The Lexical Approach
There are two key references in this area, both by Michael Lewis:
The Lexical Approach, 1993 and Implementing the Lexical Approach: Putting Theory
Into Practice, 1997
You should refer to those for more detail. What follows is only an overview and, like all such things, does injury in trying to be concise.
TheoryLanguage consists of grammaticalised lexis, not
lexicalised grammar
|
Chunks |
The theory behind the Lexical Approach hinges on the concept of
language chunks. Briefly, a language chunk is a group of words
which are habitually found together. That definition will include
common collocations such as air-conditioning + unit, dry + stone +
wall, steering + wheel, hope for the best, certain people and thousands of others, of course, as well
as fixed expressions such as
I would like
Do you mind if
How are
you
etc.
(For more,
go to the guide to collocation, linked in the list of related guides at
the end.)
However, it will also include groups of words which are not normally seen
as collocations such as
look at
just last week
tomorrow afternoon
upside down
right way up
out of sorts
have you heard ...?
etc. Some of these
are multi-word verbs, some fixed phrases and some would be called
idioms. They all count as lexical chunks and are deployed as
single ideas.
It has long been recognised that native speakers have a huge pool of
language chunks to draw on which serve to cut down on the time we
need to process thought into language and this aids fluency. The
theory is that we don't speak fluently by using our knowledge of grammar
and then slotting in the appropriate words to make the sense we choose.
What we do is select prefabricated chunks and use our knowledge of
grammar in a subsidiary, management role to help the language along.
This is what Lewis means by Language consists of grammaticalised lexis, not
lexicalised grammar.
Can you pick out the language chunks in the last paragraph? Click here when you have a few noted down.
There are lots and lots:
- native speakers
- to draw on
- to cut down on
- The theory is
- slotting in
- appropriate words
- use our knowledge
and so on.
There are a number of terms for chunks like these and you'll come across expressions such as binomials, semi-fixed / fixed expressions, holophrases, polywords and so on.
In spoken language in particular, where time pressures on speakers are high, the use of prefabricated lexical chunks oils the wheels and allows speakers to produce fluent language. It is also the case that grammar rules are often relaxed in informal speech and we are, therefore, more dependent on lexis. For example, how did the speakers in this dialogue produce their utterances?
- Did you pick up the shopping on your way home?
- No, I clear forgot about it. Sorry about that.
Click here when you have some ideas.
In this exchange, it seems likely, according to the theory that
the speakers did not set out with a grammatical utterance in mind.
To apply the grammar rules in this tiny exchange would simply take
too long.
For speaker A, the application of grammar rules would be something
like:
- form a question with an operator do in the right tense (did)
- insert verb plus adverb particle (pick up)
- choose the article determiner for known reference (the)
- form a prepositional phrase (on ... way)
- insert correct genitive determiner (your)
- choose a zero article as this is the generic use of home in this case
and for speaker B:
- insert negative
- form past tense of forget
- select the correct preposition (about)
- insert adverb to emphasise the verb but remove the -ly ending
- insert appropriate pronoun (in the singular because shopping is uncountable)
- insert apology with the adjective sorry, ellipting the copular verb and the subject (I am)
- insert appropriate anaphoric reference demonstrative pronoun to refer to the previous clause (that)
To do all that would require a lot of very fast processing.
What is probably happening is that both speakers have stored
expressions such as
on your way home as
on + genitive + way home
or
pick up the shopping as
pick up + article/determiner + uncountable
noun.
Processing the language this way, in prefabricated chunks allows the
speaker to form accurate language quickly without the need to apply
complex grammar rules.
Learners at all levels, the theory implies, can do that, too.
Semantic relationships |
Clearly, relationships such as collocation will be
central to this approach but it goes beyond that.
For example,
if we take an expression such as
pick up the shopping
we
can readily see that a whole range of mass nouns can be substituted
with the same grammatical relationship to the verb and there are
also some other verbs that can have the same relationship with
up as pick has. Furthermore, the pronoun
reference will remain the same as will the matter of separability (we
can't have *pick up it etc.). We can, therefore,
have:
- pick up the furniture / information / gossip / news
- pick it up
- gather up the crowd / data / flock
- gather it up
- clear up the mess / spillage / rubbish
- clear it up
and so on.
This close relationship between lexis and
grammar (colligation) is another key element of the Lexical Approach.
For more on colligation,
go to the guide, linked at the end.
Delexicalised verbs |
Allied to the idea above about the relationships between grammar
and lexis is the concept that some verbs are 'delexicalised'.
The most common of these in English are:
do | have | get | go | make | put | set |
take
A glance at the dictionary entry for any of these will tell you
what's odd about them. For each of these verbs, an online
dictionary entry runs to several pages. In the case of get,
for example, the online Oxford dictionary lists:
receive | experience | contract | attain |
fetch | prepare | find | travel by | obtain | contact | reach
and so on before we even get to (!) the uses with prepositions and
adverb particles.
(Source: https://www.oxforddictionaries.com/definition/english/get)
What is meant by delexicalisation, in Lewis' view, is that the verbs in question
do not carry an intrinsic meaning. What happens is that they
acquire their meaning from the noun phrase with which they collocate.
Many of these verbs refer to performing an action so we get, for
example:
hold a meeting
run a meeting
give permission
take a look
take a chance
make apologies
take pity on
throw a party
do wonders
give thanks
etc.
There is no semantic choice to be made here in terms of which verb
collocates with which noun phrase.
Naturally, all these verbs also have lexical meanings. For
example:
hold = grasp in the hand
run = move quickly / flow
give = present
throw = project through the air
take = move to another place
and so on.
The point being made is that
in these uses they are delexicalised.
Given this demoralising range of meanings, it is little surprise
that learners sometimes despair of ever understanding what the verbs
mean (because, in fact, they don't mean anything much).
All the verbs in the list exhibit the same phenomenon.
The problem for teaching then is to encourage our learners to notice
the variety of meanings that any of these verbs can exhibit and
notice, too, that it is second half of the chunk, such as
make
the beds
do the cooking
get the message
have a bath
go bad
put
in place
set the clock
take a minute
etc. which actually carries the meaning.
The use of delexicalised verbs is often semi-opaque. That
is to say that while we can easily see the meanings behind, for
example:
do the cooking
make dinner
the choice of the verb is not to do with its meaning. It
is the meaning of the noun phrase which determines the meaning of
the clause and the choice of verb may seem random.
However, opacity is not an on-off phenomenon because some verbs may
be seen as semi-delexicalised. For example in:
give thanks
pay a compliment
take an interest
set an example
etc., there is some sense in all cases of the verb's meaning
although it is still very difficult to guess which verb will form
the appropriate collocation.
It is also the case that even usually delexicalised verbs can be
used metaphorically so, for example:
Take the car to the garage
is the normal lexical use of the verb meaning moving to another
place, but:
Take a hammer to it
has a wholly different, figurative meaning of:
Hit it with a hammer.
If you dislike the term 'delexicalised verb', an alternative is
'empty verb' and you can get a list of them
here.
There is also a lesson for B1 / B2-level learners on delexicalised
verbs
here (new tab).
The concept of delexicalised verbs is not a new one (although the
term is new) because the idea of prime verbs in languages has been
around for a while. The prime verbs in English are usually
listed as:
be | bring | come | do | get |
give | go | keep | make | put | take
and 5 of those are common to both lists.
Besides the delexicalised nature of many of these verbs in certain
collocations, they are also the verbs which are basic to most
idiomatic language and which often take the place of more formal or
synthetic verbs. So, for example:
We can render ... | ... as this with a prime verb |
He appeared suddenly | He was suddenly there |
They have raised four children | They have brought up four children |
He attended the meeting | He came to the meeting |
I executed her instructions | I did as she told me |
I arrived at the hotel late | I got to the hotel late |
I handed in my essay | I gave my essay in |
He travelled to New York | He went to New York |
Please retain the receipt | Please keep the receipt |
I prepared dinner | I made dinner |
She garaged the car | She put the car in the garage |
I caught the train | I took the train |
Communicative efficiency |
Schmitt, 2000:400, states the case like this:
There is a good psycholinguistic basis for believing that the mind stores and processes these chunks as individual wholes. The main reason stems from the structure of the mind itself. It can store vast amounts of knowledge in long-term memory, but is only able to process small amounts of it in real time, such as when one is speaking. In effect, the mind makes use of a relatively abundant resource (long-term memory) to compensate for a relative lack in another (processing capacity) by storing a number of frequently-needed lexical chunks as individual whole units. These can be easily retrieved and used without the need to compose them on-line through word selection and grammatical sequencing. This means there is less demand on cognitive capacity
If this is true, then learners will be better able to communicate
if they are equipped with lexical chunks rather than grammatical
knowledge. And there's another reason: lexical chunks form an
integral part of functional language use.
For example, expressions like
Can you let me have ... ?
are
both produced and perceived as if they were single lexemes, cutting
out the need for the speaker and
the listener to process them word by word. The learner who
produces this does not need to understand its constituent parts, nor
is it necessary to understand the meaning of any of the five words
when heard. All that is needed is the knowledge that it is a
polite request for something to be given to the speaker.
Theory of learning |
Once the chunk has been acquired, goes the theory, the learner
can analyse it at leisure and notice that can may be
replaced by other modal auxiliary verbs such as would to vary the meaning
and let me have can be replaced by any number of phrases.
This means we can generate, e.g.:
Would you pass the salt?
Could you open the door?
and so on, virtually ad infinitum.
Thus the grammar is acquired through the medium of lexical-chunk
learning. In other words, grammar is not learned by combining
small units into longer ones but by breaking longer units down into
smaller ones: an analytic rather than a synthetic process.
A lexical syllabus |
Once the focus is taken away from language structure and grammar
and placed on the lexical systems of the language, the next decision
that needs to be made is what to include in a teaching programme.
This is not an uncontroversial area and different practitioners of a
lexical approach (whether whole- or half-hearted) will draw up lists
in quite different ways.
Nevertheless, there is some consensus that the following will form
the content of the syllabus.
- words / lexemes
This is not surprising but nor is it a great departure from all syllabuses which must by nature include some attention to the lexicon of the target language.
What is different is that this is the starting point for a syllabus, not an add-on module. - polywords
These are mini-expressions such as by the way, upside down, back and forth, out and out, insofar as and so on.
Many analyses (including the one on this site) will refer to these as fixed idioms but a lexical approach broadens the definition to include phrases which are transparent in meaning, unlike most idioms, but are treated as single concepts. That is, of course, also part of the definition of a lexeme.
Many analyses place multi-word verbs in this category whether they are phrasal, prepositional or phrasal-prepositional verbs. - collocations
These are familiar to most people in the profession and may be where the commitment to a lexical approach starts and stops.
This will include items such as:
vested interest
heavy rain
fully sure
ocean-going liner
and thousands more weak, strong or semi-fixed collocations blurring into compounds as the last example shows. - institutionalised utterances
These are exemplified as expressions such as:
We'll see
I could use some help
How will it end?
What's the matter?
and thousands of other clichéd expressions which are used as single items to free up cognitive processing space for other content. - sentence frames
These are ways of constructing more complex thoughts around a structure already acquired and re-used many times. Examples include:
This is more ... than you ...
... combined with .... make(s) ...
... is pretty much the same as ...
and so on.
The difficulty here is not identifying routinised sentence frames; it is selecting those most frequently found in the language and most useful for learners to internalise. Corpus studies can help considerably. - text frames
These are an extension of sentence frames but operate at the level of longer texts. They might include:
Let me tell you about ... Well, firstly, of course, and then ... but in the end ...
In this essay I will set out ... but will exclude from consideration ... The starting point is ... Now we need to consider ...
and so on.
Such text frames are of particular use to those who need to write more formally and adhere to conventional information staging patterns in specific genres.
Practice |
If you accept all, or even some, of the theory outlined here, it
will make sense to you to focus learners much more on lexical
chunking than on the grammar of the language.
Implementing a lexical approach, in full or as an adjunct to a more
traditional approach does not imply a major shift in methodology or
techniques in the classroom. What does alter dramatically, is
the content and focus of the teaching and learning.
It is, so to speak, a change in pedagogical mind sets, not a new
methodology.
Noticing |
It is clearly undesirable to try to teach the many thousands of lexical chunks that a native speaker commands so proponents of a lexical
approach lay emphasis on noticing. Learners need to
notice that certain combinations of words perform single functions
or represent single ideas as in the examples above. Once they
have done that, they can then go on to analyse the chunk and in
doing so acquire the grammar.
Noticing can be encouraged in a number of ways (and there is
a guide to noticing on
this site, linked at the end in the list of related guides):
- Teacher-led noticing
The teacher deliberately uses models in the classroom which are rich in transferable lexical chunks and highlights these as the teaching / learning targets for the lesson. Typically, this is done through a listening or reading text with the items emphasised in some way.
The learners go on to using the chunks in authentic, communicative tasks and finally to analyse them. - Student-led noticing
With practice and some familiarity with the approach, learners can become more independent and take examples of language in use to focus themselves on the potential lexical chunks it contains.
Then they can incorporate them into their own production and go on to analyse them at leisure. - Reformulation
The teacher can reformulate student output (either spoken or written) to focus the learners on lexical chunks. So, for example, the production of
I cooked the dinner
I arrived at the station
or
I received a present
can be reformulated more naturally as
I made the dinner
I got to the station
and
I got a present
respectively.
By the same token, the production of
I'm going to tell you about ...
can be reformulated as
Have you heard that ...
and so on.
Criticisms |
A number of criticisms have been made. Among them are:
- Implementing a lexical approach will produce learners whose speech is limited to a range of clichés and who will not have the language means to deal with new or unexpected topics and functional demands.
- The lexical approach ignores the way second languages are actually learned and that is by understanding the nature of grammar and using this competence to produce novel and accurate utterances. Acquiring lexical chunks is part of this, not the origin of it.
- Claims for the efficacy of a lexical syllabus are not supported by empirical evidence.
- A lexical approach is, in fact, already included in most communicative language-learning approaches and there's nothing new here.
- A good deal of the theoretical work on a lexical approach is simply re-labelling the already well known. Just calling some verbs delexicalised does not alter fundamental issues of colligation and collocation, for example.
- What Lewis and others are actually describing is a long-attested tendency in many languages to grammaticalise words, bleaching them of their lexical meaning and converting them to function words. In other words, nothing new is being described that warrants a change in teaching approach. For examples of how this has happened in the case of words like will, indeed, let, go etc., see the guide to the roots of English, linked below.
The most telling criticism of The Lexical Approach is however,
that it is not an approach at all, leave alone a methodology.
It has very little to say about how to structure teaching in terms
of a settled syllabus (although the effort is being made, see above), almost nothing to say about how to teach
above the level of raising learners' awareness of the usability and
efficiency of language chunks and little that one could call a
theory of learning (other than that it involves some serious
memorisation of items rather than rules).
It is, as Richards and Rodgers (2001:138) note,
an idea in search
of an approach and a methodology.
Related guides | |
empty or delexicalised verbs | for a list in PDF format |
a lesson on delexicalised verbs | this is in the learners' section and opens in a new tab |
collocation: essentials | for the simpler guide to this area in the initial plus section |
collocation | for a more detailed guide in the in-service area |
colligation | what it is, how it differs from collocation and some implications for the classroom |
idiomaticity | for more about lexicalised phrases, fixed expressions and so on |
noticing | for more on a key teaching technique |
lexis | the link to the guides to understanding lexis |
roots of English | for a discussion of grammaticalisation |
methodology | the link to the methodology index |
References:
Lewis, M, 1993, The Lexical Approach, Hove, UK: Language
Teaching Publications
Lewis, M, 1997, Implementing the Lexical Approach: Putting Theory
Into Practice, Hove, UK: Language Teaching Publications
Richards, J, and Rodgers, T, 2001, Approaches and Methods in
Language Teaching (2nd edition), Cambridge: Cambridge
University Press.
Schmitt, N, 2000, Key Concepts in ELT: Lexical Chunks,
ELT Journal 54(4): 400-401, Oxford: Oxford University Press