Since I have some time, now, I will present you some features of my latest
creation, which hasn't a true name, yet, but only a temprorary one: Sirmave
(pron.: /Sirmav@/), from the name of the province of my conworld where it
was originally spoken, Sirám /Sir&m/. The phonology is quite stable, now;
the morphology and other parts of the grammar, on the other hand, are still
quite sketchy.

The written language is quite different from the spoken one, thus I'll try
before to deeply describe the phonology, and then the orthography.

Sirmave has a quite rich phonetic inventory: there are many allophonic
variations which enrich the sound of the language.

I have used Kirsh. IPA with some allowances (as /*/ marking an alveorar


There are 15 vowel sounds (6 simple vowels + 5 allophones and 5 nasal
vowels); native linguists still don't agree whether nasal vowels are
independent phonemes or allophones of normal vowels appearing before the
phoneme [n] + another consonant. (Could *you* help me to grasp this

              simple                nasal
high    i            u           e~       u~
mid      e Y    o               E~    o~
low           a                         a~

Allophonic variations:
[a] becomes /&/ in closed stressed syllables (this change generally happens
amongst peasant in the country; educated people don't allow their own mouth
pronounce such vulgar sounds : )
[e] and [o] become /E/ and /O/, respectively, in closed syllables
[i] becomes /I/ word finally
[u] becomes /y/ in post-stressed syllables
[Y] becomes /o/ in unstressed syllables

If we consider nasal vowels as allophones of simple vowels appearing before
the phoneme [n] + another consonant, then we'd have:
[a] becomes /a~/
[e] becomes /E~/
[i] becomes /e~/
[o] becomes /o~/
[u] becomes /u~/

In the everyday speech ending unstressed vowels tend to vanish: henster and
mutienvar are generally pronounced /'hE~st@*/ or /'E~st@*/ and /mu'tjE~v@*/
(notice that the ending is written different: a relic of the old times);
only educated people would pronounce them /hE~ste*/ or /FE~ste*/ and


There are 19 consonant phonemes and small amount of allophones; this means
that the phonetic (not phonemic) inventory ammounts to 25 consonant sounds.

                lab    dnt    vel    uvu       glot
stop         p b    t d    k g    q
fric           f v     s         Q                h
affr               tS dZ dz
nasal         m        n
lateral                l
flap                    r

Allophonic variations:

voiced consonants in final position are devoiced
[g] in final position is pronounced /x/
[Q] in final position is pronounced /x/ as well (but it occurs really
[n] before velar and uvular sounds is pronounced /N/
[r] in clusters and word finally is pronounced /*/ (it becomes an
[l] in consonantal clusters becomes /l./
[l] in intervocalic position becomes /L/ or even /j/; this change occurs
only in the country.
[dz] in C_V position is pronounced /ts/
[s] before [i] is pronounced /S/

There are few cases of gemination, most of whom disappearing. The allowed
geminates are: -ss-, -mm-, -nn- and -rr-, present only in V_V position.


The most common raising diphthongs are:

in unstressed syllables you can find also:
/ja/, /je/, /jo/, /ju/;

Falling diphthongs:
/aj/, /ej/, /oj/, /Yj/, /uj/
/aw/, /ew/

Phonological constraints:

Most syllables are CV or CVC in form; there are no restrictions on what
(single) consonants can appear initially or finally.

Syllables may also begin with

a stop + r
fricative f + r, s + v or f
s + an unvoiced stop, f, m, or n
s + a stop / f + r

Syllables may end with

r or l plus a stop, f, or s
n plus a stop

[s] disappeared in V_C position about 300 yrs ago; a phenomenon similar to
that happened to French. Be careful that in initial position this didn't

masten > maten because [s] is in V_C position, but:

henster > henster, because [s] is in C_C position, and

strecár > strecár, because [s] is in _C position.

[s] is obviously still there in learned words (especially scientific ones).
If there had been a word *astronomea (which can't exist, because on the
planet where Sirmave is spoken noone ever brought Greek words), I think that
wouldn't have undergone the process.

<can't resist, let me play with this...>
On the other hand a vulgar word *astronomíl, originally meaning 'little
astronome', once it had gained the meaning 'wizard, warlock' in the popular
speech would become atronomíl, then syncope to tronomíl and finally loose
its l which is really unstable in the region and become tronomí; then again
syncope and become tronmí > trommí > tromí; in some scattered villages [tr]
generally pronounced /t*/ becomes /tS/, and the infamous word becomes cxomí,
a word which could be accepted in the language and whose origin would surely
cause Sirmavian linguists and etymologists terrible headaches : )))

[l] in the very old times disappeared in consonant clusters (as it happened
in French - animal + s becomes animaux via animals - or in Dutch - old >
oud); in Simave there are, anyway, some -lt-, -ld-, -lc- etc. clusters; this
because the cited syncope phenomenon caused some syllables to collapse when
the VlC > VwC change was already finished. Thus we'll have alda > auda, but
alida > alda. Notice that the [l] in the second case is pronounced /l./,
which is bound to become /w/ as well; the status of the phoneme [l] has
indeed always been particularly unstable in the language. In some dialects
it could even disappear in intervocalic position (as it did in the
natdialect of Ligury, here in Northern Italy).


The phonemes I had represented above are translitterated this way:

              simple                nasal
high    i            u           in       un
mid      e ö    o               en    on
low           a                         an

[Y] in unstressed syllables is pronounced /o/, and written as such; this
means that if you find a <ö> in the word, that is surely stressed.

                lab    dnt    vel    uvu       glot
stop         p b    t d    k g    q
fric           f v     s         y                h
affr               cx gx z
nasal         m        n
lateral                l
flap                    r

[tS] and [dZ] are translitterated c-dot and g-dot (as the Gaelic letters);
alas I can't use Unicode on this list (not yet, not yet : ( ), so I've used
the ASCII Esperanto code which is quite cool.
Another thing I find cool is the velar voiced fricative written <y>.


Stress generally falls on the penultimate syllable. A syncope phenomenon
levelled a lot the irregularities present in the older system ('CV1CV2CV3
often became 'CV1CCV3). Irregular stresses are marked with a grave: Sirám
(probably from an older Scirammus; then the -us ending disppeared, but the
stress remained there).

The first part's done. Dunno when I'll have time for the second.

Comments are welcome,