Posts tagged conversation

Most of my teaching in 2020 and 2021 was done online. Like many people, I had to learn how to teach a lot of my face-to-face classes in online contexts. My institution used Zoom for this, and even in 2022 and beyond I will continue to teach some classes online. I think this is great. So many of my students used to have a horrific commute just to attend a single class, or would miss classes because of unavoidable commitments. Of course, there are many ways studying online is a terrible experience too. One particularly fraught online experience is classroom discussion. An experienced teacher can manage a classroom full of group discussions pretty effectively by moving around the room and keeping an eye and an ear on each group. Students usually feel sufficiently private at tables, but also have the buzz of the classroom to help the discussion along. Online classes take all of that away from everyone.  

Over the last two years I’ve developed a way of managing breakout room that makes the best of a bad situation. My students regularly tell me my classes are some of the best they’ve done online, and when I share this method with colleagues they tell me they’re going to borrow it – so I thought I would write it up in case you are also still managing group work in online tutorials and workshops.  

Step 1: Set the context

I don’t make my students have cameras on (you’re making a lot of assumptions about people’s access to technology and privacy for online study), but I explain to them the benefits of seeing colleagues faces. I also encourage students to use the chat box and to play around with the reaction button to help me calibrate to the mood of the room. If you get good at running an online classroom with multiple streams like this, you know some of the multitasking joy of doing live broadcast radio! But I l also get this isn’t for everyone.  

Step 2: Giving students control

I let people select their own breakout group, which is an option in Zoom. I then explain how they should sort themselves out. The number of rooms I set up and how I assign them vary depending on the size of the class, but let’s say I have a group of 20 to keep numbers manageable. I set up 5 breakout rooms and tell students the following:

• The final room is the quiet room: no one speaks, you just work through the activities in a companionable silence.  
• Rooms 3 and 4 are for groups of friends to move into. Start messaging them now to arrange to work together!
• Rooms 1 and 2 are for people who want to meet new people or chat with a random group of people in the subject. Start with room 1 and when there are four people in there start room 2. 
  

I work backwards to flag that I’m committed to supporting the most exhausted students and it’s not an afterthought, but also I like the last room being like the quiet space up the back of the library.

For people in rooms 1-4 the maximum is 4 people. Four is a good number for conversation – more than that and the conversation starts to fragment into smaller groups, which is hard to hold together in an online space.  

Step 3: Running rooms

Online classes need much tighter guidelines and expectations than face-to-face classes, since it’s harder to change the flow of things. I make sure tasks are very clear and the time they’ll be in the rooms is also very clear to them. I also tell students I expect them to spend the first 2-3 minutes at least on small talk (not in the silent room!), online classes can really miss that human connection sometimes.  

Sometimes we have to do a bit of on-the-fly juggling, if two groups vie for the same breakout room or there are more people in the quiet room than usual – but often students get the hang of this after a few weeks and get good at managing the breakout rooms themselves.  

I also let students know that after 10 minutes I’ll come to each room to answer questions and check in, or they can ping me or return to the central room to ask questions too. Sometimes I’ll use something like Google Docs or Slides where I can see students adding content in real time, which gives me an idea of who might need more help or encouragement.  

Why this set up?

This setup for breakout rooms gives students a little more control over their learning experience.

• Some students are tired or find online classes incredibly stressful. I like providing them the chance to just get through the material, with a bit of solidarity and a central place I can check in with them.
• Some students really miss the chance to sit at a table with their friends and catch up. I know I do!
• Some students miss the chance to get to know new people at university. Some of the friends I made in undergraduate tutorials are still some of my closest friends today, I like to give my students some small chance of having that experience too.  

The important thing I’ve noticed is: it’s not the same students in the same rooms each time. Some weeks even the most sociable students need a break, or sometimes your friends didn’t make it that week and you’re up for meeting someone new.

This is my default for running breakout rooms, but I do shake things up occasionally. I’ll randomly allocate students to breakout rooms for some tasks, telling them I’m intentionally shaking things up and giving them some icebreaker tasks. I usually only make groups of 3 in these weeks to give them more time to get to know each other.

Managing breakout rooms is one of those times where thinking like a has helped me improve my teaching. Thinking about breakout rooms as interactional spaces, and how we generally prefer to interact has helped me refine my classroom management.  

If you have any success using–or refining(!)–this process, please let me know!

I have had more than one grandparent with dementia. This post is a series of personal reflections about the way conversation changes when someone you know well loses the ability to remember, and how linguistics has given me a framework for dealing with this. I don’t know anything about dementia other than these experiences, and there are whole sets of emotional and logistical challenges of living with dementia, or a loved one with dementia that I’m not going to address here.

Dementia presents a challenge when you’re having conversations, because the person with dementia loses the ability to track what’s been covered in a conversation - perhaps over the course of 15 minutes, perhaps sometimes in 15 seconds. This creates loops and eddies in conversational topics. It also breaks down one of the fundamental features we require in communication; the belief that our conversational partner is a cooperative one.

Cooperation is important because otherwise a well-flowing interaction is, on the surface, a series of non-sequiturs as the person speaking makes a leap and their conversational partner(s) makes the leap with them. Why are we talking about Uncle Desmond? It must be relevant to the conversation about Aunt Muriel we were just having. A person with dementia can feel from their perspective that they’re being cooperative but from your perspective they’re telling you the same story about Uncle Desmond’s first wife for the 3rd time in 15 minutes.

I’ve noticed that different family members deal with the destruction of the facade of cooperation in various ways; the helpful one tries to turn the conversation to other topics, the sanguine one lets the story play out a fourth time, the pragmatic one tells my grandmother every time that she’s already told this story and that, in fact, she has dementia. Each is trying to rebalance the illusion of cooperation, and each has varying degrees of success which I’m sure relates to my grandmother’s relationship to them, her personal condition, and everyone’s mood on any given day.

Another group we see constantly breaking the sense of cooperation is children. Kids are tiny randomness machines and keeping them on conversational tracks can be a lot of work (and work that extends beyond the years of just learning to speak grammatically). The thing with children is that we expect this from them to some extent. The challenge with conversing with a family member with dementia is that we have years, often a lifetime, of conversational rhythm with them.

The ways in which we expect conversational cooperation were broken into four main categories by Paul Grice. He called these maxims, which I always think of in the ‘guide’ sense rather than ‘rule’. They can help clarify the ways that it can feel difficult to maintain a conversation with someone who has dementia. The first that comes to mind for me is always the maxim of relation, because it does not seem relevant to return to the same story of Uncle Desmond again without new reason, and because this might lurch suddenly into a new conversational track about a cousin you haven’t talked about for 15 years. I’m sure that if I made transcripts of conversations my family have with my grandmother I’d find examples of various different ways that our conversational expectations are challenged in these small talk moments. Knowing about conversational cooperation has made me far more relaxed about approaching these conversations, in one of the more unusual ways my linguistics training has giving me a weird sense of peace about a process I have no control over.

This is where my thoughts on this topic had arrived at. I was talking to my colleague Tonya Stebbins about these experiences the other day and she introduced me to the concept of validation theory. Validation theory is an approach to interactions with people who have dementia that starts from the assumptions that you need to consider the emotive content of what they’re saying, rather than the informational content.  This allows you to engage with their emotional state. There’s an overview at this website, but I’m yet to dig into any literature on the efficacy of this approach.

To illustrate validation theory with an example I still remember from when I was very young and we were visiting my great grandmother: When a person with dementia is worried about her young grandson getting off the bus after school, even though she’s sharing these concerns with that grandson who is actually a middle-age man visiting with his own small children, the idea is to not simply point this out to her, but to attend to the underlying anxiety that has surfaced as this particular and temporally disjunct concern. Regardless of the fact that it’s been decades since her grandson stopped going to school on the bus, taking a conversational approach that spoke to her concern would be a way to address whatever anxiety she had. 

What I immediately like about validation theory is that it re-balances the onus of cooperation in the conversation. In this approach, the conversational partner without dementia is responsible for ensuring they are not failing at the maxim of relevance by attending to the implicit anxieties, worries or joys in the conversational eddies of the interaction. 

Chatting to my grandmother is both one of the most normal and most challenging things. It is as once unremarkably familiar and exhaustingly surreal. I’m grateful that a linguistic perspective on conversation has given me some coping mechanisms for navigating the more surreal moments.

See also:

• Lingthusiasm episode 11: Layers of meaning - Cooperation, humour, and Gricean Maxims
• Hamilton, Heidi E. (2019). Language, Dementia and Meaning Making. Navigating the Challenges of Cognition and Face in Everyday Life.  Palgrave Macmillan.
  

It’s really astonishing that human conversation happens at all. People respond to questions more quickly than they should be able to think of their answers, and our decisions about whether someone is being helpful or not can be based on millisecond differences in their responses.

Nick Enfield’s 2017 book How We Talk: The Inner Workings of Conversation summarises some of the key work he has undertaken on how conversation works, with colleagues at the Max Planck Institute for Psycholinguistics. The book situates this work in some of the key research from the field of Conversation Analysis in the last 50 years. Topics covered include looking at how silences of different lengths gets interpreted, how people repair their speech as they go along, and how ‘um’ in English has correlations in enough other languages that Enfield refers to it as a ‘universal’ word. All of this is framed throughout the book with the metaphor of the ‘conversation machine’, which is a delightful commercial model of the ‘interaction engine’ in Nick’s 2006 book with Stephen Levinson (which I found tremendously helpful back when I was conceptualising my PhD research).

There’s lots to like about this book, and the presentation of research. I particularly like the commitment to ensuring that it is not only the ol’ regulars like English that are included in analysis. You’ll get to learn about the differences in conversational features in languages like Lao (Laos), Murrinhpatha (Australia) and Siwu (Ghana) too - or perhaps more astoundingly, the similarities between them.

This book is a great example of how research can be presented in a clear way that is engaging for a non-expert reader. It’s for the quick-minded non-expert, who is ok with acquiring (or ignoring) a bit of jargon along the way, but if you want an exercise in good pop-science writing, sit down with one of these chapters and the original research on which it is based. I did hope the larger question of ‘why this work?’ would be answered; Enfield is diligent about making sure all of the researchers are given a research-area title, but it is intriguing to ponder why conversation intrigues psychologists, linguists and sociologists alike. 

If you did linguistics but never got to study Conversation Analysis, or you want a whistlestop tour of some of the most interesting work to come out of the field in the last couple of decades, this book is certainly worth a visit.

Buy: Bookshop.org affiliate link, Amazon affiliate link

[Thanks to Basic Books for the review copy. I also purchased a copy when Nick launched the book during the ALS conference in December because I couldn’t wait to get my hands on a copy.]

When discussing human communication, researchers will often put speech over here (*gestures to a space called ‘LANGUAGE’*) and gestures over here (gestures to a space called ‘OTHER STUFF’). A recent paper from Joe Blythe, Kinngirri Carmelita Mardigan, Mawurt Ernest Perdjert and Hywel Stoakes looks at how speakers of Murrinhpatha discuss directions, and argues that pointing is not just an ‘additional’ feature of discussing directions, but a necessary part of the language. The paper is in Open Linguistics, which means it’s open access and free to download.

Murrinhpatha is a languages spoken in Wadeye and nearby communities, in the north of Australia. There are around 2700 speakers, and it is actually gaining speakers, becoming the main language of an area where there used to be others. The language does not have the kind of directoinal vocabulary we’re used to, things like above, next to, to the left of, north, south,etc. They have ‘left hand’ and ‘right hand’ but these are almost exclusively used to refer to the hands themselves, literally. There’s also words that translate as ahead and behind and that’s about it.

So, how do speakers of Murrinhpatha talk about directions? Blythe and the team wanted to know, so they set up a version of a classic experiment, where one person knows a location but can’t say what it is, and the other person has to guess. Normally this is done on a map, but as they were all real and local places, they had people sit and talk about them. This is actually a neat methodology, because in the local culture (and many indigenous cultures of Australia), there are often taboos on talking about specific places - so Murrinhpatha speakers have to negotiate these kind situations all the time.

Instead of the wealth of direction terms you’d get in English, the team found that people used a number of strategies, including referring to specific landmarks where possible, demonstratives (similar to English, ‘there’ and ‘here’), and a whole lot of pointing. That pointing almost always happens oriented in real space towards the location.

The researchers acknowledge that as more speakers gain access to mobile phones, and people are face to face less, the language may need to shift to accommodate this. They are keep to point out that a lack of directionals isn’t a ‘fault’ in the language:

“In suggesting that Murrinhpatha is virtually ‘directionless’, we wish to defeat the connotation that the language has any sort of deficit. Although the lexicon and grammar seem not to allow ternary frames of reference, Murrinhpatha speakers retain a precise spatial orientation and are in no way hamstrung by the apparent absence of abstract direction terms. By expressing directional vectors deictically (gesturally and/ or as implicated through discourse anaphora) they readily make themselves understood, despite potential complications arising from culturally specific restrictions on particular placenames.” (p. 155)

[FYI - I used to work in the same department as Joe and Hywel back when I as at The University of Melbourne. They are lovely people]

Joe Blythe, Kinngirri Carmelita Mardigan, Mawurt Ernest Perdjert, Hywel Stoakes. 2016. Pointing Out Directions in Murrinhpatha. Open Linguistics 2: 132–159. [OPEN ACCESS - FREE TO DOWNLOAD]

My friend Greg posted on my Facebook wall recently with a specific request for this post on the blog. He requested that Superlinguo explain why people persist in using the word “obviously” in their speech, when their phrase would be just as clear without it – apparently it is driving him crazy and we can’t have that. Words shouldn’t drive you crazy, people.

The example given came from a TV interview with an Australian cricketer at the end of a recent match, where the player representing the winning side commented, “Well, obviously we played well…”. (Sidenote here: the speech of sportspeople being interviewed after a game or race has a lot of fertile material for analysis, but we won’t go down that particular path today. Hopefully we can find time to delve into this one day soon.)

Greg was justified in noting that the above example sentence using “obviously” would be “just as clear without it”. That’s because in this case the word didn’t necessarily have any lexical meaning to add. But it still did a job. It served as a discourse marker.

A discourse marker is a sound, word or short phrase that doesn’t add any additional meaning to the sentence, but has a different function, for example: topic changes, reformulations, discourse planning, stressing, hedging (mitigating a word or phrase’s impact) or backchannelling (reminding the other speaker/s that there are two lines of communication present).

In English we have a number of common discourse markers, including “basically” “actually” “like” “I mean” and “okay”. Here in Australia we are big users of the “yeah, nah” discourse marker too.

The functions that discourse markers undertake can be classified into three broad groups: (a) relationships among (parts of) utterances; (b) relationships between the speaker and the message, and © relationships between speaker and hearer [source]. 

In the case of “Well, obviously we played well….” it could be that the discourse marker manages the relationship between the speaker and the message, specifically as he does some discourse planning, i.e. thinks about how he’s going to phrase the response.

It could also be a case of hedging, where the phrase “we played well” is downplayed by indirectly referring to the winning scoreline as the ultimate arbiter of success. It manages the relationship between the speaker and hearer, and allowed the cricketer to mitigate the impact of his positive assessment by deflecting it to the score, thereby avoiding bragging (not socially acceptable at all in Australian culture – we’ll do a future post on the phrase “tall poppy syndrome” one day soon for all you international readers!).

Either way, I think that we need to accept that we’ll hear words in others people’s speech that seem extraneous or inefficient. But they are doing a job! And on some level we all do this as we try to keep interactions fluid or refer to the wider context of what is being said. 

I hope this assures Greg that there’s no defect in our Australian cricketers. They are having a good summer so far and if the current series against Sri Lanka is anything to go by, we’ll be hearing more TV interviews from our winning Aussie cricketers soon.

Telstra is one of the main providers of mobile services in Australia. They have had a reputation for poor customer service (but then as far as I can tell, most major telcos in Australia are roundly criticised for their customer service, so you really can’t read too much into that).

In a novel attempt to make contacting them easier Telstra now have a 24x7 “live chat” service, where customers can chat with Telstra staff in am instant message style program. A friend recently used the service to fix a problem with voicemail. One of the upsides of using this service is that you can keep a record of exactly what has been said. This is how this interaction went, I’ve given both of the participants pseudonyms:

Enrique: As I understand your concern is regarding messagebank?

Ramona: Yes thanks

Enrique: You have reached the right person, I’ll be happy to assist you.

Enrique: May I have your mobile number?

Ramona: 04********

Enrique: May I have your date of birth just for verification.

Ramona: 10/11/19**

Enrique: Thank you for all your information, I just need a moment in order to provision the changes to your service. Won’t be too long.

Ramona: OK

Enrique: I have completed the process for you.

Ramona: OK thank you!

Enrique: It was my pleasure to assist you.

Enrique: Please feel free to come back and chat with us anytime, as we are always happy to assist you. You can click on the end chat button whenever you are ready.

Functional! Efficient! Only problem is that it was the most artificial IM I’ve ever seen. We spent a bit of time in the lab debating whether it was just some well tuned AI on the other end of the interaction. Everything on Telstra’s pages indicate that the ‘live chat’ function involves real people on the other end of the line, and there is often a 'wait period’ if all of the live chat staff are busy. So it looks like this is a human being with a set of scripted dialogue options (and as a CompSci friend pointed out once, it’s often easier to just pay a room full of lowly waged humans than build a competent piece of software to interact with humans).

It’s not that I’d suggest to Telstra that Enrique sign off with “k, all done 4 u - txk bai” - but IM has its own genre, and no human in the history of IM has bothered to type “ I just need a moment in order to provision the changes to your service.” I haven’t found to many other transcripts for such interactions on the net but this one indicates that staff aren’t always stuck with such stiff dialogue options. In that first one you see the use of i instead of I, and the numeral 1 instead of one. 

I’m all for companies trying to find better ways to help their customers, but they just need to wind down the formality a little so their scripts don’t sound quite so robotic.