in the
Gulf of Lyon successfully support the Army in Marseilles to
Piemonte?"
There are no more units on the board to consider. Now let's start
resolving the to-do list items in our ordered and stacked list. First
lets look at the to-do list at note that it really is a depiction of
dependencies, of what needs to be resolved before some other situation
can be resolved:
"02--Does the Fleet in the Gulf of Lyon successfully support the Army
in Marsellies to Piemonte?"
"01--Who takes Piemonte, if anyone?"
Note that in the above list, "02" is at the "top" of the list,
it has the highest number of 02, and it must be resolved before moving
down into the list to resolve the lowest item in the list numbered 01.
We pull off the card from the top (by the way, note that the uppermost
or top card always has the highest numbering, in this case "02"),
and it reads, "02--Does the Fleet in the Gulf of Lyon successfully
support the Army in Marsellies to Piemonte?" Since no military unit
attempts to move into the Gulf of Lyon, then the French fleet in the
Gulf of Lyon does not have its support cut; therefore, it does indeed
support the French army's movement into Piemonte.
We now take our next (and last) item off the to-do list: "01--Who
takes Piemonte, if anyone?" At this stage, we now know that the
French have a force of 2 and the Italians only a force of 1, therefore
the correct adjudication is that the French army moves from Marseilles
to Piemonte.
In short, we say that we cannot resolve the situation in Piemonte
immediately because it is dependent upon whether or not the French
fleet in the Gulf of Lyon has offerred support which was not cut.
Let's define an ad-hoc term: "hot spot." A hot spot or
emphasized province on the map exists when
i. a unit has been ordered to move into a province that was unoccupied
at the very beginning of the turn,
ii. two units have been ordered into each other's province (for
instance, French fleet in Mid-Atlantic to West Mediterranean, and
Italian Fleet in West Mediterranean to Mid-Atlantic),
iii. a unit X has been ordered into a province which at the very
beginning of the turn already contained a military unit Y such that
unit Y did not receive a move order or else upon adjudicating unit
Y's move order it was found that for one reason or another unit Y's
move order could not be executed.
Here is a situation which I have scratched my head over (because it is
easy for a novice to get confused by all the rules).
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marseilles to Piemonte.
Italy:
Fleet in Piemonte to Gulf of Lyon.
The hot spot province is the Gulf of Lyon, because in this particular
instance, an military unit is not moving out of the Gulf of Lyon and
another military unit has orders to move into the Gulf of Lyon. Note
that the Piemonte province is not a hot spot (at this time) because we
have not adjudicated whether or not the army in Piemonte succesfully
moved out of that province or not.
Let's also use the ordered and stacked to-do list concept to see if
that still works. We start off at the hot spot and ask the following
question which will be added to our to-do list: "01--Who, if anyone,
ends up occuping the Gulf of Lyon?"
We then look around the board and determine that no other orders need
to be considered for us to resolve the situation in the Gulf of Lyon.
Therefore, we pull that item from our to-do list and read it:
"01--Who, if anyone, ends up occupying the Gulf of Lyon?" The
answer is that there was insufficient force for the Italian fleet to
dislodge the French fleet, so the Italian fleet stays in Piemonte and
the French fleet stays in the Gulf of Lyon.
There is now a new hot spot on the board: Piemonte. This is because
the French army in Marseilles has been ordered into a province
containing a military unit which currently has no orders to move
(because it tried to move, failed, and is now holding).
We now write up a new item for the to-do list which also is the first
item in the to-do list because the list is now currently empty:
"01--Who, if anyone, occupies Piemonte?"
The answer to this question is dependent upon whether or not the French
fleet in the Gulf of Lyon was able to support the French army's
movement. So, we had this to our ordered to-do list: "02--Did the
French fleet in the Gulf of Lyon carry out its support order (that is,
was its ability to support not cut)?"
There are no other units on the board to consider, so we remove the top
item from the to-do list and attempt to answer it: "02--Did the
French fleet in the Gulf of Lyon carry out its support order (that is,
was its ability to support not cut)?"
Due to our memory, we know that the Fleet in the Gulf of Lyon was
indeed attacked. However, there is a specific rule which states that
if the unit, S, giving support, is supporting an attack into province
P, and an attack from province P into unit S fails, then the support
offered by unit S is not cut. So even though the French fleet was
indeed attacked, in this particular scenario, its support of the
army's movement is not cut.
We now take the last item from our to-do list: "01--Who, if anyone,
occupies Piemonte?"
We know that the holding Italian army in Piemonte has a strength of
one. We know that the attacking French army supported by the French
fleet has an attacking strength of two. Therefore, the correct
adjudication is that the Italian army in Piedmonte is dislodged, and
the French army in Marseilles moves into Piemonte.
Please note that I certainly have not attempted to prove mathematically
that following any set of procedures (such as focusing on the current
hot spot and using an ordered, stacked to-do list) will always
correctly adjudicate all non-paradoxical Diplomacy positions. Has this
proof (or disproof) been made by anyone yet: more general, has it been
proven that a specific set of steps are guaranteed to correctly
adjudicate any scenario that might arise in a game of Diplomacy?
Therefore, these tools (focusing on the current hot spot and using an
ordered, stacked to-do list) are at this time experimental in nature
(at least as far as I am aware).
Let's consider a more complicated scenario next.
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marsailles to Piemonte.
Fleet in Tyrhennian Sea to Tuscany.
Italy:
Fleet in Piemonte to Gulf of Lyon.
Fleet in Tuscany supports Fleet in Piemonte to Gulf of Lyon.
Fleet in Roma to Tyrhennian Sea.
France's tactics, at least to a novice like me, seem sound. France
fields an initial force of 2 into Piemonte (we don't know the actual
force until we consider Italy's moves). France attacks Tuscany in
case the Italians in Piemonte hold and Tuscany supports Piemonte.
Let's pretend that before Italy wrote out its orders, Italy received
one-hundred percent accurate intelligence of all of France's orders.
For some reason, let it be assumed that Italy has decided that it wants
to take the Gulf of Lyon immediately. So, Italy orders the fleet from
Piemonte to the Gulf of Lyon with support from the fleet in Tuscany.
Now, however, let's pretend that Italy is new to the game. Italy
notices that the French orders for the fleet in the Tyrhennian Sea are
to attack Tuscany and thus cut Italy's planned support. In error,
the novice Italian player orders the fleet in Roma to "intercept"
the attacking French fleet from the Tyrhennian Sea prior to it reaching
Tuscany so that Tuscany's support will not be cut.
So, this example is just a reminder that even though it may look like
the fleet in Roma, due to the geography of the map, might
"intercept" the French fleet before it reaches Tuscany, the reality
is that the map contains joined provinces. Thus, nomatter what the
Italian fleet in Roma does, the French fleet will attack the Italian
fleet in Tuscany and at the very least (depending on everyone's exact
orders), cut the support being offered from Tuscany.
Let's see if the scenario given above (and repeated just below) will
create problems for the preliminary, experimental tools (the hot spot
identifier and the ordered, stacked to-do list) we have created so far.
Scenario:
France:
Army in Marseilles to Piemonte.
Fleet in Gulf of Lyon supports Army in Marsailles to Piemonte.
Fleet in Tyrhennian Sea to Tuscany.
Italy:
Fleet in Piemonte to Gulf of Lyon.
Fleet in Tuscany supports Fleet in Piemonte to Gulf of Lyon.
Fleet in Roma to Tyrhennian Sea.
.. . .
THIS ARTICLE IS NOT YET COMPLETED. But, any comments and corrections
you have on what I have written so far is much appreciated. Thanks.
Facebook
Twitter
Instagram