=============================  Appeal 858a  =============================


Panelist:                               favor
Decision:                               REVERSE


Panelist:                               elJefe
Decision:                               REVERSE


Panelist:                               Steve
Decision:                               REVERSE

========================================================================

History:

Appeal initiated:                       06 Mar 1996 06:24:19 GMT
Assigned to favor (panelist):           06 Mar 1996 14:43:54 GMT
Assigned to elJefe (panelist):          06 Mar 1996 14:43:54 GMT
Assigned to Steve (panelist):           06 Mar 1996 14:43:54 GMT
favor moves to REVERSE:                 07 Mar 1996 16:35:22 GMT
Steve moves to REVERSE:                 08 Mar 1996 02:07:30 GMT
Final decision (REVERSE):               08 Mar 1996 16:52:32 GMT
elJefe moves to REVERSE:                08 Mar 1996 16:52:32 GMT

========================================================================

Panelist favor's Arguments:

This is a difficult one, touching on an important issue that I
think has not been heavily treated by previous Courts.  The
question at issue is exactly what constitutes a conflict
between two Rules for the purposes of Rule 1482.  It seems
clear to me that if a low-MI Rule attempts to change the meaning
of a word or words in a higher-MI Rule, and that attempt constitutes
a conflict, then the attempt fails per 1482.  It also seems
clear that *some* such attempts would count as conflicts,
and therefore fail ("The phrase 'abide by all', in the Rules
of Agora Nomic, shall mean 'ignore'.")

The remaining question, then, is whether or not there are *any*
changes of meaning which would *not* count as conflicts for
the purposes of 1482.  One obvious candidate is a change of
meaning which is simply a clarification of a word that is not
otherwise defined in the Rules, changing the meaning only by
making it tighter and better-defined.  For instance, a rigorous
set of Rules which defined just what an "action" is in
Game terms should be able to touch the meaning of "action"
in Rule 1011, even if the Rule doing the defining was only
MI=1, as long as this touching did not fundamentally alter
the import and functioning of the higher-MI Rule.  A
rigorous definition of what it means to have "legal force"
should be able to touch the use of that term in Rule 114,
without having to be MI=3.  If some high-MI Rule makes a
passing reference to a Nomic Week, it should be possible
for a change to the low-MI Nomic Week Rule to touch the
meaning of that high-MI Rule at that point, without
requiring a higher VI.

I therefore find that the Statement here is FALSE, because
there are circumstances under which it is not true.  While
there are many cases in which a change to the meaning of
a word *would* fail per 1482, there are also some in
which it would not, because it would not be per se a
conflict.  The electorate, and in extremis the Courts,
will have to determine for themselves where each
individual case falls on the continuum.

It would be possible to find otherwise, and by finding
FALSE I may be leaving open a loophole in the Mutability
system through which demons will escape to plague the
community.  But I believe that the danger of that is
outweighed by the desire to keep flexibility in the
Ruleset, and not require elaborate and difficult
actions (such as passing Proposals with unusual AIs)
to acheive mundane ends (such as adjusting the meaning
of a word).

========================================================================

Panelist Steve's Arguments:

I'd like to begin by thanking Favor, Eljefe and Vanyel for the
valuable contributions they have made to the discussions about
this Statement. As a result of them, I think a consensus view has
emerged about how the issues raised by the Statement should be
treated, a result which I think constitutes valuable progress.

Although the other Justices, Favor and Eljefe, will doubtless
give similar arguments to the ones I give here, I think it is
worth including my own individual take on the subject.

I think the CFJ was motivated by a desire to see prohibited a
process which I have called 'amendment by stealth'. An example
of amendment by stealth would be a Rule which stated that, for
the purposes of R1322, 'less than' is to be understood to mean
'greater than'. The question arises, could such a Rule have its
intended of effect (of totally undermining the system of Mutability
Indices), despite the fact the R1322 has MI=3, and our hypothetical
Rule only MI=1? If it could, then it would seem that high mutability
is not much protection against a radical change in the meaning
and application of a Rule.

However, a powerful argument can be made against the possibility of
such obviously pathological cases of amendment by stealth. This is
the argument given by the Caller, namely, that amendment by stealth
effectively constitutes a Rule Change by means other than those
sanctioned by the Rules, and is therefore illegal. I am inclined
to accept this argument - for obviously pathological cases.

Problems arise, however, as soon as it is noticed that, firstly,
not all cases are as pathological as the above, and secondly, that
it may not always be clear whether a case is pathological or not.
An example: let us suppose that the phrase 'as soon as possible'
was used in the Ruleset prior to its having been given a definition
in the Rules for the first time by Proposal 805 (I don't know if
that's true or not, but it might well be). After the passage of
Proposal 805, could the new R805 be effective in giving the phrase
'as soon as possible' the meaning it tried to give it, a meaning
which is considerably different from the ordinary meaning of that
phrase? In one sense, the point is moot, since our subsequent
behaviour shows that we did accept R805 as successfully defining 'as
soon as possible'. Furthermore, we have since changed the meaning
of that phrase several more times, each time without incident.
This example alone is enough to show that the Statement is false,
for here we have a case where the meaning of a word (or a phrase
in this case) can be changed by changing a Rule which might well
be of lower precedence. That is, we take R1023 as defining what
'as soon as possible' means, for all Rules which use that phrase,
of whatever precedence.

========================================================================

Panelist elJefe's Evidence:

Rule 101/0 (Semimutable, MI=3)
Obey the Rules

      All Players must always abide by all the Rules then in effect,
      in the form in which they are then in effect.  The Rules in the
      Initial Set are in effect at the beginning of the first game.
      The Initial Set consists of Rules 101-116 (Immutable) and
      201-219 (Mutable).

History:
Initial Immutable Rule 101, Jun. 30 1993
Mutated from MI=Unanimity to MI=3 by Proposal 1480, Mar. 15 1995

========================================================================

Panelist elJefe's Arguments:

I distinguish two cases in the subject of the CFJ.  Where a Rule is clearly
relying for its meaning on the ordinary natural-language meaning of a word,
that meaning ought not be "hijacked" by a lower-precedence Rule.  However,
the statement also seems to cover the case where a Rule is explicitly relying
for its meaning on the definition of a word contained or supposed to be
contained in other (possibly lower-MI) Rules.  In this case, and in most
cases where an ambiguity or omission in a Rule is clarified by a lower-
precedence Rule, the meaning is shared by that lower-precedence Rule and
subject to change when the lower-precedence Rule is changed.

Rule 101 can be our guide.  It says that players must obey the Rules
"in the form in which they are then in effect".  I would say that if a
Rule uses a term whose meaning is imprecise, or which could rely (either
wholly or partially) for its meaning on another Rule, and another Rule
supplies the missing meaning, then the "form" of the Rules requires that
we assign the meaning given in the second Rule, even if it has a lower
MI, and if that second Rule is changed then the meaning changes.

However, if a Rule uses a term whose meaning is precise in ordinary
English (e.g. "greater than"), then a lower-precedence rule cannot
change this to something different ("in Rule XXX, "greater than" shall
mean "less than"). In a case like this, the Rules would have the
"form" of a conflicting Ruleset, which would be dealt with by the
appropriate conflict-resolution mechanism.

========================================================================