The 2017 Smiley Award Winner: Idrani

The 2017 Smiley Award is presented to Idrani, an artistic language with an army of intricate orthographies created by Trent Pehrson. Congratulations, Trent!

Smiley Award 2017

Conspiratorial Cohort

In third grade, Trent had a few friends who enjoyed passing notes back and forth during class. Trent's third grade teacher was opposed to this practice, for obvious reasons, and decided to try to put a stop to it. Whenever he caught Trent and his classmates passing notes, he would take it upon himself to snatch them up and read them aloud in front of the class, hoping to embarrass these ne'er-do-wells. In order to thwart him, Trent and his friends had the idea of creating an English cipher to write their notes in, but their teacher simply treated this as a puzzle. As a matter of necessity, the cipher got cleverer and cleverer until Trent realized that the only effective way to get a note by his teacher would be to change the very language itself.

And thus began Trent's conlanging journey.

From there, Trent would go on to study other languages, and, of key importance, study calligraphy, and little by little what began as a cipher became a full-fledged conlang. Initially that language was called Salofasi, but as it grew in character, it changed its name to what we know today as Idrani.

Like many lifetime languages, Idrani began as unsophisticated experimentation, but today it stands as one of the true artistic achievements in the world of conlanging—in large part due to the extraordinary beauty, complexity, and variety of the many orthographies Trent has created to use with it.

Language Overview

Idrani is a largely suffixing, largely head-final agglutinating language, though it is better described as a language with mixed headedness. Idrani features case-marked nominals, as well as pronominal marking on the verb. Most marking is in the form of a clitic, resulting in variable placement. Here are some illustrative examples:

(Romanization note: Throughout, ' = [ʔ]; ch = [tʃ]; kh = [x]; sh = [ʃ]; y = [j]; and zh = [ʒ]. The full romanization system can be found here.)

  • Ehinpu attedab pish. "My dog likes your father." (Two Overt Nominal Arguments)
  • Atteda ehinpub pish. "Your father likes my dog."

  • Attedab pish. "Your father is liked." (One Overt Patientive Nominal Argument)
  • Ehinpub pish. "My dog is liked."

  • Khanpish. "He likes it." (Two Pronominal Arguments)
  • Nokhpish. "It likes him."

  • Khabpish. "He is liked." (One Patientive Pronominal Argument)
  • Nobpish. "It is liked."

  • Ehinpu Khabpish. "My dog likes him." (One Agentive Nominal Argument and One Patientive Pronominal Argument)
  • Atteda Nobpish. "Your father likes it."

Generally a separate construction is used with a patientive nominal argument and agentive pronominal argument, but just in case, here's what each looks like, as far as I understand:

  • Ehinpub khapish. "He likes my dog." (Standard Case Marking)
  • Ehinpu zhu khapish. "He likes my dog." (Topic-Comment)

Now that you know basically how this bit of grammar works, here's an interesting tidbit. Let's say you want to say "I like ice cream, dogs, and the color red" (an example sentence after my own heart). You know you're going to need the three nominals, our word for "like" modified with a first person agent marker, some kind of coordinator, and case marking. If you have all that stuff, it should look something like this:

  • Yateyab, inpukab, icheyib hai hepish. "I like ice cream, dogs, and the color red."

So hai is our coordinator, which comes at the end; you should recognize pish by now; and then you have our three nominals each with a -b suffix which should look familiar. Now, you can say this, and it works. But you can also do the following:

  • Yateya, inpuka, icheyi haib hepish. "I like ice cream, dogs, and the color red."

Isn't that cool? I don't recall ever seeing anything like that, but it reminds of Moten. Basically, the coordinator hai is accepting accusative marking and passing it down to all its constituents. Makes sense that the coordinator comes at the end, then, doesn't it?

There are also some interesting morphological alternations going on in Idrani that are worth noting. For example, let's say you want to say "X is an animal", where X is anything (that, presumably, can be classified as an animal). How you do it depends on the phonological character of X. Here are some examples:

  • Ehinpu'ush qadiya. "My dog is an animal." (Ehinpu = "my dog")
  • Ehaisri'ish qadiya. "My mouse is an animal." (Ehaisri = "my mouse")
  • Ehtsiksa'ash qadiya. "My bird is an animal." (Ehtsiksa = "my bird")
  • Ehstoshe'esh qadiya. "My donkey is an animal." (Ehstoshe = "my donkey")
  • Ehchigo'osh qadiya. "My teacher is an animal." (Ehchigo = "my teacher")

As you can see, the form here isn't fixed. Basically you add -' plus a copy vowel and add a tense suffix to that to get "X is", which is fascinating. There are a number of affixes in the language that operate under the same principle, either adding a copy vowel, a copy syllable, or lengthening a short vowel as a part of the affixation process. I'm always a sucker for that kind of thing.

This should be enough grammar to give you the flavor of the bird. Now to move on to what makes Idrani truly exceptional: Its many orthographies.

Thirty-Six Years of Orthographies

At the time of writing, I'm thirty-six years old. I was born in 1981, which, according to Trent's timeline, is the year the first two orthographies of Idrani were born. Since those two, he's gone on to create forty-seven other orthographies for Idrani. And those are just the ones listed on his website. And, of course, that's just for Idrani; he's done conlang and orthography work elsewhere, as well. If that weren't enough, each of these systems is unique in its precise function, with various of them being alphabets, abjads, syllabaries, logographies, and some systems being basically unattested until Trent invented them.

Before getting into the systems themselves, it's best to simply appreciate the look of the scripts. If there's anything that ties all of Trent's stylistically divergent scripts together, it's their extremely high quality. Trent's study of calligraphy clearly aided him in the creation of these scripts, as his linework is unparalleled. It's a hard thing when creating a script to get each of your glyphs to look exactly the way you want them to, but with Trent's stuff, you know that every single line looks exactly the way Trent intended. Here's a small sampling of Trent's work to give you an idea of the kind of variety he's capable of:

An example of Trent Pehrson's Mahruwa script.
Mahruwa Script (1985)

An example of Trent Pehrson's Neototem script.
Neototem Script (1985)

An example of Trent Pehrson's Hinji script.
Hinji Script (1984)

An example of Trent Pehrson's Skeryl script.
Skeryl Script (2013)

An example of Trent Pehrson's Rohami script.
Rohami Script (1995)

An example of Trent Pehrson's Nhipii script.
Nhipii Script (1985)

An example of Trent Pehrson's Ksatlai script.
Ksatlai Script (1999)

It's difficult to identify any single thread that ties all of Trent's orthographic work together. Of course, Trent has drawn a family tree for his scripts, showing how one either influenced or led to another, but there are also tendencies—or preferences, perhaps—that run throughout. For example, Trent loves to play with line thickness. It's something I noticed straight off, because it's something I'm personally very bad at, but it's really striking. For example, take a look at these glyphs from the Shyesipokhi script:

An example of Trent Pehrson's Shyesipokhi script.
Shyesipokhi Script (1998)

The glyphs are precise, yet the thickness varies in a wonderfully satisfying way. You can see the same thing in a script that's a bit less mannerly below:

An example of Trent Pehrson's Ilkoin script.
Ilkoin Script (1988)

Something like Ilkoin looks so odd, so otherworldly, and so random in places, yet it all hangs together beautifully. It's a masterful script that nevertheless seems to have no uniform weight—anywhere! Part of that has to do with how the system is constructed, which is a good segue into Trent's next preference, and our next section.

Systemic Complexity

In my book I wrote "Designing a good writing system has everything to do with the system, and nothing to do with the glyphs". Of course, designing great glyphs doesn't hurt, and Trent is superlative at that, but what really makes his writing systems fascinating is the systematicity they display.

Trent has created systems that range all the way from simple alphabetic to nearly inexplicable, but even the simple scripts have interesting quirks. In Tinzha, for example, the vowel diacritics are fairly simple:

Vowel diacritics in Trent Pehrson's Tinzha script.

You'll notice, though, that outside of [u], each vowel has four forms. These forms are employed variously depending on the shape of the consonant (or consonant group) they appear underneath. Here are some examples with a consonant followed by the vowel [ɑ]:

The sequence [ʔɑ] in Trent Pehrson's Tinzha script.   The sequence [qɑ] in Trent Pehrson's Tinzha script.   The sequence [lɑ] in Trent Pehrson's Tinzha script.
[ʔɑ]   [qɑ]   [lɑ]

Though the vowel diacritic always appears below the consonant, it can lengthen to accommodate the extra width of a particularly wide consonant, but also changes its form radically if the consonant is particularly narrow.

Also, though Trent is fond of using a diacritic to differentiate pairs of consonants (often voiced-voiceless pairs, but also some routine pairings that are unique to his systems, e.g. [q] and [ŋ] being distinguished by the same diacritic used for voicing), the way that differentiation is instatiated can itself be rather interesting. Here's a standard example, taken from Trent's Glurg script:

Half of the consonantal distinctions from Trent Pehrson's Glurg script.

Pretty simple: Add the T shape below the voiceless consonant to get the voiced version. Now here's another partial example from Trent's Orgbymme script, with the modified versions shown in red. Sometimes the modification produces a voiced consonant, but sometimes it does something different. As for the modification itself, the best way I can describe it is that the glyph gets slightly more awesome:

Some consonantal distinctions from Trent Pehrson's Orgbymme script.

But this is nothing compared to his later work.

By listing his work the way he does, you can follow Trent's evolution as a scripter. Basically, he started with ciphers, then moved onto scripts for new alphabets, then discovered features, producing featural scripts, but then was able to abstract beyond that. If the addition or subtraction of some graphic mark can indicate a difference between two sounds, then basically any type of change can indicate a difference between any two things. Once Trent realized that, his scripts started to get wild.

In 1987, Trent created his Susipokhi script, which looks cool enough on its own, but also has some radical correspondences. The script is basically an alphabet (can also be a pure abjad), but each consonant has a fat and skinny form. The vowels, then, have a variety of forms and realizations, and basically words are put together either based on what types of shapes are being used, or, in part, based on the choice of the writer.

To give you some concrete examples, this is the vowel [ɑ]:

Short version of the [ɑ] vowel from Trent Pehrson's Susipokhi script.

So is this:

Long version of the [ɑ] vowel from Trent Pehrson's Susipokhi script.

Now here's the consonant [ʃ]:

Short version of the [ʃ] vowel from Trent Pehrson's Susipokhi script.

And its fat version:

Long version of the [ʃ] vowel from Trent Pehrson's Susipokhi script.

Now you could just add the vowel after the consonant to make [ʃɑ], but you could also do this:

The sequence [ʃɑn] from Trent Pehrson's Susipokhi script.

That is the sequence [ʃɑn], with the fat version of [ʃ] followed by the skinny version of [n]. Where's the [ɑ]? It can be inferred specifically because the script moved from a fat character to a skinny character with no overt vowel mentioned. That is brilliant. It can be used for two other vowels as well, with skinny-to-skinny indicating [ɛ], and skinny-to-fat indicating [i].

This type of marking without marking led Trent to creating his various dynamic systems. These are systems whose glyphs have no fixed forms at all. Here's an example from Trent's Tumani script. The two glyphs below read [saka] in the Tumani script:

The sequence [saka] from Trent Pehrson's Tumani script.

The two glyphs below also read [saka] in the Tumani script:

The sequence [saka] from Trent Pehrson's Tumani script.

And so do the two below:

The sequence [saka] from Trent Pehrson's Tumani script.

Don't try to figure how these are stylistic variants of one another, because they're not. In fact the second pair of glyphs come from the Roman script, and the third from my own Kwandon script. What do they all have in common? The first glyph of each pair displays horizontal and vertical symmetry, has a closed loop, has no crossing lines, and has two unconnected lines. The second glyph of each pair displays horizontal but not vertical symmetry, has a crossing line, has no loop, and has two unconnected lines.

And this is how Trent's dynamic scripts work. Rather than having fixed forms, they have rules about what has to be in each shape, and that's it. It's the type of script that would only occur to someone with linguistic knowledge, who is an experienced script creator, and who has a background in cryptography—basically, someone like Trent—and he's gotten a lot of mileage out of this script type. Just compare the Selmi script and the Tumani script to see how different these things can look.

Perhaps my favorite of Trent's scripts, though, is the Pachowi script. Aesthetically speaking, the glyphs themselves have an amazing look to them:

Some glyphs from Trent Pehrson's Pachowi script.

The thing that makes the system wild, though, is its systematicity. Looking just at the phonological portion of the system, there are twenty-eight glyphs, and each glyph has four different basic realizations, depending on its diacritic. This is rather like an abugida, where the base glyph (using Devanagari as an example) क stands for [ka], and if you add a diacritic you get a different vowel, so की is [kii] and को is [ko], etc. The difference is that the four different forms you get have nothing whatsoever to do with each other! Here are some examples:

Glyph No Diacritic First diacritic from Trent Pehrson's Pachowi script. Second diacritic from Trent Pehrson's Pachowi script. Third diacritic from Trent Pehrson's Pachowi script.
First sample glyph from Trent Pehrson's Pachowi script. ɪl tɬɑ pu
Second sample glyph from Trent Pehrson's Pachowi script. dʒi kʰɑ ɣɑ li
Third sample glyph from Trent Pehrson's Pachowi script. ʃɑn ti skɛ pi
Fourth sample glyph from Trent Pehrson's Pachowi script. mɑs tθɑ ɡo

Isn't that awesome?! Reminds me of an idea I implemented for a derivational strategy in Gweydr ages ago. As you can see, it works just like an abugida, but the results are entirely unpredictable: The vowels are different; the consonants are voiced and voiceless; the syllables are both CV and VC; and some, but not all, syllables have consonant clusters. Not a single thing is predictable about this system—and we're just getting started!

In order to get at every consonant and vowel combination needed to write Idrani, there are also a series of secondary modifiers which can change the vowel of a glyph to something specific:

Nucleus conversion glyphs from Trent Pehrson's Pachowi script. It shows glyphs that turn the vowel into [ɑ]; into [i] or [ɪ]; into [u]; into [ɛ]; into [o] or [ɔ]; or flip the entire syllable around.

And then there's that last glyph that will flip the syllable around.

So let's say you wanted the syllable [ɑɹ]. First thing you'd do is find the closest syllable to [ɑɹ] that has the same consonant (and no others), but not necessarily the same vowel—or the same order. That syllable turns out to be [ɹi]. [ɹi] is one of the four realizations of this glyph:

Glyph that can read [qu], [bi], [slu], or [ɹi] from Trent Pehrson's Pachowi script.

That glyph can also be [qu], [bi], or [slu], so we'll need to add the appropriate diacritic to get the variant we want, [ɹi]:

Glyph [ɹi] from Trent Pehrson's Pachowi script.

Now we need to change the vowel to get [ɹɑ] instead of [ɹi]:

Glyph [ɹɑ] from Trent Pehrson's Pachowi script.

And finally, we need to add the reversal glyph, to get [ɑɹ] instead of [ɹɑ]:

Glyph [ɑɹ] from Trent Pehrson's Pachowi script.

And this isn't even getting into the other aspects of this script, such as the use of color, how you make coda consonants, and the presence of determinatives. Pachowi is a fantastic example of how Trent is able to take everything he knows about language, orthography, cryptography, and systems in general and combine them into something masterful. But the mind boggling thing is we haven't even touched on his REALLY complex stuff. His skill in orthography is unparalleled.

How Idrani Has Made Me Smile

I remember the first time I stumbled across Trent's site. I went immediately to the orthography page, since that's what I always do, and I saw some stuff that was impressive, but I thought to myself, "I could do stuff as good as this if I had the time..."

Then I saw this:

An example of Trent Pehrson's Suksatlai script.

In that moment I thought to myself, "Well, that's that. I'll probably never create anything as good as this one image." But I couldn't even be mad. I didn't understand it, but I knew it was brilliant, while at the same time managing to be beautiful and otherworldly. You didn't need to understand it to appreciate it, but if you went the extra mile, it was going to blow your mind. It was then that I knew Trent was something special, and someone to keep an eye on.

Since that day, Trent has gone on to do even more amazing stuff, and, I'm glad to say, has also come to be my friend. When I was putting together The Art of Language Invention, I was going to have a chapter on orthography, and I felt like there was no way I could do it without including Trent's work. I really wanted to include the image I showed you above there, hoping it would strike my readers the way it did me. I asked Trent about including it, and he instead offered to create an original piece just for my book. This is what he gave me:

An example of Trent Pehrson's Ksatlai script.

A stunning work, but as excited as I was to see it, it was even more important to me that book readers would see it. For many readers, the book would be the first time they ever heard of language creation. Furthermore, when you have an arcane topic like language creation, if it starts to get confusing, a learner will tend to jump to something that's easier to understand—like the orthography section, which is highly visual. There are plenty of more traditional script-like scripts in that section, but I wanted to include something that would blow readers away—something that would help to show them just what the conlanging community is capable of. I couldn't have asked for anything better than Trent's work. That's what I want people to see when they come looking for the best of the best.

In addition to all of this, Trent has been incredibly generous with other conlangers, both in terms of his time, and his incredible artwork. If you go to his artifacts page, there's a whole bunch of art pieces he's done for other conlangers simply because he thought it was a nice thing to do—and that list isn't even complete. Trent is an absolutely brilliant artist, and I think we in the conlanging community are very lucky he and his friends decided not be deterred by their nosy teacher in elementary school. Trent is a fantastic conlanger, and a fantastic human being, and I'm happy to award his language Idrani the 2017 Smiley Award. Congratulations, Trent!

Back to the Smiley Award Main Page

This page was last modified on Friday, August 3, 2018.
This website was last modified on .
This page can be viewed normally, as a milk or dark chocolate bar, in sleek black and white, or in many other ways!
All languages, fonts, pictures, and other materials copyright © 2003- David J. Peterson.

free counters