Wenetic Scratchpad- NP: Gods above.

Substantial postings about constructed languages and constructed worlds in general. Good place to mention your own or evaluate someone else's. Put quick questions in C&C Quickies instead.
User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Wenetic Scratchpad- NP: Gods above.

Post by Dewrad »

OK, so I've refrained from the entire scratchpad thing so far, but Bristel's IE descendant has reminded me of an idea for an IE-descendant language I've had kicking around for a while now.

Basically, this is a transitional variety between Germanic and Balto-Slavic (two of my favourite IE families), with elements of both but a character of its own. At the moment, I'm just brainstorming some ideas, but comment is appreciated.

Location and Name
Well, given the premise, this is pretty much decided for me: somewhere between the Germanic and Balto-Slavic Urheimats. So probably on the southern shore of the Baltic Sea, probably around the lower Vistula basin. Historically, this area was inhabited by a grouping recorded in Latin and Greek sources as the Veneti (no relation to those found at the head of the Adriatic), and there's some dispute whether they were Balts, Slavs, Germans or quite what. So, for the purposes of this exercise, they're now the speakers of my IE descendant. Let's call the family "Wenetic" for now.

Main Features
(Or: What I want to steal.)

Well, things that Balto-Slavic and Germanic have as common innovations will of course also be present in Wenetic. So that includes things plural oblique case endings in *-m- rather than *--, a thematic genitive singular deriving from the ablative and merger of *o and *a. Interestingly, both Balto-Slavic and Germanic have innovated two "series" of adjectives, weak and strong (pronominal in B-S, IIRC). On the other hand, the inflections of the strong forms are not cognate between the two families, so while Wenetic will definitely have two series of adjectives, it remains to be seen how they're formed.

From the Balto-Slavic side, I want satemisation and a mobile accentual system. From Germanic, a three-way distinction in the stop reflexes and apophony as a pervasive grammatical marker.

From an inflectional and syntactic point of view, I'm really not sure yet. My normal habit when doing this kind of this is to run the soundchanges on the parent's inflected forms and then see what mergers and analogies suggest themselves.

Preliminary Sound Changes
This is a bit fast and loose at the moment, and the chronology might be off in places. The soundchanges I'm most firm on are those leading to Proto-Wenetic, which dates to let's say somewhere around the first centuries BC/AD. This puts most of the distinctive features of the family in place, and is the most recent common ancestor of the two historical dialect groupings (for the sake of argument, we'll call these "Littoral" and "Ripuarian" when we get to them.)

Cowgill's Law: *h₂ and *h₃ become *g between a sonorant (one of *w y m n r l) and *w. *n̥h₃wé > *įgwí

Siever's Law: *y and *w develop an epenthetic *i or *u when following a consonant cluster. *ḱérdʰyos > *ḱérdʰiyos > *čérdeya

Decoupling of syllabic sonorants: the syllabic sonorants * * * * acquired epenthetic vowels: after a (labio-)velar consonant this was *u, otherwise *i. **dn̥ǵʰuh₂ > *dinǯʰū́ > *dįǯū́

Matasović's Law: palatalised dorsals lose their palatalisation when followed by a sonorant and a back vowel. *ḱrópos > *krápa.

Laryngeal merger and first round of loss: the three laryngeals collapse into one, denoted *h (probably representing /ʔ/). This phoneme is then lost word-initially, and between two consonants in non-initial syllables. Note that this takes place after the laryngeals have had their colouring effect on adjacent vowels. **dʰuǵh₂tḗr > *tʰuǯhtḕr > *tʰuštḕ; *dʰuh₂mós > *tʰuhmós > *tʰū́ma

RUKI, the first round: *s becomes *š (probably a voiceless retroflex fricative) when following i y u w r k(ʷ) ḱ g(ʷ)(ʰ) ǵ(ʰ). Note that this operates accross laryngeals. **ksihₓróm > *kšī́rą

Satemisation: the velarised dorsal consonants become affricates: *ḱ ǵ ǵʰ > *č ǯ ǯʰ. *h₁éḱwos > *éčwa; *bʰénǵʰu > *bʰénǯʰu > *pʰę́ǯu

Winter's law: voiced unaspirated stops become preglottalised, merging with sequences of *hC. *tégos > *téhgos > *tḗga

Hirt's Law: the accent is retracted vowels preceding *h in a closed syllable. *dʰuh₂mós > *dʰúhmas > *tʰū́ma

Second laryngeal loss: the remaining laryngeal *h is lost, leading to compensatory lengthening on any preceding vowel. These newly lengthened vowels, when stressed, are typified by a rising tone, while original stressed long vowels are typified by a falling tone. *hₓihₓlu- > *íhlu- > *ī́lu-

Osthoff's Law: long vowels become short in closed syllables where the coda consonant is a sonorant. *gʰḗr > *kʰér

First vowel merger: *o ō become *a ā. *e becomes *i in non-initial syllables. *kʷetwóres > *kʷetwáris > *ketwári

Delabialisation: the labiovelar stops merge with the plain velars. *kʷekʷlóm > *keklą́

First Deaspiration: the aspirated stops merge with the plain voiced stops when preceding another consonant or following a nasal. *bʰéndʰeti > *bʰéndeti[/i] > *pʰę́diti; *bʰréh₂tēr > *brā́tē

Not Grimm's Law: the voiced aspirated stops become voiceless aspirates. *bʰeh₂ǵós > *pʰā́ǯa

Nasalisation: nasal consonants are lost in closed syllables, nasalising the any preceding short vowel in the process: *h₁dónt- > *dąt-

RUKI, the second round: short high oral vowels become mid vowels when preceding i y u w r k g(ʰ): *kʷṛ́mis > *kʷúrmis > *kórmi

Loss of Auslaut Consonants: any remaining word-final consonants are lost. *h₂eǵrós > ā́ǯra.

Kluge's Law: nasal consonants assimilate to a preceding consonant (unless word-initial). *kolnós > *kallá

Pretty similar to Verner's Law: voiceless stops (including voiceless aspirates) become voiced unaspirated stops following an unstressed vowel. (Note that unlike in Germanic, *s is not included in this soundchange.) *ph₂tḗr > *patḗr > *padḗ

Now, if I haven't misled myself entirely, this should give a phoneme inventory somewhere along the following lines:

Code: Select all

                        labial	dental  "palatal" velar labiovelar                     front           back
nasal stops             m       n                                          high        ī i į           ū u ų
stops           [-vce]  p       t       č         k                        
/affricates     [+asp]  pʰ      tʰ      čʰ        kʰ                       mid         ē e ę
                [+vce]  b       d       ǯ         g                        
fricatives                      s       š                                  low                         ā a ą
sonorants                       r l     y                w
And the numerals from one to ten:

*h₁óinos dwṓu tréyes kʷetwóres pénkʷe swéḱs septḿ̥ h₁oḱtṓw néwn̥ déḱm̥t > *áina dwā̀u tréyi kedwári pę́ki swéči sebį́ aǯā̀u náwų déčį
Last edited by Dewrad on Sun Oct 20, 2013 4:39 pm, edited 12 times in total.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad

Post by Dewrad »

Quickly running Schleicher's Fable through the soundchanges (not making any grammatical or lexical changes here), we get the following output, which is perhaps a slightly better way to ascertain the "look and feel" of this stage:

Gorḕi áwi, késya wullā́ ne ésti, éčwą spéči: áiną kʰe gorų́ wáčʰą wéčʰątį, áinąke méčʰį pʰárą, áiną ke ǯménį ā̀ču pʰérątį.
Áwi tu eǯwaima weugí: "Čḕ aggudá mai éčwą ā́ǯątį nérį wīdįtéi."
Éčwā tu weugą́: "Čludí áwei! Čḕ kʰe aggudá, įmméi wīdįttá: nḕ, páti, áuyą e wullą́ sebi korréudi nu gérrą wéstrą néčʰi áuyą wullā́ ne ésti"
Tā́ čeǯlowā̀ áwi ā́ǯrą pʰū́gi
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad

Post by Dewrad »

OK, so I've been dicking around with the soundchanges, and have made a couple of small alterations:

Firstly, nasal vowels now occur in two delicious flavours: short and long. So long oral vowels before a nasal become long nasal vowels. (A further small change, which I admit exists primarily to keep the accusative singular and plural distinct: a short vowel followed by a nasal stop and a fricative becomes a long vowel. This feeds the above rule, so -ans- > -aːns- > -ãːs-.)

The second more makes explicit what I was already operating on anyway: stops (and *s/š) exhibit regressive assimilation for voice. So *dʰugh₂tḗr > *dʰugtḗr > tʰūktḕ. Note that this change occurs after Winter's Law.

And thirdly: I forgot to mention outcomes of dental+dental clusters. Already in PIE there was an allophonic intrusive [s] between adjacent dental stops. In Wenetic, subject to the voicing assimilation described above, the first stop is lost, leaving a cluster of *st(ʰ) or *zd. This change occurred after the first round of RUKI backing, and does not participate in it. Thus PIE *weyd-tós > PW *weistá, not **weištá. A similar change happened to the affricates *č and *ǯ before another stop: they lost the occlusive element and became simple fricatives *š and *ž.

All this means that I've felt the need to fiddle somewhat with how suprasegmentals etc. are represented in this reconstruction. While multiple diacritics on a character don't actually bother me, they seem to offend both typesetting engines and sca2. To me, something like į̄̀ is neat and logical: the ogonek denotes nasalisation, the macron length and the grave falling tone. Unfortunately, the stacking diacritics tend not to show all that well in most programs. As such, long rising vowels will be marked with a circumflex â, while long falling vowels will be marked with a caron ǎ (because they point up and down, see.) As far as I'm aware, this is fairly non-standard usage, so alternative suggestions are welcomed (unless they involve digraphs, I'm really trying to avoid them.)
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Wenetic Scratchpad

Post by Salmoneus »

Dewrad wrote: While multiple diacritics on a character don't actually bother me, they seem to offend both typesetting engines and sca2.
And also the eyes of God...

I'm eventually doing a venetic family myself, as it happens, although in my case they ARE related to the adriatic veneti, and form a band from north to south at one stage. But I've been concentrating on the southern branch of the family. [My venetics aren't particularly closely related to germanics or baltoslavics, they just migrated there - although I expect that when i get around to it I'll bring in some areal features to make them look more germanic/slavic on the surface.]

I'm seeing the Balto-Slavic more than the Germanic so far. Possibly this is because balto-slavic creeps me out for some reason.
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad

Post by Dewrad »

Salmoneus wrote:
Dewrad wrote: While multiple diacritics on a character don't actually bother me, they seem to offend both typesetting engines and sca2.
And also the eyes of God...

I'm eventually doing a venetic family myself, as it happens, although in my case they ARE related to the adriatic veneti, and form a band from north to south at one stage. But I've been concentrating on the southern branch of the family. [My venetics aren't particularly closely related to germanics or baltoslavics, they just migrated there - although I expect that when i get around to it I'll bring in some areal features to make them look more germanic/slavic on the surface.]

I'm seeing the Balto-Slavic more than the Germanic so far. Possibly this is because balto-slavic creeps me out for some reason.
Yeah, a lot of the things that make Germanic look distinctively Germanic are the results of Grimm's and Verner's Laws, neither of which happens in Wenetic. On the other hand, one of the things which makes Balto-Slavic look distinctively Balto-Slavic is satemisation, which does happen in Wenetic. At the moment, the Germanicness is pretty covert (personally, I think it all looks a bit like Lakhota fucked Lithuanian): I'm thinking that the verbal system in particular is going to owe more to Gmc than it will to B-S.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad

Post by Dewrad »

Nominal Declensions

Let's cover the good news first, shall we? Proto-Wenetic has largely done away with PIE's batshit crazy athematic ablauting paradigms, because seriously, fuck that shit man. Most athematic nouns have migrated by analogy to either the thematic class or a largely regularised i-stem class. As a rule, it is the "strong" stem found in the nom/acc which has been generalised accross the paradigm. The only remnants of PIE's athematic classes are the r-stems (pretty much limited to nouns denoting family), the s-stems (all neuters) and the n-stems (all genders). We've also lost the dual number, and a few cases: the genitive, ablative and instrumental have merged.

The bad news, on the other hand, is that soundchange has played merry hell with accentuation patterns, with most declensions possessing subclasses differentiated only by the accentuation pattern.

Now, paradigms. Let's start with the thematic o-stems. These comprise masculine and neuter nouns, which differ in their inflection only in the nominative and accusative. There are three accentual subclasses: barytonic, where the accent does not occur on the ending; oxytonic, where the accent does occur on the ending; and dynamic, where the accent can occur on either (this last class derives from oxytones subject to Winter's Law.) As examples, let us take the following masculine nouns: barytonic *éčwa 'horse', oxytonic *deiwá 'god' and dynamic *dûma 'smoke'. To illustrate neuter inflection let's use *wérǯą 'work', *čʰudą́ 'lesser spirit' and *yûgą 'yoke', representing respectively the three accent classes.

Code: Select all

              |   barytonic        oxytonic        dynamic
--------------|--------------------------------------------
singular nom  |  éčwa, wérǯą     deiwá, čʰudą́    dûma, yûgą
         acc  |  éčwą, wérǯą     deiwą́, čʰudą́    dûmą, yûgą
         dat  |     éčwāi           deiwǎi          dūmǎi
         gen  |     éčwā            deiwǎ           dūmǎ
         loc  |     éčwei           deiwéi          dûmei
plural   nom  |  éčwā, wérǯā     deiwǎ, čʰudâ    dūmǎ, yūgâ
         acc  |  éčwą̄, wérǯā     deiwą̂, čʰudâ    dûmą̄, yūgâ
         dat  |    éčwamu          deiwámu         dûmamu
         gen  |     éčwą̄            deiwą̂           dūmą̂
         loc  |    éčwaišu         deiwáišu        dûmaišu
The accentuation pattern of the dynamic declension is simple: in the dative and genitive singular, and the nominative and genitive plural the accent occurs on the inflection, otherwise it occurs on the stem.

The ā-stems are somewhat simpler, having only two subclasses: barytonic and oxtyonic. Let's use barytonic *wúllā 'fleece' and oxytonic *awegâ 'ewe' as examples:

Code: Select all

              |   barytonic    oxytonic    
--------------|--------------------------
singular nom  |     wúllā       awegâ
         acc  |     wúllą̄       awegą̂
         dat  |     wúllāi      awegâi
         gen  |     wúllā       awegâ
         loc  |     wúllā       awegâ
plural   nom  |    wúllāsi     awegâsi
         acc  |    wúllą̄si     awegą̂si
         dat  |    wúllāmu     awegâmu
         gen  |     wúllą̄       awegą̂
         loc  |    wúllāsu     awegâsu
Note that the nominative and accusative plurals are innovations. The expected reflex of PIE *-eh₂es and *-eh₂ns are **-ā and **-ą̄, which are identical to the singular. It appears that Pre-Proto-Wenetic made use of an "extended plural", much like that found in the Proto-Germanic a-stem plural *-ōziz.

Ever onwards, we have i-stems and u-stems, which have become rather numerous in PW on account of their assumption of a number of other athematic noun classes. All three genders can be found in these classes, although no distinction is made in inflection between masculine and feminine nouns, and neuters are distinct only in the nominative and accusative. Again, we have accent-based subclasses: a dynamic subclass and a barytonic subclass. For the i-stems, we have the feminine nouns *mę́ti 'mind' and *ǯnâti 'wisdom' for the dynamic and barytonic classes respectively, and for the u-stems we have the masculine nouns *ǯústu 'choice' and *tʰêlu 'baby'. Representing the neuters of their classes are *mári 'sea' and *dáru 'tree'.

Code: Select all

              |            i-stems               |             u-stems            
              |    dynamic          barytonic    |    dynamic            barytonic    
--------------|----------------------------------|--------------------------------
singular nom  |  mę́ti, mári          ǯnâti       |  ǯústu, dáru            tʰêlu
         acc  |  mę́tį, mári          ǯnâtį       |  ǯústų, dáru            tʰêlų
         dat  |   mędéyei           ǯnâteyei     |    ǯuštáwei            tʰêlawei
         gen  |    mędéi             ǯnâtei      |     ǯuštáu              tʰêlau
         loc  |    mędéi             ǯnâtei      |     ǯuštáu              tʰêlau
plural   nom  | mę́teyi, márī        ǯnâteyi      |  ǯústawi, dárū         tʰêlawi
         acc  |  mę́tį̄, márī          ǯnâtį̄       |   ǯústų̄, dárū           tʰêlų̄
         dat  |   mędímu            ǯnâtimu      |    ǯuštúmu             tʰêlumu
         gen  |   mędéyą̄            ǯnâteyą̄      |    ǯuštáwą̄             tʰêlawą̄
         loc  |   mędísu            ǯnâtisu      |    ǯuštúsu             tʰêlusu

See that shit going on there in the declension of *mę́ti? The interchange between stressed *mę́t- and unstressed *męd-? That is due to "Pretty similar to Verner's Law" (vide supra), which states that voiceless medial stops become voiced when not preceded by an accented syllable. The alternation seems to have been pervasive throughout PW, although the daughter languages have mainly generalised one form or the other.

For a declension comprising a sum total of only five fucking nouns, the r-stems are unwontedly complex, having two species of dynamic accent as well as a static barytone accent. It is probably less hassle just to present the whole lot:

Code: Select all

             |   father   daughter   mother   brother   sister
-------------|--------------------------------------------------
singular nom |   padě     tʰūktě     mâtē     brâtē     súšā
         acc |   padérį   tʰûkterį   mâterį   brâterį   súšarį
         dat |   paderéi  tʰūkteréi  mâterei  brâterei  sušréi
         gen |   paderí   tʰūkterí   mâteri   brâteri   sušrí
         loc |   padéri   tʰûkteri   mâteri   brâteri   sušéri
  plural nom |   padéri   tʰûkteri   mâteri   brâteri   súšari
         acc |   padérį̄   tʰûkterį̄   mâterį̄   brâterį̄   súšarį̄
         dat |   padermú  tʰūktermú  mâtermu  brâtermu  sušermú
         gen |   paderą̂   tʰūkterą̂   mâtrą̄    brâtrą̄    sušrą̂
         loc |   paderšú  tʰūkteršú  mâteršu  brâteršu  sušeršú

Only two declensions remain: n-stems and s-stems. Happily, both occur only in barytonic flavour: no accent mobility here! N-stems generally only occur in masculine or feminine: the sole example of a neuter n-stem is pretty irregular. S-stems only occur in the neuter gender. So here we have the masculine n-stem *pâšā 'forest spirit', the sole neuter *námį 'name' and the s-stem *népʰa 'sky':

Code: Select all

              |            n-stems             |     s-stems
              |    epicene          neuter     |
--------------|--------------------------------|----------------
singular nom  |     pâšā             námį      |      népʰa
         acc  |    pâšanį            námį      |      népʰa
         dat  |    pâšinei          náminei    |     népʰisei
         gen  |    pâšini           námini     |     népʰisa
         loc  |    pâšani           námini     |     népʰisi
plural   nom  |    pâšani            į́mmā      |      népʰā
         acc  |    pâšanį̄            į́mmā      |      népʰā
         dat  |    pâšįmu            į́mįmu     |     népʰimmu
         gen  |    pâšiną̄            į́mmą̄      |     népʰisą̄
         loc  |    pâšį̄su            į́mį̄su     |     népʰisu
And, uh, wow. I think that's it?

Oh, wait. Now we've done nouns, we can see what these people called themselves! A man would be a *wénida, several would be *wénidā. A woman would be a *wénidā, she with her sisters would be a group of *wénidāsi. Their own name for where they live might well have been *wénidą̄ lą́dowa 'land of the Wenetoi'.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
WeepingElf
Smeric
Smeric
Posts: 1630
Joined: Wed Mar 08, 2006 5:00 pm
Location: Braunschweig, Germany
Contact:

Re: Wenetic Scratchpad

Post by WeepingElf »

This language rocks. It may be "just another IE language", but a well-done one. Rock on!
...brought to you by the Weeping Elf
Tha cvastam émi cvastam santham amal phelsa. -- Friedrich Schiller
ESTAR-3SG:P human-OBJ only human-OBJ true-OBJ REL-LOC play-3SG:A

Karinta
Sanci
Sanci
Posts: 25
Joined: Tue Mar 19, 2013 3:41 pm
Location: US

Re: Wenetic Scratchpad

Post by Karinta »

Velars become palatals before the /i/ and /e/ vowels and /j/. Meanwhile, the palatals become alveolar affricates.

User avatar
Click
Avisaru
Avisaru
Posts: 620
Joined: Sun Mar 04, 2012 11:53 am

Re: Wenetic Scratchpad

Post by Click »

What on the earth is the point of your post?

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad

Post by Dewrad »

WeepingElf: thanks! It's surprising how quickly it's all coming together. I am, however, dreading verbs.

Karinta: is that a general statement or a suggestion? If the latter, thanks, but I have my own ideas on how the daughters are going to evolve.

dhok: Drawing conclusions about the nature of Proto-Indo-European from these posts is probably not a good idea.

Pronouns

I know it would be more logical to do adjectives after nouns, but adjective declension is going to rely pretty heavily on pronominal declension, so we'll do this first. First up, personal pronouns of the first and second person, and the reflexive pronoun:

Code: Select all

              |    1st        2nd        ref    
--------------|---------------------------------
singular nom  |    êǯį        tû          -
         acc  |    mę̂ǯi       tę̂ǯi       sę̂ǯi
         dat  |    mái        tái        sái
         gen  |    méni       téni       séni
         loc  |    mái        tái        sái
plural   nom  |    įgwéi      ûwei        -
         acc  |    į̄mmí       ušwí        -
         dat  |    į̄mméi      ušwéi       -
         gen  |    į̄są́        ušą́         -
         loc  |    į̄mméi      ušwéi       -
A few things to notice here. The *-ǯi in the accusative singular is cognate to the accusative *-k in Proto-Germanic pronouns. In Wenetic, it isn't limited to the personal pronouns either: we'll be seeing this again. The plural nominative forms are actually derived from the old duals, but otherwise there hasn't been a massive amount of remodelling in the paradigm. The genitive pronouns aren't used to mark possession: rather possessive adjectives are used (we'll get to those later.)

Beyond this point, I'm just throwing out some paradigms: I'm not fully sure yet how they're going to be used. Some ideas afterwards.

First off: pronominal paradigm 1. This comprises reflexes of PIE *i/e- and *ḱi/e-. A couple of things to note here: the accusative singular (in the masculine and feminine forms; of course, the neuter accusative coincides with the nominative) is formed with the same extension *-ǯi as seen above in the first and second person singular pronouns. Additionally, note the syncretism in the plural oblique cases (the same occurs in Gmc.):

Code: Select all

              |              *i/e-              |             *ḱi/e-
              |     m          f          n     |     m          f          n     
--------------|---------------------------------|---------------------------------
singular nom  |    éi         éyā        îdį    |    čéi        čéyā       čîdį
         acc  |    į́ǯi        éyą̄ǯi      îdį    |    čį́ǯi       čéyą̄ǯi     čîdį
         dat  |    émmei      ésyāi      émmei  |    čémmei     čésyāi     čémmei
         gen  |    ésa        ésyā       ésa    |    čésa       čésyā      čésa
         loc  |    émmi       ésyā       émmi   |    čémmi      čésyā      čémmi
plural   nom  |    éyi        éyāsi      î      |    čéyi       čéyāsi     čî
         acc  |    į̂          éyą̄si      î      |    čį̂         čį̂         čî
         dat  |    éišmu      éišmu      éišmu  |    čéišmu     čéišmu     čéišmu
         gen  |    éišą̄       éišą̄       éišą̄   |    čéišą̄      čéišą̄      čéišą̄
         loc  |    éišu       éišu       éišu   |    čéišu      čéišu      čéišu
And then second, pronominal paradigm 2. This comprises reflexes of the PIE demonstrative *so, the relative *yo and the interrogative *kʷo (the last has a particularly impoverished declension, not distinguishing between the masculine and feminine genders). As in pronominal paradigm 1 above, note the syncretism in the plural oblique forms; on the other hand, pronouns of the second group do not exhibit accusatives in *-ǯi. Also in this group is *ená, which declines like *.

Code: Select all

              |               *so-              |               *yo-              |         *kʷo-
              |     m          f          n     |     m          f          n     |    m/f         n     
--------------|---------------------------------|---------------------------------|----------------------
singular nom  |     sá         sâ       tâdį    |     yá        yâ         yâdį   |     ká        kâdį
         acc  |     tą́         tą̂       tâdį    |     yą́        yą̂         yâdį   |     ką́        kâdį
         dat  |     támmei     tásyāi   támmei  |     yámmei    yásyāi     yámmei |         kámmei
         gen  |     támmā      tásyā    támmā   |     yámmā     yásyā      yámmā  |         kámmā
         loc  |     támmi      tásyāi   támmi   |     yámmi     yásyāi     yámmi  |         kámmi
plural   nom  |     tái        tāsi     tâ      |     yái       yāsi       yâ     |     kái       kâ
         acc  |     tą̂         tą̂       tâ      |     yą̂        yą̂si       yâ     |     ką̂        kâ
         dat  |     táimu      táimu    táimu   |     yáimu     yáimu      yáimu  |         káimu
         gen  |     táišą̄      táišą̄    táišą̄   |     yáišą̄     yáišą̄      yáišą̄  |         káišą̄
         loc  |     táišu      táišu    táišu   |     yáišu     yáišu      yáišu  |         káišu
So, what are all these pronouns for? I honestly don't know yet. All six stems are attested in both Germanic and Balto-Slavic, but with differing uses, and I'm not entirely sure what the situation is likely to be in Wenetic. A survey, with some thoughts:
  • *i/e-: Gmc: The PGmc reflex is *iz, which gives (at least a form of) the 3rd person pronoun in a number of branches. B-S: probably not present.
  • *ḱi/e-: Gmc: The PGmc reflex is *χī, which gives the Gothic proximal demonstrative, but in most Gmc languages replaces *iz as the 3rd person pronoun (q.v. English "he"). B-S: The reflex here is *śis, which gives forms meaning "this" in both branches.
  • *so-: Gmc: As well as the definite article, this also gives the distal demonstrative. B-S: The reflex here gives a distal demonstrative in both languages.
  • *yo-: Gmc: Not attested as an independent morpheme per se, rather as part of various compounds. B-S: gives the relative pronoun *jĭže in Sl.
  • *kʷo-: Provides the interrogative pronoun in Gmc and B-S.
  • *h₁eno-: Gmc. Reflex is *enaz, which gives the Scandinavian definite article. B-S. gives the Sl. 3rd person pronoun *onŭ and Lithuanian *anàs "that".
OK, what to take from this for Wenetic?
  • at this stage, the 3rd person pronoun is still *éi, although I expect this to be lost in the daughter languages.
  • *čéi and * are the proximal/distal demonstratives respectively.
  • * retains a strictly relative function
  • * is the interrogative but (at this stage, at least) lacks relative function.
  • *ená is the definite article. The only other contemporaneous IE language with a definite article may have been Greek, but I don't see why Wenetic can't have been the innovator here. Besides, I like definite articles.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Niedokonany
Lebom
Lebom
Posts: 244
Joined: Thu Jun 28, 2007 10:31 pm
Location: Kliwia Czarna

Re: Wenetic Scratchpad- Now Playing: "Too Many Pronouns?"

Post by Niedokonany »

Why is the length retained in padḗ? Why does dûma have initial accent in the nom/acc/loc sg, but final in the dat/gen sg? Didn't Hirt's law operate there? *yo- also functions as a personal pronoun in Slavic (intermixing with *so-/*h₁eno-): OCS acc. masc. sg и, Pl. acc. neut. sg / acc nonvir. pl je.
uciekajcie od światów konających

User avatar
Chagen
Avisaru
Avisaru
Posts: 707
Joined: Thu Sep 22, 2011 11:54 pm

Re: Wenetic Scratchpad- Now Playing: "Too Many Pronouns?"

Post by Chagen »

So I guess the dual was completely ost?
Nūdhrēmnāva naraśva, dṛk śraṣrāsit nūdhrēmanīṣṣ iźdatīyyīm woḥīm madhēyyaṣṣi.
satisfaction-DEF.SG-LOC live.PERFECTIVE-1P.INCL but work-DEF.SG-PRIV satisfaction-DEF.PL.NOM weakeness-DEF.PL-DAT only lead-FUT-3P

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- Now Playing: "Too Many Pronouns?"

Post by Dewrad »

Niedokonany wrote:Why is the length retained in padḗ?
I'm not sure what you're getting at here: to my knowledge there's nothing in my sound changes that would indicate it wouldn't be?
Why does dûma have initial accent in the nom/acc/loc sg, but final in the dat/gen sg? Didn't Hirt's law operate there?
In the genitive singular it's regular: the PW thematic genitive derives from the PIE ablative, not the genitive: so something like *duhₓmṓd. In the dative singular... it's a glitch in my sca file, embarrasingly. The stress should indeed be on the initial syllable there, not the ultima. Thanks for catching that, although I do now need to run all the paradigms through the sca engine again :(
*yo- also functions as a personal pronoun in Slavic (intermixing with *so-/*h₁eno-): OCS acc. masc. sg и, Pl. acc. neut. sg / acc nonvir. pl je.
Thanks for that, I wasn't aware. That may well come in handy.
Chagen wrote:So I guess the dual was completely ost?
I was initially in two minds about keeping the dual or not, as it's retained in both Gmc and B-S. However, the soundchanges rendered the forms only marginally distinct, so I decided to jettison it in nouns and pronouns at least: depending on how they work out in the verbs, I might do something similar to Gothic and keep dual as a category marked on verbs.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Niedokonany
Lebom
Lebom
Posts: 244
Joined: Thu Jun 28, 2007 10:31 pm
Location: Kliwia Czarna

Re: Wenetic Scratchpad- Now Playing: "Too Many Pronouns?"

Post by Niedokonany »

Dewrad wrote:
Niedokonany wrote:Why is the length retained in padḗ?
I'm not sure what you're getting at here: to my knowledge there's nothing in my sound changes that would indicate it wouldn't be?
Osthoff's Law: long vowels become short in closed syllables where the coda consonant is a sonorant. *gʰḗr > *kʰér
Though having looked up the WP article ("followed in turn by another consonant"), it could be that you meant sth else than I thought.
Why does dûma have initial accent in the nom/acc/loc sg, but final in the dat/gen sg? Didn't Hirt's law operate there?
In the genitive singular it's regular: the PW thematic genitive derives from the PIE ablative, not the genitive: so something like *duhₓmṓd. In the dative singular...
But what is it here exactly that doesn't permit the retraction to the first syllable, the final plosive?
uciekajcie od światów konających

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- Now Playing: "Too Many Pronouns?"

Post by Dewrad »

Niedokonany wrote:
Dewrad wrote:
Niedokonany wrote:Why is the length retained in padḗ?
I'm not sure what you're getting at here: to my knowledge there's nothing in my sound changes that would indicate it wouldn't be?
Osthoff's Law: long vowels become short in closed syllables where the coda consonant is a sonorant. *gʰḗr > *kʰér
Though having looked up the WP article ("followed in turn by another consonant"), it could be that you meant sth else than I thought.
Yeah, I'm afraid my wording is off here (compounded by a shitty example): in PW at least, Osthoff's only applies in internal closed syllables.
Why does dûma have initial accent in the nom/acc/loc sg, but final in the dat/gen sg? Didn't Hirt's law operate there?
In the genitive singular it's regular: the PW thematic genitive derives from the PIE ablative, not the genitive: so something like *duhₓmṓd. In the dative singular...
But what is it here exactly that doesn't permit the retraction to the first syllable, the final plosive?[/quote]
Exactly so. Winter's operates here, giving something like *duhₓmṓhₓd. The accent does not retract if the original vowel already precedes a laryngeal.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- Now Playing: "Too Many Pronouns?"

Post by Dewrad »

EDIT: HAHAHA DISREGARD THIS I SUCK COCKS.

Verbs: Overview

As I remarked to dhok a few days ago, the reconstructed PIE verbal system provided by most handbooks is, by and large, pretty much useless as is for most IE conlangs. The "Cowgill-Rix" verb only really neatly resolves into the Graeco-Aryan verbal system: most other branches have reorganised it pretty much from the ground up. So this presents a few challenges for Wenetic.

Going back to my basic plan of Wenetic being an intermediate variety between Germanic and Balto-Slavic, it would make sense first of all to compare the verbal systems of the two branches and go from there: what both share will also occur in Wenetic, and then where they differ some kind of via media would be appropriate. A fine plan, which unfortunately falls down in its details.

At first glance, the two families have a great deal in common: the three-way aspectual division of PIE is given up in favour of a tense-based system divided between past and nonpast. Person and number are still marked. The three moods are reduced to two, with the subjunctive being lost and the optative retained. Athematic root verbs are generally thematised. Only Germanic retains the middle, admittedly, but I knew from the start that I'd be retaining this for Proto-Wenetic at least.

Unfortunately, it all goes tits up when we get to the past tense. For a start, I can't find any decent reconstructions of the Proto-Balto-Slavic verbal system. Slavic seems to be rather more innovative than Baltic, but both are rather impoverished in their retention of PIE categories. While both families have changed PIE's aspect-based system into a tense-based one, each has done it differently. In Germanic, the PIE perfect (actually a stative) has provided the strong past tense, while a periphrasis gives the weak (or derived) past tense. In Slavic, it appears that the PIE aorist (actually a perfective) has given the past tense, with derived verbs gaining an innovated inflection based on the sigmatic aorist.

My initial idea was to combine the two systems: strong (underived) verbs would base their past tense on the PIE perfect like Germanic, while weak (derived) verbs would have inflection based on the sigmatic aorist. I was rather pleased with such an elegant solution, until I realised that I thought it so elegant precisely because this is what happens in Proto-Celtic. There's nothing inherently implausible about it developing in Wenetic, but I wanted something less strikingly obvious to the IE family I'm most familiar with.

What to do?

Verbs: The Present Tense

Happily, the formation of the present tense is relatively easy. It distinguishes three moods: the indicative, the subjunctive (derived from the PIE optative) and the imperative (defective in that it only has 2nd and 3rd person forms). Two diatheses: active and passive. And, the normal complement of three persons in two numbers.

From the point of view of conjugation, in the present tense verbs are divided into two groups: thematic and athematic, a division which does not entirely overlap with the weak/strong division we will encounter in the past tense. There has been a great deal of analogical levelling between PIE and PW, so the distinction between thematic and athematic stems really just boils down to the vowel connecting the stem to the personal desinence in the indicative. The subjunctive endings are pretty much uniform accross both groups.

More significant perhaps is stem formant. A number of verbs mark the present tense by adding a stem formant to the verbal root. There are a total of ten of these, some more common than others. A third and final consideration is the position of the accent. Present tense verbs can either stress the root or the ending (that is, the personal desinence and any preceding stem formant). Perversely, I shall refer to these two categories as rhizotonic and katalesiotonic, respectively.

Now, I know that an archaic IE lang is going to be pretty paradigm-heavy, but really and truthfully there isn't much point to giving ten verbs fully conjugated when there's not all that much between them. Templates are what it's all about. So, let's look at the personal desinences first:

Code: Select all

     |            active                          passive
     |  indicative      subjunctive      indicative      subjunctive
=====|==============================================================
1sg  |      -ą̄              -yį              -i             -ya       
2sg  |      -si             -i            -dai -tai      -ida -ita
3sg  |    -di -ti           -i            -dai -tai      -ida -ita
1pl  |      -ma            -ima            -mastʰa         -imida
2pl  |    -du -tu        -idu -itu       -dowi -tʰowi  -idowi -itʰowi
3pl  |   -Ndi -Nti          -yį          -Ndai -Ntai        -ira
In the table above, there are a few things to note:
  • Where two forms are given, the one to the right is used with katalesiotonic verbs, the one to the left with rhizotonic verbs.
  • The capital -N- in the 3pl forms indicates nasalisation of the preceding vowel. So with the stem *pʰéra-, we have the 3pl *pʰérądi.
  • The 1sg indicative active ending *-ą̄ always displaces any stem-final vowel. The other endings do not.
Now onto stem formants, in handy list format. Remember, these are the things that come between the root and the desinence:

Thematic formants:
  1. Class T1: always rhizotonic. In the indicative, characterised by *-a- in the 1pl and 3pl and *-i- otherwise. Subjunctive *-a-.
  2. Class T2: can occur in rhizotonic or katalesiotonic variants. The indicative is characterised by *-- or *-ya- in the 1pl and 3pl, and either *-- or *-yi- otherwise. Subjunctive *-ya- or *--.
  3. Class T3: always katalesiotonic. In the indicative, characterised by *-ská- in the 1pl and 3pl and *-skí- otherwise. Subjunctive *-ská-. Note that the *s of the formant occurs as *š after velars, rhotics and high vowels.
  4. Class T4: always rhizotonic. In the indicative, characterised by *-sa- in the 1pl and 3pl and *-si- otherwise. Subjunctive *-sa-. Note that the *s of the formant occurs as *š after velars, rhotics and high vowels.
  5. Class T5: can occur in rhizotonic or katalesiotonic variants. The indicative is characterised by *-éya- or *-eya- in the 1pl and 3pl, and either *-éyi- or *-eyi- otherwise.
  6. Class T6: always katalesiotonic. In the indicative, characterised by *-syá- in the 1pl and 3pl and *-syí- otherwise. Subjunctive *-syá-. Note that the *s of the formant occurs as *š after velars, rhotics and high vowels. When applied to a root ending in a consonant or a long vowel, a paragogic *e is inserted between the *s/š and the *y.
With the athematic formants, we move onto slightly more complex territory.
  1. Class A1: roots with final consonants double this and add *-â- in the indicative and *-á- in the subjunctive. Roots with final vowels affix *-- and *--, respectively.
  2. Class A2: again, consonant-final roots double the final consonant while vowel-final roots add *-n-. Then it gets odd. In the 1sg indicative, *-áw- occurs before the desinence. In the 2sg and 3sg indicative the diphthong *-áu- obtains, which causes the 2sg desinence to become *-ši. in the plural, one finds the vowel *-ú-. Happily, in the subjunctive, you get *-- after the nasal/geminate all the way through. Mercifully, this is a very restricted category.
  3. Class A3: nice and simple: the indicative formant is *-ê- in all persons and numbers, and the subjunctive formant is *-á-. Unfortunately, this isn't a very common class.
  4. Class A4: the only athematic class to be rhizotonic. The formant is *-ā- in the indicative and *-a- in the subjunctive. Not a very common class in the protolanguage, but becomes increasingly so in the daughters.
So, by this point we should be able to form some simple sentences! Let's take a few examples, shall we?

Starting with something easy, *yépʰ- 'copulate' is a T1 root. So, for a nice sentence like 'a bull is fucking the cow' we need the 3sg desinence *-di and the formant *-i-: *taurá eną́ gǎwį yépʰidi.

Moving on, what about the T3 root *per(č)- 'to annoy'. I want the sentence 'I am annoyed by you'. So we need the formant: *-šká- in this case because the root ends in a rhotic. And then the 1sg passive desinence *-i. So: *perškái téni.

Something athematic next. The A2 root *tʰū- 'to tremble'. The sentence is 'thou tremblest, maggot!". Formant is going to be *-náu-, as it's a vowel-final root. The desinence will be *-ši-, following as it does a high vowel. And our Wenetic actor declaimeth *ā kormi, tû tʰūnáuši!.

To finish with, let's have a gander at the one irregular verb I've bothered to investigate so far (I suspect there may be more), the verb 'to be'. For obvious reasons, it lacks a passive:

Code: Select all

     |  indicative      subjunctive    
=====|=================================
1sg  |     émmi	           sáyį
2sg  |     ési             sái
3sg  |     ésti            sái
1pl  |     simá            sáimi
2pl  |     stáwa           sáitu
3pl  |     sę́ti            sáyį
Next up, the past tense. When I decide on the details.
Last edited by Dewrad on Tue Sep 10, 2013 4:48 pm, edited 1 time in total.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- Now Playing: "no time like the Prese

Post by Dewrad »

Verbs: "Reimagining the Present" and the Past Tense

The further I think about the present tense formation above, the less happy I am; particularly now I've thought through the past tense more thoroughly. While I'm broadly happy with the personal desinences, I think the stem formant categories can be reduced and rationalised rather better, with analogy playing more of a role than it does in the system above. So. Ignore (largely) everything above, and let's have a look at the following:

By the way, paragraphs in this colour indicate a digression, and can be safely ignored.

Overview

The categories marked by the Proto-Wenetic verb were as following:
  • Two tenses: the past and non-past.
  • Three persons: first, second and third.
  • Two numbers: singular and plural.
  • Two moods: indicative and optative.
  • Two diatheses: active and passive, with the latter only marked synthetically in the non-past tense.
  • Four participles: an active and a passive participle in each tense.
  • Two verbal nouns: the infinitive and the supine.
The above list of categories I feel falls nicely between the Balto-Slavic system and the Proto-Germanic, although the resemblance here is obviously closer to the Germanic side. I'm going to blame common substratal influence along the southern Baltic coast.

The inflection of the verb was broadly organised into two main categories, with a number of conjugations in each. Conventionally, these are called strong and weak verbs, according to how they form their past tenses.

A Tour of the Verb

I am somewhat shocked that nobody has called me on my perverse use of recherché terminology in my foregoing posts. I know that I tend to use less-common terms at times, but there are reasons behind this: they frequently afford greater precision. So it's all very well for me to talk about "desinences" and "ending" and to contrast "stems" and "roots", but what precisely am I banging on about? Here's a helpful diagram of *pʰarškáma 'we are wearing':

Image

So, the root is what contains the basic semantic information. In this case, it's 'carrying'. Note that the root vowel can undergo changes (ablaut/apophony) to indicate grammatical distinctions, in PW primarily tense, but also number in some cases.

The ending is simply everything after the root. This can include a derivational affix, a thematic vowel and a personal desinence, as shown here.

The stem formant (or derivational affix) might or might not convey any semantic meaning. If it does, it's often pretty opaque to the synchronic language (as here, where the PIE frequentive-iterative suffix has been added to *bʰer- 'carry', with a semantic change from 'carry around' to 'wear clothes'). The stem formant might even not be overtly pronounced. It might not be present in all forms of the verb, or it might have different forms depending on the tense. Either way, the stem formant does not have any overt syntactic role in the utterance: it doesn't mark any of the oppositions given in our list at the top.

Together with the root, these two make up the inflectible stem, to which further suffixes are added which do mark syntactic oppositions.

The thematic vowel (I apologise here for deviating slightly from classical IEist usage) is generally where mood and (sometimes) tense are marked.

Finally, the desinence is the workhorse suffix, conveying information about person, number, diathesis and tense. Treat it with respect.

(This post has gotten stupid long, so I'm splitting it. Next is strong verbs, once I've faffed around with turning my neat tab-stopped tables into space-delimited bbcode [ code ] monstrosities, then weak verbs and finally the messy residue of irregulars. Hopefully we'll get on to periphrasis by the end of the week.)
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Sevly
Lebom
Lebom
Posts: 214
Joined: Sat Mar 31, 2007 10:50 pm
Location: (x, y, z, t)

Re: Wenetic Scratchpad- NP: "no, I've totally changed my min

Post by Sevly »

Great diagram! Obscure terminology can't be clarified much better than that.

User avatar
WeepingElf
Smeric
Smeric
Posts: 1630
Joined: Wed Mar 08, 2006 5:00 pm
Location: Braunschweig, Germany
Contact:

Re: Wenetic Scratchpad- NP: "no, I've totally changed my min

Post by WeepingElf »

Wow. This treatment of the Wenetic verb betrays profound knowledge of Indo-European verbal morphology (as far as I can tell; your knowledge seems to be better than mine). Most conlangs based on PIE simply either throw out most of the complexity of the Cowgill-Rix system by assuming simplifying restructurings like the Germanic weak verbs to run their course, or just keep it the way it was in Late PIE; you work with it in a realistic and creative way. Applause!
...brought to you by the Weeping Elf
Tha cvastam émi cvastam santham amal phelsa. -- Friedrich Schiller
ESTAR-3SG:P human-OBJ only human-OBJ true-OBJ REL-LOC play-3SG:A

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- NP: "no, I've totally changed my min

Post by Dewrad »

WeepingElf wrote:Wow. This treatment of the Wenetic verb betrays profound knowledge of Indo-European verbal morphology (as far as I can tell; your knowledge seems to be better than mine). Most conlangs based on PIE simply either throw out most of the complexity of the Cowgill-Rix system by assuming simplifying restructurings like the Germanic weak verbs to run their course, or just keep it the way it was in Late PIE; you work with it in a realistic and creative way. Applause!
Thank you :) Any such appearence of profound knowledge is, I assure you, entirely illusory.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- NP: "no, I've totally changed my min

Post by Dewrad »

Strong Verbs

There are three conjugations of strong verbs, the first two of which are made up of two sub-conjugations. By and large, the strong verbs are derived from PIE's thematic verbs, but a number of athematic verbs have migrated to this class.

The case of *léik- 'leave' is an instructive and typical example. In PIE, this verb had an athematic present tense characterised by the nasal infix, so the present stem was *linékʷ-. Pre-PW had a fairly strong aversion accent-induced stem apophony within the present tense, preferring verbs to have a fixed accent and a stem largely consistent accross person and number. Those verbs where the infix could not be reanalysed as a suffix, a thematic present was generally derived on the basis of the aorist subjunctive, in this case *léykʷ-o-: in this form the accent remains on the root and the stem is the same throughout the present tense.

The first two conjugations are the asigmatic and sigmatic, which is a touch more memorable than first and second. Verbs of the sigmatic conjugation are characterised in the present by a sigmatic stem formant after the root, either -s- or -š-, depending on the preceding segment (š is used after rhotics, dorsals and high vowels). In the past tense, this stem formant is lost. The stem formant in the asigmatic conjugation is a null morpheme, which we'll represent using -Ø-.

Both conjugations have two subtypes: iotated and uniotated. The iotated subtype is characterised by a -y- between the stem formant and the thematic vowel, while the uniotated isn't. Simple. Ish. There's Siever's law to consider. If this -y- occurs after a consonant cluster, or a long vowel and a consonant, a epenthetic unaccented -e- intrudes.

The third conjugation (sorry, no handily memorable name here) is characterised by the stem formant -sk-, which takes the form -šk- after rhotics, dorsals or high vowels. Again, in the past tense, the stem formant is lost.

Time for some paradigms!

The Present Tense

Two things to be aware of for the present tense:

The present tense is characterised by two possible accentuation patterns. Retaining the perverse terminology introduced in the previous post, there are (in the present tense) katalesiotones, those verbs stressed on the ending; rhizotones, verbs stressed on the root. The latter are far more common than the former; indeed, katalesiotones are only present in the iotated subclasses of the first two conjugations, and all third conjugation verbs are katalesiotonic in the present tense, which makes things a touch simpler.

Secondly, the thematic vowel, which precedes the ending, contributes to the marking of mood. In the subjunctive, it is always a (or stressed á in katalesiotones). In the indicative, it varies according to person: before the personal desinences of the first and third persons plural it is a/á, while in the other persons it is i/í.

Below we have five verbs, one of each (sub)conjugation, fully conjugated in the present tense. Our verbs are first conjugation *pʰér- 'carry' and *wórǯ- 'make', second conjugation *lékš- 'hide' and *čáuš- 'hear' and finally *per(č)- 'annoy' for the third conjugation.

First (asigmatic) conjugation

*pʰér- 'carry', uniotated rhizotonic

Image

*wórǯ- 'make', iotated rhizotonic

Image

Note the epenthetic -e- following the consonant cluster.

Second (sigmatic) conjugation

*lékš- 'hide', uniotated rhizotonic

Image

*čáuš- 'hear', iotated rhizotonic

Image

Note the epenthetic -e- following a diphthong (equivalent to a long vowel) and a consonant.

Third conjugation

*per(č)- 'annoy' katalesiotonic

Image

Note here first off that a final root cluster (in this case -rč-) simplifies before the stem formant, here losing the second element.

The personal desinences here are not unique to the third conjugation, they are used with any katalesiotonic stem: for example, the iotated katalesiotonic first conjugation *sad- 'seat' has a third person form *sadyíti.

Again with the splitting. Next the strong past tense.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: Wenetic Scratchpad- NP: Strong verbs pt1

Post by hwhatting »

Very well though-out. Now, where is the past tense? ;-)

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- NP: Strong verbs pt1

Post by Dewrad »

The Past Tense

The past tense, on paper, is actually rather straightforward for strong verbs. Sensu stricto, there's only the one conjugation, one accentual paradigm. The passive isn't marked, so that's one fewer set of ending to worry about. It's all gravy.

The strong past tense is straightforwardly derived from the PIE perfect tense, with only minimal modifications, generally on account of analogy. Diachronically, it's the easiest class. The difficulties arise due to the vagaries of soundchange, particularly the effect of neighbouring consonants on vowels. Interestingly, as in Germanic, reduplication has largely been lost (possibly as a result of haplology, possibly as a reaction against not-quite-Verner's-law alternations of consonants following unstressed vowels).

Almost. The strong verb forms its past tense by root apophony, known and loved by IEists as ablaut. That is, the root vowel changes in a predictable way in order to indicate the past tense. However, the change of the root vowel is frequently unpredictable synchronically: as in Germanic there are distinct ablaut classes. Seven (well, eight) to be precise, determined largely (but not entirely!) by the phonetic shape of the root. Frequently, the vowels of the past tense can be predicted on the basis of the root vowel in the present tense, but this is only really reliable in the case of rhizotonic roots. For example, while one can easily predict past tense *ráitʰ- and *rid- on the basis of present tense *réitʰidi, predicting *gám- and *gim- from present *gųskéti is somewhat more difficult.

For convenience of comparison, these broadly correspond to Sweet's Germanic strong verb classes.

The seven regular classes are as follows then:

Image

In classes 3, 4 and 5, there are two unstressed variants, one with a front vowel and one with a back. The front vowel variants are by far the most common, the back vowels only occurring after w and the reflexes of the PIE labiovelars. Given that the PIE labiovelars merged with the plain velars in PW, there was not inconsiderable hesitation between the two variants after velars, and the daughter languages have generalised one or the other.

On top of these seven regular classes there's also class 8, which is essentially the ashcan class for those which don't fit in any of the others. As a rule, they have regular class 6 stressed stems, but irregular unstressed stems.
These are generally forms which in PIE had initial laryngeals or semivowels. For example, *wék- 'to speak' has the regular stressed past root *wák-, but an unstressed *waug-, from *wewkʷ-.

A slight complication is occasioned by the effects of not-quite-Verner's Law. Where the final consonant of the root is a voiceless or aspirated stop in the stressed form, it becomes the corresponding voiceless unaspirated stop in the unstressed root. This can be seen in 3rd person singular *ráitʰa and 3rd plural *ridą́de.

The personal desinences then. Recall that the passive is not synthetically marked in the past tense, so only one verb, the class 1 *réitʰ- will be given. Witness that the stressed stem is used for the singular indicative, while the unstressed stem is used in the plural indicative and throughout the subjunctive.

Image

From a diachronic point of view, these present few difficulties, being obviously derived in the main from PIE's perfect desinences. However, the 2sg and 3pl warrant special mention. The latter has been imported from the weak verb class, the desinences of which are derived from the PIE aorist with the perfect desinences added (q.v. the Celtic imperfect inflection), thus *-ą́de is from a Post-PIE *-ṇ́t-ēr. A similar origin is suggested from the 2sg, although it is worth noting that the expected outcome of PIE *(re)róydʰ-th₂e would be **ráista: it is not impossible that the strong 2sg ending has been extracted from a mis-segmentation of 2sgs with stems ending in dentals. We shall see this again when we examine the irregular verbs later.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: Wenetic Scratchpad- NP: Strong verbs pt2, the past tense

Post by hwhatting »

So no traces of Aorist or Imperfect? This part indeed looks very Germanic.

User avatar
Dewrad
Sanno
Sanno
Posts: 1040
Joined: Wed Dec 04, 2002 9:02 pm

Re: Wenetic Scratchpad- NP: Strong verbs pt2, the past tense

Post by Dewrad »

hwhatting wrote:So no traces of Aorist or Imperfect? This part indeed looks very Germanic.
If I'm honest, I guess I'd say that I've reconciled the Germanic and Balto-Slavic verbal systems by using the Germanic system for strong verbs and the Balto-Slavic for weak verbs.
Some useful Dravian links: Grammar - Lexicon - Ask a Dravian
Salmoneus wrote:(NB Dewrad is behaving like an adult - a petty, sarcastic and uncharitable adult, admittedly, but none the less note the infinitely higher quality of flame)

Post Reply