We’re coming to the end of this multi-post overview of Jaana Suviniity’s PhD thesis on the role of interactive features in lectures delivered in English as a lingua franca (ELF) – when English is not a first language for the speaker or listeners. When students rated these lectures on a scale of “challenging” to “accessible”, it became apparent that a major difference between the more or less accessible lectures was the quantity of interactive features. After giving a general overview of her data and findings, I reviewed Jaana’s findings on control acts in ELF lectures. Now I’ll take the two other interactive features she examined – questions and repetitions.
“So what does it mean now?”
The abundance of questions posed by teachers in the more accessible lectures was what first motivated Jaana to investigate interactive features. The overall figures for the six lectures she sampled helped to clarify her first impressions – in total she found the teachers asked 59 questions in the challenging lectures, while the accessible lecturers asked 179. But what kind of questions are they? Jaana started with a broad division between audience-oriented and content-oriented questions.
The most common audience-oriented questions were information checking (so did everybody understand this) and didactic elicitation, in which the teacher invites substantial input from students (so what does it mean now too low water retention what does it mean). As for content-oriented questions, these involve the audience less directly, being used to introduce information, provide examples, or organise topic changes (how about temperature so the temperature plays also important role).
In terms of overall distribution throughout the sampled lectures, there were 173 instances of audience-oriented questions and 65 examples of content-oriented questions. As you can see from Table 1, when these question types are broken down by the category of lecture, a striking difference is revealed. Among the less interactive, content-oriented questions, slightly more were found in the challenging lectures than in the accessible ones. But the audience-oriented questions were strongly concentrated in the accessible lectures, which contained 88% of all instances of audience-oriented questions in the data (see Suviniitty 2012: 117).
There are three accessible lectures in the sample, but almost all of these audience-oriented questions (150 of 152) come from just two lecturers. Since these types of questions expect some response from the audience, it’s also interesting to see how many of these actually obtained a response. As it turns out, two thirds of these questions were answered (100 out of 150 questions). So, this highly dialogic style of teaching was a prominent feature of two of the three accessible lectures in the sample, likely contributing to the students’ activation and more deeply engaging with the lecture contents.
“in in the other end of the of the piece”
The final interactional feature that stood out in the data was repetition. Jaana distinguishes between voluntary repetition, which can be seen as a teaching tool, and involuntary repetition, which is often negatively described as “repair”, “false starts”, or “dysfluency”. Prominent types of deliberate rephrasing were lexical (a verbatim repetition of a lexical item) and rhetorical, which could involve paraphrase, contrast, or didactic repetition of an audience comment.
Taking these deliberate, pedagogically oriented repetitions first, they were much more prominent in the accessible lectures, with 398 deliberate repetitions compared to 156 in the challenging lectures. At this point, this finding could be expected. But an unexpected finding comes out of the unintentional repetition data. These so-called “false starts” are conventionally associated with “dysfluency” and at least indirectly with low language proficiency. Consider this extract from one of the lectures in the data:
a real piece of wood it might be so that some of the in in the other end of the of the piece the the the we are in the fiber saturation point
(Suviniitty 2012: 153)
These repetitions and re-starts look a lot worse in print than in speech; they’re also an ordinary part of first language speaking, and we normally don’t pay much attention to them. This example comes from one of the accessible lectures, and it turns out that the unintentional repetitions were more common in the accessible lectures overall: 244 occurrences compared to 157 in the challenging lectures. Looking more closely at the most challenging and most accessible lecture in the sample, the difference is striking. The most challenging lecture contained only 16 instances of unintentional repetition, the most accessible lecture having 85 (Suviniitty 2012: 149).
So while the more accessible lectures employ more interactional features such as control acts, questions, and repetitions, it seems they are also more “dysfluent”. But as Jaana points out, these involuntary repetitions serve to buy some time for the speaker to formulate her/his thoughts, and they simultaneously provide the listeners some brief moments to process the speech that has been uttered up to that point. If unintentional repetitions are seen in this light, they could actually facilitate comprehension by providing “padding” between the main chunks of information.
But aren’t they better in Finnish?
It’s naturally assumed that lecturers must “perform” better in their first language, so we might expect to find more interactional features in the lectures in Finnish. To investigate this question, Jaana also sampled a lecture in Finnish by the most challenging and most accessible lecturer in her data. The most challenging lecturer employed a similarly low number of interactional features in Finnish as in English. But the biggest difference was found in the most accessible lecturer, who addressed 78 questions to the audience in English, but only 30 in Finnish. Most interestingly, these 78 questions included 52 acts of didactic elicitation, where the teacher invites and expects substantial input from the students. In his Finnish lecture there were none (Suviniitty 2012: 167-169).
Another interesting finding concerned deliberate repetition – the lexical and rhetorical rephrasings discussed above. Both lecturers had higher numbers of these in their English lectures, but again the most accessible lecturer showed the biggest difference, with 114 deliberate repetitions in English and only nine in his Finnish lecture. Finally, the unintentional repetitions also differed between the lecturers. Oddly enough, the most challenging lecturer had relatively few unintentional repetitions in English (16) or Finnish (9), suggesting a very fluent and planned delivery. The most accessible lecturer, on the other hand, was relatively dysfluent in both languages, with 85 unintentional repetitions in English and 99 of these in Finnish (Suviniitty 2012: 169).
What I take from these findings is that dysfluency shouldn’t be seen as something peculiar to second-language users, nor should it automatically be evaluated negatively. There are cognitive pressures for speakers and listeners alike in any language, and both tasks are buffered by naturally placed dysfluency features such as re-starts and repetitions. In a lingua franca setting, these hesitations are likely a greater aid to comprehension than an impediment (see also Mauranen 2012: 204-232 for a discussion on repetition and comprehensibility in ELF).
The big picture
To bring these variables together in a visual way, Jaana plotted her broad findings onto a chart (click on Figure 1 to enlarge) with all 22 of the lectures in her data arranged by the students’ comprehension value average, depicted as the ascending blue line in the chart. The red dots correspond to the teachers’ English skill average (as reported by the students) on the left-hand side of the chart. As you can see, there was no clear correspondence between students’ reports of accessibility and the teachers’ English skills. In fact, the two lowest English skill averages in the data were lecture no. 02, which was Jaana’s outlier case for “most challenging” lecture, and lecture no. 21, which surprisingly was the “most accessible” outlier.
So what’s the difference? Jaana finally plotted the number of interactive features in these six lectures which were examined in depth; counts are shown on the right side of the chart, and the black squares are aligned with the lectures they come from. Now we can see that not only are there more interactive features on the accessible side of the chart, but two of the challenging lectures have been rated relatively high by students for English skills. The clear difference between these and the more accessible lecture no. 21 is the abundance of interactive features that likely contribute to students’ involvement and comprehension in a lecture.
To conclude, I think Jaana’s study has done a good job of showing what can be learned with a data-driven approach. By letting the data initially speak for itself, the investigation can be narrowed to a detailed analysis of prominent features. As a detailed, qualitative analysis of six lectures in ELF, several important features have emerged that would be interesting to test quantitatively on a larger set of data. Most importantly, this study provides an important view on good practice in international higher education, where the ability to engage students in pedagogically relevant interaction might be more important than just “good English” alone.
For the rest of the story on Jaana’s research, see these previous posts:
- Part 1 – Good ELF in English-medium instruction
- Part 2 – Let’s draw some wood cells: control acts & accessible lectures in ELF
Mauranen, Anna (2012). Exploring ELF: Academic English Shaped by Non-native Speakers. Cambridge Applied Linguistics series. ISBN: 9780521177528.
Suviniitty, Jaana (2012). Lectures in English as a Lingua Franca: Interactional Features. Doctoral dissertation, University of Helsinki. Available online: http://urn.fi/URN:ISBN:978-952-10-8540-6.