Just what is our imagination? We can’t get very far in any thinking about the arts, including the art of writing, before we find ourselves using the word to refer to some kind of thing we sense is going on in our heads to produce a work others can see or hear or read. But notice where I just placed this thing we call the imagination: in our heads. That is a very modern and precarious idea, because most of the world would have most of the time in the past referred to the imagination as a power not in the head, but in the heart. Much depends on our definition of what is human. Traditionally, the head thinks and the heart feels.

Our modern dictionaries say that our imagination is “the act or power of forming a mental image of something not present to the senses or never before wholly perceived in reality” (Merriam-Webster); or the act “of creating new images or ideas by combining previous experiences” (Webster’s New World); or simply “the creative faculty of the mind” (Concise Oxford). Once again the mind. But to imagine, according to the ancients, is to give rise to, to allow to be, not to make to be: to create, not to contrive. It requires a passive and receptive disposition from the working artist, and it seems to turn away when too many questions are asked. And that is an immensely consequential idea to be clear about, because it means our artistic work is a becoming aware of goings-on elsewhere, which we learn to watch and watch over in order to understand and then represent with reason in some significant form.

It is probably correct to say that we popularly associate the word imagination first with the poets, but of course this power lies in some degree behind all creative work. Still, many poets, the imposing John Keats among them, struggled to understand the sometimes unmanageable force which compelled them to attend to a new arising of images demanding form. The American scholar and critic Clarence Thorpe, in his The Mind of John Keats (1926), says this about the English poet’s ideas across the span of an intense but brief life:

In both his letters and his poems Keats has, first to last, many things to say about the imagination. Always he understood it to be the supreme active principle in poetic composition. But how decisively with the rapid maturity of his intellect and critical judgment during the brief, fitful working period of his life did his ideas change and develop…. Imagination, with its springs in the heart rather than the head, though the head too has its place, becomes with Keats the highest and most authentic guide to truth.

That third sentence of Thorpe’s is itself a lesson in the art of writing (an inverted structure in which all of the predicate elements, normally following the verbs, precede them for emphatic isolation), but more to the point here is his reference to the heart, which Keats, along with received tradition, identified as the source and place of the imagination. Thorpe continues:

Not only is the imagination to be trusted more implicitly than reason in matters where both are operative, but there are even things clear to the imagination of which the reason knows nothing.

Again the ancient position. Striking and maintaining this balance between imagination and reason, heart and head, is the very work of artistic composition. Too much imagination will produce only fantasies about what we cannot see, but too much reason will tie us tight to only what we can. The third-century Greek philosopher Plotinus (quoted in E. F. Carritt’s Philosophies of Beauty, 1931) warns of this:

Deliberate reasoning occurs in our mortal life when the soul is uncertain and troubled and not at its best. For the need of reasoning is a defect or inadequacy of apprehension. So in the arts; when artists falter, reasoning takes the reins; but when there is no hitch their imagination governs them and achieves the work.

Plotinus’s “inadequacy of apprehension” is the English philosopher Owen Barfield’s “mere perception,” when he remarks in his The Rediscovery of Meaning (1977) that “mere perception—perception without imagination—is the sword thrust between spirit and matter.” He goes on to say that this split, or gap, is particularly Western in its vehemence, but that it is imagination which can traverse it:

Now imagination does not disregard the gap; it depends on it. It lives in it as our very self-consciousness does, in this case not as a small helpless creature caught in a trap between the two, but rather as a rainbow spanning the two precipices and linking them harmoniously together. The concern of imagination is neither with mere matter nor with pure spirit.

But are we now any the closer to understanding just what the imagination is? Perhaps not strictly so, but with Barfield’s image of a rainbow spanning the two precipices of matter and spirit, the visible and invisible, we are in the effective presence of imagination itself. And if we take Plotinus’s instruction that we not think but contemplate that image, we might get far into just what imagination is. Perhaps even so far as to feel it at hand.

***

How are we to understand the semicolon? The traditional grammars say that “the semicolon is used when the clauses are of the same general nature and contribute to the same general effect, especially if one or more of them contain commas” (Kittredge and Farley, Advanced English Grammar); that it “indicates a wider break in thought or construction than the comma, but less than the period” (Foerster and Steadman, Writing and Thinking); and that “scarcely ever will it work with any subordinate conjunction” (Carey, Mind the Stop). That’s the theory, but what does that look like in practice?

The American writer and naturalist John Burroughs wrote lyrical prose about the natural world, and in 1920 he published a series of essays entitled Accepting the Universe: Essays in Naturalism. The fifth paragraph of that work opens with these two sentences, which, aside from their philosophical insight, have much to teach us about the art of writing and the use of the semicolon:

In our practical lives we are compelled to separate good from evil—the one being that which favors our well-being, and the other that which antagonizes it; but, viewed as a whole, the universe is all good; it is an infinite complex of compensations out of which worlds and systems of worlds, and all which they hold, have emerged, and are emerging, and will emerge. This is not the language of the heart or of the emotions—our anthropomorphism cries out against it—but it is the language of serene, impartial reason.

The semicolon appears twice in the first sentence of that passage and it plays by the rules both times: the first instance separates two coordinate clauses, each of which involves the idea of good and evil, and its second appearance abuts the idea that the universe is all good up against the further corresponding idea of its being an infinite complex of compensations. A schematic of the grammatical arrangement would look like this: good and evil; good | all good; compensations, and that reveals an architectonic, or unified, design, which contributes to the meditative quality of the language.

The sentence, though, is long and demanding: sixty-five words, nine clauses—that means nine assertions, nine discrete thoughts which the writer felt compelled to tie together in one semantic breath—along with four phrases (there are even more doing other work) which establish layers of conceptual involvement as the sentence proceeds. That grand structure accounts for the literary texture the sentence weaves, and when the thought is as elaborate as we’ve analyzed, it’s difficult both to write and to read. Right there, though, the sentence offers an instructive comparison with contemporary prose, for nowadays we would be inclined either to chop this heavy original into a number of lighter, distinct sentences, or to retain the phrases and clauses but replace the semicolons with commas.

This latter revision is a common problem in much of simplified, contemporary writing, and if we decided upon it, the sentence would unthread and collect in strands at the bottom of our attention. It is important, therefore, to note and underline the observation those traditional grammars regularly make about using the semicolon: when one of the clauses joined together has internal punctuation of its own (usually commas, but also parentheses and dashes), we best bind it with the next stronger piece of punctuation, which is the semicolon. Here it is helpful to remember that the ascending order of strength among the three most common marks is: comma, then semicolon, then period. If a clause has internal commas of its own, then a semicolon will signal to the reader the larger groups which are to be read as coordinate in meaning. This prevents the reader from falling into a confusion of uncorrelated ideas.

Meaning is coordinate in grammar when elemental structures (word, phrases, and clauses) match across a statement in expressing corresponding ideas. Establishing coordinate meaning is the job of coordinating conjunctions, one of which is but, which alerts the reader that what is about to follow is to be understood in some degree of contrast or even direct opposition to what has just been read. Coordination and subordination also involve the logic of the meaning, a topic beyond our present scope; we should note in passing, though, that what we are saying, its logic, can never be separated from how we’re saying it, its rhetoric. Change the form and we change the angle from which our readers will see the ideas we’re meaning to call to their attention. Perspective is essential to our vision of the truth.

In our example, the conjunction but intends to mark the opposition between our compulsion to separate good from evil in most everything, and the truth, as Burroughs sees it, that the universe is all good, not just in places but in its entirety. The adjectives good and evil are coordinated in opposition to the adjective good. The semicolon immediately before the conjunction has the effect of pausing the reader’s attention to absorb what was just said, because the unaccustomed assertion about the replete goodness of the universe risks shaking the credibility of the writer. This is the reason for the subsequent phrases and clauses in the first sentence which substantiate that claim, and even more so is it the purpose of the sentence which follows: to affirm outright that to believe this is not starry-eyed sentimentality, but the truth when seen dispassionately and rationally.

All this about the semicolons, but we have not said anything about the three dashes in this passage—still another topic altogether. We should note, however, that the sometimes heard advice to always use dashes in pairs is not correct—as the first sentence of Burroughs’s passage and as the sentences in this paragraph illustrate. What is good advice, though—and what I have flagrantly disregarded here—is not to overdo the use of dashes. Alas—but to every choice its purpose.

***

A curious reply has become more common lately, it seems, in congressional hearings broadcast to the public. A question is posed by a committee member, and a witness replies, that is a hypothetical and I can’t answer that, or that is a hypothetical, and it is not worth discussing because it will never happen. What is a hypothetical and are such replies defensible?

A hypothetical is a conditional statement in which we connect a condition to a result: if it is sunny tomorrow, I will go to the beach. Every hypothetical contains an if-clause (called the antecedent), which names the assumption on which something else depends, together with a then-clause (called the consequent), which states the result or outcome arising from that condition. The conjunction then is often, as in our example, unstated for stylistic reasons. The two clauses may appear in reverse order for rhetorical purposes (I will go to the beach tomorrow if it is sunny), but whether the condition appears first or last, a hypothetical gives us a way to step off the familiar empirical stage of determined facts and enter a theoretical world where we can think and plan freely, without the commitments and inevitabilities that are part of the hard and fast world of events and their consequences.

We would have little reasoning and even less science without hypothetical thinking, and so to reply that one will not answer a hypothetical because it is a hypothetical, implying thereby that such a statement is no more than devious suggestion, is mistaken in every way. It is logically possible to attribute that motive to the questioner, but it has nothing to do with posing a hypothetical as such. Anytime we plan something, we are involved in hypothetical reasoning, though we may not realize that because we often do not explicitly state the condition (we might say, for example, I could go tomorrow and leave it at that, without adding the condition if it doesn’t rain). And the second reply, that a given hypothetical is unworthy of a response because the condition set down will never happen, is again indefensible on logical grounds, but that response requires a closer look at the structure of the hypothetical sentence to understand why.

To make an assertion, whether affirmative or negative, we put down a subject and join it to a predicate. If I say that it is sunny or I will go to the beach, you are in a position to agree or disagree with the truth of either of those statements which I am declaring as fact, either present or in all likelihood to be present. That is what I, as speaker, believe to be true, and what you, as interlocutor, are in a position to agree or disagree with; we both believe in truth, and we both believe in the possibility of language to symbolize it. Such statements are called categorical propositions, and they are the means by which we define our experience and build up a rational understanding of it. Note that these statements do not include the if and then because they are being asserted independently as facts in their own right.

When, however, we place a categorical proposition within the if-then frame of a hypothetical (properly called a hypothetical, or conditional, proposition), we are not, strangely enough, asserting the independent truth of either the antecedent or the consequent. We are only asserting a connection between two assertions, each of which we are considering not as factual, but as theoretical. If it is sunny tomorrow does not mean that it will in fact be sunny tomorrow, and to say that I will go to the beach does not mean that in fact you will see me sitting under an umbrella on the sand. The only assertion that a hypothetical proposition like this makes is the almost mechanically inevitable association between a certain condition and a certain result—not the existing truth of either. We can speak of the truth or falsity of a hypothetical, but what is true or false is the connection between the two assertions—when this happens, that will happen—not the assertions taken each by itself.

So to reply that the condition of a certain hypothetical will never happen is to miss the entire point of what a hypothetical statement is meant to make possible: to think things out imaginatively, that is, in our minds, where it is much easier to change circumstances than in the brick-and-mortar world. But we can be patient with hypotheticals only on two presumptions: that no one of us has the entire truth about anything at all, and that all of us want first and above all the truth. Evasion is not truth-seeking, and that is all we’re doing when we refuse to climb into the imagination and see what will or could result when or should something else happen. Everything is possible in the imaginative, theoretical world, but we shouldn’t want everything that could possibly happen, to happen. Thus the importance of the hypothetical.

***

How a sentence unfolds determines what effect it will have on its readers, and a good measure of the art of writing involves shaping and placing the three basic elements of word, phrase, and clause to compose a suitable order. The questions who is my audience? and why am I writing this? remain the two fundamental considerations we must be able to answer before and after composing our first draft in order to find that suitable order—before, in order to set off in the right direction and after, in order to ensure we are staying on the path to our conceptual destination.

We can use the two traditional terms loose and periodic to determine the shape of a sentence we have composed and decide whether or not to revise it. A loose sentence places the independent clause first, and lets the associated detail follow in its train. Someone under formal questioning, for example, may reply: I cannot answer that question because of ongoing litigation and advice of counsel. Here, the independent clause opens the statement, and a prepositional phrase follows (the two words because of constitute one compound preposition). A sentence of this shape conveys a straightforward frame of mind, one prepared to state boldly, without caution or qualification, the central assertion to be made: I cannot answer that question.

A periodic sentence, however, begins with the qualifying information, and the statement becomes more intricate because it delays the primary assertion by presenting the reasons first. The periodic arrangement of our example sentence would then produce this pattern: Because of ongoing litigation and advice of counsel, I cannot answer that question. The two structural units of the sentence are simply reversed; the prepositional phrase now initiates the statement, but the two objects of the preposition (ongoing litigation and advice of counsel), each of which names a cause for not answering the question posed, act as clarions announcing the impending arrival of the conclusion to be made by the independent clause. Shaped this way, a sentence appeals more to reason than to the confidence (or perhaps overconfidence) of the speaker.

A further technical term, abridgment, can be helpful in reshaping a sentence in revision. To abridge something is to shorten it by omission, and in grammar we refer to abridged clauses, meaning that a word or phrase of some sort, complete in its own right, is taking the place of a clause, which by definition would otherwise present a subject together with its finite verb. We could, therefore, regard the prepositional phrase in our example sentence, whether in the loose or periodic version, as an abridgment of these two clauses: because litigation is ongoing and because counsel so advises. The compaction of the original prepositional phrase has now been expanded into a full subject-predicate articulation, striking the notes of cause and reason even more dominantly: Because litigation is ongoing and because counsel so advises, I cannot answer that question. In most circumstances, this design would overshoot the mark and risk impertinence, but we have it in our kit nonetheless should the need for bold assertion bordering on defiance be necessary.

The scholar and historian Richard Weaver wrote an excellent textbook on writing entitled A Rhetoric and Composition Handbook (1974), and he says the following on the topic we have been discussing here:

One does not make either loose or periodic sentences an object of writing. Ordinary attention to the requirements of a subject will tell one when a thought is best presented in a loose structure and when in a periodic or suspended one. We discover, as a matter of practice, that the loose sentence tends to be more common and that the periodic sentence is better for those less frequent occasions when we wish to arouse the reader’s attention by suspension of meaning. When we are revising manuscript for the purpose of improving the general effectiveness of our style, however, it can be quite helpful to identify the two types. If the writing strikes us as dull and inert, it may be that we have adhered monotonously to the loose sentence…. But if, on the other hand, the writing appears artificial and overdone in its striving for effect, we may have fallen into a habit of writing periodic sentences.

Weaver’s first sentence is important to understand. He means that we should not sit down with the deliberate intention to write a loose construction here and a periodic one there. That would only produce what is aptly called a studied style, something thought out, self-conscious, and therefore awkward and unnatural. It is instead, he says, only with attention to the subject—and to the likely audience and intended purpose—that our written voice will fall into those forms most befitting its meaning. Never can we do this always at the first draft, and so these technicals of the art of composition are there for us to apply knowingly in revision, when on rereading we have recognized that a sentence or a passage could present itself more precisely and appealingly. And that, as Weaver says, is a matter of practice, not strategy.

***

Life is movement. It acts, it reacts, it changes, and what appears to remain the same is just that—an appearance. That is why art, which tries to imitate life, must replicate life’s movement in its constructions in order to affect, to move, its viewers or listeners or readers. How to do this is a matter of understanding the techniques of the particular art one is executing.

In the art of writing, those techniques include understanding the classes of words, called the parts of speech, and then knowing how to combine those parts in such a way that action, or at least some approximation toward it, appears to be embedded in the structure of the sentences we compose. Nouns and verbs, of course, are the two principal parts of speech, but there is more going on with each of those classes than what we probably remember from our school days: nouns name things and verbs show action. Yes, but there’s more.

Nouns name things, but in doing so, they’re often freezing an action. If I speak of the removal of an old tree in my backyard, the noun removal is a way to refer from some theoretical distance to the direct action of removing an old tree. Removal is termed an abstract noun, because we don’t apprehend its meaning with one of our five senses, but conceptually with our mind. Removing, on the other hand, is termed a gerund; it refers to the action associated with the verb to remove, and so choosing the gerund in revising a sentence is one way to write toward action. If in my first draft I wrote the removal of that old tree is scheduled for next week, my first revision might be removing that old tree is scheduled for next week.

There’s something clunky in that revision, though, and it has to do with the verb phrase is scheduled. Verbs have voice, and English verbs have two voices, active and passive. If the subject of a verb is an agent, the doer of the action, then the verb is said to be in the active voice. And since we’re trying to write toward action, we should more often than not prefer the active voice. In the sentence removing that old tree is scheduled for next week, the gerund removing is the grammatical subject of the verb is scheduled, and the agent remains unstated: who exactly is going to schedule the removing? Recognizing that, we can now take our original sentence one more step toward depicting action by supplying an agent: I scheduled the removing of that old tree for next week.

That too, though, seems unwieldy, and the reason has to do with the fact (and this is very common) that I am fixating on the wrong action, on scheduling rather than on removing. If I now convert the gerund removing into a transitive verb (a verb which has a direct object) and attach to it an agent-subject, the meaning comes alive: I will remove that old tree next week. This syntactical formula of agent-subject + transitive verb + direct object constitutes the ground level, the foundation, of precise and active English. And if one objects that this version overshoots the mark, that the agent-subject is not, in fact, the person who will literally do the work of removing that old tree, then we have in our toolkit what is called a causative construction, where the agent-subject is bringing about an action which another agent will perform: I will have that old tree removed next week.

Notice that with the word removed, called a passive participle, a remnant of the passive voice has returned to the statement: it suggests that the tree will be removed by that second agent who is not stated as such. But that might be, given context and purpose, as far as we can get toward all-out action in this sentence (though further revisions are possible: I scheduled a landscaper to remove that old tree next week. A landscaper will remove that old tree next week). And such a limitation is important to realize. Our writing toward action is really a counsel of perfection, something we always try to achieve, but an achievement we know cannot always be had, because the art of writing is trying to represent the essentially active world in the essentially passive, linguistic world of rationality, which ultimately can do no more than construct concepts. But therein lies that good tension which constitutes all art and all creation.

Technique stands midway between what we want to say and saying it clearly. The art historian Kenneth Clark, in a collection of essays entitled Moments of Vision, refers to having “at the back of our minds the memory of certain devices by which sentences and paragraphs can be held together.” These devices constitute a basic knowledge of how a language works, and that body of knowledge is available to and understandable by anyone who wishes to write carefully. And we wish to do that when we care enough about the truth of the ideas we find ourselves impassioned to write about. Clark continues: “I do not suggest that we apply these devices consciously; but as the notion of what we want to say begins to form itself in our minds, it will, if we have any sense of style, be accompanied by a feeling that our thoughts must conform to certain sequences of shapes.”

Our conscious study of grammar and composition, in other words, should recede to an unconscious second nature which will intelligently inform the choices we make as we revise our composition. That way we work knowledgeably and reasonably, and what we write will show it.

***