Glo[ʔ]alisation and the Glo[b]alisation of Corpora

One of the main distinctions between British English and U.S. English is the varying pronunciation of the voiceless, dental stop [t] between vowels. While in many dialects of British English, this tends to be realised as a glottal stop ([ʔ]) so butter is pronounced "buh'uh", speakers of American English generally pronounced intervocalic [t] as a [d]-like sound ("budder"). Those American visitors to the Sandwich Shop of Dreams never could work out what I wanted to spread on their sandwich when I asked whether they would like "butter or mayonnaise"; their best guess was that I was saying mustard.

However, on a couple of podcasts recently, I have noticed this spreading across the pond. Tyler Cowen, of Marginal Revolution, for example, glottalises his [t] as evidenced in this EconTalk podcast. Similarly, on today's NY Times Science Times podcast, Frederic Riccardi, a "health insurance counselor" was doing the same thing, from about five minutes in (e.g. "Manha[ʔ]an"). The latter is Spanish-English bilingual so perhaps this played some role, although I can't think of any comparable processes in Spanish.

Neither speaker does this all of the time but I lack the time at the moment to work out the phonetic contexts in which they glottalise their [t]. Nor can I think of any reasons why this change should have come about; that said, I have previously noted that explanations of linguistic change almost always tend to be post hoc and aren't predictive.

Google Scholar doesn't seem to have found any studies of this; however, it occurred to be that the spread of podcasting has created whole new corpora of data for linguists to study. No more Observer's Paradox-ridden participant interview! No more Labovian department store harassment! Of course, I don't know where Cowen or Riccardi grew up or any of the other information about them that may be helpful when determining the scope of a language change (if they lived abroad for a period of time, what other languages they speak, etc.). Nor am I a good enough dialectologist to say confidently, on the basis of their accents alone, where they are from (although Cowen's vowel raising (about > [uh-'bʌwt] (with the vowel of put followed by a [w])), make me think he is either Canadian or from New England.