Haeckel’s famous theorem that ontogeny recapitulates phylogeny can be translated as: the history of every individual reenacts the evolutionary history of the species. If the theorem is correct, it means that we can learn about the individual by studying the history of the species, and we can learn about the species by studying the history of the individual. The theorem thus also suggests that every individual’s development of the syntax of his language probably reenacts, in general outline, the historical development of the language. We are concerned, however, only with the individual’s linguistic development, his increasing mastery of the structures of his language.
Volumes of research have been published on this area, but most of it, especially that of the last twenty years, is irrelevant to what goes on in our schools. David Crystal, for example, summarizes a famous experiment by Carol Chomsky in which she examines children’s ability to distinguish the meanings of "hard to see" and "easy to see":

What Carol Chomsky did was to present a group of children, aged from five upwards, with a blindfolded doll, and ask them "Is this doll easy to see or hard to see?" If the children said "Easy to see" then it is argued that they have learned the distinction; but if they said "Hard to see," and amplified their comment (upon request) by for instance, "Because it’s got a blindfold on," then it is argued that they had not learned it. The Chomsky results showed that before the age of six this distinction was hardly ever learned, whereas after seven it was known to nearly all children in the sample. (Child 49)

     The number of complex structures which children somehow figure out for themselves is awesome. I remember, for example, listening to my four-year-old son, speaking about a light, and saying, "Turn on it," instead of "Turn it on." Why children go through this stage, and how they work themselves through it, again pose complicated linguistic and developmental questions. But these questions are not "pedagogical": children work their way out of them, all on their own, without any other instruction than access to conversations, just as they work their way through and out of "I readed a book" and "I cutted up the paper."

     There are many things we can’t teach, and shouldn’t try to. This does not mean that the developmental research of the last two decades is totally valueless: it has provided and confirmed some important general principles of language development, and, in such fields as teaching the learning disabled, it even provides many practical suggestions. But for the normal child, studies which involve linguistic development in pre-school children are irrelevant. For relevant studies, we must return to the sixties and seventies and the work of such people as Kellogg Hunt, Roy O’Donnell and Walter Loban.

     The work of these researchers centered on the child’s syntactic development between the ages of five and eighteen. Hunt was looking for a unit by which we could measure linguistic "maturity," i.e., a unit with which we could gauge a child’s syntactic development. His famous "T-unit" is in fact, simply a main clause defined as including all subordinate clauses, and Hunt convincingly demonstrated the validity of the T-unit as the "basic" unit of measurement. I emphasize "basic" because Hunt himself notes, as we will see, that it is not the only gauge. Simply stated, Hunt counted the number of words per main clause in the writing of students in different grades and found that the average number increased steadily from grade to grade. Hunt’s study was corroborated by another by O’Donnell, to give the following figures, as summarized by O'Hare (22):
 

The figure for "Adults" is from Hunt’s study of essays published in The Atlantic and Harper’s. Although there are numerous questions surrounding such statistical studies, the basic conclusion remains valid: if samples of writing are taken from students (of similar socio-economic status) at different grade levels, the average number of words per main clause will show an increase similar to that above. Likewise, a study of passages from a variety of published writers will almost always result in a count of approximately twenty words per main clause.

     Since words per main clause is the primary validated measure we have for the linguistic development of children from the third grade on, any study of such development should begin with this measure. The basic question -- What accounts for the increasing length of main clauses? -- can be divided into two sub-questions: 1.) what principles are at work?, and 2.) is there any specific sequence in the acquisition of syntactic structures?

     The principles of syntactic development are not aspects of syntax as much as they are of psychology. They are, so to speak, mental attitudes, or mental drives, inherent in every individual throughout his lifetime. The primary, initial principles are EXPANSION and REDUCTION. Somewhere around age twelve these principles are joined by SUBORDINATION, at least on the syntactic level.

     Expansion, reduction, and subordination describe the mental process of acquiring and handling the relationships among isolated bits of knowledge. Expansion involves attempting to indicate the inter- relationships among a greater number of "facts": the child’s "We went fishing. We played catch." soon becomes "We went fishing and we played catch." The young child’s overuse of "and" is generally acknowledged, but it does demonstrate an early form of expansion. One could, depending on one’s view of life, say that humans are lazy, or one could say that they are economical: in either case, an expanded form is gradually reduced. "We went fishing and we played catch" may become "we went fishing and played catch," the unnecessary words being deleted. Reduction is thus the general tendency toward economy. As the child’s mental picture of the world grows more complex, the various elements in it are arranged by importance, cause, etc., and the principle of subordination is added to expansion and reduction.

The Principle of Expansion

     The main form of expansion, as suggested by the example with "and," involves compounding -- the use of coordinating conjunctions. Here again, our topic divides into parts: 1) which conjunctions, and 2) what is being coordinated. As for the first, we know that children start with "and" to coordinate main clauses, but little is known about when they start to use "or" or "but." (Teachers in K-4 could conduct studies to answer this and similar questions for themselves.) As for what is being compounded, our initial example already indicates that a compounding of finite verbs may involve the deletion of a compounded sentence. I mention this simply to emphasize that the three principles do not exist in separate boxes but rather continually interrelate.

     Perhaps the first important pedagogical question that can be asked about compounding is: which of the principle parts of the sentence (subjects, finite verbs, or complements) do children begin to compound first? Does

 Mary and Jane played ball.
    S     &    S   /  V    /  DO

 Mary went for a walk and played ball.
   S   /   V                     &     V    / DO

 Mary played baseball and football?
   S   /   V    /    DO      &    DO

     The tendency to compound continues as children develop new syntactic constructions. A child would first develop the use of subordinate clauses, and then begin to compound such clauses. The same is true for infinitives, gerunds, gerundives, appositives, and all the other constructions. But compounding, as "I went with mommy and daddy" suggests, is only one of the ways in which main clauses are expanded. Another is by means of what Roy O’Donnell calls "formulas."

     Unfortunately, O’Donnell does not develop the concept of "formulas" in any detail, perhaps because they tend to wreck the conclusions of statistical studies. He simply states:

there was a group of items that appeared more than sporadically in kindergarten speech but were used from about three to ten times oftener by seventh graders. At various levels, there were significant increments in their use. These would appear good candidates for identification as generally later acquisitions. They were noun modification by a participle or participial phrase, the gerund phrase, the adverbial infinitive, the sentence adverbial, the coordinated predicate, and the transformation- produced nominal functioning as object of a preposition.
     Theoretically, it seems reasonable to suppose that these constructions (unless acquired as formulas) would be mastered relatively late. Transformational grammar derives them all by application of deletion rules, and some of them indirectly from their sources by way of strings that could more directly yield subordinate clauses. Thus, The man wearing a coat . . . may be more difficult than The man who was wearing a coat . .., and A bird in the tree . . . more difficult than A bird that was in the tree . . . Noting that noun clauses did not vary much in frequency after the first grade, while participial modifiers of nouns were used by seventh graders three times as often in speech and nearly eight times as often in writing as they were used by kindergarten children, we may contend that such clauses (The dove saw that the ant was drowning) are easier to manage and earlier added to the child’s repertory than is the reduction of them to a single participial modifier (The dove saw the ant drowning). (92)

     A "formula" is a set phrase which a child -- or adult -- learns as a single unit. When my son was five, he dictated a "story," which I wrote down. The 146-word, 23-main clause story includes only one subordinate clause: "When he got home." It is no indication that he had even begun to develop a sense of subordinate clauses, for, if we think about a five-year-old’s mental experience, we quickly realize that he must have heard "When ________ get(s) home" thousands of times. "When we get home." "When daddy gets home." "When you get home." "When mommy gets home." The clause was thus absorbed as a unit--it is no more an indication of the child’s mastery of subordination than his ability to tell time by the clock is an indication of his mastery of the concept of time. (See Lovell's "The Concept of Time.")

     The affect of formulas on linguistic development requires and deserves further study, especially since it is possible, if not probable, that every new construction in an individual’s repertoire first appears as a formula. The appositive, for example, appears in "your uncle Bob." "It is true that . . . " is the formulaic beginning of the delayed subject. The common expression "all things being equal" is a noun absolute. Formulas are, of course, absorbed by the child from his linguistic environment. Thus, children of the more educated are more linguistically advanced, as are children who are read to and those who develop a habit of reading. 

     Speech, even the speech of the educated, tends to be elliptic, the context supplying the meaning. "Bread," in speech, suffices for "Where is the bread?" or "Pass the bread." or "Should I buy bread?" In writing, on the other hand, the full syntactic pattern is filled out, and, we might add, some constructions, such as the gerundive and the noun absolute, are more frequently met with. Walter Loban has made some interesting studies of the differences between the acquisition of speech and writing, and notes, for example, that 

from grades one through seven the oral average words per unit tends to be slightly higher than the written average. In grades seven through nine a rapprochement seems to be occurring, and in grades ten through twelve longer units occur in writing. (34) 

     But once again our topic, amoeba-like, divides on us, for we must distinguish between the formulaic acquisition of syntactic structures, and the ability to combine these structures once they have been acquired. David Crystal, for example, is speaking of the former when he writes:

     If I had to put a figure on it, I would say that 5-year-olds have acquired over 80 percent of the grammatical structure of their language.
     To verify this impression of competence, there is a simple exercise which anyone can carry out, given a tape recorder and patience. You find a tame 5-year-old, and record a sample of speech, on his home ground, under no pressure, when he is happily playing or chatting to someone he knows. You transcribe the recording, and go through with your pencil at the ready, looking for errors of grammar. My experience is that there will be pages of transcription with no pencil marks on them at all. You then compare the sample with the range of structures found in adult grammar. The language will be immature and limited in range, undeniably; but it is extremely competent. ("Past" 23)

     The distinction between formulaic development of constructions and the ability to combine them is a little harder to see, but perhaps it can be suggested by the following question: would a child who was brought up in a totally oral environment, a child who had never read -- or been read to -- spontaneously write a sentence such as the following from James Baldwin’s "Equal in Paris":

     I was then living on the top floor of a ludicrously grim hotel on the rue du Bac, one of those enormous dark, cold, and hideous establishments in which Paris abounds that seem to breathe forth, in their airless, humid, stone-cold halls, the weak light, scurrying chambermaids, and creaking stairs, an odor of gentility long long dead.

     What is important here is that, as I shall attempt to show in Chapter Eight, students have problems, not with individual constructions, but with keeping the connections straight when they combine them. (Sentence-combining exercises fail because they have been used without a theory of natural development and thus ride over students's problems like a bulldozer - forcing the students to combine more, but not helping students consciously understand what they are doing and why.
 

The Principle of Reduction

     As our example of "We went fishing and played catch" suggests, expansion, whether through compounding or through formulas, cannot go very far without the trimming tendency of reduction -- expansion of the verb here results from the reduction of the main clause-- "and we played catch." Both Hunt and O’Donnell emphasize the importance of reduction, especially for the later stages of syntactic development. Hunt describes the principle so neatly, and so nicely explains some of its implications, that I’ll let him speak. He is using as his example the reduction of "Moby Dick was a whale. The whale was very strong." into "Moby Dick was a very strong whale":

     Important to this study is the effect of rewriting two clauses as one clause with a modifier. That process involves the reduction of one whole clause (the whale was very strong) to a nonclause structure (very strong whale), in this instance as a modifier of a noun. The subject-predicate relationship disappears, and so one clause ceases to exist. The modifier is shifted to another clause. In effect, two clauses have been consolidated into one.
     Note the effect of this reduction and consolidation on our statistics. Originally no "modifier of a noun" appeared in either clause. Now one does, an adjective modified by an intensifier. Originally we had two clauses averaging 4.5 words long (counting "Moby Dick" as one word). Now our average clause length (for one clause) is 6 words. This gain of 1.5 words in average clause length is exactly the amount of gain that occurs, on the average, between fourth and eighth grades.
     There is another far more impressive gain, however, to be seen in this example. That is the gain in succinctness. What was said originally in 11 words is said, as a result of consolidation, in 6 words. Almost half the original wordage was waste. If that kind of gain in succinctness continues over the years, then the amount of meaning expressed, on the whole, in a thousand words of twelfth grade writing will be many times that expressed in a thousand words of fourth grade writing.
     We need not suppose that the older student thinks in short simple clauses which he then revises, consolidating them into longer and longer clauses and longer and longer T-units. There is, however, good reason to believe that when he was younger he thought in those terms. At least we can be sure that he wrote in those structures. It takes him many years to learn to write -- perhaps to learn to think -- in terms of longer clauses and T-units with vastly more intricate relationships in each one. We cannot see the process of consolidation at work, but we can see the results of it over the years. (Grammatical 105-106)

     It has been pointed out that with advances in grade, the children often increased significantly their use of certain types of structures that can be identified as transformations involving deletion rules. It might reasonably be proposed that growing power to manipulate syntax is better measured by relative use of such structures than by subordination indexes. (98)

     Reduction usually involves two processes, one of which, DELETION, has already been introduced. Whereas deletion denotes the trimming away of, to paraphrase Hunt, wasted wordage, the other, EMBEDDING, denotes the placing of the remainder into another main clause structure. This distinction is important because, as Hunt notes, "we cannot see the process of consolidation at work, but we can see the results of it over the years." What we see, what we can count, are the embeddings. Looking at a finished piece of writing, we can never know how the words first came to the writer. Did I, for example, first write:

As we look at a finished piece of writing, we can never know how the words first came to the writer.

The Principle of Subordination

     In a broad sense, every reduction is a subordination. To use Hunt’s example:

a.) Moby Dick was a whale.
b.) The whale was very strong.
c.) Moby Dick was a very strong whale.
d.) The whale Moby Dick was very strong.

Hunt stops at (c), but (d) indicates that (c) is only one of two possible reductions. An interesting experiment would be to give children (a) and (b) and ask them to combine them such that they get rid of one "was." Among the youngest children who were able to follow the instructions, would more produce (c) or (d)? Here, however, my point is that (c) emphasizes Moby’s whaleness and subordinates his strength, whereas (d) emphasizes the strength, subordinating the whaleness. (For more about this, see the discussion of Jespersen’s concepts of nexus and junction in Chapter Two.) Subordination in this broad sense thus begins concurrently with reduction.

     In the narrower sense with which we are primarily concerned, subordination is limited to clauses -- the subordination of one whole predication to another. As we have seen, subordinate clauses appear, as formulas, in the speech of most five-year-olds, and it is not at all surprising to see a few in a child’s early writing. Hunt and O’Donnell even concluded that subordinate clauses per main clause are a good index of syntactic maturity. The following table indicates the results of their studies, but adds the average yearly increase or decrease in number of subordinate clauses per 100 main clauses:
 

The final column suggests two interesting questions about these statistics. First, there are the tremendous surges in grades 4 and 8. Unfortunately, there are no comparable statistics for grades 9, 10, and 11. Is the 6.5 annual increase between 8th and 12th grade spread out evenly? Or is most of it concentrated in 9th, with a retrenchment in 10th similar to that in 5th? Walter Loban, whose study includes a separate analysis of adjectival, noun, and adverbial clauses for three groups of students, rated "high," "random," and "low," notes that "the High group’s growth in adjectival clauses is centered mainly in junior high school, grades 7, 8, and 9." (52) And he notes that grade 8 is "a year in which all groups show a large increase in the use of adjective clauses," whereas grade twelve is "the year of the Low group’s large increase in adverbial clauses." (53) On noun clauses, he notes that "for all groups the upward spurt in the use of all other noun clauses occurs at grade eight in written language rather than at grade seven as in oral language." (55) Loban’s research thus suggests that the 12 point surge in the table for eighth grade is continued into ninth, and his charts even suggest that this surge is followed by a retrenchment. (52-54)

     If we now stand back and look at our profile of the development of subordinate clauses, we see a surge in fourth grade, followed by a retrenchment, and a similar surge in eighth and ninth, probably also followed by a retrenchment. To what can we attribute these surges, and why are they important? Throughout this book I have been arguing for further research -- we don’t have many of the answers. But to suggest why such research is important, perhaps I should provide some tentative, extremely hypothetical answers. The surge in fourth grade may be due to the students’ introduction into their writing of the formulaic subordinate clauses that have been part of their oral linguistic repertoire. In most schools, students do not even begin writing, as opposed to "copying," until the third grade. As with every normal person entering on a new experience, they keep it short and simple. They also probably keep their "writing" close to the sentences which they have been accustomed to copying, for their spelling exercises, or for their handwriting exercises -- short sentences with few subordinate clauses. By fourth grade, of course, they are already experienced veterans -- they free themselves from their spelling book models and introduce "their own language," including, of course, their formulaic subordinated clauses. The retrenchment in fifth grade may be an indication that some of these formulaic clauses are already being reduced to less-than-clause structures.

     If my hypothetical explanation is correct, and, in the absence of any others, I have no reason to believe that it is not, it implies a major shift in our attitude toward the materials which we present students. The general attitude has long been that materials should be written "downward," both in vocabulary and sentence structure, such that the student can easily understand the content. Thus, Trevor Gambell, writing about what high school teachers think about grammar, notes that

Teachers also found problems with texts which employed multiple choice questions with subordinate clauses. Students had difficulty determining the main idea of the sentence and thus the question; the subordinate clause led to ambiguity and confusion. Obviously if students have problems reading questions with subordinate clauses, they will be reluctant to use such structures in their own writing. (43)

     But such continual downward shifting is precisely what students do NOT need. Which, after all, is more important -- the content of a particular course, or the ability to read? If, as many of us believe, education should be a life-long process, if we want people to read outside the classroom and after graduation, shouldn’t we teach them, if they need it, how to decipher subordinate clauses? And, I would suggest, this process should begin in kindergarten. I’m not suggesting that kindergarten students should have James Baldwin or Aristotle read to them, and I realize that what I’m suggesting does indeed involve a complicated process for the educational community and especially teachers, but "reading levels" should be concerned with complexity as well as with simplicity.

     One way of determining an appropriate level of complexity would be to analyze the syntax of the typical child’s environment. Parents do not always talk down to their children. It is not at all inconceivable that children frequently hear, in response to a request to do something, a statement such as:

When we get to Funworld, if it’s not crowded and your mother says you can spend your allowance, I’ll take you there.

     My point, of course, is that teachers, even the best, can’t teach nearly as well as children can learn. At birth every child starts from zero comprehension and, by age five has deciphered, from his linguistic environment, the majority of syntactic structures and a huge vocabulary. Then he gets to school, where he is often met either by a watery syntax and vocabulary or by a syntax and vocabulary that is far beyond him. (See Chapter Nine.) Obviously, although sometimes it is precisely the obvious that needs stating, not all materials should be of equal syntactic difficulty. If the material is to be tested for content, and I would suggest that less emphasis be placed on content in the early grades, then the syntax should be simple and transparent. But there is no reason why leisure reading, of which there should be more, and such things as handwriting exercises to be copied should not reflect the normal syntax of the child’s environment. A place to begin would be with that spurt of subordinate clauses in fourth graders’ writing that initiated this discussion. The nature of those clauses should be studied, and, since we know that comprehension precedes production, similar clauses should be introduced into children’ initial reading and copying exercises.

     My discussion of a bump on a statistical chart has been rather lengthy and highly theoretical, i.e., not backed by sufficient research, but again, I have tried to suggest why syntactic analysis is pedagogically important, what kinds of questions it may lead to, and what the implications of these questions might be. The syntactic theory necessary for such analysis, moreover, is not at all complex. We are simply looking at subordinate clauses.

     If the surge in 4th grade may be attributed to the releasing in writing of the child’s repertoire of formulated subordinate clauses, the surge in 8th and perhaps 9th is probably more directly associated with the students’ development of logical subordination. I cannot prove this, and hence will leave the question for the educational developmentalists and psychologists, but I would like to suggest an important pedagogical implication of the surge. A major complaint, especially of seventh and eighth grade teachers, is that students write fragments and don’t punctuate sentences correctly. The usual response of seventh and eighth grade teachers is to give students more and more exercises in correcting fragments and in correct punctuation. But most of these exercises involve sentences that are far less complicated that the sentences which the students are trying to write on their own. As John Mellon noted:

it may very well be the case that conventional grammar study fails to promote growth of syntactic fluency not because of the usage practice which it features, but rather because of the hundreds of simply structured and altogether childish sentences which it employs for parsing exercises. (63)

     Part Two of this book suggests an entirely different approach to helping students avoid fragments and faulty punctuation, but I would suggest that the problem as perceived by most seventh and eighth grade teachers is not really a problem at all. Seventh grade is the beginning of that surge that statistically appears in eighth grade. As students experiment with clause structures, they will make mistakes, and the more they experiment, the more mistakes they will make. But pedagogically these "mistakes" may be no more important than the young child’s "cutted the paper" or "readed the book." The "problem," in other words, is not the fragments or faulty punctuation. These are signs of development. The real problem is the individual students’ development and mastery of subordinate clauses. Left to themselves, most students would probably work their way out of the fragments and bad punctuation, just as they worked themselves out of "readed." Ironically, it is the students who have the most problems who are usually presented with the greatest number of bad examples.

1. Does the chapter make a convincing case  a) that natural syntactic development does occur beyond age 5, and b) that studying that development should be helpful in the teaching of grammar?

2. The ideas presented in this chapter are not easy to follow. Which of the ideas should be better explained?

3. Can you think of other examples of the principles of expansion, reduction, and subordination?

4. How is the theory presented here expressed in  the grammar that is being taught in your schools? Assuming that this theory is right, how can the grammar that is being taught be changed to reflect natural syntactic development?