On Writing Well

Zinsser's opening chapters identify the enemy as clutter: the unnecessary word, the circular construction, the pompous frill, the adverb carrying meaning the verb already contains.

His claim is that clutter is not a stylistic flaw but a failure of thinking, because a writer who has not decided what they mean produces qualifications and padding in place of the decision. The remedy is therefore not to write more elegantly but to know more precisely what is being said, which is why the book is more about thought than about sentences.

Clutter is not a style problem. It is the residue of a decision the writer has not yet made.

Devmark’s reading — not a quotation from the book

His practical instruction: examine every sentence for what can be removed without loss, and remove it. He demonstrates on his own manuscript pages, reproduced with his cuts visible.

Those facsimile pages are the most effective teaching device in the book, because they show a professional writer removing roughly a third of his own draft and improving it. The demonstration is far more persuasive than the instruction, and it establishes that heavy cutting is normal practice rather than evidence of a weak first attempt.

His two positions that appear to conflict and do not. Write for yourself, because you cannot know the reader's taste and trying to guess it produces hedged prose. But respect the reader's intelligence, time and attention, because they will leave the moment you stop earning them.

His reconciliation is that you please yourself on matters of taste and serve the reader on matters of craft — clarity, structure, pace — which is a clean distinction and one most writing advice muddles.

His chapters on vocabulary: prefer the concrete to the abstract, the short word to the long, the active to the passive, and the specific noun to the qualified general one.

His warning against style as something applied afterwards — a layer of personality added to clear prose — is the most important of these. His position is that style is what remains when the clutter has gone, and that attempting to add it produces the opposite of what the writer wants.

His structural chapters. Unity of tense, of person, of attitude, decided before writing. A lead that compels the next sentence and the next. An ending that arrives rather than trails off, ideally sooner than the reader expects.

His advice that the perfect ending frequently occurs one paragraph before the writer stopped is specific, testable against any draft, and correct often enough to be worth applying as a routine check.

The middle section covers particular kinds of nonfiction: interviews, travel writing, memoir, science and technology, business writing, sports, criticism, humour.

The science and business chapters are the most useful for readers outside journalism. His principle for explaining technical material — lead the reader by one step at a time, and never assume the step just taken has been absorbed — is a better account of technical writing than most books devoted to it.

His chapters on writing about one's own life, added and expanded in later editions, and clearly the subject that mattered most to him in his final years.

His guidance is to narrow rather than broaden — write about one strand of a life rather than its whole — and to be a reliable narrator about oneself, which means including the material that does not flatter.

His closing arguments concern the disposition the work requires: confidence, enjoyment, the willingness to be yourself on the page, and the discipline to keep working when the sentences will not come.

His insistence that writing is a craft learned by doing rather than a gift is consistent throughout and is the reason the book functions as encouragement as well as instruction.

Its examples are drawn largely from mid-century American magazine journalism and its cultural references assume a reader of that period. It has nothing to say about writing for screens, for search, for collaborative documents, or for readers who will not read past a paragraph.

Its conception of the writer is also individual and literary, and it does not address the documents most people actually write, which are produced by several people for a specific institutional purpose.

The stripping method, applied as a separate pass: write to find out what you mean, then cut to make it readable, and never attempt both at once.

And the ending check, which costs one minute and improves a surprising proportion of drafts by removing the paragraph in which the writer explains what has already been shown.