Guard against poor legal research with these 3 writing practices
Bryan A. Garner. (Photo by Winn Fuqua Photography)
Not long ago, the partners at a law firm told me of an embarrassingly disastrous trial. Two years before the trial, an associate was asked to research a critical point about liability. Responding by email, the associate said that the answer was clear: The law provided that such-and-such. The two lead trial lawyers, together with their considerable team of associates, framed the discovery and their trial theories accordingly. Only at the end of the trial did they learn that the associate’s research was fundamentally flawed. Two appellate decisions reaffirmed their belated lesson. The associate had overlooked an exception to the “simple and unambiguous answer”—an exception that clearly applied to the facts of their case.
Somehow, everybody on the trial team had unquestioningly accepted the associate’s research on a pivotal issue. As a result, the firm faced a gargantuan malpractice lawsuit. Scary, isn’t it?
They wanted to know whether I knew of ways to prevent this type of problem. The answer is a qualified yes: While there are no panaceas, certain protocols can minimize the risks of suboptimal research. Here are the three crucial points.
Resist the we-don’t-write-legal-memos-anymore mentality.
Since the rise of email in the 1990s, many senior lawyers have tended to say, “We don’t waste the client’s time and money with formal legal memos. We go straight into litigation documents—motions and briefs.” It’s as if they’re saying that reporting your research is a waste of time.
“Why do I need it reported?” they might respond. “Just email an answer. Or forward the relevant authority to me.”
My response to this mindset is that the “report”—a legal memo—isn’t just make-work. It memorializes your understanding about a given legal problem. It becomes the basis of a good legal argument in litigation documents and oral presentations. And if it’s done well, it becomes part of a repository of legal knowledge within a practice group.
But with emails, you’ll too often end up with chains that read like this:
Partner
I need to know whether a “sale” to Jensen actually took place on the evening of July 18.
Associate
Yes. See Conn. Gen. Stat. Ann. § 30-102 (attached).
Concededly, this cryptic exchange might do the job of supplying the senior colleague with the required information. The answer might be correct. But there’s no way of assessing its correctness from the words themselves. That’s the problem.
Imagine, though, that the associate had responded more fully. The exchange might have gone like this:
Partner
I need to know whether a “sale” to Jensen actually took place on the evening of July 18.
Associate
Meaning of “Sale” Under the Connecticut Dram Shop Act
Question presented: The Connecticut Dram Shop Act (§ 30-102) requires a “sale” from bartender to patron. A bartender for our client, Ye Olde English Pub, sold two rounds of beer to a group of eight. Witnesses have declared that one member of the party, Jensen, was visibly intoxicated. Jensen did not pay for the drinks. Did the rounds of beer constitute a “sale” to each of the eight guests, including Jensen?
Answer: Almost certainly. Connecticut courts have interpreted “sale” under the act to mean “the purveying or furnishing of alcohol” to a person or group, any one of whom is visibly intoxicated. Last year, the Connecticut Supreme Court held that circumstantial evidence alone can be enough to prove a sale—on facts more tenuous than we have here.
Attached are the statute itself and the three most relevant cases, with crucial passages highlighted. I’ll be happy to write a more formal memo if you like.
With this answer, the associate has taken the trouble to be explicit. She hasn’t answered lazily and superficially by simply hitting “reply” and reporting her findings without stating the issue.
Require that all research assignments be put into writing.
Imagine that same research assignment delivered orally. After a meeting with the client, the partner tells the associate to find out whether a “sale” took place. The associate might well mishear or forget a critical detail relevant to finding the right answer. In that situation, the associate’s first hypothetical response is incomprehensible to all but the assigning lawyer (who will comprehend it for only a matter of days at most). By contrast, the second response, with the question presented and the brief answer, is fully understandable to every conceivable reader. And it’s more likely that any error will be caught.
The point is that you’ll predictably elicit better research if all assignments are reduced to writing. Whenever a senior lawyer puts an assignment in writing, it’s likely to be more focused and more lucid. Ideally, the culture in a law office will insist on written assignments.
Require that all research memos be comprehensible to everybody at the firm, not just the assigning lawyer. That is, avoid unstated premises.
Any research that’s conducted within a law office should be reported in a way that’s comprehensible to everyone who might read it. Stated differently, a research memo should be accessible not just to those initially involved but also to secondary readers—people who may later take over the file.
To discover the meaning of a research memo, you must understand the question it answers.
It’s no good to read this question: “Can Johnson recover multiples of back wages and attorney fees under Maryland’s wage statute?” What’s the basis for that question?
Nor is it helpful to begin with a long explanation of facts before that question. A detailed factual statement at the outset is a turnoff. It requires full study by the busy reader, when it might require little more than a glance.
The best tactic is to integrate a factual predicate into the issue statement, like this:
Johnson and his employer agreed that when Johnson’s employment ended, he would be paid salary, insurance premiums and guaranteed bonuses within six months, unless he left his job without good reason. Johnson quit his job but has not been paid all sums due. Can he recover treble damages and attorney fees under Maryland’s wage statute?
Only after understanding this question can we hope to understand the summarized answer:
Probably not. The statute allows a plaintiff to recover up to three times the wage due plus reasonable attorney fees and other costs if (1) the plaintiff is entitled to the wages, and (2) the employer did not withhold the wage because of a bona fide dispute. Johnson’s employment contract provides that he will not be paid salary or insurance premiums if he leaves his job without good reason. There appears to be a bona fide, good-faith dispute about whether he had good reason. For Johnson to recover treble damages or attorney fees under § 3-507.2, he would have to demonstrate …”
What are the lessons here?
1. Whenever you’re reporting research, you’re preparing a legal memo—even if it’s in the form of an email. To say “We don’t do memos” is like saying “We don’t research the law”—or perhaps “We leave our research unreported and unrecorded.”
2. Written assignments are markedly superior to oral ones.
3. You’re never reporting your research just to the person who has commissioned it: If other readers can’t understand your report, then its utility is drastically diminished. And your answer is more likely to be ill-considered or even wrong because of any possible number of misunderstandings. Which can lead to malpractice. And to scary tales like the one that begins this piece.
This story was originally published in the October/November 2024 issue of the ABA Journal under the headline: “Explicit Answers Needed: Some lawyers don’t write memos anymore, and that’s a bad idea.”
Bryan A. Garner is the chief editor of the new 12th edition of Black's Law Dictionary, the most comprehensive law dictionary ever published. He is also the author of Legal Writing in Plain English and The Redbook: A Manual on Legal Style.
This column reflects the opinions of the author and not necessarily the views of the ABA Journal—or the American Bar Association.
