Five ways to improve your judgment writing[1]

Professor J Raymond[2]

The following article, by internationally renowned legal writing consultant James Raymond, sets out five clear principles to hone judgments.

Identify the issues before you start writing

Issues don’t blossom out of facts: lawyers and judges have to fashion them. The best time to do that is before the trial even begins. Meet with solicitors and counsel if possible, and have them agree on the questions that need to be determined in trial. Obviously, these questions change as the trial progresses, and you will need to change your plan accordingly. It is far better to start with a tentative plan at the beginning of the trial, rather than wait to the end, when the true issues may be obscured by lots of subsidiary issues that may turn out to be irrelevant.

Arrange the issues in a sequence that makes sense

Sometimes the issues are so independent of one another that you can arrange them in almost any sequence whatsoever. In other cases, however, there is a threshold issue, or a dispositive issue, and it must be dealt with first. In all cases with more than one issue, it is important to foreshadow them, preferably in a bullet-pointed list before the end of the first page, or as close to that point as possible. Once you have made this list, use headings that track these issues. This will enable your readers to find a way to the part of the judgment that might concern them as precedent, or in constructing an appeal.

Analyse the issues by using an appropriate pattern of analysis

In questions of law, it is usually preferable to express the losing party’s position first, and then to explain the flaw that you have determined in the losing party’s position. For questions of fact, it is often preferable to begin with the evidence of the party who has the burden of proof, followed by the evidence from the opposing party. Once you have laid out this evidence, indicate which side you prefer, and why. Try to give objective factors for making this determination, such as evidence from cheque, credit card, or telephone records, or evidence that undermines one party’s credibility, such as prior inconsistent statements or evasiveness in answering questions.

Write a beginning that provides the context for understanding the issues

A good beginning indicates who did what to whom, or who was arguing about what, before anyone set foot in court. Give an overview, not a detailed narration of the facts. Avoid cluttering this overview with parenthetical aliases (for example, “hereinafter called”), or with citations that serve no purpose at this point. The beginning may suggest where the judgment is heading, or it may conceal the result entirely — this is a choice that the judge must make with each individual case. Don’t waste your first paragraph on uncontested matters or procedural history that is no longer relevant.

Write a conclusion that recapitulates your analysis

The last part of a judgment is a good place to recapitulate your reasons for the benefit of those readers likely to skip the body of the judgment. It is also a good place to bolster your findings with arguments from consequence — that is, by mentioning all the bad effects that would flow from a contrary judgment. If the result is unlikely to be a popular one, the last paragraph is a good place to assure readers that remanding a case does not result in freeing an unsavoury character, but simply in a new trial that follows long-established rules of due process.

From beginning to end, a judgment is best written in language that ordinary people can understand. Each judgment contributes to the credibility and authority of the judiciary as a whole. No judgment is unimportant.



[1] Originally published in the Commonwealth Judicial Education Institute Report, June 2006, Halifax, Nova Scotia, and reprinted with permission. Published in (2007) 19(5) JOB, updated 2021.

[2] Professor Emeritus, the University of Alabama, former editor of College English; President, The International Institute for Legal Writing + Reasoning.