‘If I were you…’:
Language Standards and Corpus Data in EFL
Periodically,
‘standard English’ becomes a topic of lively controversy that staunchly
resists being reconciled. To appreciate why, we might identify two incompatible
conceptions. In the exclusive
conception, standard English is a pure medium with stringent rules and precise
boundaries; all non-standard usages and variations are uniformly classified as errors.
The ‘standards’ are perceived to be undergoing a decline unless
strong measures are taken.
The
exclusive conception encourages TESOL and TEFL to emphasise a strict division
between ‘standard’ versus ‘non-standard’ English. In practice, teaching
and learning are typically arranged to minimise the occurrences of non-standard
English even when the learners’ fluency is insufficient to produce standard
English on their own initiative. So the initiative is denied by enlisting them
in imitative production, such as repeating or writing down the teachers’
utterances, or reading out samples of English from textbooks or worksheets. The
samples sustain their ‘standardness’ by being patently simplified and
uniform, like these I found in real textbooks:
[1]
I hear with my ears.
[2]
The red rose is fine.
[3]
Useful knowledge is desired.
[4]
I like a bedroom with green walls.
Insofar
as this variety of standard English is not found in the everyday language use of
native speakers, I propose to label it non-authentic English, and to
question whether exposure to it, however prolonged, can be realistically
expected to build fluency for the authentic English used by native
speakers.
In
the inclusive conception,
Standard English is one variety bordering upon and overlapping with a family of
‘World Englishes’. Any non-standard usage which widely occurs in some
variety is not an error but a variation, and should be understood from
its social, historical, or geographical motivations. The ‘standards’ are
perceived to be undergoing not a decline but diversification, due to
swift and massive increases in the size and distribution of the world-wide
population of prospective learners of English. If a decline does ensue, then
chiefly because this population cannot gain sufficient exposure to standard
English that is also authentic English.
The
inclusive conception encourages TESOL and TEFL to emphasise the family
resemblances among varieties of English. One bundle of these varieties comprises
‘learner Englishes’, which, though partially non-standard, are nonetheless
systemic. Their errors are natural variations within a
transitional hybridised system where control is shared between the system of
English and the system of the native language. Here are some UAEU fourth-year
student data from self-paced writing assignments, which I have observed to be
typical:
[5]
The UAE desert rich with green plants [= The UAE desert is rich with green
plants]
[6]
Most of people in the world speaking English [= Most people in the world speak
English]
[7]
Teacher may be dirty from head to toes since the chalk. [= A teacher may be
dusty from head to foot/from top to toe because of the chalk.]
[8] Being a politician is appropriate for women because they like argument and debate and can talk hours. [= talk for hours]
These
students are predictably compensating for their limited fluency in authentic
English by extrapolating from their native Arabic. The latter language lacks
‘be’ in the Present [5]; 1
Webster’s Seventh uses Participles in place of Finite Verbs on many
occasions [6]; lacks the Indefinite Article [7]; and uses Nouns in the
Accusative Case as Adverbs of Time [8]. In the collocation ‘head to toes’
[7], the student-produced variation is more logical than Standard English, which
says ‘head to foot’ (why not ‘feet’?) or ‘top to toe’ (why not
‘bottom’?) (COBUILD 670, 1544).2
The
inclusive conception defines the long-range task of TESOL and TEFL as promoting
a convergence between multiple systems: between learner Englishes and standard
English. The convergence must be gradual because of the sheer size and
complexity of the task; but these factors should seem less daunting if we have
maps of the intermediary stages to identify signals of progress.
The
exclusive conception, in contrast, can provide no maps of progress insofar as it
views these stages as arbitrary fluctuations in the levels of error production.
This view totally overlooks the systemic quality of learner Englishes, witness
this magisterial pronouncement: ‘very few speakers limit their aberrancies to
the widely shared features; each individual typically adds in his own speech a
large and idiosyncratic collection of features’ (Prator, 1968: 464). Such a
view reflects a deplorable lack of knowledge and respect regarding learner
Englishes, and does not hold up in the face of systematic documentation (as in
Granger [Ed.], 1998).
The
same view distracts us from appreciating the impact of non-authentic English on
the learning process. The simplified, uniform English thought appropriate for
non-natives readily strays over the border where non-authentic English becomes
non-standard English too, as in these samples (again from real textbooks):
[9]
I differentiated milk from water.
[10]
The plane fell but the loss is small.
[11]
Be polite so that you could be acceptable.
[12]
Would that world peace is permanent.
The
authors were non-native speakers, but then so are many teachers of EFL, some of
whom, finding these samples in a textbook, might let them pass. My point is that
an easy acceptance of non-authentic English can dull our sensitivity for
standards of English that are far more subtle than the routine issues in
pronunciation, orthography, and grammar.
I
would make the same point on a higher level when non-authentic English is used
in theoretical courses in linguistics, such as syntax and semantics. There, the
motive of the textbook authors is different: the linguistic analysis is so
complicated and artificial as to be practicable at all only for simplified
samples of English like the evergreen ‘the man hit the ball’. But the
outcome is much the same: learners are exposed to sentences which do nothing to
enhance their fluency in authentic English. One textbook I encountered for a
fourth-year university course in semantics demonstrated the conceptions of
‘analytic’, ‘synthetic’ and ‘contradictory sentences’ with data
including these:
[13] John’s nine-year old brother is a boy.
[14] The fly was on the wall, so the wall was under the
fly.
[15] John is taller than himself.
[16] That girl is her own mother’s mother.
Here,
the samples are non-authentic because semantics proposes to study meaning whilst
‘deliberately excluding any influence of context or situation of utterance’
(Hurford and Heasley, 1996: 91). Preference goes to samples that neatly divide
up between obviously true and obviously false. But the division is irrelevant to
everyday conversation and also to the tasks of TEFL. We would justly feel absurd
animating our students to go about uttering things like [13-14] and not uttering
things like [15-16].
Authenticity
is a problem whose subtlety and importance have yet to be fully recognised.
Samples which reflect the systemic qualities of learner English would be
authentic in terms of that system, as different from the isolated and truly
‘idiosyncratic’ occurrences when learners panic or make wild, random
guesses. But the difference may not be easy to recognise unless we have large
and systematic samplings of learners Englishes (see now Granger [Ed.], 1998).
Moreover, our students produce many samples which, though not manifesting any
errors or violating any ‘rules’, are not authentic standard English, e.g.
(more UAEU student data):
[17]
I am about to hate my major [= coming to hate my major]
[18]
I will succeed if luck is present [= if I am lucky/if luck is on my side]
[19]
I ask God to make me achieve my ambition [= I pray that God may grant me the
achievement…]
[20]
And we as women, our message is to rear our children excellent rearing. [= And
for us women, the message is to rear our children excellently.]
These
subtler problems again imply a tension among the ‘standards’ of English and
the standards of the home language. Sample [12] reproduces a construction
considered elegant in Classical Arabic, namely the Accusative Absolute, where a
Verb takes a redundant Participial Object formed from the same stem (‘rear a
rearing’). Standard in Arabic too is making one Noun Phrase at the start of a
sentence be the Topic and another be the grammatical Subject, also found in
[12]. What appears to be careless errors is in fact the result of careful
attention to transposed standards.
2.
THEORY AND PRACTICE OF LARGE CORPORA IN EFL
By
my line of argument so far, learners of EFL, and some non-native teachers of EFL
too, suffer not from exposure to non-standard English, but partly from exposure
to non-authentic English and partly from lack of exposure to authentic standard
English. Similarly, the real danger is not that the standards of English are in
a decline but that those standards may be to a large extent unrecognised and
inaccessible. The world-wide population of non-native learners cannot encounter
enough authentic standard English to gain the intuitive control over its
standards that native children achieve. Standard English is represented by a
uniform and simplified variety of non-authentic English; standards are
exclusively defined in terms of hard and fast ‘rules’ for every occasion.
Under these conditions, the chances of genuine success heavily favour learners
who have extensive outside exposure to authentic English, e.g., through
satellite television or personal computers with Internet access.
I
am using the term ‘standards’ here in a programmatically inclusive sense
which I hold to be justified and
realistic to the degree that these standards are documented by large sets of
authentic English and not just asserted out of personal attitudes about
correctness or propriety. They are sustained by the preferences of fluent
speakers or writers for certain arrays of lexical and grammatical choices from
among the immensely larger set of theoretically possible choices These include
not just the choices stipulated by ‘grammar’ and ‘vocabulary’ in their
routine senses, but also stylistic and rhetorical choices, as well as choices
stipulated by genre, register, and text types. These standards co-ordinate
sets of choices, so that what is chosen at one point makes certain choices
at other points more probable. Only the more obvious and regular standards are
reflected in textbooks relying on non-authentic English illustrated above, such
as the major patterns of the English Noun Phrase. So, learner performance tends
to feature non-authentic English too, which, however dull, seems safe.
Where
the standards are comprehensively reflected is in suitably searched and sorted
attestations of authentic English in large corpora. Their potential for language
teaching is thus steadily gaining recognition, witness some recent collections
and edited volumes (e.g. Botley, Glass, McEnery, and Wilson [Eds.], 1996;
Wichmann, Fligelstone, McEnery, and Knowles [Eds.], 1997; Burnard and McEnery
[Eds.], 2000; Lewandowska-Tomaszczyk, and Melia [Eds.], 2000; Ghadessy, Henry,
and Rosebery [Eds.], 2001). From a
practical standpoint, the ‘convergence’ envisioned by Leech (1997)
between ‘language teaching’ and ‘language corpora’ was probably
inevitable in view of the manifest practical value of corpora for ‘data-driven
learning’ (Willis, 1993; Johns, 1994). The value is plainest in areas of study
where we are compelled to work with a great deal of data, such as style
(Jackson, 1997), register (Biber,
1994), genre (Carne, 1996), and of course literature (Kowit and Carroll,
1991; Louw, 1997).
From
a theoretical standpoint, however, we are still far from a convergence in our
thinking about how corpus data can or should transform both theory and practice
of language teaching (cf.
discussions in Aston, 1995; Barlow, 1995; Tognini Bonelli 1996). Corpus data
have long been used in studies of language, but these were not designed for the
goal of teaching non-natives. Such was true of corpus data for research in 19th-century
philology on European dialects (e.g. Wencker, 1887-95) and in 20th-century
linguistics on American varieties of English (e.g. Kurath, 1949) and on Native
American languages (e.g. Sapir, 1922). This work was eminently practice-driven;
theory sporadically took shape in higher-level statements about language types,
as in Sapir’s (1921) ambitious ‘classification’
of languages.3
As
some writers on corpus work have remarked (e.g. McEnery and Wilson, 1996),
corpus studies underwent a period of eclipse in the 1960s and 1970s.
‘Descriptive’ methods were substantially displaced by ‘generative’ ones;
and linguistics transformed its subject matter away from data sets in particular
languages (plural, count noun) over to a single theory embodied in language
(singular, mass noun) (cf. Beaugrande, 1991, 1998). As an integral step in this
transformation, it was argued that ‘attempting to state methods of analysis
that an investigator might actually use, if he had the time’, for
‘constructing the grammar, given a corpus of utterances’, must ‘fail to
provide answers to many important questions about the nature of linguistic
structure’ (Chomsky, 1957: 51ff).
For
us, the key reservation to recall was that ‘the corpus
of observed utterances’ ‘obtained by the linguist in his field work’ is
‘finite and somewhat accidental’ (Chomsky, 1957: 15, my
emphasis). This reservation was accepted at face value without noticing
that it holds for every set of observations and every set of data in every
science. Whatever a science or scientist has ‘observed’ must be
‘finite’; and ‘data’ are, both by definition and by etymology, ‘the
given’, and cannot be other than finite.
And
linguistics is after all not a science of the infinite. If a language were an
‘infinite set of sentences’ (Chomsky, 1957: 13),
then the act of uttering or comprehending a sentence would require infinite
search times. Moreover, ‘performance’ would be related to ‘competence’
in purely accidental ways, just as, in the familiar parable, a roomful of
chimpanzees with typewriters would, in infinite time, write the works of
Shakespeare. (They would also type the British National Corpus and the COBUILD
Bank of English.) Such fanciful quibbles and quiddities inhere in the
mathematically proper meaning of the ‘infinite’, which supplies poetic and
philosophical labyrinths for writers like Jorge Luis Borges but merit no place
in the theory of language.
A
language must rather be a very large but always finite set of data. This
set can never all be observed any more than can the set of particle collisions
in physics or the set of supernovas in astronomy. Nor can the whole set be
consistently described by any single definition of ‘sentence’, which is a
reliable unit only in reference to the clause structures of written language
(Beaugrande, 1999). And least of all could such a set be represented by the
non-authentic sentences of ‘the man hit the ball’ type preferred by
generative linguists for their complicated analyses, as I noted in section 1.
Scientists
who work with very large data sets must manage a trade-off between breadth
(how much data a theory can describe) and depth (what degrees of detail
and precision the description can achieve). Early corpus studies of familiar
languages (e.g. on speech varieties) could aim at the sweeping breadth of
Wencker’s ‘language atlas’ or Kurath’s ‘word geography’ because
the structure of the language was in any case under control. But corpus studies
of an unfamiliar languages, e.g. the Yana language of Northern California
described by Sapir, had to concentrate on depth to work out 'fundamental
elements' of
the structure under control, and the breadth was
correspondingly limited. Yet if this ‘limitation-in-principle
to classification and organization of data’ from a ‘corpus of observed
speech’ ‘establishes’ the ‘inadequacy’ of the description
(Chomsky 1965: 15, 67), then ‘adequacy’ must have some odd meaning, as we
will in fact see in a moment. I
cannot understand.
Now,
if a language were an infinite set, then its description would entail an
infinite breadth that flattens out our depth to an infinite shallowness, and our
description (completed in infinite time, by the way) would capture only
infinitesimal detail and precision. In practice, generative linguistics evaded
its own ‘infinity’ argument against corpus studies by ‘assuming that
the set of grammatical sentences is somehow given in advance’ (Chomsky 1957:
18, 54, 85, 103). Breadth was hypothetical,
built into the theory by focusing on the ‘ideal
speaker-hearer in a completely homogeneous speech-community, who knows its
language perfectly’; and on ‘language universals’ ‘stated only in
general linguistic theory as part of the definition of the notion “human
language”’ (Chomsky, 1965: 4, 6, 117), Breadth in the sense I suggest did
not seem to figure on the agenda; Chomsky’s well-known Aspects was
presented with just 24 non-authentic English sentences (or ‘transformations’
of these).
Science
also enlists technologies to cope with accidents in our data, most
crucially at frontiers where we can’t yet distinguish the accidents from the
regularities. We scan the collisions in linear accelerators for evanescent
particles at the frontiers of physics; we train our telescopes on invisible
planets affecting nearby stars in detectable ways at the frontiers of astronomy;
and we peruse our monitors or print-outs for units or patterns of a language
attested in very large corpora at the frontiers of linguistics.
All
across science, the more significant the potential for accidents, the greater
the breath we should seek, and the more we deploy those technologies that
increase breadth without seriously decreasing depth. We may thereby push down
the significance of any particular accident (or set of accidents) by measuring
its probability. Should the probability remain high, then we are dealing with a
regularity that had been mistaken for an accident.
The
return of large corpora of authentic language to the centre of language study
thus impels us to reopen the whole discussion of theory and practice. In that
spirit, Sinclair (1999a, 1999b) has recently probed the concepts of ‘observational’,
‘descriptive’, and ‘explanatory adequacy’ introduced by
generative linguistics (e.g. Chomsky, 1964, 1965). The first two of these had
been stoutly affirmed in early corpus linguistics, although not under such
programmatic labels. Observation was an operation of recording data so as to
sustain validity and rigour despite the lack of technology. Description was a
thorough presentation of the observed facts, e.g., by drawing maps to locate the
distinct dialect forms in the respective regions.
Explanation
remained a separate and sporadic issue, since questions about why a language
assumes a given form or evolution were rightly judged intractable. At most,
linguists hoped that explanations would eventually arrive, viz.: ‘back
of the face of the history are powerful drifts that move language, like other
social products, to balanced patterns’; ‘perhaps psychologists of the
future’ will find ‘the ultimate reasons’ (Sapir, 1921: 122). The real
theoretical significance of early corpus work adhered in the programmatic
acknowledgement of the importance of languages or dialects that had hitherto
been regarded as curiosities or degradations, much as learner Englishes are
regarded in some quarters today.
The
generative approach expressly disavowed observational
adequacy by announcing that the ‘observed use of language’ ‘surely cannot
constitute the subject-matter of linguistics’ (Chomsky, 1965: 4). So
'descriptive' and 'explanatory adequacy ‘ were consigned to goals
requiring no observation, namely
‘describing the
intrinsic competence’ and ‘intuition’ of the ‘idealized native
speaker’ (Chomsky, 1965: 34, 26). Yet despite ambitious claims,
intuition is a weak, opportunistic technique. Instead of actively observing
data, one passively rates the plausibility of data one produces for the
occasion. To judge by the samples in published studies, intuition heavily
favours what I have labelled non-authentic data. And just as non-authentic
samples cannot be an adequate basis for learning the language, neither can they
be an adequate basis for describing the language
nor for explaining ‘language’ in the abstract.
In
the new corpus linguistics, observational adequacy is our operational
front end, where we depend most crucially on our techniques and technology.
Large corpora offer us such immense breadth
that depth can be managed only by incorporating our techniques into our technology
— harnessing the computer for the descriptive stage as well. The key question
here is what mode and degree of depth to look for. We cannot simply re-open the
programme of early corpus linguistics insofar as we do not share its goals of
describing either unfamiliar languages or dialects of familiar languages. Still
less can we embrace the programme of recent non-corpus linguistics in quest of
‘linguistic universals’, which are not even expected to fit large data sets,
viz.: ‘if some remarkable flash of insight were suddenly to yield the
absolutely true theory of universal grammar’, ‘it would be at once
“refuted” by innumerable [infinite!?] observations from a wide range of
languages’ (Chomsky, 1980: 2).
The
programme of corpus linguistics might do well to shelve the principle of
explanatory adequacy as long as the conception of ‘explanation’ eludes and
operational definition. A more tractable principle at present would be applicatory adequacy: how far our work is found suitable and
productive for relevant applications. So far, the most successful applications,
which have in fact quickly become the industry standard, have been achieved in
reference works, such as dictionaries of words, idioms, phrasal verbs, and so
on. Selecting entries by their frequency of attestation renders the breath of
coverage fully operational, and finally trims off gratuitous arcane or archaic
expressions for ordinary meanings still found in conventional dictionaries, such
as ‘operose’ for ‘involving much labour’, or ‘monopsony’ for a
‘market condition with one buyer and a large number of sellers’ (Random
Webster 948, 877).2 English has accumulated a peculiar mass of
these terms, which are especially non-authentic in never being used in real-life
conversation.
In
parallel, depth become operational as more precise information on usage, such as
noting which Verbs are frequently used only in the Active (e.g. ‘elude’) or
only in the Passive (e.g. ‘construe’) (COBUILD 458, 302).4
We do not assert that using such Verbs the other way round counts as
‘error’, but that it is not expected. In my own 10-million-word corpus of
British and American writers dating roughly between 1750 and 1920,5 I
found only 2 out of 74 uses of ‘elude’ in the Passive:
[21]
they lessen the consumption; the collection is eluded; and the product to
the treasury is not so great (Alexander Hamilton)
[22]
My importunities would not now be eluded (Charles Brockton Brown)
But
21 out of 59 uses of ‘construe’ were found in the Active, not just in the
current sense of ‘interpret’ [23] but also the senses of ‘translate’
[24] or ‘interpret something into something else’ [25] — neither of which
I would use nowadays.
[23]
This behaviour in her niece the good lady construed
to be an absolute breach (Fielding)
[24]
he recalled the shrewd northern face of the rector who had taught him to construe
the Metamorphoses of Ovid in a courtly English (Joyce)
[25]
She’s an excitable, nervous person: she construed
her dream into an apparition (Charlotte Brontë)
Such
findings highlight the historical dimension of authenticity, the more so for EFL
programmes that focus on literature, which I shall discuss further on.
Corpus-based
reference works also offer an operational measure of applicatory adequacy for
description in terms of suitability for a real audience, including non-native
speakers. Compare these definitions:
[26]
hydroponics: the growing of plants in nutrient solutions with or without an
inert medium to provide mechanical support (Webster’s
Seventh 408)2
[27]
hydroponics is a method of growing plants in water rather than in soil (COBUILD
714)
In
conventional dictionaries, definitions have been authored by specialists in the
field. [26] was evidently composed by a botanist — technically correct but
accessible only to other specialists, who would understand how an ‘inert
medium’ can ‘provide mechanical support’ to a crop of tomatoes (and who
would know the meaning of ‘hydroponics’ anyway). [27] clears away the
technicalities and explains the essentials for ordinary people.
To
be sure, dictionaries represent the most thoroughly practical application of
corpus studies. The theory is sparse and straightforward: a language can be
represented by a subset of expressions whose occurrences in a very large corpus
reach a specified cut-off point; that quantity of occurrences is sufficient to
determine the meaning; and the definitions are to be illustrated with authentic
data, which are ‘examples of good practice’ for ‘speaking and writing the
English of today’ (COBUILD xv). The dictionary can impose authenticity
without having to explain its nature nor defend it against theoretical or
applied linguists who deal in non-authentic data.
Yet
the theoretical implications of authenticity surely extend much further. We must
decide whether corpus studies will be fitted to established descriptions and
categories of linguistics; or whether the foundations of linguistics will have
to be revised in light of corpus studies (Tognini Bonelli, 1996; Sinclair 1999).
As we know from the work on ‘scientific revolutions’ in the philosophy of
science since Kuhn (1970), a theory is not displaced by observation alone but
only by another theory which fits observation better, or which enables new and
important observations (e.g. Kuhn, 1970). Now, if the factor of authenticity is
so crucial for observation and description as I am suggesting, then linguistics
and applied linguistics should brace themselves for a major scientific
revolution whose repercussions will inevitably be felt in TEFL. In exchange,
TEFL can offer our best resources for measuring the applicatory adequacy of new
theories.
The
upcoming ‘paradigm shift’ can be predicted to transform the entire concept
of a language: not a static system of units (phonemes, morphemes,
phrases, sentences, etc.) but a dynamic system of relations. Instead of a dichotomy between
‘langue and parole’ or ‘competence and performance’, we can recognise a dialectical
cycle between combinability (language as potential system) and combination
(text as actual system). And instead of separating a ‘grammar’ of
‘rules’ from a ‘vocabulary’ (or ‘lexicon’) of ‘words’, we can
explore the unified lexicogrammar for the typical grammatical
combinations called colligations
and the lexical combinations called collocations.
Some
of these conceptions and terms have been with us for quite some time, but their
‘revolutionary’ impact centres on fully recognising authenticity to be the
fundamental precondition and constant requirement for observation and
description to achieve applicatory adequacy. Authenticity is first of all an empirical property of data certified by
their occurrence in a context of situation. But authenticity can become a
property of an applicable theory or description only if some challenging
problems can be solved.
I
should emphasise at once that these problems do not arise from principled
weaknesses inherent in corpora, despite what is consistently alleged by
those who oppose the use of corpus research in applied linguistics in language
teaching (e.g. Widdowson 1991). Rather, they are problems which have been
inherent in language research and language teaching all along but which corpus
studies allow us to recognise and formulate. The corpus raises questions rarely
posed, let alone answered, in mainstream linguistics, such as:
(a)
How big is a language?
(b)
What is its ratio between uniformity and diversity?
(b)
What is its ratio between regularity and accident?
(d)
How much data is enough for the description of a language?
Instead,
linguistics has propagated reassuring abstractions and generalities, e.g.: ‘language’
is ‘like a dictionary of which identical copies have been distributed to each
individual’ (Saussure, 1966 [1916]: 19); or ‘linguistic theory’ is
‘concerned’ with a ‘completely homogeneous speech-community’ (Chomsky,
1965: 4). Language is declared to be uniform without even bringing forth
authentic data. Corpus studies deny us this easy reassurance.
I
would make a similar point for applied linguistics and TEFL. The problems in
setting up and applying corpora reflect prior indecisions about the world-wide
mission and audience of EFL insofar as these have been sustained by using
non-authentic English to project a simplified uniform vision of the language to
be taught the same way to everybody, everywhere — and from the same textbooks.
Now that corpora are accessible to the teaching and learning of English for
specifically ‘academic and professional purposes’, we find ourselves
perplexed by the shift to authentic English. The shift is so difficult not
because (as some have argued) corpus data do not represent the English language,
but because the concept of ‘authentic data representing a language’ has not
yet been operationally defined in language pedagogy.
Yet
we may achieve major progress through a principle we might call dialectical
resolution, whereby the problem arises from the same source that will lead
toward its eventual solution. Corpus studies have been exposing problems that
remained implicit or excluded in non-corpus studies of language; but the corpora
themselves provide the raw materials for finally resolving those problems.
The
toughest problem is unquestionably the diversity that comes with authenticity. How far is any one set of data relevant
to another set in the corpus, or to the corpus as a whole? Which sources of data
deserve to be represented, and in what proportions? In theory, diversity might
be mastered by breadth. At each progressive jump in size, the distinctions
between accidents and regularities will become more precise; some presumed
accidents will turn out to be regularities, and vice-versa. Sinclair (1999a) has
recently aired the prospect that the ‘generic or reference corpus’, such as the Bank of English, ‘currently
approaching 500 million words’, will be ‘large enough to smooth out many of
the idiosyncrasies of individual authors and texts’. But of course the concept
of ‘idiosyncrasy’ is itself an unsolved problem within the larger problem of
diversity.
A
brief demonstration may be helpful here. In a previous analysis of the uses of
the Verb ‘warrant’ (Beaugrande 1996),
I analysed the 228 occurrences in the Bank of English (then at 220 million
words) and found just 4 in colligation with First Person Subject, used when you
want to indicate you feel sure about something though you can’t point to
actual facts. Now, my 10-million-word corpus of British and American writers
returns 198 occurrences, of which fully 75 show this colligation. The
proportions seem dramatic until we notice that 25 of those are from Fielding’s
Tom Jones alone, e.g.:
[28]
I warrant you will never repent having the money into his hands.
[29]
Why, you thought, sir, I knew nothing of the matter, I warrant you, about Madam
Sophia.
This
corpus displayed an authorial idiosyncrasy which would be mistaken for a
regularity if we did not attend to our sources.
Again
in theory, breadth of observation should move on a stable upward curve as
data are added; but the issue in fact hinges on how new data relate or compare
to previous data. Since the ‘generic corpus’ should reasonably seek out as
much diversity as is representative of authentic English, its significance or
information value gains little from adding more data of same type. This problem
applies especially to mass media, such as the plentiful newspapers conveniently
posted on the Internet. Their diversity as data is restricted in being authored
by a relatively small, well-trained group of writers, and being edited by an
even smaller group. The total effects of these restrictions upon the
representative qualities of the data are yet to be assessed. I would also note
the massive frequencies I found in the BoE in July 1994 of key-words like
‘death’, ‘kill’, ‘murder’, ‘massacre’, ‘shooting’,
‘robbery’, and ‘rape’, reflecting the morbid interests of mass media
more than the frequencies of authentic English at large.
Evidently,
breadth of coverage does not match the size of the corpus, but must be factored
out between size and diversity in relation to authentic English. Here,
dialectical resolution could apply: having admitted the problem, we use corpora
to get it under control. Daunting labours await us in accounting for our
intuitive distinctions among the English used by prominent speakers or writers
(e.g., politicians, novelists, columnists), or by media (e.g., newspapers,
magazines, chat sites), or by professions (e.g. doctors, lawyers, scientists).
No doubt we will discover factors that render the use of corpora in language
teaching more complicated but also better secured.
This
point applies most trenchantly to the use of literature. Literary texts are
widely used in EFL programmes without an operational account of the relation of
literary English to other Englishes, particularly to those of the learners.
Since current definitions of style in stylistics accentuate the personal and
special qualities, literary English should surpass all others in its diversity,
and also in its remoteness from non-authentic English. Quite plausibly, learners
face a daunting jump in order to access literary English without adequate
fluency in authentic English.
Again,
dialectical resolution could apply. Learners could work with a series of corpora
of authentic English expressly designed to promote the fluency needed for
literary English, which could in turn be approached through a series of literary
corpora arranged in terms of their accessibility. Some principled decisions
would be required about which literary texts or text types merit study within
the limits of a given programme format. Yet that this approach could materially
increase the range of literature we could effectively cover in practice.
If
the breadth of an applicable description creates tough problems, depth creates
even more. Already at this stage, we can see the hopelessness of any cut-off in
depth analogous to the cut-off in breadth derived from relative frequencies in
such applications as dictionaries. How deep a description should extend will
fluctuate sharply across a very large corpus. Frequency of occurrence is by no
means a reliable indicator of appropriate depth, though some working ratio may
eventually be determined.
For
depth, our greatest problem is undoubtedly our categories. The available
categories in linguistics and EFL are a mix of traditional Latin-based grammar
(e.g., ‘indicative’, ‘intransitive’) with various modern approaches,
sometime descriptive (e.g., ‘constituent’, ‘verb phrase’) sometimes
generative (e.g., ‘second language acquisition’, ‘universal grammar’).
Not surprisingly, they do not constitute a unified description either in theory
or in practice; and corpus studies keep turning up gaps.
I
shall end this section by illustrating one noteworthy gap. So many combinations
in authentic English carry attitudes of good and bad, or pleasant and
unpleasant, or approval and disapproval, as to constitute what I proposed to
called ‘standards’. No terms are widely established; we can’t use
‘positive’ (which could mean certain) and ‘negative’ (which is a
category of the Verb Phrase). I have chosen ‘ameliorative’ and
‘pejorative’ which, though also from Latin and a bit academic, can be
reserved for attitudes alone.
One
class of relevant data are Adjectives that do not supply the attitudinal quality
of their Head Noun, but only highlight a quality the Noun would imply by itself
(cf. (Tognini Bonelli, 1993: Sinclair 1999). In my corpus of British and
American writers, ‘beautiful’ often occurs with things I at least would not
expect [30-34], whereas ‘ugly’ often occurs with things that could hardly be
otherwise [35-39] and never with anything I’d expect to be beautiful.
[30]
One night there was a beautiful electric storm (Willa
Cather)
[31]
My heart quite fails me when I think how I might have lost that beautiful
luncheon-basket. (Kenneth Grahame)
[32]
‘Hound never ran on a more beautiful scent’,
responded the scout, dashing forward (Fenimore Cooper)
[33]
We must go and visit our beautiful suburbs of
London (Thackeray)
[34]
I come quite over-powered. Such a beautiful
hind-quarter of pork (Austen)
[35]
he opened his jaws, rolled back his lip in an ugly
snarl (Zane Grey)
[36]
what an ugly monster it was! Only his horned
head belonged to a bull (Hawthorne)
[37]
with a bang an ugly black imp appeared and
croaked a reply (Alcott)
[38]
‘He were an ugly devil’, cried a third
pirate, with a shudder; ‘that blue in the face, too!’ (Stevenson)
[39]
it was as ugly gaping wound as surgeon ever
saw; more than two feet (Melville)
‘Beautiful’
was five times as frequent, and many occurrences were collocated with things
that could be or not be so, the most numerous, perhaps inevitably, being ‘girl
- lady - woman’ — in the discourse of fiction, after all.
In
corpus data, the same word may be found to carry both attitudes. In some
collocations, the Adjective ‘serious’ has the ameliorative meaning
‘significant’ or ‘sincere’ [40-43], and in others the pejorative meaning
‘grave’ or ‘alarming’ [44-46]. If the Noun is vague, like ‘thing’,
the pejorative dominates [47-48].
[40]
She did it constantly, with such a serious
enthusiasm that he grew fond of watching her
[41]
The discussion carried me far afield in perhaps the most serious
economic reading I have ever done (Jane Addams)
[42]
If Mr. Goodwood were interested in Isabel in
the serious manner described by Miss
Stackpole he would not care to present himself at Gardencourt (Henry James)
[43]
you have actually given her reason to flatter herself that you had the most serious
designs in her favour. (Fielding)
[44]
The English Government took an extremely serious
view of the matter (Strachey)
[45]
something has occurred of a most unexpected and serious
nature; but I am afraid of alarming you (Austen)
[46]
I found the undertaking even a more serious task
than my fears had led me to imagine (Poe)
[47]
the first time life ever struck Jones as a really serious
thing was when the Dean told him he must leave school. (W.E.B. DuBois)
[48]
To some it seemed that now that they were in actual possession of it, freedom
was a more serious thing
than they had expected (Booker T. Washington)
As
remarked in the source text for [48], ‘freedom’ can seem pejorative to
slaves who are too old or weak to find new jobs.
In
this section, I have tried to show that advocating the introduction of corpora
as authentic English into EFL programmes in no way implies a brisk optimism that
overlooks substantive problems in the relation between theory and practice. By
themselves, the practical benefits should be easy to grasp. TEFL shifts from a
frontal, teacher-centred exercise in the production of correct though
non-authentic sentences over to a learner-centred joint exploration and
appreciation of authentic texts. Teachers are freed from tedium of inventing and
writing out data, and from any residual anxieties about their own command of
standard English. And even everyday classwork can lead to the discovery of
subtle and previously unnoticed ‘standards’ as a tangible and creative
achievement.
But
in the wider context, these benefits need to be secured through strenuous
theoretical work. In this exploratory stage, those of us who are most involved
with corpora in teaching are also the most compelled to assess the problems. In
contrast, people who are not involved and yet publicly oppose the use of corpora
on some abstract principles are the least qualified to do so. The burden of
proof should rest upon them to demonstrate how exposure to non-authentic English
can nurture fluency in authentic English.
3.
PRACTICAL CORPUS WORK FOR STUDENTS
The
United Arab Emirates as a whole is a patchwork of ESL and EFL environments. The
region was a ‘British Protectorate’ from 1820 to 1971, but, prior to the
discovery of oil in 1958, the British presence was minimal. No attempts had been
made to establish British schools or cultural centres, nor to encourage the
spread of the English language. Today, the larger towns and cities are strongly
ESL, but mainly among the enormous communities of expatriates from Asia, Europe,
and America. Among the actual UAE citizens, who constitute between a fourth and
a third of the total resident population (1996 estimate), Arabic is clearly the
dominant language.
The
student body of the United Arab Emirates University is almost entirely composed
of UAE citizens, aside from a small contingent from other Arab regions, such as
Egypt, Jordan, and Syria, whose families are employed here. Apart from the
modest portion who have attended English-medium schools, the fluency level is
closer to the EFL than the ESL environment.
To
increase their exposure to authentic English, our students are using WordPilot©,
a corpus-based resource program developed by John Milton at the Hong Kong
University of Science and Technology (Milton 1999). It can be used to access
on-line displays of authentic usages for specific expressions and combinations
while reading or writing in a programme like WORD. Having selected a suitable
corpus, such as ‘British academics’, students can click on a doubtful word
in their own text and see some authentic examples. They can see, for instance,
that people ‘differentiate’ not by separating mixed substances like ‘milk
from water’ in sample [9], but by developing or recognising a quality that
identifies one kind as distinct from another, as in [49-50].
[49]
This interest in work differentiates the workman from the criminal (Veblen)
[50]
Shakespeare differentiates his heroes from
his villains much more by what they do than by what they are (Bernard Shaw)
A
regular exercise in my ‘semantics’ course is to query a key-word and explain
the different meanings it can assume in context. For example, here are some
student responses to the key-word ‘fine’:
[51]
he tried to raise her self-respect with fine
clothes and flattery (Emily Brontë)
=> elegant
[52]
Harriet was short, plump, and fair, with a fine
face, blue eyes, light hair (Austen) => delicate
[53]
I heard rain strike earth in fine needles of
water (Joyce) => thin
[54]
Mr Elton has not such a fine air and way of
walking as Mr Knightley. (Austen) => dignified
[55]
‘A fine husband you are!’ said Mrs Glegg
scornfully. (George Eliot =>) worthless
[56]
Things are come to a fine pass when one
sister insults the other! (George Eliot) => dreadful
[57]
Don’t trust them fine-talking men from the big city. (George Eliot) =>
smooth, flattering
[58]
I shall come and see your mother some fine day. (Alcott) => some
indefinite future day
[59]
we get a fine day, and then down comes a
snapper at night. (Thomas Hardy) => sunny
Most
students were not familiar with the collocation ‘fine day’, and, in an
amusing cultural contrast to British usage, some guessed it must be a cool and
cloudy day — an apt guess here in one of the hottest desert regions on earth,
where sunny days reach 50° Celsius. They also did not know what a ‘snapper’
could be in [56], nor did I — surely not a nocturnal fish of the species Lutjanidae.
We did not find it again in the corpus, but we did find ‘cold snap’, e.g.:
[60]
he closed my carriage door one sleety day during the cold snap of February
ninety-three (Joyce)
A
‘snapper’ would be a regional variation among Hardy’s rustic farm hands.
The
collocation ‘fine day’ was found to have subtle social functions when we
queried it in the corpus, rather like zooming in on some detail of a visual
scene with a camera lens. These data proved helpful:
[61]
One day Parson Thirdly met him and said, ‘Good morning, Mister Everdene;
‘tis a fine day!’ ‘Amen’, said
Everdene, quite absent-like, thinking only of religion when he seed a parson.
(Hardy)
[62]
there stood the Queen in front of them, with her arms folded, frowning like a
thunder-storm. ‘A fine day, your Majesty!’ the Duchess began in a low, weak voice.
(Carroll)
[63]
‘Good morning, Mr. Watty; it’s a fine day
for walking, isn’t it?’ Seeing that the stranger still lingered, Mr Lowten
shut the door in his face. ‘There never was such a pestering bankrupt since
the world began, I do believe!’ (Dickens)
Quite
irrespective of the weather, you can greet somebody with ‘it’s a fine day’
in order to appear sociable and agreeable, though you may not get the effect you
wanted [61]. As useful variations, you can use the greeting to mollify a hostile
encounter [62] or hint that an unwelcome encounter should come to a speedy end
[63].
Another
interesting exercise is to examine the meaning of major terms within a single
text. In Jane Austen’s Pride and
Prejudice, we searched and sorted out
all collocations of ‘pride’ and ‘proud’. These ones concerning Mr Darcy
began to suggest a pattern:
[64]
everybody says that he is ate up with pride
[65]
He is not at all liked in Hertfordshire. Everybody is disgusted with his pride.
[66]
he was discovered to be proud, to be
above his company, and above being pleased;
[67]
His character was decided. He was the proudest,
most disagreeable man in the world.
[68]
It is wonderful, — replied Wickham, — for almost all his actions may be
traced to pride.
[69]
she tried to remember something of that gentleman’s reputed disposition, when
quite a lad, […] and was confident at last that she recollected having heard
Mr. Fitzwilliam Darcy formerly spoken of as a very proud,
ill-natured boy.
We
noticed how Darcy’s pride was typically stated as circumstantial opinions of
unidentified persons, such as ‘everybody’ [64-65]. Just because he failed to
be sociable at one assembly, his proudness was instantly and irrevocably
‘discovered’ and ‘decided’, where the colligation with Passive Verbs and
the collocations ‘ate up’ and ‘most…in the world’ subtly deconstruct
the force of what is asserted as absolute truth [64-67]. Then we encountered the
personal testimony of Mr Wickham, whose sincerity we knew to be worthless [68].
When Elizabeth repeats Mr Wickham’s account, the circumstantial quality
becomes too elaborate to be easily ignored: Mrs Gardiner rummages in her memory
and ‘at last’ dredges up a foggy ‘recollection’ of another unidentified opinion about a ‘reputed
disposition’ [69]. The data led us to see how the reader is positioned
to believe Mr Darcy proud by getting implicated in the ‘prejudice’ harboured against him by Elizabeth Bennet, whose
perspective is subtly entwined with that of the ironic narrator.
No
doubt the students could have dug this insight out the professional literary
criticism on the novels of Jane Austen. But they learn and enjoy much more by
working it out themselves from the actual language of the text. They have not
merely produced a key to appreciating this particular novel and its epigrammatic
title. They have hit upon a practical object lesson in the technique of irony.
And, they have discovered ways to appropriate literary English despite its
remoteness in time and place.
4.
PRACTICAL CORPUS WORK FOR TEACHERS
For
the present, EFL can provide indispensable practical assistance in identifying
the modes of data in and about authentic English that corpus studies could
reasonably provide. Indeed, progress toward applicatory adequacy can be achieved
only through a sustained co-operation of corpus studies with language pedagogy.
One
familiar problem arising in a course in ‘English grammar’, put to me by
colleagues at an African university in December 1998, prompted me to consult the
COBUILD (Collins Birmingham University International Language Database). The
database, popularly called the ‘Bank of English’, is the world’s largest
computerised data corpus, then containing 329 million words of running text, of
which 20 million were spoken data. It draws upon a range of contemporary spoken
and written sources, including: British and American books; newspapers (Times,
Independent, Guardian, Today, Wall Street Journal, New Scientist, Economist);
magazines (e.g., Esquire, Good
Housekeeping); ephemera such as letter-box mailings (e.g., YMCA appeal for
homeless people, Friends of the Earth Tropical Rainforest Campaign), radio
broadcasts (British Broadcasting Corporation in the UK and National Public Radio
in the US); and recordings of conversations.
The
problem in English grammar concerned the variation in usage between ‘if
I was…’ and ‘if I were…’ followed
by a Noun or Pronoun. Some colleagues insisted that: ‘if I was’ is
incorrect, and only ‘if I were’ is correct. The advice I found in the
voluminous Grammar of Contemporary English
and the Comprehensive Grammar of the
English Language proved ambiguous. This ‘were’ is described there as the
‘singular past subjunctive form’, sustained ‘by convention’ for ‘the
idiom “if I were you”’ (Quirk, Greenbaum, Leech, and Svartvik, 1972: 747;
1985: 1094). However, those same Grammars
concede: ‘the subjunctive is not an important category in contemporary English and is normally replaced by other constructions’ (1972:
75); and ‘the subjunctive in modern English is generally an optional and
stylistically somewhat marked variant of other constructions’ (1985: 155). If
so, then what remains of the subjunctive is preserved by certain colligations
and collocations — by standards rather than rules.
The
1972 Grammar proposed a distinction
between ‘hypothetical conditions conveying the expectation that the conditions
will not be fulfilled’ versus
‘open conditions leaving unresolved the question of the fulfilment of the
condition’ (1972: 747, their emphasis). The ‘open’ ones ‘have also been
termed “real” or “factual”’‚ and the ‘hypothetical’ ones
‘“unreal” or “counterfactual”’ (1985: 1092). The ‘were-subjunctive’
was said to be ‘restricted to one form’ and to be ‘hypothetical in
meaning’ (1972: 77). Yet ambiguity emerged again in the warning that ‘both
the past subjunctive and the past indicative are possible for hypothetical
conditions’ and are ‘occasionally used in formal contexts’, as in ‘“If
it was/were to rain, the ropes would snap”’ (but is that formal?); still,
‘the subjunctive is preferred by many in formal contexts, especially in formal
written English’ (1972: 748; 1985: 1093f). We shall see the corpus data
telling a different and more interesting story.
The
Bank of English returned 2061 lines for ‘if I were’ and 2876 for ‘if I
was’; at least both usages are truly alive and well. For purely practical
reasons, I decided to start with examining and classifying roughly 10% of these.6
During this work, I noticed that ‘if I were
you’ appeared in 20 lines, and ‘if
I was you’
in only three lines, whilst other usages with Pronouns after ‘were’ or
‘was’ were quite rare. To check these proportions, I requested all lines
from the Bank of English where ‘if I were’ and ‘if I was’ were followed
by any of the Pronouns ‘you, he/him, she/her, they/them’. This time I got
back 402 lines, and sure enough the frequencies across the entire corpus were
drastically uneven. No less than 282 lines attested ‘if I were you’, whereas
‘if I was you’ trailed at 37. The rest, those having Third Person Pronouns,
were at best marginal, some hovering between 10 and 20 and some close to or
equal to zero. After eliminating a few false alarms7
(e.g. ‘if I was her dog’), I got these
totals:
if I was he
0
if I was she 2
if I was they 0
if I was him 17 if I was her
5
if I
was them 11
if I were he
3
if I were she 0
if I were they 1
if I were him 18 if I were her
6
if I were them
10
These
figures indicate that a usage commonly recommended for ‘standard English’ in
EFL textbooks — ‘if I were’ + Subject Pronoun — is no longer secured in
authentic English. The old Subjunctive ‘were’ is surviving much better than
the presumably standard Subject Pronoun after it, and was found to colligate
with the Object Pronoun roughly as often as did the Indicative ‘was’.
The
data also indicate that the applicable standard is not the distinction between
‘hypothetical’ versus ‘open conditions’ proposed in the two Grammars
cited above. To be precise, all cases
are ‘hypothetical’, since ‘I’ can never be anybody but ‘I’. Nor did
the data show ‘were’ being preferred over ‘was’ when the prospect that
‘I’ might be another person was particularly improbable. For example, I
found both forms for scenarios of grandly imagining to be one of the Royal
Family [70-71], but also for ones of prosaically imagining to be one’s own
sister [72] or a worker for another company [73].
[70]
In regard to Diana, Joan Collins offers this suggestion: ‘If I were her I would come out here to LA, hire the biggest agent and
get $25 million to do one film.’
[71]
he wanted to tell the prince ‘what a fool he was to let Diana go’. He said:
‘She is a beautiful woman and my favourite. If
I was him I’d beg her to come back.’
[72]
I was surprised because my sister is not the submissive type of wife who obeys
whatever her husband says. I thought if I were
her, I would just put my kids in the car and not care what my husband
says.
[73]
Tim, a construction worker who supports Target, shook his head in disgust. ‘If
I was them, I would go somewhere else’.
The
data also indicate that being ‘preferred in formal contexts’ cannot be the
applicable standard. I found it in numerous contexts which were decidedly
informal, e.g.:
[74]
If I were you, I’d move on this real soon and come up with something, or
you’re going to be too late.
[75]
‘A young chickabiddy like you’s not done for yet. You know what I’d do if
I were you? I’d make a pact with myself to succeed to spite the beggars!’
[76]
He drank eleven shots before he could feel the influence of the alcohol. He
ordered his twelfth and the bartender told him: ‘If I were you, I’d get some
air’. ‘I can pay you’, Cross told him. ‘That ain’t the point’, the
bartender said.
The
applicable standards are rather to be found in the social functions of real-life
discourse. Among the major functions we find a triad of closely related discourse
moves or (to use the older term) speech
acts. Advice is given when
the speaker (or writer) has the friendly intention of suggesting what the hearer
should do. A warning is issued when
the speaker has the (more or less) friendly intention of pointing out potential
bad consequences for the hearer. A threat is
made when the speaker has the unfriendly intention of frightening the hearer and
coercing some action to be done or avoided.
In
some data, we can clearly distinguish which of these three discourse moves or
speech acts was intended, e.g., the advice in [77-78], the warnings in [79-80],
and the threats in [81-82]. Notice again the lack of ‘formality’.
[77]
‘I’d get some sleep if I were you. You’ll need to be up at six to catch
the early morning flight from Heathrow.’
[78]
If I were you I’d keep pestering them. Because sooner or later a job will come
up.
[79]
The builder looked at it and said, ‘I hope you’re not thinking of filling
that thing with water. I wouldn’t if I were you — it’ll go through the
floor.’
[80]
‘Colonel Sharpe won’t dare kill you in my ballroom, because I won’t let
him. But if I were you I’d give him his wife back and find yourself someone
more suitable.’
[81]
I wouldn’t come home if I were you. You should stay away from my patch. There
are people who know that you grassed8 and if they know where to find
you
[82]
‘The blood of the mob is up! If I were you, I’d clear out of town now with
as much as you can carry, because you’ve been found out!’
Sometimes
the move was explicitly named:
[83]
‘Well, if I were you’, she adds by way of some unsolicited advice, ‘I’d watch out for
that girlie of yours.’
[84]
‘I wouldn’t go in there if I was you’,
warned the young man in the office. ‘You’ve no idea what fish meal
smells like when it’s being dried.’
[85]
‘You’re a fortunate man, but becoming a real irritant. I wouldn’t put too
much faith in that chain, if I were you.’ ‘If you threaten
me or use any force, I shall inform the police.’ ‘They might be a while
getting here’.
For
the hearer, the advice may not be much appreciated [86]; or the warning or
threat may not have the intended effects [87].
[86]
when your having your therapeutic whinge about your last date from hell they
don’t just listen sympathetically, they wade in with remarks like, ‘If I
were you…’ and ‘Once you’ve really experienced love, like me, you’ll
realise…’ Why won’t they just shut up?
[87]
‘Gordon’s going to want your ass in a sling for this one. If I was you,
Wade, I’d move to Florida. Tonight.’ ‘But you’re not.’ ‘Nope, I’m
not. Thank Christ.’
Perhaps
all three discourse moves risk being perceived as irritating presumptions that I
know what’s good or bad for you better than you do. If so, saying ‘if I were
you’ or ‘if I was you’ lessens the risk by seeming to interchange roles
and implying: ‘I’m not telling you what to do, I’m just telling you what I
would do if I happened to be in your place’. The two colligations could thus
offer some means of saving face for the hearer who heeds the move with actually
being commanded, or for the speaker who gets ignored or challenged. Further
face-saving might be achieved with such reservations as ‘you know what I’d
do?’ [75], or ‘I hope you’re not thinking of doing that’ [79].
As
we might predict, the contexts and situations differed when the item following
‘if I were/was’ was not ‘you’.
The speaker can still issue advice, warnings, and threats, but without having to
address the intended recipient. We might call these moves playing out a scenario:
freely imagining what would be the case if the speaker were in somebody else’s
place, however fanciful — say, if you were a consultant ‘advising the
Government’ of Britain [88], or the American president forming the
‘cabinet’ [89], or the ‘prime minister’ of Australia ‘negotiating the
budget’ [90].
[88]
Once people have realised the tunnel is still there and there’s the chance of
a price-cutting bonanza, they will see that its ownership doesn’t matter. If
I was advising the Government I would tell them to tough it out
[89]
Interestingly enough, if I was choosing him for
a Cabinet slot, I would have put him at HUD, Housing and Urban Development.
Instead, Clinton’s putting him in Agriculture.
[90]
I can only imagine the outcry if I was prime
minister and I was negotiating the Budget with the Business Council of
Australia a week before the Budget was due to be brought down.
Conspicuously
popular was advice given about sports. You can play out a scenario of being the
star racing ‘driver’ Michael Schumacher deciding where to ‘sign’ [91],
or the star boxer Mike Tyson doing something as far out of character (for him)
as ‘reflecting upon life’ [92]. Your scenario can even elevate you into
being the ‘football coach’ not just for Brazil but for every team in the
whole world [93].
[91]
‘Ferrari’s decision to sign Schumacher is the right one’, Prost added.
‘He’s the best driver; however, if I were he, I would have stayed with
Benetton for another year.’
[92]
But the Briton believes Tyson should not be in too much of a hurry to lace up
the old gloves again, and warned: ‘If I was him and had spent so long inside I
wouldn’t rush into anything. I’d want to reflect on life and enjoy my
freedom.’
[93]
I was criticised when I said if I was coach of Brazil I’d make them a better
team but I only said that because I believe my way is best. My style is not only
suitable for Norway but football everywhere.
For
the English-speaking press, these wishful scenarios serve the function of
keeping the immense audience of sports fans listening or reading during the
times when actual sports events are not in progress. The audience can play out
their own scenarios and identify with their heroes by hearing how other people
do so.
A
more prosaic major source of scenarios in the was family matters with their many
expectations and deliberations about what one could or should do, e.g., when
your ‘sister is not the submissive type of wife’ back in sample [72].
Families seem preoccupied with making sure that ‘family life goes on, no
matter what’ [94], e.g., when people are ‘feeling awful’ or ‘ashamed’
or having ‘rows every day’ [95-96], or when parents are trying to control
their adult children [97].
[94]
I couldn’t cope with it if I was the hysterical type. But like most women, I
have made my mind up that my family life must go on, no matter what
[95]
These aren’t very kind thoughts of mine, particularly as I know how awful
I’d be feeling if I was her. Her life revolved round my father and she now
wants it to revolve round me.
[96]
my sister she’s never once had a holiday because he didn’t earn a decent
wage. I told him I’d be ashamed of that if I were him. After that there were
rows every day and when we weren’t rowing we didn’t speak at all.
[97]
[female speaker:] twenty-three and his parents won’t let him come and he [male
speaker:] God / that’s awful [female speaker:] abides by that / if I were him
I’d just say I’m going out [male speaker:] hitch into Stafford and get on a
train / [female speaker:] yeah
As
conceded by the Grammars, the
distinction between ‘hypothetical conditions’ versus ‘open conditions’
certainly does not match the distinction between subjunctive and indicative. The
Bank of English data displayed a subtle range of ways for indicating that
something is or is not the case, or might or might not be, and so on. Usually,
these matters were decided by the context but were sometimes were strategically
left undecided.
‘Open
conditions’ were well attested with
the implication ‘if it was the case, as it well might be’:
[98]
‘Do you think that having a personal mobile phone would increase your feeling
of security?’ ‘Yes, if I was out alone at
night’
[99]
I told her that I should be back by the end of the week with any luck, and that
I would communicate any change of plan to the office if
I was delayed.
In
other contexts, the implication was: ‘if it was the case, and in fact
sometimes it was’:
[100]
if I was drinking with mates, I had to drink
them under the table as a matter of principle. I would do anything or try
anything to show how big I was.
[101]
he was physically abused when he was young, but was not aware it was illegal:
‘I thought if I was cruelly treated, if I was
tortured, maybe it was right, maybe it
happened everywhere’.
In
a few contexts, the implication was evasive: ‘if it was the case, and maybe it
was’:
[102]
Geography it’s okay like but I could go home
[…] and get out my sister's worksheet you
know like if I was having problems
[103] I knew that Mike was highly sexed and