Passives in D’ni

korovev · **Posted:** Fri Sep 01, 2017 11:04 am

Yeah, those things :shock:

So, a discussion about passives popped up in Facebook, and while re-reading Domahreh’s First Survey (specifically, passives and participles), I pared down a set of rules I’ve been using:

-in : used as past participle, opposing -al, as an attribute (after ošanin);
-ij : used as a verb modifier, turning any active form into passive (after ril-dol-gel-en-ij);
Exception to the above: if the phrase is used as a subordinate, the passive is then formed as ken (conjugated as needed) <root>-ij (after the Kenen Gor);
I avoid the kokenen ko...-en form.

These may not cover all the examples in the corpus, but having something to work with is nice. They may be obvious to some, but might be useful for beginners.

A few points:

in the Kenen Gor there’s rilbokenet verenij, but the other part of the pair is rilbotemaet hevtíos, so to keep the parallelism, a ‘wrong’ form is used.
A few transitive verbs might take a passive meaning, in absence of a direct object and using the progressive form (e.g. rebišta kodolasaen; hevtí met dohúrít); possibly a ‘formal’ alternative for #2. Too little data, so I stick with the less ambiguous form.
I just don’t know what to do with the kokenen koséen form (why would e.g. retiwa kodoséen not work there?). Even less data, so I just avoid it.

KathAveara · **Posted:** Fri Sep 01, 2017 12:07 pm

I hadn't actually considered that rule for #3, and otherwise I agree there entirely.

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

Korovev's summary got me to thinking about the sentences in D'ni that seem to have a "passive" meaning when translated into English, but which don't have either of the "passive" suffixes -in or -ij.

One of these is the "by-line" on Aitrus's Map, which we hypothesized long ago to read shuhlen te telookahm aytruhs and to possibly translate as "drawn by Guildsman Aitrus."

I took another look at the map itself to refresh my memory, and noticed (as I think has been noticed before) that the letters in this part of the map are in a slightly different style from those in the captions throughout the rest of it; in particular the distinction between "straight" stems and "curved" stems is more subtle, because they all have some curvature at the top. Also considering the fuller vocabulary that we now have, I think that what this line actually says is:

sholen te telookahm aytruhs 'prepared by Guildsman Aitrus'.

Shorah

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

This raises the obvious question of why an idea that is typically expressed by a “passive” in English seems to be conveyed in D’ni without any explicit indication of this passivity in the verb construction? A clue to the answer lies in another text on the map.

The verb form doreesloen means literally ‘he/she/it is dissolving’ which is used transitively in the sentence where it occurs, with a direct object “the rock’s outer layer.” The 3rd person singular subject ending refers back to the preceding clauses in the sentence which describe a lizard that spits acidic saliva. Presumably one or both of the following are valid Dni sentences: rem’lah reesloen prad ‘the lizard dissolves rock’; poahnton reesloen prad ‘its saliva dissolves rock’.

In the following sentence on the map, about the lizard drinking the resulting solution of saliva and rock, the same verb is used in another form lereesloeet which means literally ‘they have dissolved’. This is an intransitive usage of the verb, so there is no object, and it is in a relative clause introduced by mot; but the subject ‘they’ clearly refers to “the minerals, plants, and small organisms” mentioned immediately before, which the lizard drinks along with its saliva. Again presumably it would be valid to say re’irvantee lereesloeet ‘the minerals have dissolved’; and most likely also re’irvantee lereesloeet trepoahnt ‘the minerals have dissolved in the saliva’.

The crucial point of course is the way the clause mot lereesloeet was translated into English idiom: “that have been dissolved.” And if we translate the literal sentence re’irvantee lereesloeet trepoahnt ‘the minerals have dissolved in the saliva’ using a similar English idiom we get: “the minerals have been dissolved by the saliva.” From the D’ni persective the two English translations (active vs. passive) do not convey a siginificant difference in meaning, and so D’ni does not make a distinction.

Shorah

korovev · **Posted:** Sat Sep 02, 2017 11:36 pm

Wait, would that mean that D’ni is ergative-absolutive?

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

There is no case-indicator to differentiate "ergative" (agent subject) from "absolute" (object or patient subject); so no, D'ni does not seem to be an ergative-absolute language in the usual sense.

Shorah

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

While the distinction between transitive and intransitive might serve for a substantial class of D’ni verbs to differentiate what is expressed by “active” vs. “passive” in English; there would still be contexts where one cannot tell whether the intended construction is transitive or intransitive and this would need to be expressed by another distinction. One example is a verbal adjective or “participle” used to characterize a noun.

Thus we might have poahnt reesloahl ‘dissolving saliva’ vs. prad reesloin ‘dissolved rock’. Also note that the former can be expanded to poahnt pradreesloahl ‘rock-dissolving saliva’. So apparently the “active” participle phrase with suffix -ahl corresponds to the transitive sentence poahnt reesloen prad ‘saliva dissolves rock’, and the “passive” participle phrase with suffix -in corresponds to the intransitive sentence prad reesloen ‘rock dissolves’.

Shorah

Talashar · **Joined:** Sun Dec 08, 2013 12:24 pm **Posts:** 91

korovev wrote:

Wait, would that mean that D’ni is ergative-absolutive?

For D'ni to be ergative, transitive objects would need to be identical to intransitive subjects in terms of some grammatical feature like case-marking or verb agreement. The fact that (like in English) a subset of verbs have a semantic identity between the roles of their transitive objects and intransitive subjects isn't sufficient.

Khreestrefah wrote:

Thus we might have poahnt reesloahl ‘dissolving saliva’ vs. prad reesloin ‘dissolved rock’. Also note that the former can be expanded to poahnt pradreesloahl ‘rock-dissolving saliva’. So apparently the “active” participle phrase with suffix -ahl corresponds to the transitive sentence poahnt reesloen prad ‘saliva dissolves rock’, and the “passive” participle phrase with suffix -in corresponds to the intransitive sentence prad reesloen ‘rock dissolves’.

This is likely to be broadly true, but there may be some wrinkles with how ahl is applied to individual words. We don't know whether or not tseemah can be used in an intransitive clause, but tseemahahl is an example of an ahl "passive participle" (assuming tsahroo is a typical Adjective and isn't doing anything funny to the thematic roles).

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

I agree with Talashar. What I have described so far can only be part of the picture. Many verbs are like reeslo in that the subject of an intransitive construction corresponds in function with the object of a transitive construction. But there are also verbs in D’ni where intransitive subjects function the way we would expect transitive subjects to function as well.

For example rehevo kroen gahreesen gahederen terthtes ‘the swarm moves, eats, and sleeps as a group’. Clearly reesen used intransitively cannot be translated “is eaten.” The question then is how does D’ni distinguish the transitive use of a verb like reesen from the transitive use of a verb like reesloen where they seem to have “opposite” meanings in relation to the corresponding intransitives.

Again the answer can be found in the sentences on the map. Another verb that we might expect to behave like reesen ‘eats’ is glahsen ‘drinks’ and we have a sentence that begins: gormot glahsen ahrepoahnt translated “it then drinks the saliva.” The D’ni is similar in syntax to the English translation except for the fact that the object ahrepoahnt begins with a preposition ah.

It is difficult to explain the meanng of this preposition, except that it goes with the noun that is the direct object in the English translation, and that it is used when the object could be left out and the verb would still have the same meaning. An example of a verb used both with and without an object is rem in pod rehmehn ah lehm ‘each flows ink’ vs. rekooahn remen ‘the stream flows’.

A verb in this subclass could not be used in its intransitive construction as a "passive" in the way other verbs can, and so there might have developed a special way of expressing a similar concept, which we should not expect to automatically apply to all verbs. The verb gel 'write' may belong to this subclass, so dolgelenij might be an example.

Shorah

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

Does the use of rildolgelenij actually confirm the rule used by Korovev:

Quote:

2. -ij : used as a verb modifier, turning any active form into passive.

Or does it point to a different rule?

The translation of the sentence khahpo rezunu rildolgelenij gahth is “perhaps the ending has not yet been written.” The active equivalent of this translation in English would be: “perhaps nobody has written the ending yet.” And khahpo rilrov legelen ahrezunu gahth would be the D’ni for this. Presumably the adverbs are grammatically optional in both sentences and we can see that the basic pattern of an “active” vs. “passive” sentence with a verb like gel would be:

rilrov legelen ahrezunu ‘nobody has written the ending’
rezunu rildolgelenij ‘the ending has not been written’.

We can make these postitive by eliminating ril thus:

erthrov legelen ahrezunu ‘somebody has written the ending’
rezunu dolgelenij 'the ending has been written’.

So apparently the syntactic transformation from “active” to “passive” involves these steps:
1) the indefinitve subject is eliminated;
2) the accusative preposition ah is removed from the object;
3) the object is moved into subject position;
4) the prefix do- is incorporated into the verb;
5) the suffix -ij is added to the verb.

To illustrate this let’s use a verb with the "active" in the past tense instead of perfect; then we get:

erthdil koreesen ahrem’lah ‘something ate the lizard’
rem’lah kodoreesenij ‘the lizard was eaten’.

Notice that this illustrates why with a verb like rees step (4) is not enough to distinguish the “passive” sense of the verb, because without the suffix -ij from step (5) we would have:

rem’lah kodoreesen ‘the lizard was eating’.

This is only true for verbs like gel and rees where the direct object is grammatically optional. If a verb requires a direct object to be grammatical, then the “passiveness” of the verb could be unambiguously expressed with just the incorporation of do- as in:

gen kodormahdhen aytrus ‘Gehn defeated Atrus’
aytrus kododormahdhen (te gen) ‘Atrus was defeated (by Gehn)’.

Here the form with do- cannot be interpretted as intransitive progressive, presumably, because the verb cannot be intransitive; and therefore the grammaticalized sense of “passive” would be unambiguous without an -ij. This may be the explanation of a sentence like:

rebishtah kodosayen kahgesh b’mes erth lenahokh yahrtee sen
‘the tunnel was originally designed to require a journey of three days’.

Shorah

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

korovev wrote:

...

-ij

ril-dol-gel-en-ij

ken

-ij

Kenen Gor

...

in the Kenen Gor there’s rilbokenet verenij, but the other part of the pair is rilbotemaet hevtíos, so to keep the parallelism, a ‘wrong’ form is used.

...

In the last point I think Korovev is onto something :!:

But to understand why we need to reconsider what is going on fundamentally with a passive clause as compared to an active one.

aytrus koreesen ah redoo
‘Atrus ate the food’ (active)
vs.
redoo kodoreesenij te aytrus
‘the food was eaten by Atrus’ (passive).

The prominent difference is that in the active the actor is the subject of the clause; but in the passive a noun other than the actor is the subject, and it is “moved” into subject position in front of the verb.

But when the object in the active sentence is a pronoun we get a slightly different result if we just apply the rules for forming a passive verb.

renahvahtee rilbovereneet set
‘the masters will not mollify us’ (active)
vs.
rilbodoverenetij trenahvahtee
‘we won’t be mollified by the masters’ (passive).

Notice that the passive does not have the emphatic placement of the subject in front of the verb that occurs in a language like English where subject pronouns are separate words. D’ni makes up for this by using an explicit emphasis with an initial phrase that is roughly equivalent to the colloquial English “it will be us”; which when rendered in grammatical D’ni becomes simply bokenet ‘we will be’. And this means that the future tense and 1st plural person no longer need to be marked on the passive verb-form itself which is reduced simply to verenij.

rilbokenet verenij trenahvahtee
“it won’t be us mollified by the masters.”

Shorah

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

Regarding the clause renezetahn kenen k’teshij trelenahokh kevoth

Talashar wrote:

I think this is our first example of a ken + -ij passive that includes a demoted te argument. (Presumably the active equivalent of the clause would be relenahokh kevoth k'teshen renezetahn.)

To put this in context here are some examples of different passive constructions, organized according to syntactic function. Note that the category labels are just for convenience of presentation.

Adjectival passive:
rekor oshahnin okh pahtstee oglahn "the lost book of ancient cities"

to met kenen bahvahnin gah bojikhahen bahvahnin ‘this place is [hidden] and will [remain hidden]’

Passive infinitive:
kenen gor khrekahntintahntee b’ken shentomeij ‘it is time for the [oppressors] to be taken from’

kenen gor khrezithahthtee b’ken elonij ‘it is time for the lowest ones to be raised’

Present passive:
renezetahn kenen k’teshij trelenahokh kevoth "the reader is entertained by the journey of another"

botahgahmeet roo kenet rilvoohee ahtmayij ‘they will know that we cannot be [stopped (?)]’

Future passive:
rilbokenet verenij ‘we will not be mollified’

Past passive:
mot kokenen kohooren terthbroon ahnotahm "that was found in a lava tube"

retiwah kokenen kosayen t’telooknahvah gahrten "the shaft was designed by Surveyors Guild Master Garten"

rebishtah kodosayen kahgesh b’mes erth lenahokh yahrtee sen "the tunnel was originally designed to require a journey of three days" 

loymaht terelin ril kodobahrelen rebishtah kodolahsahen trefilahdh "though contact wasn’t made the tunnel was sealed at the top"

Perfect passive:
khahpo rezuhnuh rildolgelenij gahth ‘perhaps the ending has not yet been written’

Conditional passive:
khoy hevtee met dohooreet pahl rifoonemah roo ril kokino ‘if these words are found then remember that I did not [surrender (?)]’.

Passive absolute:
sholen te telookahm aytruhs ‘prepared by Surveyors Guildsman Aitrus’

Shorah

Khreestrefah · **Joined:** Tue Apr 06, 2010 5:01 pm **Posts:** 163

Talashar wrote:

Khreestrefah wrote:

Thus we might have poahnt reesloahl ‘dissolving saliva’ vs. prad reesloin ‘dissolved rock’. Also note that the former can be expanded to poahnt pradreesloahl ‘rock-dissolving saliva’. So apparently the “active” participle phrase with suffix -ahl corresponds to the transitive sentence poahnt reesloen prad ‘saliva dissolves rock’, and the “passive” participle phrase with suffix -in corresponds to the intransitive sentence prad reesloen ‘rock dissolves’.

This is likely to be broadly true, but there may be some wrinkles with how ahl is applied to individual words. We don't know whether or not tseemah can be used in an intransitive clause, but tseemahahl is an example of an ahl "passive participle" (assuming tsahroo is a typical Adjective and isn't doing anything funny to the thematic roles).

This is right if tsahroo is in fact an adjective. But if so then Gehn's sentence lacks a conjunction to connect the two clauses:
kodoteegeet rilte reahreeuhtahv ‘they were working without the protection’
and
koneetsahvayeet tanuhth iglahrno ‘they suffered temporary [blindness]’.

We would expect these to have some sort of connection with a meaning something along the lines of “so that” or “as a result of which” and this could be the function that tsahroo serves, since it could contain the subordinating conjunction roo ‘that’.

If so then the participle tseemahahl is part of the second clause, and presumably modifies the subject of the immediately following verb ‘they suffered’.

The D’ni sentence would be translated something like:
kodoteegeet rilte reahreeuhtahv tsahroo tseemahahl koneetsahvayeet tanuhth iglahrno
‘they were working without protection so that needing it they suffered temporary blindness’

Note that D’ni does not require the pronoun “it” to indicate that the antecedent “protection” is the object of “needing” since the conjunction roo can be understood as the object of a following verb; as in biv roo míruh ‘all that I [desire]’. We could paraphrase the D’ni syntax as: ‘they were working without the protection as a result of needing which they suffered’.

That being said, Talashar is right that the "active" participle in D'ni sometimes corresponds to a "passive" one in English, as in dayjee fahlah’ahl 'folded path, labyrinth'. These seem to be cases where the intransitive meaning of the verb corresponds to the passive of the transitive meaning; so that e.g. 'folding path' and 'folded path' used metaphorically don't differ much in meaning.

Shorah

Passives in D’ni

Who is online