Justin B Rye 1995-2001
glossary sounds numbers nouns adjuncts pronouns verbs phrases syntax coinages examples lexicon

SECTION X: COINAGES

Xa - Compounding

New words can be derived from existing dictionary entries either by adding specialised word-endings (see suffixing, below) or by merging together whole words in the process called compounding.  This is the mechanism that produces such somewhat arbitrary English words as "waterfall", "beachfront" or "babysit".

The rules governing this language's compounding system are similarly flexible: words which occur as a set phrase can simply merge together into a single longer word, often dropping grammatical endings and/or introducing <-a-> to avoid illegal strings of consonants:

<barin ach·es> "a fish's egg" --> <barinach> "fishegg(s)"

<apas reisen·es> "a cat's eye" --> <apasareisen> "cat's-eye"

<nuchar ana> "go with" --> <nucharana> "accompany"

<ji epa> "cause to become king" --> <jiepa> "crown"

Placement of stress in these compounds is usually regular, ignoring the stress of the constituent words; <jí> plus <epá> is <jiépa>.

Demonstrative adverbs like <nullu> ("here" - see VId) and unhyphenated number-words like <sachasoikh> ("ninety" - see IIIc) are also examples of compounding.

Xb - Suffixing

Many special endings are available to turn existing words into related concepts, often changing word categories (turning verbs into nouns or the like):

<-(u)ru>

Meaning: associative adjective - "-al, -ish"

Form: <-ru> after any vowel, <-uru> after any consonant

Converts: postposition or noun to adjective

<entich> "air" --> <entichuru> "aerial, of the air"

<naithin> "after" --> <naithinuru> "latter" (also as adverb, "afterwards")

<ómme> "mother" --> <ómmeru> "maternal"

<-(a)ga, -(o)khá>

Meaning: characteristic adjective - "-ful, -y, -ous"

Form: unpredictable - often both are possible

Converts: verb or noun to adjective

<entich> "air" --> <entichaga> "airy, supplied with air"

<ulasu> "know" --> <ulasuga> "knowing, knowledgable"

<taís> "water" --> <taisokhá> "watery, water-filled"

<-e, -é>

Meaning: "abstract" noun - "-ness, -ity, -ation"

Form: steals stress in short words

Converts: verb or adjective to noun

<sifulu> "die" --> <sifulué> "death"

<chargi> "white" --> <chargie> "white(ness)"

<nutuekh> "poor" --> <nutuekhe> "poverty"

<-(a)gé>

Meaning: "degree" noun - "-ness, -th"

Form: itself a compound of <·aga·e>

Converts: adjective to noun

<rasek> "strong" --> <rasekagé> "strength"

<molnie> "tall" --> <molniegé> "height"

<karegon> "heavy" --> <karegongé> "weight"

<-(a)t, -ot>

Meaning: "result" noun - "-age, -ture"

Form: <ot> commoner near <o> or <u>

Converts: verb to noun

<karthek> "write" --> <karthekat> "scripture, writings"

<sifulu> "die" --> <sifuluot> "a corpse"

<dortila> "break" --> <dortilat> "fracture, fragments"

<-ién, -én>

Meaning: "agent" noun - "-er, -or, -ist"

Form: <én> used after <i>; n.b. irregular (IVd)

Converts: verb to noun

<karthek> "write" --> <karthekién> "writer"

<eota> "trade" --> <eotaién> "trader"

<ortothi> "command" --> <ortothién> "commander"

Xc - The Syllabificator

This is the random-syllable-generator I used to fill out the lexicon; I include it in this package because that's simpler than trying to describe the full set of abstract phonotactic constraints, and because it's as near as I can get to "releasing the source code".

DICE

Each randomly generated syllable needs five dicerolls (using standard six-sided dice) - call the results "A", "B", "C", "D" and "E" and go through the tables below.  Normal words should go through the system three to five times; very common words may have fewer syllables, and perhaps irregular stress.

INITIAL CONSONANT

Plug the first two dicerolls into the following table:

	A=1	A=2	A=3	A=4	A=5	A=6
B=1	<m>	<n>	<n>	<r>	<l>	-
B=2	<p>	<t>	<t>	<ch>	<k>	-
B=3	<p>	<d>	<t>	<j>	<g>	-
B=4	<b>	<d>	<s>	<s>	<s>	-
B=5	<f>	<f>	<th>	<th>	<kh>	-
B=6	-	-	-	-	-	-

The dashes of course indicate "no initial consonant", though if a previous syllable ended in a consonant, that can be a stand-in initial letter for this syllable.

VOWEL

A simpler table:

	C=1	C=2	C=3	C=4	C=5	C=6
---	<i>	<e>	<a>	<a>	<o>	<u>

If this vowel creates a string of two identical vowels (e.g. <aa>, <ii>) or if it creates a string of three "open" vowels (e.g. <eae>, <oea>) then go back and insert an <n> at the start of the current syllable.

FINAL CONSONANT

For a provisional syllable-final consonant, use the last two dicerolls:

	D=1	D=2	D=3	D=4	D=5	D=6
E=1	<m>	<n>	<n>	<r>	<r>	<l>
E=2	<m>	<n>	<n>	<r>	<r>	<l>
E=3	<p>	<t>	<t>	<ch>	<k>	<k>
E=4	<f>	<th>	<s>	<s>	<s>	<kh>
E=5	-	-	-	-	-	-
E=6	-	-	-	-	-	-

If this is the last syllable of the word, then that's it finished.  On the other hand if there's a following syllable then some of the above results are modified:

RULE ONE (for E=1 or E=2):

The consonants <m>, <n>, <r>, <l> are modified before a following consonant:

<m> becomes <n> before <n>, <t>, <d>, <ch>, <j>, or <th>

<n> becomes <m> before <m>, <p>, <b>, or <f>

<r> becomes <l> before <l>

<l> becomes <r> before <r>

RULE TWO (for E=3 or E=4):

The consonants <p>, <t>, <ch>, <k>, <f>, <th>, <s>, <kh> are dropped completely unless they're word-final.

MONOSYLLABIC EXAMPLE:

Single-syllable word with dicerolls A-B-C-D-E = 3-6-4-4-3:

initial <->, vowel <a>, final <ch> unmodified: result <ach> ("egg")

POLYSYLLABIC EXAMPLE:

Four-syllable word, rolling 1-1-5-4-2 / 2-1-5-2-3 / 2-6-1-2-1 / 1-5-3-2-5:

first syllable <m>-<o>-<r>

second syllable <n>-<o>-<t> (but the <t> is thrown out)

third syllable <->-<i>-<n> (but the <n> becomes <m>)

fourth syllable <f>-<a>-<->

which makes the full word <mornoimfa> ("mor-NOYM-fa"), available for use as a random vocabulary item.

SECTION XI: Examples