Re: [legaldocml] Finding the Right Term for "Resolutions" and Un-Official Laws

[Fabio Vitali <fabio@cs.unibo.it>]

Dear Grant, 

I feel I am not reaching you. 

Il giorno 13/feb/2013, alle ore 05.22, Grant Vergottini ha scritto:

> <Legislative Documents.png>
> I have prepared this simple diagram to describe the documents I come across as I travel from one jurisdiction to another. It is not a California Diagram or even a US diagram. It represents what I have come across in my most recent work. I am surprised by its consistency from Hong Kong to the US Congress. The words might differ in one way or another, but by an large, the concepts are all quite similar. Simple research on Wikipedia seemed to be in agreement with this too.

This IS very interesting, and reassuring.

> The boxes in green represent where I believe there are holes with Akoma Ntoso. In all cases I have come across, drafts are written as bills or resolutions rather than as "drafts". Because they're not published as drafts, there is little need for a <draft> tag. The UK might beg to differ though.

On the contrary, I believe that your green boxes are NOT holes with Akoma Ntoso. Non-positive code titles are <act>, there is no doubt about it. Non-binding resolutions are neither acts nor bills, and you should use element <doc> and then disambiguate in the metadata, or, as per my proposal of yesterday, we could introduce a more specific document type with a better sounding name whose content model is essentially <doc>. 

> I would not characterize non-binding resolutions them as "non-normative", which I take to mean "not prescriptive".

"Normative" in this context simply means "containing text that might become law sooner or later". If you don't like the term, we shall use a different one. A document of type B (or bill, shorthand) is a document containing text that, in the intention of the author of the Work, should or could be considered to become law sooner or later. 

This has nothing to do with its official status: a proper bill already received by the assembly is a <bill>, but also a draft bill as being written by a MP, or by a lobby, is still a <bill>, and the draft being written by a madman scheming to become the dictator of Orange County and prohibiting the production, sale and consumption of banana-flavored doughnuts would still be a <bill>. 

Thus you would use <bill> whenever you have a document that a) looks like a bill, and b) is meant (by the AUTHOR) to be or become a bill. 

> That term is better left to academic and standards contexts. The term "non-binding" seems to better convey the sense that any recommendations contained within aren't enforceable. It's a well used term and well understood.

No. This is very wrong. Binding and non-binding are attributes of the role of a document, rather than of the document. For instance, your schema details documents only at the end of their lifecycle, so we are missing their role in previous steps of the lifecycle, when they were NOT YET binding: every binding document has been non-binding for a time, e.g., while it was being discussed but was yet to be approved. 

Normative and non-normative characterization (where normative is used to mean "seemingly containing text posing as law" and "seemingly not containing text posing as law") are a more sensible distinction. You don't like non-normative as a term, we use something else: documents "looking like acts" (LLA) are <act>s, and documents "looking like acts but not approved" (LLABNA) are <bill>, and documents "not looking like acts" (NLLA) are something else, e.g. either <doc>s or a more specific proposal as per yesterdays mail from me.

> As far as I can tell, any resolution can propose law. But if that resolution invokes the power to enact that proposal, by using an enacting formula, that that resolution is a bill and the laws it proposes become law upon enactment. Without the enacting formula, the proposed law is merely a recommendation. This is why a constitutional amendment shows amendments to law, but leaves out the enacting formula as the legislature lacks the power to use it on the Constitution. Thus, that document becomes a non-binding resolution - a recommendation to the people. It also appears that many jurisdictions (in the US at least) have chosen to omit the preamble from bills in recent decades. When I look back 100 years or so, bills always seem to have preambles and are much more difficult to distinguish from other resolutions.

One more reason to use the LLA, LLABNA and NLLA test, rather than binding/non binding. Whenever the attribution to a category is a fine matter best left to fine scholars, it should not be used to build a fundamental structure of our vocabulary: my interpretation could be different than yours and we would be fighting about the root element of the document rather than some inner level metadata. 

This does not mean that we should ignore or give no room to the binding/non binding information bit, but only that it should not be the first and most important decision a manifestation author (i.e. the marker of the document) takes when converting into Akoma Ntoso a document: there are places in the metadata sections that are much more appropriate for this kind of fine scholarly diatribe.   

> So a bill is a resolution, but a resolution isn't necessarily a bill. But while technically both non-binding resolutions and bills are "Resolutions" in the general sense, the more common general term used is "measure" or simply "legislation". Measures are either "Non-Binding" resolutions or are "Bills". 

Fine. This means that we will use neither "resolution" nor "measure" for the new NLLA types of documents. 

> So I would vote to either use the terms "bill" and "resolution" if there are to be two terms or to use a single term "measure" which is the established all encompassing term.

I believe that everywhere in the world being able to distinguish between documents LLABNA and NLLA is important, and I will vote against collapsing them into a single document type, "measure" or whatever. If we decide to go with two separate document types, then I am fine with any terminological decision, and I will abstain from criticizing any actual proposal, as long as they are short, common nouns. 

Your proposals in fact seem to fit the bill, but an objection on the other hand is that, in your schema, bills and non-binding resolutions are two types of resolutions, and therefore adopting the term "resolutions" to refer only to what you call "non-binding resolutions" seems a stretch. But again, if this decision satisfies you and everybody else, I will not object. 

> Regarding the name for a non-authoritative composition of law, as with non-positive law titles, I am more concerned that the concept exist than about the name given.

Similarly to the binding/non binding distinction, the positive/non positive seems to be a fine matter of discussions. I believe that the LLA / NLLA distinction is more solid and appropriate. 

A non-positive title looks like a positive title, it is called in the same way, and there may be nothing in the structure, in the content and in the wording of a document that discloses its positiveness or lack of it. Since this information is outside of the document, it is METADATA and it is best left in the metadata section. 

Again, this is consistent with the choice taken a long time ago about non-authoritative consolidations of <act>s also being <act>s, even if they were created editorially, and therefore are "not positive". 

Non-positiveness can be deduced very simply by looking at the FRBRExpression metadata. If the author of the FRBR expression is NOT the #legislator, or equivalently the identifier of the legislative body officially entrusted with the job of generating positive law, then the document is NOT positive, no matter what the root element says. A non-authoritative consolidation of a statute will have the legislator as the author of the Work and "not the legislator" as the author of the Expression, and a non positive code will have the author of both the Work and of the Expression as "not the legislator". Of course, if we really wish to point out the positiveness of a text, we could add a metadata item (at the appropriate levels of the FRBR hierarchy). 




All said, this is my proposal for discussion later today: 

1) We introduce a new document type for documents officially produced by a legislative body that are not legislative in nature, i.e., the green box marked "Non Binding Resolution" in your schema. This element could be named "Resolution", "Measure" or, to avoid misunderstandings, something else such as "Statement" (see also item 2 of this proposal). 

2) We add an optional "name" attribute to elements "act", "bill", "statement" and really to all elements designating document types, with an open list of values where one can specify the local name used for this specific document. Therefore one would be able to express things such as: 
     <act name="statute"> ... </act>
     <act name="title"> ... </act>
     <act name="title"> ... </act>
     <bill name="draft bill"> ... </bill>
     <statement name="resolution"> ... </statement>

3) We add a new optional element to the FRBRWork element to refer to what you call "binding"/"non binding". It would allow "true"/"false" values. Since I know my chicken, I fear that the term "binding" will end up having different meanings in different countries, so I propose here to call it "prescriptive" instead (of course I would be fine with "binding", if we so vote, as I do not care for terms). 

4) We add a new optional element to both FRBRWork and FRBRExpression elements to refer to what you call "positive"/"non positive". It, too, would allow "true"/"false" values. Again, I know my chicken, I fear that the term "positive" will unknown in some countries, so I propose here to call it "authoritative" instead (of course I would be fine with "positive", if we so vote, as I do not care for terms). 

For instance, a positive title of the US Code would be as follows: 
   <act name="title">
     ...
     <FRBRWork>
        <FRBRAuthor value="#legislator"/>
        <authoritative value="true"/>
        ...
     </FRBRWork>
     ...
   </act>

, while a non-positive title of the US Code would be as follows: 
   <act name="title">
     ...
     <FRBRWork>
        <FRBRAuthor value="#olrc"/>
        <authoritative value="false"/>
        ...
     </FRBRWork>
     ...
   </act>

, and a non-authoritative consolidation of a statute after modifications would be: 

   <act name="statute">
     ...
     <FRBRWork>
        <FRBRAuthor value="#legislator"/>
        <authoritative value="true"/>
        ...
     </FRBRWork>
     <FRBRExpression>
        <FRBRAuthor value="#editor"/>
        <authoritative value="false"/>
        ...
     </FRBRExpression>
     ...
   </act>

What do you think? 

Ciao

Fabio 

> 
> 
> -- Grant
> ____________________________________________________________________
> Grant Vergottini
> Xcential Group, LLC.
> email: grant.vergottini@xcential.com
> phone: 858.361.6738


--

Fabio Vitali                            Tiger got to hunt, bird got to fly,
Dept. of Computer Science        Man got to sit and wonder "Why, why, why?'
Univ. of Bologna  ITALY               Tiger got to sleep, bird got to land,
phone:  +39 051 2094872              Man got to tell himself he understand.
e-mail: fabio@cs.unibo.it         Kurt Vonnegut (1922-2007), "Cat's cradle"
http://vitali.web.cs.unibo.it/