Trinitarian: Difference between revisions

From Linguifex
Jump to navigation Jump to search
(giving up and merging)
Line 1: Line 1:
== Phonology/Orthography ==
{{Infobox language
|name          = Weddish
|nativename    = Vediš
|pronunciation = /ˈve(ː).dɪʃ/
|speakers      = 0.01
|date          = 2014
|familycolor = Indo-European
|fam1 = Indo-European
|fam2 = Germanic
|fam3 = West Germanic
|fam4 = High German
|fam5 = Yiddish
|fam6 = Basque/Hebrew
|creator      = [[User:aquatiki|Robert Murphy]]
|created      = 2013
|setting      = Jewish intermarriage / Systematic Theology
|posteriori    = Yiddish
|script        = [[w:Hebrew script|Hebrew]]
}}
'''Weddish''' (''Weddish'': '''װעדיש''', ''X"Q'': '''וֶדִש''', ''Romanization'': '''Vediš''') is a constructed, ''a posteriori'', naturalistic [[auxlang]], made from Yiddish with heavy influences from Hebrew, English, German, and some Basque.  It has ergative-absolutive [[Linguistics:Morphosyntactic alignment|morphosyntactic alignment]] and a pervasive yet symbolic use of the dual.  It is meant to promote the institution of marriage, foster better communication between persons, and improve the tenor of Systematic Theological discussions.  It is well-suited as an auxlang for Jewish intermarriage.


{| class="wikitable"
The language was created in 2013 by [[User:aquatiki|Robert Murphy]] as part of an assignment at [[w:Covenant Theological Seminary|Covenant Theological Seminary]] for Professor Jerram Barrs.
|+ Consonants
 
== Philosophy ==
First and foremost, there is the Creator-Creature distinction.  That means, God is wholly other than the Universe.  Second, human beings are made in the image of God.  This means that we are persons -- like God is -- and our ''agency'' is our single-most important feature.  Third, we reflect the image of God as females '''''and''''' males.  The marriage bond is God-created and a fundamental part of our identity in this life.  Hence it is, that we may divide the world into: actors, non-actors, and actions.  Stated grammatically, this list becomes: ergative nouns, absolutive nouns, and verbs.  Furthermore, ergative nouns may be divided up into married and non-married actors, which we will mark with the ''dual'' or not.
 
We have said that ontologically speaking, there are ultimately only two (God and not-God).  However, the 800lb. gorilla in the room -- philosophically speaking -- is abstraction.  Since before Pythagoras, abstract nouns (such as numbers, "goodness", etc.) had been held by the Greeks to be ontic.  Westerns betray their affinity to Greek ideal, classifying humanity as ''homo sapiens'' - "thinking man".  We seek to disenthrall ourselves from this metaphysic and banish abstract nouns from our language.  Numerals are adjectives.  Infinity is that which we cannot see the end of.  Universals are the same as aggregates ("all" is the same as "sum" ... but not "some"!).
 
The Language Creation Society (as fine institution) waves a flag of the Tower of Babylon.  Unlike the hubristic man of the initial chapters ''In the Land of Invented Languages'' (Arika Okrent's chronicle), we do not believe we can undo by human effort what God has done.  Our aim is for Jews and Christians to discuss the truth ''better''.  The status of English as a ''lingua franca'', German as a language of science, and Hebrew as a holy language suggests that an Indo-European language is still the best option for an ''a posteriori'' auxlang.  Rather than compete with those vital languages, however, it seems most prudent to build upon a base off them all at once, utilizing a language that is already as eclectic as English, yet built off German and Hebrew.  That language is Yiddish.
 
== Phonology ==
Weddish has 25 consonantal sounds, which is typologically average <ref>http://wals.info/chapter/1</ref>, and common in Europe as well as the Middle East.  English speakers will find it to be common, apart from the lack of <tt>/w/</tt> and the ubiquity of <tt>/x/</tt> (like the ''ch'' in ''Bach'' or ''loch'').  Weddish has 6 vowels, which is also average<ref>http://wals.info/chapter/2</ref>, as is the resulting consonant-to-vowel ration<ref>http://wals.info/chapter/3</ref>.  This is typologically equivalent to Yiddish and Hebrew, but far less than German or English.
 
{| {{Table/bluetable}} style="float:left;"
|+ '''Consonants of Weddish'''
! colspan="9"|Consonant phonemes
|-
! colspan="2"|
! style="text-align: center;"|Labial
! style="text-align: center;"|Alveolar
! style="text-align: center;"|Post-Alveolar
! style="text-align: center;"|Dorsal
! style="text-align: center;"|Glottal
|-
! colspan="2" | Nasals
| '''מ''' <tt>/m/</tt>
| colspan="2" align="center" | '''נ''' <tt>/n/</tt>
| * <tt>/ŋ/</tt>
|
|
! Labial
! Dental
! Alveolar
! Palatal
! Velar
! Glottal
|-
|-
! Nasal
! rowspan="2"|Stops
| '''Mm''' /m/ || '''Nn''' /n/ || || || * /ŋ/ ||
!<small>voiceless</small>
| '''פ''' <tt>/p/</tt>
| colspan="2" align="center" | '''ט''' <tt>/t/</tt>
| '''ק''' <tt>/k/</tt>
| '''א''' <tt>/ʔ/</tt>
|-
|-
! Voiced Stop
!<small>voiced</small>
| '''Bb''' /b/ || '''Dd''' /d/ || ||  || '''Gg''' /g/ ||
| '''ב''' <tt>/b/</tt>
| colspan="2" align="center" | '''ד''' <tt>/d/</tt>
| '''ג''' <tt>/g/</tt>
|  
|-
|-
! Voiceless Stop
! rowspan="2"| Fricatives
| '''Pp''' /p/ || '''Tt''' /t/ || || || '''Kk''' /k/ || '''Aa''' /ʔ/
! <small>voiceless</small>
| '''פֿ''' <tt>/f/</tt>
| '''ס''' <tt>/s/</tt>
| '''ש''' <tt>/ʃ/</tt>
| '''כ''' <tt>/x/</tt>
| '''ה''' <tt>/h/</tt>
|-
|-
! Voiced Fricative
! <small>voiced</small>
| '''Vv''' /v/ || '''Zz''' /z/ || '''Jj''' /ʒ/ || || '''Qq''' /ɣ/ ||
| '''װ''' <tt>/v/</tt>
| '''ז''' <tt>/z/</tt>
| '''זש''' <tt>/ʒ/</tt>
|  
|
|-
|-
! Voiceless Fricative
! rowspan="2" | Affricates
| '''Ff''' /f/ || '''Ss''' /s/ || '''Cc''' /ʃ/ || || '''Xx''' /x/ || '''Hh''' /h/
!<small>voiceless</small>
|
| '''צ''' <tt>/ts/</tt>
| '''טש''' <tt>//</tt>
|
|
|-
|-
! Approximant
! <small>voiced</small>
| || '''Ll''' /l/ || rowspan="2" | '''Rr''' /ɹ~r~ʀ/ || '''Yy''' /j/ || '''Ww''' /w/ ||  
|
| '''דז''' <tt>/dz/</tt>
| '''דזש''' <tt>//</tt>
|
|
|-
|-
! Trill
! colspan="2" | Approximants
| '''ẞß''' /ʙ/ || || || ||
|
| '''ל''' <tt>/l/</tt>
| '''י''' <tt>/j/</tt>
| '''ר''' <tt>/ʁ/</tt>
|
|}
|}


{| class="wikitable"
 
|+ Vowels
{| {{Table/bluetable}} style="margin-left:10px; float: right;"
|+   '''Vowel phonemes in Weddish'''
!
! style="text-align: center;"|Front
! style="text-align: center;"|Central
! style="text-align: center;"|Back
|-align=center class=small
|-align=center
! High
| '''י''' <tt>/i/~/ɪ/</tt>
|
|
! Front || Mid || Back
| '''ו''' <tt>/u/~/ʊ/</tt>
|-
! High
| '''Ȧȧ''' /i~ɪ / || || '''Ăă''' /ʊ~u/
|-
|-
! Mid
! Mid
| '''Áá''' /ɛ~e/ || '''Ää''' /ə/ || '''Åå''' /ɔ~o/
| '''ע''' <tt>/e/~/ɛ/</tt>
| <tt>/ə/</tt>
| '''אָ''' <tt>/o/~/ɔ/</tt>
|-
|-
! Low
! Low
| || '''Àà''' /a/ ||
|
| '''אַ''' <tt>/ɐ/~/ä/</tt>
|
|}
 
{| {{Table/bluetable}} style="float:right;"
|+ '''Diphthongs in Weddish'''
!
! +y
! +w
|-
! a
| '''ײַ''' = ay
| '''אַו''' = aw
|-
! e
| '''ײ''' = ey
|
|-
! o
| '''ױ''' = oy
| '''אָו''' = ow
|}
Voices is contrasted in both plosives and fricatives, like Yiddish and English<ref>http://wals.info/chapter/4</ref>.  Vowel nasalization and rounding are not phonemic<ref>http://wals.info/chapter/11</ref>
 
<br clear="both" />
 
There are several issues in the pronunciation of individual sounds.  The rhotic of Weddish is either alveolar or uvular<ref>As in Hebrew, uvular may be seen as the most prestigious form: http://wals.info/chapter/6</ref> and may be anything from a flap, to a trill, to an actual approximant.  No R-colors vowels are permitted.  Words that begin with a vowel are separated from a prior open syllable by a glottal stop.  The velar nasal only occurs when an "n" is assimilated in place of articular before or after an "x", "k", or "g", in a syllable coda<ref>http://wals.info/chapter/9</ref>.  '''ng''' is pronounced <tt>/ŋg/</tt>, not just <tt>/ŋ/</tt>.  '''L''' is typically dark (aka "velarized") except before '''i'''.  '''Ayen''' is always romanized '''e''', but signifies the schwa in unaccented syllables.
 
In the dialect of the Americas, central vowels retain a color of their original/short form.  Elsewhere, they are all central, except <tt>/a/</tt> before glottals and <tt>/ɪ/</tt> before labials.  Another dialect difference is that '''c''' and '''dž''' are pronounced <tt>/θ/</tt> and <tt>/ð/</tt><ref>http://wals.info/chapter/19</ref>.  However, the rhotic is still not retroflex!
 
=== Orthography ===
Weddish written in the Hebrew alphabet, after the standard of YIVO Yiddish.  There is a one-to-one correspondence between grapheme and phoneme, except for three digraphs and one trigraph.  Weddish also has its own Romanization scheme, largely Slavic in appearance.  In it, <tt>/ʃ/</tt> is written '''š''', <tt>/ʒ/</tt> is written '''ž''', <tt>/j/</tt> is written '''y''', <tt>/ts/</tt> is written '''c''', <tt>/tʃ/</tt> is written '''č''', <tt>/dʒ/</tt> is written '''dž''', and <tt>/ʁ/</tt> is written '''r'''.
 
If the syllable after a diphthong begins with a vowel, the off-glide of the diphthong is doubled as the onset of that next syllable, '''''without being written again'''''.  Thus '''zeyer''' is pronounced <tt>/zey.yer/</tt>.
 
As in Hebrew, five letters have "final" forms, when they occur at the end of a word.  These forms do not affect pronunciation at all. 
{| {{Table/bluetable}}
! Initial/Medial
| '''מ'''
| '''נ'''
| '''פֿ'''
| '''צ'''
| '''כ'''
|-
! Final
| '''ם'''
| '''ן'''
| '''ף'''
| '''ץ'''
| '''ך'''
|}
 
Alphabetical order is '''alef''', '''alef pasex''', '''alef komac''', '''beys''', ('''veys''',) '''giml''', '''dalet''', '''dalet zayen''', '''dalet zayen šin''', '''hey''', '''vov''', '''gvovayin''', '''šurek''', '''zayen''', '''zayen šin''', ('''xes''',) '''tes''', '''tes šin''', '''yud''', '''yud xirik''', '''gyudayin''', '''gyudayin pasex''', '''vov yud''', '''xof''', ('''xof dageš''',) '''lamed''', '''mem''', '''nun''', '''samex''', '''ayen''', '''pey''', '''fey''', '''cadek''', '''kuf''', '''reyš''', ('''sin''',)  '''šin''' (, '''tav''', '''sav''').
 
When necessary to avoid confusion, <tt>/u/</tt> can be precisely specified with a '''וּ''', called a '''šurek'''.  <tt>/i/</tt> can be invoked as '''יִ''', that is a '''yud xirek'''. 
==== Others ====
Yiddish has many loanwords from Hebrew and Aramaic which are written using the Hebrew abjad in the Semitic way.  Weddish, however, writes these words out according to its own orthographic conventions.  However, in traditional setting it is sometimes necessary to use the ancient letters.
{| {{Table/bluetable}}
! Lošn Koydeš Letter
| '''בֿ'''
| '''ח'''
| '''כּ'''
| '''שׂ'''
| '''ת'''
| '''תֿ'''
|-
! Equivalent
| '''װ'''
| '''כ'''
| '''ק'''
| '''ס'''
| '''ט'''
| '''ס'''
|}
 
There is also a highly ornate style of writing Weddish, called '''xtiv qoydeš''' (holy writing, abbr. x"q) which featuring Hebrew roots spelling according to the ancient tradition and vowels written as diacritical marks ("points") around the consonants.  In this style, '''v''' is usually written as '''ו''' and '''y''' as '''י'''.  The other vowels are as follows, with the '''א''' written only for necessity's sake:
{| {{Table/bluetable}}
! Standard
! X"Q
! Roman.
|-
| '''אַ'''
| '''אַ'''
| '''a'''
|-
| '''אָ'''
| '''אָ'''
| '''o'''
|-
| '''ע'''
| '''אֶ'''
| '''e'''
|-
| '''י'''
| '''אִ'''
| '''i'''
|-
| '''ו'''
| '''אֻ'''
| '''u'''
|-
| '''ײ'''
| '''אֵ'''
| '''ey'''
|-
| '''ײַ'''
| '''אֱ'''
| '''ay'''
|-
| '''ױ'''
| '''אֹ'''
| '''oy'''
|-
| '''אָו'''
| '''אֳ'''
| '''ow'''
|-
| '''אַו'''
| '''אֲ'''
| '''aw'''
|-
| ø
| '''אְ'''
| <tt>/ə/</tt> or syllabic
|}
== Phonotactics ==
Weddish phonotactics are inherited from Yiddish, which is among the more permissive in the world<ref>http://wals.info/chapter/12</ref>.  While they do not rise to the level of Georgian or Salish, they are nevertheless daunting for new learners.  Gemination only becomes phonemic across word boundaries.  Consonant clusters are spontaneously broken up across syllables in order to make codas less complicated and, if necessary, onsets more so.
 
=== Syllabic Consonants ===
Liquids and fricatives may all be said syllabically.  Apart from the inseparable prepositions '''l''' and '''m''', and the proclitic conjunction '''v''', syllabic consonants all occur at the end of a word.  In an unstressed syllable, syllabic sonorants and syllables with a reduced vowel are indistinguishable.  In stressed syllables, no vowel is written, the onset and coda are optional or may consist of a single stop.
 
=== Onsets ===
{| class="bluetable lightbluebg  mw-collapsible mw-collapsed" style="text-align:center; vertical-align:middle; width:560px;"
! colspan="25" | Onset Consonant Clusters
|-
|
! b || g || d || dz || dž || h || v || z || ž || t || č || y || x || l || m || n || s || p || f || c || k || r || š
|-
! b
| || '''bg''' || '''bd''' || || || || || || || || || '''by''' || || '''bl''' || || || || || || || || '''br''' ||
|-
! g
|| || || || || || || '''gv''' || '''gz''' || || || || '''gy''' || || '''gl''' || || '''gn''' || || || || || || '''gr''' ||
|-
! d
|| || || || || || || '''dv''' || '''dz''' || || || || '''dy''' || || '''dl''' || || '''dn''' || || || || || || '''dr''' ||
|-
! dz
|| || || || || || || '''dzv''' || || || || || '''dzy''' || || || || || || || || || || ||
|-
! dž
| || || || || || || || || || || || '''džy''' || || || || || || || || || || ||
|-
! h
| || || || || || || || || || || || '''hy''' || || || || || || || || || || ||
|-
! v
| || || || || || || || || || || || '''vy''' || || '''vl''' || || || || || || || || '''vr''' ||
|-
! z
| '''zb''' || '''zg''' || || || || || '''zv''' || || || || || '''zy''' || || '''zl''' || '''zm''' || '''zn''' || || || || || || '''zr''' ||
|-
! ž
| '''žb''' || '''žg''' || || || || || '''žv''' || || || || || '''žy''' || || '''žl''' || '''žm''' || || || || || || ||  ||
|-
! t
| || || || || || || '''tv''' || || || || || '''ty''' || '''tx''' || '''tl''' || '''tm''' || '''tn''' || '''c''' || || '''tf''' || || '''tk''' || '''tr''' || '''č'''
|-
! č
| || || || || || || '''čv''' || || || || || '''čy''' || || || || || || || || || || ||
|-
! y
| || || || || || || || || || || ||  || || || || || || || || || || ||
|-
! x
| || || || || || || '''xv''' || || || || || '''xy''' || || '''xl''' || '''xm''' || '''xn''' || '''xs''' || || || '''xc''' || '''xk''' || '''xr''' || '''xš'''
|-
! l
| || || || || || || || || || || ||  || || || || || || || || || || ||
|-
! m
| || || || || || || || || || || || '''my''' || || '''ml''' || || || || || || || || '''mr''' ||
|-
! n
| || || || || || || || || || || || '''ny''' || || || || || || || || || || ||
|-
! s
| || || '''sd''' || || || || '''sv''' || || || '''st''' || '''sč''' || '''sy''' || '''sx''' || '''sl''' || '''sm''' || '''sn''' || || '''sp''' || '''sf''' || || '''sk''' || '''sr''' ||
|-
! p
| || || || || || || '''pv''' || || || '''pt''' || || '''py''' || '''px''' || '''pl''' || || '''pn''' || '''ps''' || || '''pf''' || || '''pk''' || '''pr''' || '''pš'''
|-
! f
| || || || || || || || || || || || '''fy''' || || '''fl''' || || || || || || || || '''fr''' ||
|-
! c
| || || '''cd''' || || || || '''cv''' || || || || || '''cy''' || || '''cl''' || || '''cn''' || || || || || || '''cr''' ||
|-
! k
| || || '''kd''' || || || || '''kv''' || || || '''kt''' || || '''ky''' || '''kx''' || '''kl''' || || '''kn''' || '''ks''' || || || || || '''kr''' ||
|-
! r
| || || || || || || || || || || ||  || || || || || || || || || || ||
|-
! š
| || || || || || || '''šv''' || || || '''št''' || '''šč''' || '''šy''' || '''šx''' || '''šl''' || '''šm''' || '''šn''' || || '''šp''' || '''šf''' || || '''šk''' || '''šr''' ||
|}
There are three-consonant clusters allowed that begin with '''s''' or '''š''' plus a voiceless stop plus a liquid: '''spl''', '''spr''', '''str''', '''skr''', '''skl''', '''špl''', '''špr''', '''štr''', '''škl''', and '''škr''' but not '''stl''' or '''štl'''.
 
American's should take care with '''dr''', '''tr''', '''štr''', and '''str''' not to "africatize" the cluster.
 
=== Codas ===
Final '''t''''s and '''c''''s devoice any other code consonants.  In writing, it may look like there are therefore combinations not possible on the chart below, but they are pronounced devoiced.
 
{| class="bluetable lightbluebg  mw-collapsible mw-collapsed" style="text-align:center; vertical-align:middle; width:560px;"
! colspan="25" | Coda Consonant Clusters
|-
!
! b || g || d || dz || dž || h || v || z || ž || t || č || y || x || l || m || n || s || p || f || c || k || r || š
|-
! b
| || || '''bd''' || '''bdz''' || || || || '''bz''' || '''bž''' || || || || || || || || || || || || ||  ||
|-
! g
| || || '''gd''' || '''gdz''' || || || || '''gz''' || '''gž''' || || || || || || || || || || || || ||  ||
|-
! d
| || || || || || || || '''dz''' || '''dž''' || || || || || || || || || || || || ||  ||
|-
! dz
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! dž
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! h
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! v
| || || || || || || || '''vz''' || '''vž''' || || || || || || || || || || || || ||  ||
|-
! z
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! ž
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! t
| || || || || || || || || || || || || '''tx''' || ||  ||  || '''c''' || || || || || || '''č'''
|-
! č
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! y
| || || || || || || || || || || || || || || || || || || || || ||  ||
|-
! x
| || || || || || || || || || || || '''xt''' || || || || || '''xs''' || '''xp''' || || || '''xk''' || || '''xš'''
|-
! l
| '''lb''' || '''lg''' || '''ld''' || '''ldz''' || '''ldž''' || || '''lv''' || '''lz''' || '''lž''' || '''lt''' || '''lč''' ||  || '''lx''' || || '''lm''' || '''ln''' || '''ls''' || '''lp''' || '''lf''' || '''lc''' || '''lk''' || || '''lš'''
|-
! m
| '''mb''' || || '''md''' || '''mdz''' || '''mdž''' || || || ='''mdz''' || ='''mdž''' || || || || || || || || ='''mps''' || '''mp''' || ='''mpf''' || || || || ='''mpš'''
|-
! n
| || '''ng''' || '''nd''' || '''ndz''' || '''ndž''' || || || ='''ndz''' || ='''ndž''' || '''nt''' || '''nč''' || || ='''nkx''' || || || || ='''nc''' || || || '''nc''' || '''nk''' || || ='''nč'''
|-
! s
|  ||  ||  ||  || ||  ||  ||  ||  || '''st''' || '''sč''' ||  ||  ||  ||  ||  ||  || '''sp''' ||  || '''sc''' || '''sk''' || ||
|-
! p
| ||  ||  ||  ||  ||  || ||  ||  || '''pt''' || '''pč''' ||  ||  ||  ||  ||  || '''ps''' ||  || '''pf''' || '''pc''' || '''pk''' || || '''pš'''
|-
! f
| ||  ||  ||  ||  ||  || ||  ||  || '''ft''' || '''fč''' ||  ||  ||  ||  ||  || '''fs''' || '''fp''' ||  || '''fc''' || || '''fk''' || '''fš'''
|-
! c
|-
! k
|  ||  ||  ||  ||  ||  ||  ||  ||  || '''kt''' || '''kč''' ||  || '''kx''' ||  ||  ||  || '''ks''' ||  || '''kf''' || '''kc''' ||  ||  || '''kš'''
|-
! r
| '''rb''' || '''rg''' || '''rd''' || '''rdz''' || '''rdž''' || || '''rv''' || '''rz''' || '''rž''' || '''rt''' || '''rč''' ||  || '''rx''' || || '''rm''' || '''rn''' || '''rs''' || '''rp''' || '''rf''' || '''rc''' || '''rk''' || || '''rš'''
|-
! š
|  ||  ||  ||  ||  || ||  ||  ||  || '''št''' ||  ||  ||  ||  ||  ||  ||  '''šs''' || '''šp''' || || || '''šk''' ||  ||  
 
|}
|}
There are also ''liquids plus stop plus homorganic, alveolar fricative'': '''lps, lbz, lks, lgz, rps, rbz, rks, rgz.
== Suprasegmentals ==
Stress is predicable, if one knows the root of a word.  The first syllable of the root receives primary stress, with secondary stresses proceeding out like ripples on a pond to every other syllable, forwards and backwards.  (The major exception is the dual, which moves the stress of a word with an odd number of syllables.)  The default rhythm of Weddish is ''trochaic'': '''stressed'''-unstressed.  Neither vowel length nor stress is phonemic. Long vowels indicate stress. If the word is long, one of the first three syllables must have primary stress.  Prefixes and suffixes all have an underlying vowel which is expressed or repressed in order to maintain the rhythm pattern.  Two syllables with reduced vowels may not follow each other.  Polar and interrogative questions are both marked by a rising tone at the end of the utterance.
== Syntax ==
{{ClassMeter
|Name = Weddish
|NativeName = וועדיש
|Type = fusional
|Alignment = Ergative-Absolutive
|Head = Second
|Tonal = no
|Declined = no
|Conjugated = yes
|Genders = 2
|NCase = no
|NNumber = yes
|NDefiniteness = no
|NGender = yes
|VVoice = yes
|VMood = no
|VPerson = yes
|VNumber = yes
|VTense = no
|VAspect = yes
|Phonology= 100
|NounCases= 50
|NounDef= 80
|NounNumbers= 90
|NounGender= 50
|VerbPerson= 50
|VerbNumber= 50
|VerbAspect= 50
|VerbTense= 100
|VerbMood= 10
|VerbVoice= 30
|AdjCase= 60
|AdjNumber= 30
|AdjDef= 80
|AdjGen= 90
|AdjComparative= 10
|AdjSuperlative= 10
|AdjGen= 10
|Supine= 20
|Gerund=0
|Participle= 20
|Infinitive= 20
|Modality= 20
|Words= 100
|adjective =  final
|adposition = initial
|adverb= final
|article= initial
|relativeclause = initial
|nounclause = mixed
|order = SVO
}}
Weddish aims to appeal to English speakers.  While the verbal-system is somewhat new, the noun-system should be easy.  Nouns are not inflected, but pronouns do have unique forms that show what part of speech they can be.  Like German, however, articles do inflect.  Like Hebrew, there are two noun genders.  The masculine is most often animate, while the feminine is not.  Adjectives do not inflect unless substantive. 
[[File:MorphSyntAlign2.png|left]]Purposefully chosen to stimulate thinking, Weddish has an ergative-absolutive morphosyntactic alignment.  Most languages in the world consider the actor of transitive verb and the subject of an intransitive verb to be equivalent.  The object of a transitive verb is special in these systems.  It can be promoted to the subject via the passive voice.  Normally, it must come after the verb.  Weddish treats the ''object'' of a transitive verb and the subject of a transitive verb the same, called the "absolutive case".  Actors of transitive verbs are specially in Weddish, called the "ergative case".
<br />


=== Phonotactics ===
It would be tempting to classify Weddish as VSO (subject-veb-object) like Hebrew, but that's not quite right.  It is, in fact, a V2 language, which means the verb always wants to come second.  (Discourse particles and few other things do not count towards calculating where "second" is, and entire phrases are taken as a whole when counting this place.)  Because core cases are marked almost solely by word order, the actor of a transitive verb (the ergative case) must come before the verb, i.e. first in the sentence.
Initial
* Single phonemes except /ŋ/


Final
The V2 Principle is carried throughout Weddish, to the point where it might be labeled a "head second" language.  This is not a recognized typology, since languages are either head-initial, head-final, or mixed.  In compound nouns, the head is second.  For auxiliary verbs, the head verb is second.  For nouns with attributive modifiers, the article comes first, but adjectives come after the noun, making it head-second as well.


=== Number ===
English, Hebrew, Yiddish and many other languages have two numbers: singular and plural.  Weddish (like Arabic) has three: singular, dual, and plural.  Obviously, the dual is for two of something, and the plural therefore means three or more.  However, in regards to persons, the dual is used on married people, even if only one of them is being spoken about.  Exceptions can be made in every case except the ergative, which is reserved for the spouses to use on each other.  This distinction does not apply in the third person for people not present.
Weddish also distinguishes whether actions were done as individuals or all as a group.  It also possible to add "associates" of a noun to it. 
=== Copula ===
The verb "to be", "to become", and "to have" are all copulas in Weddish.  That means they all use only the absolutive case, never the ergative.  However, "to be" and "to have" are more like "to equal" and "to exist".  "I have shoes" is literally "Shoes exist to me".  This can be easier for Far East Asians to learn than Westerners. 
<br clear="both" />
== Morphology ==
== Morphology ==
As in Semitic language, words are based on tri-consonantal roots.   That is vowels, are inserted within groups of three letters to form nouns, verbs, and adverbs. Four letters are called "weak", and they turn into vowels in predictable ways.   '''Y''' also equals /i/, '''W''' also equals /u/, '''A''' also equals /a/, and '''H''' also equals /e/.
=== Case ===
{| {{Table/bluetable}} style="float:right"
! colspan="4" | Genitive
|-
| colspan="4" style="text-align:center;" | '''פֿון'''/'''fun'''
|-
! Dative || Ablative || Partitive || Equative
|-
| style="text-align:center;" | '''ל־'''/'''l-'''
| style="text-align:center;" | '''ב־'''/'''b-'''
| style="text-align:center;" | '''מ־'''/'''m-'''
| style="text-align:center;" | '''ק־'''/'''k-'''
|}
 
By default, all nouns are in the ''absolutive'' case.  But, if they are placed '''''before''''' the verb, then they are said to be in the ergative case, though their morphology is unchanged. The only exception is the masculine singular/dual definite article changes in the ergative case.  Linguists call these two case the "core cases" of a language, since they are fundamental.  There are five additional cases --- called "non-core" cases --- in Weddish that are also very important.  Unlike many languages that have ''suffixing'' case marking, Weddish has ''prefixing''. This is because they are derived from Hebrew Inseparable Prepositions (IP's).  Phrases in the non-core cases either relate to the verb (and are hence, adverbial), or are in a noun phrase.  In relation to nouns, the core cases are all seen as greater specificity ''within'' the genitive case.
 
Non-core cases all fall under the umbrella term "genitive".  The generic genitive is not a case ''per se'', but a preposition (meaning, a separable preposition).  An expression like '''dos line fun gelt'''/''the love of money'' is even more ambiguous in Weddish than in English.  It may mean ''the love belonging to money'', ''the love in/by money'', ''the love from/composed of money'', or ''the love as/according to money''.  After a genitive phrase has been established or is implicitly understood, the phrase may be into a compound noun using the "head-second" structure.
{| {{Table/bluetable}}
|+ '''Case, Articles, and IP's'''
!
! colspan="3" | Definite
! Indefinite
! rowspan="2" | Anarthrous
|-
!
! <small>m.sg/dl</small>
! <small>f.sg/dl</small>
! <small>pl</small>
! <small>sg/dl</small>
|-
! Erg.
| '''{{C|der}}'''
| rowspan="2" | '''{{C|dos}}'''
| rowspan="2" | '''{{C|di}}'''
| rowspan="2" | '''{{C|a}}([[Contionary:an|n]])'''
| rowspan="2" | ø
|-
! Abs.
| '''{{C|dem}}'''
|-
! Dat.
| '''{{C|lem}}'''
| '''{{C|ler}}'''
| '''{{C|li}}'''
| '''{{C|lawn}}'''
| '''l-'''
|-
! Abl.
| '''{{C|bem}}'''
| '''{{C|bos}}'''
| '''{{C|bi}}'''
| '''{{C|bawm}}'''
| '''b-'''
|-
! Part.
| '''{{C|mem}}'''
| '''{{C|mos}}'''
| '''{{C|mi}}'''
| '''{{C|mawm}}'''
| '''m-/{{C|min}}'''
|-
! Eq.
| '''{{C|kem}}'''
| '''{{C|kos}}'''
| '''{{C|ki}}'''
| '''{{C|kawm}}'''
| '''k)-'''
|}
=== Number ===
Weddish verbs conjugate for three numbers (singular, dual, and plural), but nouns inflect '''''eleven''''' different ways! However, these myriad ways can be easily understood as the optional adding of "associates" to a noun, and distinguishing between masses of individuals and collectives (one forest vs. many trees).  The following table is color-coded to show verb conjugation in the singular (light background), dual (purple), and plural (brown).
 
{| {{Table/bluetable}}
! || Singular || Dual || Plural
|-
! Distributive
| '''פֿרײַנד''' <br /> '''fraynd''' <br /> ''a friend''
| style="background: MediumPurple;" | '''פֿרינדײַים''' <br /> '''frindayim''' <br /> ''two friends each''
| style="background: BurlyWood;" | '''פֿרײַנדין''' <br /> '''frayndin''' <br /> ''friends each''
|-
! Collective
|
| style="background: MediumPurple;" | '''געפֿרינדײַים''' <br /> '''gefrindayim''' <br /> ''a couple of friends''
| '''געפֿרײַנדין''' <br /> '''gefrayndin''' <br /> ''a group of friends''
|-
! Distributive <br /> Associative
| style="background: BurlyWood;" | '''פֿרײַנדז''' <br /> '''frayndz''' <br /> ''a friend and associates each''
| style="background: BurlyWood;" | '''פֿרינדײַיםז''' <br /> '''frindayimz''' <br /> ''two friends and associates each''
| style="background: BurlyWood;" | '''פֿרײַנדינז''' <br /> '''frayndinz''' <br /> ''friends and associates each''
|-
! Collective <br /> Associative
| '''געפֿרײַנז''' <br /> '''gefrayndz''' <br /> ''a group of a friend and associates''
| style="background: MediumPurple;" | '''געפרינדײַימז''' <br /> '''gefrindayimz''' <br /> ''a group of two friends and associates''
| '''געפֿרײַנדינז''' <br /> '''gefrayndinz''' <br /> ''a group of friends and associates''
|}
 
The dual ending is unique, in that is shifts the accent pattern of the root to itself.  It may be written '''-áyim''' to indicate that shift.  This shift triggers vowel reduction of of the previous syllable, if it is a diphthong (cutting it down to its first vowel).
 
Forms lacking the collective plural endings are automatically distributive. 
=== Pronouns ===
==== Independent Personal ====
Absolutive independent personal pronouns are most commonly used with ø-copula clauses to show predication.  Such sentences are distinguished from those with the "to be" verb, which show absolute identity, as opposed to mere attribution.  '''Gu Yidiš'''/''We are Jewish'' vs. '''Big Džonzez'''/''We are (the) Jones's''.
 
{| {{Table/bluetable}}
! colspan="2" |
! Ergative
! Absolutive
! Dative
! Ablative
! Partitive
! Equative
|-
! rowspan="3" | 1
! <small>sg</small>
| '''{{C|ix}}'''
| '''{{C|mix}}'''
| '''{{C|mir}}'''
| '''{{C|bix}}'''
| '''{{C|mix}}'''
| '''{{C|kix}}'''
|-
! <small>dl</small>
| '''{{C|nanxu}}'''
| '''{{C|gu}}'''
| '''{{C|gir}}'''
| '''{{C|bug}}'''
| '''{{C|ming}}'''
| '''{{C|kowg}}'''
|-
! <small>pl</small>
| '''{{C|undz}}'''
| '''{{C|mir}}'''
| '''{{C|undz}}'''
| '''{{C|bu}}'''
| '''{{C|minu}}'''
| '''{{C|ku}}'''
|-
! rowspan="3" | 2
! <small>sg</small>
| '''{{C|du}}'''
| '''{{C|dix}}'''
| '''{{C|dir}}'''
| '''{{C|bed}}'''
| '''{{C|mind}}'''
| '''{{C|ked}}'''
|-
! <small>dl</small>
| '''{{C|stu}}'''
| '''{{C|stuk}}'''
| '''{{C|stire}}'''
| '''{{C|bist}}'''
| '''{{C|minst}}'''
| '''{{C|kist}}'''
|-
! <small>pl</small>
| '''{{C|ir}}'''
| '''{{C|ayx}}'''
| '''{{C|ayx}}'''
| '''{{C|bikm}}'''
| '''{{C|mint}}'''
| '''{{C|kat}}'''
|-
! rowspan="4" | 3
! <small>m.sg</small>
| '''{{C|er}}'''
| '''{{C|'im}}'''
| '''{{C|inen}}'''
| '''{{C|bo}}'''
| '''{{C|mino}}'''
| '''{{C|ko}}'''
|-
! <small>f.sg</small>
| '''{{C|zi}}'''
| '''{{C|es}}'''
| '''{{C|aya}}'''
| '''{{C|ba}}'''
| '''{{C|mina}}'''
| '''{{C|ka}}'''
|-
! <small>dl</small>
| '''{{C|bera}}'''
| '''{{C|hura}}'''
| '''{{C|hav}}'''
| '''{{C|bav}}'''
| '''{{C|minav}}'''
| '''{{C|kav}}''
|-
! <small>pl</small>
| '''{{C|zey}}'''
| '''{{C|cey}}'''
| '''{{C|čire}}'''
| '''{{C|bouč}}'''
| '''{{C|minč}}'''
| '''{{C|kač}}
|}
==== Interrogative ====
{| {{Table/bluetable}}
|
! Erg. || Abs. || Gen. || Dat. || Abl. || Part. || Eq.
|-
! Persons.
| {{C|mi}} || {{C|ver}} || {{C|vermenc}} || {{C|vermen}} || {{C|bimi}} || {{C|mimi}} || {{C|komi}}
|-
! Impers.
| {{C|ma}} || {{C|vos}} || {{C|fun vos}} || {{C|vu}} || {{C|vi}} || {{C|vat}} || {{C|ven}}
|}
The interrogative pronouns do not inflect for person, number, or gender.  Linguists would say they are animate and inanimate, though Weddish grammar calls them "personal" and "impersonal".  They are identical to the relative pronoun (just as in English) and must match their antecedent in animacy, but not in case.  Instead (just as in English) they indicate their new role in the relative clause.
==== Affixes ====
Like Hebrew, Weddish uses enclitic forms of pronouns to indicate several things.  On verbs, pronominal suffixes mark the absolutive argument of the clause.  On nouns, they mark a genitive relationship.  Pronominal prefixes are used exclusively on transitive verbs to mark the ergative argument, and are obligatory.  Weddish is not pro-drop, and an affix on both ends is required on transitive verbs.  Remember, there are no ambi-transitive verbs in Weddish.  Use of the independent personal pronouns when the person has been specified on either end of the verb is considered emphatic.
 
{| {{Table/bluetable}}
! Person || # || Suffix || Prefix
|-
! rowspan="3" | <big>1</big>
! sg
| '''-(n)i'''
| '''ni-'''
|-
! dl
| '''-(u)g'''
| '''gu-'''
|-
! pl
| '''-(n)u'''
| '''na-'''
|-
! rowspan="3" | <big>2</big>
! sg
| '''-(e)d'''
| '''de-'''
|-
! dl
| '''-(e)st'''
| '''sti-'''
|-
! pl
| '''-ti'''
| '''ta-'''
|-
! rowspan="4" | <big>3</big>
! m.sg
| '''-o'''
| '''ro-'''
|-
! f.sg
| '''-a'''
| '''ya-'''
|-
! dl
| '''-av*'''
| ø
|-
! pl
| '''-(ay/i)č'''
| '''čay-'''
|}
 
=== Verbs ===
Weddish verbs do not conjugate for tense, only aspect.


{| class="wikitable"
==== Aspect Ablaut ====
|+ Common Shapes on JST
{| {{Table/bluetable}}
! Perfective || Imperfective
|-
|-
! Form || Long || Name || Meaning
| -ei- || -i-
|-
|-
| J̀S̀T̈T̆ || jasattu || 1cs-Perf. || I justified
| -au- || -ai-
|-
|-
| J̀S̀T̈Ǹ || jasatna || 1cp-Perf. || We justified
| -ou- || -u-
|-
|-
| J̀S̀T̈T̀ || jasatta || 2ms-Perf || You (m) justified
| -e- || -a-
| J̀S̀
|-
|-
| J̀S̀T̈Ṫ || jasatti || 2fs-Perf. || You (f) justified
| -o- || -oi-
|}
==== Voices ====
# Causative: š/že-
# Reflexive: hit/hid-
# Antipassive: V-u
# Mixed
==== Non-finite ====
# participle -ing
# infinitive absolute: u-V
# infinitive construct: ge-
==== Incorporation ====
On the Mithun scale<ref>Mithun, Marianne. 1984. The Evolution of Noun Incorporation. ''Language'', Vol. 60, No. 4. pp. 847-894.</ref>, Weddish does type-I and type-II noun incorporation.  This means 1) I picked berries -> I berry-picked, and 2) I washed his face -> I face-washed him.
 
=== Derivation ===
 
==== Compounding ====
When the relationships between nouns is genitive, and it has already been stated or can easily be implied, compound nouns may be formed.  For example, ''a field for football/soccer'' may become '''fusbolfeld'''.  (Note the loss of abstraction suffixes.)  Suppose it was an Australian rules football field.  Would could make '''fusbolfeldeoystralie'''. Lastly, If one wanted to add that it is '''mgroz'''/''composed of grass'', this could become '''פֿוסבאָלפֿעלדעאויסטראליעגראָז'''/'''fusbolfeldeoystraliegroz'''.  Words with greater than four parts are deemed colloquial.  Word order is "head second",  with the first specifier coming at the very front.
 
==== Abstract Nouns ====
All nouns in Weddish are inherently concrete.  Two levels of abstraction are possible through suffixation.  The first signifies the ''practice'' of one or more persons.  The second signifies the ''understanding'' of the practice.  Both are available in both genders, with the masculine form referring to a person (of either gender), however, the "practice"-form occurs much more often in the masculine and the "understanding"-form occurs much more often in the feminine.  Remember, this are no truly abstract nouns in Weddish, for they do not exist.
 
{| {{Table/bluetable}}
! Suffix
! "Tennis"
| gloss
! "Peace"
| gloss
|-
|-
| J̀S̀T̀ || jasata || 3ms-Perf. || He justified
! ø
| '''a tennis'''
| ''a game of tennis''
| '''a šolem'''
| ''a season of peace''
|-
|-
| J̀S̀T̆ || jasatu || 3fs-Perf. || She justificed
! rowspan="2" | -ay
| '''dos tenisayo'''
| ''his tennis game''
| '''šolemaya'''
| ''her (practice of) peace''
|-
| '''dem tenisayt'''
| ''the tennis player''
| '''dem šolemayt'''
| ''the peacemaker''
|-
! rowspan="2" | -šaft
| '''dos tenisšafte'''
| ''the game of tennis'' / "tennisology"
| '''šolemšafte''
| ''peace know-how''
|-
| '''a tenisšaft'''
| ''a tennisologist''
| '''a šolemšaft'''
| ''a student of peace''
|}
|}
== Discourse ==
=== Formality ===
=== Particles ===
<references />
[[Category:Languages]][[Category:Conlangs]][[Category:Germanic languages]][[Category:Semitic languages]][[Category:A posteriori]]

Revision as of 04:44, 30 January 2014

Weddish
Vediš
Pronunciation[/ˈve(ː).dɪʃ/]
Created byRobert Murphy
Date2013
SettingJewish intermarriage / Systematic Theology
Native speakers0.01 (2014)
Indo-European
  • Germanic
    • West Germanic
      • High German
        • Yiddish
          • Basque/Hebrew
            • Weddish
SourcesYiddish

Weddish (Weddish: װעדיש, X"Q: וֶדִש, Romanization: Vediš) is a constructed, a posteriori, naturalistic auxlang, made from Yiddish with heavy influences from Hebrew, English, German, and some Basque. It has ergative-absolutive morphosyntactic alignment and a pervasive yet symbolic use of the dual. It is meant to promote the institution of marriage, foster better communication between persons, and improve the tenor of Systematic Theological discussions. It is well-suited as an auxlang for Jewish intermarriage.

The language was created in 2013 by Robert Murphy as part of an assignment at Covenant Theological Seminary for Professor Jerram Barrs.

Philosophy

First and foremost, there is the Creator-Creature distinction. That means, God is wholly other than the Universe. Second, human beings are made in the image of God. This means that we are persons -- like God is -- and our agency is our single-most important feature. Third, we reflect the image of God as females and males. The marriage bond is God-created and a fundamental part of our identity in this life. Hence it is, that we may divide the world into: actors, non-actors, and actions. Stated grammatically, this list becomes: ergative nouns, absolutive nouns, and verbs. Furthermore, ergative nouns may be divided up into married and non-married actors, which we will mark with the dual or not.

We have said that ontologically speaking, there are ultimately only two (God and not-God). However, the 800lb. gorilla in the room -- philosophically speaking -- is abstraction. Since before Pythagoras, abstract nouns (such as numbers, "goodness", etc.) had been held by the Greeks to be ontic. Westerns betray their affinity to Greek ideal, classifying humanity as homo sapiens - "thinking man". We seek to disenthrall ourselves from this metaphysic and banish abstract nouns from our language. Numerals are adjectives. Infinity is that which we cannot see the end of. Universals are the same as aggregates ("all" is the same as "sum" ... but not "some"!).

The Language Creation Society (as fine institution) waves a flag of the Tower of Babylon. Unlike the hubristic man of the initial chapters In the Land of Invented Languages (Arika Okrent's chronicle), we do not believe we can undo by human effort what God has done. Our aim is for Jews and Christians to discuss the truth better. The status of English as a lingua franca, German as a language of science, and Hebrew as a holy language suggests that an Indo-European language is still the best option for an a posteriori auxlang. Rather than compete with those vital languages, however, it seems most prudent to build upon a base off them all at once, utilizing a language that is already as eclectic as English, yet built off German and Hebrew. That language is Yiddish.

Phonology

Weddish has 25 consonantal sounds, which is typologically average [1], and common in Europe as well as the Middle East. English speakers will find it to be common, apart from the lack of /w/ and the ubiquity of /x/ (like the ch in Bach or loch). Weddish has 6 vowels, which is also average[2], as is the resulting consonant-to-vowel ration[3]. This is typologically equivalent to Yiddish and Hebrew, but far less than German or English.

Consonants of Weddish
Consonant phonemes
Labial Alveolar Post-Alveolar Dorsal Glottal
Nasals מ /m/ נ /n/ * /ŋ/
Stops voiceless פ /p/ ט /t/ ק /k/ א /ʔ/
voiced ב /b/ ד /d/ ג /g/
Fricatives voiceless פֿ /f/ ס /s/ ש /ʃ/ כ /x/ ה /h/
voiced װ /v/ ז /z/ זש /ʒ/
Affricates voiceless צ /ts/ טש /tʃ/
voiced דז /dz/ דזש /dʒ/
Approximants ל /l/ י /j/ ר /ʁ/


Vowel phonemes in Weddish
Front Central Back
High י /i/~/ɪ/ ו /u/~/ʊ/
Mid ע /e/~/ɛ/ /ə/ אָ /o/~/ɔ/
Low אַ /ɐ/~/ä/
Diphthongs in Weddish
+y +w
a ײַ = ay אַו = aw
e ײ = ey
o ױ = oy אָו = ow

Voices is contrasted in both plosives and fricatives, like Yiddish and English[4]. Vowel nasalization and rounding are not phonemic[5]


There are several issues in the pronunciation of individual sounds. The rhotic of Weddish is either alveolar or uvular[6] and may be anything from a flap, to a trill, to an actual approximant. No R-colors vowels are permitted. Words that begin with a vowel are separated from a prior open syllable by a glottal stop. The velar nasal only occurs when an "n" is assimilated in place of articular before or after an "x", "k", or "g", in a syllable coda[7]. ng is pronounced /ŋg/, not just /ŋ/. L is typically dark (aka "velarized") except before i. Ayen is always romanized e, but signifies the schwa in unaccented syllables.

In the dialect of the Americas, central vowels retain a color of their original/short form. Elsewhere, they are all central, except /a/ before glottals and /ɪ/ before labials. Another dialect difference is that c and are pronounced /θ/ and /ð/[8]. However, the rhotic is still not retroflex!

Orthography

Weddish written in the Hebrew alphabet, after the standard of YIVO Yiddish. There is a one-to-one correspondence between grapheme and phoneme, except for three digraphs and one trigraph. Weddish also has its own Romanization scheme, largely Slavic in appearance. In it, /ʃ/ is written š, /ʒ/ is written ž, /j/ is written y, /ts/ is written c, /tʃ/ is written č, /dʒ/ is written , and /ʁ/ is written r.

If the syllable after a diphthong begins with a vowel, the off-glide of the diphthong is doubled as the onset of that next syllable, without being written again. Thus zeyer is pronounced /zey.yer/.

As in Hebrew, five letters have "final" forms, when they occur at the end of a word. These forms do not affect pronunciation at all.

Initial/Medial מ נ פֿ צ כ
Final ם ן ף ץ ך

Alphabetical order is alef, alef pasex, alef komac, beys, (veys,) giml, dalet, dalet zayen, dalet zayen šin, hey, vov, gvovayin, šurek, zayen, zayen šin, (xes,) tes, tes šin, yud, yud xirik, gyudayin, gyudayin pasex, vov yud, xof, (xof dageš,) lamed, mem, nun, samex, ayen, pey, fey, cadek, kuf, reyš, (sin,) šin (, tav, sav).

When necessary to avoid confusion, /u/ can be precisely specified with a וּ, called a šurek. /i/ can be invoked as יִ, that is a yud xirek.

Others

Yiddish has many loanwords from Hebrew and Aramaic which are written using the Hebrew abjad in the Semitic way. Weddish, however, writes these words out according to its own orthographic conventions. However, in traditional setting it is sometimes necessary to use the ancient letters.

Lošn Koydeš Letter בֿ ח כּ שׂ ת תֿ
Equivalent װ כ ק ס ט ס

There is also a highly ornate style of writing Weddish, called xtiv qoydeš (holy writing, abbr. x"q) which featuring Hebrew roots spelling according to the ancient tradition and vowels written as diacritical marks ("points") around the consonants. In this style, v is usually written as ו and y as י. The other vowels are as follows, with the א written only for necessity's sake:

Standard X"Q Roman.
אַ אַ a
אָ אָ o
ע אֶ e
י אִ i
ו אֻ u
ײ אֵ ey
ײַ אֱ ay
ױ אֹ oy
אָו אֳ ow
אַו אֲ aw
ø אְ /ə/ or syllabic

Phonotactics

Weddish phonotactics are inherited from Yiddish, which is among the more permissive in the world[9]. While they do not rise to the level of Georgian or Salish, they are nevertheless daunting for new learners. Gemination only becomes phonemic across word boundaries. Consonant clusters are spontaneously broken up across syllables in order to make codas less complicated and, if necessary, onsets more so.

Syllabic Consonants

Liquids and fricatives may all be said syllabically. Apart from the inseparable prepositions l and m, and the proclitic conjunction v, syllabic consonants all occur at the end of a word. In an unstressed syllable, syllabic sonorants and syllables with a reduced vowel are indistinguishable. In stressed syllables, no vowel is written, the onset and coda are optional or may consist of a single stop.

Onsets

Onset Consonant Clusters
b g d dz h v z ž t č y x l m n s p f c k r š
b bg bd by bl br
g gv gz gy gl gn gr
d dv dz dy dl dn dr
dz dzv dzy
džy
h hy
v vy vl vr
z zb zg zv zy zl zm zn zr
ž žb žg žv žy žl žm
t tv ty tx tl tm tn c tf tk tr č
č čv čy
y
x xv xy xl xm xn xs xc xk xr
l
m my ml mr
n ny
s sd sv st sy sx sl sm sn sp sf sk sr
p pv pt py px pl pn ps pf pk pr
f fy fl fr
c cd cv cy cl cn cr
k kd kv kt ky kx kl kn ks kr
r
š šv št šč šy šx šl šm šn šp šf šk šr

There are three-consonant clusters allowed that begin with s or š plus a voiceless stop plus a liquid: spl, spr, str, skr, skl, špl, špr, štr, škl, and škr but not stl or štl.

American's should take care with dr, tr, štr, and str not to "africatize" the cluster.

Codas

Final t's and c's devoice any other code consonants. In writing, it may look like there are therefore combinations not possible on the chart below, but they are pronounced devoiced.

Coda Consonant Clusters
b g d dz h v z ž t č y x l m n s p f c k r š
b bd bdz bz
g gd gdz gz
d dz
dz
h
v vz
z
ž
t tx c č
č
y
x xt xs xp xk
l lb lg ld ldz ldž lv lz lt lx lm ln ls lp lf lc lk
m mb md mdz mdž =mdz =mdž =mps mp =mpf =mpš
n ng nd ndz ndž =ndz =ndž nt =nkx =nc nc nk =
s st sp sc sk
p pt ps pf pc pk
f ft fs fp fc fk
c
k kt kx ks kf kc
r rb rg rd rdz rdž rv rz rt rx rm rn rs rp rf rc rk
š št šs šp šk

There are also liquids plus stop plus homorganic, alveolar fricative: lps, lbz, lks, lgz, rps, rbz, rks, rgz.

Suprasegmentals

Stress is predicable, if one knows the root of a word. The first syllable of the root receives primary stress, with secondary stresses proceeding out like ripples on a pond to every other syllable, forwards and backwards. (The major exception is the dual, which moves the stress of a word with an odd number of syllables.) The default rhythm of Weddish is trochaic: stressed-unstressed. Neither vowel length nor stress is phonemic. Long vowels indicate stress. If the word is long, one of the first three syllables must have primary stress. Prefixes and suffixes all have an underlying vowel which is expressed or repressed in order to maintain the rhythm pattern. Two syllables with reduced vowels may not follow each other. Polar and interrogative questions are both marked by a rising tone at the end of the utterance.

Syntax

Weddish
וועדיש
Progress: 43%
Type
fusional
Alignment
Ergative-Absolutive
Head direction
Initial Mixed Final
Primary word order
Subject-verb-object
Tonal
No
Declensions
No
Conjugations
Yes
Genders
2
Verbs conjugate according to...
Voice Mood
Person Number
Tense Aspect

Weddish aims to appeal to English speakers. While the verbal-system is somewhat new, the noun-system should be easy. Nouns are not inflected, but pronouns do have unique forms that show what part of speech they can be. Like German, however, articles do inflect. Like Hebrew, there are two noun genders. The masculine is most often animate, while the feminine is not. Adjectives do not inflect unless substantive.

MorphSyntAlign2.png

Purposefully chosen to stimulate thinking, Weddish has an ergative-absolutive morphosyntactic alignment. Most languages in the world consider the actor of transitive verb and the subject of an intransitive verb to be equivalent. The object of a transitive verb is special in these systems. It can be promoted to the subject via the passive voice. Normally, it must come after the verb. Weddish treats the object of a transitive verb and the subject of a transitive verb the same, called the "absolutive case". Actors of transitive verbs are specially in Weddish, called the "ergative case".


It would be tempting to classify Weddish as VSO (subject-veb-object) like Hebrew, but that's not quite right. It is, in fact, a V2 language, which means the verb always wants to come second. (Discourse particles and few other things do not count towards calculating where "second" is, and entire phrases are taken as a whole when counting this place.) Because core cases are marked almost solely by word order, the actor of a transitive verb (the ergative case) must come before the verb, i.e. first in the sentence.

The V2 Principle is carried throughout Weddish, to the point where it might be labeled a "head second" language. This is not a recognized typology, since languages are either head-initial, head-final, or mixed. In compound nouns, the head is second. For auxiliary verbs, the head verb is second. For nouns with attributive modifiers, the article comes first, but adjectives come after the noun, making it head-second as well.

Number

English, Hebrew, Yiddish and many other languages have two numbers: singular and plural. Weddish (like Arabic) has three: singular, dual, and plural. Obviously, the dual is for two of something, and the plural therefore means three or more. However, in regards to persons, the dual is used on married people, even if only one of them is being spoken about. Exceptions can be made in every case except the ergative, which is reserved for the spouses to use on each other. This distinction does not apply in the third person for people not present.

Weddish also distinguishes whether actions were done as individuals or all as a group. It also possible to add "associates" of a noun to it.

Copula

The verb "to be", "to become", and "to have" are all copulas in Weddish. That means they all use only the absolutive case, never the ergative. However, "to be" and "to have" are more like "to equal" and "to exist". "I have shoes" is literally "Shoes exist to me". This can be easier for Far East Asians to learn than Westerners.


Morphology

Case

Genitive
פֿון/fun
Dative Ablative Partitive Equative
ל־/l- ב־/b- מ־/m- ק־/k-

By default, all nouns are in the absolutive case. But, if they are placed before the verb, then they are said to be in the ergative case, though their morphology is unchanged. The only exception is the masculine singular/dual definite article changes in the ergative case. Linguists call these two case the "core cases" of a language, since they are fundamental. There are five additional cases --- called "non-core" cases --- in Weddish that are also very important. Unlike many languages that have suffixing case marking, Weddish has prefixing. This is because they are derived from Hebrew Inseparable Prepositions (IP's). Phrases in the non-core cases either relate to the verb (and are hence, adverbial), or are in a noun phrase. In relation to nouns, the core cases are all seen as greater specificity within the genitive case.

Non-core cases all fall under the umbrella term "genitive". The generic genitive is not a case per se, but a preposition (meaning, a separable preposition). An expression like dos line fun gelt/the love of money is even more ambiguous in Weddish than in English. It may mean the love belonging to money, the love in/by money, the love from/composed of money, or the love as/according to money. After a genitive phrase has been established or is implicitly understood, the phrase may be into a compound noun using the "head-second" structure.

Case, Articles, and IP's
Definite Indefinite Anarthrous
m.sg/dl f.sg/dl pl sg/dl
Erg. der dos di a(n) ø
Abs. dem
Dat. lem ler li lawn l-
Abl. bem bos bi bawm b-
Part. mem mos mi mawm m-/min
Eq. kem kos ki kawm k)-

Number

Weddish verbs conjugate for three numbers (singular, dual, and plural), but nouns inflect eleven different ways! However, these myriad ways can be easily understood as the optional adding of "associates" to a noun, and distinguishing between masses of individuals and collectives (one forest vs. many trees). The following table is color-coded to show verb conjugation in the singular (light background), dual (purple), and plural (brown).

Singular Dual Plural
Distributive פֿרײַנד
fraynd
a friend
פֿרינדײַים
frindayim
two friends each
פֿרײַנדין
frayndin
friends each
Collective געפֿרינדײַים
gefrindayim
a couple of friends
געפֿרײַנדין
gefrayndin
a group of friends
Distributive
Associative
פֿרײַנדז
frayndz
a friend and associates each
פֿרינדײַיםז
frindayimz
two friends and associates each
פֿרײַנדינז
frayndinz
friends and associates each
Collective
Associative
געפֿרײַנז
gefrayndz
a group of a friend and associates
געפרינדײַימז
gefrindayimz
a group of two friends and associates
געפֿרײַנדינז
gefrayndinz
a group of friends and associates

The dual ending is unique, in that is shifts the accent pattern of the root to itself. It may be written -áyim to indicate that shift. This shift triggers vowel reduction of of the previous syllable, if it is a diphthong (cutting it down to its first vowel).

Forms lacking the collective plural endings are automatically distributive.

Pronouns

Independent Personal

Absolutive independent personal pronouns are most commonly used with ø-copula clauses to show predication. Such sentences are distinguished from those with the "to be" verb, which show absolute identity, as opposed to mere attribution. Gu Yidiš/We are Jewish vs. Big Džonzez/We are (the) Jones's.

Ergative Absolutive Dative Ablative Partitive Equative
1 sg ix mix mir bix mix kix
dl nanxu gu gir bug ming kowg
pl undz mir undz bu minu ku
2 sg du dix dir bed mind ked
dl stu stuk stire bist minst kist
pl ir ayx ayx bikm mint kat
3 m.sg er 'im inen bo mino ko
f.sg zi es aya ba mina ka
dl bera hura hav bav minav 'kav
pl zey cey čire bouč minč kač

Interrogative

Erg. Abs. Gen. Dat. Abl. Part. Eq.
Persons. mi ver vermenc vermen bimi mimi komi
Impers. ma vos fun vos vu vi vat ven

The interrogative pronouns do not inflect for person, number, or gender. Linguists would say they are animate and inanimate, though Weddish grammar calls them "personal" and "impersonal". They are identical to the relative pronoun (just as in English) and must match their antecedent in animacy, but not in case. Instead (just as in English) they indicate their new role in the relative clause.

Affixes

Like Hebrew, Weddish uses enclitic forms of pronouns to indicate several things. On verbs, pronominal suffixes mark the absolutive argument of the clause. On nouns, they mark a genitive relationship. Pronominal prefixes are used exclusively on transitive verbs to mark the ergative argument, and are obligatory. Weddish is not pro-drop, and an affix on both ends is required on transitive verbs. Remember, there are no ambi-transitive verbs in Weddish. Use of the independent personal pronouns when the person has been specified on either end of the verb is considered emphatic.

Person # Suffix Prefix
1 sg -(n)i ni-
dl -(u)g gu-
pl -(n)u na-
2 sg -(e)d de-
dl -(e)st sti-
pl -ti ta-
3 m.sg -o ro-
f.sg -a ya-
dl -av* ø
pl -(ay/i)č čay-

Verbs

Weddish verbs do not conjugate for tense, only aspect.

Aspect Ablaut

Perfective Imperfective
-ei- -i-
-au- -ai-
-ou- -u-
-e- -a-
-o- -oi-

Voices

  1. Causative: š/že-
  2. Reflexive: hit/hid-
  3. Antipassive: V-u
  4. Mixed

Non-finite

  1. participle -ing
  2. infinitive absolute: u-V
  3. infinitive construct: ge-

Incorporation

On the Mithun scale[10], Weddish does type-I and type-II noun incorporation. This means 1) I picked berries -> I berry-picked, and 2) I washed his face -> I face-washed him.

Derivation

Compounding

When the relationships between nouns is genitive, and it has already been stated or can easily be implied, compound nouns may be formed. For example, a field for football/soccer may become fusbolfeld. (Note the loss of abstraction suffixes.) Suppose it was an Australian rules football field. Would could make fusbolfeldeoystralie. Lastly, If one wanted to add that it is mgroz/composed of grass, this could become פֿוסבאָלפֿעלדעאויסטראליעגראָז/fusbolfeldeoystraliegroz. Words with greater than four parts are deemed colloquial. Word order is "head second", with the first specifier coming at the very front.

Abstract Nouns

All nouns in Weddish are inherently concrete. Two levels of abstraction are possible through suffixation. The first signifies the practice of one or more persons. The second signifies the understanding of the practice. Both are available in both genders, with the masculine form referring to a person (of either gender), however, the "practice"-form occurs much more often in the masculine and the "understanding"-form occurs much more often in the feminine. Remember, this are no truly abstract nouns in Weddish, for they do not exist.

Suffix "Tennis" gloss "Peace" gloss
ø a tennis a game of tennis a šolem a season of peace
-ay dos tenisayo his tennis game šolemaya her (practice of) peace
dem tenisayt the tennis player dem šolemayt the peacemaker
-šaft dos tenisšafte the game of tennis / "tennisology" 'šolemšafte peace know-how
a tenisšaft a tennisologist a šolemšaft a student of peace

Discourse

Formality

Particles

  1. ^ http://wals.info/chapter/1
  2. ^ http://wals.info/chapter/2
  3. ^ http://wals.info/chapter/3
  4. ^ http://wals.info/chapter/4
  5. ^ http://wals.info/chapter/11
  6. ^ As in Hebrew, uvular may be seen as the most prestigious form: http://wals.info/chapter/6
  7. ^ http://wals.info/chapter/9
  8. ^ http://wals.info/chapter/19
  9. ^ http://wals.info/chapter/12
  10. ^ Mithun, Marianne. 1984. The Evolution of Noun Incorporation. Language, Vol. 60, No. 4. pp. 847-894.