Archive for November, 2009

Drake alphabet.

Sunday, November 29th, 2009

All right, so I’ve managed to get the issues with the Drake alphabet I mentioned in my last post sorted out.  I was actually pretty much right for the most part, even in the confusing areas.

Here is a table of the letters involved, along with their corresponding letters in Syriac, which looks like it has the most resemblance to Drake out of the several Semitic scripts I looked at.  I still don’t know for certain what the official immediate precursor to the Drake letters was—I’m certain I’ll have a page or two of all the letters evolving somewhere in my notebooks—but the Syriac was actually a great help on this; comparing it to Drake gave me less doubts than comparing Drake to the Semitic tree in general… except that I realized that |z| and |Z| didn’t have anything at all to correspond to.



















gimel with dot above







dalet with dot above


[h], 0









waw with dot above














[j], [iː]





















nun with dot above





not attested











not attested













shin with dot above









This table as a PDF (with appropriate fonts)

So |z| and its dotted form |Z| are certainly borrowed letters; they are either the Greek zeta or the Kirumb zéta, which is technically the same letter.  (I lean towards the latter, though it may be unlikely.)

As far as the phonetic values go, I found some old sound change files and confirmed that |š| and |Š| really were [ç] and [ʃ] respectively—the former descends from a Proto-Drake *ɬ  and the latter retains its original *ʃ value.

I think |Z| is most likely [ʝ], not [Ê’] as David Salo suggested, partly because [ʝ] would fill the blank in the [tç] [ç] [dʝ][—] grid, partly because [ʃ] is already an outlier and apparently lacks a voiced form in the protolanguage as well.  Furthermore it’s only attested in one word so far, ?[çemeʝaː] “six”, where the root [ezaː] “four” usually has [z]; however in this case the [z] is following a [ç], suggesting some kind of undocumented consonant harmony may sometimes be in place—if, again, the |Z| in the word is not an error.  (It’s possible that both [z] and [ʝ] are valid values.)

New Newspaper Article

Wednesday, November 25th, 2009

A new article on the Na’vi language of James Cameron’s Avatar has been posted in TCL’s Newspaper articles section.

Another article of note has also been found online, although this isn’t a conlang so it won’t be added to TCL. This one has to do with Cockney rhyming slang on bank machines. The article, from the Times Online, has a great image of the machine’s screen with choices like “Sausage and Mash with Receipt”.


26C3 Conlanging 101 preview – Dec 10 7-8pm PST

Monday, November 23rd, 2009
I'm going to be doing a preview / test run of my Conlanging 101 talk for 26C3 at Noisebridge, Thursday Dec 10 ~7-8pm PST.

I'll try to make sure it's simulcast @ or . I'll also have some mechanism to comment live in a way that I'll see - either irc on freenode #lcs or skype - so remote people will get to participate.

So please come by or get online at the time. Any suggestions for improving it will be appreciated.

26C3 Conlanging 101 preview – Dec 10 7-8pm PST

Monday, November 23rd, 2009
I'm going to be doing a preview / test run of my Conlanging 101 talk for 26C3 at Noisebridge, Thursday Dec 10 ~7-8pm PST.

I'll try to make sure it's simulcast @ or . I'll also have some mechanism to comment live in a way that I'll see - either irc on freenode #lcs or skype - so remote people will get to participate.

So please come by or get online at the time. Any suggestions for improving it will be appreciated.

An Article and a Link

Saturday, November 21st, 2009

Two new resources have been added to TCL:

The first is an article from the University of Minnesota’s Minnesota Daily newspaper posted in Newspaper Articles. It’s about the local Minnesota company that wrote the app for the Klingon Dictionary (but also include information about d’Armond Speers’ 3-year “experiment” in speaking Klingon to his son).

The other resource, posted in the Linguistics Online Resources page, is a handy online tool for converting IPA characters into HTML: θæŋk ju


Registration closing for 2009

Tuesday, November 17th, 2009

Registration for new participants for 2009 will close Wednesday, November 18th.  At that time I will compile the mailing addresses and email everyone with the addresses to send the cards to.  If you want to change your preferences but are not able to, please email  the exlead account at for help.

Have fun!  Remember, we are exchanging anything in the realm of: holiday cards, concultural holiday cards, postcards, concultural postcards, or any other cardlike concultural artifact in your conlang!

Drake alphabet puzzle.

Monday, November 16th, 2009

One of the hazards of working with a language that you don’t work with often is that you tend to forget things.  In many cases, this is generally not a problem so long as you have time to look things up, but with one of your own constructed languages it may not be so easy.   I was doing some work on Drake for the first time in a while and came across a font file I made for it, several years ago.  I had no information about the alphabet outside of the font and one small Shoebox dictionary that used it.

On the bright side, I was able to figure out the phonetic values of almost all the characters from the dictionary file.   The two I am not entirely sure of as yet are the character mapped to “Å¡” and the character mapped to “Å ”.  Also, I have one character mapped to “Z”, and it’s the same character as “z” but with a diacritic; however, in the original dictionary it has the same value as “z’”—and only appears in a word from whose root all other derivations have ‘z’.

But the phonetic values of the letters aren’t important at the moment.  I was working on putting this up on FrathWiki so I won’t have to hunt for where the answers are in the future, and realized I had no idea what the alphabetical order is.

Here I’ll pause to show you the script I’m talking about.


The first row is the script, the second row the “transliteration” — or at least the character the letter is mapped to — and the third is the recovered phonetic value.  Yes, it’s technically an abjad.

I probably won’t be able to discover alphabetical order until I can figure out the source of each letter.  I know the alphabet as a whole comes from the same family of old Semitic scripts as Phoenician and Greek did;  letters with diacritics come immediately after the base letter in the alphabetical order, but outside of that, the order should be the standard aleph-beth-gimel.

Anyway, some of the letters are easy to recognize, whether from their shape, or because the phonetic value is unambiguous:  ’ is aleph, l is lamed, m is mem, n is nun, etc.  Some others, less so.  (I’m not sure how I mapped the ‘emphatic’ and non-emphatic consonants to Drake characters. And where is ayin?)  Anyone want to take a stab at identifying these letters?

James A. Garfield, Part II

Sunday, November 15th, 2009

Yes, that James A. Garfield. A while ago, The Conlanging Librarian posted some information on James A. Garfield. Well, something that many people don’t know about the man who would become the 20th President of the United States was that the courtship with his future wife, Lucretia “Crete” Rudolph, began (in a series of letters) as a back-and-forth discussion on the value of studying ancient languages and why the “diversity of languages was given”. You’ll now find relevant quotations from these letters, dating from January 1854, posted at The Conlanger’s Library in Quotations under Language and Specific Natural Languages. Enjoy!

I’m going to 26c3!

Sunday, November 15th, 2009
So, I submitted a talk to 26C3 on "Conlanging 101". And it actually got accepted (with flight paid!).

Which means I get to go to Berlin and attend an awesome hacker conference, and all I have to do is give a talk I'll enjoy. W00tage.

The talk is a) an intro / overview to conlanging (like an expanded version of my lightning talk at Toorcamp - plus b) conlanging-by-crowd-committee where I'll (albeit very quickly) go through actually making a new language on the spot good enough to translate part of the Babel text (like a super-compressed version of my Berkeley DE-Cal class). Plus workshop afterwards.

Most likely what I'm going to do is spend ~10 minutes giving a fairly fast-paced overview, ~35-40 going through making a language on the spot, and anything left over on Q&A.

a) got ideas for how to improve the talk / workshop? (German speakers: anything I should be aware of, or any good jokes to make? Ich spreche keine Deutsch. :()
b) anyone else going?
c) anyone in the area (even vaguely, as in Western Europe; I'm somewhat tempted to make it an excuse to visit other places in Europe) who's either interested having a visitor or willing to host me for a couple days? (if yes, please email me directly)

Specifier clarification, possession, and other changes

Wednesday, November 11th, 2009
The world of Koa has been fairly quiet for ages now, but the last few weeks have brought some important progress (starting on our honeymoon, on Sant'Erasmo). Here goes:

1) For the last time, w and y are not fated to be Koa phonemes. Just looking at the new words that contain them in my lexicon after a year and a half to sober up made this clear: sewe, yomo, wohi, yuhu, wene, uye, maya, yume, mawa, paye, weo, yaye and yoki are so obviously not Koa words. Somehow I have to remember this and not go through any more angst on this issue.

2) The genitive phrase, including for pronominal possession, is now formally always ka X Y, where Y is the possessor. There currently remains a little bit of wiggle room for phrases like ni mama, etc., but the jury's still out. In any event, responding to a question I asked several years ago, pronoun objects will certainly not be acceptable preverbally: *ni se loha, etc. It's not so bad, really -- ka talo ni has a certain elegance to it.

3) After much consideration, during which for a brief period I was leaning heavily towards a Hungarian-style "my X exists" structure, I've come to the conclusion that verbal possession needs to be via prepositional phrase, using either me or la (final decision pending). So, for example:

this cat is mine = ti neko i me ni
this is my cat = tika i ka neko ni (note: why not tika i neko ni? let's come back to this)
I don't have a cat = na neko i me ni (or hu neko i na me ni, logically? yuck)
do you have a cat? = ei a/hu neko i me se?

There are some unresolved questions about usage here, but in general I think this will work well. In the main, my concerns are about the negative/interrogative, and pertain to the existential construct in general -- I keep wanting to say something like na tai neko me ni, which is clearly in violation of everything everywhere. I think what's happening is the collision of the logical design of the language with human language intuition; hopefully they won't end up being too difficult to reconcile.

4) Speaking of which, I think I've finally got the specifier system figured out. A lot of it had to do with realizing that the hu/po predicate logic design isn't necessarily all that relevant to human linguistic needs; what I've done is to give that meaning to these particles in conjuction with an article, but to give them a more pragmatic/specifier-type when immediately preposed to a noun. Here, then, are all our current specifiers.

ka neko = the cat(s), refers to something already on the discourse stage
ti neko = a deictic subset of cats, either of those already on stage or a new set being raised to the stage
a neko = a/some cat(s), refers to specific animals someone has in mind which are not yet onstage but are being raised
hu neko = a/some cat(s), nonspecific referents, with no intent to raise to the stage
po neko = cats in general
ko neko = the abstract idea of being a cat: cathood, felinity, etc.
na neko = no cat
ke neko = what/which cat
ni neko = my cat (optional shortening of ka neko ni)

Many of these can be be increased in specificity by adding a definite or indefinite article, as follows:

tika neko = (how does this differ from ti neko?)
huka neko = one/some of the cats already onstage/in the given set
hua neko = a/some cats, out of all the cats that exist
huti neko = one/some of the cats in the deictically indicated set
poka neko = all of the cats onstage/in the given set
poa neko = every cat, period
poti neko = all of the cats in the deictically indicated set
naka neko = none of the cats onstage/in the given set
keka neko = which of the cats onstage/in the given set? which of these cats?
nika neko = this set of my cats, these particular cats of mine

In the same vein, all of these (and a few more with -a) also stand alone as pronouns:

tika = this one, these ones
tia = this [stuff, idea], all this
huka = one/some of them, someone
hua = something
huti = one/some of these
poka = all of them, everyone
poa = everything
poti = all of these
naka = none of them, no one (nahuka with same meaning)
naa = nothing (nahua preferred?)
keka = which one?
kea = what?
nika = mine, this one of mine

5) There's an unanswered question about ditransitive clauses. Where both the accusative and dative arguments are full nouns, they form what is formally a genitive phrase if listed in dative-accusative order with no preposition:

ni si ana ka mama a neko
1SG=PERF=give DEF=mother INDEF=cat
"I gave my mother a cat" or "I gave the mother of a certain cat [to someone]" opposed to unambiguous

ni si ana a neko la ka mama
1SG=PERF=give INDEF=cat DAT=DEF=mother
"I gave my mother a cat"

For this reason I was about to say that the dative-accusative order is only acceptable when the dative argument is pronominal, but then the ambiguity above isn't really pragmatically worrisome. For the moment I'm going to assume that context and animacy will nearly always resolve this, and leave both options open.

6) Negative clauses: how are they done? To answer an old question, for the moment I'd like to leave all the possibilities available. They all make sense, and the "double negative," while perhaps logically confusing, would never be pragmatically so. Therefore, "I didn't see anyone" could be translated in any of these ways:

ni si nae na(hu)ka
ni na si nae huka
ni na si nae na(hu)ka

7) A small but important matter: what is the word for "cat?" When I was in Europe I was feeling pleased with my choice of neko in homage to Japanese, but checking my dictionary a few weeks later I remembered that I already had a word for "cat," sene. I feel torn in my allegiance; we'll need to come back to this one.