Ah, Nick, each choice is buggered no matter which way one turns. That eternal torment seems to be the crucible that makes it impossible for language ever to rest.
The voiced-syllable-initial-th at the beginning of this, that, and the other (_not_ "thing") is a characteristic of the function words of English, inherited from a time when there was a bound prefix with a vowel, which led to the voicing of the then-intervocalic-th. Psycholinguists (such as our colleague Morten Christiansen) tell us that children make remarkable use of such clues during the learning process, because these allow them to make category distinctions of the function words from mere nouns, just based on sound. The "ter" removes the cue. I would expect Gloria not to mind any such collateral damage. Of course, I have been beaten into submission enough times by linguists to know that Their Correctness is an axiom, and all views of the world must be altered so as to lead to it (or else), but I actually like leaving words within the systems where they did most of their formation. Thus, if one understands that the "Coel" in coelocanth should be said "ko-eel", then one realizes that this is a fish named for its hollow spines, from Greek koilo, which in romance devoiced the k to eventually become "hollow", and it is a latin C (followed by o -hence hardened) to refer to a greek k, because scientists use latin spellings for greek roots. Right? If one wants true carnage, there is no better shop of horrors than stressing. How really should one say "hydrogenase", in a language that accents the last syllable and then trys to map forward from the end of the word in iambs? Given that this enzyme catalyzes reactions with "hydro-gen", which was a nice word, accenting both the water and the making. Of course, chemists _hate_ you if you say hy"-droh-gen-ase' with its two reduced and un-accented syllables, since everybody knows it _has_ to be hy-drah"-gen-ase'. (And it does feel wrong, though I persist in it anyway.) I think this is the reason we are seeing an interesting phoneme shift in English just now, which drives Murray crazy. This is the younger people who say "processeez". But it is clear why. The 'cess' in Process, though secondary, is still stressed, and it is very hard to follow such a stressed syllable with an "ez" in which the vowel is reduced to a neutral, and both syllables end in similar fricatives that, to make matters worse, differ in voicing. If one makes the "ez" into "eez", then one can de-stress "cess", to get prah"-cess-eez', which follows the iambic pattern, separates the fricatives, and gives one a long vowel in which to restart the voicing. I haven't tried that explanation on Murray to see if it makes this any less painful for him. Eric ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
