Saturday, July 27, 2013

Phonology idea

Do a statistical survey of the distribution of different vowels and consonants in some common form of English. The survey would basically give a two-dimensional array, one dimension corresponding to the individual sounds and the other to frequency at different positions in words.

Now, alter the distribution significantly, so that the sum of each sound's array is roughly the same, but the tops and troughs swap place. After this, recalculate the sum for the frequency for each of the columns (corresponding to frequencies at a position), so they add up to one. (Also, try maintaining the sum of all vowels vs. the sum of all consonants in a given column somewhat intact.)

The result would be a language whose sounds are the same as English, but the way they are used is significantly off. I do think using a restrictive and relatively English-like set of consonant clusters might make sense, but who knows. 

Usage for this thing: naming languages in fiction.

Friday, July 26, 2013

Detail #50: a type of numeral

At times, trying to come up with numerals different from the normal ordinals, cardinals, groups-of-N, and distributives and whatnots is a thing I try my hand at, but seldom manage.

It hit me that one natural context for numerals is when discussing time, and it could very well be possible for a language to have specific forms for the numerals for N:th day (or even N:th general unit of time), or for N days, to such an extent that the time unit generally can be left out.

four-TIME-[some oblique case, plural]: on four days
 four-TIME-[some oblique case, singular]: on the fourth day (of the month)
four-TIME-[some other oblique case, singular]: four days from the time inferred from the context
four-TIME-[some other oblique case, plural]: every fourth day

Meanwhile, if the language treats "many" and "how many" as numerals, in giving them ordinal forms as well, this could provide some further cool stuff. Finally, diminutives would possibly be used to form, say, hours and minutes or some analogous time units.

I would further like for it to lack the nominative and accusative, so that talking of four units of time as a subject or objects requires normal ordinal or cardinal numerals.

This idea might enter Tatediem.


Monday, July 22, 2013

Detail #49: Case system detail / Vocatives

Imagine a language with vocatives and with case congruence on adjectives. Further imagine the language has some kind of intensive or superlative adjective formation, either of indo-european style or just some kind of more general intensifying meaning.

Now, it would be rather natural that vocatives never use non-intensive/non-superlative adjectives; the actual meaning-change involved would be somewhat negated by the presence of the vocative, though, as even a 'good' friend would be called 'best friend!' in such a construction, and everyone was aware of the grammatical rule there. Thus, lexical items would have to be used instead to distinguish intensities instead of morphological devices.

One could even imagine a language where there are no explicit vocative markers, except the restriction on non-superlatives in vocative NPs. One could of course have some (morphologically) superlative/intensive dummy adjective for this role too, if no other adjective is present?

Sunday, July 21, 2013


Tatədiem is a conlang I have been, well, not working on per se, but coming up with occasional ideas for over about eight years now. I have decided that all further conlanging on it will take place on this blog.

A short summary of Tatədiem

Tatədiem is an agglutinating language with some fusionality. Its noun system distinguishes three numbers - singular, dual, plural, as well as mass nouns. The singular and plural forms distinguish, further, definite and indefinite forms, whereas the dual is only used with definite forms. Mass nouns do not distinguish definiteness.

Further, definiteness and number fuse with case as well, combining with ergative, absolutive, *instrumental, partitive, dative (only definite), and locative. Mass nouns and indefinite plurals merge ergative, absolutive and partitive, indefinite singulars merge absolutive and ergative, as well as partitive and locative.

Certain female names are morphologically dual and so are some place names. Almost all of these have some syncretism with the singular, and the occasional one with the plural.

The nouns have a bantu-style gender system, but with the gender markers more restricted in when and where they appear. Basically,

  • an indefinite noun does not have gender markers
  • adjectives with a definite noun have gender-number congruence, and as complements they also take the partitive case.

As for the verb, it can show gender agreement with the subject, the object, and the indirect object. Certain auxiliary verbs can show agreement with some other things, such as instruments, locations and recipients. Certain aspects and voices are marked by simply adding an auxiliary with some congruence in place, and leaving the regular verb with canonical marking.

This system may sound a bit redundant and rich, but I intend to break it sufficiently that an interesting system will be obtained.

Saturday, July 20, 2013

The Isolating Conlang: Implicit adjectives for adpositions

In some languages, adjectives can take complements - e.g. afraid of bears, proud of himself, full of shit, ... In English, almost all such complements take 'of' as their preposition, and some take 'by' (notably past participles). In other languages that have similar complements, a wider selection of prepositions may occur, and The Isolating Conlang (henceforth TIC) is an example of such a language.

In TIC, the number of adpositions in general is large. In part, this is due to two historical processes of preposition grammaticalization having occurred: to some extent, verbs have turned into prepositions, to some extent nouns have turned into prepositions.

Now, sometimes, the association between an adjective and a preposition is particularly strong, and in these cases, omitting the adjective is very common. With the adjective omitted, it is possible to have a copy of the preposition appear at the onset of the sentence or before the verb. Obviously, this is somewhat of a lexical thing, and so a short appendix listing the meanings of adjectives that can be omitted is supplied.

Appendix: Some prepositions in TIC, some with implicit adjectives

ḱesp: on, n. head
suspicious, careful, → der lə ḱesp in "I am (keeping tabs) on him"
tsim: by, with, n. group
similar to  ir kewen tsim subul 'he runs like a calf'
der lə tsim kmtok → I group with my brother/I am similar to my brother
convinced, agreeing: ar lə tsim in → you are (agreeing) with him 
dəlts: in collusion with, with, also an adverb
Dəlts! (Do! Join! You must!)
der lə delts in → I and he collaborate
em in kwale dəlts em vupi then he plays together with (people) then self → sometimes he plays (music) together with others, sometimes by himself
kip: for, especially when doing something that is someone else's obligation or by someone else's request
eager, servile, subordinate
ner: for, to, especially when handing over or giving things to someone
charitable, helpful 
semag: equipped with, having, n. hand, v. carry
skilled, talented → ar lə semag hilhil: "you are (skilled) with the flute"
in kwale semag hilhil → he played the flute skillfully
ragad: inside,
has a peculiar inverse meaning with the adjective full: full of [content] inside, mehar ragad [content]
gad: in
occupied with, fortunate with, 
gad der lə gad məniḱ : I have bountiful catch of fish, I am fortunate with a catch of fish
gad der lə gad sauttm : I am busy with business
segad: covered with, n. pelt, v. lie under, cover with
occupied with, bored with, 
kpem: under, covered by, n. roof, tent
pregnant → wa lə kpem in, "she is (pregnant) by him"
This also can be used as an adverbial to express who the father is with any pregnancy-related verbs, wa səd gahar kpem in ≃ she gave birth to his son, we səd ruhal kpem in ≃ she gave birth to his daughter, wa tom tekka kpem in ≃ she is about to give birth to his first child, wa tom tekka ≃ she is about to give birth to her first child
kvat: moving in the direction of, (located) somewhere along the way to
agreeing with, supportive of (human object), → ar tom kvat in? are you in agreement with him? 
hungry for, eager for, interested in → ar lə kvat qurpa, "you like fights." 
kev: moving in the direction of, v. 'fetch, approach'
quick to, habitually   → in lə kev qurpa, he is quick to find a fight; in lə kev kmtok, he always goes to his brother (for help), in wəl kev sauttm astm-ḱe, he is quick to (make) business (out of) anything (literally "he have quick business anything")
sarə: along, located somewhere along the way to, n. road.
gug lə sarə → the house is along our path
xuk: at the house or abode of [complement], (there is a similar cognate noun, gug)
related → in lə wa lə xuk or in lə xuk wa, "they (he and she) are related", or "he is related to her". Due to cultural reasons, relatedness is not necessarily considered reflexive. 

seve: about, (from noun meaning 'speech')
right in lə seve əm "he is (right) about it"
 esteemed for  → seve ar lə seve kwale hilhil, he is (esteemed) for his flute playing

Thursday, July 18, 2013

Detail #48: Some congruence blocking things

I considered making a few suggestions along the lines of "no congruence for interrogative pronouns" and so on, but changed my mind a bit.

Let us imagine a language where there is an explicit congruence marker for every person, and there is a specific present tense stem (or a full-on lack of infinitives or anything like that; let us assume there's little to no way of confusing the congruenceless verb from an infinitive). Now, we could come up with a lot of relatively easy kinds of things I have already probably hinted at, here ordered by abstractness:

  • certain specific pronouns (interrogatives) or 
  • certain lexical items
  • certain null-subjects
  • indefiniteness
  • aspect-like things
However somehow, all these things seem rather similarly complex - all the suggestions this far are in a way rather focused on one particular constituent or semantic idea. Let us try and step a bit out from the verb phrase / clause, and consider interactions between participants and actions on a more general level, some more pragmatic stuff perchance.

How about

  • distant past combined with either lack of tangible results, imperfectiveness or non-differentiable objects
  • uncertain future or present combined with indefinite object
  • speaker having distinct lack of enthusiasm for the reported information
  • past-known-by-hearsay combined with intransitivity
  • partially or entirely repeated speech?
  • certain specific groups of verbs under circumstances specific for each group

Detail #47: A superlative construction

Use either all or every (if your language distinguishes them), in dative or comparable form as a particle, with a regular adjective. Works well if your language does comparisons using dative (or comparable), but could also work if that is not the usual construction, as this may be a fossilized form.

E.g. "Mount Everest is the tallest mountain" -> Mount Everest is a tall mountain to everyone
This car is the cheapest -> this car is cheap to all
The sharpest knife is not sharp enough -> to everyone the sharp knife is not sharp enough/?the sharpest knife is not sharp to anyone

Tuesday, July 9, 2013

Barxaw: Relative clauses

There are several strategies for forming relative clauses or pragmatically similar constructions in Barxaw.

1. Embedded clause with external pronoun
This is a bit similar to something like this:
He never set foot in the house that he had built -> He never set foot in it, he built the house.
However, the pronoun is somewhat specific to this construction (consisting of a gendered pronoun and a particle otherwise used with topicalized pronouns, "na"). The noun in the embedded clause that is coreferent with the pronoun gets the same particle on it, or is fronted if possible. This kind of construction does not appear much with noun phrases that are not clause-final.

2. Embedded clause with no specific particles introducing it, gapping for subjects and objects, and resumptive pronouns otherwise. Generally, TAM auxiliaries are not used in this kind of relative subclause.

Some restrictions apply for the word order with gapped objects.

3. Clause coordinated with the main clause:
he never set foot in the house yet it was he that built it
4. Embedded clause introduced by the same pronoun as in #1, but with the noun remaining in the matrix clause.
He never set foot in the house, it-na he built.
Unusual, but occurs. Only used with subjects and objects of the embedded clause.

Monday, July 8, 2013

Detail #46: Collective numbers, inalienable possession and derivative nouns

A derivative morpheme that forms collectives - 

five.COL = a quintet, a group of five
cow.COL = a herd of cows
By itself this is not particularly interesting, but let us add that the language has inalienably possessed nouns, such as
brother, sister, wife, husband, father, mother, uncle, nephew, aunt, ...
hand, arm, leg, belly, 
funeral, wedding, baptism (or similar), confirmation (or similar), 
home, farmland, village, ...

Now, since these are inalienably possessed, there is no way of speaking about, say, villages in general or brothers in general or anything like that. Derivative morphology can be applied to non-possessed stems, though, and thus you can get brother.COL, sister.COL, hand.COL, etc. If this language has plural congruence with collectives, this can be violated to mark that it's really just the general case of hand or brother or funeral that is discussed.

With family terminology, though, collectives normally refer to the general case - collective siblings are any set of siblings, for instance. Double plural congruence on the verb, in this case, could signify that several collectives are involved.

If, on the other hand, the language does not use plural congruence with collectives, plural verbs would mark that several collectives are involved. To mark that it is just one single item in the collective, the number one before the collective would signify the number.

Monday, July 1, 2013

Detail #45: A kind of participle

This sort of ties into short sketch #2, and is but one of the participle-like verb forms occurring in that language.

When this participle is used as an adverb it marks whether the main verb phrase is being carried out by someone who has received or is receiving permission, help, prohibition, advice, support, collaboration, opposition, resistance etc. Verbs that do not usually express related ideas may sometimes be somewhat reinterpreted, so say.PRTCPL would be understood as 'with permission having been expressed orally', or if the context so suggests, 'with spoken resistance or opposition' or even 'met by rudeness'

Obviously, negative forms may be common, as people do speak of doing things without permission, with advice, etc :
we went there un-permit-prtcpl-ly: we went there without having been given any permission
we went there ban-prtcpl-ly: we went there in violation of the ban on going there
we built it help-prtcpl-ly: we built it with some help 

As a complement or an argument of a verb, it carries similar meaning:
house is us.dat un-permit-prtcpl : we have no permit for building the house
 king.dat is help.prtcpl.acc : the king needs help
king.dat is help.prtpl.nom . the king has help

Further, we get forms like:
you are go.prtcpl: you have permission to go
you are buy.prtcpl my house: you have permission to buy my house
they are enter.prtcpl this place at any time: they may enter this place at any time
Thus, polite imperatives also are formed using these. Not all verbs, however, can take this form.