TOP DOWN & BOTTOM UP ESSAY

In recent years there have been two major approaches to explaining the listening process –
rather unfortunately called the top-down and bottom-up approaches. The bottom-up approach
sees comprehension as a matter of listeners first decoding (or understanding) the smallest
elements of what they hear – the sounds. /p/ is recognised as being /p/ and not /b/, /i:/ as
being /i:/ and not /i/ or /e/ and so on. These sounds are then combined and the individual
words are decoded – the listener recognises that s/he has heard /pi:t/ and not /pit/ /bit/ /bi:t/
/bi:d/ or some other word. The words are then combined into sentences and the listener works
out the meaning of /pi:t/ : as in I saw Pete yesterday or I bought some peat for the garden. To
this will be added recognition of features such as intonation and so on, until we finally reach
the non-linguistic context.

The top-down approach starts from the opposite end : it sees understanding as starting from
the listener’s background knowledge of the non-linguistic context and of working down
towards the individual sounds. Listeners will actively interpret what they hear in terms of
their understanding of the situation and the world in general. For example, imagine I tell you :

McKenzy brought me another present today. It was too late to save it so I buried it in the
garden. I think I’m going to have to put a bell round his neck.

You will certainly understand all the words in this passage, but do you understand the
meaning? Think back to what happened as you read. The first sentence probably went quite
smoothly. But there was more than just decoding of words going on. Without your even
being aware of it, subconscious expectations were forming in your mind based on your
knowledge of the world – McKenzy is probably a friend, probably a man as only the surname
is used, the present will be something nice etc. The existence of these presuppositions is
shown by the fact that you probably did a double-take when you got to the second sentence –
buried it? Eh?? And at that point you will have started to search quite consciously for the
meaning.

Maybe by the end you’d worked it out. If so, then notice that it was your knowledge of the
world which helped you understand – not what’s in the text. Or maybe you’re still in the
dark. I can help you by giving you some contextual or situational knowledge : McKenzy is
the name of my cat. Combine that with your knowledge of the world (the habit cats have of
bringing their owners “presents” of half dead birds and mice which they’ve caught, and the
fact that the noise of a bell will prevent the cat from creeping up on them unheard) and you
have the meaning of the passage.

However much help you did or didn’t need, you can see that in understanding the passage a
lot more was going on than just passively decoding the sounds (or in this case letters, as you
were reading it - but the principle is the same) then the words, then the sentences. Your mind
was working actively to interpret the passage, and using a large amount of non-textual
information to do so. And how easy it was will depend on how close to the forefront of your
mind that information was. If, as you read the passage, your cat was sitting on your lap, you
probably tuned in immediately. If you have never owned a cat, it may have taken longer.
In recent years it has been the chief approach to listening comprehension in the EFL
classroom, and has led to teachers telling students things such as You don’t need to
understand every word, What would you expect him to say? or Try and identify the main
ideas and guess the rest.

I’m not trying to suggest that this is not a valid approach. It is. The switch to a top-down
approach was a necessary change from the exaggerated bottom-up approach which in some
cases remained current in foreign language teaching as late as the 1960s. In this approach, the
learner’s listening ability was seen as being evidenced by his or her ability to take down a
dictated paragraph in exactly the same form as it was read out, or to answer detailed
comprehension questions on a written passage read by the teacher. With no exposure to the
natural features of spoken language, and with no training in the type of listening strategies
emphasised in the top-down approach, it was little wonder that even supposedly advanced
level students returned from their first trip to Britain, the States etc saying I didn’t understand
a word anyone said!

Since the mid-20th century, things have changed. The advent of the tape recorder meant that
recorded dialogues could be used in the classroom for the first time. Research into the spoken
language and the invention of video as well as audio tape led to the use of authentic and semi-
authentic materials which incorporated the type of features of the spoken language which I
focused on in Why Don’t They Understand? And finally, research into the listening process
itself led to our recognition of the importance of top-down processing.

But have we gone too far? Students still come back from their trips saying I didn’t understand
a word and still frequently hate doing listening comprehension in the classroom. However
often the teacher says If you’ve been able to answer the questions then you’ve understood the
text, they don’t really believe it. They know there were chunks that they were unable to
decode, and feel insecure – in the worst case scenario losing confidence in the teacher or the
course.

Why is this happening? I would suggest that for a while we went too far over to the other
extreme, and have often forgotten that, even if they start from the top, students still need to
get to the bottom. Listening lessons have tended to stop half-way. We help students apply
knowledge of the world and contextual knowledge to the text. We encourage them to focus
on what they do understand rather than what they don’t by teaching them to focus on key
words and infer connections. All this is valid and necessary. But what about the rest? I would
argue that if we are going to help students improve their comprehension, we do need to focus
on what they don’t understand. For example, if the item that blocked comprehension was a
weak form we need to help them analyse the pronunciation features so that they will be more
ready for it the next time. For while it is true that native speakers don’t “hear” every word
either, there’s a difference. If, as a native speaker I hear a sequence of sounds something like
umgernaseeyimlader or eelerbinsurprised, I have no trouble decoding those sentences as I’m
going to see him later and He’ll have been surprised, even though I can’t be said, for
example, to have heard the words to, will or have – they simply weren’t there. But as a native
speaker I have a non-conscious knowledge of both the phonological features and the
grammar of the language. The first, for example, means that I expect him to be pronounced
/im/, and know that if /im/ is preceeded by a vowel, the linking consonant /j/ will be inserted.
So when I hear what is apparently “yim”, I have no problem decoding it as him. Decoding
“er” (the schwa sound) as have is also a matter of knowing that have is often pronounced like
this in a unstressed position, but it is also helped by my knowledge of grammar : I know that
if I’ve decoded will and been, then have must be in the middle, whether I hear it or not.

Non-consciously then, my native-speaker brain is working actively to interpret the sounds

The problem for our learners of course is that they don’t have native speaker competence and
therefore their brains can’t “fill in the gaps” like this. Which is why, using the top-down
approach, we need to help them develop other listening strategies to the full, to help them to
compensate. However, by also taking a bottom-up approach, I think we can help them
improve their ability to decode sounds , words and phrases

I’ll exemplify this in the next article in this series by looking at a possible structure for a
listening lesson which incorporates both approaches. It starts with a top-down approach, but
then moves on to activities aimed at improving students ability to decode sounds, words and
phrases “bottom-up”.