IRAC, REA, Where We Are Now...

Keynote Address

Capital Area Legal Writing Conference

February 26, 2011

As of 2011, teachers of legal writing have come a long way. In 1975, I surveyed 80 law schools to learn what courses they offered in legal writing. I found a wasteland: few courses, fewer requirements, no Programs.

By 1984, there were a great many courses and some Programs, but no identity for the professionals doing the teaching or the field being developed. The founding of the Legal Writing Institute in that year should be noted as a watershed moment. Legal writing teachers found they were not as alone as they felt. That first national LWI conference brought hope. Over the years that hope grew and flourished. We began to feel ourselves professionals, no longer mere hirees. We discovered we shared the same problems and the same challenges. We began talking to one another. We stopped inventing multiple wheels.

Now, in 2011, we are legion. Of the 197 law schools I visited on the Internet (with the able help of Marjorie Pendergraft, to whom my thanks), every single one offered at least one required course in writing, many had multiple courses, many had multiple requirements, and most seemed proud and responsible to be offering this support to future legal professionals. The LWI is twenty-seven years old and still growing. We have come a long way.

I am more than grateful, honored, and humbled to receive this year’s Golden Pen Award. An important part of that award is this opportunity to publish in Legal Writing: The Journal of the Legal
Writing Institute. I would like to use it to take a look at some of our common practices and to suggest where we might do well to go from here.

Our Perhaps Too Tasty Alphabet Soup

Together, as a group of professionals, we have created a teaching device for first-year legal writing courses that has gained wide acceptance: We have seemingly solved the problem of teaching the structure of a legal argument. The first widespread embodiment of this solution appeared under the acronym of IRAC. Under the guiding hand of IRAC, our students have been taught that a legal argument ought to begin with the Issue, continue with the Relevant law, proceed to the Application of facts, and conclude with the Conclusion. Its efficacy seemed clear: Depositing the right information in the right places will allow the argument for your side to be perceived with clarity and judged according to its inherent persuasiveness.

IRAC held sway for a long while. But then critics arose who found flaws in the structure and produced a host of acronymic counter-structures — CRAC, CREAC, MIRAT, IDAR, ILAC, TREACC, CruPAC, ISAAC, CRRACC, and (with a nod to the present President) BARAC, to name only the most prominent amongst them. What does all this creative fervor tell us? That we are coming closer to the One True Structure, which, when we find it, will solve “the problem” of legal writing and of teaching legal writing? I think not. It demonstrates something quite the opposite: That there is not and cannot be a single structure that is the right answer to the question of how argumentative thought is best conveyed from the mind of a writer to the mind of a reader. These organizational structures are both necessary and dangerous, both supporting and defeating. As with any good idea or good invention, they can all be used for harm as well as good. Even Penicillin, if used at the wrong time or in the wrong context, can kill.

I fear we may be creating for our students, by our over-reliance on these alphabetic formulas, something of a Procrustean bed. Procrustes was a character from Greek mythology who lived just off the road from Athens to Eleusis. Each evening he would offer a weary traveler the hospitality of his home for the evening. While the food and wine might have been fine, the sleeping arrangements were not. Procrustes insisted that his guest fit perfectly into an iron bed he had constructed. If the guest was too tall, Procrustes would cut off the guest’s feet and legs to make him the right length, from which the poor fellow died. If the guest was too short, Procrustes would stretch him on a rack, until, once again, from internal hemorrhaging, the poor fellow died. “One size fits all” was not, in this situation, advantageous to the customer. Are we in danger of suggesting to our students that one shape fits all legal arguments? It is not the use of IRAC and the alphabetical like that troubles me; it is the potential intellectual
abuse.

I find two major problems with relying so heavily on this kind of pedagogical and strategic device. First, used with rigidity, these structures can breed a false sense of security in students: Since they merely need to arrange their material into the instructor-approved schema in order to succeed in their legal writing class, repeating this process in the world after law school, they might believe, will surely produce the same positive results. Second, our dependence on highlighting these matters of gross structure has led many of us as teachers to de-emphasize or perhaps ignore completely the most fundamental and essential skills of writing, without which even our best students might still remain bad writers. Those skills have to do with the forming and controlling of sentences and paragraphs. Ironically, because of our rapt attention to rhetorical structure, our profession is ignoring the rhetorical teaching of language skills.

An Ancient and Persistent Pedagogical Problem of Audience

Let me back up a few years in the educational progress of our students to consider a debilitating problem they had to face when they were undergraduates. In the title of a 2005 article, I posed the question, “Why so many bright students and so many dull papers?”¹ In the article, I asked what we expect to gain by assigning so much writing to students in courses in English or History or Psychology — or in any field you could name. Look at the reality. Students, for the most part, find the writing of all these papers a heavy burden. Teachers, for the most part, find the reading of these papers a heavy burden. If the teachers “do their job” and write prodigious comments on these papers, the teachers, for the most part, consider that a heavy burden. And then, sad to say, those comments are usually no better than about 15 percent efficient in helping the students, because of a number of reasons: A great many of these comments are never read; those that are read are often contextualized by the grade or are inflated or deflated in effect by the student according to the psychological needs of the moment; and the stream of commentary, created interlinearly as the instructor perused the paper, is not experienced interlinearly by the student but rather in a continual flow that makes it strangely into a “text,” which it was never intended to be. It is a problematic process: The students don’t want to write; the teachers don’t want to read; the teachers are burdened by commenting; and the comments don’t do enough good for the students. Why do we slavishly continue with this flawed procedure? Because that’s the way it has always been done. If this is not the proper way to do it, just think of all the time and energy that have been wasted in the lives of generations of teachers and students. That would be hard to face.

What causes this problem that is poisoning these academic waters? I think the cause is the nature of the audience for all this written work: That audience is a fake audience. In the real world, when a professional writes something, the professional is, at least momentarily, presumed to be an expert on the matter. People read to find out what the author knows that they perhaps do not yet know. We have a technical term in the field of Rhetoric to describe this relationship: We call it “communication.”

But it is a lie or a fantasy to suggest that students, writing for teachers, are primarily engaged in the rhetorical act of communication. They do not think that, having spent two days in the library, they are now “experts” on this matter; nor do they think their task is to fill full the empty vial of teacher with the milk of knowledge. They think, perhaps too severely, quite the opposite: They think teacher knows 100 percent of what can be known of this matter. The students’ rhetorical task then becomes not the productive one of “communication,” but rather the more burdensome, narrow, opprobrious rhetorical act of “demonstration.” The students’ job is to demonstrate to teacher that he or she now controls 4.8 percent or 5.3 percent of that which the teacher knows 100 percent. (4.8 percent is a B+; 5.3 percent is an A-.) As long as there is no “real” audience — no audience that has a legitimate need to know what the writer has to say — the writing can quickly deteriorate into being perfunctory, lifeless, and, if not actually thoughtless, at least not primarily thought-dependent.

With this in mind, I offer an observation we might well find sobering: Of all the fake audiences students ever encounter, the one most predominantly fake is the teacher of first-year legal writing. Please do not be insulted by this statement. It’s not in you; it’s in the rhetorical and pedagogical situation. What is the nature of the assignment of that first moot court brief? Most of us think up a challenging fact situation that produces an interesting issue. Then we hand the students all the cases they need from which they can then create a brief. We teach them where to put what — by IRAC or CruPAC or whatever structure has most won our approval. And then, knowing full well what the perfect response to this assignment would look like if written by us, we grade them on how close they came to that preconceived ideal solution. We actually do present ourselves as knowing 100 percent of the content for this particular intellectual challenge. As an audience, we are, sad to say, completely fake: We do not need to know anything they are assigned to tell us.

You can see, then, why depending more and more on their ability to get their Is and Rs and As and Cs into the right cubby-holes poses a real danger. We are forgetting to teach them how to write; we are teaching them primarily how to assemble.

Here is a brief example. Let us say we consider the issue of proximate cause as a limitation of negligence to be the number one most important consideration in this brief. A student who fails to achieve that insight might fail the assignment. But a student who has figured that out and places that issue in the correct location in the assigned structure can receive full credit for that intellectual feat whether the writing of that paragraph was skillful and powerful. The very sentence in which the concept is named might receive the mini-equivalent of an “A,” no matter how well or poorly that sentence is written. The mere existence of the issue on the page fulfills the “writing” requirement.

The way a sentence is written makes all the difference in the world to a reader who does not already know everything there is to be known about its intellectual possibilities. I was married for a decade to one of the finest appellate judges in the State of Texas. She told me that in twelve years on the bench, 95 percent of the briefs she had to read were of little or no help to her in resolving the issue at hand. I assure you that many of those inadequate briefs cited lots of the right cases and raised many of the right issues. They might even have placed the Issues and the  Relevant law and the Application of the facts and the  Conclusions in all the right IRACy locations. But the briefs and their authors had failed to aid her in thinking about all that material. Good writing helps good readers to do good thinking. Good writing tells good readers how they should go about putting all those materials together, from which perspectives thoughts are to be viewed, and how to make the relevant connections between individual pieces of information and legal theories. All of this comes not from the mere inclusion of information, but from its being articulated and arranged in ways that lead the reader — perhaps even force the reader — to think about the material in precisely the way the writer wants the reader to think. That has to do with controlling the construction of sentences and paragraphs — on which too many of us are spending too little of our time. It is to this which we must turn our attention if we are truly to fulfill our own assignment — to teach these bright young adults how to write in the world of the law.

Let me share with you a dis-spiriting, almost frightening moment I had in my law school office last semester. It demonstrated to me how overly far we have gone in our relying predominantly on IRAC-related teaching. For many years I gave lectures to the first-year class at Duke Law School on what I call the Reader Expectation Approach to language. After they heard the lectures, they were free to sign up for office appointments — two or three students for each appointment — during which I promised to analyze their writing styles. With the ability to refer to these principles, it usually takes me about twenty-two minutes per student to demonstrate what deep-rooted habits they have developed in their formation of sentences. Some of those habits are good for their readers; one or two or sometimes three are not. Changing those habits is difficult, but that is where salvation lies. Last semester two young women appeared for an appointment. (I invite multiple students simultaneously because they learn to perceive these habits in a different way when the prose is someone else’s than when it is their own.) They both complained about having to locate material in the highly formulaic way they had been taught — with re-iterated umbrella statements at the beginning of every sub-unit and precise re-iteration of statements under repeated umbrellas and sub-umbrellas. I looked at their prose in search of their habits and found — to my astonishment, and I would even say my horror — that I could find the writers nowhere present in their own prose. Actually, it was not their own prose. It was simply a prose re-ordering of the materials their instructor had produced for them and had required them to include in their pages. All of their sentences came from the assigned readings; all their arrangement proceeded from the instructions they had received on umbrella construction. The students were nowhere to be found.

Difficult Audiences and Bad Advice

Legal writing is the hardest writing of all to do well — not because the material is more difficult than Physics or Philosophy or Fractiles, but because the audience is so often entirely hostile. When a physician writes something, everybody bends over backwards to try to discern what the good doctor was trying to say. But for what audience is a lawyer writing? For a senior partner, who says, “Nothing gets out of this firm without being subjected to my acid test”; or for a judge, who has in her other hand a document arguing conclusions the exact opposite from yours but based on the same facts; or, perhaps worst of all, for opposing counsel who, completely cognizant of what you intended to say, will bend over backwards to prove that the text does not say that or says something different or perhaps says even the exact opposite. That is truly a hostile audience.

To gain the ultimate control possible over such audiences, a writer must achieve the ultimate control possible over sentences and paragraphs, forming them to defend as best they can against the efforts of critical, uncooperative, or even hostile readers. Before suggesting how we might attempt to do this, let me make it clear what I think is wrong with some of the ways far too many writing teachers have tried to do this for 200 years now. Not knowing what else to do, they have spent much of their efforts on producing “correctness” — or perhaps more accurately, on avoiding “error.” The rules of grammar and spelling create ascertainable rights and wrongs, through which the authoritarian teacher can maintain control of and superiority over the students. Little else having to do with language lends itself so readily to the digital neatness of right and wrong.

Aside from this hyper-attention to error, and with all the best of intentions, we have limited ourselves to a slavish reinforcement of a compiled string of maxims — age old now (go check out an eighteenth century writing textbook to see what I mean) — that have supported our need for authority and have even made us feel helpful. Here are the major pronouncements from this long-accepted litany of writing instruction, with my opinion of each.

Avoid the passive. Wrong.

To make it better, make it shorter. Wrong.

Never allow a sentence to exceed 29 words. Wrong.

Write the way you speak. Wrong.

To see if your writing is good, read it out loud. Wrong.

Avoid the use of the verb “to be.” Wrong.

Every paragraph should start with a topic sentence. Wrong.

The worst of these old war horses are the prohibition of the passive and the urgency to reduce the number of words in a sentence. Both of these are seriously misguided and misguiding.

Complex thought cannot be adequately expressed in English without a skillful control of the passive voice. The passive is not only as good as the active: It is better than the active in all situations in which the passive does a better job than the active. It becomes our task to teach what those situations are and how to handle them.

Long, good sentences are still good sentences; short, bad sentences are still bad sentences. Sometimes you can improve a sentence by making it longer.

So if we cannot turn with confidence to the 200-year-old tradition of old maxims, where can we turn to find direction and leadership in the quest for the production of excellent prose? It would seem reasonable to turn to those good people who for more than a half-century now have formed, led, and expanded the Plain English movement in this country. Those earnest people have had the right goals in mind. As a result of their efforts, thirty-seven states have created some kind of Plain Language legislation. I agree wholeheartedly with their goals. I praise their energy and their efforts. Unfortunately, I often disagree with many of their methods. Let us take a look.

The Methodology of Plain English

Such good work has been done in the field of Plain English, spurring improvements in so many areas. If you want to look at some of the best of this, read Joseph Kimble or Veda Charrow or David Mellinkoff or Robert Benson.² Most of what they say makes eminent sense and has had a substantial effect on legal documents written for a non-lawyer audience.

I also find admirable the attempt at a definition of the term in our new federal Plain Writing Act of 2010:

The term “plain writing” means writing that is clear, concise, well-organized, and follows other best practices appropriate to the subject or field and intended audience.³

But it leaves us to ponder what “best practices” have come to be established. Concerning those, I have great many concerns.

Of the states that have enacted some sort of Plain English statute, a number define the term as broadly as the federal law quoted above. While those terms are not objectionable, many feel they have too little effect because they set no specific standards for measuring the “plainness” of English. As a result, other statutes go into extraordinary detail in an attempt to control the readability of language by the use of statistics. Yet still others employ some “best practices” formulae that should have been laughed off the stage a half century ago.

The statutes that try to control the language by restricting the number of words per sentence or the number of syllables per word tend to have one flaw in common: They rely on a false dichotomy. Or perhaps I should say they use the concept of dichotomy falsely. The dichotomy they establish is one of “bad” versus “good.”

Bad Legal Writing                      Good Legal Writing

Characteristics:                                        Characteristics:

1                                                                1

2                                                                2

3                                                                3

4                                                                4

They then, with impressive confidence, require us as writers to avoid the bad characteristics and embrace the good, which will result, they insist, in our producing “Plain English.”

What could possibly go wrong with such a process and such a program? It would all go wrong, of course, if the wrong “characteristics” were chosen. And indeed, the well-intentioned Plain English people who created those statutes have done just that. They have tended to identify not the causes of the disease but rather its
symptoms. Think of this in terms of curing a disease:

Bad Health from a Cold                 Good Health (No Cold)

Symptoms:                                                 Symptoms:

coughing                                                      no coughing

sneezing                                                      no sneezing

We would not profit from a doctor telling us, “Never cough nor sneeze, and you will never have a cold.” It is a different matter altogether if we deal not with the symptoms but with the causes:

Bad  Health from Lung Cancer                  Good Health (No Lung Cancer)

     Causes:                                                       Causes:

smoking                                                       no smoking

genetic disposition                                       no family members have cancer

The doctor can now be of some help: “Although you cannot do anything about who your parents were, you can at least avoid smoking cigarettes. Do that, and your risk of lung cancer decreases.”

The well-intentioned Plain English people in many states have, for lack of knowing anything more certain to do, identified a number of supposed symptoms of bad writing and treated them as if they were the causes. Let us look briefly at three notably specific statutes.

Connecticut’s Plain English law for consumer contracts⁴ tries to be even-handed. It allows the author a choice of definitions: One definition cites the symptoms by name; the other calculates them by numbers.⁵ The former includes the following in its list of characteristics of Plain English:

• short sentence and paragraphs; ⁶
• “everyday words”;⁷
• the use of personal pronouns;⁸
• “simple and active verb forms”; ⁹
• “type of readable size”;¹⁰
• ink which contrasts with the paper;¹¹
• section headings, subdivisions, and boldface captions;¹²
• spacing between paragraphs and sections, and from the borders;¹³ and
• “It is written and organized in a clear and coherent manner.”¹⁴

This list sounds familiar: Most of us have been taught to believe that these are the opposites of the characteristics of bad writing. But, one might ask, how short are “short sentences?” How many days a week must one use a word for it to become an “everyday word”? How small a font is “readable”? The legislators in Connecticut must have been debating these issues, since their alternative definition in the statute — (and you can choose which one you think best suits your document) — tries to answer some of these questions with statistics. It gets remarkably specific; your prose will qualify as “Plain English” only “if it fully meets all of the following tests.”¹⁵ Remember those words “fully” and “all.” Here are some of the items on Connecticut’s list with which they have drawn lines in the linguistic sand:

(1) The average number of words per sentence is less than twenty-two; and

(2) No sentence in the contract exceeds fifty words; and

(3) The average number of words per paragraph is less than seventy-five; and

(4) No paragraph in the contract exceeds one hundred fifty words; and

(5) The average number of syllables per word is less than 1.55; . . . .¹⁶

A failure to meet any one of these five constitutes a failure for the document as a whole.

Such a tidal wave of specificity must give us hope that someone from or hired by Connecticut has actually discovered how to produce readable documents. Personally, I find it breath-taking. With all the writing and reading I have done in my life, I had never once considered whether the prose I was producing or consuming exceeded an average of 1.55 syllables per word. To my surprise, when I investigated this oddly precise statistic by a random selection of texts, I found it did indeed seem to be something of a dividing line — but not so much between good prose and bad, or the readable and the unreadable; it tended to distinguish between the simple and the sophisticated. Complex matters often require a greater number of multi-syllabic words. (The syllables-per-word ratio of the previous sentence is 2.0 — way above the allowed 1.55. Even if you substitute “long” for “multi-syllabic,” it is still an unacceptable 1.6: “Complex matters often require a greater number of long words.”) Syllable counts signify an attempt to control vocabulary. Good writing depends on far more than the assembling of non-challenging words. (The ratio for the previous sentence is 1.58. Still not good enough.)

The low numbers for words or syllables are by themselves no guarantee that prose will be clear: Just remember the last time you tried to assemble something you bought that had the instructions translated inadequately from the original Japanese. All those short words, short paragraphs, and reduced syllable content — and there you sat, still confounded.

Pennsylvania has a fair-sounding Plain English law for consumer contracts.¹⁷   It articulates a “Test of Readability” that includes the regular list of suspects — short sentences, no passives, and no technical terms.¹⁸   Its subsection 2205(d)(1)(I), part of the “Test of Readability” section, requires a contract to have a statement that contains the following:

(i) A general description of the property that may be taken or affected by reason of a security interest or contract if the consumer does not meet the terms of the contract. The statement is not required to list all possible exemptions. As it may apply, the following statement may be used: “If you do not meet your contract obligations, you may lose your house, the property that you bought with this loan, other household goods and furniture, your motor vehicle, or money in your account with us.”

This subsection in Pennsylvania’s test of readability fails to meet the objective standards of readability of the Connecticut statute: The sentences average 27.7 words, well above the statutory 22; the paragraph contains 83 words, exceeding the required 75-word average by more than 10 percent; and the syllables/word average is just a touch high at 1.554 instead of 1.55. But recall the language of the Connecticut subsection: A consumer contract is to be considered written in plain language only “if it fully meets all of the following tests.” Even the minimally excessive syllable per word count would, by itself, have condemned this document.

Notice also the number of passives in this passage from the Pennsylvania statute that itself warns against the use of the passive: “may be taken”; “may be . . . affected”; “is not required”; and “may be used” — four passives in a paragraph that contains eight verbs.

Where prose is concerned, drawing any statistical line in the sand is likely, sooner or later, to create havoc. It is not that Connecticut has chosen the wrong numbers; it is rather that no set of numbers could possibly be the correct one.

Well-written sentences longer than 22 words — all the way up to 200 words — can ring clear as a bell. A badly constructed 10-word sentence can cause major confusions. Sentences with a single clause tend not to exceed an average of twelve to fifteen words. Sentences with two clauses quite commonly extend the word count to at least the mid-20s. Lawyers often have the need to write two-clause sentences: Perhaps they are trying to compare two concepts or two facts or two legal requirements by juxtaposing them; or perhaps they are presenting two distinct chronological narrative units that need to be considered as one continuous action; or perhaps they are constructing a part 1 and a part 2 to a thought, both of which must be present in the same syntactic unit to convey to the reader the unity of that two-part thought. If we examine the sentence that precedes this one, we find some notable statistics. It is eighty-one words long. It has four sub-parts, the last of which by itself is thirty-six words. Its syllable per word ratio is 1.57. Did you have significant difficulty in making your way through it the first time? The eighty-one words are supportable because of the sentence’s structure: I gave you four different places to come to a complete halt; I warned you in the first unit (with a colon) that other units would be coming along shortly; I gave you signs that the sub-units would all be parallel to each other; I put the right kind of information in the right places. (That sentence had sixty-three words.) If a writer knows how to care for readers, by understanding well enough the nature of the reading experience, the number of words matters little if at all. And if a writer fails to understand what readers need and where certain information should appear to best send the right interpretive signals, a ten-word sentence can breed a dozen different interpretations in a dozen readers.

Connecticut was trying hard, but the hope that Connecticut’s standards will at least produce prose that is better than the dreaded Legalese is insufficient justification. Nor does it suffice that this statute might work well in many or even most cases. There are better ways to accomplish its goals.

Far worse than Connecticut’s law is Florida’s Plain English law for insurance policies.¹⁹ In addition to the familiar calls to avoid complexity and length and passives, it requires the following show-stopper:

The text [must] achieve a minimum score of 45 on the Flesch reading ease test as computed in subsection (5) or an equivalent score on any other test comparable in result and approved by the department.²⁰

Rudolph Flesch (1911–1986) was another laborer in the Plain English fields who fought the good fight nobly and produced a great deal of spirited writing in favor of clear and energetic prose. But he left us with the Flesch Reading Ease Test. That test, so precise in appearance, with its calculations to the third decimal place, might be functional as an axe, but it should never be used as a scalpel. Here are the calculations it requires us to make:

1. Count the number of words in the document.

2. Count the number of sentences.

3. Divide the latter into the former to get a words-per-sentence ratio.

4. Multiply that ratio by 1.015 to produce product “X.”

5. Count the syllables in the document.

6. Divide the number of words into the number of syllables to get a syllables-per-word ratio.

7. Multiply the syllables-per-word ratio by 84.6 to produce product “Y.”

8. Add products X (step 4 above) and Y (step 7 above) together, and subtract that sum from 206.835. The final result is your Flesch score.

9. If your Flesch score is below 45, your document fails the test.

The idea that “45″ has some magic qualities to it should have been dismissed in derision decades ago. But Florida uses this test, as does Minnesota, ²¹ and Alabama,²² and probably others I have not yet encountered.

Here is one striking instance of its failure as a scalpel. I was called upon to serve as an expert witness in a substantially large case — 75,000 employees in a class action against one of the nation’s most recognizable huge institutions. The case concerned the Company’s notices, sent to all the employees, to inform them there would be changes in the way their retirement benefits would be calculated. The legal question was simple: Were those ten pages of documents written in plain enough English for all those employees to understand them? I was convinced they were.

I subjected them to the Flesch test. The documents failed, scoring just over forty-two. Then I noted that the Company’s name included five syllables. None of those employees had a five-syllable reading experience when they saw that name. They had a familiar one-syllable kind of reading experience: It read “Boss.” In the ten pages, that name appeared fourteen times. If that name were counted as one syllable instead of five, the Flesch test result zoomed to over forty-eight, a happily passing grade. The scalpel had failed.

Where to Go from Here

The problem here is not simply that this test or some other test is not as mathematically/analytically precise as it need be. The problem is rather that these tests explore only the symptoms of the disease, not its causes. Bad prose may often be laced with lengthy sentences and be sprinkled with a great many passive constructions, but those elements did not cause the writing to be bad. Bad writing comes from not understanding the relationship in English between its structure (created and controlled mostly by society) and its substance (created and controlled mostly by the individual writer). Since we now understand these matters rather well, it is time we taught them to our students.

The bottom-line question concerning the quality of writing is simply this: Did the reader receive what the writer was trying to send? If the answer to that question is “yes,” the writing was good enough. If it is “no,” the writing was not good enough. And it matters very little, or perhaps not at all, how dazzling or impressive the prose seemed to be along the way. It is up to us, then, to discover how readers go about their process of interpretation. The key to that process, I have come firmly to believe, lies in the structural location of the major pieces of information in a sentence. Where a word appears in a sentence will control the way those words will be used by a reader. We know this intuitively as readers; we must come to know it consciously as writers.

Readers of English know where to look for what in a sentence. For more than thirty years, I have taught these “Reader Expectations” in law firms, corporate legal departments, and government agencies across this country — and in scientific research institutions around the world. This work began in the consultant firm of Clearlines, led by Joseph Williams and initially including Gregory Colomb, Frank Kinahan, and myself. There is now a good deal of our writing available that explains the approach in varying amounts of detail. Here are the main sources:

1. Joseph M. Williams’s ground-breaking book, Style: 10 Lessons  in Clarity and Grace.²³

2. Expectations: Teaching Writing from the Reader’s Perspective.²⁴ This is the book I have written for those who wish to teach this Reader Expectation Approach. It explains the principles in substantial detail and offers a good deal of advice for coping with its pedagogical challenges.

3. The Sense of Structure: Writing from a Reader’s Perspective.²⁵ This is the textbook I derived from the same Reader Expectation Approach. Although it does not focus on documents particular to legal writing, it deals with all the major problems lawyers — and all other professionals — encounter when they try to control the English language.

4. For a shorter exposition of the major points of this approach, see my article, The Science of Scientific Writing, co-authored with Judith Swan.²⁶

Understanding these principles and learning how to teach them can change your students’ writing, permanently. If adopted by statutes at the state and federal level, this approach could solve the Plain English problem. Why am I so sure of this? For two reasons. First, the federal government of Canada now uses these principles in the production of all its federal legislation. Second, in that case I mentioned above, with potentially billions of dollars at stake, the plaintiffs based their argument entirely upon the standard Plain English linguistic details, while I argued the case entirely from a Reader Expectations perspective. I filed a 120-page report demonstrating the documents were clearly written and then an 86-page response to the 26-page word-counting report from the expert for the other side. Four weeks before trial — after more than two years of litigational efforts — the plaintiffs withdrew their claim with prejudice and without asking the company for a settlement. As you probably know, this is almost unheard of. When asked by the judge why they were taking this unusual action, the lawyers for the plaintiff responded, “We can’t beat his argument.”

Epilogue

We have come so far since 1975. Formal writing instruction exists at most law schools. Writing faculty are gaining increasingly greater control over their continued employment, are slowly growing in respect from the rest of the faculty, and are forging consistently more productive bonds with their students. We have birthed and nurtured and matured professional associations — especially the Legal Writing Institute — from which we gain support, identity, and a quintessential sense of collegiality. We have become perhaps overly expert concerning the structure of argument. But now we have to add the missing piece — and I would argue it is the single most important piece: We must become teachers of language, teachers of rhetoric, teachers of writing. Knowing what we now know about how sentences and paragraphs function, we must give this knowledge — or any other relevant and helpful knowledge we can produce — to our students, with which they can better make their way in the legal world that awaits them.

© 2011, George D. Gopen. All rights reserved. George D. Gopen is Professor of the Practice of Rhetoric at Duke University, where he was for many years the University Writing Program’s first Director. He has taught langu
age and literature for forty-one years in English Departments and law schools and for thirty-three years, he has taught his Reader Expectation Approach to the language as a consultant to lawyers and scientists around the world. He is the 2010 recipient of the Legal Writing Institute’s Golden Pen Award, a lifetime achievement award for contributions to the field of legal writing.

1.  George Gopen, Why So Many Bright Students and So Many Dull Papers?: Peer-Responded Journals as a Partial Solution to the Problem of the Fake Audience, 16 WAC J. 22 (Sept. 2005) (available at https://wac.colostate.edu/journal/vol16 /gopen.pdf).

2.  Any of the publications of these four vibrant minds will be instructive. Here are some of their highlights. Veda Charrow, What Is “Plain English” Anyway? (Doc. Design Ctr. 1979); David Mellinkoff, The Language of the Law (Resource Publications 2004); Robert W. Benson, The End of Legalese: The Game Is Over, 13 N.Y.U. Rev. L. & Soc. Change 519 (1984–1985); Veda Charrow & Robert P. Charrow, Making Legal Language Understandable, 79 Colum. L. Rev. 1306 (1979); Joseph Kimble, Plain English: A Charter for Clear Writing, 9 Thomas M. Cooley L. Rev. 1 (1992).

3.  Plain Writing Act of 2010, Pub. L. No. 111-274, 124 Stat. 2861 (2010) (codified in 5 U.S.C. § 301).

4.  Conn. Gen. Stat. § 42-152 (2011).

5.  Id. at § 42-152(b)–(c).

6.  Id.at § 42-152(b)(1).

7.  Id.at § 42-152(b)(2).

8.  Id.at § 42-152(b)(3).

9.  Id.at § 42-152(b)(4).

10. Id.at § 42-152(b)(5).

11. Id.at § 42-152(b)(6).

12. Id.at § 42-152(b)(7).

13. Id.at § 42-152(b)(8).

14. Id.at § 42-152(b)(9).

15. Id. at §42-152(c) (emphasis added).

16. Id.at § 42-152(c)(1)–(5).

17. Pa. Stat. Ann. tit. 73, §§ 2201–2211 (West 2011).

18. Pa. Stat. Ann. tit. 73, § 2205.

19. Fla. Stat. § 627.4145 (2010).

20. Id. at § 627.4145(1)(a).

21. Minn. Stat. § 72C.09 (2010).

22. Alabama has opted for the Flesch test in its Pattern Jury Instructions. See Arthur J. Harris et al., “The Plain English” Project of the Alabama Pattern Jury Instructions Committee—Civil, 68 Ala. Law. 369 (2007).

23. Joseph M. Williams & Gregory Colomb, Style: 10 Lessons in Clarity and Grace (10th ed., Longman 2010).

24. George D. Gopen, Expectations: Teaching Writing from the Reader’s Perspective (Pearson Longman 2004).

25. George D. Gopen, The Sense of Structure: Writing from a Reader’s Perspective (Pearson Longman 2004).

26. George D. Gopen & Judith Swan, The Science of Scientific Writing, 78 Am. Scientist 660 (Nov.-Dec. 1990).