Bringing Children up in Conlangs: Your Conlang is Less Well Adapted than You Think

Natural languages have existed for quite a while. Due to this simple but obvious fact, they have also been subject to linguistic evolution. Linguistic evolution is quite different from the evolution of multicellular organisms. I am not going to talk about 'strong languages' supplanting 'weak languages', since this is not what I am thinking of at all - I am thinking of how the parts of a language themselves are selected for.

Languages consist of many bits and pieces. These bits and pieces do not really have an existence of their own. They exist as pathways in neural networks, viz. the brain. When an infant learns to speak, it observes the linguistic productions of other humans, and tries to identify the patterns in these productions. This we call learning, and learning is adjusting the weights in the pathways in the brain.

What relevance does this have to the evolution of a language? Well, clearly, patterns will be smoothed out in some sense. Our brains are fairly similar, but they are not identical copies. One brain might not spot the same pattern as another, and thus fail to generalize it. Thus, over time, only patterns that most brains catch will survive. Thus, the fact that some feature is in the grammar of a natural language is not only testament to this feature being learnable, but to it being some kind of local optimum for learnability - other similar patterns that are less learnable will turn into that pattern.

Of course, diligent practice - in a formalized setting - can make a less learnable pattern dominate, but that requires quite some effort. In part, some of the 'prescriptive' grammar ideas we hear are taught in such ways so as to enforce some rather unusual meanings.

Further, when a natural language lacks methods of expressing something that real life makes necessary (or at least favourable) to be able to express, a method for expressing it will soon appear; gaps are filled quickly, and good contenders for filling the gap will survive. And they will filter through the speech community. Thus, huge gaps in human interactions will not exist for long, and patterns for generalizing them will fill it out faster.

When you speak your conlang to your child, you'll end up having to invent a whole lot more of this on the run that when you speak your native language. And you won't have the time to go and jot down what you thought up, and chances are you won't remember what approach you took, thus making it quite likely you'll end up with an inconsistent hodgepodge, making it hard for the child to be able to rely on the linguistic stimulus it is exposed to.

When you've designed your language consciously, it has not gone through this smoothing process either - thus it may have features that are unlikely to be parsed the way you prescribe them to be parsed by a first-language-learner, or it might have features that are just cognitively unlikely to work out - phonological distinctions used in ways that are (too) hard to resolve, morphological and syntactic things that cannot really be figured out without formal teaching, etc. We don't know how complicated things an average child can be expected to be able to figure out.

In real languages, much of the redundancy we see appearing (so-and-so, it went ..., so-and-so, he went ... etc) are attempts at improving the likelihood that the hearer gets enough data right. If no one ever had heard things wrong or indicated that he didn't catch that word, it's a bit unlikely we'd go and waste time and effort adding a lot of extra syllables here and there. However, we do add them - and it seems this helps reduce the amount of mishearing and so on. However, if you have a formalized grammar that you've struggled to internalize, you'll be quite likely to think that the formalized version you have is not to be adjusted by such tricks - you'll stick to the levels of redundancy in the language you've made. What if the level of redundancy is insufficient? Are you even likely to realize this from your child's reactions? Notice that the way real languages do this is by being a large distributed algorithm where - in some cases dozens, in some cases hundreds, in some cases millions of people are involved in adjusting the parameters of redundancy and in testing out people's hearing. Some of them do have too little redundancy, some quite enough.

But with your child, it's quite likely you won't have the support of a huge community of other people randomly fine-tuning it like that, and it's hard to say whether your interactions - especially early on - will give a sufficient idea of whether the language has enough redundancy (or contrarily, way too much).

This is of course not all there is to it, there's even more similar arguments that can be presented.

[I have answered some comments to this post in this new post]

[This is part of a series regarding why I would advise against anyone bringing up their kid in a conlang]

Detail #140: A Link and riffing on its contents

Aszev wrote, some time ago, a pretty good summary and introduction to systems of answering polar questions. It is a good read, and really condenses the matter very efficiently.

What could be some fun way of going slightly beyond this? We could of course try to come up with some 'new' table not present in the typology offered by Aszev.

1) Right-left flip on the Three-Form System
Conflate positive answers to negative and positive questions, distinguish negative answers to negative vs. positive questions. An obvious extension. Why it doesn't seem to be attested surprises me a bit. This is a three-way system mirrored left-right.

	pos!	neg!
pos?	yes	nay
neg?	yes	nope

2) Pseudodiagonal Languages
The agreement system gives a nice diagonal system.

	pos!	neg!
pos?	yes	no
neg?	no	yes

But we could obviously make one of the diagonals not conform to letting it be diagonal.

	pos!	neg!
pos?	yes	no
neg?	no	nope

Or alternatively make one of the nos deviate.

	pos!	neg!
pos?	yes	no
neg?	nope	yes

3) Doing other things with this:
We could imagine a language that takes two types of this, and basically marks for both. Of course, let's not have the marking have similar exponents at all!

We take an agreement system (and apply that to presence of person inflections on the verb, for instance), and a negation particle that follows the English two-form system:

	pos!	neg!
pos?	yes, congruence	no, no congruence
neg?	yes, no congruence	no, congruence

We could maybe use some other exponent for this, who knows? Maybe the things I currently mark 'no congruence' get a peculiar subject case? Or maybe something like:

	pos!	neg!
pos?	yes, verb	no, did not verb
neg?	yes, did verb	no, not verb

There's endless possibilities. Of course, these will interact with other negation in general, and possibly one might want to use some of these as negation (or affirmation)-approaches in other circumstances. Maybe the markings given by the neg?-pos! cell encodes a strong affirmation, pos?-pos! encodes a reaffirmation, neg?-neg! a strong negation, pos?-neg! something else, so there's use for this outside of the slightly limited context of polar questions.

We could of course combine some other systems in some neat ways like this as well.

Detail #139: Continuous verb aspect, reduplication and a twist on reduplication in general

The reason I present this idea in the order I do is that this is how I came up with it - following this path of thinking.

Ok, so doing something along the lines given below is quite obvious:

I run run ≃ I habitually run

I think think ≃ I habitually think

etc

However, how about forming an intermitten aspect by non-continuous reduplication, i.e. by placing some other arguments between the two verbs:

I run a bit run ≃ I sometimes run, I run, then I don't, then I run again

I sing songs sing ≃ I sing every now and then

So basically, weakening the effect of reduplication by increasing the distance between the reduplicated elements. We could maybe extend this to plurals by reduplication:

house burn.pres = a|the house burns

house house burn.pres = houses are burning 

house burn.pres house = a few houses are burning

Detail #138: An ergative twist on topic-marking

Let us imagine a topic-prominent language that has a marker for its topics - say a suffix. Now, when the topic is the only noun present, it is unmarked.However, when a bare non-topic is present, disambiguation is necessary and the topic gets marked - however, several other types of noun-phrase may remain unmarked - subjects, objects, possessors and indirect objects. However, if all other present non-topics are marked by adpositions or explicit case markers, the topic may be unmarked.

Barxáw: The Noun Phrase

Barxáw noun phrases and prepositional phrases have certain complications that may be of interest.

In general, and this goes for all types of determiners and attributes, 'less' heavy attributes go to the left of their heads, heavy attributes to the right. Heavy attributes include subclauses and prepositional attributes -

túλxà ðo q^hmaní - official of other (ambassador)

túλxà ðo Íst^haŋ - ambassador from Ístaŋ

 Íst^haŋ túλxà - an Ístaŋian officer 

 sartè dásqu - warm drink (generally heated berry juices)

t^hárg sartè dásqu - spicy warm juice (generally herbs rather than what English denotes by spices)

 sartè prùk - warm blood (a term for warm fruit or berry juices flavored with meat broth)

bɛ̀n díp - running water (here, a verb works adjectively; the actual verb is not 'run', but rather 'flow', but the meaning is closer to the English phrase 'running water')

The order of adjectives tends to be pragmatically determined - adjectives that 'restrict' the type of noun, rather than describe it, are closer to the noun. Adjectives that describe desired qualities rather than necessary qualities are further from it, etc.

There are two kinds of adpositions - prepositions proper and Wackernagelpositions. Wackernagelpositions occur as the second word in the phrase. 2ndpositions could be a reasonable name as well. Thus:

ðetrú kaw - along the side

q^hmaní kaw ðetrú - along the other side

Counter words appear just before the noun, unless a 2ndposition gets in the way.

sún um ðìnt - (made) of (counter: bushel) hay; sún is a proper preposition

evé sit múlɛ́ - (counter: congregation) by permission from court; sit is a 2ndposition.

Numerals and other quantifiers go just before the counter. Most nouns require counters with prepositions, but a fair share of exceptions exist.
Complex numbers (i.e. having more than two parts) go to the right of the noun. Simple large numbers, such as 'thousand' or 'hundred' can also be moved to the right to emphasize the number or make it mean something like 'untold hundred(s)/thousand(s)'.

The 2ndpositions revert to normal prepositions in embedded adpositional phrases -

malí um sún ðìnt - house (made) of (bushel) hay

ðəlín sit evé múlɛ - license (≃ letter) by permission of court

Some of the wilder things - nouns or adjectives extracted from the phrase, etc, also revert 2ndpositions to prepositions. However, those will be reviewed in a separate post.

Detail #137, pt4: The Genitive Participle

Again, the genitive is not (normally) a verbal argument, so its use suffers from much the same problems as the comparative object participle does. However, there are some things we can do with it:

• 'locking' which argument(s) can be possessed; something high up in the accessibility hierarchy seems reasonable. Subjects and objects at most? We might not even need to distinguish when it's a passive or active - maybe that's lexically restricted by the verbs themselves, maybe it's assumed that context or type of noun involved disambiguates.
• In some languages, certain verbs take genitive arguments for subjects or objects, c.f. Finnish 'täytyy' ('has to', 'must') takes genitive subjects, Russian бояться ('to fear') takes genitive objects (which for animate masculine is identical to accusative, but for other nouns it's distinct from the accusative). For a language that has lots of genitives all over the place, look at Icelandic. (It has all its cases all over the place.) For such a language, verbs that take genitive arguments could reasonably well take such a participle too. gen-бояться-participle = who is feared, gen-täytyy-participle = who has to. Let us involve such a thing in the language.

So, we have a few exceptional verbs, and the rest we deal with according to the previous approach: contextual or lexical disambiguation. However, possessing something is also in a way having control over it. So, 

• fish(verb)-gen.participle John - John, who is an authority on fishing? John, who decides on issues pertaining to fishing? John, who is the boss of the fishermen? John, who is a damn good fisherman
• pray-gen.participle priest - the priest who has the prayer - the priest who conducts us in prayer, etc

Thus, essentially, the genitive participle becomes a bit causative-like, but with complications: not one who causes a thing, but one that has power over a thing. Thus, the term for a doctor is heal-gen.participle, a mathematician is a count-gen.participle (unlike an accountant, who is count-active.participle). Thus also socially powerful persons have genitive participles for their occupations even if a regular active or passive or even recipient participle would be semantically and syntactically reasonable.

Some verbs, however, where the type of subject or object is very clear, the implication is more genuinely genitive-like:

• Tim was birth-gen.participle = Tim's wife has just given birth

Almost invariably, die-gen.participle marks the death of the spouse; if the noun is a clan, it is the death of a clan leader - dead-gen.participle clan = a clan whose leader has died. A number of participles mark family-related things when the genitive participle is used, and the hierarchy tends to be one of political significance. Likewise, verbs that normally have cattle as subjects - give birth to calves, give birth to sheep, run off (of cattle), since the type of subject can be assumed and large portions of the population own cattle.

Detail #137, pt3: The Comparative Object Participle

Ok, we've established that relativized genitives and relativized objects of comparison are somewhat exceptional. Let us however do something with them anyway. This post will deal with the objects of comparison, a fourth installment will deal with the genitive participle.

I do not propose that this is a feasible source for such a morpheme, but it'll be easy to remember when seeing it): -than for 'object of comparison participle' - 'John's runthan Eric' - Eric, who John runs faster|more|etc than.

Verbs have associated qualities or quantities, and if no other quality or quantity has been called attention to in the wider context, the associated quality or quantity is assumed. Thus runthan ≃ who x runs faster than, eatthan ≃ who x eats more than, singthan ≃ who x sings more beautifully than. If both are subjects, X can be a genitive attribute of the participle (as in the example given above), if both are obliques or datives, X is in the appropriate case. Genitives can also appear, e.g.

Eric's John's runthan dog

In this case, the dog's speed could imaginably be compared to John's, but the language's grammar assumes that like is compared to like. I.e. Eric's [....] dog is compared to some other dog, and clearly that is John's. With pronouns, genitive forms are not used for proper possession but with participles and with verbs that take genitive arguments. Possessive non-genitive pronouns exist for all persons. To make this more clear, pronouns sometimes are used with the noun extracted. The possessive pronoun is not inflected further for case -

Eric his(poss) John his(poss) runthan dog - Eric's dog, which John's is faster than

Eric his(poss) John his(gen) runthan dog - Eric's dog, which John is faster than

However, such a specific use of a participle may seem somewhat odd, and to justify its existence we probably need it doing way more in the language. So, let us look into what it means without any arguments 'to the left' (let us use that as a shorthand for 'arguments', as opposed to the noun to which the participle is an attribute, and 'to the right' as a shorthand for 'the head noun'; these refer only to the position used in the English-based metalanguage I am using to describe the system).

"runthan John".

If the language conflates participles and gerunds, this might make sense - "runthan John is difficult" - Running faster than John is difficult. "What is your goal in life? Singthan Johnny Cash!"

Omitting the noun 'to the right' generalizes the comparison or specifies it as being somewhat reflexive - i.e. x needs to be better than it is.

In a nominal use the participle 'runthan' does not mean 'the one that jumps the highest', but rather a thing or person who is held as a standard. Thus,

we need knowthan.plur - we need people who are skillfull enough 

As participles, of course, these may encode other types of subclauses and embedded verbs as well -

my livethan this I will be happy - if I survive this, I will be happy

his-gen finishthan, he gets cigarettes - whoever finishes first gets cigarettes 

need chicken.acc dothan - the chicken need to be done better (it is raw)

need house soldthan - the house must be sold at a better price (than offered)

 Some lexemes that are verbs in English, such as exceed, suffice, extinguish, win, survive, outsmart, etc could easily just come up as participles in a language with this type of participle:

livethan = surviving, survival (as in survive a person, but maybe by extension survive the duration of some event)

competethan, finishthan = winning, winning over

smartthan = outsmarting

killthan = killing as many or more of x than there are x, thus slightly violating the 'compare like with like'-state of the language, as this is not 'kill as many or more than x kills'. As stated, voice slightly breaks down if we try to translate this into a single voice, so that's a thing we just have to live with.

lastthan = outlasting, exceeding, sufficing

bethan = outlasting, exceeding, sufficing

Since these only exist as participles, they require auxiliaries when they encode the main content of the clause (or alternatively, the language permits using participles as predicates without auxiliaries.)

These are not finished ideas, but rather hints at where one could imaginably go with a system like this.

Detail #137, pt2

An easy way of implementing these things would be for the relativized-genitive and relativized-comparative-object participles to combine with the embedded role:

John, whose dog terrorizes the mailman

dog mailman terrorize.active_participle.genitive_participle John

For this kind of thing, though, it might seem more natural to construct it something like:

with the mailman-terrorizing dog, John

his mailman-terrorizing dog, John

Which isn't quite a participle that affects John, but rather makes a kind of apposited possessee affected by a regular participle. Thus, we will temporarily leave the genitive-participle for now and return to it later with some slightly odd things to do with it.


And for the comparative object, we get some really nice problems restating this in any nice way. One probably needs a few specific "than"'s - quantitative, qualitative, etc

Phoebe, who no one has solved more challenging calculus problems than

solve-active-participle/quantitative-than Phoebe

Tanya, whom they respect more than anyone else

no-one else respect_passive-participle/qualitative-than Tanya

One option could of course be to compare the participle per se, but this kind of turns the argument structure around: the respecteder Eric ('the more respected Eric' - not 'the Eric who X is more respected than'). Of course, the object of comparison may also be oblique, fucking stuff up even more properly.
However, these are  somewhat ugly solutions, and both can lead to the participle carrying huge loads of arguments - even subjects.

So, in the third installment, we'll try and find some fun things to do with these that partially preserve the idea of a than-participle and a 's-participle without fully enabling *all* the syntactic freedom of relative clauses in English but enabling some fun unexpected new usages.

Detail #137, pt 1: Participles and the relativization accessibility hierarchy

Let us consider the two participles of English - the present and past participles. These clearly correspond more closely to voice distinctions than tense distinctions, really. They also somewhat correlate with relativization in obvious ways.

The accessibility hierarchy of relativization might be a source of inspiration, so let's have a quick look at that!

Subject > Direct Object > Indirect Object > Oblique > Genitive > Object of comparative [wikipedia]

Some of these are pretty obvious as far as being remapped from relative pronouns to participles - notice that this is not a statement about what is possible in English, but about more cross-linguistically possible things, the use of English constructions is a meta-linguistic device in this case.

subject: x, who|which verbs → verbing x

object: x, who|which [is verbed]/[someone verbs] → verbed x

indirect object: x, to whom|which is verbed → verbed x (does not work in all varieties of English, and might not work with all verbs that permit it in those varieties that have it)

oblique: x, [prep] who|which someone verbs → prep-verbed x

At the next point, however, the pattern seems to break down. In part this is because the genitive does not directly relate to the verb in any reasonably sense - the possessee may well be any type of constituent in the relative clause, compare:

John, by whose authority this would never have happened.

Erin, whose dog was a crazed, murderous wolf-like menace.

Jacob, whose children the Egyptians enslaved.

There, we have the possessee in oblique, subject and object positions. I suspect the same hierarchy actually might apply to these as well - as in, a language that permits all the way up to relativizing genitives might have restrictions on what types of role the possessee may have, and the restriction follows the same implicational hierarchy as the usual hierarchy, with the exception that the last item might be further down.

And that brings us to that last item. Objects of comparison. These are just as un-bound to any particular syntactic function as are genitives:

any house in which more people live than in this house has to be crowded

John, whose house this house is bigger than

apples, which pears are sweeter than, ...

beef, which people eat more often than veal

So, genitive- and comparative-relativizing participles seem somewhat odd. In the next post, I will present some ideas as for what could be done with them.

Detail #136: Interrogative Pronouns as Highly Morphologically Defective

This probably exists.

Imagine a language with a fairly rich case system (say 6+), and with next to no difference between what animates and inanimates are marked for. Basically, a completely symmetrical case system (such languages do exist), and next to no traces of a noun class system. (Except in the bit we will come to in a moment).

The personal pronouns may be an exception - with fewer or more cases than the regular nouns, but those are not what we'll look at now. Instead, let's have the interrogative pronouns 'who' and 'what' display some weird behaviors: both entirely lack case marking.

'Who' and 'what' also are the only place in which an animate-inanimate distinction is made in the language. Due to the fairly restricted circumstance in which the animate-inanimate distinction appears, it's not particularly fixed - animals sometimes fall on the animate side, sometimes not, and at times even greater variation happens.

Due to the omission of case marking on these pronouns, they have their own tendencies regarding what cases can be implicitly assumed. These are, of course, statistical tendencies.

Who is more often subject, genitive or beneficiary than what is. What is more often an object. Of course, in the absence of a nominative subject, either of them can be assumed to function as a subject. What also tends to be parsed as instrumental (and thus also as 'how').

Bryatesle Toponymy

Bryatesle toponymy has a few quirks of a somewhat grammatical nature, and I figure the toponymy more generally might be of some interest to conlangers.

Many toponyms have morphemes that correspond to the type of geographic feature involved, much like in your average language. Seldom are these morphemes independent lexemes, however:

-irg (m) = valley, c.f. aren (n), valley

-uxik (m) = bay, c.f. saun (m), bay

-rab (f) = forest, c.f. ramn (m), leafy forest, dugen (f), coniferous forest, kuvaļ (m), forest on a hill, sdas (n), small forest of any type, restricted by geographical things (i.e. on an island or otherwise 

 -xeş (f) = lake, c.f. dinye (f), lage (m), raşņ (n) - different types of lakes

-kik (n) =  hill, c.f. şirp (m), ţarig (n)

When the name is not just one noun, but two nouns in apposition or an adjective and a noun in apposition, the noun forms are used. The noun forms also are generally used when talking about the natural geographical objects as opposed to the places themselves.

Compounds with these can be inflected for case, and also part of larger nouns:

kaŗexeşa pirḑu = market at Heron lake

kar-e-xes-a = heron-(compound form)-lake-DAT market-neut

Towns where significant markets are held yearly may simply be named 'pirdu' with a variety of determiners. The hierarchy of town sizes and significances goes a bit like this:

dirḑu (N) - a village that is only seasonally inhabited 

ţalaņ (M) - village 

birga (M) - coastal village  

ŗiḑkir - a small village in the vicinity of a town 

kimŗa (F, has -a in the dative) -  a village with stationed military presence

vamiŗ (F) - a village with a minor temple (up to maybe 700)

pirḑu (N) - a village or small town with yearly markets or religious festivities that attract significant numbers of people. May be described as any of the previous nouns when the market or festivities is not used, but may also be used to stress the fact that some such regular event occurs.

teņḑ - a medium-sized community with walls, and likely both some military presence and a minor temple. (Numbers of inhabitants between 500 and 5000.)

gri teņḑ - a significant-sized community with walls, military presence and temple. Some administrative significance in the empire's system is also assumed in a gri tend - professional courts, some industries, etc.

One peculiarity with regards to Bryatesle is the use of 'left' and 'right' (şaņud and teņex). This generally refers to sides of a river as seen relative to the downstream direction. If only one side is settled, or significant, it is adjectiveless. Sometimes, different definiteness or possessedness is assumed by the nouns:

byņḑŗiteņ - the rapids (variations on this name occurs along pretty much each river)

teņex byņḑŗili - right rapids

This particular pair is just one such pair - other rapids may be unmarked for definiteness, might be singular, etc - and the other member of the same geographical feature may be unmarked. However, if both sides at the rapids are inhabited, a situation along the lines of

 teņex byņḑŗi(li) - right rapid(s)

şaņud byņḑŗi(li) - left rapid(s)

may occur.  A few place names instead consist of left and right as nouns (tën, san), generally with some nominal possessor or other attribute:

Şgebi teņune - Sgeb's right

Kŗali şaņune -  Kral's left

In this case, these refer to the left and right sides of the valley where Kartun Sgeb's and Mern Kral's armies countered a barbarian invasion. Sgeb held the right side (delineated by the river along the middle of the valley), Kral held the left side, and the barbarians did not break through - both the generals, however, died.

Bŗaḑuņ teņune - left of rock

Bŗaḑuņ şaņune - right of rock

These are two big rocks situated outside the harbour of the capital. One to the left of the usual route, the other to the right. The names maybe most closely would be translated as 'the left (side), (made) of rock, the right (side), (made) of rock'. A much smaller skerry between the two sometimes is referred to, jocularly, as

bŗaḑuņ begune - middle of rock

or rather the 'median made of rock'. A few isles in some rivers also have similar names.

leiḑşţaņ begune - the median of the bend

sţagar begune - town of the middle, (a name for the centre of the capital city) 

More absolutely, compass directions are used - nivix (m) north, maziņ (m) south, gurde (f) east, ņexs (n) west. These are generally in the partitive oblique case, niver, mazeŗ, gurdeŗ, ņeşyr.  The forms given above are rather adverbs indicating 'direction'. Actual nominative forms do not exist. They precede the proper toponym.

It is not particularly unusual for toponyms to use case morphology in nonstandard ways - often the dative or the ablative also are used for subjects with toponyms.

Ḑylemxi - out.dat.pl, 'to the outs', outlying islands or small permanent shelters on outlying islands

Madiţy, maḑuŗşi - coast.abl.sg, coast.abl.pl =  permanent shelters on not particularly distant islands

ḑugna - woods.dat.pl 'to the forests', generally the designation of small permanent shelters for hunting or gathering in forests

kauvux - fight-oblique.part, 'to/of a/some fight', battlefields

In other cases, the marking varies depending on dialect; many dialects simply do not inflect these further, some go and inflect them in regular manners except for the exceptional omitted nominative. The particle used to mark neuters as transitive subjects are used in situations where clarity is required.

A number of placenames are obscure, and originate with languages cognate to Ćwarmin and to Dairwueh.

Detail #135: A set of copulas

Pretty much every conlanger, I bet, knows that Spanish has two copulas. Copying the Spanish model would be an easy way of getting more than one copula, but let us go further than that, and invent an entirely different set of distinctions.

First, let's have a copula that basically correlates pretty well with English 'is' in basic statements of the nature X is Y. Our distinction will not deal with the aspectual or temporal nature of how the subject (X) has its quality (Y) - so not at all like Spanish.

Instead, we'll have a look at how the subject's having its quality interacts with the rest of the sentence.

Let us consider, for instance, comparisons - less than, more than, like, as, in the manner of, ... all these take verbs that are derived from demonstrative pronouns. Which demonstrative pronoun depends on a few things.

If the thing compared to is in the presence of the speaker, and the subject is not, the verb is derived from this:

your dad this_ed skilful you.CASE = your dad was as skilful as you are

If the object of comparison is absent, the verb is derived from that instead.

 you that(verb) beautiful Marilyn Monroe.CASE = you are as beautiful as Marilyn Monroe.

 This also happens when there's no actual object of reference - x is later > x that_s late. The complement can be specified by adverbs signifying 'less' or 'more'.
 The comparison can also be a description - I that(verb) strong to carry it - I am strong enough to carry it. The regular copula never has such descriptions.

Further, we have the causative versions of both the regular copula and the demonstrative copulas. These are formed like regular causatives, but have a slightly different meaning - the subject has a quality and this causes something. This is a bit similar to how 'so' is used in the 'yo momma so fat ...' genre of jokes.

she is.caus beautiful every woman get jealous

This is somewhat similar to the descriptive comparison given above, but tends to be a) more causal, b) much less conditional and way more indicative.

Thus, one central distinction could be pointed out thus:

he is strong - he is strong

he thiss strong - as you can see from what he's doing, he's strong

he thiss strong John.CASE - he is as strong as John, whose strength you can see for yourself

he thats strong - you can, based on previous anecdotes given, conclude how strong he is

he thats strong to carry it - he is strong enough to carry it

he was.caus strong to carry it - he was strong enough and did carry it

he thated strong to carry it - he was strong enough to carry it (but we're not saying he actually did carry it)

he is.caus strong to carry it - he is strong enough, he can (keep on) carry(ing) it

Detail #134: A language with very reduced role marking and reconstituting the role marking in a weird way

Imagine a language in which there is no 'preferred' voice. However, there are any number of circumstantial voices (voices that promote an oblique argument to subject). All non-subjects are marked in an identical fashion. An additional tendency is that indefinite nouns go to the right (unless they are subjects).

Thus, say,

'mom cooks beans in the new pot' -> se mom cooks.active e new pot e beans,
'mom cooks the beans in a new way' -> se mom cooks-active e beans e new way
'the beans are cooked by mom in a new way' -> se beans cooks-passive e mom e new way
'the beans are cooked by mom in the new pot' -> se beans cooks-passive e new pot e mom
'the new pot is cooked-in beans in a new way' -> se new pot cooks-some.circumstantial-voice e beans e new ways

etc.

Contextual information and lexical information (about what types of nouns tend to appear with what kind of role) will thus be very important for unerstanding an utterance. Of course, longer utterances may be easier to parse due to more hints at who is doing what. One could imaginably also have adjectives that are basically participles-with-voice that co-refer with the verb, and permit disambiguation when needed:

'mom cooks e beans e new in-cooking pot'
'mom cooks e being-cooked beans e pot'

etc.

Here we get an interesting possibility - pro-verbs! Maybe different main verbs take different markers such as this, also permitting us to conflate several verb phrases.

'dad hunts and cuts up and gathers, mom tends and cooks e done-passive* deer, e had-passive** used-passive**** chickens, e taken-passive*** used-instrumental_circumstantial**** herbs
* referring back to "hunts" or such
**referring back to "tends"
***referring back to "gathers"
****referring back to "cooks"

Of course, different verbs have different "verbal class", and thus different auxiliary-participles will refer to them, helping to correlate nouns with the role they have for different verbs. Nouns whose use is expected to be understood without such an auxiliary participle are left unmarked.

At this point, I've basically ruined the idea this post started out as, and built a typologically quite unlikely thing. However, I like the looks of it.

Detail #133: A Reused Thing from Germanic

Consider how Germanic languages (and some other languages, for that matter) deal with impersonal constructions:

it rains

it's late

it's clear now that the plan couldn't have worked

it kept getting colder outside

Germanic languages that have V2 (and English, that had V2 once) might have come up with having a dummy pronoun there so that the verb could remain the second constituent of a clause. Now, other languages of western Europe also have gone through stages of V2 (French and Spanish, for instance). I do not know how other V2 languages deal with it, nor am I aware of to what extent dummy pronouns appear in other kinds of languages except by influence from V2 languages (i.e. many Finnish dialects have 'se sataa' instead of standard 'sataa' for '(it) rains').

How about having a variety of dummy pronouns from different genders in a more Bantu-like gender system, and having these be lexically determined by the verb or by the construction used?

Typological Questions and Attestation

Every now and then, people post typological/universal-related questions on conlanging fora and groups, such as 'is so-and-so reasonable/possible/attested in languages'.

Oftentimes, people will respond by pointing to languages whose names you've never heard. This is of course not all that much of a problem ­­- the fact that I had never heard of Choroté before starting this article does not make Choroté less relevant for understanding all the wonderful things human languages can do.

Some other languages that have the same property - viz. that I had never heard of them before starting this article are less relevant for understanding all those wonderful things, however. Two of them are Mis Hio and Rangyayo. What makes them less relevant? They are conlangs. They tell us what a conlanger can imagine, not what is possible within the range of what a human speech community can maintain over generations.

So, when answering whether so-and-so is possible, think a bit. Either make it clear somehow that it's a conlang you are posting about (less known real natural languages might benefit from having their geographical location or family included), or even try to figure out whether the question benefits at all from an answer that includes conlangs.

The fact that a conlang has a feature only means the feature can be imaginable and is not somehow self-contradictory. I could very well construct a conlang in which the phonemes /e/, /i/, /f/ and /r/ have associated integer values, and all well-formed words for different word classes or different forms have the sums of the numbers associated with those phonemes add up to something, modulo something else.

This is possible, but it doesn't inform us whether humans actually can learn and speak such a thing.

Thus, me pointing out that my conlang Blarghhargh has that feature is not helpful at all to understanding anything about real languages. Let's grant whoever asks a question of typological nature some consideration - let's evaluate whether what our imaginary languages do is relevant to whoever asks the question. Your conlang is not an attestation of anything than the power of imagination. And if you post about your conlang, try to make it clear that it's a conlang - no one can know the names of all the conlangs, and no one can know the names of all the natural languages.

Let's help other conlangers keep their typological real-world data untainted from imagination.

Dairwueh: Some Morphological Tables

The Dairwueh verb is not excessively synthetic:

Finite Verbs with Congruence
	Isg	IIsg	IIIsg	IIIsg2	Ipl	IIpl	IIIpl
indicative, present	-as	-er	---/-o(fem)	*	-uan	-uše	-uni
indicative, past	-is	-eb	-iŋ	-e	-ad	-abe	-ari
affirmative irrealis	-əym	-əur	-əyi/-əvo	-əy	-əgan	-əgaš	-əŋan
negative, present	-šna	-(ə)vne	-šne	-š	-ušen	-ušen	-šne
negative, past	-eyš	-eyš	-eyš	-š	-eyšin	-eyšin	-eyšin
passive, present	-ŋor	-ŋor	-ŋa	-ŋ	-ŋan	-ŋan	-ŋa
Periphrastic Passive forms
passive, neg. pres.	erb- + passive neg participle
passive, irrealis	ŋey- + passive irrealis participle
passive, past	ŋe- + passive affirmative participle
passive, neg.	erb- + passive negative participle

The final consonant of the infinitive (and possibly the vowel before it) is removed, and the suffix is added. Notable here is that the realis distinguishes {past, present} x {positive, negative}, whereas irrealis distinguishes {affirmative, negative}. 

Infinitive:  no particular morpheme, but almost exclusively ends in consonants.

The negative does not distinguish realis from irrealis, but does distinguish past from present.

* is formed in a rather different way - if the last syllable of the infinitive is bimoraic (a post on moras in Dairwueh will appear at some point), the IIIsg2 is identical. Otherwise, the IIIsg lengthens the last syllable. The rest of the III_sg² forms are suffixed to that stem.

The participles are formed as follows:

active
	affirmative
		present	-un
		past	-ar
		irrealis	-umuš
	negative
		present	-šun
		past	-eyš
passive
	affirmative		-šəŋ
	negative		e-___-šor
	irrealis		e-___-šis

The negative and irrealis participle when inflected for cases other than the nominative change slightly: -šor and ₋šis shorten to -šr- and -šs- before suffixes that begin in vowel, e.g. ekanšsar - 'to he who may rise'.

Case forms

	masc sg	fem sg	neut sg	masc pl	fem pl	neut pl
nom	*	-(r)i	-e/-a	*	***-(t)a	-a (a few nouns have -e in both sg and pl)
acc	-na	-nu	-e	-ivna	***-(t)ar	-a
dat	-ar	-(r)ir	-n	-ivit	***-(r)it	-ivit
gen	-at	-ra	-ŋa	-ŋa/-edin**	***-(r)in	-ivit
loc-instr	-ŋa	-at	-ŋa	-ŋa/-eder**	***-(r)ar	-ŋa

* the masc sg and pl nominative have a plethora of lexically determined suffixes. Some nouns even permit a variety of suffixes for these forms, forms that sometimes convey extra information. The -iv-/-ed- part of the masc pl suffixes and the -er- part of the fem pl suffixes can be replaced by such suffixes as well to convey the same information.

**-edin/-eder are exclusively used with nouns referring to humans. 

*** a similar lengthening as that which happens with infinitives takes place before applying the suffix

The adjective will be laid out in a separate post.

Toponomy Shenanigans

A treatise on toponymy for a conculture could be interesting, and could offer some interesting possibilities for coming up with ideas to develop regarding the history of that conculture. Some small hints that can lead to some interesting results:

• post-glacial rebound can readjust coastlines quite drastically over time; nautical designations (bays, sounds, beaches) easily remain even after the land's been dry for centuries. (I know this because I grew up in such a region; I don't even react if a place whose name means 'thislake' or 'thatbay' is completely devoid of water and more than 20 miles from the closest coast. If a hill goes by that name I'd be a bit concerned, though.)
• at places where the coastline is receding, hills will remain as islands longer than surrounding land, thus names referring to hills may appear in an archipelago.

Of course, population movements often also leave behind all kinds of incomprehensible roots that can survive for centuries. (C.f. the old European hydronymy.)

Further, what morphemes do you use? Would 'left' and 'right' as common parts of names make sense? Does ownership have a significant influence? Inheritance customs, maybe? Are people named for places or places named for people? Is X-on-Waterway or X-by-Road or somesuch common? (They seem to be common in France, Britain and the Germanic Continent, but less so in Scandinavia.) Do you have something like 'Upper' and 'Lower' and what do they signify? Need not be genuinely vertical distinctions!

Is the presence of some type of religious building significant in the categorization of hamlets, villages, towns and cities?

Barxáw: Prepositions as Pragmatical Markers

Barxáw uses its prepositions with its verbs to tie together an utterance. We will first look at the six most important prepositions and their meaning when used with NPs. Some other prepositions have previously been introduced (kur and jà, k'e), but these are not used with verbs.

sún at, by, on

ðo  for, on behalf of, in front of,

gìλ from, about, connected to or related to in some sense

tùð above, on top of, around, covering, the origin of something in a non-motion sense (thus is a contrasting pair with gìλ)

sít under, by, at, inside, by permission from, in 

kàw along, with, as, than

When used in a pragmatical function these are placed just before the verb. 

Paiǵim c'arr nugu dór. Ekxíŋ sún hìlge dór tùð Ugin.

Paiǵim refuse to sell cattle. Ekxiŋ therefore acquire cow(s) from Ugin.

Ráð san tɛ̀n. Gérru kàw xwíu t'áć.

Rain pour much. Harvest soon grow big.

The first and/or the last verb in a unit of thought is often marked by one of these prepositions, and this tells us how it relates to the previous unit of thought or the next one.  There's slightly distinct meanings associated with these different locations:

initial:

sún - just generally introducing a statement. This basically means that this is a central part of an utterance, or that the unit of thought is an independent utterance. 'ku sún ebah', 'we are here to eat (we booked a table)'. Answers to questions, neutral introductions of intent, etc.

 ðo - emphasizes the verb. 'ku ðo ebáh' - 'we for eat' - "We're eating, for crying out loud, (don't talk about this or that)", alternatively 'we're here to eat! (so let's eat!)', '

gìλ - marks that there is coming more that might contrast with or add to what is said. 'ku gìλ ebah', 'we are eating (and maybe also having our drink on)'. The added stuff may be implicit or might be explicitly stated. 'ku gìλ génð ebah , ku káw dećà.' '(we have eaten, and want to pay now)'

túð - marks contextual information / exposition. 'ku túð ebah, ke Ðiám ke í bané is adbè. ...' 'We went to eat, (see), Ðián had birthday, ...'

káw - marks that something contrasting is probably coming up.

sít - marks either speculation or counterfactual/irrealis reasoning. 

The two last ones also have the same meaning when at an utterance-final verb. However, the four others have a somewhat different meaning.

Sún affirms strongly the accuracy of the utterance or the causal connection between the previous sentences.  Ðo basically has a function like '...and that's why so-and-so' - it doesn't affirm the conclusion (unlike sún), it affirms the causes leading up to it. It's truth-affirmation is a bit less, more along the lines of 'or so I reckon'. Gìλ ends a list of a series of events, and codes for 'that's what I recall, at least'. Túð marks irrealis stuff. 

A Link

Speculative Grammarian (and other related zines archived on the same site) is one of the finest linguistics satires. The occasional articles also delve into miniature conlanging, although for satirical purposes obviously.

One of my favourite articles deals with numbers in the Moundsbar language.

However, reading that article reminded me that there's a site elsewhere with lots of notes on Moundsbar, and some other linguistics thrown in as well.  These are classics of their genre.

Ćwarmin: Complements

Although Ćwarmin does have a verb that can serve as a copula, it is not used very often.

The copula is invariant with regards to subject person. In the present, it is au. In the immediate past, it is amca, in the far past its aun. There is also a negative version, ainik, amśik, aunik.

The following circumstances call for a copula:

• an adjectival complement that is not in the nominative-complement case

Ćwarmin adjectives can have their meanings altered by case suffixes. The particular changes in meaning are described here. Nominative complements signal prevailing or constant qualities of something. Even nominative complements require a copula.

• a complement phrase that is headed by a postposition
• a complement that is a heavy phrase, a phrase that has an embedded verb or consists of lots of words with some structure
• Confirming or affirming a statement about something. Pəćitiś dostur au - spirit-plur.acc.def pay.homage-past-1sg is - I do pay homage to the spirits.
• often, the negative copula is used as a negation particle even in transitive utterances.
• when the tense is relevant and not easily inferred from context
  
• with an implicit quality, or a coordinated or contrasted quality (Ešpək suŕurŋaca l Tuśu au. Eshpək is a widower, as is Tusyu.

Normally, the complement goes sentence-finally if no copula is present. If a copula is present, though, the complement precedes it.

Other verbs that take complements are

dəšip - to be named. The name is generally in the complement case, although the verb also can signify membership in a group. In the latter case, the complement is in the general ablative. The negative case can signify both not being a member or not being called something.

xuvop - to become. Complement case signifies becoming, instrumental signifies intensification for some already present quality. Negative signifies cessation of some quality.

daval - to considered oneself something

Object complements generally only are in the object complement case or the accusative case. Therefore, transitive verbs with complements often have more related forms that code for things that cases would code for with the intransitive verb, i.e.

davlap - to consider someone something (complement goes closer to the verb than the object does)

davlakol - not to consider someone something