An Introduction to Applied Linguistics

particularly enlightening study was carried out by Ross (1997), who tested the 
view that listeners at different levels of L2 proficiency adopt different processing 
strategies. The task Ross set his Japanese students was to listen to a recorded 
message in English and match it with one of a number of icons. He then asked 
them to introspect about the reasons behind their icon selection. He found 
that weaker listeners tended to focus on a key word, produce an initial mental 
model and stick to it, without searching for confirming clues; the more proficient 
listeners also identified the key word but continued actively to search for further 
clues in the rest of the message. 

---

190 An Introduction to Applied Linguistics
Introspection studies are open to three main criticisms. First, the demands of 
on-line reporting may lead listeners to listen differently from normal. Second, the 
data obtained can be greatly influenced by the listeners’ skill in verbalizing mental 
processes, especially if the self-reporting is done in the L2. Third, listeners’ reports 
may reflect prior knowledge, rather than their listening. These last two problems 
can be reduced, for example, by allowing subjects to report in L1, or by selecting 
unfamiliar topics, but researchers using introspective approaches have had greater 
difficulty in addressing the problem of interference in normal comprehension 
processes. 
Retrospection
An alternative method of throwing light on listening is retrospection, in which 
the listener is asked to recall the experience of comprehending some time later, 
usually prompted by memory support such as reviewing a recording of the 
original conversation. Wu (1998) used a retrospective approach to investigate 
the relationship between linguistic processing and listeners’ use of background 
knowledge. Chinese learners of English were played a three-minute text twice – 
once all the way through as they completed multiple-choice questions on content 
and then in sections. After each section, they were asked to recall their route 
to comprehension and their strategies for dealing with problems. Wu concluded 
that linguistic (bottom-up) processing was basic to successful comprehension; 
failure or partial success at the linguistic level can lead listeners to allow schematic 
knowledge to dominate their interpretation – as in other studies we have 
mentioned (Ross, 1997; Tsui and Fullilove, 1998). 
As with introspection tasks, there is a risk of ‘contamination’ in retrospection: 
subjects asked to recall how they understood a text may elaborate what they 
actually understood first time. However, as with the listening models we discussed 
earlier, the three methods summarized in this section – observation, introspection 
and retrospection – need not be mutually exclusive. Applying them in judicious 
combination is probably the best approach to finding out how individuals listen 
and how they deal with comprehension problems. 
From Theory to Practice: Issues in Teaching L2 
Listening 
The point of contact between theory and application is to be found in the work 
on learning strategies. We have already mentioned Mendelsohn’s strategy-based 
approach. Field (1998: 12) suggests what he calls a ‘diagnostic approach’, in which 
a listening lesson would involve pre-listening, listening and then an extended 
post-listening session ‘in which gaps in the learners’ listening skills could be 
examined and redressed through short micro-listening exercises’. Despite minor 
differences, Mendelsohn and Field are both advocating teaching learners how to 
listen.
Berne (1996, 1998) examined the relationship between the theory and the 
practice of acquiring listening competence, and found that – other than the 
mutual interest in listening strategies mentioned above – there was a disturbing 
mis-match between researchers’ and classroom practitioners’ interests. Better 
communication between the two sides is essential to help close this gap (for 
further discussion on this point, see Mendelsohn, 1998, 2001). 

---

191
Listening
Difficulty Factors in Listening
An examination of traditional listening comprehension materials for L2 learners 
shows that the selection of texts was not systematically based on criteria of 
difficulty. This resulted in a situation in which the materials used to teach listening 
were often unsuitable, and the tasks assigned after listening were inappropriate 
for the text or for the particular needs of the learners in question. Attention was 
usually given primarily to the appropriacy of the ‘topic’ rather than to other 
aspects. 
Research over the past number of years has attempted to define which 
factors contribute to making a particular listening passage difficult or easy to 
comprehend. Building on literature surveys such as Rubin (1994), Brown (1995a) 
and Lynch (1998), Buck (2001) distilled researchers’ findings into the following 
list of characteristics that affect listening:
INPUT CHARACTERISTICS
Language
• Speech rate.
• Unfamiliar accent.
• Number of speakers.
• Similarity of voices.
• Use of less frequent vocabulary.
• Grammatical complexity.
• Embedded idea units.
• Complex pronoun reference.
Explicitness
• Implicit ideas.
• Lack of redundancy.
Organization
• Events narrated out of natural time order.
• Examples preceding the point they illustrate.
Content
• Unfamiliar topics.
• Number of things and people referred to.
• Unclear indication of the relative importance of protagonists in the text.
• Shifting relationships between protagonists.
• Abstract content.
Context
• Lack of visual or other support.
TASK CHARACTERISTICS
Tasks tend to be more difficult when they require:
• Processing of more details.
• Integration of information from different parts of the text.
• Recall of gist (for example, writing a summary) rather than exact content.
• Separation of fact from opinion.
• Recall of non-central or irrelevant details.
• A delayed response, rather than an immediate one.
(Adapted from Buck 2001: 149–151.)
Table 11.3 Sources of difficulty in second language listening 

---

192 An Introduction to Applied Linguistics
Authenticity of Text and Task 
Authenticity of Text
The debate over the use of authentic materials in language teaching has occupied 
methodologists for at least three decades (Geddes and White, 1979; Porter and 
Roberts, 1981). ‘Authentic’ in this context is generally defined as ‘not designed 
or recorded for non-native speakers, or for language learning purposes’. Early 
in the professional debate over authenticity of learning materials, Widdowson 
proposed a separation of two different aspects of language in use: ‘genuineness’ 
and ‘authenticity’ (Widdowson, 1979). He argued that that a text was ‘genuine’ 
if it contained the sort of language typical of that genre in actual use, and 
that it did not matter (for learners or teachers) whether it had occurred in real 
communication. The term ‘authentic’, on the other hand, Widdowson reserved 
for the appropriacy of the response from the listener or reader. In other words, 
genuineness was related to text; authenticity was related to task. A text could 
therefore be genuine even if it had been ‘invented’ for teaching purposes rather 
than ‘discovered’ in actual use.
Although most classroom practitioners would agree that authenticity is 
desirable (Rings, 1986; Field, 1998; Mendelsohn, 2001), some have gone too far 
in their demand for authenticity. Recently Richards criticized what he called 
the ‘myth’ of authenticity (Richards, 2007); echoing Widdowson’s article of 
three decades ago, Richards argued that it is unrealistic and unnecessary to base 
L2 instruction on ‘authentic’ texts, since authentic listening texts are largely 
unusable, given the logistical difficulties of recording conversations, as well 
as the legal and ethical problems of getting informed consent of the people 
recorded. Both Widdowson and Richards take the position that authenticity is 
the ‘end’ of language teaching, but need not be the ‘means’. Listening teachers 
need to bring learners to the point where they can understand, interpret and 
respond to L2 listening input in the way that the original speaker intended. 
However, we should not be over-concerned with finding real texts; ‘realistic’ 
texts will do just as well, provided they are used in a way that helps learners to 
respond to them appropriately.
Authenticity of Task
With the advent of a primarily communicative focus in L2 instruction, many 
materials developers and teachers have aimed to make learning tasks as realistic 
as possible. One example of this is the ‘information gap’ task, which cannot be 
completed unless the learners share the relevant different bits of information in 
their possession. However, similar caveats are required here to those stated in the 
discussion of text authenticity above: that is, it may be necessary to help learners 
approach fully authentic tasks gradually, rather than trying to make tasks lifelike 
from the beginning. 
Strategy instruction 
Strategy instruction is not revolutionary or faddish. Neither is it something that 
should be viewed as an ‘add-on’ to what happens in the listening lesson. Strategy 
instruction is at the root of teaching learners how to tackle a listening text. It 
involves showing learners the clues to getting at meaning when gaps in their 

---

193
Listening
competence make that difficult. Moreover, strategy instruction can contribute 
significantly to learner autonomy.
Mendelsohn (1994), as part of his strategy-based approach, offers examples of 
strategies to determine setting (S), interpersonal relationships (I), mood (M) and 
topic (T) (SIMT) arguing that this facilitates comprehension. Setting relates to 
‘where’ and ‘when’, interpersonal relations relate to ‘who’, mood and atmosphere 
relate to ‘how’ things are being said (the tone), and topic relates to ‘what’ is being 
said and ‘why’. One example of a strategy to determine mood and atmosphere 
is training students to listen for changes in voice quality. They might hear the 
following dialogue twice: the first time with normal voice, the second, with 
‘whispery voice’: