Interview with Anne Wichmann

Pragmatic meaning is a very important aspect of human communication, but what makes it interesting is its mysterious and slippery nature. Most nonlinguists tend to think of meaning only in terms of what words mean, but in pragmatics one must account for what people mean. It’s interesting because it is invisible: we know it’s there, but we can’t often identify why it’s there. I also think that pragmatics is interesting from a methodological point of view. There are different approaches to pragmatics. There is Austin and Searle’s Speech Act Theory, derived from philosophy, which has become more of a sociological-ethnographical approach to pragmatics. This approach highlights what speakers intend rather than what listeners interpret, but it also requires a profound understanding of the context in which things are spoken. There is also the cognitive approach of Relevance Theory, although it worries me that this approach is often based on introspection and invented data. Then there is the strict conversation analysis approach which focuses on close detail and only accepts evidence from participants’ responses. In my experience, the analysis can be so close that it is very difficult to stand back from individual examples to make generalizations. In my own work, I have mostly used speech corpora, in other words transcribed recordings of spoken language, allowing me to look at the frequencies and distribution of data, along with the immediate context and co-text. There are many areas that are interesting for pragmatics, ranging from basic speech acts to turn-taking and the whole area of politeness and especially impoliteness. Also, it’s worth looking at pragmatics from a developmental perspective, i.e., how children learn. For example, children often take things literally, so if you say on the telephone “Is your mummy there?”, they think it’s a question, rather than a request to speak to the mother. How and when do they learn this? So pragmatic meaning is interesting from the perspective of child language acquisition. There is also a lot of interest in historical investigations

which look at language diachronically, such as how speech acts have changed over time.
For my part, I have worked extensively on prosody and meaning. Prosody in intonation languages such as English (unlike tone languages such as Mandarin) does not convey propositional meaning but has other functions, including the expression of pragmatic meaning. Sometimes this is by means of phonological choices: for example, the choice of a rising contour or a falling contour can determine whether an utterance is understood as a statement or a question, a request or a command. Similarly, the contour chosen when apologising with 'sorry' may determine how genuine it sounds. What is interesting is that listeners recognize these meanings without necessarily knowing why.
Then there is the case of backchannel responses in conversation. When somebody is speaking, the other person often says "uhm", "yes", "right", with a rising pitch, encouraging the continuation of the conversation, and indicating interest. However, if such a response is uttered with a falling pitch, it seems to signal two things concurrently: one, I hear what you're saying and two, I'd like you to stop now.
Other effects are created, not by phonological choice, but phonetically, such as varying pitch height or range. For example, when reading aloud, it is possible to indicate a new paragraph by using an extra-high pitch at the start of a sentence. Choice of pitch level is also important in conversation. Speakers tend to mirror the person they are talking to, both with gestures and pitch: matching the (relative) pitch level of the person they are speaking to can indicate a level of affinity. All these examples show that the voice plays an important part in conveying meaning.
One question that is currently of interest is what constitutes a charismatic speaker. I have recently been looking into the use of the voice in different kinds of performance, from newsreading to liturgy, especially the use of pitch range, from the very wide and charismatic use of pitch in political speeches to the very narrow, almost monotone voice typical of liturgical speech or poetry reading.

How Do Cross-cultural Issues Come into the Picture?
The way we convey pragmatic meaning varies according to language and culture, and there is therefore considerable potential for misunderstanding. Of course, speakers of all languages ask questions and make requests, but how they do so politely, such as by using different degrees of indirectness or different forms of address, is not universal.
What I find interesting to investigate cross-culturally are the non-verbal aspects of communication, because these are generally unconscious. We can choose our words, but it is harder to monitor our body language or tone of voice. For example, speakers vary in how much they exploit variation in pitch: in some cultures, a narrow range is neutral, but to others it can sound lacking in interest or emotional engagement. Conversely, a lively speaking style, using a wide fluctuation in pitch, might be normal in one culture and sound exaggeratedly emotional in another.
Different cultures also have different tolerances to silence. Some find silence in a conversation unsettling, while others are comfortable with it. Overlapping speech is also perceived differently -in some cultures, overlapping turns may be regarded as collaborative, whereas in others they are seen negatively, as interruptions. There are also differences in the use of body language: in some cultures, it is polite to look at a more powerful interlocutor while in others it is polite to avert your gaze. There are even cross-cultural differences in when it is considered appropriate to smile at someone.
Why does this matter? The potential for cross-cultural misunderstandings is of course an issue for language teaching, both for teachers and for students: teachers of multicultural classes may misinterpret student behaviour, and students of a language need to learn not just words and grammar, but how to employ them in different contexts and situations. Cross-cultural pragmatics is also critically important in high-stakes situations: interpreters in a court of law or in political negotiations may be torn between mediating literal meaning and what they believe is the speaker's intention. Presiding judges may also misinterpret a defendant's behaviour -for example, by misinterpreting silence before replying to a question as evidence of guilt.
While some cross-cultural differences may simply be noticed as unusual, or perhaps impolite, they can clearly have serious consequences, as the example of political or legal situations shows. For this reason, I find cross-cultural misunderstanding, particularly of the non-verbal kind, extremely interesting.

What Advice Could You Provide for Contrastive Pragmaticians with Interest in Pragmatic Meaning and Prosody?
This is a difficult question. My first advice for anyone working at the interface between research areas would be to work closely with other people who complement your own expertise. I have been lucky enough to work with other researchers who had expertise in areas where I felt less confident. This means that if you are not a prosody expert, you need to work with someone who understands the prosody of the language you are working with, and its relationship to pragmatic meaning. This includes knowing the effects of both phonological choices and phonetic variation, and also being aware of the potential for crosscultural differences and misunderstandings. It is also very important to reflect on one's methodology and the data to be used. I have often used data from spoken corpora, with recordings made in relatively natural situations. Obviously, when people collect corpora, they don't do it without asking for permission, so this removes some of the naturalness, but it's as natural as you can get without recording secretly (and unethically). For comparability, however, it is sometimes easier to use carefully designed experiments. Much experimental work involves people reading a script containing the elements you want to study. If you are interested in speech acts you might have small dialogues containing, say, requests such as "Can you pass me the salt?" in one language and the equivalent sentence in another language, and compare the prosody. This method takes you further away from spontaneous behaviour, but the advantage is that one can compare like with like and use instrumental methods to make precise measurements.
I would advise young prosody researchers to learn to annotate, or, if their focus is elsewhere, to become familiar with annotation systems. This involves knowing the phonology well enough to understand phrase boundaries, pitch contours and other features. Annotation is aided by modern software for instrumental analysis, but it still relies heavily on auditory analysis -the acoustic record has to be interpreted by a listener. Your choice of method entails a consideration of what counts as evidence for your claims. In an experimental study, it may be a precise analysis of the data, or the perceptions of listeners. With naturally occurring data you need to decide whether you're going to use your own intuition, that of other listeners, or whether you are only going to consider the response of the participants themselves.
Pragmatics is a very broad and complex discipline. Comparing different languages and their expression of pragmatic meaning involves another layer of complexity. Adding an analysis of non-verbal information, such as prosody, adds even more complexity, and researchers are often tempted to focus on very narrow notions of pragmatics, such as speech acts, in order to keep their research manageable. However, the more we restrict our data to what is manageable and measurable, the further we move away from natural speech. There is always a balance to be considered, depending on the aim of the research and how the results are to be used.