"Used to to"

I had an interesting conversation today with William, my partner of four years. He shares my fondness for language, if not to the same extent. We got to talking about my family's use of Pidgin, among other things how it seems my parents communicate for the most part in Pidgin, while my sister and I generally speak English, though we speak to our parents in Pidgin. My sister says her sons don't speak Pidgin in general. This is further evidenced by William's assessment this past week that Kealaula, my oldest nephew, doesn't speak Pidgin at all. While I agree in part, I also maintain that I heard him use some idiomatic phrases from Pidgin ("stay" for the stative verb "to be," i.e. "Where you stay?" Where are you? and also "get" for the possession, as in "I get two dollars." I have two dollars). I was, however, hard-pressed to find solid and consistent examples of this.

The title for this post references such an idiomatic expression used in my family and church in Hilo, HI, if not everywhere in Hawai`i. English speakers use the idiom "used to" to mean "to be accustomed to" and the same goes for Pidgin. The difference lies in the intrusive superfluous "to." (This might be almost comparable to the extra "is" found in utterances such as "The problem is is that I don't care.") The usage is the same.

We were debating what could lead to the superfluous "to." I considered two approaches, the second more extensive than the first. The second probably concerns too many hoops to be credible, but I wanted to follow this thread to its logical end. What it boils down to is how Pidgin resolves consonant clusters. I originally thought mixed voicing ("use" has a voiced final consonant, the [z] sound, while--I thought--"-ed" has a voiceless consonant. In IPA, the underlying form might be transcribed /ju:z+d/) was the culprit, but the more I thought about it, the less I thought that "-ed" could be underlyingly voiceless; it's probably voiced. This was actually the bulk of our conversation/debate, where I tried to reason it out, unsuccessfully.

So what I actually proposed is vowel epenthesis, inserting a vowel to separate two like sounds. But this approach yields a form very different than what we might expect. Considering the underlying forms /ju:zd tu/, a complex (onset/coda) resolution could easily yield [juzdutu] (where the quality of the inserted vowel mirrors that of either the preceding or following vowel, for reasons as yet unknown to me). However, what actually surfaces in often more like [justutu], where the voicelessness of "to" regressively assimilates through the preceding consonants. This would require (in derivational terms) crucial rule order of regressive consonant voice agreement before vowel epenthesis.

All of this seems too complicated for one single nuance, none of the foregoing evidenced in other idioms or borrowings. Which brings me to the adhoc solution that a friend, Jeff, suggested, he having just finishing LING 200 and passing with flying colors:): When the idiom was "borrowed in" from English, nonnative speakers considered the "to" as part of the phrase "used to" and not part of the infinitival verb that followed, which would inevitably require the infinitival "to," as in "to be," "to have," etc. The end result would yield sentences like "I'm used to to live here"* Except that's not what we say. What we would say is "I'm used to to living here."

In light of this, are we back at square one? The intuition seems to be that the superfluous "to" retains something for Pidgin speakers that simple "used to" does not. But what?