Conlanging: Codon, Part Two.

Part one (introduction and how to structure the vocabulary) can be found here.

Part three (a continuation of vocab building) can be found here.

This week, we’ll generate some common words and decide on what concepts those words will represent.  A disclaimer: I love language but I’m no linguist.  If something in here is incorrect, feel free to let me know in the comments.


A phoneme is the smallest part of a language: a unit of sound.  Usually there are countless phonemes in a language: just try going back to the beginning of this sentence and figuring out how many unique sounds show up.  Codon only has four phonemes, two vowels and two consonants, representing the four amino acids of RNA: Adenine, Guanine, Cytocine, and Uracil.  Therefore, our phonemes will be long A, G, C (pronounced “sss”), and U (pronounced “ooo”).

In RNA, these building blocks are paired together, and we’ll do the same with Codon. Unfortunately, they don’t go together in neat little vowel-consonant packets, so we can’t follow that perfectly.  Instead we’ll pair A with G and C with U.  This turns our phonemes into syllables:

AG (“agg”)
GA (“gah”)
CU (“soo”)
UC (“oos”)

And from there, we can form words.   We now need a list of all sixty-four combinations of letters, which will be the entire vocabulary of Codon. You can either write it out manually or do a simple bit of computer code (Python for me, thank you Reddit user from ages ago) to get this:


…etc.  The full list will be below, as we go through each section of vocabulary.  When we’re done, I’ll upload a full dictionary and grammar book.

Nouns (15 words)

We’ll tackle the largest section first, and the most challenging.  We are now faced with the task of distilling every noun in the universe– every person, place, or thing that ever existed– into 15 concepts.

When we get to grammar, we’ll talk about combining these concepts to create more specific vowels.  But for now, we want to cover all our bases as best we can.  The list isn’t going to be perfect, because I’m doing this armed with a Bachelor’s in English and Google, but it should at least be usable.

According to Wikipedia, the 15 most-used nouns in the English language are time, person, year, way, day, thing, man, world, life, hand, part, child, eye, woman, and place.  Many of these nouns are related: for example: “man”, “child”, and “woman” can all be modified forms of the word “person”.  For that matter, so can “eye” and “hand”: they are part of a person. To further dilute the word “person”, who are we to say what constitutes a “person” or not?  “Life” seems like it could be modified to make “person” So six words have now been condensed into our first:

AGAGAG: life form / life

“Year” and “day” can also be modified forms of “time”, which can be our second:

AGAGGA: time

And nearly anything can be a modified version of the word “thing”, so that makes the list.

AGAGCU: thing

Remaining: world, part, place.  These are all good words to have, although I feel I should modify the concept of “world” to be even broader: rather than using “world” in the sense of “planet”, it will mean “world” in the sense of “sphere of being, awareness, or consciousness”.

AGAGUC: sphere of awareness / world
AGGAAG: part
AGGAGA: place

This leaves us with 9 words left.  The next ten words on the list are: work, week, case, point, government, company, number, group, problem, fact.  “Week” is already a no-go: it’s a part of “time”.  “Work” in the sense of physics makes sense: the act of exerting a force, mental or physical.  But it might make more sense as a verb than a noun.  “Case” is vague: Merriam-Webster defines it as “a set of circumstances or conditions”.  That’s a good one, but we’ll make it vaguer: a set of things can be narrowed down with grammar.

AGGACU: a set

“Point” is very similar to “place”; “government” and “company” can be modified from “life form”.  Which leaves us with “number”.  Ah.  We have no room for a number system.  For the sake of space, we’ll decide Codon uses binary, leaving us with the need for only 10 numbers:

AGCUAG: zero

And now we have “group”, “problem”, and “fact”.  “Group” can be “multitude”, and “problem” seems like it could be distilled from “fact” and maybe a version of “wrong”.  “Wrong” in itself can be a negated form of “good”, and I’d like to keep this language positive, so:

AGCUGA: multitude
AGCUCU: fact
AGCUUC: good

We have now run out of Wikipedia.  With three words left, I’ll add in some of my own:

AGUCAG: electromagnetic spectrum
AGUCGA: intelligence
AGUCCU: shape

These words seem incredibly vague.  They are incredibly vague.  They get less vague when you allow for compound words, which is how we’re gonna cheat: “planet” can be made from place-world (aggauc aggaga), or “news” from multitude-fact (agcuga agcucu).  Heck, we can even get Hitchhiker’s Guide to the Galaxy on this language and name a Hooloovoo: multitude-intelligence-color-life (agcuga agusga agusag agagag).  If we wanted to get really specific we could use binary to give the frequency of blue on the electromagnetic spectrum, but I don’t particularly want to type out all those aggaus-es and agcuag-s.

As I mentioned back in the first Codon post, this is a completely impractical language. It’s just so much fun!

Verbs (9 words)

Nouns do not a language make.  Our Hooloovoo wants to be able to go out and do something.  Fortunately, the list I linked above has the top 25 nouns in the English language too. The first couple won’t be surprising:

AGUCUC: to be
GAAGAG: to have
GAAGGA: to do

After that: say, get, make, go, know, take, see, come.    “Go” and “come” are clearly linked, as are “take” and “get”. And most of these can be watered down to “to do”: “to make” can be “to do matter”, “to say” can be “to do an electromagnetic spectrum (insert wavelengths of sound here)”, as can “to see”. “To know” can be “to have intelligence”. So:

GAAGCU: to move
GAAGUC: to transfer

Next: think, look, want, give, use, find, tell, ask, work, seem, feel, try, leave, call.  “To think” can be “to do intelligence” and “use” can be “do”, “look” is the same as “see”, “give” falls under “transfer” (as does “find”).  “Tell” and “ask” can also be versions of “say”.  “Try” goes under “do”, “leave” goes under “move”, “call” goes under “say…”

So: work.  We’ve talked about this one.

GAGAAG: to expend energy with the intent of change

Leaving: want, seem, feel.

GAGAGA: to desire
GAGACU: to appear to be as something else
GAGAUC: to express an emotion

There.  That rounds out our verbs.  I’m a big fan of “to desire”: I love that “want” is the same sound a baby makes.

Prepositions (8 words)

A preposition is a “locator in time and space“.  In English, the most common prepositions are, according to our faithful Wikipedia link: to, of, in, for, on, with, at, by, from, up, about, into, over, after, beneath, under, above.

Step one, combine!

“To” is a good generic directional, and the negated form of “to” can be “from”. “Of”, “in”, and “for” are pretty decent on their own.  “On” can be merged with “above”.  “With” can be merged with “at”, as can “by” or “beside”.  “Under” and “beneath” are the same, and can be expressed by negated forms of “over”/”above”.  The same goes for “up”.  “About” as in “the content of information” can be kept as-is.  “After” is a negated “before”.

Let’s see what that leaves us with…

GAUCGA: above
GAUCCU: about
GAUCUC: before

I guess that works.

Words, words, words

The shape of the language is starting to come together!  Now that we’ve got the hang of this, we’ll deal with the more obscure and smaller parts of speech next week: determiners, adjectives and adverbs, pronouns, etc.  (Originally this was going to be all one post, but it got a bit long.)

Conceptual languages like these are amazingly fun.  I’ll leave you with a challenge: as a thought exercise, think up a language concept.  It could be anything from Morse code written in crochet to a fleshed-out version of Flatula (a language for which I could not find the relevant clip on Youtube).  What’s your crazy language idea?


4 thoughts on “Conlanging: Codon, Part Two.

  1. Pingback: Conlanging: Codon, Part One. | The Evening Ramble
  2. Pingback: Conlanging: Codon, Part Three. | The Evening Ramble
  3. Pingback: Conlanging: Codon, Part Four. | The Evening Ramble
  4. Pingback: Conlang Lite: The Language Facade | The Evening Ramble

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s