So I've created a very functional language with only 32 words,

[Daniel]

I would like to point out that I never said that just using the name of the animal was wrong, just that it would be less clear to some people.

It is unclear if you don't know the word. It is clear if you do. If you always circumlocute rather than name, you will never be entirely clear although you will be somewhat clear. If someone doesn't know the word "tree", then saying "tall green things" is better than nothing, but simply teaching them the word "tree" is much more effective in the end.

The thing is, we learn words for things we talk about, so there is no inefficiency. Memorizing the whole dictionary would probably be a waste of time, but learning words for things we talk about is useful. If you don't know what a "gerenuk" is, then you don't need a word for it. But once you know what it is, having the word is useful.

Theories about cognition and categorization would suggest that having words is useful. Some have even claimed that knowing a word as a label allows us to create a cognitive category. I'm not sure I'd go that far, but it's certainly useful and helps us understand specific concepts clearly.

As for the circumlocution being optional, that again reminds me of Ithkuil as I linked above. So the idea is not new at all, and you might find that comparison relevant. Maybe Ithkuil is what you're looking for, after all.

But I'm not sure how useful it is to have this sort of highly analytical speech, even optionally. In English it is very easy to say "tall green things" instead of "trees", if we want to. We circumlocute (and describe) all the time. It isn't quite as short/systematic as what you have suggested, but that also isn't necessary, because we replace indirect references to common concepts with specific words. So we don't say "tall green things" specifically because we would prefer the more efficient "trees" when we talk about them a lot.

[panini]

There is a fundamental problem that you are saying things that, as a linguist, I cannot make sense of, because you're using words that have specific meanings in linguistics (though even sometimes outside of linguistics), and you are apparently using them in a technical way that we can't understand. This is what makes it extremely difficult for linguists to explain to others what we do (sociologists have the same problem), since others don't understand our words. "Sophistication" is not a technical linguistic term, so I can't understand what you're saying about level of sophistication without an explanation of the concepts and its relevance to grammar. A first step is to say that the concept relates to the number of elements in a syllable (we would never talk of "a" being a "less sophisticated" syllable than "flaf", but now I see that you've redefined "sophistication"). You say that "level of sophistication is about "how many parts it has", but your explanation is not consistent with that, because you also say that the first "level" is the vowel and the second is the initial consonant, and yet you are talking about single elements in both cases. So I conjecture that you don't mean "how many parts", you mean "which part", where the nucleus is most basic, then onset, then coda. And finally you talk of "the vowel modifier" – I have no idea what that is, but if it means "an additional vowel", then it's not even clear that you are using "syllable" in the standard linguistic way, and perhaps you really mean "foot".

I suggest that when you present the structure of this language, you should start with a factual statement of what the directly perceptible elements are – the things that people can directly experience. That would either be the use of letters, or the use of sounds. People cannot directly experience a meaning or a grand design plan for a language. They can experience a sound. Which brings us to your question:

As for the second bit, it was a little unclear to me what exactly you meant when you started talking about all those other notations, so I'll just do what else you asked, and go into full detail how a sentence would be derived.

I'm going to assume that even if you haven't thought of language in such terms, you can at least understand the mathematical underpinnings of natural language. At the most basic level, any utterance is a string of units – "sounds". The set of sounds employed is determined by the language. Hawaiian has 13 sounds – /p t ʔ m n w h l a e i o u/. The word [honu] 'turtle' is one such word, [hoku] 'night of the full moon' is another. Any word of Hawaiian can be described as some permutation of the 13 sounds, so there are mathematically 13 possible 1-sound words and 169 possible 2-sound words. As it happens, there are also rules defining possible words where "p" is not actually a possible word, nor is "um", so you get a subset of the possible sound permutations being words. There there are theoretically possible sound combinations that simply are not words of the language (I don't know Hawaiian so I can't make up an example).

Natural language is organized around the notion of economy of elements, so that if "p" is a sound in the language, you use "p" in a lot of words. There may be exceptions (e.g. "p" exists in some dialects of Arabic in only a handful of words; ʍ in English is used very infrequently, and not at all by some speakers), but the general pattern is that if you have a sound r, you can use it widely consistent with general distributional rules.

In your system, you seem to have 30 sounds: [a b d ð ə ɛ f g h i j l m n o p r ʀ ɾ s ʃ t u v w z ʒ γ θ χ] (I'm interpolating to IPA based on typical SAMPA-like spellings and your descriptions) – 6 vowels and 24 consonants. If we define two simple rule for possible syllables (every syllable must have a vowel; syllables may have at most CV, which is more restrictive than your actual language), then there are 150 possible one-syllable words given this inventory. But you only have 32 words – the use of sounds in the language is inefficient, and you could eliminate many of those sounds. In addition, it is clear that your language allows much more complicated syllables than just CV: for every additional consonant that you can add, multiply the number by 24. That is, the fewer limits there are on permutation of elements, the fewer elements are needed. If you want to have 32 syllables, you only need 5 positions and 2 elements; or, you can trade in positions for elements (if you have 3 elements you only need 4 positions, with 4 elements you need just 3 positions, etc). I was offering you a way to reduce the number of basic elements, since you only want 2^5 syllables.

Now, the fundamental question for your language project is, why would it be desirable to have a language with only 32 "words"? The unspoken answer is that, given a system of rules, you can express an infinity of concepts using very few resources – the language would be very efficient. The problem is that Truespeak is very inefficient: it uses way too many basic elements, given the basic rules of combination.

This point is aimed at the "why" issue – why would you build a language like this? I still do not understand what the rules of the grammar actually are, especially since you severely restrict meaning-units to just 32 things. This is completely contrary to the nature of natural language, where any functional organization of existence can be efficiently expressed as a single conventional sign, made up of a few conventional units.

[Aldebaran]

There is a fundamental problem that you are saying things that, as a linguist, I cannot make sense of, because you're using words that have specific meanings in linguistics (though even sometimes outside of linguistics), and you are apparently using them in a technical way that we can't understand. This is what makes it extremely difficult for linguists to explain to others what we do (sociologists have the same problem), since others don't understand our words. "Sophistication" is not a technical linguistic term, so I can't understand what you're saying about level of sophistication without an explanation of the concepts and its relevance to grammar. A first step is to say that the concept relates to the number of elements in a syllable (we would never talk of "a" being a "less sophisticated" syllable than "flaf", but now I see that you've redefined "sophistication"). You say that "level of sophistication is about "how many parts it has", but your explanation is not consistent with that, because you also say that the first "level" is the vowel and the second is the initial consonant, and yet you are talking about single elements in both cases. So I conjecture that you don't mean "how many parts", you mean "which part", where the nucleus is most basic, then onset, then coda. And finally you talk of "the vowel modifier" – I have no idea what that is, but if it means "an additional vowel", then it's not even clear that you are using "syllable" in the standard linguistic way, and perhaps you really mean "foot".

I suggest that when you present the structure of this language, you should start with a factual statement of what the directly perceptible elements are – the things that people can directly experience. That would either be the use of letters, or the use of sounds. People cannot directly experience a meaning or a grand design plan for a language. They can experience a sound. Which brings us to your question:

As for the second bit, it was a little unclear to me what exactly you meant when you started talking about all those other notations, so I'll just do what else you asked, and go into full detail how a sentence would be derived.

I'm going to assume that even if you haven't thought of language in such terms, you can at least understand the mathematical underpinnings of natural language. At the most basic level, any utterance is a string of units – "sounds". The set of sounds employed is determined by the language. Hawaiian has 13 sounds – /p t ʔ m n w h l a e i o u/. The word [honu] 'turtle' is one such word, [hoku] 'night of the full moon' is another. Any word of Hawaiian can be described as some permutation of the 13 sounds, so there are mathematically 13 possible 1-sound words and 169 possible 2-sound words. As it happens, there are also rules defining possible words where "p" is not actually a possible word, nor is "um", so you get a subset of the possible sound permutations being words. There there are theoretically possible sound combinations that simply are not words of the language (I don't know Hawaiian so I can't make up an example).

Natural language is organized around the notion of economy of elements, so that if "p" is a sound in the language, you use "p" in a lot of words. There may be exceptions (e.g. "p" exists in some dialects of Arabic in only a handful of words; ʍ in English is used very infrequently, and not at all by some speakers), but the general pattern is that if you have a sound r, you can use it widely consistent with general distributional rules.

In your system, you seem to have 30 sounds: [a b d ð ə ɛ f g h i j l m n o p r ʀ ɾ s ʃ t u v w z ʒ γ θ χ] (I'm interpolating to IPA based on typical SAMPA-like spellings and your descriptions) – 6 vowels and 24 consonants. If we define two simple rule for possible syllables (every syllable must have a vowel; syllables may have at most CV, which is more restrictive than your actual language), then there are 150 possible one-syllable words given this inventory. But you only have 32 words – the use of sounds in the language is inefficient, and you could eliminate many of those sounds. In addition, it is clear that your language allows much more complicated syllables than just CV: for every additional consonant that you can add, multiply the number by 24. That is, the fewer limits there are on permutation of elements, the fewer elements are needed. If you want to have 32 syllables, you only need 5 positions and 2 elements; or, you can trade in positions for elements (if you have 3 elements you only need 4 positions, with 4 elements you need just 3 positions, etc). I was offering you a way to reduce the number of basic elements, since you only want 2^5 syllables.

Now, the fundamental question for your language project is, why would it be desirable to have a language with only 32 "words"? The unspoken answer is that, given a system of rules, you can express an infinity of concepts using very few resources – the language would be very efficient. The problem is that Truespeak is very inefficient: it uses way too many basic elements, given the basic rules of combination.

This point is aimed at the "why" issue – why would you build a language like this? I still do not understand what the rules of the grammar actually are, especially since you severely restrict meaning-units to just 32 things. This is completely contrary to the nature of natural language, where any functional organization of existence can be efficiently expressed as a single conventional sign, made up of a few conventional units.

Alright, you've brought up a lot of good points throughout your response and I thank you for taking your time to do so, I understand some issues with the way I explained the language (I'm not always good at explaining these sort of things), so let me address a few points in detail:

First, about the why. This is a language created for philosophically-minded fictional entities of God-like properties, and the main purpose of the language is to create a more immersive and "realistic" world, that is, to show us that this race of very abstract and unhumanly beings have a language that fits their nature. I've created many other languages (much more humanly and efficient at describing human things), but just to clarify, I didn't create this language to be efficient for us. I know it's inefficient for us.

Now, just a little background about the language itself. If you don't want to read the story relating to this language then by all means, skip this paragraph. In a world made of nothing, a light forms. In this light, a universe's worth of reality begins to form. We shall call it Bire. This being, the all that it is, chooses to split itself into, well, itself and the rest of the universe. Long story short, it gets bored of being alone as God, and births two sons. They do the same, repeating again and again until a family forms. Inspired by this idea of "two" and the fundamentality it has in his own history and future, Bire creates a language perfect for describing the things he loves and knows true, Truespeak. Now, his many sons didn't necessarily agree with him; and though some rebelled, others to remain silent, they all did one thing in common: create their own dialects to the language of their Forefather.

Now, these other dialects (if you've read the above paragraph) are different in many ways from the first dialect (Birespeak), but fundamentally are the same. So far I've explored most of two other dialects (Dezspeak and Lithspeak), and intend (and have intended for a few weeks now) to make posts detailing them as well. They are more geared towards humanly communication rather than philosophical banter.

As it stands, most of them have more than two (I'm using your words here) elements (anything from 3 to 7), and a fitting number of positions (one dialect, canonically, even has as many as 1,953,125 particles! (I'm PROBABLY not going to be writing about it)).

So that's the first thing.

The second thing I want to say regards the idea of syllables, consonants, et cetera. I see I've misused terms from linguistics in a different context (apart from a few like sophistication, which I know are not linguistic terms), so I'll try to explain in an even more fundamental manner.

A syllable in Truespeak is, to put it simple, a phonetic movement. As long as I can fit all the sounds into one movement of the mouth, it's a syllable. Now, this means that even something like bzrowmz (which contains 6 sounds) is a singular syllable, since it can be said in one movement of the mouth.

Now you said that truespeak doesn't utilise every combination of sounds within it, but I'd disagree (on the most part). Simply put, every sound can be used by itself, still have meaning, and still be pronounceable regardless of which other sounds you put around it. The only matter is that by having a series of sounds like bpbtptbp, even though it still means something, it splits into syllables in a very unpretty manner, which is why I'd usually use bawb'pyib to express the same particles.

Regarding the matter of the sounds themselves, Truespeak actually has much more than 30 sounds as you've analised, just at higher levels of sophistication (see my other topic on the matter, discussing the fifth level). Even at four levels, there are still more, since it seems to me that you've left out two vowel sounds (as you've said that there are six while there are in fact eight) and the vocalised x (gh).

As for vowel modifiers, the only explanation I can think of which is any simpler than anything before is this:

Without the vowel modifier -y: The English word "Bat".
With the vowel modifier -y: The English word "Bite".

Going by the definition of syllalbe I've explained above, both of these are one-syllable words.


I hope all this helps.

[Aldebaran]

I would like to point out that I never said that just using the name of the animal was wrong, just that it would be less clear to some people.

It is unclear if you don't know the word. It is clear if you do. If you always circumlocute rather than name, you will never be entirely clear although you will be somewhat clear. If someone doesn't know the word "tree", then saying "tall green things" is better than nothing, but simply teaching them the word "tree" is much more effective in the end.

The thing is, we learn words for things we talk about, so there is no inefficiency. Memorizing the whole dictionary would probably be a waste of time, but learning words for things we talk about is useful. If you don't know what a "gerenuk" is, then you don't need a word for it. But once you know what it is, having the word is useful.

Theories about cognition and categorization would suggest that having words is useful. Some have even claimed that knowing a word as a label allows us to create a cognitive category. I'm not sure I'd go that far, but it's certainly useful and helps us understand specific concepts clearly.

As for the circumlocution being optional, that again reminds me of Ithkuil as I linked above. So the idea is not new at all, and you might find that comparison relevant. Maybe Ithkuil is what you're looking for, after all.

But I'm not sure how useful it is to have this sort of highly analytical speech, even optionally. In English it is very easy to say "tall green things" instead of "trees", if we want to. We circumlocute (and describe) all the time. It isn't quite as short/systematic as what you have suggested, but that also isn't necessary, because we replace indirect references to common concepts with specific words. So we don't say "tall green things" specifically because we would prefer the more efficient "trees" when we talk about them a lot.

The post above this one (my reply to Panini) explains pretty well some points I was trying to get across. It could explain a few things.

Now, we're really just starting to repeat ourselves here. I agree with everything you say, it's just that the matter you're bringing here isn't really something that Truespeak (this dialect, at least (reference prior post)) was designed to do efficiently, as it was was designed far better for three other things, banter, philosophy and philosophical banter. If you were to try and form words of those sorts using Truespeak, you'd see where the true power of the language lies. In a way, we can mostly rely on 01011 for specific animals and non-abstract things.

Thanks as always.

[Daniel]

If this is fictional, then it seems interesting and creative. Fiction should still resemble reality, though, to some extent, so make sure it works out in that sense. But if the goal is to create a language of beings that seem somewhat disconnected from our reality or perhaps do not categorize the natural world like we do, then that might work out well. If these beings are god-like, though, I would wonder why their language appears less direct/specific than ours, or why they would need to be efficient if they are omniscient/omnipotent, etc. But since we're moving away from the reality of human language, my objections are indeed less relevant.

[Aldebaran]

If this is fictional, then it seems interesting and creative. Fiction should still resemble reality, though, to some extent, so make sure it works out in that sense. But if the goal is to create a language of beings that seem somewhat disconnected from our reality or perhaps do not categorize the natural world like we do, then that might work out well. If these beings are god-like, though, I would wonder why their language appears less direct/specific than ours, or why they would need to be efficient if they are omniscient/omnipotent, etc. But since we're moving away from the reality of human language, my objections are indeed less relevant.

I think you'll like the dialects of the other language I've made, Commonspeak, a lot more since it is a far more human language (used by fictional humans).

Even then, the other dialects of Truespeak are simpler and more straightforward (as I've already said), so I'm curious to see what you have to say on them once I make the posts regarding them. (There's also a hybrid which crosses between the many dialects, but I'll get to that one later).

As for why a God-like being would need an efficient language, the answer is that Bire appreciates beauty, and sees efficiency as beautiful. I mean, it also makes more sense to make an efficient language than an inefficient one.

Thanks as always.