The best test is the one that is inexpensive and conclusive. A priori languages are the easiest to test because it is easy to find someone who does not know them. Euroclones can be tested by Europeans if we use a computer to scramble the vocabulary. Grammar is a small part of a good auxlang, so it does not matter if the testers are familiar with the auxlang's grammar. The test must be written in such a way that a computer program can easily scramble its auxlang vocabulary.

The vast majority of people who use auxlangs do not have the time to learn vast vocabulary. A perfect auxlang is the one that makes it possible to express any idea with a small vocabulary. For example, if you do not know the word "hospital," you improvise a new compound word: "medical building." A realistic auxlang test mimics this improvisation. The most popular auxlang, the Special English used by the Voice of America has a vocabulary of 1500 words.

Here is an example of inexpensive and conclusive test of Esperanto:

  1. Make a primer of Esperanto that includes its grammar and a vocabulary of about 400 words. (The optimum vocabulary size is the one that results in the greatest score differences between the auxlangs and yet can convey much more than baby talk.)

  2. Pick up two persons of average intelligence (IQ = 100) who speak the same language, for example Mandarin, but do not speak any european languages or euroclones. They are called testers.

  3. Write a test consisting of 10 Esperanto sentences. The sentences must be much more complex than baby talk. Each sentence includes one word that is not defined in the vocabulary. The undefined words are rather common, and they are not proper names or names of flora or fauna.

  4. Record the 10 Esperanto sentences on a tape recorder or a similar device.

  5. One of the testers, called passive tester, has four hours to read the primer, to listen to the Esperanto recording, to translate the recording into Mandarin, and to write it down.

  6. The other tester, called active tester, has four hours to read the primer, to read the Mandarin text, to translate it into Esperanto and to write it down.

  7. The test is graded. Its score equals total number of glyphs written by the active tester divided by the total number of sentences that were translated without errors. (The test favors terse auxlangs.)

If the testing discredits popular auxlangs, their aficionados will discredit the testing with the ferocity of religious fundamentalists, for they believe that languages are not merely tools, but, like mother tongues, they are religions that must be defended.

Andrew Nowicki