Cryptopian - Inventing Languages Just For Fun

Gary J. Shannon
Created Dec. 13, 2004
Last Updated May 3, 2019

What This Site is About

This is a website about homemade languages. A little knowledge of liguistics would be helpful, but realistically, you probably know more about linguistics than I do.

Quoting from my post to the Conlang Mailing List on 17 DEC, 2008:

My first conlang was around 1952, as best I recall, and was inspired by Pig Latin. I called it Pig Russian (Igpaya Ussianruski). Rather than the one transform rule of Pig Latin, it had 26 different rules depending on the first letter of the original word. Being a trivial relex, it had no grammar of it's own.

I discovered Esperanto around 1958 or 1959, but I didn't care for it. A few years later, in high school, I took Latin for a year and German for two years. That gave me exposure to non-English grammars and made my several unfinished conlangs less like a simple English relex.

In college, majoring in computer science, I took more German, a semester of Russian, and two semesters of ASL sign language. While taking sign language I devised simple system of "pictographs" representing each sign as I learned it. Over the course of those two semesters my pictographic writing system became so complete, and I became so fluent in it that I took class notes easily in the conlang.

Many years, and three children later, all my conlang notes were lost in a house fire, including around 2,000 file cards with my pictographic dictionary. From time to time I still go back to trying to reconstruct my lost pictographic conlang.

Some Constructed Language Projects

Tazhu - The Tazhu Corpus Project (Last Update: Jan. 9, 2005)

To quote from my journal regarding the creation of Tazhu from the text of the 1879 edition of McGuffey's Reader:

Dec. 23, 2004 PM. I used to keep some sort of "things to do" list for my conlang projects, but now the only item on my "to do" list is translate the next sentence in the reader. If I keep doing that the rest of the design process takes care of itself.

Soaloa - A Language with a Peculiar Grammar (Last Update: Feb. 10, 2010)

This constructed language was first created in 2004, and I've fiddled with it from time to time, but it's still a very long way from being "complete". (Whatever that might mean for a language.)

Pop - A Language with a Peculiar Grammar (Last Update: Jan 18, 2014)

This constructed language was an offshoot of Soaloa (mentioned above), and is a much more practical implementation of the basic concept.

Txtana, A 30-Day Constructed Language (Last Update Nov. 21, 2010)

This page is where I kept a record of my attempt to construct an entire new language in 30 days. The first 22 days are at the link above. Day 22 and on is at This Link.

Ai Basata (Last Update May 4, 2009)

Yet another constructed language. This was an attempt to build a really systematic grammar.

The Alice Project (Last Update Apr. 21, 2019)

Here is another corpus-based conlang project, similar in most respects to the old Tazhu corpus project. The idea is to take an English text, in this case, Lewis Carroll's Alices' Adventures in Wonderland and to translate it into a new created language by coining each new word and each new gramatical feature as needed. Just as in the old Tazhu corpus project, "...the only item on my 'to do' list is translate the next sentence..." in the corpus. "If I keep doing that the rest of the design process takes care of itself."

Some Conlanging Resources

An Illustrated McGuffey's First Reader (Last Update: Apr. 3, 2010)

This is always been one of my faovrite go-to translation projects for a new conlang.

Some Conlanging Resources

86,580 Computer Generated Words (Last Update: May 3, 2019)

Here are over 86 thousand words generated by computer. The morphology is (C)V(C) or (C)V(n)(C)V(C), where (C) is an optional consonant, V is a vowel, and (n) is the actual letter "n", optional. The consonants are drawn from the set {D, F, K, L, M, N, P, S, T, SH, CH} and the vowels are from the set {A, E, I, O , U}.

This is a flat text file with the words listed in a single column.

A Computer Generated Lexicon (Last Update: May 3, 2019)

Using the words from the above computer generated list, and the top 20,000 words from the Brown Corpus I paired them up with the shortest computer words being paired with the most frequent corpus words. The reulst is a more or less complete dictionary that could be used to relexify English text.

A Specialized Computer Generated Lexicon (Last Update: May 3, 2019)

This is the same sort of computer generated lexicon as the above Brown Corpus lexicon, exceot that it uses only the words actually found in Alice in Wonderland. That makes this a specialized lexicon especially well-suited to relexification of that one book.