wolfson_73_veryvery.wav  2 "very"s

¥ first "very" 
Two possible decision tree paths
first answer:
Is there a boundary-related tone? no. 
Is there durational lengthening? yes. 
Is it appropriate to an iP boundary?
N: "no", therefore 2 --not long enough for a 3, so not 3m 
second possible answer:
is there a boundary-related tone?  yes: maybe a flat tone
Is there durational lengthening? yes.
iP boundary? not sure: could be 3 with flat, or 3m
Discussion points: 
¥ Should we look on either side of boundary? There are theoretical grounds for 2, i.e. there is downstep across these words.
¥ very strong rhythmic cues: strong parallel in rhythmic structure suggests same break type: short pause between "very"s seems rhythmically motivated (compare hypothetically "varying, varying, firmly established..." would not have pauses) This argument suggests 2 breaks  
RESPONDENT A #1: I agree.  
ANS #2: GreatÑbut could you tell us whether itÕs the 3- or the 2 analysis that makes more sense to you? (If itÕs the 2, do you agree that we should keep 2 breaks for examples like this (which would otherwise appear to have downstep across iPs?)
RESPONDENT A #2: What makes sense to me is that the boundary after 'very' is not 'ip' (i.e., 3 with phrase accent). I agree that there's strong rhythmic cues and the boundary is larger than 1. But not sure if the boundary is 2 or 3. I'd like to see more examples of '2' to answer the question. If there's no 2, the boundary after 'very' is a good example of '3m' becausee there's downstep across 'very's and the boundary after 'very' is long for BI 1.    This example is similar to 'Iraqi' example in ToBI - the rhythmic structure with downstep connects a sequence of lengthened words.  So, it would be good to see an example of '2' from a different context. 
RESPONDENT B #1: We hear this utterance as containing a series of downstepped High peaks. If we accept that downstep can only occur within an ip, then we must assign a break index of 1 (or 2) at each word juncture. But we agree that the juncture between words in Òvery Ð very -  firmly - established Ð areaÓ is greater than a typical level 1 break. If we label according to our sense of disjuncture, then we would assign a level 3 break at each of the junctures in this portion of the phrase, but then we would have no way to capture the downstepping F0 contour. This is a genuine mismatch! Forced to assign a label, we would assign L+H* on both of the ÒveryÓ words and on ÒfirmlyÓ, and each of these three words is followed by a L- / 3. ÒestablishedÓ has a H* (no downstep) followed by level 1 break, and ÒareaÓ has H* followed by L-L% / level 4. As for our L+H* accents on ÒveryÓ ÒveryÓ and ÒfirmlyÓ, we think we need these to account for the late peak.
RESPONDENT C #1: Yes, this is a good example of mismatch. Nice downstepping across ÒveryÓÕs. I would go with 3m, although it would be nice to talk more about 2. I agree itÕs not a huge disjuncture, but if you can have a 3 in the ToBI example ÒI (3) means insertÓ, then I think you can have a 3 here. IÕm not sure I hear ÒfirmlyÓ as downstepped. I would opt for H*/!H* on these, instead of the bitonal.