Between Restrictive and Nonrestrictive: 
Amplifying Clauses

Brock Haussamen
Bridgewater, N.J. 
A paper presented at the
Sixth Annual Conference of the NCTE Assembly for the Teaching of English Grammar,
July 27 & 28, 1995.
Pennsylvania College of Technology, Williamsport, PA. 

My talk today will focus mainly on restrictive and nonrestrictive elements but I want to set this topic in the context of the larger questions and issues about grammar that we have been considering. I see these as three.

General Questions

The first is that we need to keep working our way toward a body of grammatical knowledge that is accurate and informative. People refer to grammar as if it were a set, static, and self-evident body of knowledge. They say, and we say, that students ‘learn grammar’ or that we ‘teach grammar,’ but the knowledge that is taken for granted by that term is varied, and in the case of many of its basic terms and ideas, extremely old. Many of the terms for the parts of speech and syntactic principles go back to the middle ages or the ancient Greeks. No wonder it is difficult to pull school grammar books away from the notional definition of a noun; that definition has two thousand years of Western culture behind it. So part of our on-going job is to give these terms and ideas of grammar renewed attention, rigorous scrutiny, and revision where necessary. My talk will illustrate this, and this is an issue that interests me a great deal.

The second general question is, of all the grammar that we do know, what portions do we teach to students? What do young writers need to know about grammar, and why, and how can we verify that we are coming up with sensible answers here? This is the general question that raises all the heat in discussions about grammar. My talk will illustrate this problem also, but I want to point out here that I think that one reason why we have difficulty getting a handle on it is that it involves more than the concerns with writing and language that are always brought up. Writing behavior is in part social behavior, and so grammar is a social issue, a matter of social class and even political orientation, and I think these are the aspects of grammar that we should not avoid if we are going to have thorough discussions about why we are teaching it.

The third and final question is that once we decide what portions of grammar our students can benefit from, how do we teach such grammar effectively? What strategies work, and which don’t? How might we deflect some of the enormous influence that the current grammar publishing establishment carries in order to open the way for new approaches? The question of teaching strategies has been dealt with in a number of presentations here.

A History of Restrictive and Nonrestrictive

Now to turn to the more concrete issue of the contrast between restrictive and nonrestrictive modifiers. I offer this discussion as an example of the kind of examination and consideration for revision of common grammar terms that we need more of.

The contrast between restrictive and nonrestrictive has historically been one of the most successful conceptualizations of recent grammar theory. It is relatively young as grammar terms go, dating back to Goold Brown’s first use of the term restrictive in 1823. The term emerged from discussions of the lightening of punctuation. That is, in the 17th and 18th centuries, ALL relative clauses were bordered by commas. An example from a 1785 book on punctuation: "Never open your heart to persons, whom you do not know." But ten years later, in 1795, Lindley Murray included in his great grammar text a mention of an exception in the changing practice of the time. ‘When two members are closely connected by a relative, restraining the general notion of the antecedent to a particular sense, the comma should be omitted." Murray’s word, restraining, was replaced by Brown’s use of restricting in his version of the rule shortly after, and the new term stuck. So the inconvenience of having a positive term, restrictive, refer to an absence of punctuation arose from the description of conditions under which certain traditional commas should be left out. Without shifting patterns of punctuation in the 18th century, we might not have the terms at all.

One sign of the success of the two terms is that they have spread from conventional grammar to linguistics, a field which has been very careful about its terminology. Linguistic grammars use the terms, as conventional grammars do, to describe not only relative clauses but modifiers of all kinds in their relation to the term they modify. The main idea is that all modifiers have one of two qualities--they are either essential, tightly bound, defining, and not separated by punctuation, or they are unessential, parenthetic, loosely bound, and separated by commas.

Amplifying Clauses

A couple of linguistic works have pointed out briefly that this duality is not as neat as it appears, but I think that the problem calls for more attention. The two terms are, I believe, polarities, not categories. The conventional grammar books give the impression that all modifying clauses fall under one heading or the other, but many seem to me to fall in between. That is, not all clauses are either defining and essential on the one hand, or parenthetical on the other, as the handbooks state. Many clauses contain information that does not restrict or define the antecedent, yet these clauses are important, essential, sometimes even primary information in the sentence. I will refer to them as amplifying clauses.

Let me give some examples. One type of amplifying clause is that which amplifies an adjective that precedes the antecedent noun. Such clauses are mildly redundant and are very common in speech and informal writing. Some examples from students:

Without the marked modifiers, the relative clauses would be much more restrictive: "a schedule that does not allow many deviations." With the underlined modifiers, however, the nouns have been restricted and defined, and yet one cannot say the relative clauses after them are therefore nonrestrictive to the extent that they are unessential. We might characterize these sentences by saying that the writer has spread the task of description over both the general adjective coming before the noun and the more detailed clause coming after it.

More frequently, however, there is no preceding adjective, the antecedent itself is identified intrinsically or in context, and the amplifying clause makes an important comment about it. Some examples from The New York Times:

And from students: Because we are so accustomed to the concepts of restrictive and nonrestrictive, these examples may seem at first to fit into one category or the other. And it is true that they may each be closer to one pole or the other. That is my point--that the terms represent poles. But in each case the current definitions as presented in the handbooks do not fit easily if at all; even when the antecedent appears to be indefinite, as in a closed session, the following clause does not restrict or define as much as it comments and describes. Even when the antecedent is fully specified, as in the bottle of lce-nine, the following clause is essential to the sentence and not parenthetic; the sentence would be very different without it. These amplifying clauses are generally not separated by commas; this makes sense because although they are not restricting the antecedent, they are still closely linked to it.

The concept of the amplifying clause would not be interesting if the term non-restrictive meant just what it said--anything other than restrictive. But it is not defined that way; it means instead a modifier that is nonessential, and the reason it is defined in this somewhat extreme way has to do with punctuation: the term has its roots in, and has remained primarily, guidance about the use of the comma. If a clause is judged to be nonrestrictive, the practical consequence is that commas separate it from the antecedent. But as long as restrictive and nonrestrictive serve as punctuation guides, they are compromised as precise descriptions of the syntax or semantics of modifiers, which include gradations in the relationship to the antecedent. If we want to retain the present terminology as a guide to commas use, a concept such as the amplifying clause would be helpful to describe a great many sentences.

Conclusion

This sort of reexamination of the familiar terms of grammar is an important process, and I hope that my book Revising the Rules: Traditional Grammar and Modern Linguistics might serve as an informative starting point for a fresh look at a wide range of grammar concepts.

But such reexamination quickly leads us to the second question I raised at the beginning: Which topics of grammar do we pass along to students, and why do we choose those? The reconsideration of restrictive and nonrestrictive modifiers that I have presented should lead us to consider not only their accuracy as grammatical concepts but also their pedagogical usefulness. Some might feel that the amplifying clause should be added to the mass of abstractions already heaped on some students. But I think that the practicalities of punctuation are the real writing issues here, and that we should consider telling students in writing classes to put commas around extra, nonessential information and letting it go at that, dropping from those classes and their textbooks the baggage of the restrictive and nonrestrictive concepts. We can retain them of course in grammar and linguistics courses, but we as grammarians perhaps need to become better than we sometimes are at distinguishing the grammar information that we could teach writers from the grammar information that writers really need.


[Editor's Note: Revising the Rules (1993, ISBN: 0-8403-9032-7) is available from Kendall/Hunt Publishing Company, 4050 Westmark Drive, Dubuque, Iowa 52002.]