Message-Id: <199709191137.GAA28616@locke.ccil.org> Reply-To: HACKER G N Date: Fri Sep 19 06:37:18 1997 Sender: Lojban list From: HACKER G N Subject: Re: Morphology X-To: Don Wiggins X-cc: Lojban List To: John Cowan In-Reply-To: <0EGR0035Z2E74Q@newcastle.edu.au> X-Mozilla-Status: 0011 Content-Length: 1067 X-From-Space-Date: Fri Sep 19 06:37:18 1997 X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU On Fri, 19 Sep 1997, Don Wiggins wrote: > Could someone explain to me how the following stream should be resolved: > > lo-re-spa-ji-pci > > This could be either "lo re spajipci" (the two plant-chickens) or > "lo respa jipci" (the reptile chicken). I know it's not a great example, but > how is a CV cmavo prevent from falling off a gismu that internally has a > permissible initial consonant pair. > Another example I've heard is "ca dzukla ma" versus "cadzu klama". Even I wondered at first how to disambiguate the speech stream "cabdei cerni" from "cabdeicerni". But all of these examples can be effectively disambiguated in the speech stream by listening to the primary stress in the words. "lo re spajipci" is "lo re spaJIPci", whereas "lo respa jipci" is "lo RESpa JIPci". "ca dzukla ma" is "ca DZUkla ma", whereas "cadzu klama" is "CADzu KLAma". Finally, "cabdei cerni" is "CABdei CERni", as opposed to "cabdeicerni", which is "cabdeiCERni". Which is to say that there's more to disambiguating Lojban sounds than the consonant clusters alone. :) Geoff