You Knew Just What to Say

This week we’re looking into the realm of social linguistics, or sociolinguistics. Specifically, “register” and “turn taking”. Imagine you are in an elevator and the person next to you is in track pants and an old t-shirt, she has her hair in a loose bun, and she is talking about her “wild night” on the phone. She hangs up and she begins to speak with you. How would you speak with that person? What would you talk about, and what would you avoid talking about? How would what you talked about, and how you spoke change if she got another call, and her “wild night” had been at a hospital, and you realized she was an ER doctor?

Professor Maite Taboada from Simon Fraser University is going to explain what goes on in our heads when we navigate what to say to people we meet, and how we should say it. What Professor Taboada studies is essentially this: why and how people know what to say in a conversation. This could be a conversation removed from the other speaker(s) like in an email, or text message; however, it could also be one happening in-person and “live”.

I find it fascinating how people (hopefully myself included) know what to share when, and are quickly able to phrase it to make a conversation go smoothly.

I’ll let Professor Taboada explain why she’s so fascinated as well.

[My sister is a master conversationalist]

[My sister is a master conversationalist…and laugher]

Michael Iannozzi : First of all, what is discourse? And what is register?

Maite Taboada : These are two big questions. In linguistics in general, we think of discourse as language in context. That tends to mean that it anything bigger than a sentence. Other areas of linguistics focus on sound, words and their structure, sentence structure, and meaning. You could say that discourse starts where the other disciplines stop, because discourse focuses on how the structure and meaning of sentences make sense in the context where they are used.

Register ties in with the notion of context. As speakers, we all know that certain things are more appropriate or successful in one context than another. We also know that certain interactions take place in chunks or steps. That knowledge is knowledge of register. For instance, we know that conversations usually have a series of greetings as openings, and a series of turns as closings. Those are steps in the conversation. We also know that you can write “U” for “you” in a text message, but that you probably shouldn’t do that in a letter of application for a job. Register, then, is that complex relationship of language, the purpose of our language exchanges, and the context of situation in which those exchanges take place.

MI : In a conversation, who decides the appropriate register?

MT : Nobody really decides. Speakers adjust to the appropriate register based on the reason for coming together and talking, and based on their previous knowledge of similar situations.

MI : How does the register and discourse style used between two speakers change as they come to know one another better?

MT : There are often subtle changes over time, and also in the course of a single conversation, that have to do with establishing common ground, and also with gaining familiarity with each other. Researchers in sociology have also studied the process of accommodation, whereby conversation participants start adopting each other’s terms, structures and even pronunciations, sort of settling on a common way of saying things.

MI : Do people tend to have a personal “baseline” or “default” register or discourse style?

MT : No, register changes with the context. Think of all the situations you go through in the course of a day. Getting on public transit, purchasing goods, greeting colleagues or fellow students at the beginning of the day, work meetings, classroom interactions, meetings with friends, sports and cultural activities… Whenever you use language in those daily experiences you are employing a different register. And I have only mentioned spoken interactions. Every email, memo, assignment, letter or text you write is part of a different register as well. In each of those cases, you know roughly what steps are needed to accomplish the action, and what language is most appropriate.

MI : How are people perceived who use what the other considers to be the incorrect or unsuitable register for a given conversation?

MT : There is an obvious breakdown of communication, or at least a feeling of something not being quite right. Think of the person who boos during a speech, or the professor who rambles during a lecture. The participants feel that the behaviour is not fulfilling the purposes of the interaction.

This is more obvious with children. Part of the process of learning language in childhood has to do with learning the appropriate registers, and the steps to go through in a conversation. That is why we insist children ask for things properly, that they say goodbye, and that they address people in certain ways. We are teaching them the right steps and the right language.

MI : What is turn taking?

MT : Turn taking is the process of negotiating who is speaking at any given time. Researchers have studied turn taking for a long time, and there is a set of very simple rules that describe the process. Basically, at certain points in the speech (places where a unit of communication, usually a sentence, is finished), there are three options: the current speaker may continue; the current speaker may select the next speaker; or another speaker may self-select. The rules are deceptively simple, because the really tricky thing is how the intention to continue talking or to self-select is indicated. We often signal the intention to continue talking with “uhm” and “uh”, which sort of buy us time to formulate the next thought. We can also raise our hand, as if to say “I’m not done yet”. Other speakers may start speaking, thus interrupting, or raise their hand, or otherwise use facial and body language to indicate the desire to speak. The range of signals is quite wide, and changes from culture to culture, but we somehow learn to negotiate it.

In some contexts, of course, turn taking is heavily regulated. Think of Parliament, where the Speaker (funny term, because they don’t speak that much!) assigns turns. Or a classroom setting, where the instructor usually asks that students raise their hands, and points to the person who is supposed to speak next.

By the way, the expressions “taking the floor” and “holding the floor” [both used in studying turn taking] have their origin in parliamentary settings, where the person who wanted to speak had to come to the floor in the middle of parliament.

MI : Do gender differences play a role in the decision of register and topics? Do people of different genders approach turn taking differently?

MT : There are heaps of research on gender differences in speech. My own conclusion is that there is no simple answer to this question, and you will find research that equally proves that women interrupt more often and that men interrupt more often. It really depends on the context, and I do think that there are gender differences in how women behave at work meetings, for instance, as opposed to casual conversation. But I don’t think you can issue blanket statements like “women speak more” or “men interrupt more”.

MI : How do you collect speech of different registers in a scientific way?

MT : Data collection is an important part of the research process when you are dealing with naturally-occurring data. If we want to focus on a specific register, the best approach is to recruit people and have them speak in the particular context in which we think the register occurs.

For instance, right now one of the PhD students, Emma Mileva, is investigating the language of alternative therapy medicine. She has approached a number of practitioners, and in cases where the practitioner approves, it is also difficult to persuade the patients to allow recording to take place. She has managed, however, to collect some data and is finding interesting phenomena. Her hypothesis is that alternative therapy interactions sit somewhere between traditional medical interviews and psychotherapy sessions, in terms of the language and power relations that they exhibit.

MI : Where do you hope to take your research next?

MT : I have actually left spoken language aside for now, and I am concentrating on written language. In the last few years, I have been interested in the language of evaluation: how we express opinions, and how we intensify or mitigate them. One current concern is to measure the impact of context on words conveying evaluation. Take, for instance, the word “great”. On a scale of 1 to 5, where 1 indicates negative, and 5 very positive, I could say that “great” is maybe a 4, that is, a word that is quite positive. Now, what I want to know is what happens when you say “It could be great”. Something has changed, right? “Great” is not a 4 any longer. A similar change happens in constructing negative sentences, “It isn’t great”; or verbs expressing beliefs, such as “I believe it’s great”; or conditionals “It’s great if all you want is a quick read” (when describing a book).

In my most recent project, I built a software system that takes reviews of movies, books and consumer products and extracts evaluation from them, assigning a numerical value to the review, the same way that reviewers assign stars to their reviews online. What we found is that we could reliably predict evaluation and opinion in most cases, but that the examples I mentioned above [maybes and negative phrases] made the task much more difficult.

This is fascinating research into how we express our opinions, and how that happens online. So far, I have worked with long-ish texts, such as reviews, but a big area to tackle is micro-blogging, such as Facebook comments or tweets. At the other end of the spectrum, I am also investigating formal writing, such as newspaper editorials. I am looking forward to situating all those different types of language, and capturing how evaluation is expressed in each case.

MI : What has been your favourite part of your research so far?

MT : All of it is fun! I get to study what I think is our most human trait, the fact that we speak and use language to communicate. It is quite remarkable that we use language every day, yet most of us are unaware of its inner workings and all the possibilities it offers.


A sincere thank you to Maite Taboada from Simon Fraser University for speaking to me about this area of linguistic research.

We use language, as Prof. Taboada said, almost constantly throughout each and every day. That can make it easy to forget the brain power that goes into not just formulating thoughts and expressing them, but also into knowing how to express those thoughts to fit the context or social situation you find yourself in.

We’ve all had situations where we’ve felt we were “on a different wavelength” from the person with whom we were speaking, or that they were perhaps sharing TMI (Too Much Information). A lot of that comes down to having a different understanding of the context, and therefore register, best suited for a given conversation.

Knowing how to employ the right register in a situation can avoid an awkward conversation, and if Professor Taboada’s research can help avoid awkwardness, then it is surely worthwhile.


Take care eh,


Michael Iannozzi