Having decided to embark on the pointless and Quixotic task of creating my own conlang, the obvious question was what to call it? I chose to call it Pixie out of whimsy. It's simple, easy to remember, and if you can have languages for elves, goblins, Vulcans and Klingons, why not a language for Pixies too?

As I'm new at this, I thought I'd start a thread so other people can offer ideas and suggestions that I might not have thought of otherwise. Also so I can share it with others (assuming anyone is interested).

So if you have any advice for me, or ideas that you'd like to see made part of a language, please feel free to speak up. I might not follow all the advice and suggestions I'm given, but I'll at least think it through.

I'm also planning to put details of the language on a website.
The address is: http://www.deadsquirrel.net/pixie

At the moment it's mostly got the same information I'm providing here, although it does have additional information about the Pixie calendar, and more details and history of the Pixie numbering system. (I'm not putting that information here, because this post is going to be far too long as it is.)

It's also got a True-Type font for the Pixie script:
Download: http://www.deadsquirrel.net/pixie/files/Pixie.ttf


Here's what I have so far...

Symbols: See first attachment.

Letters

The sixteen letters shown above will be be given names consisting of a consonant followed by a vowel (I haven't named them yet).

When the letter is written in the "high" position, it represents the consonant from it's name, and when written in the "low" position, it represents the vowel from it's name. Essentially, there are thirty-two phonemes represented by sixteen symbols.

The Pixie letters may be represented by the Latin letters from K to Z. Upper-case represents the "high" position, and lower-case the "low" positon. Letters are written from left to right.

Numbers

A hexadecimal numbering system is used.
The top-left numeral in the chart represents zero.

Punctuation

The most common punctuation marks are shown in the chart, with the dots being the most significant. One dot represents a short pause (similar to a comma), two dots represents a long pause (similar to a period/full-stop, but does not necessarily represent the end of a sentence) and three dots represents an indefinite pause (indicating the end of the body of text).

The presence of three dots at the end of a body of text should not be confused with the use of ellipsis in English. The symbol that most closely fills the role of ellipsis is the "trailing off" symbol which can be seen at the far right of the top line of punctuation marks.

The "trailing off" symbol is used within a sentence to suggest an absence or omission, or at the end of a paragraph to suggest that the text continues elsewhere, either in the following paragraph after a dramatic pause, or in a different location indicated by the sentence that immediately follows the paragraph. In cases where ellipsis would sometimes be used to indicate a pause, using two dots to indicate a long pause should be used instead.

Introductory Glyphs

In the Pixie language, every body of text begins with an introductory glyph. These glyphs are used as a cue to give the reader an indication of the nature of the text that follows. These glyphs are unvoiced, never read out loud. They are closer in nature to punctuation than letters.

Newspaper articles, short stories, letters, diary entries, chapters of books, etc, should always begin with an introductory glyph and end with an indefinite pause.

Headings, dates, bylines, addresses, short notes, ect, do not need to be preceeded with an introductory glyph.

The introductory glyphs shown above are used as follows:

1. General-Purpose neutral
2. Personal, non-intimate
3. Personal, intimate
4. Translation or oral tradition
5. Raw record of facts & figures
6. Dictation, transcription or record
7. Recollections of past events
8. Fiction


Character Set: See second attachment.

Pixie Character Set

At the dawn of the digital age, America created the ASCII character set as a standard for text communication in English. This was shortly followed by other nations creating variations of this adapted for use with their own language. Naturally, Pixies did the same with their language. (Just because they're tiny magical fairytale creatures doesn't mean they can't use modern technology.)

Immediately apparent is the usefulness of this chart for romanticizing the Pixie language. One needs only to locate the Pixie symbol on the chart (yellow) and substitute the ASCII equivalent (blue).

Pixie text files that use this chacter set use the PXT filename extension. Since the contol codes are the same as ASCII, these files can also be opened and edited with a standard ASCII text editor.

If you examine the chart, you may notice that the PICSII symbols equivalent to the letters A to F are identical for both upper and lower-case. This is deliberate, so that hexadecimal values represented as letters in English will be displayed with the correct Pixie symbol regardless of case.

At first glance it may appear that the PICSII symbols represented by the letters K to Z are also identical to for both upper and lower case, but closer inspection will show that the symbols are positioned at different heights.


Syntax

I want to get the phonemes settled before I start worrying about sentence structure, but I recently read this...


From: http://conling.wikia.com/wiki/Linguistic_Universals

Since it's most common to put the subject before the object, I thought about bucking the trend by putting the object before the subject. Trying it out, I think I like the result. The sentences have an odd Yoda-like feel to them.

For example, instead of "John will kick the bucket", you'd say "Kick the bucket will John".

Then I had another idea, to seperate the tense from the verb.

"John will kick the bucket" / "Kick the bucket will John".
"John kicked the bucket" / "Kick the bucket did John".
"John is kicking the bucket" / "Kick the bucket does John".

Repeated or ongoing action could be indicated by pluralizing the verb.

"John was busy kicking the bucket" / "Kicks the bucket was John occupied."
"John will be kicking the bucket for an hour" /
"Kicks the bucket for an hour will John."

It could be an interesting way to set up a language.

So, basically, you're looking at a default word order of: V O Aux S. Here "Aux" stands for "auxiliary". Though your suggestion about "pluralising" the main verb makes me wonder if you aren't actually proposing a combination of verbal noun and light verb.

Another possibility to consider is simply having widespread use of topicalisation in your language. The ordering of object and subject would be determined by news value. In the overwhelming majority of topic-prominent languages, items of less news value come first, but you could always buck the trend here as well and do the opposite.

That's what I was thinking of, but since then I've thought about it in the context of telling stories, and realized that an SO structure has advantages over an OS structure, which is probably why SO is more common.

For example, when you hear or read a sentence like "Kermit the Frog slipped on a bannana peel" you already have an image of of Kermit in your mind before you get to the bit about slipping on a bannana peel, so as you hear/read that part you can easily imagine him crying out in surprise and flailing his arms wildly. But if the sentence were "The banana peel was slipped on by Kermit the Frog", you have no face or form to attach to the image of slipping on the banana peel until after you've passed that point, so the sentence has less impact.

But maybe we could have a language that has no default subject/object order? The ordering of the verb and auxiliary verb (thanks for suggesting the term, I had no idea what to call it) could determine which noun was the object and which noun the subject.

Either O Aux V S or S V Aux S would be valid.

"Sally was swallowed (by a) wolf" or "(A) wolf swallowed (by) was Sally"


I'm going to do a bit of reading up on those things before I understand what you mean here (thanks for the links). I've got a lot of learning to do.

For the alphabet, I think I've figured out which vowels an consonants I'll be using, but I've still got to sort out how I'm going to pair them up. Choosing sixteen consonants was the easy part, but finding sixteen vowels was tricky, as I didn't want to use vowels that sounded too similar to each other. Some of the ones I've chosen still sound a bit too close for comfort to my ear, but that's probably because I'm not used to distinguishing between them.

I'll probably post a list of letter names tomorrow. I'm hoping someone familiar with IPA will be able to verify that I'm not making awkward combinations. (For some of the vowels I'm relying on sound recordings and IPA descriptions, and may not be pronouncing them correctly myself, so I don't fully trust my judgment on that.)

There was actually an interesting crosslinguistic study done a few years back where speakers of English, Spanish, Mandarin, and Turkish were asked to relate simple actions using gestures alone. All of them strongly preferred to mime the agent first, followed by the patient, then the action. So for "Kermit slipped on a banana peel", the overwhelming choice would've been to mime Kermit first of all, followed by a banana peel, then the action of slipping.


So the auxiliary likes to be next to the object but can come before or after it? Pretty bizarre.


Sixteen vowels is a bit crazy if you're distinguishing them by vowel quality alone (e.g. height, backness, roundedness, etc.). Not that crazy, however, since English already has fourteen distinctive vowels and we don't even have front rounded vowels like /ø/ or /y/ which are common enough in other languages (particularly European languages). But you could also add in distinctions of length, tone, nasality, etc.

Sixteen consonants and sixteen vowels is an interesting setup for a phonemic inventory. Most natural languages have about 20-30 consonants (substantially fewer in the Austronesian family and substantially more in the Caucasian family) and about 5-10 independent vowels and diphthongs. As previously suggested, nasalizing would help create a full set of 16 vowels, and you might also consider adding diphthongs.

There are a very few OSV and OVS languages, mostly found in South America. You may want to look at Xavante to see how it works in practice. Oddly, there is also one well-known OVS conlang: Klingon.

Here's my first draft of the alphabet. Any feedback would be appreciated.

Kk = /kʌ/ (voiceless velar plosive, open-mid back unrounded)
Ll = /lɤ/ (voiced alveolar lateral approximant, close-mid back unrounded)
Mm = /me/ (voiced bilabial nasal, close-mid front unrounded)
Nn = /na/ (voiced alveolar nasal, open front unrounded)
Oo = /bʉ/ (voiced bilabial plosive, close central rounded)
Pp = /pɛ/ (voiceless bilabial plosive, open-mid front unrounded)
Qq = /dɞ/ (voiced alveolar plosive, open-mid central rounded)
Rr = /ɹɔ/ (voiced alveolar approximant, open-mid back rounded)
Ss = /so/ (voiced alveolar fricative, close-mid back rounded)
Tt = /tɶ/ (voiceless alveolar plosive, open-mid front rounded)
Uu = /fu/ (voicless labiodental fricative, close back rounded)
Vv = /gɑ/ (voiced velar plosive, open back unrounded)
Ww = /wi/ (voiced labial-velar approximant, close front unrounded)
Xx = /ʒɨ/ (voiced postal-veolar fricative, close central unrounded)
Yy = /ʎɐ/ (voiced palatal lateral approximant, fairly-open central unrounded)
Zz = /ʃɘ/ (voiceless postalveolar fricative, mid-central unrounded)

Note: There are sixteen symbols. Consonants are represented by writing the symbol in the "high" position (indicated here by upper-case), and vowels are represented by writing the symbol in the "low" position (indicated here by lower-case). Letter names are written by drawing the symbol in the high then low position.


I was thinking that the main Verb should be adjacent to the Subject, but in retrospect it might be better to insert the Auxiliary between the Subject and the Verb.

So make that: S Aux V O or O V Aux S

(Sorry about the typo in my previous post, but as you obviously realized, by S V Aux S I meant S V Aux O)

I'll probably be back with more ideas after I finish looking through a book that's been gathering dust on my bookshelf for several years. It's called "The briefest English grammar ever!", By Ruth Colman, Designed for English speakers who didn't learn grammar at school, particularly those now learning another language via a method based on grammar.

It's only 35 pages long, so it's about time I got around to reading it.


That's definitely a possibility. I'll have to look into the IPA modifiers I can use for this.

Thinking over, allowing both S/O and O/S would make it harder to understand, and any hypothetical population using it would likely stick with just one word order for consistency and simplicity. So I may as well just make S/O the official default since I've already decided it's more useful than O/S. I'm also going to drop the idea of an auxiliary, and just use prefixes and suffixes instead. This will help keep sentences less cluttered.

Which leaves me with the SVO sentence structure. Not very exotic, but at least I understand why I'm doing it this way, instead of simply using it because that's what I'm used to. I suppose I can get more creative in other ways.

I started making up pronouns, prefixes and suffixes, but realized that I'd done a poor job of picking phonemes, so I've decided to start over with assigning sounds to symbols. I've cut down the vowels to eight, with the addition of eight shortened versions of these vowels that can also be used in combination to form diphthongs. I've rearranged the letters to make it easier to remember; you can get a nice rhythm going while reciting the alphabet as: KkLl MmNn, OoPp QqRr, SsTt UuVv, WwXx, YyZz.

I've even made up a convenient chart showing symbols and pronunciation...

What is is your native dialect of English? On the vowel side, half of your example words don't work for me at all.

Australian.

But even for me the vowels aren't quite right, which is why I've put the disclaimer down the bottom that some of the examples are approximations only. I couldn't always come up with examples that were exactly right, so I put down examples that sounded similar to my ear.

Of course, I'm also assuming that the recordings I'm listening to are accurate. I'm using the Paul Meier charts (http://www.paulmeier.com/ipa/charts.html), which I saved to my computer.

Doing a bit of comparison, the vowels on that chart seem to sound slightly different to the ones on the IPAlab chart I just found (http://web.uvic.ca/ling/resources/ipa/charts/IPAlab/IPAlab.htm), although that might just be because the recordings in the first chart stretch them out longer.

Hmm... going by that second chart, the examples like moo and shoe for /ʉ/ seem to be worse than I'd thought (not that I thought it was a great match to begin with), and it's probably the same for other sounds as well. Maybe the difference in length is affecting the way I perceive the sounds? I'm going to have to re-do the pronunciation guide.

What would you recommend I do about the pronunciation guide when I can't think of any English terms that match the sound exactly? Leave out the examples that aren't quite right, change the sound to match the example? Or just use the IPA symbols with no examples?

I'd like to have some kind of description or example to explain how they sound without needing to listen to an audio recording.

I see, my problem was due to four things...

• Me. I seem to have trouble identifying isolated vowel sounds accurately.
• The recordings, which confused things further by stretching the sound out far beyond what would be normally heard in spoken words.
• My notebook computer with tiny little cheap speakers.
• Me again, for not bothering to turn off the TV while sorting out the sounds.

But here's the fixed pronunciation guide...