How to replace English? What an audacious question; English is fantastically well established. I think what we need to figure out first is how to make a constructed language succeed. As you can see from many of the comments here, that's really hard for one simple reason: most people don't believe
one can succeed, not even geeks. Nor do very many people get excited about the prospect of replacing English. I think many people don't even want
one to succeed. And when it comes to something like language, whose success is built upon network effects, that's a huge problem. (Also a problem is all the preconceived notions people have about language, like "all languages must be [highly] irregular" or "interlanguages can't succeed because they don't evolve" or "constructed languages can't be easy to learn because any popularity will cause them to instantly devolve into a mess like English" or "English is easy, so we don't need a constructed language" or "I don't like constructed languages because they are devoid of culture and soul" or "Different languages do not take different lengths of time to learn" or "Picking up a language that no one actually speaks is difficult, since it has no purpose." There is just so much BS to push against!)
For a little while I started designing a language tentatively called Lengwish, the idea of which was to be an interlanguage for the Americas, that would use English and Spanish, and French and Portuguese to a lesser extent, as vocabulary sources, with other languages used in cases when the available vocabulary from these languages doesn't work well enough to due ambiguity or other issues. (Why "for the Americas"? Simple, it's just that I've studied Spanish for five years and would like to learn French.) I planned four purposes for it to serve:
1. To teach the basic grammar and vocabulary of English or Spanish. Learning a
natural language is very hard work, especially at the beginning when you have
no hope of being fluent until years later. In contrast, you can be fluent in
Lengwish in less than a year. That means it's more fun, because you can feel
yourself making progress quickly. Since its vocabulary is derived mainly
from English and Spanish, it's a useful "stepping stone", especially for
those that don't speak any European languages, to help learn one of the
languages of the Americas. Since English is more popular than Spanish, the most
common words tend to be derived from English rather than Spanish. In rare cases,
a word is taken from other sources (Mandarin, Novial) when there is no simple
English or Spanish word for a simple concept.
Lengwish is designed to be very software-friendly, so that automated tools can
help you learn it, through underlining of errors, instant translations, and
2. To learn to learn languages. It is fairly well-known that a person can learn
a third language more quickly if they already learned a second language, even
if the second and third languages are not related to each other. Learning
Lengwish can teach you some of the skills you will use to learn other languages,
such as understanding grammar, and the ability to translate meaning rather than
3. To be a translation medium. Unlike English, Spanish or any other language with
two thousand years of complicated baggage, Lengwish is clear and relatively
unambiguous. There are thousands of English words and phrases that have
more than one meaning. Consider this: what does the English term
"free software" mean? "free" has two meanings, "free as in freedom" or "given
without a charge" ("charge" itself has several meanings, but in this case I'm
talking about a price or a fee). In some technical circles, "free software"
specifically refers to "free as in freedom": freedom to see the source code,
freedom to redistribute, freedom to modify; free software can be obtained
"for free", but freedom is just as important. "freeware", in contrast, is
software offered without a charge; it is not free-as-in-freedom, and often
includes unwanted extras like advertising or "crapware". However, a layperson
may not realize that there is a difference, and think that "free" just means
There are several reasons why languages are hard to translate, but the single
biggest reason is that so many words have multiple meanings. Because Lengwish
is clear and relatively unambiguous, it is much, much easier to translate
automatically. When completed, Lengwish will be an excellent language for
writing documents in multiple languages at once: just write your document in
Lengwish and a computer will translate it to other languages with much higher
reliability than if you had written it in English (perfection is impossible,
but we can come close). Also, translation can be done instantly offline, for
free, no need for a commercial tool or an online tool like Google Translate.
4. To be an interlanguage for international communication. This, of course, is the
purpose for which languages like Esperanto were designed, but it's extremely
hard to convince anyone to adopt an interlanguage for altruistic reasons, which
is why this goal is listed last. Most likely, Lengwish can only succeed as an
international language if it first succeeds for some other purpose. When I tell
people about Esperanto, their first question is "where is it spoken?" Since
Lengwish is so easy to learn, if any country decided to teach Lengwish to all
its schoolchildren, this question for Lengwish would quickly have an answer,
and I'm convinced that this would allow it to spread all over the world, as
people would quickly see its value for overcoming the language barriers.
However, for now it's hard to imagine politicians anywhere that would have the
courage to suggest Lengwish as even an optional course.
For now, those who find Pig Latin to be too obvious can enjoy using Lengwish as
as a "secret language"--speak it with your friends, baffle your enemies.
I haven't gotten very far with the design yet - it's pretty challenging to have an easy-to-learn phonetic spelling system that doesn't completely mangle either the spelling or the pronounciation of most of the source English words.
Anyway, because of the "network effects" problem, I think it's crucial to design a language that offers value to people even if no one else speaks the language. That's why I've been thinking about features like reliable automated translation (note that the language must be specifically designed to translate easily to a small number of specific other languages, in order to make accurate translation practical in a small open-source effort). Similarly, using English vocabulary is attractive if it means the language can be marketed as a stepping stone to learn English.
You know that there are many interlanguage designs already, right? Before designing a whole language from scratch, you should take some time to make sure that the kind of language you want hasn't already been designed, and in any case you should study what has been done before. (I've been meaning to do more of this myself!)
It's kind of weird that the link to 'Loren Chorley' is broken but anyway, I might like to join a group that is designing a language, in order to help keep the design software-friendly, to write programs to be used during the design process (e.g. a dictionary manager to help manage vocabulary and detect conflicts), and if all goes well, to write programs to help people learn the language (interactive lessons, syntax highlighter, grammar checker, etc.) You can find my email on the bottom of the front page of loyc.net.