Longing for a universal language is a dream of mankind since antiquity, such as the Biblical story of Babel. In the human history, many languages (such as, Greek, Latin, Arabic or English) claimed to be a universal language with the political or economical supremacy for a short period of time, especially in the area that its political power could reach. Nonetheless, a few languages do act as trans-national and trans-racial literary language, such as the Chinese written language in China, in Vietnam, Korea and Japan for centuries. However, there are, at least, two difficulties for any natural language to become a true universal language.

1. No natural language is easy. Less than 10% of people can truly master their mother language to a scholastic level. In general, the difficulty of learning another natural language as a second language is about 10 times harder than learning the mother language. Thus, even if we all accepted politically that one particular natural language (such as, English) is the lingua franca, the illiteracy rate for this language would have still been higher than 85% worldwide.
2. Just as all the de facto world languages owe their status to historical political supremacy, the suggestion of a given natural language as a universal language has strong political implications, and the major world powers will never be agreeing such an agreement. Thus, the best hope for a universal language, if ever possible, is by choosing an insignificant language or a constructed one, such as Esperanto.

With these realities, a universal language, if any, must be:

• as a second language for all people, and
• as a constructed language.

1. Criterion one (C1): Its scope and capacity must be in par, at least, with one natural language.
2. Criterion two (C2): It must be mastered to a literacy level similar to the language skill of a 12th grader on his/her mother language by an average person in 100 days with 3 hours of study a day, that is, a total of 300 hours of study.

The verification of C2 is quite simple in principle. As soon as its construction is completed, a few volunteers can either confirm or disprove it. The major issue is of how to construct it.

The biggest difficulty of a language is the vocabulary, as the foundation of the vocabulary of the most of natural languages is practically arbitrary. Why are the four letters (L, O, V, E) that means love? There is no way of any kind that we can find out the meaning of "love" by dissection or decoding of those four letters. They mean "love" because of "You told me so!". Otherwise, the string "love" is just a blob. Thus, learning a language must learn thousands or even half a million of those blobs together with their "You told me so!", especially for someone who learns them as a second language without the benefit of being already able to speak those blobs. Of course, a pure (100%) root words system with all vocabulary that are composed of only from those root words (no exception), which is also self-revealing of those word meanings, can eliminate the above-stated vocabulary difficulty. Yet, this root word system idea is still having, at least, two difficulties:

1. Can such a root-word vocabulary system be constructed? How to select those root words? How many roots must the system have? If the number of roots goes over one thousand, the benefit of a root word system will be significantly reduced.
2. A language is much more than just vocabulary. A language becomes more difficult to learn when the learner must learn to make distinctions that he is not used to making. For a non-English speaker, it could be quite difficult for him if the grammar of this universal language (u-language) contains accusatives, mandatory tenses, tones, noun/adjective agreement, etc.. In this sense, the grammar of this u-language must encompass (or, not significantly different from) all grammars of different natural languages. Yet, can this be achieved?

If we cannot resolve these two difficulties, we probably can never pass the Criterion two (C2) with any constructed system. Yet, what is the guiding light for resolving these issues? Fortunately, we do know a fact. Norwegian is easy to a Swede because it is practically a mere dialect of his own language, while Norwegian is not easy in itself as it would be very difficult to an Oriental. The dialects of Chinese are mutually unintelligible, which would take even a gifted European at least three years to learn to speak one of them while it takes a Chinese person only about six months to learn another dialect. Thus, with this fact, if all natural languages are dialects of this u-language, then it can be learned in 300 hours of study by all different people who speak different mother languages. Of course, this is a big "IF." However, we can re-state the Criterion 2 as below:

RC 2: If language A is a u-language, then all natural languages must be dialects of this language A.

Postulate 1: Language A is a known natural language. Language B (either natural or constructed) is a dialect of Language A. For a person whose mother language is language A, he can master language B within three months to a level similar to a 12th grader's language ability of his/her mother language.

If all natural languages must be dialects of this u-language, it must be the mother language of all those natural languages, that is, they are all grown out from the mother. Thus, in every baby language, it must consists of two parts, the part that is inherited from the mother and the part of some new growth (the bells and the whistles). Then, the task of constructing a u-language becomes a task of searching for the mother language of all natural languages.

Seemingly, the comparative linguistics could be of a great help on this task. However, the major interest of comparative linguistics is on the genetic relationship between languages that are members of the same language family, with the emphasis on phonological and the lexicon. Thus, there is not much to compare about between Arabic and Chinese on their lexicon and their phonology. Thus, the current study of comparative linguistics is of no use for our task of finding a mother language for Arabic, Chinese and English, if such a mother, indeed, exists. That is, we must invent a new methodology for this seemingly impossible task, and the best way of tackling this issue is the reverse-engineering.

If such a u-language (as the mother of all natural languages) does exist, it should be in every its baby language genetically, and we should be able to find its genetic codes from any one of its baby language, without doing any comparison between languages. If such a technique can be developed, I will call it "Begetting the mother from her baby" (or BMFB in short), and I am making the following proposal:

1. The attributes of a natural language (such as, English) are listed as Ar(1), Ar(2), ..., Ar(n).
2. If Ar(m) can be substituted with a different mechanism U(m) without any change to the system, U(m) will be put into a bag called "Mother bag" and Ar(m) will be placed into a bag called "Baby bag."
3. If an Ar(x) cannot be substituted in any way, it will be placed into both bags.
4. After we replaced all Ar(n) with U(n), if possible, we filled up two bags, the mother bag and the baby bag.

Listing some major attributes of English language might not be a terribly difficult job. Yet, listing all necessary attributes of English exhaustively might not be an easy thing to do. After all, what are the necessary attributes of a language? Without knowing the answer of this question, we are as a blind man riding on a blind horse. Fortunately, there are a few toy languages (the formalized languages) which do constitute as language while their scopes are small enough for us to investigate their structure and all their necessary attributes in their entirety.

The smallest toy language (formal system I) has only four symbols (an identity symbol =, and three individual constants, a1, a2, and a3). Although this System I is a genuine language system, it is too small of a system to convince the general public that it is, indeed, a language system.

1. An identity symbol, =
2. Five connective symbols (logical constants), {no (negation), or (disjunction), and (conjunction), if...then (conditional), if and only if (biconditional)}
3. Two parenthesis symbols, ( , )
4. Two quantifier symbols, { for some, for all}
5. Infinite number of individual symbols, which again are subdivided into two groups:
   • v1, v2, v3,..., as individual variables,
   • c1, c2, c3, ..., as individual constants.

• related to other symbols,
• related to things that is referring, denoting or connoting,
• related to the using, application of the things named by the symbols.

1. "term" of T (language T) is either a variable or an individual constant.
2. "formula" of T:
   • a predicate of T followed by a term is a formula of T.
   • any logical constant or quantifier together with a formula is also a formula of T.
3. "sentence" of T is a formula of T in which no variable is free (undefined).
4. "expression" of T is a linear string of symbols.

1. The formation rules -- how is the linguistic unit formed:
   • expression (a string): operation of concatenation.
   • subject - predicate structure.
   • propositions
   • indexical signs: personal pronoun, tensed verbs, etc..
2. Rules of inference -- how is a linguistic unit read or how can it move around in T:
   • rule of symmetry
   • rule of transitivity
   • rule of detachment
   • rule of generalization

Although this toy language T above is a genuine language, its scope is quite small in comparison to a natural language, as the main interest of any natural language is about the meaning of sentences. In a syntactical system, syntax, as only a symbol or a token, does have an innate meaning for itself while it has no extensional application in a sentence. How a syntax is used or applied in a sentence and how the meaning arises from an application belong to the field of semantics. In short, syntax concerns the truth-value of the formula while semantics concerns the meaning of the sentence. The linguistic definition of semantics is as below:

A syntactical language T becomes a semantical system when rules are given in its metalanguage M which determine a Necessary and Sufficient truth-condition for every sentence of the language, and the truth-condition of every sentence in M is provable.

1. Descriptive semantics of natural language
2. Pure semantics of the analytical study of formal language.

• theory of reference -- denotation, intension
• theory of meaning -- connotation, extension

1. The sentence itself (the sentence token) -- being uttered or written as inked marks on a paper, it is composed of some symbols.
2. The mental idea (the intention or the proposition) of the speaker -- which is suppose to be carried by this sentence token.
3. The understanding of the speaker's proposition by a reader -- this requires a shared understanding of those symbols' denotation (its reference) and connotation (a meaning beyond its direct reference).

Proposition is a position that a person holds on an issue or an object after his judgement (or an intentional act) on them. Yet, the linguistic proposition consists of two parts:

• a mental act (proposition act) which is directed toward some objects or some events
• the meaning of an expression (proposition token) that is pointed out by the object or the event

1. Subject -- the one who made this proposition
2. Predicate -- a linguistic symbol that expresses the proposition act (judgement or intention)
3. Object -- a linguistic symbol that points out the object which is the target of the proposition act

• Propositional verbs -- judge, think, believe, ...
• Cognitive verbs -- know, see, hear, taste, smell, etc.

By concerning only forms and their relations, a syntactic system is always timeless. A semantic system which is defined as above (with the meanings as the central issue) does not truly concern about spatiotemporal issues as most of the propositions are also timeless. Thus, the space-time position of a sentence must be dealt with a new mechanism, the pragmatics. Pragmatics is the study of formal languages containing indexical terms, such as, tensed verbs, pronouns, demonstrative, etc.. In fact, pragmatics is simply the extension of the semantical truth-definition to formal languages containing indexical terms, for the truth-value of a sentence for relating to both the person asserting the sentence and his space-time position.

Now, this toy language T can be clearly and definitely described as consisting of the followings:

1. A syntactic system:
   • a list of symbols:
     • logic symbols:
       • one identity symbol, =
       • five connective symbols
       • two quantify symbols
       • two parenthesis symbols
     • infinite number of individual symbols:
       • individual variables
       • individual constants
   • Formation rules (terms, formulas, sentence, ...)
   • Rules of inference (for truth-value of sentences)
2. A semantic system (propositions, subjects, predicates, objects, etc.)
3. A pragmatic system (indexical signs -- tensed verbs, pronouns, demonstrative, etc.)

1. Grammar
2. Rules of inference

However, there is a significant difference between a natural language and this toy language T. The following sentences are non-sense and meaningless in T while they could be very meaningful in a natural language.

1. Type one -- tautological
   • Now is now. (non-sense in T)
   • When is the best time to do it? Now, now is now. (meaningful in natural language)
2. Type two -- illogic
   • Red is green. (false and non-sense in T)
   • When red is green, the Sun will rise up from West. (meaningful in natural language)
3. There are many more such examples.

• Language T
• A fictitious machine -- F - machine.

1. Grammar
2. Rules of inference
3. F - machine

1. List of symbols:
   • inflected vocabulary
   • a set of punctuation marks
2. Formulation rules:
   • word order -- a word string from concatenation
   • Subject - predicate
     • Descriptive
       • active
       • passive
     • Subjunctive
     • Exclamatory
3. Semantics -- Propositions (subjects, predicates, objects, accusatives, etc.)
4. Pragmatics -- indexical terms (tensed verbs, pronouns, demonstrative)

• Create three new verbs -- do, be and not
• All English verbs will be used as nouns.
• The way of substitution will be as follow:
  • Original sentence: I sing a song.
  • Substituted sentence: I do a sing a song.

• I had eaten dinner when you came. (the original sentence)
• (I eat dinner when you come, papf), the substituted sentence in a pair structure. The S-body is "I eat dinner when you come), the S-tag is papf (past perfect tense).

• a = sentence type (descriptive, subjunctive, exclamatory)
• b = voices (active, passive)
• c = tense
• d = numbers
• ...

• b-word (having w-body without a w-tail)
• i-word = b-word + w-tail

• I love you
• You love I

• You love-I
• Love-I you

1. a unique meaning
2. an array of 5! (five factorial = 120) combinations
3. a Google outcome. With a Google data base, these five words can produce a big google outcome.

• With hyphen -- there is a word order for the phrase
• With parenthesis -- there is no word order for the phrase. (I, go-school) and (go-school, I) are the same.

1. For vocabulary:
   • i-words and b-words, paired word structure
   • transformed all verbs into action-nouns with three new verbs (do, be, not)
2. For sentence:
   • paired-sentence structure (S-body, S-tag)
   • word-phrase method to reduce the power of word order

Now, if we choose the mother bag English as the u-language, the criterion one (C1) has been met automatically as the mother bag is identical to the natural English (the baby bag) structurally. The only differences are some English grammar which are mechanized, that is, jobs are done by a formalized grammar table and a machine. For example, a sentence of the mother bag below,

{(If I have money, I have 10 house), (subjunctive, past, number)}

1. Most of English grammar is formalized as a table which can be learned in one or two days. The learner does not need to apply those English grammar word by word in a sentence but chooses a S-tag from the table and places it at the end of the sentence. Then, a computer can print out a proper English sentence if he chooses to do so.
2. For inflected words, only the noun form is required in this u-language. All the verbs are treated as action-nouns. That is, the required vocabulary for this u-language is about 10% from the original English, which is 90% reduction. However, can this reduction enough for this u-language meeting the C2 for all the non-English speaking people?

• Method 1: Replacing all English noun words (the w-body) with a true (100%) root-word system.
• Method 2: Making all natural languages dialects of this u-language.

1. For vocabulary:
   • i-words and b-words, paired word structure
   • transformed all verbs into action-nouns with three new verbs (do, be, not)
2. For sentence:
   • paired-sentence structure (S-body, S-tag)
   • word-phrase method to reduce the power of word order

1. For vocabulary:
   • There are only b-words, no i-words, nor verbs. All verbs are b-words in the mother proper.
   • All (100%) b-words of English will be replaced with words which are composed of from only 240 root words as root-word strings. These 240 root words are not English but are specially designed for the universal language.
     Note: The words of many natural languages are patterns of temporally ordered sound types, and meaning of a word does not attach to particular activities, sound, marks on paper, or anything else with a definite spatiotemporal locus. The meaning of those words is agreed by a linguistic community. That is, it will take a great effort to learn those words. On the contrary, the meaning of all b-words of this Universal (Mother Proper) can be read out from the string of the root-words.
2. For sentence:
   • All (100%) formation rules of language T or English (word order, subject-predicate, etc.) will not be used. The only formation rule is word-phrasing of b-words with hyphen and parenthesis.

• Beginning with the mother bag English,
• Only English b-words are replaced with universal b-words.
• The i-words of English:
  • Was: i-word (English) = b-word (English) + inflection
  • Is: i-word (U (English)) = b-word (U (mother proper)) + inflection (English)
• Nothing else of the mother bag English is changed.
  • Formation rules: U (English) = mother bag English = natural English

1. Beginning with the natural language of English
2. From the natural language of English, we get mother bag English.
   Natural English = mother bag English (structurally identical)
3. From the mother bag English, we get the Universal (Mother Proper), a presumed universal language.
   U( mother proper) has its own vocabulary which is composed of from 240 root words in my design.
4. From U( mother proper), we get U( English). The b-word (English) is replaced with the b-word U (mother proper).

• the mother bag English is a dialect of natural English,
• U( English) is a dialect of mother bag English
• U( English) is also a dialect of U( mother proper).

With the same BMFB procedure, we can construct U (Russian), U (German), U (Arabic), U (Chinese), etc.. Then, is it now reasonable to propose another postulate?

Postulate 2: The U (of any natural language) is a dialect of the U (Mother Proper) .

The true Universal Language consists of the followings:
1. The Universal (Mother Proper) -- U (mother proper)
2. The U (natural languages); dialects of the U (Mother Proper)
   • U (English) <---> mother bag English
   • U (Russian) <---> mother bag Russian
   • U (Chinese) <---> mother bag Chinese
   • ... others

1. list of symbols:
   • conceptual words only -- b-words (mother proper) composed of from only 240 root words, no i-words nor any kind of inflection.
   • punctuation marks -- the same as English
2. Formation rules:
   • with two types of word-phrasing
     • with hyphen -- having word order
     • with parenthesis -- having no word order
   • all other English grammar are excluded
3. rules of inference -- the same as English
4. fictitious machine -- the same as English

1. Referential theory -- every word (a linguistic token) always has one non-linguistic object in the real world as its reference, such as the word token "s-t-a-r" corresponds to the star in the sky. For unicorn (a fabled creature) , there is still a picture of this animal on paper.
2. Ideational theory -- every word token marks a representation of an idea. Communication is successful when my utterance arose in you the same idea which led, in me, to its issuance.
3. Linguistic community theory -- a word token, the bearer of meaning, is a relatively abstract entity. Thus, the word token that one uses lose its meaning if one misuses it. A word is a common possession of a linguistic community, and it has the meaning it has by virtue of some general facts about what goes on in that community.

1. the innate meaning (the syntax meaning)-- it arises from its composing root words, and everyone who knows those 240 root words can read its innate meaning from the face of the word token.
2. the meaning from its usage (the semantic meaning) -- this needs to be learned during the usage of the language, similar to the linguistic community theory.

1. 240 root words for the U (mother proper);
2. 300 first generation words (b-words) for the U (mother proper) and for the U (English);
3. 2,000 words U (mother proper)/natural English dictionary (coming soon),

Most of previously claimed universal languages, such as Esperanto, are spoken languages with a lesser emphasis on the written part. While learning a new spoken language is not easy, especially without a speaking environment as a constructed language will face, learning a new written language under such a circumstance is going to be much harder. Even for English, people who use English as their native language do not know how to spell difficult words, since they basically know English as a spoken language. On the contrary, the U (mother proper) is a silent language. All its root words are ideographs and are silent. Any b-word of U (English) will be pronounced the same as the b-word of English. In fact, the b-word of U (Arabic), identical to the b-word of U (English) in word form, will be pronounced the same as the b-word of Arabic. That is, learning the U (mother proper) and U (English) needs not putting up an effort of learning a new spoken language. This unique feature of the U (mother proper) will further assure its meeting the criterion 2.

However, the U (mother proper) is also a spoken language. I did design 300 sound modules which are the generation 1 words, that is, they are the grandfather of many descendant words. They can be used as sound roots for those descendant words. However, I did not provide any sound for those sound modules, as they can be assigned by the users. That is, the spoken part of this U (mother proper) is yet to be finished by the using community.

With the above analysis, the U (mother proper) does meet both the C1 and C2. If anyone has doubts about it, it is always testable, especially for C2.

Futhermore, this U (mother proper) can be the base of a true auto-translation machine. While the b-word of Arabic and the equivalent b-word of English are having different word forms, their corresponding b-word of U (mother proper) could be the same word. Thus, an auto-translation machine can be constructed as follow:

1. Word of English ----> b-word of mother bag English + w-tail
2. b-word of mother bag English ----> b-word of U (English) + w-tail
3. b-word of U (English) = b-word of U (Arabic)
   w-tail (English) ----> w-tail (Arabic)
4. b-word of U (Arabic) ----> b-word of mother bag Arabic
5. b-word of mother bag Arabic + w-tail (Arabic) -----> Word of Arabic

• the syntax (formal) path -- word to word translation
• the semantic (meaning) path -- synonym translation
• cultural path -- considering the culture difference
• situation path -- considering the situation difference