Preface

This proposal is a third draft of the lexicon building proposal, which has been modified to clear up mistakes, take into account what we have seen in testing the system in operation (successfully, I believe) and generally make suitable for public viewing and use.

The basic structure of this document is to look first at what I see as being the basic structure of the NGL language, in order to be able to specify what things this proposal pertains to and what things it doesn't. Then I will propose in detail a procedure for building the general lexicon in an orderly fashion. The theoretical structural portion should be viewed as "non-official", more of a preamble to explain the conceptual framework from which the rules were constructed. The portion that is binding is the rules themselves.

This document represents a re-thinking of the way in which we add new words to the language. The hope is that this proposal will facilitate an orderly and friendly effort of language building that can easily accomodate any project member who wants to propose an addition or a change to the vocabulary. It is hoped that growth of the lexicon, while slow, will be both continuous and well-thought-out. This proposal takes considerable inspiration from the status quo that had already seemed to have developed, codifying it to try and remove confusion. The only major change is in the way modules are viewed; under this proposal, the amount of vocabulary belonging to modules is minimal, and any item that can be is to be subsumed into the general vocabulary, where it can be generally useful and productive.

Before I begin, I want to stress that the idea of this proposal is not to mire us in legalism. What I'm hoping for is to make a set of guidelines for an orderly, but friendly and cooperative, vocabulary building effort that everone can feel comfortable participating in. If we succeed in working in this spirit, it is hopeful that we will only rarely have to argue over rules. I'm hoping that this proposal, by being orderly, but at the same time having its basis in friendly agreement and consensus, will reduce the scope for argument and help keep things friendly and cooperative.
 
Naesverig,
 
Stephen DeGrace
1999 wenasu 29.e

Further note: Jack Durst provided an excellent critique of this document. I have taken too long to get it online as it is, and do not at this time feel there is time to do a complete reworking (especially since some points may be in dispute and so there would be a certain amount of debate inherent in any reworking). Such a revision I feel is something that should be done at a future date, preferably sooner rather than later. In the meantime, in order to give the reader a balanced view of the opinions expressed in the group with regard to the ideas presented herein, you can read Jack Durst's commentary in its entirety (it is not long and definitely worth reading, and not just 'cause he was nice to me ;-) ) by following this link.

1.0    A View of the Structure of the NGL Language

The language is seen in this conception to basically have two parts; the core and the modules.

1.1    The Core:

The core consists of the General Grammar plus the General Vocabulary.

1.1.1    The General Grammar

The general grammar consists of the generic forms, such as generic tense and generic number, plus any modular grammar that is ratified as core. For example, if the word order, case marking and number modules pass the current ratification vote, they shall become core grammar. To use the NGL language, you need to know all the core grammar plus some form of the proposed modular grammar for any parts of the core grammar not yet sufficiently filled in with ratified modules (as with verb systems). It appears to be the case de facto that modules that don't win ratification (i.e., lose out to a competing proposal for core grammar) as core grammar can remain on as modules. These modules are alternative grammar, however, and should not be required learning for most speakers. This conflicting alternative grammar remains modular even if ratified. Unratified modules are proprietarily under their proposer's creative control. General grammar becomes common property upon ratification rather than proprietarily under its proposer's control, as it was before ratification.

Note: my vision of the general grammar differs somewhat from that expressed by Jack. Jack has said that the general grammar should be thought of as consisting of generics and "pick a module" statements. This discussion is somewhat outside the realm of the general vocabulary proposal, and in any case, I rather suspect that our difference of opinion in this area does not have much practical effect of the day-to-day operation of the language, i.e., it doesn't make much difference whether ratified modular grammar remains modular and is related to the core grammar by "pick a module" statements in the core grammar, or are actually subsumed into the core grammar, as I believe.

1.1.2    General Vocabulary

The subject of this proposal. General vocabulary is everything except for morphemes essential to the functioning of some non-ratified module and for words derived by the unique modular rules (such as IUPAC chemical names or kinship terms). General vocabulary is common property. Each item has its own individual status and is subject to individual review (morphemes provisionally accepted as a block are subject to individual vetting at pleasure).

Forms derived using general vocabulary morphemes are also part of the general vocabulary. It is taken as a general principle at this time that derivations are to be subject to evolutionary standardisation, rather than standardisation by debate and decree. While one may record derivations in one's personal lexcons, naturally, the Main Lexicon shall not record derivations.

This notwithstanding, it has come to pass that some derivations have been treated under general vocabulary rules, in order to enforce standards in some key areas such as standard derivations for country and ethnicity terms, and standard professional vocabulary. We will have to reconsile this "fuzzy" treatment of derivations, but in the meantime it should be understood that common practice actually allows some manoeuvering room where the official acceptance or derivations is concerned. As a rule of thumb, it should be considered that derivations can be treated under these rules, but since the rules generally are not menat to deal with derivations, one should present a convincing reason for making an exception.

NOTE: All morphemes have one and only one owner, either the General Vocabulary or a module (if one can be said to be owned in two places, these are more properly two homonyms than the same word - one can be changed without affecting the other). Only a bare minimum of morphemes should be considered modular, those essential to and restricted to the functioning of some module, and those generated by modular derivation.

1.2    Modules

These consist of proposals for grammar, morpheme-generation, or general principles of the language (as I gather Jack's style proposal may be considered a module of sorts).

1.2.1    Modular Grammar

Consists of proposed modules (e.g., VTT, traditional, and PVS verb systems are all modular grammar), and modules which failed to win ratification as core grammar for some specific area of the NGL language but remain on as modular alternative grammar. This grammar is proprietary, as the proposer is in control of the form that goes before the electorate.

1.2.2    Modular Vocabulary

As stated above, consists only of morphemes that are essential to and restricted to the functioning of some module, and those that are generated by unique modular rules.

2.0    The Rules for Building the General Vocabulary

First and foremost, proposals to modify the general vocabulary, either to add to it or to modify it in some other way, concern themselves only with morphemes, usually not with words that are derived from general vocabulary morphemes.

The rules of this proposal are only in effect during the design phase of the NGL project.

The basic concept is to try and handle as many morphemes as possible amicably and without resorting to ratification votes. Votes would only need to be held to settle intractible disputes. Morphemes accepted under this system but not ratified are automatically ratified upon the end of the design phase of the NGL project. This system requires a spirit of maturity, co-operation and compromise on the part of the participant of the NGL project, something which I very much believe we are capable of. A secondary, but no less basic, concept of this system is that no morpheme should be unreasonably sheltered from debate. At a minimum, proposed morphemes require endorsement from at least one other group member. This system will produce a slow progression, but a reasonable one, I feel.

Every morpheme in the General Vocabulary has a status which is recorded in the lexicon with the entry for the morpheme. What status a particular morpheme gets assigned to it depends on the progress of the debate regarding it.

2.1    Morpheme Status Assignments

This is a description of what each morpheme status assignment in the General Vocabulary means in basic terms. Note that to fully understand what each assignment means, you really need to refer to the description of the process of debate and acceptance, section 2.2.

PROPOSED: Any morpheme which is proposed and which has not yet acquired any other status.

PROVISIONAL: A morpheme which is proposed but to which the interested parties cannot come to an agreement on. Provisional morphemes are recorded but have no official status (and are of lesser prestige than accepted or ratified morphemes), and are not automatically ratified at the end of the design phase. This category exists in order to allow people to play with morphemes and try to demonstrate their usefulness, and to make these morphemes accessible, without giving them undue status and without having to resort immediately to the measure of placing the morpheme on the ratification list.

Morphemes that have been placed on the ratification list are treated as provisional until such time as a vote is taken to clear the list.

ACCEPTED: Any morpheme which has been proposed, seconded, and for which debate has ceased amicably (i.e., there remain no further objections to its inclusion in the General Vocabulary). Accepted morphemes are protected, because they cannot be modified without the acceptance or ratification of a proposal to do so. Accepted morphemes are ratified automatically when the group votes to end the design phase.

PROVISIONALLY ACCEPTED: A category which exists to facilitate making use of large blocks of morphemes that may be too difficult to indiviually vet one by one. Provisionally accepted morphemes are treated as accepted morphemes in every way in usage, and are ratified upon the end of the design phase. However, in order to uphold the principle that no morpheme should be sheltered from scrutiny, if anyone shall object at any time to a particular provisionally accepted morpheme, its status is reduced to proposed.

RATIFIED: Ratified morphemes are those in dispute which have been placed on the ratification list and gone to vote, and which have been passed by vote. These morphemes should be treated as having somewhat more official status than accepted morphemes and may only be changed by another vote.

Morpheme status is recorded in a field of the general lexicon, usually along with the initials or some other identifier of the person who proposed it. The purpose of recording the proposer of the (few) proposed morphemes that are recorded, and the provisional morphemes, is so that if a morpheme languishes for a long time, we know whose right it is to drop it or ask it to be put on the ratification list (i.e., the proposer of a proposed or provisional morpheme has certain rights over its fate, so we need to remember who that is). The reason for recording who proposed an accepted morpheme is a politeness issue. If somewhere down the road somebody proposes to change or delete some accepted morpheme, and its proposer is still around but away for a while, maybe on vacation, it would be polite to wait until they got back and had a chance to look at the proposal (and maybe object to it if they want) before debate on the proposal were closed. In order to be able to exercise this sort of politeness, it would be helpful if the proposer of any particular accepted morpheme were recorded.

The proposed abbreviations for the status levels to go in the general lexicon are: proposed (proposed - full name because these should not often need to be recorded, and we don't want them potentially confused with provisional), prov (provisional), acc (accepted), prov acc (provisionally accepted), rat (ratified). The identity of the proposer need not be recorded for ratified morphemes. Examples: A provisional morpheme proposed by me would be recorded as "prov SD." An accepted morpheme proposed my Julian Morrison would be recorded as "acc JM."

2.2    The Process of Lexicon Building

All debate starts with a proposal. There are three things which may be proposed and dealt with under these rules:

1. Addition of a morpheme to the General Vocabulary
2. Deletion of an accepted morpheme from the General Vocabulary
3. Modification in form or meaning of an accepted morpheme in the General Vocabulary

Note that provisionally accepted morphemes may be dealt with under these rules without having to put forth a proposal, merely by challenging them. Provisional morphemes are under the control of their proposers, and lack official status in any case. Ratified morphemes are beyond the scope of these rules.

Note also that one is allowed to propose synonyms, so long as they be considered separate morphemes and not modifications of existing morphemes (i.e., the proposal does not delete the existing morpheme with that meaning).

2.2.1    Proposal

Anyone may propose the addition of a new general morpheme, or the modification or deletion of an accepted general morpheme. A proposal to modify may constitute either or both of a proposal to modify the form (i.e., spelling, sound) of a word or a proposal to modify the meaning/definition of a word. The proposal must be posted in an acceptible public NGL forum. The aim of proposal is to seek accepted status for the proposal.

2.2.2    Debate and Seconding

After the proposal, debate should ensue, either cursory or complex depending on how controversial the proposal is. Obviously, a proposal may be modified by the proposer as a result of debate. THE PROPOSAL CANNOT BE ACCEPTED UNLESS IT IS SECONDED by at least one other group member besides the proposer. This means that you obviously cannot get an proposal accepted if you can't get at least one other person to endorse it. The purpose of this is to ensure that proposals have some kind of base of support (as opposed to apathy) beyond the proposer, and to ensure that proposals cannot be accepted simply because nobody bothered to debate them.

2.2.3    Acceptance

2.2.4    Provisional Acceptance

This is a mechanism that exists to allow the admission of large blocks of vocabulary without having to individually scrutinise each and every item right away. Provisional acceptance has to be moved by someone. The proposer or anyone else may move provisional acceptance for a block proposal. The motion to provisionally accept must be seconded, and cannot pass if anyone objects. If there is objection to provisional acceptance, the morphemes can only be dealt with on a one-by-one basis. Provisional acceptance for any reasonable block of vocabulary should not be too objectionable, since it allows the convenience of using the morphemes as if they were accepted, without any of the protection of accepted status, i.e., the morphemes are not protected from scrutiny later on. If any individual provisionally accepted morpheme is challenged at any time, its status is reduced to proposed, and its fate can only be resolved by regular debate.

2.2.5    Deferring as Provisional

If debate on a proposal to add a morpheme cannot be resolved, the proposer may have the morpheme placed in the General Vocabulary with the status of provisional. The purpose of this is to make the morpheme generally accessible, allowing the proposer to try and demonstrate the morpheme's utility in actual use. The proposer may ask to re-open debate on the morpheme addition at any time. If debate still cannot be resolved, the morpheme is placed on the ratification list (see 2.3.6). Morphemes remanded for ratification are treated as provisional until such time as time may be found to hold a vote to clear the ratification list.

NOTE: The proposer always has to option to drop a proposal before it is accepted or ratified.

2.2.6    Remanding for Ratification

A ratification list is kept for proposals made under this system which could not be agreed upon by all parties in debate. It is desirable to keep this list small, so amicability and a spirit of mutual respect and compromise are necessary among members of the group in order to make this system function efficiently. The ratification list should be kept by a volunteer (possibly the moderator or the keeper of the lexicon), and available for viewing upon request; the list should be cleared of business on a reasonably regular basis.

NOTE: The design phase of the NGL project cannot officially end until the ratification list is cleared. So part and parcel of the process for voting whether to end the desing phase of the NGL language and consider it officially launched must be the clearing of the general vocabulary ratification list.

2.2.7    Proposals for Deletion and Modification

These types of proposals may only attempt to delete or modify morphemes having accepted status in the General Vocabulary. Like proposals to add morphemes, these types of proposals must be seconded and have debate cease amicably in order to be accepted. If the proposal is accepted, the morpheme in question is modified or deleted as per the proposal. If one person objects and states that they will not change their mind, the proposal is considered defeated (exactly as would happen if the proposal were one to add a morpheme). The proposer then has the option to have the proposal placed on the ratification list, where it will be dealt with by vote, when the list is cleared.

2.3    Illustration of the Overall Process

For the addition a morpheme:

1. A proposal is made.
2. Debate ensues. Somebody seconds the proposal in some form agreed on by the proposer.
3. If debate ceases amicably, the morpheme is accepted.
4. If somebody digs their heels in, proposer may either drop the idea, or have it listed in the lexicon as provisional.
   1. After a reasonable trial period, the proposer re-opens debate or drops the proposal.
   2. If the proposal is not dropped, and agreement still cannot be reached, the proposal is placed on the ratification list.

3.0    Proposal in practice

General morpheme proposals can be part of modules (not actually part of the module, but rather, something that is done concurrently with the module), part of the vocabulary list for a composition, or completely independent. If part of a composition or a module, the only rule (more of a strong suggestion than a rule) is that the proposal be clearly indicated and organised, to facilitate debate. In making modules, probably most morphemes should be dealt with as general vocabulary proposals; there needs to be a clear justification, i.e., that the morphemes are only usable within the scope of the module or that the module would suffer unacceptibly by generalising these morphemes, in order to claim a morpheme as owned by the module and not the general vocabulary. Very, very few morphemes should be sheltered from standing on their individual merits because they are part of a module; morphemes should be judged on their own merits, not those of the module to which they belong.

An example is the colour module. Under this system, the eight basic colour morphemes proposed with the system are proposals for general vocabulary. Then, two separate modules propose two distinct protocols, mine and Jack's, for using specific subsets of this pool of morphemes, depending on what level of sophistication is desired. The modules themselves consist only of their respective protocols, plus the unique colour-combining morpheme {-ni-`} which both use.

An example of a module which proposes a large set of morphemes which don't fall under this system and have to be dealt with on the merits of the module as a whole is Jack's kinship system. This is because the kinship morphemes are derived regularly, rather than being arbitrary, using rules unique to the module, and so it would not be useful or practical to deal with each combination as a general vocabulary proposal.

4.0    The Status of Exisiting Morphemes

This section deals with what happened to existing morphemes when this system and its classifications were implemented, and is included in case it is needed to resolve any future issues. It has been agreed that the Ogden set proposals shall all be given accepted status at the start of the employment of these rules, as will all existing vocabulary items at the time of the inception of the rules that had achieved general and uncontroversial acceptance through some sort of process of debate similar to the spirit of the above rules. Morphemes in some kind of use which are controversial should be treated as provisional, and we can deal with them from there.

5.0    Conclusion