Detail #187: Onwards with #182

Let us consider idea #182 and add some complications. Let's posit that the language further has some verbs where the subject has exceptional prepositions.

Thus, all the following pairs of structures signify the same subjects and objects, i.e. NOUN2 is an object in all of them, NOUN1 a subject. The / bit only serves to remind the reader of this, and is not a marker that is present.

NOUN1 hits NOUN2
NOUN2 hits subj.prep. NOUN1


NOUN1 has NOUN2
NOUN2 has at NOUN1

NOUN1 sees NOUN2
NOUN2 sees for NOUN1

Tatediem: The Noun Classes (Obsolete Post)

This post is obsolete

Although the noun class system is pretty stable in most of the languages related to Tatediem - Tatediem itself belongs to one of the branches where the system has actually tended to wear down. Tatediem itself is conservative for its branch, keeping a number of classes - some languages that have spread into even closer proximity of Dairwuo-Bryateslean languages, and even more so Ćwarmin-Ŋʒädär-languages, have reduced the noun classes even further. 

There are basically seven noun classes, each of which has singular, plural, dual and mass forms. The least "typical" class is the grammatical class, which mainly is used on nouns with various grammatical functions rather than actually denoting referents. Two other classes are familiar from Indo-European and Semitic as well: feminine and masculine.

These contain some non-human, even inanimate nouns. This, mainly, seems to be by loans from more northernly relatives, where the gender system has been further reduced.

The proto language for the highest branch to which Tatediem belongs had something along 34 noun classes - counting singulars, plurals, duals and mass nouns each as separate classes. There isn't clear one-to-one relationships between these classes: some singulars have nouns appear in four to six different plurals, some plurals catch nouns from multiple singular classes, etc. The dual-plural relation seems somewhat stable, though. 

A graph illustrating the relationships between Proto-Tatediem noun classes is given below. Numbers 1 to 10 are singular classes, 11 to 15 are plurals, 21 and 22 do not distinguish plurals from duals, 24 and 25 are plurals for which dual marking is impossible (because the nouns tend to appear in large groups and no one would want to single out two out of those groups), 26 is uniquely dual (and marks some pretty specific nouns, and for many of them the plural marker has very peculiar meanings: such as 'my parents' or 'a pair of siblings (inclusive) vs. (exclusive)'. 28 through 34 are mass markers.

Coloured lines are to be interpreted as follows: nouns of class 1 or 2 with 28 as their mass class have 11 (and 16) as their plural (dual) class. The blue line with 2-11-17 are for nouns that are class 11 and 2, but lack  class 28 altogether - they have class 17 plurals. 32-14 implies that class 5 nouns with class 14 plurals have class 32 mass marking (or vice versa), and the same goes for 33-22 with regards to 6. The dotted lines that appear here and there are not very common, but have at least a dozen or so nouns. Examples of odd class-jumpers can be found, e.g. náhaŋ, "goose", which goes through classes 29-5-13-20 and also class 32 as goose-meat. However, such oddballs have been largely lost in Tatediem.

1 is primarily masculine, 2 primarily feminine. 28 denotes groups of humans - i.e. tribes, occupations, nationalities, degrees of holiness or status, etc. This may also create abstract nouns such as "kingship" or even "political power".

The bit around 1 and 2 has remained fairly stable in Tatediem, but also creating a complete parallelism between feminine and masculine.

28A - 1 - 11 - 16
28B - 2 - 12 - 17

Classes 3 and 4 (with attendant classes) converged in Tatediem, merging 29, 30 and 31 into a single mass class, and 13 and 14 to a single class - with the same effect in 18 and 19. These are then Neuter 1, singular, dual and mass. Some traces of the different mass classes can be found in retentions of tonal distinctions in the first syllable of certain words, due to the previous class prefix for 29 having had a "migrating" tone, that got expressed on the first syllable after the prefix,, therefore now written dwe.`- in reconstructions. Some class 5 nouns with class 15 plurals migrated to Neuter 1.

Some nouns in Neuter 2 retain class 6's lack of a separate attendant dual class.  Other than that, it is mainly the result of 5 and 6 (with attendant classes) collapsing, with just a few class 4 nouns migrating, and about a third of class 7 (with attendant classes).

The remainder of Class 7, as well as classes 8 and 9 (with attendant classes), which previously had a rather large set of vocabulary in them basically very shuffled into two classes, one keeping some class 7 and 8 morphology, the other some class 8 and 9 morphology. The two new classes were likewise "reshaped" to the same singular-dual-plural-mass structure that all other classes had - these are classes 3 and 4.

Finally, the 10, 26, 23 and 27 classes were spread out over the the other classes. Remants of these can be found in formerly class 26 nouns having high tone in the first syllable of their stem in their plural forms even if there's other tones there otherwise.

This leaves the types of nouns prevalent in the different classes rather open-ended:

Neuter 1 has many animals (from classes 3, 4 and 5), some religious festivities, possessions of status, clothing, body parts
Neuter 2 has some animals (from classes 5 and 6), many tools, many plants, spans of time, rocks and minerals, cuts of meat, vessels for seafaring or for travel on land
Neuter 3 has things of geographical extent, abstractions based on geographical concepts, weather-related nouns specific to the drier part of the year, some tools (mainly of wooden type),metals, fruits
Neuter 4 has things of of geographical extent, weather-related nouns specific to the wetter part of the year, many house-related nouns, metallic tools, wooden materials, diseases, containers

Liquids and powders seem spread over all classes' mass number, with individuated bodies of the liquids in the other numbers.

The grammatical class seems to have appeared out of the use of class 10 markers for certain grammatical particles.

The subclasses of 1 and 2 that are formed with ŋwu- and ku- prefixes seem to be a development that has happened after proto-Tatediem, but the isogloss for it does not cut neatly with any branch division in the larger family, and is thus probably a regional feature.

Tatediem: Relative Clauses

Relative clauses in Tatediem use slightly different constructions depending on what argument is relativized. Relativized subjects are fairly simple: replace the subject prefix with an object prefix - or a possessive prefix for animate, transitive subjects.

For other arguments, there is a wackernagel particle that takes the object congruence of the relativized argument on it, -byim. In addition, the verb may take agreement for the grammatical class, either in the subject or object slot, depending on which one is free. The morpheme always is the form usually used with objects.

Verbs are turned into infinitives and gerunds by use of the grammatical class as well. Infinitives can carry subjects as well as objects. Both have objective prefixes, however, and there is a manner-prefix -ngkia- that signals infinitiveness. So essentially the template is gram-subj (as obj)-obj-ngkia-stem. An empty subject or object slot is normally not marked, but a strategy for distinguishing which is omitted exists - restating the subject as owner of the gerund or infinite, or for objects having -byim appear with object congruence after the gerund or infinitive.

Tatediem: Descriptive Comitative Possession

In Tatediem, it is possible to form phrases along the line of "with his arms bare" using the following pattern: cong₁-ADJ def₂poss₁-NOUN₂ where 'cong' is the congruence class marker of the possessor, and poss- the possessive prefix of that class.

This construction is most common with the kinds of nouns that typically are inalienable, but not restricted only to them. It is maybe more similar to the Latin version, nudae lacertos. Syntactically, the closest in English would be something like 'nude of the arms',although that seriously scrambles the Latin cases.

Ćwarmin: A Regional Quirk in Forming Large Numbers

Most Ćwarmin dialects have calqued their way of forming numbers beyond a thousand from Bryatesle. However, in the far west, there is a community that went through a relatively rapid modernization of their economy with very little contact with either Bryatesle or Dairwueh, but rather with several other groups. This rapid modernization also brought with it the need for an ability to form bigger numbers, and since none of their trading partners were particularly dominant, they evolved one of their own instead of borrowing it.

Notice that the dialect has some sound changes as well as some lexical differences from the version I usually describe. Below I give examples for the tens and the thousands.

Observation: nine divided by six is one and a half.

dusso = ten
eyse tara dusol = one ten-part = 1.5*10, rounded up = 20
mey tara dusol = 2* 1.5 * 10 = 30
mey tara e dusso = two ten-parts and ten = 30 + 10
sićey tara dusol = three parts, rounded up = 50
nurwa tara dusol = 60
nurwa tara e dusso = 70
meŋgə tara dusol = 80
seŋgə tara dusol = 90

kurcuw = thousand
eyse tara kurgwal = one piece of thousand = 1.5 * thousand, then rounded up = 2000
mey tara kurgwal = two pieces of thousand = 3000
mey tara e kurcuw = two pieces and thousand = 4000
sićey tara kurgwal = three pieces = 5000
nurwa tara kurgwal = four pieces = 6000
nurwa tara e kurcuw = four pieces and thousand = 7000
meŋgə tara kurgwal = five pieces = 8000
seŋgə tara kurgwal = 9000

So essentially, "ten with a sub-base of rounded-off 1.5". The same pattern for hundreds can be obtained by substituting dusso with peknə and dusol with pektəy. A smaller number can be placed to the right of these numbers, i.e.

sićey tara dusol sićey = 53
three part ten three

Sometimes, larger numbers appear on the left hand side, giving, for instance

seŋgə dusso tara dusol = 16*1.5*10 = 240.

The rounding is always upwards.

Detail #186: Inverse Habitual Marking

Imagine a language where some verbs are inherently habitual, and some inherently punctual or momentane or such. The same marker serves the role of deriving the other meaning. However, with a few verbs, the verb itself carries no inherent aspect. The aspect is instead derived from contextual cues, and the aspect marker operates with regards to whichever aspect the contextual cues implied. 

Further, the language does have other aspectual, temporal, diathetical* and modal affixes. These are also contextual clues - but the position of the inverse habitual marker with regards to the position of other morphemes in the verb complex can influence the parsing as well.

Some important clues:

• with generic plural subjects (i.e. a species or class of things taken as a whole), habituals are always expected, but the interpretation is then that e.g. 'members of this class do so and so', without any necessity that the individual do it habitually - just that it's a trait of the class to do so at least once.
• passive voice tends to imply non-habitual, except with a handful of verbs (sexual ones, food-related ones and work-related ones)
• past tense in combination with perfect tends to imply non-habitual
• past tense in combination with imperfect defaults to habitual, except with a handful of verbs (for definite subjects, ones that can only be carried out once or are unlikely to happen often in a lifetime: be born, die, lose virginity, be blinded, succumb to plagues, drown, marry (for women subjects)
• if a person is mentioned with his social class or title, it is more likely to be parsed as habitual
•  ... definiteness, social views, types of action, etc, may all influence this; I see too many possibilities to start listing them all, and if I listed them all I'd end up with every verb having this. Come up with your own lists!

The combination of future tense and some evidential is parsed as follows:

X verb-fut-evidential  → I think X will verb in the future, because he habitually verbs already

X verb-fut-inverse-evidential → I think X will habitually verb in the future, because he habitually verbs already

The difference in meaning here of course is subtle: the first example simply deals with some particular future time, the latter with future taken more widely. 

* voice-related

A Review: The Indo-European Controversy: Facts and Fallacies in Historical Linguistics

A Review: The Indo-European Controversy: Facts and Fallacies in Historical Linguistics (Asya Pereltsvaig, Martin W. Lewis)

A few years ago, a team of researchers lead by Russell Gray and Quentin Atkinson presented a mathematical model for the spread of language families. Applying this model in reverse to the Indo-European languages supported the Anatolian hypothesis, a minority position on the location of the Indo-European urheimat.

For some reason, this was widely published in media, and the paper Mapping the Origins and Expansion of the Indo-European Language Family appeared at the same time in the journal Science. Gray and Atkinson have made very vocal and powerful claims about their findings: 'decisive support [for the Anatolian hypothesis]' is among the various things they have said about their own work.

Pereltsvaig and Lewis go over the results in depth, and find them highly lacking: they find numerous problems in the geographical spread that it presents, including multiple instances where the sanity of the model is excruciatingly questionable. They present the evidence we have against the Anatolian hypothesis (and even more clearly, the evidence we have against the Gray-Atkinson version of the Anatolian hypothesis) and all the difficulties it brings with it, as well as the evidence we have for the Steppe hypothesis.

They also present the evidence that has lead most linguists to accept Steppe Hypothesis instead. 

The argumentation is persuasive and clear, well-nigh undeniable. This leads to an important question: how did Science let a paper that is so rife with unsound historical linguistics pass peer review? It turns out that linguists did peer review it, and Science ignored their judgment, because their negative comments did not pertain to the maths of the model - clearly, having a mathsy model is a guarantee that the mathsy model is correct in Science's view?

Publications such as Business Insider either repost bad science from the Gray-Atkinson team, or add their own even worse spin to it. Consider their version of the Gray-Atkinson animated map. This is, allegedly, how "Language" spread across Europe. In linguistics, "Language" signifies the general phenomenon, the fact that humans can communicate in a complicated system. So if we are to take Business Insider's video title seriously, this is how the ability to speak spread in Europe, and all the current language families were the first languages spoken in their areas. 

Pereltsvaig and Lewis point out a very real problem: other scientists apparently do not take linguistics seriously, and we are facing a rise of armchair philosophers who disdain empiricism in favour of cute models (at least when going outside of their own field - i.e. Gray and Atkinson probably understand how to be scientific in their own field, but when working with language, the computational model seemingly blinds them to empirical facts). 

This, in turn, is coupled with the modern phenomenon of clickbaiting, where the most attention-attracting claim is more likely than other claims to pull in ad money, and thus scientific claims are propagated online not by their likelihood of being accurate, but by how tittilating they are. This is a genuine problem, and needs to be curbed.

Pereltsvaig's and Lewis's book is less combative than this review, although at times it does take vigorous swings at the Gray-Atkinson teams publications. It is a good read, and gives a lot of information about historical linguistics and especially Indo-European historical linguistics. A certain glimpse into issues in the philosophy of science can also be gleaned. It is well written: both clear, enjoyable and relevant.

Ćwarmin: A Detail about the Ćwarmin Infinitive

I have previously given a short overview of the Ćwarmin infinitive, but more details about the use thereof will sporadically appear.

The infinitive is not the same as the verb stem, but has some infinitive suffix on it. There are several infinitive suffixes, and some of them have different implications, cf:

hacam - hacan - haćjul - to think

rigən - ridjel - ridjen - ridjin - to hurry

 The {hacam, hacan, haćjul} triplet shows three different suffixes, all with slightly different connotations: hacam - to believe something, hacan - to think about something, haćjul - to plan. The finite verbs conflate these. The {rigən, ridjul, ridjen, ridjin} four-tuplet distinguishes rigan - to hurry in regards to some thing, to be anxious for something, ridjul - to hurry with the intent to achieve something, ridjen - to be in a hurry so as not to miss a thing, ridjin - to hurry at the expense of the results

Now, these infinitives can appear sentence-initially as a topicalized ~adverb to specify which particular meaning is intended, while a finite form of the verb goes sentence-finally:

hacan bec terəś hacac? - what are you thinking about?
hacam bec terəś hacac? - what do you believe?
haćjul bec terəś hacac? - what do you plan to do?
bec terəś hacac? - what do you ... [any of the above]?

This is not a very common usage, but occurs when confusion is overwhelmingly likely or specifying which meaning is crucial.

Detail #185: Inverse Number and Pronouns

Inverse number is a pretty neat thing found in the Tanoan family. Each noun class has an associated number or two, whose marking for that class is zero. The remaining number(s) are marked by a number marker.

Now, with personal pronouns this could open up some fun: of course, the simple way would be to have personal pronouns either follow some number class, so e.g. "I" is unmarked and "we" is marked. We could also, of course, leave the personal pronouns out of it altogether. However, fun could be had:

Let's have both singular and plural pronominal roots:

er : I
mur : we

Let's have the inverse number marker -xi, and have this apply to both of those as well:

erxi : we
muxi : I

There should be some subtle difference though: maybe erxi expresses my role in it as that of a representative of some group, whereas muxi represents my group as an extension of my will - i.e. I and my army, or the like. 

Another alternative could use these for deference and the like.

Many Tanoan languages seem to have classes that, for instance, mark dual with the inverse marker, and singular and plural both go with the zero marker. Tanoan languages also have a single number be the default, marking two numbers with the same marker. There is no guarantee that the marker or absence thereof only goes on "neighbouring" numbers.

If the language further has a dual number, a dual pronominal root could make stuff even more fun, but there may be parsing restrictions there, such as 'dual.inv' only parsing as singular (or only as plural, it's up to the conlanger, obviously) on pronouns. The parsing may of course be different for different persons as well (and different for pronouns of different noun classes too!)

Detail #184: Head-Marking Taken a Step too Far

In a rather materially limited culture where also some form of incorporating polysynthesis operates, it is imaginable that some verbs may develop suppletive forms for certain verb-object or verb-subject combinations. For certain things, it might even be possible that there never was a noun, or that the noun has been lost, and the only way to refer to some thing or class of thing is by a limited set of verbs that mark that class by suppletion.

Would we notice if this already has happened in the languages we speak?

Bryatesle: Universal Quantification (and the Scope of Negation)

Universal quantification in Bryatesle has some peculiarities with regards to scope, at least when compared to most European languages. "Universal quantifier" is basically fancy for 'all' (or 'each' or 'every', although some differences exist between them; turns out 'each' and 'every' are more similar to each other than to 'all', but no more on that for now). Consider what happens with negation:

not all men are tall

I do not have all the volumes of that work

If we want to make a statement where we actually deny having each and every one of them, English requires using a different pronoun altogether -

I do not have any volumes of that work / I have no volumes of that work

no men are tall

In Bryatesle, it is possible to convey both of these meanings by use of negation and 'all'. These are distinguished by the use of the negative congruence case marker. The word 'all' is bar in the least marked form. It is inflected as follows in the plural (the use of bar in the singular is somewhat different, and is not dealt with it in this post) :

pl:
nom: bares (bara with feminines)
acc: barku (bares with neuters)
dat: barsa (barse with masculines)
abl: barti (bara with neuters)

With negation and no negative congruence marker on the head noun, it is parsed as the first examples in English:

da bares keng-er sdruf mii
not all       men-pl   tall    stand-3sg.animate
but not all men are tall

This is parsed as a statement that does not rule out the existence of tall men, but it does assert the existence of at least some who are not tall.

However,

da bares keng-ute sdruf mii
not  all      men-pl.neg tall stand-3sg
but no man is tall, but all men are untall

It is common for negated plural 3^rd person subjects to have a singular verb form. Also, as might not have been stated very clearly in posts about Bryatesle this far, it lacks a verb that directly corresponds to 'to be', but rather uses other intransitive verbs that often convey some extra information, such as the form of the quality expressed (i.e. tallness is upwards, so one stands tall, etc.)

Not all dialects use the partitive as the negative congruence marker - in fact, some have a suffix -tav instead. Dialects that entirely lack a negative congruence marker exist, and they have dealt with this in different ways, i.e. generally having some other case marker on the noun or even on 'bares' to distinguish how the negation affects the quantifier.

Detail 183: Multiple Lexically Conditioned Aspect Markers of Varying Syntactic Type

We're used to relatively predictable aspect markers of a somewhat clearly "unified" type. Let us play with a slightly less unified system!

One obvious way from which we could derive perfectiveness could be a dummy object. I.e. 'we ate' → ~we were eating, 'we ate the food' → ~we ate (perfective). For transitive verbs, 'food' might appear as a determiner for the object:

we ate moussaka → ~we were eating moussaka
we ate the food moussaka → ~we ate the moussaka

Now, different verbs take different nouns like this, and they are grammaticalized and generally speaking monosyllabic.

we walk demonstrative-lative → ~ we arrive

we walk → ~ we are currently walking

We can also have other noun types:

we eat our fill, they received their portion, etc

In such constructions, the aspect particle gets person congruence!

Sometimes, verbs serve this role: we speared killed the animal.

Now, we can imagine a system where aspect is marked by lots of lexemes in a varied and wildly unpredictable system, where also different subjects may trigger different markers, i.e. 'we ate the food moussaka', but 'the livestock ate the fodder grass', so you can have multiple interacting implications of different strengths.

Detail #182: Tools, Verbs and Culture

Let us consider a culture where certain kinds of tools are in some use, but their inner workings are not well understood by most who use them. We may have locks, butter churns, fire stones, compasses, lenses and other similar things belonging to this category.

Now, it'd be somewhat easy to ascribe some manner of agency to such tools, and this could easily be reflected in verbs pertaining to their use. I.e. a lock being locked might be formed by a reflexive verb with the lock as its subject, rather than a transitive verb where the person who locks it is the subject. Maybe all verbs pertaining to these are reflexive, take direct objects as obliques and the user of them as dative.

Detail 181: An Adposition

A simple, subtle, yet slightly unusual adposition ­­­­– whose existence in some actual natural language you can take for granted – that struck me as an idea today, and which never has occurred to me previously, is this:

An adposition that marks subjects, but only if they're in unexpected positions with regards to word order.

Where this might get interesting is if the language has transformations that change word order, i.e. fronting the verb for polar questions. Should the new position be considered the expected location of the subject in questions? Both ways seem reasonable: verb fronting dislocated the subject, but with regard to the clause type, the subject is very much where it should be (it's rather the verb that has been dislocated).

What further could get really interesting about this is adposition stranding, as well as intransitive use of the adposition (i.e. no noun in the entire clause which could be its 'subject' so to speak). It could of course also have some kind of restriction like never appearing with intransitive verbs, making the language slightly more ergative – or never appearing with existential verbs? Or some set of verbs with which it appears even when the subject is where it should be.

Further, one could imagine it being used with nouns uttered in isolation, e.g. when used as answers to questions and the like. Maybe some pronouns have incorporated it into their forms, or maybe pronouns are immune to its effects – i.e. never are marked by it.

Large-Scale Conlanging: Options to Universal Hierarchies

If you want to go full-on megalomaniac with regards to conlanging, how about designing the language universals (both absolute and universal) for the world of your conlangs!

However, we could start out with some things that would end up somewhat different from human languages (or maybe these things exist, but no one's come up with the idea of analyzing things in terms of these exact structures?) Let's first look at hierarchies.

Hierarchies are a popular thing in universals research, and a good one is the relativization accessibility hierarchy:

subjects ← objects ← ...

But how about cyclical hierarchies? Here, the implications would need a restriction - a circle of "true implications" would obviously lead to either all features or none of them. However, let's use some other "operator".

A₁ ⇴² A₂ ⇴² A₃ ⇴² A₄ ⇴² A₅ ⇴² A₆ ⇴² A₁ ...

The superscript 3 signifies the length of the implication chain. Basically, ⇴^x should be read as 'if a language has A_n, but not A_{n - 1}, then it also has A_n+1 and A_n+2 and ... and A⇴^n+x.

Notice that implication only goes one way: if we're dealing with ⇴² and the hierarchy described above, the language could very well have A₁ to A₅ - it only tells us how short the chain at least is, not how long it maximally is.

A language could also imaginably have several discontinuous bits of a long universal cycle. We can also imagine different superscript-operators in different parts of the hierarchy. An extra possibility would be 'bolded' operators as well - if another chain of properties reaches a bolded operator, it is forced to continue at least as long as the bolded operator says. (Alternative, there could be bolded and nonbolded superscripts - bolded ones apply for long chains, non-bolded ones for a chain that is within its 'natural' reach.)

Detail #180: Nasal Affricates

Consider the articulation of an affricate - a stop with a slow release, such that the release essentially produces a fricative. 

Now, what prevents us from having a nasal airstream during the time of closure? (It will weaken once the closure opens a bit, though, so that bit actually helps us a bit in turning this into a meaningful thing)

/n͡z/ or even /n̥͡s/ could be pretty neat phonemes. These seem realistic enough - at least nz - to actually maybe exist somewhere, so if someone knows whether they're attested, please comment!

Detail #179: Connegative verbs and names

Some Uralic languages have a set of verb forms called 'connegatives'. Generally speaking, these languages also have a negative auxiliary (much like English "don't/doesn't", but inflecting for more persons and not necessarily taking the infinitive). In Finnish, the connegative verbs are identical to other verb forms, mostly: the connegative present verb is identical to the singular imperative:

mene! (go!)
ei mene (doesn't go)
menee (goes)
menen (I go)
en mene (I don't go)

The past tense is formed by using the past participle instead: 

saapunut mies (the man who (has) arrived)
mies ei saapunut (the man did not arrive)
mies saapui (the man arrived)
mies on saapunut (the man has arrived (literally "is arrived")
et saapunut (you didn't arrive)
saavuit (you arrived)
olet saapunut (you have arrived)
et ole saapunut (you have not arrived)

In most of the Uralic languages with such a negative auxiliary, the imperative is formed by a suppletive negative auxiliary. In Finnish, it is älä/älkää(/älkäämme/älköön/älkööt)(the forms in parenthesis are somewhat unusual, 1pl, 3pl, 3sg). In the second person singular , the connegative is identical to the regular imperative, again.

Älä mene! (Don't go)

However, with plurals and the third person singular, it is a unique form:

älkää menkö (don't y'all go!)
älköön tulko (don't he come! as an optativey thing)

Finally, the passive has a connegative that simply removes part of the passive suffix -tAAn (present), -tiin (past) and obtains -dA (or -tA) (present passive connegative, sometimes identical with the infinitive), -tU (past, also the past passive participle).

I am not all that sure how other Uralic languages deal with this, but let's go and imagine a system slightly different from that of Finnish. We posit an explicit set of forms - connegative imperative, connegative past and connegative present (possibly some TAM's connegative forms are constructed, however, by reusing other forms, much like how Finnish reuses the imperative and the past active participle here). Now, we further add these restrictions: these forms only ever appear as part of the verbal complex with negative particles and they originate with deverbal forms of some kind (possibly having some formerly case morphology on them).

So, we suddenly have a set of almost-nouns, that when they are used indicate negation. This could be used for something. Let's be weird and use them for personal names! 

Fear.CNEG-IMPER-PLUR "(do not, ye all) fear"
Fall.CNEG-PRES "(does not) fall"
Conquer.CNEG- PASS "(un)conquered"
Deceive.CNEG-PRES-SG "(does not) deceive"
Suffer*.CNEG-PRES-SG "(does not) suffer"
Surrender.CNEG-IMPER-PLUR "(do not, ye all) surrender"

*as in "to suffer a fool", or such.

This might be a pretty unusual naming scheme, I figure.