Chapter 3

GRAMMAR AND GRAMMARS

Introduction

Since we ended our last chapter with a conceptual examination of the passive, for the sake of comparison I want to examine five ‘text book’ explanations of the passive found in an assortment of college level grammars. Again, it would be most natural for a teacher confronted with a grammar problem to first of all consult whatever references might be available.

So what I'd like to emphasize in this chapter is that ‘doing grammar’ should be independent of any one specific grammatical text or model. Doing grammar, in my sense of the term, is to take an active, creative and open-minded interest in language, particularly sentence syntax and its relationship to meaning. This definition doesn’t mean that you can’t make choices and have strong preferences about texts and theories. However, language studies are in a state of constant flux and every position has its strengths and limitations. So ultimately, we have to make do with what we have. Hopefully, this situation should make us generous at the very least. To me, a good or useful grammar text does not have to be flawless. It can serve long and well if the reader is aware of its strengths and weaknesses and uses it accordingly.

A Brief History of Grammar

Before we begin our examination of the five presentations of the passive, let's briefly review a bit of grammatical history. Traditional grammars are called traditional primarily because they generally assume that grammatical ‘facts’ are established by tradition, by previous usage. And with them, a grammatical ‘fact’ takes on the force of an imperative. In other words, if this is the way things were done in the past, this is the way we ‘ought’ to do them in the present.

The very first western grammar was written on the structure of Greek by a scholar named Thrax because he abhorred the changes that were occurring in the aristocratic dialect of his class. He felt that by codifying the rules of his language he could curb the tide of this deterioration. Although Thrax's work appears to be a simple attempt to describe the structure of Greek, his motives were clearly normative or prescriptive in nature. On the other hand, due to the fact that all languages share certain categorical features in common, his descriptions of Greek, e.g. his naming of the parts of speech, remain the backbone of most grammars even to this day.

Introduction

Since we ended our last chapter with a conceptual examination of the passive, for the sake of comparison I want to examine five ‘text book’ explanations of the passive found in an assortment of college level grammars. Again, it would be most natural for a teacher confronted with a grammar problem to first of all consult whatever references might be available.

So what I'd like to emphasize in this chapter is that ‘doing grammar’ should be independent of any one specific grammatical text or model. Doing grammar, in my sense of the term, is to take an active, creative and open-minded interest in language, particularly sentence syntax and its relationship to meaning. This definition doesn’t mean that you can’t make choices and have strong preferences about texts and theories. However, language studies are in a state of constant flux and every position has its strengths and limitations. So ultimately, we have to make do with what we have. Hopefully, this situation should make us generous at the very least. To me, a good or useful grammar text does not have to be flawless. It can serve long and well if the reader is aware of its strengths and weaknesses and uses it accordingly.

A Brief History of Grammar

Before we begin our examination of the five presentations of the passive, let's briefly review a bit of grammatical history. Traditional grammars are called traditional primarily because they generally assume that grammatical ‘facts’ are established by tradition, by previous usage. And with them, a grammatical ‘fact’ takes on the force of an imperative. In other words, if this is the way things were done in the past, this is the way we ‘ought’ to do them in the present.

The very first western grammar was written on the structure of Greek by a scholar named Thrax because he abhorred the changes that were occurring in the aristocratic dialect of his class. He felt that by codifying the rules of his language he could curb the tide of this deterioration. Although Thrax's work appears to be a simple attempt to describe the structure of Greek, his motives were clearly normative or prescriptive in nature. On the other hand, due to the fact that all languages share certain categorical features in common, his descriptions of Greek, e.g. his naming of the parts of speech, remain the backbone of most grammars even to this day.

The same struggle against linguistic change was imitated by Thrax’s Roman successors who wrote their grammars to arrest the contamination of their aristocratic dialect of Latin. Although Latin ceased to be used as a spoken language, European scholars inherited these grammars centuries later, and, with the Renaissance, ‘classical’ Latin became the official written language of government and serious scholarship while the living, indigenous languages were regarded as common and vulgar.

Needless to say, the first grammars of European languages such as Italian and French were largely based on classical Latin grammars, the most influential of which was written by a Spanish scholar named Sanctius. The first grammar of English was written in Latin by Bishop Wallis who was determined to show that English was, in fact, a proper, civilized tongue by describing English in terms of Latin grammatical categories, whether they fit or not. The analysis of the English infinitive is one of the legacies of this translation.

Ironically, the very force which the early traditionalists abhorred, linguistic change, often gave them the power to remake language in their own image. Generally speaking, the normative grammarian had little use for change unless it is a change ‘back’ to his preordained view of how things were when people were civilized and knew how to use language ‘properly.’

Traditional grammar is obviously not very scientific. In fact, its methodology is antithetical to the fundamental tenets of science. It is for this reason that it became the natural target of all those who were trying to put the study of language on a ‘scientific’ basis. A group of grammarians who emerged in the late l9th century (and were later known as structuralists) became the sworn enemies of the traditional prescriptive and often elitist approach to language. And in many cases they had very good reason to insist that traditional concepts and methods be done away with entirely.

They argued that statements generated by the traditional method could not even be proven or disproved because they could not be ‘tested’ empirically. After all, one can measure and record a sound wave but how does one measure a meaning? For example, let’s look at the passive again. Most traditional grammars would state that the only difference between the active and the passive voice has to do with whether the subject of the verb is acting or is acted upon. Note that this analysis implies that meaning relationships (the logic of the sentence) remain the same in spite of the differences in grammar. So if we say,

1) John ate the cat.

2) The cat was eaten by John.

the logical subject in both cases remains ‘John’ while the logical object remains ‘the cat’ so (1) and (2) mean the same thing. The traditionalists would assert that the similarity in meaning between actives and passives is something we all ‘know.’

However, the structuralists would argue that from a ‘scientific’ point of view, (1) and (2) are very different types of syntactic constructions. They don’t look anything alike. Any connection between the two would be in terms of what they mean. And since meaning is a mental concept and since mental concepts can’t be observed and measured, nothing definitive can be said about the relationship between the two sentences other than the fact that they share certain vocabulary items.

It seems that the structuralists were disciples of the belief that ‘true science’ consisted largely of data collection and the classification of that data into categorical hierarchies. And as we just noted, they would admit as scientific data only that which could be observed and counted or measured. Hence, their approach had clear limits. But because they tried to work without social prejudice and bias, they did make lasting strides in the area of phonetics and, to a limited extent, the study of syntax. However, they had serious difficulty in relating sounds and sentences to actual meanings.

On the other hand, to their credit, they were adamant about the principle that there were no such thing as a ‘primitive’ culture or a ‘primitive’ language and so they spent much of their time studying American Indian and other non-Western languages. Benjamin Lee Whorf and his dedication to Hopi is a fine example of this school of thought.

It is ironic that in the nineteenth century there were also traditional scholars who became genuinely interested in grammar for its own sake and, therefore, sought to study and describe language ‘as it was’ and not as some special group or class thought it should be. While largely utilizing introspection and intuition and not entirely free from normative assumptions (they continued to focus on the dialects of the upper classes or the language of formal literature), they did attempt to be ‘objective’ and unbiased according to the informal scholarly models of their day. In particular, they did not simply believe that previous usage was necessarily better and many did recognize linguistic change to be the natural process it is.

Although their work was not what we would call ‘scientific’ today, traditional scholars of language like Jespersen, Onions, Sweet and Poutsma

were able to document the standard literary language of the period with admirable thoroughness. Unlike their prescriptive forerunners who were primarily concerned with usage, and unlike the emerging ‘scientific’ linguists who were limited by their definitions of empirical evidence, these scholars were very much concerned with meaning, with what various syntactic forms meant to those who used them.

It soon became obvious that if language is a system for using sounds and symbols to communicate meanings, then the study of meaning has to be a significant part of any grammar or theory of language which hopes to be empirically sound. Hence, the transformational-generative grammarians of the 1960's combined many of the descriptive techniques of the structuralists with the formalism of mathematics and the intuitive interpretations of the traditional grammarians to form the basis for an entirely new approach to the study of language. It was these innovative researchers who saw that the scholarly traditionalists relentlessly unpacked many of the complexities of the language which had been previously been ignored by other grammarians, scientific and traditional alike.

Since I have inherited the intellectual legacy of the transformationalists, I too believe that scholarly traditional grammarians had many genuine theoretical insights into the nature of language, insights that would have to be considered by any complete theory of language in the future. (For a further discussion of the theoretical assumptions underlying my approach, see Appendix 3.)

With these brief historical observations in mind, let's now to turn to examine the passive voice as presented by five texts which have been used in modern classrooms.

Ehrlich and Murphy: English Grammar

Let’s start with the basic text for 110J, a fairly standard college teaching grammar which I have used for many years. I am sorry to report that most of my students dislike Ehrlich and Murphy because they find it much too difficult, even though the authors regard it as a ‘review’ of high school grammar. The two major reasons for this problem are that most of what average college students have learned about grammar dates back to junior high or even elementary school, unless they’ve had to study a foreign language. And second, even those who wish to teach usually don’t have an interest in grammar because most of their rather limited experiences with it have, like my own, been largely negative. So even though the authors of this text don’t assume

that the student has had any special training in grammar, what little they do assume is still beyond the range of most college students. Hence, what Ehrlich and Murphy call ‘review’ is for many a first time introduction to the subject.

Now when discussing the passive under the heading ‘Voice,’ Erhlich and Murphy simply tell us the following:

They then give several examples of passive constructions:

He is found

This paragraph is all we are told about the passive, so I suppose we would have the right to conclude that it must be a pretty simple construction. I don’t think that the vast majority of teachers of English know any more about the passive than what is provided here except to say that in writing the passive should be avoided whenever possible because it is awkward.

G.O. Curme: English Grammar

Let’s turn our attention to one of the best traditional scholarly grammarians, G.O. Curme. In this single volume handbook, which is in Barnes and Noble’s College Outline Series, he begins by defining the active and passive in the following way.

Curme makes the same distinction between logical subject and grammatical subject that we’ve noted above. And when he discusses the formation of the literary or standard written English passive, his explanation is equally familiar. But when he begins to take up what he calls colloquial forms, we find some very interesting observations and we can see how Curme is not intent on defending traditional usage for its own sake.

Our house gets painted every year.

John tried to cheat but he got caught at it.

As I passed by, my coat got caught (act) on a nail. but

I had to stop since my coat was caught (state) on a nail.

This discussion is, to my mind at least, fascinating, because Curme is being extremely sensitive to the way the grammar ‘feels’ (we assume that he is relying a great deal on his own intuitions here). He also shows a sincere respect for what he calls colloquial (usually spoken) as opposed to literary (written) forms, which is typical of his approach.

Unlike many traditional grammarians, Curme seldom champions a literary form for its own sake. As he states in his preface, ‘... English grammar is represented, not as a body of fixed rules, but as part of an evolutionary process ...’ This openness to change is evident as he seems to suggest that the passive is not simply a ‘surface’ form but is an underlying concept which has various and often quite dissimilar syntactic realizations. Note, for example, his discussion of ‘become:’

He became seized with a profound melancholy.

This newspaper has already become widely read.(52)

What I find so ironic is that many of my students dislike Curme because they insist that all he is doing is creating unnecessary classifications and making everything more complicated than it has to be. I would suggest that Curme is not ‘creating’ or ‘making up’ anything. What he is doing is bringing actual usage into a traditional grammatical discussion instead of relying on textbook examples and promoting a set of rules which don't work for any but the most obvious cases.

I am not, of course, saying that Curme’s analysis is all inclusive or definitive, but it certainly forces us to think about what grammatical passivity is and this is precisely what a good grammar should do. Our next text, at least from my perspective, tries to do too much and consequently does too little.

Martha Kolln: Understanding English Grammar

Martha Kolln’s popular text tries to integrate traditional, structural and transformation material into a single system and thus satisfy everyone. But aside from her introduction of the passive transformation, she gives a fairly traditional analysis of the passive. The tone of her explanation is similarly traditional in that the reader has the feeling that what is being presented is some definitive, objective truth.

Again, even though Kolln does make a general attempt throughout her book to make her approach appear ‘flexible,’ I would say that aside from the use of the word ‘transformation,’ there isn’t much here that can’t be found in the traditional accounts. She apparently believes that what people want are specific rules and yet she also knows that they want ‘the latest thing,’ the latest grammatical ‘technology’ (linguistic jargon, diagrams and such). So what she does is give her reader traditional rules dressed up in the language of modern linguistics. Unfortunately, this method of ‘sprucing up’ traditional concepts with the trappings of contemporary linguistics has become the dominate strategy of most recent teaching grammars.

D.T. Langendoen: Essentials of English Grammar

We have already used examples Langendoen in the last chapter. I want to examine it in more detail here because the approach I use in this text has been greatly influenced by linguists like Langendoen, Charles Filmore and others working in the area of ‘case grammar’ back in the 1970’s. Langendoen makes it very clear in the preface to his little book that grammatical analysis is less like a science and more like an art and that the purpose of language instruction should be to develop an attitude of appreciation for language in students.

In keeping with the process model of theory construction, Langendoen

doesn’t offer his material as ‘fact’ or ‘the final word.’ Instead, he talks of ‘impressions,’ ‘observations,’ ‘assumptions’ and ‘arguments.’ Let me give you a brief example of his analysis of the passive which was, incidentally, presented to a select group of high school teachers during a summer institute at Ohio State University.

131. John was startled by a loud noise.

132. This house was lived in by George Washington.

Langendoen presents his arguments as steps in a path towards discovery, not as ‘fact’ or ‘rule.’ This attitude is nowhere clearer than in his analysis of exceptions to the point most often made by grammarians that all transitive constructions can be passivized. Kolln notes that the passive transformation has exceptions such as ‘have’ and ‘cost’ and then goes on as if they are not worth considering. Langendoen uses these exceptions to make a much deeper point about the grammar.

136. Mary surprised John.

137. John was surprised by Mary.

138. Mary surprised herself.

139. *Mary was surprised by herself.

140. Mary was surprised at herself.

What makes Langendoen’s account so ‘sophisticated’ from a theoretical sense is that the reader is always reminded that grammatical explanations as opposed to certain grammatical facts are functions of the particular theories or models in terms of which the explanations are made. Hence, his explanations are self-consciously presented as model explanations.

What he tries to offer are analyses consistent with the given ‘facts’ as well as the basic assumptions of his model, but he never even suggests that these explanations are complete, definitive or true for all time. Therefore, he is quick to point out places where his own method breaks down.

Before I leave Langendoen, however, I want to look at his concept of a grammatical ‘fact.’ Generally speaking, what he means by a fact has to do more with the data than with his explanations. In other words, it is an empirical ‘fact’ that native speakers of English would say that (137) is grammatical and (139) is not. However, it is not a fact that everyone would call (137) a passive. The active/passive distinction itself is not a fact; it is a way linguists have of modeling the relationship between various types of sentences. Hence, from a theoretical point of view, grammatical rules and explanations are not ‘facts.’ They are used to account for facts.

Of course, once one starts working with a particular model, the model goes on to define its own set of internally consistent ‘facts.’ For example, it is a fact of Langendoen’s model that the preposition ‘at’ does not appear in agent expressions which are associated with the preposition ‘by.’ For this reason, he states that (140) is not a passive sentence at all. It is a also a ‘fact’ of his model that he can accept what real people (native speakers) actually say is grammatical or ungrammatical as his basic data. Other grammarians, particularly those labeled ‘traditional,’ might not accept this method as a means of determining grammaticality.

Patricia Werner: Mosaic 1

Our last example is a text which is part of an elaborate series directed at students for whom English is a second language; the set includes reading, writing, and speaking/listening modules in addition to the one on grammar we are considering. The reason I am including Werner here is that her analysis combines the transformational and traditional approaches in a format which must explain concepts that are often overlooked in books written for native speakers. Hence, we might expect a lot more in the way of explanation.

And we are not disappointed. Werner presents the most elaborate set of examples and rules as to how to form and use the passive that we have yet considered. Let me start where she starts with the heading: The Passive Voice: Simple Tenses.

Most passive verbs (verbs that take an object) can be used in the passive voice as well as in the active voice. In sentences in the active voice, primary focus is on the subject (the agent or doer of the action). To give primary focus to the object, use the passive voice. The passive voice is used in both spoken and written English, and is used frequently in technical writing. (353)

She then goes on to give instructions as to when to use various forms of the passive. Her next heading clearly shows her attempt to integrate transformational material into her program: ‘by + Agent’

The next case she deals with is the ‘Anticipatory It and the Passive Voice.’ She states:

Werner gives as examples:

It was said, ‘The earth was flat.’

It was said that the earth was flat.

It was feared that the telephone was going to change people’s lives. (359-360)

Finally, she shows how the passive is formed in the perfect and progressive and with modals. (We will discuss modals in Chapter 5.) There is no further commentary on the passive or its meaning. She only offers examples, readings and exercises. Werner feels that these provide the context necessary to give the student an idea of the meaning of the constructions she is discussing.

Now if we compare Werner’s view of the passive with Ehrlich and Murphy’s, we can see that hers is clearly process oriented. The passive comes from changing the active in certain ways. The ‘by + Agent’ expression

expression need not appear in the sentence; it can be deleted. The anticipatory It can be used to change passive sentences even further. We come away with the idea that language is a active, fluid communication tool over which the user has a great deal of control. Rather than an arbitrary sense of right and wrong, Werner’s text emphasizes creative options. She begins with what the user wants to say and then matches those intentions with certain syntactic structures. This emphasis on meaning is part of what she means by referring to her approach as a content-based grammar.

On the other hand, there are none of Langendoen’s counterexamples or anomalies. For Werner is writing a text for people trying to learn how to use the American standard dialect. In other words, her goals are practical; she must give her students a feeling of certainty so that they can use her recommendations in actual communication situations. Langendoen, on the other hand, assumes that his readers understand how to use the dialect and is attempting to help them learn something about its complexity. His aims are basically theoretical in nature, particularly since he is addressing himself to those who are already teaching professionals. In a sense, he is trying to show his students how to ‘do grammar,’ how to deepen their understanding of the language they already know how to use. The differences in these aims can be clearly seen in the differences in their exercises. A number of Langendoen’s problems have no precise answers; they are primarily designed to force students to think.

Summary

So of the five analyses, the two which seem to be the most compatible with 110J would be Langendoen and Curme because both are concerned with meanings and the underlying features of the language. In a sense, both are explorers who are not primarily concerned with reassuring their readers but with the language itself. It is obvious that they go about their tasks in different ways. Langendoen, like most modern grammarians, is constrained by the rules of empirical science. Curme is essentially elaborating on the body of extent traditional grammar which has a preestablished legitimacy. Hence, an explanation from Curme and an explanation from Langendoen are very different sorts of logical creatures. And yet, we have said that they are compatible, and we have also said that 110J itself utilizes both types of explanations. How is this possible?

Perhaps I am belaboring the frustrating aspects of doing grammar. But I firmly believe that one of the reasons we as a culture don’t take language study more seriously is that we don’t think it’s very much fun. I firmly believe that once you start to really get into all of this and look forward to playing

around with syntactic problems and the like, you’ll find that a whole new universe has been opened up to you, like Marine World Africa USA. And experience has taught me that with enjoyment will come an increased versatility at all levels of language performance, from creative writing to editing and speaking.