Characters and actions

A letter about a planning application contained this sentence. It's grammatical, correctly spelled, and only twenty words long.

Determination of the application was undertaken following consideration of representations received during the consultation period by officers with delegated authority.

Read it twice and you can work out roughly what happened. That work you just did is the subject of this lesson, and the point is that you shouldn't have had to do it.

Now the same event, in eleven words:

Officers with delegated authority decided the application after reading the responses.

One thing did get dropped, and it's worth naming: the consultation period. Everything else is still there. The representations became "responses", the delegated authority survived intact, and the determination became "decided". If the period matters, put it back, and the sentence still runs to fourteen words: "after reading the responses sent during the consultation". What changed isn't that anything was cut. It's that somebody is now doing something.

The move that does most of the work

Joseph Williams built his book Style around two principles, as everyone who has read it reports, including the sources this course has read. Gopen and Swan reach the same place from the reader's side in their seven structural principles, where they appear as numbers three and five. Put together they say:

Make the main character of your sentence its grammatical subject, and put that character's action in the verb.

Before you take that as a rule, take their own warning with it. The same paper says "None of these reader-expectation principles should be considered 'rules.'" They're describing what readers expect, not issuing prohibitions, and a description of an expectation can be worth violating on purpose. Lesson 8 is about the difference, and this lesson's last section is about the cases where you should.

A teacher of writers and two describers of readers landing in the same place is better evidence than either alone. Be careful how much you make of it, though: Gopen and Swan wrote nine years after Williams, and this course hasn't read Williams's own text to see whether he's behind them. What you can say is that the principle turns up independently of any one author's taste, which is more than most writing advice can claim.

The mechanism is easy to state. A reader meets a sentence expecting a story, and a story needs somebody doing something. If your subject is a person or a thing that acts, and your verb is the acting, the reader gets the story for free. If not, they have to assemble it from the nouns, and they will, and it costs them.

That's what happened to you in the planning sentence. "Determination was undertaken" gave you a subject that isn't a character and a verb that isn't the action. The real action, deciding, was sitting inside the noun determination. The real character, the officers, was parked at the very end behind the word by. You reconstructed the sentence in your head, and every sentence like it spends the reader's attention on assembly instead of on what you're telling them. That's the cost, and lesson 1 is about how little of it you have.

Diagnose before you rewrite

Almost everyone starts rewriting straight away, shuffling words until it sounds better. That produces an improvement you can't repeat.

Do it in two steps instead, and write the first step down.

Step one, name the character and the action. Who or what is doing something, and what are they doing? Answer in ordinary words, not in the sentence's own vocabulary.

Step two, put them in the slots. Character becomes the subject. Action becomes the verb.

Here are the first three done for you.

Now do the next two yourself, in writing, before you open the answer.

Look at the third row and at the risk sentence. The character came out as "somebody unnamed", and that isn't a failure of the diagnosis. It is the diagnosis. Those sentences don't say who is implementing or who assessed, and you can't put a character in the subject slot until you decide who it is. That's the whole point of the harder example below.

Nominalisation, and why the -ion rule is only a detector

The noun that swallowed the verb has a name. A nominalisation is a verb or an adjective turned into a noun: decide becomes decision, fail becomes failure, implement becomes implementation, reduce becomes reduction.

GOV.UK's guidance gives working writers a cheap detector for these. It says to watch words ending in -ion and -ment, because they "tend to make sentences longer and more complicated".

Take that for exactly what it is. It's a place to look, not a verdict. Government, equipment, condition and region all end that way and none of them is a buried verb. And plenty of nominalisations end some other way: failure, analysis, growth, belief, refusal. The ending finds you candidates faster than reading can, which is why it's worth knowing. Then you check each one by asking whether there's a verb hiding in it and a person who should be doing it.

The pair that is not about the passive

Slow down for this one, because it corrects something almost everybody has been taught.

In 1918, William Strunk wrote in The Elements of Style that "a common fault is to use as the subject of a passive construction a noun which expresses the entire action, leaving to the verb no function beyond that of completing the sentence." His example pair:

A survey of this region was made in 1900.

This region was surveyed in 1900.

He prefers the second. Now look carefully at both.

That's a useful thing to carry, and lesson 8 comes back to it properly. A great deal of advice that arrives labelled "avoid the passive" is really about buried actions, and the two get confused because they often show up together. They're separate problems with separate fixes, and if you only know the passive rule you'll fix the wrong one.

Be fair to the other side while you're here. The same GOV.UK page this lesson took its detector from does tell writers to "use the active voice", and it gives a reason: the active "is more direct and puts the focus on the user". That's a defensible position and it isn't what this section is arguing against. The claim here is narrower, that the particular fault Strunk named is about buried actions rather than about the passive, and that treating the two as one thing makes you fix the wrong one. Whether "avoid the passive" survives contact with the evidence as a general rule is lesson 8's question.

The harder example: when the fix forces a decision

Reports about things that went wrong are full of sentences like this.

Following identification of the discrepancy, escalation was delayed pending clarification of the reporting requirements.

Three nominalisations, no character anywhere. Diagnose it and every row of the table comes out "somebody unnamed".

Try to rewrite it and you hit a wall at once, because you can't write "X delayed escalating" without knowing who X is. The sentence isn't merely badly arranged. It's arranged so that the question doesn't come up.

Sometimes that's carelessness and sometimes it isn't. A sentence with no character can't be checked against anyone, and the version that names one can. Compare:

The duty manager knew about the discrepancy on Tuesday and did not escalate it until Friday, because the reporting rules were unclear.

Same events. Now there's a person, two dates, and a reason. Every one of those is checkable, and one of them may be wrong, which is exactly what the original protected.

When to leave it alone

Undoing every nominalisation is a worse rule than the one it replaces, and it's the commonest overcorrection once this idea lands.

When the abstraction is genuinely the subject. "Inflation fell to three percent" is a sentence about inflation. Rewriting it as "prices rose more slowly" changes what the sentence is about. If your topic really is the decision, the failure or the reduction, then the noun belongs in the subject slot, and lesson 5 will show you the deeper reason why.

When the noun is a fixed term. Planning permission, risk assessment, performance review are names of things. Unpacking them into verbs produces a sentence that's technically clearer and unrecognisable to the people who have to act on it.

When it links two sentences. If you've just written about the committee deciding, then opening the next sentence with "That decision..." puts old information first, which is doing a job. Lesson 5 is entirely about this, and it's the commonest reason a nominalisation earns its place.

When naming the character would be false precision. If it genuinely isn't known who assessed the risks, "somebody assessed the risks" invents a certainty the writer doesn't have. Write that it isn't recorded.

What people get wrong

"Nominalisations are bad." They're a normal and necessary part of English. The problem isn't the noun; it's a sentence whose action has left the verb, leaving was made or took place behind. Diagnose before you judge.

"This is a rule about words ending in -ion." That's a detector with a good hit rate and plenty of false positives, both ways. The test is whether there's an action buried and a character displaced.

"Passive voice is the real problem here." Strunk's own pair disproves it, since both his sentences are passive and he preferred one of them. Buried actions and the passive are different faults that travel together.

"Rewriting is the first step." Rewriting is the second step. Naming the character and the action in ordinary words is the first, and skipping it is why people's rewrites are inconsistent.

"Being concrete means using shorter words." Concreteness is about whether a reader can picture who is doing what. "Utilise" for "use" is a vocabulary problem and a small one. "Determination was undertaken" is a structural problem and a large one, and every word in it is short.

Practice

Where this goes next

You now have the move that does most of the work at the sentence level. The next lesson is its companion and takes about half the time. Once you've got the character into the subject slot, the reader wants the verb almost immediately, and a great many otherwise well-built sentences fail on that one point.

Then lesson 5 goes underneath both of them, and answers the question this lesson kept deferring. Why does an abstraction sometimes belong in the subject after all, and why does a paragraph of good sentences sometimes read as a list of unrelated facts?

Go deeper

• Gopen and Swan in full. Principles three and five are this lesson; the other five are the rest of the course, and the closing section on why none of them are rules is worth more than most books on writing.
• Strunk's The Elements of Style at Project Gutenberg, public domain. Rule 10 is two pages and contains all three pairs, including the one in this lesson's quiz. Lesson 8 comes back to what the rest of the book is worth.
• Joseph M. Williams, Style: Lessons in Clarity and Grace, Part Two, Clarity. The book this principle is usually credited to, and the one this course has not read. If you read one thing beyond this course on sentences, read that part.
• GOV.UK on clear language, where the detector comes from, alongside a dozen other rules written for people who have to get it right at scale.

Sources

• Joseph M. Williams, Style: Lessons in Clarity and Grace, Part Two, Clarity, on making central characters the subjects of verbs and using verbs to name their actions. This course has read the book's structure and its two central principles as consistently reported by others, not the text itself, which is why no wording is quoted from it and no lesson number is cited.
• George D. Gopen and Judith A. Swan, "The Science of Scientific Writing", American Scientist 78 (1990). Principle three, place the person or thing whose story a sentence is telling at the beginning, and principle five, articulate the action of every clause or sentence in its verb.
• William Strunk Jr., The Elements of Style (1918), rule 10, for the survey and mobilization pairs and for the sentence naming the fault, "a common fault is to use as the subject of a passive construction a noun which expresses the entire action".
• GOV.UK, "Use clear language", for the practical detector, that words ending in -ion and -ment "tend to make sentences longer and more complicated". The same page tells writers to "use the active voice" because it "is more direct and puts the focus on the user", which this lesson quotes rather than omits, since the passive section argues a narrower point than that guidance does.