NGL: My Proposed Verb System, Version 0.1

[Jack Durst]

---------- Forwarded message ----------
Date: Wed, 10 Jun 1998 06:55:04 GMT
From: Stephen DeGrace <ephi...@thermopylae.net>
Reply-To: c72...@captain.morgan.ucs.mun.ca
Subject: NGL: My Proposed Verb System, Version 0.1

Hi all,

I am going to post the first version of my proposed verb system now, a
bit earlier than I thought I might. I will handle revisions on a
section basis until I reach version 1.0. When I get to this level in
my personal copy, and if the system isn't torn to ribbons and
generally dead in the water by then :-), I will post 1.0. I am certain
that mistakes and inconsistencies slipped by me despite my best
efforts to nip them in the bud; I apologise in advance for those, and
plead a tight schedule.

This document is available as a WordPerfect-format binary attachment
(with a bit fancier formatting...) if anyone so desires.

This document is best viewed with a fixed-pitch font.

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

A Proposal for a Verb System for NGL

PVS Version 0.1

1.0 Introduction

First off, the proposal needs a snappy name so that it may be referred
to in a non-cumbersome fashion. I will call it PVS, for "Proposed Verb
System." Uninspired, but literal; it will serve.

This system grew out of a thought experiment (energised by my exposure
to NGL) on how a verb might possibly be structured in an old and
highly derivatised language (but not so old a language as Chinese...).
It has its roots in ponderings of mine going back a while about the
nature of change in languages, particularly the observation that loss
of inflection seems to be a more-or-less common trend over time in
languages belonging to civilised people. It is based particularly on
the (I suppose) somewhat related observation that in the English
language, distinct among Western European languages that I know about,
many verbs have lost their power to form on their own explicit
interrogatives or negatives, needing the help of the verb "to do."
I.e., you can say "you eat...", however, "eat you...?" or "you eat
not...", while understandable, are not really correct in modern
English. So I wondered, if the course of this trend was eventually for
most verbs to need "to do" for _all_ their functions, making "do" a
sort of verb marker, what might the verb system look like? Of course,
in constructing this model, I indulged in a few non-English-based
fancies...

In constructing my model I wasn't originally seriously proposing a
verb system for NGL, although I hoped that some elements of my
proposal might find favour as potentially useful for NGL and get
adopted. I did borrow from what I knew of NGL, especially from its
sound system, because my encounter with NGL formed the source of my
inspiration to attempt this experiment. Because it wasn't really for
NGL per se, my system assumed that sentences would have rigid word
order; as well, I made up a bit of "pseudo history" for interest's
sake, and as part of that pseudo history I made up the broken fourth
conjugation, an innovation I hereby dispense with in my serious
proposal.

This is not to say that I think that my verb system is completely
unsuitable and unadaptable for NGL, on the contrary, I would not waste
your time and mine with this otherwise enjoyable exercise if I didn't
think that it could be seriously adapted or at least that something
good could be taken from it. I believe that PVS has the power and
sophistication to serve L1 users well, and as well I see it as readily
adapting itself to "pseudo-pidginisation" by L2 speakers wishing to
avoid the hard forms. While the latter is really not a direct
advantage to L1 speakers, it may help to expand the language's
currency, which in itself is an indirect aid to L1 speakers. Look at
the advantages that first-language users of high-currency languages
like French, German, Spanish, Zhongwen, Hindi, etc and most especially
English enjoy by virtue of large currency, including L2 popularity. L1
power is God in NGL, I understand that, and I concur heartily with
that, but if you can get some L2 ease out of these forms as well, it
is not to be sneezed at.

I believe that PVS may readily be adapted to free word order, with the
caveat that the order of particles within the verb is rigidly
fore-ordained. I don't see the necessity of maintaining order within a
verb as being especially onerous. Looked at one way, it's no worse
than insisting on a default order of inflections agglutinated to a
verb.

I don't mean to imply that I am presenting my system as an attempt to
create an "evolved" or superior system, because, for one thing, I make
a point never to look down on any language or dialect as inferior or
look up to any language as superior. I just happened to be thinking
about change in language (evolution in the strict biological sense as
change with time, rather than the popular sense as "improvement"...)
and this system popped out of my ponderings. I don't claim that it is
necessarily better than anybody else's system, and considering its
origins in a thought experiment completely unrelated to optimisation,
I certainly don't make claims of optimal efficiency. I can't even
claim I was trying to make a prediciton about the course of any
language, either, because I indulged my own "creativity" quite freely.
However, that said, I think my darling little baby (ha ha:) is a good,
effective system, and I think it may have some advantages for NGL,
these being:
* It is deceptively easy, IMO, to learn the rudiments and make
yourself understood. Tricky things like aspect, mood, voice and
displaced tenses aren't necessary to make yourself understood at a
crude level. This deceptive ease makes it easier to suck people in.
IMO, the rudiments of the system can be grasped quickly, and
sophistication can follow at a more leisurely pace.
* It is capable of a fine level of precision and sophistication
in the hands of a sophisticated, experienced user (or, of course, a
first-language user), IMO again.
* It mirrors the verb system of the highly-successful English
language in these advantages, IMO. One of the advantages of the
English language, aside from its tremendous currency in wealthy and
powerful nations, is that (barring spelling...) it is comparatively
easy to learn English well enough to make yourself understood (because
of the comarative paucity of inflection and gender - the testimony of
a couple of Chinese friends of mine who have also tried to learn
Russian seems to bear this observation out) but at the same time
advanced users are capable of a large degree of precision and elegance
(well, IMHO of course). I believe that these qualities are useful ones
to emulate for the verb system of NGL.

I will endeavour to prove in this proposal that PVS does indeed
possess the advantages of ease of initial learning and depth of
possible sophistication. In fairness, however, I must state that I am
absolutely not trying to imply that anybody else's system lacks the
advantages I claim for PVS, or has them in less abundance. It is not
my purpose to knock anybody else's system. PVS may prove inferior in
these qualities to another proposed system. Here I will simply present
PVS and leave the reader to judge its relative merit. Regardless
whether this proposal is ultimately judged to be right for NGL (which
I realise is a separate question from the proposal's merit as a verb
system for some other hypothetical language) and ultimately regardless
of the objective merit of the end product of my efforts, I am enjoying
and deriving much fulfillment from this exercise and will feel the
effort put in to be intrinsically worth-while. That said, I do
sincerely hope and work for people's approval and will be extremely
gratified if even a few ideas are taken from this work, find popular
support, and incorporated into NGL.

Some of the terminology used in this proposal will not be correct,
standard terminology. This is due to my insufficient and incomplete
grasp of liguistic terminology and concepts. In order to compensate
for this deficiency, I will endeavour to define any terms I use so
that correct alternatives may be more easily proposed by others.

Conceptually, here is how I envision this verb system and its role
within NGL. Verbs, with a few exceptions (more about these and my
justification for them later), conjugate by preceding them with a verb
marker or markers that contain explicit information about the time
(tense), as well as implicit or explicit information about the aspect,
mood and voice of the verb. Verbs that "conjugate" in this fashion
shall be referred to as "non-conjugable" verbs (since they need help)
and denoted Vnc. A verb may be preceded by a whole string of particles
which are written as independent words, and which, because they are
capable of standing alone as verbs, may be referred to as "helping
verbs," or "helpers." Even the verb marker, "te," which translates as
"is doing" (more on the meaning-realm of te later) is itself capable
of being a stand-alone verb as well as an indicator for another verb.
Verbs are by default participative (aspect), definite (mood) and
active (voice). The stative aspect and the definite mood are rendered
by adding a suffix on to one of the helping verbs (more on this
later). The passive voice is rendered using a helping verb. I will try
to justify my choices when I deal with these aspects of the verb in
detail below.

The verb itself, e.g. :eat:, and all of its preceding particles, shall
together be considered the verb of the sentence (coin: superverb). The
order of particles within the superverb is rigid, but the position of
the superverb is free.
So in the English sentence:

Vespa had been eating cod tongues
---superverb---

"had been eating" is what I am referring to as the superverb, and in
my system it had rigid internal order and is not allowed to be
scrambled or split up (except rarely to admit an adverb, if this level
of freedom proves desirable - see 8.0). You could also perfectly
legitimately say (assuming the language allows this word order, of
course, I'm not 100% sure...):

Toung[acc.pl.] cod[gen.pl.] had been eating Vespa[nom.sing.]
---superverb---

When the speaker employs rigid word order, it is understood that the
noun phrase preceeding the superverb is the superverb's subject. When
the speaker employs loose word order, it is understood that the
subject of the verb within a clause will be a noun in the nominative
case within the clause.

PVS verbs do not inflect for person, gender or number. While the
subject can be omitted by a speaker if he/she feels that it should be
obvious from context what the subject is (as should occur only
rarely), it will have to also be obvious from context what the person,
gender and number of the subject is. When the subject is explicitly
stated, which will occur in almost all cases, these qualities of the
subject will have to be explicitly stated, i.e. a pronoun or noun will
be needed that gives information about person, gender(?) and number.

That said, the PVS system could conceivably be modified to include
inflection for person and number (and gender?).

I will here coin terminology for the positions in the superverb. This
is so that the correct location for helpers can be readily referred
to. The last part of a superverb is the "most specific part" and the
initial helper is the "least specific part." This works on the
principle that in constructing a verb we try to organise particles
from the most general to the most specific, i.e., we start with the
helper that gives the least information about the action in question
and end with that which tells us the most (the verb itself). I will
refer back to this principle in constructing my examples below. There
are competing organising principles which may take precedence,
primarily that particles giving information about voice, aspect and
mood go as near as possible to the front of the superverb, a principle
designed to warn the reader/listener as soon as possible that the verb
is non-standard. I will continue to use my invented terminology for
now, however, because it has a "right feel" to it, to me. As I deal
with each type of particle, I will be sure to be clear where it goes.

In constructing examples I will try and use as much English as
possible. In translating, I will put the superverb in brackets, and
any English word within the brackets will be placed in colons. In any
case where I feel it necessary to demonstrate how the verb would work
with free word order, I will place the cases of nouns in square
brackets following the nouns, and this will be intended to show that
the case is being marked explicitly in the real NGL sentence.

I will mostly refrain from coining vocabulary in this proposal, with
one important exception. I will need to procoin names for the helping
verbs and grammatical particles of the PVS system in order to be able
to deal with them in examples. However, any and all of these items may
be replaced with better Tokcir vocabulary items by those more
knowledgeable than I about the language's vocabulary.

One final caveat. In constructing a verb system, one is limited by the
fact that the verb is, IMHO, inextricably intertwined with the
language's idiom when put into practice. Without a vocabulary, I think
that it is impossible to definitively state how some issues of how an
idea will be expressed will be resolved, and so, many examples will
necessarily invent a hypothetical idiom. The point being that whatever
verb system NGL eventually adopts, it will at some stage, probably
after its first vocabulary is released and people start employing it,
undergo a process of revision and refinement in all its grammatical
forms as idiom developes.

I will try and provide many in-text examples to illustrate principles.

1.1 Summarised System Overview

PVS is a system that looks at all text and oration of almost any sort
as being essentially narrative, and tries to pinpoint the action of
the verb within the relative framework of the narrative. The basic
structure is the superverb, which consists of a string of one or more
helping verbs followed by the all important VERB, a word like "go" or
"want." The verb tells you the type of action. The particles preceding
the verb give two streams of information simultanously which work to
precisely describe the meaning of the superverb constructed from that
verb. One stream is tense. Each helping verb has tense, and the
sequence of tenses across the helping verb situate the level of the
narrative in the past, present or future and tell us whether the event
in question happens in the narrative-time's past, present or future
(see Appendix B and sec. 3). Simulataneously, the nature of the
particle chosen to mark the verb can add finer distinctions to the
verb such as whether the subject is able to perform the action of the
verb, the choice of initial particle can tell you whether the subject
is the perpetrator or the recipient of the action, and the choice of
optional suffixes can render the verb hypothetical or definitive, or
tell you whether the action being talked about is discrete or
repetitive. A verbal idiom exits for changing a demonstrative sentence
into an explicit interrogative. Finally, adverbs may be employed to
give greater precision to how near or far in time an action is.

2.0 Tense, Aspect, Mood, and the Verb "te":

This section will deal with the forms of the verb marker "te," its
independent meanings, and with its use as a marker and tense particle
for that class of verbs that can't inflect for time on their own (but
which might optionally bear inflections for person and number if
people feel it is necessary for the system to have this capability),
the verbs I will refer to for want of a better name right now as
"non-conjugable" verbs (Vnc). Knowledge of the verb "te" is the
foundation of being able to use almost every other verb in NGL in the
PVS system. This leads into a discussion of aspect and mood. I
envision that aspect will be especially commonly used, and therefore
necessarily intertwined into any discussion of the PVS system.

2.1 The First Conjugation (V1): Te

We will start by showing the paradigm (such as it is) for the first
conjugation of the handful of conjugable verbs, referred to here as
V1. V1 has only one member: te. Te inflects for only one thing: tense.

te:
pres te - is doing
past ta - was doing
fut to - will be doing
Infection is obviously by internal sound change rather than by
agglutinative endings.

On its own, the verb "te" corresponds very closely in meaning to the
English verb "to do." In addition, it is the universally necessary
helping verb, acting as the marker for the verb and situating the verb
in time, either relative to the present of the speaker/narrator or
relative to a time referenced by the speaker/narrator which is in turn
relative to this individual's present (this refers to "displaced
tenses," more about which will be discussed later). On its own as the
most specific part of the superverb (i.e. in cases where it is not
followed by a verb such as "eat" or "go"), te has a broadly general
idea of action, and as such may serve as a stand-in for another verb
which is understood by context. This is particularly useful in
answering questions. For example, in response to the question:
"Do you play sports?"
You could say:
"I (tean)."
Instead of simply "yes" or "Yes, I (tean :play:) sports."
More usefully, in response to:
"What sports do you play?"
You _could_ say:
"I (tean :play:) baseball and hockey."
Or, since "play" can be understood, you shorten it to:
"I (tean) baseball and hockey."
Features of this verb, such as the choice of the stative aspect, will
be confusing here, but will hopefully become clear below. The
important point is that the verb to be marked can in some cases be
omitted and its meaning taken up by te, "do."

Since "te" marks a verb, it has a tendency to try and verb anything
that follows it, generating confusion if you're using te as a sort of
"pro-verb," (think: "pronoun") standing in the place of a more
specific verb. Sometimes this is okay -- in the exaples above making
"baseball and hockey" a verb with the understood connotation of "play"
does not at all detract from the meaning. However, there are a couple
of ways to eliminate confusion where the possibility exists for it. If
te is acting as a "pro-verb" and there is a noun following it that you
absolutely don't want verbed, you can place a declensional ending on
that noun, (e.g., an accusative ending). This makes it clear that the
noun is not being verbed, as only a noun with the ending dropped is
allowed to be verbed. Another way to avoid confusion is to use te to
mark itself, i.e., use the form "te te," where the first te is
conjugated to show time and the second is treated like a Vnc and left
in the present tense. So you avoid any feel that the nouns "baseball
and hockey" are being verbed by saying:
I (tean) baseball[acc.] and hockey[acc.].
or:
I (tean te) baseball and hockey.

2.1.1 Conjugation of "Non-Conjugable" Verbs (Vnc) with Te

Knowing this simple verb form "te-ta-to", you have the means to make
yourself understood with any almost other verb, although achieving
sophistication requires considerably more effort. Te is by default
participative, definite and active (more on these below), and any verb
it acts on has these qualities as well. So the conjugation of the verb
":eat:" is:
:eat: :
pres te eat - is eating
past ta eat - was eating
fut to eat - will be eating
Getting back to the "least specific/most specific" terminology, in "te
eat," for example, the least specific part is "te" and the most
specific part is "eat." The qualities referred to by "te" are
qualities that any verb could possess, i.e. it is unspecific about the
nature of the verb; but only one verb means :eat:, therefore it is
more specific.

The verb "te" situates the action directly in the time being referred
to, not before it or after it (i.e. it generates "imperfect" tenses).
The "present" to which the past and future is relative is always from
the point of view of the speaker, narrator, ect., except for some of
the displaced tenses (more on these later). If the verb is
present-tense undisplaced active participative definite (i.e., would
take only "te") and if the speaker is using a word order that puts the
subject at the beginning of the sentence and right before the verb (so
that it will be obvious that the word to follow is expected to be a
verb), the subject itself can serve as the verb marker, and the "te"
can be dropped at the user's disgression. So
:I: te :eat:.
can also be written as:
:I: :eat:.

2.2 Aspect and Mood

Now for a discussion for the two shadings of the verb that are
achieved by suffixing particles to helping verbs, that is, aspect and
mood. These two are conditions of the entire superverb, not just
portions of it, so any given superverb can only have one particle for
aspect and one for mood. The reason for this is that I simply don't
think the human mind can wrap itself around the complicated structure
that would ensue if you had, say, three helpers in front of the verb
and all three can be either participative or stative. I think it
simplifies matters considerably to think of aspect and mood as being
conditions of the superverb as a whole, and I don't think that this
restriction significantly limits the power of the system to present
meaning. Aspect and mood markers always are suffixed to the least
specific part of the verb, for no other reason than to make it clear
up-front that the verb is non-standard in some way. Aspect takes
precedence over mood, so that if a superverb is both stative and
conditional the stative marker comes before the conditional marker.

2.3.1 Aspect

A verb can have one of two aspects: participative and stative. A verb
is by default in the participative. To render it stative, the suffix
"-an" is attached to the least specific part of the verb (i.e., the
left-most helper). Think of the subject of the sentence as a hunter
carrying a rifle and with a shotgun slung over his back; he's out to
bag him a few verbs. The participative is like the rifle: it hits the
action at some discrete point within the action. Since the hunter is
carrying the rifle and it's loaded, he doesn't have to do anything to
use it; just point and fire. The stative is like the shotgun. It not
only obliterates the action, it also can do in a whole _flock_ of
actions :-). To use it, the hunter has to sling his rifle and grab
for the shotgun, reach for his "-an" as it were.

So the participative describes a discrete action. It funtions like "am
eating" in the sentence:
I am eating eggs.
You can think of it as hitting a specific instant in the action.
The stative is more nebulous, describing a set of actions (or, by
metaphorical extention, a state of being - hence the name). It
functions like "eat" in the sentence:
I eat eggs for breakfast every Wednesday.

I chose to make aspect a distinct feature of PVS verbs because I think
it is an interesting feature which might be fun to play with and
explore. Further, I feel that making distinctions like that of aspect
extends the power of the language to state things precisely without
relying on context. While complete and exhaustive precision is not
always necessary or desirable, IMO, given the power of context to
simplify the task of expressing meaning, it is still necessary for a
language to be explicit in at least some distinctions in order to give
it subtlety and power. What these distinctions are need not be the
same for every language, but you need some, IMO. I have chosen aspect
as a distinction for PVS, and I think that on the whole it serves to
enrich the system.

I chose to make the participative the "ground state," the opposite of
the way it's done in English, partly to see what that would look like,
partly because I have developed a certain aesthetic preference for it,
and also partly because I wanted the default forms of the verbs to be
a literal and concrete as possible, as I think that that will be an
aid in picking up the rudiments of the language.

So here's what the verb :eat: looks like in the stative:
pres tean eat - eat
past tan eat - ate
fut toan eat - will eat
"toan eat," for example, is the verb form you want for the sentence "I
will eat eggs Wednesday mornings [i.e. every Wednesday morning] for
breakfast." Putting "toan eat" into that sentence: "I (toan :eat:)
eggs Wednesday mornings for breakfast."

"to :eat:" is the form you want for the sentence "I will eat eggs for
breakfast this Wednesday"; the sentence becomes "I (to :eat:) eggs for
breakfast this Wednesday," which is more literally translated as "I
will be eating eggs for breakfast this Wednesday."

Getting back to the example about sports, you probably noticed the
ending -an on the verb in that example and wondered why it was there.
Looking at the example again (with slight modification):
I (tean :play:) baseball.
in response to the question "What sports do you play?" The speaker is
not asking what you are doing right this minute, he wants to know
something about a presumably oft-repeated aspect of your life, which
is the realm of the stative. The respondant uses the present stative
because in the current time playing baseball has been a repetitive
action and the speaker has every expectation that it will continue to
be a repetitive action, i.e. that he is still within the set of
basball games that God set out for him when he was born ;-).

2.3.2 Mood

There are two moods: definite and conditional. Verbs are by default
definite. To make a verb conditional, you suffix "-me" to the least
specific part of the verb. Any sentence that says a thing definitely
did or did not happen or asks if it did or did not happen, is
definite. All of the examples given above, for example, are definite.
The conditional mood, by contrast, establishes the feeling that the
consequence of the action is in question, or alternatively, that the
action is hypothetical. Here are two examples of the use of the
conditional:

"I would like a glass of water" (in context, a polite request)
I (teme :want: :receive:) a glass of water.
literally: I would be wanting to receive a glass of water.
"want receive" is in the conditional because we are attempting to be
polite. We are not presuming to know whether or not our desire will be
fulfilled by the listener. In giving this example, I am not stating
how polite terms would absolutely _have_ to be constructed in NGL
using PVS; I am only giving an example of how the conditional _might_
be employed in PVS for that purpose.

The main function of the conditional is to make the a statement
hypothetical, like the function of the word "would" in English. E.g.:
"I would buy a gun if my Momma would let me"
I (teme :buy:) a gun if my Momma (teanme :allow:) me.
literally: I would be buying a gun if my Momma would allow me.
The PVS sentence, unlike the English, gives both verbs a very "right
now" feel. "teme buy" is in the present participative, because I want
it to have to feel of "go out right now and buy." "teanme allow" is in
the present stative because I want to give a more nebulous feeling of
needing Momma's permission as a state of being in the here and now,
rather than a feeling that I'm hoping Momma will physically walk in
right here and now and give her permission. I used the present rather
than the past because I want to give a feeling of wanting Momma's
permission in the here and now, rather than wishing she gave it at
some previous juncture. I expect that many more shades of meaning
could be generated depending on the context and the meaning the
speaker desires. By no means is the form I give the only conceivable
correct form. I would expect that over time, the language would
develope a sense of what possibilities make better style than others,
and in the meantime good selections can be made logically.

2.4 Conclusion Regarding Aspect and Mood

To conclude this section, let me say that I realise that aspect and
mood are occasionally fuzzy things in PVS, allowing many options and
shadings and often a number of correct or at least understandable ways
to communicate some specific action or condition. This feature is
deliberate. By including aspect and mood, I want to give verbs a
built-in source of subtlety and slipperiness (and, perhaps conversely,
precision), allowing users of PVS within NGL the ability to eventually
develop a set of verbal idioms as fine, elegant, complex and varied as
that possessed by any natural language. That said, I will of course
gladly entertain any proposal to modify, eliminate, or even expand
aspect and mood.

2.5 Note to Gerald

Jerry, I swear to God I didn't borrow the te-ta-to sequence from you.
When I made that up for my pseudo-joke system, I didn't even know that
you once used a similar (te-ta-ti) sequence for particles doing
something very similar. I considered removing the coincidence entirely
in the serious system and getting some new markers, but in the end I
decided to keep them. The coincidence is just too intriguing, and I
was kind of attached to the words themselves. _Was_ it a coincidence,
I wonder, or is there something either inherent in language or
inherent in our English language that made us think of such similar
forms? Anyway, it's neat.

3.0 The Second Conjugation, the Verb "'e", and Displaced Tenses

This section deals with the second conjugation verbs (designated V2),
which has two members, "'e" (roughly, "to have") and "se" (roughly,
"to be"). Although these verbs are very close to the verb "te" in
form, they have an important difference from te when used as the most
specific part of a superverb rather than as helpers. This difference
is major enough that I consider it justified to group these separately
from te for ease of learning. This section also concerns itself with
displaced tenses (what I also think of as perfect and predictive
tenses), which often need the verb 'e as a necessary helper. The
passive voice, which employs the verb se, is dealt with in the next
section.

3.1 Conjugating V2 Verbs

The V2 verbs compare rather strongly to "te." Like te, 'e and se are
both self-marking, i.e., they require no marker in order for it to be
absolutely clear that these are verbs. Like te, they both have broad
meanings so that in probably the vast majority of situations where
either could be applicable, speakers will likely select a more precise
verb as the most specific part of the superverb. (Although, that said
I expect se will enjoy considerably more independent use than 'e.)
They both function as helpers, and have functional meanings which are
metaphorically related to their independent meanings but not exactly
the same as their independent meanings.

The big difference between V1 and V2 is in default aspect. V1, that
is, te, is by default participative, and gives this participativeness
to any verb it precedes either explicitly or implicity - this means
that all superverbs ending in "te" or Vnc are by default
participative, hence the existence of a marker to render verbs
stative. By contrast, V2 verbs are by default stative, but do not
communicate this stativeness to any verb they help (think of it this
way: the rule is that in determining the aspect of the superverb you
look at the last helper, and at least implicitly the last helper will
_always_ be te). The reason for this distinction is that while I
wanted participativeness to be the default for the PVS system as a
whole (and hence for te), I see stativeness as being more compatible
with the independent nature of 'e and se and therefore likely to be
the most-used form. This irregularity, in other words, is justified as
an irregularity which may slightly inconvenience learners but which
will serve as a shortcut for native speaker.

In order to render a V2 verb which is the most specific part of the
superverb into the participative aspect, you simply mark it with "te."
So you get "te 'e" and "te se" forms. These forms, like the "te te"
forms mentioned earlier, treat the 'e or se as a Vnc, and thus the
terminal 'e or se is not allowed to conjugate (be in any other tense
except present) in these forms.

Example:
"I was having a bad day."
I (ta 'e) a bad day.
Note: I realise that this idiom would probably _not_ be used in NGL!
One would probably use a verb meaning more like "to experience..." I
use this example only to show what a participative 'e might look like,
and what _sort_ of idiom might partake in it.

Example:
"Billy-Ray is being an unmitigated prick."
Billy-Ray (te se) an unmitigated prick.
Contrast this with:
Billy-Ray (se) an unmitigated prick.
The first form says what Billy-Ray is being right now, it makes no
claim that this is is usual nature, in fact, he might actually be
quite nice most of the time. The second form, however, is saying that
Billy-Ray was born a prick, is a prick, and will most likely die a
prick. Well, perhaps not quite _that_ strong, but it does make claims
about his essential character rather than his specific behavior.

Here is the second conjugation and aproximate translations:
'e:
pres 'e - have
past 'a - had
fut 'o - will have
se:
pres se - is
past sa - was
fut so - will be

3.2 The Independent Meanings of the V2 Verbs

This section treats the meanings of the V2 verbs in those cases where
a V2 verb is the most specific part of a superverb.

3.2.1 The Verb "'e"

The verb 'e, like the verb te, takes in a very broad variety of
meanings. It can have the sense of "to have," "to possess," "to own,"
"to hold," "to have located on/within," "OBJ is located at SUB," etc.
However, it is rarely used, unless the explicit sense is absolutely
clear from context and the speaker/writer wants to take a bit of a
shortcut. Ordinarily, if one wanted to express one of these particular
ideas, one would select a word with precise meaning and not use 'e.
However, 'e does have some importance as a shortcut and as well a
tremendous importance as a grammatical marker for the displaced tenses
(tenses one might say are perfect or predictive - the justification
for this later). "'E"'s function as a grammatical marker is, however,
rooted in its independent meaning, and so to understand its function
we need to have a conception of its meaning. It will help to think of
it as meaning broadly "have."

3.2.2 The Verb "se"

"Se," unlike te or 'e, actually, has a rather specific meaning and
would be quite highly used. Its meaning is something like that of the
verb "to be," with a couple of small differences. Nevertheless, se is
also a fairly important grammatical particle, a metaphor of its
meaning being a justification for its employment as the marker for the
passive voice (it also has the simultaneous ability to displace
verbs). This will be examined in detail when we treat the passive
voice in section 4. What se does is to say that its subject (a
nominative of some sort) has the quality of the verb's object (an
accusative or an adjective). Note: unlike the Latin "sum," se is _not_
an equal sign! It does not connect two nominatives which are
interchangeable. So:
Bill[nom] (se) the president[acc] - Bill is the president
is subtly different from:
The president[nom] (se) Bill[acc] - The president is Bill
Clinton
Se attaches more importance to its subject, and this makes a
difference in how you answer a question. If you are asked "What does
Bill do?" it is more correct to say:
Bill[nom] (se) the president[acc].
but if you are asked "Who is the president?" by some ignorant doofus,
it is more correct to say
Bill Clinton[acc] (se) the president[nom].
unless, of course, you don't just slap him instead.

I justify making this admittedly subtle and often perhaps irrelevant
point on the basis that I feel it adds to the information that is
conveyed by the verb and thus adds a bit of subtlety to the system.

When using se to apply a adjective to a noun, as in the sentence "the
sky is blue," I move that it should be necessary to precede the verb
with the subject and follow it with the adjective, or at least to
place the adjective right on the heels of se.

3.3 Displaced Tenses Using "'e"

First let us look at the use of the verb "to have" to express perfect
tenses, i.e., tenses in which the action of the verb takes place
before the time being talked about and which carry the connotation of
completeness (hence "perfection" in oldest and most literal sense of
the word without any connotation of superiority). It is nearly
universal in Western European languages, at least, to use the
language's equivalent of "to have" in order to express the idea of
perfection in a verb, an idiom that I borrow in order to generate
PVS's displaced tenses.

I say _nearly_ universal, BTW, because I know of at least one and
maybe two exceptions. In the Scots Gaelic language (and probably,
therefore by extension in the Irish Gaelic language which is its very
close cousin) there is no explicit verb "to have." Part of the
meaning-realm of "to have" is taken up using the idiom that
such-and-such a thing is "at" whatever thing has it. So "I have the
book" in Scots Gaelic is "Tha an leabhar agam" - approximately "The
book is at me." I'm not at all certain how Gaelic takes care of tenses
like the pluperfect and so on, but one can bet that like the rest of
the Gaelic langugae it is bound to be eccentric :-).

But that's neither here nor there. Think about what it means to "have"
an action that took place in the past. Unfortunately, to talk about
it, I'm going to have to use it, so this will be a little bit
difficult to discuss clearly. I only ask that you stretch your
imagination and try to see the logic of saying that if the action was
completed before the time of the narrative, i.e., exisits in the
narrative's past, then you possess or "have" that action. If you have
completed an action before the time you're talking about, in a sense
you _possess_ that action as part of your personal history, say. If
you "have done" a thing, right here and now, your status with regards
to that action is that you have completed it. If you "had done" a
thing, at some time in the past that you're talking about, you had
completed the action in question; "had done" makes the action
complete, but references against some past time, useful in narrating
events that occured in the past to put the events _before_ the past
time you're talking about. For these perfect tenses, the verb "to
have" is being used in an almost metaphorical sense to create tenses
where the action is displaced backwards in time from the narrative's
"present." So you can say that the pluperfect ("had done...")
references a time in the past and then places the (completed) action
in the "past's past." If we look at the examples of "have done,"
(perfect) "had done," (pluperfect) and "will have done," (future
perfect), you can see the relative roles of the tense of the verb "to
have" versus the tense of the verb it modifies, in these examples "to
do." "To have" references the time we're talking about, and the verb
itself, being in the past tense, puts the action in that time's past.

The verb "'e" is used to serve a similar, but even broader, role in
PVS. In discussing its use as a helping verb, we ignore its true,
independent meaning for a while and dispense with metaphors. The
metaphor is employed essentially as a justification for the verb's
grammatic function as a helper. Instread, we will examine 'e entirely
from the point of view of its function.

In a superverb, 'e is to be considered less specific that te, i.e., 'e
comes before te in the superverb. What 'e does is references a time
and says "this is the time we're talking about." The form of the verb
te following it and preceding the verb is then considered relative to
_that_ time and not the speaker's present. 'E compounds _displace_ the
action from the time being talked about, creating what I shall refer
to as "displaced tenses." I will list all the combinations that are
possible between te and 'e and their functions. There is but one major
criterion in deciding whether a combination will be allowed or not:
redundancy. If a combination creates a tense that is exactly
equivalent to an existing simpler form, that form, although obviously
understandable, will be disallowed as being very poor style.

3.3.1 'E + "te" Combinations

The combinations "'e te," "'a te," and "'o te" will not be allowed.
Think about what they mean. "'E te" says that the action takes place
in the present's present, "'a te" says the action takes place in the
past's present, and "'o te" says the action takes place in the
future's present. The present's present is exactly equivalent to
saying "te," the past's present is exactly equivalent to saying "ta,"
and the future's present is exactly equivalent to saying "to." To my
mind, all these combinations are describing on-going actions and
cannot be distinguished from the simpler te forms.

3.3.2 'E + "ta" Combinations

These are highly useful work-horse tenses that are broadly similar to
the perfect ("'e ta"), the pluperfect ("'a ta"), and the future
perfect ("'o ta"). However, to refer to these tenses I will invent a
new English terminology that I will use occasionally which I feel
better reflects what these word combinations are doing in PVS. I shall
refer to the time being talked about as present, past or future, and
the action of the verb as being backward-displaced ("ta") or forward
displaced ("to" - more on this later). So the tense rendered "'a ta" I
shall refer to occasionally as the "backward-displaced past," for
example.

Here are some examples of the backward displaced tenses in action.

Backward-displaced past (pluperfect):
Hillary had known about Bill's lovers for some time.
---------
Hillary ('a ta :know:) about Bill's lovers for some time.
We use the pluperfect because we are establishing background for a
story in the past and don't want to deal with Hillary's current
knowledge of Bill's lovers regardless of whether it is there or not.
With the verb "to know" the choice of participative or stative is
tricky, since knowing is very much a state of being. I elected to use
the simpler participative form and throw the rest into the lap of
context (which you're allowed to do); it would be more grammatical,
however, IMO, to say "'an ta :know:."

Backward-displaced present (perfect):
Sergei has washed the dishes.
----------
Sergei ('e ta :wash:) the dishes.
The context of the above example is that Sergei has finished some
specific act of dishwashing - if we wanted to say that yes, Sergei has
gotten his hands soapy once or twice in his life, we'd use the
stative.

Backward-displaced future (future perfect):
After paying what I owe the government, I will have lost all I own.
--------------
After paying what I owe the government, I ('o ta :lose:) all I own.

3.3.3 'E + "to" Combinations

Contrary to what you might think, PVS can actually make use of these
tenses, which I think of as being "predictive." What they're saying,
to my mind, is that you "have" a future act - in other words, that you
are destined to it. These tenses give an extraordinarily strong sense
of certainly, and so I expect the forward-displaced past, ("past
predicitive," let's say) to see more use, because the speaker can have
actual knowledge about what was going to happen after some time in the
past - from the speaker's point of view, in other words, the act was
"destined." (note: see section on "'e to:" below; I have since revised
my thinking a bit on these forms) Here are explanations and examples
where appropriate of all three possible combinations. They are similar
in function to the English expressions "is going to," "was going to,"
and "will be going to."

"'E to":

At first I was going to disallow this combination, on the basis that
"'e to," the present's future, seems pretty damn similar to "to," the
future. However, on some thought I have decided to revise this and
admit the form "'e to"; I can even think of a pretty good use for it.
Consider a second the meaning behind the metaphor used to create
displaced tenses with 'e when you apply it to "to." What it's saying
is that you _have_, that you _possess_ a future event, that it is your
destiny - a high degree of certainty. Whence this certainty? For the
"'a to" form the source of the certainty is obvious - hindsight. With
sufficient knowledge of the events being spoken of, saying with
certainty that a thing definitely happened _after_ the level in time
of a past-tense narrative (see below) is pretty easy. But what
distinguishes possessing a future event in the present from saying
that an event happens at a future narrative time? To me, it's the
element of rock-hard certainty that the "'e to" form has over and
above the "to" form, certainty which comes from "'e to"'s metaphorical
meaning. If the event in question is coming very soon, you can have a
large amount of confidence in it, thus you can say you "have" it,
you're destined to it. And what is the implication of this certainty?
To me, a future event possessing this level of certainty is a future
event in the _near_ future, rather like the French idiomatic form
"nous allon gagner," "we're going to win" approximately, although the
French conveys a much more explicit sense of immediacy than the
English.

So, "'e to" is the immediate future tense, for use in a present-time
narrative. Its sense can also be made more explicitly immediate by
using an adverb with the meaning :in a short time:. More on this in
section 8.

Nota bene: all this said there is still a bit of a meaning-overlap
between "to" and "'e to." Because of the desire to avoid redundancy,
_"to"_will_be_used_anywhere_it_can_. I expect and hope that "'e to"
will be a comparatively rare form, but nevertheless, it's there.
Here's an example of where one might _possibly_ use it.

Example:
"I am going to take out the garbage [just now] before we head out to
the game."
Verb translation:
I ('e to :remove:) the garbage before we (to :leave:) for to go to the
game.

"'A to":

This combination puts the act in the future of some past time, and
strongly implies in most contexts I can imagine that the predicted act
happened after the time being talked about but before the speaker's
present. Since the speaker can have factual knowledge of that time, he
can be very certain of what lay in that time interval. Here's an
example (highly fictional :-) complete with context.
The context:
"Especially after his infamous bender that ended up causing Ł50,000
worth of damage to downtown Richmond, nobody would have expected that
this young man, this George Washington, would one day become President
of the USA."
The sentence:
This young man would one day become President of the USA.
Verb translation:
This young man ('a to :become:) President of the USA.

Actually, the "'a to " form translates very closely as "was going to."

"'O to":

Aside from its obvious value to psychics in making ever more
convoluted predictions, the future predictive is useful in making the
sequence of future events a bit more explicit. Like the "'e to" form
it possesses a sense of immediacy that comes from its metaphor, but
unlike "'e to," this immediacy is not as strong. The reason for this
is that "'o to" doesn't have a competing form, the way "'e to" has
"to"; since it is the only combination available to express the
future's future, it is able to stretch out pretty much infinitely far
into future time, if you need it to. For that reason, it more so than
the present predictive form "'e to" will often take temporal adverbs
to clarify how _far_ it is into the future.

Example:
"Trust me, you will be just going to leave, when what do you know, the
phone will ring."
Sentence:
"You will just be going to leave."
Verb translation:
You ('o pes to :leave:).
Note: "pes" is a proposed adverb of time. See section 8.0.

4.0 The Passive Voice with the Verb "se"

The passive voice is a means of reversing the flow of action from
subject (nom) to object (in this case an instrumental object, what
Latin would render in the ablative I do believe) in a sentence, rather
like the enzyme reverse transcriptase reverses the normal flow of
information from DNA to RNA in cells. It alows the subject of the
sentence to be the passive recipient of the verb's action rather than
its active perpetrator. One can argue that the passive voice is not
necessary, and certainly there are ways to get aound it. If there is
an explicit object from which the action flows, as in:
"Cyrus was eaten by a wolf" (maybe "Wolf[abl.] Cyrus[nom] was eaten")
you can simply reframe it as an active sentence, as in:
"A wolf had eaten Cyrus" (maybe "Wolf[nom.] Cyrus[acc] had eaten")
If the actor is not clear, as in:
"The reagents were refluxed in THF for 3 hrs."
you could perhaps use a vague pronoun like "one" to get the same idea:
"One refluxed the the reagents in THF for 3 hrs."
There are, however, some advantages I see to the passive voice that in
my opinion justify its inclusion in PVS. As in the first example, the
passive voice allows you to frame the *topic* of your conversation,
say Cyrus' whereabouts, as the *subject* of the sentence, even if this
subject is actually the _recipient_ of the action. This allows, IMO,
greater range and subtlety of meaning. The passive voice allows you to
say that a action took place without saying anything about the actor
at all and without needing a "fuzzy actor" for times when it isn't
convenient or possible to specify the actor, as in the second example.
Passive voice is a feature in a number of important languages, English
and German I can think of off the top of my head, and therefore
allowing access to it may make speakers of these languages a bit more
comfortable with NGL. Finally, I would contend that the NGL idiom will
almost certainly provide ways around using the passive if users want
to avoid it, but I expect that it will be a valued feature of the
language for those who do use it, and I don't anticipate it being very
hard for users to comprehend and use the passive voice correctly with
a little practice, should they feel so inclined.

I have elected to use an idiomatic helping verb to mark the passive
voice, rather than a particle with function but no independent
meaning, or a suffix. I do this because in my opinion using a particle
that is primarily functional rather than meaningful in its role in the
verb but which has independent meaning in and of itself makes it
easier to gain an intuitive grasp of the passive and what it means.
The idiom I chose, as with the active displaced tense, closely mirrors
that employed in English. To render the passive, you use a form of the
verb "se" as the least specific part of the verb. Both se and 'e can
be used in the same superverb (more on this later in section 5), and
in that case, I put forth the convention that "se" shall be considered
less specific that "'e," even though this point is definitely
arguable. The reason for this choice of order, is 1) to make it clear
as early as possible in the formation of the verb that it is of a
non-standard form - this reasoning is similar to that which puts
aspect and mood modifiers on the least specific part of the verb, and
2) to reduce monotony by increasing the number of ways a superverb can
start legally.

4.1 "Se" As a Marker for the Passive: Simple Possible Forms

The use of se to render the verb passive can be to an extent
intuitively grasped if you consider what it might mean to _be done_
(se ta) the action of a verb rather than to _do_ the action of the
verb (te). I contend that it can be envisioned that if you _gain the
quality of_ the action of a verb, (which is a closer translation of
"se" that saying you "are"), then you are the _recipient_ of that
verb's action.

In the ensuing discussion, we will deal with the function of se as a
marker rather than with its meaning. Its functions are twofold.
1) Se at the beginning of a superverb indicates that the subject
of the verb will the recipient of the action of the verb rather than
its perpetrator, i.e., it changes the voice of the verb from the
default active to the passive voice.
2) Se functions exactly like 'e in that it references a time from
which the tense of te is relative. In other words, "se te" forms can
be displaced tenses all on their own.

The rule for disallowed combinations with se is exactly the same as
that with 'e; no redundant forms are allowed.

4.1.1 Listing of Forms and Permissions for Use

se te - allowed - pres. imperfect passive
sa te - allowed - past imperfect passive
so te - allowed - fut imperfect passive

These forms, which were not allowed with 'e due to redundancy, are
allowed with se. They have the function of rendering the verb passive
without displacing it in time from the time being talked about.

se ta - allowed - perfect passive
sa ta - allowed - pluperfect passive
so ta - allowed - future perfect passive

se to - allowed - present predictive (see below and
3.3.3 on "'e to" forms)
sa to - allowed - past predictive passive
so to - allowed - future predictive passive

"Se to" is a form parallel to "'e to" - a present predictive form with
a strong sense of certainty, and attendant with that, a strong
connotation of immediacy which can be made explicit by using it with
the adverb "pes" (proposed only - see 9.0) Like "'e to" it may
_occasionally_ be useful, especially in establishing a sequence of
future events, but in any case where one feels one can in good
conscience employ it "so te" should be the preferred future imperfect
passive form, just as "to" is preferred over "'e to."

4.1.2 Examples of the Use of the Passive

Imperfect Forms:

Example:
"Right before my eyes, Jimbo was being eaten by a rabid zebra."
Sentence:
"Jimbo was being eaten by a rabid zebra."
Verb Transalation:
Jimbo (sa te :eat:) by a rabid zebra.
Here's an example using non-fixed word order:
Jimbo[nom] zebra[abl.] rabid[abl.] (sa te :eat:).

Example:
"Although he has the intelligence of a lobotomised ox, Cyrus will be
given a scholarship at the ceremony tonight because he's seven feet
tall and can play basketball."
Sentence:
"Cyrus will be given a scholarship."
Verb translation:
Cyrus (so te :give:) a scholarship.
Or maybe:
Scholarship[acc.] (so te :give:) Cyrus[nom.]
Notes:
The verb "so te :give:" means something more literally like "will be
being given." You could express this same idea actively using
"receive," I know. However, the point remains that the passive voice
does expand your options, which is important for writing. Note the
actor is not really specified in this sentence. The passive gives an
advantage here because using the passive the actor need never be
specified.

Perfect Forms:

Example:
"Oh, Officer, has there been any news about my darling Billy-Ray?"
"Ma'am, I'm afraid your husband has been killed by a pack of hungry
beavers."
Sentence:
"Your husband has been killed."
Verb translation:
Your husband (se ta :kill:).
Notes:
I realise you could just as easily say "beavers have killed your
husband." However, I think that this sentence is enriched by the coice
of the passive, because it allows poor Billy-Ray, the topic of the
conversation, to be put in the nominative. This gives a nice option
for dramatic flow, would you not agree?

Example:
"By the end of the day, 50,000 cows had been turned into steaks and
shoe leather."
Sentence:
"50,000 cows had been turned into steaks and shoe leather."
Verb translation:
50,000 cows (sa ta :make:) into steaks and shoe leather.
Notes:
This is a place where in my opinion the passive is clearly superior to
its alternatives. The passive lets you make it clear that this act is
clearly somebody's fault, the cows didn't just become that way of
their own accord, while at the same time allowing you to not specify
who the rascals are or use "they." The argument is that with the
option of the passive voice available, sophisticated users will be
able to create more intricate and varied works of language with little
aditional learning or effort required. Anyway, I just like the
passive.

Predictive Forms:

Example:
"At the time the passengers abandoned the sinking ship in the frigid
waters of the North Atlantic, nobody suspected that very few of them
would eventually be rescued."
Sentence:
"Few would be rescued."
Verb translation:
Few (sa to :save:).
Notes:
The English sentence is a bit ambiguous; is it that nobody was
thinking they wouldn't be rescued and, by God, they were eventually
rescued after all (justifying those God-fearing folks'
pluck/stupidity), or did nobody at the time suspect the awful and
inevitable truth? (In either case they suffered from a startling lack
of imagination, but it serves for the example.) In English it needs a
bit of clarification; the form is rather hypothetical sounding. The
PVS past predictive passive (or passive forward-displaced past tense,
if you like) makes absolutely no bones about it, in the translation I
have given. They are all going to die. The past predictive is
mercilessly inevitable.

5.0 Multiply Displaced Tenses

The system of displaced tenses contains the possibility of talking
about an action more than one "step" removed from the time you're
talking about. This is done by using a string of "e'"'s in the
superverb. The superverb can theoretically contain an infinite string
of "'e"'s, but it may only contain one se, and if it contains a se the
se must be the first in the series in accordance with the principle of
putting information like voice at the beginning of the superverb. I
will shortly give an example of how a multiply displaced tense might
be used. I envision that using multiply displaced tenses to establish
sequence of events will be a very rare occurence, and so I propose to
not deal with it exhaustively here, but rather wait until questions
about it come up. There are a couple of rules that can be used to help
determine if a multiply displace form should be considered "possible"
or not:
1) If, when you work out what the sequence means, it turns out to be
equivalent or very nearly equivalent to something simpler, the long
sequence is disallowed and the simpler form should be used instead.
2) If the string is long and/or convoluted enough to be significantly
difficult to follow, you really ought not to use it; instead you
should try and find a simpler way to say what you want.

Example:
This is part of a narrative, where the narrator is telling a story in
the past tense, and is here going on a little excursion into the
past's past, as it were, and using the pluperfect.
"At the time Jimbo had begun his life of crime, he had had achieved
straight A's in school"
Verb translation:
At the time Jimbo ('a ta :begin:) his life of crime, he ('a 'a ta
:achieve:) straight A's in school.
Notes:
In the form "'a 'a ta :achieve:", the first 'a references the time in
the past the speaker is referring to, the second 'a displaces the
action one level back into the past, to the level of "begin" in the
first verb, and the ta displaces the action one step further back in
the past than that (how far will have to be evident from the context)
and causes the verb ":achieve:" to "fire," as it were (i.e., marks
it). Even this short sequence is a bit hard to follow; you can see,
I'm sure, why I say that writers and speakers will generally not use
many of these forms or use them often. Nevertheless, they are
available.

6.0 The Third Conjugation (V3)

The third conjugation (V3) is my attempt to deal with those verbs that
add a connotation about the possibility or imperativeness of the verb
they modify. The verbs I'm thinking of, like "can" and "must," are
verbs that _always_ either explicitly or implicitly modify a verb. In
designing the PVS system, I wanted to have a reasonably economical and
elegant way of expressing these concepts, and at the same time I
wanted something flexible, so that if I'd missed anything it would be
easy to add to the system to cover it. In pursuit of these goals, I
decided to retain my third conjugation in modified form, as an
expansible, "spandex" conjugation that can easily add more verbs of
this sort with any variety of subtle meanings, as needed. As you
gather, then, I decided to treat these concepts with verbs, and to
make them, like the V1 and V2 verbs, able to take a verb directly and
conjugate for time.

The way V3 verbs basically work, is that they are self-marking as
verbs, and like te, they mark whatever follows them as being a verb.
In fact, they have all the functionality of te and more, replacing te
in the superverbs they modify. I propose two initial members of the
conjugation, a number I would expect would probably be easily made to
grow if PVS were adopted, as the needs of the language for modifiers
of this sort became clearer.

6.1 The V3 Verbs and Their Independent Meanings

V3 contains two verbs as the initial set: gul (approximately "is able
to") and zol (approximately "has to"). On their own they are neither
really participative or stative, although in the same manner as te
they confer a sense of participativeness on the verb they modify and
can be rendered stative in the usual fashion. Like te, they can
sometimes be used to stand in for a verb that is understood to be
there; more on this function later.

The paradigm: (gul - "can")
pres gule - can
past gula - was able to / could
fut gulo - will be able to

Inflection is aggluinative rather than internal, allowing new mambers
to be added to the class easily. It is optional to drop the inflection
for the present tense. "Gul" is understood to mean the same thing as
"gule." The only time you _have_ to use "gule" rather than "gul" is
when you need to put a stative ending on gul, i.e., "gulean," or
"guleanme." I will use "gul" and "zol" in the examples in this
section, but I could have just as easily used the long forms "gule"
and "zole." I should think that the short forms would generally be
preferred, but on the other hand, it's possible that people might be
more comfortable with the longer forms that are more parallel with te.
For now I leave open the question of which is more proper.

The two verbs I chose were selected deliberately to have simple
meanings. The realm of "gul" should be restricted as closely as
possible to the physical possibilty of an act, not whether the act is
permissable. "Zol" should be limited to a sense that the act is
absolutely imperative from the point of view of the speaker. Given how
closely these sorts of words cut to the essential character of the
vocabulary of the language (i.e., what sort of broad or narrow
distinctions do you make, and how do you think of these sorts of
words), I feared to try to add more to this conjugation for fear of
running into conflicts.

6.2 Functions of V3 Verbs

V3 verbs can stand in for a verb they modify (modifying it implicitly
rather than explicity) if it is understood from context what the verb
must be.
So in answer to:
"Can you have made 300 widgets by Friday?"
You could answer with a simple "yes" or "no," with an explicit answer
like:
"Yes I (gulean :make:) them."
or simpler:
"Yes I (gulean)."
Notice I use the present tense; this is legitimate because the abilty
to do the thing asked exists in the present as a state of being.
However, we could explicitly move the ability to do the action into
the future perfect, mirroring the question, and achieve a very similar
and perfectly legitimate meaning:
"Yes I ('o gula :make:) them."
or:
"Yes I ('o gula)."
In the first of these two, to illustrate the function of the verb,
first the 'o references the future as the time we're talking about,
then the gula shifts the action of the verb into that time's past and
stipulates that we are saying that the subject has the ability to
perform the action in question. Finally we say what the action in
question is.

Gul always refers to the _subject's_ ability, and zol always refers to
the _subject's_ requirement. When a gul-containing verb is in the
active voice, it says that the subject is able to perform the action
of the verb. When it is in the passive, the gul says that the subject
is able to _recieve_ the action of the verb.
So in the active:
"Wesley can eat a horse."
Wesley (gulean :eat:) a horse.
But in the passive:
"Wesley can be eaten by a horse." [possibly he's made of grass?]
Wesley (sean gul :eat:) by a horse.
Note that both forms are using the stative. We're talking about a
state of being here, not implying that a specific opportunity exists
for either party to chow down, and by using the stative we make that
explicit.

6.3 Stacking of V3 Modifiers, and V3 in Displaced and Passive
Forms

One can employ a V3 to modify another V3 (giving forms like "have to
be able to"). The modifying verb goes before the V3 it modifies. Time
displaces in the normal fashion. The only rule is that the combination
has to make sense; "to be able to have to" makes little sense, so it
is hard to imagine when you'e ever use "gul zol."

6.3.1 Simple Displaced Forms: (gul)

Imperfect Forms:

'e gul(e) - disallowed - redundant (see 3.3.1)
'a gul(e) - disallowed - redundant (see 3.3.1)
'o gul(e) - disallowed - redundant (see 3.3.1)

Perfect Forms:

'e gula - allowed - perfect
'a gula - allowed - pluperfect
'o gula - allowed - future perfect

Predictive Forms:

'e gulo - allowed - see 3.3.3
'a gulo - allowed - past predictive
'o gulo - allowed - future predicitve

6.3.2 Simple Passive Forms: (gul)

Imperfect Forms:

se gul(e) - allowed - present
sa gul(e) - allowed - imp. past
so gul(e) - allowed - future imperfect

Perfect Forms:

se gula - allowed - perfect
sa gula - allowed - pluperfect
so gula - allowed - future perfect

Predictive Forms:

se gulo - allowed - see 3.3.3 and 4.1.1
sa gulo - allowed - past predictive
so gulo - allowed - future predictive

6.3.3 Stacked Forms: (zol gul)

Imperfect Forms:

zol(e) gul(e) - allowed - present
zola gul(e) - allowed - imp. past
zolo gul(e) - allowed - future imperfect

Perfect Forms:

zol(e) gula - allowed - perfect
zola gula - allowed - pluperfect
zolo gula - allowed - future perfect

Predictive Forms:

zol(e) gulo - allowed - see 3.3.3 and 4.1.1
zola gulo - allowed - past predictive
zolo gulo - allowed - future predictive

6.3.4 Stacked Passive Forms:

Doubly displaced forms are possible with passive stacked V3 forms, but
will _not_ be considered here. Such forms will require more mental
gymnastics than I feel currently warranted relative to the actual
amount of use that I anticipate for them (which is small). In
presenting the passive forms, the forms shown shall represent the
preferred way _not_ to create double displacement.

Passive Imperfect Forms:

se zol(e) gul(e)
sa zol(e) gul(e)
so zol(e) gul(e)

Passive Perfect Forms:

se zola gul(e)
sa zola gul(e)
so zola gul(e)

Passive Predictive Forms:

se zolo gul(e) - see 3.3.3 and 4.1.1
sa zolo gul(e)
so zolo gul(e)

It shall be considered good form to avoid double displacement by
having the third particle in the present tense, putting the verb in
the time of the particle preceding it.

6.4 Examples of V3 Forms

Example:
"Why has Charlie had his jacket on all evening?"
"He has had to be able to leave at a moment's notice."
Sentence:
"He has had to be able to leave at a moment's notice."
Verb translation:
He (zol gula :leave:) at a moment's notice.
Note: Charlie's still around and has his jacket on, hence the perfect
rather than the pluperfect.

Example:"Why was Charlie carrying around that goddam cell phone?"
"He had to have been able to be reached because he's on call."
Sentence:
"He had to have been able to be reached because he's on call."
Verb translation:
He (sa zola gul :contact:) because he's on call.

Example:
"I am carrying around this canteen in case I get thirsty."
Verb translation:
"I (te :carry:) this canteen for when I (zolome :drink:).
Notes:
This is an interesting form, and not at all the only way the thought
could be expressed. I chose this particular idiom for "would have to
drink" because it illustrates a possible use of zol and because it
shows a very interesting use of the conditional mood. The sense of the
subordinate clause is quite hypothetical; the sense of "when + -me
(fut)verb" aligns with "in case of" quite well. Equally you could say:
....for when I (tome :thirst:).
I could, of course, make it explicit that the canteen covers any
number of incidences of thirstiness by using the stative, "zoloanme"
and "toanme." However, since my meaning is quite clear either way, and
since I intend this for informal speech, I choose to be a bit lazy
with my forms and use a form that is shorter but not as precisely
accurate.

Example:
"Since Jose will have been payed, he will have been able to buy
groceries."
Verb translation:
Since Jose (so ta :pay:), he ('o gula :buy:) groceries.

6.5 Final Note on V3 and Pronounciation

With the V3 verbs, which can be as much as four syllables long, maybe
longer than that if any two-syllable short-form verbs are added to the
class. The forms we have now are:
zoleanme, zoloanme, guleanme, and guloanme
I find these difficult to imagine pronouncing with only a stress on
the first syllable. Thus, for these words of four syllables or more, I
propose adding a second stress on the -an. So you get:
zňleŕnme, zňloŕnme, gůleŕnme, gůloŕnme
although you would not actually write this explicitly, it would be
understood. With the fictional two-syllable V3 addition "joteg," you
get a finer mess. For these, I propose leaving the stress on the first
syllable and adding one to the an for -anme forms:
jňtegeŕnme, jňtegŕnme, jňtegoŕnme
and move the stress to the second syllable for all other four syllable
forms:
jotčgean, jotčgoan, jotčgeme, jotčgame, jotčgome

7.0 Negation and Interrogation

7.1 Negation

Negation will be handled by the simple expedient of preceding the
superverb with the negation particle "no." So:
"Vespa had been eating cod tounges."
Could be:
"Tounge[acc.pl] cod[gen.pl] Vespa[nom.] ('a ta :eat:)."
in the positive, and:
"Tounge[acc.pl] cod[gen.pl] Vespa[nom.] (no 'a ta :eat:)."
in the negative.
When the reader reads or the listener hears "no," he is automatically
given to understand that a verb must follow, and that verb will be
understood to be negative.

7.2 Interrogation

Verbs (more properly, _sentences_, but the verb is too mixed up in the
process of interrogation to ignore interrogatives when discussing
verbs) are understood to be inherently demonstrative. They can be
rendered interrogative by some of the techniques that appear to be
currently agreed upon (listed exhaustively for rigour):
1) Tone (rendered ż? in text)
2) Tag questions
3) Presence of -ne compounds
One somewhat-less-certain status method I've proposed:
4) Placing "ke," what I would have in this function serve as a
"general tag," at the end of the sentence. In this function, it serves
as an explicit question marker, like the Chinese "ma" or the Japanese
"ka" do when placed at the end of a sentence in those languages.

As for the other proposed explicit way to make a question, the i no V
idiom, while I find it unobjectionable for other verb systems, I
rather dislike it for this system. It just doesn't feel right to me. I
would propose a different idiom for PVS to replace this.

Let us ponder the WH- compound-creator, the suffix "-ne," and its
meaning. In these compounds the morpheme has the sense of "what"... so
if "tid" is "time," "tidne" is "when." But here I want you to consider
a deeper meaning for -ne... something that can't precisely be
translated as a word, any more than the English past marker -ed can be
really thought of as a word, but more something that can be translated
as a function. Just a plain morpheme, if you like. The meaning I have
in mind for it is for ne to serve as a marker for interrogation. Think
about what this does to a noun if you suffix "-ne" to it... If "time"
is "tid," "tidne" is "tid[interrogative]" or "żtid?"... translated
crudely into our beloved English language ;) which lacks a particle
with quite such a function, I propose you get "what time" or "when".
The compound, with its emergent meaning derived from the meaning of
its parts, then functions in a sentence just as has been discussed.

Now, when a listener hears or a reader reads a V1 or V3 word of some
type, automatically he's expecting to hear the verb right after. But
what if he hears "ne" instead? To me, this would immediately create an
interrogative feel...

So the idiom I propose for Vnc is thus: in order to make a sentence
explicitly interrogative within the verb, you insert the interrogative
particle "ne" between the verb marker, be it a V1 or a V3, and the
verb. This listener/reader, hearing/seeing "ne" instead of a verb, and
knowing the PVS idiom, automatically says to himself, "A-ha, this is a
question!" and then listens for the verb. This is not ambiguous,
because the meaning of "ne," while required to be understood on a
deeper level than that of "ne = what..." remains entirely consistent,
and its function is simply determined logically by its position. If
you hear it after a verb marker, it's making the verb interrogative.
If you here it suffixed to some noun, you know it's rendering that
noun into a WH- question word. The decision to use "ne" as a separate
particle immediately preceding the most specific part of the verb is
made on the basis of what feels right (see next paragraph).

The only possible inconsistency I see here is that of -ne being a
suffix when used for question words versus ne being a free
interrogative word to make a verb interrogative, and to me this
trifling thing is entirely tolerable. Look at it this way - in every
case, you're using ne essentially to modify something preceeding it,
either a noun in the case of WH- words, or a verb marker in the case
of verbs. The reason I don't feel comfortable writing "ne" as a suffix
of the marker has to do with the way I hear the sound of the verb in
my mind. Let us make up the nonsense-verb "zutavi." Take:
żtean ne zutavi?
I hear the marker, tean, begin on a highish tone on the stressed first
syllable, and then decline as the word wears on, to a significantly
lower tone at the end. Ne is not particularly stressed, and it occurs
with a tone that is on the low sweep generated by the preceeding
marker. This in a sense gives it a "auditory attachment" to tean.
However, I hear "ne" being held for slightly longer than a last
syllable of a word would ordinarily warrant, and very slightly
separated from the preceeding marker in carefully inunciated speech.
So while the way you say the "ne" marker kind of "puts it on to" the
preceding verb marker, the slight spacing and the extra length of the
vowel tend to separate it slightly and draw attention to it. Following
the interrogative marker, the tone rises again with the stressed first
syllable of the verb. The reasons I hear it this way: 1) because
attaching the "ne" to the verb marker by making it part of the same
tone and not stressing the "ne" puts the interrogative modification on
the preceeding verb marker, which I find intuitively both satisfying
and necessary; 2) because separating it a little with spacing and
vowel length draws attention to the particle and by extension to the
interrogative nature of the verb - after all, to ask a question is to
seek attention; and 3) because the abrupt rise in tone and stress with
the beginning of the verb serves to clearly mark the beginning of the
verb.

If one inunciates clearly, there should be no mistaking "ne" for the
beginning of a verb.

However, I _do_ acknowledge verbs innocently starting with ne- as a
_possible_ source of confusion in some situations. While most of the
time people are really quite good at telling where word boundaries
are, and while this ought to be especially easy with NGL's handy
stress system, in times of hurry or stress, I could see where
confusion _could_ arise. This is why I am supplying a remedy.
Ordinarily it is not necessary to explicitly mark a sentence as
demonstrative, that is automatically implied. However, I propose a
particle "ve," which has the deep-meaning of [demonstrative], and
which may be inserted between the marker and the verb to make the verb
explicitly demonstrative (even though the sentence as a whole may be
rendered interrogative by other means. Let us look at an example.
Fictional Vocabulary:
netoba - to commit immoral acts with livestock
toba - to buy livestock
Example:
Ma McSweeny comes goes out into the pasture looking for Billy-Ray, on
account of he's wanted on the phone. Upon seeing him she says:
"Dear Sweet Jesus, Billy-Ray, you (ta netoba) sheep!?"
Translation:
"Dear Sweet Jesus, Billy-Ray, you were committing immoral acts with
that sheep!?"
However, this is a stressful, not to mention hurried, situation; what
our intrepid young zoophile might _actually_ hear would be:
"Dear Sweet Jesus, Billy-Ray, you (ta ne toba) sheep?!"
Translation:
"Dear Sweet Jesus, Billy-Ray, were you buying that sheep?!"
Obviously this is the sort of misunderstanding we want to avoid at all
cost. So what Ma McSweeny _could_ have said was:
"Dear Sweet Jesus, Billy-Ray, you (ta ve netoba) sheep!?"
Which makes it _explicit_ that the verb itself (although not
necessarily the sentence as a whole, as in this case tone makes the
sentence somewhat interrogative anyway - although, really, you must
admit, given what Ma has just seen, it's rather a rhetorical question)
is demonstrative, and that therefore we're talking about immoral acts,
not purchase. Of course, "ve" makes the verb _strongly_ demonstrative
- handy for those situations when you've caught someone in a
compromising position and feel the need for extra emphasis.

Of course, I can think of an even worse situation that might arise.
Example:
Farmer Bob is at a livestock show. He is examining a pig
speculatively. Cyrus comes along. He asks:
"Farmer Bob, you ('e to ne toba) pig?"
Translation:
"Farmer Bob, are you going to buy that pig?"
But Farmer Bob is a bit hard of hearing, and Cyrus doesn't really
articulate all that well. What Farmer Bob hears is:
"Farmer Bob, you ('e to netoba) pig?" [tonal question]
Translation:
"Farmer Bob, are you going to commit immoral acts with that pig?"
Farmer Bob breaks both of Cyrus' legs. He's a touchy guy, that Farmer
Bob.

I really have no remedy for Cyrus. He really should have inunciated
more clearly, or chosen his words with more care. One cannot fix all
the world's troubles; Cyrus had better not come to _me_ with his
hospital bill.

Note on choice of tense for the preceding example: I chose the present
predictive because intuitively I felt it fit the narrative better than
the future imperfect. I realise this is at least very close to being a
violation of the principle I laid down in 3.3.3 about avoiding the use
of this tense. However, in this case I offer two justifications. One
is the aforementioned aesthetic feel of immediacy I felt the passage
needed. And if all else fails, I remind you that that there Cyrus as
was talkin', he's a reeel iggernint feller, he don't know no good
grammer... ;->

8.0 Important Adverbs of Time

A useful feature in a language is a short-cut method for saying
approximately how long ago an event occured or how soon it is expected
to occur. This is particularly so with recent time. Some languages
handle this using colourful and interesting idioms; in French (English
examples following... English is also idiomatic in this fashion), for
example, you can put an event in the more-or-less immediate future
using expressions like "je vais manger...," "I am going to eat" or "je
viens de manger," "I just ate." I propose to do the same thing, but
instead using adverbs to modify the verb. Here are the four I propose:

pes - in a short time
ulat - a long time from now
deh - a short time ago
oteg - a long time ago

Adverbs of time, when not placed inside the verb, modify the first
time-level that they can be applicable to. In order to modify a deeper
time level, you simply insert the adverb into the verb, immediately
preceeding the particle giving the time level to be modified.
Example:
In an interview for a criminal investigation, taking place long after
the infamous deeds took place:
"What had he been doing before you walked in?"
"He had just been molesting the goat, Officer."
Sentence:
"He had just been molesting the goat, Officer."
Verb translation:
He ('a deh ta netoba) goat, Officer.
Note:
In this context, how far in the past the walking-in took place is
clear to the speakers from the larger context of the situation. The
respondant wants to indicate that the molestation had been completed a
very short time before he had walked in, so he places "deh" in front
of the ta to make it clear that it is the molestation narrative level
whose time is being clarified, not the walking-in level. If he had
said:
He deh ('a ta netoba) goat, Officer.
The respondant is saying something more like "He just a moment ago had
been molesting the goat," in this case. Putting it this way restricts
the walking-in to having happened a short time ago, and clearly this
violates the established context of an interview taking place months
after the event.

9.0 Some Obvious Possible Modifications

Changing "-an" to "-n"

The ending "-an" creates two vowels next to eachother whenever it goes
with te, 'e or se in any of their forms. Aside from the contracted
forms tan, san, 'an, my system does nothing to rectify this. Indeed
this is intentional, because I think it sounds pretty. However, be
that as it may, forms like soan and 'ean may be considered to be in
intolerable conflict with the principles upon which the sound system
operates. In that case, I would propose shortening "-an" to "-n" when
the stative marker follows a vowel, giving:
ten 'en sen
tan 'an san
ton 'on son

Admitting more V3 verbs

The two V3 verbs in the system right now are purposefully
insufficient. It would eventually be a good idea to admit more, such
as a verb for "ought to." (The English verb "may," equivalent to the
German verb "dürfen," but somewhat crippled :-), could, I think, be
rendered by using the NGL verb for :allow: and using it in the
passive.

Adding inflection for person and number and gender

Optionally, I think it should be possible to admit inflections to the
system for any and all of person, number and gender. These should, I
think, go on the most specific part of the verb, and on their own
would be sufficient to mark the verb as present tense participative
definite - making the word "te" particularly redundant with inflected
verbs, so that it would be able to be dropped much of the time. Of
course the past and future forms of te would be needed to render those
tenses.

Adding more modifying suffixes

Possibly "-an" and "-me" could be supplemented with more suffixing
modifiers giving still finer shades of meaning.

APPENDIX A:
Glossary of PVS Morphemes

Translations given are approximate.

te - (V1) is doing; present verb marker
ta - (V1) was doing; past verb marker
to - (V1) will be going; future verb marker
'e - (V2) has; present referencer
'a - (V2) had; past referencer
'o - (V2) will have; future referencer
se - (V2) is; passive marker/present referencer
sa - (V2) was; passive marker/past referencer
so - (V2) will be; passive marker/future referencer
-an - stative marker
-me - conditional marker
no - negator
gul - (V3) is able to, also marks verb and supplies tense
zol - (V3) has to, also marks verb and supplies tense
-e - V3 present tense suffix
-a - V3 past tense suffix
-o - V3 future tense suffix
-ne,ne - interrogative marker, not suffixed when used with
verb markers for interrogative of verb. Suffixed to
nouns to give WH- question words (7.2)
ve - explicit emphatic demonstrative marker, used between
verb marker (V1 or V3) and verb (7.2)

Order of precedence of suffixes:
V3 tense endings - "-an" - "-me"

Proposed Supplementary Adverbs:

pes - (av) in a short time
ulat - (av) a long time from now
deh - (av) a short time ago
oteg - (av) a long time ago

Words We Can Live Without (7.2):

netoba - (Vnc) bugger (an animal)
toba - (Vnc) buy (an animal)

APPENDIX B:
Common Tense Sequences and Overall Resulting Tense

In this appendix, I will denote the tenses that result from combining
helpers in different tenses. The overall tense is independent of the
nature of the helpers in the sequence - it depends only on the tenses
of the members of the sequence. Therefore, in this section "e" will be
used to denote "present tense particle," "a" will be used to denote
"past tense particle," and "o" will be used to denote "future tense
particle."

e - present imperfect (happening now)
a - past imperfect (was happening then)
o - future imperfect (will be happening)

e e - pres. imp. (for passive imperfect forms)
a e - past imp. (ditto)
o e - fut. imp. (ditto)

e a - perfect (has [just] done)
a a - pluperfect (had done)
o a - future perfect (will have done)

e o - present predictive (is going to do)
a o - past predicitve (was going to do)
o o - future predicitve (will be going to)

e e e - pres. imp. (for form with a V3 modifying a V3

+ verb)
a e e - past. imp. (ditto)
o e e - fut. imp. (ditto)

e a e - perfect (ditto)
a a e - pluperfect (ditto)
o a e - future perfect (ditto)

e o e - pres. pred. (ditto)
a o e - past pred. (ditto)
o o e - fut pred. (ditto)

a a a - plupluperfect (had had done - I used this in
one example)
email: c72...@captain.morgan.ucs.mun.ca
lose the captain to email me

[Jack Durst]

I'm not going to quote his proposal, since it's rather long, but instead
address the issues which were apparant to me on the first reading.

1. Incomplete compatibility with existing core grammar and other systems:

Since you haven't seen all the debates, I'll give you a quick background,
The grammar of NGL is devided into two parts, the core grammar and the
modules. By nessecity, all core grammar is shared by all modules, which
means that modules must conform with certain rules to insure
compatibility. One of these rules is that the core grammar form must
always be meaningfull in every proposal. The core grammar form is known
as the generic.

a. The accepted (but not ratified) generic for verbs is marked for
person and number, this should be made meaningfull.
b. The accepted rules of word order allow inflectional particles
to
follow as well as preceede the verb, and allow adverbs to
intervene between verbs and inflections.
c. modules must self-mark by use, I can see places where yours and
Jerry's would colide
d. There is a derivational morpheme -je which turns adjectives
into
stative (in the linguist's sense) verbs, the use of se in that
context is unnessicary.
e. There are already a whole suite of morphemes for deriving all
classes of words from varrious roles in relation to the verb.
f. The only accepted cases in NGL are nomative, accusative,
dative.
everything else is accomplished with preposition+generic. (Except
genitive, which uses the idiom of adding a person marker to the
thing
possessed)
g. You needn't re-propose the question and interrogation set here,
it's already been accepted into core grammar in the form you
proposed
earlier. NOTE: I thought ke was the question marker!

2. Terminology:

I must be unusually dense, either that or my Linguistics training is
causing cognitive interferance, can you explain the stative/participative
distinction again? It seems *almost* like you mean perfect/imperfect but
there're all sorts of wierd connotations and your use of 'e for this
listinction later really makes it hard for me to grasp.

What, exactly do you mean by conditional? It is a mood in real languages,
but it's much more restricted than you use the term for; what you're using
seems more like an irrealis (the proper term for what you describe as a
general hypothetical tense).

The typical term for what you call an "immediate future" is a non-past
tense, as these are usually also emphatic presents in languages which have
them. What You use as V3 would be modal verbs in normal jargon. They add
mood to the other words; however their existance begs some interesting
questions about your other reference to mood.

3. General concerns and comments:
I like your use of te as a proverb (BTW: That is the correct term) and it
goes well with existing conventions. Your system seems to be rather
powerfull, but I still have trouble (on the gut level) mixing verbs of
different tenses and aspects to get a flexable mood/tense/aspect, maybe a
little more explanation would help. I just can't see how the subtilties
develop. Why not use the same tense markers throughout, with just the
stem to distinguish V1 from V2? It seems like there would be no
difficulty.

Why give the particles meaning at all? yes I know you feel odd about it,
but, trust me, it works better that way. Sure, it's a nice shortcut, but
it leaves some odd loopholes in the language. I'd have to run a
cross-check to be sure, but it seems like it would act strangely in
general use.

NOTE: Have dosn't universally shift actions to the past, in Latin it made
the compound future tenses.

I like the displaced tanses. It adds a lot of power to the system with
very little overhead. I'd like for the 'e te forms to have some idomatic
meaning, just to fill in the paradigm, but otherwise, I'm pleased.
However, how does this work with se te? I can't see the contrast. You
were unclear about which tense carried the reference at first, I had to
re-read. It would work better if te carried the referent tense, since the
system feels like it wants it that way.

Your explanation of the voice system went completely over my head (And
I've studied languages with free word order and voice linguistically)
please explain again. Your examples make sense, but I don't quite grasp
how you intend to do it. Technical explanation is in order. The role
structure of verbs can be changed at whim through derivation (with
semantic shift, but only a small one); NGL already has word fronting and
emphasis markers, so moving something to the nominative isn't really
nessicary given the existing markers. Many languages avoid passive forms;
I do like the predictive passive, though.

The multiply displaced tense idea works, but has this feeling of
non-concreteness to me which seems to clash with the rest of the system.
Your first rule is unnessicary, since the shorter form will almost always
occur to a person before the longer form; the second makes it sound like
you're having second thoughts about including it at all. I say keep it.

Couldn't te be considered a V3 verb? As a linguist, if I found it in a
language, I'd almost certainly class these two sets together as a complete
set of modal verbs (with te being irregular) with the following paradigm
(using correct terminology):
te Indicative
gul Permissive/potential
zol Advisory/nessecitive
... fill in the chart ...

Your admixture of mood and voice seems odd to me, could you explain how
you got this? How does your use of the term subject differ from your
understanding of nominative case? (In linguistics, they're the same) In
most languages, the mood would apply to the verb as a whole, anyway.
Explain more about your stacked mood verbs.

COMMENT: Your V3 section is overlong, given the subject matter: you make
your point twice. The section on passives is overshort.

I like your temporal adverbs; this, however is not the place for them.
They should be proposed seperately as vocabulary items (under core
grammar) instead of being tied to a proposal for a specific system. I ask
that we accept them as lexical items.

Sincerely,
Jack Durst
Sp...@sierra.net
[this posting written in Net English]

[Jack Durst]

---------- Forwarded message ----------
Date: Thu, 11 Jun 1998 23:20:04 GMT
From: Stephen DeGrace <ephi...@thermopylae.net>
Reply-To: c72...@captain.morgan.ucs.mun.ca

Subject: Re: NGL: My Proposed Verb System, Version 0.1

On Wed, 10 Jun 1998 20:44:00 -0700, Jack Durst <sp...@sierra.net>
wrote:

>I'm not going to quote his proposal, since it's rather long, but instead
>address the issues which were apparant to me on the first reading.

Jack, thank you seriously for your comments. You are tough but fair,
and I am deeply grateful for your superior knowledge of linguistics.
My ability to communicate linguistic concepts is woefully incomplete,
making it difficult for me to compose acceptible proposals regardless
of the quality or lack thereof of my ideas. The discipline you impose
is therefore very welcome, if not always easy on the pride.

This whole experience has been good. There is no way to learn about
something quite like getting in there and getting your hands dirty. I
feel that being able to make proposals has sped my progress in
learning of the nature of the NGL project and about linguistic
proposals generally, and I am grateful to you and everyone else in the
group for having patience with me.

>1. Incomplete compatibility with existing core grammar and other systems:
>
>Since you haven't seen all the debates, I'll give you a quick background,
>The grammar of NGL is devided into two parts, the core grammar and the
>modules. By nessecity, all core grammar is shared by all modules, which
>means that modules must conform with certain rules to insure
>compatibility. One of these rules is that the core grammar form must
>always be meaningfull in every proposal. The core grammar form is known
>as the generic.

Okay, I realise I missed a lot, and I realise that I have an excessive
tendency to leap in with both feet. Thanks for taking the time here to
set me straight.

> a. The accepted (but not ratified) generic for verbs is marked for
> person and number, this should be made meaningfull.

Okay, I would do this by placing the inflection for person and number
on the verb itself rather than on any of the markers.

> b. The accepted rules of word order allow inflectional particles
>to
> follow as well as preceede the verb, and allow adverbs to
>intervene between verbs and inflections.

Having any adverb admissable into the sequence of markers doesn't
create any problem. Further, so long as the order of the markers is
allowed to remain rigid, I don't see a problem with them being able to
follow a verb.

> c. modules must self-mark by use, I can see places where yours and
> Jerry's would colide

Okay, I'd need to work with Jerry on this. In any case, I'm not even
certain if my system ought to be considered for acceptance as a
module, on reflection.

> d. There is a derivational morpheme -je which turns adjectives
>into
> stative (in the linguist's sense) verbs, the use of se in that
> context is unnessicary.

Okay, as you write later, giving the particles meaning may not have
been such a hot idea anyway; since se amd 'e are probably not needed
to perform actual functions, their functions could be revoked.

> e. There are already a whole suite of morphemes for deriving all
>classes of words from varrious roles in relation to the verb.
> f. The only accepted cases in NGL are nomative, accusative,
>dative.

Thanks. Actually, reviewing the grammar cheat-sheet a little while
earlier this evening, I did notice that... if for some reason I put
out another version of my proposal, I shall be sure to take out the
genitives.

> everything else is accomplished with preposition+generic. (Except
> genitive, which uses the idiom of adding a person marker to the
>thing
> possessed)
> g. You needn't re-propose the question and interrogation set here,
> it's already been accepted into core grammar in the form you
>proposed
> earlier. NOTE: I thought ke was the question marker!

All right, I withdraw this section of the proposal, it was in poor
taste and I apologise for it. I propose instead that inserting a no
between the te and the verb can serve as the i no V 's manifestation
in PVS. I suppose some earlier criticisms that I should have put
behind me in a more adult fashion still stung...

>2. Terminology:
>
>I must be unusually dense, either that or my Linguistics training is
>causing cognitive interferance, can you explain the stative/participative
>distinction again? It seems *almost* like you mean perfect/imperfect but
>there're all sorts of wierd connotations and your use of 'e for this
>listinction later really makes it hard for me to grasp.

Think rather that it is my lack of linguistic training that is causing
interference. I lack much of the words I need to explain myself. If
you are able to correct my mistakes at any point in this regard, it is
very much appreciated. Anyway, I'll try and explain
stative/participative again.

The sequence of particles that I have going before the verb serve to
establish the verb's position in time. The aspect marker, which is
suffixed to the first particle in the string no matter how long the
string is, tells us whether the action is a single, discrete, specific
case, or whether the verb refers to a set of cases. So the
participative:
We te play baseball
says that the speaker is actively engaged in playing a game of
baseball, right this very moment. So mom tells Jimmy to come inside,
and he replies that he can't because he and his friends are playing
baseball - this is an example of the use of the participative.
Whereas:
We tean play baseball.
says the speaker and the group of people he is including (not
including the person he's speaking to - is this a use of the first
person paucal?) play baseball a number of times - perhaps they belong
to a leage. The act of playing baseball referrs to an unspecified
number of incidences, i.e. games, and since the present tense is used,
the speaker fully expect there to be games in his future. With an
imperfect stative verb, the time being talked about is situated
somewhere in the middle of the set of actions the verb describes
rather than in the middle of a discrete action as in the
participative.

Does that help? What have I missed?

>What, exactly do you mean by conditional? It is a mood in real languages,
>but it's much more restricted than you use the term for; what you're using
>seems more like an irrealis (the proper term for what you describe as a
>general hypothetical tense).

Then maybe that's a better thing to call it. Thinking about it, what I
want is a way to make the action of the verb hypothetical.

>The typical term for what you call an "immediate future" is a non-past
>tense, as these are usually also emphatic presents in languages which have
>them. What You use as V3 would be modal verbs in normal jargon. They add
>mood to the other words; however their existance begs some interesting
>questions about your other reference to mood.

Well, perhaps a better way to deal with my so-called conditional mood
would be to make a V3 verb, lets call it "xi," that serves as a verb
marker to make the verb hypothetical.

>3. General concerns and comments:
>I like your use of te as a proverb (BTW: That is the correct term) and it
>goes well with existing conventions. Your system seems to be rather
>powerfull, but I still have trouble (on the gut level) mixing verbs of
>different tenses and aspects to get a flexable mood/tense/aspect, maybe a
>little more explanation would help. I just can't see how the subtilties
>develop. Why not use the same tense markers throughout, with just the
>stem to distinguish V1 from V2? It seems like there would be no
>difficulty.

First of all, I would propose, from looking at your criticism, to
combine V1 and V3 into a single conjugation of verb markers V1 with
the single irregular member te. V1 marks the verb, conveys modal
information about it, and marks the verbs position in time. V2 is used
for specifying narrative levels, and in addition contains the verb
"se" which has the double function of marking a narrative level and
making the verb passive rather than active.

I'll try and give a little more explanation of how verb tense are
constructed from sequences of particles with simpler tenses.

There are two temporal functions the particles preceding the verb have
to do. One is to pinpoint the level in time of the narrative, and the
other is to position the action of the verb relative to that narrative
level. If a verb is modified only by a V1, both the narative level and
the action of the verb are placed in the time specified by the V1. If
the V1 is preceded by another particle, whether a V2 or a V1 modifying
the V1 marking the verb, that preceding particle specifies the
narrative time, and the V1 then places the action relative to that
time. If these particles are preceded by another, the first particle
specifies a narrative time, the second specifies a narrative time
relative to that narrative time, and the third, marking, verb
specifies the position of the action of the verb relative to that
narrative time.

The actual identities of the helpers in the sequence does not matter
to the tense of the verb. Only the sequence of the tenses of the
helpers is important in determining the overall time of the verb.
Let's take "I was going to call you." The narrative time, the level of
time that serves as the narrative's present, is located in the
speaker's past. However, the action of the verb takes place in that
time's future. Let us say that our speaker is making an excuse to his
girlfriend about why he never called, and will shortly be trying to
make an excuse. That would, in my view, make the action hypothetical
(you could restate the sentence, "I would have called you, but..."),
so I elect to use the V1 marker xi. Since the action is displaced from
the narrative time, however, I need a helping verb to specify the
narrative time. Since the voice is active, I use 'e. To set the
narrative level in the past, I use 'a. To put the hypothetical action
in that time's future, I use xio. So, using fixed word order, I get:
Om 'a xio :call: mac.

Question, Jack: am I allowed to leave off the person and number
inflections of call under some circumstances? Or should that have
been:

Om 'a xio :call:om mac.

Please pardon my slowness.

>Why give the particles meaning at all? yes I know you feel odd about it,
>but, trust me, it works better that way. Sure, it's a nice shortcut, but
>it leaves some odd loopholes in the language. I'd have to run a
>cross-check to be sure, but it seems like it would act strangely in
>general use.

Okay... I need to think more about this.

>NOTE: Have dosn't universally shift actions to the past, in Latin it made
>the compound future tenses.

The way I envisioned it (and I realise I did not make it at all easy
to extract this from my lengthy monolog), if you *have* a past action,
it is complete, if you *have* a present action, the action is ongoing
(which is why the narrative marker + verb marker combinations with the
verb marker in the present are considered redundant. They are only
used when you have a string of markers, for example when one V1 is
modifying another V1 modifying a verb, or when the verb is passive,
and you need to keep from creating displaced tenses when you don't
want them), and if you *have* a future act you are predicting the act
will come to pass. When you *have* an act in the narrative's future,
and that does not place the act itself in the speaker's past, this
"having" implies a degree of rock-hard certainty that the act will
take place, for example if the act is in the more easily-predicted
near future.

>I like the displaced tanses. It adds a lot of power to the system with
>very little overhead. I'd like for the 'e te forms to have some idomatic
>meaning, just to fill in the paradigm, but otherwise, I'm pleased.

I sort of need for 'e te to equal te. That way, se te can basically
equal te[passive], or xi zol can be a simple present tense form. That
said, I see no reason why the definitively redundant 'e te forms
cannot develope idiomatic meaning over time, especially since those
form are currently unused under my system.

>However, how does this work with se te? I can't see the contrast.

se te = te[passive]. If you want to make the verb passive, you have to
use se as the first particle in the string. But as an economy measure,
I have se simultaneously marking the passive and marking the narrative
level. This is real handy if you want to render the perfect, say "se
ta," because you don't need any more helpers in order to pinpoint the
narrative. But in order to be able to make a string of helpers with
time-marking powers able to be non-displaced, I needed redundant time
combinations that would give you imperfect, undisplaced tenses from a
sequence of markers. I get my solution from the idea that if you have
a present action, then that action is ongoing, which is exactly the
same as saying that you're doing the action. So following a
narrative-marking particle of any sort with a present tense marker
makes the action of the verb happen right at the narrative time. Here
is a list of two-marker combinations that I say are equal to a single
marker:
e + e = e
a + e = a
o + e = o
The combination of any narrative time marker with a present tense verb
marker always gives the effect of making the verb have the narrative
marker's tense. It says that from the point of view of the time I'm
talking about, the verb is _not_ displaced.

What needs to be clear here is that tense is established by the
sequence of tenses of the markers. These are _all_ past predictive:
"I was going to call, but..."
Om 'a xio :call:, :but:...
"I was going to be beaten up."
Om sa to :beat up:.
"I was going to have to be able to swim."
Om zola gulo :swim:.

You see that I gave all helping verbs that aren't actively engaged in
marking the verb and its position relative to the narrative time the
ability to mark narrative time. I feel that this feature is very
economical. However, it creates a need for combinations that come out
as being equivalent ot the non-displaced imperfect tenses.
Fortunately, placing the verb marker in the present tense following a
narrative marker with the time you want to refer to performs this
function admirably. Say in that last example you want to say, "I am
having to be able to swim," for some reason. This tense is the present
imperfect, which in the simplest sort of verb would be rendered by
just "te." However, we have two helpers in this case, and we need them
both. So we say:
Om zol gul :swim:.
This type of form is exactly equivalent to the tense of se te.

>You
>were unclear about which tense carried the reference at first, I had to
>re-read. It would work better if te carried the referent tense, since the
>system feels like it wants it that way.

Admittedly, the system as I have it makes the verb marker able to take
on two possible temporal meanings depending on whether it is alone or
when it is preceded by other temporal markers. When the marker is
alone, it references the narrative time and then places the action of
the verb in that narrative time. When it is preceded by another helper
or helpers, however, those helpers reference the temporal level and
the marker is changed in function to placing the action relative to
the narrative time.

>Your explanation of the voice system went completely over my head (And
>I've studied languages with free word order and voice linguistically)

I envy you and I respect you. Seriously.

>please explain again. Your examples make sense, but I don't quite grasp
>how you intend to do it. Technical explanation is in order. The role
>structure of verbs can be changed at whim through derivation (with
>semantic shift, but only a small one); NGL already has word fronting and
>emphasis markers, so moving something to the nominative isn't really
>nessicary given the existing markers. Many languages avoid passive forms;
>I do like the predictive passive, though.

I know it, about avoiding the passive; I was lectured enough in school
about avoiding the passive voice. That, however, I categorically
refuse to do when writing in English. I like the passive voice,
sometimes it's just the right tool for the job, IMO. Passive is
particularly important in science, BTW, escecially in my field of
chemistry. Passive voice is used _extensively_ in the literature, and
speaking for myself, my lab notes are written almost entirely in the
passive. It's what we were taught was good form. I fully realise that
you can say just about any passive sentence actively somehow, and I
realise that many or most users will almost always couch things so as
to avoid using the passive. Nevertheless, I feel strongly that an
optional passive voice is a healthy thing for NGL to have. As you have
stated (paraphrasing), NGL's purpose in life is to be powerful much
more so than it is to be simple. I think adding an option to the
language that allows the nominative of the sentence to be the passive
recipient of the verb's action rather than its active perpetrator has
advantages. I submit that there would be a strong tendency for some
speakers to want to put the noun most central to the topic of the
conversation in the nominative case; allowing the passive voice allows
speakers to have the option to preserve this tendency while still
talking about an event which happened _to_ the subject. Furthermore,
allowing the passive voice allows one to be completely ambiguous
regarding the nature of the action's perpetrator. English teachers
seem to abhor this, but I find it an occasionally very useful device.
It comes in particularly handy in writing my lab notes, as I have
mentioned.

Well, anyway, of course I could live without the passive if the
feature were rejected, and I realise that with NGL's flexibilty, you
may well be able to produce examples of forms that handle the cases I
have mentioned just as well as the passive would. Nevertheless, I am
trying to make a case for the inclusion of the passive voice, because
I just like passives.

>The multiply displaced tense idea works, but has this feeling of
>non-concreteness to me which seems to clash with the rest of the system.
>Your first rule is unnessicary, since the shorter form will almost always
>occur to a person before the longer form; the second makes it sound like
>you're having second thoughts about including it at all. I say keep it.

Okay. I was ambiguous about it because I saw there being a huge number
of cases, and I despaired of tackling them all; however, I think I
will sit down and think more seriously about the role of multiply
displaced tenses.

>Couldn't te be considered a V3 verb? As a linguist, if I found it in a
>language, I'd almost certainly class these two sets together as a complete
>set of modal verbs (with te being irregular) with the following paradigm
>(using correct terminology):
> te Indicative
> gul Permissive/potential
> zol Advisory/nessecitive
> ... fill in the chart ...

Oh, I _like_ this! This is so neat! I propose a V1 of modal verbs, as
you have here, and a V2 of secondary helpers se and 'e. I further add
the modal V1 "xi" to the list to take care of my conditional mood (as
above).

>Your admixture of mood and voice seems odd to me, could you explain how
>you got this?

I'm not certain where the problem lies; if the mood is hypothetical,
and the voice is active, then we are saying that it is hypothetical
that the subject will perform the action. If the mood is hypothetical
and the voice is active, then we are saying that it is hypothetical
that the subect will have this action performed on it. To me it seems
intuitively obvious. I think I have misunderstood the question; could
you explain further what seems odd?

>How does your use of the term subject differ from your
>understanding of nominative case? (In linguistics, they're the same)

I wasn't sure if an uninflected noun which derives its function from
its position and the prepositions associated with it could count as
having a "case," so I was a bit leery about just saying "nominative,"
in case that would be considered a crass usage when used in reference
to fixed word order sentences. Your clarification on this point is
gratefully accepted.

>In
>most languages, the mood would apply to the verb as a whole, anyway.
>Explain more about your stacked mood verbs.

Okay, the first modal modifies the mood of the second modal, which in
turn modifies the mood of the verb. The order of the modals is thus
important to the overall mood transmitted to the verb. "zol gul," "to
have to be able to," is different from "gul zol," "to be able to have
to." the only rule with stacking is that the first invariably
modifies the second which in turn modifies the verb, and the compound
has to overall make sense. I would deal with the admissability of
sequences of V1's on a case-by-case basis.

Further, at the same time as the V1 modals carry modal information to
the verb, they have implicit tense which simulataneously carries
temporal information to the verb. In the case of a stacked modal with
two components, the first marker's time is the narrative time, and the
second marker's time is the action's time relative to the narrative
time. It works the same as any sequence of modifiers, you look at the
order of tenses of the modifiers, regardless of what the modifiers
actually are, and from that you work out the narrative time and the
displacement from narrative time if any.

>COMMENT: Your V3 section is overlong, given the subject matter: you make
>your point twice. The section on passives is overshort.

Okay, thanks for this comment. If I re-release, I will try and correct
this.

>I like your temporal adverbs; this, however is not the place for them.
>They should be proposed seperately as vocabulary items (under core
>grammar) instead of being tied to a proposal for a specific system. I ask
>that we accept them as lexical items.

I am deeply sorry for this and other impropieties I have committed,
and thank you for your magnamimity.

I have a mini-proposal for a verb system module for NGL that is based
off the PVS proposal and inspired be Jack's many helpful and
insightful (if occasionally bruising to the pride :-) comments. I will
detail it in a separate post, but suffice it to say the idea is to
create a set of optional modal verbs for NGL which have the
simultaneous power to displace verbs from their time given in their
native NGL inflection, allowing a different optional route to
predicitves and pluperfects, and allowing the referencing of complex
narrative levels.

Sincerely,

Stephen

[Jack Durst]

On Thu, 11 Jun 1998, Stephen DeGrace wrote:

> On Wed, 10 Jun 1998 20:44:00 -0700, Jack Durst <sp...@sierra.net>
> wrote:
>

> >I'm not going to quote his proposal, since it's rather long, but instead
> >address the issues which were apparant to me on the first reading.
>

> Jack, thank you seriously for your comments. You are tough but fair,
> and I am deeply grateful for your superior knowledge of linguistics.

Thank you. Believe me, it's all training, when I started I couldn't even
clearly tell the difference between an adverb and an adjective. You'll
pick it up in time; especially working on projects like NGL.

> This whole experience has been good. There is no way to learn about
> something quite like getting in there and getting your hands dirty. I
> feel that being able to make proposals has sped my progress in
> learning of the nature of the NGL project and about linguistic
> proposals generally, and I am grateful to you and everyone else in the
> group for having patience with me.

It's nice to know I finally have someone on the group who thinks like I
do! Both Julian and Jerry are better at thinking out the ramifications of
things; I'm good at playing with morphemes.

> > a. The accepted (but not ratified) generic for verbs is marked for
> > person and number, this should be made meaningfull.
>

> Okay, I would do this by placing the inflection for person and number
> on the verb itself rather than on any of the markers.

That's exactly the existing system; so there'll be no trouble.

> > b. The accepted rules of word order allow inflectional particles
> >to
> > follow as well as preceede the verb, and allow adverbs to
> >intervene between verbs and inflections.

> Having any adverb admissable into the sequence of markers doesn't
> create any problem. Further, so long as the order of the markers is
> allowed to remain rigid, I don't see a problem with them being able to
> follow a verb.

Actually, I'd treat the string of markers like a single word in syntax,
since the order of the particles is already fixed, and allow the adverb on
either side of it as is done with vector particles.

> > c. modules must self-mark by use, I can see places where yours and
> > Jerry's would colide

> Okay, I'd need to work with Jerry on this. In any case, I'm not even
> certain if my system ought to be considered for acceptance as a
> module, on reflection.

Well, keep working on it, it'll get there.

> > e. There are already a whole suite of morphemes for deriving all
> >classes of words from varrious roles in relation to the verb.
> > f. The only accepted cases in NGL are nomative, accusative,
> >dative.
>

> Thanks. Actually, reviewing the grammar cheat-sheet a little while
> earlier this evening, I did notice that... if for some reason I put
> out another version of my proposal, I shall be sure to take out the
> genitives.

Cool. Just making sure you knew. There's a reason the cheat-sheet is
made to fit on one side of a sheet of paper...

> > g. You needn't re-propose the question and interrogation set here,
> > it's already been accepted into core grammar in the form you
> >proposed
> > earlier. NOTE: I thought ke was the question marker!
>

> All right, I withdraw this section of the proposal, it was in poor
> taste and I apologise for it. I propose instead that inserting a no
> between the te and the verb can serve as the i no V 's manifestation
> in PVS. I suppose some earlier criticisms that I should have put
> behind me in a more adult fashion still stung...

I understood that, but the existing ordering rules would already perscribe
exactly that order, so it would be redundant and would risk introducing
inconsistantcies.

> >2. Terminology:
> >
> >I must be unusually dense, either that or my Linguistics training is
> >causing cognitive interferance, can you explain the stative/participative
> >distinction again? It seems *almost* like you mean perfect/imperfect but
> >there're all sorts of wierd connotations and your use of 'e for this
> >listinction later really makes it hard for me to grasp.
>

> Think rather that it is my lack of linguistic training that is causing
> interference. I lack much of the words I need to explain myself. If
> you are able to correct my mistakes at any point in this regard, it is
> very much appreciated. Anyway, I'll try and explain
> stative/participative again.

I'll try. I haven't seen a language with features like this, so even
*I*'m not sure of all the terminology.

> establish the verb's position in time. The aspect marker, which is
> suffixed to the first particle in the string no matter how long the
> string is, tells us whether the action is a single, discrete, specific
> case, or whether the verb refers to a set of cases. So the

Ah! I see. this is indeed aspect; but is usually combined into other
systems of aspect. When they occur alone, the proper terms are punctual
(for participative) and continual (for what you call stative). I get it
now, thanks.

> Does that help? What have I missed?

That helps a lot. I was interpreting your perfect/imperfect distinction
as containing something which it does in most languages, but dosn't in
your proposal because this is covered by the continual/punctual
distinction elsewhere.

> >What, exactly do you mean by conditional? It is a mood in real languages,
> >but it's much more restricted than you use the term for; what you're using
> >seems more like an irrealis (the proper term for what you describe as a
> >general hypothetical tense).

> Then maybe that's a better thing to call it. Thinking about it, what I
> want is a way to make the action of the verb hypothetical.

That would be using an irrealis mood.

> would be to make a V3 verb, lets call it "xi," that serves as a verb
> marker to make the verb hypothetical.

That makes a lot more sense.

> First of all, I would propose, from looking at your criticism, to
> combine V1 and V3 into a single conjugation of verb markers V1 with
> the single irregular member te. V1 marks the verb, conveys modal
> information about it, and marks the verbs position in time. V2 is used
> for specifying narrative levels, and in addition contains the verb
> "se" which has the double function of marking a narrative level and
> making the verb passive rather than active.

OK, that makes sense. I'd accept that.

>
> I'll try and give a little more explanation of how verb tense are
> constructed from sequences of particles with simpler tenses.

<snip>
OK, that's a much better explanation and I suggest you use it in your next
version of the proposal.

> Question, Jack: am I allowed to leave off the person and number
> inflections of call under some circumstances? Or should that have
> been:
> Om 'a xio :call:om mac.

Number is modular in NGL, as usual, the generic is -0 (linguist's
shorthand for no ending) and refers to a non-specific or precise number
(eg. òl ka'am - an animal, ol ka'a - 1 animal) and can always be used
wherever number is required. The person ending must always be used in the
1st and 2nd persons; but the nominative pronoun can be dropped. The most
reduced form of the sentence is thus:
'a xio :call:o mac.

> >NOTE: Have dosn't universally shift actions to the past, in Latin it made
> >the compound future tenses.
>

> The way I envisioned it (and I realise I did not make it at all easy
> to extract this from my lengthy monolog), if you *have* a past action,
> it is complete, if you *have* a present action, the action is ongoing
> (which is why the narrative marker + verb marker combinations with the
> verb marker in the present are considered redundant. They are only

> want them), and if you *have* a future act you are predicting the act
> will come to pass. When you *have* an act in the narrative's future,
> and that does not place the act itself in the speaker's past, this
> "having" implies a degree of rock-hard certainty that the act will
> take place, for example if the act is in the more easily-predicted
> near future.

Ah! I see. this makes sense now. Part of my role is to question reasons
and since the languages you know tend to have a classic have=past
connection, I was concerned that you might not have known about it.

> >I like the displaced tanses. It adds a lot of power to the system with
> >very little overhead. I'd like for the 'e te forms to have some idomatic
> >meaning, just to fill in the paradigm, but otherwise, I'm pleased.
>

> I sort of need for 'e te to equal te. That way, se te can basically
> equal te[passive], or xi zol can be a simple present tense form. That
> said, I see no reason why the definitively redundant 'e te forms
> cannot develope idiomatic meaning over time, especially since those
> form are currently unused under my system.

OK then, we'll leave it up to the native speakers.

> >However, how does this work with se te? I can't see the contrast.
>

> se te = te[passive]. If you want to make the verb passive, you have to
> use se as the first particle in the string. But as an economy measure,
> I have se simultaneously marking the passive and marking the narrative
> level. This is real handy if you want to render the perfect, say "se
> ta," because you don't need any more helpers in order to pinpoint the
> narrative. But in order to be able to make a string of helpers with

I see. I think hadn't read your discussion of passives yet when I wrote
that and was a bit confused by a specific line. I actually like the
economy of it in this sense.

> You see that I gave all helping verbs that aren't actively engaged in
> marking the verb and its position relative to the narrative time the
> ability to mark narrative time. I feel that this feature is very
> economical. However, it creates a need for combinations that come out
> as being equivalent ot the non-displaced imperfect tenses.

At least it works well. I like the system with this explanation, but it
was a little fuzzy the first time through.

> >You
> >were unclear about which tense carried the reference at first, I had to
> >re-read. It would work better if te carried the referent tense, since the
> >system feels like it wants it that way.
>

> Admittedly, the system as I have it makes the verb marker able to take
> on two possible temporal meanings depending on whether it is alone or
> when it is preceded by other temporal markers. When the marker is
> alone, it references the narrative time and then places the action of
> the verb in that narrative time. When it is preceded by another helper
> or helpers, however, those helpers reference the temporal level and
> the marker is changed in function to placing the action relative to
> the narrative time.

OK. It makes sense once you read it all the way through, but it
definitely gave me a bump the first time through. I think it was the
order in which you presented this part. A little re-write to make it
clear which carried the tense early on will do away with the problem.

> >Your explanation of the voice system went completely over my head (And
> >I've studied languages with free word order and voice linguistically)

> I envy you and I respect you. Seriously.

Thank you. It's nice to know that that class on verbal morphology paid
off.

> >nessicary given the existing markers. Many languages avoid passive forms;
> >I do like the predictive passive, though.
>

> I know it, about avoiding the passive; I was lectured enough in school
> about avoiding the passive voice. That, however, I categorically
> refuse to do when writing in English. I like the passive voice,

[...]

> to avoid using the passive. Nevertheless, I feel strongly that an
> optional passive voice is a healthy thing for NGL to have. As you have
> stated (paraphrasing), NGL's purpose in life is to be powerful much
> more so than it is to be simple. I think adding an option to the
> language that allows the nominative of the sentence to be the passive
> recipient of the verb's action rather than its active perpetrator has
> advantages. I submit that there would be a strong tendency for some

> talking about an event which happened _to_ the subject. Furthermore,
> allowing the passive voice allows one to be completely ambiguous
> regarding the nature of the action's perpetrator. English teachers
> seem to abhor this, but I find it an occasionally very useful device.

This is a pretty good explanation of why voice is a generally usefull
thing. Now, the counter explanation of why it isn't nessicary in NGL:

1. Free word order: with free word order the natural tendency
is to place whatever is most important first or last, regardless of it's
case. Having a passive for this purpose (its main one in English) is thus
pointless; and what's less makes the cases ambiguous.
2. Easy derivation: In languages with free word order and voice,
the voice changes are almost always used to allow for one or more of the
main roles of the verb (agent (=nominative/ergative case) patient
(=absolutative case) or beneficiary (=dative case)) to be optionally
ommitted. This is the case of passives in Swedish, Latin, and Russian,
and of antipassives in Basque (ergative languages are naturally passive).
Since you can change the role structure of a verb in NGL with only a very
simple derivation to add or remove any of the three elements this too is
unnessicary.

Since both of its major functions are already covered, there is no extra
use in having a passive.

> Well, anyway, of course I could live without the passive if the
> feature were rejected, and I realise that with NGL's flexibilty, you
> may well be able to produce examples of forms that handle the cases I
> have mentioned just as well as the passive would. Nevertheless, I am
> trying to make a case for the inclusion of the passive voice, because
> I just like passives.

If you like passives, you really ought to consider adding an entire voice
system with active, passive, and middle voices.

> >The multiply displaced tense idea works, but has this feeling of
> >non-concreteness to me which seems to clash with the rest of the system.
> >Your first rule is unnessicary, since the shorter form will almost always
> >occur to a person before the longer form; the second makes it sound like
> >you're having second thoughts about including it at all. I say keep it.
>

> Okay. I was ambiguous about it because I saw there being a huge number
> of cases, and I despaired of tackling them all; however, I think I
> will sit down and think more seriously about the role of multiply
> displaced tenses.

Cool.

> Oh, I _like_ this! This is so neat! I propose a V1 of modal verbs, as
> you have here, and a V2 of secondary helpers se and 'e. I further add
> the modal V1 "xi" to the list to take care of my conditional mood (as
> above).

Good idea. I like this.

>
> >Your admixture of mood and voice seems odd to me, could you explain how
> >you got this?
>

> I'm not certain where the problem lies; if the mood is hypothetical,
> and the voice is active, then we are saying that it is hypothetical
> that the subject will perform the action. If the mood is hypothetical
> and the voice is active, then we are saying that it is hypothetical
> that the subect will have this action performed on it. To me it seems
> intuitively obvious. I think I have misunderstood the question; could
> you explain further what seems odd?

I think it was your use of the term conditional which made things come out
oddly in my mind.

> >most languages, the mood would apply to the verb as a whole, anyway.
> >Explain more about your stacked mood verbs.
>

> Okay, the first modal modifies the mood of the second modal, which in
> turn modifies the mood of the verb. The order of the modals is thus
> important to the overall mood transmitted to the verb. "zol gul," "to
> have to be able to," is different from "gul zol," "to be able to have
> to." the only rule with stacking is that the first invariably
> modifies the second which in turn modifies the verb, and the compound

OK, that makes sense. I think it was your phrasing which had me off the
first time.

> has to overall make sense. I would deal with the admissability of
> sequences of V1's on a case-by-case basis.

Why not just state that all of them are valid but that some of them have
idomatic meanings?

> the verb, they have implicit tense which simulataneously carries
> temporal information to the verb. In the case of a stacked modal with
> two components, the first marker's time is the narrative time, and the
> second marker's time is the action's time relative to the narrative
> time. It works the same as any sequence of modifiers, you look at the
> order of tenses of the modifiers, regardless of what the modifiers
> actually are, and from that you work out the narrative time and the
> displacement from narrative time if any.

Cool. That seems like it'll work.

>
> >I like your temporal adverbs; this, however is not the place for them.
> >They should be proposed seperately as vocabulary items (under core
> >grammar) instead of being tied to a proposal for a specific system. I ask
> >that we accept them as lexical items.
>

> I am deeply sorry for this and other impropieties I have committed,
> and thank you for your magnamimity.

No problem. It's hard joining a group and being the new guy, and we'll
do our best to help you integrate.

> I have a mini-proposal for a verb system module for NGL that is based
> off the PVS proposal and inspired be Jack's many helpful and
> insightful (if occasionally bruising to the pride :-) comments. I will
> detail it in a separate post, but suffice it to say the idea is to
> create a set of optional modal verbs for NGL which have the
> simultaneous power to displace verbs from their time given in their
> native NGL inflection, allowing a different optional route to
> predicitves and pluperfects, and allowing the referencing of complex
> narrative levels.

Cool idea! I'm looking forward to it. If you're using my system, though,
e-mail me with all your questions first, because I have a feeling we'll be
creating inconsistantcies without doing this.

Sincerely,
Jack Durst
Sp...@sierra.net
[this posting written in Net English]

...