Book Review Volume 63:1

Book Review — Recension critique

* Professor and Dean of Law, University of New Brunswick (formerly Associate Professor, University of Saskatchewan).

Table of Contents

Review of Edward Berry, Writing Reasons: A Handbook for Judges, 4th ed (Toronto: LexisNexis, 2015), pp 158. ISBN: 978-0-433-47964-2.

Edward Berry’s Writing Reasons,[1] though aimed at judges who provide written reasons for their decisions, deserves a wider audience. With this version, it may get just that. Originally published in 1998 and self-published for the first three editions, this delightful and highly instructive handbook has now been published by LexisNexis. While I lament the loss of some of the third edition’s form—its cover, its elegant typesetting, its clever Shakespearean epitaphs leading off each chapter (it still has epitaphs, but now mostly from other sources)—I laud the substantive changes and the decision to leave the book’s basic format intact. That format is one in which the author—an emeritus professor of English and long-time leader of judicial writing workshops—goes from macrocosm to microcosm, continually imparting wisdom along the way and asking readers to test how well they’ve imbibed it through end-of-chapter exercises and answer keys.

Berry’s first macrocosmic point is context. It is the driving theme of the first three chapters—“Context First”, “Introductions”, and “Organization”—and of much of the rest of the book. The notion that information needs context isn’t hard to understand, says Berry, yet it is often forgotten for two general reasons. The first is that, in working through a problem, we tend to write for ourselves rather than our readers. The second reason, closely related to the first, is that we assume reader expertise or knowledge that doesn’t actually exist. Journalists are aware of both tendencies and work hard to overcome them. Two further reasons for forgetting context apply to judicial writing, says Berry: legal training and the traditions of legal communication. The facts-law-application-conclusion sequence that law schools attempt to drill into students may lend itself well to legal precision; as a presentation method, though, Berry asserts that it often fails to respond to readers’ needs. This is even truer when considering that the audience for judgments is not only lawyers, but parties to a case (at least one of whom is often now “self-represented”[2]) and, indeed, the “public at large”.[3] Such readers need more context so that they can better understand the legal concepts and terminology and anticipate how and why judgments unfold in the way that they do.

Example being better than precept, I’ll provide one. Consider the following introduction to a judgment:

This is an appeal from a judgment of the Court of Appeal for Ontario, 2011 ONCA 482, 107 OR (3d) 9, affirming a decision of the Ontario Superior Court of Justice per Himel J, 2011 ONSC 1500, 105 OR (3d) 761, granting the respondent’s application for an order declaring that life support could not be removed from her husband without her consent, and that any challenge to her refusal to consent must be brought before Ontario’s Consent and Capacity Board pursuant to the Health Care Consent Act, 1996, SO 1996, c 2, Sch A (“HCCA”). For the following reasons, we affirm the decisions below and dismiss the appeal.

In fact, this is a hypothetical introduction to an actual Supreme Court of Canada decision. I provide the real introduction below, but let me first defend my hypothetical one, then critique it à la Berry.

In some ways, this introduction is not only defensible, but representative of many introductions to appellate judgments. The style and structure is especially common in the U.S. federal courts, but also in other jurisdictions. In a few sentences (just two here), the court explains the procedural history of the case and says how it will dispose of it. The key issues, though not labelled as such in this example, can be inferred: they have to do with whether a patient’s spouse must consent before the patient’s life support is removed and with whether a refusal to consent must be challenged before a special tribunal or board acting under a provincial statute.

However, this compaction is achieved at the cost of clarity, especially for the lay reader. Legal readers—lawyers, judges, law clerks, and law students slogging through the task of briefing cases—might appreciate having the procedural context handed to them at the outset in highly crafted form. But for parties and public-at-large readers, the important thing is substantive context, largely missing here. The questions “Why should I read this?” and “What does this mean?” go unanswered. And this isn’t the only problem. The first sentence, at ninety-five words, is longer than the average sentence by a factor of almost five, making it nightmarish to read.[4] It is clogged with citations—fewer, actually, than many legal sentences—and uses jargon like “respondent” and “application”. The passive voice has also crept in—we learn that life support “could not be removed”—and we are left to wonder: removed by whom? These things further detract from the sentence’s readability and even its accountability. The second sentence raises a more difficult question: should an introduction announce the decision as well as the issues? Berry says that if readers need a road map, “why not announce the final destination in advance?”[5] But he then considers whether, and in which cases, this is a good or a bad idea. For example, he notes that doing so might induce a reader to stop reading or treat the reasons that follow as “mere rationalizations, afterthoughts produced to justify a verdict arrived at by mere prejudice.”[6] Ultimately, he eschews a single or formulaic answer to this question on the basis that judgment writing “involves not merely legal logic but psycho-logic.”[7]

Here, then, is the actual introductory paragraph to the case, Cuthbertson v. Rasouli:

This case presents us with a tragic yet increasingly common conflict. A patient is unconscious. He is on life support—support that may keep him alive for a very long time, given the resources of modern medicine. His physicians, who see no prospect of recovery and only a long progression of complications as his body deteriorates, wish to withdraw life support. His wife, believing that he would wish to be kept alive, opposes withdrawal of life support. How should the impasse be resolved?[8]

This paragraph, written by Chief Justice Beverley McLachlin, vividly exemplifies what Berry calls the focus first strategy for creating a context for details and reasons.[9] It is actually the first paragraph of a four-paragraph overview, so it can be thought of as an introduction to an introduction. To set the stage for a reader’s willingness to accept the reasons that follow (yes, Berry is unapologetic about the persuasive function of reasons for judgment), the Chief Justice cultivates a sympathetic, collaborative, or problem-solving tone, in contrast to my hypothetical version, which might be described as aloof, authoritative and legalistic. Notably, she does not announce the decision in the first paragraph, though she does so by the third one. The rest of the overview deals with the same things that my version addresses—the statute, the board, etc.—and while her treatment of them is more expansive, it provides just enough information to give the necessary context for understanding the case. And for wanting to read it.

After the first three chapters, Writing Reasons devotes Chapter 4 to “Conciseness”, which Berry distinguishes from brevity. Berry writes that “[b]revity is only about saving words; conciseness is about making every word count.”[10] Since clarity is the overarching goal, extra words can sometimes help, as with contextual sentences, transitional words or phrases, or occasional summaries. But wordiness is a problem in judgments, and it takes effort to achieve conciseness there as in any form of writing. “I would not have made this letter so long,” wrote Blaise Pascal, “but for not having had the time to make it shorter.”[11] Berry identifies three areas where judgments frequently need concision: evidence, party positions, and quotations. He has some suggestions here. Evidence, for example, rarely merits its own section, and if you do create one, you are likely to be tempted to fill it up, recounting testimony in a plodding, witness-by-witness fashion. Instead, says Berry, use issue-driven structures, which tend to discipline the presentation of evidence and align the necessary components of it with the issues to be decided. Creating sections that outline party positions can also be tempting; after all, this is how cases are presented to adjudicators. But they also encourage wordiness (and, I would add, conclusory reasoning or even insufficiency of reasons). Berry suggests that one remedy is to focus on the losing party: the reasons may often align with those of the winning party, in which case repeating that party’s position becomes unnecessary. Before using quotations, Berry suggests that we ask two questions: first, are the exact words essential; and second, would a paraphrase be less effective than the original. Berry has in mind quotations to authorities like cases and statutes, but his advice applies as well to other material, such as quoting from affidavit evidence and counsel submissions. Some quoted text, even in block form, can add vitality to a judgment, but large amounts can deaden it and even encourage ‘cut-and-paste’ writing and litigation about the sufficiency of reasons.[12] In any case, says Berry, block quotations should generally be introduced with a statement that either focuses the quotation or summarizes its point in the judge’s own words.

Chapters 5 to 8 of Writing Reasons—“Paragraphs”, “Sentences”, “Words”, and “Punctuation”—continue the macrocosmic-to-microcosmic approach, addressing such things as transitions and topic sentences; structure and length; balance; active versus passive voice; and the idiosyncrasies of commas, semicolons, colons, parentheses, and dashes. Berry manages to do this without seeming dogmatic and by using examples, recognizing that there are various ways of doing things. He might even approve of this paragraph. Though the first sentence is very long, it uses punctuation to create syntactical breaks; “em dashes”[13] to mark an interjection; semicolons to signal a list; and commas within one of the list items. It also attempts to follow the principle of “parallel structure” (the general format of my review is a chapter-by-chapter outline of the book).

Chapter 9, “Widening the Audience”, is one of my favourites. The section on prejudicial language, some of which can be quite unconscious, is especially good. Berry avoids a “checklist” of words that are likely to offend. This is in part because the common recommendations on such lists have already been absorbed into the language; more fundamentally, though, checklists invite us to substitute one word for another instead of searching for underlying principles that can be applied to sundry situations and withstand the test of time. Berry suggests the following principles or strategies: (i) invite parties to define their own identities (e.g., despite changing norms, “some women might prefer Mrs. to Ms.”[14]); (ii) mention a distinguishing characteristic only when it is pertinent, and show its pertinence immediately; (iii) avoid merging the person with the characteristic in a dehumanizing way (compare, e.g., “AIDS victim” to “a person with AIDS”[15]); (iv) if a racial or ethnic categorization is pertinent, prefer the specific to general (e.g., “a ‘Haida’ would probably prefer that designation to ‘Aboriginal person’”[16]) (and, I would counsel, avoid the absurdly vague “racialized person”[17]); (v) be wary of stereotypes, both explicit and implicit (“the appointment of Justice Janet Marshall will bring a refreshingly sympathetic face and nurturing demeanour to the bench”[18]); and (vi) “stay current and Canadian.”[19] If there is one way in which this chapter could be improved, I think it would be in providing techniques for gender-neutral writing that are neither awkward (‘s/he’) nor grammatically contested (‘a judge should choose their words carefully’).[20] In the meantime, one of the best resources for the legal writer on that subject is the British Columbia Law Institute’s short report on managing personal pronouns.[21]

In Chapter 10, “Developing a Personal Style,” Berry first addresses the antipathy that some legal readers might feel towards the word “style”. Substance is what matters, the argument goes, and brooding over style can interfere with substance—drawing attention to the judge instead of the parties and the issues to be decided. This argument, not entirely without merit, has recently led one commentator to frame a Supreme Court judge’s stylistic efforts as a form of “judicial arrogance”.[22] But as Berry notes, citing Benjamin Cardozo, form and substance are inseparable: “[t]he strength that is born of form and the feebleness that is born of the lack of form are in truth qualities of the substance.”[23] The advice to attend to style, then, becomes advice on how the manner of an argument may best support its matter. Berry provides four examples from different levels of Canadian courts to encourage this kind of reflective approach to style.

The last chapter, “Revising”, is in some ways the most important one. Berry gives some “simple advice” based on “sad experience”.[24] First, “[w]hen drafting, write to think; when editing, write to be read.”[25] Second, “[w]hen drafting, resist editing, especially at the micro-level.”[26] Simple to state, perhaps, but hard to put into practice. Berry provides a seven-step program to implement these two tips, starting with macro questions like how much information readers need to make sense of the reasons; how issues should be framed and analyzed; and micro considerations such as paragraph structures, sentence variations, word choice and tone.

I said at the outset that Writing Reasons deserves a wider audience. Administrative tribunal members are obvious candidates for that audience: from the point of view of sheer numbers alone, their decisions affect the day-to-day lives of many more people than decisions of courts, and Berry’s advice transposes well to the writing of administrative decisions. Berry also makes the case that judges who give mostly oral judgments can benefit from some of the ideas in the book, and provides advice specifically for them.[27] Lawyers who diligently craft written submissions would also do well to imagine themselves in the judge’s role and revise with that role in mind; Writing Reasons would be a good place to start. Finally, law schools, which traditionally have focused on other forms of legal writing—research memos, opinion letters, factums, pleadings, and essays—might want to consider the benefits of teaching judgment writing. Using the third edition of Writing Reasons, I did that for the first two years in an upper-year seminar course on the written judgment—and I have recently continued the practice with the fourth edition of Berry’s excellent book.


*     Professor and Dean of Law, University of New Brunswick (formerly Associate Professor, University of Saskatchewan).

Ó John C. Kleefeld 2017

Citation: (2017) 63:1 McGill LJ 191 — Référence : (2017) 63:1 RD McGill 191

[1]     Edward Berry, Writing Reasons: A Handbook for Judges, 4th ed (Toronto: LexisNexis, 2015).

[2]     See generally National Self-Represented Litigants Project, online: <>, archived at, and particularly, Julie Macfarlane, “The National Self-Represented Litigants Project: Identifying and Meeting the Needs of Self-Represented Litigants: Final Report” (May 2013) at 33–34 online: National Self-Represented Litigants Project <
/uploads/2015/07/nsrlp-srl-research-study-final-report.pdf>, archived at
-BTXK (noting, among other things, that self-representation in provincial family courts is now consistently at or above 40 per cent, and in some cases far higher).

[3]     Berry, supra note 1 at 3, citing R v Sheppard, 2002 SCC 26 at para 55, [2002] 1 SCR 869.

[4]     See LA Sherman, Analytics of Literature: A Manual for the Objective Study of English Prose and Poetry (Boston: Ginn & Co, 1893) at 256–62, online: Internet Archive <>; William H DuBay, “The Principles of Readability” (2004), online: <
ility02.pdf>, archived at Both works stand for the proposition that, other things being equal, readers understand shorter sentences better than longer ones. Sherman, a professor of English literature, was the first to statistically document a progressive shortening of sentences over time by counting average sentence length over a range of literature per 100-year periods. His study of pre-Elizabethan sentences based on representative authors yielded an average length of 63 words (Robert Fabyan); by Elizabethan times, this had shortened to 45 words (Edmund Spenser, Richard Hooker); by Victorian times, to 28 words (Lord MacAulay, Thomas De Quincy); and in Sherman’s own time, 23 words (Ralph Waldo Emerson, William Ellery Channing). DuBay estimates that sentences now average about 20 words (ibid at 10). For comparable studies finding a drop in average sentence length over time, see e.g. Mark Liberman, “Real Trends in Word and Sentence Length” (31 October 2011), Language Log (blog), online: <>, archived at (analyzing US presidential inaugural and state of the union addresses); Alan G Gross, Joseph E Harmon & Michael Reidy, Communicating Science: The Scientific Article from the 17th Century to the Present (New York: Oxford University Press, 2002) at 171 (finding that average sentence length in scientific prose has declined from about 60 words to 27 words since the 17th century).

[5]     Berry, supra note 1 at 22.

[6]     Ibid at 23.

[7]     Ibid.

[8]     Cuthbertson v Rasouli, 2013 SCC 53 at para 1, [2013] 3 SCR 341.

[9]     See Berry, supra note 1 at 3–8 (describing focus first writing as involving “the addition of a context at the outset that will clarify the significance of the narrative details” at 4).

[10]    Ibid at 58.

[11]    The quip is often attributed to Mark Twain, but he apparently borrowed it from Pascal. See Garson O’Toole, “If I Had More Time, I Would Have Written a Shorter Letter” (28 April 2012), Quote Investigator (blog), online: <
ter-letter>, archived at (noting that the quip appears to be a translation of Pascal); Blaise Pascal, Les Provinciales, annotated edition by Charles Louandre (Paris: Charpentier, 1875) at 330, online: Internet Archive <> (in a 1657 lettter, Pascal wrote: “[j]e n’ai fait celle-ci plus longue que parce que je n’ai pas eu le loisir de la faire plus courte”). See also Twain Quotes, “Letters”, online: <
Letters.html>, archived at (noting Pascal’s authorship of the quote and its frequent misattribution to Twain). A similar quote appears in Writing Reasons, supra note 1 at 57, where Berry attributes it to a “Confucian scholar”. If that is correct, the quip has demonstrated a remarkable resilience, staying intact after translation from Chinese to French to English over a couple of millennia.

[12]    I cannot resist a disquisition here. Whereas Cicero was concerned with the sound effects of his defence speeches—the “well-knit rhythm of prose,” he called it (See Cicero, Brutus: Orator, translated by HM Hubbell, revised ed (Cambridge, Mass: Harvard University Press, 1962) at 447, online: <>)—today’s lawyer is more apt to think of how her written submissions look on a page—or on the judge’s iPad. If they look very good, the judge might use them. And if the lawyer has provided them electronically, they may end up being incorporated into the judge’s reasons, perhaps without attribution. In this setting, the lawyer is usually delighted, assuming the judge uses the submissions to find in favour of the lawyer’s client. But what about those on the other side—might they not be left with the impression that the judge has failed to fairly consider the evidence and the arguments? In recent years, the Australian, English and Canadian courts have all had something to say about this, though reaching somewhat different conclusions.

The Australian Full Court (the appellate division of the Federal Court of Australia), in LVR (WA) Pty Ltd v Administrative Appeals Tribunal, [2012] FCAFC 90, 128 ALD 489 provided an extensive analysis (146 paragraphs) of the issues associated with copying submissions. The Court held that, at least in the circumstances of that case, the failure to attribute the copying to the submissions of one side was fatal to the case. Approximately 95 per cent of the Tribunal’s reasons had been taken from the written submissions of the Commissioner of Taxation. After the decision, the Commissioner then reviewed five other appeals that were headed to the Federal Court and agreed to have all of them go back for a rehearing. See the consent orders in the joined appeals, styled as Palassis v Commissioner of Taxation (No 2), [2012] FCA 955, 136 ALD 91.

The English Court of Appeal considered this issue in Crinion v IG Markets Ltd, [2013] EWCA Civ 587, [2013] All ER (D) 272 (May). The case involved a father and son, Tommy and Declan Crinion, who had run up large debts in accounts with a futures trader, IG Markets (IG). IG successfully sued them for recovery of the debts, and most of the judgment—94 per cent, by the Crinions’ estimate—was copied and pasted from the submissions of IG’s counsel. The Crinions appealed solely on the perception that the judge had abdicated responsibility by slavishly adopting the submissions as his own conclusions. One of the appeal judges, Sir Stephen Sedley, noted that technology has made it “seductively easy” to do what the judge did but “embarrassingly easy” to show what he had done. However, he found that enough independent words could be “teased out” of the judgment—some heading changes, a few additional sentences—to satisfy the legitimate demand for independent reasoning. The other judges agreed, though their reasons carry a strong aura of disapproval and a wish that “a judgment like the one now before us will not be encountered again” (ibid at para 40).

The leading Canadian case, Cojocaru v British Columbia Women’s Hospital and Health Center, 2013 SCC 30, [2013] 2 SCR 357, was an action for a negligent birth. The mother’s uterus had ruptured, resulting in brain damage to her child and a $4 million damages award after a 30-day trial. Of the 368 paragraphs in the trial judgment, 321 were almost verbatim from the plaintiff’s closing submissions. The case went to the BC Court of Appeal, where two of three judges reversed the trial judge on the basis of an apparent lack of independent analysis. On further appeal, the Supreme Court of Canada unanimously restored most of the trial judgment, with some variation to its substance. Writing for the Court, McLachlin CJC noted a long practice of judicial copying, with or without attribution, and said that neither the copying nor the failure to attribute sources, without more, answer the ultimate question—whether a reasonable person would conclude that “the judge did not put her mind to the issues to be decided” (ibid at para 31). The Court concluded that while it would have been “best practice” for the trial judge to have used his own words, he had actually rejected some of the plaintiff’s submissions and, in the 47 paragraphs that were his own, had added enough independent material to satisfy the “reasonable observer” test (ibid at paras 69, 74).

[13]    Not to be confused with the slightly narrower en dash (–) or even narrower hyphen (-). See The Punctuation Guide, “Em dash”, online: <>, archived at; Wikipedia, “Dash”, online: <>, archived at

[14]    Berry, supra note 1 at 110.

[15]    Ibid at 111.

[16]    Ibid.

[17]    See Jonathan Kay, “Stop calling people ‘racialized minorities.’ It’s silly and cynical”, National Post (11 July 2014), online: <>, archived at But see, contra, Ontario Human Rights Commission, “Racial Discrimination, Race and Racism (Fact Sheet)”, online: <>, archived at (“[r]eco-
gnizing that race is a social construct, the Commission describes people as ‘racialized person’ or ‘racialized group’ instead of the more outdated and inaccurate terms ‘racial minority’, visible minority’, ‘person of colour’ or ‘non-White’”); and Louise Brown, “U of T gets personal with staff to track race, gender data”, Toronto Star (9 July 2016), online: <
-with-staff-to-track-race-gender-data.html>, archived at (describing a University of Toronto survey that asked staff to ask if they identify themselves as “persons of colour” or “racialized persons” and then drilling into deeper levels of specificity).

[18]    Berry, supra note 1 at 111.

[19]    Ibid.

[20]    On this point see Jeremy Butterfield, ed, Fowler’s Dictionary of Modern English Usage, 4th ed (Oxford: Oxford University Press, 2015) sub verbis “their”, “they, their, them”, “he or she”.

[21]    See British Columbia Law Institute, Gender-Free Legal Writing: Managing the Personal Pronouns (Vancouver: BCLI, 1998), online: <>, archived at

[22]    See Alice Woolley, “The Problem of Judicial Arrogance”, Slaw (20 October 2016), online: <>, archived at (referring to Justice Russell Brown’s opening paragraph in Canada (Attorney General) v Igloo Vikski Inc, 2016 SCC 38, [2016] 2 SCR 80). I disagree with Woolley, though I also disagree with Justice Brown’s opening sentence: “In wintertime ice hockey is the delight of everyone.”

[23]    Berry, supra note 1 at 118, citing Benjamin N Cardozo, Law and Literature and Other Essays and Addresses (New York: Harcourt, Brace and Company, 1931) at 6 [emphasis added by Berry].

[24]    Berry, supra note 1 at 127.

[25]    Ibid.

[26]    Ibid.

[27]    Ibid at 113–14.