How is Technical Creativity Possible?

Yastreb Natalya Andreevna Doctor of Philosophy Associate Professor, Director of the Institute of Social and Human Sciences, Vologda State University 160035, Russia, Vologda region, Vologda, Lenin str., 15, of. Department of Philosophy nayastreb@mail.ru

IntroductionThe problem of technical creativity for the philosophy of technology is the cornerstone.

Ontological, epistemological, value and many other contexts converge in it, in connection with which the understanding of the essence and mechanism of technical creativity in a particular philosophical concept can be considered as some kind of test for the efficiency of the theory. The actualization of this problem is also largely connected with the radical expansion of the technical sphere. First, computer programs and virtual objects were added to machines and mechanisms, and now augmented reality systems and man-made virtual worlds forming a "metaverse" also belong to the technical sphere. The question of how something new appears out of nothingness, what are the sources, boundaries and limits of technical creativity becomes extremely important, which determined the problems of this study. Within the framework of this work, we will use the terms "creativity" and "creativity" as synonyms, and under the word "machine" we will understand all technical systems, including computer programs.

Creativity as an impossible activityThe basis of any definition of creativity is that it consists in creating a new one.

The philosophical problematization of creativity is based on the paradox according to which "the new does not arise from the old, since it differs significantly from it, but it cannot arise from nothing, since then it remains incomprehensible" ^{[1, p. 19]}. In this regard, creativity is often described using various variants of the terms unthinkability and impossibility. Thus, A.Y. Nesterov and A.I. Demina draw attention to the fact that creativity, as a generation of the new, "as a rule, is associated with the disidentification of the unthinkable and the impossible, with negation procedures as ways of accounting for reality by human consciousness" ^{[2, p. 84]}. At the level of thinking, creativity presupposes the appearance of something that was not thought of before, that is, literally unthinkable, and at the level of embodiment – the appearance of something new, including what was not possible before, that is, impossible.

A similar approach is proposed by M. Boden, who distinguishes two types of creativity, namely, the incredible and the impossible ^[3]. Incredible creativity is an activity that gives something new to the individual. She also calls this type psychological, or P-creativity. Psychological (P) creativity does not bring anything fundamentally new, but, nevertheless, it has value, since it contributes to the self-realization of a person, obtaining skills of activity, solving tasks facing a person. If there is an invention of ideas that have not been realized before, through "mapping, research and transformation of conceptual spaces to create new ideas," we are talking about "impossible" creativity. Boden calls such creativity historical, or H-creativity. Historical (H) creativity leads to the appearance of something that did not exist before, it is an invention in the literal sense of the word. It is not limited to methods, techniques and algorithms, and we do not have a "systematic explanation of H-creativity, there is no theory that explains everything and only H-creative ideas" ^{[3, p. 77]}. Impossible creativity is essentially a search, research and transformational activity.

It seems that the paradox of the impossibility of the emergence of a new one can be resolved at least in relation to technical creativity. Without pretending to explain the essence of creativity as a whole, within the framework of this work, an attempt will be made to explain the possibility of technical creativity as the invention of subjectively new (P-creativity) and as the emergence of objectively new (H-creativity). Sources of technical creativity

The production of technical objects can be aimed at reproducing already invented and implemented objects, or it can involve the development of new models, parts, devices and systems.

If creativity is understood in a broad sense as the creation of something new, then technical creativity can be defined as the creation of objects that have qualitative differences from all previously created artifacts. In this paper, the result of technical creativity is understood as the creation of a technical object. It is important to understand that a technical object is a system that allows you to implement some functions necessary for the subject, based on natural cause-and-effect relationships.

The creation of any technical object is a practice in which the actions of the subject, such as cognition, creativity and construction, and the actions of the environment represented by natural and artificial objects, including previously created technology, are combined. Within the framework of the subject-centered approach, the source of creativity is a person in whose imagination a new thing appears ^[4]. With such an anthropocentric approach, an active role in the act of creation is assigned only to a person, and the source of the new is called imagination, thinking, insight and other cognitive functions of the subject. This approach is opposed by the deterministic concept of technology, which gives priority to the internal logic of technological evolution in the emergence of new devices ^[5]. Thus, M. McLuhan considers technology as "an active logos that transforms both people and their world" ^{[6, p. 98]}. At the same time, a person is considered as a participant in creation, and not a source of creativity.

Analyzing the history of technology, you can see that there are confirmations of both approaches in it. Looking at the machines created by Leonardo da Vinci, even after several centuries, we understand that it is unlikely that many of them would have appeared if not for the genius of their author. At the same time, complex modern machines, from cars to supercomputers, cannot be invented in their pure form, their design is based on a huge series of artificial systems that make their existence possible. To explain technical creativity, it is necessary to take into account both factors, that is, conscious human activity and the influence of the environment, including previously created technical devices. These devices can act in two important qualities, as elements for creating more complex systems, and as tools for the production of other objects. Instrumental nature of technical creativity

The foundations for understanding the instrumental nature of technical creativity were laid by Aristotle, who interpreted craft as giving matter a form that existed before in human consciousness. To do this, the master uses both his body and various tools. However, with this approach, creative activity takes place in consciousness, and tools play a passive role, only helping the new to materialize ^{[7, 8]}.

Another approach is suggested by P.K. Engelmeyer, who urges not to limit the work of a creative person to the sphere of imagination. He writes that "three things are needed for an invention: 1) guess, 2) knowledge and 3) skill" ^{[9, p. 9]}. The creative act in his soul acts as the first act of creating a new technical object, this is "one third of the invention". The second act of invention, the act of knowledge, "receives the future invention in the form of free-standing parts." At this stage, the idea is transformed into a diagram, concept, drawing, model or other form, thereby filling in the gaps in the prototype of the new object. It is no longer intuition that comes to the fore, but thinking. The neoplasm here is the development of an invention plan. The final, third, act consists in the implementation of the plan of the invention, and its result is the creation of a design. This stage, which requires skill from the creator first of all, with all its complexity in terms of creativity, is the easiest, since the inventor can use many existing machines to implement his idea.

In the scheme proposed by Engelmeyer, we see the stages of the emergence of a new one, in which creative activity is maximal at the beginning, at the guessing stage, and then its contribution to the result decreases. The role of reason and skill, on the contrary, at the beginning of the process gives way to the primacy of creativity, then increases and becomes the leading one. Also, although Engelmeyer considers technical creativity as a collective action in general, the role of a particular individual is of the greatest importance in the first act. In the latter, on the contrary, the result usually depends on the participation of many subjects.

It is important to understand that this refers specifically to the three-act structure of creativity, which can be implemented at any stage of the creation of the invention. Both during the development of the scheme and during its implementation, new solutions may be needed, each of which will appear as a result of the same three acts of invention. Engelmeyer's process of technical creativity, in modern terms, is fractal in nature, and he himself describes it in the form of a whole bunch of grapes, where three acts are piled on top of each other ^[9].

Where do new creative tasks come from at the stage of implementing an invention? An important role in this process is played by the tools and materials with which the creator works: "The work itself tells the worker the most suitable form for the tool ..., both in the worker and in the very last savage, some creative act is performed at the moment when he asks himself: is it not better to bend the handle of his braids?" ^[9]. To answer the question of whether tools can play an active role in creativity, let us turn to the concept of D. Gibson, who introduced the concept of affordance (opportunity) to describe the interaction of the subject with the environment ^[10]. The environment, Gibson writes, offers the subject the possibility of action. Affordance is considered by him as a property possessed by an object, and which informs the user about possible actions, while it is turned "in both directions – both to the outside world and to the observer" ^{[11, p. 209]}.

The tool as an instrument of activity offers a person certain ways of action that can both limit the author's capabilities and contribute to the emergence of certain solutions ^[12]. Gibson's affordances exist in the environment, including in the absence of the subject. The activity of the subject contributes to the realization of certain opportunities that initially exist in the world. Creativity here can be considered as a response and behavior that occurs as a reaction to the affordances in the environment. Affordances cause deviations in the usual forms of interaction between a person and material objects, which may be the reason for creativity ^[13]. The world surrounding the inventor, including tools and tools, thus have an active influence on the creative process, defining and limiting the actions available to the author.

Thus, the environment provides the subject with many possibilities of action, many of which cannot be implemented without the participation of the subject. At the same time, technical creativity acts as an act of realizing those opportunities that exist in the environment, but were not actualized earlier in a natural way or by other subjects. At the same time, the creation of a new object leads to the emergence of new opportunities (affordances), which determines the internal logic of technical development. The active role of the subject is manifested in the allocation and implementation of not all opportunities, but only a number of them in accordance with the goals and needs of the subject. With this approach, there is no ontological difference between P-creativity and H-creativity for the subject, since in the structure of his interaction with the environment, both options will consist in the realization of previously unrealized possibilities. The difference between them will be subjective, since P-creativity creates something that is new for one subject, and H-creativity – for the whole of humanity. Limitations of technical creativity

The realization of the possibilities of creating new technical objects, both tangible and intangible, occurs under a number of restrictions. At the level of ideal objects, it seems that such limitations are conceivability and logical consistency. The requirement of logical consistency in the technical sphere has its own specifics, which A.N. Pavlenko defines as a condition of consistency "in the being of the possible" ^{[14, p. 139]}, as opposed to consistency in the being of the conceivable, for example in mathematics. The technique is mathematically determined, therefore, it must meet the requirements for mathematics itself. However, logical consistency does not mean technical feasibility at all. Therefore, in the process of engineering design and construction, the researcher is no longer dealing with logical, but with physical prohibitions. So, for example, the theory of relativity asserts the impossibility of transmitting a signal at a speed exceeding the speed of light in a vacuum, and classical thermodynamics – the creation of a perpetual motion machine of the second kind. Therefore, the second condition is physical consistency. At the level of material objects, creativity can be understood as the realization of those possibilities that exist in nature, but do not manifest themselves without human participation for one reason or another. At the same time, the creative activity of the subject is an act of transformation of potential being into actual. Its non-subject boundaries are set, at least, by the laws of physics, which potentially allow the existence of a multitude of objects that do not arise in nature without the participation of the subject, but at the same time do not give the opportunity to create anything.

The question of physical prohibitions is extremely complex due to the fact that they are always imposed within the framework of an accepted theory and can be transformed in the process of scientific development. This moment determines the need to highlight the third condition — paradigmatic consistency associated with the concrete historical stage of the development of science and the prohibitions imposed by the current theory, within which technical knowledge is carried out, as well as historical and socio-cultural circumstances. As a result, technology is prohibited by "logic (together with the mathematics behind it) and history (a specific level of knowledge and its corresponding values in a specific era)" ^{[14, p. 147]}.

The condition of praxeological permissibility is due to the fact that, since technical knowledge is initially practically oriented, it matters not only its truth (or reliability), but also efficiency, the possibility of activity. Technical knowledge cannot be justified only by theoretical methods. Even if all the requirements of consistency are met, only the synthesis of knowledge in a real object can serve as a criterion for assessing its reliability, because "only practice constructs the world of real objects" ^{[15, p. 29]}. Machine creativity

The proposed approach to understanding technical creativity can be applied to the analysis of the creativity of artificial intelligence systems.

The question of whether machines can create something new, posed at the time by A. Lovelace, helped not only to understand the limits of artificial intelligence capabilities, but also to problematize the very concepts of creativity and novelty. In the history of artificial intelligence, one can find both apologists for creative artificial intelligence ^[16] and opponents ^[17].

Traditionally, the idea that machines are not capable of creativity is justified by reference to the algorithmic nature of the machine, suggesting that the computer only implements the methods of action laid down in its program, without inventing them. However, if we think within the framework of the approach proposed above to understanding creativity as the realization of the possibilities available in the "subject-environment" system, then we can see that the machine also realizes the possibilities that are available to it. Of course, people will have more opportunities (affordances) due to the presence of various means of interaction with the environment and a developed cognitive apparatus, however, it can be hypothesized that the more ways available to them for machines to act, the more likely they are to implement affordances that have not been used by people before. That is, from a formal point of view, we have no reason to impose restrictions on the machine's ability to create.

However, it is necessary to take into account the fact that a person's creative activity consists not just in getting new, but also in realizing goals and obtaining value ^[18]. In recognizing an activity as creativity, an important role is played by its assessment by the subject. The decision on what to consider creativity and what not to consider is made by the person and society on the basis of determining the value of the result. We can talk about the conventionality of this decision in the sense that, as a rule, we do not have objective strict criteria for distinguishing creativity and other practices of creating something. We must evaluate at the same time the novelty of the result, including in comparison with previously obtained objects, and the creative nature of the activity.

If we evaluate creativity by results, then, apparently, we will have to recognize the possibility of machine creativity. Generative algorithms create music, texts, images, and give answers to questions. The composition "in the style of Bach", written by the DeepBach neural network (https://www.youtube.com/watch?v=QiBM7-5hA6o&ab_channel=SonyCSL ), is a new piece of music, can perform aesthetic functions and have value for the subjects. The works created by artificial intelligence successfully pass the "musical Turing test", when judges, including experts, comparing two objects, try to determine which of them is created by a person and which is generated by an artificial intelligence system. Either we must recognize the composer's work, done in the manner already found, as non-creative, or accept the idea that machines are capable of creating not just new, but also works of art that are significant for man and society.

Another argument against machine creativity is related to the peculiarities of creative activity itself. However, the hidden nature of creativity, manifested in the inability of the subject to understand the entire chain of obtaining a result, in the spontaneity and unconsciousness of the process of the emergence of ideas by analogy, can also be attributed to the machine. Neural network models that are not programmed, but are trained by examples, also do not, as a rule, give the opportunity to trace the process of forming the result. Data about the task (input parameters) and the result (output parameters) are available to the subject, and the signal passes through the inner layers of the neural network "in the dark". This is especially true for models based on deep learning. Modern neural networks can "hide" the process of the appearance of a result not only from a person, but also from themselves, which is used for self-assessment of the level of creativity of their own results. In this regard, an example of adversarial generative neural networks is indicative, which have in their structure a module for generating new data and a module for evaluating their novelty. The task of the generator is to create such a data package that will be indistinguishable from those in the available neural network sample. The second module, the discriminator, must find the generated content in the data package. The interaction of these modules is based on the principle of competition. Both modules are trained at the same time, that is, they do not work according to a ready-made algorithm. Creative neural networks have started to be developed on the basis of generative neural networks (https://arxiv.org/abs/1706.07068 ), whose task is to create new artistic styles. The purpose of the generator in this case is to create works that the discriminator will consider art, but will not be able to attribute to any of the styles known to him. The presence of a built-in system for evaluating creativity, a kind of internal "reflexivity" of such networks allows them to reach a new level of creativity in comparison with previous content generation systems.

Machines, like people, realize the opportunities that the world around them and their own cognitive apparatus (algorithms and system architecture) provide them, and we have no strict arguments that would prohibit machines from P-creativity or even H-creativity. Structurally, their creativity is similar to human activity, although it differs in scale and quality. The main difference between their creativity and human creativity is the lack of goal-setting and work in a limited subject area. If experts are trying to solve the problem of machine creativity going beyond the limits of the subject area set by a person ^[19], then the question of goal setting remains unanswered. It is the problems of setting tasks and the meaning of activity that seem to be the demarcation line between human and machine creativity at the moment. This gives us hope that natural intelligence will still be in demand and loaded with tasks. Conclusion

Thus, in ontological terms, technical creativity can be considered as the realization of those possibilities that are conditioned by the peculiarities of the environment, in particular, physical laws, but due to various circumstances are not actualized without the participation of the subject.

At the same time, technical evolution should be considered in the unity of its own internal laws and human activity. Internal patterns are due to the fact that the emergence of new devices creates new opportunities (affordances) for constructive human activity. The role of the subject in this case is to actualize the potential, that is, the practical realization of those opportunities that appear in the "man-environment" system. Such activity is available not only to humans, but also to machines, the results of which can be new both for the systems themselves (P-creativity) and for humanity (H-creativity). In this sense, machines are becoming an effective tool for generating content, but existing artificial intelligence systems do not have the possibility of personal goal-setting. Since the choice of opportunities that will be actualized is greatly influenced by the goals, needs, values and attitudes of the subject, in order to answer the question of whether machines are capable of creativity, it is necessary to study more deeply the problem of setting tasks posed by P.K. Engelmeyer and to model the process of problematization of the subject area itself.