And folks... time for Proto-Deithas

Museum for the best conlanging and conworldery threads. Ask mods to move threads here.
Post Reply
Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

And folks... time for Proto-Deithas

Post by Neek »

[Update] 17 August. Good news everyone! Voices and verbs have received an overhaul. Please contribute with some thoughtful feedback!

When creating a group of languages, it's never sometimes good to focus too much on a proto-language. This was how my gem, Hanawenzo, was made--with an idea of a proto-language, but not one in particular. This resulted in some rather unique problems, such a constant dissatisfaction with the verbal system. It also create the problem of word building: which words are native, which ones are borrowed?

While I had a sketch of a protolang, it never was much good. So instead, after toiling, I have worked on a new constructed language, the new copy of proto-Deithas.

Phonology and orthography

The consonant phonology contains the following sets:
Nasal: [m n ŋ] m n ŋ
Pre-nasal stops: [mp nt ŋk] mp nt ŋk
Stops: [p t ʈ c k ʡ] p t ṭ j k ħ
Voiced stops~Voiced Fricatives: [b~β d~ð g~ɣ] b d g
Voiceless fricatives: [ɸ s ʃ ʂ x ʜ] f s z ṣ x h
Trills: [ʙ r я] v r ḫ
Approximants: [ɻ j w] ṛ y w
Laterals: [l ʟ] l ł

High vowel: [ɨ] i
Mid vowel: [ə] e
Low vowel: [a] a

The language contains a primary series of bilabial, dental, and velar sounds. It contains a secondary series within the retroflex, palatal, and epiglottal series, and a trace within the postalveolar series. Its vowel system is a three-point vertical series, which means that height is only distinguished; there seems to be phonemic distinction between roundedness, nasality, or length. These factors seem to be a property of the daughter languages.

Prosody is based on a stress-accent, which is grammatical and phonemic. By that, certain roots are distinguished solely on the placement of the accent, and so are grammatical forms. Generally, words are stressed on the second-to-last accent in their least marked form.

The sonority scale is built by rating the following series (lowest to highest):
1 Voiceless plosives [p t ʈ c k ʡ]
2 Voiced Plosives~fricatives [b~β d~ð g~ɣ]
3 Voiceless fricatives [ɸ s ʃ ʂ x ʜ]
4 Prenasalized stops [mp nt ŋk]
5 Nasals [m n ŋ]
6 Lateral approximants [l ł]
7 Trills [v r ĝ]
8 Approximants [ɻ j w]
9 Vowels: [a ə ɨ]

Phonotactics are built upon the sonority scale to build the syllables:

Code: Select all

_Onset_
{
	{
		{fricatives} AND
		{
			{voiceless plosives} OR
			{voiceled plosives~fricatives}
		} AND {
			{approximants} OR
			{laterals} OR
			{trills}
		}
	} OR {
		{pre-nasalized stops} OR
		{nasals}
	}
}

_Nucleus_
{vowel}

_Rime_
{
	{prenasalized stops} OR
	{nasals} OR
	{
		{
			{approximants} OR
			{fricatives} OR
			{laterals} OR
			{trills}
		} AND {
			{voiceless plosives} OR
			{voiced plosives~fricatives}
		}
	}
}
A few rules to this set up, however:
* Any element, other than the nucleus, or sub-element, may be zero. This has the possibility of allowing a syllable to be just a vowel.
* No syllable can begin and end with a pre-nasalized stop.
* Epiglottal sounds are errant, and can oftentimes occupy invalid positions. The reasons for this are generally unknown.

There is some evidence that nasals and approximants (including laterals) can be syllabified, which can appropriately modify how a syllable is made--however, these instances are rare or extremely typed in a singular instance, and should not be taken as the rule.

In the case where a lateral, approximant, or nasal is syllabified regularly, a notation is used:
* A syllabified [w], [y] is always written, in this transcription, as u, i.
* In all other instances, it is written with an undercircle. As the board won't let me use combining characters with consonants, this leads to some trouble. For the purpose of the board, I've left them unmarked. Where necessary, I will specificy if the consonant is vocalic.

Basic morphology

Proto-Deithas' words are built from roots, stems, and affixes:
* A root is a bare word, which contains a meaning. A root has three possible functions: a. Verbal, b. Nominal, c. Verbal and nominal.
* A stem is used to derive the root into a meaning or function, generally through a classifier or productive derivation.
* Affixes exist as prefixes (such as augments or reduplication), infixes (such as the tau, sigma, or khi infixes, or the pre-nasalized prospective forms of verbs), or suffixes (such as numeric suffixes on nouns).

Nominal morphology

Nouns are any word which are formed from a nominal or nominal-verbal root, or derived from verb through a derivational process. Nouns are defined by function: They can stand in the subject and object position of a verb, become a constituent argument in a noun phrase through the use of a particle; they are assigned into a class, either animate or inanimate, which tends to be natural; animate nouns express one of two cases, either nominative or oblique, which is defined through stress (inanimate nouns only have the nominative case); and lastly, nouns can receive a numerical suffix that confers numerical marking, but has an indefinite impact.

In the conjectured pre-Proto-Deithas, the case marking formed as a result of the rules governing accent: The rising pitch could occupy the ultimate or penultimate syllable. The oblique was formed with a simple suffix, generally assumed to be -e:
Nom: **kʷóy-ùn-tʸ, one whose occupation requires both hands, slave, servant; Obl. **kʷɔi-ún-tʸè. When the vowel system collapsed along with the pitch-accent, the rising pitch became the primary stress, and the oblique marker was eroded away. This led to the shift in the accent:
Nom: kwéyunṭ, slave, servant; Obl. kweyúnṭ.

This situation, however, only applied to nouns of more than one syllable. In the instance of monosyllables, there appears to be at most six ways, four of which being exceedingly common, two of which being extremely rare, in the formation of the oblique. The four most common ways all utilize the same principle: Adding an additional syllable, and through analogy, stressing that syllable:

- By adding a classifier suffix to a noun; this applies to nouns that do not have a classifier suffix already: Nom: *dléyb, tail, Obl. *dleybwá (*-wa is a classifier implying an extension of a body. It tends to occupy limbs or body parts that extend outward, arms, legs, fingers, etc.)
- By applying a diminutive to a noun; this applies to nouns where the diminutive will not cause a shift in meaning: Nom: *łéw, eye, Obl. łewéy.
- By applying an augmentative to a noun; this applies to nouns where the augmentative will not cause a shift in meaning: *préḫ, lizard, Obl. preḫmáw.
- An epenthic vowel (which seems to be a remnant of the old case marker), Nom. *káḫyṭ, god, goddess, *kaḫyṭé.

Rare, irregular formation:
- Reduction of the vowel and syllabification of an inherent consonant. This only applies if the consonant can be syllabific, that is, either *m, n, ŋ, ṛ, l, or ł: Nom. *ṭáŋk, bear, Obl. *ṭŋˊk. This rare formation appears mostly in borrowings.
- In some instances, the oblique case suffix in pre-Proto-Deithas, when dropped, caused a lengthening of vowels. Long vowels, at the same time, were breaking apart into diphthongs, with one element becoming a desyllabic version of itself: **drʸá-e > **drʸâ > **drʸaə > *dṛáh, with the nominative in *dṛá. This formation typically occurs in open monosyllables, but has low frequency in closed monosyllables.

Nouns generally receive a suffix for number. These numerical suffixes possess a strong indefinite connotation, so are better refered to as indefinite articles. Their absence, however, implies a strong definite connotation. In animate nouns, there are three such articles: Singular, dual, and plural. In inanimate nouns, the singular is formed with an indefinite article, but the plural is formed through a much different means.

There are two types of indefinite articles, depending on environment. These environments are open syllables and closed syllables.

Animate nouns:
* Singular:
** Open: *-ḫz, *adáḫṭṛa-ḫz, a cougar, one cougar.
** Closed: *-eḫz, *ṭáŋk-eḫz, a bear, one bear.
* Dual:
** Open: *-wn, *adáḫṭṛa-wn, a pair of cougar, two cougars.
** Closed: *-un, *ṭáŋk-un, a pair of bears, two bears.
* Plural:
** Open: *-mh, *adáḫṭṛa-mh, some cougars, a many number of cougars, cougars.
** Closed: *-mh (syllabified m): *ṭáŋk-mh, some bears, a many number of bears, bears.

One would expect that the number markers would alleviate the formation of obliques in monosyllables. This tends to not be the case; the oblique in monosyllables is still formed as previously, plus the number marking.

Inanimate nouns:
* Singular:
** Open: *-ḫn, kálṭa-ḫn, a city, one city.
** Closed: *-ḫn (syllabified n): *ṛéwb-ḫn, a horn, one horn.

There tends to be a frequency that the classifier is lost when the indefinite is added. It appears that thematically, this suffix used to be a classifier: *méḫwa, breast > *méḫ-ḫn (syllabified n), a breast, one breast.

The plural of such nouns are formed irregularly, yet still retain an indefinite impact through analogy with animate nouns:
* If the noun possesses a classifier, that classifier is reduplicated in full: *dwádag, egg (*-ag is a loose classifier implying domestication) > *dwadágag, some eggs, a clutch of eggs.
* Nouns not derived from classifiers may receive one, then reduplicate it *át, digit, finger > *atwáwa, some fingers.
* In other instances, the least sonorous consonant and nucleus vowel in the root are reduplicated: *bséŋk, gift > *bebséŋk, some gifts.

Some animate nouns tend not to form their plurals with articles. They tend to follow the inanimate rules for forming a plural, plus one additional method. The reasoning for this unknown.
* The suffix *-ey (which is not tied to the diminutive) is used: *súḫt land > *súḫtey, some lands, lands.

Adjectival morphology

Adjectives are formed from any type of root, but typically nominal. Their function is a transparent and attributive from nouns. They do not possess any true morphological distinction from nouns. Adjectives, when modifying a noun, do not indefinite articles, or show even a nominative-oblique distinction. Adjectives can occupy the same position as a noun, in a substantive, rather than attributive, form. In this instance, it is identical to a noun.

Adjectives generally precede the noun they modify: *mpít kléppeḫn, a new stone.

There is a comprehensive system for creating comparatives. Comparison of adjectives works on one axis: Positive, comparative, and superlative. There exists three separate ways of forming the superlative, though it appears there is no thematic relationship to the adjective and the morphological process. That is, an adjective may express any of three superlatives.
Comparative: -int, -nt, *jábasint, fuller.
Superlative 1: Reduplication of least sonorous consonant + root vowel, jajábas, most full.
Superlative 2: Same superlative 2, with -ant, -nt, jajábasant, most full.
Superlative 3: -isk., jábasisk, most full.

Pronominal morphology

Pronouns are a closed list of words that bear an anaphoric relationship to a stated or understood antecedent. Morphologically, pronouns act as nouns but differ slightly on two factors:
1). Pronouns do not take indefinite articles; their dual and plural forms are reminiscent of the inanimate singular, and animate dual and plural indefinite articles.
2). The obliques are formed as per monosyllables, but show no thematic correlation with nominal lexemes.

The root forms are common in use, even if the number is known and emphasis is not required:
* 1st - *ehm
* 2nd - *yat
* 3rd - *dwe, *sám, *ḫém.

The oblique is formed differently for each pronoun:
* 1st - *ehmé
* 2nd - *yaht
* 3rd - *dú, *smˊ, *ḫmˊ.

In 3rd person pronouns, three are given. The first one is given as the most common, and the latter two have a proximate/obviative distinction, respectively. The distinction between animate and inanimate of these two lie in function, context, and number. In terms of function, an inanimate noun will not express an oblique distinction. One can deduce the animacy by context of the verb, or through anapohra, and finally, the inanimate form does not express number.

The number markings, as stated, are reminiscent of the indefinite articles:
Singular: *-ḫn, the syllabific of the n depends on the environment.
Dual: *-un (or in the case of third person nominative/1st and 3rd oblique, *-wn.)
Plural: *-i or -y.

The similarity of these suffixes is attributed to their formation:
* Singular *-ḫn in pronuons is formed from pre-Proto-Deithas **-ēn, a variant form of **ē, one. The inanimate singular indefinite article, however, is formed from **-ēne, a now defunct classifier.
* The dual are remarkably similar, because they are formed by suffixing pPD **ūn, two, to the word.
* *-ey, and this variant, *-i (Possibly from -ē) seems to have been used as a collective suffix, and has gained a meaning of plurality.

The most common 3rd person pronoun, *dwe, is possibly derived from an older interrogative adjective, the interrogative meaning having been eroded away. This is realized by the fact that the interrogative pronoun is derived from an extension of this root and the root for interrogative adverbs. The interrogative adjective/pronoun is dwéne. Despite having an always indefinite meaning, this noun is never receives an indefinite article, nor is it marked for number.

There is a combination of the root + aḫ to express possesison. The second form listed is a reduced form, and it is never stressed. The West-Deithas languages utilized the reduced forms as nominative prefixes, but North-Deithas languages tend to keep them separate.

1st - *éhmaḫ, maḫ
2nd - *yáhtaḫ, yaḫ
3rd - *dwáḫ, duḫ
Whose/of which - *dúnaḫ, ntaḫ

There also exists anaphoric adjectives, which can therefore be used as pronouns:
Any - *mé
Either (of two) - *ámp
Every - tyén
No one - *kí
Other (of two) - *e-slént
One -
Sole - *ṣké
Some - *pré
Such - *ha
Which (of two) - *áŋ

When used as adjectives, as previously stated, they have no special forms. When used as pronouns, they can receive number markings as per pronouns (rather than indefinite articles). They are noted for forming their oblique/plural with the classifier -ṭ (which is the same as the derivational suffix, -ṭ, which suggests agency).

Adverbs and nominal et cetera

Proto-Deithas possesses a two axis correlative system (this, that), and expresses them in location of (here, there), movement towards (hither, thither), movement from (hence, thence). The proximate/obviative distinction is releative to topicality. That is, "here" is relative to, generally, the subject. If unspecified, then it tends to be that of the speaker.

Proximate
* This, now - *sám
* Here - *sm-íy-ṭ
* Hither - *sm-íy-i
* Hence - *sm-íy-aḫ
* Now, at this time - *sán-tr (syllabic r)
Obviative
* That, then - *ḫém
* There, at that time - *ḫm-iy-ṭ
* Thither - *ḫm-íy-i
* Thence - *ḫm-­íy-aḫ
* Then, at that time - ḫen-tr
Interrogative:
* What, when? - *dwé
* Where? - *dw-íy-ṭ
* Towards where? - dw-íy-i
* Whence?, from where? - dw-íy-aḫ
* At what time? dwé-tr

In terms of sám, ḫem, and dwé, which share both the temporal and locative adverbial meanings, distinction is refined with the use of both syntax and qualifiers. Temporal adverbs tend to precede the verb, while locative adverbs tend to follow it. When they are used for relative adjectives or prornouns, they are typically fronted, which causes some confusion. The phrase ap sám, ap ḫem, and ap dwé (thus this, thus that, thus who?, thus what?) are prefered.

Nouns are not just modified by adjectives--particles can also provide relationships to the clause or to loosely to another word. Such particles are reduced verbs or verbs in their bare roots. These particles generally precede the constituent noun phrase they are modifying. Unlike a more familiar class such as prepositions, these particles operate on a clausal level--they either precede the verb they modify, or the constituent phrase, however modifies the phrase. This is similar to Homeric Greek prepositions, which contained a more adverbial meaning.
*ṣew (*ṣéw-tey, to follow), alongside, behind.
*diw (*díw-tey, to sit, to stay), at, on, in
*ahw (*hemmáḫ-tey, to walk), from
*iṛd (*iṛd-tey, to use, to do, also -iṛd, tool classifier), using, possessing, having

There is, however, a particle that provides a relationship between two constituent nouns or noun phrases: aḫ. This particle is either understood as a possessive prefix, or a genitive suffix. The possessing noun is put into the oblique case, while the possessed noun is in either the nominative or oblique case, dependent on whether it is in the subject or object position of the verb.
*eslenṭ aḫ dáḫag, another's dog.

In some instances, this particle can modify the possessing noun, rather than the possessed:
*aḫ eslentéḫn dáḫag, someone's dog.

The confusion arises in dialectalism; in some dialects, it is a mark of being possessed (such as West-Deithas languages, which utilize it as a prefix). North-Deithas languages instead utilize it as a suffix on the possessing noun, forming the genitive case. South-Central Deithas and Isolate languages have no distinction: In Azwrd, it is a prefix to the possessing noun. In the Yeyet-J`an group, it is marked on both for agreement.
Last edited by Neek on Wed Aug 18, 2010 1:03 am, edited 8 times in total.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

Verbal Morphology

Verbs are any words that are formed from a verbal or nominal-verbal root, or derived from a noun through a derivational process. Verbs are defined by function: They are any words that accept an Aspect-Mood suffix for finite meanings; possibly accept an agreement marker; alter valency in set patterns; are modified by adverbs; can be chained with other verbs to alter meaning.

The verb generally agrees with the subject, as expressed generally by a prefix. This form of agreement is irregular, and is not present in all verbs. There are three systems at work: Primary secondary, and the intransitive. More will be given about intransitive person formations later.


There are three moods:
* Indicative - Stated in a primary clause; factually valid; or is garnered from a direct source.
* Subjunctive - Stated in a secondary clause; factually dubious; or is garnered from indirect sources.
* Hortative - A command given. A statement made that has a force of impact.

There are three aspects:
* Imperfective - An action described as a shape, with or without a relevance to its impact.
* Perfective - An action described as a point, without a relevance to its impact.
* Retrospective - An action described as a point, with relevance to its impact.

The primary person prefixes are used in the imperfective and perfective forms. Primary prefixes agree with the first, second, and third person in the singular, with one prefix used for a plural subject; the 1st and 2nd person prefixes are merged, and the third differentiates between definite and indefinite objects:
Singular:
1st/2nd persons: me-
3rd person definite: ḫa-
Plural: ḫi-.
3rd person indefinite singular and plural: ø- (Null)

The secondary endings, generally used in the retrospective, are much more robust, but also lends itself to corruption through sandhi:
Singular:
1st person: m-
2nd person: r-
3rd person definite: a
3rd person indefinite: ø- (null)
Plural:
1st person: im-
2nd person: ir-
3rd person definite: ḫay-
3rd person indefinite: i-

Especially in the singular of the secondary series, a vast array of alterations can happen. The following are the most common forms:
* m{p,b,t,d,k,g} > mp, nt, ŋk; *m-díw-tey > ntíw-tey.
* rt > ṭ; *r-tzéw-tey > ṭzéw-tey.
* im-, ir-, and y- tend to force t > ṭ and k > j with the loss of the i. i-krék-tey > jrék-tey, mi-tzéw-tey > mṭzéw-tey.

This sandhi process does not seem to be productive, so borrowings or new innovations do not tend to generate these situations.

The aspect and mood are combined to form a singular suffix, which are unique between the imperfective and perfective; the imperfective and retrospective tend to share the same endings, but their stem is different.

The imperfective form is the least marked form of the verb. That is, the bare stem forms the stem for the imperfective mood. The other moods have different formative tactics:
Perfective
* The perfective stem is generally the same as the imperfective.
* Nasal-root stems tend to convert the initial nasal to a pre-nasalized plosive in the perfective.
* Some verbs contain a an infix found only the imperfective. This is lost in the perfective.
Retrospective
* The retrospective generally is formed by reduplicating the stem with the least-sonorous consonant plus the root vowel.
* Nasal-root stems tend to share the same root in the retrospective as the imperfective.
* Some verbs do not differentiate the retrospective from the imperfective.

This seems to be a learned system; that is, there is no thematic context to determine how a verb produces its stems, and therefore is memorized.

The active voice utilizes a special set of endings, that at time agreed with a third person object. This distinction was lost, however, and has merged with the aspect/mood endings. The active endings were the third person object endings (there does not seem to be a numerical consideration in object endings, or had there been one, it has been eroded away). The detransitive voice, however, utilizes the archaic object endings in their fullest in agreeance with the subject. This perculiarity has led some analysts to believe that the pre-Proto-Deithas was at least a split-ergative language.

The endings of the active voice in the imperfective/retrospective I are formed with the suffix -t, and the vowel following (if present) represents the mood: Indicative -tey, Subjunctive -tay, and Horative -ṭ (from older -ty). This is by no means the only means of the forming the mood-aspect markers, but where the root suffix alters, the endings do not: -yey, -yay, -yi exists based on the second person, and an incomplete paradigm based on a presumable separate root for the 1st person pronoun exists: -pey, -pay, with no hortative. These alternate forms, however, are exceedingly rare.

The perfective endings are a bit more difficult to trace down. They were formed with a root suffix separate from the object marker, -ṣ. It has been argued that this suffix root has merged with the personal marker (possibly second person singular, yielding **rs). The aspect markers, however, are far more alien: Indicative -ṣew, -ṣlu (unknown presence of -l-, perhaps from previous -o-), and a naked -ṣ for the hortative. There doesn't seem to be any alteration as in the imperfective with separate person markings left over as the basis; this does not mean to imply that such endings could not possibly exist, but that they were so much in disuse from analogy from this form, that analogy has left no evidence. Even in verbs that possess an irregular imperfective/retrospective set of endings, their perfective is the same as others.

The retrospective II endings are based on the perfective, sans the mood root suffix.

A complete paradigm of the most common endings are as follows:

Code: Select all

                  Indicative     Subjunctive     Hortative
Imperfective      -tey          -tay             -ṭ
Perfective        -ṣew          -ṣlu             -ṣ
Retrospective I   -tey          -tay             -ṭ
Retrospective II  -u            -lu              -i
As stated before, the imperfective/retrospective I endings were based on the 3rd person singular object ending that were left intact in the aspect-mood endings. In the detransitive voice, there exists a full system used to agree with the subject. Also as previously stated, there is no numerical distinction.

The first person is formed with *e-, which is desyllabifed into *h, yielding indicative -hey, but the subjunctive is -ela if the verb ends in a consonant, but -hla if it ends in a vowel. There appears to be no hortative for this form. The perfective of this is formed with -s- plus the endings as listed: -shey, -sela.

The second person is formed with *r-: Indicative -rey, Subjunctive -ralz (in analogy to third person definite), and the second person hortative is a syllabic -ṛ. These endings change in the perfective: -ṣey, -ṣalz, with no hortative present.

The third person definite is formed with *ṭ-, possibly the remnants of the plural being used in place of the singular that was left in the imperfective/retrospective endings: Indicative -ṭeyz (reinforced with the dummy object -z), Subjunctive -ṭalz; the hortative endings seems to be the same as the active hortative. The perfective endings are formed with -s-: -sṭeyz, sṭalz. There is again, no hortative.

The third person indefinite is rather straightforward, lacking a personal suffix as the previous, but utilizing the dummy object *-z attached to the standard imperfective endings: Indcative -eyz, Subjunctive -ayz, Hortative -iz. The perfective forms are the same as the definite; the reasoning for the lack of this distinction here is unknown.

Thus, the full endings are as follows:

Code: Select all

              Indicative     Subjunctive     Hortative
Imperfective/Retrospective
   1st        -hey           -ela            ***
   2nd        -rey           -ralz           -ṛ
   3rd-def    -ṭeyz          -ṭalz           -ṭ
   3rd-indef  -eyz           -ayz            -iz
Perfective
   1st        -shey          -sela           ***
   2nd        -ṣey           -ṣalz           ***
   3rd        -sṭeyz         -sṭalz          ***
Last edited by Neek on Tue Aug 17, 2010 12:13 am, edited 2 times in total.

User avatar
roninbodhisattva
Avisaru
Avisaru
Posts: 568
Joined: Sun Mar 05, 2006 11:50 pm
Location: California

Post by roninbodhisattva »

Neek wrote:An explanation of the uses and formation of the voices, basic syntax.
Ah yes, this is exactly what I was going to ask for.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

Valency and Voice Marking

As we've already seen Proto-Deithas inflectional forms and have an understanding of how verbs are formed, we can procede forward on a new distinction, that of valency. Proto-Deithas has a morphosyntactic alignment of Nominative-Accusative. This means that the nominative case expresses the core arguments of intransitive verb and the agent of a transitive verb, with the object as the oblique case distinct. Though inanimate nouns only appear in the nominative case, the position of the argument has a morphosyntactic factor.

In terms of Proto-Deithas, intransitive verbs are monovalent and transitive verbs are at least divalent. Verbs are capable, however, of decreasing or increasing the valency of a verb through a morphological process which, for convenience, is called voice. There are a total of 8 voices: Active, Avalent, Benefactive, Causative, Detransitive, Malefactive, Middle, and Reflexive, though some analysts consider the specialized functions of certain voices independent, therefore increasing the number.

It should be noted that these voices can be layered. Where a voice will increase the valency, a voice can be applied to the verb to decrease it at the same time. They are therefore not true voices, but are called so as they affect the overall valency of the given verb.

Active voice

We've already discussed the formation and existence of the active voice. It is the standard voice of intransitive and transitive verbs.

Detransitive

The detransitive voice is the second most common voice, where the verb loses its object argument, whether it is unimportant to express or omitted--that is, there is no implicit object to be expected. This voice therefore reduces the total number of arguments of a polyvalent verb down by one: Transitive becomes intransitive.

The detransitive uses a separate form of agreement than the active voice; rather than prefixing the subject, it has that suffix embedded in the aspect-mood marker. This marker receives the stress:
*ḫa-gwím-tey dwe maḫ manwṛˊh, he marries my sister > *gwim-ṭéyz dwe, he marries, he gets married.

Detransitizing may cause a lexical shift:
*me-xíw-tey ehm yáht, I hear you > *xiw-héy ehm, I listening, I am listening.

The detransitive has one function abscribed to it that is beyond the typical scope of such a voice: The passive voice. In the passive voice, the subject is dropped, and the object is raised to the subject. The passive is sometimes considered a separate voice, as demonstrated by its formation. The 3rd person forms of the detransitive are used as the primary marker for the passive, and do not change to agree with the subject. Instead, the personal prefixes are used to agree with the subject. The stress of the passive verb rests on the second-to-last syllable:
*ḫa-gwím-tey dwe maḫ manwṛˊh, he marries my sister > *ḫa-gwim-ṭéyz maḫ manwṛˊh, she is married.
*me-xíw-tey ehm yáht, I hear you > *me-xíw-ṭeyz yat, you are heard.

Benefactive and Malefactive

The benefactive and malefactive are two polyvalent voices: That is, unlike the detransitive, can be applied on any verb with any number of inherent valencies, to increase the valency of the verb. The two voices increase the valency by one, adding someone who in some respect benefits or is hindered by the action.

The benefactive creates an oblique argument that is benefited from the action. It is formed with the root suffix, -e-, placed before the aspect-mood marker, and generally receives the accent. The additional argument is placed after the object, the word order being Verb Subject Object Beneficiary.
*me-bséŋk-tey ehm ntyágṭ yáht, I give the food to you > *me-bseŋk-é-tey ehm ntyágṭ yáht renwṛˊh, I give you food for (your) son.

The malefactive, however, creatse the opposite effect: The argument is someone who negatively benefits from the action. It implies harm or depletion, that something is being done to hinder the argument. It is formed with the suffix -ane-:, with the stress on the final syllable:
*me-sdétir-tey ehm ntyágṭ, I cook the food > *me-sdetir-ane-téy ehm ntyágṭ yáht, I cook food *from you, or, I cook food in spite of your presence (i.e., there may not be enough for you--tough!).

The malefactive, plus the hortative, creates a prohibative. The verb is negated, and the beneficiary argument becomes the agent of prohibition:
*ḫa-gwím-tey dwe maḫ manwṛˊh, he marries my sister > *ḫa-gwim-ane-téy dwe maḫ manwṛˊh ehmé I forbid him from marrying my sister!

The detransitive voice may be layered onto these two voices. By doing so, one removes the object argument, not the beneficiary object.
*me-bséŋk-tey ehm ntyágṭ yáht, I give the food to you > *bseŋk-e-héy ehm yáht, I give for you.
*me-sdétir-tey ehm ntyágṭ, I cook the food > *sdetir-ane-héy ehm yáht, I cook in spite of you.

In some instances of the the detransitive voice, the beneficiary argument is lost. In the malefactive, this creates a negative meaning within the verb, where the subject is inherently hindered by the action, and alters the lexical meaning of the verb in question. This is known as the negative detransitive.
*nyag-ane-héy ehm, I cannot eat, I starve.

This meaning exists, through analogy, on intransitive verbs:
*me-yéŋ-tey ehm, I breath > *yeŋ-ane-héy ehm lénał, I drown (I cannot breath because water prohibits me from breathing).

The reflexive cannot be used along with the benefactive; the middle voice is used instead.

Causative

The causative voice is a polyvalent voice, like the benefactive or malefactive. Unlike the benefactive or malefactive, this can convert an intransitive verb into a transitive one. In intransitive verbs, the subject of is raised to the agent, and an object is created. The general meaning is, "to cause to x". This is formed with the the root suffix, -ja-, which receives the stress.
*kréṛd-tey pré, someone is dead > *ḫa-kreṛd-já-tey dwe préṭ, he kills someone (causes to die).

The causative in transitive verbs works a little different. The subject is promoted to an agent of causation, while the initial object is promoted to the agent of the action, and a new object is created as the object of the verb. This creates a trivalent verb.
*ḫa-gwím-tey sam dwé, he marries her > *me-gwim-já-tey ehm smˊ dú, I cause him to marry her, I arrange him to marry her.

As can be seen, the causative voice often causes a lexical shift: Die > to cause to die, kill; marry > to arrange someone to marry someone; to drink > to serve a drink; to see > to cause to see, to make someone understand, teach.

The detransitive is also used with this voice. When used on a bivalent causative, there is no change syntactically, but there is a change in meaning:
*ḫa-kreṛd-já-tey dwe préṭ, he kills someone > *kreṛd-já-ṭeyz dwe préṭ he has someone killed.

In ditransitive verbs, the agent of causation and the object of the verb remain, but the agent of action is lost.
*me-len-mí-tey ehm méłłen I drink beer > *me-len-mi-já-tey ehm dwé méłłen, I serve him beer > *len-mi-já-hey ehm méłłen, I serve beer.

The newly created object cannot be dropped; instead, it can be replaced with pré, someone, something.

The reflexive can easily be applied to the causative, either direct or indirect:
*ḫa-kreṛd-já-tey dwe préṭ, he kills someone > *ḫa-kreṛd-já-tey-si dwe, he kills himself, he commits suicide.
OR > *kreṛd-já-ṭeyz-si dwe, he has himself killed, he sacrifices himself.

The middle can also be applied to causatives. In this instance, the middle root suffix -e- follows the causative root suffix, and desyllabifies into -h-.

Reflexive and middle

The reflexive and middle do not alter valency as greatly as the other voices. In the reflexive, the subject becomes the object as well; in intransitive verbs, this results in the subject being the same as the indirect object (if applicable--not all verbs can become reflexive), and in transitive verbs, this can retain or lose the object, depending on the meaning. The reflexive is formed with the suffix -si. The stress does not move in the verb.
*ḫa-káḫy-tey dwe préṭ, he blesses someone > *ḫa-káḫy-tey-si dwe, he blesses himself.
*ḫa-gráf-tey dwe préṭ, she nurtures someone > *ḫa-gráf-tey-si dwe, she nurtures herself, she rests.
*me-bseŋk-tey ehm seṛkág dú, someone gives me the chicken > *me-bséŋk-tey-si ehm seṛkág, I give myself the chicken, I take the chicken.

The middle is a bit more difficult of a voice to pin down. The middle is formed with the root suffix -e-, similar to the benefactive, but retains the final suffix -s.

The middle voice has three primary functions: Egoist, reciprocal, and autocausative.

Egoist middle is similar to the benefactive, however is reflexive in this function; the subject directly benefits from the action being performed. This meaning is much more difficult to translate into English.
*me-bseŋk-é-tey-s ehm seṛkág yáht, I give the chicken to you (I benefit from giving this to you).

The reciprocal middle is used to express that the action is performed against the subject, which is generally plural. Unlike the reflexive, this implies that the action is formed against each of the subject by each of the subjects.
cf. *ḫi-kreṛd-já-h-tey-s dwey, they kill each other, versus *ḫi-kreṛd-já-tey-si dwey, they kill themselves.

The middle, with the detransitive, forms an autocausative meaning. That is, the action becomes the experiencer--this only applies to verbs where an experiencer may be valid.
*krek-é-ṭeyz ħwáŋkey, the window breaks.

The two avalent voices: Avalent and locative

The avalent and locative are two voicse that lack any valency; that is, this form of reduction reduces the valency to zero; there is no object, and there is no subject. Indirect objects, etc., may still apply.

The avalent is a statement of general fact, that the action is being performed without regard to any argument. This is formed with the root suffix -ḫiṣ-, from the verb of the same name (which is also used to form the perfective and retrospective), and utilize the third person indefinite detransitive endings, instead of the typical endings found in verbs. This is generally used in weather verbs, but also in certain periphrastic constructions.
*len-ḫíṣ-eyz, it is raining.
*ħent-ḫíṣ-eyz, it is harvesting, there is harvesting.

The hortative is used as a general cohortative.
*len-mi-ḫíṣ-iz, Let's drink!

The locative, however, is used to verbalize locations: This is where X occurs, where "this" or "here", like any form of deixis, is relative to topicality. It is formed with the root suffix -yi-, which receives the stress. This suffix, however, causes a shift in in the endings, that are easily determined with sandhi: -eyz becomes -hiz, -ayz becomes -ḫiz. There is no locative hortative.
*ħent-yí-hiz, here is harvesting.

Both the avalent and locative are useful in compounding and in forming adverbs of manner from verbs. The locative is typically used as an adverb in verbs of motion.
ḫa-tzéw-tey kreṛd-ja-ḫíṣ-ey dwe smˊ, he catches him with murderous intent.
ḫa-hemmáḫ-tey len-yí-hiz dwe, he walks while it is raining.

As shown by the examples, the avelent and locative can be used in conjunction with the causative voice; no other voice is compatible with these voices.
Last edited by Neek on Wed Aug 18, 2010 1:02 am, edited 3 times in total.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

Bump. Syntax still on its way. Updated the orthography, handled a few issues in the noun section, created section on adjectives, correlatives and other stuff, and added in a few pronouns.

Questions/Comments? Feedback or questions would be appreciated.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

Seriously, ZBB? No comments? Quality 'lang here!

Also, the Lexicon is now available, Lexicon of Proto-Deithas @ Dick.

User avatar
Xephyr
Avisaru
Avisaru
Posts: 821
Joined: Sat May 03, 2003 3:04 pm

Post by Xephyr »

That there sure is one quality Lexicon you got there, Neek.
"It will not come by waiting for it. It will not be said, 'Here it is,' or 'There it is.' Rather, the Kingdom of the Father is spread out upon the earth, and men do not see it."
The Gospel of Thomas

User avatar
Morrígan
Avisaru
Avisaru
Posts: 396
Joined: Thu Sep 09, 2004 9:33 am
Location: Wizard Tower

Post by Morrígan »

Cool voice stuff. Kinda makes me re-evaluate Voice in Kuma-Koban, where I've assumed that Causatives are derivational, but I now realize they could very well be inflectional.

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

Post by Salmoneus »

Some remarks on the first post:

- I'm curious about the near-total regularity of POAs being spoiled in the unvoiced stops and fricatives. Have the additional POAs been gained, or have the other series merged them? Or is it the result of cluster-simplification (ie anything voiced is from a cluster itself, and there were only two elements per cluster, so that things could be voiced or at the new POA but not both - but then, what clusters would produce those POAs?)

- Why is the case marked by moving the accent? The obvious answer would be that the oblique marker was an affix that has been lost, but that was there long enough to move the accent from the antepenult. However, why then does the accent not move when adding a numeric suffix?

- Why do disyllabics stress the first syllable and not the last?

- When exactly are the numeric suffixes used - you seem to suggest it's not just about definiteness. Also: isn't that the wrong way round? When things are indefinite, it's more likely that you don't know the number, and thus couldn't specify it, whereas for a definite, you always know the number.

- What are these 'reduced forms' of the possessives - when are they used?

- 'hn' is clearly a suffix to make things anaphoric - so what are those root words? And why is it used to make words like 'any' and 'every' and so on, which aren't anaphoric?

- What does 'other' mean? Is it 'THE other' or 'AN other', as they're totally different meanings? Similarly: 'a sole' or 'the sole'?

- Given that there's no clear distinction between adjectives and nouns, why is there both nominal 'what' and adjectival 'which'? Couldn't they just use 'what' in an adjectival way? I'm assuming that 'which' is the interrogative adjective - or is it the descriptive relative pronoun? It would be helpful if you could give the meanings more fully than just an English word, which often have many different uses.

- What are your 'correlatives' correlating with? Correlatives are pairs of words. Are yours meant to be deictic adverbs? Or can they perhaps function as adjectives/nouns as well? If they're adverbs, do queries regarding them take a different interrogative method? [Eg, in English, spatial deictic adverbs answer 'where', 'whither' and 'whence' questions, but temporal deictic adverbs answer 'when' questions, and most other adverbs answer 'how' questions.]

- Why is 'now' grouped with the spatial deictics? Where are there not ablative and allative forms of it? You don't have to have them, obviously - English doesn't, for a start - but it's worth considering.

- Can any of these 'correlatives' introduce dependent clauses? If not, do they have any equivalents that can? Do these trigger different behaviours in the main clause? For instance, "now" can introduce clauses: "Now I am earning money, we can buy a house"; but "then", as you translate your obviate equivalent to your 'now', cannot - it must be replaced by 'when', and triggers the main verb to be subjunctive and perfect (if the meaning is not habitual): "When I was earning money, we could have bought a house". "Here" and "there", on the other hand, both need to be replaced by "where", but they can be used for clauses that are being contrasted: "here I'm a valued employee; there I'm surplus to requirements". How does all this work in Proto-Deithas?

- More generally, how does the Proto-Deithas deictic framework operate in these cases? What's the difference between proximate and obviate, and how does it interact with the positions through time of the speaker, listener, the protagonist of the story and any of the arguments of the verb, and are there ways to alter this? How far away is obviative, and where is it measured from?

- I'm unclear about your particles. They modify verbs, so they're adverbial, fine, just like in English... except, where does their object come from? Alongside WHAT, from WHERE, using WHAT? Are you trying to say that these are applicative particles that change the role of the object (like prepositional compounding in English)? Or are they valency-altering particles that introduce another object? Or are they deictic/anaphoric particles that define the relation of the verb to an unstated additional item, as they often are in English? If so, are they deictic, anaphoric or both, and how is their referant defined? Can they ever be used in non-referential ways to create phrasal verbs, as their English equivalents often can? ["I walked along", for example]

- Your possessive particle: you say it can be understood as either a prefix or a suffix: how? Which is it really? You write it as though it were neither. When you have nominal phrases, how does it cope with those - does it modify the phrase or the constituents? WHEN does that altered word order you mention occur?


-----

General comments: I think that sometimes you could do with a bit more clarity and precision in your explanations, and less reliance on English translations, as this can obscure the unique subtleties of the language.
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
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Post by Salmoneus »

On the verbs:

- 'person' is a type of 'agreement'

- it's a little odd that even with so little nominal case, there's still no verbal agreement. It's possible - it's not too dissimilar from English - but still a bit surprising.

- some more terminological confusion: you mean "perfect", not "perfective", and you mean "perfective" (probably) rather than "prospective" (different thing altogether).

- I don't understand "In most nouns, the perfective and imperfective forms are the same" - nouns have no aspect. Or at least, you've not mentioned that they have aspect. Typo for 'verb'? But THEN you've said that most verbs have DIFFERENT forms for the two...?

- I see no 'mood infix' mentioned again.

- Isn't that about twice as many voices as in any real language?

- Are the fused AM suffixes really both purely Cartesian and purely unique?

- How do other semantic aspects get coded? Eg, are habituals perfective or imperfective? Iteratives? How are inchoatives, etc, dealt with? Does the aspect system interact with pluraction?

- How are other moods conveyed?
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
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Post by Salmoneus »

- Why do you say there is a 'direct' case if it does not contrast with the nominative?

- 'there's a maximum of 11 voices a verb can express' - what? It can be in eleven voices at the same time?

- why do you call it the 'gnomic' when it has no gnomic element?

- Again, I'm struck by the extreme cartesian nature of the suffixing. For instance: given that an imperative implies a second person agent, what is the difference between saying "stand! (active)" and "stand! (gnomic)"?

- How is the 'locative' a voice? It seems like a short-hand way of predicating certain properties of locations. But in shorthanding it, surely you have to address issues of deixis? To where does the "here" refer? And can this only be used in exclamations?

- How often will people be saying "I command that you not have in the past continually eaten, because I command that there was no food!"?

- I don't know what the negative detransitive is, but it isn't a voice. It seems to be or include an aspect, a sort of post-terminative or 'depletive'.

- How do you deal with other negation? Non-depletive detransitives: "I didn't eat [although there was food]"? Or negative non-detransitives: "I didn't eat the pork [because there was none]."?

- The normal detransitive is an antipassive, isn't it? It's quite rare in the case of nom-acc languages, because it's not very useful. In this language, does the antipassive have semantic import, or is it purely a grammatical marker to indicate that no object need be expected?

- It's not clear to me: do benefactive arguments still require case or preposition marking? Likewise the patients of causatives?

- How are the passive and middle voices different in use?

- How are the reciprocal and egoist functions of the middle voice distinguished?

- How do you put a reflexive into another voice?

-
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
eodrakken
Lebom
Lebom
Posts: 225
Joined: Fri Sep 13, 2002 11:11 pm
Location: Green Mountains
Contact:

Post by eodrakken »

Salmoneus wrote:Again, I'm struck by the extreme cartesian nature of the suffixing. For instance: given that an imperative implies a second person agent, what is the difference between saying "stand! (active)" and "stand! (gnomic)"?
What does Cartesian mean in this context?

User avatar
ayyub
Sanci
Sanci
Posts: 44
Joined: Sun Oct 09, 2005 8:23 am
Location: USA

Post by ayyub »

eodrakken wrote:What does Cartesian mean in this context?
As in a Cartesian product, producing a full table without gaps.
Ulrike Meinhof wrote:The merger is between /8/ and /9/, merging into /8/. Seeing as they're just one number apart, that's not too strange.

User avatar
vec
Avisaru
Avisaru
Posts: 639
Joined: Tue Sep 16, 2003 10:42 am
Location: Reykjavík, Iceland
Contact:

Post by vec »

You say that roots are either verbal, nominal or both. Does this mean that a verb cannot be derived from a noun based on a nominal root?

When adjectives are declined, can they head NPs?

Can pronouns refer to VPs? If not, are there some other tools to do so? (As in: "I was happy at the time, which is a good thing".

The temporal correlatives—do they not have any kind of temporal anaphora besides past/present? How are until now, until then and from now on, from then on expressed?

Regarding the verbal system—cool stuff—I'm all about figuring out valency propery in languages. It seems to me to be one of the most important aspects of verbs, and yet, it's only recently come to the attention of mainstream linguists. Intransitive/transitive without anything else is of barely any use at all in most languages. I really love the idea of the avalent voices. What was your inspiration here?

Regarding your choice of names, I find detransitive a good choice, rather than anti-passive, which always confuses me. Again, where did you get this term from? I've never heard it before. However, negative detransitive is confusing to me. Is it actually negative or am I misunderstanding terribly? Also, your use of the noun want is quite Edwardian. I would love to have some more examples of the middle voice, its use is not obvious from your explanation.

Finally, I would like to second Salmoneus' (Salmonei?) first three comments regarding POAs, case marking (which does not seem to be synchronous with the way it works in verbs). I know you are making a proto-language, but this seems so novel, you kind of have to understand how that came about. Was it that the accent used to be fixed, then after the elimination of some affix, jumped? How can you make it so that the verbal system's use of the accent as a marker and the nominal system's use, look like they could come from the same source? Right now, they look a teensy bit random.

But a lovely start, I must say.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

First, I want to thank you, Sal, for the very good, in-depth criticism. I'll do my best to be as thorough in the reply.
Salmoneus wrote:Some remarks on the first post:

- I'm curious about the near-total regularity of POAs being spoiled in the unvoiced stops and fricatives. Have the additional POAs been gained, or have the other series merged them? Or is it the result of cluster-simplification (ie anything voiced is from a cluster itself, and there were only two elements per cluster, so that things could be voiced or at the new POA but not both - but then, what clusters would produce those POAs?
Were one to postulate a Pre-Proto-Deithas, the plosives would have been /p t k/, plain and simple. */kj/ yields rather frequently /c/ (there are no instances of /kj/ or /gj/, for instance, cf. *vék-ja > *véja). /c/ is also an allophone of /dj/ in unstressed syllables, *íṛd-ye > *íṛje. /c/ is also acquired in borrowings.

/ʈ/ has a bit more phonological uncertainty, though there are a few definite. Post-tonic, intervocalic */tj/ seems to generate them rather frequently, as did /rt/ (a quick sampling suggests /rt/ does not occur with any frequency, so it's safe to assume such). /ʈ/ does not seem to have come from borrowings.

The epiglottal series is a bit trickier. They have developed from complex consonant clusters, such as /kxʷ/ (which yielded /ʜ/, or /ʜʷ/), or developed from de-syllabific of open back vowels during the vowel collapse. In one of the more popular theories put forth, Pre-Proto-Deithas had a much more complex vowel system which also including length and complex phonation. Long vowels collapsed in patterns of V > VV_^ / _:, i.e., ā > aḫ; this process also occurs in unusual vowel diphthongs: *ae > ḫe. (It would be argued that stridentification of any vowel would seem unlikely, and multiple theories of Pre-Proto-Deithas vowel systems have been put up, but without any data backing it up, would be difficult to even guess near correctly.) This compensatory method seems highly irregular, and there might have been processes that de-lengthened a vowel before this change took place.

Voiced consonants were developed from a). intervocalic plosives (most common), b). initial complex clusters (second most common), c). raising of complex plosives in certain conditions (i.e., */pj[vowel][+front]/, /kʷ[vowel][+back]/ (I hope I'm describing that properly). When I have more data, I can describe more thoroughly the conditions that would bring about the development of these sounds.
- Why is the case marked by moving the accent? The obvious answer would be that the oblique marker was an affix that has been lost, but that was there long enough to move the accent from the antepenult. However, why then does the accent not move when adding a numeric suffix?

- Why do disyllabics stress the first syllable and not the last?

- When exactly are the numeric suffixes used - you seem to suggest it's not just about definiteness. Also: isn't that the wrong way round? When things are indefinite, it's more likely that you don't know the number, and thus couldn't specify it, whereas for a definite, you always know the number.
Accentuation as case marking came from length compensation plus a pitch-accent system. Pre-Proto-Deithas possessed a pitch-accent, whereas the pitch must rise up and fall, but no mora could not follow the falling pitch. Thus, **pápàt > **papátè > **papât for the oblique. When the vowel system collapsed into a vertical system, losing vowel length (save for irregular compensatory lengthening), and the accent system, the rising pitch became an accent, so the theoretical word **pápàt > *pábat, and **papât > *pabát for the nominative and oblique respectively. Monosyllabic words seem to have compensated differently. This would also explain why disyllabic nouns are stressed on the first syllable.

The underlying process that defines definitiveness and the conjunction of numerical suffixes are purely conextual, so it would seem. In nouns where number is explicit (*win dwád, two eggs), the number is entirely unnecessary. As you stated, the number markings aren't necessary if you know the number, so it wasn't (normally) stated. There is sufficient data to suggest, however, that the singular number marking, when used as an indefinite marking, implies the number isn't known. It was more than likely the same case in Proto-Deithas.
- What are these 'reduced forms' of the possessives - when are they used?

- 'hn' is clearly a suffix to make things anaphoric - so what are those root words? And why is it used to make words like 'any' and 'every' and so on, which aren't anaphoric?

- What does 'other' mean? Is it 'THE other' or 'AN other', as they're totally different meanings? Similarly: 'a sole' or 'the sole'?

- Given that there's no clear distinction between adjectives and nouns, why is there both nominal 'what' and adjectival 'which'? Couldn't they just use 'what' in an adjectival way? I'm assuming that 'which' is the interrogative adjective - or is it the descriptive relative pronoun? It would be helpful if you could give the meanings more fully than just an English word, which often have many different uses.
The reduced forms of possessives seem to be alternative forms. There might have been a difference in their usage, but Proto-Deithas's daughter languages seem to not note a difference (i.e., some branches prefer the reduced forms for their possessive pronouns, while others prefer the expanded.) The best guess that can be made is the reduced forms are contractions.

-hn, despite being used for inanimate nouns, is used as the primary number marking for pronouns. I should state that they are shown to explicitly show wich words use them (though it isn't necessary--I should only state that *dwéne that is like a noun in this respect.). I apologize for the confusion. The number marking that is used for personal pronouns is also used for other pronominal entities.

I apologize for the brevity, and will note that *áŋ, which, is used to describe one of a set, as opposed to *dwéne, which is an interrogative pronoun/adjective.
- What are your 'correlatives' correlating with? Correlatives are pairs of words. Are yours meant to be deictic adverbs? Or can they perhaps function as adjectives/nouns as well? If they're adverbs, do queries regarding them take a different interrogative method? [Eg, in English, spatial deictic adverbs answer 'where', 'whither' and 'whence' questions, but temporal deictic adverbs answer 'when' questions, and most other adverbs answer 'how' questions.]
The system itself produces adverbs, though spatial sám and ḫém can be used as adjectives and pronouns--I'll return through them and ensure that's explicit. There should be a set of queries, and I'm amazed I forgot them. Also, "correlatives" seems like a bad term. I'll be sure to change that. Thank you.
- Why is 'now' grouped with the spatial deictics? Where are there not ablative and allative forms of it? You don't have to have them, obviously - English doesn't, for a start - but it's worth considering.

- Can any of these 'correlatives' introduce dependent clauses? If not, do they have any equivalents that can? Do these trigger different behaviours in the main clause? For instance, "now" can introduce clauses: "Now I am earning money, we can buy a house"; but "then", as you translate your obviate equivalent to your 'now', cannot - it must be replaced by 'when', and triggers the main verb to be subjunctive and perfect (if the meaning is not habitual): "When I was earning money, we could have bought a house". "Here" and "there", on the other hand, both need to be replaced by "where", but they can be used for clauses that are being contrasted: "here I'm a valued employee; there I'm surplus to requirements". How does all this work in Proto-Deithas?
These adverbs, as I'll now call them instead of correlatives, are used to introduce dependent clauses. The distinction between spatial and temporal lies not within the morphology, but its syntactical placement. Temporal adverbs tend to precede the verb, while spatial adverbs tend to follow it, acting as an argument for the verb (in some instances, the adverb can be hypercorrected and placed into the oblique case.) The ablative and allative uses in a temporal sense seems invalid--you'd be better using a more specific temporal adverb. I'll be sure to include those.
- More generally, how does the Proto-Deithas deictic framework operate in these cases? What's the difference between proximate and obviate, and how does it interact with the positions through time of the speaker, listener, the protagonist of the story and any of the arguments of the verb, and are there ways to alter this? How far away is obviative, and where is it measured from?
The deictic pronouns are light on explanation--and I'll have to expand that. The proximate and obviate distinctions are relative to topicality; generally, the subject of the primary clause, or if none stated, such as in an avalent form of a verb, the speaker. The topicality of this can lead to confusion, and possessing the adverb helps reduce ambiguity. Proximate seems to imply close to the topic of the argument, generally within sight, while the obviative is out of scope of sight. In distance wise, the distinction is relative to the scope; if the proximate is used to describe a room, the obviative is anywhere outside the room; if it's used for the clearing in a forest, or the burrough of a city or a village proper, the the obviative is what's beyond the clearing in the forest, outside the burrough in the city, or the land outside which the village resides.
- I'm unclear about your particles. They modify verbs, so they're adverbial, fine, just like in English... except, where does their object come from? Alongside WHAT, from WHERE, using WHAT? Are you trying to say that these are applicative particles that change the role of the object (like prepositional compounding in English)? Or are they valency-altering particles that introduce another object? Or are they deictic/anaphoric particles that define the relation of the verb to an unstated additional item, as they often are in English? If so, are they deictic, anaphoric or both, and how is their referant defined? Can they ever be used in non-referential ways to create phrasal verbs, as their English equivalents often can? ["I walked along", for example]
Particles are rather obscure in that sense. Consider it an expansion of Homeric Greek's prepositions--they are truly adverbs, but can be used to modify the whole clause rather than individual arguments. They are used weakly and rarely in the addition of adding syntactical arguments, and more deictic/anaphoric particles that define the relationship of the verb to an unstated argument. I'll have to do some research before I can describe their restrictions.
- Your possessive particle: you say it can be understood as either a prefix or a suffix: how? Which is it really? You write it as though it were neither. When you have nominal phrases, how does it cope with those - does it modify the phrase or the constituents? WHEN does that altered word order you mention occur?

-----

General comments: I think that sometimes you could do with a bit more clarity and precision in your explanations, and less reliance on English translations, as this can obscure the unique subtleties of the language.
The possessive particle is simply there. Daughter languages treat it either way--some prefix it to the noun being possessed, some suffix it or prefix to the possessing noun. This development seems to have come from the ambiguity of its meaning--to the speakers, it seems there was confusion whether it modified the possessor or the possessed. The settling of this ambiguity is dialectal, and therefore key to the development of the daughter languages. None of the constructions I give are invalid, dependent on how the speaker believes the word to be.

I'll be more clear in my next write-up, and I'll work on attributing more comments later on. Because as it stands, my girlfriend has returned from work. Thank you again for the indepth write-up.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

Salmoneus wrote:On the verbs:

- 'person' is a type of 'agreement'
- it's a little odd that even with so little nominal case, there's still no verbal agreement. It's possible - it's not too dissimilar from English - but still a bit surprising.[/quote]

The former is a typo. To the latter, syntax governs syntactical argument moreso than the nominal case--therefore case isn't as important. Is there a correlation between verbal agreement and case distinction in nouns?
- some more terminological confusion: you mean "perfect", not "perfective", and you mean "perfective" (probably) rather than "prospective" (different thing altogether).

- I don't understand "In most nouns, the perfective and imperfective forms are the same" - nouns have no aspect. Or at least, you've not mentioned that they have aspect. Typo for 'verb'? But THEN you've said that most verbs have DIFFERENT forms for the two...?
I'm curious, moreso, how I screwed up the definitions. I double checked Rick Harrison's article on Verb Aspect, and amazed myself that I managed to screw it up. I'll go ahead and correct those (and as to not screw them up again, perfect will be labeled retrospective.)

The statement, "In most nouns...", I mean verbs. It is a typo.
- I see no 'mood infix' mentioned again.

- Isn't that about twice as many voices as in any real language?

- Are the fused AM suffixes really both purely Cartesian and purely unique?
The 'mood' infix that I mention is poorly described: The indicative, subjunctive, and imperative are respectively *-e-, *-a-, *-ø [+long]. The imperfective/perfect 1 system utilizes the suffix *-ti (yielding *-tey, *-tay, *-ṭi < **-tī.) The subjunctive is formed with the same vowels, but using the suffix **-ṣu: *-ṣew, *-ṣlu < **-rsou, *-ṣu < **-rsū.

I would be lying if I said the Aspect-Mood suffixes are purely unique; what's not described here are the passive endings, which I just sort of gloss over later in there. The existence of these endings are a bit of mystery as to their whole uniformity--there's no reason than there should only be one type of verbal ending. The current popular theory suggests that -tey is a personal ending that was re-analyzed as an aspect marker; the main problem with this theory is that there is little data to back it up, and that it attempts to justify Hanawenzo's usage of *-tu (3rd person singular ending, which traces back to PD *-tey).

Some verbs to exhibit percularities. While the perfect 2 endings are for a small class of words, generally those in the *-mi- stem, and some pre-nasalyzed stops, about 25% of those verbs have no imperfective form (i.e., **álb-tey, to know, is given improperly as the imperfective dictionary form, yet the verb only exists in the retrospective and perfect; it should be *álb-u.)

There may be smaller irregularities in the system as a whole outside of the perfect 2 with no imperfective, but those are simple irregularities, and without the proper data and diachronics, there's no telling where they came from.
- How do other semantic aspects get coded? Eg, are habituals perfective or imperfective? Iteratives? How are inchoatives, etc, dealt with? Does the aspect system interact with pluraction?

- How are other moods conveyed?
Semantic aspects get encoded either through context, adverbs, or periphrastically:

Adverbs are rather straight forward. Generally, such aspects are expresed temporally, so fall before the verb:
Habitual: ṭéŋk dyeḫ-ṣew-sí éhmeḫz, alwas talk-PERF-REFL I.nom-PLURAL, we have always talked.
Iterative: *tyén évay yí-tey éhmeḫz jargṣtyé, every day go-IMPERF I.nom-PLURAL shop.DIR, every day we go to the shop.

Periphrastics are a bit more complex, generally using one verb in an infinite sense (which hasn't been discussed. Verbs lacking a mood-aspect marker are placed are deverbalized, and can be treated as an adverb, adjective, or noun; context describes which occurs. When modifying another verb, such periphrastically, it acts as an adverb, going before the verb). The avalent gnomic can also be used in these situations. Generally, both are semantically link:
Inchoative: *pḫét ħálk-ṣew dwé,, sleep lie.down-PRFV he.sing, He lied down to sleep, he started to sleep.

I'm not sure what you mean by pluraction--you mean if multiple actions are described on different aspects, how do the aspects interact?
- Why do you say there is a 'direct' case if it does not contrast with the nominative?

- 'there's a maximum of 11 voices a verb can express' - what? It can be in eleven voices at the same time?

- why do you call it the 'gnomic' when it has no gnomic element?
The "direct" case isn't too dissimilar to the nominative form of animate nouns. Describing it as direct, rather than nominative, is meant to show that it has no other possible forms. It is a little misleading, and for simplicity, it will be refered to as the nominative.

What that means, "there's a maximum of 11 voices a verb can express', is that it can be conjugated in to one of 11 voices at most--I'll reword that. It's rather obvious that intransitive verbs cannot become passive nor detransitive. The avalent gnomic voice has a gnomic quality--it describes a condition that is true. It, much like in daughter languages, has a limited use, either as the object of a verb or in affirmations. Other daughter languages have lost the valent constraint, and it has become a tense.

The problem with describing them as voice is rather misleading--but there are few words in linguistics to describe a grammatical function that reduces or increases valency. The closest word that approximates it is "voice." Plus, valencizers sounded awesome, but entirely unprofessional ;). For the total number, I'd rather blame the analysts, who were overzealous on describing functions of voices (i.e., detransitive as a function of the passive form, the negative detransitive as a function of the malefactive form) as separate voices. You can slice them either way.

Where the analysts were liberal on the voices, they were far too conservative on the moods--I'll be sure to include that distinction.
- Again, I'm struck by the extreme cartesian nature of the suffixing. For instance: given that an imperative implies a second person agent, what is the difference between saying "stand! (active)" and "stand! (gnomic)"?

- How is the 'locative' a voice? It seems like a short-hand way of predicating certain properties of locations. But in shorthanding it, surely you have to address issues of deixis? To where does the "here" refer? And can this only be used in exclamations?

- How often will people be saying "I command that you not have in the past continually eaten, because I command that there was no food!"?
The cartesian nature isn't entirely true--especially not with the verb given, *díw-. I'll present a better verb that could potentially be used in those positions.

There are certain functions of the imperative that go beyond the scope of commands. They exist in dependent and subordinate clauses, and not just restricted to indirect speech. The imperative can have less force than the Indo-European imperative, though not always--in dependent clauses, it can have an impact slightly greater than an optative.

The locative is rather unique in its status, and I agree--it's hardly a voice (despite influencing the valency of a verb), but something else entirely. Its deixis is proximate to topicality, as an adverb would. While one might say, "There is drinking," the imperative could mean, "Let there be drinking!", without an indication of person or object. How this qualifies as a voice--well, I'm not going to justify it. I'm just going to include the disclaimer.

(In the development of Proto-Deithas, it might seem that the locative was a means to encode verbal data onto localized nouns.)
- I don't know what the negative detransitive is, but it isn't a voice. It seems to be or include an aspect, a sort of post-terminative or 'depletive'.

- How do you deal with other negation? Non-depletive detransitives: "I didn't eat [although there was food]"? Or negative non-detransitives: "I didn't eat the pork [because there was none]."?
I won't dispute the negative detransitive as a 'depletive'.

The other forms of negation are rather easy. *Kí can be used as an adverb: kí na-nyág-ṭeiz ehm, I had not eaten. The same can be said without a detransitive applied to the argument in question: na-nyág-tey ehm kí griḫtyím[/i], I had eaten no pork. This still contains the ambiguity of whether or not you have eaten. Such a distinction can further be made if necessary.
- The normal detransitive is an antipassive, isn't it? It's quite rare in the case of nom-acc languages, because it's not very useful. In this language, does the antipassive have semantic import, or is it purely a grammatical marker to indicate that no object need be expected?
The normal detransitive is an antipassive. While I agree it's rare and rather useless, it's implicit on transitive verbs without an explicit object. There are theorie that suggest that PPD was ergative-absolutive, but there's little evidence (aside from an antipassive) to suggest it. It does have certain uses, but those'll be covered later.
- It's not clear to me: do benefactive arguments still require case or preposition marking? Likewise the patients of causatives?
Neither do. Grammatical syntax still dictates it:
In a standard verb with two arguments, we see Verb Subject Object. In a causative verb, the initial object becomes the patient, and a new object is added at the end. Benefactive operates the same way:
Stanard: Verb Subject Object.
Causative: Verb Subject (Agent) Patient.
Benefactive: Verb Subject Object Beneficiary.

While it hasn't yet been described, the causative can undergo nearly any valency reduction available; that is, one may omit the agent from the verb with the passive. One may also omit the object with the detransitive. The reflexive cannot be put into another voice, as you ask at the end of your post.
- How are the passive and middle voices different in use?

- How are the reciprocal and egoist functions of the middle voice distinguished?

- How do you put a reflexive into another voice?
The passive simply removes the subject and promotes the object. The middle voice contains various contextual uses not defined by any function in the morphology, such as reciprocal, egoist, and unaccusative. It would be good to go into them all, which I will once I start verbal morphology (I'd rather keep one in suspense).

That concludes this article. I promixed my girlfriend I'd be in bed an hour ago, so I might want to make good with that promise.
Last edited by Neek on Wed Aug 04, 2010 10:11 am, edited 1 time in total.

User avatar
vec
Avisaru
Avisaru
Posts: 639
Joined: Tue Sep 16, 2003 10:42 am
Location: Reykjavík, Iceland
Contact:

Post by vec »

So I'll get my response later today? :)

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

vecfaranti wrote:You say that roots are either verbal, nominal or both. Does this mean that a verb cannot be derived from a noun based on a nominal root?
Not directly. A nominal-verbal root can be used as a noun without any derivation, a verbal root cannot be used as a noun without derivation, and a nominal root cannot be used as a verb without derivation. Luckily, we are not short of derivation (the most common verbalizer is *-mí-, which creates the process related to X.
When adjectives are declined, can they head NPs?
Yes. Adjectives can be declined when they're being used a nouns.
Can pronouns refer to VPs? If not, are there some other tools to do so? (As in: "I was happy at the time, which is a good thing".
Technically, pronouns can't really refer to VPs. Adverbs, however, can. When using an adverb to refer to a VP, the verb it's modifying doesn't need an argument:
*ḫéntr dedvéjtey ehm, ap ḫíṣṭéy dveka
there REDUP-be.pleased-PRF I.nom, thus to.be-IMPRF well.
At that time I was pleased, thus it was well.

(To be, being an existential verb, doesn't necessarily drop syntactical arguments. More than likely because it's suffixed onto verbs to express avalent gnomics.)
The temporal correlatives—do they not have any kind of temporal anaphora besides past/present? How are until now, until then and from now on, from then on expressed?
Such phrases would be compounds of the particles, or pure adverbial phrases:
*ṣusám, until now. (alongside this point)
*ahw-ṣusám (whether ahw contracts in compounds of compounds is unknown), from now on, from alongside this point.
*ṣuḫén, until then. (alongside that point)
*ahw-ṣuḫén, from then on, from alongside that point.

I'll be sure to include a few more.
Regarding the verbal system—cool stuff—I'm all about figuring out valency propery in languages. It seems to me to be one of the most important aspects of verbs, and yet, it's only recently come to the attention of mainstream linguists. Intransitive/transitive without anything else is of barely any use at all in most languages. I really love the idea of the avalent voices. What was your inspiration here?
The inspiration came from wanting to shy away from a wholly trans-intrans defining language or an ergative-absoluti. I wanted to give the language a few more voices than necessary--something beyond active, passive, and middle, and something unique in the scheme of things. It was a risk that language might become kitchen sinkish, which is hardly an easy risk to take.
Regarding your choice of names, I find detransitive a good choice, rather than anti-passive, which always confuses me. Again, where did you get this term from? I've never heard it before. However, negative detransitive is confusing to me. Is it actually negative or am I misunderstanding terribly? Also, your use of the noun want is quite Edwardian. I would love to have some more examples of the middle voice, its use is not obvious from your explanation.
Calling the detransitive the antipassive would have been misleading this language as an ergative language. The idea came from a discussion with vlad on #isharia. Vlad was explaining that Nahuatl marks unexpressed arguments with te-/tla- (personal and object dummy prefixes). The terminology, however, I chose for its descriptiveness--and at the time, I hadn't realized it was called detransitive.

The negative detransitive, as Salmoneus points out, is a form of 'depletive.' It detransitizes the verb, so the object is removed, but is replaced with an object, of which there is none. As it is a function of the malefactive, the subject of the verb is hindered from the lack of an object. Some of these meanings, therefore, possess a idiomatic meaning:
*nyág-tey to eat > *nyag-áne-tey, to starve.
*gi-gwím-tey, be have been married > gi-gwim-áne-tey, to be divorced.

And here are some examples of the middle voice in action:

Autocausative middle: krék-tey ehm ħwáŋkey, break-IMPRF I window, I broke the window > krék-tey-si ħwáŋkey, the window broke. (As opposed to the passive, which would mean, the window was broken). This is useful in verbs where the object is generally an experiencer.

Reciprocal middle: bséŋk-tey-si dwe-ḫz dṛá, give.IMPF.MID he-PLURAL flower, they gave each other flowers. This contrasts with the reflexive with reciprocality. The reflexive would imply, they gave themselves flower, that is, the flowers were given back to each of them from them (so one might assume it would mean, they each took flowers), while the middle would imply that they gave flowers to each other. This is generally noticed because the subject is plural.

Egoist middle: bséŋk-ṣew-si ehm dṛá yát, ḫem pḫét-ṣew-si ehm-ḫz, give-IMPF.SUBJ-MID I you, then sleep-IMPRF.SUBJ-MID I.plural, if I give you flowers, then we might sleep together. (The idiomatic meaning of the reciprocal middle of "sleep" is rather obvious, however, the subjunctive of the initial clause indicates there might be a condition. The adverb, ḫem, suggests a temporal chain of events not revealed. Outside of that, there is a benefit of me giving flowers--that I might sleep with you.)
Finally, I would like to second Salmoneus' (Salmonei?) first three comments regarding POAs, case marking (which does not seem to be synchronous with the way it works in verbs). I know you are making a proto-language, but this seems so novel, you kind of have to understand how that came about. Was it that the accent used to be fixed, then after the elimination of some affix, jumped? How can you make it so that the verbal system's use of the accent as a marker and the nominal system's use, look like they could come from the same source? Right now, they look a teensy bit random.

But a lovely start, I must say.
The oblique, as explained, is from the loss of a short vowel, which marked the older case. This loss lengthened the final vowel, which caused the second vowel to remain stressed.

For the verbs, we'll look at the reflexive, benefactive, middle, and causative. Each appears virtually the same, despite the accent:
Middle: *pḫét-tey-si.
Causative: *pḫet-téy-si.
Reflexive: *pḫet-tey-sí.

In Pre-Proto-Deithas, these were separate morphemes:
Middle: **paét-tèy-s.
Causative: **paet-e-téy sù-.
Benefactive: *paet-é-tey
Reflexive: **paét-tèy sî.

When the vowel system collapsed, a few things happened. /u/ and /i/ merged into /ɨ/ (of either length). The falling pitch was lost, and the rising pitch became the primary stress on the word. In the reflexive, this stress was retained on the reflexive pronoun, which became a suffix to the word. The causative seemed to be a separate construction, where the stem received a derivation, **-e-. The word that followed it was marked with **su, which marked the agent (so a ditransitive causative verb was "Subject Verb **su Agent Object", or so internal reconstruction suggests.) This particle was fused onto the verb, instead of the noun it modified, altering the stress pattern. The benefactive was the opposite move, analyzing the the new -si as an enclitic rather than a suffix, and therefore retaining the stress on the derivational -e-. (It's not entirely sure how the semantic difference between the two forms arose, as they arose from the same form in Pre-Proto-Deithas).

The middle is a bit harder to justify, as it should end in *-s, not -si. It would seem the same analogy that regularized the roots in these voices may be responsible for *-si in the middle.

The avalent gnomic and locative follow a much more basic pattern: A verb is suffixed after the stem. The avalent gnomic has *-ḫiṣ-, from the root of to be, and the locative has *-yi-. These forms are stressed on the second syllable.

I'll have to explain the theoretical development of the passive/detransitive, and the malefactive/negative detransitive at a later time.

User avatar
vec
Avisaru
Avisaru
Posts: 639
Joined: Tue Sep 16, 2003 10:42 am
Location: Reykjavík, Iceland
Contact:

Post by vec »

Sounds very cool. So what are your plans for the descendants of this language?

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

My plans? Well, I have a few things to work on:
1). Larger starting vocabulary. Currently, I have 261 independent lexemes, plus one reconstructed proper noun of a city. I'd be more comfortable once I have 600-700 words, then I can reduce that number in the daughter languages.
2). Culture and archeology. I want to work on cultural concepts, specifically relationships and proto-mythology, then work on archeological concepts. What time frame were they in? What technology did they possess? This will assist in developing a rudimentary history.
3). Once those are done, I can work on differentiating the dialects, to create the separate branches (North, West, South-Central, and two separate isolated dialects that possess no aerial features--perhaps more).

I can't really answer 3 yet, as I've had to scrap what aerial features made up the dialects when I started this project. It'll take some time to figure that out, but when I know, I'll post it.

Neek
Avisaru
Avisaru
Posts: 355
Joined: Mon Sep 30, 2002 12:13 pm
Location: im itësin
Contact:

Post by Neek »

Again with the double posting! I have completed the redraft of the basic verbal morphology, including new details. Specifically, I took out the verb paradigm charts because they were false. I justified the existing endings with a bit of history.

But a much larger note is a piece of criticism left by Salmoneus: The lack of agreement has been justified. I have included agreement prefixes, and a bit of detail behind them. I also justified the formation of the detransitive case: At one point, pre-Proto-Deithas utilized object suffixes to agree with objects. This has eroded into the detransitive voice, leaving behind the object suffixes that agree with the subject.

So "I have eaten fish":
mentyágtey ehm praṣé.

But, I have eaten is:
ntyaghéy ehm.

(Or in some crazy instance of hypercorrection:
**mentyaghéy ehm, using the personal markers as a circumfix!).

I'll be sure to include more details in the update of voices, where I actually differentiate between true voices and functions of voices, which will reduce the number of voices to a realistic total.

Post Reply