Re: DIS: CFJ 3813 proto-judged FALSE

[Kerim Aydin via agora-discussion]

On 2/20/2020 3:08 PM, Gaelan Steele via agora-discussion wrote:
> CFJ 3813 asks us whether Agora is a contract. As noted by the caller, CFJ 
> 3706 (Feb 9-14, 2019) asked a similar question. It was judged TRUE by G.
> 
> As I see it, the questions before me are:
> 1) Was CFJ 3706 judged correctly?
> 2) Has there been a change in circumstances since then that would affect that 
> ruling?
> 
> I’ll start with the second question. It appears that the version of the rule 
> in question there was 8139/20. (The court admonishes G. for failing to 
> include revision numbers in his rule citations, by the way.) The current 
> rule, 8139/22, is largely similar, with the following changes:
> * Lowered the minimum party count to one.
> * "may make an agreement” -> "may publicly make an agreement”
> * “agreement between parties” -> “consent of all parties” throughout
> * The new stuff about provisions being public and the body/annex distinction
> * A requirement that things be permitted "explicitly and unambiguously”
> * Clarity in wording about asset changes
> 
> Nothing here changes the fundamental nature of what a contract is. The 
> addition of “publicly” is closest, but the agreements that form Agora were 
> all (AFAIK) made publicly, so that doesn’t change anything.
> 
> Therefore, this CFJ is TRUE iff 3706 was correctly judged TRUE. This leaves 
> open the question of whether I should overturn that judgement. I certainly 
> *want* to—Agora as a contract is unintuitive and inconvenient to deal with as 
> Notary, and results in contract rules applying to Agora, which is weird. But 
> can I?
> 
> There is one point in 3706’s logic that at least isn’t clearly to me. I’ve 
> reproduced it below: {
> - R1742 states that “Any group of two or more consenting persons (the
> parties) may make an agreement among themselves with the intention that
> it be binding upon them and be governed by the rules. Such an agreement
> is known as a contract.”  This simply applied the label “contract” to
> this type of agreement.
> 
> - Therefore, the “contract” label applies to Agora.
> }
> 
> The R1742 quote in question could be interpreted in one of two ways:
> 1) It simply describes an existing thing that people can do "may make an 
> agreement among themselves with the intention that it be binding upon them 
> and be governed by the rules” and gives it a name.
> 2) It defines a new type of game action (forming a contract), analogous to 
> creating a proposal, calling a CFJ, or any other action that is only possible 
> because a rule defines it as such.
> 
> G. went with #1 in his judgement, but did not explain eir decision or address 
> the possibility of the second interpretation within the judgement. Before the 
> judgement, G. initially seemed to believe #2 
> (https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-discussion/2019-February/053314.html):
>  {
> Here, contracts are defined by how people MAY create them.   This rule is
> not written retroactively (e.g. "all agreements that people have made in the
> past that fit these criteria are defined as contracts"), but such that it
> takes a specific, initial act of agreement to create contract.  The Rules
> predate this creation mechanism.  And Agora was not re-created by this
> creation mechanism since then.  Therefore it is not a contract, in the R1742
> sense.
> }
> 
> D Margaux responded with something vaguely resembling #1 
> (https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-discussion/2019-February/053315.html):
>  {
> I don’t see why it needs to have been created as a contract to be a contract. 
> To me, it seems like “contract” is merely a term defined to encompass a 
> certain class of entities (viz., agreements between players that are to be 
> governed by the Rules). That class appears on its face to include Agora 
> itself. Not sure why under the text of the Rule it makes any difference that 
> Agora (and the agreement(s) that created Agora) preexisted the class defined 
> by the Rule. Still seems like Agora is a member of that class of entities.
> }
> 
> I agree with G’s initial judgement: the word “may” implies that we’re 
> defining a class of action, not just naming an existing class of entity. 
> Therefore, Agora is not a contract, because it was not created by the R1742 
> mechanism, because it didn’t exist yet. I find FALSE.

I was convinced by D. Margaux's arguments, but also this in R869:
>
>      A person, by registering, agrees to abide by the Rules. The Rules
>      CANNOT otherwise bind a person to abide by any agreement without
>      that person's willful consent.

This phrasing strongly, strongly suggests that registering in Agora is an
agreement.  This language goes back to January 2014.  And there was no
explicit definition of "agreement" in the rules beyond this in 2014 (e.g.
in this ruleset:
https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-official/2014-February/010620.html).
 In common definitions, a binding agreement is a contract ("contract: a
written and spoken agreement...").  So starting in Jan 2014 at least, it
was reasonable to see "Agora = agreement = contract".

Now R1742 history was re-enacted on July 1 2018 with this text:
>  Any group of two or more consenting persons (the parties) may
>  make an agreement among themselves with the intention that it be
>  binding upon them and be governed by the rules. Such an agreement
>  is known as a contract.

and "definition and continuity of entities" would seem to apply as the new
definition in keeping with the old definition, regardless of creation
method (e.g. if proposals used to have a different mechanism for being
created, changing the creation mechanism doesn't make the old things "no
longer proposals").

So this was part of my thinking on that previous judgement, but as you
point out, I glossed over them or never wrote them in a final draft.

HOWEVER, there was a wrinkle I missed in this progression in the earlier
judgement.  Which is that contracts were defined under an entirely
different set of rules (rules 2520 - 2527 from Oct 22, 2017 - July 1
2018).  The full set of rules can be found in this SLR:
https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-official/2017-November/012035.html

Here are some quotes:
>      A contract is a ruleset-defined entity embodied in text. An entity
>      can only become a contract through the appropriate ruleset defined
>      procedures.
and:
>      A contract is subservient to the rules. Although a contract may
>      specify obligations or powers beyond those created by the rules, a
>      contract may not override the rules

This, examining now, pretty clearly breaks the "all agreements are
contracts" link for the Rules, because contracts are subservient to the
rules.  R869 means the ruleset was still an "agreement" but not of the
"contract" type.

So when R1742 was re-implemented (replacing R2520-2527), we still ended up
with "Agora = agreement" and "contract = agreement" but the link "Agora =
contract", while not inconsistent with the R1742 text, was never re-created.

-G.