> [mailto:[log in to unmask]] On Behalf Of Risto Kupsala > In Pandunia's grammar the rule is called predicate-object pairing. For > clarification, predicate is the generalization of verbs and adpositions, > and object is the generalization of direct and indirect objects and > oblique case. > > In Pandunia predicate+object = object+predicate and therefore the > adpositions can be used both as prepositions and as postpositions. The > only thing that matters is delimiting Po and oP pairs so that you know > which predicate and which object go together. > > In SVO (i.e. sPo) word order with prepositions a normal expression is: > s(Po)(Po)(Po)... > In SOV (i.e. soP) word order with postpositions a normal > expression is: > s(oP)(oP)(oP)... > > Such expressions delimit themselves, which is quite handy. > > Mixed word orders such as SVO with postpositions (expression type: > s(Po)(oP)(oP)...) could require some means for predicate-object pair > delimitation. Anyway, the implications for parsing are > obvious: it's not > going to be trivial. > > > Regardless, even inflected > > languages have some sort of preference for a certain word order. For > > example, it's technically possible to rearrange words in Russian, > > but the general convention is SVO, but SOV with pronoun objects as > > is common in Romance languages. Shifting the words around would > > still be grammatically correct, but may change the focus so there is > > a shift in the meaning of the whole sentence. In Romance languages, > > adjectives can change meanings depending upon whether they are > > placed before or after the noun they qualify, so it's not really > > just a stylistic choice. > > It's an anomaly, an idiomatic feature, which should have no > implications for world IAL planning. Probably not idiomatic so much as a legacy since the parent language, Latin, was an SOV language. Apparently a shift to SVO that hasn't yet fully materialized. > > In some poetic styles, you may see > > adjectives follow the nouns in English, but I wouldn't recommend > > trying to speak like that. > > That's because you have been conditioned to the standard English word > order(s). In a new language like Pandunia such conditioning > doesn't have to happen. Yes, I've been conditioned to English ordering. I've been studying a lot more Spanish lately. It doesn't bother me, nor present any problem if adjectives follow nouns. What is confusing is that a lot of adjectives can be placed either before or after the noun, but there is a change in meaning that goes along with the change in position. Imagine the confusion for speakers of languages like these who will hear the same words in different orders. > > And as much as some of use like to use > > Chinese as an example of how "easy" Chinese grammar is because it > > lacks inflections and such, it's stricter word order is responsible > > for making that happen. > > Chinese word order is not strict in this concern. Mandarin > has a feature called *topic prominence*, which you surely know. Right, but their word order is much stricter than that of European languages. Even Western languages have traces of inflection that aid in freeing up word order to some degree. Moving words to the beginning is still a form of "marking" the word whether it be for topical purpose, focus, or function within the sentence. A "free" word order is going to cause speaker A to do one thing, while speaker B interprets it according to his perception based upon a different model. The point of making a language is to provide a language model that everyone can follow knowing that others are doing the same, thus knowing what to expect and where to expect it. > .... > In Pandunia I have selected free word order and accepted the risk of more > difficult parsing (in some cases) as a consequence. That's the trade-off I > made. Well worth it. Not just more difficult parsing (which I don't see a big issue with), but actual misinterpretations. I wouldn't want to try and read a set of technical instructions in a language where I can't clearly tell how the words relate to each other.