The Nomadic Language Family

[Khemehekis]

Aspect
The part of verbal morphology referred to as aspect in terms of proto-Nomadic is more like a grammaticalised day marking system. There were four aspects: continuous, hodiernal (same day/to-day), cratinal (following day/to-morrow) and perendinal (two days time/the day after to-morrow). These are glossed as CONT, HOD, TOMRW and 2DAY, as cratinal and perendinal are non-standard terminology (if anyone knows of the standard terminology, please tell).

The standard word for "cratinal" is "crastinal".

[VaptuantaDoi]

The sub-branches of the Nomadic family

To-day I’ve been working on the lexicon, so there’s nothing much to show, but I’ll still inflict some ramblings on you by describing in more detail the sub-branches I defined in the first post. Here’s the family tree, updated:

Code: Select all

                         Proto-Nomadic (PN)
                                |
              +–––––––––––––––––+–––––––––––––––––––+––––––––––––––––––+
              |                                     |                  |
         True Nomadic (PTN)                 False Nomadic (PFN)        |
              |                                     |                  |
        +–––––+–––––+             +–––––––––+–––––––+––+––––––––+––––––+
        |           |             |         |          |        |      |
     Eastern     Central      Fisherman  Pig-farmer Settlers Bushman   |
     (PEN)       (PCN)        (PF)       (PPF)      (PS)     (PB)      |
     =A          =E           =H         =Wátreṣe   =Óẹ      =S        |
     =B          =F           =I         =O         =Úkux    =T    =Isolate
     =C          =unnamed     =J                    =R
     =D                       =K
                              =L

The overall time fram
Proto-Nomadic was spoken from around 6000 years ago to 4000 years ago; times here are given as approximate numbers NF, a system in which 0 is about 5000 years ago. The most likely scenario is that PFN broke off from PN at 0NF, then PN continued for about a thousand years to give PTN at about 1000NF. PFN split up at about 1500NF and PTN split up about 2000NF. The daughters of the PFN languages split at various times around 3500NF and the PEN and PCN daughters both split around 4000NF; it is now about 5000 NF. The most recent divergence was between the Wátreṣe variants around 4500NF.

1.1 True Nomadic languages
These are defined ultimately by not initially leniding */g/. The */s/ and */ɾ/ allophones which probably existed in PN. The TN languages are defined by their non-concatenative morphology and somewhat more conservative phonology. As PTN emerged considerably later than PN, it has fewer defining characteristics. There is often less lenition and consonant loss, while vowels are less stable.

1.1A Eastern Nomadic languages
PEN had an emerging consonantal root system. This was suppressed in one daughter (A) but expanded in the others. It also underwent considerable vowel changes and loss, which resulted in the morphological variation, as well as the definining consonant shift of */t k d g/ → */θ h t k/, not affecting */s ɾ/. In all descendants, */θ/ was either shifted to /f/ (A and C) or /t/ (B), or underwent a shift of */θ s/ → /s ʂ/ (D). A created a regular system based mostly on the alternation of the final syllable. B and C kept the system but with sound changes muddying the regularity. D is remarkable for becoming practically monosyllabic, removing almost all non-accented vowels with no compensation.

1.1B Central Nomadic languages
PCN underwent a shift of */d g/ → */n ŋ/, while also merging diphthongs to long vowels. PCN itself had relatively normal morphology, but in all three of its daughters, the whole thing collapsed. The unnamed language I’ve worked on above gave itself thirty vowels and thirteen tones which alternate, albeit very regularly, in verbal morphology. E and F did something very strange, where they added the vowel of the affix to the root of the verb and removed the consonants, but left the tone alone. This means that they have to be analysed as polysynthetic and redundantly marking most aspects of verb morphology on the subject, where the tones are dumped. I don’t really care if this is naturalistic, cause I’m going to do it anyway.

1.2 False Nomadic languages
These languages are not spoken by nomads, but rather by fishermen, pig-farmers or agriculturalists. They are all defined by the shift of */g/ to */h/, which gives them a consonant inventory of /t d k s h ɾ/ retained in full only in the isolate fifth branch (which is a special case). They are on the whole characterised by consonant loss and lenition, although this is true to a varying degree. PFN gained front rounded vowels (not in the isolate) and long vowels and four tones (to a lesser extent in the isolate).

1.2A Fisherman languages
The Fisherman languages labialised *h then debuccalised *k and *s; this system was pretty much retained in daughters. I don’t know what special feature this branch is going to have yet.

1.2B Pig Farmer languages
The Pig-farmer languages added two phonemes, /ɣ/ and /z/, which were subsequently mostly lost in descendants. There was quite a bit of vowel loss, although only in specific environments. Most interestingly, some offglides deriving from vowel breaking were fortitioned to /s z ɣ/, making those consonants quite common. The two languages in the family are the Wátreṣe dialect chain and O. Wátreṣe went quite mad adding retroflexes, palatals, labials and linguo-labials, giving some varieties nearly twenty phonemes. The vowel systems are also unusually small, with three or four vowels, their contrasting properties being bled onto the consonants. In O, many interesting, secret things happened.

1.2C Settler languages
The Settler languages were in a way the most successful, with by far the most speakers. There are three varieties, which form something of a variant chain running from Úkux to Óẹ, although with some discernable breaks. PS most famously shifted */t/ to */dʒ/ for literally no reason, unconditionally. Also the front rounded vowels were mid-centralised and then derounded, giving a very lop-sided system. Úkux didn’t do much interesting. Óẹ underwent more consonant loss in many positions. Both palatalised /k/ before front vowels. The PFN consonant system is almost preserved, although only coincidentally. The third language, R, was slightly more conservative but added some glottalised plosives.

1.2D Bushman languages
These are of no relation to the San (also referred to as Bushmen) of Africa, and are only named such because they live in the bush and they are men (and women). PB added ejectives and implosives through very obscure processes. PB is the most conservative grammatically, preserving most morphological systems semi-intact. The consonants changed considerably from PN, but the vowels are mostly unchanged. I don’t know about the specific languages S and T.

1.3 The Isolate
There is also a family-level isolate I am tentatively referring to as Olasa, which branched off at some point between 7000 and 3500 years ago. It is either the descendant of a sister of PN, a descendant of a branch of PN at the same level as PFN and PTN, or a daughter of PFN. There is also a chance it is a complete isolate, reflecting the pre-Nomadic languages of the island (I, the creator, know which one is the case but if I told you it wouldn’t be interesting). In any case, it has been highly influenced by the PFN languages, to the extent that most of the grammar and words have been borrowed from the Settler languages. However, from the words which are evidently retained (for some time at least), it is notable for sharing many changes - too many to be coincidental - with PFN, but at the same time being the most conservative member of the family in only sibilising */t/ allophonically before /i/. The only really interesting thing about it is that it is useful for the phonological reconstruction of PN.

---

Aspect
The part of verbal morphology referred to as aspect in terms of proto-Nomadic is more like a grammaticalised day marking system. There were four aspects: continuous, hodiernal (same day/to-day), cratinal (following day/to-morrow) and perendinal (two days time/the day after to-morrow). These are glossed as CONT, HOD, TOMRW and 2DAY, as cratinal and perendinal are non-standard terminology (if anyone knows of the standard terminology, please tell).

The standard word for "cratinal" is "crastinal".

Ah, thanks. I think that was at first a typo then I copied it off my own work without checking.

[VaptuantaDoi]

I've been slowly working on the very complicated Hssïz verbal system, but that's not finished yet. Instead, I'll post something about this wierd language isolate completely unrelated to all the Nomadic languages but in considerable contact with them. It's called Chijapó (a completely random exonym) and it is very strange. It only has between seven and eleven phonemes depending on how you analyse it and how you count it. If you consider [ʔ] to be purely allophonic and vowel harmony to be a suprasegmental, then the phonemes are /p tʃ h ɲ e o a/. If you count all six results of vowel harmony and /ʔ/, it has /p tʃ ʔ h ɲ e o ɛ ɔ a Ø/. Generally I'd say it has ten, /p tʃ ʔ h ɲ e o ɛ ɔ a/. This isn't a very naturalistic inventory, but meh. Unlike all the Nomadic languages, Chijapó will display only limited ergativity, agglutinitivity and only concatenative morphology, and it has three degrees of vowel length, a phonemic nasal, vowel harmony and no phonemic high vowels. It's spoken by about four hundred people on an offshore island in the north-west. There are two villages where Chijapó is spoken are [unnamed], with 250 speakers, and [also unnamed] with 150 speakers. Hopefully the language sounds sort of North- ~ Meso-American; the long vowels and tones being more Northern and the CVC roots being more Meso-American.

Phonology
Chijapó has a very unusual phonemic inventory (more so than the Nomadic languages), in both the consonants and the vowels. There are only five consonants (/p tʃ ʔ h ɲ/) and five vowel qualities (/e o ɛ ɔ a/). The vowels have in addition five contrastive tones and three degrees of length. There are thus seventy-five vowel forms. It is also notable for the lack of /t/ or velars, and the phonemic nasal.

Consonants
Plosives: /p ʔ/
Affricate: /tʃ/
Fricative: /h/
Nasal: /ɲ/

They are written as in the IPA except /tʃ ɲ/ are č n. These are very general qualities, as would be expected from the small inventory. The following allophony takes place:

• /p/ is phonetically [p ~ b ~ ɓ ~ m ~ ɸ ~ k]. The allophones [b ɓ] occur between two vowels, [m] in a syllable preceding one containing /ɲ/ or root-finally and [ɸ] in a syllable preceding /h/. [k] occurs occasionally word-initially before /e/. Elsewhere, the allophones [p], [​b​] and [ɸ] occur in free variation.

• /ʔ/ displays relatively little variation. It may be lightly preaspirated [ʰʔ ~ hʔ] when word-initially after a pause. It might be a product of vowel hiatus and represent a zero-space in the CV(C) syllable template, although vowel-initial borrowed words are generally prefixed with either /h/ or /ɲ/, rarely /ʔ/. It is also retained when the following vowel surfaces as [Ø].

• /tʃ/ varies between [tʃ ~ ʃ ~ s ~ dʒ ~ ɟ]. Intervocalically, [dʒ ɟ] are most frequent. In an initial syllable followed by /h/, it is most often [ʃ ~ s]. The allophones [tʃ], [ʃ] and [​s​] occur elsewhere in free variation. It also has a raising effect on near-high vowels. Word-finally it is [c].

• /h/ may be any non-coronal non-sibilant fricative, varying between [h ~ ħ ~ χ ~ x ~ ç ~ f ~ ɸ], although most commonly occuring are [h x ɸ]. It is generally velar or palatal [x ç] before /e/, velar or uvular before /ɛ/, uvular or radical [χ ħ h] before /a/ and labial [f ɸ] before /ɔ/ or /o/. Word-finally it is generally velar [x] or glottal [h].

• /ɲ/ becomes velar [ŋ] before /a/ and labial [m] before /o/. It has similar vowel-raising effects to /tʃ/. Root-finally it is [ɲ] or [ŋ].

The “default” value is somewhat arbitrary. In most cases the initial allophone has been given precedence, or if there are multiple then the one from which the others can be most easily derived is used (i.e. /p/ rather than /b/ for the labial plosive, as [ɸ] is more easily derived from /p/ than it is from /b/).

Vowel qualities
Chijapó has under most analyses a five-vowel system, although there is a case for a sixth vowel with no realisation. There are no high vowels in the system but a height distinction in the mid vowels, similar to Momu. Although the vowels /e o/ are allophonically [​i ​​u​​​​​​] under some circumstances and never lower than [e o], the most common realisation is high-mid and this requires the least number of rules to accomplish.

High-mid vowels: /e o/
Low-mid vowels: /ɛ ɔ/
Low vowel: /a/

The high-mid vowels are written i u and the low-mid vowels e o. There is some limited allophony, mostly concerning the high-mid vowels. These become [​i ​​​​u​​​] when following /tʃ/ or /ɲ/, and the sequences /eha oha/ become [iha uha]. Elsewhere, [​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​i​​​​​​​​​​​​​​​​​​ u] occur occasionally in free variation, but [e o] are the most common realisations. The low-mid vowels /ɛ ɔ/ are slightly centralised to [ɛ̠ ɔ̟], or even rarely [ɜ ɞ]. The low vowel /a/ is unspecified for frontness and varies freely between [æ̞ ~ a ~ ɐ̞ ~ ɑ], with front-central [a] being the most common. It is never (or at least, very rarely) higher than a true low vowel. The existence of a zero-realised vowel is discussed in the ATR harmony section.

Length
Chijapó has three degrees of vowel length. These are arbitrarily classified as short, semi-long and long; I could also have called them normal, long and overlong (which would be more traditional) but I didn’t. The semi-long vowels are about twice as long as the short vowels and the long vowels three times as long. This is reflected in the orthography; short vowels are single (i u e o a), semi-long vowels are doubled (ii uu ee oo aa) and long vowels are tripled (iii uuu eee ooo aaa). Long vowels may end slightly more raised than they begin; /eː oː ɛː ɔː aː/ are often [eɪ̟̯ oʊ̠̯ ɛe̯ ɔo̯ aɐ̯], although this is inconsistant and more common with contour tones.

Tone
There are five tones; three level (low, mid and high) and two contour (rising and falling). The exact realisation of the contour tones varies a lot; the only unifying aspect is that they become higher or lower. Thus the rising tone may be [˩˨], [˩˧], [˩˦], [˧˥], [˨˥], etc. The most common realisations are [˩˧] (rising) and [˦˨] (falling). The mid tone is unmarked (e aa uuu), the high tone with the acute (é áá úúú), the low with a grave (è àà ùùù), the rising with a caron (ě ǎǎ ǔǔǔ) and the falling with a circumflex (ê ââ âââ). There will likely be some tone allophony but I haven’t decided on it yet. Tones in Chijapó do not have any grammatical usage as they do in some Nomadic languages; they are purely lexical.

±ATR harmony
Chijapó displays simple ATR harmony. All the vowels in a word must be either /e o a/ or /ɛ ɔ/. Thus there are no words where /e/ occurs with /ɛ/ or /ɔ/, no words where /ɛ/ occurs with /e/, /o/ or /a/ and so on. Suffixes and words in compounds change their vowels to apply to these harmonic rules. The –ATR forms are /e o a/; their correspponding +ATR forms are /ɛ ɔ Ø/ (i.e. /a/ is lost; Ø representing no phoneme rather than capital ø). It is thus necessary for the purposes of vowel harmony and syllable structure to construct the null-vowel /Ø/. However, the phonemic status of this is questionable as it is never realised as a phoneme. It is not enough to say that /a/ is lost when a word is +ATR, as in compounds where a +ATR word assimilates to –ATR, /a/ consistently appears between two surface consonants.

Syllable structure
All Chijapó roots must begin with a consonant and contain one vowel. All verbal and nominal roots also end in a consonant other than /ʔ/. The maximum syllable structure is [consonant][vowel][p,tʃ,h,ɲ]. Within compounds, CC clusters do occur.

Edit: I'm thinking of having this gender system

1) Inanimate
​ 1a) inanimate (contains all non-living things and abstract ideas)

2) Animate
​ 2a) royalty (used only for respected people and members of the royal family)
​ 2b) dead animate (things which were once alive but are now not; e.g. wood, leather, corpses)
​ 2c) domesticated animals (including humans)
​ 2d) young of animals or humans
​ 2e) fish or other marine life
​ 2f) mythological animals
​ 2g) foraging animals (animals which are not domesticated but eat scraps from humans such as foxes)
​ 2h) some animals (a left-over class)
​ 2i) animals associated with diseases (including cockroaches, rats, etc.)
​ 2j) animals which appear in swarms (midges, small insects, locusts)
​ 2k) representations of animals (including statues, paintings, rocks which look like animals)
​ 2l) some more animals (another general left-over class)
​ 2m) animals associated with mischief (monkeys, small children and criminals)
​ 2n) flies and bees

This is of course based on the Celestial Emporium of Benevolent Knowledge which Ser mentioned. I thought it could be easily adapted to make a naturalistic wierd gender system.

[DesEsseintes]

2c) domesticated animals (including humans)

Amazing

[VaptuantaDoi]

2c) domesticated animals (including humans)

Amazing

Ahaha, I got to the end of it before realising that humans weren't included on the list, so I put them in the one which seemed most fitting. I'm going to have to keep it now . I technically could have put then in "those belonging to the Emperor" but I'm not really sure if there will be an emperor.

[VaptuantaDoi]

I'd just like to note that although this isn't 100% dead, I'm going to completely revise the descendants until I'm happy with them which may take some time due to laziness. I want to make the descendants slightly more homogenous and give them more realistic morphology. I was rushing too fast to make wierd sound changes and stuff without actually trying to make something I liked, so I'll be focusing much more on the individual languages... hopefully. There won't be anything new for a while or very possibly ever.