Chapter 1. The Appeal as a Clock

The sixteen records about appeals never ask whether the decision below was wrong. This chapter shows what they ask instead: whether, and how far, a judgment is final while an appeal is possible or under way, what stops in the meantime, which act starts the clock, and where the record lies. It adds to the debate over what appeals are for a genre in which the time of appeal is the whole subject, and it shows how the records turn that time into open questions without answering them.

1.1Data and method

The data are PRO SE's public library as it stood when it was read. The library list gives 483 questions in four sections: 122 on scope, 121 on documents, 120 on timing and 120 on the courtroom. All 483 are marked synthetic. The list carries the notice "General information, not legal advice" and an editorial reference date, which it says is not a case date and does not confirm current law. Each question's page data were fetched and kept as snapshots, which agree word for word with the library's English text. The snapshots hold, in English and Korean, five fields that this study reads: the question; the evidence, which says what material is examined; the readings, always two; the judgment, which in this library is itself a question; and the revisit condition, which says when the matter would be looked at again. The English text is PRO SE's own; nothing was translated.

The selection rule is one sentence: every question in the Timing section whose English question contains the word appeal, appealed or appeals. The rule gives 16 records. It is lexical and confined to one section so that the records share an angle of view, the calendar of a case. The library has 24 questions with the word in all. The rule leaves out 7 in the Documents section and 1 in the Scope section; 5 of those 8 ask about grounds of appeal, the facts found below, a breach of law or what the appeal court decides. It also leaves out 9 Timing questions that name a retrial, a challenge or an objection without using the word. Because the rule draws on the Timing section, an emphasis on time is partly built into the selection; what the records do with time is not, and that is what this study reads. The sixteen records run to 2375 English words, 148 on average, with 32 readings in all. All sixteen carry the source tags court and law, none has a publication date of its own, and 14 say in their evidence field that they are practice materials.

Some properties belong to the studio's form and would be found in any record. Every one of the 483 has exactly two readings, every judgment is written as a question and every record has a condition for looking again. These are the library's shape, not findings. The sixteen keep the library's most common phrasing more often than the section does: across the Timing section, 86 of 120 records (72%) open both readings with "One interpretation holds" or a close variant, and here all 16 do; across the section 79 records (66%) open their condition for looking again with "The matter is looked at again" or the same words with analysis in place of matter, and here 14 do. To see how much of the wording is repeated phrasing, the study measured, for each field across all 483 records, the share of five-word runs that recur in at least 25 records. In the sixteen the share is 0% in the questions, 2% in the evidence, 3% in the readings, 1% in the judgments and 16% in the revisit conditions. By the 30% threshold this series of studies uses, no field here is formula, though the revisit conditions come closest.

Every count in this paper comes from a script that reads only the snapshots: whether a record names an error or the merits, finality, a stay, service or notice, the case record or a level of court; whether its two readings divide over counting together or apart; what its condition for looking again turns on. The counts describe the records, not courts or parties. This study draws no fact about any law, court or person from them, and nothing in it is advice about whether, when or how to appeal. What it studies is the frame.

1.2The judgment held open

The first finding is an absence. No record among the sixteen names an error, a mistake, a wrong decision, the merits of a case, a breach of law, the facts found below or a reversal. Four mention a statement of grounds, but only as a document with a period or a service; none says what grounds might be. Only three name the parties' setting, a debtor company with suppliers, and one calls it a payment dispute; none says what was decided between them or why anyone might think it mistaken. The decision below is present in every record and examined in none. Each record is a synthetic practice material, published by PRO SE as general information, not legal advice, and the absence belongs to that genre. It is still a choice of frame, and a different one from the debate's: what the records hold up for inspection is not the content of the judgment but its state.

The state that matters most is finality. Five records name a judgment becoming final, and three ask about it in the question itself. The record on a disputed withdrawal of the suit is the sharpest. A suit has been withdrawn after the final judgment, the withdrawal is challenged, and the question is what happens to the period for appeal. The second reading says:

The sentence holds two judgments in a chain. One judgment's finality waits on another's, and the period for appeal is measured from a starting point that does not yet exist. The first reading answers differently, that the original period keeps running while the dispute continues and runs anew once the second judgment is final. Neither says anything about what the original judgment decided. The record's question in place of a judgment then names the condition under which the decision would have to be made:

The phrase "on the record as it stands" carries the frame of the whole set. The judgment's finality is read off the record at a moment, and the moment is not yet settled. The record leaves the question open.

Finality can also be divided. The record on part of a judgment asks whether the rest becomes final when only part is appealed. Its two readings put the division and its limit side by side:

Here the appeal does not reopen the decision; it cuts the judgment into a part that may be closed and a part that stays open, and a possibility not yet used, the other side's cross appeal, keeps the line between them uncertain. The judgment that would follow asks: "While a cross appeal is still possible, how far should the final part be taken to extend?" Its revisit condition looks to whether the case record moves to the higher court. Finality, in this record, has a size that changes with time and with the record's location.

The record on withdrawal or waiver makes the same point through two acts that end an appeal. Its readings are not rival accounts of one thing; both open "One interpretation holds", and each describes a different act:

Both sentences end in the same words, "the judgment becomes final". What divides them is the act and its moment, and the judgment the record poses accordingly asks only for facts of time and place: "Which of the withdrawal and the waiver was filed, and when, and in which court was the record at the time?" An appeal here is a state a case can leave by two doors, and the record's interest is in which door, when, and with the file where.

If the appeal holds a judgment open, the next question is what else stays open with it. Three records name a stay, a halt or a stop, or execution or performance going on, and the same three name enforcement, execution or the performance of a plan. Their first readings point the same way: stopping what the judgment set in motion takes something beyond the appeal. The record on enforcement pending appeal concerns a judgment declared provisionally enforceable; its first reading holds that a way opens to halt enforcement if the appellant applies for a stay and makes a prima facie showing, and its second reading adds a limit:

The record on an approved rehabilitation plan says, in its first reading, that the plan takes effect as soon as the approval decision is made "and that its performance continues as a rule even if an appeal is filed." The record on an appeal or objection against a ruling of the execution court says that execution goes on "unless a stay order or an interim measure is made." In all three the appeal and the stopping of what it attacks are separate things, and a second act, an application, an order, is needed to join them. The question in place of a judgment in the first of them asks exactly that:

Read beside the debate, these records show something the models of error and oversight do not need to show. Shavell's reply to Schwartz, Reply to a Comment on "The Appeals Process as a Means of Error Correction" (1996), answers criticisms of the assumption that the appeals judge decides only on what happened at trial; this study reads that assumption as one that puts the life of the judgment between trial and appeal outside the model's frame. In the records, that interval is the subject. The judgment below is neither confirmed nor corrected; it is held, partly or wholly, in a state between open and closed, while enforcement may go on beside it. Spitzer and Talley, Judicial Auditing (1998), describe review as a higher authority auditing a lower one. The records show the audited decision continuing to act on the parties while the audit is pending, and they make the question of whether it should stop a separate one, for the court to order.

1.3The clock, the record and the levels of court

The second finding concerns what the clock runs from. Twelve of the sixteen records name service, notice or a ruling being made known, and nine name a starting point, counting, a period running or a period running anew. A period, in these records, is never only a length; it is a length from some act, and in the records read below the act is the court's: a service, a notice, a sending of the record. The record on the court's delay asks whether a party's appeal period is delayed when service by the court is delayed. Its readings could hardly be shorter:

Two clocks are in view, the court's and the party's, and the readings divide over whether they are linked. The record's question in place of a judgment then drops the dispute and asks for two facts: "When do the actual service and the notice each take place?" The record presents the time of appeal as tied to acts done by others, at moments that have to be found out.

That division, together or apart, recurs. In eight records the two readings divide over whether two periods, two acts or two effects are counted or judged together or separately. The record on two notices, about a final appeal, sets out the alternatives in their plainest form:

The record on an order to correct a defective notice of appeal stages the same choice between the time for appeal and the period set for the correction. One reading holds that a defect corrected within the period set for correction saves the notice of appeal "even if that period ends after the time for appeal"; the other holds:

The question that follows asks "By reference to what point in time" it is decided whether the notice of appeal escapes dismissal. Fourteen of the sixteen questions in place of a judgment name a point, a time, a start, when, before or first. Three ask what should be asked, settled or looked at first: the record on skipping a level asks what must be asked first before a written agreement to skip the intermediate appeal is given effect; the record on two statements of grounds asks what should be looked at first to tell apart how a case ends when no statement is filed and the exceptions; and the record on a re-appeal asks:

The question is about the order of questions. The appeal reaches the reader as a sequence of determinations, each of which must be made before the next can be, and none of which concerns whether the ruling under challenge was right.

The record has a place as well as a time. Seven records name the case record or an appeal record, where it is, or its sending or moving to another court. The record on a respondent's turn on final appeal asks when the respondent files an answer and until when it may file a cross appeal; its second reading ties the clock to the form of the higher court's hearing:

The record on withdrawal or waiver, read above, ends its condition for looking again with the most complete reset in the set: "If the location of the record changes or a new withdrawal or waiver is filed, everything is checked again from the start." The record treats where the file lies as a fact to check, and a change in its place sends the whole check back to the start.

The levels of court appear in five records, and they appear as differences in time. The record on two statements of grounds asks whether the periods for the statement of grounds on appeal and on final appeal are treated alike. Its second reading says:

Cameron and Kornhauser, Decision Rules in a Judicial Hierarchy (2004), recall that, with litigants fully informed after trial, the best hierarchy under reasonable conditions has three tiers, and weigh hierarchies by the errors they reduce. The records name tiers too: an intermediate appeal that parties may agree to skip, a final appeal, a re-appeal against a ruling of the appellate court. This study reads the record above as locating the difference between tiers in what happens when a period is missed, not in what each tier is for. Jung, Appellate Review of Judicial Fact-Finding Processes and Decisions (1997), argues that appellate control is built on a particular procedural structure and shaped by the legal instruments available. The respondent's turn and the two statements of grounds show a structure of that kind from below, as the reader would meet it: a court that may sit without oral argument, and a period that allows extension set apart from the period on final appeal.

The parties appear in the same register. Three records name the other side, a respondent, a successor or the main appeal. The record on a cross appeal asks when one can be filed after the appeal period has passed, and its two readings, both opening "One interpretation holds", describe two different cases rather than two rival views:

Read together, the two sentences present the other side's appeal as a condition of one's own: what one party may still do is made to depend on what the other has done and when. Daughety and Reinganum, Appealing Judgments (1998), model a losing defendant who chooses whether to appeal and a court that learns from the choice. The records show a choice whose availability itself depends on time and on another party's acts, and which they describe without saying which way it should go.

What reopens the matter completes the picture. Fifteen of the sixteen conditions for looking again turn on a record, a service record, a date, a notice, a filing or a ruling coming to light or changing, and twelve name a change. None turns on the merits, an error or what was decided below. One condition reopens the matter when a calculation becomes contested:

What would make PRO SE look again at its own question is, in every case but one, a ruling, a filing, a notice, a date or the record coming to light or changing, and in that case it is a dispute about counting. The records place the appeal's uncertainty in the calendar and the file, not in the judgment.

This section adds to the debate a description of the appeal from inside its clock. Shavell, On the Design of the Appeals Process (2009), asks when an appellant's chance of success justifies review and which door to review fits. The records stand before that question: they ask whether the door is still open, from what moment it was opened, and in which court the file now lies. They leave each of those questions open too.