Problem setting. Tort law allows to reallocate the costs of accidents so that the aggrieved person has an opportunity to shift damage he or she sustained onto someone else. But this reallocation is available if and only if there is a sufficient justification for it. As J. Coleman, S. Hershovitz and G. Mendlow nicely explained it: you cannot make your mass mine just because you want so [1]. Therefore, in order to succeed in tort suit plaintiff has to prove the existence of preconditions of liability in tort, which are: wrongfulness, damage, causation and fault. Wrongfulness means that the defendant has infringed some interest protected by law. The concept of damage denotes that the plaintiff has sustained some harm. Finally, causation indicates that but for the defendant's behavior the damage would not have occurred. The notion of fault presupposes that three above mentioned requirements do not suffice; instead, in addition there has to be something else in order to justify the imposition of liability onto the defendant. Why is it so and what should it be? - These are the questions addressed in this article.

Recent research and publications analysis. Fault principle is an essential feature that distinguishes modern law from ancient primitive law. In academic literature it is generally recognized that in primitive law person was hold liable for any damage she caused regardless of culpability (i. e. regardless of whether the person's actions were blameworthy or not) [2, p.64]. In other words, the sole necessary and sufficient precondition of liability in tort was the causal link between the tortfeasor's behavior and the damage suffered by the aggrieved person. This feature of primitive law was noticed by R. von Jhering in his famous work The Spirit of Roman Law at the Various Stages of Its Development [3, pp. 19-111]. Further explanation can be found in H. Kelsen's Pure Theory of Law [4]. The reason for the strict liability regime (as we would call it today) was the fact that primitive man did not distinguish between the principle of causality and the principle of imputation [4, pp.82-85]. Thus primitive man interpreted the nature in a normative animistic way: all the natural events were treated either as a reward or as a punishment for human's actions. And since the crucial difference between human's actions and acts of nature went unnoticed there was no need to inquire into the relation between harmful consequence and human's will. H. Kelsen writes that “ [t] he decisive step in the transition from a normative to a causal interpretation of nature, from the principle of imputation to the principle of causality, consists in man becoming aware that the relations between things (as distinguished from relations between men) are independent of a human or superhuman will, or, which amounts to the same, are not determined by norms - it consists in man becoming aware that the behavior of things is neither prescribed nor permitted by any authority" [4, pp.84-85].

Paper objective. The objectives of this article are as follows: (1) to reveal why the fault principle is considered to be morally superior to no-fault liability in primitive law; and (2) to find out the essence of fault in modern tort law and then to express the concept of fault in the most precise manner possible, namely through math formula.

Paper main body. The moral strength of fault principle consists in the fact that modern law blames person only for those violations of law that person commits freely, i. e. while enjoying freedom and therefore being able to make conscious choice. As G. W. Hegel stated it, a human is a subject of law only as a free being [5, p.36]. Thus, the fault principle is deeply rooted in the very concept of law. The interrelation between fault and freedom was noticed by A. Tunc, who wrote: “ [t] he law of tort, if it takes fault as a criterion of liability, recognizes the freedom of a man, his responsibility, his capacity of behaving in a social or in antisocial manner, and his ability to choose between good and evil. According to his decisions and behavior, it absolves him of responsibility or imposes upon him a civil sanction” [2, p.65].

Law is a deontic rule appealing to free will, because only what is physically possible can be morally demandable. The demand is absurd both when it is impossible to satisfy it and when it is impossible not to satisfy it. The rule demanding not to submit to gravitation is absurd as well as the one demanding to submit to. That is why any deontic rule makes sense only if there are some alternatives, which constitute a space of freedom. In such a context the very existence of law is contingent on freedom of human's will. It is the human's freedom that allows to judge human's actions. Thus, provided that we consider tort law as a set of rules prohibiting infliction of damage and establishing liability the refor, it is fair to state that fault is a precondition of tort liability specifically because freedom is a precondition of the very law's operation.

The concept of fault is an inverted concept of freedom (“dark side” of the freedom): while the existence of freedom in the present enables the law to operate, fault is an affirmation that the freedom existed in the past (at the moment of inflicting damage). From this point of view, it is easy to draw a distinction between wrongfulness and fault: while the former means that the fact at hand contravenes the law the latter means that the demand to avoid this fact was operative in the circumstances.

While establishing fault the court investigates whether the tortfeasor was free at the moment of infliction of damage. Fault denotes that formally wrongful act was committed freely. Since establishing fault is conducted after the wrongful act has been committed (it is conducted within judicial proceedings, which constitute backward-looking research), the inference follows that fault is an ex post conclusion of freedom.

The proposed definition of fault is completely based on the concept of freedom, which has been treated by philosophers in numerous ways. Therefore, the proposed definition of fault will be rather vague unless we provide a clear concept of freedom.

For the purposes of present research freedom has to be defined as the ability to choose from among several alternative ways of acting. Thus, tortfeasor is at fault if (while inflicting damage) he was free to choose from among several (at least two) alternative ways of acting at least one of which was harmless. Put in math terms it is read as follows: a tortfeasor is at fault if the infliction of damage was committed in the situation W, where the number of available alternatives (how to act) A (W) added up to not less than two items (a_n), and at least one of those items was lawful (wouldn't have caused the infliction of damage) (R), i. e. in the situation W there is fault if

Г A (W) = {, a₂,., a_n}, where n > 2; [Зa є A (W): R (a).

This formula demonstrates that fault is present whenever there is a missed opportunity of the right choice. If there is no such an opportunity, there is no fault. Moreover, the greater the opportunity (of the right choice) was, the greater the fault is. The fact that the tortfeasor has acted in a particular harmful manner constitutes wrongfulness; the fact that he could have acted in a different manner constitutes fault.

The similar idea has been stated by K. Larenz who worded it in the following way. “We speak of fault only where a person can individually be held responsible for the act or omission. We blame the person because he could have and should have acted differently in the specific situation, because he has acted wrongfully even though he could have acted properly had he applied the necessary care or attention or good will. The freedom in the sense of the ability to act differently on the one hand and the violation of the obligation on the other constitute the elements of fault" (as cited in [6, p.142]).

In short, fault means that at the moment of the infliction of damage the tortfeasor was in the situation (situation W) of free choice. But the crucial question is: when can we say that such a situation is in evidence? In other words, what are the necessary elements of a situation of free choice?

First of all, choosing is possible only if there is someone who is to choose. Thus the first element of free-choice-situation is a person. But in accordance with the law not every individual is deemed to be able to choose consciously; for this reason, there has to be not just somebody, but somebody who is capable of making conscious choices. In other words, there has to be legally capable person, which is the person who is conscious of her actions and has control over them. Indeed, if the person lacks mentioned features her choices cannot be regarded as free choices and therefore there is no blameworthiness in her actions. Incapacitated persons cannot be held liable for damage they caused specifically because they cannot be at fault.

It is noteworthy that this principle was clearly stated in Ancient Roman law. Thus, in Justinian's Digest we can find the following words: “ [t] herefore we ask whether an action under the Lex Aquilia will lie where an insane person causes damage? Pegasus denies that it will, for how can anyone be negligent who is not in his right mind? This is perfectly true. Hence an action under the Lex Aquilia will not lie; just as where an animal causes the damage, or where a tile falls from a roof. Again, if a child causes any damage the same rule applies" (Dig.9.2.5.2).

One can choose only if there are some options. Thus, the second necessary element of free-choice-situation is a variative range, i. e. number of available alternatives at least one of which will not cause the damage. This variative range {a,,, a₂,. a } depicts the scope of opportunities that is available to tortfeasor and is dependent on extrinsic objective factors (such as the law of nature, technological level, actions of third persons etc.). It seems quite obvious that the tortfeasor is not at fault if in order to avoid inflicting damage he had to do something that was technically impossible at the moment. Another factor that can exclude harmless way of acting out of the number of available alternatives is irresistible force (so called acts of God). For example, if the defendant's vehicle is washed away from the motorway by the mudflow wave and as a consequence the plaintiff's fence is destroyed, it is clear that there is no fault, because the harmless option (not to swerve from the motorway) was excluded by the irresistible force.

What else is needed, apart from the chooser and the alternatives? Let's imagine the following hypothetical situation. A person is offered to press one of the three buttons; though the person is not told that pressing the first button will annul the person's bank account, pressing the second one will transfer the title to all person's property to the state; and pressing the third one will grant the person two million USD. Will it be a free choice in the mentioned sense? Of course, no, because free choice presupposes knowledge of the relevant circumstances. In other words, free choice is always an informed choice. Knowledge of factual and legal consequences of every option constitutes sine qua non of free choice. For this reason, there is no fault if person could not know (a) what kind of consequences would be entailed by her actions or (b) that the very way of her behavior was per se illegal. Whether the tortfeasor knew the juridical consequences of his actions - this question can be easily answered due to the presumption that everyone knows the laws properly promulgated.

Whether the tortfeasor could predict the factual consequences of his behavior - this question is way more complicated. It is also known as the problem of foreseeability and remoteness of damage. As a general rule, one should be liable only for those consequences of his actions that he can foresee. If the damage caused could not have been reasonably foreseen, then the tortfeasor is exempted from liability on the ground of so-called casus.

It has to be emphasized, though, that while assessing the foreseeability of damage court does not try to find out what the tortfeasor actually foresaw, but rather what the tortfeasor had to foresee as a reasonable and circumspect person. In other words, the objective criterion applies based on what can be expected of everyone in the similar situation.

The elements of free-choice-situation that have been mentioned so far do not suffice to define fault though, because sometimes all the elements being present, the tortfeasor nevertheless cannot be deemed to be at fault. This is the case, where the tortfeasor could have avoided inflicting damage, but at excessively heavy cost. For instance, when in order to avoid damaging neighbor's cheap property the tortfeasor had to risk his own life.

Perhaps the most straightforwardly this idea was expressed by Judge Learned Hand in the famous case United States v. Carroll Towing Co., 159 F.2d 169 (2d. Cir. 1947). According to Judge Learned Hand, if we denote as P the cost that has to be incurred in order to avoid infliction of damage; D - the amount of damage; and L - probability that damage will occur, then the tortfeasor is at fault whenever P < DL.

The Hand's formula is based on a very sound idea. Every human's action can be treated as an act of exchange where something is given away (X) in order to get something else (Y) (for instance, when you study at the university you invest your time, efforts and money (X) in order to get necessary qualifications (Y)). From this perspective one acts reasonably if Y > X, i. e. if the value of what has been received exceeds the value of what has been given away. In the context of tort liability, it is the tortfeasor's investment in the precautions (necessary to avoid infliction of damage) that is given away; and it is the preclusion of damage that is received. As Coleman, Hershovitz and Mendlow stated it “a precaution is reasonable when it is rational; a precaution is rational when it is cost-justified; and a precaution is cost-justified when the cost of the precaution is less than the expected injury (the latter being the cost of the anticipated injury discounted by the probability of the injury's occurrence) ” [1].

Thus, it is not enough if among the available alternatives {a₁, a,. aj there is one harmless; in addition, this one harmless option has to be reasonable. Otherwise choosing of this harmless option cannot be expected. Moreover, law and economics theory proves that even if tort law held person liable irrespectively of reasonability-criterion, persons nevertheless would not take unreasonable precautions, because it would be more cost-efficient not to take those precautions but rather to inflict damage and compensate it afterwards [7].

Having regard to the above we have to supplement formula (1) with one more element. A tortfeasor is at fault if the infliction of damage was committed in the situation W, where the number of available alternatives A (W) added up to not less than two items (a_n), and at least one of those items was lawful (wouldn't have caused the infliction of damage) (R) and reasonable (S), i. e. in situation W there is fault if

I A (W) = {{,a₂,.,a_n}, where n > 2;

[3a є A (W): R (a) л5 (a). (²⁾

The reasonability-criterion is recognized in Principles of European Tort Law (hereinafter - PETL). Pursuant to Art.4: 101 PETL a person is liable on the basis of fault for intentional or negligent violation of the required standard of conduct; and according to Art.4: 102 PETL the required standard of conduct is that of the reasonable person in the circumstances. Moreover Art.4: 102 PETL provides important guidelines of how to model the behavior of the hypothetical reasonable person. Namely, it is stated that the reasonable person's behavior depends, in particular, on the nature and value of the protected interest involved, the dangerousness of the activity, the expertise to be expected of a person carrying it on, the foreseeability of the damage, the relationship of proximity or special reliance between those involved, as well as the availability and the costs of precautionary or alternative methods.

Therefore, according to PETL fault is established by comparing the tortfeasor's behavior with the behavior of hypothetical character (reasonable person): what has been actually done by the tortfeasor is compared with what could have been done by a reasonable person in the same circumstances.

The similar approach is followed in common law as well. Thus, in a prominent case Blyth v Birmingham Waterworks Company (1856) 11 Ex Ch 781 Baron Alder - son defined the negligence in the following way: “ [n] egligence is the omission to do something which a reasonable man, guided upon those considerations which ordinarily regulate the conduct of human affairs, would do, or doing something which a prudent and reasonable man would not do. The defendants might have been liable for negligence, if, unintentionally, they omitted to do that which a reasonable person would have done, or did that which a person taking reasonable precautions would not have done”.

Moreover, digging deep shows that this approach was invented in Ancient Rome, though the standard of comparison at that time was called slightly different, namely diligens paterfamilias. In the Justinian's Digest one can find the words of Paulus: “it is negligence when provision was not made by taking such precautions as a diligent man would have done” (Dig.9.2.31).

Unlike the PETL and common law Ukrainian Civil Code does not include the reasonability-criterion in the definition of fault. Under the Art.614 of Civil Code of Ukraine a person is not at fault if she proves that she has done everything in power. Taken literally this provision requires to take every possible step to prevent infliction of harm irrespective of whether particular step is reasonable or not. And though some of the Ukrainian scholars argue that reasonability-criterion is implied due to the operation of general principles of civil legislation the character of a reasonable person remains hardly known to national legal science.

Moreover, it has to be noted here, that Art.614 is contained in Chapter 51 of Civil Code of Ukraine that is titled as “Legal Consequences for Breach of Obligation. Liability for Breach of Obligation”. For this reason, taken literally this article matches only situation where the defendant breaches some already existing (e. g. contractual) obligation. In particular, para.1 of the Article provides that “a person is not at fault if she proves that she has done everything in power to perform an obligation in a proper way”. Since Chapter 82 (which addresses torts) does not contain any definition of fault at all, many scholars argue that Art.614 mutatatis mutandis applies to tort liability as well. In particular, for the purposes of tort liability one should modify the provision of Art.614 in the following way: “a person is not at fault if she proves that she has done everything in power to prevent infliction of damage”.

Although the case-law on the issue is quite equivocal there are decisions where courts refuse to apply Art.614 to torts. In one of its decisions the Supreme Economic Court of Ukraine stated: “ [a] s to the appellant's allegation that the court did not take into consideration Art.614 CC of Ukraine, this allegation is overruled because the mentioned article deals with obligational relationship that was not present between the plaintiff and the defendant" [8]. Moreover, the Supreme Economic Court of Ukraine emphasizes that the distinction has to be drawn between contract damages and tort damages. So if the damage is caused by the breach of contract court cannot resort to provisions of Civil Code on torts and vice versa: if the damage is caused by tort court cannot resort to provisions on contract damages [9]. As long as this approach is taken it is inevitable to conclude that Ukrainian civil law does not have any definition of fault for the purposes of tort liability, notwithstanding fault is expressly required as a precondition of liability in tort. From this point of view, the expansive interpretation of Art.614 Civil Code of Ukraine may be seen as the least-evil solution.

As a result, pursuant to Civil Code of Ukraine fault matters for both tort and contract liability, but the definition of fault is provided for the purposes of contract liability only. Approach proposed in academic literature (to interpret the fault definition extensively) poses an important question of whether the essence of fault is the same both in tort and in contract law or, conversely, there has to be two completely different concepts of fault for the two realms of civil law.

The answer to this question seems to be as follows. The concept of fault is the same in both tort and contract law: fault means that a person (either tortfeasor or contracting party) has not done something that (a) could have been done and (b) was reasonably expected of her. However, what is reasonably expected of everyone (as a potential tortfeasor) starkly differs from what is reasonably expected of contracting party, for it is obvious that we are entitled to expect way more of the one who has promised something to us than of a random passer-by. First and foremost, it means that Hand's formula does not apply to contract liability. Being interpreted in the context of contracts Hand's formula would mean that contracting party is not at fault (and, therefore, is exempted from liability) whenever the cost of performance exceeds the value of counter-performance. But that is not the case, because according to the principle pacta sunt servanda even if performance of the contract becomes more onerous or less profitable than it was expected the contracting party nevertheless has to perform its obligations under the contract. Thus, for instance, according to Art.6: 111 (1) of Principles of European Contract Law “ [a] party is bound to fulfil its obligations even if performance has become more onerous, whether because the cost of performance has increased or because the value of the performance it receives has diminished”.

Therefore, the necessary flexibility of the definition is provided by the reasonability-criterion, meanwhile the general concept of fault (as the omission to do something that could have been done and was reasonably expected) applies for both tort and contract liability.

Conclusions of the research. The concept of fault in tort law is based on two pillars: freedom and reasonableness. When a court concludes that the defendant is at fault it means that the defendant could have avoided inflicting damage had he behaved as a reasonable person. In other words, the concept of fault denotes that that the tortfeasor had a choice, i. e. that there were several alternatives available at least one of which would not have caused the damage and was reasonable. Although the concepts that are employed hereinabove to define fault (such as reasonableness, foreseeability, remoteness) call for further thorough research; this article, grounding on moral justification of fault principle, provides general theoretical concept of what the fault is. The definition of fault contained in the Art.614 of Civil Code of Ukraine is inspired by Soviet-era instruments; notwithstanding the current Civil Code of Ukraine was adopted in 2003, the definition of fault in Art.614 exactly repeats the wording of Art.71 of Fundamentals of Civil Legislation of the U. S. S. R. and the Union Republics (1991). Therefore, the definition of fault in Ukrainian legislation has to be improved so that it includes the reasonability-criterion.

References