Good Legal Writing: of Orwell and Window Panes

46 University of Pittsburgh Law Review 149 (Fall 1984)

Pamela Samuelson[+]

Copyright 1984 by Pamela Samuelson

George Orwell once wrote that `[g]ood prose is like a window pane.'[1] What I take Orwell to have meant by that remark is that when people read good prose, it makes them feel as if they've `seen' something (whatever the author was trying to convey) more clearly.[2] Put another way, if a writer can induce his or her reader to feel that the reader would have come to the same conclusion that the author reached had the reader done his or her own investigation of the subject matter, the writer has achieved a kind of `window pane' effect on the reader.

There are certainly many styles of successful writing. Some charm by exotic imagery, others by suspense, some even by subtle obfuscations. However, the style of writing I have found to be most successful for legal analyses is the sort which Orwell's comment conjures up: flawlessly clear, lucid, and enlightening. As I have worked with law students on supervised writing projects, I have noticed that lucidity does not come naturally to most law students, perhaps because they have been forced in their legal studies to read so much bad writing that they mistake what they've read for the true and proper model. I have also noticed that simply directing someone to be lucid doesn't often help him or her to become so. Over time I have developed a set of ideas about how one can make one's legal prose become like a window pane. In this Essay, I will share those ideas with you.

You will soon notice that this Essay is addressed chiefly to an audience of inexperienced legal writers, most specifically to those law students whom I teach. More experienced writers may find the Essay mildly entertaining, perhaps even picking up a tip or two for improving their own writing technique. I hope that at least some of my readers who are experienced writers, especially those who try, as I do, to teach others to write, will find a fellow traveller's exposition helpful to further their own thinking about the mysterious process of learning to write well.

Despite what you may infer from the detailed comments within, I do not believe in a formulaic approach to legal writing.[3] There is no one `right' way to organize, analyze, or write about any legal issue. [4] There are thousands, if not millions, of `right' ways. My aim in reviewing other people's writing is to help them find the analytic approach, organization, and style that will work for the ideas they want to convey. To put it another way, I try to help each of them find his or her `voice' as a writer.[5] No one can learn to write well by aping another person's style. I tell my students: The first thing you must do when you write is be yourself; the second thing you must do is learn to appreciate and enhance your natural strengths as a writer (yes, you have them); and the third thing you must do is understand and work on removing whatever weaknesses are undermining your strengths. I see my role as facilitating this process of self-discovery.

Now for the framework of this Essay: In Section I, I set forth what are for me the six paramount rules of good legal writing. These have general applicability to the whole of one's writing. This is followed in Section II by a set of fourteen additional rules, which pertain, for the most part, to specific aspects of a legal analysis. Sections III and IV briefly advise inexperienced writers on how much research to do for a legal analysis and on strategies for overcoming writer's block. The final section suggests an approach to editing other people's work and communicating with the writers about their work. This approach, in my view, is the one most likely to be successful in yielding the intended result of improving the quality of the writer's next draft. As you'll see, I view an editor's job as well done when he or she has rendered him or herself obsolete. (Now that I have given you this structure, I feel free to be a bit jaunty and jocular, hoping to lead you further into my web.)


1. Have a Point

In order to write a good legal analysis, you've got to have a point (that is, a thesis) you want to make.[6] In reviewing student papers, I have noticed two sorts of problems students have with the seemingly simple point of having point. One is not to have a point at all. There are two main subcategories of this problem: one is being very descriptive about one's topic,[7] that is, surveying rather than analyzing it; the other is demonstrating the author's ambivalence about the issue.[8] Not having a thesis is unacceptable. So is being ambivalent about what your thesis is. Go ahead. Take a risk. Bite the bullet. Choose your thesis and see where it takes you.[9]

The other sort of problem--one to which enthusiasts, in particular, are prone--is to have twelve or thirteen points. It is very easy, as one begins to get familiar with a subject, to get captivated by the myriad issues which any reasonably interesting subject will raise. One may find it difficult to resist the temptation to put all of the interesting issues in at least the first draft without tying them together in a single thematic structure.

I tell my students: Very few people can write in any depth about more than one major issue at a time, and you probably aren't one of them. From my perspective, the aim of this exercise is to explore one thing (your thesis) in depth. While it may very well be possible to weave some of the more closely related tantalizing issues into the fabric of your paper (or, better, into the footnotes), you must make them work in service of your thesis. However sharp your insight, if it doesn't advance your argument, drop it.

2. Get to the Point

Quite as important as having a point is getting to it with reasonable dispatch. This means that you should tell your reader what the point is in your first paragraph, if possible, or at the most by the end of the second page.[10] It means that you should start your analysis of the thesis on page two or three, not on page twenty or thirty. It means that you should remind your reader of your thesis as you go along by such means as section titles and transitional sentences.[11] Think of your thesis as the spice of your paper. Just as you wouldn't put all the spice for your dinner into your dessert, don't put all of your analysis in the latter third or quarter of your paper.

3. Adopt a Structure For Your Analysis That Will Allow You to Integrate the Facts, Court Analysis, and Policies Into the Body of Your Argument

The reason many law students seem to take so long to get to the point is that they have adopted what I deem to be an artificial structure, often a structure which could be used for any paper, no matter what its thesis. If this has been your habit, think about breaking it. It's time to try a more unified, more integrated, and more analytical structure. It is time to learn to get to the point on page two. It is time to learn to develop a structure which can be the structure only for that paper because it's based on the particular analysis you've developed to support your thesis.

What do I mean when I say law students tend to adopt an `artificial structure?' Here's an example:

I. Introduction (Often this is an introduction to the general subject matter of the paper. If it mentions the paper's thesis, it often does it in a sentence tucked away at the bottom of page three or five in an inconspicuous place.)

II. Background (This is often an overview of some of the historical context of the problem and/or of legal concepts which are to be discussed in the analytic section fifteen pages later. It often reads like an encyclopedia or a dispassionate general treatise on the subject.)

III. Facts (Often this is a recitation of `the facts' of a particular case which will be under the analytic microscope in Section V. It often meanders through the odds and ends of the case, often giving many more facts than is necessary to make the analytic point.)

IV. Court Decision(s) (This is usually a report on the trial and any appellate court's holdings--usually all of the issues, not just the ones pertinent to the discussion--and sometimes also of arguments rejected or accepted and any titillating dicta, pertinent or not.)

V. Analysis (Finally. The author now attacks court A for this and court B for that, repeating in the process all the arguments or holdings discussed in the previous section, the facts of the next previous section, and the concepts of the background section, and inadequately developing all of them in terms of their relationships to one another.)

VI. Policy (If law students mention the policy implications of their thesis at all, the policy issues are trotted out in a page or so near the end, usually held up for display as if they were the recently severed head of John the Baptist. Horrors!)

VII. Conclusion (This is often a rather mechanical repetition of the kernel of the analysis from Section V. It rarely contains anything new or interesting.)

What's so bad about this structure, you may ask. Lots of things, I will answer. The worst thing about it is that it's the sort of structure that doesn't start getting to the point until page twenty.

A concomitant bad effect is that along the route to your analytic section, your reader is very likely to get distracted by the unfocused discussion of history, concepts, facts, and the like. Your reader isn't your mother who might be willing to read everything you write with fascinated attention.[12] Even if you've told your reader what your thesis is in the first page or so, your reader will have forgotten it by the fifth page of your meandering history. And by the time you get to the analysis itself, your reader will have forgotten the background stuff mentioned in your earlier sections. Unless you repeat in Section V what you said in Section II, a diligent reader may have to turn back to Section II and reread it, which again is a distraction from the smooth development of your argument.

This brings me to a third bad thing about the kind of structure I've been discussing. It requires a lot of repetition as glue. Repetition is all right if it's necessary, but if you organize an analysis well, there is little or no need to repeat yourself.

A fourth ill effect of this structure is that the potential power of your argument is dissipated by having been so severely chopped up into bits and sprinkled throughout the paper. If the historical context of argument A is discussed on pages four and five, the facts on seven, the courts' holdings briefly on eleven and thirteen, your analysis of them on sixteen, and the policy implications on twenty-two, the forcefulness of the argument will be less than if all are integranted into one section of seven pages. A highly diligent reader will feel forced to do the work for you, but most readers will say, `Ho, hum, I'm glad I got through that mess, but what was it about?'

Remember that the whole point of legal writing is to persuade your reader of your thesis, so you shouldn't structure your paper to impede your ability to persuade. Integrate, and your prose will be like a window pane.

4. Break Your Analysis Up Into Its Component Parts and Develop Them Separately, But In An Organized Way

One can only analyze an issue in an organized way. There are a great many ways to organize the analysis of any thesis, but all good ways have this in common: One's analysis must be broken down into component parts, and each element must be examined and developed in an orderly and integrated way. The structure of your paper should reflect the basic components of the argument which has to be made to support your thesis. It's hard to be more specific than this about how to break down the analysis into its component parts because not only does each thesis demand a different set of components, but each writer on the same thesis might use different components or might order his or her components differently.

Do make an outline before you begin to write. It needn't be a formal outline with I, A, 1, a iii's in it, but it should break the argument up into the basic components and it should reflect the order in which the strongest case for your position can be made. Play around with the ordering of the elements. Ask yourself why what you put first needs to be there.

Try not to have more than three basic elements in your argument. Why, you ask. Simple. Human beings are limited creatures. They get bored or confused if you make six, twelve, or twenty arguments at once. People are pretty good at holding onto two or three basic arguments or components of arguments at the same time. So pander to them; pick what's most important to say in the text and put everything else in footnotes (or in a folder for your next book).

Allow yourself to revise your structure as you go along if, as you write, you find yourself dissatisfied with your initial organization. Reread the paper after you've written it to see if the organization you tried has been successful. If not, try to rework it in outline form before rewriting the whole paper.

5. Adopt a Measured Tone

The importance of tone is often overlooked by neophyte writers.

To decide what kind of tone your paper should have, think about whom you want to persuade of the point you're making in the paper. When you know to whom the piece is addressed, you will (hopefully) be able to figure out what tone would be most likely to faciliate your goal of convincing that person. If you keep this audience in mind throughout your paper, you will be able to keep the tone constant.

In legal papers, it is generally a good idea to adopt a tone of measured rationality, as if you were saying `let us reason together on this issue.' Target the paper as if the audience were a reasonably intelligent and diligent judge who until now has had little or no exposure to the issue on which you write but who is about to make an important decision on it.[13] Assume this judge has some serious reservations about the position you take, but has not yet made up his or her mind on the subject. Your job is to anticipate and meet all of his or her objections and other concerns. It is more effective to do this not in a defensive way but by incorporating those considerations that seem to cut against your position within the framework of your affirmative analysis of the issue.[14]

Too forceful or strident a tone will make the reader feel bludgeoned. This will not only make the reader angry, but it will also distract him or her from perceiving the merit of your argument. Even your supporters, who may look to your piece for intellectual support, may be offended by too harsh a tone.

Too colloquial or journalistic a tone will make the reader think you are not really taking the issue seriously, or are not capable of taking the issue seriously, except perhaps in an emotional way. Too wishy-washy, vague, or ambivalent a tone will cause the reader to doubt whether he or she can gain anything by reading what you have to say about the matter.

The tone of a paper is one of the primary things which will make a reader decide whether to trust or distruct the writer. If the tone is wrong, most readers will stop reading the paper.

6. Be Concrete and Simplify Whenever Possible

Concreteness is related to tone, but it is worth emphasizing as a separate concern. My advice is this: Whenever there is a choice between saying something in an abstract way or saying it in a concrete way, opt for the concrete expression. Or, if you feel you must speak abstractly, at least give a concrete example to illustrate the abstract point. Abstraction and reification seem to be occupational hazards of the legal profession,[15] but one can fight one's impulses to be abstract. Generally, it is wise to do so.

A friend once told me that Albert Einstein said: `Make things as simple as possible--but no simpler.' That is a basic rule of science; it should be a basic rule of writing as well. It should apply to everything about your writing from your theories, to the thread of your argumentation, to your descriptions, and to the language you use to express your ideas.

Almost any subject can be made to seem to be unremittingly complex. The world is, in fact, a complicated place. However, we can understand the world (or one of its facets) only by simplifying it. The more complicated your argument is, the harder it will be for your reader to follow it. That's why if there's a simpler way to develop your argument, use it.[16]

Don't try to impress your reader with fancy words and long, complex sentences structures. (An occasional fancy word is okay.) When you finish writing, go back over your text and prune the sentence structure and wording. We all tend to be a bit prolix (is this a well-placed fancy word?), especially in our early drafts. At least five to ten percent of your text can usually be trimmed without loss of meaning. Often, the excision of that excess five to ten percent will, in fact, enhance the effectiveness of your communication.


1. Write a Strong Introduction to the Paper

The introduction to a legal analysis is like the overture to an opera. It should introduce the audience to and prepare it for the major themes that will recur throughout the work.[17]

Switching metaphors, an introduction should be a kind of roadmap of the terrain you are planning to visit on the journey you have set for your reader. Suspense works in detective fiction, but it's a lousy technique in legal writing. Readers want to know where you're headed. They want to know your destination (your thesis) and your intended route to it (the component elements of your analysis in the order you'll develop them). Your readers will be more inclined to trust you if you pay minimal attention to this need they have. A good introduction prevents anxious readers from being distracted by the question, `Will the author discuss this or that issue?'

The introduction is also your best chance to make the reader excited about the topic and your thesis about it, so don't blow it. Don't just tell the reader that the subject is important or interesting; demonstrate that it is. Your reader should finish the introduction with the thought `hey, what a snappy idea; I can't wait to read about it.' A good introduction gets your reader prepared to work along with you. He or she anticipates what you will say as the paper unfolds its argument.

2. Use Meaningful Titles to Introduce Each Section of Your Paper

First of all, use section titles. It keeps the reader awake and helps the reader get oriented about what to expect. Apart from the introduction and conclusion, section titles should make reference to the aspect of your argument addressed in the section. (Titles such as `facts,' `background,' and `analysis' are not meaningful titles.)

You need not make section titles into full sentences about what you intend to `prove' in the section, but use section titles to remind your reader of the purpose of the section and where it fits into your overall argument. On the other hand, don't assume that the reader has read the title or that the reader has treated the title as the first sentence of a new paragraph. Section titles are section titles, not part of the text.

Two other things: Be affirmative in your statement of the purpose of a section. `No' or `not' should not appear therein if you can think of a way to say it well without the negation. And avoid like the plague cute or journalistic titles to sections or subsections.[18] Only the most subtle of puns is acceptable.

3. Make Your Case Discussions As Thorough and Yet Brief As Possible

Disembodied rules extracted from cases are virtually worthless. To understand why a rule makes sense, one must understand the particular circumstances in which the rule was applied and the considerations which attended that application. Therefore, my rule is: Never mention a case in the text of your paper unless you describe at least briefly the context from which the principle you are interested in emerged. If it is an important case, more of the circumstances should be discussed. If the context is not worth discussing, the case is not worth mentioning in the text, although it may be worth a footnote.[19]

4. Have a Strong Opening Line into Case Discussions

There are few things more boring than a twenty or thirty page paper consisting of page after page of a case discussion in which each paragraph begins: `In A v. B . . .'; `In C v. D . . .'; `In E v. F . . ..' By the third one, the reader feels like saying `who cares?'

Add a little zip to these paragraphs by a strong lead-in sentence. Use the opening line to give the reader some clue about what's interesting (or whatever) about the case. In other words, what's the point of discussing the case? Tell your reader as you open your discussion of it. Then he or she will work along with you to make the point.

5. Make Transitions Smoothly

One often writes paragraphs in chunks. The first rough draft may be an assemblage of these chunks. One's main concern when writing a rough draft is to get the chunks into some semblance of order. One way to tell whether the chunks are in proper order is by trying to write transitional sentences to tie one chunk to the next. If you can't find some way to connect the ideas developed in one paragraph to those being developed in the one that follows, it is likely that you have left out some component of the argument.

As you begin to polish the piece, continue to work on your transitions from paragraph to paragraph and from point to point. The paper should flow, not start and stop abruptly like a train on a milk run.

6. Resolve Issues as You Go Along and at the End of the Paper

Another transition problem, but one worth emphasizing separately, occurs when a writer goes back and forth about pros and cons of an issue and then goes on to another issue. Your reader needs a sense of closure on one issue before you go on to the next or a sense that closure is coming. In concluding your discussion of an issue, pull all the strands of your argument together and give that part of the piece a resolution. The same thing should be done at the end of the whole paper. It is not pedantic to tell your reader from time to time what you have already shown and what you are about to show.

7. Foresee and Address Respectfully the Arguments That Might Be Made in Opposition to Your Thesis

Although I urge my students to take a position on the issue about which they are writing, I do not want them to act as though they are advocates writing a brief. A brief is written to argue the strongest case for one side of an issue. In a brief, the writer, consciously or unconsciously, undertakes to minimize the difficulties with his or her own position and to ignore or minimize the strengths of his or her opponent's arguments. The brief writer typically feels that it is his or her opponent's job to develop these positions. If the opponent fails to do so, the brief writer feels no sense of loss.

I tell my students: I want you to be as fair to the strengths of your imaginary antagonist's arguments about the matter as you are to your own position's strengths. You should also be prepared to show the weaknesses of your position, as well as the weaknesses in arguments that would be made against you. Your paper can and should incorporate all of these things. You may be surprised to discover that your paper will be the stronger for its inclusion of these differing perspectives and its frank admission of its limitations.

Anticipate what critics of your thesis would say about your position (if they haven't already made their criticisms known) and incorporate consideration of their concerns into your paper in an affirmative way. That is, don't pose your critic's concerns in his or her terms and then argue the negative. Rather, restructure the critic's points in such a way that any weaknesses in them are apparent. Your reader should gain insights about the problems with your critic's position from your reformulation of the criticisms. Also, guard against setting up straw men. That is, don't oversimplify the position of your opponent and attack only the oversimplified version. It weakens your position in your reader's eyes.

8. Make Use of Footnotes for a Variety of Purposes

Footnotes can be useful for a wide range of purposes. Experience has taught me that most people are aware of only a few of them. Although some functions of footnotes will be familiar to you,[20] it seems worthwhile to iterate some additional functions of footnotes which many of my students in the past have not known:

a) to give the reader an instance illustrating a generalization you make in the text (that's what the `See, e.g., X v. Y . . .' is about);

b) to send the reader to source materials about a particular subject;

c) to demonstrate additional depth in your research and to give credit to those whose thinking has influenced you;

d) to demonstrate the depth of your understanding of the complexities of a subject;[21]

e) to raise issues which are related to the main thesis of your paper but which would seem digressions if developed in the text; and

f) to raise and address some potential objections your reader may have about your thesis that you judge do not need to be developed at length in the paper, but cannot be ignored.

Footnotes allow you to keep your text trim, while still being attentive to your reader's need to be assured of the depth of your understanding of the complexity of an issue.

You should never assume that your reader will read your footnotes so don't put any truly important points in the footnotes. Use footnotes to elaborate, not to make your argument.

9. Never End a Paragraph or a Section With a Quotation

A quotation at the end of a paragraph or a section looks lazy, as if you couldn't summon the energy to finish saying what you intended to say. You must tie any quotation into your argument and emphasize the point for which the quotation has been included in the text. Remember that the quotation arose in a different context than your paper; it must be worked into your context.

10. Keep Your Quotations Short and to the Point

Long quotations look like a lazy way to make a paper long. A good general rule is that you should have no more than ten continuous lines of indented single-spaced quotation in your text at any one time. Apart from suggesting author laziness, the problem with long quotations is that they invite readers to skim. Remember that it's hard to read single-spaced text. In a long quotation, the average reader is likely to read the first four and last four lines. If the point is buried in the middle of a twenty-line indented quotation, the reader is likely to miss it altogether.

11. Don't Track the Language of an Article or Case You Are Discussing Any More Closely Than is Necessary to Convey the Ideas with Precision

Occasionally, it is tempting to track closely the language of a case or an article. The more technical the subject matter, the more beautiful the prose, the stronger is the temptation. Resist the temptation. It's plagiarism. Frequent footnotes acknowledging the source does not make it something other than plagiarism. You won't have truly analyzed an issue until you incorporate it into your own terms of expression.

12. Be Attentive to Proper Form

One of the things you have to adjust to if you want to be a lawyer is that form counts. In legal writing, form is part of the substance. If you are sloppy about your form, your reader will assume you may well be sloppy as to the substance as well. (In my experience, this is usually a good inference.)

A few particular form rules: (1) Always cite cases and statutes in proper `bluebook' form. It is your responsibility to teach yourself the form rules. (2) Never (well, almost never) split infinitives. (3) Spell all the words in your piece correctly. (4) Don't get cute about personal pronouns. (`He' or `he/she' or `he or she' is ok; `she' is too cute.) (5) Don't construct run-on sentences.

Your seventh-grade English teacher may have been a bore, but he or she was right to insist that you pay attention to spelling, grammar, diction, and the like. You can't afford to take the chance of your argument being dismissed because of your sloppiness about the little things.

13. Have a Healthy Balance to Your Paper

How much introduction, background, case discussion, analysis, etc. is needed in any paper depends on its subject. A good rule of thumb is to keep your introduction to about five percent of your text ( 3/4 of a page for a fifteen page paper; one page for twenty pages); your conclusion to five to ten percent of the paper (five percent if you're only summing up; ten percent if you're suggesting a new approach); and equal quanta per internal argument (twenty-five to thirty percent per issue if you have three arguments sections; forty to forty-five percent per issue if you use two). Watch for imbalance in your paper. After you write your first draft, read it over and ask yourself whether you need to do more development of one part or need to cut another. If you can't tell if it's imbalanced, ask a friend to read it.

14. Consider the Policy Implications of Your Thesis

Questions you should address include the following: What effect would the rule you propose have on the way the world works? What effect would a contrary rule have? Is the intended effect achievable? Is the effect you seek to bring about worth the cost? What consequences besides the ones you intend are likely to result from the rule you espouse?

The reason your legal analysis should consider the policy implications of your proposal should be obvious. Words have power. They often serve as a basis for action, and, when carried out, they have an effect on the society of which you are a part. You are about to cease being a student (whose words rarely have much effect on anything but their grades) and about to become a professional whose words and actions have effects on real people--not only on your clients and their opponents but on other people in society. I think you have a responsibility to use your words to do good and to persuade others you are doing good. Also, people don't adopt rules just because they are clear or consistent; they adopt rules because they make sense in ordering human affairs.


A question almost all inexperienced writers have that they almost never have the temerity to ask is: How much research should I do? I know you'd like to ask me this question, and surprise, I'll even answer it. The answer is simple: Do research adequate to the task you've undertaken.

You will no doubt protest, `but that tells me nothing at all!' Don't be so hasty. It does tell you something. It tells you that different topics require different amounts of research. Usually, although not always, it's possible to have a pretty good sense at the start as to how much or how little research a particular topic will require.

If you decide to tackle a very difficult topic, you will have taken on a commitment to do the research adequate to the task. That may mean you'll have to do more work than one of your colleagues, but remember, you chose the topic. Choosing a topic about which there is little in the traditionally recognized literature doesn't necessarily mean you'll have it easy. In that case, you will be expected to find everything there is on the topic, make the most of what there is, and be more creative in your analysis than your less adventurous colleagues.

At a minimum, you must read all the pertinent cases, treatises, law review articles, other commentaries, legislative history, and whatever else you can get your hands on. What if there is an unmanageably large volume of material to read? One response is to abandon your thesis and look for a new one. Another response is to refine your thesis in such a way that you can pare down the volume of pertinent material. A third--the one your dread--is to go ahead and get started. Ambition is a great thing if you've got the energy to satisfy it. (I advise you to try the second strategy before consigning yourself to a new search or six years in the library.)

What to read first is another important question, and I have advice about this as well. The temptation is often to go to secondary or tertiary sources first (that is, annotations describing cases or treatises) and read them to find out what's going on. While I urge my students to comb through these sources to find cases and references to other source materials, I believe quite firmly that the best and most original papers are written when the writer reads the primary sources first and reads them very thoroughly before getting into the secondary sources. For most legal analyses, this means reading the cases first, and the law review articles, treatises, etc., second. If you read the cases first, your mind will be fresh and unencumbered by the analytic constructs of other authors. You'll have a chance to form your own opinion about the cases, to be attentive to certain details or arguments other writers may have overlooked, and to formulate an analytic framework from your fresh impressions. If you read the treatises or law review articles first, they are likely to color your vision. They may well give you a handle on what the issues are. But the cost you're likely to pay is that you'll use the other writer's analytic perspective as a selection mechanism when you read the cases and may lose the chance to come up with a more original approach.

When to stop researching is also an important concern of yours. The truth is that just as it is impossible to overknead bread, it is impossible to do too much research for a paper. As with kneading, at some point you've done enough to make the bread have the proper texture. It is at that point you begin to fashion the loaf.

To know when to stop researching, you really have to be honest with yourself about what kind of person you are. Are you the sort of person who tries to scrape through life with the most minimal exertion? If so, push yourself a little harder to do research for your paper. Or are you the sort of person who tends to take out his or her anxiety about committing words to paper by setting off on an unfruitful search for the `perfect' case, or worse, for the tenth source to support a minor point? If you are this sort of person, then work on holding your impulses somewhat in check. Find the important cases and deal with them. Put your anxiety to work on writing.

I have two main criticisms about my students' research: It lacks depth, or it lacks breadth.[22] When I talk of lack of depth, I generally mean one of two things: either that you have discussed or referred to too few of the pertinent cases about the issue or that you have unduly restricted the scope of the question, and thereby missed cases relatively close to the point.

When I speak of a need for greater breadth, what I am likely to mean is this: To make a convincing argument about a point it may be necessary to refer to sources other than cases or law review articles. If you have restricted your discussion to these sources, your paper lacks breadth. I will urge you to hunt out other types of sources.


We have all had the experience of facing a blank pad of yellow paper with an equally blank mind. If this happens to you as you try to write the introduction (or whatever) to your analytic paper and it lasts long enough to make you fear you've caught the dread disease `writer's block,' don't panic. It's not terminal. The best way to overcome it is often to try getting started a different way from the one you initially chose.

Everyone has a different set of techniques for making a breakthrough on a writing project. A technique that may work for you may not work for me and vice versa. Still, you may be interested to know about a few things I do when I get stuck in the muck when I'm writing.

If I've made an outline (and that's the first thing to do if you're stuck), I go back over it to see if there is something in there about which I feel I can write at least a paragraph. If so, I do it. I continue to write in chunks until I've found a narrative line to tie a set of chunks together. When I've done this, I usually find I've broken the ice and can begin writing the earlier parts of the paper.

Or if there is a case that is of particular importance to my paper, I start writing about that. I start with a description of the facts.[23] Then I write about the court's analysis and finally about my view of the implications of the case. Again, this usually breaks the ice and gets me going.

At other times I find it helpful to jot town a set of ideas about one part of my argument (in somewhat greater detail than my outline would show). Then I monkey around with ordering the jotted ideas; then I try to write a paragraph or more using this order.

If none of this works, I try to find someone who will let me explain what I am trying to say in my paper.[24] I try to remember what I said or I ask my friend to take notes. (It can also help to tape it if one can talk without feeling self-conscious because of the machine.) Then I write up what I've just said. Of course it won't be perfect, but it's a start. Most people find it easier to work from a rough draft than from nothing.

Sometimes talking to a friend can also help you detect if there's some missing link in your argument. Often what's blocking your writing is that you unconsciously know there's some unsolved problem with your analysis. Once you find it, you may be able to solve it, and that may break the block.[25]


Everyone feels a little naked when another person reads something he or she has written. This is particularly true if the reason for allowing the other person to read it is to find its imperfections. Every writer knows there are the prose equivalents of warts, scaly skin, or a roll of fat around the waistline of the body of his or her writing. If you are asked to edit someone's work, think of your role as similar to the job of a plastic surgeon who has been asked to remove a wart or the nurse at a weight reduction clinic who has been consulted to help a fat person lose weight. You will not be doing your job properly if by your silence you communicate to your clients that they really haven't got the problems they came to you to have fixed, whether those be warts, rolls of fat, or lumpy muddy prose.

On the other hand, if your enthusiasm for your job impels you to a fevered description of the multiplicity of defects you see, the listener may be so devastated by the criticism that he or she will be paralyzed and unable to summon the motivation or confidence to remedy the situation. The hard thing, then, is to find a way to communicate enough of a positive response to make the person feel as though there is something about his or her writing to be salvaged and enough of a negative response to make the person feel that much work remains to be done.[26] Here is what I tell my students about my role as editor and how they should take my comments.

It is unlikely that anyone will ever have read your papers with as much care as I do. Whatever else you may feel after getting my comments, you are unlikely to feel neglected. However good your paper is, you are likely to get extensive comments from me. It is important that even the best students know how much further they could develop their analyses. The extensiveness of my comments should be taken not as a sign of my displeasure, but as a sign of how engaged I've become in your work and as an indicator of how rich are the opportunities for further exploration. It will only be the outstandingly good or bad papers that will leave me speechless.

My comments will range from scoldings about split infinitives to questions about your interpretation of a case or statute or to disquisitions about issues you raise which I think deserve fuller treatment. If you take my comments in the way I wish you to take them, you will do two things: (1) Respond to each of them not by automatically making changes as I have directed (or as you think I have directed), but by making a judgment whether some change I have suggested makes sense to you. Don't make the change unless you are persuaded that there is an ambiguity which needs to be corrected or that the question I raise needs to be addressed. Just be prepared to defend your decision if I challenge you about the issue a second time. (2) Treat my remarks as an invitation to go back over your own writing with a critical eye. Use my remarks as an aid to your reexamination of your writing; look for things I missed or things I only hinted at. Use the insights you gain from this process to rebuild your paper. A friend once told me that his dissertation adviser would heavily edit two or three paragraphs of his dissertation draft and then simply direct my friend to `continue on in this fashion.' You should do the same.

Understand that the point of my remarks is to help you develop your own self-critical capacities, to make you your own best editor. When you have gotten to the point that you can read a page or two of your own prose and anticipate what I will say about it (or better still, think of things I should have said about it but didn't), you don't need me any more.

Well, that will do for preliminaries. Now get to work and have fun.[27] Yes, at the same time.

[+] Associate Professor, University of Pittsburgh School of Law. B.A., M.A. 1971, 1972, University of Hawaii; J.D. 1976, Yale University.

[1] G. ORWELL, Why I Write, in A COLLECTION OF ESSAYS 320 (1954).

[2] Think of how many of the words we use to describe good writing that are derived from words referring to light or related to light imagery: lucid, enlightening, elucidating, illuminating, dazzling, brilliant, reflective, polished, to name a few.

[3] That is, I do not believe every paper should be organized and written in the same way: First, do X, then do Y, then do Z. People get bored by formula- structured work. A writer's aim is to keep his or her reader interested.

[4] On occasion I suggest to a writer whose work I'm editing an organization or approach I might use if I were writing on the topic. I do this mostly to help the writer think of alternatives when I see a problem with the existing organization. To the extent I can, I try to work with the writer on the organization and mode of analysis he or she has chosen for the piece.

[5] . I also find it necessary to tell my students to be aware of my objectivity. There are many ways to analyze an issue and many acceptable outcomes on particular issues. I tell them I will be neither harder nor easier on them if they take the position they think I might espouse. My only concern in reviewing their work is to help them build the best possible argument for whatever position they decide to take. Even harebrained ideas can be made to look respectable. I also tell my students that for perhaps the last time in a long while, they will be able in their papers to be above the adversarial fray of well-heeled interests. I ask them to disregard what their past, present, or future professors, employers, or clients would think about the issue. Write about the topic, I say, as if it was your responsibility to decide what to do about the problem. Even better, write as if you were explaining to someone who would make the decision why the course of action you recommend or the way of looking at the problem that you have adopted, makes the most sense of all the possible alternatives and is the one most conductive to the best welfare of society. See Section II, Rule 14, regarding consideration of the policy implications of one's thesis.

[6] It is often necessary to do some research to refine one's approach to a topic enough to have a thesis about it. I usually require my students to write a summary of the thesis of their paper in the first few weeks of the term in order to prod them to formulate one. I tell them they'll be surprised how much easier it is to be research and organize their thoughts about a topic when they decide what point they want to make.

[7] In A v. B the court held X; in C v. D the court held Y. The cases are in conflict. Conclusion: too bad, or there is a need for the issue to be resolved, but I don't care to help with it.

[8] Now I think X; but then there's Y; and yet Z makes me think something else; so maybe I should go to Hawaii and forget about it.

[9] I tell my students it's okay to change their minds about their theses if further research or thought reveals a gaping hole in the roof of the argument.

[10] See Section II, Rule 1, regarding my advice on strong frist paragraphs.

[11] See Section II, Rules 2 & 5, regarding section titles and transitions.

[12] For those of you whose mothers don't have this attitude, isn't it true that you'd like her to feel this way?

[13] Someone who strongly favors or is adamantly opposed to your position won't need your reasoned analysis to make up his or her mind.

[14] See Section II, Rule 7, regarding the advice on respectfully addressing your opponent's arguments.

[15] There seems to me to be two main reasons for the typical law student's lust after abstraction. One is that they have spent their first year of law school reading judicial opinions which themselves are replete with abstractions. The second is that in reading the cases for the purpose of extracting abstract principles, law students become intoxicated by the principles and mistake them for the real thing in the cases which are the situations.

[16] Scientists will also tell you that simpler explanations are also more likely to be true.

[17] It should be an overture only to your opera. Read your first paragraph after you've written it. Make sure it is specific enough that it could only be used to introduce your thesis.

[18] See Section I, Rule 5, regarding the importance of tone.

[19] At a minimum, you should say who sued whom for what, whether there has been a judgment or only a preliminary ruling on the matter for which you are mentioning the case, and whether the court made a ruling on the point for which you're citing the case or only incidentally commenting on it.

[20] Among the ones students usually know or ought to know already are: (a) to make reference to the page and volume citation of a case or the title and section number for a statute named in the text; (b) to let the reader know exactly where a quote can be found in the original source; (c) to recite the pertinent provisions of a statute which is paraphrased in the text.

[21] It may be unnecessary and unwise to put all the details of, say, the history of an issue or the facts regarding a technology in the text, yet your reader may want to know you've done your homework.

[22] Both of these problems may be apparent if you outline your argument and line up whatever support there may be for each element of it. If you have six sources for the first point, but only one for your second point, it may be a good idea to hunt for more help for the latter.

[23] Usually I overdo the facts in a rough draft, but it's easy to pare them down to the essentials later and it makes me feel like I'm getting the writing underway (and that's the whole point, isn't it?).

[24] Nonlawyers are better for this job than lawyers or law students because talking to a nonlawyer will make one more likely to explain in a simple, straightforward, and relatively organized way.

[25] I have discovered that many of my students don't know the difference between a rough draft and a first draft. I tell my students that when I ask them for a first draft, I am not asking for what is literally their first draft. Reorganize your real first draft, I say, rework it, and rephrase it as much as you can before giving it to me to read. Only when you feel you have exhausted your ability to make progress in reworking the paper on your own, is it time to submit the draft to me. The more they give me, the more likely they are to engage me, and the more likely I am to be able to be helpful to them in reworking the paper to achieve the peak of persuasiveness.

[26] The Scylla of this process is the natural arrogance, self-confidence, or resistance of the insecure to hearing any criticism; the Charybdis is the tendency to distort any criticism that is made. The natural insecurity of the writer will magnify every criticism out of proportion to the editor's real intent. The editor must steer a course between these dangers.

[27] As D. H. Lawrence once said: If it isn't any fun, don't do it.