Archive for June, 2009

LCC2 – Clint Hutchinson – Universal Semantic Markers

Tuesday, June 23rd, 2009


It’s funny.
Many of the conlangers in the audience during Clint’s talk (myself included) were immediately fascinated by the machine Clint used when he was a court reporter. In fact, if you look at the progress bar on the video, you’ll notice that Clint’s talk ends when the bar is about a quarter of the way through. Most of the video is actually his question and answer session, and most of the questions are about Machine Shorthand.

Conlangers, though, by their very nature, find language fascinating, so I suppose this shouldn’t be surprising. After all, in Machine Shorthand, we conlangers in attendance were presented with a radical representation system that most of us, I’d wager, had never encountered. There’s a fantastic moment on this video somewhere after the halfway point where Clint shows exactly how one types “stop”, and how it shows up on the paper, and after he finishes his explanation, everyone in the audience says, “OHHHHH!!!” It’s priceless.


As for the actual content of Clint’s talk,
the system he presents (and to see it, since you can’t see the whiteboard in the video, take a look at the LCC2 program) makes sense, given his background. Clint has a degree in linguistics, learned several languages with different scripts, and worked as a court reporter, and what you have in his system (this is on page 43) is a kind of classificatory that would be very useful to a court reporter who knew a bit about language and linguistics.

For a conlanger, it does, I admit, make me wonder…

What if (and bear with me if this has already been done or pondered already) one created a kind of meta language which could be translated automatically into a conlang? It might use some markers similar to what Clint has come up with, but, of course, they would have to be expanded, but if it worked, it would be quite handy for someone with four or more languages.

Here’s what I’m envisioning. Let’s say you have a sentence like “I saw a bird”. That would be translated into four of my languages below as follows:

  • Kamakawi
    • Ka mata ei i fuila.
    • /past-new.subject see 1sg.pronoun object bird/
  • Zhyler
    • Ivželer matlarum.
    • /bird-accusative see-past-1sg./
  • Kelenala
    • Ma yu ay yo.
    • /1sg.pronoun past see bird/
  • Njaama
    • Lí k’óó!á wa tekaané sá.
    • /object bird 1sg.pronoun perfect-see 3sg.pronoun/

Now, if you’re just dealing with these four languages, you don’t need a lot of information. Basically, in the meta language, you encode the lowest common denominator for each sentence. The code would look something like this:

V[“see”, past, perfect; N1(exp), N2(stim)], N1[“1sg. pronoun”, 1st person, singular, exp, subject(new), definite], N2[“bird”, 3rd person, singular, stim, patient, indefinite]

Yes, this seems like overkill, but here’s the payoff: If you enter that, the idea is it will automatically generate the correct translations in each of those four languages.

The way it work is this. Take one feature, the “subject(new)” feature. For Zhyler, all it needs to see is “exp” and “1st. person” and it’ll know how to deal with it, so it’ll see “subject(new)” and ignore it. The same with Kelenala and Njaama. Kamakawi, though, will see “subject(new)”, and know that the marker out in front will have to be ka and not ke or kae.

Obviously, it’d be much more complicated than this (for example, each language would have to have an entire set of rules just to interpret this information, and then to get the words in the right order), but if it worked…man! You could write a text in this semantic meta language and it could be instantly translated into dozens of one’s own conlangs!


As a final thought,
I love the way the LCC casts such a wide net in terms of presenters and attendees. We get all different types of conlangers, and others who aren’t conlangers but are creative, interesting, and interested people. The result is a kind of bubbling cauldron of linguistic creativity that gets to bubble up and boil over for a weekend. It’s both inspiring and a lot of fun.

This video is part of the 2nd Language Creation Conference, held at UC Berkeley on July 7-8, 2007, and hosted by Language Creation Society.

We would like to add closed captioning / subtitles to all the videos from LCC2, including this one. If you are willing to help, install Subtitle Workshop, and email your transcribed .sub file to conference@conlang.org. In return, you’ll get credit and a free copy of the DVD with this video.