The concept of holistic group liability

Abstract The role of an individual in a group depends on the role of other individuals, context, ideas, and other factors; the individual in a group is never only the individual. If this is confirmed with experiments in social psychology and logic, it is undermined in a liability regime that still largely treats individuals as individuals. This paper thus examines how the present liability forms deal with the individual in groups. Terms such as “take into account the circumstances of the case”, “establish shares in the damage caused”, “the weight of fault”, “the weight of the consequences flowing from the action”, and “fairness” could serve to assert group liability if the latter is previously well documented in social psychology and then accepted by legal scholarship. To better adapt, address and legislate the individual’s liability in groups, this document provides new research directions in the field of groups, based on the new concept of “holistic group liability” introduced, in addition to the existing collective liability, which still treats the individual as a mere part of the group, rather than as a new value arising from relationships with other members, contexts, concepts, and ideas.

What is the 'crowd', how does it acquire the capacity to influence so decisively the mental life of the individual, and what is the mental change it imposes on the individual (Freud, 2007, p. 9)? ABOUT THE AUTHOR Mirko Pečarič has obtained Master of Science and PhD at the Faculty of Law, University of Ljubljana. He has passed also the national law exam. In 2013, the Government of the Republic of Slovenia appointed him as the State Secretary at the Ministry of Education, Science and Sport. As the full professor, he teaches the courses of Administrative Law, Regulatory Techniques, Public services and Introduction to Law. He wrote a series of legal opinions and requirements for the assessment of the constitutionality of laws; he is the author of four monographs and numerous scientific articles. The research reported in this paper relates to wider concepts of co-production of public services and values in the public administration.

PUBLIC INTEREST STATEMENT
People act differently in a group than they do as individuals. If this is confirmed with experiments and it is also clear to us as a matter of logic, the latter is undermined with a liability regime that still treats the individual as an individual, irrespective of his or her different role in the group. It is not the man's free will and reason, but also his unconscious and instinctive side, that is in control in a crowd. This leads to conflicts with our conception of man and to legal problems. In this regard, this paper addresses the variety of an individual's inclusions in a group vis-à-vis the present forms of liability. Based on the established discrepancies, the paper proposes new research directions on collectives for the legal arrangement of liability to better align these two areas.

Introduction
In an increasingly global, interconnected, and complex society, the action of individuals in groups and organisations is all the more necessary. There is more truth in saying that "two heads are better than one" than can be seen at first glance. Great things usually happen with the help of other people, and consequentially the ability to perceive, adapt, and change societal patterns is more important than any, however brilliant, individual decision. Thus, understanding social change predisposes to change in groups: Individual autonomy, integrity, and other human rights can only be considered in the context of the existing society. If this is logical, on the other hand, a community is still "logically" focused mainly on its reductionistic parts, on individuals. This holds true especially in law, which is not fully accustomed to relations, to subjectivepsychological and objective-structural-sociological influences (between the society and an individual), created in the wider social context of time and space. As the mere presence of other persons can be a form of social influence, this paper focuses on links between the collective forms of uniformity, obedience, and conformity and the corresponding legal liability 1 that does (not) address the former.
Any form of organisation achieves better productivity than the individual can. Although in organisations the "we" stance is more important than the "I", it seems legal liability law still insists on the latter; it is not accustomed to collective events, in the "winning the hearts and minds of others". But who are others and how can a person's actions be legally understood if he is part of a group in its relations, combinations and interactions? The willingness of one person to cooperate and to promote the overall success of a group is based on the various external and internal social influences and social powers. Cooperation (common preferences) and competition (divergent preferences) within and outside groups, social dilemmas (situations in which short-term selfinterest is in conflict with long-term collective interests), and questions about an individual's role in a group 2 are this paper's elements, based on which the individual's role in the group is compared to the legal regime that governs his or her liability.
Psychological and experimental insights into crowds diminish the individual's self-responsibility: It is not the man's free will and reason, but also his unconscious and instinctive side that is in control in a crowd. This leads to conflicts with our conception of man and to legal problems, for which someone can be held responsible (Malik, 2010). In this regard, this paper addresses the variety of inclusions of an individual in a group compared to the present forms of liability. Based on the established discrepancies, the paper proposes new research directions on collectives for the legal arrangement of liability to better align these two areas. In this line of thought, the next two sections examine collective behaviour and other people's influence on the individual's behaviour. This gives various options on the individual role in the group, where social influences and social powers are ubiquitous. In light of the described examples in these two chapters about the conduct of individuals within and outside groups, different types/combinations of liability will be presented in section four to compare the substantive relevance between the legal institution of liability and the conduct of individuals within a group. Section five deals with specific forms of social control (dialogue, nonconformism, and system structure) that can be used to build on the legal system of liability "in the short term". To align legal liability more closely with the behaviour of individuals in a group, Section six gives some new directions for research on collective legal liability, followed by a conclusion in Section seven.

Social mimicry
One of the fundamental characteristics of working in groups is social mimicry; it means a universal occurrence where one person unintentionally replicates the movements of another person. This is also known as the chameleon effect and/or the mere perception of another's behaviour that automatically increases the likelihood of engaging in that behaviour oneself (Chartrand & Bargh, 1999). Bond calls the power of others our innate socialness that holds huge sway over how we think and act, propelling us to both high achievement and unthinking cruelty. We are beholden to our peers, even when we think we're calling the shots (Bond, 2014). Empathy and other emotions 3 (which are tightly connected with decision-making 4 and cooperation with others), relationships (emotional or other connections between people), communication (imparting or interchange of thoughts, opinions, or information by speech, writing, or signs), expectations, social institutions, values, roles, norms etc. are essential building elements of social structure that are not possible in isolation. Bon (2001) was one of the first scholars of crowd psychology (it considers the individual as a member of a larger mass of people who organise themselves into a crowd at a particular time and place and for a particular purpose); in 1895, he argued the crowd acquires a "collective soul", independent of the individual, which becomes a source of folly rather than wisdom. Crowds are characterised by mass hypnosis, impulsiveness, volatility, and irritability; they are gullible and can be influenced by suggestions. They are also characterised by exaggeration and overuse of emotions, intolerance, dictatorship, and conservatism. Mobs are driven primarily by unconscious instincts, not by reason. The masses are ruled by the contagion of opinions and the fascination with the leader, who uses rapid suggestion, example, affirmation, repetition, and dissemination, all of which lead to prestige as a form of domination. 5 For him, masses are a form of abuse, which is why there can be no just and good mass uprisings. Even today, we know the term "mob" as a label for a disorderly, uninformed, uncultured crowd of people, close to Le Bon's mad crowd. 6

Crowd and public
Tarde, another great researcher of the masses, shares some of Le Bon's views on the regressive behaviour of the masses, but not his pessimism about the rule of the masses. He distinguishes between the physically present crowd (e.g., in the theatre, on the street) and the mental, ideologically connected public that is emerging with the development of print media and means of communication (Tarde, 2010). Unlike Le Bon, he does not consider this to be the age of crowds, but of the public towards which crowds evolve. In both crowds and public, he distinguishes between female and male groups, between groups of young and old, groups that believe in something (which are convinced and fanatical) and those that want something (which are passionate and despotic). The rural masses are more aggressive than the religious ones; the political masses (especially the urban ones) are the most passionate and the most furious; fortunately, they are versatile, moving easily from contempt to adoration, from excessive anger to excessive joy. The economic or industrial crowds are more homogeneous than others; the aesthetic or artistic crowds are the least tolerant because of the subjectivity of their opinions. The masses have four levels of passivity or activity. These are: expectant (e.g., the crowd in the theatre before the curtain rises), attentive (the same crowd after the curtain rises, lectures), demonstrative (the meaning of symbolism, slogans), or active (companies, sects, associations). Disordered or spontaneously (naturally) formed crowds (where the individual abilities and consciousness of individuals are lowered below the average) eventually become disciplined or ordered (artificial) publics, where the individual abilities and consciousness of individuals can be raised above the average with the help of a suggestive (intelligent) leader. The role of the leader is essential here: a confident leader seeks prestige, he is the mirror of the crowd, which imitates him.

Charismatic leader
The third big name in crowd psychology is Freud (2007); he gave special attention to charisma, identification with the leader. He explained the emergence of the drive for adoration and the association of the crowd with the leader. The individual's passion for self-adoration is said to originate from his narcissism, and it can only be the leader of the crowd who replaces the libidinal moment of self-love with love for the other, that is, for the leader (identification with and idealisation of the leader). The leader is a bonding material that binds the members of the crowd together. In this line of thought, Festinger et al. (1952) presented the idea of deindividuation, i.e. a group's condition where its members do not pay attention to other individuals as individuals, and therefore members do not feel like they are compared to others. This results in a reduction of members' inner restraints, and they are thus freer to indulge in behaviour from which they are usually restrained. Zimbardo (1969) also claims deindividuated or impulsive behaviour lowers the threshold of normally restrained behaviour; this more likely occurs when group members acquire feelings of anonymity, where "here and now" time orientation is adopted, and where people outside of the group are dehumanized. In the 1950s, Asch showed that people often accept the majority opinion, even when it is utterly wrong. The opinion of a majority of one's peers (whom one wants to please for one social reason or another) is enough to change one's mind, without the involvement of an official person (Asch, 1952).

Crowds' spontaneousness and context
Social psychology has tried to disprove Le Bon's notion that people in crowds are just a mad mob, as well as its counterpart-deindividuation theory. The mandatory element of the leader has not been confirmed in the yellow jackets' movement that sporadically began with the online petition . . . this spontaneousness also stands for the fall of the Berlin Wall or the Arab Spring. Le Bon himself seems to have overlooked the usefulness of the French Revolution, which took place on his soil and was triggered by a discontented mass of people. The classical psychology of crowds, based on group coherence and groupthink, has been confronted with the issues of social context and basis on which crowds operate, of participation, and the content of the very acts committed in the group. Based on the analysis of St. Paul's riots (in Bristol on 2 April 1980), Reicher spots the difference between crowd behaviour moulded by social identity (actions through the network) and crowd behaviour that may mould social identity (action from the network): "there is an assumption that crowd episodes are inferior [but] the present perspective stresses how sophisticated crowd actions are in the sense of being an accurate expression of an ideological understanding of the social world" (Reicher, 1984, p. 19).

A sense of belonging
When dealing with groups, it is important not to overlook social identity theory, which focuses on a personal sense of belonging to the same social category with other members with whom members share some emotional involvement (self-esteem, value connotations, in-group (us) and a degree of social consensus about that group and their membership in it vis-à-vis the out-group (them). Based on the mental processes of social categorisation, identification, and comparison, people tend to positively evaluate "us" vis-à-vis negatively "them" (Tajfel & Turner, 1979). Therefore, understanding the contexts, dynamics, and meanings of crowds is important to understand public disorder (Waddington & King, 2005).

Variety of stimuli
The social identity perspective recognises that people's characteristics, background experience, group inclusiveness, diversity, and/or the entire spectrum of identities and consequential identification with the group play an important role in shaping attitudes, values, beliefs, and behaviour (Reynolds et al., 2003). People reflect themselves and consequentially act not only from their characteristics but also from socially determined notions of "nation, profession, class, common ideas". The presence of others shifts decision-making about risky or deviant behaviour; there is significant variation in perceived risks, costs, and benefits when the only change was the number of other people involved in the act (Mcgloin & Thomas, 2016). This point towards a situated interactional analysis of group and intergroup dynamics, rather than merely trying to read the events in terms of broad background grievances or ideologies (Stott et al., 2017). Canetti (1981) is to the author's knowledge a single scholar, who focused on the importance of commands in the crowd. A command is part of our general biological 7 and social structure and our intimate human relationships. A command addressed to a large number of people makes a crowd of them, and, in as far as it succeeds in this, it does not arouse fear. There are commands addressed to single individuals and many. In a crowd, the command quickly (due to their density) spreads horizontally from a single individual to finally everyone. As far as liability, he uses the metaphor of a gazelle being caught by a lion while other gazelles stop running and graze on (similar to this being a religious sacrifice). It is a form of sacrifice that originates in the crowd and procures a respite for other crowd members. Canetti mentioned that every command consists of momentum (to act) and sting (to be punished otherwise), but his focus was not on various types and degrees of liability that could be focused on the crowd or its particular or collective members.

Collective wisdom
The concept of collective wisdom (Dyer, 2020;Galton, 1907;Landemore & Elster, 2012;Surowiecki, 2005;Tovey, 2008;Woolley et al., 2015) claims crowds are not folly but can exhibit high intelligence. Collective wisdom is not based on masses gathering in the marketplace but on the spontaneous (nowadays increasingly mathematical) aggregation of otherwise individual and independent opinions that do not easily fall under the influence of the loudest and most emotive speakers. Groups of individuals can act collectively in a way that reflects signs of wisdom. One such value, even a fundamental value of civilisation, is democracy or the rule of the people. A society in which true democracy is not developed is also more susceptible to "shock doctrines", where people, because of the collective trauma caused by major disasters or crises, resort to actions that would otherwise have been unthinkable before (Klein, 2007). Direct democracy can be understood as the most basic example of collective intelligence, where "people are always right." Nowadays, this wisdom is relatively conditioned by the electoral system and its combinations. In its basic idea, collective intelligence is not just a modern possibility made available by the advent of computers or the Internet, but a combination of the actions, behaviors, preferences, or ideas of a group of people to create new steps, to gain new insights into a subject. The difficulty of doing what we want to do collectively may lie in the invisibility in advance of the effects, from the point of view of each individual. The collection of responses from a large group of people allows statistical conclusions to be drawn about the group that no individual can know, that are not and cannot be known in advance. The co-creation of new conclusions by independent stakeholders is an essential part of what collective wisdom is about. Caution must be exercised regarding opinions: although they are always context-dependent in time, place, situation, and people, they are more robust if they are responsive and adequately processed and statistically processed in real-time and space by the active, numerous, and diverse members of the community. 8

Crowd dynamics
Understanding the dynamics of crowds is thus more complex than simply labelling them as stupid.
In light of all of the above, it is clear that individuals as members of a group act differently than individuals. Each of them identifies with the crowd and recognises himself as the equal member, or as part of "something bigger" than himself, and at the same time as a separate entity compared to other groups or individuals. Acting in a group can be positive or negative, and the outcome depends on the constantly changing context, personal preferences, and individual behaviour, which-in conjunction with the actions of others-can create new value that individuals did not even have before or even less foresee. Can this also be present between individuals? One could expect in such relationships that there is an equality of arms, where two persons can equally influence each other. The next section thus deals with the (in)direct influence of other people outside groups.

Equality of arms
In the dialogue between two individuals, their equality is usually evident, which manifests itself as a settlement, as a consensus of wills, where each concedes to the point of common agreement. In more protracted proceedings, one part may make concessions at the expense of the other, provided the reverse is the case in the next instance (tit for tat). The result of such cooperation is a mutual benefit; outside such cases, one side usually wins and the other loses, or they do not even engage in a "duel" with each other. These combinations are valid provided that two individuals are reasonable and at least approximately equal.

Uncritical compliance
Conversely, uncritical compliance with the orders of official or superior persons, i.e. institutions, over (unconsciously) subordinate people can be present. This was confirmed by Milgram's experiment (Milgram, 2009). Citizens, as subordinates, are less critical of orders or instructions from officials, and the latter sometimes behave inappropriately out of proportion in a power environment (Zimbardo's prison experiment, i.e. prison as a sadistic institution; Zimbardo, 2008).

The underlying systemic structure
We all love heroes, but according to American psychologist Zimbardo (2008) (known for the Stanford prison experiment), the banality of heroism is equally as likely as the banality of evil. Both occur in special situations where circumstances play the most important role for an individual, in his or her decision (not) to take measures or action. From a value-based view on system management, one needs to realise that there is "greater power to make evil from good within the System [rather than in the behaviour of individuals] as a complex of powerful forces creating the Situation. Ample evidence from social psychology supports the concept that situational power prevails over individual power in given circumstances" (Zimbardo, 2008, p. x). It matters how the systems are organized, how their underlying structure is formed, activated, and connected. They form the nature of information and show how the representation of information influences people's mental frames/maps and thus also decisions. As humans are also an example of a system and its embeddedness in other systems, the findings from the field of systems also apply to humans, to their input and output, to their processing, to their monitoring and their interventions in the environment according to their desired goals. It thus really matters how the systems are arranged. Their design and placement thus also affect the liability process itself.

People behave differently in a group than they do as individuals
The Romans knew this in the saying "senators are good people, it is the senate that is a mean beast" (senatores boni viri, senatus mala bestia). In light of the examples described in the previous two sections, the individual's actions in a group must be considered differently from those done in his isolation, since as a part of a group he takes some of the characteristics of the group that he does not have as an individual. Our socio-cognitive processes of perceiving and acting ourselves around others are different from those in isolation. If this is self-evident, the understanding of personal legal liability is nevertheless still individual-oriented: "an obligation or willingness to accept responsibility or to account for one's actions" (Merriam-Webster, 2021, emphasis added). Individual blame, responsibility, or credit, as known in law, for the actions (acts and omissions) of a group is not the most appropriate concept for understanding the "real" happenings in groups.

The relations produce new value
What happens in groups is a system of connections that create a new value that parts do not have in themselves. If we always refer to the context of a matter of interpretation, this is also the case with the role that social context plays in shaping group behaviour and attitudes. If this can be logical without too much difficulty, there is something else (the unconscious) at work, when people acknowledge this in the case of human actions. If this stands on a general level, it is not so far as far as liability is concerned. As a society, the people want to have a guilty individual, they want a "sacrificial lamb" (someone must be guilty) for actions that this individual would probably not have committed as the individual.

Barefoot law
Law, even at a theoretical level, has not yet developed the concept of group liability for acts committed by individuals as part of a group (e.g., as heroes, heroines, or as suicide bombers, martyrs). In such cases, the individual steps outside himself and acts in the context of broader collective goals that cannot be said to be positive or negative per se. In both cases, the perspective may be the same, i.e. the survival of another person, a group, an entire nation, its customs or religion, only the context is different. The converse is also true, that the law has not developed a concept of group responsibility for the individual.

Types of liability
Noncontractual liability for damages is divided into fault (subjective) and strict liability (liability without regard to fault). Whoever causes damage to another is liable to compensate for it, unless he proves that damage was caused without his fault. However, liability for damage to things or activities from which a greater risk of harm to the surrounding environment arises is imposed irrespective of fault. Fault is a fundamental prerequisite for subjective liability for damages when the injured party causes the damage intentionally or negligently. The closest to group liability is Article 136 of the Slovenian Obligations Code, which defines nonliable persons: "whoever, because of a mental disorder, a mental health problem or any other cause, is incapable of reasoning, shall not be liable for the damage he causes to another. Whoever causes harm to another in a state of temporary indiscretion shall be liable unless he proves that he was not at fault in bringing about that state. If a person has been put into such a state through the fault of another, the person who has put him into such a state shall be liable for the damage". More than unconsciousness, unreasonableness as a result of, for example, hypnosis, but also a compulsive force or a threat from a third party, is relevant here. The prerequisites for liability for damages include the unlawfulness of the conduct (a harmful fact), damage as a negative consequence, a causal link between the unlawful conduct and the damage suffered, and fault (intent or negligence and reasonableness). From what has been said so far, it is evident that for an individual to act as part of a multitude, it is particularly difficult to establish his or her fault, causation, and reasonableness. Between the individual and the group-as in all other two-part cases-there are four, not only two, combinations present: i) individual, ii) group, iii) individual-group, and iv) group-individual. In all four cases, the purpose and consequences of an action are different, and hence also liability and gratitude vary accordingly. In all four cases, the context is different, as social positioning creates various cultural, organisational, and institutional relations. In the absence of a clear demarcation between these four combinations, the entities of the individual and the group cannot be understood correctly. Otherwise, or in the present binary division, people find themselves in an amplifying spiral of delusions and departures from what is going on.

Individual-to-individual
In the first case, the individual is an independent person, completely left to and responsible for his own actions (entrepreneur, loner, egoist, contemplative thinker). Beliefs held by the individual are understood as his interpretation of the world (naive realism). In order to avoid liability, he rarely shares his real, deep sense of reality with other people (he does not tell everything he means); to calm his own self-esteem, he rarely questions his world view, and makes little effort to create a plausible alternative on his fault. Interactions with another individual are seen as a social influence that is "equally powerful in both directions" (one person can influence another and viceversa) with the more persuasive person winning (with no change of mind), using foot-in-the-door, lowball and door-in-the-face techniques. The liability for damages is well regulated here, while public liability is not: for example, the Rome Statute of the International Criminal Court states that the Court has jurisdiction (only) over natural persons (Article 25). Similarly, the Criminal Code sets the limits of criminality for perpetrators and participants, treating each person in his or her individuality, within the limits of his or her intent or negligence, as well as personal relationships, characteristics, and special circumstances that by law exclude guilt or criminality or permit remission, reduction, or increase of the penalty that may be taken into account only with respect to the offender or participant in respect of whom such relationships, characteristics, and circumstances are established. Liability for acting in a group is treated here as an act of each individual, or as his contribution to the common conduct (aiding, abetting, complicity-anyone who incites, assists, or commits an act with another by knowingly participating in the commission of the act or otherwise contributing decisively to the commission of the act is to be treated in the same way as the perpetrator himself). For acts committed by an individual, the law provides an effective means of enforcing his or her liability, so no additions are needed here.

Group-group
In the second case, the group acts as a well-oiled, cohesive machine (a group of firefighters, soldiers, policemen, athletes, paramedics), but also as a chaotic crowd, which through its interests also pursues communal goals (Smith's invisible hand) against another group; it also manifests itself as a degree of mental or other development, a common good denominator (teamwork, cohesion, bureaucratic organisation, acting based on group values) but also a bad result (herd mentality, groupthink, group polarization, path dependence, enclaves of like-minded people or a shared illusion of unanimity, the banality of evil); responsibility in the legal sense is not wellconceived here, or developed. This is because the convictions that one has in a professional group are reinforced in proportion to the similar opinions of the group members and the longer the practice has been carried out (in the style of "that's the way we do it"), and the same applies to membership in other groups. This second example is a typical case of inter-institutional behaviour of several public authorities or public/private institutions, where, based on many interactions between individual authorities, public officials and inter-alia, a new value is created that no one can fully own alone. As a result, even in the event of negative consequences, no one feels responsible or shifts responsibility to other people or bodies.
In the case of group liability, the law is familiar with criminal law 13 and the liability of legal persons for damages, which (in the case of criminal liability) does not apply to certain legal persons governed by public law (a county or local authority) and to public bodies, as these are not legal persons. As regards liability for damages, a company is liable for its obligations with all its assets, and the law also provides for the so-called "disregard of legal personality", while the state or its institutions are also liable for damages, with certain specificities about individuals. The liability of groups can also be political, in particular, when the electorate punishes a political option for the actions of its members.

Possible solutions
In cases of large-group behaviour, it is difficult, if not inappropriate, to hold individuals individually responsible, as individual actions are always relationally caused by their behaviour as part of a group. However, negative consequences can be prevented and positive ones improved by having a systemic regime that works based on indicators and continuous monitoring (data-driven policy), which records deviations and makes necessary corrections (we could call this a system of "living or self-driven, responsive authority"). For the right degree of liability and legitimacy of groups, good knowledge of the basic features of group functioning is needed beforehand; until these findings are translated into legal frameworks, other solutions from other scientific disciplines can be used. Despite the fact that decision-making is the lifeblood of every organization and the central focus in the practice of evidence-based management (OECD, 1996(OECD, , 2011(OECD, , 2012Rousseau, 2013) 14 attempts are still made to solve problems in the classical regulatory frames developed in the 18 th -19 th century, notwithstanding the known methods developed in other fields like the mathematical Bayes theorem (Ayres & Nalebuff, 2015;Carrier, 2012;Finkelstein & Fairley, 1970;Hacking, 2001), statistical sampling (Dodge & Romig, 1959;Fiedler & Juslin, 2006;Schilling & Neubauer, 2009;Wetherill, 2013), decisional analysis (Edwards, 1999;Edwards et al., 2007;Keeney & Raiffa, 1993;Raiffa, 1994), and risk analysis (Black, 2012;Franklin, 1998;Hood et al., 2001;Molak, 1996;Slovic, 1996), and the already mentioned collective intelligence. In the latter context, in particular, we could talk about the concept of a "living" information platform, which could be managed by the decision maker, with real-time and spatial participation of other relevant institutions (real-data driven policy), not only on all potential negative forms of work but also on all matters requiring public opinion. Based on the findings on group behaviour and the influence of other people on individual behaviour presented in the second and third sections of this article, a new concept of holistic group liability could be introduced, along with the existing concept of collective liability, which still too often treats the individual as a mere part of the group.

Holistic group liability addresses new values that (spontaneously, unconsciously) emerge from relations among the group's members, the context of time and space, concepts, needs, ideas imagined, real consequences and other universal elements of groups or networks that lead individuals to act differently than they would as individuals outside the group.
This is where social psychology should first come to a common understanding of what this new concept entails, to be able to implement it in the liability regimes. 15 As holistic collective liability emerges in relations, the latter can be better managed through a system that records these relations and the unified opinion of the (expert and general) public that would provide a relevant response to these relations. Both can be done through a platform that would be based on proactive work that could prevent negative consequences in the relevant area and establish ways to address specific forms of behaviour. This kind of operations room and/or an institution's cockpit (a control centre 16 ) could strike a balance between institutions and freedom (between diminished economic justice and civil liberties), through regulation, i.e. through a platform that seeks to follow Polanyi's fundamental idea that "in a well-ordered society, the right to non-conformity must be institutionally secured" (Polanyi, 2008, p. 366). Integration of the different aspects of the implementation of activities in a single IT tool is only a matter of time; it should be supported from the definition of the problem to its solution in the end. The real value would be in establishing the relationship between factors that can be controlled and those that cannot be; this would be followed by setting up a way of monitoring controllable factors (as the uncontrollable ones cannot be controlled). Liability cannot be controlled (it is determined by the rules), but we can control to a greater extent the practice that constitutes the actual state of legal norms concerning these forms of liability. The creation of an IT tool would allow an ongoing evaluation of effects and a feedback loop in the system that would secure the appropriate and rapid actions to prevent liability and potentially to change the rules if needed. A social problem can best be solved within society.
To this end, an information platform could seek common patterns of action through information interactions from many sides and at many levels. Through cooperation, soft boundaries, and information flow, this could be a suitable approach to increase the level of democracy in the process of collectively understood public services. It seems that at first an underlying systems structure-which also fits the mentality of the highest decision-makers-has to be changed before changes can be implemented.

Individual-Group
In the context of two groups, social influence is exercised when one part depends on the other to the extent that the superior reasonably anticipates the subordinate's modus operandi. Social power can be present in the relationship between the individual and the group, and vice-versa. In the third case, the individual acts for the group or other people that he has in mind (benefactors, philanthropists, politicians, civil servants, heroes, martyrs, influencers, or officials); he acts based on personal virtues as well as the official authority. The law here relies on the classical system of liability (because it is still supposed to be about active individuals) and on the legal institutes of rewards, recognition, force majeure, or mitigating circumstances. It is characteristic of a hero, a saint, or a martyr to be an ordinary individual outside of "exceptional circumstances". Their "ordinariness" is therefore not the appropriate level to understand their actions in extreme situations (but it is appropriate in the first case of the individual as the individual); the appropriateness of understanding them is possible only by looking at the group or other people. In this role, the individual is still fully identifiable and is understood in the fullness of his rights and obligations-it is his actions that are done with a view towards the wider whole or community of people.
According to the Slovenian Obligations Code, a fault is established when the injured party causes the damage intentionally or recklessly. In the context of shared liability (Article 171), if the injured party has contributed to the damage or caused it to be greater than it would otherwise have been, he or she is entitled to a proportionately reduced amount of compensation. Liability for another is regulated only for persons with mental disabilities and mental health problems, parental liability for the acts of children, the liability of others for a minor in the case of supervision by a guardian, school, or other institution, and the liability of employers. 17 In labour law, tort law, in particular, the Roman institute of respondeat superior ("let the master answer") is also known, which is nearly closest to the role of the individual within the group/institution (a similar institute was used by the Nuremberg war crimes tribunal, known as command responsibility). A principal/leader is liable for the wrongful acts of a subordinate worker if these acts happen in the framework of the former's directions (qui facit per alium, facit per se-what is done through another, it is done by me). Usually, an employment relationship must be established and the act must be done within the scope of employment, while the various national legal arrangements to be applied to this institute also vary.

Group-individual
In the fourth case, the group acts with the individual in mind (social transfers, institutional assistance to individuals in need, education, social welfare, legal protection, protection, but also automatic data processing, algorithmic decision-making on people's rights, facial recognition, motion recognition; as in the third case, the responsibility of a group or groups of people for actions against the individual has not yet been developed). If the third example of the individual as a hero or martyr from the point of view of the group is known from ancient history, the fourth example of the group with the individual in mind is, on the one hand, more familiar, and, on the other hand, this very familiarity-even though it is in plain sight-may make it potentially dangerous: if social transfers, institutional assistance to individuals, the rule of law, and the welfare state are familiar to people, this does not yet tell us anything about the effectiveness of such an arrangement. Similarly, the development of information technology hides new pitfalls-the collection of large amounts of data and the resulting data banks, the development of computing power, and the progress of algorithms are leading society in at least two directions (again, we actually have four combinations): any development of a new thing (such as the primordial discovery of fire) can be used for good (here too we are dealing with side-effects) or for evil (even a bad thing can be good for something) purposes. Just as the Industrial Revolution led to a great leap forward in human development, it also led, on the other hand, to the exploitation of factory workers and the impoverishment of the rural population at the expense of entrepreneurship and banking. Since everything, no matter how good, has undesirable side-effects or unintended consequences, this is likely to continue to be the case in the future as far as algorithms, artificial intelligence, machine learning, and other discoveries are concerned. The only difference with the past could be that we may be more aware of this, that many processes can be monitored, measured, and reacted to (the good or slightly less good Covid-19 measures are a typical example of this).
Among specific cases of liability, the Slovenian Obligations Code 18 includes liability for acts of terrorism, public demonstrations, or manifestations, 19 and the liability of an event organiser. 20 Joint and several liabilities, or the liability of several persons for the same damage, is also stipulated, 21 where each individual, as part of a group, is liable to pay a full amount of damages if it is not possible to determine their respective shares of the damage caused. Regarding the recovery of the payer, under Article 188, joint and several debtors who pay more than their share of the damage may require any other debtor to reimburse them for what they have paid for it. The amount of each debtor's share shall be determined by the court according to the gravity of his fault and the gravity of the consequences resulting from his actions. If the debtors' shares cannot be ascertained, each shall bear an equal share, unless justice requires a different decision in the particular case. The Obligations Code refers to "taking into account the circumstances of the case", "ascertaining the share of the damage caused", "the gravity of the fault", "the gravity of the consequences resulting from the action", and "equity" regarding the recovery of damages for actions in a group, but these concepts are not well developed in the context of holistic collective liability. This is true also for the responsibility of States for acts otherwise committed by their (individual) official institutions or employees. The European Convention on Human Rights thus allows that the exercise of certain human rights can only be restricted by the law if this is necessary for a democratic society for reasons of national security or public safety, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others. As signatories to the Convention, member states have committed to respect the final judgment of the Court in any case in which they are parties (who are liable to an individual plaintiff for a violation of his or her human rights), even if, in reality, the acts in question were committed by their (individual) official institutions or employees. In any event, the act in question is one committed in the exercise of official authority, which is why the individual is not acting in the capacity of an individual, but in the capacity of an official.

Dialogue
Social control is used to control social power. Hegel's master-slave dialectic (Hegel, 1998) illustrates that it is not self-evident and necessary that one consciousness recognises the other: the first imposes itself on the other. For fear of death, the other recognises the first as master and the self as a slave. This creates a form of consciousness, but not its freedom: it can only be realised through the other's self-consciousness, through the reciprocity of self-consciousnesses, when the consciousness of the other is recognised as equal. Here are hidden the first means of control: mutual communication, dialogue. As reciprocity is not enough to create self-awareness, communication also has its pitfalls (the aforementioned pitfalls of information stunts, persuasion techniques): social psychology has shown through numerous experiments that our beliefs are influenced by several behaviours that people consent to because they have been asked to do something by a person who has at least a modicum of authority, and that consent demands to calm down our consciousness (this is somewhat less pronounced in the case of submission, where the individual changes his or her mind in public but stays with his own internally, and more pronounced in the case of conversion, where the person changes his or her mind in private, under pressure from a group leader). This is known as cognitive dissonance. Dialogue is genuine when the two speakers base their dialogue on the rules of communication, establish facts, make assumptions from them, and implement and acknowledge the arguments. The ability, the style of good communication, of rhetoric is particularly important here. It is therefore productive to direct people to think more independently/on their own, in the style of an adapted Russian saying, doveryay, no (inogda) proveryay (trust, but [sometimes] verify). 22 Just as the underlying structure of a system that creates the context or circumstances can create a hero or a villain, it can also help to create clear relationships between individuals in a group: we are closer to this if communication is top-down and vice-versa (the top, the leadership of the group and each individual in it), inward (relationships within the group) and outward (relationships to other groups), and relationships are better defined through the accessibility of information, the participation of members, and their joint decisionmaking. If speakers have genuine expectations, they are willing to agree and make win-win decisions, this can be seen as a form of social control.

Nonconformism
Regarding nonconformity, cognitive dissonance theory is relevant, according to which people who are confronted with some form of subordination (and thus tension-dissonance) eventually adopt new beliefs or adjust their own beliefs according to their own (subordinate) behaviour (Festinger, 1957). Such tension is paradoxically reduced by the superior who "grants freedom", who gives the subordinate the choice to "freely" decide whether or not to follow the superior's request. The subordinate is here even more personally engaged: "if agents endowed with authority treat the freedom of subordinate agents in the right way, then this not only does not turn the subordinate agents into autonomous beings, but it even leads them to regard their subordination as a value" (Beauvois, 2000, p. 15). Most people think they can resist peer pressure; yet remaining independent against the will of the majority is harder than people think. Conformism is acceptance and adherence to social or group norms for personal comfort or gain and has its roots in the concern for one's survival, for one's popularity, for finding a mate, for the preservation of the species, and is especially relevant when emotions are active. This could be one of the reasons for uncritical (or at least less critical) acceptance of peers, colleagues, and superiors, even for taking orders under threat of fear of something bad. Categorical refusal of orders is also counterproductive, as is, on the other hand, group cohesiveness leading to the immutability of common opinion. The middle ground is usually somewhere in between.

System's structure
The importance of the underlying structure that "provides situations" and the IT platform that could administrate it was already mentioned. Despite the many examples of conformist behaviour, some individuals nevertheless did not emulate most of the conformists in Milgram, Zimbardo, and Asch experiments. This nonconformism can also be understood as a failure to agree with apparent arguments, as a form of genuine communication. What is the reason that nonconformists did not act like Eichmann? If the difference between a hero or a coward is perhaps less in light of the suddenness of the situation, but perhaps in the underlying context 23 and especially the system in which persons find themselves, then it could be valuable to give more emphasis to the underlying system's structure. A system that perceives relationships, shows trends, "catches/provides context", and corrects behaviour in line with underlying goals can be seen as a form of social control, but it can be also a form of social development. In addition to the Zimbardo situation and/or the system mentioned, the new science of networks is also focused on the structural properties of a networked system (Newman et al., 2011) that has its own rules of behaviour.

Directions for research on holistic group liability
Interest in the structural properties of networks has shown some supposedly universal (regardless of network) properties, where the network begins to exist independently of its parts; the same is true for the social structure and individuals in it. Individuals within a crowd are drawn to it and follow its dynamics. The crowd's characteristics cannot be explained by looking at the individual. Crowd phenomena occur more often in situations of uncertainty in the absence of knowledge or information. On such occasions, people on average support a positive decision in otherwise value-neutral situations, not in the ratio of 50:50 but of 62:38, which is the Fibonacci ratio Phi (0.618; Prechter, 1999). People respond more to narratives than to facts and data; the more compelling the story, the more it erodes critical-thinking skills (Green & Brock, 2000). Malik (2010) among the general characteristics of crowds enumerates the unity of mind, direction or goal, emotion over reason, biological and sociological foundations of crowds, and situations of uncertainty-if there is not adequate knowledge of a certain situation or when conditions are very unstable, people tend to herd and thus follow what others do. There is a new science of social prediction called Socionomics (Elliott, 2019;Prechter, 1999), focused on human behavioural patterns that can predict the course of society more accurately. It bases social trends on an endogenous human herding tendency manifesting itself in a predictable pattern called the Wave Principle.
Some networks also do not exhibit a bell curve but a power-law degree distribution, where some nodes have exponentially more connections than others (heavy-tailed degree distributions), and/or where the numerous tiny events coexist with a few very large ones (Barabasi, 2002). Here, the preferential attachment process ("the rich get richer" or the Matthew effect) or self-fulfilling prophecies are also situated. When people know what other people are doing (e.g., download popular songs), they can follow them regardless of their taste or inclination. Even when most of an individual's cares, aversions, skills, manners, opinions, anticipations, and ideas are known, this still would not suffice to predict much about collective behaviour (Watts, 2011). Despite the highly clustered world, everybody on this planet is separated by only six other people (six degrees of separation; Watts, 2004Watts, , 2018, known as the small-world effect. Small-world networks are the class of networks in which any node could be reached from any other on average in only a few steps. In the presence of triads (sets of three nodes), "the relationship between any two nodes is potentially (and frequently actually) affected by the existence of other nodes and by the presence of ties flowing to and from those nodes" (McLean, 2016, p. 31); this holds even if these nodes and ties connect people.
In addition to all the differences mentioned above, the liability of group actions and thus of individuals in it can be-having the underlying system/network structure that acts differently than individuals as particular units, or regardless of them, to which unconscious, herding people's acts could be added-examined in further research. Possible paths could be to focus on the types of cooperation in groups: two forms of cooperative behaviour (discretionary-people engage in behaviour that is not directly required by the rules or norms of group membership, and mandatory-people engage in behaviour that is required by the group rules or norms), two functions of cooperative behaviour (the difference between the performance of desired and undesired behaviours by group members), and two types of motivation for behaviour (the basic human desire to maximize gain and minimize the loss in interactions with others and internal motivation, the latter being composed of attitudes and personal values; Tyler & Blader, 2013). There is also a degree/ difference: (i) to which a group's decisions lead to favourable outcomes for all members and ii) contrary, iii) to which a group's decisions lead to favourable outcomes for a particular member and iv) contrary. This probably also affects the group's results. Regarding the subjective probability of a group member, his or her probability or willingness to cooperate can be assessed with the help of Bayes theorem (i * ii/i * ii + iii * iv) if the mentioned four parts were/are stated in numbers. Groups also probably behave differently than the individual when results and procedural justice (the fairness and the transparency of the processes by which decisions are made) and distributive (fairness in the distribution of rights or resources, criteria, frame, system) and retributive justice (fairness in the punishment of wrongs) are considered. Individual behaviour in a group also depends on the attractiveness, unity, size, status, expertise, and rituality of the group, identification with the group, i.e., social attractiveness (identification with a specific group or institution compared to other groups or institutions), interpersonal attractiveness with other members of the group, insecurity (people are more prone to look to and accept the actions of others as correct in the face of uncertainty), and similarity (people amplify their meanings/actions when they observe similar meanings/actions of other people, maybe even more so when other people are similar to us). Based on the mentioned differences between the individual's acts when he acts alone and when in the group, causation and fault (intent and negligence)-as the necessary elements of subjective liability-are put under a heavy test (there is an open question whether they exist as such in groups and/or how they can be established [or replaced with something else] in the first place).

Conclusion
People act differently in a group than they do as individuals. These are various actions that would not be done as individual acts. The individual's role in the group is interdependent with the role of other individuals, so the individual should not be considered only as an individual. If this is confirmed with experiments in social psychology (some of them were also mentioned in this paper) and it is also clear to us as a matter of logic, the latter is undermined with a liability regime that still treats the individual as an individual, irrespective of his or her different role in the group. Regarding liability for acts committed in a group, demands such as "take into account the circumstances of the case", "establish a share in the damage caused", "the gravity of the fault", "the gravity of the consequences flowing from the action", and "justice" are indeed mentioned, but they can serve well to enforce collective liability only when social psychology can well establish the individual's role in the group. A fundamental ethical rule of research is that the integrity and dignity of the people being researched should not be compromised. The same should apply for legislation, regulations, rights, and duties when individuals would be liable according to their real role in the group when their subjective mental conditions are better determined. In regard to holistic group liability that ex novo emerges from relations, we have not come much further than Roman law, where it was known that the actions of one individual could affect the whole group, where the individual is still understood (only) as an individual. If the division or combination between the group and the individual is not resolved in the future, the status quo will continue. 24 It is not a question of advocating a greater or lesser degree of liability; it is about a different kind of liability that follows the actual role of the individual in the group, a liability that considers the relative weight of the individual in the group vis-à-vis other individuals, a leader, ideas, contexts, mental states, and consequences. The role of the individual may be small, but the problem of the butterfly's wing effect as a precursor to the storm is also initially in its insignificance and invisibility. When we look at the approaching storm, we no longer think of this effect and look elsewhere for reasons. Therefore, what is the individual's role in the group that can cause social movements and what are other supporting elements here? The insignificance and invisibility of individual actions can eventually turn into a real social movement. It is also important to be aware that such individual acts can create errors and side effects in person's actions, which must be critically evaluated in light of other possible alternatives. The role of the individual is different in the group, beyond or below the proportions of his or her personality and character. It is first a question of calibration of the role of the individual in the group (and beyond), and then the question of the most appropriate liability. Only then can an appropriate answer be given to Freud's question from the paper's introduction, only then can liability demand what is legitimate, appropriate, necessary, and realistically proportionate between the benefits and the harm caused. This is, after all, a fundamental requirement of the legal principle of proportionality.

Funding
The Slovenian Research Agency, Grant/Award Number: Grant No. P5-0093 (Development of an efficient and effective public administration system).

Disclosure statement
No potential conflict of interest was reported by the author(s).

Notes
1. It can be a long debate on the differences between responsibility, accountability, and liability, but for the purposes of this paper, responsibility means duty to respond, accountability duty to report or answer, and liability duty to compensate. The paper's focus is on the latter, while it is tightly connected with the second in the form of legal (not political) accountability.
2. To what extent does group membership decrease cognitive or rational abilities of individuals and increase irrational reactions? To what extent does it reduce an individual's sense of identity, individuality, and autonomy and lead to anonymity in a group? To what extent do social norms, the group's goal, and the context of the situation influence the individual's behaviour in the group?. 3. Emotions are formed in the amygdala (a small, two-sided structure in the temporal lobe of the brain), which causes emotional learning and evaluation, particularly in provoking emotional responses to fear or aggressive stimuli (Adolphs et al., 1998). This involves all those reactions that people have even before realising what is good in terms of their self-preservation instinct and evil in terms of maintaining stereotypes and discrimination that are located in the amygdala (Phelps et al., 2000). Actions in a group are probably more subject to emotion than to reason. 4. Personal engagement is, along with using reason, also always connected to emotion; in fact, "there is no inner contrast between emotion and reason: emotion is the main source of motivation steering us towards certain goals; it can steer a considerable power of reason towards the goals that it gives rise to" (Simon, 1997, p. 91). 5. Given this, it is not surprising that Lenin, Stalin, Mussolini, and Hitler embraced his concepts of hypnotic action, anonymity, and the collective mind (Bond, 2014). 6. His assertion that the crowd is always inferior whatever its composition is also interesting: "in a crowd people are always evenly matched, and if it were a general question, the vote of forty academians would be no better than that of forty water-carriers [. . .] as far as social problems are concerned, in which a number of unfamiliar aspects are intertwined, the people are all, more or less, equally ignorant" (Bon, 2016, pp. 183-184). 7. Commands are older than speech. If this were not so, dogs could not understand them. Animals can be trained because they can be taught to understand what is required of them without understanding speech (Canetti, 1981, p. 303). 8. Multiple perspectives give a better view of all things. Try to see with one eye and then open the other-this is the difference we are talking about; this difference is what makes a community of people. If we see each thing as separate and divisible, as a function, an organisational unit, an individual's knowledge, skills, and competences, it is harder to see solutions based on holism, on bringing the parts together into larger units, groups, into the bonds that bind us into a family, a living community, an organisation, a country. This is what I become when I start to realise that I am part of something bigger than myself. 9. Nudges are initiatives that maintain freedom of choice while also steering (by choice architects) people's decisions in the right direction (from setting a printer's default options to providing organs for transplant operations; Sunstein & Reisch, 2019;Thaler & Sunstein, 2008). 10. Error of perception results from a lack of attention to an unexpected object, so it goes by the scientific name "inattentional blindness": we experience far less of our visual world than we think we do (Chabris & Simons, 2010). 11. The whole is not only greater than the sum of its parts, but the nature of the whole fundamentally alters the parts. Asch stated: "Most social acts have to be understood in their setting, and lose meaning if isolated. No error in thinking about social facts is more serious than the failure to see their place and function" (Asch, 1952, p. 51). 12. Bureaucracy or the rule of an intricate system of bureaus in which no men, neither one nor the best, neither the few nor the many, can be held responsible could appropriately be called the rule by Nobody. If, in accordance with traditional political thought, we identify tyranny as a government that is not held to give an account of itself, rule by Nobody is clearly the most tyrannical of all, since worker to a third party at work or in connection with the work, unless it can prove that the worker acted as he or she ought to have acted in the circumstances. The injured party may also claim damages directly from the worker if the worker intentionally caused the damage. Anyone who has compensated the injured party for damage caused by the worker intentionally or through gross negligence shall be entitled to recover from the worker the amount paid. 18. Official Journal of the Republic of Slovenia, No 97/ 07. 19. Article 156: for damage caused by death or personal injury as a result of terrorist acts, and in the event of public demonstrations or manifestations, the country or the person who should have prevented damage under the law is liable. 20. Article 157: The organiser of a meeting of a large number of people, whether indoors or outdoors, shall be liable for damages for death or personal injury resulting from extraordinary circumstances that may arise on such occasions, such as the movement of crowds, general disorder, and the like. 21. Article 186 (1) All persons involved shall be jointly and severally liable for damage caused by several persons.
(2) The instigator, the aider and abettor, and the person who assisted in preventing the persons responsible from being discovered shall be jointly and severally liable with them.
(3) Those who caused the damage but acted independently of each other shall also be jointly and severally liable for the damage caused if it is not possible to establish their respective shares in the damage caused.
(4) Where there is no doubt that the damage has been caused by one of two or more specified persons who are in some way related to each other, but it is not possible to determine which of them caused it, those persons shall be jointly and severally liable. 22. The same applies to good supervision (quis qustodiet ipsos custodes-who will supervise the supervisors) and an organisational culture that encourages a diversity of views. 23. The notion of context is relevant, but it could be at some point similar to "do no harm"-it does not bring us closer to the causes of one's (non-) behaviour. Perspectives can be better known in the context of other things, and thus also future doings (as imagined consequences), relations, and their combinations in an event in which actors find themselves (if they recognise it as such-or are just the mute observers of "someone helplessly lying on a pavement"). 24. The Roman jurist Seneca illustrated this by saying qui non vetat peccare, cum possit, iubet (He who does not forbids not sin, although he is able to, commands; Stojčević & Romac, 1989, p. 429).