Humanising Language Teaching
Year 5; Issue 6; November 03

"The Cambridge Grammar of the English Language"

reviewed by Joan Solà
(Professor of Catalan at Barcelona University)

The Cambridge Grammar of the English Language (CGEL), by Rodney Huddleston and Geoffrey K. Pullum (Cambridge University Press 2002) is a significant achievement, which I consider among the best in the languages I know.

An ambitious present-day grammar can be approached in two or three ways. It can be done by a small team of specialists or by a large group of specialists directed by a small team. The CGEL is authored by two specialists but draws on 13 “collaborators”. Its uniqueness lies in the fact that the “authors” sign all the chapters, both or one of them or both and a “collaborator” or one and one or more “collaborators”. This method assures a maximum coherence throughout the work because both authors are always responsible for the results of each chapter. Coherence is one of the biggest problems of a collectively written grammar as regards the doctrine, the terminology and the material aspects (typos and others). It's true that coherence is not essential in this type of work: the reader is looking more for modernity, the depth and reliability of the content than uniformity of thought. It is even interesting that some aspects be approached from different angles. However, it is also true that coherence is a very positive and desirable quality. In the CGEL this also implies taking a stance on theoretic aspects, not possible in other works.

The CGEL has a very select bibliography – less then seven pages – but no books are cited in the work. This procedure is only advisable in such a well-known and well-studied language as English.

Perhaps the main characteristic of CGEL is its insistence on the systemic nature of grammar points. This notion implies a clear differentiation between “categories of form” and “categories of meaning” as well as laying out the empirical foundations on which it is built. One clear case of this is the verb.

Taking into account that the “categories of form” in the English verb appear without inflexion, that is, through auxiliaries, the CGEL establishes four formal systems for the language with examples from the verb “to go”.

1. The “primary forms”: the past (went) versus the present (goes)
2. The “secondary forms”: the perfect (have + participle: has gone) versus no perfect (unmarked: goes)
3. The aspect: the progressive (be + gerund-participle: is going) versus non-progressive (unmarked: goes)
4. The mode: modal (modal auxiliary + plain form: can go) versus no modal (unmarked: goes)

Thus the number of “forms” of the English verb is reduced to six, when some grammars counted 120 or even 150. This grammar says, for example, that English has no form for the future (so that “He will take or “I shall go” are not future “forms”: the future “meaning” is provided by the modal meaning of will or shall).

INNOVATIONS

Many of the more notable originalities of the CGEL are related to the question of system, which in turn implies the notion of “centrality” (taken from cognitive theory). Let's look at an outstanding example.

The prepositions, adverbs and the conjunctions are categories, which have always baffled grammarians (and we are still far from a definitive explanation). The CGEL takes a “strong” position here: it reduces drastically the adverbs and the conjunctions and widens considerably the class of prepositions, based on “central” properties or prototypes of this category, which are the following:

1. the capacity to take a noun phrase and sometimes a sentence as object (“before the war”; “provided (that) it doesn't rain”
2. The capacity to head a prepositional phrase in the function of “predicative adjunct” ([Ahead of the ship], there was an island.” Apart from the fact that many prepositional phrases can also be objects in the real sense (“Put it in the drawer.”)
3. In certain cases the capacity to be modified by right and straight (“They pushed it right under the bed.” “She ran straight upstairs.” On this basis the following words are also prepositions: as, away, because, but, contrary, down, here/there, home, like, now, owing, than, underneath east, north, etc.

The division of constructions in “canonicals” (or basics) and non-canonicals. In the construction It would be pointless [to resist], the subject (in parenthesis) is a sentence. In these constructions what is “normal” in English (and Catalan) is that the subject is placed behind the verb (in 96% of cases) but the CGEL studies them under the “non canonicals”. This is purely a methodological, procedural, question, with no other implication.

Catalans will have a special interest in one of the aspects of the CGEL: “Our intention is to describe not to prescribe” says the first page, describe standard English or “Present-day English” (from the end of the second world war until today); that is the English “accepted by the majority in the countries where it is the language of government, education, the radio and television, advertising, entertainment and other public uses.” Then the authors immediately set about making their position clear on the relationship between description and prescription. In essence descriptive manuals say how the language is and prescriptive manuals say how it should be. This dichotomy in a language like English and from a starting point of the definition of standard given by the CGEL is a misrepresentation. The authors of the work reject it outright because they say that the prescriptive tradition (the classic “manuals of usage” or “style”) is flawed in many ways, as we shall now see.

In the first place these manuals are based on personal tastes. There are some which state that the construction “centre around” is incorrect in this example: The workers' demands centred around pay, though they recognise that “perhaps the correct construction centre on is more usual”. Pullum and Huddleston qualify this verdict, as “aesthetic authoritarianism “and protest” grammar is neither a question of taste or aesthetics. Secondly, they confuse the register “informal” with “ungrammatical”. Thus they also qualify as incorrect the following use of who instead of whom (which is used in a more formal style) because it is the object and not the subject: “It's clear who they had in mind.” Our authors ask themselves on what basis one can condemn a form which is widely used and respond that they reject the“ dichotomy between «the spoken» and «the grammatical»". The third mistake made by many prescriptive manuals is that of adducing impertinent arguments, “alien” to English. One case in point is the aforementioned who which is based on a norm in Latin (the language on which European grammars are based) but not in English. But there is another, exactly opposite, case. Most English speakers say 'It's me.' but “a powerful normative tradition” says that only 'It's I.' is correct since me is accusative and I is nominative and that after the verb be there must follow a nominative. However the CGEL replies that that was Latin, not English: “The mistake here is to suppose that what happens in Latin also happens in English.”, an error that still prevails in many manuals in many languages.

Prescriptive authors also turn to other debatable arguments. For example they condemn the prepositional use of due (“Due to the rain, the match was cancelled.”), for the same reason that it has been condemned in Catalan until now: because due “is” an adjective. Reply: due “was” an adjective, but today it “is” also a preposition “unquestionably sanctioned by use”. We Catalans have a lot of material for thought, in this case and in the others mentioned. For example some have condemned no només [not least] because it sounds bad. And a bitter lesson: traditional condemnations have been accepted, in English and in Catalan. From so much condemnation of the concordance of haver-hi [to happen] (one of the most extended and original phenomena in Catalan), now many who study the language (!!) loathe this concordance. However we must understand that the ideological basis of the CGEL is indisputable: what “dictates” in English (a language which is solid, autonomous, uncontaminated etc.) is the opinion of the majority of the users as we defined them above. We cannot apply the same criteria with the same certainty to any language, for example to Catalan.

In this last part I'd like to comment on one of the most spectacular sections of this important grammar. Thanks to the help of Geoffrey K. Pullum and Enric Vallduví, I hope to explain it clearly. The authors say explicitly: “One of the aspects in which in this work we have differed most clearly from traditional grammars is that we do not recognise a verbal form for the future in English.”

The traditional grammar establishes a system of three “tenses” and three corresponding “forms”, illustrated by take: past (took), present (takes) and future (will take). The CGEL proposes that, even though there are several ways of indicating the idea of future, “there is no grammatical category which we can analyse as particularly future”, and so the opposition is simply two terms: past – present (or past – not past).

In the first place, will (which has a past in would) is part of the same system of modal verbs as want (past wanted), shall (should), can (could), etc.: all these forms are used in the same contexts, in present or past, (taking into account that want is followed by to: will take, like want to / shall / can take; and would take, like wanted to / should / could take.

Secondly, will / would integrates perfectly in a harmonious system of forms: present: take (without auxiliary), has taken (with perfect auxiliary), is taking (with progressive auxiliary), will / can / may take (with modal auxiliary); corresponding pasts: took, had taken, was taking, would / could might take. Consequently if will take were the future form of take, what would would take be? (something strange like “the past of the future of take?) and what would be can take, shall take, might take, etc.? That is, if we give will take this role as future, then we muddle the whole system.

Meaning is no less important. The difference in meaning between a verbal form alone or preceded by will (or by another modal auxiliary) is a question of modality. Let's look at that.

“That will be the doctor.” “They will meet in the final in May.” are weaker, more qualified statements than “That is the doctor.” and “They meet in the final in May.” (This is a fact.) The other modals qualify in different ways: “ He will have left already.” and “He may have left already.” (Perhaps he has left.) describe a situation in the past; “He will be in Paris now.” and “He may be in Paris now.” Describe a situation in the present; and "He will see her tomorrow" and "He may see her tomorrow" describe a situation in the future. The forms will and may of these examples are present verbal “forms”: they describe the modal quality of the event at the moment that the speaker states his case. Will and would are very compatible with situations which still have to happen, but also with present and past situations. Languages (such as French or Catalan), which have formal futures, can certainly oppose these futures with other resources to express these qualifications. But, “if we look at the verbal system of English without prejudging that the ideas of past, present and future have to correspond inevitably with a three “form” system, then all the testimony is in favour of grouping will, shall, may can, must, etc. under one type of auxiliaries” says CGEL.