Tetral

tetral is an artistic language designed to be used by cyborgs and robots in science-fiction stories. Therefore, it is also highly logical, in order to sound very computer-like. The language is designed around a base 4 numeral system.

=Phonology= tetral name originates from the amount of symbols in its basic system: four. These symbols correspond to four figures (0, 1, 2 and 3), and have several ways of being represented - either on paper, orally, or any by any mean to represent a base-4 number. This is a big advantage of tetral : you can commumicate either by speaking, writing, sending smoke signals, slapping on the wrists ...

Another advantage of this system is that it is easily encoded on a computer, since a computer can only encode numbers (there is a one-to-one mapping between base-4 numbers with one figure and binary numbers with two figures). Hence, tetral text is very compact on a computer.

In this document, we will use the sword writing system The sword  alphabet represent 0 as a vertical bar, 1 as slash, 2 as a backslash and 3 as a dot centered in height. Here is the table :

tetral figures are usually put together in group of 3 called codons. The codons are the actual basic units in grammar. Let's just glimpse a bit in it, and note that there is two types of codons : the structural codons are codons composed of three times the same figure, the regular codons are the 60 others.

In the sword writing system, the three figures constituing codon are written vertically (downwards), wheras sequence of codons is written horizontally (from left to right), albeit the direction of text and codon may vary from a document to another one. Let's just give a sample of text written with the sword alphabet.



Its numeric table is

The fellowing example shows another alphabet based on the same principle, but with different symbol for figures, which are more suited to be printed with a stamp. Little translation or rotation of a character does not affect its meaning, therefore it is very easy to write tetral manually with stamps.



=Grammar=

Regular words
These words are written with the sixty codons that contains at least two differents figures. There is no further subclassification on their role. For instance, can be translatted either as the adjective "large", the verb "enlarge" or the nominal expression "a large object" depending on the context it is used with.

If you know a bit of lojban, notice that there is no clear distinction such as selbri (relation word) and sumti (argument word) in tetral. In tetral every word can be used as an argument or a relation or both. (Though the distinction seems at bit blurrier in Lojban than what I explained) A regular word just carry its meaning, and have nothing implied with the sentence structure. So, in tetral, the link between a word and one of its parameter is not encoded in the relation word (In Lojban, every selbri defines what each argument means based on their position in the sentence).

Example Tree :

Structural words
These words are written with the three codons, and . They do not carry any meaning (or almost none). The list of structural words is very confined, but half of these words are the most used ones, since they are use to structure the sentences.

The blank world
This word is written. It is used to separe other types of words. Its role is similar the latin space, but is translated by a comma  in latin form. The blank word between two words is sometimes mendatory, sometimes facultative.
 * You must use it between two regular words or two structural words.
 * You shouldn't use it between a regular word and a structural word.
 * The first rule can be eluded between two closure structural word ( [[File:333.png]][[File:111.png]] or [[File:333.png]][[File:222.png]] ) words.

Sentence structure
Sentences are articulated as arborescences where a node corresponds to a word and its children correspond to its qualifiers or the words it refers to.

Computer science terminology will be reused here. The topmost word is called the root. The depth of a word (or node) is its distance from the root. Put in another way, the root have a depth equals to 0 and every other nodes have the depth of their parent plus one.

The height of an arborescence (or sentence) is the maximal depth of the nodes composing it.

Parameters
Parameters are words used to refine the meaning of an other word, like adjectives and relative propositions do in English. For example red colour refines the meaning of apple. The parameters of a word a given in the parameter list, right after the word they are supposed to modify. The parameter list begins with and ends with. Structure :

Example : A red apple

The meaning of a word can only be modified through its parameters.

Links
A link word is a word changes the meaning of the relation between a word and one of its parameter. The link is put in front of the parameter it applies to, and is separed from it with. When a parameter has no given relation, a "to be" relation is assumed i.e. the parent word is also one the entity given by the parameter. An explicit relation changes this meaning.

A link word can be any kind of word, but a small set of word are used in 80% of cases.

structure :

Example : I'm looking at you



Note : means "owned by". In the case of an action, the owner should be understood as the subject of the action. Similarly, means "owns" and refers to the direct object in when relating with an action.

A link can have it own parameters itself, although this is usually used in very specific cases as it can make the sentence unclear. The most common example is to specify whether a link applies only at the beginning or the end of an action. Example : to come from home.



Note : Since origin of a place is often used (in motion verb for instance), there exists the word that means  on its own.

Collections
In every languages it is common you have to refer to several entities with the same properties or actions. One obvious solution is to repeat the properties to each entity, but this is cumbersome. English skips repetion by allowing to stack different subjects before the verb (the last word separated by and). In tetral you cannot stack words one after the other to group them. Indeed such stacking just enumerate words and any parameters put after would apply only to the last of them. Remember : there relations are only between a parent word and its children. Tetral add a new concept, a collection wich basically allows us to group various entities. Collections can group word entities to give them the same parameters. But it is stronger than that. It can group several parameters and give them the same link with their parents as it can group several links and give them the same parameter.

A collection is started with the codon and ends with codon.

Examples :

green apples, melons and pears. (Each of them are individually green.)

Looking at an apple and a melon. (Looking one after the other, or a the same time, ... The order is unspecified.)

Looking at myself.

Please note that nothing prevents a word from having its own parameters list inside a group. You can also also nest groups but it sounds a bit weird and its usage is strongly discouraged when uneeded. I may be justified when you have very specific parametrisation of word in a large collection. Like in the following example, where is a parameter applied to two entities, and  is a parameter applied the the three entities.

A collection is sometimes referred as collection list, to remember it is written down as a list like the parameters list.

Closure
Some sentences are very huge arborescence that can go deep. Some conveniance tools and some good practise is needed to keep track of opened lists. The first good practise is obviously to avoid to build large height tree.

As this is not always possible to reduce height and the height of the tree is hardly predictable before saying it, there is a second good practise : add useless blank word after the opening and before the closing of a depth-4 list (or at a multiple of 4 depth).





But the previous exemple has something special more. It use the pattern with one parameter nested several times. This grows up the depth of the tree as fast as adding meaningfull words. This is innefficient. Hence, a new connector, which like the relation connector, do not growths the depth is added. It is. It is a simple shortcut to a one-word parameter list. is equivalent to. Please do not mistake this shorcut with a relation. A relation specify the link between two words, and is used to actually links the two words. can be used with relations, though. is equivalent to. Do not mistake with, wich is a parameter linked to its parent by the relation. Note also that  can never mean.

Variable-length words
=Vocabulary=

Common variable identifiers
=Example : Babel=