Chapter 7
Introduction to the Scientific Knowledge of the Mind
7.1. Research program
7.1.1. Profession of pragmatic faith
The aim of my research program is to increase human potential in general and the human capacity for development in particular. The justification for the model presented here is pragmatic: its function is primarily to support the goal of cognitive augmentation of the species. I therefore make no claim to deducing my hypotheses from absolutely true axiomatic principles or infallible logical reasoning. I think my hypotheses are relevant because of their results: they establish a reasonable basis for scientific knowledge of the mind, knowledge that as much as possible uses the new digital possibilities for ubiquitous recording and calculation.
7.1.2. Initial questions
The Internet is already enhancing our individual and collective cognitive processes: it gives us access to huge quantities of multimedia data in real time, expanding our capacity for memory and perception. It also enables us to communicate and coordinate ourselves to a degree unknown to previous generations. Although the digital medium is gradually gathering in all the works of the mind accumulated by humanity over the centuries1, although it contains the vast majority of our contemporary thought, and although it has become the preferred context for our exchanges and transactions, it still does not offer us a readable image of the functioning of our collective intelligence. Yet all the information is there, ubiquitous, ready to be processed by means of constantly increasing calculating power. The data can certainly be found and analyzed at the level of documents or well-organized series of documents (for example, by ontologies), but their overall cognitive dynamics remain opaque. How can we model the cognitive processes of online creative conversations while improving their knowledge management? How can we transform the Internet into a rigorous observatory of economic, social and cultural phenomena that promotes human development? In short, how can we use all the resources of the digital medium to enhance collective intelligence?2 The answers to these questions, which I have been asking since the early 1990s, obviously require a scientific theory of human collective intelligence. Before the IEML model presented in this book, we had no such theory.
7.1.3. Instruments
Since the late 20th Century, it has been clear to me that the development of the digital medium has been creating new conditions for the scientific modeling of symbolic cognition. The modeling tool is no longer “the computer”, but the interconnected set of symbol-manipulating automata, an evolving society of agents that is rapidly growing. The data to be manipulated to simulate cognition are no longer contained in one clearly delimited database; they spring up in the huge multimedia hypertext of the Web: a global reservoir accessible anywhere, fed constantly by the multifaceted activities of Internet users and myriads of distributed input devices. Thus my research has dealt with a way of modeling human symbolic cognition that would make full use of this instrument of observation and calculation now available to us.
During the Renaissance, in the new communication environment opened up by printing (an instrument for reproducing and disseminating ideas), the invention of the telescope and the microscope (material instruments of observation) and calculus (a symbolic instrument of computation) expanded the horizons of cosmology and physics. Similarly, the digital medium’s potential for communication, recording and distributed calculation enables us today to expand the horizons of the cognitive sciences. At the same time, we need to envisage improving the technical tools through science: improving the scientific modeling of cognition could give the digital medium the transparency and reflexivity it still lacks in 2011, at the time I am writing.
7.1.4. Subject-object
One of the first questions that arose for me was that of the subject of cognition. Who or what is thinking? One of the first answers that comes to mind is “the brain”. This, however, was not the answer I opted for. While I do not doubt that the brain is a biological medium essential to human cognition, modeling “the brain” and modeling human symbolic cognition are two different things. Even if we only consider the physical/biological medium of symbolic cognition, an isolated brain, or even a human body, is insufficient. The production of symbolic thought requires at least a society in a natural and technical environment. A society most often exists at the confluence of several cultural traditions. There is therefore no individual subject of cognition that is not immersed in broader sociocultural cognitive systems from which this individual subject receives languages, customs, values, tools, etc. Although it is manifested in a personal reflexive consciousness or intelligence that is indisputably individual, symbolic cognition is necessarily inscribed in a collective cultural field. I consider the human brain to be a basic cognitive processor, but I believe that symbolic cognition emerges only from the interconnection of brains implementing cultural “programs” in a coordinated fashion. In the rest of this book, I will call the network of human brains that cooperate in using symbolic systems based on a material culture the Cortex.
The main object of my scientific quest, like the subject that is capable of knowing this object, is none other than the mind — human symbolic cognition considered in its dynamism and its specific content, independently of its technical/biological media (although, obviously, such media are a necessary condition for its very existence). The distributed socio-semantic processes designated by the word mind include infra-personal, personal, collective, conscious and unconscious cognitive processes on all time scales, with the understanding that symbolic (and therefore cultural) systems operate on all levels and at all scales3. Throughout this text, the word mind designates the sphere of communication between the functions of symbolic cognition. As we will see, the IEML model of the mind ensures the computability and interoperability of these functions.
7.1.5. Method and result
The initial postulate of my whole undertaking is that the mind lends itself to scientific modeling. This means that, through inevitable abstractions and simplifications, it is possible to describe the human mind using a coherent system of calculable functions. Starting from this original intuition, my research involved developing a formal model of the mind that met the requirements of contemporary scientific method and that as much as possible used the reservoir of data and calculating power of the digital medium. When I obtained a Canada Research Chair in Collective Intelligence at the University of Ottawa in June 2002, I threw myself body and soul into an extended research process, a kind of intellectual marathon that, at the time, I never imagined would last 10 years. I was working under exceptional conditions: my teaching load was reduced and I had guaranteed funding, which I have mainly used for expert collaborators. During these 10 years, in order to solve problems encountered along the way, I had to improve my knowledge and skills in computer science, mathematics, linguistics and graphic design. I read articles in the cognitive sciences, but also many on philosophy, mainly the classic texts of various traditions. As I explained in the introduction to this book, throughout these years, five or six key elements (sign, being, thing, virtual, actual, emptiness) served as my Ariadne';s thread. I represented and combined them in all sorts of ways using a range of software until I obtained a satisfactory version of the IEML language, which will be described in Volume Two. It goes without saying that this research was carried out “organically”, with countless trials and errors, periodic returns to the same problems slightly modified or refined, and no guarantee that I would finally reach a favorable outcome. The main result of my work is a scientific advance in the study of human symbolic cognition: the development of the IEML semantic sphere — a system of coordinates for representing the mind as a unique, infinite nature describable in calculable functions. This semantic sphere is the mathematical/linguistic framework of a digital Hypercortex that will make it possible to observe and simulate human cognitive processes.
The presentation below is a simplified logical reconstruction of my research process rather than a detailed history of my trials and errors. The constraints of print publishing oblige me to present this work in two volumes, but readers should understand that the two volumes form a whole and that many aspects of IEML language will only be revealed in Volume 2, in particular the dictionary, the rules of grammar and the semantic topology. The complexity of the model I am presenting here necessitates a certain amount of repetition: each chapter concentrates on one specific aspect, but refers to certain elements of the whole in order to make it comprehensible. The introduction to Part 2 presents a synopsis of the model that will be developed in the rest of Volume 1. The reader will be able to refer to it when an overview is required.
7.2. The mind in nature
7.2.1. The uni-duality of communication nature
The nature whose structure I am now going to describe is neither absolute nor eternal. The word nature here is a technical term whose limits are defined by two conditions of validity. First, this nature emerges through symbolic cognition. It thus does not pre-exist our species in the course of biological evolution. I have no idea what this nature would be independently of its reflection in human consciousness. Second, the nature I am going to discuss appears to human knowledge from a scientific perspective. The meaning of the expression scientific knowledge is precisely the question in Part 2, and it will be revealed gradually. I want to emphasize from the outset that forms of knowledge other than scientific knowledge can obviously lead to other representations of nature and other visions of the world in general, all of them just as relevant as mine in their own domains of validity.
7.2.1.1. Virtual and actual spheres of communication
I am starting from the principle that nature is communication, i.e. that messages carrying information are exchanged in it. We can distinguish two main spheres of communication: actual (“matter” in ordinary terms) and virtual (“mind” in ordinary terms). By actual I am referring not to any particular substance, but to the sphere of communication in which messages are perceptible phenomena. Similarly, by virtual I do not mean a substance, but rather a sphere of communication whose messages are intelligible (and thus invisible) forms or concepts. Concepts are received, manipulated and transmitted through processes of symbolic cognition. Virtual and actual imply each other, since the medium of invisible messages – or signifieds – can only be visible or perceptible in general. Signifieds are necessarily presented to our senses through perceptible signifiers, whether through direct perception or in imagination, fiction or dreams. At the same time, the perceptible forms of the phenomenal world can only appear to us carried by the medium of symbolic cognition because when these forms are apprehended they are necessarily categorized and integrated into some narrative or theory: they have a meaning (see Table 7.1). Thus, in the virtual sphere of communication – or the nature of the mind – the (semantic) messages are invisible and the media are visible. On the other hand, in the actual sphere of communication – or material nature – the messages are visible and the media (the processes of symbolic cognition) are invisible.
Table 7.1. Medium and message in the nature of communication
| Virtual | Actual | |
| Invisible | Message | Medium |
| Visible | Medium | Message |
7.2.1.2. Actual space-time
The science of material nature situates communication between perceptible phenomena in a system of space-time coordinates. Leaving aside relativistic effects and string theory, actual space is presented here in a three-dimensional geometric form, while time is presented as a pure linear or sequential succession. This space-time of science is a useful convention for the coordination of human activities and the functional (mathematical) description of communication among material phenomena. The system of space-time coordinates is in no way a spontaneous datum of experience. It is obviously a relatively recent acquisition of cultural evolution, whose main advantage is that it leads to universal, calculable, interoperable representations of physical phenomena. Scientific and technical activity has conquered or constructed this space-time through many centuries of labor, and it has gradually been integrated into daily life and common representations through techno-social institutions such as clocks, calendars, time zones, maps, GPS, laboratories and networks of measurement. Hypothetically, in the material nature described by science, communication takes the form of causal circuits. As complex as they may be, these circuits are formed entirely within the system of space-time coordinates. Consequently, a cause necessarily precedes its effect. Goal-oriented, or teleonomic, behaviors do exist, but they emerge from feedback loops or automatically executed programs, and can therefore always be reduced to temporal sequences in which causality (and therefore communication) circulates from the past to the future.
7.2.1.3. Virtual space-time
Let us now imagine a science of the mind. What would the system of coordinates, the basic framework of communication of the virtual sphere, look like? It is clear, to begin with, that three-dimensional geometric space is not adequate for the localization of concepts. No one can say where justice or truth is located in three-dimensional space, although we can point to physical places and times in which these concepts are actualized. Nor can we say that they are located in our brains, since no close observation of these brains will ever show anything but neurons, circuits of excitation and discharges of neurotransmitters in synapses. We will never observe concepts. I admit that it is impossible to think about concepts without a working brain, but we cannot deduce from this that concepts are located in the brain (in the sense that we say that the neurons are located in the brain). Neurons and concepts belong to two different spheres of communication. In the rest of this book, I will show that a system of coordinates of the mind must be presented as a hypercomplex network of interrelated concepts. For example, in this system of coordinates, the concept of justice is related to the concepts of injustice, balance, equality, law, decision, innocence, guilt, retribution, etc. The concepts are interconnected in relationships of meaning in a fractal tangle of semantic circuits with a structure that is very different from that of the geometric space that coordinates the actual sphere.
What about time? It is clear that the universe of meaning is not organized according to a simple sequential temporality. Although thoughts follow one after another sequentially in our experience, each thought, at the moment it occurs, also resonates with previous thoughts along complex semantic and affective circuits. In the mind, communication drives the transmission and transformation of meaning. Its operation is not causal but interpretative. Symbolic cognition sets down, organizes and reorganizes relationships of meaning in a dynamic interpretive memory within which experiences of life and learning in turn act on our understanding — and therefore the meaning - of past events. In the world of ideas, the virtual past of memory can be affected by its future. Distinct from material causality, narrative time governs the development of meaning. The virtual time of memory - the dynamics of meaning - is woven, unraveled and rewoven in hypertextual narrative patterns. Semantic communication propagates simultaneously in the multiple branching of the narratives and theories that structure the reticular universe of symbolic cognition. Far from being sequential or simply tree-like, the channels that carry semantic communication are organized in rhizomes that sprout and branch out in all directions of the mind. While in terms of the visible medium of the mind (see Table 7.1), the reading and writing of signifiers there is always a “before” and an “after” on an irreversible sequential line, the invisible message of meaning is organized in rhizomes in the living duration of memory.
7.2.1.4. The interdependent co-emergence of the virtual and actual spheres
Since nature is communication, we must now consider the relationship between the virtual and actual spheres of nature. As we have seen in Table 7.1, the medium of each of these spheres is the message of the other. On the one hand, the virtual world of meaning cannot exist without the biological medium of signification - actual phenomena. On the other hand, the perceptual phenomena of the actual sphere are defined by processes of symbolic cognition that actively construct the meaning of these phenomena, even when the perceptual forms are experienced as the direct result of pre-existing realities. There is no physical nature4 without conceptual categorization or affective polarization, and no spiritual or intellectual world without perceptible signifiers or a biocosmic medium. In the nature of communication, mind is not the opposite of matter, but its partner: virtual and actual, noumenon and phenomenon, physical sphere and metaphysical sphere co-emerge interdependently, with each actively needing the other in order to exist. With the complex uni-duality of communication nature established, we now need a description of mind that is as scientific as the description of matter that has been achieved in physics, the molecular sciences and biology.
Before going any farther in this direction, let us pause for a moment at the mysterious crossroads where virtual and actual communicate and exchange with each other: the human presence.
7.2.2. The uni-ternarity of communication nature
The human species is at the “center” of nature as described here because it is thus far the only conscious carrier of the ideal forms that exist in the virtual sphere, and the only species capable of contemplating these abstract forms and using them skillfully to act in the actual sphere. We are able not only to say and understand that “this” represents “that”, but also to manipulate “this” and “that” in complex and systematic ways while maintaining the trace and the active memory of the correspondence between “this” and “that”. In addition, we know that there is “someone”, an interpreting subject, for whom “this” represents “that”. In fact, without such an interpreter, it would be impossible to conceive of a correspondence between a signifier and a signified. Meaning cannot be something objective that resides simply in material phenomena. To fully exist, the virtual sphere of meanings requires cultural conventions, symbolic systems and socialized individuals capable of interpreting signs according to the appropriate conventions. In my technical vocabulary, I say that our species carries the ternary relationship sign (S)/being (B)/thing (T), i.e. that it produces and reproduces interpreters (B) for whom there are signs (S) evoking concepts and referring to (virtual or actual) realities (T) according to contexts and “rules of the game” that are infinitely varied. The human interpreters are capable of playing in all kinds of ways with the interpretive triad being/sign/thing. This basic generator of meaning is what creates the specificity of the human presence.
In the nature of communication, the human presence (“now”) simultaneously generates two temporalities: one that links perceptible phenomena in matter, and one that links ideas in the mind. Through sensory-motor experience, presence changes into sequential time in the actual sphere. Through the semantic experience of symbolic learning and thought, presence changes into hermeneutic and narrative memory in the virtual sphere. Presence is projected into the geometric space of the physical cosmos in the actual sphere, and into the semantic topology of the world of ideas in the virtual sphere. The nature of communication in its totality, both actual and virtual, emanates from the human presence.
This presence that mediates between the “Heaven” of symbolic cognition and the “Earth” of material bodies directs the “ascending” currents of virtualization and the “descending” currents of actualization. In the center, the communicative presence animates an ontological “breathing”: in the movements of “breathing in” — or virtualization — phenomena signify, and in the movements of “breathing out” — or actualization — meanings are manifested and embodied. Virtualization reflects visible light into invisible forms and actualization projects the invisible light of ideas into the sensory world. At the nexus of this reciprocal translation, presence functions as an affective Moebius strip that reciprocally transforms the medium into the message, and the visible into the invisible. The human presence appears at the center of nature like a source of non-dual existential light, impossible to grasp, preceding the distinction between visible and invisible, virtual and actual.
Figure 7.1. Nature of information and communication: mind, presence, matter
Figure 7.1 shows the double series of concentric translucent spheres that reflect and reveal the existentiating light of presence. In the sphere of the actual, the burning nucleus is made up of a transitory human organism and its sensory-motor activities. Within this nucleus, communication is dense and rapid. The living human body is itself surrounded by a second concentric sphere, a hot magma made up of other human bodies, tools, machines, buildings, infrastructure, media and networks with which it interacts and which organizes its relationship to the material world. The techno-cultural ecosystems of this second concentric sphere are obviously very varied and constantly evolving. A third, relatively cold sphere surrounds the magma of material culture: the cosmic envelope. This material cosmos is primarily made up of the terrestrial biosphere (the least cold layer of the envelope) and, beyond it, physicochemical layers, within which there are levels of astronomic, planetary, molecular, atomic and quantum complexity. For our physical science, there is a single universal material cosmos that envelops material cultures (and, through these cultures, human bodies). Science describes the cosmos through calculable functions, using the system of space-time coordinates that unifies it. At the same time, however, it is understood that the complexity of the physical cosmos is inexhaustible by our finite science.
Let us now analyze the virtual sphere, which has the same structure in three concentric envelopes as the actual sphere. The burning nucleus of the virtual sphere is a hypercomplex metaphysical form: intelligence, or the individual mind, which may be defined as a process of construction of memory driven by personal learning. Within this nucleus of individual intelligence, communication is dense and rapid. Individual intelligence is dependent on the human body of the actual sphere. The hot magma of collective intelligence surrounding the nucleus is made up of “language games”, symbolic systems and cultural conventions in which the individual intelligence participates. This second concentric sphere of evolving symbolic systems represents the immaterial dimension of culture and corresponds to the magma of material culture in the actual sphere. Symbolic systems organize the relationship of the individual intelligence to the world of ideas and enable it to coordinate its learning and memory with other individual intelligences. The hot magma of the evolving symbolic systems is itself contained in a colder sphere: the envelope of the mind, in Figure 7.1, designated the “symbolic mechanism”. This envelope contains the potential for symbolic manipulation carried by the human species and the set of the ecosystems of ideas generated by this potential. Its temporality — that of memory — distinguishes it radically from the temporal phenomena of the physical cosmos.
Although it is possible for purposes of analytical description to distinguish zones, spheres and levels in nature, I would like once again to emphasize its unity: mind and matter are spheres of communications and their respective concentric subspheres are inextricably contained in each other. In addition, the virtual and actual spheres are interdependent. At the center of nature, the human presence simultaneously illuminates the visible and the invisible; it implies one in the other and it reciprocally explains matter and mind, geometric space and the complex topology of semantic circuits, sequential time and interpretive memory.
Physical nature and the nature of the mind are two interdependent images of one and the same nature of information and communication. Just as the physical cosmos can be described in calculable functions using a system of space-time coordinates, the world of human ideas can be described in calculable functions using a system of semantic coordinates: the IEML semantic sphere.
7.3. The three symbolic functions of the cortex
Figure 7.2. Projection of the cortex in the digital medium
How does the human presence generate the world of ideas, which is unknown to other animals? To answer this question, we have to go into the “factory of the mind”, which I call the Cortex. The term Cortex here is a technical term that designates the actual dynamics of symbolic communication among the brains of living human beings. It should be kept in mind that the Cortex is not a static entity, but a dynamic process. Collective human intelligence emerges from the interaction between the Cortex and nature. The symbolic cognition specific to human beings is additional to the non-symbolic cognition the human species shares with the other animal species and it reorganizes that non-symbolic cognition. The operation of the Cortex, the symbolic dimension of human cognition, may be described in terms of the dialectical interaction of three types of manipulation: (i) manipulation of symbols, or signifiers, which corresponds to the syntactic function; (ii) manipulation of concepts, or signifieds, which corresponds to the semantic function; and (iii) manipulation of data, or referents, which corresponds to the pragmatic function. The Cortex in Figure 7.2 shows the dialectical unity of these three functions.
7.3.1. The syntactic function
We can only think or form representations of general categories by using systems of symbols: languages, writing systems, icons, etc. I should point out here that symbolic forms can appear to any of our senses or to any combination of these senses. Communities of the deaf have developed sign languages. Systems of religious symbols and rituals in general can involve songs, dances and “multimedia” physical environments of all kinds. The point to remember is that the abstract thought that is specific to humans necessarily has to operate through signifying sensory representation. The human mind is capable of processing these signifiers in very elaborate ways. It is the syntactic function of symbolic cognition that expresses our capacity to break down, arrange and rearrange complex signifying structures.
We know that human beings are capable of respecting the syntax of very complex languages through imitation, even without having formally learned their grammars. The existence of the syntactic function also explains our ability to use abacuses and number systems, and therefore to perform calculations with numbers. If we consider operational movements and tools as symbols to be combined, the syntactic function also explains the technical development that distinguishes humanity. Finally, social games with varied symbols and complicated rules are practiced in all cultures, whether these games are “purely playful” or are “serious”, such as family, political, legal and economic games.
7.3.2. The semantic function
The manipulation of symbols is obviously not a goal in itself. Its role is to support the semantic function, the manipulation of concepts, signifieds and categories. This manipulation of concepts is not limited to logical reasoning, but also includes games of opposition, complementarity, analogy, derivation and linguistic composition between the signifieds, including all the refinements of dialogue and narration. The semantic function explains both our capacity to produce and to comprehend (in the etymological sense of “take together”) conceptual architectures that can be indefinitely complex. We can transform arrangements of symbols, so we can also carry out all kinds of transformations on the architectures of concepts represented by these arrangements. Just as the syntactic function is based on the discipline of grammar, the semantic function has often been studied under the term dialectic, in the sense of a very general ability to organize relationships among concepts. Dialectical ability involves breaking down, synthesizing, transforming and ordering signifieds in relevant structures.
7.3.3. The pragmatic function
7.3.3.1. Interpretation, memory, action
The manipulation of concepts is not a goal in itself, either. The very concept of the relevance of conceptual architectures implies a situation, real or fictional, in which signifieds are related. Concepts categorize sensory data according to a practical intention, whether the data are perceived, remembered or imagined. Just as symbols are used for manipulating concepts, concepts are used for manipulating data or percepts. The pragmatic function accompanies the immersion of the thinking subject in the temporality of memory and action. With regard to memory, perceptual data are organized according to their conceptual meaning and their affective value for the subject. To navigate in memory, the pragmatic function draws a rhizomatic graph of conceptual and affective relationships among perceptual data. With regard to action, the pragmatic function categorizes percepts according to the subject's goals and maintains compatibility with the subject's emotional and conceptual memories. As the name indicates, the pragmatic function aims primarily for effective action. However, just as in medicine the effectiveness of the treatment is subordinate to the accuracy of the diagnosis; the effectiveness of the action of the pragmatic function is subordinate to the refinement and relevance of the conceptual and affective interpretation of the data.
7.3.3.2. Ideas
By categorizing percepts and attaching an affective value to the categorized percepts, the pragmatic function produces ideas. Ideas are organized in ecosystems. They are connected by the semantic relationships of their concepts and the sensory relationships of their percepts, and they exchange their affects. We can thus analyze ideas in three interdependent components:
– a sensory datum, or percept (T);
– an affect (B); and
– a concept (S).
7.3.3.3. Pragmatics and general rhetoric
Just as the art of the syntactic function is grammar and the art of the semantic function is dialectic, the art of the pragmatic function is rhetoric. Rhetoric, in fact, includes both an art of memory and an art of effective symbolic action. Rhetorical skill organizes data to be retained, both for the orator and the audience. If the ideas are not imprinted in the mind of the orator, how can they be in the minds of the audience? It is by controlling the ideas in memory on the basis of their conceptual, emotional and perceptual dimensions that rhetorical skill ultimately controls the data of the situation. We can generalize from special rhetoric (the art of persuasion) to an expanded rhetoric that uses social conventions, as well as the ideas and emotions of a community, to ensure the maximum effectiveness of symbolic action.
7.3.4. The sign (S)/being (B)/thing (T) dialectic of symbolic cognition
The syntactic capacity to manipulate symbols serves the semantic capacity to manipulate concepts, since we cannot apprehend abstract categories except through the medium of signifiers. In turn, the semantic function serves the pragmatic function of manipulating data (or percepts), since concepts qualify and designate realities, organize memory and, through the affective force of ideas, act on social contexts. Furthermore, memory, semantically organized by the pragmatic function, obviously serves as a medium for syntactic function, since we could not retain and manipulate so many symbols if they had no conceptual and practical relevance. In a sense, the pragmatic function is central, since the semantic and syntactic functions are only justified by their pragmatic use. In another sense, however, the semantic function is the highest, since there would be no pragmatic function at all if the concepts were not there to give meaning to data and situations. The percepts and the affects of ideas draw their meaning from the concepts. Finally, we can consider the syntactic capacity to manipulate signifiers as the root or source of symbolic cognition, since without it we would be reduced to animal cognition: there would be no language, technology, culture or reflexive intelligence. We therefore have a ternary dialectic — sign (S)/being (B)/thing (T), each pole of which is both clearly distinct from the other two, since it occupies a different function, and is absolutely dependent on the other two, since none of the three functions can carry out symbolic cognition separately. This tripolar dialectic should be thought of in its generality as a symmetrical interaction among three roles, which may be played by different conceptual actors depending on contexts, disciplines or intellectual tradition. The linguistic version of this tripolar dialectic connects the signifier (S), the signified for an interpreter (B) and the referent (T).
Figure 7.3. Sign/being/thing dialectic in symbolic cognition
Figure 7.3 provides some variations on the tripolar dialectic of symbolic cognition5. We can see that the syntactic function permits the manipulation of symbols, expresses the computational faculties of human cognition and is studied in the generalized art of grammar. The semantic function controls the manipulation of concepts, explains the faculties of linguistic representation of human cognition and is studied in the art of the dialectic. Finally, the pragmatic function produces ideas, controls the manipulation of data, organizes interpretative human memory and is studied in general rhetoric6. The S/B/T dialectic of ideas (concept/affect/percept) is internal to the pragmatic function.
It should be noted that the virtual/actual dialectic is itself indissociable from the sign/being/thing dialectic. Indeed, only symbolic cognition, because it is reflexive, can distinguish between the concrete actuality of bodies and events inscribed in the space-time continuum and the virtuality of possibilities, abstractions, concepts and ideas envisaged by the mind.
7.4. The IEML model of symbolic cognition
7.4.1. The semantic sphere: the mathematical basis of the IEML model of the mind
In constructing the IEML model of symbolic cognition, I had to meet two major requirements. First, the model obviously had to take into account the functioning of the human Cortex, respecting its major connections and its in principle unlimited capacity for the manipulation of symbols, concepts and data. Second, my model had to make maximum use of the calculating power, ubiquity and interconnection of the digital medium. This second condition meant not only that each of the major functions of symbolic cognition actually had to be calculable by logical automata, but that they had to be interoperable. In the IEML model, this general interoperability is provided by the semantic sphere, which operates as a universal system of mathematical coordinates of the mind.
7.4.2. The Cortex, the Hypercortex and the semantic sphere
The Hypercortex must be clearly distinguished from the semantic sphere. The Hypercortex is a technical mechanism capable of reflecting a simulated image of the cortical process of symbolic cognition. This image is thus, like the Cortex it reflects, a dynamic, evolving process. The semantic sphere is the system of mathematical coordinates, the virtual grid used by the Hypercortex to reflect the image of the Cortex. The relationship between the semantic sphere and the Hypercortex is therefore a relationship between a scientific instrument of observation (the Hypercortex) and the projection system by which it is organized (the semantic sphere). In short, as suggested in Figure 7.2, the Cortex is the dynamic object reflected by the Hypercortex against the background of the semantic sphere, while the digital medium is the almost-unlimited source of the data and calculating power used in the process of reflection.
7.4.3. The Cortex, the Hypercortex and the mind
I would now like to distinguish between the mind and the Cortex. The Cortex designates the process of symbolic communication among human brains that supports actual collective intelligence. This collective intelligence is augmented by the media and systems of signs developed in the course of cultural evolution. The human mind has always included a dialectic between its Cortex and the intellectual technologies available to support, augment and reflect its symbolic functions (we only have to think of the role of libraries). While the emergence of the digital medium is very recent on the scale of human history, it is nevertheless a sign of a convergence, an accelerated evolution and reciprocal multiplication of multimedia capabilities and symbolic codes. The role of the semantic sphere is to bring consistency to this movement in order to augment the Cortex. Once the semantic sphere enables the Cortex to be reflected in a Hypercortex that presents it with the scientific image of its own functioning, its operations will become both more precise and more powerful. The Hypercortex mobilizes the media, systems of signs and intellectual technologies that have always augmented the Cortex, but it does so by making maximum use of the calculating power, ubiquitous communication and access to data that characterize the digital medium. The mind — that is, human cognitive power seen from the theoretical perspective of its open evolution — therefore results from a reflexive dialectic between the Cortex and the Hypercortex.
7.4.4. General structure of the IEML model
In order to provide the Cortex with the mirror in which it will be able to observe its hypercortical image, the scientific theory of the Cortex and the technical plan of the Hypercortex must follow the same general model of the mind. Without structural isomorphism between the functions of the Cortex and those of the Hypercortex, the latter would not reflect actual symbolic cognition and the Cortex would not be enhanced by a new reflexive capacity on the scale of collective intelligence. That is why the IEML model of the mind is transversal to the Cortex and the Hypercortex.
Figure 7.4 shows a dialectic in six sections, with the three lower (actual) sections representing the Cortex and the three upper (virtual) sections representing the Hypercortex. On the left, the two sign (S) sections represent the syntactic function. In the center, the two being (B) sections represent the semantic function. On the right, the two thing (T) sections represent the pragmatic function. Since I have already talked about the dialectic of the Cortex and the Hypercortex and the three functions of the mind in general terms, I will now focus on the three virtual areas of the Hypercortex, i.e. on the automatable representation of the syntactic, semantic and pragmatic functions of the mind. Since the IEML model of the mind must be calculable, I will borrow a metaphor for its general structure from computer science.
Figure 7.4. The IEML model of the mind
S: The function of manipulation of symbols is modeled in an abstract semantic machine, the syntax of IEML, which has been demonstrated to be capable of computing the giant graph of the semantic sphere.
B: The function of manipulation of concepts is modeled in a calculable metalanguage based on the syntax of IEML. The grammar rules and dictionary of this metalanguage function as a linguistic operating system of the machine: they translate the paradigmatic and syntagmatic graphs of the IEML semantic sphere into natural languages.
T: The function of manipulation of data is carried out by the applications of the IEML semantic machine, which are called IEML games or collective interpretation games. These games make it possible to freely organize the digital memory according to the universes of discourse and values of creative conversations: they produce interoperable ecosystems of ideas.
7.4.5. IEML as machine: formal properties
7.4.5.1. Toward a universal semantic calculus
Contemporary computers are already capable of automatically manipulating symbols and data. What has not yet been achieved, in my view, is the capacity to automatically manipulate concepts on a large scale, systematically and interoperably, i.e. across disciplinary, cultural and linguistic differences. The possibility of automatically manipulating concepts using a general method (rather than a multitude of ad hoc methods, as is done today in 2011) would open the way to a universe of automatic manipulation of data according to their meaning. Let us assume that the data would be categorized according to universally calculable semantic metadata. Then a society of automata (interoperable “services”) could interpret and filter the digital data using the mechanism provided by these metadata. The problem thus is to design a method of encoding concepts that would make a universal semantic calculus possible. Such a problem is particularly difficult to solve for two reasons. First, machines are notoriously blind to semantics: computers only “understand” the syntax and formal rules for manipulating symbols. Second, the natural languages in which concepts are normally encoded are irregular: there is no point trying to find a systematic, calculable correspondence between syntactic forms and conceptual meanings. I solved this problem by inventing a symbolic system in which syntactic functions and semantic functions are strictly parallel: all the semantic relationships among concepts encoded in IEML correspond to calculable syntactic relationships among IEML texts.
7.4.5.2. The three modules of the IEML machine
The IEML semantic machine consists of three modules.
– First, a generative syntax produces a regular (in the mathematical sense) language in which each text (each USL) is the variable of a transformation group. This means that texts in the regular IEML language can be produced, recognized and transformed automatically.
– Second, an algorithm uses the grammar rules and multilingual dictionary of the linguistic operating system of IEML to assign a meaning in natural languages to the IEML texts (the USLs).
– Third, all semantic relationships (paradigmatic and syntagmatic) between USLs are calculated automatically using a giant hypercomplex graph in which each node and each link is translated into natural languages: the semantic sphere. Like the USLs that are the nodes and links, the circuits that make up the semantic sphere are the variables of a transformation group. This means that the circuits of the IEML semantic sphere can be produced, recognized and transformed automatically. The formal properties of IEML include a “semantic topology” defining the circuits of the semantic sphere and their transformations7.
As it supports a model of the mind, the IEML semantic machine generates a virtual universe that is practically infinite and inexhaustibly complex. It also supports a scientific model of the mind, so this machine fulfills strict conditions for symmetry and calculability.
7.4.6. IEML as metalanguage: semantic properties
7.4.6.1. STAR: The linguistic operating system of the IEML semantic machine
The IEML syntax, which I have already described, can be considered an automatic writing system, and the IEML semantics, which I will now discuss, may be considered a linguistic interpretation of that writing. The linguistic operating system of the IEML machine is called STAR (Semantic Tool for Augmented Reasoning). The role of STAR, which is no small thing, is to provide the IEML machine with data in natural languages that will enable it to produce the meanings of the USLs. These meanings are based on (i) the paradigmatic relationships among the terms in the multilingual STAR dictionary; and (ii) on the STAR grammar rules that define the syntagmatic relationships among these terms in the USLs. Its mechanical syntax (the semantic machine) and its automatic semantics (STAR) make IEML a calculable metalanguage that can be used as a bridge language between natural languages in the digital medium. Any IEML text can be converted automatically into a semantic network that is readable in natural languages, and vice versa. This means that an IEML text written using an interface in the writer's own language will be able to be read in all the languages supported by the multilingual IEML dictionary (at this time, the IEML dictionary supports only French and English).
7.4.6.2. IEML as a human language
Envisaged as a language, IEML lies at the intersection of human languages and computer languages. Like human languages, it is primarily suited for the expression of signifieds or concepts. As we will see, the structure of IEML has many similarities with that of natural languages. Its grammar has layers of increasing complexity: phonemes, morphemes, words, sentences, texts, etc. Its terms and propositions are distributed in verbal, nominal and auxiliary classes. Finally, its textual units can play many distinct grammatical roles: subject, object, genitive, etc. Unlike other human languages (whether natural or artificial), however, the semantics of IEML is entirely calculable in the form of circuits of paradigmatic and syntagmatic relationships, and its expressions can be the operands of unions, intersections or differences (it is a symmetric transformation group). I should add that the IEML language is not made to be spoken, but rather to be read and written using an interactive computing platform, with access to relevant data, and the availability of all kinds of interfaces8 that this requires.
7.4.6.3. IEML as a computer language
Like computer languages, IEML can be manipulated automatically. IEML is neither a data format (such as PDF, HTML, XML, RDF or OWL) nor a programming language. It is not a data format, because its main purpose is to express concepts; it is a real language, with verbs, nouns, cases, sentences, etc. To clarify: there is a word in IEML for justice: *k.o.-n.o.-'** but there is obviously no translation of the word justice in XML, RDF or OWL, because XML, RDF and OWL are not languages but data formats9. Moreover, IEML can be used with any data format imaginable10. It is not a programming language, since its purpose is not to give instructions to a logical automaton. On the other hand, since IEML is calculable, the texts (USLs) and corresponding semantic circuits can be generated and processed at will, using existing programming languages. However, this does not in any way exclude the possibility that programming languages or user-friendly applications for non computer specialists could be designed especially for the manipulation of IEML texts (USLs) and semantic circuits.
7.4.7. IEML as a universe of games: pragmatic properties
7.4.7.1. The hermeneutic functions and the production of ideas
The main IEML applications are collective interpretation games (CI games). These games automate the production of ecosystems of ideas using hermeneutic functions (see Figure 7.5). As we saw above11, an idea can be modeled by the combination of a concept, an affect and a percept. The collective interpretation games represent the concept by a USL, the percept by a URL and the affect by a semantic current in the circuits corresponding to the USL. The production of ideas can be broken down into two operations: (i) categorization connects a USL to a URL; and (ii) evaluation determines the semantic current. It is understood that the same data (the same URLs) can be categorized and evaluated differently using many different hermeneutic functions.
Figure 7.5. Collective interpretation games and their hermeneutic functions
We can distinguish two major types of hermeneutic functions:
– those whose input variables are data (addressed by their URLs) not categorized in IEML; and
– those whose input variables are ideas categorized in IEML by USLs.
Functions whose input variables are data may be considered functions of perception, and their products, phenomenal ideas (or actual ideas).
Functions whose input variables are ideas may be considered functions of thought, and their products, noumenal ideas (or virtual ideas). A function of thought corresponds to a theoretical or narrative interpretation of phenomenal ideas.
7.4.7.2. The interoperability of IEML games
It should be recalled that USLs are automatically translated into circuits of the semantic sphere by the IEML machine and that all ideas can thus be associated in meta-circuits conducting the semantic current. In addition to their functions of “writing” circuits of ideas on the voluminous mass of data, the CI games will obviously have functions of “reading” enabling searching, filtering and navigating in a hermeneutic memory holding a multitude of ecosystems of ideas. Although their rules may be different, all the CI games are interoperable because they exist in the shared universe woven by the circuits of the semantic sphere. The hermeneutic functions operate on the same types of variables: URLs for the addresses of data, USLs for the addresses of concepts categorizing the data, and flows of current in the semantic circuits corresponding to the USLs to evaluate the data. We can thus imagine game engines that are compatible and capable of calling upon a large number of interoperable hermeneutic services, so that the players can join forces and compose games at will.
The users of the IEML semantic sphere can participate simultaneously in many CI games with different rules of perception and thought. In a sense, each person and each IEML game organizes the semantic sphere and, beyond it, the digital data of the Web, from a distinct perspective. I note once again that good and evil, loss and gain, creation and destruction of value are variables in IEML games. These variables will be defined in different ways by different games. As it is modeled in the IEML semantic sphere, augmented collective intelligence is thus both decompartmentalized (through semantic interoperability) and radically polycentric (because of the existence of an open/multitude of distinct CI games). The digital medium is today fragmented by competition among commercial platforms, difficulties of automatic translation between natural languages and the large number of incompatible systems of metadata and ontologies. The semantic sphere could transform the digital medium into a perspectivist memory, however, a hermeneutic monadology in which creative conversations will be able to interpret each other freely and collaborate effectively without giving up their original points of view.
7.4.7.3. IEML games and knowledge management
The IEML system of semantic coordinates makes it possible to have transparent communicative interaction among individuals, among creative conversations organized in CI games, and between games and individuals. A CI game functions as a social system of knowledge management, an abstract machine that permits the players to produce and collaboratively weave a shared memory in the form of an ecosystem of ideas. The actual individuals, who remain the ultimate sources and destinations of all collective intelligence, will be able to influence the games in which they choose to participate in order to optimally accumulate and use their own memories. “Personal interpretation engines” will permit them to dynamically organize their personal knowledge management12 and guide their learning while participating in different games, thus exploiting the cognitive interoperability provided by the semantic sphere.
7.5. The architecture of the Hypercortex
Figure 7.6 presents the general architecture of the Hypercortex. In the back, ubiquitous multimedia interfaces establish the relationship with the Cortex. On the left, the IEML semantic sphere represents the virtual “wing” of the Hypercortex. On the right, the Internet (the logical sphere) represents its actual “wing”. In the center, creative conversations control and coordinate the activity of the two “wings”. In front, the reflexive consciousness of their collective intelligence is aimed at the synergy of actual collective intelligence and human development.
Figure 7.6. General plan of the hypercortex
7.5.1. The Internet
With the Internet:
– (S) The mechanical heart of the Internet is the global network of interconnected computers: an actual society of automata that manipulate symbols. With its system of physical addressing of local calculators, the Internet functions as a big global logical machine capable of distributing its calculations over all the electronic and optical processors that possess IP addresses.
– (B) Surrounding the Internet's society of automata is the Web's system of physical data addresses (URLs). As it is a universal addressing system, the Web makes it possible to interconnect all data. It makes all digital data into a single multimedia hypertext document that is constantly growing and being reorganized.
– (T) Web applications, the production tools and vehicles for navigating in the voluminous hyperdocument, organize ecosystems of data suited for use in creative conversations.
7.5.2. The IEML semantic sphere
Let us begin by defining the IEML semantic sphere: it is a specialized system of mathematical coordinates for markup and simulation of ecosystems of ideas. By extension, we will consider the semantic sphere to contain everything it organizes. It forms the virtual “wing” that balances the actual “wing” of the Hypercortex:
– (S) This is the heart of the semantic sphere is the IEML machine. It is a society of virtual automata capable of producing, transforming and measuring the circuits of the semantic sphere.
– (B) Surrounding the IEML semantic machine is the system of virtual addressing of the IEML metalanguage: each USL encodes a distinct concept. This system of semantic addressing processes the concepts as variables of a symmetric transformation group and organizes their interconnection in semantic circuits. The concepts form the nodes, and the syntagmatic and paradigmatic relationships among concepts form the links of this gigantic network. Since each concept is translated into natural languages, the IEML metalanguage functions as a bridge language between natural languages and symbolic systems.
– (T) Finally, collective interpretation games enable creative conversations to create and manage their ecosystems of ideas. Ideas are created by folding back the IEML metalanguage onto the Web, i.e. by categorizing data (URLs) using concepts (USLs). The interpretive operation that produces ideas also determines the semantic currents that connect the URL–USL pairs. Out of the interoperable set of ecosystems of ideas emerges a monadological hermeneutic memory in which all semantic perspectives are symmetrical.
7.5.3. Interdependence of the semantic sphere and the Internet
The virtual and actual “wings” of the Hypercortex are in a relationship of dialectical interdependence and mutual reflection. The semantic sphere is dependent on the Internet, because the IEML semantic machine is activated by the actual logic circuits (electronic, optical, etc.) of the Internet, since the USLs have Web addresses and the collective interpretation games are Web applications. It goes without saying that the virtual semantic sphere of IEML needs the processors and storehouses of the actual data of the Internet. At the same time, the semantic sphere operates as an automaton, reading/writing semantic circuits on the fluctuating mass of data. Using this symbolic automaton, creative conversations transform the opaque “grey matter” of the Internet into a Hypercortex capable of reflecting collective intelligence.
7.5.4. New perspectives in computer science and the human sciences
The project of building the Hypercortex implies a significant shift in research and teaching in computer science as it is practiced in the early 21st Century. Since the late 1950s, artificial intelligence (AI) has always been considered the most “advanced” perspective in computer science. The contemporary undertaking of the Web of data can be considered the extension of the AI project to the new environment of the Web. I think research and teaching in AI, which is still indispensable, should be included within the broader perspective of augmented intelligence (collective and reflexive intelligence), of which the Hypercortex is now emblematic.
It will likely be in the human sciences that the most radical questions will be raised. Research and teaching in these fields will sooner or later have to draw conclusions from the following three facts: (i) an enormous mass of data on society and culture is increasingly available ubiquitously in the digital medium; (ii) accessible calculating power is constantly growing; and (iii) with IEML, we now have a theoretical tool that enables us to exploit the growth in online data and calculating power to methodically observe the object of the human sciences, symbolic and social cognition. The Hypercortex should be seen as a project for a great observatory for the humanities and social sciences, comparable to the cyclotron for physicists or the conquest of space for space agencies. Adoption of this new instrument will significantly increase the potential of the “sciences of the mind” while solving the huge problems of knowledge management they face today as a result of their disciplinary and theoretical fragmentation.
7.6. Overview: toward a reflexive collective intelligence
The overview in Figure 7.7 corresponds to the central “vanishing point” in Figure 7.4 and to the “head” in Figure 7.6. I will use it in the rest of this volume as a conceptual map in order to situate the themes of the chapters. The map illustrates the reciprocal dynamics of information between the actual collective intelligence that determines the effective human development of a community (cortical cognition, at the top in the diagram) and the scientific image of this actual intelligence as reflected in the Hypercortex (at the bottom). The symmetrical reflexivity between cortical and hypercortical cognition is organized by creative conversations.
Figure 7.7. Overview
The Hypercortex can be considered an instrument for the scientific observation of collective human intelligence, which belongs to the nature represented in Figure 7.1 and reflects that nature in its own way. The aim of the entire IEML undertaking is to create the conditions for the scientific observation of collective human intelligence. Why? Because only scientific knowledge of this collective intelligence can lead to its systematic, rational augmentation and thus, ultimately, to the acceleration of human development. With respect to the humanist credentials of such a research program, it should be clear that the observation of collective intelligence can only be reflexive observation — i.e. self-observation — guiding the autonomous development of the communities concerned.
With respect to the scientific credentials of the research program based on IEML, the knowledge obtained by hypercortical observation is hypothetical, transparent and calculable. It is hypothetical since it is based on freely chosen interpretation functions, the effects of which can be explored at will and can be challenged at any point. It is transparent since the data on which it is based are public and the functions using these data are explicit and interoperable. Finally, this knowledge is hypothetically calculable, since it is based on a system of coordinates — the semantic sphere — especially designed to simulate human symbolic cognition automatically.
The digital medium, with its flows of data, its distributed calculating power and its ubiquitous multimedia interfaces, provides our observation instrument with its fundamental technical support. The semantic sphere is contained in the digital medium. As we saw above, the formal or mechanical nucleus of the semantic sphere, the IEML semantic machine, calculates its generation, transformations and measurement. The metalinguistic dimension of the semantic sphere makes it a universal system of semantic coordinates, capable of addressing and interconnecting concepts expressed in natural languages. The metalanguage can thus serve as a common semantic grid for a hermeneutic memory in which creative conversations freely drive and maintain their ecosystems of ideas through collective interpretation games. It is precisely these ecosystems of ideas that represent the collective intelligence of creative conversations.
The flight of the Hypercortex is headed toward the vanishing point of a collective human intelligence that is increasingly conscious of itself and its interdependent links with nature.
1 For a computer engineering perspective, see [SAL 2008].
2 On the concept of collective intelligence, see [KAP 2009, KAP 2010, LÉV 1994b, NGU 2009, TOV 2008].
3 It seems to me that my approach is compatible with the so-called “4E” (“Embodied, Embedded, Enactive, Extended”) philosophy of cognition as illustrated by Harry Halpin, Andy Clark and Michael Wheeler in their article “Towards a philosophy of the Web: Representation, enaction, collective intelligence” [HAL 2010].
4 It should be noted that from an etymological perspective, physical nature is a pleonasm, since in Greek, physis actually means nature. In ancient Greek thought, physis (nature) was contrasted with nomos (law, human convention).
5 We will study others in Volume 2 of this book.
6 On the subject of the trivium (grammar, dialectic, rhetoric) as the backbone of the Western intellectual tradition from Ancient Greece until the 16th Century and beyond, see [MAC 1943]. On the parallels between the trivium and the major articulations of semiotics and linguistics, see [RAS 1990].
7 The mathematical definition of semantic circuits as well as the proof of the calculability of their generation, transformation and measurement, will be presented in Volume 2 of this book. Meanwhile, see [LÉV 2010b].
8 These interfaces can be in natural, iconic or visual/tactile languages, 3D simulation, augmented reality, etc.
9 The lack of a distinction in English corresponding to that between the French langue and langage, and the use of the word language to designate data formats may lead to confusion.
10 There is already a parser for IEML expressions, which automatically translates USLs into XML format; see [LÉV 2010d].
11 See section 7.3.3.2.
12 On personal knowledge management, see section 4.2.1.