Log in

No account? Create an account

Previous Entry | Next Entry

Любителям языков: Ithkuil

Собираю по крупицам остатки моего сайта, уничтоженного Гуглом. Остатки очень жалкие. Собственно почти всё, что сохранилось - это вот эта статья Джона Кихады - создателя искусственного языка Ithkuil. Язык Ithkuil - это самый сложный (практически - невероятно сложный) из когда-либо созданных искусственных языков, и, по всей видимости, самый хорошо продуманный. Джон Кихада написал эту статью по моей просьбе. Там же было ещё и очень большое, интереснейшее интервью с ним, но оно пропало безвозвратно.

После истории создания Ithkuil'а следует подробнейшее описание процедуры перевода первого предложения "Анны Карениной" на Ithkuil. Любители языков получат огромное удовольствие.

History of Ithkuil by John Quijada

I’ve been fascinated with foreign languages since about age 11, when, believe it or not, I saw a photograph of a Moscow street scene in a book about the Soviet Union. There was an advertising sign in the photo – it was the first time I’d ever seen Cyrillic writing. I was instantly fascinated and began reading every book in my local public library about every language I could find, including Russian. At age 15, I read the books of J.R.R. Tolkien and like so many others, was fascinated at the idea that he had made his own languages. My brother and I began creating our own rudimentary languages just for fun, the first ones being little more than relexified English with a few “exotic” features from other European languages added in.
At university, I studied linguistics, and it was there that I gained the knowledge and tools for understanding how languages are structured. However, the more languages I read about and studied, the more I came to realize how inefficiently most languages function in terms of conveying thought in both a precise and a concise manner. I also noticed that certain languages were “better” than others in the manner that they were able to convey a specific task. For example, I noticed how elegant and efficient the three-letter root structure of Semitic languages like Arabic and Hebrew were as a means of building words compared to European languages. I noticed how the perfective versus imperfective verbal aspect of Slavonic languages like Russian were able to convey certain verbal distinctions easily which languages like English had to use whole phrases to convey.
In other cases, I found certain languages that grammaticalized thoughts that most other languages did not (such as the “4th person” distinction of certain American Indian languages). I also fell in love with all the exotic and difficult-to-pronounce consonant sounds of Caucasian languages like Abkhaz and Ubykh, as well as the numerous vowels of Uralic and Altaic languages.

The idea came to me that I might try to create a language which “combined” the most efficient and interesting features of all the languages I was familiar with. And so the seed of the language which eventually evolved into Ithkuil was born. The more I worked on it over the years, the more of my own ideas went into it, as opposed to simply borrowing ideas from existing languages. Then, during the 1980s, I discovered the writings of the new “cognitive” school of linguistics that was beginning to arise in the United States. Specifically I discovered the book “Metaphors We Live By” by George Lakoff and Mark Johnson, as well as Lakoff’s book “Women, Fire, and Dangerous Things.” These books opened up a whole new level of understanding regarding the relationship between human thought and language which led to a major overhaul and expansion of the language I was working on in my spare time. It was at this point that I got the idea that I would try to accomplish the seemingly incompatible goals of creating a language capable of conveying much more information than natural human languages do, while simultaneously being more brief and concise than natural human languages (previously, all versions of the language were very “long-winded”).

The work was slow, painstaking, and often frustrating. Between the mid-1980s and the mid-1990s, I abandoned the work at least a dozen times out of frustration (aggravated by the fact that every time I thought I was getting nearly finished, I would decide to scrap about 75% of what I had and start over with better ideas). Sometimes several months would go by before I would pick up where I left off. The breakthrough came around 1996 when my interest was reignited by discovery of the writings of two other cognitive linguists, Guilles Fauconnier and Len Talmy. Their research and findings on the structure of human thought and language renewed my interest in wanting to finish the project.
By 1999, the grammar was about 95% finished and it was time to start finalizing the script and the lexicon. Interestingly enough, it was at this point that I discovered Robert Heinlein’s description of “Speedtalk” from his novella “Gulf.” At first I was surprised and mildly disappointed that someone had already thought of doing what I was working on, but then I realized that Speedtalk was only a shallow attempt which worked at the morpho-phonological level of language only, while the language I was creating took the same principles and applied them to all the different structural levels of language. It took me another four years to reach the point where I felt the work was complete enough to show the world, then it took me another year to write it all up formally on the computer (the language was developed entirely using pencil and paper) and create a website. The website was posted to the world on January 28, 2004. The name of the language was not finalized until 2003. Previously it was called Aekuith for about ten years, but I had to rename it because changes to the grammar had caused the word Aekuith not to mean anything anymore. Prior to Aekuith, the language had so many different names, I can’t even remember them all.

I am very grateful, surprised, flattered, and humbled by the interest that hundreds of people around the world have shown towards Ithkuil. It tells me that many intellectually-minded persons have an innate understanding that language is the most direct window we have into the workings of the mind, and of expressing the contents of the mind, and that the language we use to express thought, therefore, should be a precision instrument. Ithkuil is designed to show the world the unfortunately wide gap between how a language could be used to express thought and the grossly inadequate ways existing human languages actually do express thought.

When I started working on this language in 1978, it was with the full intention that I (and anyone else who might be interested in it as an intellectual exercise) would learn and speak it. However, over time, the language became so complex that I realized its primary purpose was no longer to be a language that anyone would necessarily learn to speak, but rather to function as a cautionary reminder to all those who would dare think (or claim) that existing human languages (usually one's own) are the ultimate means of expressing human thought. By showing the world just how a human language COULD express what's really going on inside one's brain at a cognitive level, and by seeing the vast gap between what such a language is capable of expressing versus what actually gets expressed in natural languages, hopefully such persons would come to reflect with humility what a hopelessly inadequate job natural languages make of the process. While I appreciate the efforts of those few persons who have written me to tell me they want to (or are trying to) learn to speak Ithkuil, I don’t believe its primary purpose is to necessarily be learned and spoken, but rather to be reflected upon and studied for its value as a way of seeing just how much of what is going on at a cognitive level never actually gets expressed in real languages.

Step-By-Step Procedure For Translating
The First Sentence of Anna Karenina Into Ithkuil
by John Quijada

NOTE: I am using the English translation of the sentence instead of the Russian original as my translation source:

All happy families resemble one another, each unhappy family is unhappy in its own way.

STEP ONE: Semantic Analysis of Source Sentence

Before beginning translation, it is important to carefully analyze the actual meaning of the sentence from a cognitive standpoint, i.e., what is the cognitive intent of the sentence as opposed to how it reads stylistically? This is important because Ithkuil must convey the underlying cognitive intent of the sentence. In this sentence, one might initially interpret the sentence to suggest that there are two different kinds of families, one happy, the other unhappy, and that happy ones can’t ever be unhappy or vice-versa. But in fact, all families are capable of being happy or unhappy. Therefore, the Ithkuil translation will reword the sentence to convey that when families are happy, they resemble one another, whereas when they are unhappy they don’t resemble one another. Secondly, we must analyze what is meant by the word ‘resemble’. In English, ‘resemble’ refers to similarity in visual appearance or observed behavior. However, the intent of the sentence is that the similarity of behavior is intrinsic and occurs whether or not there is anyone available to visually observe it. Therefore, the Ithkuil translation must avoid use of any concept such as ‘resemble’ to imply visual observation of behavior. The behavior simply is similar or dissimilar, regardless of whether there is any visual comparison made. The Ithkuil translation will therefore simply distinguish between similarity in the manifestation/behavior of happiness in families versus dissimilarity in the manifestation/behavior of unhappiness in families. Finally, in the original sentence, the second clause is syntactically disguised as a co-equal main clause, however, cognitively it is a dependent clause in a contingent relationship to the first half of the sentence. The “missing” conjunction should be ‘whereas’ or ‘as opposed to’ or ‘while on the other hand’ since it is being contrastively compared to the first clause. Ithkuil provides a case-frame (the COMPARATIVE) for exactly these kinds of contrastive comparisons. Therefore the Ithkuil sentence will employ such a case-frame when translating the second half of the sentence. So at this point, what we will be translating into Ithkuil reads something like “All families manifest happiness similarly, whereas they manifest unhappiness dissimilarly.”

STEP TWO: Separating the Lexical from the Morphological

At this point, the original sentence must be analyzed to determine what Ithkuil lexical roots/stems will be necessary, as opposed to those words/concepts in the original which can be translated via morphological components instead of lexical roots/stems. We can see we are making a statement about families and are qualifying those families as happy or unhappy. Therefore it is likely we will need Ithkuil lexemes for ‘family’ and ‘happy/happiness.’ We have already determined above that we do not need a word for ‘resemble’ and words like ‘all’, ‘one another’, ‘each’, ‘in its own way’ carry little intrinsic meaning by themselves until they are used in conjunction with the main participants (nouns and verbs) of the sentence. We know from studying Ithkuil morphology that all of these words will be handled by morphological, not lexical means when translating. Therefore, we only need two main lexemes: ‘family’ and ‘happy/happiness.’

STEP THREE: Analysis/Derivation Using Ithkuil Morphological Categories

Those who have studied the Configuration and Affiliation categories of Ithkuil morphology might recognize at this point that the concept ‘family’ is a composite sort of concept, indicating an entity composed of sub-components or sub-members, each one of whom is a ‘family member.’ Ithkuil tends not to lexify as discrete stems those words that are gestalt entities comprised of sub-components. Rather it is the sub-components that are lexified into stems and then morphologically modified via the various Configuration and Affiliation categories. The Ithkuil lexicon contains a word han meaning a nuclear family member, derived from the root H-N referring to the various sanguine members of a nuclear family (parent, child, mother, father, son, daughter, etc.). A family consists of various such members, each of whom are not identical in their attributes (e.g., gender, age, personality, talents, etc.). A gestalt composite composed of non-identical members who are not physically interconnected, is shown by the AGGREGATE [AGG] configuration in Ithkuil. Likewise, the resulting gestalt entity ‘family’ is more than the sum of its parts from a cognitive standpoint (i.e., it is capable of participating in events or being described in ways irrespective of its individual members), therefore it should be placed in the COALESCENT [COA] affiliation. Finally, the particular sentence being translated talks about families in a timeless cultural sense, describing families in an archetypal fashion. Such a sense can best be captured by use of the AMALGAMATE [AMG]context in Ithkuil. So far, the Ithkuil word for ‘family’ we are fashioning has been declined for Configuration, Affiliation, and Context. However, there are six remaining morphological categories for nominal formatives in Ithkuil. We must consider each of these:

PERSPECTIVE: The sentence is comparing every family against every other family, thus it is speaking of each family as a single discrete entity, not a collective concept. This is reinforced by the use of the word ‘all’ (which in the Ithkuil translation will appear as the Inclusive suffix [INL] in ninth degree, meaning ‘each/every’). Thus the correct perspective is the default MONADIC perspective.

EXTENSION: Each family is being considered as an entire discrete entity within the context of the sentence, therefore the default DELIMITIVE extension is appropriate.

FOCUS: The sentence does not have a specific component being conveyed as new information in comparison to already known background information, therefore the formative is UNFOCUSED.

DESIGNATION: The stem han is in the default INFORMAL designation, as indicated by its default stem vowel.

ESSENCE: The sentence refers to all families in the real world, not families in a hypothetical or contrary-to-fact context, therefore the default NORMAL essence is appropriate.

CASE: The word ‘family’ will be the “subject” of a verb complex meaning ‘manifests happy behavior’. The appropriate semantic role for the subject of a non-causal descriptive state is as CONTENT, shown by the default OBLIQUE [OBL]case in Ithkuil.

The concept ‘happy/happiness’ is given in Ithkuil by Q-Ḑ which literally means ‘degree of happiness/joy’.

This stem is derived from the root Q-Ḑ ‘DEGREE OF POSITIVE EMOTIONAL SATIETY/FULFILLMENT’. Qualitative concepts such as happiness are usually lexified in Ithkuil as a spectrum or range which then takes the Sufficiency [SUF] or Extent/Degree [EXN] suffix to indicate what particular amount/extent/degree of the quality is manifested, anywhere from none up to an overabundance. In this case, we will use the [EXN1/6] suffix indicating ‘an above-normal degree of’. Adjectives such as ‘happy’ don’t exist in Ithkuil; their equivalent is usually shown using a verbal formative declined into either the STATIVE [STA]conflation or the DESCRIPTIVE [DSP] conflation. The choice between using STATIVE or DESCRIPTIVE conflation for this translation is subtle. I have chosen [DSP] because it captures more of the sense of describing the nature of the family itself (which is what I think the author was trying to convey) rather than simply a state they are in. Verbal formatives are complex structures in Ithkuil, inflecting for 22 different morphological categories. So, besides the DSP conflation, the remaining 21 categories must be analyzed, as follows:

CONFIGURATION: Conceptually, ‘happiness’ is not a gestalt composite, therefore the default UNIPLEX value is appropriate.

AFFILIATION: The UNIPLEX configuration implies a default CONSOLIDATIVE affiliation.

PERSPECTIVE: The sentence is asserted as being true at all times, a “law of nature” about families, therefore the appropriate perspective is the NOMIC [N].

EXTENSION: The state of happiness being discussed is not described as having specific beginning or ending points, therefore the appropriate extension is PROXIMAL [PRX].

FOCUS: The sentence does not have a specific component being conveyed as new information in comparison to already known background information, therefore the formatives are UNFOCUSED.

CONTEXT: Happiness as an emotional state exists irrespective of opinions, cultural ideas, social conventions, etc., therefore, the default EXISTENTIAL context is appropriate.

DESIGNATION: The stem is Q-Ḑ in the default INFORMAL designation, as indicated by its default stem vowel.

ESSENCE: The sentence refers to happiness as a real-world state, not in a hypothetical or contrary-to-fact context, therefore the default NORMAL essence is appropriate.

ILLOCUTION: The sentence constitutes a propositional speech act, an utterance which is either true or false and can be either believed or disbelieved. Thus, the default ASSERTIVE illocution is appropriate.

VALENCE: The state of happiness is being compared between families, implying PARALLEL valence (i.e., one participant is engaged in or experiencing the same activity or state as the other). However, because the verbal formative will be carrying the [SIM1/8] suffix (meaning ‘very similarly’), the valence relationship between one family and another is already implied. This means that overtly showing the PARALLEL valence in this sentence is unnecessary (or at least optional).

VERSION: The utterance describes a non-hypothetical end-in-itself and is not goal-oriented, therefore the default PROCESSUAL version is appropriate.

FORMAT: The descriptive conflation is a primary conflation, not a derivation (i.e., a secondary conflation), therefore the showing of Format is inapplicable.

MODALITY: There is no modal or hypothetical content in this sentence, therefore there is no modality.

LEVEL: While one might consider translating this sentence using Ithkuil’s EQUATIVE level, its use would imply that families manifest happiness exactly the same way (as opposed to similarly). This is not the intent behind the word ‘resemble’ and therefore use of the [SIM1/8] suffix (meaning ‘very similarly’) is more appropriate. Therefore, the default INDETERMINATE level is appropriate.

CASE-FRAME: As previously mentioned, the second clause of this sentence will employ the COMPARATIVE [CMP] case frame to signify a “whereas X” or “as opposed to X” type of relationship.

VALIDATION: Since the statement is not an empirically verifiable statement, but is written as being a truthful assertion, the implied source of information behind the statement is as an inference on the part of the author based on his own intuition or feelings. Therefore the appropriate validation is the INFERENTIAL.

PHASE: The statement does not refer to any repetitive or iterative situation, therefore the default CONTEXTUAL phase is appropriate.

SANCTION: The statement constitutes an ontologically subjective assertion or allegation expressing an opinion or belief, potentially subject to challenge or refutation. Therefore, the appropriate sanction is the ALLEGATIVE [ALG].

ASPECT: The sentence contains no additional aspectual information.

MOOD: The sentence is an assertion of an alleged fact, therefore the default FACTUAL mood is appropriate.

BIAS: The utterance contains no overtly subjective attitude or bias.

STEP FOUR: Diagram and Assemble the Ithkuil words

At this point, we can create a sequential analytical diagram of the morphological structure of the sentence showing the stems, plus all their associated non-default morphological categories, thus:

When putting these elements together to render the final Ithkuil sentence, I take the option to reorder the suffixes and reverse some of them into their Consonant+Vowl forms for the sake of phonetic euphony.

Here then, is the end result, accompanied by the most natural sounding literal translation into English:

All families are happy in the same way, while being unhappy in their own way.

STEP FIVE: Converting to Ithkuil Script


( 5 comments — Leave a comment )
Mar. 30th, 2009 07:46 am (UTC)
Чёрт!!!! это ведь про http://odzillazine.jot.com/ ? :(
Mar. 30th, 2009 07:56 am (UTC)
Про него самого :(
Mar. 30th, 2009 09:40 am (UTC)
Безвозвратно, говорите? А это не оно ли?
Mar. 31st, 2009 01:12 am (UTC)
Mar. 31st, 2009 03:32 am (UTC)
я всегда считал что надо APL толкать с обычную речь :)
( 5 comments — Leave a comment )