Lojban
Lojban (AFI /ˈloʒban/) is a constructed language, more accurately a logical language, based on first-order logic (predicative logic) created by the Logical Language Group in 1987. Its predecessor is Loglan, the original logical language created by James Cooke Brown.
Language development was started in 1987 by the Logical Language Group, which attempted to achieve the purposes of Loglan, as well as complement the language by making it more practical and freely available. After a long period of debate and testing, the foundation was completed in 1988 with the publication of The Complete Lojban Language.
The name Lojban is a portmanteau of loj and ban, which are short forms of logji (logic) and bangu (language), respectively. Lojban is a spoken language, used for communication between people. Even being able to express the most complicated logical concepts, it is highly flexible. Depending on the degree that the speaker wishes, it can resemble a natural language or a programming language, or other constructed languages, being semantically ambiguous, poetic, precise or neutral.
The main sources of the basic vocabulary are the six most widely spoken languages in the world: Mandarin Chinese, English, Hindi, Spanish, Russian and Arabic, chosen to increase the familiarity of the words to people with diverse linguistic backgrounds. The language took components from other constructed languages. A notable example is the set of indicators in the Láadan. Also Toki Pona and Esperanto are similar to Lojban to some extent.
In this article, explanations of its grammatical aspects will be based almost entirely on The Complete Lojban Language; the latin mode will be used for spelling.
History
Origins (1955-1987)
Lojban's predecessor, Loglan, was invented by James Cooke Brown in 1955 and developed by the Loglan Institute. The Loglan was initially conceived as a means to examine the influence of language on the subject's thinking (a thesis known as the Sapir-Whorf hypothesis).
As Brown began to claim copyrights to language components, the activities of the Loglan community were restricted. In order to avoid this control, a group of people decided to start a separate project, departing from the Loglan lexical base and reinventing the entire vocabulary, resulting in the current Lojban lexicon. This group of people established in 1987 the Logical Language Group, based in Washington DC. They also won a lawsuit over whether they could call "Loglan" to your version of the language.
The Freeze Period (1997-2002)
After the publication of The Complete Logical Language, it was expected that "the documented lexicon would be established and the combination of lexicon and comprehensive grammar would be frozen for a minimal period 5 years as the use of the language grows" As stated, this period, which has been officially called the "freeze," expired in 2002. Lojban speakers are now free to construct their own words and expressions, and to decide where the language goes.
Main features
Lojban retains many of the features of Loglan:
- Its grammar is based on the logic of preaching, designed to express complex logical constructions with precision.
- It has no irregularities or ambiguities in spelling or grammar. This gives rise to great clarity for your computer syntactic analysis.
- You talk exactly the way you write. For example, there are several words equivalent to the sign ? used in different languages to point out a question at the end of it. In Spanish there is no spoken form of this symbol, the questions are indicated by a change in the intonation at the beginning of the prayer to question it.
- It is designed to be culturally neutral.
- Its basic morphemes were taken from common or combined elements of the six most spoken languages (at the time of its elaboration): Mandarin, English, Hindi, Spanish, Russian and Arabic, reconstructed according to the phonetic and grammatical norms.
- It allows systematic use and learning compared to natural languages.
- It has an intricate system of indicators that effectively communicates attitudes and emotions.
- Simplicity is not a criterion of its design.
- Although the initial objective was to investigate the Sapir-Whorf hypothesis, the Lojban community has additional objectives, such as:
- Generic research on linguistics.
- Research in artificial intelligence and communication with machines.
- Better human interaction with computers, ontology storage and automatic translation of natural languages.
- Possible use as an international auxiliary language
- Use of language for education.
- Personal creativity.
Grammar
Phonology
Lojban uses the Latin alphabet (there are other ways to write it, but they are rare). The Lojban alphabet consists of 17 consonants, 6 vowels and 3 auxiliary characters ('.) that are not punctuation marks. Most of them have a single pronunciation, although some alternative pronunciations are allowed to facilitate pronunciation for speakers of different languages. It also has 16 diphthongs (but no triphthongs). The apostrophe (') is used when two vowels do not form a diphthong, a character that is usually pronounced /h/.
Lojban is written in all lowercase letters, that is, sentences do not begin with a capital letter. However, capital letters are used to indicate accents of words that are outside the norm. The Lojban letters and their respective pronunciations are shown in the following table. AFI symbols without parentheses show preferred pronunciations (others are similar).
consonant | vowels | auxiliary characters | ||||||||||||||||||||||||
representation AFI | b | ()) | d | ()) | g | ()) | k | l ()) | ()) | n (,,), 한) | p | r ₡,, | s | t | (β) | x | z | a | e/ | i | o ()) | u | ♫ | h (θ) | . | , |
Latin character | b | c | d | f | g | j | k | l | m | n | p | r | s | t | v | x | z | a | e | i | or | u | and | ' | . | , |
Cyrillic character | . | ш | д | ф | г | . | к | ll | м | . | . | р | с | т | в | х | ♪ | ♪ | е | и | ♪ | у | ONE | ' | . | , |
In the case of the "r" there is no preferred sound, any rhotic sound is equally acceptable.
In principle, there can be many more orthographic systems as long as they satisfy the required unambiguities. Some of the reasons for this flexibility would be the following:
- The Lojban is defined by its seals, therefore a system of representation can be appropriate to represent the Lojban as long as its symbols are assigned in such a way that the audiovisual isomorphism is maintained.
- The Lojban pretends to be as culturally neutral as possible, therefore it is not crucial to choose any particular system of representation (e.g., the Latin alphabet). Some lojbanparlantes have extended this notion until they claim that a native system should be sought for language.
This article will always use the Latin alphabet, as it is the most familiar to Spanish speakers.
The 16 diphthongs with /AFI/ are:
ai /aj/ | ei/display/ | oi /oj/ | au /aw/ | |
ia /ja/ | ie /j transformation/ | ii/ji/ | io/jo/ | iu /ju/ |
ua /wa/ | ue /w transformation/ | ui /wi/ | uo /wo/ | uu /wu/ |
iy /jambi/ | uy /wambi/ |
Numerals
In Lojban, numerals are read by digits. There is a word for each decimal digit from zero to nine. Numbers greater than nine are written by joining two or more of these words in the same order in which the decimal digits of the number to be represented are written. For example, 28 is said rebi. The numbers from ten to fifteen can be represented in this way or by the words that are assigned their respective digits (A, B, C, D, E, F) in hexadecimal numeration.
Number | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | A | B | C | D | E | F |
Lojban | No. | pa | re | ci | Vo | mum | xa | ze | bi | so. | dau | fei | gai | jau | rei | vai |
Morphology
Lojban has three word classes: predicative words, structural words, and nouns. Each of them has unique properties that identify them in such a way that you can recognize which of these categories each word belongs to. These categories have their own subdivisions with their own morphological characteristics.
Syntax and semantics
Grammatical structures "are defined by a set of rules that have been shown to be unambiguous in computer tests". These grammatical rules are effectively called a "machine grammar". From this machine grammar the syntactic characteristics of Lojban are derived:
- Each word has exactly a grammatical interpretation
- Words have exactly one way to relate to each other.
However, these standards are achieved with some care:
It is important to note that new lojbanparlants will not be able to speak 'perfectly' by starting to learn Lojban. In fact, they may never speak 'natural' in a conversation in Lojban. No English speakers always speak in a perfect English in a spontaneous conversation; Lojban's speakers will also make grammatical mistakes when speaking quickly. However, they may speak or write in an inambiguous way if they are careful, which is difficult if not impossible with a natural language.
Nick Nicholas and John Cowan. What Is Lojban? II.3
Expressions in Lojban are modular; short and simple expressions can be nested in longer sentences in such a way that the parts manifest a grammatical unity. This mechanism allows expressing concepts of great logical complexity.
Its typology can basically be considered Subject Verb Object. However, almost any can be used:
- mi prami do (SVO)
- mi do prami (SOV)
- do se prami mi (OVS)
- do mi se prami (OSV)
- prami fa mi do (VSO)
- prami do fa mi (VOS)
Expressions that all equal "I love you" ({my prami do})
This flexibility is related to the purported ability of the language to be able to express as many natural language expressions as possible, based on a single system of positions. The meaning of the sentence {mi prami do} is determined by {prami}, which has its own "place structure"; a specific semantic relationship between {my} and {do}; when the relationship of places between {mi} and {do} changes, the meaning of the sentence changes as well. The previous examples show the resources that exist in Lojban to change the order of words without altering the semantic structure.
As benefits of a logical language, there are a large number of logical connectives. These conjunction words take different forms depending on what they connect, another reason Lojban expressions are often precise and clear.
Multiple predicate words can be concatenated to reduce the semantic range of a sentence. In skami pilno, "computer user", the word skami (computer) modifies the broader meaning of pilno (user) to form a more specific concept. As you can see, the order in which the words are affected is from left to right, different from the more usual way in Spanish (from right to left).
Examples
Common Phrases
Lojban | Verbatim significance | Spanish |
---|---|---|
coi/co'o | [sighs]/[goods!] | Hello. |
pe'u | [please!] | Please |
ki'e | [thank you] | Thank you. |
.u'u | [Repentance!] | Sorry. |
xu do se sanbau/jbobau | [true-false?] are you a Spanish speaker/Lojban | You speak Spanish/Lojban? |
ti/ta | This | This |
my na jimpe | I understand. | I don't understand. |
Go'i | the previous statement | yes/the above statement is true |
na go'i | [false] the previous statement | no/the previous statement is false |
la'u ma | a quantity [what] | How many? |
ma jdima | which price | How much is it? |
ma stuzi lo vimku'a | What is the position of a toilet? | Where's the toilet? |
Text
Article 1 of the Universal Declaration of Human Rights:
Lojban | ro remna cu se jinzi co zifre je simdu'i be le ry. nilselsi'a.e lei ry. selcru.i ry. se menli gi'e sezmarde.i.ei jeseki'ubo ry. simyzu'e |
Analysis | [ro (all) remna (human)] cu (es) "se jinzi (is innately) co (of type) zifre (libre) je (y) simdu'i (very equal) be (en) le ry nilselsi'a (human dignity) e (y) lei ry selcru (human rights)). |
Spanish | All human beings are born free and equal in dignity and in rights. They are endowed with reason and conscience, and they must behave fraternally with each other. |
The parsing shown here is a simplified version of the parsing that the jbofihe tool produces of Lojban text.
Some Unique Lojban Expressions
- .oiro'o bu'onai pei
[physical pain!] [end of emotion] [?]
Don't you have pain anymore? - mi nelci ko
I like [imperative] you
Make me thank you! - le cukta be'u cu zvati ma
The book [necessity] is in which
I need the book, where is it? - ko ga'inai nenri klama le mi zdani
[imperative] you [me-in-feriority-social] within [of type] go home related to me
I'd honor you coming into my house. - le nanmu cu ninmu
One or more specific things I describe as "men" is a woman
The man is/is a woman.
Tongue Twister
- Lo'u lu le la li'u le'u
- he crisa srasu cu rirci crino
- tisna fa la tsani le cnita tsina lo tinci tinsa
- the bab. zbasu loi bakyzbabu loi bakygrasu
- mi na djuno le du'u klama fa makau la makaus. makau makau makau
A poem in Lojban (audio)
Comparison with other artificial languages
Logan
Loglan is now a generic term that refers to both James Cooke Brown's Loglan, and all languages derived from it. Since the organization created by Dr. Brown, The Loglan Institute (TLI), still calls their language Loglan, it is necessary to mention that this section refers specifically to the language of the TLI, and not to the family of languages derived from it.
The main difference between Lojban and Loglan is their vocabulary. A Washington DC breakaway group, which would later form the Logical Language Group (LLG), decided in 1986 to reinvent the entire Loglan vocabulary in order to avoid Dr. Brown's copyright claims. After a lengthy legal battle, this copyright claim was dismissed. But by then, the new vocabulary was already a fundamental part of a new language that was called by its supporters Lojban: A Realization of Loglan.
The root word set was the first part of the vocabulary to be reinvented. The words in Lojban were made with the same principles as those in Loglan; that is to say, they were chosen from among candidate forms according to how many sounds they had in common with their equivalents in some of the most widely spoken languages in the world, which was multiplied by the number of speakers of the languages with which the words had letters. in common. The difference that this reinvention produced was that the root words reflected certain changes in the number of speakers of the languages. This resulted in words that had fewer English borrow sounds and more Mandarin Chinese borrow sounds. For example, the Loglan word norma is equivalent to the Lojban word cnano (Chinese 常, pinyin cháng), meaning &# 34;normal" in both languages.
Lojban and Loglan still have essentially the same grammar. The simplest declarative sentences could still be translated word for word between both languages; but its grammar differs in some details, and in its formal foundations. The Lojban grammar is defined primarily by the YACC linguistic definition formalism, with a few prerequisite processing rules. Loglan also has a machine grammar, but it is not yet final; it is based on a relatively small number of sentences that have remained intact for decades, which take precedence in the event of a grammatical conflict.
There are also many differences in the terminology used to talk about the two languages. In his writings, Brown used many terms based on the English, Latin, and Greek languages, some of which were already established with meanings slightly different from those used by Brown. On the other hand, Lojban freely borrowed many terms from itself. For example, what linguists call roots or root words, loglanists call primitives or prims, and Lojbanists call it gismu. The lexema of Loglan and the selma'o of Lojban have nothing to do with the linguistic meaning of lexema. It is rather a part of speech, a subdivision of the set of grammatical words, or particles, which Loglanists call small words and Lojbanists cmavo. Loglan and Lojban use grammatical constructions called metaphors and tanru respectively; these are not true metaphors, but a sort of modifying-modifying relationship similar to the adverb-verb relationship in Spanish. A loan word in Loglan is simply called a loan; but in Lojban we say fu'ivla.
In the new Lojban phonology, the consonant q and the vowel w were removed, and the consonant h was replaced by x. The consonant ' (apostrophe) was added with the value of /h/ in the International Phonetic Alphabet, however some speakers use the sound /θ/)
Lojsk
It was conceived by Ari Reyes, heavily influenced by Loglan, Lojban, Universal Networking Language (UNL), Esperanto, Visual Basic, Dutton's Quick Words, Ceqli and Guaspi. It is designed to be more monosyllable oriented. This feature could make Lojsk even more sensitive in noisy environments than Lojban.
Voksigid
Voksigid, created by an Internet working group led by Bruce R. Gilson, aims to build a predicative language different from others. Its syntax is somewhat influenced by Japanese, and its vocabulary is based mainly on European roots. Loglan and Lojban both use word order to mark parts of a predication, but since remembering which position refers to which meaning may be beyond most people's memorization capacity, Voksigid was designed to overcome that problem. It uses a great variety of propositions with specific semantic contents to mark the roles of the arguments in using the order of the same to indicate it.
Whospi
Guaspi is a descendant of Lojban and Loglan that uses Chinese-like tones to mark grammatical structures. It was developed by Jim Carter. By using tones instead of structure words, and by cutting the predicates from two to one word, Carter has solved a minor problem in Guaspi's predecessors: it takes a lot of syllables to say things.
Community
On the Internet
The activities of the Lojban community take place largely through the Internet:
- Lojban.org: A site maintained by its users that is a reflection of the lojban community outside the LLG.
- Lojban IRC (irc.freenode.net #lojban): Located on the Freenode network (IRC). A web interface can be used as an alternative to IRC clients.
- Lojban mailing list: A mailing list dedicated to talking and learning about language (oriented to beginners).
- jbovlaste: The editing interface of the official dictionary of Lojban. Created by Jay Kominek, updated by Robin Lee Powell. New words can be proposed in Lojban with definitions and examples, or voting for other experimental words.
- jbobac: A web forum with discussions made mainly in sound files.
- samxarmuj/The Lojban Moo: A multi-user environment similar to ancient adventure games. A guide is provided here.
- le jbopre pe lj's Journal: A community blog.
- lojban-valsi Archived on March 10, 2013 at Wayback Machine.: A mailing list that sends a word a day at Yahoo! Groups.
- jbotcan: A community of people who practice their Lojban, ask questions and propose ideas related to the Lojban.
- uikipedias: Wikipedia in Lojban, where discussions can be done in English.
The Logfest
Lojban-speaking gatherings have been organized in the United States since 1990, called Logfests. It is mostly informal, lasts a weekend, with the only scheduled activity being the annual meeting of the LLG. Those who cannot be present can still get involved via IRC. Activities are anything attendees want to do: Lojban conversations, lectures, technical discussions, or socializing.
Population
The total number of Lojban speakers is unknown.
According to Lojban.org, some places with significant concentrations of Lojban speakers are:
- Australia, Israel, United States
Frappr.com also shows that as of March 2009, people from the following countries have taken an interest in the language:
- Argentina, Canada, China, Estonia, France, Ireland, Mexico, Netherlands, New Zealand, Norway, Russia, Thailand, Ukraine, United Kingdom, and Venezuela.
It has been noted that there is generally little participation from Hindi speakers, despite the fact that the etymological nature of Lojban vocabulary is heavily influenced by this language.
The following are some of the notable personalities who have contributed to the development of Lojban:
- Bob LeChevalier (aliases) lojbab): President of the LLG.
- John Cowan: author of The Complete Lojban Language.
- Jorge Llambías (aliases) xorxes): one of the most involved lojbanparlantes and has made several translations. It is also a prominent figure on the mailing list, helping beginners with language.
- Matt Arnold. epkat): one of the most active lojbanparlants. He has been contributing to the translation project and software development.
- Nick Nicholas (alias) nitcion): Australian linguist, was the first speaker to speak the language fluently. He's written a lot on lojban including Lojban for beginners in co-authorship with Robin Turner.
- Robin Lee Powell camgusmis): The current administrator of lojban.org. It provides the bandwidth with which the site is served. He has also written several materials in Lojban.
- Robin Turner: An English philosopher. He's the co-author. Lojban for Beginners.
Literature and Resources
Literature
Like most languages with few speakers, Lojban lacks a large body of associated literature, and its creative potential has not been fully developed (for example, the true potential of its attitudinal system is considered to be It will not be developed unless you have a community of children who have grown up in a multicultural Lojban-speaking environment). However, massive knowledge resources, such as the Lojban Wikipedia, may be able to expand the lexical horizon of the language.
The Lojban texts currently available are mainly concentrated on Lojban.org, although there are independent Lojban blogs as well. The Lojban IRC channel (or its archive) has a collection of Lojban expressions, but their grammatical correctness is not guaranteed. These materials available on the Internet include both original works and translations of classic pieces in the field of natural languages, from poetry, short stories, novels, to academic essays. This has been likened to a chrestomatic project to produce a collection of translated texts with the aim of showing a wide variety of examples of the language with a variety of genres and styles. Some examples of the works that are available:
- Alice in Wonderland by British-Irish author Lewis Caroll
- The Mil and A Night (a classic Arabic account).
- The Moon is a Severe Love (Chapter 1) by science fiction author Robert A. Heinlein
- The Prophet by the Lebanese author Kalil Gibrán
- Small things by the author of short stories Raymond Carver
- The Man and the Serpent by the author of short and satirical stories Ambrose Bierce
- The book by the author of terror H. P. Lovecraft
- Legend of Zelda (a fantastic NES game) by Nintendo [chuckles]required]
Learning resources
Aside from language practice, some members of the community and the LLG have developed projects to help students. The Complete Lojban Language, the last word in all aspects of Lojban, is one of them, completed in 1997. Some of the projects, with different degrees of development are:
- Grammatical analysers:
- jbofi'e (by Richard Curnow)
- Official Grammatical Analyst (Robin Lee Powell)
- valfendi (by Pierre Abbat)
- Dictionary/base of data:
- jbovlaste (by Robin Lee Powewll)
- Reference Database (by Matt Arnold)
Contenido relacionado
Category:Korean writing
Venezuelan languages
Hanja