PLAUSIBLY ELVISH
================
APOLOGIES FOR THE .TXT FILE
---------------------------
Man, I'm sick of HTML. All those
tags and
s and
- s just to get some pretty colors and fonts? No thank
you. I can't even tell the difference between a capital I
and lowercase L in most fonts, and let me tell you, that is
bad, *especially* for conlangs. How are you supposed to
learn a word in a language you don't know yet when you can't
even tell what the letters are?
So. Plain text. Sorry about the ugly font. Maybe someone
will invent a better monospace font one day and then we'll
all just use that.
WHAT IS "PLAUSIBLY ELVISH"?
---------------------------
It's the working title of my new constructed language
(conlang), which is designed to sound like Tolkien's Elvish
language Sindarin but have an unambiguous syntax.
Why Sindarin? Well, I like Sindarin. Always have. Why the
unambiguous syntax? That one's a bit harder to explain. I
guess when I design "naturalistic" languages I always feel
a bit like I'm fumbling around in the dark--like I don't
*really* understand the rules behind the example sentences
I'm writing. There's always a nagging doubt in my mind that
my language is not internally consistent. I'm a programmer
by trade, so that particular type of doubt *really* bothers
me.
If you find the idea of learning a "logical" language
offputting--don't worry, I do too. Lojban[1], this ain't. My
intention is not to make an inhuman or robotic-sounding
language; just the opposite.
So I hope you'll find that the "logical" features of
Plausibly Elvish (henceforth PE) are helpful and
mind-expanding rather than annoying and limiting.
[1]: https://mw.lojban.org/papri/Lojban
WORDS
-----
The central concept of PE is that every word refers to a
type of relationship or interaction between things. This is
very different from English, where some words just refer to
things standing by themselves--we call those "nouns". In PE,
any word can act like a noun, a verb, or an adjective,
grammatically speaking--but the definition of each word is
always expressed in terms of a relationship.
Philosophically, the PE worldview rejects the idea that
"things" exist on their own, independent of context. The
context is what brings "things" into existence, by making it
useful/meaningful to draw distinctions between them. Of
course, contexts themselves are formed from groups of
things, which are in turn defined by their own contexts...
there's no base case for this recursion, but as you look
deeper and deeper, the recursive ripples do converge into
stillness. I believe this is similar to the Buddhist concept
of *co-dependent arising*.
So, what are these "relationships" defined by PE words?
The most obvious type of word that expresses a relationship
corresponds to a verb in English: a word like "eat" defines
a relationship between the eater and the thing eaten.
But even in English, a lot of nouns are actually defined by
relationships. Take a word like "ancestor". Someone can only
be an ancestor in relation to some other person or people.
Or a word like "king". A king is only a king because he has
subjects. Or perhaps "hero". A hero is only a hero *to*
someone else.
PE allows you to leave the participants of these
relationships implicit, just English does. You can say "I
eat" without specifying what you are eating. But you have to
inflect the word specially for that, so there's always that
reminder that you're trading precision for brevity.
I AM, YOU ARE
-------------
With that introduction out of the way, let's look at some
of the most basic phrases in PE: introducing yourself!
Im Arwen
1ps.v1 Arwen.v1.n1
"I am Arwen"
The format of the examples in this book, BTW, is:
where a "gloss" is kind of like a translation, but with
a bunch of linguistic annotation describing exactly what
each word does in the sentence.
Even in this simple example, there's a lot of syntactical
subtlety to unpack. So let's go through it.
The verb of the sentence (or "predicate" to be perfectly
precise[1]) is, surprisingly, what we're used to thinking of
as a pronoun: the first-person singular (abbreviated 1ps)
*Im*. In PE, the first-person singular pronoun names a
relationship defined to have one participant. We say that it
has a *valence* of one: one other word can grammatically
attach to it, and that other word *must* be present for the
sentence to be grammatical.
In glosses, I'll abbreviate valence numbers as *v1*, *v2*,
etc.
The second word is a name, *Arwen*. This also names a
relationship that has one participant, so it has valence 1.
However, what we are referring to in this sentence is not
the abstract state of "Arwenness" ("Arwenity?"), but a
person: "she who is Arwen". In order to express this, we
have to specify that we are using the word *Arwen* to refer
to the first (and only) participant in the Arwenness
relationship.
That's what the *n1* in the gloss does. *n1* means we're
"naming" (hence the n) the first participant in the
relationship. This doesn't correspond to any overt marking
on the word "Arwen", but conceptually, it's there. It *is*
overtly marked on other words, which we'll see later.
That was a lot of grammar for one simple example. Don't
worry if you didn't get it. More examples will follow, and
at some point it will click.
Let's look at another example: turning the first sentence
on its head.
Arwen-d im
Arwen.v1 1ps.v1.n1
"Arwen is me"
Remember, the word "Arwen" expresses a relationship, so it
can function as a predicate! The structure of this sentence
is the same as the last one, but "Arwen" and "Im" have
switched places. The only notable difference is that the
word "Arwen" now takes a suffix "-d". What's up with that?
Remember in the last example how "n1" was in the gloss, but
didn't correspond to anything overt in the PE sentence?
Well, now we're overtly marking the *absence* of "n1" on
"Arwen". That's what the "-d" suffix does.
This may seem strange, but it's very convenient. As you'll
see later, "n1" comes up a *lot*--basically any time we want
to use a word as a noun. If we had to apply a suffix to
every word to nounify it, that suffix would be repeated
*everywhere*.
Astute readers may notice that the word "im" has not changed
its form through all of this. Neither the presence nor the
absence of "n1" is marked on "im". That's just because "im"
is irregular.
Let's see another example:
Le Gildor
2p.v1 Gildor.v1.n1
"You are Gildor"
This has the same structure as the earlier example, but now
we're using the second-person pronoun "le".
As before, we can reverse the word order:
Gildor-ed le
Gildor.v1 2p.v1.n1
"Gildor is you"
Here, the absence of "n1" on Gildor is marked by "-ed",
which is the form that "-d" takes after "r". The addition of
"n1" to "le" is not marked: "le", like "im", is irregular.
[1]: the distinction is that "verbness" is an intrinsic
quality of a word--something a word has or doesn't have,
even when it's just sitting in the dictionary not being
used. However, "predicate" is a role that a word plays in a
sentence. A word doesn't have "predicateness" or
"non-predicateness" unless and until you use it.
EXPERIENCERS
------------
It may seem strange to you that I've been saying that words
like "Arwen" and "Gildor", which we think of as names,
define relationships. They're relationships with one
participant?? That's not a relationship!
In PE, every word has an "experiencer", which defines the
other half of these "one-sided relationships". When not
explicit, the experiencer is assumed to be the
speaker or writer of the sentence.
So in a sentence like "Im Arwen", every word has an implicit
experiencer. When Arwen says "Im", she's referring to
herself; When Gildor says "Im", he's referring to himself.
If the two of them are conversing, they use "le" to mean the
other person. The implicit experiencer is what enables these
apparent asymmetries to exist.
However, we can make the experiencer explicit if we wish,
using the particle "go".
Im go im Arwen
1ps.v1 exp 1ps.v1.n1 Arwen.v1.n1
"What I experience as me is Arwen"
Or perhaps:
Im Arwen go im
1ps.v1 Arwen.v1.n1 exp 1ps.v1.n1
"I am what I experience as Arwen"
Of course, the "im" in "go im" is another word, another
participant in a relationship, whose experiencer is left
implicit. So we can go a level deeper:
Im Arwen go im go im
1ps.v1 Arwen.v1.n1 exp 1ps.v1.n1 exp 1ps.v1.n1
"I am what what I experience as myself experiences as Arwen"
And perhaps even:
Im go im Arwen go im go im
1ps.v1 exp 1ps.v1.n1 Arwen.v1.n1 exp 1ps.v1.n1 exp 1ps.v1.n1
"What I experience as myself is what what I experience as myself experiences as Arwen"
I'm reminded of Monty Python:
"so she said, she said, she said, the dead crab she
said, she said."
This gets absurd pretty quickly, which is why the
experiencer is normally left implicit. This is another case
of recursion with no base case, and it has deep
philosophical implications for how we define the concept of
"self". It is, in the words of Douglas Hofstadter, a
"strange loop".
It's much more useful to be able to mention *other people*
in the experiencer role.
Im go Gildor Arwen
"According to Gildor, I am Arwen"
Im Arwen go Gildor
"I am Arwen according to Gildor"
"I am Gildor's concept of what Arwen is"
You can put the experiencer at the beginning of the
sentence, in which case it modifies the whole sentence.
Go Arwen, im Gildor
"According to Arwen, I am Gildor"
Verbs
-----
In general, the word order of a sentence is
verb-subject-object.
Some verbs have no subject or object. We say they have a
valence of zero.
Hwil
wind.v0
"It is windy"
Some verbs have only a subject.
Losto Gildor
sleep.v1 Gildor.v1.n1
"Gildor sleeps"
Others take a subject and object.
Banna Gildor nen
drink.v2 Gildor.v1.n1 water.v1.n1
"Gildor drinks water"
In this sentence, the fact that *banna* "drinks" takes a
subject and an object is indicated in the gloss by *v2*:
it has a valence of 2.
Sometimes, we want to leave the object implicit--perhaps
we don't even know what it is. The way to do that is by
adding a suffix, usually -r, to the verb.
Bannar Gildor
drink.v2.i2 Gildor.v1.n1
"Gildor drinks"
The *i2* in the gloss corresponds to the *-r* suffix, and
means the second valence slot of the verb *banna* is left
implicit.
We can also leave the subject of a verb implicit.
Banwen nen
drink.v2.i1 water.v1.n1
"Water is drunk"
And finally, everything can be left implicit:
Bant
drink.v2.i1.i2
"Drinking happens"
Remember when I said that every word expresses a
relationship? Here's a good example of that:
Banna Gildor laethwen
drink.v2 Gildor.v1.n1 milk.v2.n2.i1
"Gildor drinks milk"
Take a look at the gloss for "milk". "milk.v2.n2.i1"? What
the heck is that?
It turns out that the word "milk" has a valence of 2. The
first slot is the animal that produces the milk. The second
slot is the thing that's produced. So if you want to refer
to "milk" in general instead of "cow's milk" or "goat's
milk", you have to leave slot 1 implicit--hence the *-wen*
suffix.
The gloss also confirms, via the "n2", that what we're
naming with the word "laeth" is the milk, not the animal
that produced it.
If we do mention the producer of the milk, slot 1 is no
longer implicit, so we drop the *-wen*.
Banna Gildor laeth megli
drink.v2 Gildor.v1.n1 milk.v2.n2 bear.v1.n1
"Gildor drinks bear's milk"
Badass!
The general principle is that whenever you use a word, all
its valence slots must be "accounted for" somehow. Either
they're filled by other words, as is usual for verbs, or
they're looped back into the meaning of the word with
"n1" or "n2" markings, or they're marked as implicit.
NEGATION
--------
Negation of a verb is accomplished by making that verb the
subject of a special negating verb, *ú*. *Ú* means roughly
"to be false" or "to not be so".
This causes *soft mutation* on the verb after *ú*. Mutation
is a topic that we have not yet covered. For now, note that
soft mutation causes the initial *b* of *banna* to change to
*v*.
Ú vanna Gildor laeth megli
false.v1 drink.v2 Gildor.v1.n1 milk.v2.n2 bear.v1.n1
"Gildor doesn't drink bear's milk"
It may seem strange that the verb (actually, an entire verb
phrase starting with *vanna*) can be the subject of another
verb. In English we actually do something similar, when we
say "it is not the case that Gildor drinks bear's milk."
However, we have to introduce the sub-clause with "that".
There's no word corresponding to this usage of "that" in PE,
because none is needed; the grammar is unambiguous without
it, because of the way valence works.
Often, the verb *ú* will be attached to the following verb
with a hyphen. This is a stylistic convention.
Ú-vanna Gildor laethwen
false.v1 drink.v2 Gildor.v1.n1 milk.v2.n2.i1
"Gildor doesn't drink milk"
You can use an experiencer with *ú*, to express doubt. The
*ú* can't be hyphenated to the following word in this case.
Using an experiencer also cancels the soft mutation.
Ú go im banna Gildor laethwen
false.v1 exp 1ps.v1.n1 Gildor.v1.n1 milk.v2.n2.i1
"I don't think Gildor drinks milk"
*Ú* can also be used to negate things that we don't usually
think of as verbs.
Ú-Arwen
false.v1 Arwen.v1.n1
"non-Arwen"
*Ú-Arwen* means "someone/anyone who isn't Arwen; non-Arwen
people".
While this may not seem that useful, it can be helpful for
expressing subtle distinctions between negative sentences.
Ú-'wanno Arwen
false.v1 disappear.v1 Arwen.v1.n1
"Arwen doesn't disappear"
The sentence above just talks about Arwen; it says nothing
about whether other people disappear or not.
Gwanno ú-Arwen
disappear.v1 false.v1 Arwen.v1.n1
"*Arwen* doesn't disappear"
In the sentence above, on the other hand, we're saying
that while Arwen does not disappear, others do. In English,
we'd probably express this by putting more emphasis on the
word "Arwen," to communicate that we're contrasting her with
other people.
SUMMARY OF AFFIXES AND PARTICLES
--------------------------------
-d/-ed : added to a name to make it into a predicate, by
removing its implicit *n1* marking.
go : introduces an experiencer
-r : added to a valence-2 word to leave slot 2 implicit.
This effectively makes verbs intransitive.
-wen : added to a valence-2 word to leave slot 1 implicit.
This turns verbs into the passive voice.
Dropping the valence suffix of a verb (usually -o, -a, -ta,
-na, -tha, or -da) leaves all slots implicit.
Inflections for v0 Nounlikes: