A Few Observations on Minimalist Languages

[janKipo]

Not really. tp is all about dealing with a limited vocabulary that can function in any role and is vague enough to be used in a variety of context to convey appropriately precise meaning in that context. Ygyde seems to want to create new words for each situation, to make them as precise as possible with a minimum of context, and to assign each a restricted syntactic role. Given these very different blueprints, there seems little to say to one another. I will clue you in, however, that the quest for precision is always a doomed one, beginning with the unambiguity of the primes.

[oligo]

How transparent are these compounds?...

I am a newbie trying to understand Toki Pona. (I am much more familiar with Ygyde and other oligosynthetic languages.) As far as I can tell, there is fundamental difference between Toki Pona and Ygyde. Every word of Toki Pona has about a dozen different meanings. Ygyde is different because there is only one valid meaning of every of its 90 morphemes. Another fundamental difference is grammar. I have IQ of 125, but I can't understand Toki Pona's grammar. On the other hand, you can speak Ygyde without knowing its grammar. You will sound funny, but you will be understood. Ygyde in Ygyde means "noun simple language." I am sure that Ygyde is more simple than Toki Pona.

Compounding is just one of derivational strategies and compounding isn't all that much of a panacea, nor is POS markers.

Nothing is a panacea, but compounding and POS markers are useful.

Does mouth-water mean delicious (as it obviously does in English) or does it mean spit (and I can't remember what conlang uses that)...

In Ygyde delicious = amamaby = "adjective good good food"

[oligo]

Ygyde seems to want to create new words for each situation, to make them as precise as possible with a minimum of context, and to assign each a restricted syntactic role.

True.

I will clue you in, however, that the quest for precision is always a doomed one...

I agree. We will never reach precision or perfection, but we should strive for both.

[janMato]

How transparent are these compounds?...

I am a newbie trying to understand Toki Pona. (I am much more familiar with Ygyde and other oligosynthetic languages.)

Best way to understand it is to attempt to write and read some of it.

As far as I can tell, there is fundamental difference between Toki Pona and Ygyde. Every word of Toki Pona has about a dozen different meanings. Ygyde is different because there is only one valid meaning of every of its 90 morphemes.

I'm skeptical. Loglan may (or may not) have exactly one parse tree for each sentence (i.e. is not ambiguous), but "Would you like to come up for some coffee?" (said after a pleasant date) can mean at least two things. How does one say that in Ygyde in a way that achieves the desired results? My most recent attempt to write a conlang in 30 days was derailed because it was so easy to generate the words, yet so laborious to assign suitable meaning to them, even if they were compounds or stems and prefixes or what have you.

Another fundamental difference is grammar. I have IQ of 125, but I can't understand Toki Pona's grammar. On the other hand, you can speak Ygyde without knowing its grammar.

No, you'll be speaking pidgin. And I mean it in the sense of, you're speaking your L1 with the vocabulary of L2. (I know some people think that pidgin and creole are reserved for phenomena that happen on slave plantations)

Anyhow, there is nothing stopping you from using your innate ability to learn languages just by reading tp text. Your brain is smart and can extract most the the grammar you need from reading lots of text. (after memorizing as many phrases as you can of course, vocabulary counts.)

You will sound funny, but you will be understood. Ygyde in Ygyde means "noun simple language." I am sure that Ygyde is more simple than Toki Pona.

Any relex is simple. [I've recently decided relexes don't bother me, they're kind of like English condialects.]

Compounding is just one of derivational strategies and compounding isn't all that much of a panacea, nor is POS markers.

Nothing is a panacea, but compounding and POS markers are useful.

In tp, you force POS readings by position, although not always 100% of the time. It follows a noun? it is a adjective. It follows a pi? It can be a noun, but not always. It follows li? If it isn't a predicate, it is a verb. It's in the subject slot? It starts with a noun. It directly follows a preposition (other than lon)? It is a noun. lon phrases often are of the sort mi lape lon sewi kasi suli. I sleep on top the tree. The lon sewi is parseable as either 2 preps in a row or a prep + a noun.

Does mouth-water mean delicious (as it obviously does in English) or does it mean spit (and I can't remember what conlang uses that)...

In Ygyde delicious = amamaby = "adjective good good food"

So what is spit in Ygyde?

[oligo]

So what is spit in Ygyde?

to spit = ibagupa = "verb top liquid anatomical part of a multicellular animal"
i = verb
ba = top
gu = liquid
pa = anatomical part of a multicellular animal

You can quickly derive the noun spit/saliva from the verb: saliva = obagupa = "noun top liquid anatomical..."

You can find all these words in the dictionary: http://www.ygyde.neostrada.pl/ygyded.htm

[janMato]

You can quickly derive the noun spit/saliva from the verb: saliva = obagupa = "noun top liquid anatomical..."

You can find all these words in the dictionary: http://www.ygyde.neostrada.pl/ygyded.htm

Hmm. Knowing the one, I'd have never guessed the other. I would have guessed water on the brain, hydrocephalia. Non-opaque compound phrases are hard to make and at the end of the day the best you can hope for is something that is transparent to some people or at least pretty sounding. ljóspera, Icelandic for "light-pear" for light bulb would be a good example, that if it isn't transparent, at least it is pretty.

[janKipo]

These comments are pretty much talking past one another, largely because tp and Ygyde are so different in almost every respect. To try and compare them, particularly evaluatively, is, thus, an almost impossible task. The most we can do is compare their parts and see how they do various things.
I haven't looked at Ygyde lately, so don't remember the details. But, in broad outlines IIRC, there are 90 (15 consonants times 6 vowels?) morphemes, each a (C)V syllable, each of which has two distinct meanings, one or the other of which is uniquely selected by other morphemes in a given word (defined morphologically, by counts and initials). The words can be arranged fairly freely because the POS shown in the initial morpheme of each restricts the way that various pieces can interact, though there are preferred and excluded patterns. The meanings of the morphemes is fairly precise (narrow semantic scope -- though how narrow or not is not necessarily specified by the usual translations or discussions I have seen). The way the words are constructed from these atoms limits the ways in which the component morphemes may be related semantically, leaving some cases at least fairly open. As a result, while it is easy to create new words, it is sometimes difficult to have them understood without explanation -- although something close may be arrived at from contexts. But ideally context should be dispensable and each word clear from its own construction.

tp has five vowels and 10 consonants (counting, as I did above,the null case). It forms syllables (C)V(n), though some are disallowed for various reasons, so the total will be around 90 syllables. None of them has a meaning on its own per se, though some of them are words and get meaning at that level. There are about 120 words (some flux at the moment) a handful of which serve purely grammatical purposes, so a much smaller stock than Ygyde to begins with (as the other features so far are smaller). The meanings of these words are very broad, covering a large number of English words each, including some not obviously related to each other (this is not ambiguity or even vagueness except as viewed from some other language, in tp each word is as exact as can be). Thus 'toki', for example, covers just about everything in communication: one medium (as well as the general notion of media), the message. the language in which the message is framed and so on (we are constantly discovering new English notions already embedded in tp words). The aim of the language is not to be precise in every context (including null), but to be adequate to convey the intended message in any given context. Much of the apparent complexity of tp is due to working in the null context; once away from that, conveying the desired message becomes much simpler (see Mato's wine example as a nice case -- though you have to fill in the context imaginatively). Aside from the function words, no tp word has a part of speech in an exclusive sense: it has a home POS, whose meaning explains the meaning of that word in other than home constructions -- but it can play any role in any construction (in theory; in practice a few roles are restricted: modals and prepositions and maybe some others). Because of this flexibility, the structures and roles must be tightly defined, so there is only one surface pattern, SVO, with its parts clearly delineated and the heart of each of these is a NA construction, left grouping with possible inclusion of already formed NA strings. The result is a grammar that can be put into a very few lines (see pckipo.blogspot.com. first page, for an example). Learning to work with it naturally takes a little practice, but obviously not as much as a language with a more complex grammar -- unless it is essentially a relex of L1.

So you see that there is virtually nothing similar between tp and Ygyde, so no grounds for evaluative comparisons.

[oligo]

So you see that there is virtually nothing similar between tp and Ygyde, so no grounds for evaluative comparisons.

Both languages rely heavily on compound words/phrazes and they struggle with ambiguity. It may seem that bigger morpheme set would be desirable, but the additional morphemes would not be used frequently. Furthermore, the bigger the morpheme set, the longer the morphemes. Nearly all speakers hate long words, so the morphemes must be short. Ygyde has 4x90 predefined words (time-words, colors, numbers, and prepositions) in addition to its 2x90 morphemes. All the morphemes have consonant-vowel structure.

It is not possible to construct a conlang having perfectly self-explanatory compound words. For example, it is not possible to make a compound word meaning 'piano' that is self explanatory. It would take a short article to explain what the piano is and how it differs from the harpsichord. Nontheless, the compound words are useful because they are easy to memorize.

Yes, Standard Ygyde has 15 consonants and 6 vowels. There is also Long Ygyde, which has 3 vowels, 8 consonants, and 3-letter-long morphemes.

[janKipo]

I am relieved to hear that Ygyde has gotten away from it original claim of precision and moved back to "good enough for government work." I do gather, however, that the aim is still to work in a relatively context-free environment, more suitable for writing. tp started with the contextual definitions and has stuck to that, despite being also in a written context for the most part, even though it is a language natural for face-to-face spoken communication. So, for now, it has essentially the same problems as Ygyde; and, I assume, that, when spoken usage gets going, the usage of complex expressions will diminish in both.

[oligo]

What is the purpose of Toki Pona?

I used to believe that its purpose is to express emotions. I searched in the dictionary (http://tokipona.net/tp/ClassicWordList.aspx) for words describing emotions: mourning, painful, passive, patient, nausea, discomfort, unconvenient, trust, immoral, noble, pleasure. I have found none, but all of them were in the Ygyde dictionary.