The early research carried out by Cobuild and other corpus linguists produced dramatic results showing how native speakers' intuitions about the English language are often incorrect. The rapid growth in computing technology in the 1980s allowed linguists to obtain significant amounts of real language data drawn from large computer-held corpora and when this data was studied, it was clear that any true and accurate description of English must be firmly rooted in direct observation of facts, rather than the unreliable mix of impressions, prejudices, unconscious assumptions and idiosyncrasies which shape our intuitions about language.
The Cobuild dictionaries therefore
They're just trying to score brownie points with politicians
The boss is pleased -- that's another brownie point
One of these is an authentic example taken from a corpus. Can you spot which one it is?
It's often difficult to identify made up examples, particularly if you are not a native speaker of English! In the Bank of English corpus, the singular brownie point occurs only 6 times while the plural brownie points occurs 76 times. So the example above which shows the plural form is showing a more natural use of the phrase.
And the most frequent verb which is used with brownie points in the Bank of English is score. Earn is the second most frequent.
The dictionary which presents the second example of the two above is really not showing the learner how the phrase is most typically used in modern English.
One reason why dictionaries that claim to based on a corpus do not use
real examples is because they do not have enough examples to choose
from! The Bank of English is over 300 million words and growing month
by month. This is important, because it allows our dictionary editors
to select one or two real examples from a wide range of instances in
the corpus. That way they are able to choose examples which are
authentic, typical and yet not confusing or difficult for a non-native
speaker. If the corpus is small and restricted it is likely that you
will not have any good examples available to select: so you are forced
to invent an example to meet the need.
Secondly, we tell the user in plain and simple English that the word
is "an informal word". We think this is clearer than using another
special code (such as (infml)) somewhere hidden in the
dictionary entry.
Don't bother trying to compare Cobuild's entry for cutesy with
other dictionaries for learners.....this word seems to have been
missed from most of them.
Text based on material produced by Cobuild.Explanations rather than definitions
Here is the Cobuild explanation for the word cutesy:
If you describe someone or something as cutesy, you dislike
them because you think they are unpleasantly pretty and sentimental;
an informal word.
Why didn't we give a shorter definition? Such as:
cutesy: pretty and sentimental.
Because the short definition (while it saves space in a printed book)
leaves out too much important information about this word. First, we
tell the learner that this word shows a particular attitude taken by
the person who uses the word. If I call a child fair-haired
then native English speakers will not make any assumptions about my
attitude towards the child: the adjective fair-haired contains
no PRAGMATIC force. If we wrote a definition of cutesy like the
second one shown above, the learner of English might describe a
friend's new baby as cutesy without being aware that this will
probably cause offence! Some dictionaries simply place some special
code (such as (oft. derog.)) to indicate this pragmatic
meaning. In the Cobuild dictionary we explain this fully in the
definition. By the way, we also put a large PRAGMATICS label alongside
this explanation in the extra column in the margin.