Manúskta

Consonants
Consonants written as digraphs and trigraphs are pronounced as a single sound, but are not phonemes in their own right. Underlyingly they are sequences of two phonemes that can be split up by infixes.

The voicing contrast in fricatives (central and lateral) is gradually being lost in the language. The only place where it is consistently maintained is word initially. Furthermore fricatives cannot be found in syllable codas. If one is written, it is not pronounced, but puts breathy voice on the preceding vowel.

Voiceless lateral fricatives can also be pronounced as voiceless lateral approximants at the same point of articulation. All lateral fricatives can also optionally be pronounced as clusters of fricative+lateral.

Consonants in square brackets [] are only found in formal speech. Consonants in round brackets are only found in some dialects.

It is debatably whether the glottal stop is a phoneme in its own right. It only occurs between vowels that would be next to each other because of affixing or compounding. It never occurs morpheme internally. However it is written in the orthography with a hyphen.

The uvular approximant written as g is in free variation with a trill, and a voiced stop. It can also be velar or pharyngeal for some speakers.

Ejectives, implosives, glottalised sonorants, consonants with faucalised and breathy voice as well as voiceless aspirated stops and affricates can all be heard in Manúskta. However these can be predicted from their position relative to accented syllables and breathy voiced vowels. Therefore they are not phonemes and are not listed.

Vowels
All diphthongs listed are centering diphthongs that start at the position labelled on the diagram and center to schwa. Ee and Oo are normally monophthongs, except in syllables with long accent.

There are 12 vowels in total, 6 with modal voice and 6 with breathy voice. These are grouped together in pairs, with one member of each pair having modal voice, and the other having breathy voice. There are many phonological and grammatical processes that change a modally voiced vowel to its breathy voiced counterpart, and vice versa. These pairs are listed below, with modally voiced vowel first, followed by it's breathy voiced counterpart:

a - IPA: /a/ X-Sampa: /a/ <-> aa - IPA: /a̤ə̤/ X-Sampa: /a@_t/

ae - IPA: /ɛ/ X-Sampa: /E/  <-> e - IPA: /e̤/, X-Sampa: /e_t/

i - IPA /ɨ/, X-Sampa: /1/ <->  ee - IPA: /i̤/, X-Sampa: /i_t/ (in syllables with long accent, this is a falling diphthong [ie] instead)

u - IPA /ɵ/, X-Sampa: /8/ <-> ua - IPA: /y̤a̤/, X-Sampa: /ya_t/

o - IPA: /o/, X-Sampa: /o/ <-> oo - IPA: /ṳ/, X-Sampa: /u_t/ (in syllables with long accent, this is a falling diphthong [uo] instead)

ao - IPA: /ɔ/, X-Sampa: /O/ <-> oe <span style="color: rgb(50, 61, 79); font-family: 'Lucida Grande', 'Trebuchet MS', Helvetica, Arial, sans-serif; ">- IPA: /o̤ɛ̤/, X-Sampa: /oE_t/

Vowels with breathy voice are pronounced with a markedly lower pitch than vowel with modal voice. Also Manúskta has a phonation spreading process that works in a similar way to tone spreading. Breathy voice can spreading from the vowel in one syllable to the vowel in the preceding syllable, as long as none of the intervening consonants are "blocking". Blocking consonants are those that were historically voiceless. They are written in the orthography as p, t, k, f, th, s, h and - (the glottal stop). This is why the word for 1 is aw, IPA: [aw̰], X_Sampa: [aw_k], the word for 100 is awhaanztî, IPA: [awxa̤ə̤tʼɨː], X-Sampa: [awxa@_tt_>1:] (even though the following vowel is breathy voiced, the h blocks the phonation spreading) but the word for 110 is aawmwuastî, IPA: [a̤ə̤wŋʷy̤a̤tʼɨː], X-Sampa: [a@_twN_wya_tt_>1:]. There is nothing to stop the breathy voice on the vowel "ua" spreading to the previous vowel, since neither "m" nor "w" are blocking consonants.

Initial Consonant Mutation
Word initial clusters of fricative + voiceless stop are normally ejectives at the point of articulation of the stop. The only exception to this is if there is another ejective later in the word. In this case the cluster is pronounced as a tenuis stop. Under normal conditions, word initial voiceless stops are aspirated, however, in a similar fashion to the ejectives, they become tenuis if there is another aspirated stop later in the word. A similar thing happens with voiced stops at the beginning of words, which become implosives unless there is another implosive later in the word, in which case they become tenuis stops. Likewise, word initial voiceless fricatives (central and lateral) become voiced if there is another fricative later in the word that is pronounced as voiceless (if it is written, but elided or pronounced as a voiced fricative, then this process doesn't apply). A few speakers glottalise all word initial sonorants, except if there is another glottalised sonorant later in the word.

Suprasegmentals
<p style="border-style: initial; border-color: initial; font-style: normal;">Each word in Manúskta contains exactly one accented syllable. There are two types of accent, Short, and Long (which lengthens the vowel it is on). Word final syllables with no coda consonant (note that -y and -w count as coda consonants) can only have Long accent.

<p style="border-style: initial; border-color: initial; font-style: normal;">If a vowel with short accent is immediately followed by a stop, then that stop becomes implosive (with the exception of k and g which become glottalised uvular approximants IPA: [ʁ̰], X-Sampa: [R_k], owing to the difficulty of making an implosive in the back of the mouth). e.g. cot - kót, IPA: [kʰɔɗ] X-Sampa: [k_hOd_<]. to stay overnight - pák, IPA: [pʰaʁ̰], X-Sampa: [p_haR_k]. If the stop is written b, d or g, then it itself is pronounced the same way, but the vowel is changed to breathy e.g. bottle - kóed, IPA: [kʰo̤ɛ̤ɗ], X-Sampa: [k_hoE_td_<], vegetable - páag, IPA: [pʰa̤ə̤ʁ̰], X-Sampa: [p_ha@_tR_k]. If the stop is written p, t or k, then the vowel can never be breathy.

<p style="border-style: initial; border-color: initial; font-style: normal;">Also breathy voice can never spread onto (or through) syllables with short accent e.g. cots - kótuaz, IPA: [kʰɔɗy̤a̤], X-Sampa: [k_hOd_<ya_t], not *kóetuaz.

<p style="border-style: initial; border-color: initial; font-style: normal;">If a vowel with a short accent is immediately followed by s and then a voiceless stop, then the s is not pronounced and the voiceless stop becomes ejective e.g. Manúskta is pronounced IPA: [mʲanɵkʼta], X-Sampa: [m_jan8k_>ta].

<p style="border-style: initial; border-color: initial; font-style: normal;">If a vowel with a short accent is immediately followed by a sonorant, then that sonorant acquires creaky voice and also blocks the spreading of breathy voice e.g. human - mánuas, IPA: [mʲan̰y̤a̤], X-Sampa: [m_jan_kya_t]. Here we can see that the vowel in the second syllable has breathy voice, which would normally spread through n (as it is a sonorant) were it not for the fact that n has acquired creaky voice.

<p style="border-style: initial; border-color: initial; font-style: normal;">If the vowel with a short accent is immediately followed by a fricative and then a sonorant, the fricative is not pronounced and the sonoarant acquires creaky voice. However the vowel acquires breathy voice e.g. Muslim - móozlim IPA: [mʲṳl̰ɨm], X-Sampa: [m_ju_tl_k1m]. If such a sonorant is a vowel, then a glottal stop is inserted before it e.g. Pacific - paséefik IPA: [pʰazi̤ʔɨ̰k], X-Sampa: [p_hazi_t?1_tk].

<p style="border-style: initial; border-color: initial; font-style: normal;">In contrast to vowels with short accent, vowels with long accent often modify the pronunciation of the preceding consonant, in addition to lengthening the vowel e.g. if they are preceded by a sequence of fricative + voiceless stop, then the fricative will be deleted and the stop will become an ejective e.g. twenty seven - twenskông, IPA: [ʈʰe̤nkʼoːŋ], X-Sampa: [t`_he_tnk_>o:N]. Otherwise a voiceless stop before a vowel with long accent will become aspirated e.g. eggplant - makûax, IPA: [m ʲ akʰy̤ːa̤], X-Sampa: [m_jak_hya_t:]. As mentioned previously, fricatives are normally voiced between vowels, but a long accent on the following vowel blocks this for f, s and h e.g. to collect tax - pasêeh, IPA: [pʰasi̤e̤], X-Sampa: [p_hasie_t]

<p style="border-style: initial; border-color: initial; font-style: normal;">The pronunciation of the long accent in other cases depends on whether or not the vowel is breathy voiced. Voiced fricatives, voiced stops, nasals and approximants (i.e. all the non blocking consonants) all become breathy voiced before a vowel with long accent and breathy voice. The vowel in this case acquires a "peaking" tone that starts low, rises to mid and then falls to low again e.g. zoo - zôoh, IPA: [z̤ṳ̂ː], X-Sampa: [z_tû_t:]. Furthermore the breathy voice can spread to preceding syllables in the same word e.g. to visit - tám, IPA: [tʰam̰], X-Sampa: [t_ham_k]. This word by itself clearly has no breathy voice, but when we look at the verb meaning to visit a zoo - taamzôoh, IPA: [tʰa̤ə̤m̤z̤ṳ̂ː], X-Sampa: [t_ha@_tmz_tû_t:] we can see that the breathy voice has spread to the first syllable.

<p style="border-style: initial; border-color: initial; font-style: normal;">If a syllable with long accent begins with a voiced fricative, a voiced stop, a nasal or an approximant, but there is no breathy voice on the vowel, then the vowel acquires a sharply rising tone and the consonant is pronounced with faucalised voice e.g. twelve - lô, IPA: [l̬ǒː], X-Sampa: [l_vǒː].

<p style="border-style: initial; border-color: initial; font-style: normal;">Note that aspirated stops, ejectives, implosives and voiceless fricatives are each considered special categories of consonant in Manúskta. A word can have a maximum of one phoneme from each category. If it would have more than one, then all except the last are changed to tenuis stops (or voiced fricatives in the case of f, s and h) e.g. to visit - tám, IPA: [tʰam̰ ʲ ], X-Sampa: [t_ham_k_j], to make a surprise attack- kyâp, IPA: [cʰa:p], X-Sampa: [c_ha:p], to travel a distance to make a surprise hit and run raid - tamkyâp , IPA: [tam ʲ cʰa:p], X-Sampa: [tam_jc_ha:p]. Here the first t would normally be aspirated, but because there is an aspirated palatal stop later in the word, it is deaspirated to a tenuis stop.

<p style="border-style: initial; border-color: initial; font-style: normal;">Manúskta has an extremely productive process of producing nouns from verbs and vice versa by changing the location of the pitch accent. This process originated with English stress derived nouns (e.g. to inVITE vs. an INvite), but was borrowed into Manúskta and became much more productive e.g. to invite- inváoyt, IPA: [ɨnvʲɔj̤], X-Sampa: [1nv_jOj_k] vs. an invite (e.g. an invitation) - ínvaoyt IPA: [ɨn̰vʲɔj], X-Sampa: [1n_kv_jOj], a convict - káonvikt, IPA: [kʰɔn̰vʲɨk], X-Sampa: [k_hOn_kv_j1k] vs to convict - kaonvíkt IPA: [kʰɔnvʲɨʁ̰], X-Sampa: [k_hOn_kv_j1R_k].

<p style="border-style: initial; border-color: initial; font-style: normal;">It is very common for the type of accent to shift from short to long or long to short when it is moved from one syllable to another. The reason for this is that, once upon a time, every syllable in a word in Manúskta could be long or short. However a sound change occurred that shortened all the unaccented vowels. An example of this is the following pair of words: contrast (noun) - kóntraast, IPA: [kʰɔn̰ʈa̤ə̤t], X-Sampa: [k_hOn_kt`a@_tt] vs. to contrast - kaontrâast, IPA: [kɔnʈʰa̤ə̤ːt], X-Sampa: [kOnt_ha@_t:t] Note that this is another example of the rule that Manúskta does not like words with more than one aspirated consonant. In the verbal form, the initial k would normally be aspirated, but is tenuis because there is another aspirated stop later in the word.

Phonotactics
<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">Manúskta does not like syllables that are more complex than CVC, where V is a vowel and C is a consonant sound. For the purposes of phonotactics, clusters like -tr- behave as single sounds even though, underlyingly, they are clusters of two consonants. It is also worth noting here that, in most dialects, "r" and "l" are deleted in syllable codas. Most Manúskta speakers also do not like retroflex and palatal nasals (especially palatal nasals after back vowels) in syllable codas, and pronounce them as alveolar/dental nasals instead. Also, in syllable codas, labio-velar and uvular nasals are usually merged into plain velar nasals.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">Similarly, there is a restriction on the range of stops permitted in syllable codas. Retroflex, palatal and labio-velar stops are not normally allowed here. Note that all of these consonants are written with digraphs. In syllable codas, they are prononced as though the second consonant of the digraph didn't exist e.g. ky becomes k, and dy becomes d.

Rules for simplifying illegal consonant clusters
<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">Manúskta orthography contains many instances of consonant clusters that would be forbidden by the phonotactics specified above (sometimes formed by affixation and compounding). These are simplified to permissible clusters by applying a series of rules. These are applied in layers, and keep going until only permissible syllable shapes are obtained (maximum CVC):

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">1) Delete all fricatives unless they come immediately before a vowel (this may lead to breathy voice being put on the preceding vowel, as long as there is no intervening voiceless consonant to block it). <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">2) Excluding digraphs listed in the consonant chart above, begin deleting all approximants other than approximants immediately after a vowel. This rule starts working from right to left. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">3) Begin deleting all stops in the cluster. Start working from right to left. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">4) Begin deleting all nasals in the cluster. Start working from right to left.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">Take for instance the transitive verb skrúym - to scream at someone or something. Both the initial and final consonant clusters as written are not allowed in Manúskta. They are simplified as follows:

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">a) The initial s does not come before a vowel. It is therefore deleted (though it makes the k ejective). We now have [kʼɻɵj̰m ʲ ], which is not yet permissible, so we go to the next step. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">b) The [ɻ] is not immediately after a vowel, so it gets deleted. However the [j] is, so it is retained. We now have [kʼɵj̰m ʲ ], in which the syllable onset is now phonotactically permissible, but the coda is not. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">c) The final [m ʲ ] is deleted. We now have [kʼɵj̰] which is permissible.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">Now let's say we want to make the verb antipassive. To do this, the infix "it" is inserted after the first underlying consonant in the onset of the final syllable, this gives sitkrúym which simplifies as follows:

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">a) The initial s comes before a vowel this time. It is therefore not deleted. To start with we have [sɨtkɻɵj̰m ʲ ], where both the medial and final clusters are not yet permissible, so we go to the next step. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">b) The [ɻ] is not immediately after a vowel, so it gets deleted. However the [j] is, so it is retained. We now have [sɨtkɵj̰m ʲ ], with a coda that is still not phonotactically permissible. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">c) The final [m] is deleted. We now have [sɨtkɵj̰] which is permissible.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">ow say we prefix it with the deontic mood prefix waona-, to make woenaaskrúym (remember that the s is deleted and puts breathy voice on the preceding vowel, with spreads through "n" to the first vowel in the prefix). Also not that the k is no longer word initial, so it is not ejective.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">a) The s is deleted, so we initially have [wo̤ɛ̤na̤ə̤kɻɵj̰m ʲ ]. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">b) The [ɻ] is not deleted, because the consonant cluster that it is in conforms to Manúskta phonotactics. Nothing happens to the final [j] for similar reasons as before. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">c) The final [m ʲ ] is deleted.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">Now let's say that we want to put both the antipassive infix and the deontic prefix on the verb. First waona- is prefixed, since the mood inflexion takes place before voice inflexion. So first we have woenaeskrúym like before. However the antipassive infix -it- attaches to the first underlying consonant in the onset of the initial syllable. For the purposes of infixing, Manúskta grammar would divide a word like this up as woenaes.krúym. Therefore the written form is woenaeskitrúym.

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">a) tr coalesces into [ʈ], and the s is deleted with the same breathy voice effect as before. So now we have [wo̤ɛ̤na̤ə̤kɨʈɵj̰m ʲ ]. <p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">b) The [m] is deleted for the same reasons as in the other words. So the word is pronounced [wo̤ɛ̤na̤ə̤kɨʈɵj̰m ʲ ].

<p style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;">So when is the m actually pronounced? Well Manúskta has an extremely productive process that suffixes -ar (pronounced [a] except in the few dialects that allow [ɻ] in codas) to create a noun from a verb meaning "the one who does ...". So woenaeskitrúymar means "the one who I hope screams", and is pronounced [wo̤ɛ̤na̤ə̤kɨʈɵj̰m ʲ a].

<h3 style="border-style: initial; border-color: initial; font-style: normal; border-style: initial; border-color: initial;"> "Creaky" Stops

In the word final position, after a modally voiced vowel. b and d are devoiced to p and t respectively. Some speakers also change g to k in this context. If this happens the preceding vowel gets creaky voice and falling tone e.g. tâod - to fry, IPA: [tʰɔ̰ːt], X-Sampa: [t_hO_k:t]. Of course if something like suffixation means that this is no longer the final phoneme of the word, then this process does not apply e.g. tâodar - fryer, IPA: [tʰɔːda], X-Sampa: [t_hO:da] (note the lack of creaky voice on the first vowel).

This applies even if a written consonant has been deleted to make the syllable fit Manúskta phonotactics e.g. straight - ngây, IPA: [ŋa:j], X-Sampa: [Na:j] has modal voice and high level tone, whereas easy - ngâyd, IPA: [ŋa̰ːj], X-Sampa: [Na_k:j] has falling tone and creaky voice.

Number
Manúskta nouns are unmarked in the singular, and inflect for dual number (exactly two), collective (a group e.g. a pride of lions) and plural (more than two, or a large amount in the case of uncountable nouns). How this is done depends on whether the noun is monosyllabic, or consists of two or more syllables.

For monosyllabic nouns, the dual form is created by reduplicating the rime of the syllable and placing it after the noun e.g. pedal boat - slâey, IPA: [ɬɛːj], X-Sampa: [KE:j] vs. two pedal boats - slâeyaey, IPA: [ɬɛːjɛj], X-Sampa: [KE:jEj]. The collective form is made by suffixing -stowd as is the word for a fleet of pedal boats - slêystowd, IPA: [ɬe̤ːjto̰w], X-Sampa: [Ke_tto_kw]. The deletion of the -s puts breathy voice on the preceding vowel, and the -d at the end of the word makes the final vowel have creaky voice and falling tone.

This can work the other way as well, where reduplication means that a fricative is pronounced in the dual form, where it was not pronounced in the singular form (with a change in phonation on the vowel) e.g. bank (of a river) - fâasy, IPA: [fa̤ə̤ːj], X-Sampa: [fa@_t:j] vs. a pair of banks (of a river) - fâsyaay, IPA: [faːʝa̤ə̤j], X-Sampa: [fa:j\a@_tj].

The way in which the plural form of monosyllabic nouns is formed depends on the coda. If it is written with a voiceless consonant, the suffix -teh is added e.g. banks (of rivers) - fâasyteh, IPA: [fa̤ə̤ːjte̤], X-Sampa: [fa@_t:jte_t]. If not, then what happens depends on the phonation of the vowel. If it is breathy then the suffix is -myaah, and if not the suffix is -dae e.g. boats (more than 2) - slâeydae , IPA: [ɬɛːjdɛ], X-Sampa: [KE:jdE].

For words with more than one syllable, things are totally different. The dual form is formed by reduplicating the onset and nucleus of the accented syllable, and placing the reduplication immediately before the accented syllable e.g. condominium - kaondomínyum, IPA: [kʰɔndomɨɲ̰ɵm], X-Sampa: [k_hOndom1J_k8m] vs. a pair of condominiums - kaondomimínyum, IPA: [kʰɔndomɨmɨɲ̰ɵm], X-Sampa: [k_hOndom1m1J_k8m]. For collective number, add the infix -oev immediately after the first consonant of the accented syllable e.g. a set of condominiums - koendoomoevínyum - IPA: [kʰo̤ɛ̤ndṳmo̤ɛ̤vɨɲ̰ɵm], X-Sampa: [k_hoE_tndu_tmoE_tv1J_k8m] (note how breathy voice can spread onto multiple syllables if there are no blocking consonants)

The way in which the plural form of words with more than one syllable is formed depends on the final syllable. If the vowel does not have breathy voice then pluralisation is simple. Simply write a -z immediately after the last vowel the word. The consonant itself is silent, but its deletion puts breathy voice onto the preceding vowel. As usual, this can spread as far as there are no blocking consonants e.g. condominiums (more than 2) - kaondomínyuazm, IPA: [kʰɔndomɨɲ̰y̤a̤m], X-Sampa: [k_hOndom1J_kya_tm].

If the final vowel of the word has breathy voice, then pluralisation is more complicated and involves metathesis. The rime of the final syllable is "cut", and "pasted" immediately after the first consonant of the accented syllable. Then the suffix -uaz, is added to the word e.g. collage káolaazr, IPA: [kʰɔl̰a̤ə̤], X-Sampa: [k_hOl_ka@_t] is pluralised to collages (more than 2) - kaazráoluaz, IPA: [kʰa̤ə̤ʐɔl̰y̤a̤], X-Sampa: [k_ha@_tz`Ol_kya_t].

To illustrate this, here are some words in their singular, dual, collective and plural forms. There is one word for each pluralisation pattern.

English: man (monosyllabic, voiced coda, non-breathy vowel)

Singular: mâen, IPA: [m̬ʲɛːn], X-Sampa: [m_v_jE:n]

Dual: mâenaen, IPA: [m̬ʲɛːnɛn], X-Sampa: [m_v_jE:nEn]

Collective: mênstowd, IPA: [m̬ʲe̤ːnto̰w], X-Sampa: [m_v_jE:nto_kw]

Plural: mâendae, IPA: [m̬ʲɛːndɛ], X-Sampa: [m_v_jE:ndE]

English: child (monosyllabic, voiced coda, breathy vowel)

Singular: klêx, IPA: [kʰe̤ː], X-Sampa: [k_he_t:]

Dual: klêxex, IPA: [kʰe̤ːɣe̤], X-Sampa: [k_he:Ge_t]

Collective: klêxstowd, IPA: [kʰe̤ːto̰w], X-Sampa: [k_he_t:to_kw]

Plural: klêxmyaah, IPA: [kʰe̤ːmʲja̤ə̤], X-Sampa: [k_he_tm_jja@_t]

English: animal (monosyllabic, voiceless coda)

Singular: sátw, IPA: [saɗ], X-Sampa: [sad_<]

Dual: sátwatw, IPA: [saᶑat], X-Sampa: [sad'_<at] (has a retroflex implosive).

Collective: sátwstowd, IPA: [saᶑto̰w], X-Sampa: [sad_<to_kw]

Plural: sátwteh, IPA: [saᶑte̤], X-Sampa: [sad_<te_t]

English: star (non-monosyllabic, non-breathy final vowel)

Singular: ngôysaw, IPA: [ŋ̬oːjzaw], X-Sampa: [N_vo:jzaw]

Dual: ngongôysaw, IPA: [ŋ̬oŋ̬o:jzaw], X-Sampa: [NoN_vo:jzaw]

Collective: ngoevôysaw, IPA: [ŋo̤ɛ̤v̬ʲoːjzaw], X-Sampa: [NoE_tv_v_jo:jzaw]

Plural: ngôysaazw, IPA: [ŋ̬oːjza̤ə̤w], X-Sampa: [N_vo:jza@_tw]

English: rotten object (non-monosyllabic, breathy final vowel) [note: I have underlined the part of the word that "moves" in the plural form]

Singular: ráol uasy , IPA: [ɻɔl̰y̤a̤j], X-Sampa: [r`Ol_kya_tj]

Dual: raoráol uasy , IPA: [ɻɔɻɔl̰y̤a̤j], X-Sampa: [r`Or`Ol_kya_tj]

Collective: roeváol uasy , IPA: [ɻo̤ɛ̤vʲɔl̰y̤a̤j], X-Sampa: [r`oE_tv_jOl_kya_tj]

Plural: r uasy áoluaz, IPA: [ɻy̤a̤ʝɔl̰y̤a̤], X-Sampa: [r`yaj\Ol_kya_t]

Note that the dual and plural forms are not used when the object is explicitly counted using a numeral e.g. for "two men", mâen tô is used, not *mâenaen tô, and to say "three rotten objects", ráoluasy trî is correct, not *ruasyáoluaz trî. Also note that is is perfectly possible to pluralise any noun, even if it would not normally be considered countable. In such a case it merely indicates a large amount of the said thing e.g. water - wótar, IPA: [woɗa], X-Sampa: [wod_<a] vs. a large amount of water - wótaazr, IPA: [woɗa̤ə̤], X-Sampa: [wod_<a@_t]. Often the dual forms of such uncountable nouns are used as well, usually with highly idiomatic meanings e.g. the dual form of water, wowótar means the place where two rivers meet.

Possession
Manúskta distinguishes alienable and inalienable possession in many cases. Both are indicated by a prefix on the possessed noun. Often the shape of this prefix depends on whether it is followed by a consonant or a vowel. If so, the form used with a consonant is given first. Here are the pronominal prefixes (which would correspond to English "my", "your" etc.).

Singular

1st person: maoy-

2nd person: yo- / yor-

3rd person alienable: hees- / his-

3rd person inalienable: hu- / hur-

3rd person inanimate possessor: it- / its-

Dual

1st person exclusive: meys- / maeys-

1st person inclusive: stu-

2nd person: vu-

3rd person alienable: pez- / pezr-

3rd person inalienable: thud-

3rd person inanimate possessor: thaah- / thah-

Plural

1st person exclusive: aew- / aewr-

1st person inclusive: yay-

2nd person: zyu-

3rd person alienable: êz- / êzr-

3rd person inalienable: rôesl- / râosl-

3rd person inanimate possessor: hâ-

If one noun possesses another, then the normal possessive construction places the possessed noun before the possessor (like in the English "of" construction). The possessed noun is prefixed with i- / y- if the possession is alienable and na- / n- if it is inalienable. If the possessor has been fronted to the beginning of the sentence as a topic, then instead the possessed noun is prefixed with zo- if the possession is alienable, and sweh- / swaeh- if it is not.

Note that when these affixes attach to a noun that is monosyllabic (after any pluralisation), then the pitch accent shifts to be on the prefix. Whether the prefix will have short or long accent is unpredicatable (it depends on whether the prefix had a long or short vowel in the proto-language) and depends on the prefix e.g. the 1st person singular prefix maoy- will always have short accent, leading to words like "my pedal boat" - máoyslaey, IPA: [mɔj̰ɮɛj], X-Sampa: [mOj_kK\Ej] whereas the 2nd person singular prefix yo(r)- will always have long accent, leading to words like "your pedal boat" - yôslaey, IPA: [j̬oːɮɛj], X-Sampa:[j_vo:K\Ej]. As mentioned before, this pitch accent only applies to words that are monosyllabic after any pluralisation processes have taken place. So the word for "my pedal boats (more than 2)" will not be *máoyslaeydae, but maoyslâeydae, IPA: [mɔɬɛːjdɛ], X-Sampa: [mOKE:jdE].

Here is a list of the variant forms of each possessive prefix (some of the forms vary slightly from the unaccented forms):

Singular

1st person: máoy-

2nd person: yô- / yôr-

3rd person alienable: hées- (breathy voiced regardless of whether it is followed by a vowel or consonant)

3rd person inalienable: hú- / húr-

3rd person inanimate possessor: ít- / íts-

Dual

1st person exclusive: múy- / múys-

1st person inclusive: stû-

2nd person: vû-

3rd person alienable: sôwn-

3rd person inalienable: thûd-

3rd person inanimate possessor: thâah- / thah-

Plural

1st person exclusive: âew- / âewr-

1st person inclusive: yáy-

2nd person: zyû-

3rd person alienable: êz- / êzr-

3rd person inalienable: rôesl- / rôasl-

3rd person inanimate possessor: hâ-

Normal Alienable Possessive Prefix: í- / y-

Normal Inalienable Possessive Prefix: ná- / n-

Topicalised Alienable Possessive Prefix: zó-

Topicalised Inalienable Possessive Prefix: swéh- / swáeh-

Case
There are only two grammatical cases in the language, that I will label Absolutive and Ergative even though Manúskta differs from a classic ergative-absolutive language like Basque. Absolutive case is unmarked, and the Ergative is marked with the prefix baoy- e.g.

homkwa kák Mary báoy.John

yesterday kick Mary ERG-John

Yesterday John kicked Mary

However if the subject of the transitive verb is more animate than the object, both nouns take the Absolutive Case e.g. assuming Rover is a dog, and therefore less animate than John (a human), we would say:

hômkwa kák Rover John

yesterday kick Rover John

Yesterday John kicked Rover (a dog).

The Ergative case is also used like a Genitive marker on nouns that possess other nouns e.g.

í.maeyk báoy.John

ALNBL.POSS-comrade ERG-John

John's comrade

(Note that the inalienable prefix ná- can be used instead to make the word námaeyk. This gives the sense of eternal, undying comradeship)

However if the possessor is more animate than the posssessed noun, then the possessor takes the Absolutive case instead (i.e. it is unmarked).

i.wótar John

ALNBL.POSS-water John

John's water

Also Ergative Case is never found on nouns which are otherwise inflected in any way e.g.

í.maeyk stû.klex

ALNBL.POSS-comrade 1INCL.DUAL.POSS-child

our child's comrade

Articles
Whilst in English definite and indefinite articles are separate words (albeit clitics), in Manúskta they are fully fledged affixes (prefixes that come before the possessive prefixes). Like the possessive prefixes, they often vary in shape depending on whether they are followed by a consonant or a vowel. If this is the case then the form before the consonant is written first. They normally are not accented, except if the noun is monosyllabic and there are no intervening affixes between the article and the noun root. The accents only remain on the prefixes in this case, and in all other cases are lost. Here is a list of the articles (note that there is only a distinction between singular and plural, with no dual forms, when referring to two of something, the plural is used):

Indefinite Singular: ées- / ís- (Indefinite articles are used when referring to something that the listener is assumed to not have any information about.)

Indefinite Plural: úys-

Semi-Definite Singular: áed- (Semi-Definite articles are used when commenting on or asking for information about something that your interlocutor has already brought up in the discussion. If you brought it up in the discussion yourself, then you use a definite article instead. They are also used when bringing up something in a discussion that you assume your listener has already heard of.)

Semi-Definite Plural: âaz-

Definite: ú- / íy- (This is used when referring to something that you yourself have already brought up in the discussion, or something for which you expect that both you and your listener should know exactly what you're talking about. It is the same for both singular and plural nouns).

Hypothetical Singular: á- / áen- (This is used when introducing anything anything hypothetical into the discussion. It is also used to mean "any one of" the noun taking the affix. The next time the object is mentioned, it will take the indefinite article).

Hypothetical Plural: Unmarked (This has a similar usage to the hypothetical singular.)

Nonspecific: sam- (Used to refer to something, usually in passing, that you have no information about, and that you don't expect your listener to have any information on either).

One important point to make here is that articles are not found on possessors, nor on nouns which are the grammatical subject of a sentence. The only exception to this is when the noun is topicalised e.g. the following phrase could be grammatical:

<p style="border-style: initial; border-color: initial; font-style: normal; margin-top: 1em; margin-bottom: 1em;">is.í.maeyk stû.klex

<p style="border-style: initial; border-color: initial; font-style: normal;">INDF.SG-ALNBL.POSS-comrade 1INCL.DUAL.POSS-child

<p style="border-style: initial; border-color: initial; font-style: normal;">a comrade of our child's

But the next phrase would never ever be grammatical, as it is placing an article on a non-topicalised possessor:

<p style="border-style: initial; border-color: initial; font-style: normal; margin-top: 1em; margin-bottom: 1em;">* í.maeyk ees.stû.klex

<p style="border-style: initial; border-color: initial; font-style: normal;">ALNBL.POSS-comrade INDF.SG-1INCL.DUAL.POSS-child

<p style="border-style: initial; border-color: initial; font-style: normal;">one of our children's comrade

Adjectives
Adjectives do not form a seperate class of words in Manúskta that are distinct from nouns. Any noun can be used as an adjective by placing it after the noun it modifies. Adjectives never inflect for case, possession, number or definiteness e.g.

báoy.satw klêx

ERG-animal child

a young animal

Demonstratives
Manúskta has a 3 way system of demonstratives: this - nêeh', that - nánt and that over there (yonder) '''- nônd. When used as modifiers, they follow the noun they modify e.g. klêx nêeh''' - this child. When they are used as nominals themselves, they can inflect for number, possession, case and definiteness just like regular nouns e.g. éesnant - "one of those" has an indefinite article, whereas únant - "that one" has a definite article.