Brief report: some phon-related talks at IPrA 2017 in Belfast (July 16-21)

Hello, hello! Hope you are all doing well. Life over here is hectic, as usual, and a few days ago I decided I had to take a holiday after a year (or more) of basically working non-stop, so I'm back at work now, with renewed energy, for a while, at least.

Two weeks ago I attended the fantastic International Pragmatics Association conference in Belfast (hashtag to look up lots of live-tweeting: #ipra2017). I was there in Antwerp in 2015 (see report on panel on prosody, parts 1 and 2), and it was great to be back in this craze of six days of talks and papers and coffee breaks with colleagues, friends, and "academic crushes" (!). I have to admit I was a bit emotional as well, because two years ago at this conference all I was dreaming of was my PhD at York with all this inspiring people, and today, here I am! (Dreams do come true, folks, you just need to work really hard...only to keep working hard!)

I have to say that the number of phonetics-related talks was disappointingly small (see the book of abstracts here), and there was so much overlap (15 concurrent sessions or more!) that I could not see all the presentations I would have liked to see. From an Interactional Linguistic/Conversation Analytic perspective to the study of phonetics/prosody, three universities stood out: University of York & University of Sheffield (UK), and Universität Potsdam (Germany), and I am going to briefly comment on some of the talks I have seen. After my report of each talk, I have included my own remarks and reflections in italics.

Richard Ogden: "The actions of peripheral linguistic objects: clicks"

Undoubtedly, Ogden is one of the people who has studied clicks as a resource in interaction most widely (see a past lecture here). In this presentation, Ogden addressed issues such as how co-participants themselves establish what actions (if any) can be ascribed to clicks. Ogden studied clicks pre-, mid- and post-positioned in TCUs, and also standalone ones, and he has found them to do interactional work for the regulation of turns or sequence (as a "metronome" for the handling of transition space across turns, for example), and/or the display of affect. What is interesting about clicks is that they are not easily manipulated prosodically, but they may co-occur with peaks of embodied activity: examples shown included raised eyebrows, shifts in body posture, and tracing gestures.  In this particular presentation, clicks followed by other continuations (pre-positioned clicks) and standalone clicks treated as a form insufficient response were studied. Participants were found to make sense of clicks in interaction and/or orient to them through their positioning, their deployment in frequent collocations and usages, and through iconicity in their mirroring in next position, or the marking of incipient speakership.

Some of the conclusions after some careful analysis of data and participant orientation (that is, not through analyst-imposed categories but through participant behaviour in the next turn, as we do in CA) has led Ogden to conclude that:

• clicks in these positions must be deliberate
• the lack of accompanying verbal material makes participants reliant on other practices to interpret them
• clicks are interpreted by participants in interaction through: positioning, multimodality, dialogic practices such as rhythm or mirroring.

Clicks are really exciting, and we use clicks and percussives so very often in speech even though they are not phonemic in English or Spanish as they are in many African languages. I have to admit I "fish" for clicks when I listen to my friends' WhatsApp recordings...fascinating!

Traci (Curl) Walker: "The differential design of other-repetition in repair initiation: does form follow function, or function follow form?"

As one of the organisers of the panel that discussed the function-form conundrum (because yes, this is a massive debate in linguistics!), Walker studied other-repetitions in response position. She reported some of the findings in the use of repetitions in interaction, such as the initiation of repair (because of acceptability issues), the checking of understanding, and the display of surprise. Walker identified two types of other-repetitions in her study: framing, and prefacing ones.

Framing repetitions consist of a sequence in which B repeats a part of what A has said, and A repeats/completes the rest. These were found to be hearing or understanding checks, mostly (suspending the display of understanding), and these have been cases where the first turn-compoments are picked up in repetition by interlocutors. Walker found these to be produced with the following prosodic parameters, completion-inducing in a way:

• tempo: slower than first saying
• final syllable lengthening
• flat pitch contour (rises < 2ST)
• articulatory features: audible release of final plosives, glottal constrictions rare

Prefacing repetitions, on the other hand, include situations where  B repeats a part of what A has said, but B keeps on talking, and in this case, turn-final components are the ones repeated. Most of these cases were followed by an explicit request for repair or further information. Prosodically, they were found to be produced with anticipatory phonetic design, thus:

• tempo: slower than 1st mention
• quieter in loudness
• falling in pitch

Walker concluded that "form and function should be one, joined in spiritual union (sic)".

The form-function debate is a huge thing, and it affects me deeply as a former trainer in Applied Phonetics for speakers of English as a foreign language. The thing is that as teachers we rely a lot on form-function mappings, and for many years, I have felt very guilty when trying to systematise intonation because many of these mappings only work in restricted contexts, and after all, the real "meaning" of linguistic features in interaction is dependent on participant orientation, something we very hardly focus on. 

When analysing conversation and asking my students to decide on intonation patterns, I have tried (with varied degrees of success) to get them to focus on what co-participants do after something has been uttered, in order to trace how a certain stretch of talk has been interpreted (oriented to), what type of action has been ascribed to it. But I know I have sadly fallen into many of the form-function connections that Walker criticises (which, to be fair, ALL textbooks are filled with), due to the need for simplication for teaching purposes. In my defence, at least I know I have not "sinned" with contextless text analysis: context and genre were always at the heart of my intonation-teaching practices, however lacking in other things they may have been.

 The teaching of prosody in ELT and the training of teachers, in my humble opinion, would really benefit from following a conversation-analytic, inductive, qualitative approach (something I will rant about in my upcoming PTLC conference presentation!).

Susanne Reinhardt: "Tying next turns to question-answer sequences: How links between linguistic forms contextualize different kinds of sequence continuation"

Susanne is a PhD student from Potsdam, and has made a very interesting presentation on format tying in post-expansions after question-answer sequences (if you want to know the basics about post-expansions, see this chapter by Sidnell here). She described how clusters of linguistic features helped disambiguate between minimal or non-minimal post-expansions. Reinhard has found that minimal post-expansion initiators had narrow pitch span, and general lowering of pitch values in relation to the question turn, as well as some degree of prosodic matching with the question turn. Non-minimal post-expansion initiators, on the other had, had a wider pitch range, a medianization of pitch values (moving towards the median), and a clear contrast between the question-answer pair, and the third position, initiator of the expansion.

Dagmar Barth-Weingarten: "Discourse units in English interaction: a prosodic-phonetic perspective"

This was one of my favourite talks in the conference (together with other non-phon talks by Pillet-Shore, Raymond, Hoey, and Heritage). Barth-Weingarten discussed the bases of what she proposes in her new book, a boundary-based approach (rather, a "cesura"-based approach) to the analysis of speech "segmentation" (being particularly careful here not to discuss the notion of "units", which Barth-Weingarten somehow argues against).  The presenter made a very thorough review of different approaches to chunking, and discourse and conversation units, and focused on the similarities and differences between them, as well as the problems, some of which stem from the use of "monologic" data, or the mapping of chunks with syntax or specific pre-determined phonetic parameters. Since units in the end are an epiphenomenon of segmentation in talk as it happens, in order to avoid analyst-imposed categories, Barth-Weingarten proposed focusing on what creates discontinuities in talk, that is, what clusters of features appear to break the flow of talk (which, she states, are in fact what co-participants orient to in the online processing of talk, rather than the units) and how these prosodic parameters display differing degrees of cesuring. By looking at data, real recorded interaction (non-monological), the presenter identified a number of parameters that have a role in marking different degrees of cesuring, including pitch movement, voice quality, tempo, rhtyhm, articulatory features, settings and processes (especially in terms of release, aspiration and glottal constriction), and pauses. 

I see and love the points and criticism made in the presentation and the book to the different theories of "chunking" (tonality). I have done corpus work on this, and I have found syntax-information structure and prosody mappings in pseudo-monological speech (lectures, stories, some speeches). Most of the materials I have developed for my students on this work quite well for these generic types, but it's completely true that when it comes to talk-in-interaction, chunking is fuzzy, and less explicitly (at least) rule-governed. 

At times I think that in ELT we see texts, even oral texts, as finished products. We forget about the essential value of online, emergent speech as it develops. In my intonation lessons, I always felt guilty when asking students to work on texts, because of the unnaturalness of having to see something already developed and treat it as if it was developing. But then again, we need to make peace with the fact that there are different tasks and activities meant to teach and test different things, and having to do transcription or reading aloud work for intonation practice, if applied properly, is a way of testing what was taught. However, if I had the chance to re-plan my intonation courses, I am sure I would go for activities that foreground this developing, online planning aspect of speech, which I did, to a certain extent, with my "first-sight speaking" tasks (emergent, prompt-induced speech tasks) in some advanced courses. 

The following are talks I would have liked to see, but couldn't (refer to the book of abstracts to read about them!):

• Elena Becker: "Closing telephone conversations: The role of prosody"
• Marjorie H Goodwin "The intertwining of touch, prosody, and voice"
• Adriana Cáldiz "Prosody and evidentiality: About how some intonational features pertain the roots of discourse in the Spanish of Buenos Aires"
• Brown & Prieto: "Multimodal (Im)Politeness"
• Jesús Romero-Trillo: "The pragmatics of prosody in intercultural communication"

All in all, my IPrA 2017 experience was fantastic, and it spurred on me further feelings of passion about what I do and what I would like to do with my research, while at the same time stirring some of those feelings of inadequacy that come with taking some distance from one's teaching and thinking about what one would have liked to do better, or differently. I wrote a short commentary of what this conference was like for me as a postgraduate student at the ROLSI (Research on Language and Social Interaction) blog here, if you are interested.

******

I am aware that my posts lately may have been a bit too technical, and yes, at the moment this blog seems to be a way of trying to make sense of English phonetics in real life (now that English is so readily available for me everyday), and my own past practices as a pronunciation teacher. I am sure I have done many things right, and if I see what my students have been able to achieve, and how many of them actually liked the subject and approached it passionately, I am sure I must have done quite a few things well. However, I am still trying, in my head, to find the best possible way to introduce the prosody of English to speakers of other languages based on what happens in real life (and not on the analysts' head as I am sad to see in many textbooks!), and I pretty much guess this will be an ambition to keep me going for a couple of decades.