The 2016 Smiley Award Winner: Ilaini
I am very pleased to present the 2016 Smiley Award to Ilaini (formerly known as Valdyan), an a priori language created by Irina Rempt. Ilaini is a language created for the world of Valdyas, which later served as the setting for a role playing game, a novel, and a wealth of concultural detail. Irina is a long time conlanger, and I'm delighted to give her language Ilaini the 2016 Smiley Award. Congratulations, Irina!
http://www.valdyas.org/irina/valdyas/ilaini/
For My Own Enjoyment
Early on in Irina's website's FAQ, she addresses the relationship between her ongoing role playing games set in the world of Valdyas and the world itself. Of it she writes, "The games exist for the sake of the world, not vice versa". This is a sentiment many conlangers will find familiar. As with Tolkien, the language and the world was a jumping off point for Irina, which led to a novel, and an impressively large and long-running series of role playing campaigns.
Examining Irina's progress on any one aspect of her creation (the world, the role playing game, the fiction, the language) makes one wonder how she had any time to work on any other aspect of it. The output is truly impressive, and the world, which has been developing for many years now, has a ton of fascinating detail (I'm always a sucker for monetary systems). Today we focus on her language, Ilaini, which was known for many years in the community as Valdyan.
Introducing Ilaini
Ilaini is a dependent-marking, mixed-headed SOV language. Ilaini has two genders that basically fall under the categories animate (common) and inanimate (neuter), as in Swedish. Nouns decline in six cases (the nominative, genitive/ablative, dative/illative, accusative, locative/instrumental, and vocative), and verbs have four aspects and inflect in four tenses, with some other miscellaneous forms (e.g. the imperative, which never seems to fit neatly into a paradigm slot). The semantics of the verb paradigm are quite interesting. Here's a comparison of the four aspects:
(Romanization note: Throughout, y = [ɨ]; j = [j]; ch = [x]; c = [k]. Diphthongs are a little more complex.)
- Lyase estean chalut. "The woman saw a fish." (Imperfective Aspect)
- Lyase estean chalenut. "The woman had seen a fish." (Perfective Aspect)
- Lyase estean chalesut. "The woman was about to see a fish." (Inceptive Aspect)
- Lyase estean chalynut. "The woman looked at a fish." (Punctual Aspect)
You can see with the last example, what began as an inflectional aspectual distinction has taken on more of a derivational role. The current state of Ilaini is an interesting one, as we can see the change happening before our eyes, for though the punctual aspect can be used with any verb, it can also take additional aspects, as shown below:
- Lyase estean chalynenut. "The woman had looked at a fish." (Perfective Aspect)
- Lyase estean chalynesut. "The woman was about to look at a fish." (Inceptive Aspect)
And though the forms are rather compositional when written out, in practice, there are a number of rules that govern where certain consonants can be dropped, the result being some unpredictable verb forms. Based on what I've seen, it looks as if certain "weak" vowels are dropped in the second syllable of tetrasyllabic forms (stress is usually word-initial). Here are some examples (note that consecutive vowels represent diphthongs, whose forms aren't identical to two consecutive vowels of the same quality):
| Expected Form | Actual Form | English |
| chalynenut | chalyenut | "s/he/it had looked" |
| chalynesut | chalyesut | "s/he/it was about to look" |
| chalesaye | chalsaye | "you (plu.) are about to see" |
| daysenesut | daysesut | "it started to rain" |
Presumably, though, one wouldn't see contraction in a form like bastesaye, "you (plu.) are about to beat", as the sequence [sts] is illicit (though I'd be curious to know if duchsaye, from duchesaye, meaning "you (plu.) are about to esteem" would work, with the sequence [xs]. Update: I've heard from Irina that it actually contracts to dushaye, with /xs/ becoming [ʃ]!). It's not clear what would happen with a form like talaynenan, "I have whispered", as the ay diphthong tends to be pronounced long (especially in Southern dialects). Though unstressed, it certainly wouldn't be lost, so I wonder if it'd be possible to see the form talaynean. I suspect it would be, but that's just my intuition.
Another cool aspect of the grammar is the reflexive particle le. It bears a clear etymological relationship with the third person pronominal series (compare lean "s/he", lei "he", lye "she", lin "it"), yet it doesn't inflect for case, and its distribution is extremely limited. First, here is its standard usage:
- Lyase chalynut. "The woman looked." (Intransitive)
- Lyase estean chalynut. "The woman looked at a fish." (Transitive Non-Reflexive)
- Lyase le chalynut. "The woman looked at herself." (Transitive Reflexive)
There is no change in the form of the reflexive for different persons, genders, or numbers: it's always le (in some dialects, at least). Now here's how it interacts with clausal negation:
- Lyase na chalynut. "The woman didn't look." (Intransitive)
- Lyase estean na chalynut. "The woman didn't look at a fish." (Transitive Non-Reflexive)
- Lyase na le chalynut. "The woman didn't look at herself." (Transitive Reflexive)
Now that is cool! Basically what we have here is a reduced pronoun en route to becoming an inflectional—or even derivational—prefix. As a bit of lexical material grammaticalizes, it generally becomes phonologically reduced (compare le to the other third person pronouns), semantically reduced (notice how it's no longer simply third person, but works with any person, gender, and number), and it also begins to form a tighter unit with its head (as evidenced by its unusual position for an object). This is a textbook example of grammaticalization, and it's awesome!
Here are a couple more fun quirks with le. First, it can be used as a resumptive pronoun. There's a complex example on Irina's site, so I'll try to create a simpler one to illustrate.
- Lyase ansinat. "The woman thinks." (Intransitive)
- Esten lyasea chalynut. "The fish looked at a woman." (Transitive with Nominal Object)
- Esten lye chalynut. "The fish looked at her." (Transitive with Pronominal Object)
- Lyase esten le chalynut ansinat. "The woman thinks the fish looked at her (= the woman)." (Resumptive)
- Lyase esten lye chalynut ansinat. "The woman thinks the fish looked at her (= some other female person)." (Non-Resumptive)
- Lyase esten liarne le chalynut ansinat. "The woman thinks the fish looked at itself." (Reflexive)
In the last example above, liarne is used for emphasis, to make it clear that that the embedded verb is reflexive, and the le isn't being used as a resumptive pronoun. Very cool!
It's also worth pointing out that le can crucially alter the meaning of words. For example, rada means "to swear", and le rada means "to decide". In the section on reflexive pronouns, Irina provides this fascinating example: le tisa le rada "to decide to serve oneself" (where tisa means "to serve"). In one phrase, we have le serving in a grammatical capacity as a reflexive pronoun, and also serving in a derivational capacity to render the meaning "to decide"! This is the type of stuff I love discovering in both natural languages and constructed languages.
Case by Case
Those who know me know I love case languages, and love fancy case forms. There's a lot of fun stuff going on with Ilaini's case system.
First, several case distinctions collapsed due to the merger of long and short vowels. This caused Ilaini to lose 3 cases, dropping its total from 9 to 6. Two of the mergers make a lot of sense, semantically (the genitive and ablative merging, along with the dative and illative merging), but the last is really interesting: the locative merging with the instrumental. Granted, the first would usually be used with locations, and the second with objects and maybe persons, so there's little chance of confusing the two (e.g. who would translate Reshien halun as "I sang with a forest" as opposed to "I sang in a forest"), but how cool!
Also, similar thematic endings in certain declensions result in further mergers for some words. For example, the nominative and vocative often take the same form (they're only routinely distinguished in the first declension in the singular), and often the locative/instrumental ends up sharing a form with the vocative as well. Where the vocative and locative/instrumental are different in the singular, there is often syncretism between the locative/instrumental singular and the vocative plural—but only in some declensions. There's a lot of texture here that's interesting to pore over.
The forms themselves are very well constructed, especially in the singular. Each declension class is familiar, but distinct in one way or another. Compare this partial list of nominative and genitive/ablative singular forms (note: there is no formal distinction between adjectives and nouns):
| Declension Class | Nominative Singular |
Genitive/Ablative Singular |
English Gloss |
| Class I | hinla | hinlei | "lark" |
| arle | arlei | "truth" | |
| hanea | hanei | "youth" | |
| Veray | Veryi | Place Name | |
| Class II | colsen | colsein | "key" |
| ochin | ochein | "broken" | |
| varzas | varzeis | "island" | |
| dilyn | dileyn | "event" | |
| Class III | mudh | mudhein | "health" |
| asel | aslei/aselei | "faded" | |
| lau | laui | "moon" | |
| Class II-III | men | mein/menei | "gift" |
| rhin | rhein/rhinei | "boat, ship" |
This is just a sampling of two case/number combinations, but I love the result. If you look at the genitive/ablative singular forms, you get the idea: there's some sort of ei in there if possible; if not, some sort of palatal thing going on. How it's reified, though, depends on the character of the word. I also love that you have some variation in the forms. This is what we have with certain plural forms in English, and even certain verb forms, depending on the dialect (e.g. does "houses" have a [z] in the root or not? And does it make a difference if it's a plural noun or a present tense verb with a third person singular subject?).
I love seeing this type of variation in a conlang—especially when there's a reason for it to exist. In this case, with the Class III form above, there's the tension between the "weak" syllable that we see drop in certain verbal forms, and in the Class II-III forms, whether the root is inviolable, as in Class III, or whether they're able to count as Class II forms. The Class II-III paradigm is worth a look, anyway, as Irina captures very well the distinction between forms which are impossible, forms which are in free variation, and forms which are rare but still possible. It's a great piece of conlanging!
The Starlings' Song
Though the Smiley Award is intended to recognize a conlang, I simply couldn't talk about Ilaini or Irina without talking about the Conlang Relay. For those who are unfamiliar, the Conlang Relay is probably the conlang community's most enduring and proudest tradition (well, aside from arguing about how we all pronounce some particular English word), and it all started with Irina Rempt.
Translation exercises have always been a fun way to test out a conlang and to compare one conlang to another, but in the summer of 1999, prompted by a request from Sally Caves, Irina initiated what would become the very first conlang relay. She sent a text in her language Ilaini to Sally to translate into her language, Teonaht. This caused Irina to think of the game Telephone, where one player whispers a short message to another, and that one to another, and so on, until the original message has become mixed up at the end of the line. This is all done in the players' native language, of course. Well, thought Irina, what's the fun in that? Why not instead add the complexity of decoding a text written in another person's conlang, and then translating it into one's own and passing that on?
And so a tradition was born.
The very first conlang relay was based on Irina's text Hanleni Halsen: The Starlings' Song. This is the text below:
Hanleni halsen varyenan laynat
Daysinen verein idanla le listat
Havien hinla laziena forat
Culea rachleni arlea a chalat?
The song of the starlings speaks of heroic deeds
In the morning rain the heron does its laundry
In the night the lark worships the stars
Who sees the true nature of birds?
After it had passed through 17 other conlangs, the text came back to Irina, which she retranslated and presented to the Conlang-L:
Rachla moy arnei halsean halesit
Rozein rachlei jat foyin morhiyis sali
sudine dir arnei shonean shonynesyit
Halla jat havein laziena numena laynesit
Valan alea halsean jat chynesit.
The large bird shall sing a song of itself
This bird of the river's feathers shall dance,
between the clouds of some ground, a dance of themselves
This song-bird shall speak of the night's awesome stars
Every king shall hear this song.
The results were so hilarious, and the game so amusing, that it wasn't long before a second relay followed the first—and then a third, and a fourth, and a fifth, and just this past autumn, we saw the twenty-third relay. Irina has participated in over a dozen relays (including the most recent, at the time of writing), and maintained for many years the official Conlang Relay Listserv, which was dedicated to nothing but facilitating conlang relays.
But it doesn't stop there. Unlike some other traditions that have existed on the Conlang-L or in other communities but not gone further, the idea of the Conlang Relay has spread well beyond its origins. There are conlang relays on Tumblr; on Reddit; on the ZBB; on the CBB; and it's become a celebrated biennial tradition at the LCC. Even in content, there have been variations, with two Inverse Relays (conlangers learn another conlanger's language, and translate from their own language into the conlang they've learned), Romlang-only relays, an orthographic relay, Jeffrey Henning's scheduleless relay, a daughterlang relay... The sky's the limit; I'm sure we haven't seen the last type of conlang relay.
All of this, though, began with Irina's Hanleni Halsen. It was the beginning of a wonderful tradition (and a lot of fun!), and Irina not only initiated it, but helped it to grow and flourish with the Relay List. It's been a wonderful gift that the whole conlanging community has gotten to enjoy, and for it she deserves a thunderous round of applause.
How Ilaini Has Made Me Smile
The word for "to urinate" is dusha. I live for terrible puns like this.
More seriously, Ilaini has a lovely script. It's an alphabet with a number of special glyphs for diphthongs (something I'm quite fond of), and it fits together superbly. Here's a nice example:
I've always thought that dragon was rather fetching.
Also, I'm endlessly charmed by this etymological chain:
- rhin (n.) boat, ship (Root)
- rhina (v.) to float, to sail (Verbalization)
- tarhina (v.) to mess about with boats (Diminutive)
- tarhinla (n.) duckling (Avian)
Finally, I'll note quickly that many of us conlangers (and yes, I'm including myself) rarely take the time to speak our languages aloud, as opposed to reading and writing them—and if we do, we rarely take the time to record ourselves speaking them. This is why I thought it was especially wonderful that Irina took the time to record herself singing Hanleni Halsen. I've always found it to be wonderfully beautiful. You can hear it below (or if it doesn't work, right click here to download an MP3 file or here to download an OGG file):
There's really no quantifying phonaesthetics, but I've always found Ilaini to be a very beautiful sounding language. In addition, it's served as an inspiration to the conlanging community for...geez, more than 20 years at this point (we're getting old!). Irina's site is a wonderful place to get lost for a while to learn about Valdyas and its inhabitants, and I'm proud to present Irina and her Valdyan language with the 2016 Smiley Award. Parinay, Irine!