Saturday, 23 February 2013

Moten Part X: Surdéclinaison, Other Patterns and Isolated Cases

As promised in my previous post, this article is the last one of a series discussing surdéclinaison (the first one is available here, and the second one here). In this article I am going to describe more restricted patterns, as well as isolated cases that don't fit in general categories. This post should be slightly longer than the previous one, but it should be entertaining!

Complements of Comparisons

A long time ago, I introduced the degrees of comparison in Moten, and I explained then that I couldn't describe how to form the standard of comparison (the complement typically introduced by "than" in English), as I hadn't yet described the grammatical structures needed to form such a complement. This was, of course, because complements of comparison in Moten make heavy use of surdéclinaison!

But first, let's give you a quick reminder: the degrees of comparison in Moten refer to a series of prefixes and circumfixes used on nouns to indicate:

  • The comparatives of superiority ("-er/more"), equality ("as/as much/as many") and inferiority ("less/fewer");
  • The superlatives of superiority ("-est/most") and inferiority ("least/fewest");
  • The intensifiers/excessives of superiority ("much/many/very/very much/a lot/too/too much/too many") and inferiority ("not very/not very much/little/few/too little/too few/not enough").

All those forms can take a complement, although what the complement means and how it is constructed depends on which structure we are talking about:

  • The complement of comparatives indicates the standard of comparison, and is introduced in English with "than" or "as". It can be a full clause ("he is as tall as I remember") or a noun phrase ("he is taller than me", or more pedantly "he is taller than I"), possibly starting with a preposition ("he goes more often to the beach than to the mountain");
  • The complement of superlatives describes the group from which the element in question is highest or lowest in quality, and in English can be a noun phrase introduced with "of" ("he is the tallest man of the house") or a relative subclause ("this is the smallest cat (that) I've ever seen");
  • Intensifiers don't really have complements (in the sentence "he is very tall for his age", "for his age" is not really a complement of the intensifier. Rather, it's an adverbial phrase meaning "with respect to the standard of his age", and it can be used without intensifier: "he is tall for his age" is just as correct a sentence), but excessives can. The complement of excessives indicates the goal that is missed due to the excess or lack, and in English can be introduced by "for" and be a noun phrase ("there is not enough left for me") or an infinite clause ("this is too high for me to reach"), or can be an infinitive ("this is too heavy to carry").

In Moten, the constructions are different, but the meaning is the same.

Let's start with the comparatives and the superlatives, as they both use the same construction to form their complements. In Moten, the complement of a comparative or a superlative is always introduced with the originative prefix go-. When the complement is a full clause, this prefix go- is used like any other prefix to form an adverbial subclause. Both methods described in the previous section are valid, although by and large the most common one is to directly add the prefix to the auxiliary of an independent clause.

However, this is not enough. Adverbial clauses, like adverbial phrases, can only be used to complete a verb, never a noun. However, at least in Moten, the complement of comparatives and superlatives is considered a noun complement. So we still need to convert the adverbial subclause into a noun modifier. This is done exactly like adverbial phrases as explained in the previous post, i.e. by over-inflecting the adverbial subclause in the genitive case (note that the dependent form of the auxiliary is not used in this case: once it has undergone surdéclinaison, an auxiliary is treated like any nominal for further surdéclinaison). In other words, in order to convert a clause into a comparative or superlative complement, you need to over-inflect it twice, the first time with the originative prefix go-, and the second time with the indefinite singular genitive case! Here are a few examples to illustrate how this works:

Umpi kolen gdan jelojmastu|l gojdvoj pepludegun ige: This house is smaller than I remember (literally: "This house has more smallness than (it) is remembered by me". Although hardly recognisable, gojdvoj is simply ito over-inflected in the originative and the genitive case. Notice the verb jelojmastu|l, which corresponds to the verb "to remember", but with an opposite argument orientation: the thing remembered is the subject, while the person remembering is the object, making the verb more exactly equivalent to "to be remembered by". I will discuss this verb and similar ones in a future posts).

Len pe|laz gojdvoj badi pesezgedono ito: This is the fastest dog I know (literally: "This is the fastest dog than (I) have seen (them)". Notice how the relative subclause in English is replaced by an adverbial subclause turned noun modifier in Moten. The difference between the two is that there is no expectation that the head of the subclause has a function in the subclause itself. Rather, here the subclause just describes the group the head is part of, and it is just Moten's pro-drop nature that makes it look similar to the English relative clause. But in fact, if we were to use a more complete subclause, this would be, rather redundantly: Len ba|zin pe|laz gojdvoj badi pesezgedono ito, i.e. literally "This is the fastest dog than (I) have seen dogs").

While it is fine to know how to form subclauses completing comparatives or superlatives, in general the complement of comparison is a single noun phrase (possibly one with a preposition). This is possible in Moten as well, but such forms are actually derived from subclauses, so it was necessary for me to describe them first in order to make this part understandable.

The principle is simple: the standard of comparison must be described using a subclause in Moten. However, the ability to omit anything that can be understood by context is still valid, so when the subclause closely parallels the main clause it's possible to omit everything except the different bits. And in this case, this includes omitting the verb altogether! (in a future post, I will show that omitting the verb of a clause is actually more common than I made it seem in previous posts, at least in an informal speech register. For comparisons though, it's routinely done in every register) When this is done, the marks of the subclause (the originative prefix and the genitive case) cannot be omitted though (they are needed to indicate the role of the remainder in the main clause), so they get relocated on whatever is left (normally a noun phrase, which ends up over-inflected twice over).

This description may seem overly complicated, but the described phenomenon is actually rather simple, especially if you see it in action. So I will illustrate it by giving a few examples starting with a fully specified subclause and carrying on with its shortened form:

E|lon kolam luvami koka|se gojgvej peftin ige: That woman is taller than her husband is (literally: "That woman has more tallness than her man has". Here the subclause is already shortened slightly. A full, but overly redundant construction would be luvami koka|se fedin gojgvej, literally "than her man has tallness". Notice the noun fin: "summit, top, great height, tallness", and the use of ka|se: "man" with a possessive pronoun to mean "husband") -> E|lon kolam luvami gokokazvej peftin ige: That woman is taller than her husband (literally: "That woman has more tallness than her man". Here, as you can see, the comparative complement is formed by simply omitting the verb of the comparative subclause. The remaining phrase is kept in its original form, here the instrumental, and is over-inflected in the originative and genitive, resulting in the triply inflected gokokazvej).

E tinesa umpi molen izunlaj gojdvoj tina petunedano ipelda|n ito: You can see here the biggest room that is in this house (literally: "(You) are seeing here the biggest room than the rooms are in this house". In practice, tinesa would be omitted here, leading to a sentence similar to the English translation. The noun tuna used here means "big" or "big size") -> E umpi gomolve|n tina petunedano ipelda|n ito: You can see here the biggest room in this house (literally: "(You) are seeing here the biggest room than in this house". Here the only thing not omitted in the subclause is the locative phrase, which is then over-inflected in the originative and the genitive case).

Bazlo molen o|so bazlo molam izunlaj gojdvoj peo|so izunlaj ito: There are more cars in this town than there are cars in that town (literally: "In this town more cars are situated than cars are situated in that town". This sentence shows how the comparative construction can be used even when what is compared is used as a noun rather than an adjective) -> Bazlo molen gomolvami peo|so izunlaj ito: There are more cars in this town than in that one (literally: "In this town more cars are situated than in that one". This example may be the clearest of them all: the main clause and the subclause were nearly identical, and the only thing that has not been omitted in the subclause is the different pronoun).

Although this structure is superficially similar to the English usage, there is a small difference worth mentioning. In English, "than" is often treated as a preposition, leading to sentences like "he is taller than me" (instead of the prescriptivist "he is taller than I"). In Moten that's never the case: even when the verb of the subclause itself is omitted, it is considered to be present "in spirit", and the comparative construction with a single noun phrase is still treated as a subclause (in this way, Moten is more similar to prescriptivist English than to actual spoken English). In particular, this means that one cannot use a reflexive pronoun in Moten in the same way as it is used in the following example: "I was part of something greater than myself". Since vike always refers to the subject of the very clause it's in, in the case of comparative constructions it refers exclusively to the subject of the comparative subclause, which wouldn't make sense in the example above. Rather, the example I just gave can be translated simply as Koga gokogvaj tamut pejuva|n judosun etok, literally: "I was a part of a greater something than I". The strength of the English word "myself" is translated here by repeating the pronoun in both the main clause and the comparison.

Now that we've seen how to form the complements of comparative and superlative constructions, the time has come to look at the complements of excessives. Luckily, although those complements work slightly differently from the complements of comparatives and superlatives, the main structure is the same, so I won't have to go through a long description again. Basically, the basic form of the complement of an excessive is a subclause, like the complement of a comparative or superlative. But rather than being introduced with the originative go-, this subclause is introduced with the final te-, which indicates goal. Like the complement of comparatives though, this subclause must then be turned into a noun modifier, so it is over-inflected in the genitive case. In other words, subclauses completing excessives are identical to subclauses completing comparatives or superlatives, with the only difference being the replacement of go- with te-. Here are some examples to illustrate this:

Ibutaj kolsos ga itmamej tejdvoj peniptizun ige: Those chairs are too heavy for me to carry (literally: "Those chairs over there have too much heaviness so that I carry (them)". Notice the use of the infinitive ibutaj: "to sit" to refer to something facilitating sitting, i.e. "chair". The noun ipiz means "heaviness, high weight" while the verb itmamej means "to carry, to hold, to bear").

Ba jo|zemej tejdvoj lentamun e izunlaj ito: There's not enough here for you to eat (literally: "Too little (of it) so that you ingest (it) is here". Notice how one can indicate the degree without explicitly mentioning what is in excessive degree by using the pronoun tamun: "some, one". This works also with comparatives and superlatives, although we will see in a future post that such usage is not as common as one might think).

Forming a complement corresponding to a single infinitive is as simple as omitting everything in the subclause except the verb itself:

Ibutaj kolsos itmamej tejdvoj peniptizun ige: Those chairs are too heavy to carry (literally: "Those chairs over there have too much heaviness so that (I) carry (them)").

Forming a complement with a single noun phrase is done just as it is done with comparatives and superlatives: omit everything in the subclause that can be inferred by context, including the verb, and move te- and the genitive case to whatever's left. Using the example directly above, one can then make the following sentence:

Tebvaj lentamun e izunlaj ito: There's not enough here for you (literally: "Too little (of it) so that you (ingest it) is here").

Notice that in this usage, when the noun phrase refers to an animate being (a person or an animal, usually), the final prefix te- can be replaced with the benefactive prefix |la-. This is never done for entire subclauses though. For instance, the example above could also be:

|Labvaj lentamun e izunlaj ito: There's not enough here for you (literally: "Too little (of it) for your benefit is here").

So that's all there is to say about the complements of comparatives, superlatives and excessives. The main thing to remember is that they rely strongly on surdéclinaison, as both a functional prefix and the genitive case are necessary to build them correctly.

Other Nominalisations

So far, all the nominalisations done by surdéclinaison we've seen involve nouns in the genitive case (possibly itself the result of surdéclinaison) and verbs in the dependent form (notwithstanding the special case of adverbial subclauses). Those are indeed common and very productive, but by no means the only way to form nominals by over-inflection. Indeed, quite a few other inflected noun forms can be nominalised through over-inflection, although most of them are not nearly as productive (or at least more restricted in meaning) than the nominalisation of the genitive. I'll mention here three such patterns, which are relatively common and reasonably productive.

A long time ago, on my post about numbers, counters, dates and time, I mentioned the word negesizdan: "week", adding that I would eventually explain how this neologism was formed. Time for me to do so.

As I explained then, the noun siza: "(calendar) day" cannot normally be used for durations, but indicates the day of the month. There is one exception: when a compound number+siza is used in the accusative case (optionally with the functional prefix of time di-), it indicates that the action lasted for a continuous period of calendar days. For instance:

Bdan imasizdan izunlafodo|n ito: He's been looking for you for three whole days (literally: "(he) is trying to locate you for three calendar days". Notice how the verb izunlafotoj: "to search, to look for, to try to locate" is in the imperfective aspect rather than the perfect aspect as in English, since this sentence does not indicate the result of something happening in the past, but rather a current event that happened to start in the past).

Using imasizdan here emphasizes that the action happened continuously during that period of time (hence the translation "for three whole days"), unlike alternatives like imagdomun, which also means "for three days", but doesn't necessarily mean that the action happened continuously during that period (in particular, since gom literally means "the period of the day between sunrise and sunset", it implies that the action stopped during the night, while imasizdan implies continuous action throughout day and night. Yes, you can exaggerate in Moten as well as in any other language!).

Such an expression can, like any other, be over-inflected in the genitive in order to complete a noun (for instance: imasizduva|n zunlaz: "a three-day stay", for instance at a hotel), and such a form can itself be nominalised by surdéclinaison: imasizduvea|n: "the three-day one".

However, the expression itself can be nominalised as is, without having to over-inflect it in the genitive first. This pattern forms nouns that refer to generic calendar periods that don't already have a name (like siza itself, mune: "month" and ada: "year"). In particular, it can be used to form Moten equivalents of periods of time that have a special name and meaning in various languages but not in Moten. Hence negesizdan (literally "seven-day period") for "week", or for instance getolsizdan (literally "fourteen-day period") for "fortnight".

One thing to remember that there is nothing special about negesizdan and getolsizdan in Moten. That's to say, those words are no more culturally important than, say, |simsizdan: "six-day period" or genisizdan: "ten-day period" (a "week" of the traditional Chinese calendar). However, since they correspond to relevant periods of time in our Western world, C.G. and I use them more often than other periods.

By the way, this pattern is not restricted to siza. Any counter related to time and date can be used in the same way. For instance, with pele: "minute", one can form gevelbelden: "quarter of hour" or imagenipelden: "half hour". With mune, you get for instance imamunden: "trimester" and |simunden: "semester". And with ada, you naturally have japujadan: "century" and senadan: "millenium".

While the above pattern is still relatively generic, the one I'm going to talk about now, while productive, is also very restricted in meaning and use. Still, it's interesting to look at.

Back when I first described the functional prefixes, I may have given the impression that the benefactive |la- and originative go- were only used with persons, while the final te- and causative |zu- were only used with objects and concepts. And while it's true that |la- and go- are often used with animates and te- and |zu- with inanimates, it's by no means the whole story (as the use of go- for the complement of comparisons, regardless of the animacy of the complement, shows). One day I'll go back to the use cases of the functional prefixes (which are quite varied), but for now let's focus on one particular case, which is the basis of the surdéclinaison pattern I want to talk about.

As I've explained before, when used with a person the benefactive |la- can indicate the person who is given something, or the person for the benefit of whom the action is done:

Nanageduzun |laba negesizdan kun eganeo joplej ito: I'll give you the book next week (nanaguz is the participle of |nanagi: "to write", but is mostly used to mean "book". Notice the expression negesizdan kun (dj)eganeo, literally "on the second week after an unspecified origin". Context makes it clear that it means "next week". Notice also how the perfective present with an indication of time corresponds to the future tense in English).

Tinedan |laga izu|lebi egek: (He) cleaned the room for me.

However, the benefactive can also be used with inanimate concepts and objects, in which case it indicates that the action is done in order to enhance, improve or upgrade that object or concept. Here's an example to illustrate:

Poltuz amla kolen |lajumbvude|n ito: This new door will improve the house (literally: "this newly-acquired door is one to enhance the house". Using te- instead of |la would have resulted in a neutral statement instead: poltuz amla kolen tejumbvude|n ito: "this new door is for the house". Notice also the expression |lajumbvude|n, which is the adverbial phrase |lajumpej: "for the sake of the house" over-inflected twice, first in the genitive case to make it a modifier phrase, then nominalised and put in the accusative case as atom requires. This kind of expressions is relatively common in Moten).

While there is still nothing special about this, things start getting interesting when the noun with the benefactive prefix refers to a body part. In that case, the resulting inflected form can be nominalised directly by surdéclinaison (without first putting it in the genitive). The noun formed that way has a very specific meaning: rather than indicating something generic that in some way "enhances" the body part, it refers specifically to a piece of jewellery that is commonly associated with the body part in question. For instance, from poma: "neck", one can form |lapoma: "necklace". From jespoma: "wrist" (literally "hand neck", with jez meaning "hand", "arm"), one gets |lajespoma: "bracelet". From mensin: "ear" (literally "cup", used metaphorically to refer to the outer ear including earlobe), we get |lamensin: "earring". And from ipe|lastu|l: "finger" (basically the infinitive of a verb meaning "to show"), one forms |lajpe|lastu|l: "ring".

"Wait, what about zanej?" you may ask. Good question! As it happens, there's no reason why the pattern above should be the only way to name jewels, and indeed, some pieces of jewellery have alternative names. In general though, the terms based on body parts are the ones that are most commonly used. The exception is, indeed, the finger ring. Because the word |lajpe|lastu|l is so long, and a bit of a tongue twister, the much shorter and easier to pronounce synonym zanej is used much more often. In any case, it's an interesting pattern to remember, despite its restricted use.

The last pattern I want to mention is slightly different from the previous ones. It may also be at the same time the most restricted one in terms of usage, and the least restricted one in terms of productivity. Indeed, unlike previous patterns and actually all surdéclinaison as I've described it so far, it can be used with any word in any form, including particles! (and that includes clitics as well) In fact, it can even be used with things that aren't even words, like phrases or sentence fragments!

Until now, we've seen how to talk in Moten about things, events, actions, situations, etc. To do so, I have written about Moten, in English. But what if I wanted to translate those blog posts into Moten? What if I wanted to talk about Moten in Moten? To do so, we need the ability to quote words and sentence fragments in Moten itself, something we can't seem to do yet. I did mention how to handle direct speech once, but that method (repeating the direct quote as a separate sentence and referring to it further using a pronoun) does not really work well when talking about words. What we need is a method to embed a sentence fragment in another sentence, in a way that allows case marking so that we know the function of that fragment in the sentence.

In Moten, this is actually done in a relatively simple way: by nominalising the word or sentence fragment using the infix article -e- (basically, treat the fragment as a nominal stem and put it in the definite form). The result can then be inflected as if it were any other definite noun. In writing, this is usually accompanied by quotes, while in speech there is sometimes a slight pause before and after the nominalised form, especially when it's a sentence fragment.

There are three main uses to this pattern:

  • Embedding quotes in other sentences;
  • Mentioning citation forms of words;
  • Using interjections adverbially.

The most common use of this pattern is to embed direct quotes (usually short ones, more rarely full sentences) within a sentence, as an alternative to the usual quoting method described above. Here are a few examples:

Mudutun isej etok? — 'Mejtedon' isej etok: "What did (he) say?" "(He) said 'hi'." (literally: "(He) said the 'hello'", with mejto, a particle meaning "hello, hi" nominalised by -e- and put in the accusative case to become the object of isej: "to say").

'Kovepe|ne' lugen penegipedizno ive|zaj ito: "Sorry" seems to be the hardest word (literally: "The 'sorry' looks like the hardest word". In this case, due to the shape of the word vepe|ne: "sorry", the surdéclinaison isn't visible, and since that word is also a noun meaning "apology", the sentence could also mean: "The apology seems to be the hardest word". Context and simple common sense make the meaning clear in this case. Notice also the form penegipedizno, from negipiz: "difficult, hard to do", put in the superlative of superiority, definite and in the accusative singular. It parses as pe-negip<e><d>iz-n-no. It's actually a case where the morphophonemic rules of Moten are slightly unclear about how to treat the suffixes -n and -no together, given that without the superlative the word would become negipedizun: "the difficult". Maybe due to the influence of this form, a Moten speaker might be heard saying penegipedizuno instead of the form above. They are in free variation, although the first one is still the one most commonly used).

Luvosi 'us vajaguz ite|zon' bunes ito!: His "I don't know" are annoying (me) (literally: "(I) have become annoyed by that one's '(I) haven't learned'". Here what is quoted is actually a full clause: us vajaguz ito: "(I) don't know". It is treated as a single word and nominalised with the article, and then declined in the accusative plural. Its status as a nominal is cemented by the use of the genitive phrase luvosi: "that one's" on it. Notice also the verb ibunesi: "to become annoyed by". It is similar to the English verb "to annoy", but with opposite orientation, and refers to the creation of bother, hence the use of the perfect aspect to indicate that the annoyance is currently existing).

Slightly less common, for the simple reason that talking about one's own language isn't a common activity for many people, is the use of the pattern with citation forms:

'Kozunlea' tolugden ito. Me|lo 'kojzunleaj' neglugden ito. 'Kome|leo' samlugden ito: "Place" is a noun. "To be at", on the other hand, is a verb. "But" is a particle (literally: "The 'place' is a nominal. The 'to be at' however is a verb. The 'but' is a particle". While the first two citation forms are indistinguishable from normally inflected nominals, the last one inflects the normally invariable particle me|lo, which means it has to be surdéclinaison in action here. Notice also the infinitive. As I mentioned a while ago, the infinitive is used as citation form for verbs. But to be truly used as a citation form in a Moten sentence, it needs to be definite. Finally, notice the words tolugen, neglugen and samlugen. They are the Moten names for respectively nominal, verb and particle).

'Kojteo' 'atevomi' beldegun ito: "Is" is a form of "to be" (literally: "The 'is' is a shape of the 'to be'". The noun be|leg means "form, shape, figure, appearance").

Komotenku|leju 'house' de lugeden komut isej ito? — 'Umpedin': "How do (you) say 'house' in Moten?" "'Umpi'" (literally, the answer is "The 'house'", in the accusative case as it is basically an abbreviation of 'umpedin' isej ito: "(you) say 'umpi'". Notice the expression 'house' de lugeden, literally "the word 'house'". The word de is a coordinating particle, which like opa can often be translated as "and". Its main difference with opa, though, is that while the latter coordinates nouns that have different referents, the former coordinates nouns that have the same referent. In other words, while the phrases ka|se opa ge|sem and ka|se de ge|sem both mean "a man and a father", in the first case the phrase refers to two separate persons, one a man and the other a father, while the second refers to a single person, who is both a man and a father. Another common way to translate de is as "that is" or "that's to say", although the most common way to translate it in English is by plain juxtaposition. And this is the case in the expression 'house' de lugeden, where de is used to indicate that "house" and "the word" have the same referent, i.e. "the word 'house'". This is done because while Moten speakers are happy to over-inflect even their interjections, they aren't as willing when it comes to foreign words and phrases that have not been borrowed and adapted to the Moten phonology. Those just cannot be inflected. Since uninflected phrases cannot have a grammatical function in a sentence, the solution is to coordinate the foreign word or phrase with a Moten word like lugen: "word", which can take inflections. Since only the last of coordinated phrases needs to be inflected, this solves the problem: the foreign word can stay uninflected, while still having a well-defined function in the sentence).

Finally, this pattern can be used to include interjections and onomatopoeia into a sentence, with a well-defined syntactic function (rather than the paralinguistic function those usually have). Mostly this is done for purposes of quoting (see above), but when the nominalised interjection is put in the instrumental, it can be used as an adverb of manner, describing the action in a metaphorical way. A few examples will help understand how this usage works:

Gevomi genegedaj tekafe kokinkean ipenlazdu|lun ito!: (He) keeps inviting (me) for coffee, everyday at 5, like clockwork! (literally: "(He) keeps inviting (me) each day at 5 o'clock, for a coffee, like the tick tock!". Here we see the onomatopoeia kinkan, representing the sound of a clock ticking, over-inflected in the instrumental definite to indicate an almost mechanically recurring event. It can be interpreted positively, to mark punctuality, or negatively, to mark routine. Notice also the verb ipenlastu|l: "to invite", conjugated in the imperfective aspect as the sentence refers to a habitual event)

Los koni|sej otedon joknezde|n etok: He kept droning about his car (literally: "He kept talking about the car like the blah". Here the onomatopoeia used is ni|si, the sound of continuous speech, which can mark boredom. The verb is joknesej: "to recount, to narrate, to talk about").

Kozutejuun izektin etok: (He) was following as still as a mouse (literally: "(He) was following like the ...". Even an onomatopoeia that does not strictly follow the phonotactic rules of Moten, like zutuun, the mark of absolute silence, can be over-inflected. The verb here is izeki: "to follow, to go after").

One word of warning: this usage is considered very informal. It's used in speech among friends, but it should be avoided in writing and in formal situations. I'll devote a future post on the subject of language registers in Moten.

Isolated Cases

We've nearly reached the end of this already far too long post. But before I conclude, I'd like to show you a few more expressions that are constructed using surdéclinaison. The difference between these expressions and the ones I've shown so far is that these are not productive at all. They are exceptional phrases that do not form patterns. Unlike the cases presented above, you cannot emulate these to form other expressions. But it's still useful to know they exist.

In terms of meaning, all those expressions are basically adverbial phrases, although some of them are actually disjuncts, i.e. they do not directly modify the sentence, but instead show the speaker's attitude towards it. According to C.G., a Moten speaker treats them as single units (the closest thing Moten has to true adverbs) but is aware of their internal structure, which is always transparent.

Here are a few examples of such constructions:

  • tekojses: actually, in fact (this is the noun isis: "truth", in the instrumental definite, over-inflected with a final prefix. The form kojses itself means "really, truly");
  • godetun: from now on (here we have the pronoun et: "this time, now", in the accusative case and over-inflected in the originative);
  • tegoga: personally, for my part (this is the pronoun ga: "I", put in the originative and then in the final form. It's used as a disjunct, and interestingly it keeps its form even in indirect speech, although it then refers to someone else than the speaker);
  • dikovo|sedan: normally, when things go well (the noun vo|sa: "beauty, appropriateness", in the instrumental definite, over-inflected in the accusative case, with the temporal prefix di-. Despite the temporal prefix being nominally optional, in this case a construction without temporal prefix is never used);
  • |negdin kodidon: while we're at it, at the same time (a construction difficult to translate in English, it expresses the idea of doing something at the same time as something else, because the two activities belong with each other, or at least one facilitates the other. It's the verb |negi: "to do", conjugated in the imperfective aspect, and then made into a temporal subclause, by over-inflecting the auxiliary in the accusative case with the prefix di-, itself over-inflected in the instrumental. The result is still treated as an adverbial subclause, and can actually take participants. For instance: ba |negdin kodidon: "while you're at it" or umpedin |negdin kodidon: "since we're busy with the house". This expression is somewhat informal).

There are many more of those, and I will point them out when we meet them in future posts.

What's Next

Pfew! In total, this took much longer than I expected! That said, surdéclinaison is at the heart of many grammatical constructions in Moten, so it deserves to be treated extensively. I do hope you enjoyed the ride!

Now that I've finished discussing about surdéclinaison, there is not much Moten grammar left to describe. Next post will be about the last bits of morphology I haven't looked at in depth yet: derivation and compounding, i.e. how to make words out of other words. Compared to surdéclinaison, this is a relatively simple subject, so I expect one post on it will be enough. And I will do my best to keep it at a reasonable length!