Summary
The attribution of the burden of proof differs from the degree of proof, and the relevant precedent held that the plaintiff bore the burden of proving that the cause of the fire lay in the negligence of the company's employees.
2.1The Standard by Which the Path Opens
The mere conclusion that the evidence is insufficient does not determine who bears that disadvantage. Proof of fact in civil matters follows not the certainty of natural science but the standard of high probability spoken of in precedent. For a state that fails to meet this standard to lead directly to defeat, it must be determined both which requirement shall be regarded as unfulfilled and who shall bear the disadvantage of that non-ascertainment. Accordingly, what is asked now is which requirement fails when the fact remains ultimately unascertained, and what the standard is by which the passage at that stage opens. The discussion that follows proceeds on the premise that the question of measuring the degree of proof and the question of fixing the person to whom the disadvantage is attributed are different questions.
The precedent in question holds in its Summary of Decision that "the burden of proving that the cause of the fire lay in the negligence of the company's employees was on the plaintiff" (Supreme Court Decision 2008Da6755 decided October 28, 2010). Here, unlike language asking whether facts have been adduced sufficiently to admit a finding of negligence, the words "burden of proving" indicate a rule of allocation designating which party bears the disadvantage if the fact is not established. The party designated by the language is the plaintiff, and the fact in question is the requirement fact that the cause of the fire lay in the negligence of the company's employees. The scope read in this section is limited to the designated phrase of the Summary of Decision.
What the above precedent confirms is that the attribution of the burden of proof is not the same thing as the degree of proof. The former is the question of how to treat the gap to whose disadvantage when proof of a given requirement fact is ultimately absent, and the latter is the question of whether the evidence presented has reached high probability. These two must be distinguished. Meanwhile, Civil Act Article 750 provides that "a person who, by an unlawful act committed intentionally or negligently, causes harm to another is liable for damages" (Civil Act Article 750), establishing a structure in which the party seeking to establish liability for damages must satisfy the requirements of intent or negligence, unlawful act, harm, and causation. Accordingly, the entrance to this remedy opens when the performance, parties, and conditions required are specified in the claim statement, the right of the selected form of action can be defended, and at the same time the question whether the requirement facts are proved and the person who will bear the risk of their non-ascertainment are exactly connected.
The strongest counterargument arises from the standpoint of fairness. The claim that it is always fairer to place the burden on the party holding the materials points out that if the party with the materials remains silent, the ordinary rule of burden of proof entrenches substantive inequality between the two sides. The loss identified by this counterargument is real. Where access to information is imbalanced, formally identical requirements of proof may leave harsh consequences only on the side without information, and that result in turn fixes the position of the side lacking materials. Such disadvantages must not be treated lightly. Nevertheless, the mere fact of being the possessor of evidence cannot transfer every statutory burden. The attribution of requirement facts must be assessed together with the structure of the requirements set by law, and cannot be uniformly altered on the single circumstance of holding the materials.
According to the synthetic scene, plaintiff A's explanation of the cause was not sufficiently supported, but defendant B's alternative explanation was not proved either. In this scene, all that can be adjudicated is the fact that neither explanation sufficiently supported the requirement facts. Plaintiff A's failure to prove the cause falls within the case where proof of the requirement fact is ultimately absent, but that does not mean that defendant B's alternative explanation is accepted as true. What is to be recorded in the adjudication record, however, is the requirement-by-requirement allegations, evidence, statutory presumptions, findings of fact, and assessment of the burden of proof. Since no year, amount, or deadline start date is set in this scene, no conclusion premised on such figures can be drawn. Moreover, even with the same non-ascertainment, a different applicable rule of burden of proof alters the attribution of the risk of defeat.
The conclusion this section permits is that when the fact remains ultimately unascertained, the requirement that fails is the requirement fact exactly connected to the attribution of the burden of proof, and that the passage to this remedy opens when the specification of the required performance, parties, and conditions and the connection to the allocation of the burden of proof are both in place. Accordingly, conclusions not permitted are to view the defeat of a particular party as immediately fixed merely because the evidence is insufficient, and to view every statutory burden as passing to the other party solely because one held the evidence. The state of insufficient materials remains undetermined, and the same is true where the person bearing that risk has not been identified.
2.2The Conditions That Block Relief
The question now is the conditions under which the remedy passage closes. That is, where a core requirement fact is ultimately unproved and the risk of that non-ascertainment is borne by the claimant concerned, it must be determined which statutory requirement fails so that the remedy passage closes.
The relevant precedent, addressing the proof concerning the cause of the fire, holds that "the burden of proving that the cause of the fire lay in the negligence of the company's employees was on the plaintiff" (Supreme Court Decision 2008Da6755 decided October 28, 2010). Here the party is the plaintiff seeking damages, the act is the proof of the fact that the cause of the fire lay in the negligence of the company's employees, and the condition is the circumstance that the responsibility for that proof was placed on the plaintiff. This is language read limited to the Summary of Decision, and does not draw in the reasons for judgment or the detailed facts of the case.
The same sentence places two determinations in different positions. The first is the degree of proof, namely how well the plaintiff's explanation was supported; the second is the attribution of the burden of proof, namely who bears the risk of that non-ascertainment when proof is ultimately not made. The burden of proof is a rule assigning the disadvantage borne when a fact is unascertained. On the premise that the attribution of these two must be distinguished, the author takes into account the language that, in the case of tort (unlawful act), "a person who, by an unlawful act committed intentionally or negligently, causes harm to another" bears liability. Civil Act Article 750. Accordingly, if the requirement fact of negligence remains as a risk of attribution on the claimant's side, at least one of the statutory requirements is unfulfilled and the remedy passage closes.
The strongest counterargument is the claim that it is always fairer to place the burden on the party holding the materials. If the party with the materials remains silent, it is said, the ordinary burden of proof entrenches substantive inequality. This discomfort is real, and in cases where information is concentrated on one side, the side without materials can hardly even put forward an explanation. Yet the author makes clear that this equitable demand cannot be accepted in full. The loss identified by the counterargument is left intact, but that loss alone does not justify a transfer beyond the statutory requirements.
This fact alone does not specify which requirement failed. Only where the rule of attribution places the responsibility for proving negligence on the plaintiff, in a structure like the above precedent, does A bear the risk of that non-ascertainment. Matters for which it is not specified which fact is unascertained, or who bears that risk, are left undetermined so far as the records remain.
The conclusion this section permits is limited to the point that, where a core requirement fact is ultimately unproved and the risk of that non-ascertainment is borne by the claimant concerned, the remedy passage closes. What further explanation would suffice, or whether a separate requirement must run in parallel, is not determined here. It is taken as a boundary, however, that the mere fact of being the possessor of evidence cannot transfer every statutory burden.
2.3The Boundary That Remains When There Is No Data
When evidence is insufficient, the disadvantage appears to fall as it is upon the party lacking proof, but viewing it that way alone does not explain why that disadvantage is attributed to that party. What this section addresses is precisely what the ground of that attribution is, and further how to deal with the state in which, for want of materials, the fact itself cannot be ascertained.
The original text on this issue reads, "along with the reason that the burden of proving that the cause of the fire lay in the negligence of the company's employees was on the plaintiff" (Supreme Court Decision 2008Da6755 decided October 28, 2010). Here "burden of proving" is not a matter of degree meaning how far until the court can ascertain the fact; it is a matter of the rule designating the party who must prove that fact. This language designates the plaintiff as the party to prove that the cause of the fire lay in the negligence of the company's employees.
What the above language shows is that the attribution of the burden of proof is a separate stage of determination, distinct from the degree of proof. Here the burden of proof means the rule bearing the disadvantage when a fact is unascertained. Whether high probability has been reached is a judgment of degree, whereas who bears the non-ascertainment of a fact that has not reached that degree is a question of attribution under a rule. Only by not mixing these two can one accurately grasp what burden this precedent placed on the plaintiff.
Even so, the counterargument that it is always fairer to place the burden on the party holding the materials is persuasive. The concern that if the party with the materials remains silent the ordinary burden of proof entrenches substantive inequality cannot be avoided. Still, the mere fact of being the possessor of evidence cannot transfer every statutory burden. The attribution of the burden of proof is determined not merely by the location of the materials but in accordance with the structure of the requirement facts, and to transfer on the fact of possession alone is to lose the ground of that attribution.
In this scene where neither explanation has been ascertained, the lack of materials is recorded only as a lack of materials and is not, in itself, reduced to the success or failure of any party. Only where the proof status of the requirement fact and the person to bear the risk of its non-ascertainment are exactly connected is the attribution of that risk determined. Matters for which it is not specified which fact is unascertained, or who bears that risk, are left undetermined, as distinct from impossibility.
Ultimately, what this section permits is an attribution judgment in the case where the party to bear the risk of non-ascertainment has been specified; what it does not permit is a method of immediately fixing the attribution of defeat on the mere existence of insufficient materials. Leaving matters for which the materials of determination are unspecified as undetermined is not a reservation of adjudication, but an accurate noting of the state in which the ground of attribution does not yet exist.
2.4The Records Needed for Determination
On the premise that proof of fact in civil matters follows not the certainty of natural science but the standard of high probability spoken of in precedent, whether proof has reached that standard and who bears the disadvantage of ultimately failing to reach it are divided questions. Accordingly, if the preceding stage discussed the degree of proof, here it must be asked which requirement fails when the fact remains ultimately unascertained, and what records are needed for that determination.
As the original text responsive to this question, Supreme Court Decision 2008Da6755 decided October 28, 2010 contains the phrase "for the reason that the burden of proving was on the plaintiff." The party designated by this language is the plaintiff, the act in question is the proof showing that the cause of the fire arose from the negligence of the company's employees, and it presupposes the condition placing the responsibility for that proof on the plaintiff's side. What the precedent specifies is not the degree of proof or the quantity of evidence, but to whom the burden of proving is attributed. Civil Act Article 750 provides that "a person who, by an unlawful act committed intentionally or negligently, causes harm to another is liable for damages." The purpose of reading the provision here, however, is limited not to explaining all the requirements for the constitution of liability for damages, but to confirming to whom the proof of that requirement is entrusted in a claim that takes negligence as a requirement fact.
Connecting this ground with the conclusion, the stage determination is made only on the set records. The Records here mean the requirement-by-requirement allegations, evidence, statutory presumptions, findings of fact, and assessment of the burden of proof; this fixes the type, and does not refer to physical records collected in any particular case. The reason these records are used to determine the stage is that the judgment asking the degree of proof and the judgment fixing the person who will bear the disadvantage of non-ascertainment are different questions. If the former is a matter of distinguishing whether high probability has been reached, the latter is a matter of fixing to which claim the effect of non-reach is attributed when the standard is not reached. If the two judgments are treated as the same, the non-establishment of the fact and the allocation of disadvantage to a particular party become mixed. Not filling in absent records with conjecture likewise flows from the same distinction.
One who opposes this approach says that it is always fairer to place the burden on the party holding the materials. The discomfort targeted by this counterargument is real. Where the side holding the principal materials in a lawsuit keeps silent, the other side may end up bearing the risk of defeat without obtaining material for rebuttal, and formal allocation of burdens appears to leave that gap as it is. The conclusion nevertheless stands. The inequality identified by the counterargument is a consideration that must be resolved, but the fact of possession alone cannot be taken as a ground to overturn the whole of the already established rules on burden of proof.
Viewed in relation to this scene, plaintiff A's explanation of the cause was not sufficiently supported, but defendant B's alternative explanation was not proved either. Placing these two facts in the records confirms the state in which the proof of the plaintiff side's requirement facts did not reach high probability, while at the same time the defendant side's alternative explanation does not remain as an ascertained fact. This determination is described only in the light of the records of requirement-by-requirement allegations, evidence, statutory presumptions, findings of fact, and assessment of the burden of proof. Since the scene fixes no year, amount, or judgment outcome, success or failure and actual recovery results cannot be ascertained. Nor does the scene state who obtained which materials by what route, so a discussion of burden transfer premised on the fact of possession by the holder of evidence remains an item that cannot be tested on this scene alone.
The conclusion this section permits is that the stage is determined only on the set records, and that absent records are not filled in with conjecture. Conversely, it is not an admissible determination to assert outright that some requirement failed without specifying the records that fix the attribution of the burden of proof, or to regard statutory burdens as transferred on the sole fact of holding evidence. Insufficient materials are a circumstance of indeterminacy and do not lead immediately to inability to obtain relief; whether it is effective is left as a matter to be confirmed when the attribution of the risk of defeat arising from unascertained facts is exactly connected with the proof status of the requirement facts and the person who will bear the risk of their non-ascertainment. The contrast condition that, even with the same non-ascertainment, a different applicable rule of burden of proof alters the attribution of the risk of defeat tests this boundary.
2.5Circumstances That Change the Decision
In this scene, if neither explanation has been confirmed, which requirement fails when the facts remain ultimately undetermined? Moreover, what change would reverse the existing stage determination? Because the state of insufficient evidence alone does not determine whether the plaintiff or the defendant bears that disadvantage, it is necessary to examine separately who bears the uncertainty.
There are two textual grounds for answering this question. First, Civil Act Article 750 provides: "A person who has caused damage to another by an unlawful act arising from intent or negligence shall be liable to compensate for that damage" (Civil Act Article 750). This language merely indicates the requirements for liability for damages to arise; it does not directly determine who must prove those requirements. Supreme Court Decision 2008Da6755 decided October 28, 2010 cited, in its gist, the reason that "the responsibility to prove that the cause of the fire lay in the negligence of the company's employees rests with the plaintiff." Here, the act of placing the cause of the fire in the negligence of the company's employees is the constituent fact, and the plaintiff is designated as the party who must prove whether that act occurred.
At this point, the degree of proof and the allocation of the burden of proof must be distinguished. The former concerns the extent to which the evidence presented supports the constituent fact; the latter is the rule governing who bears the disadvantage when support ultimately fails. What Supreme Court Decision 2008Da6755 decided October 28, 2010 assigned to the plaintiff was not an evaluation that the proof was insufficient, but the burden of proving the constituent fact of negligence. Only once the allocation of this burden is fixed can one say which requirement failed in this scene, where the plaintiff's explanation of causation was not sufficiently supported. The mere fact that the evidence is insufficient does not identify which requirement failed, or why.
The strongest counterargument against this allocation is that it is always fair to place the burden on the party that holds the materials. It is also clear that when the party holding the materials remains silent, the ordinary burden of proof entrenches substantive inequality. Yet one cannot shift every statutory burden merely because a party possesses the evidence. The inequality the counterargument identifies does not disappear, but what remains is that that inequality alone does not change to whom the proof of a given requirement is allocated in a lawsuit. A shift of the burden occurs only when a rule prescribed by law operates separately.
The fact placed in the synthetic scene is a single one: neither explanation has been confirmed. This scene merely shows a state of uncertainty; it does not contain any change that would alter the allocation of the burden of proof. If it is confirmed that the applicable burden-of-proof rule has changed, the allocation of the risk of losing changes accordingly. Thus, from the scene alone, it remains unspecified who bears that risk. The stage determination is completed only when it is ascertained, through the connection, which constituent fact is undetermined and who bears the risk of that uncertainty.
What this section permits is to examine separately the state of uncertainty and the allocation of that risk; what it does not permit is to fix the loss of either Plaintiff A or Defendant B on the sole ground that the evidence is insufficient. Items that cannot be determined remain undetermined, but the boundary is maintained that, once a designated change is confirmed, the existing blocking and effect determinations are discarded and the determination is redone from the stage itself.
2.6The Strongest Counterargument and the Answer
In this scene, the counterargument is raised that it is always fair to place the burden on the party that holds the materials. This examines what answer that counterargument gives to the question of which requirement fails when the facts remain ultimately undetermined.
The precedent in question states which side bears the burden of proof for the explanation of causation. Supreme Court Decision 2008Da6755 decided October 28, 2010 describes it as "on the ground that the responsibility to prove that the cause of the fire lay in the negligence of the company's employees rests with the plaintiff." The party fixed by this language is the plaintiff; the act fixed is the proving that the cause of the fire lay in the negligence of the company's employees; and the allocation of that responsibility is fixed upon the plaintiff. The determination of allocation is treated as a matter separate from whether the plaintiff actually held more evidence.
Connecting this language with the conclusion yields the distinction that the allocation of the burden of proof differs from the degree of proof. Supreme Court Decision 2008Da6755 decided October 28, 2010 held that the responsibility to prove that the cause of the fire lay in the negligence of the company's employees rested with the plaintiff. The degree of proof does not determine which fact a rule applies to and upon which side; the allocation determines the direction of that disadvantage. Accordingly, if the two are mixed, one cannot read which requirement fails.
The counterargument that it is always fair to place the burden on the party that holds the materials targets the loss that, when the party holding the materials remains silent, the ordinary burden of proof entrenches substantive inequality. Such inconvenience and inequality cannot be ignored. What determines the shift of the burden is not the fact of possession but the rule of the burden of proof.
In the scene, the explanation of causation of Plaintiff A was not sufficiently supported, and the alternative explanation of Defendant B was not proved either. The allocation of the burden of proof differs from the degree of proof, and the precedent in question held that the responsibility to prove that the cause of the ignition lay in the negligence of the company's employees rested with the plaintiff. The per-element assertions and evidence, legal presumptions, findings of fact, and determinations of the burden of proof are the records. Matters that are unspecified as to which fact is undetermined or who bears that risk remain undetermined.
The conclusion this section permits is that the allocation of the burden of proof is distinct from the degree of proof, and that the language of the precedent allocating the burden of proof to the plaintiff is connected to a fixed conclusion. The conclusion this section does not permit is the determination that all statutory burdens are shifted merely by the fact of being an evidence holder. Unless the rule of allocation and the degree of proof are read together, one misses which requirement fails.
2.7The Link to the Application and the Dispositive Part
In this scene, if there is not enough material to determine which side's explanation is true, on what basis must the court conclude its determination? Taking up the conclusion of the preceding section that all statutory burdens cannot be shifted merely by the fact of being an evidence holder, this section now asks which requirement fails when the facts remain ultimately undetermined. To answer that question, one must examine separately who bears the disadvantage and what has not been proved.
The language of the gist of the precedent answering this question is settled as follows: "In addition, on the ground that the responsibility to prove that the cause of the fire lay in the negligence of the company's employees rests with the plaintiff," (Supreme Court Decision 2008Da6755 decided October 28, 2010). The party indicated by this phrase is the plaintiff, and the act imposed upon him is the proving of the fact that places the negligence of the company's employees as the cause of the fire. The responsibility borne by the plaintiff is the disadvantage that arises when, among the constituent facts of the claim for damages, the parts corresponding to causation and negligence are not proved. The precedent is read as allocating this disadvantage to the plaintiff, and no further factual circumstances are taken as the basis of this section.
The line connecting this language with the conclusion of this section lies in distinguishing the allocation of the burden of proof from the degree of proof. The burden of proof means the rule for bearing the disadvantage when a fact is undetermined, and differs from the question of which side's explanation appears more plausible. Proof of facts in civil matters takes as its standard not scientific certainty but a high degree of probability; yet when facts remain that fall short of that standard, the rule to fill that gap must be determined separately. The precedent's recognition that the burden of proof rests on the plaintiff is a determination at a different stage from the evaluation that the plaintiff's explanation was insufficient, and only after first fixing to whom the failure of proof of a constituent fact is allocated is the propriety of the claim decided.
The strongest counterargument lies in the proposition that it is always fair to place the burden on the party that holds the materials. The criticism that, when the party holding the materials remains silent, the ordinary burden of proof entrenches substantive inequality is felt especially heavily in lawsuits where access to information is uneven. The allocation of the burden of proof is a matter that follows the structure of the constituent facts and the direction of the burden prescribed by law, and the burden does not shift merely because the opposing party holds the materials. Maintaining this boundary, one nonetheless does not deny the risk of inequality the counterargument identifies; one merely holds that that circumstance does not constitute a ground for shifting the burden.
Returning to the synthetic scene, only the facts that the explanation of causation of Plaintiff A was not sufficiently supported and that the alternative explanation of Defendant B was not proved either are confirmed. When these two states are placed together, what the court can confirm is that neither explanation has been confirmed, and from this scene alone it is not revealed who holds more materials or what evidence has not been submitted. Accordingly, where a key constituent fact remains ultimately unproved and the claimant bears the risk of that uncertainty, the requirements of that claim are recorded as unmet. Even with the same uncertainty, if the applicable burden-of-proof rule differs, the allocation of the risk of losing changes; but one cannot assert the opposite allocation by drawing in facts not set in this scene.
What can be confirmed in this section is, when proof of a constituent fact ultimately falls short, to which requirement that failure is recorded, and that when that disadvantage is allocated to the claimant, that requirement remains unmet. What cannot be confirmed is the outcome of the win or loss of a real case or the result of recovering damages, and that gap cannot be filled with facts absent from the synthetic scene. The remaining task is to examine how the structure of this allocation aligns with the premises received from the preceding section.
2.8The Scope This Section Undertakes
Even after it is settled that proof of facts in civil matters follows not scientific certainty but the standard of a high degree of probability stated in the precedent, the explanation of causation of Plaintiff A was not sufficiently supported, while the alternative explanation of Defendant B was not proved either. In this state, when the evidence falls short of a conclusion, what remains is whose claim is blocked and why.
For this question, the relevant original text is the phrase "In addition, on the ground that the responsibility to prove that the cause of the fire lay in the negligence of the company's employees rests with the plaintiff," (Supreme Court Decision 2008Da6755 decided October 28, 2010). The party indicated by this language is the one bearing the responsibility to prove; that act is the proof of the causal fact; and the condition is the satisfaction of the requirement of damage caused by negligence. As to the content of the tort (unlawful act), Civil Act Article 750 provides: "A person who has caused damage to another by an unlawful act arising from intent or negligence shall be liable to compensate for that damage."
Connecting this ground with the conclusion, what fails is the element of negligence among the constituent facts of the claim for damages, and what is blocked is the plaintiff's claim. Here, the burden of proof means the rule for bearing the disadvantage when a fact is undetermined, and this allocation is distinguished from the degree of proof. The degree of proof asks at what height of probability a determination is reached, whereas the burden of proof is the question of who bears the loss of the state of uncertainty that does not reach that height. The precedent in question held that the responsibility to prove that the cause of the ignition lay in the negligence of the company's employees rested with the plaintiff.
If this distinction is ignored, there arises the counterargument that it is always fair to place the burden on the party that holds the materials. Control over or access to the evidence does not itself change the allocation of the burden of proof for a constituent fact, and the inequality the counterargument identifies remains a matter to be addressed through other avenues of relief.
In the synthetic scene, the facts relevant to this question are that the explanation of causation of Plaintiff A was not sufficiently supported and that the alternative explanation of Defendant B was not proved either. With these two uncertainties placed side by side, if the burden of proof for the constituent fact of negligence is allocated to the plaintiff, the plaintiff's claim is blocked. Applying the test that, even with the same uncertainty, a change in the applicable burden-of-proof rule changes the allocation of the risk of losing, since the scene itself cannot determine which rule applies, this item remains undetermined.
The conclusions this section permits are the distinction that the allocation of the burden of proof differs from the degree of proof, and the narrow determination that, once the party bearing the burden of proof of a constituent fact is fixed, the disadvantage of that uncertainty is determined. What this section does not permit is the determination that the burden is shifted merely by the status of being an evidence holder, and the determination that the failure to prove the alternative explanation is at once the falsity of that explanation.
2.9Application of the Text of the Provisions and the Holdings of Precedents
증거가 모자라다는 사실 판단 하나만으로는 그 손해를 어느 쪽이 감수해야 하는지까지 정해지지 않는다. 이 점을 받아 사실이 끝내 불명일 때 어떤 요건이 실패하는가라는 물음을 이번 단계에서 다룬다. 불명이란 어떤 사실의 진위가 재판 과정에서 확정되지 못한 상태를 말하며, 이 상태에서 청구가 기각되는 경우 그 불이익이 누구에게 귀속되는지가 문제로 남는다. 저자는 이 물음을 판정할 때 해당 원문 문언만을 근거로 삼는다는 전제를 세우고, 그 문언이 가리키는 주체와 조건을 밝히는 방식으로 검토를 진행한다.
해당 원문 구절은 「아울러 화재의 원인이 회사 직원들의 과실에 있음을 증명할 책임은 원고에게 있다는 이유로,」(대법원 2010. 10. 28. 선고 2008다6755 판결)이다. 이 문언은 손해배상청구의 요건사실 가운데 가해자의 과실에 해당하는 사실을 누가 증명해야 하는지를 특정하고 있다. 문언상 증명할 책임을 진 주체는 원고이고, 증명의 대상은 화재 원인이 갑 회사 직원들의 과실에 있다는 사실이다. 이 구절은 판결요지 [2] 중 지정된 부분이므로, 판결이유의 다른 사실관계나 이 문언 밖의 논거는 이 절의 판정 근거로 삼지 않는다. 한편 손해배상책임의 근거 조문인 「제750조(불법행위의 내용) 고의 또는 과실로 인한 위법행위로 타인에게 손해를 가한 자는 그 손해를 배상할 책임이 있다.」(민법 제750조)는 배상책임이 성립하는 요건을 규정하고 있을 뿐, 그 요건사실이 불명일 때의 불이익을 누가 감수할지에 대해서는 직접 말하지 않는다.
이 두 문언을 함께 놓고 보면 증명책임의 귀속과 증명의 정도를 구별해야 할 이유가 드러난다. 민법 제750조가 요구하는 것은 고의 또는 과실로 인한 위법행위와 손해의 발생이라는 요건의 충족이며, 대법원 2010. 10. 28. 선고 2008다6755 판결이 정한 것은 그러한 요건 중 과실 부분이 끝내 입증되지 않았을 때 그 불이익을 원고가 부담한다는 규칙이다. 이 규칙은 어떤 증거가 얼마나 설득력 있는지를 묻는 증명의 정도와는 별개의 문제로, 민사의 사실 증명은 자연과학적 확실성이 아니라 판례가 말하는 고도의 개연성의 기준을 따른다는 전제 아래에서 작동한다. 이 기준을 충족하지 못해 사실이 불명으로 남았을 때, 실패하는 것은 해당 요건사실의 증명이며, 그 결과를 감수할 사람이 누구인지는 별도로 정해진 증명책임 규칙에 따라 결정된다.
자료를 가진 당사자에게 항상 부담을 지우는 편이 공평하다는 반론은, 자료를 가진 쪽이 침묵하면 통상적 증명책임이 실질적 불평등을 고착한다는 점을 지적한다. 이 반론은 강한 설득력을 지닌다. 상대방이 정보에 접근하기 쉬운 상황에서 형식적으로 증명책임을 분배하면, 실질적 진실 규명이 희생되고 불이익이 힘의 불균형을 따라 배분되는 결과가 생길 수 있기 때문이다. 그러나 앞의 답은 증거 보유자라는 사실만으로 모든 법정 부담을 전환할 수는 없다는 것이다. 해당 원문은 증거를 누가 쥐고 있는지에 따라 책임이 옮겨진다고 말하지 않고, 특정 요건사실의 증명 책임이 원고에게 있다고 선을 그었다. 공평의 요청을 받아들여 증명책임 배분을 조정할 근거가 별도로 존재하더라도, 이 절의 범위 안에서 증거 보유 상태만으로 법정 부담이 전환된다고 보기는 어렵다. 반론이 지적하는 불평등의 위험은 지워지지 않으며, 다만 증거 보유라는 사실이 모든 부담의 이전 사유가 되지는 않는다는 한계만이 확인되는 것이다.
이 장면은 원고 A의 원인 설명이 충분히 뒷받침되지 않았지만 피고 B의 대체 설명도 증명되지는 않았다는 것이다. 이 장면에서 양쪽 설명 모두 요건에 해당하는 사실의 증명이 끝내 완성되지 않았다. 이 경우 대법원 2010. 10. 28. 선고 2008다6755 판결이 정한 문언을 적용하면, 가해자의 과실에 해당하는 사실이 끝내 증명되지 않은 결과 그 불명의 위험을 원고가 부담하는 방향으로 귀속이 정해진다. 다만 이 장면은 실제 사건의 재현이 아니며, 원고 A와 피고 B가 각자 어떤 자료를 가지고 있었는지, 어느 쪽이 침묵했는지는 설정된 바 없다. 자료의 보유 여부나 제출 태도가 어느 쪽이었는지에 따라 결론이 달라지는 부분은 이 장면만으로 판단할 수 없으므로 미확정으로 남는다. 같은 불명이라도 적용되는 증명책임 규칙이 달라지면 패소 위험의 귀속을 바꾼다는 조건부 검사를 적용하면, 규칙이 바뀌는 경우의 결과는 이 장면의 사실로서 확정된 것이 아니다.
이 절이 허용하는 결론은 해당 원문 문언만을 불명확한 사실에 따른 패소 위험의 귀속의 판정 근거로 쓰며, 그 문언이 특정한 주체와 사실에 대해 정한 증명책임의 귀속을 적용하는 데 그친다는 것이다. 이 문언을 넘어서 판결이유의 다른 논거나 다른 항목으로 확장하는 것은 허용되지 않으며, 증거 보유 사실만으로 법정 부담이 전환된다고 보는 것도 허용되지 않는다. 한편 이 절은 증명책임의 귀속이 누구에게 가는지를 정할 뿐, 그 귀속에 따라 실제로 어떤 구제가 이루어지는지까지 확정하지 않는다. 기록이 부족해 어떤 사실이 불명인지, 누가 그 위험을 부담하는지가 특정되지 않은 경우에는 미확정으로 남으며, 이는 곧바로 구제 불능을 뜻하지 않는다.
2.10Fine Distinctions Within the Stage
앞선 단계에서 사실이 끝내 불명일 때 어떤 요건이 실패하는지를 밝혔다면, 지금은 그 불명의 불이익이 누구에게 귀속되는지를 따져야 한다. 불충분한 증거라는 사실 상태가 곧바로 어느 당사자의 패소로 이어지지는 않기 때문이다. 증명의 정도와 부담의 귀속은 서로 다른 문제이며, 그중에서도 단계 안에서 다시 구분할 부분을 무엇으로 삼아야 하는지를 정해야 한다.
이 물음에 대해 판례는 증명책임의 귀속을 특정한 당사자에게 지정한 바 있다. 대법원 2010. 10. 28. 선고 2008다6755 판결은 "화재의 원인이 회사 직원들의 과실에 있음을 증명할 책임은 원고에게" 두었다. 여기서 화재라는 표현은 해당 판례의 요지가 다룬 사안의 표기이고, 이 글의 합성 장면과는 별개다. 중요한 것은 증명할 책임이라는 문언이 원인 사실의 증명 소재를 가리킨다는 점이다. 이 책임은 손해배상 청구의 요건인 고의 또는 과실로 인한 위법행위에 관한 민법 제750조의 요건사실 가운데 원인에 해당하는 부분에 붙는다.
그 근거를 살펴보면 증명책임은 사실이 불명일 때의 불이익을 부담하는 규칙이며, 이 규칙의 작동 여부는 두 가지 다른 조건으로 나누어 판단한다. 요건사실의 증명 여부와 그 불명의 위험을 부담할 사람이 정확히 연결된 경우에는 이 규칙이 작동한다. 반면 어떤 사실이 불명인지 또는 누가 그 위험을 부담하는지 특정되지 않은 경우에는 작동 여부 자체가 미확정으로 남는다. 증명의 정도를 묻는 문제와 부담의 귀속을 묻는 문제를 섞지 않아야 하는 이유가 여기에 있다. 증명의 정도가 판례가 요구하는 고도의 개연성에 미치지 못한다는 사실만으로는, 누가 그 불이익을 감수하는지가 자동으로 정해지지 않는다.
한편 자료를 가진 당사자에게 항상 부담을 지우는 편이 공평하다는 반론이 제기될 수 있다. 이 관점은 자료 보유자가 침묵하는 동안에 통상적 증명책임이 실질적 불평등을 고착한다고 지적하며, 그 손실과 불편을 진지하게 겨냥한다. 그러나 이 글의 결론은 그러한 불평등을 해소할 수 없다는 뜻이 아니다. 단지 증거 보유자라는 사실만으로 모든 법정 부담을 전환할 수는 없다는 조건을 지킬 뿐이다. 핵심 요건사실이 끝내 증명되지 않았고 그 불명의 위험을 해당 청구자가 부담하는 경우에는, 보유자가 자료를 내놓지 않았다는 사정만으로 그 위험의 귀속이 뒤집히지 않는다.
이 장면에서는 원인 설명이라는 핵심 요건사실이 증명되지 않았고, 그 불명의 위험을 원고가 부담한다고 지정된 경우에 속한다. 따라서 이 장면만으로는 자료를 가진 쪽이 누구인지가 밝혀지지 않았으므로, 증거 보유자를 특정하여 부담을 바꾸는 판단은 이 글의 범위 밖에 있다. 다만 적용되는 증명책임 규칙이 달라지는 것이 확인되면 패소 위험의 귀속이 달라지는데, 이는 귀속 자체가 불변하는 것이 아님을 보여준다.
결국 이 절이 허용하는 결론은 명시한 조건에 한정된다. 단계 안의 세부 구분은 작동하는 경우·막히는 경우·미확정인 경우 조건으로만 나누며, 이 세 조건 밖의 판정 기준을 새로 만들어서는 안 된다. 어떤 요건이 실패했는지를 밝힌 뒤에 추가 단계를 발명하여 부담의 귀속을 거기서 다시 가르는 것은 금지된다. 허용되지 않는 것은 증거 보유라는 하나의 사정으로 모든 법정 부담을 일괄 전환하는 것이며, 이는 이 글의 경계이기도 하다.
2.11Synthesis of Effectiveness, Blockage, and Undetermination
그러나 확신이 미치지 못했다는 사실만으로 어느 쪽이 질 것인지가 자동으로 정해지지는 않는다. 증명책임은 사실이 불명일 때의 불이익을 부담하는 규칙을 뜻하고, 전 절은 증명책임의 귀속이 증명의 정도와 다르다는 점만을 확인하고 넘어왔다. 이어서 사실이 끝내 불명일 때 어떤 요건이 실패하는가 하는 물음이 남는다.
이 물음에 답할 수 있는 문언은 요지 속에 이미 정해져 있다. 판례는 "아울러 화재의 원인이 회사 직원들의 과실에 있음을 증명할 책임은 원고에게" 있다는 이유로 판단했다(대법원 2010. 10. 28. 선고 2008다6755 판결). 여기서 '증명할 책임'이라는 표현은 입증의 노력이나 성실함을 요구하는 뜻이 아니라, 그 사실이 끝내 확인되지 않았을 때 어느 당사자가 그 불이익을 감수해야 하는지를 가리키는 것으로 읽힌다. 주체는 원고이고, 대상은 과실로 인한 손해의 원인 사실이며, 그 귀속은 청구가 인용되기 위해 필요한 요건사실과 정확히 맞물려 있다.
이 두 단계가 분리되어야 하는 이유는 각각이 다른 문제를 결정하기 때문이다. 민법 제750조는 "제750조(불법행위의 내용) 고의 또는 과실로 인한 위법행위로 타인에게 손해를 가한 자는 그 손해를 배상할 책임이 있다"(민법 제750조)고 규정한다. 과실과 손해 사이의 인과관계는 배상 책임을 발생시키는 요건사실이고, 그 사실이 불명일 때의 불이익은 증명책임의 배분에 따라 귀속된다. 요건사실의 미확인은 실패의 내용이고, 증명책임의 귀속은 그 실패를 어느 당사자에게 머무르게 하는지를 결정한다. 요건사실의 증명 여부와 그 불명의 위험을 부담할 사람이 정확히 연결되면 이 구별은 작동한다.
가장 강한 반론은 자료를 가진 쪽이 침묵하면 통상적 증명책임이 실질적 불평등을 고착한다는 점에서 나온다. 자료를 가진 당사자에게 항상 부담을 지우는 편이 공평하다는 주장은, 증거에 대한 접근이 이미 불균등한 상황에서 기존 배분이 불리한 쪽을 영원히 묶어 둔다고 지적한다. 이 점을 부정할 수 없고, 실제 소송에서 자료의 편재가 결과에 그림자를 드리우는 것도 부인하기 어렵다. 소유한 자료의 양은 절차적 판단의 여러 요소 가운데 하나일 뿐, 요건사실과 증명책임의 연결 자체를 대체하지 않는다.
합성 장면에서 확인되는 사실은 하나다. 여기서 끝내 확인되지 않은 사실은 원인이고, 그 사실의 불명에 따른 불이익을 부담할 사람으로 원고 A가 지정되어 있다면 판례의 문언과 같은 구조가 성립한다. 이 경우 불명은 작동하여 원고 A의 청구를 막는 방향으로 귀속된다. 다만 어떤 사실이 불명인지, 또는 누가 그 위험을 부담하는지 특정되지 않은 경우에는 판정이 미완이다. 장면만으로는 원고 A의 증명책임 귀속 근거가 확정되지 않으므로 그 부분은 미확정으로 남는다.
이 절이 허용하는 결론은 셋의 구별이다. 증명 여부와 위험 부담자가 정확히 연결되면 귀속이 작동하고, 핵심 요건사실이 끝내 증명되지 않은 채 그 위험이 해당 청구자에게 귀속되면 청구는 막히며, 위험 부담자 특정이 안 되면 미확정이다. 같은 불명이라도 적용되는 증명책임 규칙이 달라지면 패소 위험의 귀속을 바꾼다는 사실은 이 구별이 사실의 양이 아니라 규칙의 정함에 서 있음을 보여준다.
2.12Conclusion Carried Over to the Next Section
이렇게 양쪽 모두 확정되지 않은 상태에서 앞서 살핀 결론은 사실 인정의 정도가 자연과학적 확실성이 아니라는 점이었다. 그렇다면 지금 남는 물음은 그 정도가 모자랐다는 판단만으로 어느 요건이 실패한 것인지, 그리고 그 실패를 기록에 어떻게 남기는지이다. 사실이 끝내 불명일 때 어떤 요건이 실패하는가?
해당 문헌은 증명책임의 주체를 가리키는 문언을 담고 있다. 해당 판례는 "증명할 책임은 원고에게 있다는 이유로,"라고 적시한다(대법원 2010. 10. 28. 선고 2008다6755 판결). 이 구절은 증명의 정도를 묻지 않고, 확인되지 않은 채 남은 요건사실에 대해 누가 그 상태를 감당하는지를 정한다. 원고는 손해배상 청구에서 자신의 청구를 뒷받침하는 요건을 주장하고 증명하는 자리에 있고, 회사 직원들의 과실이라는 원인 사실이 그 요건에 해당한다.
이 문언을 통해 증명책임의 귀속과 증명의 정도가 서로 다른 단계임이 드러난다. 전자는 어느 요건이 확정되지 않았을 때 그 불이익을 누가 지는가를 정하는 규칙이고, 후자는 그 요건을 확인하기 위해 필요한 개연성의 높이를 말한다. 사실 인정의 기준이 고도의 개연성이라 하여 증명책임의 귀속까지 달라지지 않고, 귀속이 정해졌다고 하여 증명의 정도가 높아지지도 않는다. 요건사실의 증명 여부와 그 불명의 위험을 부담할 사람이 정확히 연결될 때에만 패소 위험의 귀속이 확정된다.
이 구분에 반대하는 쪽은 자료를 가진 당사자에게 항상 부담을 지우는 편이 공평하다고 말한다. 다툼의 대상이 되는 요건사실이 무엇인지, 누가 그 요건을 주장하고 있었는지가 정해지지 않은 채 자료 보유 사실만으로 부담이 옮겨지면, 증명책임은 자료의 위치를 따라가는 것이 아니라 청구의 구조를 따라가야 한다는 원리를 놓친다.
합성 장면으로 돌아가 보면, 원고 A의 원인 설명과 피고 B의 대체 설명이 모두 확정되지 않았다. 이 상태에서 판정할 수 있는 것은 그 범위뿐이다. 원고 쪽 요건이 증명되지 않았다는 점은 분명하지만, 피고 B의 대체 설명이 참이라는 결론은 이 장면에서 얻을 수 없다. 어떤 사실이 불명인지, 누가 그 위험을 부담하는지가 특정되지 않은 상태로 남는 범위에서는 확정할 수 없다.
정리하면, 사실이 끝내 불명일 때 실패하는 것은 증명책임의 귀속이 정해진 바로 그 요건사실의 증명이며, 그 불이익은 귀속 규칙이 가리키는 당사자가 감당한다. 확정되지 않은 채 남은 것을 피고 쪽 설명의 진실로 바꾸어 읽는 것은 허용되지 않는다. 증명의 정도에 관해 확정한 것은 사실 인정의 기준이고, 누가 불명을 지는가는 별도의 규칙이 정한다. 남은 것은 그 둘을 구별한 채 법률 논쟁을 미증명의 문제와 뒤섞지 않는 일이다. 법률상 해석이 다투어지는 상태를 사실의 미증명으로 처리할 수 있는가?