In this article I want to give a very short sketch of the philosophical project of this blog. Each of the following sections outlines one direction of research. I want to expand each of these sections into one or several separate blog articles and later – based on research of the relevant literature – into academic level philosophical articles. This sketch is not complete in any way. It is – a sketch, a collection of ideas to be expanded (and probably in some instances, corrected).
First a few words about what this blog is not about: what I am calling “creativistic philosophy” must not be confused with creationism. This blog is not about religion, mysticism, esoteric beliefs or anything in such a direction.
Instead, this project is based on hard science, but tries to outline a new paradigm for several disciplines. Basically, it is about a theory of creativity (based mainly on the work of the mathematician, computer scientist and philosopher Kurt Ammon). In one of my next articles, I am going to post a bibliography of relevant literature about this. While Ammon’s work, providing the core of the paradigm, has been developed to the level of academic science, my own work, as outlined in this blog, is in a stage of developing philosophy. However, I see a potential to develop these ideas to the level of academic science in several of the areas mentioned below.
2 The Theory of Creative Systems
2.1 Cognition as information processing
The starting point is a view of cognitive processes as information processing. We are receiving information through our senses, processing and transforming it, storing some of it in our memories and control our muscles. These processes might be viewed as processes of computation that can be modeled as execution of programs. However, the generally accepted model of computation is based on the concept of algorithms. These can be thought of as programs that can be completely described by a formal theory, i.e. a finite system of axioms and rules of inference. Common formalizations of the concept of algorithms include “Turing machines”, “Recursive functions”, the “Lambda Calculus” and some others.
2.2 Algorithms vs. creative programs
Kurt Ammon and I think that this model of computation is too restrictive. Ammon has presented a proof that programs are possible that can develop out of the scope of any given formal theory about them. Such programs are not algorithms. They develop and change over time, based on their inputs. There is no complete formal theory describing what such a program will do. Each theory that can be formed about it is incomplete. Ammon shows that such programs can calculate functions that are not Turing-computable, i.e. for which no algorithm exist. So the common assumption that every program is an algorithm is wrong. (see AN EFFECTIVE PROCEDURE FOR COMPUTING “UNCOMPUTABLE” FUNCTION). Algorithms appear as components of creative systems but such systems as a whole cannot be viewed as algorithms.
2.3 A definition of creativity – incompleteness
Ammon defines creativity as the ability to calculate non-Turing-computable functions. A system that can do this must evolve over time (i.e. have a history or biography) and cannot be described formally by any single theory, i.e. it cannot be understood completely. For each theory you give to describe what it is doing, it can develop out of the scope of that theory. You can extend the theory so that it covers the new cases but the resulting theory is incomplete again. See also THINKING OUT OF THE BOX and INSPIRATION.
2.4 The hypothesis of creative systems
Based on these ideas Ammon develops the hypothesis that programs exists that can, in principle, develop any kind of knowledge based on a sequence of inputs and that human cognition can be thought of as such a system. Both human beings and groups of human beings (cultures, communities, scientific communities etc.) may be viewed as creative systems in this sense.
2.5 Creativity and Reflexivity
In the article (Reflexivity and Creativity) I have sketched the connection between creativity and reflexivity.
3.1 The development of knowledge
Creative systems can develop new cognitive structures not contained in previous stages of their development. For any given cognitive structure in such a system, there might exist a simpler system that can develop that particular structure if exposed to appropriate inputs. The simpler system will tend to be less economic or less efficient.
3.2 There are no synthetic judgments a priori
However, this means that it is not necessary to assume any sophisticated structure to exist in the beginning. In particular, there are no synthetic structures a priori in the sense of Kant. Sophisticated, complex structures develop based on simple structures.
There might be innate structures but in principle, these could develop by cognition alone.
3.3 Errors and anomalies
There is always the possibility of error but also a chance for each error to be corrected in the course of the development. The knowledge (the term is not used in the sense of secure knowledge but in the sense of information stored in the system) may always contain anomalies like errors and inefficiencies. These can be eradicated during the development of the system but the knowledge is always incomplete.
3.4 Analytical spaces
Creative systems develop by the division and unification of analytical spaces, i.e. finite connections of knowledge, together with the objects this knowledge refers to. This has been sketched in (Analytical Spaces)
3.5 The trade-off between exactness and generality
There is a trade-off between explicitness/exactness of knowledge and generality of knowledge. Some analytical spaces (e.g. in physics) are very exact and explicit but special; others are very general but less explicit and less exact. Both goals cannot be achieved at the same time. It is not possible to unify all analytical spaces into one single analytical space.
3.6 Creative systems and scientific theories
Creative systems cannot be described completely by any single formal theory. This means that reality contains systems that have more properties than can be derived in any single theory. As a corollary it follows that any “theory of everything” must be computationally incomplete, i.e. while it might describe reality completely in an implicit way, it would not always be possible to solve its equations. We should expect to find systems where any given set of mathematical methods to solve instances of the theory is incomplete. Extensions can always be made but there is always the necessity to add new methods of calculation to the scientist’s toolbox. So the normal scheme to describe scientific explanations by initial conditions + laws is incomplete since the laws might contain or define functions that are not Turing-computable. You would have to add a third thing: calculation methods, and these would always be incomplete.
In (INFORMATION STORAGE CAPACITY AND CREATIVITY) I have argued that for systems with a fixed information storage capacity, it should be possible to devise a complete theory. However, if you add information storage capacity, the theory will become incomplete, so qualitatively new phenomena might happen. In physical systems, the information storage capacity can increase when energy is added or when the radius of the system expands, so in an expanding universe, new things might happen.
So, finding a theory of everything might be not so useful after all.
4 Philosophical Anthropology
4.1 A new approach to philosophical anthropology
The hypothesis that human beings can be viewed as creative systems offers a new approach to philosophical anthropology, i.e. the discipline of philosophy concerned whit the question: what is the human being? The central concept here is the idea of creativity in the sense outlined above.
4.2 Previous approaches
Previous approaches to philosophical anthropology (Scheler, Plessner, Gehlen) provided different approaches to this question. A similarity of these approaches is that they describe the human being as open to the world, i.e. not restricted to a single scheme of perception and action. The description of the human being as a creative system is compatible with this.
4.3 The human being as a creative system – the division of sciences.
The concept of creative systems means that it is not possible to describe human beings completely in terms of a single formal theory. A complete description of humans by a finite body of knowledge is impossible in principle (since humans are able to create things new with respect to any such body of knowledge).
4.4 Philosophy of Culture and History
A complete formal theory of human culture is impossible. Cultures (i.e. groups of people with the knowledge they hold), are creative systems. Since there is no complete theory about human culture and cognition, humans have history. A theory of history in the sense of a formal theory is impossible.
The knowledge, methods of research and academic disciplines concerned with human beings and their culture can never come to a closure. As a further result, human culture and history cannot be captured completely in terms of a single formal theory, justifying the division of natural sciences and cultural sciences (or humanities).
4.5 Origin and evolution of humans
The origin of humans can be described as the point in the history of our species where cognition itself became the main process of changing the cognitive structures, instead of genetic, biological evolution. I think that this did not involve the formation of any sophisticated mechanisms or “algorithms of thinking”. Instead, the processing and storage capacity of the brain increased, giving the cognitive processes more room to develop. Any sophisticated structures of thinking developed in terms of cognition, not evolution, i.e. they are part of our culture, not of our genetics.
As a result of this hypothesis, I think that the “semiotic revolution” observed by archeologists, i.e. the increase in the complexity of human culture during a certain period in the older stone age, was not the result of structural changes in the brain due to genetic mutations, but the result of cultural inventions. This might have been, for example, the cultural invention of some new constructs of language and thinking, e.g. universal quantification (“for all”) or another advance in language and cognitive representation. If all the semantic devices, representations and logical devices of our language and thinking are not hard-wired in our brains (i.e. are not genetic) but cultural, they have to have been invented (cognitively/culturally) at some point in our history, i.e. they are a product of creativity, not of evolution.
5 Cognitive Psychology
The theory of creative systems provides a new paradigm for cognitive psychology. Like in the analytical philosophy of mind, cognition is viewed as information processing, but the cognitive structures are due to change by the cognitive processes themselves. There are not fixed algorithms of thinking, no fixed representations inside the mind (a fixed “mentalese” as an internal “language of thinking”) or laws of thinking. All of these things are a result of cognition and culture and may change both in the development of cultures and in the development of an individual.
6 Artificial intelligence (AI)
The common paradigm of AI is flawed insofar it is searching for fixed structures of intelligence. There are no unchangeable fixed structures. Common AI-Systems can be viewed as algorithms, i.e. fixed programs. They are not changing themselves in response to inputs but are developed and reprogrammed by humans. As a result, for example, current speech recognition applications (like available through some smart phones), are limited to certain domains of knowledge. New such domains can be added (by programmers) but the systems are not creative themselves. They are not intelligent in the sense of the theory of creative systems.
True intelligence involves creativity at its core. As a result, an artificial intelligent system might be possible, but it would have the strange property of not being completely describable (and thus, understandable) in terms of any single, fixed theory.
It remains to be seen if an artificial creative system, if ever implemented, could be controlled so as to be technically useful or if it would develop into directions “of its own choice”. It might turn out that creative systems are technically feasible but not usable as a technology since a technology will normally match certain means with certain goals. With systems whose properties cannot be completely described, this might not be possible.
If artificial creative systems can be created, it must be investigated if these may become conscious. This might lead to novel ethical problems (see An open letter to the “Human Brain Project”).
The concept of creativity has interesting consequences for linguistics and philosophy of language.
7.1 Communication as a creative process – hermeneutics
Communication can be viewed as a creative process. A speaker might create a message containing new signs (e.g. new words or new grammar). A hearer might understand this utterance by creatively amending the gaps in the code, based on the context. So communication with incomplete codes is possible. For example, if you hear a word in a foreign language you don’t know, you might form a hypothesis of its meaning. Other examples include the adaption to an unusual dialect of the faulty language of a non-native speaker.
Communication can thus be seen as a hermeneutic process in which you start from a (possibly incomplete or partially incorrect) pre-understanding and interpret an utterance based on that. As a result, the pre-understanding (pre-existing knowledge) is modified and extended.
These processes are creative, i.e. there is no complete formal theory describing them. It is not possible to completely formalize hermeneutics. If you formalize hermeneutics, you make it special (you may view the limited language-understanding programs mentioned above as formalized hermeneutical theories. Making them formal (algorithmic) made them special and limited.
7.2 Origin of Language
If communication with incomplete codes, i.e. hermeneutics or communication as a creative process, is possible, it may start with no previous language at all. In this view, the origin of language is not the result of the creation of a sophisticated brain structure by genetic mutations but the result of inventions, i.e. creative cognitive processes. Only a sufficient amount of processing power in the brain (i.e. of nonspecific brain areas, i.e. of plasticity) was necessary for this. Language is completely a cultural phenomenon. There are no universals of language.
7.3 Language Acquisition and language Change
Language acquisition is the creative recreation of the grammatical, lexical and cognitive system of a language by a child. This process may be imperfect, leading to language change. Parent’s ability to understand imperfect utterances may lead to change-tolerance on their side, where a certain degree of tolerance might be part of the language’s system. As a result or non-perfect language acquisition and fault-tolerance in parents or other people, languages change over time. So creativity leads to innovation and split of languages.
Contrary to hypotheses put forward by linguists like Noam Chomsky, there is no fixed (algorithmic) language acquisition device (“LAD”) hardwired in our brains. Language acquisition is a creative process and the LAD is developed creatively during language acquisition. As a result, there is no basis to distinguish strictly between “performance” and “competence”. The grammaticality of sentences of real languages cannot be completely captured in formal theories like formal grammars. A grammar is always an incomplete description since the real language can be changed creatively at any time.
A complete formal semantics (like it was attempted by Donald Davidson and others) is not possible. There is no complete formal description of the grammar of any language and there is no complete formal description of cognitive processes. As a result, there is no complete formal theory describing how utterances are mapped into or influence cognitive processes and vice versa. Partial formalizations or description in terms of algorithms are possible but such descriptions are incomplete in principle. Language and hence its semantics are extensible and adaptable.
There are attempts in analytical philosophy to provide exact definitions of concepts and to eliminate vagueness. There is a tradition to view vagueness of natural language as a flaw. In contrast to that, I think that vagueness of concepts is normal and contributes very much to the adaptability of natural language. A vague concept is a concept with an incomplete definition. In each instance of using the concept, the definition might be adapted. There is no complete theory defining the adaption process. Communication (seen as a creative, hermeneutic process, see above) and cognition can use such incompletely defined concepts and still succeed. Vague concepts may be viewed as incomplete and developing analytical spaces.
See (FROM MONOCHROME TO NOISE – AESTHETICS, INFORMATION THEORY, AND CREATIVITY) as well as some of the articles on this subject in (http://asifoscope.org/category/aesthetics/).
9 History of ideas on creativity, formalization vs. creativity in the history of science
See the introduction to (http://arxiv.org/pdf/1302.1155v1.pdf) where Ammon outlines some of the scientific discussions on the topic. Based on this, but widening the scope, the history of ideas about creativity in general and the concept of creativity in Ammon’s sense should be investigated. There is a tradition (going back to Frege and Hilbert) to try to completely formalize logic, math, cognitive science and linguistics. The current paradigm of “algorithmic AI” also belongs into this tradition. This tradition has failed but parts of it became dominant in the academic world. The history of these developments should be investigated.
Additionally, the theory of creative systems should be compared to other approaches, e.g. the theory of autopoietic systems, and the relationship between these approaches should be analyzed.
10 Creativity and Mathematics
Ideas about this have been developed by Ammon himself. I refer to his publications.
11 Biological evolution as a creative process
There are certain parallels between creative cognitive processes and biological evolution. The parallels as well as the differences should be investigated.