One of the more persistent difficulties in writing is where to place a comma. Many of us were taught to put a comma where you breathe, and I am of the opinion that that well-meaning directive has caused more problems than it has solved. I say this not because the instruction is without merit, but because it is actually too subtle a technique for the majority of workaday situations we encounter. Better it would be to find an answer in the structure of what we have written.
Take, for example, this: The problem with this new refrigerator, is that it doesn’t have enough space for small jars and bottles. That’s a reasonably involved sentence, and in saying the words aloud you might very well pause for a nanosecond after the word refrigerator. And if you then remembered the advice of your good English teacher, you might next decide to storm heaven and confidently pitch your comma at predicate’s gate. Now if you were born to write English 150 years ago, you would find yourself in good compositional company, for then it was more common to do just what we are censuring here. But just what are we censuring?
A quick analysis will show us that the subject (what the sentence is saying something about) is the problem with this new refrigerator, and that the predicate (what is being said about the subject) begins with the very next word, the verb is, and continues to the period. The writer, then, has placed a comma between the subject and predicate in an effort to distinguish the two, most likely making a special effort to emphasize the assertion. But therein lies the problem. Writers of modern English, more logical than many of their more rhetorical forebears, do not want to separate subject from predicate, recognizing that the two halves are strictly correlative: every subject must have a predicate, so inserting a comma between the two is more interruptive than clarifying. And that is a decision we can come to only upon understanding the form of what we’ve written, its structure, not upon an appeal to our ear and the aural features it is meant to detect.
The working observation, then, is: do not place a comma between the subject and its predicate. But there are (and you knew there would be) exceptions. If the subject phrase is long and winding, the writer is taking some risk that the reader will grow forgetful before the predicate is reached. In such a case, a comma between the subject and predicate is, at least arguably, justified. Take, for example, this sentence I wrote in an earlier post (Balanced and Parallel): To see, as we have done, that each half of the sentence comprises an independent clause with a compound predicate, is to note evidence of parallelism. The subject phrase here begins with the infinitive to see, and the object of that infinitive is the fourteen-word noun clause beginning with the conjunction that. These two corresponding parts, moreover, are separated from each other by the parenthesis as we have done, and all that, I judged, was sufficient to justify a comma where one would not usually be placed. The point to note is that the decision was based first on clarity, not sound—and clarity is the first requirement of expository prose.
Rules for the comma abound, and there is, in fact, a rightful place for the ear to make a contribution in deciding where to place the mark. That rightful place, though, is down the road, after one has decided the issue on matters of understanding and substance. Then, should we decide that there is an appreciable danger of confusing the reader, we can turn to the ear to help sort things out. But how something sounds will never first and definitively answer matters of grammar and logic, for that is not the function of rhetoric and its organ, the ear. The relation of these three subjects, grammar, logic, and rhetoric, is complementary, and we are to orchestrate them case by case, into a harmony both disciplined and natural.