Ethical Issues of Humanoid-Human Interaction

Introduction

What is a humanoid? The suffix -oid means 'similar' or 'resembling.' The etymological root of Greek -oeides means 'having the form of' (eidos), 'like that of, thing like a__'. The Latin suffix -id means 'offspring of'. 'Descendant of' is used, for instance, in taxonomic names in biology (1). A humanoid is thus an artificial entity that looks like a human being as distinct from hominids, our natural predecessors. Do humanoids resemble more a man or a woman? This issue was discussed in the EU Project ETICA leading to the following conclusion:

"The Literature on Robotics is dominated by the term "android", i.e. the male version of humanoids and it rarely relates to "gynoid", the female version of robots. Also, there is much controversy about producing "gynoids" for sexual purposes, promoted on a market as a kind of "sex-aid". Similarly to other ICT, the field of Robotics is assumed to be the one in which gender aspects may be regarded [as] unimportant. Robotics designers and producers believe that technology is gender neutral. In this way, they neglect the fact that a large segment of consumers of these products are females who may have different expectations and needs when it comes to the use of Robotics." (2, p. 86)

According to Massimo Negrotti, director of the Laboratory for the Culture of the Artificial at the University of Urbino (Italy), artificial technology is a relation between an "exemplar" and its "representation" conceived as a process of selection based on an "observation level" of some "essential performance" (3 pp. 46-47, 4, 5). This makes a difference to conventional technology that generates machines "by combining and recombining components and sub-systems whose raison d'etre [sic] and consistency depends only on the design itself and not on the structure of the world" (3, p. 34). For conventional technology, nature is something to be dominated while artificial technology aims at reproducing a natural exemplar. Negrotti writes:

“The artificial, in conclusion, is conceived and present in the first phases of its existence as a naturoid, an object achieved by man and oriented to some natural exemplar as it is seen at a given observation level. However, it soon becomes, or reveals itself to be a technoid, that is to say, it becomes an object that exhibits characteristics which exceed those of the exemplar and either strengthens, reduces or somehow transfigures some of these, as if it had to redraw the exemplar not as it is but as it should be.” (3, p. 47)

What is a human being? According to a traditional definition, a human being is a rational animal (animal rationale). But what is human rationality or, to consider the term commonly used in computer science, human intelligence? What kind of rationality or intelligence as a form or eidos of human beings 'in-form' artificial beings? (6). And what other ‘essential performances’ of humans can be selected when creating humanoids? A tentative answer to the first question would take us on a long journey through philosophy, psychology, history of logic and the analysis of scientific rationality, to mention just a few fields. Intelligence is a question for intelligence itself. This is at first sight disappointing. Looking for unquestionably true answers is begging the question. It presupposes the answer of what it is asking for, namely that intelligence is about unquestionable answers. Last century's philosophical debate on language, emotions and rationality ― Ludwig Wittgenstein (1889-1951) being one prominent example of this debate ― has shown the limits of this concept of intelligence whose origin goes back in the Western tradition to Greek philosophy, through the rise of empirical science up to the debate on artificial intelligence, the imagining of humanoids in art and literature and "how we became posthuman" (7). Nothing is probably more controversial than agreeing on what is specific to human beings, not only with regard to other living beings but also to artificial entities among which humanoids are a prominent case. We must keep this long-standing debate in mind when dealing with humanoids resembling the human body and performing some kind of human-like intelligent actions and emotions, giving rise to ethical issues regarding humanoid-human interaction. We start with a short historical review on human intelligence and humanoids and look at some current ethical issues concerning humanoid-human interaction in the second part.

1. Human Intelligence and Humanoids

Human intelligence or knowing has knowledge or the known as its correlate - noesis and noema in Husserl's (1859-1938) terminology (8; 9). The origin of this correlation goes back in the Western tradition to Greek philosophy. Aristotle (384-322 BC) distinguishes between two kinds of knowledge (dianoia): dealing with truth and opinion, respectively. Knowledge dealing with truth can be technical knowledge (techne poietike), scientific knowledge (episteme), practical knowledge (phronesis), knowledge of the first principles (sophia), and intellectual reasoning (nous), while conjectural knowledge (hypolepsei) and opinion (doxa) deal also with the wrong (10, 1139 b 15-18). This typology can be roughly paralleled with the customary distinction between know-how, know-why and know-what (11). An important difference between Aristotle's typology and our prevailing view of truth-seeking knowledge concerns the conjectural or fallibilistic nature of science as analyzed, for instance, by Karl Popper (1902-1994) (12). Conjectural knowledge undermines the theoretical ambitions to reach unquestionably true knowledge. Furthermore, knowledge is embedded in human action. Aristotle emphasizes that knowledge does not by itself move anything, neither with regard to human virtue-oriented action (eupraxia) nor to the production of material things (poiesis), unless there is a striving combined with reasoning or a weighing-up of various possibilities (10, 1139 a-b). Weighing up possibilities of action through deliberating about the best means for achieving good life (eu zen) is based on customary rules and individual character (ethos) that implicitly or explicitly pervade social life. Human self-understanding in ancient Greek society, as represented in tragedy and reflected in philosophy, has as its core the dependence of human knowledge and action on divine actors and necessity. It changes throughout history and in different cultures. Note that to follow readily and spontaneously (hekon) a rule, in the case of classic Greek civilization, differs from what many centuries later became the will of the autonomous subject (13, pp. 41-74).

In Modernity, the human subject was conceived as separated from the world and the other human beings outside the consciousness of the individual. According to René Descartes (1596-1650), we would have no way of finding the difference between an animal and an animal-like "self-moving machine" (automates); however, this would not be the case for machines imitating humans as we have two tests that can be used in order to ensure that machines do not act by knowledge (par connaissance). The first test deals with human communication. Human-like automata may be able to use words and even perform some "bodily actions" (actions corporelles), but they would be unable to arrange words properly in order to answer meaningfully what a human being says, which is something that even the most dull persons (hébétés) can do. The second test concerns the bodily organs, which could not be artificially shaped in such a way that they allowed to behave adequately, i.e. reasonably in every situation (14, pp. 164-165). The first argument was contested by Alan Turing (1912-1954) in his seminal 1950 article "Computing Machinery and Intelligence" (15). It is symptomatic that the Turing Test consists in guessing not only about human intelligence but also about the gender difference. This takes us to the second Cartesian argument that contradicts the current understanding of human intelligence as a social phenomenon as well as intrinsically related to the body. Indeed, intelligence is intrinsically related to emotions and both are grounded in the body. According to Michael Polanyi (1891-1964), this bodily relation generates "tacit knowing" which is the basis of objective or explicit knowledge (16, 17). Phenomenology and hermeneutics have stressed the role of foreknowledge, the original belonging together of feelings and understanding and how bodily knowledge guides our actions dealing with tools in familiar environments. The Cartesian dualism between mind (res cogitans) and body (res extensa) is taken over by builders of AI (Artificial Intelligence). Hubert Dreyfus (18, 19), Terry Winograd and Fernando Flores (20), Don Ihde (21, 22) and Peter-Paul Verbeek (23), to mention just a few, have further developed the paths of phenomenology and hermeneutics in relation to technology. Human intelligence as embodied intelligence is different from mimicking human intelligence in algorithms as well as human emotions in a human-like robot: What humanoids are, depends on the social and cultural context in which they are created.
In the following we shall deal with two contexts, art and labour.

1.1 Humanoids in the Context of Art

Classical Greek and Roman art is an example of the sculptural and pictorial representation of anthropomorphic or human-like gods and goddesses and theomorphic or god-like humans. Although this is also the case in other cultures and epochs – including zoomorphic or animal-like representations of gods and goddesses, for instance, in ancient Egypt, pre-Columbian America or India – what makes this tradition particularly interesting is that originals representing human-like gods or god-like humans were reproduced in a "serial, iterative, [and] portable" way that made them ubiquitous in the ancient Western world; although no more than 2% of them survived until today (24, p. 68). It is not only the external shape of humans that is imitated (mimesis) by Greek art but, more substantially, human action (praxis) is imitated in Greek drama. The history of robots is closely related to the history of puppets and marionettes influencing the creation of robots, for instance, in Japan (25).

There is a four hundred years old tradition of humanoid automata during the Arabic-Islamic Renaissance (800-1200), among them al-Jazari's (1136-1206) Elephant Clock, the Beaker Water Clock, Automation for Carousals with mechanical slave girls playing flute, harp, and tambourine and a helmsman steering the boat with the rudder. Many of the automata created by al-Jazari were built according to Ayhan Ayteş:

"to the glory of God and to the honour of Allah's powerful representative on Earth. Many of his devices, however, had profane uses that competed with divine omnipotence. A superficial function of many was that the hydraulic automata served the purpose of making the guests at festivities drunk as quickly as possible." (26, p. 112).

This was long before Jacques de Vaucanson (1709-1782) created “The Flute Player”, Pierre Jacquet-Droz (1721-1790) “The Writer and The Musician” and Wolfgang von Kempelen (1734-1804) “The Turk” and the speaking machine.

In the 19th century, the French manufacturer, Ferdinand Barbedienne (1810-1892), invented a machine to produce miniature bronze replicas of famous antique as well as modern statues, particularly by Auguste Rodin (1840-1917), making them accessible to the bourgeoisie. With photography and other reproduction technologies, works of art lost their "aura" as Walter Benjamin remarked (27); although this was already the case with serial reproduction in Antiquity. The theomorphic transfiguration of human intelligence based on computer technology has been a subject of art, science-fiction literature and movies as well as of scientific projects throughout the 20th century, including the present idea of singularity, i.e., the emergence of a technological super-intelligence having predecessors in philosophy, myth and, religion (28, 29). It is not unreasonable to predict that humanoids will lose their aura created in movies and novels and become part of everyday life in the 21st century as happened, for instance, with radio, telephones, cars, computers and other advanced technologies. Norbert Wiener gives the following brief account of the history of automata:

"At every stage of technique since Daedalus or Hero of Alexandria, the ability of the artificer to produce a working simulacrum of a living organism has always intrigued people. This desire to produce and to study automata has always been expressed in terms of the living technique of the age. In the days of magic, we have the bizarre and sinister concept of the Golem, that figure of clay into which the Rabbi of Prague breathed life with the blasphemy of the Ineffable Name of God. In the time of Newton, the automaton becomes the clockwork music box, with the little effigies pirouetting stiffly on top. In the nineteenth century, the automaton is a glorified heat engine, burning some combustible fuel instead of the glycogen of the human muscles. Finally, the present automaton opens doors by means of photocells, or points guns to the place at which a radar beam picks up an airplane, or computes the solution of a differential equation." (30, pp. 39-40)

1.2 Humanoids in the Context of Labour

What are servants and slaves other than organoids or tool-like humans, excluded from some "essential performances" that belong apparently only to a special group of society? In a foundational text of Western thought on the relationship between masters and slaves Aristotle writes:

"Property is a part of the household (oikias), and the art of acquiring property (ktetike) is a part of the art of managing the household; for no man can live well (eu zen), or indeed live at all, unless he be provided with necessaries. As in the arts which have a definite sphere and must have their own proper instruments for the accomplishment of the work, this is also the case of the household. Some instruments (organon) are lifeless (apsycha), others living (empsycha); in the rudder, the pilot (kybernete) of a ship has a lifeless, in the look-out man, a living instrument (for in the arts the servant (hyperetes) is in the rank of instrument). Thus, too, the property of such an instrument is for maintaining life as well as the totality of them; a slave (doulos) is a living possession, and like an instrument which takes the place of all other instruments. For if every instrument could accomplish its own work, following a command or anticipating it, like the statues of Daedalus, or the tripods of Hephaestus, which, says the poet, of their own accord (automatous) entered the assembly of the Gods; if, in like manner, the shuttle would weave and the plectrum touch the lyre without a hand to guide them, the master builder (architektosin) would not need servants (hypereton), nor masters (despotais) slaves (doulon)." (31, 1253 b 25-39, revised English translation, RC)

What is particularly remarkable in this seminal text from Aristotle's Politics is the last sentence in which he considers the possibility of a non-slavery based society in case "lifeless" instruments could accomplish the same kind of work done either by a servant or by slave. After thorough argumentation Aristotle comes to the conclusion that because there is some evidence for the factual distinction between free men and slaves, this distinction is not always based on nature (31, 1255 b 5). The idea of a society free of masters and (born or factual) slaves is, indeed, Utopian. Being rooted in a mythical context, it is somehow ironical too. Nevertheless it is a provocative idea. In the Rhetoric Aristotle quotes the Sophist Alcidamas: “God has left all men free; Nature has made none a slave.” (32, 1383 b 17-18). What, indeed, makes a basic difference with regard to the production and use of technical objects in Antiquity is the meaning of labour and commercial competitive behaviour since Modernity as what binds society (33, p. 319; 34). But a society based only on market economy is to the detriment of friendship and use-value as conceived by Aristotle.

The impact of Aristotle's thinking as well as of other great thinkers in Antiquity and Modernity on the non-natural status of slavery was weak if we consider that, for thousands of years, human societies have continued to be slavery-based (35). For the purpose of this analysis, we keep in mind that servants and slaves ― Aristotle points with some examples to the difference between both ― are understood as instruments for special tasks. There are different kinds of relationships between masters and servants/slaves, humans and animals, and humans and lifeless artefacts according to different contexts. According to Aristotle, the best one is when the work is done by humans because masters and servants/slaves share "psychic" qualities lacking in non-human instruments. What is proper for a slave, as distinct from a servant, is to be the property of a master, doing with the body the things the master foresees with the mind. Ownership means, in the case of slaves, that no part remains outside the master's control (31, 1254 a 10-13). The ownership of productive "lifeless" or "living" instruments allows the master to use what the instruments produce, as well as to purchase and exchange the products, using money for the sake of profit. To strive towards an unlimited accumulation of money and things owned through money (techne chremastike) means that a human agent is committed to life but not to good life (31, 1257 b 34-42). A manager can delegate these tasks in order to dedicate himself to politics and philosophy, i.e. to another form of intellectual practice (31, 1255 b 35-38). A human being (anthropos) is by nature a "political animal" (politikon zoon) and the only living being endowed with speech (logos), allowing him ― Ancient Greece was a male- (aner) dominated society ― to reason politically about good and evil, the just and the unjust (31, 1253 a 2-18). Both "essential performances" of human beings were excluded in the social and economic context of Ancient Greece from the being of servants and slaves (36).

Slavery was de iure abolished in the 19th century, but different forms of labour appeared in industrial society with the use of machines based on steam-power and electricity that led to mass production and new forms of the division of labour under slave-like conditions. This was criticized most prominently by Karl Marx (1818-1883). In his opus magnum Das Kapital he quotes the seminal text by Aristotle and calls him "the greatest thinker of antiquity" (37, p. 278). Marx argues that neither Aristotle nor other thinkers could comprehend that machinery produces "the economic paradox, that the most powerful instrument for shortening labour-time, becomes the most unfailing means for placing every moment of the labourer’s time and that of his family, at the disposal of the capitalist for the purpose of expanding the value of his capital." (37, p. 278). But as Hannah Arendt remarks: "A hundred years after Marx we know the fallacy of this reasoning; the spare time of the animal laborans is never spent in anything but consumption, and the more time left to him, the greedier and more craving his appetites" (38, p. 133). Norbert Wiener echoes both concerns when he writes:

"Let us remember that the automatic machine, whatever we think of any feelings it may have or may not have, is the precise economic equivalent of slave labor. Any labor which competes with slave labor must accept the economic conditions of slave labor. It is completely clear that this will produce an unemployment situation, in comparison with which the present recession and even the depression of the thirties will seem a pleasant joke." (39, p. 162)

A reflection on the impact of digital technology on human labour from the perspective of Political Economy serves as a corrective to a one-sided market-oriented capitalistic view of technology that fails to address issues of justice as fairness faced by societies in the digital age (40, 41).

2. Ethical Issues of Humanoid-Human Interaction

What is interaction? According to Hannah Arendt, we are a working animal (animal laborans), a producer of things (homo faber), and a being capable of action and speech. She writes:

“Human plurality, the basic condition of both action and speech, has the twofold character of equality and distinction. If men were not equal, they could neither understand each other and those who came before them nor plan for the future and foresee the needs of those who will come after them. If men were not distinct, each human being distinguished from any other who is, was, or will ever be, they would need neither speech nor action to make themselves understood” (38, pp. 175-176).

We disclose ourselves through action and speech. "This disclosure of 'who' in contradistinction to 'what' somebody is ― his qualities, gifts, talents, and shortcomings, which he may display or hide ― is implicit in everything somebody says and does" (38, p. 179). The difference between who we are, our 'whoness,' and what we are, is the ethical difference (42). The Australian philosopher, Michael Eldred, coined concepts of interaction, interplay and whoness. He writes with regard to Arendt:

“The realm or dimension she is addressing, of "people... acting and speaking together" (38, p.198) through which they show to each other who they are and perhaps come to "full appearance [in] the shining brightness we once called glory" (38, p. 180), is not that of action and reaction, no matter (to employ Arendt's own words) how surprising, unexpected, unpredictable, boundless social interaction may be, but of interplay. it is the play that has to be understood, not the action, and it is no accident that play is also that which takes place on a stage, for she understands the dimension of "acting and speaking" (38, p. 199), revealing and disclosing their selves as who they are. On the other hand, interplay takes place also in private: in the interplay of love as a groundlessly grounding way to be who with another, where speaking easily becomes hollow” (43, p. 83)

We build our character (ethos) through social life. The difference between making things (poiesis) and building what Arendt calls "the "web" of human relationships" (38, p. 183) through human action (praxis) goes back to Aristotle. Human beings conceal and reveal who they are in a shared world. The task of an ethics of the humanoid-human interaction is, firstly, learning to see the difference between the human-human interplay and humanoid-human interaction. Human-human interplay is about mutual acknowledging or disacknowledging who we are individually and socially on the basis of shared customs, rules, values and practices. Human interplay is risky because human agents face the contingencies and openness of their past, present and future actions and interpretations and above all the risks of ongoing power play with others. A selection of human performances and moral rules can be codified in algorithms building the basis for the humanoid-human interaction. An ethics of humanoid-human interaction is, secondly, learning to see how far and in which contexts such an algorithmic selection of human interplay can make sense for the purpose at issue. This second task can be dealt with using Helen Nissenbaum's concept of privacy as contextual integrity. She writes:

“Contexts are structured social settings characterized by canonical activities, roles, relationships, power structures, norms (or rules), and internal values (goals, ends, purposes). Contexts are "essentially rooted in specific times and places" that reflect the norms and values of a given society” (44, pp. 132-133)

According to Nissenbaum, privacy is not a property of a particular kind of information, but a second-order attribute ascribed to information in a specific context. Furthermore, human interplay is constituted by situations that constantly evolve and dissolve based on the free power plays of the agents. It is based on trust, not on subordination. Trust relies on implicit or explicit norms that can be selected for the humanoid-human interaction, taking into account not only the evolution of such norms and customs but also the difference between implicit and explicit knowledge, the original belonging-together of feelings and understanding, and how bodily knowledge guides our behaviour in familiar environments (see also 45 for a review of the importance of trust from a safety point of view). According to Michael Nagenborg, there is a difference between a program and an agent. He writes:

“One major difference between a „program“ and an „agent“ is, that programs are designed as tools to be used by human beings, while „agents“ are designed to interact as partners with human beings. […] An AMA [artificial moral agent, RC] is an AA [artificial agent, RC] guided by norms which we as human beings consider to have a moral content. […] Agents may be guided by a set of moral norms, which the agent itself may not change, or they are capable of creating and modifying rules by themselves. […] Thus, there must be questioning about what kind of „morality“ will be fostered by AMAs, especially since now norms and values are to be embedded consciously into the „ethical subroutines“. Will they be guided by „universal values“, or will they be guided by specific Western or African concepts?” (46, pp. 2-3)

But, in fact, artificial agents need a battery and an algorithm in order to move by themselves as agents. Automata as defined by Aristotle are no less tools (organa) than assistants and slaves that they could replace.

An artificial moral agent, a 'what-agent,' is, in turn, not the same as a 'who-agent' capable of criticizing her own rules of behaviour as well as codes of morality guiding humanoid actions. There is a difference between morality as a set of customs and norms underlying implicitly or explicitly human interplay and their critical ethical reflection that allows human agents to question moral and legal rules. Consequently, the notion of autonomy with regard to humanoid and human agents must be differentiated;. Autonomy with regard to humanoids refers to their capacity to do what their authors/originators wants them to do in a particular context and in view of a given set of standard situations. Human autonomy means the capacity to look beyond such contexts and to be open to unexpected situations giving rise to a process of deliberating for self-care that means always caring for others. What moves humans is the awareness of the groundless interplay of their lives and the care for a good life. The human self is not an isolated and worldless subject as imagined in Modernity but a plurality sharing a common world. The same can be said with regard to other notions ascribed to humanoids such as 'learning' or 'making a decision' that the author/originator of the humanoid selects from the human "exemplar" deprived of the historical, political, societal, bodily and existential dimensions (47, see also 48). In short, humanoid-human interaction is asymmetrical in view of human-human interplay. The ethical task of designers and users of humanoids is to make this asymmetry as transparent as possible for the human agents so that they can be aware of its risks and opportunities. In the field of healthcare robots, Aimee van Wynsberghe puts this asymmetry thus:

“[...] the latest developments in healthcare robotics are those intended to assist the nurse in his/her daily tasks. These robots, now referred to as care robots, may be used by the care-giver and/or the care-receiver for a variety of tasks from bathing, lifting, feeding, to playing games. They may have any range of robot capabilities and features and may be used in a variety of contexts, i.e. nursing home, hospital or home setting. They are integrated into the therapeutic relationship between care-giver and care-receiver with the aim to meet a range of care needs of users.” (49, p. 2, my emphasis)

A key ethical issue of humanoid-human interaction is the contextual delimitation of the purpose and the field of action within which a humanoid or several of them are supposed to be assistants in interaction with humans, in accordance with some human "essential performances" selected for a specific goal. The more explicit such contextual delimitation is, the more reliance can be built up concerning the contextual integrity of what can and should be done and what not. Within such contexts, not only breakdowns in physical performance must be taken into account, but also misunderstandings and frustration in humanoid-human communication. Noel Sharkey writes:

“What would happen if a parent were to leave a child in the safe hands of a future robot caregiver almost exclusively? The truth is that we do not know what the effects of the long-term exposure of infants would be. We cannot conduct controlled experiments on children to find out the consequences of long-term bonding with a robot, but we can get some indication from early psychological work on maternal deprivation and attachment. Studies of early development in monkeys have shown that severe social dysfunction occurs in infant animals allowed to develop attachments only to inanimate surrogates.” (-50, para. 6)

These issues were analyzed in the mid-1980s by Terry Winograd and Fernando Flores with regard to software systems embedded in human conversations. "In designing tools," they wrote, "we are designing ways of being" (20, p. xi). This is particularly true when it comes to the design of humanoid assistants for various tasks embedded in different cultures. There is a variety of ethical and legal values and principles that need interpretation in view of situations in which they are supposed to govern and/or assist interaction with humans, with a lot of possibilities between governing and assisting. The selection of human actions may either conform with or contradict customs and legal rights of the users in a particular context and cultural tradition (see also e.g. 51, 52).

Care-giver humanoid robots cannot only mimic human intelligence and human communication but also human emotions (see also 53), an issue that has been object of research with regard to robots and avatars in several research projects for fifteen years. Barbara Becker (1955-2009) wrote ten years ago:

“The personalization of robots and avatars, which is in common particularly observed among children and is obviously reinforced by even primitive expressions of feelings (see Cassell 2000, Breazeal 2002, and also Dautenhahn 2004), is highly ambivalent. On the one hand the personalization of avatars can frequently lead to more rapid initiation of communication and reinforce the feeling of social involvement (Schröder et al. 2002). On the other hand, however, "emotionalized" robots or avatars suggest the existence of an emotional context on the part of the virtual agent or robot, which is a pure fiction. This brings us back to the issue of social practice (Gamm 2005): How do people, e.g. children, deal with the artificial agents that give them the impression of having feelings similar to those they have themselves? The personalization frequently observed when dealing with agents is unproblematic as long as the users can maintain a reflective distance from such attributions. Nor does this present a serious conflict for children, as long as the artificial agents have the status of cuddly toys or dolls, to which a form of personality has always been ascribed. Nevertheless, children usually had a clear perception of the artificiality of these toys. The future will show whether this changes when the artefacts show expressive reactions or can move.” (54, p. 42 emphasis added)

This issue today becomes particularly problematic, on the one hand, in case of children interacting with online toys, exposed to a deregulated digital environment if they disclose who they are to third parties. Online toys become invasive tools of children's privacy. On the other hand, humanoids can be useful as teachers' assistants or for autism treatment.

Conclusion & Future Directions

The ethical compass when dealing with humanoid-human interaction is pronounced, to raise clear awareness of the difference between humanoid-human interaction and human-human interplay. Ethical issues arise from the intersection of the natural with the artificial. What kind of decisions should a robot be allowed to make? From a philosophical point of view, it should be allowed to make decisions only within a given context, based on behavioural rules with the prior informed consent of the human agent. Humanoids' algorithmic intelligence becoming the master of the interaction should be avoided. Instead, humans should always have their own freedom in choosing options that might meet expectations and desires in facing situations not foreseen by the algorithm or in dealing with customs and norms other than the ones programmed into the algorithm. A humanoid should not be allowed to act autonomously beyond some standard situations where the risks for the human agent can be minimized. The dependency of the human 'who' on a humanoid 'what' is only ethically acceptable, if there is a who behind the what taking responsibility for the consequences of the humanoid's action on the human agent.

The task of the ethics of humanoid-human interaction is to reflect on the possibilities in-between these two poles in order to give human agents their freedom back as far as possible. Humanoid-human interaction selects some essential actions of the human-human care, switching from the pole in which such care tends to take the place of the other, to the pole that opens up a path for the other to take care of him- or herself. This raises the question as to how flexible programming algorithms are allowed to be. This question can only be answered with regard to specific contexts and foreseeable standard situations of danger for human agents. Emerging behaviours from adaptable algorithms should be monitored by the persons responsible for the contextual integrity the humanoids are supposed to be subjected to, setting a limit of their autonomy. Such monitoring of deteriorating or ameliorating possibilities and realities of humanoid-human interaction should take place regularly, particularly when human agents are physically, intellectually and/or volitionally weak. In these cases, humanoid-human interaction should be complemented or even supplemented in various degrees by human-human interplay. Finding the right balance of this intersection needs patient attention to and evaluation of changing human needs. In special cases, such as in health care, a technical and ethical monitoring could be even legally mandatory.

Should a robot be invariable or programmed to follow the cultural and ethical expectations of a specific society? In 2006, the International Review of Information Ethics dedicated a special issue to Ethics in Robotics. The editors wrote:

“Our main values are embedded into all our technological devices. Therefore, the question is: which values are we trying to realize through them? Artificial creatures are a mirror of shared cultural values. Humans redefine themselves in comparison with robots. This redefinition of living organisms in technological terms has far-reaching implications. Long-term reflections need to be developed and plausible scenarios need to be anticipated." (55, p. 1).

There is plenty of theoretical and empirical research to be done on Intercultural Roboethics. Cultures are not closed entities, but complex and dynamic systems. Ethics likewise needs an intercultural perspective. Naho Kitano writes:

“'Rinri', the Japanese Ethics. When discussing the ethics of using a robot, I have been using the term ―"Roboethic" generally in my research, but it is used in very particular ways especially at international conferences. The word for ―"Ethics" in Japanese is Rinri. However, the Japanese concept of ethics differs from the Western concept of ethics, and this can lead to misunderstandings. In Japan, ethics is the study of the community, or of the way of achieving harmony in human relationships, while in the West, ethics has a more subjective and individualistic basis. The contrast can be observed, for example, in the concept of social responsibility. In Japan, responsibility in the sense of moral accountability for one‘s action already existed in the classical period, but the individual was inseparable from status (or social role) in the community. Each individual had a responsibility toward the universe and the community. Thus in Japan, social virtue lay in carrying out this responsibility” (56, p. 80)

In her book Robotica Nipponica Cosima Wagner writes:

“"social" robots illustrate the "negotiation character of the creation and use of technological artefacts" (Hörning), which, for example includes the rejection of military applications of robot technology in Japan. On the other hand, as a cultural topos, they mirror dreams, desires and needs of human beings at a certain time and therefore have to be interpreted as political objects as well. As a source for a Japanese history of objects, "social" robots exemplify the cultural meaning of robots, the expectations of the Japanese state and economy, the mentality of Japanese engineers and scientists and, last but not least, the socio-cultural change which the ageing Japanese society is about to face.” (57, p. 2)

These examples show the complexity and relevance of research into ethical issues regarding humanoid-human interaction. This research should not deal only descriptively with differences between humanoid-human interaction and human-human interplay in different cultures, but also normatively with issues implicit or explicit in both cases, including their intersection. Ethics of humanoid-human interaction is an exciting field of research for which this contribution aims at rising awareness for taking more steps based on an ongoing intercultural robo-ethics dialogue between 'East' and 'West' (58).

Acknowledgements

I thank Michael Eldred (Cologne) for polishing my English.

References

1. Dictionary.com Unabridged, oid. (n.d.)
http://www.dictionary.com/browse/oid. Accessed 28 Feb 2017.

2. M. Rader, (ed.), A. Antener, R. Capurro, M. Nagenborg, L. Stengel, W. Oleksy, E. Just, Edyta; K. Zapędowska, I. Székely, M. Szabó, B. Vissy, Beatrix, ETICA Project Deliverable D. 3.2 Evaluation Report 2012.
http://www.etica-project.eu/deliverable-files. Accessed 28 Feb 2017.

3. M. Negrotti, The Theory of the Artificial (Cromwell Press, Wiltshire, 1999).

4. M. Negrotti, Naturoids. On the Nature of the Artificial (World Scientific Publishing, Singapore 2002).

5. M. Negrotti, The Reality of the Artificial. Nature, Technology and Naturoids (Springer, Berlin, 2012).

6. R. Capurro, Past, present and future of the information concept. tripleC, 7/2 (2009), 125-141,
http://www.triple-c.at/index.php/tripleC/article/view/113/116. Accessed 28 Feb 2017.

7. K. H. Hayles, How We Became Posthuman. Virtual Bodies in Cybernetics, Literature and Informatics (The University of Chicago Press, Chicago, 1999).

8. E. Husserl, Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie (Meiner, Hamburg, 1992) (Engl. Ideas: General Introduction to Pure Phenomenology, trans. W. B. Gibson, 1962).

9. A. D. Spear, Edmund Husserl: Intentionality and Intentional Content. The Internet Encyclopedia of Philosophy, ISSN 2161-0002, 2016.
http://www.iep.utm.edu/. Accessed Feb 28, 2017.

10. Aristotle, Ethica Nichomachea (Oxford University Press, London, 1962).

11. R. Capurro, Skeptical Knowledge Management, in Knowledge Management. Libraries and Librarians Taking Up the Challenge ed. by H.-C. Hobohm (Saur, Munich, 2004), 47-57.
http://www.capurro.de/skepsis.html

12. K.R. Popper, Conjectures and Refutations. The Growth of Scientific Knowledge (Routledlge & Kegan Paul, London, 1965).

13. J.P. Vernant, P. Vidal-Naquet, Mythe et tragédie en Grèce ancienne (La Découverte, Paris, 2001).

14. R. Descartes, Discours de la méthode, in Oeuvres et lettres. ed. by A. Bridoux. (Gallimard, Paris, 1952).

15. A. M. Turing, Computing Machinery and Intelligence, Mind 59, 433-460 (1950).

16. M. Polanyi, The Tacit Dimension (Doubleday, New York, 1966).

17. F. Adloff, K. Gerund, D. Kaldewey (eds.), Revealing Tacit Knowledge: Embodiment and Explication (transcript, Bielefeld, 2015).

18. H.L. Dreyfus, Hubert L. (1992). What Computers Still Can't Do: A Critique of Artificial Reason. (The MIT Press, Cambridge, 1992).

19. H.L. Dreyfus, What Computers Can’t Do: A Critique of Artificial Reason. (Harper & Row, New York, 1972).

20. T. Winograd, F. Flores, Understanding Computers and Cognition: A New Foundation for Design. (Ablex, Norwood, NJ, 1986).

21. D. Ihde, Bodies in Technology (The University of Minnesota Press, Minnesota, 2002).

22. D. Ihde, Expanding Hermeneutics: Visualism in Science. (Northwestern University Press, Evanston, Ill., 1998).

23. P.P. Verbeek, Moralizing Technology: Understanding and Designing the Morality of Things. (The University of Chicago Press, Chicago, 2011).

24. S. Settis, Supremely Original. Classical art as serial, iterative, portable, in Serial / Portable Classic ed. by S. Settis, A. Anguissola, D. Gasparotto, 51-72. (Fondazione Prada, Milan, 2015).

25. R. Capurro, Living with Online Robots. (2015) www.capurro.de/onlinerobots.html. Accessed 26 Feb 2017

26. A. Ayteş, Divine Water Clock: Reading al-Jazarī in the Licht of al-Ghazālī's Mechanistic Universe Argument, in Allah's Automata. Artifacts of the Arabic-Islamic Renaissance (800-1200) ed. by S. Zielinski, P. Weibel (Center for Art and Media, Karlsruhe, 2016), 100-115.

27. W. Benjamin, Das Kunstwerk im Zeitalter seiner technischen Reproduzierbarkeit. (Suhrkamp, Frankfurt a.M., 1977) (Engl. The Work of Art in the Age of Mechanical Reproduction)
http://web.mit.edu/allanmc/www/benjamin.pdf. Accessed 26 Feb 2017

28. R. Capurro, On Artificiality, ed. by Istituto Metodologico Economico Statistico (IMES) Laboratory for the Culture of the Artificial, Università di Urbino, IMES-LCA WP-15 (1995)
http://www.capurro.de/artif.htm. Accessed 26 Feb 2017

29. R. Capurro, Leben im Informationszeitalter (Akademie Verlag, Berlin, 1995).

30. N. Wiener, Cybernetics: or Control and Communication in the Animal and the Machine (The MIT Press, Cambridge, Mass., 1965).

31. Aristotle, The Politics of Aristotle (Oxford University Press, London, 1950).
http://classics.mit.edu/Aristotle/politics.1.one.html. Accessed Feb 27, 2017.

32. Aristotle, Ars Rhetorica (Oxford University Press, London, 1959).

33. J.P. Vernant, Mythe et pensée chez les Grecs. Études de psychologie historique (La Découverte, Paris, 1996)

34. J.P. Vernant, Travail et esclavage en Grèce ancienne. (Complexe, Brussells, 1994).

35. H. Joas, Sind die Menschenrechte westlich? (Kösel, Munich, 2015).

36. W. Kullmann, Aristoteles und die moderne Wissenschaft (Franz Steiner, Stuttgart, 1998).

37. K. Marx, Das Kapital. Kritik der politischen Ökonomie (Anaconda, Cologne, 2009) (Engl. Capital. A Critique of Political Economy, Transl. S. Moore & E. Aveling).
http://www.marxists.org/archive/marx/works/download/pdf/Capital-Volume-I.pdf. Accessed 26 Feb 2017.

38. H. Arendt, The Human Condition (The University of Chicago Press, Chicago, 1998).

39. N. Wiener, The Human Use of Human Beings. Cybernetics and Society. (Free Association Books, London, 1989).

40. M. Schneider, A Dialética do Gosto: Informação, música e política. (Faperj/Circuito, Rio de Janeiro, 2015).

41. C. Fuchs, Digital labour and Karl Marx. (Routledge, New York, 2014).

42. R. Capurro, M. Eldred, D. Nagel, Digital Whoness: Identity, Privacy and Freedom in the Cyberworld. (de Gruyter, Berlin, 2013).

43. M. Eldred, Phenomenology of whoness: identity, privacy, trust and freedom, in Digital Whoness: Identity, Privacy and Freedom in the Cyberworld ed. R. Capurro, M. Eldred, D. Nagel, Daniel (de Gruyter, Berlin, 2013).

44. H. Nissenbaum, Privacy in Context: Technology, Policy, and the Integrity of Social Life. (Stanford University Press, Stanford, 2010).

45. D. Araiza-Illan, K. Eder, Safe and trustworthy human-robot interaction, in Section: Human-Humanoid Interaction, Humanoid Robotics: a Referernce (Springer, London).

46. M. Nagenborg, Artificial moral agents: an intercultural perspective. International Review of Information Ethics, 7, (2007).
http://www.i-r-i-e.net/inhalt/007/13-nagenborg.pdf Accessed Feb 27, 2017.

47. R. Capurro, Toward a Comparative Theory of Agents. AI & Society 27, 4 (2012), 479-488.
http://www.capurro.de/agents.html. Accessed 27 Feb 2017

48. K.S. Lohan, H. Lehmann, C. Dondrup, F. Broz, H. Kose, Enriching the human-robot interaction loop with natural, semantic and symbolic gestures, in Section: Human-Humanoid Interaction, Humanoid Robotics: a Referernce (Springer, London).

49. A.v. Wynsberghe, Healthcare Robots. Ethics, Design and Implementation. (Routledge, New York, 2016).

50. M. Sharkey, The Ethical Frontiers of Robotics, Science, Dec 19 (2008) .
http://www.science.sciencemag.org/content/322/5909/1800.full

51. B. Miller, D. Feil-Seifer, Embodiment, Situatedness and Morphology for Humanoid Interaction in Section: Human-Humanoid Interaction, Humanoid Robotics: a Referernce (Springer, London).

52. F. Ferland, R. Agrigoroaie, A. Tapus, Assistive humanoid robots for the elderly with Mild Cognitive Impairment, in Section: Human-Humanoid Interaction, Humanoid Robotics: a Referernce (Springer, London).

53. T. Nomura, Empathy as signaling feedback between humanoid robots and humans,in Section: Human-Humanoid Interaction, Humanoid Robotics: a Referernce (Springer, London).

54. B. Becker, Social robots – emotional agents: Some remarks on naturalizing man-machine interaction. International Review of Information Ethics, 6 (2006), 37-45.
http://www.i-r-i-e.net/inhalt/006/006_Becker.pdf. Accessed 27 Feb 2017.

55. R. Capuro, Th. Hausmanninger, K. Weber, F. Weil, Ethics in Robotics. International Review of Information Ethics, 6 (2006) 1.
http://www.i-r-i-e.net/inhalt/006/Editorial.pdf. Accessed 27 Feb 2017.

56. N. Kitano, 'Rinri': An Incitement towards the Existence of Robots in Japanese Society. International Review of Information Ethics, 6 (2006), 79-82.
http://www.i-r-i-e.net/inhalt/006/006_Kitano.pdf

57. C. Wagner, Robotica Nipponica - Recherchen zur Akzeptanz von Robotern in Japan (Tectum Verlag, Marburg, 2013) Abstract.
http://www.japanologie.uni-frankfurt.de/58656291/English_abstract_Robotopia_Nipponica_CW.pdf Accessed 27 Feb 2017-

58. M. Nakada, R. Capurro, An Intercultural Dialogue on Roboethics, in The Quest for Information Ethics and Roboethics in East and West. Research Report on trends in information ethics and roboethics in

Japan

and the West. ed. by M. Nakada, R. Capurro, Research Group on the Information Society and International Center for Information Ethics,Tsukuba, Japan (2013), 13-22. ISSN 2187-6061.
http://www.capurro.de/intercultural_roboethics.html Accessed 27 Feb 2017.

Homepage	Research	Activities
Publications	Teaching	Interviews

ETHICAL ISSUES OF HUMANOID-HUMAN INTERACTION

Rafael Capurro