A claim has turned up on the Linguistics SE that двумястами and other such forms have two stressed syllables. I disagree; -мя-
seems pre-emphatic to me here, and Wiktionary is with me on this.
However, one thing the two-stresses theory may have going for it is that there are no other (standard) examples of pre-emphatic /a/
failing to reduce to [ɪ]
after a soft consonant.
And I can think of a test, but I'm not sure its result for me would be the same for most people.
Here's the test. With words that have a true secondary stress, the following cadence is possible at the end of a sentence:
...вагòноремóнтный завóд. ___ \_____ \_________ \_
Now I can only imagine saying something like двумястами годами with two pitch drops:
...двумястами годами. _______ \______ \__
Yet I'm not entirely positive that no-one ever would say it with three drops:
...двумястами годами. ____ \__ \______ \__
— which would indicate that for this speaker, there is a secondary stress on -мя-
.
Does this occur? Are there any more arguments for either side?