Valdyas is a fictional country in a fictional world. Ilaini is an invented language. I invented it for my own enjoyment and that of anybody else who enjoys it. It has no scholarly pretensions. Please don't make a fool of yourself by using my language pages for your linguistics case-work.

Valdyan scholars distinguish between laynyin coli "open sounds" (vowels) and laynyin nacoli "closed sounds" (consonants), with laynyin disi "changeable sounds" (diphthongs) in between - rising diphthongs change from closed to open, falling diphthongs from open to closed.

There is no set order for letters or phonemes. Most lists and gazetteers are ordered thematically. Mailei Halla put the letters in the order shown below when she designed the system of writing that has become the standard.

If something does need to be ordered by letter, like a list of clients, usually only the first letter is taken into account.

Speech sounds

I use Evan Kirshenbaum's ASCII IPA representation with some slight modifications. The real IPA font is SIL Doulos IPA, distributed by SIL, formerly the Summer Institute of Linguistics. For speed, I give all inline examples in either transcription or ASCII IPA. This means that you don't have to have the IPA font installed or to load a lot of little pictures. If you want to try the Ilaini font used for the examples in the table, you can find it on our constructed languages page.

Laynyin coli - Vowels

Ilaini Ilaini vowels
Transcription a y i e o u
IPA IPA vowels
ASCII IPA a i" i e, E O u

In the south <y> is still long, one of the few long vowels left after the vowel shift that happened some time between the founding of the kingdom and the time of Vegelin the Great. In some southern dialects (notably in Essle) <y> is pronounced [i:] and its length is the only thing to distinguish it from <i> [i]. When <y> or <i> is in an unstressed final syllable, for instance in many proper names, it's often flattened to [I]. Most names in -in have a variant spelling in -yn and vice versa.

Of all the vowels only <e> has two markedly different pronunciations: [e] in open syllables, [E] in closed syllables. When unstressed at the end of a word, <e> tends to shwa, but usually doesn't quite reach that stage.

Unstressed <e> at the beginning of a word (as in echain "past") is almost always pronounced [a].

Stressed <e> and <i> at the beginning of a word are usually pronounced (and sometimes written) [je] and [ji].

In the dialect of the Western Plain, <u> is fronted to [y] and speakers of this dialect tend to hear the eastern [u] as closer to [o] or even [au], especially in dialects that have retained the long vowel [u:] (for instance in most of Ryshas).

Laynyin disi - Diphthongs

Falling diphthongs

Ilaini Ilaini falling diphthongs
Transcription ai ay oy au ei ey iy
IPA IPA falling diphthongs
ASCII IPA a; A; O; a<w> e; e; i;

In most dialects <ai> is pronounced [a], always stressed, where the spelling with i serves only to show the stress. In the north, there's little difference between <ai> and <ay>, except that <ai> is always stressed, even in a normally unstressed syllable (as in the name Valain). In southern dialects <ay> is pronounced very long, [A:j], and this is the feature of southern speech most often made fun of by northerners.

In careful southern speech, the <e> of <ey> is slightly longer than that of <ei>. In the north there's no noticeable difference.

Rising diphthongs

Ilaini Ilaini rising diphthongs
Transcription ea ya ia ie ye yi
IPA IPA rising diphthongs
ASCII IPA E@, ja ja ja je je ji

The glide of <ia> and <ie> is considered to be more open than that of <ya> and <ye>, and the glide of <ea> more open still, but in practice it's hard to hear the difference. In some dialects (that of Valdis, among others, and as such considered 'civilised') <ea> is pronounced falling, much like (British) English non-rhotic air.

Laynyin nacoli - Consonants

Ilaini Ilaini consonants p-m
Transcription p b f v m
IPA IPA consonants p-m
ASCII IPA p b f v m

In western dialects (the region around Ildis) the difference between <f> and <v> is much greater than in eastern dialects, so much that someone from the west will hear eastern <v> as [f] and someone from the east will hear western <v> as [w]. Eastern [v] is often hardly voiced at all, only somewhat less tense than [f]. The same goes for [s] and [z].

Ilaini Ilaini consonants t-lh
Transcription t d th dh n l lh
IPA [IPA consonants t-lh]
ASCII IPA t d T D n l s<lat>

The sound indicated by [s<lat>] is the voiceless lateral fricative spelt <ll> in Welsh, <lh> in transcriptions of Tibetan and <hl> in transcriptions of Teonaht. It can indeed be seen as a species of <s>, though I (like the Welsh, the Valdyans and the Teonim) prefer to see it as a species of <l>.

A neat trick to pronounce it: put your tongue in position as if to say <l> and, without moving the tongue, try to say <s>.

Ilaini Ilaini consonants k-rh
Transcription c g ch gh h r rh
IPA IPA consonants k-rh
ASCII IPA k g x Q h r" r"<o>

I transcribe the [k] sound with <c> partly for historical and esthetic reasons, and partly because both <c> and <ch> tend to be palatalized before front vowels: they become [c] and [C] respectively.

There's no voiceless aspirated uvular <r> (transcribed <rh>) in Kirshenbaum's ASCII IPA, so I write it as a devoiced uvular <r>. In most people's speech <rh> has some aspiration.

The uvular [r"] and [r"<o>] have apico-alveolar allophones: (voiced) [r<trl>] and (voiceless aspirated) [r<o>].

Ilaini Ilaini consonants s-j
Transcription s z sh zh j
IPA IPA consonants s-j

In western dialects <j> at the beginning of a word is usually pronounced [dZ] (voiced alveolar affricate); in Idanyas it often disappears altogether, so that Jeran will hear his name pronounced ['eran] (or even ['e:rjan] or ['erian]) in Dol-Rayen, ['jeran] in Valdis (the standard usage) and ['dZeran] in Tal-Crun.

Note that in the dictionary [j] still has the old transcription <y> in most places; I'll fix that when I overhaul the dictonary at some unspecified time in the future.