Writing that persuades

A written argument has no way of responding to objections as they arise, which changes what it has to do.

Knowing the reader's position

The first requirement.

What do they currently believe, and why?

What would they lose by agreeing with you?

What is the strongest reason for their position?

What evidence would they accept, and from whom?

And what are they actually deciding, which is frequently narrower than the general question.

Writing without answering these produces a piece that persuades people who already agree.

The two-sided structure

Which is better supported than the one-sided version.

Research on persuasion consistently finds that acknowledging and addressing the opposing case is more effective than ignoring it, particularly with informed or initially opposed audiences.

Which means stating the strongest counter-argument, accurately, and then answering it.

Not the weakest version, which the reader recognises as a straw man and which costs you credibility.

And conceding what is genuinely valid, which is the move that makes the rest believable.

Where to put the concession

The structural question.

Early, if the reader is likely to be thinking of the objection from the start, since an unaddressed objection blocks everything after it.

After your positive case, if the objection is one they would not have raised themselves.

Never at the very end, which leaves the counter-argument as the last thing read.

And with a clear response attached, since a concession left hanging is a concession.

Evidence

What actually moves people.

Specific over general: a documented case beats an assertion about the pattern.

Numbers, presented with context and comparison rather than alone.

Sources the reader already trusts rather than ones you trust.

Concrete consequences rather than abstract principles.

And a limited number of strong points rather than a long list, since a weak argument included alongside strong ones dilutes them.

The dilution effect

Worth its own note.

Adding weaker arguments to a case tends to reduce its overall persuasiveness rather than increasing it, which is counter-intuitive and reasonably well established.

Which means cutting your third and fourth reasons if they are noticeably weaker than the first two.

The reader evaluates the case as a whole, and the weakest element is what gets attacked.

Tone

Which determines whether the argument is heard.

Contempt for the opposing view guarantees that anyone holding it stops reading.

Which means writing as though the reader is intelligent and has reasons, because they generally do.

Overclaiming invites scrutiny that a measured claim avoids.

Stating your own uncertainty where it exists increases credibility rather than reducing it.

And avoiding the rhetorical question, which reads as a debating trick on the page.

What you are asking for

The practical end.

State it explicitly and make it as small as it can be while still mattering.

A reader will move a short distance more readily than a long one, which is why asking for a trial rather than a commitment frequently works.

Name what happens next, concretely.

And distinguish between persuading someone to believe something and persuading them to do something, which require different amounts of movement.

Openings and closings

Which carry disproportionate weight.

Open with the point of contact: the problem as the reader experiences it, or a specific that they will recognise.

Not with a definition or a history.

Close with the implication and the request, not with a summary of the argument.

And the last sentence should be the one you most want remembered, which means writing it deliberately rather than arriving at it.

What does not work

The reliable failures.

Volume of argument in place of quality.

Emotional appeal to a sceptical reader.

Attacking the people who hold the other position.

Assuming the reader's motives.

Repeating the claim more forcefully.

And writing to satisfy people who already agree, which is the most common thing published under the heading of argument.

Length and the reader's patience

A practical constraint.

A persuasive piece competes with everything else the reader could be doing, which means the argument must be visible early or it will not be read at all.

Which is an argument for stating your position in the first paragraph, even though it removes the pleasure of building to it.

The reader who agrees stops there, satisfied; the reader who disagrees reads on to find out why you think that, which is exactly the reader you wanted.

Writing to someone who has already decided

The hardest case.

Public commitment makes positions considerably harder to abandon, since changing costs face as well as belief.

Which means giving the reader a way to move without conceding: new information that was not available before, a distinction that narrows the disagreement, or a change in circumstances.

"Given what we now know" is a more effective construction than "as I argued previously", and it costs nothing.