The Private Language Thesis and the types of languages

Day eleven --

Question -- from "Trailsend" -- Mmm...well, I would respond, but...it seems you did not read my prior post, or if you did, you opted not to address any of the points I raised there. I believe most of them are still valid arguments for why your criterion i, ii, and iii cannot be met universally.

Nnnope. I'm very sure I meant that your idea was theoretically impossible.

Answer -- I find your previous post on this issue, as below.

Quote from "Trailsend" -- I would suggest that a "universal language" as you have conceived it is indeed impossible, save that it becomes a redundant middle-man tongue that complicates cross-language understanding more than anything.

In that way, PB is, as others have said, somewhat redundant. It is just another language which must be "learned" like any other, and therefore would not serve the purpose of a universal language. For example, learning PB would not enable me to quickly learn Hindi, because Hindi will conceptualize ideas in a much different way than PB does, and I would essentially be starting over.

I also agree with the concerns others have mentioned about your proposed root system. It seems very redundant in places and very limited in scope. Granted, you have admitted that your root system may not be perfect, and you seem to be more interested in the theoretical notion of some small root set which could be used to derive all other concepts. (As mentioned above, I think this notion is flawed--even if you found a single root set which could represent all the concepts of one language, those representations will not necessarily be intelligible by speakers of different languages with different mindsets.) But for the sake of science, here's a test for the current root system. --End quote.

In a previous post, I did discuss an issue about the isomorphism between languages, which was brought up by you. And, you kind of agreed that if all languages are isomorphic to one another, then a true universal language becomes possible.

After the introduction of the "Private Language Thesis (PLT)" and the "Invention Theorem", the isomorphic issue can be dealt a lot more precise than before. The whole issue is really about "what is a language?" I discussed this issue at two places. 1. The main page of the PreBabel (  &nbsp http://www.prebabel.info &nbsp; ), trying to list the necessary and the sufficient conditions for a language, with the example of English. 2. A June 2007 paper at (   &nbsp http://www.chinese-word-roots.org/cwr018.htm  &nbsp; ), discussed the necessary and the sufficient conditions for a language on the metaphysical and on the ontological levels.

With PLT and the Invention Theorem, we now can address this issue in a much better details. 1. Long before the coming of human existence, our universe moved and evolved for a long time, billions years. 2. During the past a few thousand years, men began to notice or to hear some of those primordial moves and evolutions by either discoveries or by inventions, 1. Discovery -- reached with induction (observations of phenomena) 2. Invention -- reached with deduction then, they wrote them down with one of man's invention, the languages. 3. What are those primordial moves and evolutions? What should they be called? A language or a "what not"? 4. Even if we deny that they are a language, and call them as a "what not", yet they are something which can be described with the human languages. What is the difference between description and translation? If we give that a name, the "What not language," what is the grave consequence in linguistics? Can the following postulates make sense? Postulate A: Every "what not" which can be described by a human language carries a "what not language".

Postulate B: All natural laws written by human (either via discovery or invention) are just translations of the "primordial 'what not' language". In fact, many of such a translation were not very good. Newton's laws were not a good translation of that "what not", and the situation became better after Einstein re-translated it by adding the Relativity. In fact, for the "gravity what not", no exact translation is found thus far yet.

With the postulate A and B, we have discovered a new kind of language, the "Primordial what not language (PWNL)," which itself is wordless, soundless,..., and many more -less(es). Yet, they can be translated into human languages as soon as we can hear them with some means.

Seemingly, this PWNL has no intention to communicate to or with humans but governs the entire universe with its wordless, soundless "what not language." At the human level, this PWNL is not a communicative language. Communication is not its purpose while the communication plays a major role in all human languages. If this is true, we then have two types of languages. 1. Communicative language 2. Non-communicative language Then, the issue of "what is a language?" must be re-evaluated. This will be "the" central issue for linguistics. We should get back to this issue later. Yet, even without an answer for this issue, we are now able to re-visit the isomorphic issue of human languages.

Not only are the differences between natural languages great on their word form and on their grammar, but the differences between their contents are even greater. With so big content differences among languages, the isomorphic issue of comparing any two languages with one-to-one translation mapping could become less convincing. Yet, now we can select some "same" contents for any two languages and make one-to-one translation comparison. Let us fix the content to be the four parts of the PWNL (physics, chemistry, mathematics and biology). We only compare the part (the verbal part, not math language) of the language which is used to describe these four parts by any two natural languages. Can an one-to-one translation mapping be made between these subsets of these two natural languages? This is not a theoretical issue but a testable one. As far as for the world's five major languages (English, Russian, ..., etc.), they all passed this test. Thus, even if those languages were not isomorphic among one another as a whole, there are subsets of them isomorphic to one another. That is, we can, at least, build a universal language for those subsets.

Signature -- PreBabel is the true universal language, it is available at http://www.prebabel.info