In theory, every law school graduate should know something about how to write an effective appellate brief. After all, first-year legal writing classes in law school often concentrate on that skill. Moot court competitions do too. Compared to other kinds of legal work, appellate briefs seem tidy and self-contained, with a predictable structure. So they are what law schools teach.

Once in practice, regardless of law school background, trial lawyers sometimes seem to believe that no special talent or training is needed to write a good brief on appeal. The idea appears to be that what works before a jury or is acceptable to a busy trial judge should be more than adequate for an appellate court.

Despite what law students should learn and despite what lawyers think they know, appeal after appeal is lost, or at least made harder to win, because of ineffective briefs. Why? In part, because many lawyers write appellate briefs infrequently. When they do have to brief an appeal, they fail to appreciate that the job is different from much other lawyering. It poses special problems, but presents special opportunities, for advocacy.

The most common mistake made by trial lawyers is to think that they should do the same thing in the appellate court that they did in trial court. They write their jury speech and call it a brief. At best, they address the appellate judges as they would address the trial judge. At worst, they treat the appellate judges like jurors.

Such advocates bog down in irrelevant detail and empty rhetoric. Ninth Circuit Judge Alex Kozinski’s comments about oral argument apply even more forcefully to the brief: "When a lawyer resorts to a jury argument on appeal, you can see the judges sit back and give a big sigh of relief. . . . [W]e know, and you know we know, that your case doesn’t amount to a hill of beans, so we can go back there in the conference room and flush it with an unpublished disposition."

Even those who understand that a court of appeals is different from a trial court often fail to seize the opportunities for advocacy that an appellate brief offers. They may recall their early law school lessons, but they do not know and do not take (or do not have) the time to study the more sophisticated lessons that actual experience in appellate practice can bring. Their written product is formulaic. It fails to take advantage of the flexibility that an appellate brief writer has in packaging arguments to meet the needs of a particular case.

Here is what we mean by effective packaging: A few years ago, the Supreme Court considered a case that turned on the interpretation of two complex, interrelated statutes. One statute involved regulation by the FDA, and the other involved patent law. Conventional law school wisdom would have called for the brief to begin with a statement of the events giving rise to the controversy, followed by a description of the proceedings below. The winning brief did not do that. Instead, it opened with a four-page description of the statutory scheme. Not one sentence on those four pages was argumentative or even disputable. The passage alerted the Court to the statutory elements that the brief writers knew were most significant and helpful to their side. It gave the Court a framework to understand everything else the brief said — from the statement of facts through the conclusion of the argument.

Ultimately, the Court ruled in favor of the side that had taken the unconventional approach, saying that it found ‘ "the structure of the [statute] taken as a whole" ’ to be dispositive. The critical information the Court needed to rule as it did was in those first four pages. Of course, this technique is not right for every appeal (although it probably makes sense more often than not in cases turning solely on statutory construction). But it is one way an advocate can achieve maximum effectiveness while staying within the rules.

Note that we have referred to "staying within the rules." That is important. The rules are the first thing any lawyer must consider before putting pen to paper — or fingers to keyboard.

A surprising number of prominent litigators fail to read, understand, and follow the rules that govern appeals. A noted constitutional lawyer recently got egg on his face when the D.C. Circuit rejected one of his briefs because it "evaded" the court’s page limits by having too many long footnotes. The Seventh Circuit frequently writes tart opinions about such behavior.

Judges can express their disapprobation of noncompliant counsel in even more emphatic ways. Many years before he became a Supreme Court Justice, John Marshall Harlan briefed an appeal to the Second Circuit. His brief was too long, but the clerk’s office did not reject it. When Harlan’s senior colleague, Emory Buckner, stood up to present oral argument on the appointed day, Judge Learned Hand demanded to know who wrote the brief. Buckner said that he himself had merely "put [his] name on it"; he complimented his junior colleague as the author.

Some compliment: After huffing that the brief was too long and saying he would not read it, Hand threw it over the bench. It landed on counsel table with a thud. The youthful lawyer (and future Justice) sitting there was left with a queasy stomach and a sinking feeling. If you want to avoid being pelted with your own handiwork, consult and follow the rules.

A lawyer writing a brief in the United States Supreme Court need consult only one set of formal rules: the Rules of the Supreme Court of the United States, which became effective in its current form on January 1, 1990. Those rules are clearly written and easily understood, as far as they go. Experienced Supreme Court litigators know, however, that certain Supreme Court practices do not appear anywhere in the rules.

For example: If a brief writer has cited materials in the brief and wants them readily available to the Justices, but those materials cannot be included in the joint appendix (perhaps because they are not part of the record), the Clerk usually will allow copies of the materials to be "lodged" with his office. (The materials must, of course, be served on opposing counsel.) The Oxford Companion to the Supreme Court of the United States (1992) mentions this technique, but only in the book’s discussion of the Solicitor General — as if the procedure was somehow available only to the government. In fact, it can be used by any litigant who knows to ask the clerk’s office for permission to use it.

Because this useful technique (and several others) are not in the rules, a Supreme Court brief writer ordinarily should consult the leading treatise on the nuts and bolts of Supreme Court practice. Popularly known for decades as "Stern and Gressman," the book Supreme Court Practice came out in late 1993 in a seventh edition, written by Robert L. Stern, Eugene Gressman, Stephen M. Shapiro, and Kenneth S. Geller.

The lawyer writing a brief for a federal court of appeals must consult two sets of rules. One is the FRAP. Lawyers quickly learn that is not a Boston native’s term for a milk shake, but is instead an acronym for the Federal Rules of Appellate Procedure, which apply in all federal courts of appeals.

But the FRAP is not enough. A brief writer also must study the "local rules" of the court to which the appeal is being taken. Each court of appeals has authority to supplement or modify the FRAP; idiosyncratic rules abound, setting traps for unwary advocates. For example, the FRAP sets page limits for reply briefs; but the D.C. Circuit regulates brief length by the number of words. The same court also has special procedures and unusual timetables for obtaining leave to file an amicus brief. The Fourth Circuit insists that footnotes in printed briefs meet the same minimum size requirements as text. The Ninth Circuit requires that parties submit "excerpts of record" rather than the "joint appendix" more common in other appellate courts (and discussed in FRAP 30); it directs counsel to use special citation forms for documents in the district court record. The list goes on and on.

State appellate courts also usually have detailed rules. There is so much variation from state to state that it is useless to generalize, other than to repeat the basic point: find, read, and follow the rules.

One rules-related question comes up more than any other: When a case is complicated and adequate briefing will push up against the page limits, how, within the rules, can you squeeze the text you need into the allotted number of pages? Almost all appellate courts specify in their rules the maximum number of pages and the minimum size of margins. But there is considerable variation in the degree to which courts regulate type styles and sizes. Many courts will allow proportional spacing (easy to do on modern word processors) or submission of printed briefs, without imposing a page-limit penalty. These techniques, if permitted by local court rules, enable you to expand the content of a brief by 20 percent or more while still producing an attractive, readable product.

Not all courts will countenance clever format shuffling, however. When a court does have rules governing the format of a brief, obey them or be prepared to face the music. A few years ago, a petitioner seeking review of an NLRB decision in the Seventh Circuit was denied permission to file a 70-page brief. It then resubmitted the same brief "stuff[ed] . . . into 50 pages" by "a variety of typographical techniques" prohibited by the Circuit’s rules, such as 1-1/2 spacing and type smaller than the required 11 points. "[T]he lawyers, caught with their hands in the cookie jar, . . . apologized and promised not to play the same trick . . . again." But apologies were not enough. The court forcefully expressed its disapproval of the lawyers’ conduct and imposed a $1000 penalty. You probably don’t want something like that to happen to you.

Unsubstantiated rumor has it that some United States circuit judges have requisitioned special rules that printers use to measure type sizes and margins, illustrating an amateur detective’s zeal for catching lawyers who try to squeeze too much material into their briefs. Whether or not that is true, one judge has written for public consumption his reaction to a brief that chisels on the type size: "It tells the judges that the lawyer is the type of sleazeball who is willing to cheat on a small procedural rule and therefore probably will lie about the record or forget to cite controlling authority." Encountering that attitude could be worse for you than a $1000 fine.

Let us pause for a moment and consider the standard advice for how to meet page limits without disobeying or evading the rules. It is a humorless, spartan maxim: Write short briefs; be so economical and terse that no squeezing is needed. As is often the case with standard advice, these admonitions have much truth to them, but they are not uniformly correct.

Some cases do warrant short briefs. The first case that one of us argued in the Supreme Court was a relatively simple Fourth Amendment matter; the total number of pages in the petitioner’s brief, respondent’s brief, and reply brief combined was less than the 50 pages the rules allowed for a single party’s opening brief. But that is unusual. Other cases do warrant the full number of pages allotted by the rules, or (with the permission of the court) even more.

"Write short" is not a panacea. Relatively extended treatment may be necessary because the case involves an especially complex issue or because a number of issues must be presented. In some cases, a court could follow any one of several routes to the same conclusion, and the advocate must present each logical path, not knowing which the court will take. When appellate judges lament — as they frequently do — the unnecessary length of some of the briefs they see, they may not appreciate fully that a lawyer cannot, as a judge can, simply settle on a single true path to the desired result.

Our own advice on how to meet page limits is not merely to be brief. Instead, first write lean prose that makes the necessary points and avoids excessive repetition. Then, if the brief is too long, take advantage of whatever latitude the rules provide (but no more) to vary margins, typefaces, line endings, and so on. Then edit the prose to make it leaner still.

Judges may always grumble about the length of briefs, but, if you stay within the rules and write briefs that tell them what they need to know in economical prose, they usually will come around. After flinging John Marshall Harlan’s brief at him, Learned Hand eventually voted in favor of Harlan’s client. Hand even called Harlan in to tell him it was "a very good brief."

So much for format and length. What about substance? Usually the first non-boilerplate item in an appellate brief will be something called the "Questions Presented" or the "Issues Presented" or the "Statement of Issues." This section can be critical. It is difficult to underestimate the importance of clear, effective framing of the issues: In advocacy, as in life, first impressions last. Unfortunately, many briefs state the issues in a way that either impairs the author’s credibility or confuses the court’s understanding of what the appeal is about.

Advocacy has a role in drafting the questions presented, but it is a mistake — and a common one — to slant the formulation of the issue too obviously in your own favor. Consider an extreme example: Suppose your case presents a question of whether exigent circumstances entitled police officers to enter your client’s dwelling without a warrant; the police say they acted to prevent the destruction of drugs that could be used as evidence. In such a case, you should not present a question such as "Whether the Fourth Amendment has been suspended as a result of the ‘War on Drugs.’ " You may, if the situation warrants, want to suggest to the court that the search was unr easonable and that excessive zeal in the "War on Drugs" explains the government’s behavior (and the trial court’s ruling condoning that behavior) — but save the point for the argument section. If you start out so contentiously in the question presented, the court will conclude that you are unwilling — or unable — to ever be balanced. It will cast a skeptical eye on everything else you say and assume that it is all slanted. Your credibility — a key element of a brief — will be gone.

You can preserve your credibility for formulating the issues on appeal even-handedly; but there is another challenge: You must also make the questions comprehensible. If the judges cannot understand what the case is about from the initial substantive exposure to your writing — a statement they expect to be clear — they may have far less patience with the parts of your brief that may legitimately be complex.

A good brief writer can formulate clear, neutral-sounding questions but frame them in a way that tends (subtly, of course) to suggest the answer the writer seeks. The question should not present your argument, but it should express a clear point of view about the case.

An example from one of our recent cases may demonstrate the distinction. It was an antitrust case. Our opening brief (for the appellants) stated five issues presented and did so in less than half a page. We slightly loaded one of them with what we thought were helpful facts:

Whether defendant can be labeled a "monopolist" under Section 2 of the Sherman Act because it owned the only bowling center in a small area, even though uncontradicted evidence showed that defendant lacked power to exclude competition or control price.

Our adversaries took a different approach. They heavily loaded their issues presented and took five pages of their brief to state them. The first issue presented, according to our adversaries, was:

Was the finding of the jury that [defendant] possessed monopoly power in the Antelope Valley of California ("the relevant market") supported by substantial evidence when there was evidence (a) that over time [defendant’s] share of the relevant market increased and, ultimately culminated in [defendant] achieving a 100% share of such market; (b) that two competitors of [defendant] withdrew from, and no competitors entered, the relevant market; (c) that the prices charged by [defendant] for bowling services in the relevant market were higher than those charged by [defendant] in markets where it faced competition; and (d) that because of the limited availability of bowling center and equipment financing, potential competitors confronted a significant barrier to entering the relevant market?

Sometimes these things are a matter of taste. Lawyers might differ over which of these formulations is preferable, and the decision in a case is unlikely to turn on such phrasing variations. We cannot help thinking, however, that judges tire quickly of laboriously reading such detailed Questions Presented and would prefer to see the minutiae elsewhere.

Remember, the Questions Presented section is likely a judge’s first exposure to your side of the case. It is a place to provide a concise overall view of what is at stake. It is not a place to bury a judge in detail. If judges must wade through facts, the significance of which is not immediately apparent, they may have a hard time grasping what your arguments are about.

Another key to successful appellate litigation (at least for the appellant or petitioner) is to limit the number of questions presented. Here again, there are no universal rules: Two questions presented are sometimes too many and five are sometimes too few. But it is fair to say that judges are more likely to give full attention to fewer issues than to many. An appellate lawyer must resist the temptation (and the pressure from client or trial counsel) to include many issues in the hope that, somehow, lightning will strike one of them. And it is never good advocacy to present two or more questions that simply rephrase what is really a single legal issues.

Sometimes, a succinct introductory sentence or two, or even a succinct paragraph, placed before the questions presented will aid understanding of a complex case. Most courts permit this device, although relatively few advocates use it. Here is an example:

Prior to 1983, the tax code prohibited the compounding of interest on tax deficiencies or on tax overpayments. In section 344 of the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA), Congress repealed that prohibition and provided for the compounding of all such interest. Section 344(c) of TEFRA directed that these changes would apply to "interest accruing after December 31, 1982." The question presented is whether section 344(c) authorized the compounding of appellant’s tax-deficiency interest, which had completely ceased accruing on February 1, 1982.

Elegant prose? Perhaps not. But imagine how much more inelegant it would have been to cram a single sentence with enough subordinate clauses to embrace all of these ideas. The case was technical, as were the issues. The statement of the question presented broke out the main ideas into separate sentences so that the judges could understand more easily what they were being asked to decide.

We now come to one of the few absolute — but, unfortunately, often violated — rules of brief writing: The Statement of Facts should never be argumentative in tone. The Statement of Facts is for telling the court what the case is about. The argument portion of the brief is for contention about the significance of those facts. Nothing impairs a brief writer’s credibility more than an emotional, sarcastic, plaintive, or visibly one-sided Statement of Facts.

In order words, in the Statement of Facts, understated advocacy works best. A judge will be more prepared to believe that your client should win if your statement seems objective than if it editorializes. A judge will be more inclined to accept the fairness of your statement if it acknowledges the other side’s strongest points and introduces — but does not argue — the facts or concepts you will later use to counter the other side. Remember, judges are lawyers, too, who are accustomed to careful analysis of facts and authorities. If your statement presents your case in a fair but favorable light, you do not need to carry every argument all the way to its logical conclusion at that point. You certainly need not drown the reader in rhetoric.

A closely related blunder is committed by many appellants challenging adverse jury verdicts. They fail to recognize that the evidence will be reviewed on appeal in the light most favorable to the verdict — that is, most favorable to the other side. It may be appropriate to describe both parties’ evidence, but you should never present only the version favorable to you when that version has been rejected by the factfinder.

Of course, it is essential in the Statement of Facts to describe the record accurately. An answering brief that can show that you have distorted the record, or quoted material out of context, or otherwise arguably misled the court, can be devastating. The resulting loss of credibility will — you may be sure of this — undermine the reception that every other part of your brief receives.

This does not mean, however, that advocacy plays no role in drafting the Statement of Facts. Quite the opposite. Although the tone must at all times remain neutral and dispassionate, artful selection, emphasis, and organization of facts can go far to shape a reader’s perception of the case.

The trick for the appellant is to make the reader feel that the statement presents a fair description of what happened — an account of the material facts leavened with a recognition of the presumption of correctness that fortifies the factfinder’s resolution of factual disputes — yet, at the same time, have the reader come away with the feeling that the outcome of the trial court proceedings was none too sensible or fair.

Conversely, if you are the appellee, you will try to suggest that the appellant has distorted the facts, which, when correctly described, make the trial court outcome seem fair, reasonable, and almost inevitable.

One final point regarding the content of the fact statement: Every lawyer should know, though not all honor, the rule that you are limited to stating the facts contained in the case record (even though you may believe that the record is not what "really" happened). What is less often appreciated is that the statement need not be confined to the historical facts — who did what, and when. In addition, it can introduce relevant statutes, cases, and arguments to the court, as long as it presents them in a descriptive rather than an argumentative manner. It can also set forth — though carefully — what might be called "legislative facts," even though those "facts" are not part of the trial record, if they are background facts of the kind that a responsible judge would consider in determining the appropriate legal rule. Such a submission — a "Brandeis brief" — has an honorable place in American law; it is proper as long as the line between legislative and adjudicative facts is scrupulously honored.

How long should the statement be? Recall what Lincoln said about how long a horse’s legs should be: long enough to reach the ground. A statement should be long enough to tell the judges or Justices what they need to know, and no longer. Sometimes that will mean four pages of a 50-page brief, and sometimes 20 or 25.

In a case involving a plain legal issue, a short factual account may suffice, followed by a more elaborate legal analysis. In a fact-intensive case, on the other hand — a challenge to an administrative agency ratemaking decision, for example — the statement may need to be much more elaborate. It may have to set forth in some detail the relevant statutory scheme and the structure of the particular regulated industry, followed by an account of the course of agency proceedings. In such a situation, it may then be possible, building on the factual foundation that the statement has laid, to have a comparatively short legal discussion.

In general, a reader is unlikely to grow too impatient with a statement that usefully sets forth relevant facts, even at some length. However, if the statement seems to be loaded with irrelevant detail — either because it actually is full of irrelevancies or because it is so poorly organized that the reader cannot grasp the relevance of what is being said — then it is likely to receive an unsympathetic reading.

One final point on this topic: The Statement of Facts is the place to introduce the parties and to explain any shorthand you will use to refer to them, plus the acronyms that you intend to use in the brief. Such shorthand references can help keep the writing lively, which is an important goal. Rule 28(d) of FRAP specifically advises counsel to "keep to a minimum references to parties by such designations as ‘appellant’ and ‘appellee’." That advice is but one example of a larger point.

A mostly excellent brief recently filed in the Supreme Court flirted with loss of its audience in the dense prose of the very first substantive sentence of the brief:

The issue presented in this case — which arises under the Federal Service Labor-Management Relations Statute, 5 U.S.C. § 7101, et seq. ("Federal Labor Statute") — is whether the most basic policies of that Act should play any role in a major area of its administration, viz., in determining whether a union acting as the exclusive collective bargaining representative of federal sector employees — having been selected by those employees through the secret ballot electoral processes provided by federal law — is entitled to the disclosure of personnel records of bargaining unit employees when such disclosure is "necessary for the full and proper" performance of that representative’s collective bargaining functions.

The writer of that sentence asked it to do too much. The sentence introduces too many concepts without a pause. By contrast, one of the authors a few years ago had a rare opportunity to use lively prose to make his point, which was that an Arkansas highway tax unconstitutionally discriminated against interstate commerce by exempting trucks carrying agricultural products (which were by no coincidence predominantly local), while fully taxing those carrying equally heavy shipments of other commodities (which came predominantly from out of state):

There is an old riddle: Which weighs more, a ton of feathers or a ton of bricks? While many find the question deceptive at first, the correct answer, that a ton is a ton regardless of what is being weighed, becomes irrefutably clear once explained. But in enacting and now defending the NR Exemption, the State has managed to get the answer wrong — a ton of soybeans or chicken feed is treated as though it weighs less than a ton of baked beans or dog food.

The reader is sure to understand the point and may even have gotten a smile out of the arduous task of reading a brief.

Another point that frequently crops up with fact-intensive legal issues (for example, the sufficiency of the evidence to support the verdict) is whether it is best to recite all the relevant facts in the statement or save them for the argument. There is no general rule, but be aware of this: It is permissible to mention such facts briefly in the statement and then explore them fully in the argument. Often such treatment will reduce repetition and enhance the comprehensibility of your presentation. The section is called the Statement of Facts, but that does not mean all the facts you rely on must be there.

Turning to a point of general application, you should be especially careful how you refer to the court or agency below. If you are the appellant or the petitioner, you are, of course, asking the appellate court to reverse that court or agency. The appellate court knows that. It knows you disagree with the outcome thus far. It will reverse in an appropriate case. But its initial inclination, almost always, will be sympathetic to the fellow judge who had to sit through the trial or to the agency that had to sift through the entire record now being selectively quoted on appeal.

Criticism of the lower tribunal therefore should be stated carefully and objectively (for example, "the trial court did not address the ‘waiver’ issue" or "the agency’s entire response to this argument was as follows"). Although this advice may seem obvious, lawyers do the contrary often enough that Judge Kozinski has been led to write, in a passage dripping with sarcasm: "Chances are I’ll be seeing that [lower court] judge soon at one of those secret conferences where judges go off together and gossip about the lawyers. I f ind that you can always get a good chuckle out of the district judge by copying the page where he is described as ‘a disgrace to the robe he wears’ or as ‘mean-spirited, vindictive, biased and lacking in judicial temperament’ and sticking it under his nose right as he is sipping his hot soup."

Should the brief include a summary of the argument? FRAP does not require a summary of argument. Supreme Court Rule 24.1(h) does, as do the local rules of many — but not all — of the federal courts of appeals. Even when the rules do not require a summary, it usually is a good idea to write one except in the simplest cases. And it is always essential to good appellate advocacy that somewhere in the brief — if not in a summary, then near the beginning of the argument itself or even somewhere in the statement — counsel provide an overview of the position they will be arguing. Without this, it often becomes impossible (or possible only after an amount of effort that exceeds what the judges are able to spend) to understand a litigant’s exact position. If judges do not understand a litigant’s position, they may well substitute a position that is easy to understand — but is not what the party meant and is easy to rebut.

Experienced brief writers know that the summary of argument is usually written after the argument itself. The summary ordinarily should have the same structure as the argument. In our experience, the structure of the argument tends to evolve over the course of drafting and editing. Writing out a summary before writing the argument may serve the same useful function as preparing an organizing outline, but a summary written in advance rarely will be phrased and organized as well as one written after thinking through all the ideas that come up during drafting.

Often, you may wish to begin with some background or table-setting that will not be repeated in the argument section and therefore is not, strictly speaking, a "summary" of any part of the argument. In such instances, it is perfectly legitimate to combine the summary of argument with an introduction, as long as the combination of "introduction and summary of argument" is so labeled and does not cause the section to be too long (more than four or five pages).

Supreme Court Rule 24.1(h) cautions that "[a] mere repetition of the headings under which the argument is arranged is not [a] sufficient" summary of argument. The same caution surely holds true in every court that requires a summary. But it is equally important to remember that the argument headings themselves will also serve a summarizing function. Some readers of appellate briefs do skip over the table of contents and table of authorities when they first pick up a brief, but many do not, and virtually all return to the table of contents at some point when they try to understand the structure of a brief. Thus, you should pay attention to the argument headings so that when they are plucked out from the text and stand on their own, they will comprehensively, comprehensibly, and (within limits) persuasively state the party’s position.

Ironically, the most critical section of the brief — the argument itself — is least subject to general rules or advice. There are two primary determinants of the quality of the argument section of a brief: (1) the quality of the arguments available and (2) the analytical and writing skills of the lawyers involved. Nevertheless, some aspects of writing an argument are specific to the appellate process.

First, never forget the importance of organization. It is vital to organize, not only the writing, but also the theory of the case. Appellate judges know that they are setting precedents. They therefore worry about whether the theory they adopt in one case will or will not apply appropriately to slightly different sets of facts. Appellate lawyers should assist the judges by having — and expressing — clear theories with reasonably clear limits.

Unfortunately, many appellate briefs are organized in ways that do not advance an overall theory. One common but particularly unsatisfactory form of appellate brief (whatever its merit in a trial court) is to quote snippets from one precedent after another without fitting those precedents into an overall pattern. Such filings are long on cut-and-paste, but short on logic or explanation. Likewise, it is tempting (but equally ineffective) to use a brief to take a series of potshots at the opinion below (in an appellant’s brief) or the adversary’s brief (in an appellee’s brief or a reply brief), never bothering to devise an overall theory of the correct approach to the case. And it bears repeating that ad hominem criticisms of adversaries or the decisionmakers below — as opposed to their legal positions — are counterproductive.

It also is desirable to explain the client’s position in a way that makes sense from a policy (or common sense) perspective. Judges are concerned about both the institutional and the real-world consequences of the rules they adopt. Relatively few cases that reach appellate courts are controlled so squarely by precedent that the judges have no wiggle room. Accordingly, even if favorable precedent is available and you intend to rely heavily on it, write the argument in a way that gives the judges confidence that they should follow that precedent. That is far better than baldly telling them that they must follow it — and daring them to disagree.

But be careful about policy forays. You cannot just make up the law. Most appellate judges are offended by briefs that are merely naked policy arguments and that pay no attention to such familiar judicial guideposts as case law, statutory language, and (for most judges) legislative history.

Statutory language can be especially important. More than once we have edited draft briefs that contained dozens of pages of material before ever quoting the actual language of the statute being construed. That is always a mistake in a true statutory construction case — one in which the court is called on to determine the meaning of statutory language, rather than construe precedents that have infused meaning into broad statutory generalities (such as those of the Sherman Act).

If statutory language makes your position difficult, do not hide the statute at the back of your brief. The court will see such placement as a tacit admission that the statute cannot be construed your way. The judges may think you want the court to ignore the statute. If the statutory language is favorable, you have done your client even more of a disservice by not beginning with that and telling the judges that Congress has made all necessary policy choices. Judge Kozinski, in his advice on how to lose an appeal, has written: "[S]tart out by discussing policy. . . . [I]nstead of talking about what Congress did, talk about what it should have done."

An important tactical question that often confronts the drafter of an appellant’s opening brief is the extent to which the brief should provide responses, then and there, to arguments the other side may make in its brief. Should the rejoinders be saved for the reply brief instead? Remember, anticipating arguments entails some risk, especially if opposing counsel are weak; you may put ideas into their heads that they would not otherwise discover or articulate coherently. In general, however, an appellate brief that tries to hide from the adversary’s best arguments is less effective than one that confronts them. And, when the point has already been made by the trial court or argued by your adversary at earlier stages of the proceeding, you cannot expect to hide. You will almost surely want to address such hard points in the opening brief, stating the issue in your terms rather than letting your opponent set the agenda.

Content is not everything, of course. Writing also matters in an appellate brief and in the argument section especially. The point is not that judges consciously grade style or decide appeals based on which brief they think is better written. Rather, it is that judges must understand and remember your position before they can agree with it — and a stylish brief usually is more understandable and memorable.

Appellate judges are busy people. Judge Kozinski estimates that he must read 3500 pages of briefs a month. There is not always sufficient time for a judge to untangle convoluted sentences or dense prose. In addition, typographical and grammatical errors can distract from more important matters. And, if it is possible to write the brief in a lively fashion — without making the writing style itself a distraction — the reader is likelier to comprehend and remember it.

Here again, heated rhetoric and overstatement are harmful. Perhaps the most common flaw in appellate briefs is writing in emphatic, unequivocal, and conclusory terms. Such briefs, overconfident, even cocky, in tone and uninformative in content, are likely to obscure what the judges must really decide and what analytical steps are needed to reach a sound decision — especially if the weakness in the argument has been glossed over in an effort to make the position seem stronger than it is. This is not only unhelpful to the court, but injurious to the advocate’s own cause. It is far better to confront the issues coolly, honestly, and logically, guiding the reader lucidly down a path that leads to victory.

Tone matters too. In a recent, highly publicized criminal case, appellate counsel did a masterful job of identifying the issues and mustering legal and factual support for his client’s position. He did so, however, in a self-righteous tone, overstating accusations of prosecutorial misconduct, belittling the trial judge, and portraying his client as the victim of a person who, the jury had found, was herself the victim of the client’s serious criminal conduct.

The lawyer, who is prominently affiliated with an elite East Coast institution, should have been careful to adopt a respectful tone toward the midwestern state judges he was addressing. We read the briefs before the case was argued. We concluded that, if the judges thought the issue otherwise close, human nature probably would make them want to rule against the defendant because of his lawyer’s imperious tone. We are not mind readers, but we do know that the defendant lost on appeal by a 2-1 vote.

The brief of an appellee or respondent — the "bottomside" brief, in the jargon of appellate practice — has certain special features. The bottomside brief writer has the disadvantage of not being able to introduce the judges to the case and the issues; they will read the topside brief first. But there are advantages too. The party filing second has a target to shoot at: the appellant’s brief. And, except in cases involving cross-appeals, the bottomside writer has prevailed below on all of the issues before the appellate court; that litigant has the advantages that flow from having already had one decisionmaker agree with its position. The bottomside party wins if the decision below was right on the merits or if appellate issues were not preserved below.

The first item on the checklist of the writer of the bottomside brief should be to ask: Was each of the arguments now being raised on appeal properly preserved below? Were alleged instructional errors properly objected to? Were the grounds now advanced for overturning evidentiary rulings the same ones offered in timely objections at trial? In a related vein, is the legal theory urged on appeal the same one presented to the trial court and, if not, is there an advantage to be gained from the change?

Appellate rules usually give the bottomside brief writer the option of dispensing with several of the features required in the topside brief. There is rarely a need to repeat or correct the predictable recitations of the basis for jurisdiction and the nature of the rulings below. On the other hand, it is usually worthwhile in a bottomside brief to reformulate the questions presented and write a competing statement of the case.

Some appellees seem to feel compelled to go further and to tell the court at the outset that the other side has misstated the questions presented and tendered a slanted version of the facts. That can be a bad idea. The court often will know from reading your questions presented and your statement that you believe the other side’s version is either inaccurate or incomplete; you waste space and possibly goodwill by adding another sentence with an accusatory tone. However, if you can demonstrate flat out distortion, and it concerns something important, do so.

The other major difference between a bottomside and a topside brief is that the writer of a bottomside brief already knows exactly what arguments are being made on behalf of reversal. It is therefore appropriate and — because the appellee gets no reply — necessary to take on those arguments. This does not mean, however, that the brief should consist simply of a point-by-point refutation of each of the appellant’s arguments. The aim of a bottomside brief is not just to debate the other side. There also must be an affirmative and coherent statement of the reasons why the decision being appealed is correct.

Having prevailed below can also have its burdensome features. Sometimes, to put it bluntly, the decision below is bad. It may be difficult to defend in whole or in part. The topside brief will have mercilessly laid bare its central defects. The bottomside brief writer then must offer other ways to reach the same result. Occasionally, it may be wise to abandon the lower tribunal’s reasoning and substitute a different and better rationale. In essence, the appellee ends up defending, not the opinion that was written, but the opinion that should have been written.

Most times, however, it is prudent to defend the lower court’s approach and offer, in addition, either something explicitly called an "alternative" approach or an embellishment on the decision below. Of course, there are times when defending the rationale of the decision below will be the only way to secure an affirmance. In cases coming from administrative agencies, for example, the appellate court is not allowed to adopt a rationale that was not the basis of the agency’s decision; similarly, a court reviewing a jury verdict may not affirm on a basis never presented to the jury.

The recent decision in TXO Production Corp. v. Alliance Resources Corp., 113 S. Ct. 2711 (1993), represents a triumph of the tactic of presenting a new and better rationale on appeal. The highest court of West Virginia had upheld an award of punitive damages that was many times the compensatory damages; the award was therefore greatly out of proportion to the actual harm suffered by the plaintiff. The state court had opined that the case implicated no federal constitutional limit on the size of the punitive damages on the dubious ground that the defendant had been "really mean." The Supreme Court’s grant of certiorari suggested likely dissatisfaction with the West Virginia court’s rationale.

Then respondent’s counsel went to work, scouring the record and discovering a theory that, although it might have been barely hinted at before the jury, seemed more likely to persuade the nine Justices in Washington: The potential gain to the defendant from its alleged misdeeds much more closely approximated the punitive damages award than the amount tha t the plaintiff actually lost. Because counsel advanced this theory in the bottomside brief (and argument) and showed to the satisfaction of the necessary number of Justices that it had been preserved below, the West Virginia court’s judgment (but not its reasoning) was upheld.

Finally, what about reply briefs? They are optional, but it is the rare case — if any case at all — in which it makes sense to forgo the opportunity to file one. One of us once argued on behalf of the government a Supreme Court criminal case in which the petitioner simply did not bother to file a reply brief. The Court decided the case 5-4 in the government’s favor, with the unusual coalition of Justices Brennan, Marshall, Scalia, and Kennedy dissenting. It would be easy to believe that the government’s sterling written and oral advocacy assured the result no matter what the other side did, but one must wonder whether an effective reply might have swayed one of the Justices who formed the tenuous majority. It is a mystery why counsel passed up the chance to have the last word in such a close case.

The reply brief must be (relatively) short, (relatively) punchy, and selective. Sometimes it will follow the same structure as the opening brief, but sometimes it will not. What it must do, to be effective, is identify from the start one or more overall themes in the argument or arguments with the best chance of winning and explain to the court where the appellee’s brief, which it just read, went fundamentally astray.

The function of a reply brief is to respond to an adversary’s arguments. The court can look back to your opening brief as a reminder of the overall structure of your argument and to answer nagging questions. It is therefore usually unnecessary to retrace all the steps of your logic in the reply brief, and it is far more acceptable in a reply than in an opening brief to concentrate on sharply focused (but polite) debate. Sometimes, however, your adversary may have confused things so much that re-emphasizing the structure of your arguments will be the most useful thing to do in reply.

If you must put a rhetorical flourish somewhere in your briefs — and sometimes that may be useful — the beginning or end of the reply brief is the place to put it. Rhetoric turns appellate judges off when they see it as a substitute for analysis. By the time they read your reply brief, however, the judges should know that you are prepared to analyze — and have analyzed — the issues fully. Having, in a way, paid your dues, you have more leeway for a catchy phrase or metaphor at the beginning of the reply brief. This may help dramatize the central defect in the adversary’s brief, which the judge will have just read; such a phrase at the end of the reply brief may be the last word the judges read before they put down their papers.

Do not strive to write a pithy ending for its own sake, however. Litigation gives its authors and editors a style sheet that advises: "Formal conclusions are not worth the trouble. Start at the beginning, go to the end, then stop." The same goes for reply briefs.