[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >
The article on reading TP is on my blog. http://www.suburbandestiny.com/?p=845 Dropping a pi or e is a disaster for readability, except in a few cases such as if the DO or post-pi phrase is a two word pronoun, like mi mute, sina mute, because no one shuffles mi mute even when they could. Dropping the head of a proper noun phrase is not a problem at all for readability. Likewise, adding li after mi and sina. Dropping the ala X from the X ala X pattern in yes no questions doesn't affect readability at all. The complete list is longish & tedious unless you're interested in TP. Again, the only pattern I see is that when items are marked multiple times, it doesn't hurt to drop the extra marks (double marked questions, double marked pronoun possessor phrases). And back to the original question, these feature also seem like the least stable syntactic features, most likely to be lost over time. Matthew Martin --- In engelang@yahoogroups.com, MorphemeAddict <lytlesw@...> wrote: > > On Fri, Apr 15, 2011 at 4:17 PM, MatthewDeanMartin <matt@... > > wrote: > > > > > > > re: language learned from books get "reset" > > Ah, good observation. I'm reading "The Last Lingua Franca" and this > > phenomena is accredited for why lingua francas seem to be more stable, > > especially where they are being used outside of the range where that > > language is a mother tongue. [Taglish certainly looks like a counter > > example, I'm not trying to say this is a law of physic or anything] > > > > re: noisy room & redundancy > > Hmm. Maybe these are all the same problem. The problem of language change > > over time is not unlike language change as it goes through a noisy telephone > > connection. > > > As you've already mentioned, redundancy is one way to minimize the noise > problem. I don't like redundancy in my conlang projects, but they haven't > been thoroughly tested. > > > > > The ultimate in engineered protocols would be a sort of check sum. I've > > been trying to think up a checksum that could be applied to language, but I > > haven't though of one. > > > > I also have been noticing how certain mistakes in toki pona are > > recoverable, i.e. if you make the mistake the meaning isn't lost, but other > > mistakes, if you make the mistake, the sentence is gibberish. But I don't > > have thing to generalize yet. > > > I'd like to hear what you find about this, i.e., what kinds of errors are > major, and which are minor. > > > > > As Esperanto example would be eroded "n", which can be dropped without too > > much loss because people use SVO ordering in addition to accusative marking. > > > This is why in Ido the accusative '-n' was dropped unless the word order is > unusual. For Ido, it's a trade-off: either use strict (SVO) word order or > mark the object. > My projects all use strict (VSO) word order, so there are no cases. > > stevo >