Some students, especially those who learned the language orally and did not do a lot of reading, make frequent errors related to clause boundaries. These errors are serious because the clause is the primary unit in the process of reading. As words enter a reader's short-term memory, they are chunked to each other until everything that is supposed to go together is eventually chunked to the S/V/C pattern of a main clause. At the end of that clause, the writer should have used punctuation which signals a dump to long-term memory. If parts of the pattern are missing, or if the punctuation is missing or incorrect, the reader becomes confused. Although students do not need to know the names for various errors, teachers do so that they can understand the nature of the students' problems and determine what, if anything, to do about them. Because clause-boundary errors are a focus of "Cobweb Corner," my research area, the following brief explanations include links to the relevant discussions in that research material.
     To my knowledge, almost nothing that we are currently doing in our schools helps students with these problems. The reasons for that are simple. The sentences in the exercises that students are given to work with are too simple, much less simple than the sentences that the students often themselves write ( and thus have problems with). Then there are the teachers who tell students to put a period wherever they would make a "long" pause in speech. That advice is simply stupid. "Because we talk in fragments."
     The KISS Approach definitely helps students because as they analyze real sentences from randomly selected texts, they come to learn how sentences -- and punctuation -- work. We need to understand, moreover, that we cannot expect immediate results. Under pressure, as in in-class writing, students will still make mistakes, and as for out-of-class writing, we need to teach students the difference between editing and revising. Then we need to force them first to revise and then to edit.

Fragments

     As the name suggests, fragments are parts of sentence patterns that are punctuated as complete sentences. Often, fragments are the result of an overload of the writer's STM. With STM overloaded, the inexperienced writer simply puts down a period (or some other main-clause-ending punctuation mark), and then writes the rest of the main clause (often a subordinate clause) as a separate sentence. In the KISS Approach, the teacher's job is to point out to students that the fragment can probably be connected to the sentence either before or after it. If a number of students in a class are having problems with fragments, a teacher might want to spend NO MORE THAN TEN MINUTES explaining the problem and giving the class as a whole a few examples of how to make this connection. Otherwise, this problem should be dealt with on an individual basis as students either analyze or edit their writing.
     The problem with fragments appears to be most common in grades seven through nine. For anyone familiar with the research of Hunt, O'Donnell, and Loban, this is not surprising because these researchers have convincingly demonstrated that these years are the period of most intense growth in the use of subordinate clauses. Unfortunately, most English teachers are not familiar with this research, and even more unfortunately, almost no thought has been given to its implications. It is quite possible, for example, that the current attempts to "help" these students actually do more harm than good.
     We know, for example, that as young children we all said such things as "I cutted the paper," and "Turn on it" (for "Turn it on.") Even if no one EVER corrected us, we all learned the correct forms. But just as the learning of irregular verb forms is part of natural syntactic development, so is the growth of subordinate clauses. Perhaps the only really important difference is that subordinate clause growth occurs well into the school years, when teachers feel that they have to "do something." But instruction, by its very nature, is an intervention into the "natural." And what we tend to do is to impress upon students -- by the very fact that we give them these exercises to avoid fragments -- that there is a problem, but the exercises we give them do not work - for the reasons stated above. By interfering, in other words, we might well be making the problem worse -- adding both anxiety and lack of clarity to it.
     I would like to see a lot more research done on actual students' writing to determine the nature of students' fragments, their relative frequency (per main clause), and the grade levels at which they occur. I'm wondering if fragments that occur in students' writing before ninth grade should simply be ignored. (Teachers might correct them in students' writing, but not count off for them or do any instruction about them.) If students are being taught using the KISS Curriculum, they will be learning to analyze the clauses in their own writing in seventh and eighth grades. In this process, they will begin to recognize any fragments in their writing, and, as suggested above, they will have a clear context for understanding the problem -- and for fixing it. 
     We should not leave the question of fragments without noting that some fragments are totally acceptable. Currently, instruction is vague about which are acceptable and which are not, but the KISS Approach here, as almost always, relies on the psycholinguistic model of how the brain processes language: a fragment that might cause a crash is bad; one that probably will not, is not only acceptable but sometimes a sign of good writing. Good fragments usually, but not always, appear at the beginning of a paragraph, where they establish a topic or attitude that is developed in the paragraph, or at the end of a paragraph, where the reader can obviously see the coming paragraph break and will therefore not expect a completion to the fragment.
(Click here for the Cobweb Corner discussion of fragments.)
 

Comma-splices and Run-ons

       Comma-splices and run-ons are related in that two main clauses are joined by only a comma (CS) or the second main clause runs into the first with no punctuation between them (RO). These errors create the exact opposite of the problem created by a fragment. Instead of being directed to dump to long-term memory with only a partial pattern in STM, the reader has a complete pattern in STM and starts trying to chunk the words from the next pattern into the previous one. Because they don't chunk, a crash may occur. I say "may" because, as most grammar textbooks state, comma-splices are acceptable if the main clauses are short. Unfortunately, they do not say how long "short" can be. 
     KISS, relying on the psycholinguistic model, states that if the intended readers can be expected to have no problems processing the sentence, then the splices should be considered acceptable. Parallel constructions, for example, make sentences easier to process, and adults can process longer sentences than can fifth graders. This still leaves the question with a subjective answer. The KISS Approach would settle any questionable case in the student's favor, provided, of course, that the student has been taught through the KISS Approach, and therefore understands that the splices might cause readers to crash.
      The opening of Dickens' A Tale of Two Cities, by the way, was included among the exercises precisely because of his use of comma-splices and parallel constructions. For the Cobweb Corner discussion of comma-splices and run-ons, click here.
 

Incomplete Subordination

     I haven't had the time or opportunity to collect a lot of examples of it, but teachers should expect to see cases of incomplete subordination:

     Although the author Kent Scheidegger of the essay "Habeas Corpus is Abused by Convicts" relays many good examples of the abuse of this procedure, but the fallacies in which the author commits weakens his essay and argument dramatically. [See also "in which.] 

[In this case, the writer has subordinated the first clause with "although," but has retained the "but" that would join two main clauses.]