When we read, does the meaning come to us through the words we see or the sounds those words subtly make in our mind? Both, surely, but it’s the balance that counts in finding our own written voice, the style that is our own. The words we write are ultimately the replication of the sounds we make to voice meaning. Those sounds stay on the pages we compose, and so the conscientious writer pays attention to diction, the choice and arrangement of words.
We might think first that the word diction refers to how well we enunciate our words when we speak, and the term certainly does mean that. More exactly, though, diction refers to the vocabulary we choose and the grammatical structures we construct with those words to carry our meaning persuasively to the readers we conceive we are addressing. Diction is a rhetorical concern, but it works closely with grammar to shape our thoughts. It is the result that attends on the way our sentences sound, audibly to our ears or almost inaudibly to our inner ear.
The collection of techniques that determine our diction is called the rhetorical figures, and one such figure, or configuration, is ellipsis, the omission of a word which is otherwise grammatically or logically necessary. Ellipsis is a technique we employ naturally all the time when we speak, and we use it especially frequently with the word that. Now this word that is grammatically quite rich, and its holdings include no fewer than four parts of speech: pronoun, adjective, adverb, and conjunction. It’s the last of these which concerns us here, when that is used as a subordinating conjunction.
Here’s an example: He said that he’d be here an hour ago. As simple as this sentence is to understand, it’s grammatically quite involved, and it is defined, in fact, as a complex sentence because it comprises one independent clause and one subordinate clause. The independent clause is he said. The verb said is transitive, which means it requires a direct object. That object is the entire subordinate clause which follows, that he’d be here an hour ago, and we know that clause is dependent because it begins with the subordinating conjunction that. We call such a dependent clause a noun clause. Read the sentence with attention and you’ll hear a carefulness, an effort on the speaker’s part to be exact.
But how would the diction change if we designed the sentence like this: He said he’d be here an hour ago. We have now omitted the conjunction that, the word that formally introduced the noun clause in our original version, and in doing so we have changed the tone that underlies the statement. Read this sentence with attention and you’ll hear an urgency absent in the first, complete rendition. The rush now is to get to the adverbial phrase an hour ago, perhaps to emphasize or even worry over the long delay. The ellipsis, that is, has produced an apprehension appropriate to a context we might easily imagine.
Or here’s a more sophisticated example: His friends used to say that’s how they knew he’d be famous one day. The word that in the contraction that’s is not the conjunction but a pronoun; the subordinating conjunction has been omitted here (and again before he’d) in order to speed the line to replicate a more familiar, more casual way of speaking. If we restore the two ellipses, we get this explicit version: His friends used to say that that’s how they knew that he’d be famous one day. Trying a bit too hard, it seems.
Is ellipsis the right choice, then? Rhetorical designs do not answer to the criterion of right or wrong. They are responsible, instead, to the standard of appropriate or inappropriate to the effect we wish to create for a particular audience. The trick is not only to see, but hear too, what we write.
***