Last Updated 8/6/04
Return to Printable Workbooks Return to Background Essays

The KISS Perspective on
Restrictive and Non-Restrictive Constructions

     One of the areas in which pedagogical grammars like to befuddle teachers and students is the distinction between "restrictive" and "non-restrictive" "clauses." In an excellent short discussion of these concepts, Brock Haussamen notes that "The explanation of the terms restrictive and nonrestrictive occupies a sizable segment of the grammar handbooks." (91) That they do so is somewhat silly since they almost always do so in terms of clauses, but they never teach students how to identify clauses in the first place. As a result, the concepts rarely clarify anything, but they add tremendously to the anxiety and fear of grammar on the part of both teachers and students.
     The basic, either/or distinction between "restrictive" and "non-restrictive" is very simple. A "restrictive" clause restricts (limits) the meaning of the word  that it modifies:

The man who robbed the bank is still in prison.
She played for the team that won the title.
In both of these examples, we can say that the subordinate clause, identifies the word it modifies and is thus essential to the meaning of the sentence. "Non-restrictive" clauses, on the other hand, simply offer additional information: 
The man, who was wearing an orange jail suit, is still in prison.
She played for the team, which is also the team that beat the Yankees.
As Haussamen notes, in many textbooks discussions of "restrictive" and "non-restrictive" are "placed in the chapter on commas because the terms are intended to serve as a guide to punctuation." (91) The general rule is that restrictive clauses are not set off by commas, whereas non-restrictive clauses are. 
     Haussamen demonstrates that although most textbooks present these as basically "either/or" "categories," they are actually "polarities." (93) It is, in other words, easy to find simple examples of "restrictive" and "non-restrictive" clauses, but if we look are real texts, we will find that there are many clauses that are in between the two. Haussamen offers several examples, including:
The committee continued several hours of open hearings today, followed by a closed session in which the panel's members discussed a variety of classified intelligence matters.
He then notes:
Because we are so accustomed to the concepts of restrictive and nonrestrictive, the examples above 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. But in each case the current definitions as presented in the handbooks do not fit; even when the antecedent appears to be indefinite, as in a closed session, the following clause does not restrict or define, but comments on and describes. (95)
Haussamen gives several other interesting examples, but his point should be clear -- many clauses simply do not fit the textbook distinction.
     We need to stop and think about what Haussamen wrote. The distinction occupies a "sizable segment of the grammar handbooks," but as it is taught in the handbooks, it is, in many cases, incomplete. Why, then, we might ask, is it even taught, especially when students are not taught to identify clauses in the first place? Haussamen's discussion may illustrate one of the reasons. He was, I would suggest, torn between two, probably unconscious, objectives. On the one hand, he was writing for teachers of grammar, as a grammarian. At the same time, however, he sincerely wanted to help students.
     Grammarians are trained taxonomists -- they love to create classifications. And, to be accepted by grammarians, one thus creates categories. Haussamen notes, in his historical introduction, that originally, the term used for "restrictive" was "restraining." But "Gould Brown in 1823 made the same point using restrictive instead of restraining and the new term stuck." (92) Haussamen actually seems to prefer "restraining" over "restrictive," but the point here is that we could have four terms ("restraining," "non-restraining," "restrictive," and "non-restrictive") for two -- inadequate -- concepts. (And we wonder why teachers and students are confused?)  And, unfortunately, Haussamen ends his discussion by stating:
     But if we wish to continue classifying clauses using anything like the present categories, we should add a middle category -- the amplifying clause -- if the schema is to be reasonably precise. And I think, while we are at it, we could improve the nomenclature by dropping the distracting name nonrestrictive and replacing it with supplemental. The third category would remain the restrictive clause, defined as we define it now. (96)
Now we have a fifth (amplifying) and a sixth (supplemental) term involved. But we need to note that Haussamen's "we" refers to teachers, and, even more to the teachers of teachers and linguists.
    The grammarians, linguists, and the writers of textbooks thrive on this multiplication of categories. Why else would it occupy a "sizable segment of the grammar handbooks"? The confusion caused by all of this for most teachers and students doesn't seem to concern them. Note that many teachers are already confused by the four terms that textbooks use to designate the two main types of clauses -- "main" and "subordinate" and "independent" and "dependent." And here we are discussing six terms that distinguish types of subordinate (or is it dependent?) clauses. In grammarland, terms and categories multiply faster than feral cats. And the grammarians love it.
      As noted, however, Haussamen is split. He really does care about students, as demonstrated in his next-to-last paragraph:
     Of course, speaking of commas, one approach that would drastically simplify the pedagogy of all this would be to address the punctuation issue more directly and omit the categories as much as possible -- that is, to tell student writers to use commas around extra, nonessential information and let it go at that. (96)
This is a point of major importance for anyone who is truly interested in teaching students -- it clearly implies that the "restrictive" and "non-restrictive" categories are both confusing and not helpful to students.
     Haussamen makes other very important points when he notes that "restrictive" and "non-restrictive" should not be limited to clauses:
Within traditional grammar, restrictive and nonrestrictive have been expanded to refer not only to clauses but to appositives, phrases, and modifiers generally. They have served admirably to help clarify the basis for the use of commas, and bringing some rhyme and reason to the use of the capricious comma was long an elusive goal of grammarians. (92)
The first thing we should note is that Haussamen's pedagogical advice, (discussed above) effectively explains to students what the "restrictive" distinction has done for grammarians -- but without using the terms. Thus, as noted, the distinction is still not really needed for students. As Haussamen stated, "tell student writers to use commas around extra, nonessential information and let it go at that." Getting the distinction out of the textbooks, however, is not going to be easy. It is, after all, a major contributor to the bulk of the books and thus to the profits of the textbook companies. It also adds to the confusion of the students, and thus, to the felt need for bigger and still more expensive books.
     It also gives teachers something to teach. I was recently contacted by a sixth grade grammar teacher who wanted to know "what grammar he should teach" to his students. Users of the KISS Approach would probably agree that that is a somewhat senseless question. What he should teach should depend on what his students already know. But in today's world of education, students are not expected to "know" anything. As a result, teachers do not know what to teach, but the textbooks can give them a ton of concepts, including "restrictive" and "non-restrictive" to fill the vacuum. (Who cares if they are inadequate and confusing? They fill the educational vacuum and put money in the pockets of the textbook writers and the publishers.) 
     The emphasis given to "restrictive" in textbooks has led some people to ask why the KISS Approach has not (until now) dealt with it. The question often came with a subtle suggestion that the KISS Approach is inadequate or incomplete without it. But as Haussamen implied, the distinction is not needed and probably confusing. If we do teach it, we ought to do so in the broader sense also suggested by Haussamen -- it also applies to "appositives, phrases, and modifiers generally." Restrictive modifiers are essential in that they specify what the word being modified refers to. As such, they are not set off by commas.
     I have quoted Haussamen at some length not only because what he wrote is important, but also because he refers to the grammarians' attempt to tame the "capricious comma." Another question that I frequently get is -- Does KISS deal with the rules of punctuation? Some day I hope to have the time to deal with that question at length and more directly, but this site does already deal with punctuation in a fair amount of detail. It assumes, however, that third graders, where the KISS curriculum starts, already know that sentences end in periods, question marks, or exclamation points. Thus it currently focuses on the punctuation of compounded main clauses, which is a major problem for many students. Beyond that, students will best learn how to use the marks of punctuation not by studying a bunch of isolated, and often too simplistic "rules," but rather by following Haussamen's advice, by understanding the KISS psycholinguistic model, and by remembering that the purpose of punctuation is to clarify for readers what words chunk with which.

1. Revising the Rules: Traditional Grammar and Modern Linguistics. 2nd ed. Dubuque, IA: Kendall/Hunt, 1997. 91-96.

This border presents Sandro Botticelli's The Mystical Nativity
(c. 1500) The National Gallery, London
Mark Harden's WWW Artchive
[for educational use only]
Click here for the directory of my backgrounds based on art.